Recombinant Human Erythropoietin (rhEPO) in Clinical Oncology: Scientific and Clinical Aspects of Anemia in Cancer, 2nd Edition

W 5WPIUUIL:M[I6W_ZW][QIV-L :MKWUJQVIV\0]UIV-Za\PZWXWQM\QVZP-87 QV+TQVQKIT7VKWTWOa ;KQMV\QâKIVL+TQVQ...

Author: Mohammad Resa Nowrousian

47 downloads 545 Views 5MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!

Report copyright / DMCA form

DOWNLOAD PDF

W

5WPIUUIL:M[I6W_ZW][QIV-L

:MKWUJQVIV\0]UIV-Za\PZWXWQM\QVZP-87 QV+TQVQKIT7VKWTWOa ;KQMV\QâKIVL+TQVQKIT)[XMK\[WN )VMUQIQV+IVKMZ ;MKWVL:M^Q[MLIVL-VTIZOML-LQ\QWV

SpringerWienNewYork

Prof. Dr. med. M. R. Nowrousian Department of Internal Medicine (Cancer Research), West German Cancer Center, University Hospital of Essen, Essen, Germany

This work is subject to copyright. All rights are reserved, whether the whole or part of the material is concerned, speciﬁcally those of translation, reprinting, re-use of illustrations, broadcasting, reproduction by photocopying machines or similar means, and storage in data banks. Product Liability: The publisher can give no guarantee for all the information contained in this book. This does also refer to information about drug dosage and application thereof. In every individual case the respective user must check its accuracy by consulting other pharmaceutical literature. The use of registered names, trademarks, etc. in this publication does not imply, even in the absence of a speciﬁc statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use.

© 2002, 2008 Springer-Verlag/Wien Printed in Germany SpringerWienNewYork is a part of Springer Science + Business Media springer.at Typesetting: SNP Best-set Typesetter Ltd., Hong Kong Printing: fgb, Freiburger Graphische Betriebe, 79121 Freiburg, Germany Printed on acid-free and chlorine-free bleached paper SPIN: 11403838

With 122 (partly coloured) Figures Library of Congress Control Number: 2007942571

ISBN 978-3-211-25223-9 SpringerWienNewYork ISBN 3-211-83661-6 1st edition SpringerWienNewYork

Preface Since the introduction of rhEPO in clinical oncology in 1993, considerable insight has been gained into the pathogenesis, prevalence and incidence of anemia in cancer and its impact on the life of cancer patients and the course of their disease. Anemia is not only a frequent complication of cancer and its treatment but also a major factor deteriorating patients’ physical well-being and quality of life (QOL). In addition, it may be involved in the development of tumor resistance against radiotherapy and chemotherapy. A number of studies indicates a close relationship between anemia and tumor hypoxia, and shows that the latter negatively and signiﬁcantly determines the outcome of radiotherapy. Also, there is evidence suggesting that hypoxia stimulates angiogenesis and contributes to the selection of a more malignant phenotype of tumor cells with a reduced sensitivity to irradiation and chemotherapeutic agents. These ﬁndings and the impact of anemia on metabolic and organ function, as well as QOL, identify anemia as a much more serious problem for cancer patients than previously recognized. Over many decades, treatment of anemia in cancer patients has consisted of red blood cell (RBC) transfusions. However, these are of limited value in achieving a sustained and sufﬁcient increase in Hb level. Furthermore, ﬂuctuating Hb levels resulting from repeated RBC transfusions may produce intermittent hypoxia, which is considered an important factor for tumor progression. Acute, cyclic hypoxia has been shown to be even more deleterious than chronic hypoxia in selecting aggressive, apoptosis-resistant tumor cells and promoting metastasis. RBC transfusions are also associated with a number of side effects and risks, such as febrile and allergic reactions, alloimmunization, transmission of infection, iron overload and suppression of cellular immunity, which may be of particular concern in cancer patients. Furthermore, RBC transfusions, because of their risks and limitations as well as shortness of supply, are usually given at Hb levels below 8 g/dl. Numerous studies, including large prospective randomized and nonrandomized trials and meta-analyses of data from these trials have shown that patients with cancer are already seriously affected by the impact of anemia on metabolic and organ functions, exercise capacity and QOL long before such low levels of Hb are present. In addition, they have indicated that by using erythropoiesis-stimulating agents (ESAs), it is possible to achieve sustained physiological and much more effective Hb levels than with RBC transfusions. Furthermore, they have shown that such an increase in Hb level not only reduces or eliminates the need for RBC transfusions, but it is also associated

vi

Preface

with signiﬁcant improvement in exercise capacity, physical well-being and QOL of patients. The majority of these studies was carried out in anemic patients receiving chemotherapy or chemoradiotherapy and the results have identiﬁed ESAs not only as signiﬁcantly effective but also comparatively safe drugs. However, some studies have reported on a slight but signiﬁcantly increased risk of thromboembolic events associated with the use of ESAs. This has to be considered in patients with a history of such events and patients who are receiving thrombogenic treatments. Some other studies have raised the question on the safety of ESAs with regard to the outcome of anticancer treatment. These studies, however, were performed in settings other than currently approved for the use of ESAs in cancer patients and the results have been critically reviewed because of a number of methodological problems, which have been present in these studies, e.g. high initial and/or target Hb levels used and imbalances in the distribution of patients and disease-related factors determining survival between patients receiving ESAs and control patients (Chapters 17–19, 31). Another issue of concern exclusively arising from in vitro studies has been a possible stimulatory effect of ESAs on tumor cells expressing EPO receptors (EPO-R). The results of these studies are in part controversial and have become a subject of critical reviews because of methodological problems associated with the determination of EPO-R and their functionality. In many of these studies, the stimulatory effects of ESAs on tumor cells were marginal and typically achieved with extremely high concentrations of ESAs, which do not occur in the treatment of cancer patients receiving ESAs. In some studies, in-vitro concentrations used were at several magnitudes higher than the peak concentrations observed in serum of patients treated with ESAs (Chapters 3, 4, 17, 18, Addendum 1). Furthermore, numerous animal experiments have failed to show tumor growth promotion by ESAs, even when the tumor cells were expressing EPO-R. In many of these studies, ESA treatment of anemic animals signiﬁcantly improved the results of radiotherapy or chemotherapy. Nevertheless, further preclinical and clinical studies are required to better understand the beneﬁts and risks of treatment with ESAs and to use these drugs with the highest possible beneﬁt and safety for our patients. Like its ﬁrst edition, the second edition of this book aims to be a comprehensive source of information on clinical and scientiﬁc aspects of anemia in cancer and its treatment with ESAs. All chapters were updated and some new chapters were added to achieve a greater spectrum of topics and to include future developments of ESAs in other ﬁelds of clinical medicine, such as neurology and cardiology. Under certain circumstances, e.g. in preventing adverse effects of radiotherapy and chemotherapy, neuroprotective and cardioprotective effects of ESAs may be of value for patients with malignant diseases. It was again a great pleasure and honor for me, and highly appreciated, that outstanding authors, all experts on their topics, agreed to contribute to

Preface

vii

this edition and to comprehensively present the state of knowledge in their ﬁelds. My sincere gratitude is also extended to Mrs. Ch. Wartchow for her help in proof-reading and Mrs. I. Demirer and Mrs. U. Senkel-Nicklaus for their excellent organizational assistance in preparing the book. Prof. Dr. M. R. Nowrousian

Contents

Contributors

xiii

1.

Physiology of erythropoiesis U. Testa

2.

Biology of EPO and EPO-receptor C. Lacombe, and P. Mayeux

67

3.

The role of erythropoietin receptor expression on tumor cells J. Fandrey

81

4.

Problems associated with erythropoietin receptor determination on tumor cells A. Österborg

103

Deﬁnition, classiﬁcation and characterization of anemia in cancer M. R. Nowrousian

117

5.

6.

Pathophysiology of anemia in cancer M. R. Nowrousian

7.

Prevalence and incidence of anemia and risk factors for anemia in patients with cancer H. Ludwig

1

149

189

8.

Signiﬁcance of anemia in cancer chemotherapy M. R. Nowrousian

207

9.

Incidence and impact of anemia in radiation oncology J. Dunst and M. Molls

249

10. Relationship between hemoglobin levels and tumor oxygenation P. Vaupel, A. Mayer and M. Höckel

265

x

Contents

11. Tumor hypoxia and therapeutic resistance P. Vaupel and M. Höckel

283

12. Symptoms of anemia R. Pirker

307

13. Impact of anemia and red blood cell transfusion on organ function M. R. Nowrousian

317

14. Relationship of hemoglobin, fatigue, and quality of life in anemic cancer patients Z. Butt and D. Cella

369

15. When to use red blood cell transfusions in cancer patients with solid tumours? J. K. Jacob and P. J. Barrett-Lee

393

16. Pharmacology, pharmacokinetics and safety of recombinant human erythropoietin preparations W. Jelkmann

407

17. Epoetin treatment of anemia associated with multiple myeloma and non-Hodgkin’s lymphoma A. Österborg

433

18. rhEPO in anemic patients with solid tumors and chemotherapy – efﬁcacy and safety M. R. Nowrousian

449

19. Early intervention with recombinant human erythropoietin for chemotherapy-induced anemia G. H. Lyman and J. Glaspy

509

20. Recombinant human erythropoietin (rhEPO) therapy in myelodysplasia E. Hellström-Lindberg

531

21. Prediction of response to rhEPO in the anemia of cancer Y. Beguin and G. Van Straelen

541

22. rhEPO in hematopoietic stem cell transplantation G. Van Straelen and Y. Beguin

583

Contents

xi

23. Treatment of anemia with rhEPO in radiation oncology J. Dunst

615

24. Recombinant human erythropoietin in pediatric oncology C. Hastings and J. Feusner

635

25. rhEPO in surgical oncology M. J. Fontaine and L. T. Goodnough

663

26. Erythropoiesis, iron metabolism and iron supplementation during erythropoietin therapy L. T. Goodnough 27. Are there risks for use of iron in cancer patients? P. Gascón 28. Metabolic and physiologic effects of rhEPO in anemic cancer patients K. Lundholm and P. Daneryd 29. Effects of rhEPO on quality of life in anemic cancer patients S. Chowdhury, J. F. Spicer, and P. G. Harper 30. Thrombosis during therapy with erythropoiesis stimulating agents in cancer J. Glaspy 31. The effect of rhEPO on survival in anemic cancer patients T. J. Littlewood 32. From bench to bedside: Neuroprotective effects of erythropoietin H. Ehrenreich and C. Bartels

679

703

713

729

745

759

771

33. rhEPO in patients with anemia and congestive heart failure D. S. Silverberg, D. Wexler, A. Iaina, S. Steinbruch, Y. Wollman, and D. Schwartz

793

34. Cost-effectiveness of treating cancer anaemia P. Cornes

813

Addendum

851

Index

853

Contributors Barrett-Lee Peter J., Dr. Velindre Cancer Centre, Velindre NHS Trust, Whitchurch, Cardiff, Wales CF14 2TL, UK E-mail: [email protected] Bartels Claudia, Dipl.-Psych. Division of Clinical Neuroscience, Max-Planck-Institute of Experimental Medicine, Hermann-Rein Strasse 3, 37075 Göttingen, Germany E-mail: [email protected] Beguin Yves, MD, Prof. University of Liège, Department of Hematology, CHU Sart-Tilman, 4000 Liège, Belgium E-mail:[email protected] Tel. +32-4-366-7690 Fax +32-4-366-8855 Butt Zeeshan, PhD Center on Outcomes, Research and Education, Evanston Northwestern Healthcare, 1000 University Place, Suite 100, Evanston, Illinois 60201, USA E-mail: [email protected] Cella David, Dr. Center on Outcomes, Research and Education, Evanston Northwestern Healthcare, 1000 University Place, Suite 100, Evanston, Illinois 60201, USA E-mail: [email protected] Tel. +1-847-570-7370 +1-847-570-8033 Chowdhury Simon, MA, MRCP, PhD Medical Oncology, 3rd Floor Thomas Guy House, Guy’s Hospital, St Thomas Street, London SE1 9RT, UK E-mail: [email protected] Tel. +44-207-955-2933 Fax +44-207-955-4939

xiv

Contributors

Cornes Paul, Dr. Bristol Haematology & Oncology Centre, 9 Royal Victoria Park, Bristol, BS2 8ED, United Kingdom E-mail: [email protected] Tel. +44-117-928-3008 Fax +44-117-928-4409 Daneryd Peter, MD, PhD, Prof Department of Surgery, Sahlgrenska University Hospital, 413 45 Göteborg, Sweden Dunst Jürgen, Prof. Dr. Department of Radiation Oncology, University Clinic Schleswig-Holstein, Campus Lübeck, Ratzeburger Allee 160, 23538 Lübeck, Germany E-mail: [email protected] Tel. +49-451-500-6660 Fax +49-451-500-33324 Ehrenreich Hannelore, MD, DVM, Prof. Dr. Head, Division of Clinical Neuroscience, Max-Planck-Institute of Experimental Medicine, Hermann-Rein-Strasse 3, 37075 Göttingen, Germany E-mail: [email protected] Tel. +49-551-3899628 Fax +49-551-3899670 Fandrey Joachim, Prof. Dr. med. Institut für Physiologie, Universitätsklinikum Essen, Universität DuisburgEssen, Hufelandstrasse 55, 45147 Essen, Germany E-mail: [email protected] Tel. +49-201-723-4600 Fax +49-201-723-4648 Feusner James, MD, Prof. Director, Pediatric Oncology, Department of Hematology and Oncology, Children’s Hospital and Research Center of Oakland, 747 52nd Street, Oakland, CA 94609, USA E-mail: [email protected] Tel. +1-510-428-3689 Fax +1-510-601-3916 Fontaine Magali J., MD, PhD Assistant Professor of Pathology, Associate Director of Transfusion Services, 300 Pasteur Drive L235, Stanford CA 94305, USA E-mail: [email protected] Tel. +1-650-450-1459 Fax +1-650-723-9178

Contributors

xv

Gascón Pere, Prof. Dr. Division of Medical Oncology, Hematology-Oncology Department (ICMHO), Hospital Clínic, Villarroel 170, escalera 2, planta 5, Barcelona, Spain E-mail: [email protected] Tel. 34-93-2275402 Fax 34-93-4546520 Glaspy John, MD, Prof. Professor of Medicine, Division of Hematology-Oncology, UCLA School of Medicine, Los Angeles, California 90095, USA E-mail: [email protected] Goodnough Lawrence T., Prof. Dr. Washington University School of Medicine, Transfusion Services, Barnes – Jewish Hospital, 660 South Euclid Avenue, Box 8118, St. Louis, MO 63110, USA E-mail: [email protected] Tel. +1-314-362-3186 +1-314-362-1461 Harper, Peter G., Dr. Medical Oncology, 3rd Floor Thomas Guy House, Guy’s Hospital, St Thomas Street, London SE1 9RT, UK E-mail: [email protected] Tel. +44-207-955-2933 Fax +44-207-955-4939 Hastings Caroline, MD Department of Pediatric Hematology and Oncology, Children’s Hospital and Research Center of Oakland, 747 52nd Street, Oakland, CA 94609, USA E-mail: [email protected] Tel. +1-510-428-3631 Fax +1-510-601-3916 Hellström-Lindberg Eva, MD, PhD Karolinska Institutet, Department of Medicine, Division of Hematology, Karolinska University Hospital, Huddinge, Ihn 86, 141 86 Stockholm, Sweden E-mail: [email protected] Höckel Michael, Prof. Dr. Department of Obstetrics and Gynecology, University of Leipzig, PhilippRosenthal-Strasse 55, 04103 Leipzig, Germany E-mail: [email protected] Tel. +49-341-9723-400 Fax +49-341-9723-419

xvi

Contributors

Iaina Adrian, MD Department of Nephrology, Tel Aviv Medical Center, Weizman 6, Tel Aviv 64239, Israel Jacob Jayin K., Dr. Velindre Cancer Centre, Velindre NHS Trust, Whitchurch, Cardiff, Wales CF14 2TL, UK E-mail: [email protected] Jelkmann Wolfgang, Prof. Dr. Director, Professor of Physiology, Institute of Physiology, Medical University of Luebeck, Ratzeburger Allee 160, 23538 Luebeck, Germany E-mail: [email protected] Tel. +49-451-500-4152 Fax +49-451-500-4151 Lacombe Catherine, MD, PhD, Prof. Unité 363, Université René Descartes Hematology Department, Hôpital Cochin, 27 rue du Faubourg St. Jacques, 75014 Paris, France E-mail: [email protected] Tex. +33-1-4633-1409 Fax +33-1-4633-9297 Littlewood Tim, Dr. Department of Haematology, John Radcliffe Hospital, Oxford OX3 9DU, UK E-mail: [email protected] Tel. +44-1865-220364 and +44-1865-741166 Fax +44-1865-221778 Ludwig Heinz, Prof. Dr. Department of Medicine I, Center for Oncology and Hematology, Montleartstrasse 37, 1171 Vienna, Austria E-mail: [email protected] Tel. +43-1-49150-2101 Fax +43-1-49150-2109 Lundholm Kent, MD, PhD, Prof. Department of Surgery, Sahlgrenska University Hospital, 413 45 Göteborg, Sweden E-mail: [email protected] Tel. +46-31-342-2239 Fax +46-31-413-892

Contributors

xvii

Lyman Gary H., MD, MPH, FRCP (Edin) Duke University School of Medicine, Duke Comprehensive Cancer Center, Durham, North Carolina, USA E-mail: [email protected] Tel. +1-585-275-3335 Fax +1-585-276-1885 Mayer Arnulf, Dr. Institute of Physiology and Pathophysiology, University of Mainz, Duesbergweg 6, 55099 Mainz, Germany E-mail: [email protected] Tel. +49-6131-392-5203 Fax +49-6131-392-5774 Mayeux P., Dr. Université Paris-Descartes, Faculté de Médecine, Unité 363, Service d’Hématologie, INSERM, CNRS, Hôspital Cochin (U567-UMR 8104), Paris, France E-mail: [email protected] Tel. +33-1-0140516514 Fax +33-1-0140516510 Molls M., Univ.-Prof. Dr. med. Department of Radiation Oncology, Technical Ismaninger Straße 22, 81675 München, Germany E-mail: [email protected] Tel. +49-89-414-04502 Fax +49-89-414-04477

University

Munich,

Nowrousian M. R., Prof. Dr. Department of Internal Medicine (Cancer Research), West German Cancer Center, University Hospital of Essen, Hufelandstrasse 55, 45122 Essen, Germany E-mail: [email protected] Tel. +49-201-723-3127 Fax +49-201-723-5984 Österborg Anders, Dr. Department of Oncology (Radiumhemmet), Karolinska University Hospital, 171 76 Stockholm, Sweden E-mail: [email protected] Tel. +46-85-177-3385 Pirker Robert, MD, Prof. Department of Internal Medicine I, Medical University of Vienna, Währinger Gürtel 18-20, 1090 Vienna, Austria E-mail: [email protected] Tel. +43-1-40400-4422 Fax +43-1-40400-4461

xviii

Contributors

Schwartz Doron Department of Nephrology, Tel Aviv Medical Center, Weizman 6, Tel Aviv 64239, Israel Silverberg Donald S., MD Department of Nephrology, Tel Aviv Medical Center, Weizman 6, Tel Aviv 64239, Israel E-mail: [email protected] Tel. Ofﬁce 972-3-6973270 Tel. Home 972-9-8666013 Tel. Cellular 972-0522-555412 Fax: 972-9-8665715 Spicer James F., MA, MRCP, PhD Medical Oncology, 3rd Floor Thomas Guy House, Guy’s Hospital, St Thomas Street, London SE1 9RT, UK Tel. +44-20-7188-7188 Steinbruch S. Department of Nephrology, Tel Aviv Medical Center, Weizman 6, Tel Aviv 64239, Israel Testa Ugo, Dr. Department of Hematology and Oncology, Istituto Superiore di Sanitá, Viale Regina Elena 299, 00161 Roma, Italy E-mail: [email protected] Van Straelen Gaetan University of Liège, Department of Hematology, CHU Sart-Tilman, 4000 Liège, Belgium Vaupel Peter, Prof. Dr. Institute of Physiology and Pathophysiology, University of Mainz, Duesbergweg 6, 55099 Mainz, Germany E-mail: [email protected] Tel. +49-6131-392-5929 Fax +49-6131-392-5774 Wexler Dov, MD Department of Cardiology and Heart Failure Clinic, Tel Aviv Medical Center, Weizman 6, Tel Aviv 64239, Israel E-mail: [email protected] Wollman Y. Department of Nephrology, Tel Aviv Medical Center, Weizman 6, Tel Aviv 64239, Israel

Chapter 1

Physiology of erythropoiesis U. Testa Department of Hematology and Oncology, Istituto Superiore di Sanità, Rome, Italy

Ontogenesis of erythropoiesis The hematopoietic system in vertebrates requires the presence of cells that ensure a continuous production of new cells needed to replace mature blood elements endowed with only a limited life span. This continual blood cell production is ensured by pluripotent stem cells which have the unique property of both self-renewal and differentiation through progressive commitment to multipotent progenitors, then to committed progenitors and, ﬁnally, to progressively maturing precursors of the different hemopoietic lineages. Commitment of stem cells to the speciﬁc cell lineages appears not to be regulated by exogenous growth factors. Rather, stem cells develop into differentiated cell types through incompletely deﬁned molecular events that are intrinsic to the stem cell itself. Following lineage commitment, hematopoietic progenitors and precursors come increasingly under the regulatory inﬂuence of growth factors and hormones. The hematopoietic cells and, particularly, the primitive embryonic erythroid cells originate from the ventral mesoderm through a differentiation process under the control of bone morphogenetic proteins (BMP). Particularly, the effects of BMP seem to be related to the induction of the expression of hematopoietic transcription factors, such as GATA-1, GATA-2, Tal-1, LMO2 and EKLF in mesodermic progenitors (Adelman et al. 2002; Schmerer and Evans 2003). The differentiation of embryonic stem cells to primitive erythroid elements is promoted by vascular endothelial growth factor (VEGF) (Cerdan et al. 2004). In the mouse (and also in humans) during embryonic/fetal development three distinct stages of erythropoiesis have been deﬁned. The ﬁrst stage corresponds to primitive embryonic erythropoiesis and occurs at the level of blood islands developing within the yolk sac around day 7 of murine embryonic life; the distinctive feature of erythropoiesis at this stage corresponds to the erythropoietin (EPO)-independency (Wu et al. 1995). The initial erythroid differentiation from pluripotent embryonic stem cells requires a mesoderm patterning factor, such as bone morphogenetic factor 4 and an

2

U. Testa

angiogenic factor, like VEGF. Primitive erythroblasts differentiate within the bloodstream, remain predominantly nucleated at the end of their maturation and have a very large size (accordingly, they are called “megaloblasts”). The second stage of erythropoietic development corresponding to deﬁnitive erythropoiesis, occurs around day 10 of gestation and consists in a rapid and marked proliferation of erythroid elements in the fetal liver; the distinctive feature of this erythropoietic stage consists in the development of EPO, EPO receptor (EPO-R) and JAK2 dependency (Neubauer et al. 1998; Parganas et al. 1998). The third stage of erythropoietic development is characterized by the migration of hematopoietic stem cells at the level of bone marrow, where adult hemopoiesis and, particularly, adult erythropoiesis develops. A distinctive feature of adult erythropoiesis consists in its negative regulation mediated by SHP-1 phosphatase (Klingmuller et al. 1995). The human hematopoiesis begins in the second-to-third embryonic weeks with formation of mesoderm-derived blood islands in the extraembryonic mesoderm of the developing secondary yolk sac. Blood islands develop foci of nucleated erythroblasts (known for their large size as “megaloblasts”) intimately associated with and surrounded by endothelium. Yolk sac blood cells consist of nucleated primitive erythrocytes synthesizing exclusively embryonic hemoglobins (like ζ2ε2). After the onset of circulation occurring around the day 21 of development, yolk sac cells are found in embryonic blood. The fetal liver subsequently replaces the yolk sac as the main hematopoietic tissue with the appearance of deﬁnitive enucleate, macrocytic erythrocytes synthesizing fetal hemoglobin (α2γ2). Recent in vitro studies have provided evidence that hematopoiesis and particularly primitive erythropoiesis originates from embryonic bodies-derived endothelial progenitors endowed with both endothelial and hemopoietic differentiation capacities (hemangioblasts) (Wang V et al. 2004; Zambidis et al. 2005). Erythropoiesis involves the progressive differentiation starting from hemopoietic stem cells to mature erythrocytes. The differentiative steps involved in this complex differentiation process are numerous and involve ﬁrst the differentiation to multipotent hemopoietic progenitors generating mixed colonies in vitro (CFU-Mix), then to committed erythroid progenitors subdivided in early erythroid progenitors (burst forming unit-erythroid, BFU-E) generating in vitro large erythroid colonies and late erythroid progenitors (colony forming unit-erythroid, CFU-E), generating small erythroid colonies and, ﬁnally, to morphologically recognizable erythroid precursors. During this differentiation process the cells become progressively sensitive to EPO due to the appearance on these cells of the EPO-Rs. The terminal stages of bone marrow erythropoiesis occur in peculiar cellular associations named erythroblastic islands consisting of a centrally located macrophage surrounded by maturing erythroid precursors. The erythroblastic islands are considered as morpho-functional units of erythropoiesis where differentiation of CFU-E takes place (Bernard 1991). The formation of erythroblastic islands occurs in a region away from the

Physiology of erythropoiesis

3

sinusoidal endothelium and the erythroblastic islands migrate toward the sinusoids as erythroid maturation proceeds (Yokoyama et al. 2003). It was suggested that macrophages/stromal cells present in the erythroblastic islands provide an essential microenvironment for maturing erythroblasts, releasing various cytokines, and maintaining intimate contacts with erythroblasts until enucleation and release of reticulocytes. The adhesion of erythroblasts to the extracellular matrix and to the central macrophage has been shown to be important for erythroid cell differentiation and involves several adhesion and extracellular matrix molecules, such as the erythroblast membrane receptor, integrin α4β1 and its ligand VCAM-1, integrin-associated protein, ﬁbrinonectin and erythroid-speciﬁc adhesion receptor membrane protein. In this context, particularly relevant seems to be the role of the erythroid-speciﬁc intercellular adhesion molecule-4 (ICAM-4). As a binding partner of both α4β1 and αV integrins, ICAM-4 could play a multifunctional role within the erythroblastic islands: ICAM-4 – α4β1 association may mediate adhesion between adjacent erythroblasts, while ICAM-4 – αV interaction may affect binding of erythroblasts to the central macrophage (Lee et al. 2003). The compartment of bone marrow erythroid precursors is well characterized and involves a proliferative-maturative compartment, involving the maturation from proerythroblasts to polychromatophilic erythroblasts through the basophilic stage, and a maturative non-proliferative compartment involving the maturation from polychromatophilic erythroblasts to mature red blood cells through the reticulocyte stage. Most red blood cells (RBC) are released into circulation as reticulocytes and mature further over the next days to become erythrocytes. During this maturation process they lose their mitochondria and ribosomes. Consequently, they lose the ability to synthesize hemoglobin and to carry out oxidative metabolism. The end cells, erythrocytes, are uniform biconcave disks lacking a nucleus and organelles and composed in large part by a highly specialized protein, hemoglobin. The mature RBC rely on glucose and the glycolytic pathway for their metabolic needs, including the production of large amounts of 2,3-diphosphoglycerate (2,3-DPG) that reduced the afﬁnity of the hemoglobin for oxygen, thereby facilitating the release of oxygen at the tissues. Mature RBC have a relatively long life span (about 120 days), but a continuous bone marrow erythropoiesis is required to replace the constant destruction of senescent RBC. The bone marrow erythropoiesis is a highly efﬁcient system that tunes the rate of erythropoietic production to physiologic needs. In this system the physiologic stimulus is represented by hypoxia that represents the signal inducing the production of EPO, the main cytokine involved in the control of erythroid production. In spite of the efﬁciency of the erythropoietic system, a small, but signiﬁcant level of inefﬁcient erythropoiesis due to premature intramedullary death of erythroid cells occurs. Recent studies indicate that this inefﬁcient erythropoiesis is due to apoptotic death of erythroblasts. The main biochemical and molecular events occurring during the process of erythroid differentiation and maturation are brieﬂy outlined in Fig. 1.

4

U. Testa

Physiology of erythropoiesis

5

Fig. 1. Outline of the process of erythroid differentiation. In the top panel the main steps of erythroid cell differentiation are schematically represented. Initially, through a series of complex differentiation events hemopoietic stem cells (HSC) differentiate to early erythroid progenitors, called burst forming units-erythroid (BFU-E). BFU-Es generate in vitro within two weeks of culture large colonies composed by mature erythroid elements. BFU-Es in turn differentiate into late erythroid progenitors, colony forming unit-erythroid (CFU-Es), able to generate in vitro small erythroid colonies after one week of culture. CFU-Es undergo mitotic divisions to differentiate into mature erythroid cells through several maturation steps represented by proerythroblasts, basophilic erythroblasts, polychromatophilic erythroblasts and othochromatic erythroblasts. Finally, othrochromatic erythroblasts undergo terminal maturation by extruding their nuclei and progressively loosing their organelles, thus generating reticulocytes and then erythrocytes. In the other four panels the kinetics of expression of several molecules during erythroid differentiation is shown. These molecules are divided, according to their function, in three main groups represented by early differentiation antigens, transcription factors, growth factor receptors and proteins important for erythroid cell structure and/or function

䉳

BFU-E progenitors take about 14 days to differentiate in vitro forming large colonies; CFU-E progenitors take about 7 days to form colonies in vitro. CFU-E progenitors undergo 3–5 mitotic divisions to differentiate to terminal erythroid elements through different maturation steps: proerythroblasts, basophilic erythroblasts, polychromatophilic erythroblasts and orthochromatic erythroblasts. Finally, these cells undergo terminal maturation by extruding their nuclei and progressively losing all organelles: as a consequence of these processes young erythrocytes, called reticulocytes, are formed. As erythroid precursors progress in their maturation, they undergo a progressive decrease in their size, increase in hemoglobin concentration and chromatin density. In vitro studies have shown that the erythroid differentiation process from BFU-E to mature RBC takes place in about 2 weeks.

Basic mechanisms in the control of erythropoiesis In mammals, oxygen (O2) is transported to tissues bound to the hemoglobin contained within erythrocytes. The mature RBC is an anucleated cell, discoid in shape, 8 μM in diameter and extremely pliable and therefore capable to traverse the microcirculation. Normal RBC production results in the daily replacement of 0.8 to 1% of all circulating RBC in the body. The rate of RBC production is tuned to the physiologic requirement for these cells. The ensemble of the machinery responsible for RBC production is called the erythron. The erythron is composed by a pool of proliferating elements represented by bone marrow erythroid precursors and by a large mass of

6

U. Testa

nonproliferating cellular elements represented by mature RBC. The size of RBC mass represents the balance of RBC production and destruction. The rate of RBC production is directly related to the mass of bone marrow erythroid precursors. The rate of RBC production is regulated through a feedback mechanism, where the triggering mechanism is represented by hypoxia. When the RBC volume decreases below normal, the erythropoietic activity of bone marrow increases, and conversely, when the RBC volume increases above normal, the erythropoietic activity of the bone marrow decreases. These adjustments are brought about by changes in the circulating level of EPO, a circulating cytokine that contains 165 amino acid residues and four oligosaccharide chains, required for its activity in vivo. The existence of this cytokine was postulated since 1906 by Carnot and DeFlandre who, based on their experiments of induction of reticulocytosis by injection of plasma derived from a bled animal to a normal recipient rabbit, hypothesized the existence of a humoral factor, “hemopoietine”, regulating RBC production (Carnot and DeFlandre 1906). The biological activity of EPO was subsequently characterized and in 1977 human EPO was puriﬁed at homogeneity and its structure determined in detail (Miyake et al. 1985). The availability of the primary sequence of the protein allowed the cloning of the gene for EPO and the development of a transfected cell line in Chinese hamster ovary cells that provided recombinant EPO for therapeutic purposes (Jacobs et al. 1985; Lin et al. 1985). EPO is a member of the family of class I cytokines which fold into a compact globular structure consisting of 4 α-helical bundles. Its molecular mass is 30.4 kDa, although it migrates with an apparent molecular weight of 34–38 kDa on SDS-polyacrylamide gels. The peptide core of the protein is formed by 165 amino acids and is sufﬁcient for receptor binding and for the in vitro capacity to stimulate erythropoiesis, while the carbohydrate moiety (about 40% of the total molecule) is required for the in vivo survival of the cytokine. The function of the four carbohydrate chains of EPO was studied in detail: the 3 complex-type N-linked oligosaccharides at asparagines 24, 38 and 83 play a key role in the stability of the protein in circulation, while the small O-linked oligosaccharide at serine 126 seems to be devoid of any functional role. The development of speciﬁc and sensitive immunologic methods to measure EPO levels in biologic ﬂuids clearly showed that the plasmatic levels of this cytokine are inversely related to the RBC mass (i.e. hematocrit) (Erslev 1991). The normal plasmatic level of EPO ranges from 10 to 25 mU/mL. When hemoglobin levels fall below 100 to 120 g/L plasma EPO levels increase in proportion to the severity of the anemia (Fig. 2). In adults the large majority (about 85%) of the EPO is synthesized at the level of the kidneys and a minority (about 15%) in the liver. EPO can be extracted also from the salivary glands and the spleen; however, since both

Physiology of erythropoiesis

7

Fig. 2. Feedback mechanism regulating the rate of EPO production. This feedback mechanism is based on the levels of circulating hemoglobin which determine the blood oxygen tension. The level of blood oxygen is sensed by the kidney through a molecular oxygen sensing mechanism. A decrease in blood oxygen level determines an increased rate of kidney EPO production. EPO, in turn, stimulates bone marrow erythropoiesis, with a consequent increased production of red blood cells

8

U. Testa

these tissues do not possess the mRNA encoding for EPO and therefore they do not appear to synthesize this cytokine. During fetal and neonatal life the main site of erythropoiesis and EPO production is the liver. When erythropoiesis, during the fetal development, is taken on by the bone marrow the major site of EPO production is switched at the level of kidneys. When renal mass is reduced in adults by chronic renal diseases or nephrectomy, the liver cannot compensate for EPO production and anemia develops. In adults, EPO is produced by interstitial cells in the peritubular capillary bed of the kidneys (Koury et al. 1988; Fisher et al. 1996) and by perivenous hepatocytes in the liver (Zanjani et al. 1977; Koury et al. 1991). EPO is also produced in the brain where it exerts a cytoprotective effect against oxidative damage caused by hypoxic conditions. Finally, in the uterus and oviducts EPO production is induced by estrogens and contributes together with other growth factors to the development of estrogen-induced angiogenesis. The fundamental stimulus for EPO production is the availability of O2 for metabolic needs. Impaired O2 delivery to the kidney can result from a decreased RBC mass (anemia), impaired O2 loading of the hemoglobin molecule (hypoxemia) or, rarely, impaired blood ﬂow to the kidney (renal artery stenosis). The molecular mechanism involved in the control of EPO production has been in part elucidated. The main stimulus for EPO production is hypoxia. The regulation of EPO expression by hypoxia is a complex phenomenon that implies several molecular steps (Fig. 3): in a ﬁrst step hypoxia factor 1α (HIF-1α) is induced and stabilized; in a second step HIF-1α translocates from the cytoplasm to the nucleus, where it dimerizes with the arylhydrocarbon receptor nuclear translocator (HIF-1β); in a third step the HIF-1α/HIF-1β complex binds the hypoxia-responsive enhancer located in the 3′ region of the EPO gene (Michiels et al. 2002). This enhancer element is a 50-bp hypoxia-inducible enhancer that is located approximately 120 bp 3′ to the polyadenylation site (Huang and Bunn 2003). This enhancer contains a nucleotide sequence required for the binding of HIF. Many studies have been devoted to deﬁne the molecular mechanisms involved in HIF-1α activation/stabilization. On the basis of the experimental evidence so far accumulated, two models of HIF-1α activation have been proposed. The ﬁrst model is based on the observation that under normoxic conditions, HIF-1α protein is hydroxylated at its prolyl residues located in the oxygen-dependent domain by a family of prolyl hydroxylase enzymes (Ivan et al. 2001; Jaakkola et al. 2001). Hydroxylation of HIF-1α initiates binding of the Von Hippel-Lindau (VHL) protein that acts as an E3 ubiquitin ligase that then promotes the degradation of HIF-1α by proteosomes. The activity of these prolyl hydroxylases requires oxygen, ascorbic acid and iron. Therefore, prolyl hydroxylases are inhibited by hypoxia, cobaltous ions and iron chelators, explaining why HIF-1α is activated by hypoxia, transition metals, and by iron chelation (Epstein et al. 2001).

Physiology of erythropoiesis

9

Fig. 3. Mechanism of oxygen sensing through the hydroxylation of the transcription factor HIF-1α. HIF-1α protein domain structures are shown schematically: basic helix-loop-helix (bHLH), Per/Arnt/Sim (PAS), N-terminal transactivation (NAD) and C-terminal transactivation (CAD). The NAD domain is also called the oxygen dependent degradation domain (ODD) and represents the site of binding of pVHL. Under low oxygen conditions HIF activates transcription acting in combination with coactivators p300 and CBP. At these oxygen concentrations, von Hippel-Lindau tumor suppressor protein (pVHL) cannot bind to HIF. Under normal or high oxygen conditions, and Fe2+ and 2-oxoglutarate, HIF-1α is hydroxylated by FIH (Factor Inhibiting HIF) at the level of aminoacid residue Asn 803, thus preventing coactivator recruitment at the level of CAD and reducing gene activation. At high oxygen concentrations, the NAD is hydroxylated at the level of two prolyl sites: this event determines the binding of pVHL to HIF with its subsequent proteolytic degradation

10

U. Testa

VHL recognizes HIF-α subunits through two independent sites that contain the common motif LXXLAP: the hydroxylation of speciﬁc proline residues into this conserved box is required for the interaction of VHL with HIF. In addition to prolyl hydroxylases, also an asparaginyl hydroxylase seems to play a key role in the mechanism of control of HIF activity, via hydroxylation of a saparaginyl residue present in the C-terminal transactivation domain. Factor-inhibiting HIF (FIH) has been identiﬁed as the asparaginyl hydroxylase involved in this hydroxylation (Lando et al. 2002). Asparagine hydroxylation prevents the interaction of HIF-1α with the coactivators CREB-binding protein (CBP) and p300. These co-activators of the transcription, CBP and p300, are recruited by the activated HIF-1α protein and are transported by HIF at the level of the EPO hypoxia enhanxcer element, cooperating in the stimulation of the expression of this gene. The second model is based on the observation that hypoxia leads to a decreased production of reactive oxygen species, due to inhibition of nicotinamide adenine nucleotide phosphate oxidase activity. The fast diffusing reactive oxygen species oxidize and destabilize HIF-1α protein resulting in a decreased expression of hypoxia-inducible genes, including EPO. The decrease in superoxide production after hypoxia leads to HIF-1α stabilization and activation (Ehleben et al. 1997). In line with this model, extracellular superoxide dismutase acts as a repressor of hypoxia-induced EPO gene expression, thus implicating superoxide as a signaling intermediate in HIF-1α activation (Zelko and Folz 2005). There are three HIFα family members (HIF-1α, HIF-2α, HIF-3α) and three β family members (HIF-1β/ARNT1, HIF-2β/ARNT2, HIF3β/ARNT3). The structure of a HIFα protein implies the presence of four domains: basic helix-loop-helix (bHLH), Per/ARNT/Sim(PAS), N-terminal transactivation (NTAD) and C-terminal transactivation (CTAD). The NTAD and CTAD domains are involved in activation of transcription when bound to DNA in complex with a β subunit. Over 100 HIF target genes have been identiﬁed and, typically, they contain a canonical HIF-binding site 5′-RCGTG-3′. These target genes include genes that regulate erythropoiesis, angiogenesis, apoptosis and glucose uptake and metabolism. HIFβ proteins are not affected by changes in oxygen level. In contrast, HIFα proteins are highly unstable in the presence of high oxygen level owing to polyubiquitinylation and proteosomal destruction. Under low oxygen levels, HIFα subunits stabilize, translocate to the nucleus and activate transcription acting in concert with a β family member. In addition to HIF members, another protein, the von Hippel-Lindau tumor suppressor protein (pVHL), plays an essential role in oxygendependent regulation of EPO. The mutations of the gene encoding pVHL are responsible for a hereditary cancer syndrome: tumor cells lacking pVHL do not degrade HIFα subunits in the presence of O2 and then they overexpress a series of HIF-responsive genes, including VEGF and EPO (Gnarra et al. 1996). Interestingly, in two types of tumors (hemangioblastoma and

Physiology of erythropoiesis

11

renal cell carcinoma), associated with Von Hippel-Lindau disease, the coexpression of EPO and EPO-R was observed (Vortmeyer et al. 2003; Lee et al. 2005). pVHL forms a molecular complex with elongins, cullin and Ring-box1: this complex possesses ubiquitin ligase activity and polyubiquitinylates HIFα subunits in the presence of oxygen. pVHL binds to the NTAD region (also called the oxygen dependent degradation domain, ODD) of HIFα, when this region, at high O2 level, is hydroxylated in correspondence of proline residues 402 and 564. Germline mutations of the pVHL gene are associated in some rare patients to congenital polycythemias (Gordeuk et al. 2005). Targeted disruption of HIF members provided important information on their biologic role in hematopoiesis. Embryos lacking HIF-1β have defects in yolk sac vascularization and a global defect in hematopoiesis: the hematopoietic defect seems to be dependent upon absent VEGF production and cannot be corrected by exogenous EPO (Adelman et al. 1999). Mice lacking completely the HIF-1α expression die due to multiorgan malformations and developmental arrest. Examination of HIF-1α−/− mice showed the incapacity to stimulate EPO production following a chronic hypoxic stimulus, thus suggesting a role for HIF-1α in EPO production. Analysis of the in vitro differentiation of embryonic stem cells HIF-1α−/− showed a defective erythropoiesis. Finally, mice lacking HIF-2α exhibit pancytopenia. The defective hematopoietic differentiation is related to an altered bone marrow microenvironment. This defect in hematopoiesis, and particularly in erythropoiesis, cannot be rescued by addition of VEGF, but is related to a defective EPO production (Scortegnagna et al. 2005). In addition to its key role in the control of EPO production, HIF-1 plays an essential role also in the modulation of iron metabolism. In fact, among the HIF-1 targets are the genes encoding transferrin, transferrin receptor, heme oxygenase-1, erythroid 5-aminolevulinate synthase and ceruloplasmin which coordinately regulate iron metabolism (Lee PJ et al. 1997; Rolfs et al. 1997; Lok and Ponka 1999; Tacchini et al. 1999; Mukhopadhyay et al. 2000). Increased iron uptake, release of iron from the liver, plasma transport and uptake in the bone marrow are essential to sustain the erythropoietic function of EPO. Iron deﬁciency is known to induce EPO gene expression and HIF-1α protein stabilization (Wang and Semenza 1993), a phenomenon seemingly related to inactivation of the iron-dependent protein hydroxylases PHD1 to 3 and FIH. On the other hand, copper stabilizes nuclear HIF-1α under normoxic conditions, through a mechanism independent on the iron concentration (Martin et al. 2005). According to these observations it was proposed that HIFdependent gene regulation plays a key role in the regulatory network for oxygen, iron and copper metabolism, regulating the oxygen-, iron- and copperbinding transport proteins hemoglobin, transferrin and ceruloplasmin. In addition to HIF-1α other transcription factors play an important role in the modulation of EPO gene expression. Thus, the transcription factor GATA-2 lowers EPO gene transcription by binding to the EPO gene promoter under normoxic conditions (Imagawa et al. 1994; Imagawa et al. 1997).

12

U. Testa

The GATA-2 mediated inhibition of EPO gene expression could be responsible for the reduced EPO synthesis observed in patients with chronic renal failure: in fact the NO synthase inhibitor, NG-monomethyl-L-arginine (LNMMA), one of the candidate molecules responsible for the suppression of EPO synthesis observed in patients with chronic renal failure, lowers EPO synthesis by stimulating GATA-2 DNA binding (Tarumoto et al. 2000). The EPO gene 5′ ﬂanking sequences contain also binding sites for the kB nuclear factor (NFkB). Both GATA-2 and NFkB are involved in the inhibition of EPO gene expression observed in inﬂammatory conditions. The proinﬂammatory cytokines interleukin-1 (IL-1) and tumor necrosis factor-α (TNF-α) activate both GATA-2 and NFkB and, through this mechanism, inhibit EPO synthesis (La Ferla et al. 2002). Interestingly, the GATA-speciﬁc inhibitor K-7174 restores EPO production in IL-1, TNF-α or L-NMMA treated human hepatoma cell cultures and experimental mice (Imagawa et al. 2003). The oral administration of the GATA inhibitors K-7174 and K-11706 were efﬁcacious in restoring normal EPO production in animal models of anemia related to inﬂammatory diseases (Nakano et al. 2004). Interestingly, another GATA transcription factor, GATA-4, is responsible for the high EPO expression in fetal liver. In fact, the expression of this transcription factor is high in fetal liver, but low in adult liver (Dame et al. 2004). It is of interest to note that erythroid cells grown under particular cell culture conditions (i.e. in the presence of stem cell factor and IL-6/sIL-6R) are able to produce EPO according to an autocrine pattern (Sato et al. 2000). This ﬁnding explains the capacity of cord blood hemopoietic progenitors to undergo terminal erythroid maturation when grown in the presence of stem cell factor (SCF) and IL-6, but in the absence of exogenous EPO (Sui 1996).

Role of EPO, EPO-R, Kit Ligand and c-kit in the control of erythropoiesis EPO is the main cytokine involved in the control of erythropoiesis. RBC production is strictly dependent on the interaction of EPO with its single transmembrane receptor, EPO-R. EPO is secreted from kidney and fetal liver according to a molecular mechanism triggered by hypoxia (Ebert and Bunn 1991). Bone marrow erythroid cells expressing EPO-R are the main target of EPO: particularly, the late erythroid progenitors, CFU-E and proerythroblasts, where the EPO-R is maximally expressed (about 1,000 receptors per cell). The EPO-R pertains to the super family of cytokine receptors, characterized by the presence of regions of homology at the level of four conserved cysteine residues and the WSXWS motif located near to the transmembrane region. The EPO-R has a large extracellular ligand binding domain, a single transmembrane helix composed by 22 hydrophobic amino acids and a large cytoplasmic domain (Fig. 4). It is synthesized as a 62 kDa

Physiology of erythropoiesis

13

Fig. 4. Schematic representation of the EPO receptor. A single receptor chain is shown in the ﬁgure. The extracellular, transmembrane and cytoplasmic (intracellular) domains of the receptor are outlined (sizes drawn to scale). The receptor is synthesized as precursor of 508 aminoacids, while the mature receptor is composed by 484 amino acids. In the extracellular domain several important structural features are outlined: four conserved cysteine residues, a site of N-glycosylation, a ﬁbronectin-3like sub-domain and a WSXWS motif. In the intracellular, cytoplasmic domain are evidenced two proximal conserved boxes 1 and 2, one of them being involved in JAK2 binding and activation and 9 tyrosine residues playing a major role in the mechanism of EPO receptor signaling. The molecules binding to each of these Tyr (Y) residues are indicated on the right

14

U. Testa

precursor that is quickly glycosylated through a high-mannose glycosylation process. The mature EPO-R exhibits a 66 kDa molecular mass. The unliganded EPO-R exhibits an open scissor-like conformation and the dimerization induced by the transmembrane domain maintains this receptor in an inactivated state. The receptor exists as a preformed homodimer, each binding one EPO molecule. Therefore, each EPO molecule shows two binding sites, where the binding site1 has a high afﬁnity (in the nanomolar range) for one EPO-R subunit, while the binding site2 exhibits a relatively low-afﬁnity (in the micromolar range) for the other EPO-R subunit. The two binding interactions determine a high-afﬁnity interaction of EPO with the EPO-R homodimer. The interaction of EPO with EPO-R homodimer determines a conformational change of the EPO-R subunits, with consequent activation of the signal transduction machinery (Remi et al. 1999). Although the primary role of EPO and its receptor is the regulation of RBC production, EPO and its receptor have been localized to several nonhematopoietic tissues and cells, including the central nervous system, endothelial cells, the liver, the uterus and several solid tumors (Farrell and Lee 2004). The biology of EPO and EPO-R, as well as the role of EPO-R expression on tissues other than the erythron and on tumor cells are subjects of separate chapters in this book (see chapter II and III).

Molecular mechanisms of control of erythropoiesis: role of transcription factors The process of erythropoietic differentiation, as well as the whole process of hemopoietic differentiation, is orchestrated at molecular level by a complex network of transcription factors that act regulating the expression of a set of target genes. Particularly, evidence has been accumulated showing that lineage-speciﬁc transcription factors, acting together with general transcription factors, play an essential role in the process of erythroid differentiation. Among these different transcription factors GATA-1 seems to play a key role in erythroid development. This zinc-ﬁnger transcription factor, expressed in erythroid and megakaryocytic cells, binds to GATA-binding motifs present in the promoters and/or enhancers of all erythroid-speciﬁc genes (reviewed in Cantor and Orkin 2002). Three functional domains have been identiﬁed within the GATA-1 protein: an N-terminal activation domain, the N-terminal zinc ﬁnger (N-ﬁnger) and the C-terminal zinc ﬁnger (Cﬁnger). The C-ﬁnger is essential for GATA-1 function, since it is responsible for recognition of the GATA-consensus sequence and consequent binding to DNA. The N-ﬁnger plays a crucial role in GATA-1 capacity to induce erythroid differentiation. Furthermore, the N-ﬁnger mediates the formation of complexes between GATA-1 and other cofactors, such as FOG-1 (Friend of GATA-1) (Ferreira et al. 2005).

Physiology of erythropoiesis

15

GATA-1 is scarcely expressed in quiescent erythroid progenitors, but is rapidly induced when EPO induces these cells to erythroid differentiation and then progressively accumulates during erythroid maturation being abundantly expressed during all stages of erythroid maturation (Sposi et al. 1992; Labbaye et al. 1995). The highest levels of GATA-1 are observed in CFU-Es and proerythroblasts, while low GATA-1 levels are observed in BFU-Es (Suzuchi et al. 2003). GATA-1 levels decrease during the maturation from proerythroblasts to orthochromatic erythroblasts and this seems to be required for terminal erythroid maturation (Ferreira et al. 2005). The essential role of GATA-1 in the control of erythropoiesis is directly supported by the study of the hemopoietic phenotype of GATA-1 knock-out mice exhibiting a severe anemia with production of an erythroid progeny resulting in a maturational arrest at the level of proerythroblasts. Studies on GATA-1−/− embryonic stem cells induced in vitro to erythroid maturation showed that these cells failed to mature beyond the proerythroblast stage and undergo rapid apoptosis, thus suggesting an essential role for GATA-1 in erythroid survival, in addition to erythroid maturation (Weiss et al. 1994). Proerythroblast-like cells derived from GATA-1 knockdown ES cells have the ability to proliferate vigorously, but a GATA-1 level of 5% cannot sustain the gene expression required for maturation of proerythroblasts (Suwabe et al. 1998). These ﬁndings have been also conﬁrmed in a model of GATA-1 knock-down using a promoter-speciﬁc disruption of the GATA-1 gene. Therefore, high levels of GATA-1 are required in immature erythroblasts to ensure their maturation. Using a promoter interference approach it was possible to demonstrate that 5% expression of GATA-1 is insufﬁcient to support erythropoiesis during embryonic development, while 20% expression of GATA-1 is sufﬁcient to support erythroid cell maturation (Takahashi et al. 1997). Low levels of GATA-1 expression are sufﬁcient to allow the proliferation of deﬁnitive erythroblasts and to protect them from apoptosis, while they are unable to sustain their maturation (Pan et al. 2005). On the other hand, over expression of GATA-1 in erythroid cells inhibit erythroid differentiation both in vitro and in vivo (Wyatt et al. 1997; Wyatt et al. 2000). However, erythroid cells overexpressing GATA-1 differentiate normally in vivo when in the presence of wild-type cells. This intriguing phenomenon was explained by assuming that normal erythroid cells generate within the erythroblastic island a RBC differentiation signal able to overcome the intrinsic defect in GATA-1-overexpressing erythroid cells (Gutierrez et al. 2004). These ﬁndings demonstrate the importance of intercellular signaling in regulating GATA-1 activity and indicate that this homotypic signaling between erythroid cells is crucial to normal erythroid differentiation. GATA-1 is required for both primitive and deﬁnitive erythropoiesis. In primitive erythroid cells GATA-1 expression is controlled by a 5′ enhancer element called the GATA-1 hematopoietic enhancer, while in deﬁnitive

16

U. Testa

erythroid cells GATA-1 expression requires a regulatory element in the ﬁrst intron, in addition to the GATA-1 hematopoietic enhancer. According to these observations it seemed logical to assume that one or more GATA-1 target genes, which normally exert their physiological function at the proerythroblast/basophilic erythroblast stages, are strictly necessary for the survival and terminal maturation of erythroid precursors. However, surprisingly, an examination of the expression of several important potential target genes showed that they continued to be expressed at approximately normal levels in the absence of GATA-1 (Tsang et al. 1998). Among these potential target genes, one could be represented by Bcl-XL, an anti-apoptotic protein, from which the expression is strongly induced by GATA-1 (Weiss and Orkin 1995). An elegant approach for the identiﬁcation of GATA-1 target genes consisted in the micro array transcriptosome analysis of GATA-1- rescued erythroblasts. This analysis provided evidence that GATA-1 induced a complex genetic program of cell cycle control consisting in the coordinate upregulation of cell cycle inhibitors (such as p18 and p27) and repression of mitogenic genes, such as cyclin D2 and cyclin dependent kinase 6, events mainly related to GATA-1 mediated inhibition of c-myc (Rylski et al. 2003). In a second analysis, carried out on a similar cellular model, clear evidence was provided about a stimulatory effect of GATA-1 on the expression of a set of genes including known GATA-1 gene targets, but also a repressor effect on another set of genes. Notable examples of genes induced by GATA-1 are represented by FOG-1, globins, glycophorin-A, Bcl-XL: the inductive effect on FOG-1 was rapid, while the stimulatory effect on the other genes was delayed. Two remarkable inhibitory effects of GATA-1 are exerted on GATA-2 and c-myc genes. In addition to its key role in the control of erythropoiesis GATA-1 affects also the differentiation capacities of early hemopoietic progenitors/stem cells, as shown by experiments involving the enforced expression of this transcription factor in murine stem cells or in committed hemopoietic progenitors (Iwasaki et al. 2003). On the other hand, the differentiation to the granulo-monocytic lineage required mandatorily the inhibition of GATA-1 activity via binding of the transcription factor PU.1 at the level of GATA-1 binding sites of GATA target genes (Rekhtman et al. 2003). In addition to PU.1, also other transcription factors such as Ski (Ueki et al. 2004) and HERP2 (Elagib et al. 2004) are able to interact with GATA-1 and to inhibit its transcriptional activity. The loss of GATA-1 transforms primitive blood precursors into myeloid cells, resulting in a massive expansion of granulocytic neutrophils and macrophages (Galloway et al. 2005). The ensemble of these observations clearly indicates that GATA-1 acts as a master regulator of erythropoiesis. It is therefore evident that GATA-1 acts as an activator and repressor of different target genes: it represses cell proliferation and early hemopoietic genes (such as GATA-2), while it

Physiology of erythropoiesis

17

activates later during differentiation erythroid genes. These different tasks are accomplished by GATA-1 through the formation of different molecular complexes with other transcription factors (Rodriguez et al. 2005): the GATA-1/Gﬁ-1B acts early during differentiation and suppresses genes involved in cell proliferation, such as c-myc and c-myb; the GATA-1/FOG-1/MeCP1 complex also acts early during differentiation to suppress the expression of genes, such as GATA-2, involved to maintain multipotency of progenitors; ﬁnally, the GATA-1/FOG-1 and the GATA-1/Tal-1/Lbb1 complexes act at later stages of differentiation promoting erythroid-speciﬁc gene expression. In addition to bind DNA, other transcription factors exert their activity through protein-protein interactions. A prototype of this category of factors is represented by FOG-1, a zinc ﬁnger transcription factor that binds to the amino zinc ﬁnger of GATA-1 (reviewed in Cantor and Orkin 2002). FOG-1 is a polypeptide of 998 amino acids that contains nine zinc ﬁngers, from which four interact independently with GATA-1. FOG-1, abundantly expressed in erythroid and megakaryocytic cells, is co-expressed with GATA-1 during erythroid development and exerts a key role in the control of erythropoiesis: in fact, FOG-1 mice die of severe anemia caused by an erythroid maturation block similar to that observed in GATA-1−/− mice (Gregory et al. 1999). Genetic evidences based on the study of in vitro obtained or spontaneously occurring GATA-1 mutants proved that the physical interaction between GATA-1 and FOG-1 is required for the erythropoietic and megakaryocytopoietic effects of these transcription factors (Nichols et al. 2000). At the moment, the mechanisms by which FOG-1 inﬂuences the biologic activity of GATA-1 are largely unknown, but the majority of the studies suggest that the simple interaction between GATA-1 and FOG-1 is sufﬁcient to activate GATA-1 (Cantor and Orkin 2002). In vitro differentiation studies of mouse stem cells and conditional gene expression assays allowed to better deﬁne the role of FOG-1 in erythroid and megakaryocytic cell differentiation: FOG-1 exerts an inhibitory effect on the proliferation of erythroid cells, while it is required for late stages of megakaryopoiesis (Tanaka et al. 2004). Several mutually occurring mutations in the GATA-1 gene that alter the FOG-binding domain have been reported. The mutations are associated with familial anemias and thrombocytopenias of differing severity (Liew et al. 2005). Interestingly, excess GATA-1 mutant protein lacking FOG-1 binding capacity abrogates lethal anemia that is owing to GATA-1 deﬁciency, but it cannot rescue megakaryocyte differentiation (Shimizu et al. 2004). Another transcription factor pertaining to the GATA family, GATA-2 exerts also effects on erythropoiesis. In the early mouse embryo, loss of GATA-1 leads to a qualitative defect in yolk sac erythropoiesis, while loss of GATA-2 determines only a moderate quantitative effect at the yolk sac stage. At later times of development early erythroid progenitors, as well as other hemopoietic progenitors, strictly require GATA-2 for their development

18

U. Testa

(Tsai et al. 1994). The deﬁciency of both GATA-1 and GATA-2 determines a complete blockade of erythropoiesis, as well as of other hemopoietic lineages (Fujiwara et al. 2004). GATA-2 expression is down-regulated and GATA-1 expression increases as hemopoietic precursor cells differentiate into erythroid cells. Although major questions remained unanswered regarding the exact consequences of GATA-2 down-regulation, this appears to be important for differentiation, as sustained expression of GATA-2 alters erythroid differentiation. The repression of GATA-2 expression during normal erythropoiesis seems to be dependent upon a transcriptional repressor activity exerted by GATA-1, via binding to GATA sites present in a regulatory region of the GATA-2 locus (Martowicz et al. 2005). Erythroid differentiation is inﬂuenced also by other transcription factors acting either at early or late stages of the differentiation process. In this context, among the transcription factors regulating the early events of hematopoietic differentiation, an important role in the control of erythropoiesis is certainly played by Tal-1, a member of the basic helix-loop-helix (bHLH) family of transcription factors, initially identiﬁed for its involvement in some chromosomal translocations occurring in patients with T-cell acute lymphoblastic leukemia. Tal-1 expression is essential for the early developments of the primitive and deﬁnitive hematopoietic systems. Experiments on embryonic stem cells Ta1−/− clearly indicate that this transcription factor is required for the development of competence of mesodermal cells to become hematopoietic (Endoh et al. 2002). However, studies of conditional Tal-1 gene targeting have shown that Tal-1 expression, although essential for the genesis of hematopoietic stem cells, is not essential for the function of these cells (self-renewal, proliferation and commitment). Finally, Tal-1 expression is strictly required for proper erythroid and megakaryocytic differentiation (i.e. in the absence of Tal-1 no erythroid and megakaryocytic differentiation is observed) (Mikkola et al. 2003). Like other tissue-restricted bHLH transcription factors, Tal-1 binds DNA as a heterodimer with the ubiquitously expressed E protein, which recognizes the hexanucleotide sequence CANNTG found in a wide variety of eukaryotic transcriptional enhancers, particularly at the level of erythroid and megakaryocytic genes. In normal adult tissues Tal-1 expression is restricted to hematopoietic and endothelial cells and, particularly in erythroid cells, Tal-1 is initially expressed at low levels in quiescent erythroid progenitors, but becomes highly expressed throughout all the differentiative and maturative process and forms active heterodimers with E2A, participating in the activation of erythroid-speciﬁc genes (Gabbianelli et al. 2000; Gabbianelli et al. 2003). Thus, it was shown that Tal-1 is strictly required for glycophorin-A expression (Lahil et al. 2004). In fact, antisense oligomers to Tal-1 inhibit proliferation and self-renewal of erythroleukemia cells (Green et al. 1991), and forced Tal-1 expression exerts a stimulatory effect on erythroid development of normal hemopoietic progenitors (Valtieri et al. 1998).

Physiology of erythropoiesis

19

It is of interest to note that several studies suggest that the capacity of Tal-1 to inﬂuence the hemopoietic differentiation depends more on the capacity to form transcriptional complexes with other transcription factors than on its DNA-binding properties. This intriguing conclusion is directly supported by the observation that Tal-1 mutants that lack a DNA-binding domain are able of rescuing primitive erythropoiesis from in vitro differentiated Tal-1−/− embryonic stem cells (Porcher et al. 1999). However, a recent study clearly indicates that some of the effects of Tal-1 on erythropoiesis require its DNA-binding activity (Ravet et al. 2004). Furthermore, Tal-1 was shown to form transcriptional complexes with E47, GATA-1, LMO2 and Ldb1: these complexes bind to E boxes present in the promoters of some erythroid genes, thus stimulating their transcription (Xu et al. 2003). On the other hand, other transcription factors, such as Erythroid Kupperlike Factor (EKLF) and NF-E2, act late during erythroid maturation. EKLF is a zinc ﬁnger transcription factor which binds to a binding site consensus sequence 5′-NCNCNCCN-3′ present within the adult β-globin gene promoter. Binding sites for EKLF are found in close proximity of GATA-1 binding sites in the regulatory regions of several erythroid-speciﬁc genes. Mice with genetic deletion of the EKLF genes die from severe anemia at the fetal liver stage due to failure of adult β-globin gene activation. Recent studies provided evidence that EKLF could act coordinating erythroid cell proliferation and hemoglobinization (Coghill et al. 2001). In addition to its role in the control of globin synthesis, EKLF plays also an important role in the control of the expression of several genes involved in heme synthesis, such as ALA2S and PBGD (Drissen et al. 2005). Expression of the p45 subunit of transcription factor NF-E2 is restricted to selected blood cell lineages, including megakaryocytes and developing erythrocytes. Mice lacking p45 NF-E2 show profound thrombocytopenia and a number of RBC defects, including anosocytosis and hypochromia (Levin et al. 1999). NF-E2 level is low in early erythroid progenitors and progressively increases during erythroid differentiation (Labbaye et al. 1995). The low expression of NF-E2 during the early stages of erythroid differentiation seems to be related to an inhibitory effect exerted by NF-κB (Liu et al. 2003), whose expression is high in erythroid progenitors (Zhang, MY et al. 1998). In addition to these lineage-speciﬁc transcription factors, other ubiquitously expressed transcription factors play an important role in the control of erythropoiesis. Among them a peculiar role is played by the CBP, a ubiquitously expressed histone acetyltransferase, able to interact with a large number of proteins. Particularly, CBP is able to interact with GATA-1 and to acetylate it at the level of lysine residues: as a consequence of this acetylation, the transcriptional activity of GATA-1 is stimulated (Hung et al. 1999). The functional role of GATA-1 acetylation is directly supported by the observation that mutation of the acetylated lysine residues to arginine markedly reduces the GATA-1 ability to rescue erythroid maturation after transfection

20

U. Testa

in a GATA-1−/− cell line. CBP is also able to acetylate Tal-1: this acetylation destabilizes the interaction of Tal-1 with mSin3A repressor, while enhancing binding to DNA and transcriptional activity (Huang et al. 2000). Recent studies provided evidence that the transcription factor Gﬁ-B could play an important role in the control of erythropoiesis. Gﬁ-1B is a highly related zinc ﬁnger protein with a selective pattern of expression restricted to hemopoietic cells. This transcription factor possesses a typical SNAG domain and binds to the consensus DNA sequence TAAATCAC (A/T) GC (A/T). Some of the target genes of Gﬁ-1B have been identiﬁed (p21; Socs1, Socs3); in this case Gﬁ-1B acts as a transcriptional repressor. The Gﬁ-1B transcriptional factor is expressed in the bone marrow and spleen in mice, selectively in erythroid cells in chicken, while in man its expression is restricted to erythroid and megakaryocytic cells (Osawa et al. 2002). The expression of Gﬁ-1B gene in erythroid cells is under the control of GATA-1 that, in cooperation with NF-Y, acts as a transcription activator of the Gﬁ-1B gene (Huang et al. 2004). Studies of gene targeting in mice showed that Gﬁ-1B embryos exhibit delayed maturation of primitive erythroid cells and subsequently die with a failure to produce deﬁnitive erythroid cells. The fetal liver of Gﬁ-1B−/− mice contains erythroid and megakaryocytic precursors arrested in their development, while myelopoiesis is normal (Saleque et al. 2002). The role of Gﬁ-1B in erythroid differentiation was also conﬁrmed by studies of enforced expression of Gﬁ-1B gene in normal human hemopoietic progenitors, showing a marked proliferation of erythroid cells in an EPO-independent manner, associated with an inhibition of myeloid cell differentiation. Deletion of the SNAG repressor domain abolished Gﬁ-1Binduced erythroid maturation, strongly suggesting that Gﬁ-1B acts in the late stages of erythroid differentiation as a transcription repressor (Garçon et al. 2005). Although the role of Gﬁ-1B on erythroid differentiation is clear, its position in the hierarchy of hemopoietic transcription factors, as well as its functional relationship to GATA-1 and FOG-1, remains to be evaluated. Transcription factors of the AP-1 family are activated by EPO in erythroid cells and play a role in the control of apoptosis, cell proliferation and differentiation. The AP-1 family of transcription factors consists of the Fos and Jun proteins, which are known as “early response” proteins due to their up-regulation in response to extracellular stimuli regulating cell growth, differentiation and survival. C-Jun and particularly JunB is induced during early and late stages of erythroid differentiation: the late JunB expression is required for erythroid maturation (Jacobs-Helber et al. 2002). There is also evidence that c-jun may play a role in the protection from apoptosis elicited by EPO. In addition to these transcription factors that act on erythropoiesis modulating the expression of sets of target genes, other transcription factors are also essential for normal erythropoiesis, but they act through a different mechanism. In this context particularly relevant is the retinoblastoma (Rb)

Physiology of erythropoiesis

21

gene product (pRb), a key regulator of cell proliferation and differentiation. A large set of data indicates that pRb is an important controller of erythroid differentiation. In fact, Rb−/− mice die of ineffective erythropoiesis during embryonic development, with decreased numbers of BFU-E, CFU-E and deﬁnitive erythrocytes. Although subsequent studies have cast doubt as to whether these defects in erythropoiesis are intrinsic to the erythroid lineage (Jacks et al. 1992), more recent studies have largely clariﬁed the role of pRb in erythropoiesis (Whyatt and Grosfeld 2002). Thus, it was shown that pRb is greatly accumulated during in vitro human erythropoiesis and it’s silencing by RNA antisense oligonucleotides resulted in a marked inhibition of erythroid colony formation (Condorelli et al. 1995). Culture experiments using hemopoietic progenitors derived from Rb−/− mice showed that these cells have an impaired terminal maturation (Clark et al. 2004) because pRb seems to be required for erythroblast enucleation (Spike et al. 2004). Particularly, pRb is also required for the development of cells, such as macrophages that support erythropoiesis. In fact, pRb promotes the differentiation of macrophages by opposing the inhibitory function of the transcription factor Id2 on PU.1, a master regulator of macrophage differentiation (Iavarone et al. 2004). Recent studies provided evidence that the octamer-binding protein-1 (Oct-1) is required for normal erythropoiesis. In fact, Oct-1−/− mice are anemic and suffer from a lack of erythroid precursor cells. Oct-1−/− embryonic stem cells display a markedly reduced capacity to differentiate to erythroid progenitors and to mature erythroid elements (Wang et al. 2004). Gene targeting studies in mice have shown that the loss of VEGF or VEGF-RI or VEGF-RII or c-myc resulted in a blockade of primitive erythropoiesis, but this phenomenon is not speciﬁc to erythropoiesis and seems to be related to a more generalized defect of hematopoiesis. Although the inactivation of c-myb determines a generalized defect in hemopoiesis, this transcription factor seems particularly relevant for erythropoiesis. In fact, using a gene targeting approach allowing only a partial inactivation of c-myb gene expression, it was possible to show that c-myb activity is strictly required for the transition of CFU-Es to immature erythroblasts (Emambokus et al. 2003). After this stage of differentiation, c-myb expression is downmodulated through a repressive mechanism mediated by the binding of GATA-1 at the level of the c-myb promoter (Bartunck et al. 2003). In addition to transcription factors that cooperate and enhance the activity of erythroid transcription factors, other transcription factors antagonize their activity. An important example is provided by the myeloid and Blymphoid transcription factor PU.1: PU.1, when overexpressed in erythroid cells, inhibits their maturation through a molecular mechanism related to a physical interaction and a functional antagonism with GATA-1 (Zhang et al. 2000). However, at earlier stages of erythroid differentiation PU.1 promotes the self-renewal of erythroid progenitors (Back et al. 2004). In contrast, other

22

U. Testa

transcription factors, such as c-jun, antagonize the transcriptional activity of GATA-1 through induction of the expression of the basic helix-loop-helix protein HERP2: this protein acts as a repressor of GATA-1 transcription (Elagib et al. 2004). A key event during terminal differentiation of erythroid cells is the activation of globin gene transcription, leading to hemoglobin synthesis. The high expression of globin genes in erythroid cells is dependent on the presence of peculiar gene regions, locus control region (LCR), which consists of ﬁve nuclease hypersensitive sites spread over 20–30 kilobases of DNA and located upstream the β-globin like genes. The β-globin LCR contains numerous binding sites for both erythroid-speciﬁc and ubiquitous transcription factors, including E-boxes (recognized by Tal-1), CACC motifs (recognized by Kruppel-like proteins and by the Sp family of proteins), GATA motifs and MAREs (Maf recognition elements, recognized by the transcriptional activation complex NF-E2, which comprises the hematopoietic-speciﬁc p45 subunit and the ubiquitous Maf protein 18). Using quantitative proteomics the transcriptional complexes binding to the β-globin LCR have been characterized (Brand et al. 2004). Interestingly, the transcriptional molecular complexes formed at the level of MAREs sites change during erythroid differentiation: in immature erythroid precursors the p18 subunits forms heterodimer with Bach1, recruiting co-repressor complexes (including NuRD, SIN3) that generate a transcriptionally repressive chromatin structure; in maturing erythroid precursors the p18 subunit forms heterodimers with p45, recruiting a co-activating complex (including Tal-1 and CBP) that generates a transcriptionally permissive structure (Brand et al. 2004).

Anti-apoptotic mechanisms operating in erythroid cells Stat5 The signal transducer and activator of transcription (Stat) proteins have a dual role as signal transducers and activators of transcription. These proteins are latent in the cytoplasm and are activated by extracellular signaling cytokines or growth factors that bind to speciﬁc cell surface receptors (Levy and Darnell 2002). Following the interaction of these ligands with their receptors, various tyrosine kinases are activated in the cell that phosphorylate Stat proteins; phosporylated Stat proteins become active and accumulate in the nucleus to drive transcription. Stat 5 is a transcription factor present in the cytoplasm in a latent form and activated by the EPO-R, as well as many other cytokine receptors. As a consequence of EPO-R activation Stat5 binds to phosphorylated tyrosines present on the cytoplasmic tail of the EPO-R, and itself becomes phosphorylated at the level of tyrosine residues. The activated Stat5 dimerizes and

Physiology of erythropoiesis

23

translocates from the cytoplasm to the nucleus where it modulates the expression of several genes involved in the control of cell proliferation and differentiation. In addition to these genes, Stat5 activates in erythroid cells the expression of several anti-apoptotic genes and, notably, of Bcl-XL through a direct binding at the level of Stat5 binding consensus sequences present in the promoter of the Bcl-XL gene (Dumon et al. 1999; Nosaka et al. 1999; Ariyoshi et al. 2000; Levy and Darnell 2002). These effects of Stat5 on Bcl-XL gene expression represent the molecular basis to explain the antiapoptotic effects elicited by Stat5 in erythroid cell lines (Levy and Darnell 2002), while dominant-negative Stat5 induces a growth arrest and apoptosis of erythroid precursors (Chida et al. 1999). Stat5 was reported to induce also Bcl-2 mRNA, but its induction seems to be indirect. Studies on mice that lack the expression of both Stat5a and Stat5b isoforms have provided deﬁnitive evidence about the protective role of Stat5 in erythroid cell survival. Stat5a−/−/Stat5b−/− mice embryos are severely anemic as a consequence of an impaired survival of liver erythroid progenitors. Particularly, fetal liver cells derived from Stat5a−/−Stat5b−/− animals generated a low number of erythroid colonies in vitro, are less sensitive to EPO and showed a three-time increase in the frequency of apoptotic cells and a pronounced increase in the percentage of apoptotic cells when grown in vitro in the presence of EPO (Levy and Darnell 2002). In spite of the marked anemia during the embryonic life, Stat5a−/−Stat5b−/− mice exhibited at birth a moderate condition of anemia, which was progressively attenuated in animals of adult age (at adult age only about 50% of Stat5a−/−Stat5b−/− mice were anemic). However, the adult Stat5a−/−Stat5b−/− mice, in spite of their nearnormal hematocrit levels, are deﬁcient in generating high erythropoietic responses following stress stimulation (i.e. the induction of a chemically induced hemolytic anemia). The analysis of erythropoiesis in Stat5a−/−5b−/− animals anemic during adult life under steady-state conditions showed the existence of an increased pool of immature erythroblasts exhibiting reduced Bcl-XL levels and undergoing apoptosis at a high rate with respect to the corresponding cells of wt animals (Socolovky et al. 2001). In conclusion, the analysis of Stat5 knockout mice strongly suggests that the anti-apoptotic effect of this transcription factor in erythroid cells is exerted at the level of immature erythroid cells and is mediated mainly via modulation of Bcl-XL. In line with this observation, Stat5a and Stat5b and Bcl-XL exhibit a similar pattern of expression during normal erythroid maturation, with a progressive increase of expression during the differentiation from BFU-E to CFU-E and to immature erythroid precursors, followed by rapid and marked decline at later stages of maturation (polychromatophilic and orthochromatic erythroblasts) (Fig. 5). It is of interest to note that Stat5a5b deﬁciency induces also a premature death of myeloid precursors, which seems to be related to a decrease of Bcl-2 and Bcl-XL (Kieslinger et al. 2000).

24

U. Testa

Fig. 5. Kinetics of Stat5a, Stat5b, JAK-2, Bcl-XL and Bcl-2 expression during erythroid differentiation. Puriﬁed hemopoietic progenitor cells have been induced to erythroid differentiation under unilineage cell culture conditions, either in the presence of EPO alone or EPO plus kit ligand. Cell aliquots were recovered at different days of culture and Stat5a, Stat5b, JAK-2, Bcl-XL and Bcl-2 levels have been evaluated by Western Blotting analysis on samples normalized according to β-actin content

Physiology of erythropoiesis

25

Bcl-XL The protective role exerted by Stat5 on apoptosis of erythroid precursors is mainly related to the induction of Bcl-XL. Bcl-XL pertains to the family of Bcl-2-related proteins, important regulators of cell death. The Bcl-2 family members can be classiﬁed into three different groups on the basis of comparative analysis of their structure-function: (i) multidomain death antagonists (Bcl-2, Bcl-XL, Bcl-w, Mcl-1 and A1); multidomain death agonists (Bax, Box and Bok); (iii) BH3-only pro-apoptotic proteins (Bim, Bid, Bik and Bak). The ﬁrst two groups of Bcl-2 members function to protect or disrupt the integrity of mitochondrial membranes, respectively, while BH3-only proteins trigger cell death through binding to the receptor domain of multidomain Bcl-2 members, thereby mediating the inactivation of the antiapoptotic members or the activation of the pro-apoptotic members (Joza et al. 2002; Kaufmann et al. 2001). In normal cells, inactive BAX is located in the cytosol, but following an apoptotic stress, BAX is inserted into mitochondria as homooligomerized multimer, resulting in downstream mitochondrial function (Danial and Korsmeyer 2004). Bcl-XL, as well as Bcl-2, is targeted to mitochondria through its interaction with a protein, FKBP38, pertaining to the family of immunophilins. Bcl-XL and Bcl-2 reside on the outer mitochondrial membrane. The main function of the anti-apoptotic Bcl-2 family members consists in promoting adaptation and maintaining vitality of mitochondria to various types of perturbations of cellular metabolism (Vander Heiden and Thompson 2002). Particularly, the anti-apoptotic mechanism of Bcl-XL, as well as of other Bcl-2 family members, consists in the inhibition of the activity of Bid and Bax which cooperate in the formation of pores in the membrane of mitochondria, allowing the release of components of these organelles in the cytoplasm, with subsequent loss of their function (Kuwana et al. 2002). Recent studies, however, suggest a broader role of Bcl-2 in the control of apoptosis, consisting in a general control of caspase activation program independently of the cytochrome c/Apaf-1/caspase-9 apoptosome (Marsden et al. 2002). Therefore, two models have been proposed: in the ﬁrst model, BH3-only proteins can directly bind and activate Bak/Bax, and this phenomenon may be inhibited by Bcl-2/Bcl-XL-mediated sequestration of BH3-only proteins; in the second model, the binding of BH3-only proteins to their primary targets Bcl-2/Bcl-XL leads to a neutralization of these anti-apoptotic factors, resulting in the activation of Bax/Bak (Cory and Adams 2002). Gene targeting studies of anti-apoptotic Bcl-2 family members provided clear evidence that they exhibit a unique physiologic role: Bcl-2 is required for the survival of kidney, melanocyte stem cells and mature lymphocytes; Bcl-XL for neuronal and erythroid cells; Bcl-w for sperm progenitors; A1 for neutrophils and Mcl1 for zygote implantation (reviewed in Cory and Adams 2002). Recent studies indicate that Mcl1 is strictly required for the survival

26

U. Testa

of hemopoietic stem cells and hemopoietic progenitor cells and its function cannot be replaced by other Bcl-2 members (Opferman et al. 2005). In this context, particularly interesting were the phenotypic features observed in the Bcl-X gene knockout: in fact Bcl-X−/− mice die at E13, exhibiting massive cell death at the level of neuronal elements of the central nervous system and of erythroid elements present in fetal liver (Motoyama et al. 1995). The level of action of Bcl-XL in the protection of erythroid cells from apoptosis was ﬁnely investigated using mouse embryonic stem cells in which both alleles of the Bcl-XL gene were disrupted. Bcl-XL−/− embryonic stem cells were unable to contribute to the generation of deﬁnitive erythroid cells, thus indicating that the expression and function of this anti-apoptotic protein is essential for the production of deﬁnitive erythroid cells. Bcl-XL−/− embryonic stem cells generated a number of erythroid colonies similar to that originated from wt embryonic stem cells; however, a signiﬁcant proportion of erythroid precursors present in the erythroid colonies of Bcl-XL−/− cells undergo apoptosis during their maturation and do not reach the terminal stage (Motoyama et al. 1999). These ﬁndings strongly suggest that Bcl-XL is a critical anti-apoptotic regulator of erythropoiesis. The role of Bcl-XL in adult erythropoiesis was explored through the study of transgenic mice conditionally deﬁcient in Bcl-XL gene. These animals exhibited at three months of age a condition of severe hemolytic anemia, associated with platelet deﬁciency (Wagner et al. 2000). The analysis of the bone marrow of these animals showed a hyperplasia of both megakaryocytic and erythroid precursors; the rate of apoptosis in erythroid precursors of Bcl-XL deﬁcient animals was only slightly increased as compared to that observed for the corresponding cells of normal animals. This conditional knockout model was recently re-evaluated providing clear evidence that Bcl-XL plays a key role in the survival of the late-stage erythroblasts in all phases of the cell cycle, but not in early stage erythroblasts (Rhodes et al. 2005). Studies on an immortal line of phenotypically normal mouse erythroblasts provided further details about the anti-apoptotic role of Bcl-XL in erythroid maturation. These cells were maintained in an undifferentiated state by agents promoting self-renewal, such as SCF and glucocorticoids, while they were induced to differentiate by EPO. Bcl-XL overexpression allowed these cells to undergo terminal differentiation to mature erythrocytes in the absence of EPO (Dolznig et al. 2002). The molecular basis responsible for the elevated expression of Bcl-XL in erythroid cells is at the moment unknown. However, studies on the Bcl-XL gene promoter have in part helped to understand the regulation of the expression of this gene in erythroid cells. The promoter of the Bcl-XL gene used for the start of the transcription in erythroid cells is localized in 5′ in close proximity to the start initiation codon ATG. This promoter region contains several putative regulatory sites recognized by some transcription factors, including

Physiology of erythropoiesis

27

GATA-1, NF-E2 and Ets-1 and other transcription factors pertaining to the Ets family (Grillot et al. 1997). This promoter region contains also a Statbinding element, able to bind Stat-5 (Silva et al. 1999). Studies carried out on embryonic stem cells isolated from GATA-1−/− mice provided evidence that GATA-1 plays an important positive role in the regulation of Bcl-X gene expression. Particularly, it was shown through the analysis of conditional knockout models that the expression of GATA-1 in embryonic stem cells strongly induces Bcl-XL, but not Bcl-2, expression in erythroid cells and this may represent one of the mechanisms responsible for the anti-apoptotic effects exerted by GATA-1 in erythroid cells (Gregory et al. 1999). Finally, more recent studies have led to the identiﬁcation of the major site of start of Bcl-XL gene transcription in erythroid cells. This site corresponds for the human Bcl-X gene at −654 relative to the ATG initiation codon; furthermore, an enhancer element was identiﬁed at the level of the position −1804 through −1734 (Tian et al. 2003). EPO deprivation induces the activation of caspase-3, leading to apoptosis of erythroblasts (Gregoli and Bondurant 1999). Since activated caspase3 was reported to cleave Bcl-XL (Negoro et al. 2001), it is conceivable that EPO protects erythroid cells from apoptosis in part via blockage of caspase3-dependent cleavage of the Bcl-XL protein. Furthermore, it was shown that the activation of ERK1 and ERK2 by EPO up-regulates Bcl-XL expression via inhibition of caspase activities, thus resulting in the protection of erythroid cells from apoptosis (Mori et al. 2003). The kinetics of Bcl-XL expression during the process of erythroid differentiation (from early BFU-Es to CFU-Es) and maturation (from proerythroblasts to reticulocytes) was explored in detail. Few studies have explored the expression of Bcl-XL, as well as of Bcl-2, in hemopoietic progenitors (i.e. in total CD34+ cells or in fractions of CD34+ cells, such as 34+/38− or 34+/38+). Studies on human CD34+ cells have shown that both Bcl-XL and Bcl-2 are widely expressed in these cells; in contrast, in CD34+/38− cells (corresponding to early, immature hemopoietic progenitors) Bcl-XL is preferentially expressed as compared to Bcl-2 (Park et al. 1995; Peters et al. 1998). The kinetics of Bcl-XL expression during erythroid differentiation and maturation was explored using different cellular models of erythroid differentiation. Using human CD34+ cells grown either in erythroid (SCF + EPO) or in granulocytic (SCF + G-CSF) cell culture medium, it was shown that at day 4 and 8 of culture Bcl-XL expression was markedly more pronounced in erythroid than in granulocytic precursors (Dumon et al. 1999; Josefsen et al. 2000; Dolznig et al. 2001). Bcl-XL was expressed at very low levels in immature erythroblasts and its expression progressively and markedly increased during maturation up to terminal erythroblasts (Gregoli and Bondurant 1997). Using a cell culture system allowing the study of all stages of erythroid differentiation from BFU-E to mature erythroblasts it was possible to carefully study Bcl-XL expression during erythroid differentiation and

28

U. Testa

maturation (Testa 2004). In quiescent CD34+ human cells a low Bcl-XL expression was observed (Fig. 5). A marked increase in Bcl-XL expression was observed during early stages of the differentiation of CD34+ to the erythroid lineage (day 4 and day 7, when the cells correspond to a stage intermediate from a BFU-E to a CFU-E and to CFU-E/proerythroblasts, respectively); Bcl-XL levels continued to increase during early and intermediate stages of erythroid maturation, reaching peak level of expression when the majority of erythroid cells have reached the stage of polychromatophilic erythroblasts; during terminal stages of erythroid maturation (day 14–16 of culture) a marked decline in Bcl-XL expression was observed (Fig. 5). The administration of SCF to the erythroid cultures moderately potentiated BclXL expression, associated with terminal erythroid maturation (Fig. 5). In contrast, Bcl-2 expression was limited only to the early stages of erythroid differentiation and completely declined to very low or undetectable levels during the maturation of erythroid cells. Bcl-XL could exert, in addition to its role as an anti-apoptotic factor in erythroid cells, other biologic functions. Inhibition of Bcl-XL expression (by antisense transcripts) in Friend erythroleukemia cells elicited a marked inhibition of hemoglobin synthesis after DMSO induction, as well as an increased rate of apoptosis (Haﬁd-Medhab et al. 2003). Importantly, this inhibitory effect of Bcl-XL deﬁciency on hemoglobin synthesis was observed also in cells over expressing Bcl-2 (Haﬁd-Medhab et al. 2003). The inhibitory effect on hemoglobin synthesis caused by Bcl-XL deﬁciency was due to an inhibition of heme synthesis and not to a reduction of globin mRNA expression. Since in differentiating erythroleukemic cells Bcl-XL remains localized to mitochondria, it was suggested that it could exert in this organelle a dual function being involved as an anti-apoptotic factor in mitochondria integrity and in heme synthesis (Medhab et al. 2003). However, it remains to prove whether Bcl-XL exerts this dual function also in normal erythroid cells. Recent studies on conditional Bcl-XL knockout have shown that in the absence of Bcl-XL there is only a slight decrease during late stages of erythroid maturation (Rhodes et al. 2005). Furthermore, studies of enforced expression in the FDCP-Mix multipotent progenitor cell line have indicated a novel role for Bcl-XL in cell fate decision beyond cell survival. In fact, the erythroid expression of Bcl-XL in these cells was associated with induction of erythroid differentiation and prohibited generation of myeloid cells (Haughn et al. 2003). It is of interest to note that recent studies have indicated that Bcl-XL may represent a major target of some anti-tumor chemotherapeutic drugs. Therefore, Bcl-XL levels may represent a major determinant in the protection of the erythroblasts from apoptotic cell death induced by DNA damaging chemotherapeutic drugs. In spite of their structural heterogeneity and their different molecular targets, many chemotherapeutic agents kill cells by inducing apoptosis. The

Physiology of erythropoiesis

29

majority of these drugs mediate cell death through an activation of the intrinsic apoptosis pathway; this pathway is activated by different types of intracellular stresses, such as genetic damage or growth factor deprivation, and implicates the Apaf-1-dependent activation of caspase-9, following release of cytochrome C and Smac/Diablo from mitochondria (Johnstone et al. 2002). This pathway is regulated by the proapoptotic multidomain and BH3-only Bcl-2 proteins and their anti-apoptotic counterparts, including also Bcl-XL. In addition to represent one of the main molecules involved in the cell protection in the intrinsic apoptotic pathway, Bcl-XL may represent also a direct target of some chemotherapeutic drugs. In fact, it was shown that Bcl-XL could be modiﬁed by deamidation of asparagine residues, at the level of an unstructured loop separating the (1helix-BH4 domain and the α2helix-BH3 domain (Deverman et al. 2002). Bcl-XL deamidation was observed in tumor cells or in mouse embryonic ﬁbroblasts p53−/− treated with different types of genotoxic agents, including cisplatin, etoposide, taxol and γ-irradiation (Deverman et al. 2002). The deamidation of Bcl-XL has as the consequence of its inactivation, with consequent loss of Bcl-XL binding to Bim and subsequent induction of apoptosis. The same authors also showed that the hypophosphorylated active form of Rb protein inhibits deamidation of BclXL, suppressing through this mechanism the apoptosis induced by DNA damaging agents. Anemia is a common complication of cancer, often resulting in a decrease of quality of life and inﬂuencing the outcomes of patient care. The myelosuppressive effects of chemotherapy are a major cause of anemia in cancer patients. Chemotherapy-induced anemia is the result of two different mechanisms acting at the level of: (a) the bone marrow progenitor/stem cell compartment, (b) the immature erythroid precursors (proerythroblasts and basophilic erythroblasts). This last mechanism was recently clariﬁed, showing that immature normal erythroblasts are extremely sensitive to the cytotoxic effect of chemotherapeutic agents, such as cisplatin, etoposide or camptothecin, while mature erythroblasts (acidophilic erythroblasts) are almost completely resistant. Importantly, EPO also at high doses was unable to protect immature erythroblasts from chemotherapy-induced apoptosis. According to these observations it was concluded that the primary target of chemotherapy-induced apoptosis is represented by proerythroblasts and basophilic erythroblasts. Interestingly, preincubation of erythroid cells with SCF resulted in a marked protection of erythroid cells from chemotherapyinduced apoptosis of immature erythroid cells. The mechanism of this protection seems to be related to an SCF-mediated up-modulation of Bcl-2 and Bcl-XL expression, associated with a consequent inhibition of caspase activation (Zeuner et al. 2003). At the moment it is unknown whether the inhibition of chemotherapy-induced apoptosis could be related to an inhibition of Bcl-XL deamididation, possibly mediated by upregulation of hypophosphorylated Rb expression.

30

U. Testa

Hydroxyurea, another myelosuppressive agent, elicited also apoptosis of erythroid cells through a mechanism involving a strong up-modulation of the TRAIL receptor 2 (Wang et al. 2002).

Role of apoptosis in the control of erythropoiesis The mitochondrial “intrinsic” and the transmembrane “extrinsic” pathways are the two principal pathways leading to apoptosis, both of which converge on caspases, a family of cysteine proteases. The “intrinsic” apoptosis pathway triggered by intracellular injury such as DNA damage or oxidative stress controls caspase activation through the Bcl-2 gene family. In this pathway damage sensors induce transcription of Bcl-2 homology 3 (BH3) domain proteins (i.e. Noxa, Bim, Bmf). These apical regulators activate downstream pro-apoptotic Bcl-2 members (i.e. Bax, Bak), overcoming inhibition by anti-apoptotic Bcl-2 family members (i.e. Bcl-XL, Bcl-2): activated Bax and Bak trigger mitochondrial release of factors that promote caspase activation in the cytosol. One of these factors is cytochrome c, which cooperates with Apaf-1 to activate caspase-9. This apical caspase activates the effector caspases 3, 6, and 7, inducing apoptotic cell death. During this process, two other mitochondrial factors, Smac/Diablo and Omi/HtrA2 prevent IAPs from inhibiting caspase activation. A key regulator of the intrinsic apoptotic pathway is the transcription factor p53, a potent tumor suppressor that acts as a stress sensor activated in response to DNA damage, hypoxia, nucleotide depletion, aberrant growth signals and chemotherapeutics drugs. Activated p53 induces a series of biological processes and, notably, acts as an activator of the intrinsic apoptotic pathway. Several evidences indicate that p53 triggers apoptosis by activating the intrinsic pathway: (i) by inducing over-expression of Bcl-2 or its pro-survival homologues; (ii) by stimulating the expression of the pro-apoptotic proteins BAX, BIM, BID, NOXA and PUMA; (iii) by directly binding to the BAX or BAK proteins and thereby inducing apoptosis through a transcription-independent mechanism (Michalak et al. 2005). The “extrinsic” pathway is triggered by extracellular death ligands such as TNF relatives FasL and TRAIL, which signal through speciﬁc membrane receptors. This apoptotic pathway seems to play an important role in the control of erythropoiesis.

Role of cell death receptors in the control of erythropoiesis Ligands and cell death receptors pertain to the tumor necrosis factor (TNF) and TNF receptor (TNF-R) superfamilies, respectively. Members of the TNF family of membrane-bound and secreted ligands pair off with one or more

Physiology of erythropoiesis

31

speciﬁc cell surface receptors that form a corresponding family of cognate receptors. Each ligand-receptor pair is considered a system, and actually more than 40 distinct ligand-receptor systems are currently recognized (Ware 2003). Members of both the TNF ligand and TNF-R superfamilies exhibit several remarkable structural similarities. The TNF ligands are type II transmembrane proteins, characterized by an extracellular C terminus domain, named the TNF homology domain: this domain contains a conserved framework of aromatic and hydrophobic residues, responsible for the association of single molecules to form a trimer (Bodmer et al. 2002). Most of the TNF ligands are synthesized as membrane-bound proteins, but soluble forms are originated by a process of limited proteolysis. The solubilization of some ligands is associated with a considerable loss of biological activity: this is the case of the Fas ligand (Fas L) and of the TNF-related apoptosis-inducing ligand (TRAIL). The main molecules pertaining to the TNF ligand superfamily are represented by TNF-α, FasL, TRAIL, TWEAK, RANKL, APRIL, lymphotoxin-α (LTα). More than 40 TNF-Rs have been identiﬁed. The majority of them are type I transmembrane proteins, with an extracellular domain N terminus and with an intracellular domain C terminus; a notable exception is represented by the TRAIL-R3, which is anchored to the cell membrane through a covalently linked C-terminal glycolipid (Bodmer et al. 2002). Many of these receptors possess a stretch of amino acids, the death domain (DD), as part of their sequence (Bhardway and Aggarwal 2003). The DDs are required for apoptotic signaling. For some of these receptors, soluble forms may be generated through a process of limited proteolysis (TNF-R1 and TNF-R2) or of alternative splicing (Fas). The different members of the TNF-R superfamily exhibit a typical structural feature represented by the presence of cysteinerich domains, which are usually repeats containing each six cysteine residues involved in the formation of three disulﬁde bonds (Bodmer et al. 2002). The TNF-Rs exist as preformed trimers on the cell membrane before the interaction with their respective ligands. Each member of the TNF ligand superfamily binds at least one receptor of the TNF-R superfamily: FasL binds selectively to Fas; TRAIL may bind to four different membrane receptors, called TRAIL-R1, TRAIL-R2, TRAIL-R3 and TRAIL-R4; TNF-α binds two membrane receptors, TNF-R1 and TNF-R2. The majority of the TNF-R superfamily members function as membrane receptors transducing a signal following interaction with their ligands. These signaling receptors may be subdivided into two groups on the basis of the structure and function of their cytoplasmic region: one of the two receptor groups possesses at the level of this cytoplasmic region, a death domain which mediates the association between the receptor and a death adaptor protein; the other group does not possess this death cytoplasmic domain. The TNF-Rs, TRAIL-R1 and TRAILR2, as well as Fas are typical examples of death receptors. Two examples of death domain adaptors are represented by FADD (also known as MORT1)

32

U. Testa

and TRADD: FADD possesses two interfaces, one interacting with the death domain (DD) of Fas and the other one (death effector domain, DED) interacting with intracellular effector enzymes (caspases that initiate the apoptotic process) (Ashkenazi 2002). The induction of death by FADD depends on the DED rather than the DD domain. The DED of FADD recruits two DEDcontaining caspases, caspase-8 and caspase-10, to form the death-inducing signal complex (DISC), thereby initiating apoptosis. The DEDs have no enzymatic function but link participants in a signaling chain through homotypic interactions, containing other proteins with DEDs (Tibbetts et al. 2003). The principle structural feature of the DEDs in various proteins is the conserved backbone of six α-helices, a conserved negatively charged residue at position 19 and the Arg-X-Asp-Leu (RXDL) motif at positions 78–81 (Tibbetts et al. 2003). The other adaptor called TRADD (TNFR-associated death domain) can initiate apoptosis through FADD or may stimulate protein kinases involved in the control of phosphorylation cascades to induce the transcription of some immune-system modulation genes. TRADD seems to be involved in the apoptotic signaling of the TNF-R, but not of Fas and TRAILRs (Dempsey et al. 2003). Some of the receptors of the TNF receptor superfamily are decoy receptors able to compete with signaling receptors for ligand binding, thereby inhibiting their function. Examples of decoy receptors are given by DcR1 and DcR2 (also known as TRAIL-R3 and TRAIL-R4), competing with TRAILR1 and TRAIL-R2, respectively, for binding of TRAIL and DcR3 (sFas) competing with Fas for binding of FasL (Ashkenazi 2002). The physiologic role of these receptors is still unclear, but it has been suggested that they could protect some normal cells by their cytotoxic ligands. The series of biochemical events elicited by the interaction of a death ligand with its receptor has been in part elucidated and can be summarized brieﬂy as follows (Fig. 6). The engagement of death receptors by their ligands activates the extrinsic apoptotic pathway. This pathway leads, as the intrinsic apoptotic pathway, to the activation of caspases independently on p53. The activation of a death receptor by its ligand leads to the recruitment at the level of the receptor of a death adaptor, which in turn determines the rapid assembly of a death-inducing signaling complex (abbreviated as DISC), with consequent activation of the apoptosis-initiating caspases 8 and 10. Caspase-10 is an initiator caspase found in primates but not in rodents. The role of this caspase in initiating the apoptotic process is supported by two observations: a) lymphocytes from ALPS II patients bearing caspase-10 mutation are resistant to TRAIL-mediated killing (Wang et al. 1999); b) endogenous caspase-10 is activated and recruited at the level of the DISC and is capable of transmitting an apoptotic signal in the absence of caspase8 (Kischkel et al. 2001). These caspases, in turn, activate caspase-9; in turn, the activated caspase9 activates “executioner” caspases-3, -6 and -7. Activated caspases-8 and -10

Physiology of erythropoiesis

33

Fig. 6. Signal transduction through the extrinsic apoptotic pathway. Engagement of TRAIL or FasL with its cognate receptor TRAIL-R2 results in the activation of two different signaling pathways involving the activation of the apoptotic cascade (right) or a two signaling pathways, MAPK and NFκB. The activation of the apoptotic cascade involves the formation of a receptor proximal complex containing the adaptor proteins FADD. These adaptor proteins in turn recruit additional key pathway-speciﬁc enzymes, such as caspase-8, they became activated and initiate downstream events leading to apoptosis

34

U. Testa

are also able to activate by proteolytic cleavage Bid, which is responsible for the activation of the pro-apoptotic arm of the Bcl-2 gene superfamily, with consequent release of apoptogenic factors from mitochondria. Particularly, SMAC/DIABLO released from mitochondria promotes apoptosis through its capacity to bind and neutralize the inhibitor of apoptosis proteins (IAP), thus preventing the activity of these proteins as attenuators of the caspase activation. Depending on the cell type, activated caspase-8 formed at the level of the DISC induces apoptosis by two different signaling pathways. In type I cells, large amounts of active caspase-8 formed at the DISC induce direct cleavage/activation of pro-caspase-3 independently of mitochondria. In type II cells, small amounts of active caspase-8 are insufﬁcient to activate procaspase-3 directly. Instead, caspase-8 cleaves the “BH3-only protein” Bid, generating a Bid fragment that activates the mitochondrial death pathway. The activation of caspase-8 and -10 is antagonized by cellular FLICE inhibitory protein (c-FLIP), an enzymatically-inactive relative of caspase-8 and -10, that binds to the DISC. Three c-FLIP proteins are present: c-FLIPL (55 kDa), c-FLIPS (26 kDa) and c-FLIPR (22 kDa) (Golks et al. 2005). Knockdown of c-FLIPL, as well as of c-FLIPS, augments DISC recruitment, activation and processing of caspase-8, thereby enhancing effector-caspase stimulation and apoptosis (Sharp et al. 2005). These observations clearly indicate that c-FLIP functions as an inhibitor of death receptor-mediated apoptosis. In the last years, evidence was accumulated showing that death receptors and their ligands may act as negative regulators of erythropoiesis and that their action may play a physiologically relevant role in the control of the rate of erythropoiesis. Furthermore, abnormalities of these receptors may play an important role in the physiopathology of some anemic conditions.

Effects of TNF-a on normal erythropoiesis The binding of TNF-α to the TNF-R1 leads to a cascade of events, above outlined, that determines the induction of apoptosis, NF-κB and JNK activation. TNF-R1 signaling involves assembly of two molecularly and spatially distinct signaling complexes that sequentially activates NF-κB and caspases (Muppidi et al. 2004). Within the ﬁrst minutes after binding of TNF-α to its receptor TNF-R1, a signaling complex termed complex I is formed. This complex is composed by TNF-R1 itself, TRADD, TRAF2 and RIP1. Complex I transduces signals that lead to NF-κB activation. Later (i.e. >2 hours) TRADD, TRAF2 and RIP1 dissociate from the receptor and recruit FADD and caspase-8 with subsequent formation of the complex II. In conditions where the complex I is able to induce a sufﬁcient level of NF-κB activation, expression of anti-apoptotic proteins is induced and the activation of apical caspases in complex II is inhibited (Muppidi et al. 2004). In addition to activating

Physiology of erythropoiesis

35

antiapoptotic genes, NF-κB can suppress apoptosis by inhibiting sustained activation of the MAP-kinase family member Jun-kinase (JNK) (Varfolomeev and Ashkenazi 2004). Given the signal transduction pathways activated by this cytokine, it is not surprising that TNF-α is able to induce a variety of biological effects. Mouse TNF and TNF-R1 gene knockout studies have clearly shown that this cytokine plays a key role in protection against infection by bacterial, fungal and parasitic pathogens. One of the main functions of TNF-α consists in its capacity to induce the production and release of the pro-inﬂammatory cytokines, IL-1, IL-6 and IL-8. In initial studies carried out in vivo (Moldawer et al. 1989; Ulich et al. 1990) and in vitro (Akahane et al. 1987) evidence was provided that TNF-α exerted an inhibitory effect on erythropoiesis. In addition, in a phase I study of TNF-α in cancer patients a clear decrease in hemoglobin levels after 1 month of TNF-α treatment was observed. Subsequent in vitro studies, however, showed that the inhibitory effect of TNF-α on erythropoiesis at the CFU-E level was likely mediated by IFN-γ released by macrophages in response to TNF-α and not due to a direct effect of TNF-α on erythroid progenitors (Means et al. 1990; Rusten and Jacobsen 1995). However, a direct inhibitory effect of TNF-α on erythroid cells was supported by subsequent studies. In one study, a direct effect of TNF-α on BFU-E was suggested, mediated by the TNF-R1 (Rusten and Jacobsen 1995). In another study using preparations of virtually pure erythroid precursors derived from unilineage erythroid cultures it was shown that TNF-α exerts a moderate, but signiﬁcant, inhibitory effect on the proliferation of immature erythroblasts (proerythroblasts and basophilic erythroblasts), associated with a slight induction of apoptosis, in the presence of EPO. Furthermore TNF-α, like other death receptor ligands, exerted an inhibitory effect on erythroid maturation (De Maria et al. 1999). A direct inhibitory effect of TNF-α on erythroid cells was also supported by a recent study showing that: (i) both murine and human CFU-E release TNF-α; (ii) the addition of neutralizing anti-TNF-α-antibody to cultures of human CD34+ cells stimulated with EPO increased the generation of erythroid cells; (iii) the number of BFU-E colonies was higher in the bone marrow of TNF-α−/− mice than in wt mice; (iv) the addition of TNF-α to TNF-α−/− mice elicited a signiﬁcant inhibition of BFU-E colony formation, while TNF-α exerted only a slight inhibitory effect on wt mice BFU-Es (Jacobs-Helber et al. 2003). An inhibitory effect of TNF-α on the generation of erythroid cells from puriﬁed human CD34+ cells was also reported in another recent study (Xiao et al. 2002). In this study it was also shown that detectable expression of TNF-R1 was observed only during the initial stages of erythroid differentiation, while TNF-R2 expression was observed during all stages (Xiao et al. 2002). However, it is unlikely that the inhibitory effects of TNF-α on erythropoiesis could be mediated via TNF-R2 signaling, since this receptor lacks a death domain in its cytoplasmic tail. The inhibitory effect of TNF-α on erythroid maturation may also NF-κB induction (Xiao et al.

36

U. Testa

2002). In fact, NF-κB activity is high in early erythroid progenitors and then declines at later stages of erythroid maturation. In fact, low NF-κB levels are required for the induction of the expression of erythroid-speciﬁc genes (Liu et al. 2003). Taken together, these observations suggest that TNF-α exerts an inhibitory effect on erythropoiesis and this effect could be physiologically relevant. An increased TNF-α production is involved in the inhibition of erythropoiesis observed in Fanconi anemia patients (Dufour et al. 2003). An increased TNF-α production seems to be involved also in the lack of response to EPO therapy observed in some patients with chronic kidney disease. In fact, patients with chronic kidney disease who persistently fail to respond to EPO therapy, express abnormally high levels of the pro-inﬂammatory cytokines TNF-α and IFN-γ, which are known to inhibit erythropoiesis (Macdougall and Cooper 2002; Cooper et al. 2003).

Effects of Fas-FasL pathway on normal erythropoiesis Activation-induced cell death (AICD) is the primary homeostatic mechanism used by the immune system to control T-cell responses, promote tolerance to self-antigens, and prevent autoimmunity. Following activation, T cells express Fas and FasL and become sensitive to Fas/FasL-mediated autocrine and paracrine apoptosis (Krammer 2000). Growing evidence indicates that the Fas/FasL system could also play a relevant role in the regulation of hematopoiesis and, particularly, of erythropoiesis. Fas is a 45-kDa, type I cell surface protein with an extracellular domain that binds to FasL and a cytoplasmic domain that transduces the death signal. Apoptosis is executed by the engagements and co-aggregation of Fasl with the Fas receptor on the cell surface followed by a series of intracellular molecular interactions that coordinate the hierarchical activation of caspases and cell death, as outlined above. FasL protein is expressed in three distinct molecular forms: (i) membranous form on the cell surface; (ii) membranous form stored in intracellular microvesicles, which are secreted into the intercellular milieu in response to various physiologic stimuli; (iii) soluble form (sFasL) generated from the cleavage of the membranous molecule by matrix metalloproteinases within minutes of cell surface expression. Membranous Fasl is the primary mediator of cell apoptosis through formation of trimers on the cell surface. In contrast, sFasL can have pro-apoptotic, anti-apoptotic or neutrophil chemiotactic functions. Initial studies on mice with mutations of either the Fas or FasL system failed to show signiﬁcant modiﬁcations of the hematopoiesis. However, a re-evaluation of the hematopoiesis in Fas or FasL-deﬁcient mice showed a striking extramedullary increase in hematopoietic progenitors, including both erythroid and nonerythroid progenitors. Furthermore, colony forming

Physiology of erythropoiesis

37

unit-spleen (CFU-S) and colony forming unit-culture (CFU-C) on peripheral blood and in the spleen clearly increased in these mice after birth (Schneider et al. 1999). These observations clearly indicate that the Fas/FasL system can affect hematopoiesis during both the fetal and adult life. These conclusions were supported by a recent study showing that mice with mutations of the Fas gene (lpr mice) and mostly of the FasL gene (gld mice) show a pronounced increase in the number of colony forming unit-granulocyte/ macrophage (CFU-GM) (Alenzi et al. 2002). The expression of Fas/FasL during erythroid differentiation, as well as the inhibitory effects of Fas activation, has been explored in detail. Initial studies were focused to evaluate Fas expression on immature hemopoietic progenitor cells and showed low Fas mRNA expression in CD34+ cells isolated from human bone marrow (Nagafuji et al. 1995; Takenaka et al. 1996). Immature primitive hematopoietic progenitors (CD34+/38−) from human fetal liver express signiﬁcant amounts of Fas antigen, whereas the more mature progenitors (CD34+/38+) showed low Fas antigen expression (Barcena et al. 1996). Both TNF-α and interferon-γ (IFN-γ) induced a marked increase in Fas expression on CD34+ cells (Maciejewski et al. 1995; Nagafuji et al. 1995; Barcena et al. 1996). Interestingly, the expression of Fas antigen on CD34+ cells was greatly increased following the induction of the cycling of these cells with cytokines, such as SCF, IL-3 and granulocyte/ macrophage colony forming factor (GM-CSF) (Barcena et al. 1996). Other studies have explored the consequences of Fas activation in progenitor cells on the induction of apoptosis, proliferation and differentiation. Basically these studies have provided evidence that immature progenitor cells are resistant to Fas-mediated apoptosis. Particularly, studies carried out in mice have shown that Lin-/Sca+/cKit+ stem cells reveal little or no constitutive expression of Fas and are resistant to the apoptotic triggering by an anti-Fas agonistic antibody (Bryder et al. 2001). However, if these stem cells are induced to cycle by incubation with early-acting growth factors and are incubated in the presence of TNF-α, they exhibit a marked up modulation of Fas antigen expression and become sensitive to Fas-mediated apoptotic triggering (Bryder et al. 2001). These ﬁndings are also conﬁrmed by studies carried out using human CD34+ cells derived from either peripheral blood or cord blood: these cells are found to be resistant to the apoptotic triggering elicited by either soluble FasL or anti-Fas agonistic antibody (Kim et al. 2002). However, if CD34+ cells are induced to proliferate in the presence of ﬂt3 ligand, SCF and thrombopoietin, they become sensitive to anti-Fasmediated apoptosis. The resistance of quiescent CD34+ cells to Fas-mediated apoptosis seems to be related to the expression in these cells of elevated levels of c-FLIP (FLICE-inhibitory protein), a dominant negative inhibitor of caspase-8 (Kim et al. 2002). Other studies have directly evaluated Fas/FasL expression in erythroid cells at various steps of differentiation and maturation (Fig. 7). Initial studies

38

U. Testa

Fig. 7. Kinetics of TRAIL, TRAIL-R1, TRAIL-R2, Fas and FasL expression during erythroid differentiation. Puriﬁed hemopoietic progenitors have been induced to unilineage erythroid differentiation; the developing erythroid cells have been harvested at different days of culture and processed for TRAIL, TRAIL-Rs, Fas and FasL by ﬂow cytometry after labeling with speciﬁc monoclonal antibodies

based on a strategy focused on obtaining a population of CFU-E cells originated in vitro from BFU-E, showed that these cells constitutively express low levels of Fas, whose expression was potentiated by IFN-γ (Dai et al. 1998); interestingly, CFU-E-like cells possess low, but signiﬁcant amounts of FasL, whose levels are potentiated by IFN-γ treatment and by differentiation (Dai et al. 1998). The expression of Fas and FasL was explored in detail by De Maria et al. (De Maria et al. 1999) using a selective system of unilineage culture of human erythroid cells starting from puriﬁed CD34+ peripheral blood cells. Using this cell culture system it was shown that Fas is clearly upregulated during initial steps of erythroid differentiation from BFU-E to CFU-E, reaching elevated levels of expression at the stage of immature erythroblasts (proerythroblasts and basophilic erythroblasts) and remaining expressed at signiﬁcant levels up to terminal stages of erythroid maturation (orthochromatic erythroblasts) (De Maria et al. 1999) (Fig. 7). These ﬁndings

Physiology of erythropoiesis

39

were also conﬁrmed through the analysis of erythroid cells present in normal bone marrow. Fas cross-linking was effective in inducing an apoptotic response only in immature erythroblasts, while mature erythroblasts are resistant to this triggering. In contrast, an opposite pattern of FasL expression was observed, this membrane-bound ligand being expressed in late differentiating Fas-insensitive erythroblasts, mostly at the orthochromatic stage. FasL expressed on the membrane of mature erythroblasts was found to be functional in that it was able to kill Fas-sensitive lymphoblast targets in a Fas-dependent manner. Importantly, FasL-bearing mature erythroblasts displayed a Fas-mediated cytotoxicity against immature Fas-positive erythroblasts, which was in part abrogated by high concentrations of EPO (De Maria et al. 1999). According to these ﬁndings, it was suggested that erythropoiesis is regulated according to a negative feedback mediated by mature and immature erythroblasts, whereby the former cells might exert a cytotoxic Fas-mediated effect on the latter cells within the erythroblastic island (Testa 2004). It is conceivable to assume that this feedback may operate in the presence of physiologic, low EPO concentrations, while it is inhibited when EPO concentrations are high and there is the need of high rate of erythroid cell production. As mentioned above, EPO was only in part able to protect immature erythroid cells from a strong triggering of the Fas-mediated apoptosis, as it is induced by Fas agonistic antibodies. There is, however, evidence that other growth factors, like SCF, may exert a more pronounced protective effect than EPO. In this context, an initial observation was derived from the studies of Lee et al. showing that anti-HLA-DR monoclonal antibody elicited a marked upregulation of Fas on primary human bone marrow cells, thereby increasing their susceptibility to Fas-mediated cell death. SCF partially antagonized Fas-mediated apoptosis of primary cells, which suggests that SCF protects hemopoietic precursor cells from Fas-mediated apoptosis (Lee JW et al. 1997). In a subsequent study it was shown that SCF might protect CFU-E from Fas-induced apoptosis: the protective effect of SCF was signiﬁcantly more effective than the effect elicited by EPO. This protective effect of SCF involves the activation of Src-kinases (Nishio et al. 2001). In a recent study it was shown that the simultaneous addition of IFN-γ and FasL to CFU-E elicited a marked activation of caspase-8 and caspase-3 with a consequent apoptotic response; this phenomenon was greatly inhibited by SCF addition, via a mechanism involving the upmodulation of c-FLIP expression (Chung et al. 2003). In another set of studies the mechanisms through which Fas-mediated apoptosis exerts a negative control on erythropoiesis was explored in detail. These experiments take advantage on the observation that Fas triggering in the presence of high EPO concentrations elicited a low apoptotic response in immature erythroblasts, but induced a blockade of erythroid maturation. This last mechanism requires caspase activation. One of the major effects of

40

U. Testa

caspase activation consisted in the degradation of the transcription factor GATA-1; the induction of GATA-1 degradation seems responsible for the Fas-mediated blockade of erythroid cell maturation since the transduction in erythroid progenitors of a caspase-resistant GATA-1 mutant resulted in a complete protection against Fas-mediated blockade of erythroid maturation. Interestingly, also the erythropoiesis blockade elicited by EPO deprivation was in part inhibited by the transduction of caspase-resistant GATA-1 in erythroid progenitors (De Maria et al. 1999). In a second study, evidence was provided that another transcription factor, Tal-1, is an additional target protein cleaved by activated caspases in erythroid cells. The transduction in hemopoietic progenitors of a Tal-1 mutant resistant to caspase cleavage protected erythroid cells from Fas-mediated inhibitory effect on erythroid maturation and cleavage of GATA-1. Interestingly, the expression of the Tal-1 caspase-resistant mutant in erythroid cells completely protects these cells from the apoptosis elicited by EPO deprivation (Zeuner et al. 2003). These observations indicate that the apoptotic effects and the inhibitory actions on cell differentiation elicited by EPO deprivation or by death receptor triggering involve as a necessary step the caspase-mediated cleavage of two transcription factors, Tal-1 and GATA-1, whose integrity and activity is essential for erythroid maturation and survival (Fig. 8).

Effects of TRAIL on normal erythropoiesis In addition to Fas and TNF-R1, erythroid cells express other death receptors, such as TRAIL-R1 and TRAIL-R2 (De Maria et al. 1999). The expression of these two membrane receptors was higher in immature than in mature erythroblasts: particularly, the highest TRAIL-R1 and TRAIL-R2 expression was observed in proerythroblasts and basophilic erythroblasts; in polychromatophylic erythroblasts the expression of these two receptors declined, and then almost completely disappeared in late orthochromatic erythroblasts (Fig. 7). TRAIL-R3 and TRAIL-R4 are not expressed at any stage of erythroid maturation (Secchiero et al. 2004). Furthermore, it was shown that the membrane-bound form of the TRAIL was also expressed in normal erythroblasts, during all stages of erythroid maturation. Therefore, the pattern of the TRAIL/TRAIL-R system in erythroid cells was highly comparable to that observed for the Fas/FasL system. The analysis of the biological effects of TRAIL on erythroid cells showed that immature erythroid cells are sensitive to the antidifferentiative (in the presence of EPO) or cytotoxic (at low EPO concentrations) effects induced by TRAIL following its interaction with the TRAIL-Rs expressed on these cells (De Maria et al. 1999). These ﬁndings were conﬁrmed in a subsequent study, showing that hemopoietic progenitors as well as mature erythroblasts are resistant to the apoptotic effects induced by TRAIL (Zamai et al. 2000).

Physiology of erythropoiesis

41

Fig. 8. Negative regulation of erythropoiesis triggered by death receptor activation or by EPO deprivation. Both death receptor activation or growth factor deprivation induce the apoptotic cascade, with consequent activation of caspases that cleave the transcription factors GATA-1 and Tal-1. The cleavage of these transcription factors is responsible for either the maturation arrest or apoptosis of erythroid cells

The blockade of erythroid maturation induced by TRAIL-R stimulation involves the ERK1/2 activation, as suggested by two observations: TRAIL induced ERK1 and ERK2 phosphorylation; pharmacological inhibitors of the ERK pathway blocked the anti-differentiative effects induced by TRAIL (Secchiero et al. 2004). The mechanism of TRAIL-induced apoptosis of erythroid cells was also explored in experimental models of erythroid differentiation. In this context, studies in the erythroleukemia K562 cell line showed a sensitivity of these cells to TRAIL only after induction with hemin. The induction of TRAIL sensitivity was not related to the stimulation of expression of TRAIL receptors, but to a downmodulation of c-FLIP (Hetakugas et al. 2003). Interestingly, the expression of TRAIL-R1 may be upmodulated by ionizing radiations on erythroleukemic cells, but not on normal erythroblasts. According to this pattern of TRAIL-R1 modulation, TRAIL sensitizes the cytotoxic effects of ionizing radiations on erythroleukemia cells, but not on normal erythroid cells (Di Pietro et al. 2001).

42

U. Testa

An increased production of TRAIL in the bone marrow represents one of the main pathogenetic mechanisms involved in the genesis of anemia observed in myeloma (Silvestris et al. 2002) and in myelodysplastic syndromes (Campioni et al. 2005).

Effects of IFN-g on normal erythropoiesis IFN-γ is the example of a cytokine that per se does not induce an apoptotic signaling, but may induce apoptosis of erythroid progenitors through an indirect mechanism involving upmodulation of both Fas and FasL, as mentioned above. Interestingly, recent studies clearly indicate that the inhibitory effect exerted by IFN-γ involves multiple members of the TNF superfamily, including FasL, TRAIL and TWEAK (Felli et al. 2005). In fact, only the combined neutralization of FasL, TWEAK and TRAIL was able to restore erythroid cell survival, proliferation and maturation in the presence of IFN-γ, indicating the simultaneous involvement of these three ligands as effectors of IFN-γ in erythropoietic inhibition. The inhibitory effect of IFN-γ on erythropoiesis is exerted at the level of early stages of erythroid differentiation: in in vitro assays, BFU-Es are inhibited by IFN-γ more than CFU-Es (Wang et al. 1995). In addition, IFN-γ upmodulates the level of expression and activates several caspases, including caspase-8 and -3, in erythroid progenitors (Dai and Krantz 1999). Paradoxically, IFN-γ may exert a protective effect on apoptosis on erythroid cells at later stages of differentiation (Choi et al. 2000). Ceramide, an intracellular second messenger produced by sphingommyelin hydrolysis has been involved as the mediator of apoptosis induced by a number of cytokines, including IFN-γ and TNF-α. Ceramide induces an inhibition of CFU-E colony formation. Interestingly, the ceramide antagonist sphingosine-1-phosphate signiﬁcantly reversed the CFU-E colony inhibition induced by IFN-γ. These observations suggest that ceramide is one of the key mediators of the inhibition of CFU-E colony formation by IFN-γ (Dallalio et al. 1999).

Role of caspases in erythroid maturation The caspases are the central executioners of the apoptotic process. There is evidence, however, that these proteolytic enzymes may display, in addition to a role in the apoptotic process, also a function in cell differentiation/maturation. The activation of some caspases is required for the process of normal erythroid differentiation. Recent studies have provided evidence that the “spontaneous” activation of caspases was observed during late steps of differentiation in various hemopoietic lineages, including monocytic, megakaryocytic and erythroid lineages.

Physiology of erythropoiesis

43

During the process of normal maturation of monocytes to macrophages an activation of caspases 3 and 9 was observed, associated with release of cytochrome C from the mitochondria and cleavage of the protein acinus, but without induction of cell death. The blockade of the caspase activity when using synthetic cell-permeable inhibitors blocked the cell maturation process (Sordet et al. 2002). Similar observations have been made in megakaryocytic cells, where caspase activation occurred at two stages of maturation: (i) a ﬁrst step of caspase activation (caspase 3 and 9 became spontaneously activated) occurred before proplatelet activation and was limited only to some cellular compartments; (ii) a second step of diffuse caspase activation occurred at the end of the maturation process after platelet release. The ﬁrst process of caspase activation did not lead to cell death, but the second wave was associated with cell death (Botton et al. 2002). Importantly, the addition of caspase inhibitors to megakaryocytic cultures inhibited platelet release. In line with these observations transgenic mice overexpressing the antiapoptotic gene Bcl-XL exhibited impaired platelet fragmentation (Kaluzhny et al. 2002). Several lines of evidence indicate that caspase activation occurs also during the process of erythroid maturation and could play an important role in this process. In this context, an initial study showed that caspases 1, 2, 3, 5, 6, 7, 8 and 9 are expressed in erythroid cells. The level of procaspase 2, 3 and 8 were markedly higher in immature erythroblasts than in mature erythroblasts (Gregoli and Bondurant 1999). As it is expected, EPO deprivation elicited a marked increase in caspase activation. In a second study based on a peculiar cell culture system which implies a ﬁrst ampliﬁcation of CD34+ cells in the presence of IL-3, SCF and IL-6, after 7 days, erythroid progenitors selected according to the CD36 positivity were cultured or additional 7 days in the presence of SCF, IL-3, EPO and TGFβ1. The CD36+ erythroid progenitors underwent a progressive differentiation during the second week of culture. A transient caspase-3 and -7 activation occurred during erythroid maturation corresponding to proerythroblasts and basophilic erythroblasts; this phenomenon, however, was transient in that the caspase activation regressed at later stages of erythroid maturation. These caspases, transiently activated through the mitochondrial pathway, cleaved proteins involved in nucleus integrity (lamin B) and chromatin condensation (acinus) without inducing cell death. Inhibitors of caspases, such as z-VAD, added to erythroid cultures just before the moment of caspase activation elicited a block of erythroid maturation at the basophilic stage (Zermati et al. 2001). The selective blockade of caspase-3 by transfection of small interfering RNA (siRNA) directed to caspase-3 in differentiating erythroid precursors elicited the inhibition of erythroid maturation (Carlile et al. 2004). These ﬁndings have also been conﬁrmed in murine erythroblasts. Furthermore, it was shown that the overexpression of Raf-1, which prevents caspase activation, impairs erythroid maturation by reducing differentiation-associated caspase activation. An opposite phenomenon was observed in Raf-1−/− mice (Kolbus et al. 2003).

44

U. Testa

The terminal maturation of erythroblasts is characterized by enucleation, a process of nuclear extrusion, preceded by nuclear chromatin condensation, reduction of nuclear size and migration of the nucleus to the plasma membrane. In the two other cell types showing enucleation at the end of maturation, keratinocytes and lens epithelial cells, caspase activation is required for enucleation. In contrast, erythroblast enucleation does not require caspase activation. More particularly, erythroblast enucleation does not involve a marked reorganization of nuclear structures by activated caspases (Krauss et al. 2005). Therefore, erythroblast enucleation, the terminal step of erythroblast maturation, occurs without evidence of major activation of apoptotic proteins. A possible involvement of caspase in the control of erythroid cell production is also supported by the analysis of the phenotype of mice with the targeting of the genes encoding some of these caspases. In this context, the most interesting ﬁndings were originated from the analysis of caspase-8−/− mice (Varfolomeev et al. 1998). Caspase-8 deﬁciency resulted in embryonic lethality, associated with two salient features: impaired heart muscle development and congested accumulation of erythrocytes (Varfolomeev et al. 1998). These features resemble the phenotype reported in mice with targeted disruption of FADD gene (Yeh et al. 1998; Zhang J et al. 1998). Recent studies show a nonapoptotic role of caspase-8 in the control of hemopoietic differentiation (Kang et al. 2004). The analysis of the defects in the erythroid lineage observed in these animals, however, was limited to the analysis of erythrocytes and it is, therefore, impossible to know whether mice with caspase8 and FADD deﬁciency exhibit abnormalities of erythroid cell maturation, in addition to an increased expansion of the number of mature erythrocytes. It is of interest that in a recently described family exhibiting a complete deﬁciency of caspase-8, no hemopoietic abnormalities have been described (Chun et al. 1998). The discrepancy between the phenotype of caspase-8 deﬁciency in humans and mice could be related to the function in humans of caspase-10, the closest paralogue of caspase-8 (caspase-10 has no known orthologue in mice). In addition to caspases, recent studies suggest a possible role of p53 during late stages of erythroblast maturation. Particularly, an overexpression of the p53 protein was observed in late orthochromatic erythroblasts: this p53 activation may be related to the nuclear degradation occurring in these cells, without a fully executed apoptotic process because of exhaustion of caspase3/7 (Peller et al. 2003).

Apoptotic mechanisms in mature RBC The occurrence of apoptotic processes and, particularly, of a possible activation of caspases was explored also in mature RBC. These cells lack the

Physiology of erythropoiesis

45

nucleus and after a life-span of about 120 days undergo a process of senescence and are then removed from the circulation. During this process senescent RBC undergo typical morphological changes, such as shrinkage as the consequence of a progressive release of microvesicles from the cell membrane and shape transformation from a discocyte to a spherocyte form, enabling ﬁrst their recognition and then their phagocytosis by macrophages. In addition to these changes, there is a series of modiﬁcations at the level of the cell membrane, such as a progressive loss of the structural membrane protein spectrin and the loss of membrane asymmetry associated with the externalization of phosphatidylserine residues, resembling those observed in the process of apoptotic cell death in nucleated cells. These observations prompted studies aimed to evaluate whether mature RBC possess an apoptotic machinery similar to that observed in the majority of nucleated cells. In this context, preliminary experiments carried out in human RBC showed that these cells do not undergo programmed cell death when either treated with staurosporine and cycloheximide or when cultured in the absence of serum, conditions that induce apoptosis in all types of nucleated cells. According to these observations it was concluded that mature RBC do not have an apoptotic machinery and therefore do not undergo an apoptotic programme. This issue was re-explored on chicken RBC showing that serum deprivation or treatment with staurosporine and cycloheximide induces the death of these cells; however, although these erythrocyte deaths displayed many features that are typical of apoptotic cells, they are not blocked or inhibited by different types of caspase inhibitors (Weil et al. 1998). According to these observations it was concluded that chicken RBC die without apparently activating caspases. More recently, the problem was reevaluated focusing the attention to human RBC. In this context, it was observed that mature erythrocytes contain considerable amounts of caspase-3 and caspase-8, while other main components of the apoptotic machinery such as caspase-9, Apaf-1 and cytochrome C are absent. Although present at relatively high levels, caspase3 and caspase-8 were not activable by various types of proapoptotic stimuli (Berg et al. 2001). However, recent studies suggest a possible role for caspase3 in erythrocyte aging. First, it was observed that activated caspase-3 can be detected in old, but not in young RBC; second, this activated caspase-3 was able to cleave cell membrane band 3, disrupting its interaction with the peripheral membrane protein 4.2 (Mandal et al. 2003). These observations suggested that some caspases activated during RBC aging could participate in the degradation of crucial erythrocyte membrane proteins involved in the maintenance of shape and function. In line with these ﬁndings it was observed that loss of band 3 was associated with premature erythroid cell death (dyserythropoiesis) (Berg et al. 2001). Since an increase in Ca2+ concentration was associated with the aging of RBC and was considered as one of the molecular mechanisms responsible

46

U. Testa

for the senescence of these cells, additional studies explored whether the induction of an increase of Ca2+ concentration in RBC could trigger an apoptotic or apoptosis-like response. Following the induction of an increase of cytosolic Ca2+ concentration RBC underwent a rapid self-destruction process exhibiting several apoptotic features, such as plasma membrane microvesiculation, and phosphatidylserine externalization, a process which culminated either in RBC destruction or, in the presence of macrophages, in macrophage phagocytosis with subsequent destruction. This process was inhibited by cysteine protease inhibitors, but not by caspase inhibitors. Importantly, cysteine protease inhibitors allowed erythrocyte survival both in vitro and in vivo (Bratosin et al. 2001). According to these ﬁndings it was proposed that an increase in intracellular Ca2+ concentration in RBC determines the activation of the cysteine protease calpain that mediates spectrin cleavage and other morphological changes leading to cell shrinkage (Fig. 8).

Role of Deoxyribonuclease (DNase) IIa in erythroid maturation During terminal stages of erythroid maturation, erythroblasts exhibit several features typical of an apoptotic process, such as reduction of cell size, withdrawal from the cell cycle, nuclear condensation and nuclear expulsion. The progressive transition from the rapidly proliferating compartment of immature erythroblasts to the terminal slowly proliferating compartment of mature erythroblasts is regulated by growth inhibitory genes (Aerbajinai et al. 2004). Many endonucleases have been implicated in the apoptotic process and particularly in DNA digestion occurring during apoptosis, such as caspaseactivated DNase (CAD), DNaseI and DNaseII. Two types of DNase II exist in mammalians, designated as DNase IIα and DNase IIβ. DNase IIα is ubiquitously expressed, while DNase IIβ is expressed only in the salivary gland. Recent studies have shown a role for this enzyme in erythroid maturation, particularly in the process of enucleation occurring at very late stages. To evaluate this, mice deﬁcient in DNase IIα expression were generated. These animals died of severe anemia and showed a marked decrease of circulating RBC (i.e. they have only 1/10 of the physiological number of circulating erythrocytes), associated with the presence in the circulation of deﬁnitive nucleated RBC, a cell type normally present only in bone marrow and not in peripheral blood. The reduced presence of anucleated RBC in the peripheral blood of DNase IIα−/− mice could not be related to a defect in erythroid commitment, since these animals exhibited in their fetal liver a normal number of both BFU-E and CFU-E (Kawane et al. 2001). However, mutant fetal liver erythroblasts showed a normal maturation after transfer into irradiated normal mice, thus suggesting that the defect of DNase II−/− erythroblasts was not intrinsic to these cells. In fact, histological experiment suggested that the source of DNase II responsible for the rescue of the maturation of

Physiology of erythropoiesis

47

DNase II−/− erythroblasts was represented by macrophages. It was in fact suggested that central macrophages present in erythroblastic islands might represent the source of DNase IIα (Kawane et al. 2001). According to these ﬁndings it was concluded that DNase II present in macrophages was responsible for digesting nuclear DNA expelled from maturing late erythroblasts. This conclusion was directly supported by the analysis of a second knockout of DNase IIα. The generated DNase II−/− mice died at birth exhibiting at the level of the liver macrophages large DNA bodies resulting from the engulfment of undigested nuclei extruded from erythroblasts (Krieser et al. 2002). According to these ﬁndings, macrophages appear to play an essential role in erythroid maturation through a biochemical mechanism involving DNase IIα. These observations indicate that a regulated process of DNA fragmentation is strictly required for normal tissue homeostasis (Zhang and Xu 2002), and that the enucleation event occurring during late erythroid maturation is a regulated process requiring extensive signaling between erythroblasts and macrophages.

Role of AKT pathway in the apoptotic control of the erythroid lineage Phosphorylation and activation of the serine/threonine kinase PKB controls fundamental processes such as cell cycle progression, survival, apoptosis and mRNA translation. The AKT/PKB pathway involves a cascade of activation/signaling steps (Brazil et al. 2002; Djordjevic and Driscoll 2002; Nicholson and Anderson 2002; Vivanco and Sawyers 2002): – the ﬁrst step involves the activation and modulation of phosphatidylinositol-3-kinase (PI3K); PI3K may be activated through different mechanisms, involving either the binding of the complex p85/p110 (p85 is the regulatory subunit, while p110 is the catalytic subunit) to phosphorylated tyrosine residues present at the level of the cytoplasm domain of activated receptor tyrosine kinases or the interaction with members of the small GTPase family, such as Ras; – the second step consists in the generation of 3′-phosphorylated inositol lipids (PIP3) which function as classical second messengers by binding proteins that harbor PIP3-binding domains. PIP3 binds the binding domain of the serine/threonine PKB/AKT, with its consequent activation. The activated PKB/AKT phosphorylates several components of the apoptotic machinery and through this mechanism induces anti-apoptotic effects. Particularly, PKB/AKT promotes cell survival by multiple mechanisms: i) phosphorylation and inactivation of the proapoptotic protein Bad; ii) maintenance of mitochondrial integrity; iii) decrease of the expression of death genes, such as FasL and Bim, via the phosphorylation of forkhead transcription factor FKHR-L1: phosphorylated FKHR-L1 is exported from the nucleus to the cytoplasm where it is sequestered by 14-3-3 pro-

48

U. Testa

teins; iv) increase of the transcription of survival genes through the activation of NF-κB and CREB transcription factors (Djordjevic and Driscoll 2002; Nicholson and Anderson 2002; Vivanco and Sawyers 2002). Furthermore, recent studies indicate that the PKB/AKT pathway is involved in the upregulation of surviving expression observed during cell cycle entry (Fukuda et al. 2002). Survivin pertains to the family of inhibitors of apoptosis protein (IAP) that inhibit apoptosis by inactivating several caspases (Altieri 2003). Finally, PI3K is involved in Gab2 phosphorylation. Both EPO and SCF activate phospatidylinositol-3.kinase (PI3K) and its target PKB. However, compared with EPO, SCF is much more potent in inducing PKB phosphorylation (von Lindern et al. 2004). Both the catalytic p110 and the regulatory p85 subunits are required for PI3K activity. P85−/− mice are pale with a marked reduction in RBC in their peripheral blood (Huddleston et al. 2003). Erythroid progenitors BFU-Es and CFU-Es are reduced in the fetal liver of these animals and display reduced proliferative response to both EPO and SCF (Huddleston et al. 2003). A molecular balance between the regulatory and catalytic subunits of PI3K is therefore required for optimal control of erythropoiesis. Initial studies on erythropoiesis and PKB/AKT signaling were based on the use of chemical inhibitors of PI3K. In a ﬁrst set of studies, a PI3K inhibitor called wortmannin was used: basically these studies provided evidence that the inhibitor blocked the PI3K activation elicited by EPO and considerably decreased the proliferation of erythroid precursors expanded in vitro (Sui 1998). Using a more speciﬁc and potent PI3K inhibitor (LY294002), an inhibitory effect on erythroid cell proliferation was conﬁrmed, but it was also shown that PKB/AKT inhibition elicited the apoptotic death of a signiﬁcant proportion of erythroid precursors (Haseyama et al. 1999). In a subsequent study the effects of the PI3K inhibitor LY294002 at earlier stages of erythroid differentiation were explored, showing a marked inhibitory effect on the differentiation of CD34+ to erythroid precursors: particularly, the addition of the inhibitor to CD34+ cultures grown in the presence of EPO and SCF resulted in a marked inhibition of the generation of glycophorin-A+ cells (Myklebust et al. 2002). The essential role of the PI3K in the mechanism of induction of erythroid proliferation and survival was also conﬁrmed in a recent study showning that three different mechanisms equally active in erythroid cells are responsible for PI3K activation: direct association of the PI3K to the EPO-R; phosphorylation of Gab via either Tyr 343 or Tyr 401 of the EPO-R; phosphorylation of the IRS2 adaptor protein (Bouscary et al. 2003). The activation of PKB/AKT following EPO stimulation requires a normal protein kinase C (PKC) activity, as suggested by the experiments carried out with PKC inhibitors (Von Lindern et al. 2000). The Forkhead box, class O (FoxO) subfamily of Forkhead transcription factors is an important effector of PKB in regulating apoptosis and cell cycle

Physiology of erythropoiesis

49

progression. Members of this subfamily, FoxO4, FoxO1a and FoxO3a (FKHR-L1) are directly phosphorylated by PKB, leading to cytoplasmic retention and inhibition of their transcriptional activity required for the control of the expression of several genes involved in cell cycle control, cell death and oxidative stress (Burgering and Kops 2002). There is evidence that the activation of FKHR-L1 alone can recapitulate all known elements of the apoptotic program normally induced by cytokine withdrawal (Dijkers et al. 2002). FKHR-L1 is expressed in erythroid cells, and its expression progressively decreases during erythroid maturation (Kashii et al. 2000). In immature erythroid precursors, EPO rapidly phosphorylated FKHR-L1 through a PI3K dependent pathway (Kashii et al. 2000). In addition to FKHR-L1, also FoxO3a exhibits an important role in erythropoiesis. In fact, FoxO3a expression and nuclear accumulation increased during erythroid differentiation. Enforced expression of FoxO3a in erythroid progenitors accelerated differentiation of these cells to erythroblasts (Bakker et al. 2004). An additional target of PKB/AKT is the glycogen synthetase kinase-3 (GSK 3), which is a serine/threonine kinase involved in metabolic processes (glycogen metabolism) and also implicated in apoptosis regulation in primary human erythroid progenitors. The activity of GSK 3 is suppressed by EPO, as well as by SCT. Importantly, the inhibition of GSK3 using speciﬁc chemical inhibitors prevents apoptosis of erythroid progenitors by EPO deprivation (Somervaille et al. 2001). Taken together, these observations implicate at least two PKB/AKT targets, FKHR-L1 and GSK3, in the regulation of the survival of erythroid cells.

Further signaling pathways involved in the apoptotic regulation of erythroid cells In addition to the PKB/AKT, other signaling pathways seem to play a role in the apoptotic control of erythroid cells. In this context, it was shown that the transcription activator protein 1 (AP1) could be involved in the regulation of apoptosis mediated by EPO in erythroid cells. AP1 is a transcriptional complex comprising members of the Jun and of the Fos families of transcription factors. Evidence was provided both in primary human hemopoietic progenitors and in EPO-dependent cell lines that both EPO addition or deprivation induces AP-1 activity which seems to be required both for induction of erythroid proliferation or apoptosis. The analysis of the composition of AP-1 transcriptional complex in these two conditions showed an interesting ﬁnding: JunB is present in erythroid cells triggered to apoptosis by EPO deprivation, while c-Jun was present in erythroid cells induced to proliferate (Jacobs-Helber et al. 2000). Finally, a recent study showed that one of the two negative regulators of EPO-R signaling, the small cytokine-inducible SH2-domain (CIS) containing protein plays a role in the control of erythroid cell apoptosis. CIS belongs to the family of suppressor of cytokine signaling

50

U. Testa

proteins and acts reducing the Stat-5 activation elicited by EPO. The enforced expression of this protein in erythroid progenitors not only reduced their proliferation, but also induced a signiﬁcant level of apoptosis (Ketteler et al. 2003). Due to these ﬁndings it may be that the level of CIS protein is involved in the control of both the proliferation and survival of erythroid progenitors.

Summary Erythropoiesis is a complex multistep process encompassing the differentiation of hemopoietic stem cells to mature erythrocytes. The stem cells ﬁrst differentiate to early erythoid progenitors, BFU-E, then to late erythroid progenitors, CFU-E, and ﬁnally to morphologically recognizable erythroid precursors. A key event in the the late stage of erythropoiesis is nuclear condensation, which is followed by extrusion of the nucleus to produce enucleated reticulocytes and ﬁnally mature erythrocytes. During the differentiation process, the cells became progressively sensitive to EPO, which controls both the survival and proliferation of erythroid cells. A normal homeostasis of the erythropoietic system requires an appropriate balance between the rate of erythroid cell production and RBC destruction. Anumber of apoptotic and antiapoptotic mechanisms play a major role in the control of erythropoiesis both under physiologic and pathologic conditions. Withdrawal of EPO or stimulation of death receptors such as Fas or TRAIL-Rs leads to activation of a subset of caspase-3, -7 and -8, which then cleave the transcription factors GATA-1 and TAL-1 and trigger apoptosis. In addition, there is evidence that a number of caspases are physiologically activated during erythroid differentiation and are functionally required for erythroid maturation. Several caspase substrates are cleaved in differentiating cells, including the protein acinus whose activation by cleavage is required for chromatin condensation. The studies on normal erythropoiesis clearly indicate that immature erythroid precursors are sensitive to apoptotic triggering mediated by activation of the intrinsic and extrinsic apoptotic pathways. These apoptotic mechanisms are frequently exacerbated in some pathologic conditions, associated with the development of anemia (i.e. thalassemias, multiple myeloma, myelodysplasia, aplastic anemia). The considerable progress in our understanding of the apoptotic mechanisms underlying normal and pathologic erythropoiesis may offer the way to improve the treatment of several pathologic conditions associated with the development of anemia.

References 1. Adelman CA, Chattopadhyay S, Bieker JJ (2002) The BMP/BMPR/Smad pathway directs expression of the erythroid-speciﬁc EKLF and GATA-1 transcription factors during embryoid body differentiation in serum-free media. Development 129: 539–549

Physiology of erythropoiesis

51

2. Adelman DM, Maltepe E, Simon MC (1999) Multilineage embryonic hematopoiesis requires hypoxic ARNT activity. Genes Dev 13: 2478–2483 3. Aerbajinai W, Lee YT, Wojda U, Barr VA, Miller JL (2004) Cloning and characterization of a gene expressed during terminal differentiation that encodes a novel inhibitor of growth. J Biol Chem 279: 1916–1921 4. Akahane K, Hosoi T, Urabe A, Kawakami M, Takakau F (1987) Effects of recombinant human tumor necrosis factor (rhTNF) on normal human and mouse progenitor cells. Int J Cell Cloning 5: 16–26 5. Alenzi F, Marley SB, Lewis JL, Chandrasherkran A, Warrens AN, Goldman J, Gordon MY (2002) A role for the Fas/Fas ligand apoptotic pathway in regulating myeloid progenitor cell kinetics. Exp Hematol 30: 1428–1435 6. Altieri DC (2003) Validating survivin as a cancer therapeutic target. Nat Cancer Rev 3: 46–54 7. Ariyoshi K, Nosaka T, Yamada K (2000) Constitutive activation of Stat5 by a point mutation in the SH2 domain. J Biol Chem 275: 24407–24413 8. Ashkenazi A (2002) Targeting death and decoy receptors of the tumournecrosis factor superfamily. Nat Cancer Rev 2: 420–430 9. Back J, Dierich A, Bronn C, Kastner P, Chan S (2004) PU.1 determines the self-renewal capacity of erythroid progenitor cells. Blood 103: 3615– 3623 10. Bakker WJ, Blanquez-Domingo M, Kolbus A, Besooyen J, Steinlein P, Beug H, Coffer PJ, Lowenberg B, von Lindern M, van Dijk TB (2004) FoxO3a regulates erythroid differentiation and induces BTG1, an activator of protein arginine methyl transferase 1. J Cell Biol 164: 175–184 11. Barcena A, Park SW, Bonapour B, Muench MO, Mechettner E (1996) Expression of Fas/CD95 and Bcl-2 by primitive hematopoietic progenitors isolated from human fetal liver. Blood 88: 2013–2022 12. Bartunck P, Kralova J, Blendinger G, Dvorak M, Zenke M (2003) GATA-1 and c-myb crosstalk during RBC differentiation through GATA-1 binding sites in the c-myb promoter. Oncogene 22: 1927–1935 13. Berg CP, Engels IH, Rothbard A, Lauber K, Renz A, Schlosser JF, SchulzeOsthoff K, Wesselborg S (2001) Human mature red blood cells express caspase3 and caspase-8, but are devoid of mitochondrial regulators of apoptosis. Cell Death Differ 8: 1197–1206 14. Bernard J (1991) The erythroblastic island: past and future. Blood Cells 17: 5–14 15. Bhardway A, Aggarwal BB (2003) Receptor-mediated choreography of life and death. J Clin Immunol 23: 317–332 16. Bodmer JL, Schneider P, Tschopp J (2002) The molecular architecture of the TNF superfamily. Trends Biochem Sci 27: 19–26 17. Botton S, Sabri S, Zermaty Y, Guidotti JE, Hermine O, Kroemer G, Vainchenker W, Debili N (2002) Platelet formation is the consequence of caspase activation within megakaryocytes. Blood 100: 1310–1317 18. Bouscary D, Pene F, Claessens YE, Muller O, Chrètien S, Fontenay-Roupie M, Gisselbrecht S, Mayeux P, Lacombe C (2003) Critical role for PI 3Kinase in the control of erythropoietin-induced erythroid progenitor proliferation. Blood 101: 3436–3443 19. Brand M, Ranish JA, Kummer NT, Hamilton J, Igarashi K, Francastel C, Chi TH, Crabtree GR, Aebersold R, Groudine M (2004) Dynamic changes in

52

20.

21. 22.

23. 24.

25. 26. 27. 28. 29.

30.

31.

32.

33. 34.

35.

36.

U. Testa transcription factor complexes during erythroid differentiation revealed by quantitative proteomics. Nat Struct Biol 11: 73–77 Bratosin D, Estaquier J, Petit F, Arnoult D, Montreuil J, Ameisen JC (2001) Programmed cell death in mature erythrocytes: a model for investigating death effector pathways operating in the absence of mitochondria. Cell Death Diff 8: 1143–1156 Brazil DP, Park J, Hemmings BA (2002) PKB binding proteins: getting in on the Akt. Cell 111: 293–303 Bryder D, Ramsfjell V, Dybedal I, Borge DJ, Jacobsen EW (2001) Self-renewal of multipotent long-term repopulating hematopoietic stem cells is negatively regulated by Fas and tumor necrosis factor activation. J Exp Med 194: 941–952 Burgering B, Kops G (2002) Cell cycle and death control: long live Forkheads. TRENDS Biochem Sci 27:352–360 Campioni D, Secchiero P, Corallini F, Melloni E, Capitani S, Lanza F, Zauli G (2005) Evidence for a role of TNF-related apoptosis-inducing ligand (TRAIL) in the anemia of myelodysplastic syndromes. Am J Pathol 166: 557–563 Cantor AB, Orkin SH (2002) Transcriptional regulation of erythropoiesis: an affair involving multiple partners. Oncogene 21: 3368–3376 Carlile GW, Smith DH, Wiedmann M (2004) Caspase-3 has a nonapoptotic function in erythroid maturation. Blood 103: 4310–4316 Carnot P, DeFlandre C (1906) Sur l’activité hemopoietique du serum au cours de la regeneration du sang. C. R. Acad Sci (Paris) 143: 384–386 Cerdan C, Rouleau A, Bhatia M (2004) VEGF-A165 augments erythropoietic development from human embryonic stem cells. Blood 103: 2504–2512 Chida D, Miura O, Yoshimura A, Miyajima A (1999) Role of cytokine signaling molecules in erythroid differentiation of mouse fetal liver hemopoietic cells: functional analysis of signaling molecules by retrovirus-mediated expression. Blood 93: 1567–1578 Choi I, Muta K, Wickhema A, Krantz SB, Nishimura J, Nawata H (2000) Interferon gamma delays apoptosis of mature erythroid progenitor cells in the absence of erythropoietin. Blood 95: 3742–3749 Chun HJ, Zheng L, Strauss SE, Lenardo MJ (1998) Pleiotropic defects in lymphocyte activation caused by caspase-8 mutations lead to human immunodeﬁciency. Nature 419: 395–399 Chung IJ, Dai C, Krantz SB (2003) Stem cell factor increases the expression of FLIP which inhibits IFN-γ induced apoptosis in human erythroid progenitor cells. Blood 101: 1324–1328 Clark AJ, Doyle KM, Humbert PO (2004) Cell-intrinsic requirement for pRb in erythropoiesis. Blood 104: 1324–1326 Coghill E, Eccleston S, Fox V (2001) Erythroid Kruppel-like factor (EKLF) coordinates erythroid cell proliferation and hemoglobinization in cell lines derived from EKLF null mice. Blood 97: 1861–1868 Cole J, Ertoy D, Lin H, Sutiliff R, Ezan E, Guyene T, Capecchi M, Corvol P, Bernstein KE (2000) Lack of angiotensin II-facilitated erythropoiesis causes anemia in angiotensin-converting anzyme-deﬁcient mice. J Clin Invest 106: 1391–1398 Cooper AC, Mikhail A, Lethbridge MW, Kemeny M, Macdougall IC (2003) Increased expression of erythropoiesis inhibiting cytokines (IFN-γ, TNF-α,

Physiology of erythropoiesis

37. 38.

39.

40. 41.

42. 43.

44.

45. 46.

47.

48.

49.

50.

51.

53

IL-10 and IL-13) by T cells in patients exhibiting a poor response to erythropoietin therapy. J Am Soc Nephrol 14: 1776–1784 Cory S, Adams JM (2002) The Bcl-2 family: regulators of the cellular life-ordeath switch. Nat Rev Cancer 2: 647–656 Dai C, Krantz SB (1999) Interferon-γ induces upregulation and activation of caspases 1, 3, and 8 to produce apoptosis in human erythroid progenitor cells. Blood 93: 3309–3316 Dai CH, Price JO, Brunnet T, Krantz SB (1998) Fas ligand is present in human erythroid colony-forming cells and interacts with Fas induced by interferon-γ to produce erythroid cell apoptosis. Blood 91: 1235–1242 Dallalio G, North M, Worden BD, Means RT (1999) Inhibition of human erythroid colony formation by ceramide. Exp Hematol 27: 1133–1138 Dame C, Sola MC, Lim KC, Leach KM, Fandrey J, Ma Y, Knopﬂe G, Engel JD, Bungert J (2004) Hepatic erythropoietin gene regulation by GATA-4. J Biol Chem 279: 2955–2961 Danial NN, Korsmeyer SJ (2004) Cell death: critical control points. Cell 116: 205–219 De Maria R, Testa U, Lucchetti L, Zeuner A, Stassi G, Pelosi E, Riccioni R, Felli N, Samoggia P, Peschle C (1999) Apoptotic role of Fas/Fas Ligand system in the regulation of erythropoiesis. Blood 93: 796–803 De Maria R, Zeuner A, Eramo A, Domenichelli C, Bonci D, Grignani F, Srinivasula SM, Alnemri ES, Testa U, Peschle C (1999) Negative regulation of erythropoiesis by caspase-mediated cleavage of GATA-1. Nature 401: 489–493 Dempsey PW, Doyle SE, He JQ, Cheng G (2003) The signalling adaptors and pathways activated by TNF superfamily. Cytokine Growth Factor Rev 14: 193–209 Deverman BE, Cook BL, Manson SR, Niederhoff RA, Langer EM, Rosova I, Kulans LA, Fu X, Weinberg JS, Heinecke JW, Roth KA, Weintraub SJ (2002) Bcl-XL deamidation is a critical switch in the regulation of the response to DNA damage. Cell 111: 51–62 Dijkers PF, Birkenkamp K, Lam E, Thomay S, Lammers JW, Koenderman L, Coffer PJ (2002) FKHR-L1 can act as a critical effector of cell death induced by cytokine withdrawal: protein kinase B-enhanced cell survival through maintenance of mitochondrial integrity. J Cell Sci 156: 531–542 Djordjevic S, Driscoll PC (2002) Structural insight into substrate speciﬁcity and regulatory mechanisms of phophoinositide 3-kinase. Trends Biochem Sci 27: 426–432 Di Pietro R, Secchiero P, Rana R, Gibellini D, Visani G, Bemis K, Zamai L, Miscia S, Zauli G (2001) Ionizing radiation sensitizes erythroleukemic cells but not normal erythroblasts to tumor necrosis factor-related apoptosis-inducing ligand (TRAIL)-mediated cytotoxicity by selective up-regulation of TRAIL-R1. Blood 97: 2596–2603 Dolznig H, Habermann B, Stangl K, Deiner EM, Moriggi R, Beug H, Mullner EW (2002) Apoptosis protection by the EPO target Bcl-X(L) allows factorindependent differentiation of primary erythroblasts. Curr Biol 12: 1076–1085 Dolznig H, Boulmé F, Stangl K, Deiner EM, Mikulitis W, Beug H, Mullner EW (2001) Establishment of normal, terminally differentiating mouse erythroid progenitors: molecular characterization by cDNA arrays. FASEB J 15: 1442– 1444

54

U. Testa

52. Drissen R, Von Lindern M, Kolbus A, Driegen S, Steinlein P, Beug H, Grosveld F, Philipsen S (2005) The erythroid phenotype of EKLF-null mice: defects in hemoglobin metabolism and membrane stability. Mol Cell Biol 25: 5205–5214 53. Dufour C, Corcione A, Svahn J, Haupt R, Poggi V, Beka’ssi AN, Scimè R, Pistorio A, Pistoia V (2003) TNF-α and IFN-γ are overexpressed in the bone marrow of Fanconi anemia patients and TNF-α suppresses erythropoiesis in vitro. Blood 102: 2053–2059 54. Dumon S, Constantino R, Santos S (1999) IL-3 dependent regulation of Bcl-XL gene expression by Stat5 in a bone marrow derived cell line. Oncogene 18: 4191–4199 55. Ebert BL, Bunn HF (1999) Regulation of the erythropoietin gene. Blood 94: 1864–1877 56. Ehleben W, Porwol T, Fandrey J, Kummer W, Acker H (1997) Cobalt and deferioxamine reveal crucial members in the oxygen sensing pathway in HepG2 cells. Kidney Int 51: 483–497 57. Elagib KE, Xiao M, Hussaini IM, Delehanty LL, Palmer LA, Racke FK, Birrer MJ, Shamugasundaram G, McDevitt MA, Goldfarb AN (2004) Jun blockade of erythropoiesis: role for repression of GATA-1 by HERP2. Mol Cell Biol 24: 7779–7794 58. Emambokus N, Vegiopoulos A, Hamman B, Jenkinson N, Anderson G, Frampton J (2003) Progression through key stages of haemopoiesis is dependent on distinct threshold levels of c-Myb. EMBO J 22: 4478–4488 59. Endoh M, Ogawa M, Orkin S, Nishikawa SI (2002) SCL/tal-1-dependent process determines a competence to select the deﬁnitive hematopoietic lineage prior to endothelial differentiation. EMBO J 21: 6700–6708 60. Epstein AC, Gleadle JM, McNeill MI, Hewitson KS, O’Rourke J, Mole DR, Mukherji M, Metzen E, Wilson MI, Dhanda A, Tian YM, Masson N, Hamilton DL, Jaakkola P, Barstead R, Hodgkin J, Maxwell PH, Pugh CW, Schonﬁeld CJ, Ratcliffe PJ (2001) C. elegans EGL-9 and mammalian homologs deﬁne a family of dioxygenases that regulate HIF by prolyl hydroxylation. Cell 107: 43–54 61. Erslev AJ (1991) Erythopoietin. N Engl J Med 324: 1339–1349 62. Farrell F, Lee A (2004) The erythropoietin receptor and its expression in tumor cells and other tissues. Oncologist 9[Suppl 5]: 18–30 63. Felli N, Pedini F, Zeuner A, Petrucci E, Testa U, Conticello C, Biffoni M, Di Cataldo A, Winkles JA, Peschle C, De Maria R (2005) Multiple memberes of the TNF superfamily contribute to IFN-γ mediated inhibition of erythropoiesis. J Immunol 175: 1464–1472 64. Ferreira R, Ohneda K, Yamamoto M, Philipsen S (2005) GATA1 function, a paradigm for transcription factors in hematopoiesis. Mol Cell Biol 25: 1215–1227 65. Fisher JW, Koury S, Ducey T, Mendel S (1996) Erythropoietin (EPO) production by interstitial cells of hypoxic monkey kidneys. Br J Haematol 95: 27–32 66. Fujiwara Y, Chang AN, Williams AM, Orkin SH (2004) Functional overlap of GATA-1 and GATA-2 in primitive hematopoietic development. Blood 103: 583–585 67. Fukuda S, Foster RG, Pelus LM (2002) The antiapoptosis protein surviving is associated with cell cycle entry of normal cord blood CD34+ cells and modulates cell cycle and proliferation of mouse hematopoietic progenitor cells. Blood 100: 2463–2471

Physiology of erythropoiesis

55

68. Gabbianelli M, Testa U, Massa A, Pelosi E, Sposi NM, Riccioni R, Luchetti L, Peschle C (2000) Hemoglobin switching in unicellular culture of sibling burstforming units: kit ligand induces a dose-dependent fetal hemoglobin reactivation potentiated by sodium butyrate. Blood 95: 3555–3561 69. Gabbianelli M, Testa U, Massa A, Petrucci E, Morsilli O, Peschle C (2003) HbF reactivation in sibling BFU-E colonies: synergistic interaction of kit ligand with dexamethasone at therapeutic level. Blood 101: 2826–2832 70. Galloway JL, Wingert RA, Thisse C, Thisse B, Zon LI (2005) Loss of GATA1 but not GATA2 converts erythropoiesis to myelopoiesis in zebraﬁsh embryos. Dev Cell 8: 109–116 71. Garçon L, Lacout C, Svinartchouk F, LeCoudedic JP, Villeval JL, Vainchenker W, Duménil D (2005) Gﬁ-1B plays a critical role in terminal differentiation of normal and transformed erythroid progenitor cells. Blood 105: 1448–1455 72. Gnarra JR, Zhou S, Merrill MJ, Wagner J, Krumm A, Papavassiliou E, Oldﬁeld EH, Klausner RD, Linehan WM (1996) Post-transcriptional regulation of vascular endothelial growth factor mRNA by the product of the VHL tumor suppressor gene. Proc Natl Acad Sci USA 93: 10589–10594 73. Golks A, Brenner D, Fritsch C, Krammer PH, Lavrik IN (2005) c-FLIPR – a new regulator of death receptor-induced apoptosis. J Biol Chem 280: 14507– 14513 74. Gordeuk VR, Stockton DW, Prchal JT (2005) Congenital polycythemias. Haematologica 90: 109–116 75. Green AR, DeLuca E, Begley CG (1991) Antisense SCL suppresses selfrenewal and enhances spontaneous erythroid differentiation of the human leukemic cell line K562. EMBO J 10: 4153–4158 76. Gregoli PA, Bondurant MC (1997) The roles of Bcl-XL and apopain in the control of erythropoiesis by erythropoietin. Blood 90: 630–640 77. Gregoli PA, Bondurant MC (1999) Function of caspases in regulating apoptosis caused by erythropoietin deprivation in erythroid progenitors. J Cell Physiol 178: 133–143 78. Gregory T, Yu C, Ma A, Orkin SH, Blobel GA, Weiss MJ (1999) GATA-1 and erythropoietin cooperate to promote erythroid cell survival by regulating bcl-XL expression. Blood 94: 87–96 79. Grillot D, Gonzalez-Garcia M, Ekhteral D, Duan L, Inohara N, Ohta S, Seldin MF, Nunez G (1997) Genomic organization, promoter region analysis, and chromosome localization of the mouse bcl-x gene. J Immunol 158: 4750– 4757 80. Gutierrez L, Lindeboom F, Langeveld A, Grosveld F, Philipsen S, Whyatt D (2004) Homotypic signaling regulates Gata1 activity in the erythroblastic island. Development 131:3183–3193 81. Haﬁd-Medhab KH, Augery-Bourget Y, Minatchy MN, Hanania N, RobertLézénés J (2003) Bcl-XL is required for heme synthesis during the chemical induction of erythroid differentiation of murine erythroleukemia cells independently of its anti-apoptotic function. Blood 101: 2575–2583 82. Haseyama Y, Sawada KI, Oda A, Koizumi K, Takano H, Tarumi T, Nishio M, Handa M, Ikeda Y, Koika T (1999) Phosphatidylinositol 3-kinase is involved in the protection of primary cultured human erythroid precursor cells from apoptosis. Blood 94: 1568–1577

56

U. Testa

83. Haughn L, Hawley G, Morrison DC, Von Boehmer H, Hockenbery DM (2003) BCL-2 and BCL-XL restrict lineage choice during hemopoietic differentiation. J Biol Chem 278: 25158–25165 84. Hetakugas V, Poukkula M, Hetskanen KM, Kaninen JT, Sivtonen L, Eriksson JE (2003) Erythroid differentiation sensitizes K562 leukemia cells to TRAILinduced apoptosis by downregulation of c-FLIP. Mol Cell Biol 23: 1278–1291 85. Huang DY, Kuo YV, Lai JS, Suzuki Y, Sugano S, Chang ZF (2004) GATA-1 and NF-Y cooperate to mediate erythroid-speciﬁc transcription of Gﬁ-1B gene. Nucleic Acids Res 32: 3935–3946 86. Huang LE, Bunn HF (2003) Hypoxia-inducible factor and its biomedical relevance. J Biol Chem 278: 19575–19578 87. Huang S, Qiu Y, Shi Y (2000) P/CAF-mediated acetylation regulates the function of the basic helix-loop-helix transcription factor TAL1/SCL. EMBO J 19: 6792–6813 88. Huddleston H, Tan B, Yang FC, White H, Wenning MJ, Orazi A, Yoder MC, Kapur R, Ingram DA (2003) Functional p85α gene is required for normal murine fetal erythropoiesis. 102: 142–145 89. Hung HL, Lou J, Kim AY, Weiss MJ, Blobel GA (1999) CREB binding protein acetylates hematopoietic transcription factor GATA-1 at functionally important sites. Mol Cell Biol 19: 3496–3505 90. Iavarone A, King ER, Dai XM, Leone G, Stainley ER, Lasorella A (2004) Retinoblastoma promotes deﬁnitive erythropoiesis by repressing Id2 in fetal liver macrophages. Nature 432: 1040–1045 91. Imagawa S, Izumi T, Miura Y (1994) Positive and negative regulation of the erythropoietin gene. J Biol Chem 269: 9038–9044 92. Imagawa S, Yamamoto M, Miura Y (1997) Negative regulation of the erythropoietin gene expression by the GATA transcription factors. Blood 89: 1430–1439 93. Imagawa, Nakano Y, Ohara N (2003) GATA-speciﬁc inhibitor (K-7174) rescues anemia induced by IL-1beta, TNF-alpha, or L-NMMA. FASEB J 17:1742–1744 94. Ivan M, Kondo K, Yang H, Kim W, Valiando J, Ohh M, Salic A, Asera JM, Lane WS, Kaelin WG (2001) HIFalpha targeted for VHL-mwdiated destruction by proline hydroxylation: implications for O2 sensing. Science 292: 464–468 95. Iwasaki H, Mizuno S, Wells RA, Cantor AB, Watanabe S, Akashi K (2003) GATA-1 converts lymphoid and myelomonocytic progenitors into the megakaryocyte/erythrocyte lineages. Immunity 19: 451–462 96. Jaakkola P, Mole DR, Tian YM, Wilson MI, Gielbert J, Gaskell SJ, Kriegshein AV, Hebestreit HF, Mukherji M, Schoﬁeld CJ, Maxwell PH, Pugh CW, Ratcliffe PJ (2001) Targeting of HIF-alpha to the von Hippel-Lindau ubiquitylation complex by O2-regulated prolyl hydroxylation. Science 292: 468–472 97. Jacks T, Fazeli A, Schmitt E, Bronson RT, Goodell MA, Weinberg RA (1992) Effects of an Rb mutation in the mouse. Nature 359: 295–300 98. Jacobs K, Shoemaker C, Rudersdorf R, Neill SD, Kaufman RJ, Mufson A, Seehra A, Jones SS, Hewick R, Kawakita M, Shimiza T, Miyoke T (1985) Isolation and characterization of genomic cDNA clones of human erythropoietin. Nature 313: 806–810 99. Jacobs-Helber S, Wickema A, Biner MJ, Sawyer ST (1998) AP1 regulation of proliferation and initiation of apoptosis in erythropoietin-dependent erythroid cells. Mol Cell Biol 18: 3699–3707

Physiology of erythropoiesis

57

100. Jacobs-Helber SM, Ryan JJ, Sawyer ST (2000) JNK and p38 are activated by erythropoietin (EPO) but are not induced in apoptosis following EPO withdrawal in EPO-dependent HCD57 cells. Blood 96: 933–940 101. Jacobs-Helber SM, Abutin RM, Tian C, Bondourant M, Wicrema A, Sawyer ST (2002) Role of JunB in erythroid differentiation. J Biol Chem 277: 4859– 4866 102. Jacobs-Helber SM, Roh K, Bailey D, Dessypris EN, Ryan JJ, Chen J, Wickrema A, Barber DL, Dent P, Sawyer ST (2003) Tumor necrosis factor-alpha expressed constitutively in erythroid cells or induced by erythropoietin has negative and stimulatory roles in normal erythropoiesis and erythroleukemia. Blood 101: 524–531 103. Johnstone RW, Rueﬂi AA, Lowe JW (2002) Apoptosis: a link between cancer genetics and chemotherapy. Cell 108: 153–164 104. Josefsen D, Myklebust JH, Lomo J, Sioud M, Blomhoff HK, Smeland EB (2000) Differential expression of Bcl-2 homologs in human CD34+ hematopoietic progenitor cells induced to differentiate into erythroid or granulocytic cells. Stem Cells 18: 261–272 105. Joza N, Kroemer G, Penninger JM (2002) Genetic analysis of the mammalian cell death machinery. Trends Genet 18: 142–149 106. Kaluzhny Y, Yu G, Sun S, Nesswandt B, Jackson CW, Ravid K (2002) Bcl-XL overexpression in megakaryocytes leads to impaired platelet fragmentation. Blood 100: 1670–1678 107. Kang TB, Moshe TB, Varfolomeev EE, Pewzner-Jung Y, Yogev N, Jureciwz A, Waisman A, Brenner O, Haffner R, Gustafsson E, Ramakrishnan P, Lapidot T, Wallach D (2004) Caspase-8 serves apoptotic and nonapoptotic roles. J Immunol 173: 2976–2984 108. Kashii Y, Uchida M, Kirito K, Tanaka M, Nishijima K, Toshima M, Ando T, Koizumi K, Endoh T, Sawada Ki, Momoi M, Miura Y, Ozawa K, Komatsu N (2000) A member of Forkhead family transcription factor, FKHRL1, is one of the downstream molecules of phosphatidylinositol 3-kinase-Akt activation pathway in erythropoietin signal transduction. Blood 96: 941–949 109. Kaufmann JH, Hengartner MD (2001) Programmed cell death: alive and well in the new millennium. Trends Cell Biol 11: 526–534 110. Kawane K, Fukuyama H, Kondoh G, Takeda J, Ohsawa Y, Uchiyama Y, Nagata S (2001) Requirement of Dnase II for deﬁnitive erythropoiesis in the mouse fetal liver. Science 292: 1546–1549 111. Ketteler R, Maghraby CS, Hsiao JG, Sandra O, Lodish HF, Klingmuller U (2003) The cytokine-inducible SH2 domain containing protein CIS negatively regulates signalling by promoting apoptosis in erythroid progenitor cells. J Biol Chem 278: 2654–2660 112. Kieslinger M, Woldman I, Moriggi R, Hofmann J, Marine JC, Ihle JN, Beug H, Decker T (2000) Antiapoptotic activity of Stat5 required during terminal stages of myeloid differentiation. Genes Dev 14: 232–244 113. Kim H, Whartenby KA, Georgantas III RW, Wingard J, Civin CI (2002) Human CD34+ hematopoietic stem/progenitor cells express high levels of FLIP and are resistant to Fas-mediated apoptosis. Stem Cells 20: 174–182 114. Kischkel FC, Lawrence M, LeBlanc H, Virmani A, Schaw P, Gordar A, Blenis J, Arnott D, Askenazi A (2001) Death receptor recruitment of endogenous

58

115.

116.

117.

118. 119.

120. 121.

122.

123.

124.

125.

126. 127.

128.

129.

U. Testa caspase-10 and apoptosis initiation in the absence of caspase-8. J Biol Chem 276: 46639–46646 Klingmuller U, Lorenz U, Cantley LC, Neel BG, Lodish HF (1995) Speciﬁc recruitment of SH-PTP1 to the erythropoietin receptor causes inactivation of JAK2 and termination of proliferative signals. Cell 80: 729–738 Kolbus A, Blazquez-Domingo M, Carotta S, Bakker W, Luedemann S, von Lindern M, Steinlein P, Beug H (2003) Cooperative signaling between cytokine receptors and the glucocorticoid receptor in the expansion of erythroid progenitors: molecular analysis by expression proﬁling. Blood 102: 3136–3146 Kolbus A, Pilat S, Husak Z, Deiner EM, Stengl G, Beug H, Baccarini M (2003) Raf-1 antagonizes erythroid differentiation by restraining caspase activation. J Exp Med 196: 1347–1353 Koury ST, Bondurant MS, Koury MJ (1988) Localization of erythropoietin synthesizing cells in murine kidneys. Br J Haematol 95: 27–32 Koury ST, Bondurant MC, Koury MJ, Semenza GL (1991) Localization of cells producing erythropoietin in murine liver by in situ hybridization. Blood 77: 2497–2503 Krammer PH (2000) CD95’s deadly mission in the immune system. Nature 407: 789–795 Krauss SW, Lo AJ, Short SA, Koury MJ, Mohandas N, Chasis JA (2005) Nuclear substructure reorganization during late stage erythropoiesis is selective and does not involve caspase cleavage of major nuclear substructural proteins. Blood 106: 2200–2205 Krieser RJ, MacLea KS, Longnecker DS, Fields JL, Fiering S, Eastman A (2002) Deoxyribonuclease II alpha is required during the phagocytic phase of apoptosis and its loss causes perinatal lethality. Cell Death Differ 9: 956–962 Kuwana T, Mackey MR, Perkins G, Ellisman MH, Latterich M, Schneiter R, Green DR, Newmeyer DD (2002) Bid, bax and lipids cooperate to form supramolecular openings in the outer mitochondrial membrane. Cell 111: 331–342 Labbaye C, Valtieri M, Barberi T, Meccia E, Masella B, Pelosi E, Condorelli GL, Testa U, Peschle C (1995) Differential expression and functional role of GATA2, NF-E2, and GATA-1 in normal adult hematopoiesis. J Clin Invest 95: 2346–2358 La Ferla K, Reinman C, Jelkmann W, Hellwig-Burgel T (2002) Inhibition of erythropoietin gene expression signaling involves the transcription factors GATA2 and NF-kB. FASEB J 16: 1811–1813 Lahil R, Lécuyer E, Herblot S, Hoang T (2004) SCL assembles a multifactorial complex that determines glycophorin A expression. Mol Cell Biol 24: 1439–1452 Lando D, Peet DJ, Gorman JJ, Whelan DA, White MF, Bruick RK (2002) FIH1 is an asparaginyl hydroxylase enzyme that regulates the transcriptional activity of hypoxia inducible factor. Genes Dev 16: 1466–1471 Lee G, Spring FA, Parsins SF, Mankelow TJ, Peters LL, Koury MJ, Mohandas N, Anstee DJ, Chasis JA (2003) Novel secreted isoform of adhesion molecule ICAM-4: a potential regulator of membrane-associated ICAM-4 interactions. Blood 101: 1790–1797 Lee JW, Gersuk GM, Kiener PA, Beckam C, Ledbetter JA, Deeg HJ (1997) HLA-DR-triggered inhibition of hemopoiesis involves Fas/Fas Ligand interactions and is prevented by c-kit ligand. J Immunol 59: 3211–3219

Physiology of erythropoiesis

59

130. Lee PJ, Jiang BH, Chiu BY (1997) Hypoxia-inducible factor-1 mediates transcriptional activation of the heme oxygenase-1 gene in response to hypoxia. J Biol Chem 275: 5375–5381 131. Lee YS, Vortmeyer AO, Lubensky IA, Zhuang Z (2005) Coexpression of erythropoietin and erythropoietin receptor in Von Hippel-Lindau diseaseassociated renal cysts and renal cell carcinoma. Clin Cancer Res 11: 1059–1064 132. Levin J, Peng JP, Baker GR, Villeval JL, Lecine P, Burstein SA, Shivdasani RA (1999) Pathophysiology of thrombocytopenia and anemia in mice lacking transcription factor NF-E2. Blood 94: 3037–3047 133. Levy DE, Darnell JE (2002) Stats: transcriptional control and biological impact. Nat Rev Mol Cell Biol 3: 651–662 134. Liew CK, Simpson RJ, Kwan AH, Crofts LA, Loughlin FE, Matthews JM, Crossley M, Mackay JP (2005) Zinc ﬁngers as protein recognition motifs: structural basis for the GATA-1/Friend of GATA interaction. Proc Natl Acad Sci8 USA 102: 583–588 135. Lin FK, Suggs S, Lin CH, Bowne JK, Smailing R, Egric JC, Chen KK, Fox GM, Martin F, Wasser Z (1985) Cloning and expression of the human erythropoietin gene. Proc Natl Acad Sci USA 92: 7880–7884 136. Liu JJ, Hou SC, Shen CK (2003) Erythroid gene suppression by NF-κB. J Biol Chem 278: 19534–19540 137. Lok CN, Ponka P (1999) Identiﬁcation of a hypoxia response element in the transferrin receptor gene. J Biol Chem 274: 24147–24152 138. Macdougall IC, Cooper AC (2002) Erythropoietin resistance: the role of inﬂammation and pro-inﬂammatory cytokines. Nephrol Dial Transplant 17 [Suppl]: 39–43 139. Maciejewski J, Selleri C, Anderson S, Young NS (1995) Fas antigen expression on CD34+ human marrow cells is induced by interferon-γ and tumor necrosis factor-α and potentiates cytokine-mediated hematopoietic suppression in vitro. Blood 85: 3183–3190 140. Mandal D, Baudin-Creuza V, Bhattacharyva A, Pathak S, Delaunay J, Kondu M, Basu J (2003) Caspase 3-mediated proteolysis of the N-terminal cytoplasmic domain of the human erythroid anion exchanger 1 (band3) J Biol Chem 22: 4648–4655 141. Marsden VS, O’Connor L, O’Reilly LA, Silke J, Metcalf D, Ekert PG, Huang DC, Cecconi F, Kuida K, Tomaselli KJ, Roy S, Nicholson DW, Vaux DL, Bouillet P, Adams JM, Strasser A (2002) Apoptosis initiated by Bcl-2-regulated caspase activation independently of the cytochrome c/Apaf-1/caspase-9 apoptosome. Nature 419: 634–637 142. Martin F, Linden T, Katschinski DM, Oehme F, Flamme I, Mukhopadhyay CK, Eckhardt K, Troger J, Barth S, Camenisch G, Wenger RH (2005) Copperdependent activation of hypoxia-inducible factor (HIF)-1: implications for ceruloplasmin regulation. Blood (in press) 143. Martowicz ML, Grass JA, Boyer ME, Guend H, Bresnick EH (2005) Dynamic GATA factor interplay at a multicomponent regulatory region of the GATA-2 locus. J Biol Chem 280: 1724–1732 144. Means RT, Dessypris EN, Krantz SB (1990) Inhibition of human colonyforming-unit erythroid by tumor necrosis factor requires accessory cells. J Clin Invest 86: 538–541

60

U. Testa

145. Michalak E, Villunger A, Erlacher M, Strasser A (2005) Death squads enlisted by the tumor suppressor p53. Biochem Biophys Res Commun 331: 786–798 146. Michiels C, Minet E, Mottet D, Raes M (2002) Regulation of gene expression by oxygen: NF-kB and HIF, two extremes. Free Rad Biol Med 33: 1331–1342 147. Mikkola H, Klintman J, Hock H, Fujiwara Y, Orkin SH (2003) Hematopoietic stem cells retain long-term repopulating activity and multipotency in the absence of stem cell leukemia SCL/tal-1 gene. Nature 421: 547–551 148. Miyake T, Kung CKH, Goldwasser E (1985) Puriﬁcation of human erythropoietin. J Biol Chem 252: 5558–5564 149. Moldawer LL, Marano MA, Wei H, Fong Y, Silen ML, Kuo G, Manogue KR, Vlassara H, Cohen H, Cerami A (1989) Cachectin/tumor necrosis factor-alpha alters red blood cell kinetics and induces anemia in vivo. Faseb J 3: 1637–1643 150. Mori M, Uchida M, Watanabe T, Kirito K, Hatake K, Ozawa K, Komatsu N (2003) Activation of extracellular signal-regulated kinases ERK1 and ERK2 induces Bcl-XL upregulation via inhibition of caspase activities in erythropoietin signalling. J Cell Physiol 195: 290–297 151. Motoyama N, Wang F, Roth KA, Sawa H, Nakayama K, Negishi I, Senju S, Zhong Q, Fuji S (1995) Massive cell death of immature hemopoietic cells and neurons in Bcl-X-deﬁcient mice. Science 267: 1506–1510 152. Motoyama N, Kimma T, Takahashi T, Watanabe T, Nakano T (1999) Bcl-X prevents apoptotic cell death of both primitive and deﬁnitive erythrocytes at the end of maturation. J Exp Med 189: 1691–1698 153. Muppidi JR, Tschopp J, Siegel RM (2004) Life and death decisions: secondary complexes and lipid rafts in TNF receptor family signal transduction. Immunity 21: 461–465 154. Myklebust JH, Blonhoff HK, Rusten LS, Stokke T, Smeland EB (2002) Activation of phosphatidylinositol 3-kinase is important for erythropoietin-induced erythropoiesis from CD34+ hematopoietic progenitor cells. Exp Hematol 30: 990–1000 155. Nagafuji K, Shibuya T, Harada M, Mizuno S, Takenaka K, Miyamoto T, Okamura T, Niho Y (1995) In vitro expansion of Fas antigen (CD95) on hematopoietic progenitor cells. Blood 86: 883–894 156. Nakano Y, Imagawa S, Matsumoto K, Stockmann C, Obara N, Suzuki N, Doi T, Kodama T, Takahashi S, Nagasawa T, Yamamoto M (2004) Oral administration of K-11706 inhibits GATA binding activity, enhances hypoxia-inducible factor1 binding activity, and restores indicators in an in vivo mouse model of anemia of chronic disease. Blood 104: 4300–4307 157. Negoro S, Oh H, Tone E, Kusinada K, Fujio Y, Walsh K, Kishimoto T, YamauchiTakihara K (2001) Glycoprotein 130 regulates cardiac myocyte survival in doxorubicin-induced apoptosis through phosphatidylinositol 3 kinase/AKT phosphorylation and Bcl-XL/caspase-3 interaction. Circulation 103: 555–561 158. Neubauer H, Cumano A, Muller M, Wu H, Huffstadt U, Pfeffer K (1998) JAK2 deﬁciency deﬁnes an essential development checkpoint in deﬁnitive hematopoiesis. Cell 93: 397–409 159. Nichols KE, Crispino JD, Ponez M, White JG, Orkin SH, Maris JM, Weiss MJ (2000) Familial dyserythropoietic anaemia and thrombocytopenia due to an inherited mutation in GATA-1. Nat Genet 24: 266–270

Physiology of erythropoiesis

61

160. Nicholson KM, Anderson NG (2002) The protein kinase B/AKT signalling pathway in human malignancy. Cell Signal 14: 381–395 161. Nishio M, Oda A, Koizuni K, Sato I, Endho T, Tsutsumi A, Fujihara M, Ikebuchi K, Ikeda H, Koike T, Sawada K (2001) Stem cell factor prevents Fasmediated apoptosis of human erythroid precursor cells with Src-family kinase dependency. Exp Hematol 29: 19–29 162. Nosaka T, Kawashima T, Misawa K, Ikuta K, Mui AL, Kitamura T (1999) Stat5 as a molecular regulator of proliferation, differentiation and apoptosis in hematopoietic cells. EMBO J 18: 4754–4765 163. Opferman JT, Iwasaki H, Ong CC, Suh H, Mizuno SI, Akahashi K, Korsmeyer SJ (2005) Obligate role of anti-apoptotic MCL-1 in the survival of hematopoietic stem cells. Science 307: 1101–1104 164. Osawa M, Yamaguchi T, Nakamura Y, Kaneko S, Onodera M, Sawada K, Jegalian A, Nakauchi H, Iwana A (2002) Erythroid expansion mediated by the Gﬁ-1B zinc ﬁnger protein: role in normal hematopoiesis. Blood 100: 2769– 2777 165. Otsuki T, Nagakura S, Wang J, Bloom M, Grompe M, Liu JM (1999) Tumor necrosis factor-α and CD95 ligation suppress erythropoiesis in Fanconi anemia C gene knockout mice. J Cell Physiol 179: 79–86 166. Pan X, Ohneda O, Ohneda K, Lindeboom F, Iwata F, Shimizu R, Nagano M, Suwabe N, Philipsen N, Lim KC, Engel JD, Yamamoto M (2005) Graded levels of GATA-1 expression modulate survival, proliferation, and differentiation of erythroid progenitors. J Biol Chem 280: 22385–22394 167. Parganas E, Wang D, Stravopodis D, Topham DJ, Marine JC, Teglund S, Varrin EF, Bodner S, Colamonici OR, Van Deursen JM, Grosveld G, Ihle JN (1998) Jak2 is essential for signalling through a variety of cytokine receptors. Cell 93: 385–395 168. Park JR, Bernstein ID, Hockenbery DM (1995) Primitive human hematopoietic precursors express Bcl-X but not Bcl-2. Blood 86: 868–876 169. Peller S, Frenkel J, Lapidot T, Kahn J, Rahimi-Levene N, Youa R, Nissim L, Goldﬁnger N, Sherman DJ, Rotter V (2003) The onset of p53-dependent apoptosis plays a role in terminal differentiation of human normoblasts. Oncogene 22: 4648–4655 170. Peters R, Leyvraz S, Perey L (1998) Apoptotic regulation in primitive hematopoietic precursors. Blood 92: 2041–2052 171. Porcher C, Liao EC, Fujiwara Y, Zon LI, Orkin SH (1999) Speciﬁcation of hematopoietic and vascular development by the bHLH transcription factor SCL without direct DNA binding. Development 126: 4603–4615 172. Ravet E, Reynaud D, Titeux M, Izac B, Fichelson S, Roméo PH, DubartKupperschmitt A, Pﬂumio F (2004) Characterization of DNA-bindingdependent and -independent functions of SCL/TAL1 during human erythropoiesis. Blood 103: 3326–3335 173. Rekhtman N, Choe KS, Matushansky I, Murray S, Stopka T, Skoultchi AI (2003) PU.1 and pRB interact and cooperate to repress GATA-1 and block erythroid differentiation. Mol Cell Biol 23: 7460–7474 174. Remi I, Wilson IA, Michnick JW (1999) Erythropoietin receptor activation by a ligand-induced conformation change. Science 283: 990–993

62

U. Testa

175. Rhodes MM, Kopsombut P, Bondurant MC, Price JO, Koury MJ (2005) Bcl-XL prevents apoptosis of late-stage erythroblasts but does not mediate the antiapoptotic effect of erythropoietin. Blood 106: 1857–1863 176. Rodriguez P, Bonte E, Krjigsveld J, Kolodziej KE, Guyot B, Heck A, Vyas P, de Boer E, Grosveld F, Strouboulis J (2005) GATA-1 forms distinct activating and repressive complexes in erythroid cells. EMBO J 24: 2354–2366 177. Rolfs A, Kvietikova I, Gassman M, Wenger RH (1997) Oxygen-regulated transferrin expression is mediated by hypoxia-inducible factor-1. J Biol Chem 272: 20055–20062 178. Rusten LS, Jacobsen SE (1995) Tumor necrosis factor (TNF)-α directly inhibits human erythropoiesis in vitro: role of p55 and p75 TNF receptors. Blood 85:989–997 179. Rylski M, Welch JJ, Chen YY, Letting DL, Dichl JA, Chodosh LA, Blobel GA, Weiss MJ (2003) GATA-1 mediated proliferation arrest during erythroid maturation. Mol Cell Biol 23: 5031–5042 180. Saleque S, Cameron S, Orkin SH (2002) The zinc-ﬁnger proto-oncogene Gﬁ-1B is essential for development of the erythroid and megakaryocytic lineages. Genes Dev 16: 301–306 181. Sato T, Maekawa S, Watanabe S, Tsuji K, Nakashata T (2000) Erythroid progenitors differentiate and mature in response to endogenous erythropoietin. J Clin Invest 2: 263–270 182. Schmerer M, Evans T (2003) Primitive erythropoiesis by Smad-dependent signalling in post-gastrulation mesoderm. Blood 102: 3196–3205 183. Schneider E, Mareau G, Arnould A, Vasseur F, Khodabaccus N, Dy M, Ezine S (1999) Increased fetal and extramedullary hematopoiesis in Fas-deﬁcient C57BL/6-lpr/lpr mice. Blood 94: 2613–2621 184. Scortegnagna M, Ding K, Zhang Q, Oktay Y, Bemet MJ, Bemet M, Shelton JM, Richardson JA, Moe O, Garcia JA (2005) HIF-2alpha regulates murine hematopoietic development in an erythropoietin-dependent manner. Blood (in press) 185. Secchiero P, Melloni E, Heikinheimo M, Mannito S, Di Pietro R, Iacone A, Zauli G (2004) TRAIL regulates normal erythroid maturation through an ERKdependent pathway. Blood 103: 517–522 186. Sharp DA, Lawrence DA, Ashkenazi A (2005) Selective knockdown of the long variant of cellular FLICE inhibitory protein augments death receptor-mediated caspase-8 activation and apoptosis. J Biol Chem 280: 19401–19409 187. Shimizu R, Ohneda K, Engel JD, Trainor CD, Yamamoto M (2004) Transgenic rescue of GATA-1-deﬁcient mice with GATA-1 lacking a FOG-1 association site phenocopies patients with X-linked thrombocytopenia. Blood 103: 2560– 2567 188. Silva M, Benito A, Sanz C, Prosper F, Ekhteral D, Nunez G, Fernandez-Luna JL (1999) Erythropoietin can induce the expression of Bcl-XL through Stat5 in erythropoietin-dependent progenitor cell lines. J Biol Chem 274: 22165– 22169 189. Silvestris F, Cafforio P, Tucci M, Dammacco F (2002) Negative regulation of erythroblast maturation by Fas-L+/TRAIL+ highly malignant plasma cells: a major pathogenetic mechanism of anemia in multiple myeloma. Blood 99: 1305–1313

Physiology of erythropoiesis

63

190. Silvestris F, Tucci M, Cafforio P, Dammacco F (2002) Fas-L up-regulation by highly malignant myeloma plasma cells: role in the pathogenesis of anemia and disease progression. Blood 97: 1155–1164 191. Socolovsky M, Fallon AEJ, Wang S, Brugnara C, Lodish HF (1999) Fetal anemia and apoptosis of red cell progenitors in Stat5a−/−5b−/− mice: a direct role for Stat5 in Bcl-XL induction. Cell 98: 181–191 192. Socolovsky M, Nam HS, Fleming MD, Haasse VH, Brugnara C, Lodish HF (2001) Ineffective erythropoiesis in Stat5a−/−5b−/− mice due to decreased survival of early erythroblasts. Blood 98: 3261–3273 193. Somervaille T, Linch D, Khwaja A (2001) Growth factor withdrawal from primary human erythroid progenitors induces apoptosis through a pathway involving glycogen synthase kinase-3 and Bax. Blood 98: 1374–1381 194. Sordet O, Rébé C, Plenchette S, Zermati Y, Hermine O, Vainchenker W, Garrido C, Solary E, Dubrez-Daloz L (2002) Speciﬁc involvement of caspases in the differentiation of monocytes into macrophages. Blood 100: 4446–4453 195. Suwabe N, Takahashi S, Nakano T, Yamamoto M (1998) GATA-1 regulatesd growth and differentiation of deﬁnitive erythroid lineage cells during in vitro ES cell differentiation. Blood 92: 4108–4118 196. Spike BT, Dirlam A, Dibling BC, Marvin J, Williams BO, Jacks T, Macleod KF (2004) The Rb tumor suppressor is required for stress erythropoiesis. EMBO J 23: 4319–4329 197. Sposi NM, Zon LI, Caré A, Valtieri M, Testa U, Gabbianelli M, Mariani G, Bottero L, Mather C, Orkin SH, Peschle C (1992) Cycle-dependent initiation and lineage-dependent abrogation of GATA-1 expression in pure differentiating hematopoietic progenitors. Proc Natl Acad Sci USA 89: 6353–6357 198. Sui X (1996) Erythropoietin-independent erythrocyte production: signals through gp130 and c-kit dramatically promote erythropoiesis from human CD34+ cells. J Exp Med 183: 837–845 199. Suzuchi N, Suwabe O, Obara N, Imagawa S, Pan X, Motohashi H, Yamamoto M (2003) Identiﬁcation and characterization of two types of erythroid progenitors that express GATA-1 at distinct levels. Blood 102: 3573–3583 200. Tacchini L, Bianchi L, Bernelli-Zazzera A, Cairo G (1999) Transferrin receptor induction by hypoxia. HIF-1-mediated transcriptional activation and cellspeciﬁc post-transcriptional regulation. J Biol Chem 272: 24142–24146 201. Takahashi S, Onodera K, Motohashi H, Suwabe N, Hayashi N, Yanai N, Nabesima Y, Yamamoto M (1997) Arrest in primitive erythroid cell development caused in promotor-speciﬁc disruption of the GATA-1 gene. J Biol Chem 272: 12611–12615 202. Takenaka K, Nagafuji K, Harada M, Mizuno S, Miyamoto T, Makino S, Gondo H, Okamura T, Niho Y (1996) In vitro expansion of hematopoietic progenitor cells induces functional expression of Fas antigen (CD95). Blood 88: 2871–2879 203. Tanaka M, Zheng J, Kitajima K, Kita K, Yoshikawa H, Nakano T (2004) Differentiation status dependent function of FOG-1. Genes to Cells 9: 1213–1226 204. Tarumoto T, Imagawa S, Ohmine K (2000) N(G)-monomethyl-L-arginine inhibits erythropoietin gene expression by stimulating GATA-2. Blood 96: 1716–1722 205. Testa U (2004) Apoptotic mechanisms in the control of erythropoiesis. Leukemia 18: 1176–1199

64

U. Testa

206. Tian C, Gregoli P, Bondurant M (2003) Bcl-x promoter function in erythroid progenitor cells. Blood 101: 2235–2242 207. Tibbetts MD, Zheng L, Lenardo MJ (2003) The effector death domain protein family: regulators of cellular homeostasis. Nat Immunol 4: 404–409 208. Tsai FY, Keller G, Kuo FC, Weiss M, Chen J, Rosenblatt M, Alt FW, Orkin SH (1994) An early hemopoietic defect in mice lacking the transcription factor GATA-2. Nature 371: 221–226 209. Tsang AP, Fujiwara Y, Ham DB, Orkin SB (1998) Failure of megakaryopoiesis and arrested erythropoiesis in mice lacking the GATA-1 transcriptional cofactor FOG. Genes Dev 12: 1176–1188 210. Ueki N, Zhang L, Hayman MJ (2004) Ski negatively regulates erythroid differentiation through its interaction with GATA1. Mol Cell Biol 24: 10118– 10125 211. Ulich TR, Del Castillo J, Yin S (1990) Tumor necrosis factor exerts dosedependent effects on erythropoiesis and myelopoiesis in vivo. Exp Hematol 18: 311–315 212. Valtieri M, Tocci A, Gabbianelli M, Lucchetti L, Masella B, Vitelli L, Botta R, Testa U, Condorelli GL, Peschle C (1998) Enforced Tal-1 expression stimulates primitive erythroid and megakaryocytic progenitors but blocks the granulopoietic differentiation program. Cancer Res 58: 562–569 213. Vander Heiden MG, Thompson CB (2002) Bcl-2 proteins: regulators of apoptosis or of mitochondrial homeostasis? Nat Cell Biol 1: 209–216 214. Varfolomeev EE, Schuchmann M, Wallack D (1998) Targeted disruption of the mouse caspase 8 gene ablates cell death induction by the TNF receptors, Fas/Apo1, and DR3 and is lethal prenatally. Immunity 9: 267–276 215. Varfolomeev EE, Ashkenazi A (2004) Tumor necrosis factor: an apoptosis JuNKie? Cell 116: 491–497 216. Vivanco I, Sawyers CL (2002) The phosphatidyl-inositol 3-kinase-AKT pathway in human cancer. Nat Rev Cancer 2: 489–501 217. Von Lindern M, Amelsvoort P, Van Dijk T, Deiner E, Akker E, De Vries S, Willems P, Beug H, Lowenberg B (2000) Protein kinase Cα controls erythropoietin receptor signalling. J Biol Chem 275: 34719–34727 218. Von Lindern M, Schmidt U, Beug H (2004) Control of erythropoiesis by erythropoietin and stem cell factor: a novel role for Burton’s tyrosine kinase. Cell Cycle 3: 876–879 219. Vortmeyer AO, Frank S, Jeong S (2003) Developmental arrest of angioblastic lineage initiates tumorigenesis in von Hippel-Lindau disease. Cancer Res 63: 7051–7055 220. Wagner KU, Claudio E, Rucker III EB, Riedinger G, Broussard C, Schwartzberg PL, Siebenlist U, Hennighausen L (2000) Conditional deletion of the Bcl-XL gene from erythroid cells results in haemolytic anemia and profound splenomegaly. Development 127: 4949–4958 221. Wang CQ, Uduba KB, Lipschitz DA (1995) Interferon-gamma exerts its negative regulatory effect primarily on the earliest stages of murine erythroid progenitor cell development. J Cell Physiol 162: 134–138 222. Wang GL, Semenza GL (1993) Desferrioxamine induces erythropoietin gene expression and hypoxia-inducible factor 1 DNA binding activity: implications for models of hypoxia signal transduction. Blood 82: 3610–3615

Physiology of erythropoiesis

65

223. Wang M, Tang D, Liu W, Chin K, Zhu J, Fibach E, Rodgers GP (2002) Hydroxyurea exerts bi-modal dose-dependent effects on erythropoiesis in human cultured erythroid cells via distinct pathways. Br J Haematol 119: 1098– 1105 224. Wang T, Zheng L, Lobito A, Chan F, Dale J, Sneller M, Yao X, Pack J, Strauss S, Leonardo M (1999) Inherited human caspase 10 mutation underlies defective lymphocyte and dendritic cell apoptosis in autoimmune lymphoproliferative syndrome type II. Cell 98: 47–58 225. Wang V, Schmidt T, Chen S, Sharp PA, Tantin D (2004) Embryonic lethality, decreased erythropoiesis, and defective octamer-dependent promoter activation in Oct-1 deﬁcient mice. Mol Cell Biol 24: 1022–1032 226. Ware CF (2003) The TNF superfamily. Cytokine Growth Factor Rev 14: 181– 184 227. Weil M, Jacobson MD, Raff MC (1998) Are caspases involved in the death of cells with a transcriptionally inactive nucleus? J Cell Sci 111: 2707–2715 228. Weiss MJ, Keller G, Orkin SH (1994) Novel insights into erythroid development revealed through in vitro differentiation of GATA-1- embryonic stem cells. Genes Dev 8: 1184–1197 229. Weiss MJ, Orkin SH (1995) Transcription factor GATA-1 permits survival and maturation of erythroid precursors by preventing apoptosis. Proc Natl Acad Sci USA 92: 9623–9627 230. Whyatt D, Karis J, Harkes IC, Verkek A, Gillemans N, Elefanty AG, Vairfo G, Ploemacher R, Grosveld F, Philipsen E (1997) The level of tissue-speciﬁc factor GATA-1 affects the cell-cycle machinery. Genes Funct 1: 11–24 231. Whyatt D, Lindeboom F, Karis A, Ferriera R, Milot E, Hendricks R, De Bruijn M, Langeveld A, Gribnau J, Grosveld F (2000) An intrinsic but cell-autonomous defect in GATA-1 overexpressing mouse erythroid cells. Nature 406: 519–524 232. Whyatt D, Grosveld F (2002) Cell-autonomous function of the retinoblastoma tumor suppressor protein: new interpretations of old phenotypes. EMBO Reports 3: 130–135 233. Wu H, Klingmuller U, Besmer P, Lodish HF (1995) Interaction of the erythropoietin and stem-cell-factor receptors. Nature 377: 242–246 234. Wu H, Liu X, Jaenisch R, Lodish HF (1995) Generation of committed erythroid BFU-E and CFU-E progenitors does not require erythropoietin or the erythropoietin receptor. Cell 83: 59–67 235. Xiao W, Kozumi K, Nischio M, Endo T, Fujimoto K, Sato I, Sakai T, Koike T, Sawada K (2002) Tumor necrosis factor-α inhibits generation of glycophorin-A+ cells by CD34+ cells. Exp Hematol 30: 1238–1247 236. Xu Z, Huang S, Chang LS, Agulnick AD, Brandt SJ (2003) Identiﬁcation of a TAL-1 target gene reveals a positive role for the LIM domain-binding protein Ldb1 in erythroid gene expression and differentiation. Mol Cell Biol 23: 7585–7599 237. Yeh WC, De La Pompo JL, Goeddel DV, Mak TW (1998) FADD is essential for embryo development and signalling for some, but not all, induction of apoptosis. Science 279: 1954–1958 238. Yokoyama T, Etoh T, Kitigawa H, Tsukahara S, Kannan Y (2003) Migration of erythroblastic islands toward the sinusoid as erythroid maturation proceeds in rat bone marrow. J Vet Med Sci 65: 449–452

66

U. Testa: Physiology of erythropoiesis

239. Zamai L, Secchiero P, Pierpaoli S, Bassini A, Papa S, Alnemri ES, Guidotti L, Vitale M, Zauli G (2000) TNF-related apoptosis-inducing ligand (TRAIL) as a negative regulator of normal human erythropoiesis. Blood 95: 3716–3724 240. Zambidis ET, Peault B, Park TS, Bunz F, Civin CI (2005) Hematopoietic differentiation of human embryonic stem cells progresses through sequential hemato-endothelial, primitive, and deﬁnitive stages resembling human yolk sac development. Blood 106: 860–870 241. Zanjani ED, Poster J, Burlington H, Mann LJ, Wassermaqn LR (1977) Liver as the primary site of erythropoietin production in the fetus. J Lab Clin Med 89 (3): 640–644 242. Zelko IN, Folz RJ (2005) Extracellular superoxide dismutase functions as a major repressor of hypoxia-induced erythropoietin gene expression. Endocrinology 146: 332–340 243. Zeng W, Chen G, Kajigaya S, Nunez O, Charrow A, Billings AM, Young NS (2004) Gene expression proﬁling in CD34 cells identiﬁes signiﬁcant differences between aplastic anemia patients and health volunteers. Blood 103: 325–332 244. Zermati Y, Garrido C, Amsellem S, Fishelson S, Bouscary D, Valensi F, Baret B, Solary E, Herime O (2001) Caspase activation is required for terminal erythroid differentiation. J Exp Med 193: 247–254 245. Zeuner A, Eramo A, Testa U, Felli N, Pelosi E, Mariani G, Srinivasula SM, Alnemri ES, Condorelli GL, Peschle C, De Maria R (2003) Control of erythroid cell production via caspase-mediated cleavage of transcription factor SCL/ Tal-1 Cell Death Differ 10: 905–913 246. Zeuner A, Pedini F, Signore M, Testa U, Pelosi E, De Maria R (2003) Stem cell factor protects erythroid precursor cells from chemotherapeutic agents via upregulation of Bcl-2 family proteins. Blood 102: 87–93 247. Zhang J, Cando D, Chen A, Kobra N, Winoto A (1998) Fas-mediated apoptosis and activation induced T-cell proliferation are defective in mice lacking FADD/Mort1. Nature 392: 296–300 248. Zhang J, Xu M (2002) Apoptotic DNA fragmentation and tissue homeostasis. Trends Cell Biol 12: 84–89 249. Zhang MY, Sun SC, Bell L, Miller BA (1998) NF-κB transcritpion factors are involved in normal erythropoiesis. Blood 91: 4136–4144 250. Zhang P, Zhang X, Iwama A, Yu C, Smith KA, Mueller BU, Narravula S, Torbett BE, Orkin SH, Tenen DG (2000) PU.1 inhibits GATA-1 function and erythroid differentiation by blocking GATA-1 DNA binding. Blood 96: 2641–2648 Correspondence: Ugo Testa, Department of Hematology and Oncology, Istitute Superiore di Senità, Viale Regina Elena 299, 00161 Rome, Italy.

Chapter 2

Biology of EPO and EPO-receptor C. Lacombe1,2 and P. Mayeux1 1

Université Paris-Descartes, Faculté de Médecine, INSERM, CNRS, Institut Cochin (U567-UMR 8104) Paris, France 2 Service d’Hématologie biologique, AP-HP, Hôpital Cochin, Paris, France

Introduction Patients undergoing chemotherapy for cancer are at risk of developing anemia, and recombinant human EPO is an interesting alternative to replace transfusions of allogenic red blood cells in this setting. The role of EPO, a 34 kDa glycoprotein hormone, is to control red blood cell production through the promotion of survival and proliferation of the erythroid progenitors in the bone marrow. EPO is the hematopoietic growth factor which is acting speciﬁcally on the late erythroid progenitors, so-called CFU-E (colony-forming unit-erythroid). These cells correspond to the last ampliﬁcation compartment of the erythroid lineage and give rise to the erythroblasts in the bone marrow. Because the main function of red cells is to transport oxygen from the lungs to the peripheral tissues, the regulation of EPO production is an important feature of the control of tissue oxygenation. Accordingly, EPO is the only hematopoietic growth factor the production of which is regulated by hypoxia. EPO acts through a speciﬁc receptor (EPO-R) belonging to the family of the hematopoietic growth factor receptors. Activation of the EPO-R by its ligand leads to the tyrosine phosphorylation of numerous proteins into the target cell; among these proteins, some migrate to the nucleus, where they stimulate the transcription of speciﬁc target genes. This article will review the regulation of EPO production, the structure of the EPO-R and the EPO-induced intracellular signaling events. We will also describe the mechanisms of EPO and EPO-R internalization and degradation.

Role of EPO in erythropoiesis Cultures of hematopoietic progenitors in semi-solid media have shown that the main targets of EPO are the late erythroid progenitors and especially the

68

C. Lacombe and P. Mayeux

cells which are called colony-forming unit-erythroid (CFU-E) (Gregory and Eaves 1978). EPO and EPO-R gene disruptions in mice conﬁrmed that EPO stimulation was absolutely required for survival and proliferation of CFU-E (Wu et al. 1995; Lin et al. 1996). Moreover, these experiments showed that EPO stimulation was not necessary for the commitment of the progenitors in the erythroid lineage. Indeed, both early erythroid progenitors, so-called BFU-E (burst-forming unit-erythroid), and CFU-E were produced to normal levels in EPO or EPO-R null mice (Wu et al. 1995; Lin et al. 1996), demonstrating that EPO is not involved in the determination of the erythroid lineage. EPO is able to sustain the proliferation of several hematopoietic cell lines either naturally expressing the EPO-R, such as HCD57 (Spivak et al. 1991) or UT7 (Komatsu et al. 1991) or after ectopic expression of the EPO-R (D’Andrea et al. 1991; Quelle and Wojchowski 1991). These cells, as well as primary erythroid cells, undergo apoptosis after EPO deprivation, thereby showing that EPO requirement is mandatory for the erythroid lineage. However, the speciﬁcity of action of EPO on erythroid progenitors is mainly due to the fact that, during erythroid differentiation, only EPO-R are present at the cell surface. Indeed, it has been shown that ectopic expression and stimulation of other cytokine receptors by their cognate ligands allowed the proliferation and maturation of erythroid progenitors in the absence of EPO (Socolovsky et al. 1998; Fichelson et al. 1999; Ghaffari et al. 1999). Similarly, the expression of anti-apoptotic proteins such as Bcl2 delayed apoptosis, allowed partial erythroid differentiation, although it did not sustain cell proliferation (Lacronique et al. 1997; Lesault et al. 2002). Thus, the overall action of EPO is to protect from apoptosis and to induce the proliferation of CFUE progenitors rather than to play a speciﬁc role in erythroid differentiation.

Regulation of EPO production EPO production is regulated by hypoxia that leads to an increase of the level of gene transcription (Schuster et al. 1989), there are no preformed stores of EPO. The identiﬁcation of the transcription factor hypoxia-inducible factor 1 (HIF-1) as a DNA transcriptional complex has been a critical step to understand the regulation of EPO production. Afﬁnity puriﬁcation showed that HIF-1 was composed of two subunits (Semenza and Wang 1992; Wang et al. 1995; Wang and Semenza 1995). Molecular cloning of HIF-1 by Semenza and colleagues (Wang et al. 1995) showed that the DNA binding complex was made of two basic-loop-helix PAS proteins called HIF-1α and HIF-1β. HIF1β had previously been identiﬁed as the aryl hydrocarbon nuclear receptor translocator (ARNT), a molecule involved in the xenobiotic response. In contrast, HIF-1α was a new member of this family of PAS (Per-Arnt-Sim) pro-

Biology of EPO and EPO-receptor

69

teins. The mechanism of regulation by hypoxia was ﬁrst studied in hepatoma cells which produced EPO. It was further shown that identical responses could be obtained in a large array of non-EPO-producing cells and that the system of gene regulation by oxygen was widespread from mammalian to insect cells (Maxwell et al. 1993; Wang and Semenza 1993). Many genes have been identiﬁed as targets of HIF-1 function; these include, in addition to EPO, vascular endothelial growth factor (VEGF), several glycolytic enzymes, glucose-transporter 1, inducible nitric oxyde synthase, heme oxygenase and transferrin (Maxwell et al. 1993; Wang and Semenza 1993). These data show that cellular response to hypoxia is an important physiological process and that a similar mechanism for oxygen sensing and signal transduction must be shared by many tissues and cells (Bunn and Poyton 1996). The major mechanism of regulation of HIF-1α involves the ubiquitinproteasome system: HIF-1α is constitutively degraded in normoxia, while it accumulates rapidly following exposure to hypoxia (Salceda and Caro 1997). The von Hippel-Lindau (VHL) tumor-suppressor protein (pVHL) has been linked to the regulation of the transcription factor HIF-1 (Maxwell et al. 1999). Wild-type pVHL is a component of an E3 ubiquitin-ligase complex that transfers ubiquitin onto substrates to be degraded, and the α-subunit of HIF-1 is the ubiquitination target for VHL (Ohh et al. 2000). Degradation of HIF-1α under normoxic conditions is triggered by hydroxylation of a proline residue 564 located within the oxygen-dependent degradation domain (ODD) of the protein (Hon et al. 2002). A family of oxygendependent prolyl hydroxylases is responsible for the modulation of HIF stability (Epstein et al. 2001; Semenza 2001). In addition, the carboxyl-terminal transactivation domain (C-TAD) of HIF-1α is able to recruit coactivator complexes such as p300/CBP only under hypoxic conditions. This regulation also involves an oxygen-dependent hydroxylation event targeted to a conserved asparagine residue (Lando et al. 2002). The VHL gene is inactivated in 80% of sporadic clear-cell renal carcinoma; these tumors lacking functional pVHL fail to degrade HIF-1α, which stimulates the transcription of a series of hypoxia-responsive genes, among which VEGF plays an important role in tumor angiogenesis (Wiesener and Eckardt 2002). More recently, Chuvash polycythemia has been described as a congenital defect of oxygen homeostasis due to an homozygous mutation in VHL gene. An Arg200Trp substitution impairs the interaction of VHL with HIF-1α, thus reducing the rate of degradation of HIF-1α and resulting in increased expression of downstream target genes including EPO (Ang et al. 2002).

Structure of the EPO receptor EPO acts on its target cells through speciﬁc membrane receptors. They are mainly expressed at the CFU-E stage, receptor expression then decreases

70

C. Lacombe and P. Mayeux

with erythroid maturation (Mayeux et al. 1987). The number of EPO-R at the cell surface of normal or transformed erythroid cells is low: around one thousand per cell (reviewed in D’Andrea and Zon 1990). EPO-R are present on the surface of erythroid cells (Broudy et al. 1991), on megakaryocytes (Fraser et al. 1989), on endothelial cells (Anagnostou et al. 1994), neuronal cells (Masuda et al. 1993) and probably many other cell types according to the wide expression of its mRNA (Suzuki et al. 2002). The EPO-R cDNA has been cloned by A. D’Andrea (D’Andrea et al. 1989) and was shown to encode a single membrane-spanning protein of 507 amino-acids which does not possess catalytic activity in its intracellular region. This receptor belongs to the cytokine receptor family. Most of these receptors form multimeric complexes; several chains have been cloned for the receptors for IL-2 to IL7, for the GM-CSF and the leukemia inhibitory factor (LIF) receptors. The 66 kDa chain cloned in the EPO-R is responsible for intracellular signalling, since the transfection of this protein in hematopoietic cells such as Ba/F3, 32D or DA3, which do not possess EPO-R at their cell surface, allows their proliferation in response to EPO alone (Gobert et al. 1995b). However, chemical cross-linking experiments with 125 iodine-labelled EPO have detected at the surface of erythroid progenitors two additional proteins of 85 and 100 kDa, respectively, which are not recognized by anti p66 antibodies (Mayeux et al. 1991). These proteins probably belong to the receptor complex but are not able to bind to the ligand. Their cloning will be required to better understand their speciﬁc role. The ﬁxation of EPO on its cognate receptor leads to dimerization of the p66 EPO-R as shown by cristallization of the complex EPO/EPO-R (Syed et al. 1998). One EPO-R molecule binds to the ligand with a rather high afﬁnity (Kd = 1 nM), the second receptor molecule binds to the complex with a lower afﬁnity (Kd = 1 μM) and both associations give a high-afﬁnity binding (Kd = 160 pM). Thus, EPO is a bivalent molecule for the ﬁxation of the EPO-R. Recently, the presence of EPO-R has been reported on the surface of several cells not belonging to the erythroid lineage. Therefore, it appears that the anti-apoptotic role of EPO is not restricted to erythropoiesis but is enlarged to many other tissues of normal origin. The role of EPO-R expression on tumor cells is the subject of a separate chapter in this book.

EPO-induced intracellular signalling As mentioned above, the dimerization of the EPO-R after binding to one molecule of EPO leads to its activation and subsequent downstream intracellular signalling. Many groups showed that the EPO-induced activation led to the rapid tyrosine phosphorylation of a number of proteins, even though the EPO-R does not possess endogenous tyrosine kinase activity. The two Jak2 tyrosine kinase molecules are pre-associated to the EPO-R. This asso-

Biology of EPO and EPO-receptor

71

ciation between Jak2 and EPO-R occurs in the endoplasmic reticulum and seems to be required for the maturation and expression of the receptor at the cell surface (Huang et al. 2001). The ﬁxation of EPO on its receptor leads to a change of conformation of the complex; the two Jak2 molecules become positioned in sufﬁcient proximity for their reciprocal transphosphorylation and activation (Remy et al. 1999). Activated Jak2 proteins in turn phosphorylate the EPO-R tyrosine residues (Dusanter-Fourt et al. 1992, 1994). These phosphorylated tyrosines become secondary binding sites for signalling proteins containing SH2 (SRC homology 2) domains. Thus, a complex of signalling proteins is generated around the dimerized and activated receptor.

Signalling pathways activated by EPO The PI 3-kinase/Akt pathway PI 3-kinase is associated to EPO-R in response to EPO stimulation (Mayeux et al. 1993). It was ﬁrst reported that one SH2 motif of the PI 3-kinase p85 subunit was bound to the last tyrosine residue of the EPO-R (Damen et al. 1995). Other adaptor mechanisms of PI 3-kinase activation have since been described: two adaptor proteins, IRS2 (Verdier et al. 1997) and GAB1 (Lecoq-Lafon et al. 1999) are phosphorylated following EPO stimulation and associate with PI 3-kinase. We demonstrated that these three alternative pathways independently led to EPO-induced activation of PI 3-kinase (Fig. 1). Ly294002, a speciﬁc inhibitor of PI 3-kinase activation inhibits EPOinduced cell proliferation, thereby suggesting that the PI 3-kinase pathway plays an important role in the mode of action of EPO. PI 3, 4, 5 trisphosphate, a metabolite of the PI 3-kinase pathway, activates the serine/threonine kinase AKT which is known to play a major role in the inhibition of cellular apoptosis (Franke et al. 1997). We showed that PI3K signalling occurred through modulation of the E3 ligase SCFSKP2 which downregulated p27Kip1 inhibitor via proteasome degradation. Thus, the activation of PI 3-kinase in response to EPO stimulation is an important event, contributing to the inhibition of apoptosis of erythroid progenitors and required for their proliferation (Bouscary et al. 2003).

The Ras/MAP kinase pathway Ras, Raf and MAP kinase proteins are all activated by EPO (Gobert et al. 1995a). The adaptor proteins SHC and Grb2 are associated to the EPO-R, together with the tyrosine phosphatase SHP-2, which is also able to bind Grb2 (Tauchi et al. 1995). The Ras/MAP kinase pathway could be activated by EPO via several different mechanisms; however, these mechanisms have

72

C. Lacombe and P. Mayeux

Epo

Jak-2

IRS-2

Y1 Y2

GAB1

PI 3-kinase

Y8

Fig. 1. Mechanisms of PI 3-kinase activation by EPO. The IRS2 adaptor is constitutively associated to the proximal part of the intracellular EPO-R domain. After EPO stimulation, its phosphorylation leads to ﬁxation of PI 3-kinase. Phosphorylation of Y1 and Y2 tyrosine residues of the EPO-R allows GAB1 ﬁxation and its phosphorylation. After tyrosine phosphorylation, GAB1 associates with the PI 3-kinase. The third possibility of PI 3-kinase activation is its direct ﬁxation to the last phosphorylated tyrosine of the EPO-R

not been deﬁnitively identiﬁed. This pathway is also involved in EPO-induced cell proliferation (Damen and Krystal 1996).

The STAT pathway The STAT (Signal Transducer and Activator of Transcription) proteins are transcription factors activated in response to several cytokines (Ihle 1995). EPO activates the two isoforms of STAT5, STAT5a and STAT5b (Gouilleux et al. 1995; Pallard et al. 1995). The STAT proteins bind to the Tyr 343 and 401 of the EPO-R, they become phosphorylated and activated and translocate into the nucleus (Gobert et al. 1996). The role of STAT transcription factors during EPO stimulation has been a matter of debate: some reports established a correlation between STAT activation and cell proliferation (Chrétien et al. 1996), whereas others attributed a role for STAT in erythroid differentiation (Wakao et al. 1997). A double knock out for STAT5a and STAT5b genes did not lead to any major defect of erythropoiesis (Teglund et al. 1998). It was further shown that STAT5 was essential for the high erythropoietic rate during fetal development, because it bound to the promoter of the Bcl-X gene and played a crucial role in EPO-R antiapoptotic signaling (Socolovsky et al. 1999).

Biology of EPO and EPO-receptor

73

Pathways leading to signalling arrest The signal of activation which results from stimulation by a cytokine needs to be terminated by additional pathways leading to signal interruption. Several proteins have been recently isolated that play a negative role in EPOinduced signal transduction.

The tyrosine phosphatase SHP-1 SHP-1 protein is involved in EPO signalling, this protein binds both to the third Tyr of the EPO-R and to Jak2 tyrosine kinase. This association leads to Jak2 dephosphorylation and thus to the signal arrest (Klingmuller et al. 1995). Several cases of familial erythrocytosis have been reported, due to a truncation of the cytoplasmic domain of the EPO-R, which lacked the binding site for SHP-1 protein and thereby became hypersensitive to EPO stimulation in vivo (De La Chapelle et al. 1993; Kralovics et al. 1997). The erythroid progenitors derived from mice knocked out for the Shp-1 gene are also hypersensitive to EPO (Van Zant and Shultz 1989).

The Cis and SOCS3 proteins The Cis protein (for cytokine-inducible SRC homology 2-containing protein) is one of the known targets of STAT5 factor (Yoshimura et al. 1995). Cis is an inhibitor of EPO-induced cell proliferation. We showed that Cis was associated to the Tyr 401 of the EPO-R and was ubiquitinated. This ubiquitination of Cis suggests that this protein could play an active role in the sequestration of the EPO/EPO-R complex by the proteasome (Verdier et al. 1998). Indeed, we showed that the proteasome controls the down-regulation of EPO-R in EPO-stimulated cells by inhibiting the cell surface replacement of internalized EPO-R (Verdier et al. 2000). A second member of the Cis family called SOCS3 (for Suppressor Of Cytokine Signaling) has been described as essential in the regulation of erythropoiesis. However, contradictory results have been published and the precise role of SOCS3 in erythropoiesis remains to be understood (Marine et al. 1999; Roberts et al. 2001).

Role of the proteasome system The duration of activation of the EPO-R is rather short when compared to that of other cytokine receptors such as the thrombopoietin (TPO) receptor. Indeed, after EPO stimulation, the different intracellular pathways are acti-

74

C. Lacombe and P. Mayeux

vated during a very short period (up to 10 min), whereas these pathways are still activated one hour after TPO stimulation. For the EPO-R, two mechanisms involving the ubiquitin/proteasome system are responsible for the control of duration of receptor activation: ﬁrst, the stimulated receptors are internalized and degraded, second, new synthetized EPO-R are hampered to reach the plasmic membrane after EPO stimulation. We showed that, following EPO-R activation, the EPO-R is ubiquitinated at the cell surface by a Jak-2 dependent mechanism. As a consequence, there is a cleavage by the proteasome of the intracellular part of the EPO-R which probably leads to a rapid arrest of the signalling process. The remaining of the EPO/EPO-R complex is internalized and degraded by the lysosomes (Verdier et al. 2000; Walrafen et al. 2005) (Fig. 2). The E3 ligase responsible for the EPO-R ubiquitination is not yet identiﬁed.

Fig. 2. Mechanisms of EPO-R down-regulation during EPO stimulation. Ia: Upon EPO binding, the EPO-R is tyrosine-phosphorylated, triggering intracellular signaling. II: The EPO-R is ubiquitinated at the cell surface. III: The proteasome degrades the EPO-R cytoplasmic tail, removing all the phosphorylated tyrosine residues and preventing further signal transduction. IV: The cleaved EPO-R is internalized and degraded in the lysosomes. Ib: If Jak2 activation is prevented by the inhibitor AG490, neither phosphorylation nor ubiquitination of the EPO-R occur. EPO-EPO-R complexes are still internalized but the complexes are not degraded and recycle to the cell surface. This research was originally published in Blood. Walrafen et al. Both proteasomes and lysosomes degrade the activated erythropoietin receptor. Blood. 2005. 105:600–8.© the American Society of Hematology

Biology of EPO and EPO-receptor

75

The efﬁcient mechanisms of downregulation explain why the EPO-R very rapidly disappears from the cell surface of erythroid cells in contrast to the TPO receptor that is still present on the platelet surface. These mechanisms of EPO-R downregulation could also improve erythroid terminal differentiation. Indeed, erythroid terminal differentiation requires caspase and FOXO 3A activation (Zermati et al. 2001; Bakker et al. 2004). Both activations are blocked by pathways activated after EPO-R stimulation like the PI 3-kinase/Akt pathway. Therefore, the disappearance of the EPO-R favors a balance towards caspase and FOXO 3A activation, thereby leading to terminal differentiation of these erythroid cells.

Acknowledgements This work was supported by a grant from Association pour la Recherche contre le Cancer (ARC) and a grant from Ligue Nationale Contre le Cancer (LNCC).

References 1.

2.

3.

4.

5.

6. 7.

Anagnostou A, Liu Z, Steiner M, Chin K, Lee ES, Kessimian N, Noguchi CT (1994) Erythropoietin receptor mRNA expression in human endothelial cells. Proc Natl Acad Sci USA 91: 3974–3978 Ang SO, Chen H, Hirota K, Gordeuk VR, Jelinek J, Guan Y, Liu E, Sergueeva AI, Miasnikova GY, Mole D, Maxwell PH, Stockton DW, Semenza GL, Prchal JT (2002) Disruption of oxygen homeostasis underlies congenital Chuvash polycythemia. Nat Genet 32: 614–621 Bakker WJ, Blazquez-Domingo M, Kolbus A, Besooyen J, Steinlein P, Beug H, Coffer PJ, Lowenberg B, von Lindern M, van Dijk TB (2004) FoxO3a regulates erythroid differentiation and induces BTG1, an activator of protein arginine methyl transferase 1. J Cell Biol 164: 175–184 Bouscary D, Pene F, Claessens YE, Muller O, Chretien S, Fontenay-Roupie M, Gisselbrecht S, Mayeux P, Lacombe C (2003) Critical role for PI 3-kinase in the control of erythropoietin-induced erythroid progenitor proliferation. Blood 101: 3436–3443 Broudy VC, Lin N, Brice M, Nakamoto B, Papayannopoulou T (1991) Erythropoietin receptor characteristics on primary human erythroid cells. Blood 77: 2583–2590 Bunn HF, Poyton RO (1996) Oxygen sensing and molecular adaptation to hypoxia. Physiol Rev 76: 839–885 Chrétien S, Varlet P, Verdier F, Gobert S, Cartron JP, Gisselbrecht S, Mayeux P, Lacombe C (1996) Erythropoietin-induced differentiation of the human erythroleukemia cell line TF-1 correlates with impaired STAT5 activation. EMBO 15: 4174–4181

76 8.

9. 10. 11.

12. 13.

14.

15.

16.

17.

18. 19.

20.

21.

C. Lacombe and P. Mayeux Damen J, Cutler RL, Jiao H, Yi T, Krystal G (1995) Phosphorylation of tyrosine 503 in the erythropoietin receptor (EpR) is essential for binding the p85 subunit of phosphatidylinositol (PI) 3-kinase and for EpR-associated PI 3kinase activity. J Biol Chem 270: 23402–23406 Damen JE, Krystal G (1996) Early events in erythropoietin-induced signaling. Exp Hematol 24: 1455–1459 D’Andrea AD, Lodish HF, Wong GG (1989) Expression cloning of the murine erythropoietin receptor. Cell 57: 277–285 D’Andrea AD, Yoshimura A, Youssouﬁan H, Zon L, Koo J, Lodish HF (1991) The cytoplasmic region of the erythropoietin receptor contains non overlapping positive and negative growth-regulatory domains. Mol Cell Biol 11: 1980– 1987 D’Andrea AD, Zon LI (1990) Erythropoietin receptor: subunit structure and activation. J Clin Invest 86: 681–687 De La Chapelle A, Traskelin A, Juvonen E (1993) Truncated erythropoietin receptor causes dominantly inherited benign human erythrocytosis. Proc Nat Acad Sci USA 90: 4495–4499 Dusanter-Fourt I, Casadevall N, Lacombe C, Muller O, Billat C, Fischer S, Mayeux P (1992) Erythropoietin induces the tyrosine phosphorylation of its own receptor in human erythropoietin-responsive cells. J Biol Chem 267: 10670–10675 Dusanter-Fourt I, Muller O, Ziemiecki A, Mayeux P, Drucker B, Djiane J, Wilks A, Harpur AG, Fischer S, Gisselbrecht S (1994) Identiﬁcation of JAK protein tyrosine kinases as signaling molecules for prolactin. Functional analysis of prolactin receptor and prolactin-erythropoietin receptor chimera expressed in lymphoid cells. EMBO 13: 2583–2591 Epstein AC, Gleadle JM, McNeill LA, Hewitson KS, O’Rourke J, Mole DR, Mukherji M, Metzen E, Wilson MI, Dhanda A, Tian YM, Masson N, Hamilton DL, Jaakkola P, Barstead R, Hodgkin J, Maxwell PH, Pugh CW, Schoﬁeld CJ, Ratcliffe PJ (2001) C. elegans EGL-9 and mammalian homologs deﬁne a family of dioxygenases that regulate HIF by prolyl hydroxylation. Cell 107: 43–54 Fichelson S, Chretien S, Rokicka-Piotrowicz M, Bouhanik S, Gisselbrecht S, Mayeux P, Lacombe C (1999) Tyrosine residues of the erythropoietin receptor are dispensable for erythroid differentiation of human CD34+ progenitors. Biochem Biophys Res Commun 256: 685–691 Franke TF, Kaplan DR, Cantley LC (1997) PI3K: downstream AKTion blocks apoptosis. Cell 88: 435–437 Fraser JK, Tan AS, Lin FJ, Berridge MV (1989) Expression of high afﬁnity binding sites for erythropoietin on rat and mouse megakaryocytes. Exp Hemat 17: 10–16 Ghaffari S, Wu H, Gerlach M, Han Y, Lodish HF, Daley GQ (1999) BCR-ABL and v-SRC tyrosine kinase oncoproteins support normal erythroid development in erythropoietin receptor-deﬁcient progenitor cells. Proc Natl Acad Sci U S A 96: 13186–13190 Gobert S, Chrétien S, Gouilleux F, Muller O, Pallard C, Dusanter-Fourt I, Groner B, Lacombe C, Gisselbrecht S, Mayeux P (1996) Identiﬁcation of tyrosine residues within the intracellular domain of the erythropoietin receptor crucial for STAT5 activation. EMBO 15: 2434–2441

Biology of EPO and EPO-receptor 22.

77

Gobert S, Duprez V, Lacombe C, Gisselbrecht S, Mayeux P (1995a) Erythropoietin activates three forms of MAP kinase in UT7 erythroleukemia cells. Eur J Biochem 234: 75–83 23. Gobert S, Porteu F, Pallu S, Muller O, Sabbah M, Dusanter-Fourt I, Courtois G, Lacombe C, Gisselbrecht S, Mayeux P (1995b) Tyrosine phosphorylation of the erythropoietin receptor: role for differentiation and mitogenic signal transduction. Blood 86: 598–606 24. Gouilleux F, Pallard C, Dusanter-Fourt I, Wakao H, Haldosen L-A, Norstedt G, Levy D, Groner B (1995) Prolactin, growth hormone, erythropoietin and granulocyte-macrophage colony stimulating factor induce MGF-STAT5 DNA binding activity. EMBO 14: 2005–2013 25. Gregory CJ, Eaves AC (1978) Three stages of erythropoietic progenitor cell differentiation distinguished by a number of physical and biologic properties. Blood 51: 527–537 26. Hon WC, Wilson MI, Harlos K, Claridge TD, Schoﬁeld CJ, Pugh CW, Maxwell PH, Ratcliffe PJ, Stuart DI, Jones EY (2002) Structural basis for the recognition of hydroxyproline in HIF-1 alpha by pVHL. Nature 417: 975–978 27. Huang LJ, Constantinescu SN, Lodish HF (2001) The N-terminal domain of Janus kinase 2 is required for Golgi processing and cell surface expression of erythropoietin receptor. Mol Cell 8: 1327–1338 28. Ihle JN (1995) Cytokine receptor signalling. Nature 377: 591–594 29. Klingmuller U, Lorenz U, Cantley LC, Neel BC, Lodish HC (1995) Speciﬁc recruitment of SH-PTP1 to the erythropoietin receptor causes inactivation of JAK2 and termination of proliferative signals. Cell 80: 729–738 30. Komatsu N, Nakauchi H, Miwa A, Ishihara T, Eguchi M, Moroi M, Okada M, Sato Y, Wada H, Yamata Y, Suda T, Miura Y (1991) Establishment and characterization of a human leukemic cell line with megakaryocytic features: dependency on granulocyte-macrophage colony stimulating factor, interleukin 3, or erythropoietin for growth and survival. Can Res 51: 341–345 31. Kralovics R, Indrak K, Stopka T, Berman BW, Prchal JF, Prchal JT (1997) Two new EPO receptor mutations: truncated EPO receptors are most frequently associated with primary familial and congenital polycythemias. Blood 90: 2057–2061 32. Lacronique V, Varlet P, Mayeux P, Porteu A, Gisselbrecht S, Kahn A, Lacombe C (1997) Bcl-2 targeted overexpression into the erythroid lineage of transgenic mice delays but does not prevent the apoptosis of erythropoietin-deprived erythroid progenitors. Blood 90: 3050–3056 32a. Masuda S, Nagao M, Takahata K, Konishi Y, Gallyas F Jr., Tabira T, Sasaki R (1993) Functional erythropoietin receptor of the cells with neural characteristics. Comparison with receptor properties of erythroid cells. J Biol Chem 268: 11208–11216 33. Maxwell PH, Pugh CW, Ratcliffe PJ (1993) Inducible operation of the erythropoietin 3′ enhancer in multiple cell lines: evidence for a widespread oxygensensing mechanism. Proc Natl Acad Sci U S A 90: 2423–2427 34. Maxwell PH, Wiesener MS, Chang GW, Clifford SC, Vaux EC, Cockman ME, Wykoff CC, Pugh CW, Maher ER, Ratcliffe PJ (1999) The tumour suppressor protein VHL targets hypoxia-inducible factors for oxygen-dependent proteolysis. Nature 399: 271–275

78 35.

36.

37.

38.

39.

40.

41. 42.

43.

44.

45. 46.

47.

48.

49.

C. Lacombe and P. Mayeux Mayeux P, Billat C, Jacquot R (1987) The erythropoietin receptor of rat erythroid progenitor cell : Characterization and afﬁnity cross-linkage. J Biol Chem 262: 13985–13990 Mayeux P, Dusanter-Fourt I, Muller O, Mauduit P, Sabbah M, Drucker B, Vainchencker W, Fischer S, Lacombe C, Gisselbrecht S (1993) Erythropoietin induces the association of phosphatidylinositol 3′ kinase with a tyrosine phosphorylated complex containing the erythropoietin receptor. Eur J Biochem 216: 821–828 Mayeux P, Lacombe C, Casadevall N, Chretien S, Dusanter I, Gisselbrecht S (1991) Structure of the murine erythropoietin receptor complex. Characterization of the erythropoietin cross-linked proteins. J Biol Chem 266: 23380–23385 Ohh M, Park CW, Ivan M, Hoffman MA, Kim TY, Huang LE, Pavletich N, Chau V, Kaelin WG (2000) Ubiquitination of hypoxia-inducible factor requires direct binding to the beta-domain of the von Hippel-Lindau protein. Nat Cell Biol 2: 423–427 Pallard C, Fabrice G, Martine C, Groner B, Gisselbrecht S, Dusanter-Fourt I (1995) Interleukin-3, erythropoietin, and prolactin activate a STAT5 like factor in lymphoid cells. J Biol Chem 270: 15942–15945 Quelle DE, Wojchowski DM (1991) Localized cytosolic domains of the erythropoietin receptor regulate growth signaling and down-modulate responsiveness to granulocyte-macrophage colony-stimulating factor. Proc Natl Acad of Sci USA 88: 4801–4805 Remy I, Wilson IA, Michnick SW (1999) Erythropoietin receptor activation by a ligand-induced conformation change. Science 283: 990–993 Roberts AW, Robb L, Rakar S, Hartley L, Cluse L, Nicola NA, Metcalf D, Hilton DJ, Alexander WS (2001) Placental defects and embryonic lethality in mice lacking suppressor of cytokine signaling 3. Proc Natl Acad Sci U S A 98: 9324–9329 Salceda S, Caro J (1997) Hypoxia-inducible factor 1alpha (HIF-1alpha) protein is rapidly degraded by the ubiquitin-proteasome system under normoxic conditions. Its stabilization by hypoxia depends on redox-induced changes. J Biol Chem 272: 22642–22647 Schuster SJ, Badiavas EV, Costa-Giomi P, Weinmann R, Erslev AJ, Caro J (1989) Stimulation of erythropoietin gene transcription during hypoxia and cobalt exposure. Blood 73: 13–16 Semenza GL (2001) HIF-1, O(2), and the 3 PHDs: how animal cells signal hypoxia to the nucleus. Cell 107: 1–3 Semenza GL, Wang GL (1992) A nuclear factor induced by hypoxia via de novo protein synthesis binds to the human erythropoietin gene enhancer at a site required for transcriptional activation. Mol Cell Biol 12: 5447–5454 Socolovsky M, Fallon AE, Lodish HF (1998) The prolactin receptor rescues EPO-R-/- erythroid progenitors and replaces EPO-R in a synergistic interaction with c-kit. Blood 92: 1491–1496 Socolovsky M, Fallon AE, Wang S, Brugnara C, Lodish HF (1999) Fetal anemia and apoptosis of red cell progenitors in Stat5a-/-5b-/- mice: a direct role for Stat5 in Bcl-X(L) induction. Cell 98: 181–191 Spivak JL, Pham T, Isaacs M, Hankins WD (1991) Erythropoietin is both a mitogen and a survival factor. Blood 77: 1228–1233

Biology of EPO and EPO-receptor 50.

51.

52.

53.

54. 55.

56.

57.

58.

59.

60.

61. 62. 63.

64.

79

Suzuki N, Ohneda O, Takahashi S, Higuchi M, Mukai HY, Nakahata T, Imagawa S, Yamamoto M (2002) Erythroid-speciﬁc expression of the erythropoietin receptor rescued its null mutant mice from lethality. Blood 100: 2279–2288 Syed RS, Reid SW, Li C, Cheetham JC, Aoki KH, Liu B, Zhan H, Osslund TD, Chirino AJ, Zhang J, Finer-Moore J, Elliott S, Sitney K, Katz BA, Matthews DJ, Wendoloski JJ, Egrie J, Stroud RM (1998) Efﬁciency of signalling through cytokine receptors depends critically on receptor orientation. Nature 395: 511–516 Tauchi T, Feng GS, Shen R, Hoatlin M, Bagby GC, Kabatt D, Lu L, Broxmeyer HE (1995) Involvement of SH2-containing phosphotyrosine phosphatase Syp in erythropoietin receptor signal transduction pathway. J Biol Chem 270: 5631–5635 Teglund S, McKay C, Schuetz E, van Deursen JM, Stravopodis D, Wang D, Brown M, Bodner S, Grosveld G, Ihle JN (1998) Stat5a and Stat5b proteins have essential and nonessential, or redundant, roles in cytokine responses. Cell 93: 841–850 Van Zant G, Shultz L (1989) Hematological abnormalities of the immunodeﬁcient mouse mutant, viable motheaten (mev). Exp Hematol 17: 81–87 Verdier F, Chrétien S, Billat C, Gisselbrecht S, Lacombe C, Mayeux, P (1997) Erythropoietin induces the tyrosine phosphorylation of insulin receptor substrate-2: An alternate pathway for erythropoietin-induced phosphatidylinositol 3-kinase activation. J Biol Chem 272: 26173–26178 Verdier F, Chretien S, Muller O, Varlet P, Yoshimura A, Gisselbrecht S, Lacombe C, Mayeux P (1998) Proteasomes regulate erythropoietin receptor and signal transducer and activator of transcription 5 (STAT5) activation. Possible involvement of the ubiquitinated cis protein. J Biol Chem 273: 28185–28190 Verdier F, Walrafen P, Hubert N, Chretien S, Gisselbrecht S, Lacombe C, Mayeux P (2000) Proteasome regulate the duration of erythropoietin receptor activation by controlling down-regulation of cell surface receptors. J Biol Chem 275: 18375–18381 Wakao H, Chida D, Damen JE, Krystal G, Miyajima A (1997) A possible involvement of Stat5 in erythropoietin-induced hemoglobin synthesis. Biochem Biophys Res Commun 234: 198–205 Walrafen P, Verdier F, Kadri Z, Chrétien S, Lacombe C, Mayeux P (2005) Both proteasomes and lysosomes degrade the activated erythropoietin receptor. Blood 105: 600–608 Wang GL, Jiang BH, Rue EA, Semenza GL (1995) Hypoxia-inducible factor 1 is a basic-helix-loop-helix-PAS heterodimer regulated by cellular O2 tension. Proc Natl Acad Sci U S A 92: 5510–5514 Wang GL, Semenza GL (1993) General involvement of hypoxia-inducible factor 1 in transcriptional response to hypoxia. Proc Natl Acad Sci U S A 90: 4304–4308 Wang GL, Semenza GL (1995) Puriﬁcation and characterization of hypoxiainducible factor 1. J Biol Chem 270: 1230–1237 Wiesener MS, Eckardt KU (2002) Erythropoietin, tumours and the von HippelLindau gene: towards identiﬁcation of mechanisms and dysfunction of oxygen sensing. Nephrol Dial Transplant 17: 356–359 Wu H, Liu X, Jaenisch R, Lodish HF (1995) Generation of committed erythroid BFU-E and CFU-E progenitors does not require erythropoietin or the erythropoietin receptor. Cell 83: 59–67

80 65.

66.

C. Lacombe and P. Mayeux: Biology of EPO and EPO-receptor Yoshimura A, Ohkubo T, Kigushi T, Jenkins NA, Gilbert DJ, Copeland NG, Hara T, Miyajima A (1995) A novel cytokine-inducible gene CIS encodes an SH2-containing protein that binds to tyrosine-phosphorylated interleukin 3 and erythropoietin receptors. EMBO 14: 2816–2826 Zermati Y, Garrido C, Amsellem S, Fishelson S, Bouscary D, Valensi F, Varet B, Solary E, Hermine O (2001) Caspase activation is required for terminal erythroid differentiation. J Exp Med 193: 247–254

Correspondence: Catherine Lacombe, MD, PhD, Hematology Department, Hôpital Cochin, 27 rue du Faubourg St Jacques, 75014 Paris, France, E-mail: [email protected]

Chapter 3

The role of erythropoietin receptor expression on tumor cells J. Fandrey Institut für Physiologie der Universität Duisburg-Essen, Essen, Germany

Introduction Based on the successful treatment of renal anemia with recombinant erythropoietin (rhEPO), EPO was soon considered for the use in other forms of anemia, such as the cancer associated anemia. Anemia is common in patients with malignant disease (Jumbe 2002; Weiss 2003). It is either due to the myelosuppressive effects of chemotherapy and/or radiotherapy or sometimes inﬁltration of the bone marrow. In many cases, cancer patients suffer from the anemia of chronic disease which is caused mainly by functionally impaired iron availability, reduced EPO levels and direct suppressive effects of cytokines on erythroid progenitor cells (Jelkmann 1998). Experimental evidence suggested that proinﬂammatory cytokines that may be increased during the course of malignant disease, effectively inhibit endogenous EPO synthesis (Jelkmann et al. 1992). Thus, the resulting anemia impairs the survival of cancer patients by 1. making chemotherapy or radiotherapy less effective due to the reduced oxygen supply, 2. limiting the intensity of treatment due to fatigue and overall exhaustion, 3. decreasing the patients’ quality of life and thus indirectly inﬂuencing therapy and 4. impairing tumor oxygenation which causes areas of hypoxia within the tumors where angiogenesis may be stimulated or the potential for tumor growth and metastases is enhanced. The introduction of EPO as a supportive measure in the treatment of cancer patients can correct the anemia and has improved the quality of life of the patients (Vaupel et al. 2005). In addition, at least in a meta-analysis in which 200 publications were critically reviewed anemia has been identiﬁed as an independent prognostic factor for survival (Caro et al. 2001). Whether the correction of anemia has an impact on survival will be subject of other chapters in this book.

82

J. Fandrey

However, from the very beginning when rhEPO became available for therapy there has been the serious concern that a growth factor such as EPO might also promote the growth of tumor cells. Although these safety considerations were not followed by any unfavorable observations in millions of renal patients two recent publications have fueled the discussion whether EPO treatment may actually be dangerous for patients with cancer. Both studies (Henke et al. 2003; Leyland-Jones 2003) diverge from the generally positive results obtained with EPO treatment in cancer patients and have initiated a hearing at the FDA in May 2004 (FDA 2004). In addition, both studies have been extensively discussed in letters to The Lancet and some of the potential pitfalls with these studies or differences in interpreting their data have been documented in a recent publication to which the reader is referred to here (Vaupel et al. 2005). Interestingly, at the above mentioned hearing the FDA panel considered that determination of the EPO receptor status of tumors is “difﬁcult and (may) not (be) totally relevant” (FDA ODAC on EPO 2004). Still, the clinical studies by Henke et al. (2003) and Leyland-Jones et al. (2003) most probably received so much attention because a number of in vitro studies was published in which established tumor cell lines were claimed to respond to recombinant EPO with enhanced proliferation. Although the safety issue in patients can only be answered in large clinical studies that are underway (FDA ODAC on EPO 2004) it also appears necessary to selectively review some of the in vitro studies particularly with respect to the EPO receptor and its intracellular signaling pathways.

Does EPO cause proliferation of tumor cells in vitro? Westenfelder and Baranowski (2000) studied cell lines from patients with renal cell carcinoma (RCC) and also established renal tumor cell lines of human origin (Caki-2, 786-0) as well as a mouse cell line of a renal adenocarcinoma (RAG). EPO receptor mRNA and protein were detected and 125 Iodine-labelled EPO binding was found but only to a single class of receptors (erythroid progenitors have a high and a low afﬁnity receptor binding site (Philo et al. 1996). Proliferative effects of EPO, however, were weak and only observed in serum deprived cultures. In one cell line the addition of 10% serum had a more prominent effect than EPO. Whatever component(s) of the serum may have caused this effect it is conceivable that they may be present in a tenfold higher concentration in vivo and may thus be much more relevant as tumor proliferation supporting factors than EPO. Interestingly, in samples from the tumors low level expression of the EPO gene was detected while the permanent cell lines had no EPO mRNA. Thus, a potential autocrine role of EPO in renal tumors might be suggested although only a minor fraction of renal cell carcinomas (RCC) express EPO mRNA (Gross

The role of erythropoietin receptor expression on tumor cells

83

et al. 1994). However, a study by Shouval et al. (1988) which was also quoted in the work by Westenfelder and Baranowski (2000) demonstrated that EPO secreting RCC cells were less tumorigenic when transplanted into nude mice than non-EPO-secreting cells. Acs et al. (2001) suggested that either endogenous EPO or exogenous rhEPO could support the growth and survival of EPO receptor expressing breast carcinoma cells. In their study they used ﬁve breast carcinoma cell lines that are well-established but, on the other hand, have gone through many passages in culture. It is therefore questionable whether they actually reﬂect the in vivo system of a breast carcinoma. When studying these cells, the authors deprived the cells of serum which may synchronize the cells and thus make them even less suitable as a model. Extremely high concentrations of up to 10,000 mU EPO/ml were used to show increased DNA synthesis and even a 25-fold higher concentration of EPO was claimed to increase tyrosine phosphorylation of the EPO receptor protein. Interestingly, as a positive control for the EPO receptor detection Acs et al. (2001) used Hela cells which served as a negative control in the above mentioned study by Westenfelder and Baranowski (2000). In a more recent study Acs et al. (2002) proposed an autocrine or paracrine effect of EPO produced by tumor cells to promote breast carcinogenesis (Acs et al. 2002). Immunohistochemical staining of tumor cells showed cytoplasmic localization of the EPO receptor which was not reﬁned to malignant cells but also to normally appearing ductal cells. Acs et al. (2002) and in an additional study Arcasoy et al. (2002), who detected EPO receptor protein by immunohistochemical staining in breast cancer cells, both used an antibody that was directed against the C-terminus, i.e. the intracellular part, of the EPO receptor. By this procedure of immunohistochemical staining splice variants or soluble receptors will be missed (see below). Moreover, the speciﬁcity of the antibody used (Santa Cruz C20) has recently been questioned since it crossreacted with multiple proteins of sizes different from that predicted for the EPO receptor in Western blot analysis (Busse et al. 2005). More recently, Pajonk et al. (2004) used HeLa cells which were stably transfected with the EPO receptor cDNA to express the EPO receptor. They reported tyrosine phosphorylation of STAT5 which is of central importance for EPO receptor signaling in erythroid progenitors (see below; (Klingmuller 1997)) and strong activation of NFκB by EPO receptor signaling. However, NFκB activation was unusual in the way that it did not require degradation of IκBα and activation was not prevented by inhibition of proteasomal function. This mode of NFκB activation is of considerable interest since it may reveal a tumor cell speciﬁc process but it requires further studies. Although treatment with EPO of EPO receptor expressing cancer cells did not change intrinsic radiosensitivity or sensitivity to chemotherapeutics, the authors “advocate a restricted use of erythropoietin to patients suffering from erythropoietin-receptor-expressing cancers” (Pajonk et al. 2004).

84

J. Fandrey

In contrast to the above mentioned work many other authors do not ﬁnd any inﬂuence of EPO on tumor cell proliferation in vitro. Rosti et al. (1993), tested the effect of recombinant human EPO on clonogenic growth and DNA synthesis in 10 different cell lines derived from hematologic malignancies or solid tumors. They included cell lines like K-562 and HEL which both express EPO receptors. Despite signiﬁcant EPO receptor expression, concentrations up to 10,000 mU/ml of EPO did not affect the clonogenic growth of any of these human tumor cell lines. Likewise, a study by Selzer et al. (2000) conﬁrmed that normal melanocytes and human melanoma cell lines which express the EPO receptor, were not affected in their cell growth by treatment with recombinant EPO. Interestingly, the authors of this study suggested that the expression of EPO receptor may be a progression marker of the more malignant phenotype, but nevertheless did not have functional inﬂuence on tumor growth. A study with a large number of 25 different benign and malignant cell lines by Westphal et al. (2002b) revealed that most of the cell lines expressed EPO receptor mRNA and proteins, but did not respond with an increase in proliferation to treatment with EPO. In this study several cellular response markers like 3H-thymidine uptake as a measure of DNA synthesis, Northern blot for c-fos expression (as an early growth response gene) and tyrosine kinase activity were measured. Since tumor cells that were EPO receptor positive did not respond with an increase in c-fos mRNA or stimulated tyrosine kinase activity, the authors suggested that the EPO signal was probably not transduced in these cells despite the detection of EPO receptor protein in whole cell lysates. The authors thus conclude that a deleterious effect on cancer patients may not be expected. This notion is supported by the ﬁnding that several tumor cell lines that expressed high levels of EPO receptor protein as detected by Western blot showed no detectable binding of 125 Iodine-labelled EPO (Sinclair et al. 2005). In another study Westphal et al. (2002a) measured soluble EPO receptors that were produced by tumor cells in culture from well established cell lines that were also used for most of the so far mentioned in vitro studies. Many of these cells secreted soluble EPO receptors into the culture supernatant. The meaning of soluble receptors will be discussed in further detail below. Finally, a very recent study by Liu et al. (2004) aimed at investigating the potential effects of GM-CSF and EPO on tumor cells and speculated whether growth factor treatment during chemotherapy may only be detrimental in those cancers in which the tumor cells express high levels of the speciﬁc receptor. For their study the authors selected seven cell lines including two RCC cell lines of which one was also used in the previously mentioned study (Westenfelder and Baranowski 2000). Neither EPO nor GM-CSF increased the rate of proliferation which was both measured by cell counting and bromo-2′-deoxyuridine (BrdU) incorporation. Interestingly, EPO caused an increase in mitogen activated protein (MAP) kinase activity in several cell

The role of erythropoietin receptor expression on tumor cells

85

lines but did not change the status of Bcl-2, which is part of the antiapoptotic pathway. Indeed, although MAP kinase activation is part of the intracellular signaling of the EPO receptor (Klingmuller 1997), Kolonics et al. (2001) have suggested that the MAP kinase pathway may not be the principal mechanism of EPO receptor activity. Consequently, any adverse effects associated with increased MAP kinase activity which result in reduced chemotherapeutic drug sensitivity may not be an issue of EPO treated cells (Liu et al. 2004). Finally, the authors conclude that EPO therapy may be even safe in renal cancers and other tumors that show high EPO receptor levels.

EPO receptors are found on many cells It appears to be generally accepted now that EPO receptor expression is not only found in erythroid progenitor cells, but also in many nonmalignant nonerythroid cells like endothelial cells, neuronal cells, myoblasts and megakaryocytes (Farrell and Lee 2004). Expression of erythropoietin in neural tissue appears to be of considerable importance and is also of interest with respect to EPO receptor activation. More and more evidence accumulates that EPO in the brain is neuroprotective and even neuroregenerative. These effects have recently been extensively reviewed and the reader is referred to this work (Jelkmann 2005). Physiologically EPO and EPO receptors are expressed in the human brain, even very early during embryonic development (Marti et al. 1996; Juul et al. 1998; Dame et al. 2000). Recombinant EPO has been found to actively cross the blood brain barrier in animals, although much higher doses than conventionally used for the treatment of anemia were required (Brines et al. 2000). This has been conﬁrmed for humans, where doses of 5,000 units/kg were required to increase erythropoietin concentration in the cerebral spinal ﬂuid after intravenous administration (Jumbe 2002). Still systemic erythropoietin administration has been found to reduce injury from brain ischemia both in an experimental animal model (Brines et al. 2000; Siren et al. 2001) and in patients at doses similar to those used in clinical practice (Ehrenreich et al. 2002). Other beneﬁcial effects of erythropoietin on neuronal cells include an attenuated brain injury after different forms of trauma and/or toxic effects of glutamate, a representative of the excitotoxicity found in many forms of brain injury (Morishita et al. 1997; Jelkmann 2005). The role for erythropoietin in the brain is underlined by observations of mouse embryos with targeted disruption of the EPO and the EPO receptor gene. Although these animals die from severe anemia in mid gestation, an increased rate of apoptosis in the brain of these mice has been reported (Wu et al. 1995; Yu et al. 2002). In contrast, however, mice that exclusively express the EPO receptor in hematopoietic tissues using GATA-1 directed erythroidspeciﬁc expression show no apparent neurological deﬁcits (Suzuki et al.

86

J. Fandrey

2002). Thus, it is conceivable that EPO receptor signaling plays an important role in limiting the damage incurred by various neurological stresses. These protective effects could depend on different classes of EPO receptors and on intracellular signaling pathways that are different from those known in erythroid progenitor cells (Masuda et al. 1993). In this respect the work of Digicaylioglu and Lipton (2001) is of great interest. The authors reported the activation of NFκB following EPO treatment which caused a delay in apoptosis of neuronal cells induced by N-methyl-D-aspartate or nitric oxide. Very recently, it was shown that carbamylated EPO does not stimulate erythropoiesis but is tissue-protective (Leist et al. 2004). Detailed analysis based on the assumption that EPO receptor signalling is different from the known pathways revealed the association of the EPO receptor with the common β receptor (βcR) in nonhematopoietic cells (Brines et al. 2004). βcR acts as the signal transducing subunit for several other cytokine receptors like GM-CSF, IL-3 and IL-5. It remains open whether this “new” signaling from the EPO receptor plays any role in tumor cells. Whereas the effect on neuronal cells is very welcome since EPO may indeed be neuroprotective and potentially have a role in neuronal differentiation, the detection of EPO receptors and potential anti-apoptotic effects of EPO on tumor cells have raised serious concerns. Although tumor cell lines derived from breast, liver or cervical carcinoma, neuroblastoma, glioblastoma and glioma were used for these studies they had been propagated through many generations in culture (Acs et al. 2001). Tissue biopsies from RCC (Westenfelder and Baranowski 2000), breast carcinoma (Acs et al. 2001) and melanoma (Selzer et al. 2000) have also been studied. Although results from experiments with these tissue samples in general did not reﬂect the ﬁndings from cells in culture, e.g. (Westenfelder and Baranowski 2000), it was inferred that administration of rhEPO to treat the anemia in cancer patients may positively inﬂuence the growth and/or survival of cancer cells or stimulate angiogenesis to promote tumor growth. It remains doubtful whether the receptors found on tumor cells in culture or in the tissue have any functional activity when exposed to clinically relevant concentrations of EPO. Nevertheless the theoretical consideration of such an adverse effect fell on fertile soil when two clinical trials with negative outcome in the EPOtreated group of patients were reported (Henke et al. 2003; Leyland-Jones 2003). Both studies have recently been critically evaluated and it was concluded that the negative outcome cannot be unambiguously attributed to the anemia-correcting treatment by rhEPO (Vaupel et al. 2005). Even a considerable number of in vivo – mostly animal – studies does not help to clarify the role of EPO on tumors: EPO was found to induce tumor regression (Rubins 1995; Gagic et al. 1997; Mittelman et al. 2001; Mittelman et al. 2004) or improve tumor sensitivity to radiation or chemotherapy by increasing the oxygenation of tumors (Silver and Piver 1999; Thews et al. 2001). In fact, the improvement of tumor oxygenation has been one

The role of erythropoietin receptor expression on tumor cells

87

reason to consider treatment with rhEPO not only to increase the quality of life by correcting the anemia, but also to increase the oxygen capacity of the blood and thus the oxygen supply of the tumor tissue. Human tumors and metastases commonly have an inadequate vessel supply and are therefore hypoxic (Vaupel et al. 1989). Tumor tissue hypoxia can lead to poorer outcomes and is generally considered as an indicator of a negative clinical prognosis (Hockel et al. 1993). In addition, hypoxia causes resistance to radiation treatment, and a higher hematocrit would thus be predicted to be beneﬁcial. Indeed, Stüben et al. (2001) conducted a study with six groups of nude mice which had glioblastoma cell line tumors implanted. Four groups of mice were anemic induced by total body irradiation 6 h before implantation of the tumor cells. Two of these groups were pretreated with EPO to prevent anemia. Subsequently, one half of all animals was irradiated with a single dose and tumor volume was assessed. None of the EPO treated animals showed any increase in tumor growth. On the contrary, anemic mice were increased in their sensitivity to radiotherapy when their anemia was corrected by EPO treatment. In another animal study Kelleher et al. (1996) implanted DS sarcomas onto the foot of rats and a tumor associated anemia was induced. Again, treatment with EPO improved the anemia and also increased the pO2 in the tumor tissue as measured by polarographic oxygen electrodes. However, EPO administration did not inﬂuence tumor growth or showed any other adverse effects. Obviously these animal experiments are closer to the clinical situation. Still, the in vitro data together with the above-mentioned two clinical studies with potential adverse effects have fueled the discussion on the potential tumor-proliferative effects of EPO. Only in vitro studies will allow to completely elucidate the signaling from the EPO receptors inside the cell. Methodological differences and, in part, over-interpretation of in vitro data are not helpful to deﬁne the potential risk of EPO treatment in cancer patients. Therefore, in vitro studies need to very carefully address at least the ﬁve following points to help clarifying the meaning of EPO receptors on tumor cells.

What does the detection of EPO receptors on tumor cell mean? Five points to consider Signal transduction from the EPO receptor to the nucleus The EPO receptor is a member of the type I cytokine receptor super family (Klingmuller 1997). Other members of this family are receptors for growth hormone, prolactin, G-CSF, GM-CSF thrombopoietin and several interleukins. Common features shared by these receptors in the extracellular

88

J. Fandrey

domain are four spaced cysteins near the N-terminus and a Trp-Ser-X-TrpSer (WSXWS) motive located proximal to the cell membrane (Klingmuller 1997). For the EPO receptor in particular, crystallographic analyses have shown that two EPO receptor monomers are bridged by EPO as a symmetrical T-shaped dimer (Livnah et al. 1996). These analyses and data from Syed et al. 1998 clearly indicate that EPO receptor activation depends upon ligand binding to the receptor dimer. This is in line with previous observations that certain receptor dimerizing antibodies can activate the EPO receptor in EPO responsive cells or that disulﬁde bridging of β-chains due to a mutation at position 129 of the EPO receptor (R129C mutation) leads to constitutive proliferation in BAF/3 cells (Yoshimura et al. 1990; Elliott et al. 1996). However, when EPO receptors were successfully crystallized, it turned out that the receptors are found as preformed dimers caused by interaction between two ﬁbrinonectin III-like subdomains in the extracellular domain of each monomer in the absence of their ligand EPO. However, unliganded dimers have a distance of 97Å whereas binding of the ligand reduced the spacing between the 2 receptor monomers to 39Å. Thus it has been concluded that receptor activation is driven by a ligand induced conformational change which then causes Janus kinase 2 (Jak2) activation (see below). This model was nicely conﬁrmed by ﬂuorescence complementation measurements which clearly indicated that unliganded EPO receptor is preformed in the membrane and that these dimers are brought into functional proximity by EPO or EPO mimetic peptides (Remy et al. 1999). Importantly, the cytoplasmic domain of the EPO receptor lacks intrinsic signaling activity. This requires the recruitment of cytoplasmic kinases, in particular Jak2 to promote signal transduction. Upon EPO binding the conformational switch facilitates the binding of Jak2. A continuous stretch of residues in the membrane proximal region of the EPO receptor is required to activate intracellular signal transduction by recruitment of Jak2. Interestingly, the same juxta-membrane motive is critical for the transport of the EPO receptor from the endoplasmic reticulum to the cell surface (Huang et al. 2001). Activated Jak2 is then involved in activation of signal transducer and activator of transcription protein-5 (STAT5) (Ketteler et al. 2002). Activation of STATs promotes signal transmission from the cell surface to the nucleus to induce changes in the expression pattern of EPO responsive genes. One of the key targets are the antiapoptotic proteins Bcl-2 and Bcl-XL. This clearly illustrates one primary action of EPO in erythroid cells which is to inhibit programmed cell death (Koury and Bondurant 1990). In contrast, the signaling of EPO to act as a mitogen has been more difﬁcult to deﬁne. The addition of EPO to early erythroid progenitor cells supported their proliferation and was shown to activate several known effectors of mitogenesis like c-myc, mitogen activated protein (MAP) kinases and phosphoinositol-3 kinase (PI-3K). In addition to these receptor associated effectors at least 19 additional signaling factors have been described to bind to the activated EPO

The role of erythropoietin receptor expression on tumor cells

89

receptor complex (for review Wojchowski et al. 1999). Among them are further activating but also negative inhibitory molecules including suppressors of cytokine signaling (SOCS) and tyrosine phosphatases that terminate EPO signaling. Proliferation of immature red blood cells and their differentiation into mature erythrocytes are mediated by binding of erythropoietin to its cell surface receptors (D’Andrea et al. 1990). Ligand binding to the EPO receptor on the cell surface was found to accelerate the rate of receptor internalization from a half-life of 3 hours in the absence of EPO to 15 minutes in its presence (Sawyer and Hankins 1993). Now, obviously the binding of the hormone EPO takes place at the surface but EPO receptor internalization is required for EPO-mediated proliferation of erythroid progenitor cells. Truncated receptors lacking amino acid 268 to 276 were defective in their capacity for endocytosis of the EPO/EPO receptor complex and did not promote cellular growth. Thus, surface localization of the EPO receptor and the 9 amino acid stretch, the internalization motif, of the EPO receptor responsible for internalization upon binding of EPO are required for proper signaling (Flint-Ashtamker et al. 2002). Further intensive studies to determine the role of the internalization motif within the cytosolic part of the EPO receptor revealed that it is not only important for internalization but also for association of Jak2 and thus intracellular signal transduction since tyrosine phosphorylation mediated by Jak2 was abolished in mutants lacking the nine amino acids (FlintAshtamker et al. 2002). Importantly endocytosis of EPO via the EPO receptor is essentially similar in different cellular systems (Levin et al. 1998; Beckman et al. 1999). This supports the notion that the process of internalization of EPO via the receptor relies mainly on intrinsic properties of the receptor molecule and would not depend on the cell type in which the receptor is expressed. If de novo synthesis of EPO receptor is blocked by the translational inhibitor cycloheximide, EPO receptors decline by 50–60% after 4 hours of cycloheximide treatment. The absence of the internalization motif abrogated the decline. All these data were obtained in the absence of ligand indicating the considerable amount of ligand-independent EPO receptor turnover via internalizations is present at least in hematopoietic cells. In addition, particular events in the activation of proliferative signals from the EPO receptor were deﬁned. The above mentioned anti-apoptotic effects of EPO on neuronal cells appear to involve activation of NFκB through cross-talk between the Jak2 and the NFκB pathways (Digicaylioglu and Lipton 2001). This appears different from the regular signaling in erythroid cells but the signaling chain from the receptor to the target genes was successfully followed. This has not been the case for several in vitro studies with tumor cells in which activation of the EPO receptor was claimed. In most studies only parts of the signaling were studied, i.e. sometimes only tyrosine phosphorylation of the EPO recep-

90

J. Fandrey

tor protein. Interestingly, in a recent study using HeLa cells stably transfected with the EPO receptor activation of NFκB was suggested without phosphorylation and degradation of IκBα, i.e. an untypical activation mechanism of NFκB. If such “new” signaling pathway would exist in tumor cells a more complete dissection of the signaling cascade is required to draw any ﬁrm conclusions as to the potential impact of EPO on tumor cells in vivo. Moreover, the above-mentioned known components of the signaling cascade from hematopoietic progenitors should be scrutinized in tumor cells. It appears mandatory to run hematopoietic progenitors as positive controls in parallel to either conﬁrm activation of known pathways or to compare new intracellular signals with respect to receptor afﬁnity, signaling kinetics and the response of target genes. Of note, in view of the recent work on signaling in neuronal cells (Brines et al. 2004) one should be prepared to ﬁnd new modes of EPO receptor activation in tumor cells.

Soluble receptors and EPO receptor splice variance In addition to the classical membrane spanning EPO receptor with its intracytoplasmic region for recruitment of Jak2, splice variants resulting in soluble forms of the EPO receptor have been described (Yet and Jones 1993; Ku et al. 1996). Soluble forms of cytokine receptors or growth factor receptors are often expressed in cancer cells and have been implicated in various biological functions. On the one hand it is known that soluble receptors can antagonize the effect of membrane spanning isoforms, whereas in other cases they prolong the half-life of the respective ligand (Rose-John and Heinrich 1994; Heaney and Golde 1996). With respect to EPO it has been shown that ligand binding to soluble EPO receptors decreases receptor-mediated signal transduction (Shimizu et al. 1996). Therefore, a report on different receptor splice variants in human cancer cells potentially has a great impact (Arcasoy et al. 2003). The authors examined a considerable number of tumor cells and were able to detect full-length EPO receptor mRNA in lung, colon, breast, prostrate and ovary cancer cells. In addition to the full-length receptor which was also detected on the protein level, at least 5 isoforms resulting from different splicing of the seven introns of the EPO receptor gene were isolated. Interestingly, three of these isoforms resulted in truncated EPO receptor that still contained the trans-membrane domain and can be expected to be found on the cellular surface. The structures of the deduced EPO receptor proteins for the ﬁrst three isoforms had severe truncations deleting the membrane distal intracellular part of the receptor including the Jak2 docking site. Thus, these truncated forms are expected to lack functional activity. Isoform 3 of these truncated EPO receptors is identical to previously described isoform from erythroid progenitor cells that showed a dominant negative effect on EPO-induced differentiation and inhibition of apoptosis (Shimizu et al.

The role of erythropoietin receptor expression on tumor cells

91

1996). Thus, although only isoform 3 truncation of the cytoplasmic domain has been clearly associated with an inhibitory function, truncated EPO receptors isoforms 1 and 2 may likewise negatively modulate EPO-mediated signaling in cancer cells. It is noteworthy that in addition to the detection in permanent cancer cell lines these isoforms were also isolated from primary cancer tissue of breast, colon, lung and ovary carcinomas. Two forms, isoforms 4 and 5, that lack the transmembrane domain are most likely soluble forms and the authors succeeded in detecting a 26 kD immunoreactive protein in the supernatant of prostate and breast cancer cells by Western blot using an antibody that recognizes the extracellular domain (Arcasoy et al. 2003). These proteins were not detected in the supernatants of Chinese hamster ovary cells or in non-conditioned medium but corresponds to data from Westphal et al. (2002b) who found soluble EPO receptors in the supernatant of many cancer cell lines. As pointed out above, soluble receptors may be ambivalent with respect to the effect on EPO signaling, but the competition with membrane-bound receptors for EPO has been shown for hematopoietic cells (Yet and Jones 1993). Moreover, in some of the splice variants that still contain the transmembrane part of the EPO receptor, splicing also affects the immediate cytoplasmic part. This stretch of amino acids is close to the membrane spanning domain and overlaps with amino acids 268–276 that are important for internalization (see above). Therefore, some of the splice variants may be defective in internalization and thus not convey a proliferative signal. Tumor cells secrete soluble receptors and have different splice variants expressed. The issue of EPO receptor isoforms in human cancers has not been thoroughly studied. Soluble receptors in the tissue are hard to identify but may well signiﬁcantly affect the effects of any EPO, endogenous or exogenous.

Cellular localization of the EPO receptor Since the EPO receptor lacks intrinsic kinase activity, recruitment of Jak2 is essential to transmit the signal from the receptor to the nucleus. The precise orientation of critical residues in the juxta membrane motif is essential for Jak2 activation and subsequent signaling upon binding of EPO to its receptor. It is of note that the membrane proximal domain of the EPO receptor that mediates binding of Jak2 is also responsible for transport of the EPO receptor from the endoplasmic reticulum to the cell surface. To visualize intracellular trafﬁcking of the receptor Ketteler et al. (2002) generated fusion proteins of green ﬂuorescent protein (GFP) and different parts of the EPO receptor. After transient transfection of both Ba/F3 cells (serving as a model for erythroid progenitors (D’Andrea et al. 1991)) and HEK293T-cells (as a permanent tumor cell line) the authors observed different cellular localiza-

92

J. Fandrey

tion of the fusion proteins in the membrane or within the cell. Interestingly, fusion protein constructs that resided within the endoplasmic reticulum and did not reach the cell surface, were unable to trigger the activation of signaling. Signaling was assessed by detecting association of Jak2, subsequent tyrosine phosphorylation of the EPO receptor and association of STAT5, SHP-1 and p85 (as part of the PI3-kinase) with the activated EPO receptor (Ketteler et al. 2002). In addition, proliferation and differentiation of Ba/F3 cells was determined. All these clear signs of receptor activation were absent with the above-mentioned fusion proteins that stayed in the endoplasmic reticulum. Other fusion proteins showed punctuated cellular ﬂuorescence, but were also localized in the membrane and able to bind EPO as shown by 125 iodine-labelled EPO bound to the surface. Flow cytometry using EPO receptor antibodies that detect extracellular epitopes of the EPO receptor supported the notion that – depending on the construct – inactive receptors were not transported to the membrane and thus not detected by ﬂow cytometry. Collectively, the authors concluded that the extracellular, transmembrane and membrane-proximal segments of the cytoplasmic domain form an entity of which the precise orientation is essential for initiation of signaling of the EPO receptor. Although the cytoplasmic domain had some ﬂexibility in adopting an activated conﬁrmation of the receptor, fusion proteins that were retained in the endoplasmic reticulum were signaling-incompetent. On the other hand, some cell surface expressed receptors could bind EPO, but were signaling-incompetent due to disruption of the intracellular parts of the receptor. Finally, as a positive control, however, and to exclude any unspeciﬁc effects through the presence of the GFP within the fusion proteins some of the constructs were detected on the cell surface, bound EPO and initiated signaling with subsequent proliferation, although the receptor had been fused to GFP. These data are of critical importance with respect to the immunohistochemistry of EPO receptor protein that has been reported in tumor cells so far. Several studies (Acs et al. 2002; Arcasoy et al. 2002) have included immunohistochemical staining for the EPO receptor in which the protein was clearly found in the cytoplasm, partly surrounding the nucleus and most likely in the endoplasmic reticulum. From the study of Ketteler et al. (2002) and also earlier work from Yoshimura et al. (1990) it is evident that EPO receptors retained in the endoplasmic reticulum are unable to interact with the ligand and thus transmit no signal to the nucleus. In view of the recent reports of the questionable speciﬁcity of the commercially available antibodies against the EPO receptor (Busse et al. 2005) and the fact that despite the detection of EPO receptor protein by Western blot no binding of 125Iodinelabelled EPO was observed (Sinclair et al. 2005), it is of utmost importance to show functionality of EPO receptor signaling. The sole detection of EPO receptor mRNA by PCR and immunohistochemical staining for the EPO receptor protein by the currently available antibodies do not justify the conclusion of potentially proliferative effects of EPO on tumor cells.

The role of erythropoietin receptor expression on tumor cells

93

Promiscuous activation of cytokine type 1 family receptors by high concentrations of EPO Extremely high concentrations of recombinant EPO have been used for in vitro studies with different tumor cell lines. Some of the investigators used 10,000–250,000 mU/ml (Acs et al. 2001, 2003) to stimulate their tumor cells in vitro while comparable work on EPO receptor signaling on hematopoietic cells provides clear results with EPO concentrations between 100 and 1,000 mU/ml (Ruan et al. 2004). This is several orders of magnitude higher than would ever be achieved in patients by exogenous EPO even after intravenous injection. These high concentrations may be problematic with respect to speciﬁcity of the EPO effects. In earlier studies to investigate the putative effect of EPO on smooth-muscle-like cells with respect to the observed increase in blood pressure in patients with renal anemia, the effects of EPO on changes in intracellular calcium concentration and subsequent cell contraction were determined (Morakkabati et al. 1996). Upon treatment with EPO between 2,000 and 10,000 mU/ml the cells showed clear shape changes as a result of contraction and actomyosin activation after an increase in their cellular calcium concentration. However, despite highly sensitive PCR methods no EPO receptor mRNA was detected. Thus, despite the reproducible effects on cell calcium and contraction no expression of the receptor for EPO was found and it remained open whether EPO concentrations in the range of several thousand mU/ml as used in this study caused promiscuous activation of other cellular receptors of the cytokine receptor I family (Morakkabati et al. 1996).

Local concentrations of EPO in tumors In addition to the just mentioned high concentrations needed for some in vitro studies it appears mandatory to consider which concentrations are likely to be achieved in the tumor when EPO is given to cancer patients. Often large parts of solid tumors are mal-perfused and thus hypoxic as pointed out above (Vaupel et al. 1989). Consequently, EPO concentrations that can be reached in the tissue after injection, will be extremely low, deﬁnitely several orders of magnitude lower than those used in all in vitro studies. Berdel et al. (1991) used serum concentrations considered to be in the therapeutic range when patients receive EPO. None of these concentrations stimulated growth of tumor cells. Subcutaneous injection of EPO which has been recently favored in renal patients will even show more sustained plasma levels of EPO but much lower peak concentrations. Figure 1 summarizes these ﬁve issues as they were raised. The situation regarding local EPO concentration within the (tumor) tissue may be different when EPO is produced by and/or within the tumor.

94

J. Fandrey

2

soluble EPO-R

4

EPO

5

local EPO concentration

EPO-R P-

non EPO-R

-P Jak2 1

Ca2+

?

?

STAT5

NFkB

3 ER

nuclear signals

Fig. 1. The ﬁgure summarizes the open questions with respect to EPO and EPO receptor signaling in tumor cells. 1 EPO signaling depends on recruitment of several kinases and subsequent signal transduction to the nucleus. In addition, however, other signalling pathways like increase in intracellular calcium or activation of NFκB in neuronal cells have been described (see text for details). It is still unresolved whether tumor cell speciﬁc signaling pathways exist that have not been described in erythroid progenitor or neuronal cells so far. 2 Soluble EPO receptors may well antagonize membrane spanning receptors, as has already been described for hematopoietic cells (Yet and Jones 1993). On the other hand, EPO bound to soluble EPO receptors may have a longer half-life in the serum or even tissue. 3 EPO receptors retained in the ER are unable to interact with the ligand and are thus non-functional (Yoshimura et al. 1990; Ketteler et al. 2002). 4 EPO concentrations to stimulate tumor cell responses are several orders of magnitude higher than concentrations to be expected in the serum. Considering a mal-perfused tissue concentrations of EPO that can be expected in close proximity to tumor cells may be very low. Thus, the concentration of EPO within the tumor needs to be critically evaluated. 5 Non-speciﬁc effects of high concentrations of EPO in the absence of EPO receptor mRNA have been found in smooth muscle like cells (Morakkabati et al. 1996). Thus, unspeciﬁc effects, potentially promiscuous activation of other cellular receptors by extremely high EPO concentrations used in vitro have to be considered

The expression of EPO receptor and EPO mRNA in tumors of the female reproductive tract has made some authors to propose an autocrine or paracrine role of EPO supporting tumor growth (Arcasoy et al. 2002; Acs et al. 2003). Two further studies by Yasuda et al. (2002, 2003) reported that antagonists of EPO can reduce tumor growth of EPO and EPO receptor expressing tumors of the female reproductive tract. EPO alone will not cause tumors – at least not in animals – because several transgenic mice that over-

The role of erythropoietin receptor expression on tumor cells

95

express EPO in an oxygen independent fashion suffer from polycythemia due to the continuous stimulation of erythropoiesis. However, these animals do not develop erythroleukemia (Wagner et al. 2001; Madan et al. 2003). It will therefore be critical to determine local concentrations of EPO and also consider the potential occurrence of soluble receptors (see above). Moreover, if EPO synthesized within the tumor reaches a sufﬁcient concentration to play an autocrine role in tumor development and progression it is questionable whether a small increase that may arise from treatment will have any effect on the tumor cell. This does not argue against a potential role for EPO expressed in the tumor which is worth to be carefully studied. But this is an issue that is completely different from the questions whether exogenous EPO would cause any negative effects with respect to tumor proliferation.

Concluding remarks The concern that the hematopoietic growth factor and hormone EPO may affect tumor cell growth has to be taken seriously. However, until today solid evidence that EPO promotes tumor cell proliferation or delays tumor cell death is missing. Even determination of EPO receptor mRNA expression on tumor cells does not ensure that signals from this receptor are transduced to the nucleus to affect cell growth and survival. Clearly in vitro studies are needed to fully deﬁne the signaling pathways – if there are any – connected to EPO receptors in tumor cells. Moreover, however, pre-clinical in vivo studies using animal models with EPO receptor positive tumors are needed in which treatment with EPO is critically evaluated with respect to changes in proliferation, apoptosis or even changes in the phenotype of the tumors. In addition, large databases are available with millions of patients treated successfully with recombinant EPO due to end stage renal disease. These databases are an excellent source for retrospective analyses to detect potential adverse effects on tumor development or treatment. Apart from the discussion whether exogenous EPO can stimulate tumor growth directly Yasuda et al. (2002) have suggested that tumor angiogenesis may be affected by EPO. They reported that the capillary endothelium within tumors showed EPO receptor immunoreactivity and that injection of a monoclonal antibody against EPO or soluble forms of the EPO receptor into tumors reduced capillaries and caused tumor destruction in a dosedependent manner. If local intra-tumor levels of EPO are sufﬁciently high to activate EPO receptors in endothelial cells this might in fact be a new therapeutic approach supporting the efforts of anti-angiogenic therapies. However, it should again be emphasized that these concentration will most likely not achieved from outside in a mal-perfused tumor with therapeutic doses. Intracellular signaling in non-erythroid cells obviously differs from the regular pathways that have been thoroughly studied in erythroid progenitor

96

J. Fandrey

cells (Digicaylioglu and Lipton 2001; Brines et al. 2004). There is still lack of convincing evidence that classical or novel signaling of the EPO receptor is fully active in tumor cells. A very recent survey on different cellular responses towards EPO may be helpful in this respect: Korbel et al. (2005) using a proteomic approach showed a full array of signaling proteins in EPO responsive cells. Proﬁling the cellular response in erythroid progenitors could provide a “standard” of signaling pathways to be expected in an active state. It obviously remains to be studied whether most or even a subset of these signaling molecules are activated in tumor cells in vitro and also in vivo. Treatment of anemia in cancer patients by recombinant EPO has been for the beneﬁt of many patients. Great care should be taken to avoid overtreatment, i.e. EPO treatment currently should be limited to alleviating the anemia as suggested in a recent review (Vaupel et al. 2005). Based on the in vitro and preclinical data, however, it does not appear to be justiﬁed to refrain from treating cancer associated anemia by recombinant EPO according to the clinical practice guidelines for the use of EPO developed by the American Society of Clinical Oncology and the American Society of Hematology (Rizzo et al. 2001).

Acknowledgements I gratefully acknowledge the excellent secretarial assistance by Ms. Gundula Endemann.

References 1. Acs G, Zhang PJ, Rebbeck TR, Acs P, Verma A (2002) Immunohistochemical expression of erythropoietin and erythropoietin receptor in breast carcinoma. Cancer 95: 969–981 2. Acs G, Acs P, Beckwith SM, Pitts RL, Clements E, Wong K, Verma A (2001) Erythropoietin and erythropoietin receptor expression in human cancer. Cancer Res 61: 3561–3565 3. Acs G, Zhang PJ, McGrath CM, Acs P, McBroom J, Mohyeldin A, Liu S, Lu H, Verma A (2003) Hypoxia-inducible erythropoietin signaling in squamous dysplasia and squamous cell carcinoma of the uterine cervix and its potential role in cervical carcinogenesis and tumor progression. Am J Pathol 162: 1789 4. Arcasoy MO, Amin K, Karayal AF, Chou SC, Raleigh JA, Varia MA, Haroon ZA (2002) Functional signiﬁcance of erythropoietin receptor expression in breast cancer. Lab Invest 82: 911–918 5. Arcasoy MO, Jiang X, Haroon ZA (2003) Expression of erythropoietin receptor splice variants in human cancer. Biochem Biophys Res Commun 307: 999–1007 6. Beckman DL, Lin LL, Quinones ME, Longmore GD (1999) Activation of the Erythropoietin receptor is not required for internalization of bound erythropoietin. Blood 94: 2667–2675

The role of erythropoietin receptor expression on tumor cells

97

7. Berdel WE, Oberberg D, Reuﬁ B, Thiel E (1991) Studies on the role of recombinant human erythropoietin in the growth regulation of human nonhematopoietic tumor cells in vitro. Ann Hematol 63: 5–8 8. Brines M, Grasso G, Fiordaliso F, Sfacteria A, Ghezzi P, Fratelli M, Latini R, Xie QW, Smart J, Su-Rick CJ, Pobre E, Diaz D, Gomez D, Hand C, Coleman T, Cerami A (2004) Erythropoietin mediates tissue protection through an erythropoietin and common β-subunit heteroreceptor. PNAS 101: 14907–14912 9. Brines ML, Ghezzi P, Keenan S, Agnello D, de Lanerolle NC, Cerami C, Itri LM, Cerami A (2000) From the cover: erythropoietin crosses the blood-brain barrier to protect against experimental brain injury. PNAS 97: 10526–10531 10. Busse L, Sinclair A, Rogers N, Sarosi I, Van G, Elliott S (2005) Is EPO receptor over-expressed in human tumor cells? AACR 96th Annual Meeting Abstract No:4562 11. Caro JJ, Salas M, Ward A, Goss G (2001) Anemia as an independent prognostic factor for survival in patients with cancer: a systemic, quantitative review. Cancer 91: 2214–2221 12. D’Andrea A, Fasman G, Wong G, Lodish H (1990) Erythropoietin receptor: cloning strategy and structural features. Int J Cell Cloning 8 [Suppl 1]: 173–80.: 173–180 13. D’Andrea AD, Yoshimura A, Youssouﬁan H, Zon LI, Koo JW, Lodish HF (1991) The cytoplasmic region of the erythropoietin receptor contains nonoverlapping positive and negative growth-regulatory domains. Mol Cell Biol 11: 1980–1987 14. Dame C, Bartmann P, Wolber E, Fahnenstich H, Hofmann D, Fandrey J (2000) Erythropoietin gene expression in different areas of the developing human central nervous system. Brain Res Dev Brain Res 125: 69–74 15. Digicaylioglu M, Lipton SA (2001) Erythropoietin-mediated neuroprotection involves cross-talk between Jak2 and NF-[kappa]B signalling cascades. Nature 412: 641–647 16. Ehrenreich H, Hasselblatt M, Dembowski C, Cepek L, Lewczuk P, Stiefel M, Rustenbeck HH, Breiter N, Jacob S, Knerlich F, Bohn M, Poser W, Ruther E, Kochen M, Gefeller O, Gleiter C, Wessel TC, De Ryck M, Itri L, Prange H, Cerami A, Brines M, Siren AL (2002) Erythropoietin therapy for acute stroke is both safe and beneﬁcial. Mol Med 8: 495–505 17. Elliott S, Lorenzini T, Yanagihara D, Chang D, Elliott G (1996) Activation of the erythropoietin (EPO) receptor by bivalent anti-EPO receptor antibodies. J Biol Chem 271: 24691–24697 18. Farrell F, Lee A (2004) The erythropoietin receptor and its expression in tumor cells and other tissues. Oncologist 9: 18–30 19. FDA ODAC on EPO (2004) http://www.trends-in-medicine.com/May2004/ FDAepo054qp.pdf 20. Flint-Ashtamker G, Eisen-Lev R, Cohen J, Jun-shen Huang L, Neumann D (2002) Amino acid residues 268–276 of the erythropoietin receptor contain an endocytosis motif and are required for erythropoietin-mediated proliferation. FEBS Letters 518: 189–194 21. Gagic K, Campagnaro E, Laborde CJ, Edavettal M, Levine EA, Potter BJ, Racey Burns LA, Burns AH (1997) The effect of clenbuterol and recombinant erythropoietin on tumor growth and the anemia caused by the Walker 256 carcinosarcoma. Life Sciences 61: 2475–2484

98

J. Fandrey

22. Gross AJ, Wolff M, Fandrey J, Miersch WD, Dieckmann KP, Jelkmann W (1994) Prevalence of paraneoplastic erythropoietin production by renal cell carcinomas. Clin Investig 72: 337–340 23. Heaney ML, Golde DW (1996) Soluble cytokine receptors. Blood 87: 847–857 24. Henke M, Laszig R, Rube C, Schafer U, Haase KD, Schilcher B, Mose S, Beer KT, Burger U, Dougherty C, Frommhold H (2003) Erythropoietin to treat head and neck cancer patients with anaemia undergoing radiotherapy: randomised, double-blind, placebo-controlled trial. The Lancet 362: 1255–1260 25. Hockel M, Knoop C, Schlenger K, Vorndran B, Baussmann E, Mitze M, Knapstein PG, Vaupel P (1993) Intratumoral pO2 predicts survival in advanced cancer of the uterine cervix. Radiother Oncol 26: 45–50 26. Huang LJ, Constantinescu SN, Lodish HF (2001) The N-terminal domain of Janus Kinase 2 is required for Golgi processing and cell surface expression of erythropoietin receptor. Molecular Cell 8: 1327–1338 27. Jelkmann W (1998) Proinﬂammatory cytokines lowering erythropoietin production. J Interferon Cytokine Res 18: 555–559 28. Jelkmann W (2005) Effects of erythropoietin on brain function. Curr Pharm Biotechnol 6: 65–79 29. Jelkmann W, Pagel H, Wolff M, Fandrey J (1992) Monokines inhibiting erythropoietin production in human hepatoma cultures and in isolated perfused rat kidneys. Life Sci 50: 301–308 30. Jumbe NL (2002) Erythropoietic agents as neurotherapeutic agents: what barriers exist? Oncology (Huntingt) 16: 91–107 31. Juul SE, Anderson DK, Li Y, Christensen RD (1998) Erythropoietin and erythropoietin receptor in the developing human central nervous system. Pediatr Res 43: 40–49 32. Kelleher DK, Mattheinsen U, Thews O, Vaupel P (1996) Blood ﬂow, oxygenation, and bioenergetic status of tumors after erythropoietin treatment in normal and anemic rats. Cancer Res 56: 4728–4734 33. Ketteler R, Heinrich AC, Offe JK, Becker V, Cohen J, Neumann D, Klingmuller U (2002) A functional green ﬂuorescent protein-tagged erythropoietin receptor despite physical separation of JAK2 binding site and tyrosine residues. J Biol Chem 277: 26547–26552 34. Klingmuller U (1997) The role of tyrosine phosphorylation in proliferation and maturation of erythroid progenitor cells – signals emanating from the erythropoietin receptor. Eur J Biochem 249: 637–647 35. Kolonics A, Apati A, Janossy J, Brozik A, Gati R, Schaefer A, Magocsi M (2001) Activation of Raf/ERK1/2 MAP kinase pathway is involved in GM-CSF-induced proliferation and survival but not in erythropoietin-induced differentiation of TF1 cells. Cellular Signalling 13: 743–754 36. Korbel S, Buchse T, Prietzsch H, Sasse T, Schumann M, Krause E, Brock J, Bittorf T (2005) Phosphoprotein proﬁling of erythropoietin receptor-dependent pathways using different proteomic strategies. Proteomics 5: 91–100 37. Koury MJ, Bondurant MC (1990) Erythropoietin retards DNA breakdown and prevents programmed death in erythroid progenitor cells. Science 248: 378–381 38. Ku H, Hirayama F, Kato T, Miyazaki H, Aritomi M, Ota Y, D’Andrea AD, Lyman SD, Ogawa M (1996) Soluble thrombopoietin receptor (Mpl) and granulocyte colony-stimulating factor receptor directly stimulate proliferation of primitive

The role of erythropoietin receptor expression on tumor cells

39.

40.

41. 42.

43.

44.

45.

46.

47.

48.

49.

50.

51. 52.

99

hematopoietic progenitors of mice in synergy with steel factor or the ligand for Flt3/Flk2. Blood 88: 4124–4131 Leist M, Ghezzi P, Grasso G, Bianchi R, Villa P, Fratelli M, Savino C, Bianchi M, Nielsen J, Gerwien J, Kallunki P, Larsen AK, Helboe L, Christensen S, Pedersen LO, Nielsen M, Torup L, Sager T, Sfacteria A, Erbayraktar S, Erbayraktar Z, Gokmen N, Yilmaz O, Cerami-Hand C, Xie QW, Coleman T, Cerami A, Brines M (2004) Derivatives of erythropoietin that are tissue protective but not erythropoietic. Science 305: 239–242 Levin I, Cohen J, Supino-Rosin L, Yoshimura A, Watowich SS, Neumann D (1998) Identiﬁcation of a cytoplasmic motif in the erythropoietin receptor required for receptor internalization. FEBS Letters 427: 164–170 Leyland-Jones B (2003) Breast cancer trial with erythropoietin terminated unexpectedly. The Lancet Oncology 4: 459–460 Liu WM, Powles T, Shamash J, Propper D, Oliver T, Joel S (2004) Effect of haemopoietic growth factors on cancer cell lines and their role in chemosensitivity. Oncogene 23: 981–990 Livnah O, Stura EA, Johnson DL, Middleton SA, Mulcahy LS, Wrighton NC, Dower WJ, Jolliffe LK, Wilson IA (1996) Functional mimicry of a protein hormone by a peptide agonist: the EPO receptor complex at 2.8 A. Science 273: 464–471 Madan A, Lin C, Wang Z, Curtin PT (2003) Autocrine stimulation by erythropoietin in transgenic mice results in erythroid proliferation without neoplastic transformation. Blood Cells Mol Dis 30: 82–89 Marti HH, Wenger RH, Rivas LA, Straumann U, Digicaylioglu M, Henn V, Yonekawa Y, Bauer C, Gassmann M (1996) Erythropoietin gene expression in human, monkey and murine brain. Eur J Neurosci 8: 666–676 Masuda S, Nagao M, Takahata K, Konishi Y, Gallyas F Jr, Tabira T, Sasaki R (1993) Functional erythropoietin receptor of the cells with neural characteristics. Comparison with receptor properties of erythroid cells. J Biol Chem 268: 11208–11216 Mittelman M, Neumann D, Peled A, Kanter P, Haran-Ghera N (2001) Erythropoietin induces tumor regression and antitumor immune responses in murine myeloma models. PNAS 98: 5181–5186 Mittelman M, Zeidman A, Kanter P, Katz O, Oster H, Rund D, Neumann D (2004) Erythropoietin has an anti-myeloma effect – a hypothesis based on a clinical observation supported by animal studies. Eur J Haematol 72: 155–165 Morakkabati N, Gollnick F, Meyer R, Fandrey J, Jelkmann W (1996) Erythropoietin induces Ca2+ mobilization and contraction in rat mesangial and aortic smooth muscle cultures. Exp Hematol 24: 392–397 Morishita E, Masuda S, Nagao M, Yasuda Y, Sasaki R (1997) Erythropoietin receptor is expressed in rat hippocampal and cerebral cortical neurons, and erythropoietin prevents in vitro glutamate-induced neuronal death. Neuroscience 76: 105–116 Pajonk F, Weil A, Sommer A, Suwinski R, Henke M (2004) The erythropoietinreceptor pathway modulates survival of cancer cells. Oncogene 23: 8987–8991 Philo JS, Aoki KH, Arakawa T, Narhi LO, Wen J (1996) Dimerization of the extracellular domain of the erythropoietin (EPO) receptor by EPO: one highafﬁnity and one low-afﬁnity interaction. Biochemistry 35: 1681–1691

100

J. Fandrey

53. Remy I, Wilson IA, Michnick SW (1999) Erythropoietin receptor activation by a ligand-induced conformation change. Science 283: 990–993 54. Rizzo JD, Seidenfeld J, Piper M, Aronson N, Lichtin A, Littlewood TJ (2001) Erythropoietin: a paradigm for the development of practice guidelines. Hematology (Am Soc Hematol Educ Program): 10–30: 10–30 55. Rose-John S, Heinrich PC (1994) Soluble receptors for cytokines and growth factors: Generation and biological function. Biochem J 300: 281–290 56. Rosti V, Pedrazzoli P, Ponchio L, Zibera C, Novella A, Lucotti C, Della Cuna GR, Cazzola M (1993) Effect of recombinant human erythropoietin on hematopoietic and non-hematopoietic malignant cell growth in vitro. Haematologica 78: 208–212 57. Ruan W, Becker V, Klingmuller U, Langosch D (2004) The interface between selfassembling erythropoietin receptor transmembrane segments corresponds to a membrane-spanning Leucine Zipper. J Biol Chem 279: 3273–3279 58. Rubins J (1995) Metastatic renal cell carcinoma: response to treatment with human recombinant erythropoietin. Ann Intern Med 122: 676–677 59. Sawyer ST, Hankins WD (1993) The functional form of the erythropoietin receptor is a 78-kDa protein: correlation with cell surface expression, endocytosis, and phosphorylation. PNAS 90: 6849–6853 60. Selzer E, Wacheck V, Kodym R, Schlagbauer-Wadl H, Schlegel W, Pehamberger H, Jansen B (2000) Erythropoietin receptor expression in human melanoma cells. Melanoma Res 10: 421–426 61. Shimizu R, Komatsu N, Nakamura Y, Nakauchi H, Nakabeppu Y, Miura Y (1996) Role ofc-junin the inhibition of erythropoietin receptor-mediated apoptosis. Biochem Biophys Res Commun 222: 1–6 62. Shouval D, Anton M, Galun E, Sherwood JB (1988) Erythropoietin-induced polycythemia in athymic mice following transplantation of a human renal carcinoma cell line. Cancer Res 48: 3430–3434 63. Silver DF, Piver MS (1999) Effects of recombinant human erythropoietin on the antitumor effect of cisplatin in SCID mice bearing human ovarian cancer: a possible oxygen effect. Gynecol Oncol 73: 280–284 64. Sinclair A, Busse L, Rogers N, Arnold G, Hoey T, Sarosi I, Elliott S (2005) EPO receptor transcription is not elevated nor predictive of surface expression in human tumor cells. AACR 96th Annual Meeting Abstract No:5457 65. Siren AL, Fratelli M, Brines M, Goemans C, Casagrande S, Lewczuk P, Keenan S, Gleiter C, Pasquali C, Capobianco A, Mennini T, Heumann R, Cerami A, Ehrenreich H, Ghezzi P (2001) Erythropoietin prevents neuronal apoptosis after cerebral ischemia and metabolic stress. PNAS 98: 4044–4049 66. Stuben G, Thews O, Pottgen C, Knuhmann K, Vaupel P, Stuschke M (2001) Recombinant human erythropoietin increases the radiosensitivity of xenografted human tumours in anaemic nude mice. J Cancer Res Clin Oncol 127: 346– 350 67. Suzuki N, Ohneda O, Takahashi S, Higuchi M, Mukai HY, Nakahata T, Imagawa S, Yamamoto M (2002) Erythroid-speciﬁc expression of the erythropoietin receptor rescued its null mutant mice from lethality. Blood 100: 2279– 2288 68. Syed RS, Reid SW, Li C, Cheetham JC, Aoki KH, Liu B, Zhan H, Osslund TD, Chirino AJ, Zhang J, Finer-Moore J, Elliott S, Sitney K, Katz BA, Matthews DJ,

The role of erythropoietin receptor expression on tumor cells

69.

70.

71.

72.

73. 74. 75.

76.

77.

78.

79.

80.

81.

82.

101

Wendoloski JJ, Egrie J, Stroud RM (1998) Efﬁciency of signalling through cytokine receptors depends critically on receptor orientation. Nature 395: 511–516 Thews O, Kelleher DK, Vaupel P (2001) Erythropoietin restores the anemiainduced reduction in cyclophosphamide cytotoxicity in rat tumors. Cancer Res 61: 1358–1361 Vaupel P, Dunst J, Engert A, Fandrey J, Feyer P, Freund M, Jelkmann W (2005) Effects of recombinant human erythropoietin (rHuEPO) on tumor control in patients with cancer-induced anemia. Onkologie 28: 216–221 Vaupel P, Kallinowski F, Okunieff P (1989) Blood ﬂow, oxygen and nutrient supply, and metabolic microenvironment of human tumors: a review. Cancer Res 49: 6449–6465 Wagner KF, Katschinski DM, Hasegawa J, Schumacher D, Meller B, Gembruch U, Schramm U, Jelkmann W, Gassmann M, Fandrey J (2001) Chronic inborn erythrocytosis leads to cardiac dysfunction and premature death in mice overexpressing erythropoietin. Blood 97: 536–542 Weiss MJ (2003) New insights into erythropoietin and epoetin alfa: mechanisms of action, target tissues, and clinical applications. Oncologist 8: 18–29 Westenfelder C, Baranowski RL (2000) Erythropoietin stimulates proliferation of human renal carcinoma cells. Kidney Int 58: 647–657 Westphal G, Braun K, Debus J (2002a) Detection and quantiﬁcation of the soluble form of the human erythropoietin receptor (sEpoR) in the growth medium of tumor cell lines and in the plasma of blood samples. Clin Exp Med 2: 45–52 Westphal G, Niederberger E, Blum C, Wollman Y, Knoch TA, Rebel W, Debus J, Friedrich E (2002b) Erythropoietin and G-CSF receptors in human tumor cells: expression and aspects regarding functionality. Tumori 88: 150– 159 Wojchowski DM, Gregory RC, Miller CP, Pandit AK, Pircher TJ (1999) Signal Transduction in the Erythropoietin Receptor System. Exp Cell Res 253: 143– 156 Wu H, Liu X, Jaenisch R, Lodish HF (1995) Generation of committed erythroid BFU-E and CFU-E progenitors does not require erythropoietin or the erythropoietin receptor. Cell 83: 59–67 Yasuda Y, Fujita Y, Masuda S, Musha T, Ueda K, Tanaka H, Fujita H, Matsuo T, Nagao M, Sasaki R, Nakamura Y (2002) Erythropoietin is involved in growth and angiogenesis in malignant tumours of female reproductive organs. Carcinogenesis 23: 1797–1805 Yasuda Y, Fujita Y, Matsuo T, Koinuma S, Hara S, Tazaki A, Onozaki M, Hashimoto M, Musha T, Ogawa K, Fujita H, Nakamura Y, Shiozaki H, Utsumi H (2003) Erythropoietin regulates tumour growth of human malignancies. Carcinogenesis 24: 1021–1029 Yet MG, Jones SS (1993) The extracytoplasmic domain of the erythropoietin receptor forms a monomeric complex with erythropoietin. Blood 82: 1713– 1719 Yoshimura A, Longmore G, Lodish HF (1990) Point mutation in the exoplasmic domain of the erythropoietin receptor resulting in hormone-independent activation and tumorigenicity. Nature 348: 647–649

102 J. Fandrey: The role of erythropoietin receptor expression on tumor cells 83. Yu X, Shacka JJ, Eells JB, Suarez-Quian C, Przygodzki RM, Beleslin-Cokic B, Lin CS, Nikodem VM, Hempstead B, Flanders KC, Costantini F, Noguchi CT (2002) Erythropoietin receptor signalling is required for normal brain development. Development 129: 505–516 Correspondence: Univ.-Prof. Dr. med. Joachim Fandrey, Institut für Physiologie, Universitätsklinikum Essen, Universität Duisburg-Essen, Hufelandstrarse 55, 45147 Essen, Germany, E-mail: [email protected]

Chapter 4

Problems associated with erythropoietin receptor determination on tumor cells A. Österborg Department of Oncology and Hematology, Karolinska University Hospital, Stockholm, Sweden

Erythropoietin is a glycoprotein hormone which regulates erythropoiesis by stimulating proliferation, preventing apoptosis and inducing differentiation of red blood cell precursors in the bone marrow (Spivak 2005). Not surprisingly, erythropoietin receptors (EPO-R) were ﬁrst observed in these target erythroid cells (D’Andrea and Zon 1990). Various studies conﬁrm that the number of receptors located on the cell surface is low, in the range of 100–<2,000 receptors per cell (D’Andrea and Zon 1990). More recently, EPO-R has been reported in cells and tissues outside of the hematopoietic system (e.g. liver, uterus, central nervous system, heart, vascular endothelium and malignant tumors) (Farrell and Lee 2004; Jelkmann and Wagner 2004). This has led to the speculation that erythropoietin may have an additional pleiotropic role, eliciting a tissue protective response following ischemic stress or injury (Boogaerts 2006). However, reports that tumor cells can express EPO-R have led to the concern that stimulation of these receptors by exogenous erythropoietin could have a detrimental impact on the long-term outcome of patients with cancer (Khuri 2007). These studies are of important scientiﬁc signiﬁcance, but it is necessary to consider their relevance to the clinically approved use of erythropoietin in the oncology setting (Österborg et al. 2007). There are several mechanisms postulated by which erythropoietin could have an impact on tumor growth, including direct growth-regulatory effects in cells expressing EPO-R, reversal of tumor hypoxia, modulation of the efﬁcacy of anticancer drugs and modulation of angiogenesis (Österborg et al. 2007). However, a review of the published literature suggests that erythropoietin may demonstrate no effect or negative or beneﬁcial effects on tumor growth. Importantly, a functional EPO-R would be required for erythropoietin to induce direct effects and there are several problems in identifying functional EPO-R in tumor tissues; these problems are associated with poor speciﬁcity of EPO-R detection techniques, the difﬁculty in conﬁrming the membrane location of EPO-R and the requirement for very high doses of

104

A. Österborg

erythropoietin to show any EPO-R-induced stimulatory effects in tumor cells. These difﬁculties, along with contradictory results from different research groups and between in vitro systems and animal models, suggest that ﬁndings of preclinical studies should not be over-interpreted with regards to their relevance for patients with cancer (Österborg et al. 2007).

Problems with detection of EPO-R There are various techniques used to detect EPO-R in tumor biopsies and cancer cell lines, including reverse transcriptase-polymerase chain reaction (RT-PCR) to identify receptor mRNA (Kokhaei et al. 2007), and use of antibodies directed at a region of the receptor protein for immunohistochemistry and Western blotting techniques (Elliott et al. 2006). Unfortunately, none of these methods has been well validated and all have their limitations. For example, samples used for RT-PCR have the potential for contamination by non-tumor cells as it can be difﬁcult to isolate malignant cells from surrounding normal tissue. Various studies have suggested that levels of EPO-R transcripts in neoplastic cells are similar to levels in paired normal samples, suggesting that there is no selective advantage for tumors to over express the EPO-R gene (Sinclair et al. 2005). In addition, RT-PCR can only detect EPO-R transcripts and cannot show whether these lead to functional mRNA or receptor protein expression. Kokhaei et al. (2007) recently investigated EPO-R expression in tumor cell samples obtained from patients with lymphoid malignancies. Although EPO-R mRNA was detected by RT-PCR in many of the samples, no tumor stimulatory effect was observed despite culturing cells at high erythropoietin concentrations (Fig. 1). Similar ﬁndings were observed by Westphal et al. (2002) in several cell lines of solid tumor and lymphoid malignancy origin. Thus, these results underscore the importance of demonstrating functionality as well as the presence of EPO-R. There are major limitations associated with the antibody techniques used for detecting EPO-R (Österborg et al. 2007). In addition to immunohistochemistry studies being unable to differentiate adequately between receptor located on the cell surface or within the cytoplasm, there are also problems with the reagents used. Various studies have shown that the EPO-R identiﬁed in tumor cells is often located in the cell cytoplasm, meaning that the receptors are inaccessible to erythropoietin (Jelkmann and Laugsch 2007). In the study by Kokhaei et al. (2007), EPO-R was not detected on the cell surface of enriched tumor cells from patients with lymphoid malignancies, despite several of the samples being positive for EPO-R mRNA expression. Other researchers have observed similar results, showing no or extremely low levels of surface EPO-R on tumor cell lines, although EPO-R protein was identiﬁed by immunoprecipitation (Sinclair et al. 2005; LaMontagne et al. 2006; Um et al. 2007). Therefore, EPO-R protein may be synthesized by

Problems associated with erythropoietin receptor determination

105

Normal donor, B cell fraction

MCL PBMC

MM CD138+ fraction

B-CLL B cell fraction

B-CLL T cell fraction

100 bp ladder

Negative control

The same patient

Normal donor, T cell fraction

A

B Unstimulated cells

1400

Erythropoietin-stimulated cells

1200

CPM

1000 800 600 400 200 0 No epoetin

Epoetin alfa

Epoetin beta

Darbepoetin alfa

Fig. 1. A. Erythropoietin receptor mRNA detected by RT-PCR is expressed in tumor and non-tumor cell fractions from patients with B-cell chronic lymphocytic leukemia (B-CLL), mantle cell lymphoma (MCL) and multiple myeloma (MM) and enriched T and B cells from a healthy control. PBMC = peripheral blood mononuclear cells. B. Tumor cell proliferation (3H-thymidine incorporation) after culture with or without high-dose erythropoietin (100 IU/ml for epoetin alfa and beta and 500 ng/ml for darbepoetin alfa) for 5 days. Mean ± SD CPM for tumor cells obtained from eight patients with B-CLL are shown. [Reprinted from Kokhaei et al. 2007, with permission]

106

A. Österborg

Position of molecular weight markers (KDa)

100 80 60 50

769-P (kidney)

Caki-2 (kidney)

SH-SY5Y (brain)

HeLa (cervix)

MCF-7 (breast)

UT-7/EPO+ (leukemia)

Empty vector

FLAG-EPO-R

FLAG-EPO-RD 40

tumor cells but may not be transported to the cell membrane (Sinclair et al. 2007). Probably most importantly in the EPO-R debate, the commercially available antibodies directed against the EPO-R have not been adequately validated in terms of their speciﬁcity and selectivity to allow identiﬁcation of EPO-R over background staining (Österborg et al. 2007). Using Western blot analysis, a recent study conﬁrmed that most commercially available antiEPO-R antibodies lack speciﬁc EPO-R staining (Elliott et al. 2006). One of the most commonly used antibodies in identiﬁcation of EPO-R in tumor cells is the C-20 (sc-695) rabbit polyclonal antipeptide anti-human EPO-R antibody (Santa Cruz Biotechnology). This antibody detected at least three additional proteins of ∼35, 66 and 100 KDa in most tumor cell lines tested, as well as a 59 KDa protein corresponding in molecular weight to EPO-R in the known EPO-R positive cell line, UT-7 (Fig. 2; Elliott et al. 2006). Sequence analysis conﬁrmed EPO-R peptide sequences in the 59 KDa band. No EPO-R peptide sequences were apparent in the 66 KDa band and one third of the peptides in this band were from heat shock proteins. More-

Molecular size of EPO-R

40 30

Fig. 2. Western blot analysis of the erythropoietin receptor (EPO-R) protein using antibody C-20 in various cancer cell lines including the known EPO-R positive cell line, UT-7, derived from a megakaryoblastic leukemia. To aid identiﬁcation, tagged versions of EPO-R contained an 8 amino acid N-terminal FLAG sequence attached to full length EPO-R (FLAG-EPO-R) or to EPO-R with the C-terminal 40 amino acids deleted (FLAG-EPO-R ΔD40). [Reprinted from Elliott et al. 2006, with permission]

Problems associated with erythropoietin receptor determination

107

over, peptide blocking experiments conﬁrmed that heat shock protein (HSP)70–2 speciﬁcally inhibited antibody C-20 binding to the 66 KDa protein. The lack of speciﬁcity of antibody C-20 has subsequently been conﬁrmed by other research groups in a range of cancer cell lines (Brown et al. 2007; Della Ragione et al. 2007; Jelkmann and Laugsch 2007). Elliott et al. (2006) showed that the only antibody that detected EPO-R readily in Western blot analysis was M-20 (Santa Cruz Biotechnology), but even this antibody resulted in nonspeciﬁc binding in immunohistochemistry experiments. These ﬁndings suggest that publications relying on these antibodies and techniques to identify EPO-R protein expression should be viewed with caution (Österborg et al. 2007). For example, the ﬁndings of a recent publication by Henke et al. (2006), which suggested that administration of erythropoietin may induce tumor progression in patients with head and neck cancer and EPO-R-expressing tumors, have been questioned because the C-20 antibody was used in EPO-R detection (Jelkmann and Laugsch 2007; Della Ragione et al. 2007).

Clinical relevance of erythropoietin doses used in preclinical studies It is very important to consider the dose of erythropoietin used when evaluating the results of preclinical studies. In non-anemic, healthy subjects, the serum concentration of endogenous erythropoietin ranges from 0.005– 0.025 IU/ml (Österborg et al. 2007). However, the doses tested in many in vitro studies are several orders of magnitude higher than either the level of endogenous erythropoietin or the level achieved following administration of erythropoietin to patients with anemia (Table 1, Chapter 18 and Addendum 1 in this book). In most of these studies, tumor modulatory effects were typically observed only at the highest doses tested.

Effects of erythropoietin on tumor growth and survival: interpretation of preclinical studies EPO-R mRNA and/or protein expression has been reported in a variety of tumor cell lines (Acs et al. 2001; Arcasoy et al. 2005a; Dagnon et al. 2005; Dillard et al. 2001; McBroom et al. 2005; Ohigashi et al. 1996; Selzer et al. 2000; Westenfelder and Baranowski 2000; Yasuda et al. 2001–2003) and biopsy samples of human lung, head and neck, breast, endometrial, and cervical cancers (Acs et al. 2001–2004b; Amin et al. 2005; Arcasoy et al. 2002, 2005b and 2005c); however, as mentioned above, care should be taken in interpreting these results because of the limitations of the techniques used.

Cancer type

Erythropoietin dose (IU/ml)

Renal cancer cell lines and biopsies

Ovarian cancer cell line

Melanoma cell lines

Breast cancer cell lines and biopsies Cervix cancer cell lines and biopsies Breast cancer cell line Various cancer cell lines and biopsies Glioma and cervix cancer cell lines Prostate cancer cell lines

0.5–100

0–200

10 and 100

Erythropoietin (100 or 200 IU/ml) decreased sensitivity of cells to cisplatin Increased proliferation of cells in response to erythropoietin

Increased proliferation in some cell lines. No increased proliferation in some cell lines Erythropoietin increased resistance to hypoxia and dacarbazine

0–100

30

Autocrine erythropoietin signaling inhibits hypoxia-induced apoptosis Erythropoietin increased expression of anti-apoptotic genes and production of angiogenic growth factors Erythropoietin induced resistance to ionizing radiation/cisplatin

25% increase in proliferation rate of cells in response to exogenous erythropoietin Erythropoietin inhibits cytotoxicity of cisplatin

200 10 or 30

0–200

10

Effect of erythropoietin

Carvalho et al. 2005 Dunlop et al. 2006 Gewirtz et al. 2006 Kokhaei et al. 2007 Liu et al. 2004 Westphal et al. 2002

Renal cancer and leukemia cell lines Non-small-cell lung cancer cell line Breast cancer and leukemia cell lines Lymphoid malignancies (puriﬁed tumor cells) Various cancer cell lines Various cell lines 10 5

100

10

0–12.5

0–8

No signiﬁcant change in efﬁcacy of taxol, tamoxifen or adriamycin with erythropoietin Erythropoietin does not affect tumor proliferation. No erythropoietin-induced cell activation was observed No signiﬁcant change in sensitivity to cisplatin with erythropoietin Apart from in the positive control, UT-7, erythropoietin did not induce proliferation in any of the cell lines tested

Increased sensitivity (pro-apoptotic effect) to chemotherapy in the presence of erythropoietin No erythropoietin induced proliferation

Studies suggesting erythropoietin has neutral or tumor inhibitory effects

McBroom et al. 2005 Westenfelder et al. 2000

Belenkov et al. 2004 Feldman et al. 2006 Kumar et al. 2005

Acs et al. 2004 Batra et al. 2003

Acs et al. 2003

Acs et al. 2001

Studies suggesting erythropoietin has tumor-promoting effects

Reference

Table 1. Examples of preclinical studies examining effects of erythropoietin in cancer cell lines and/or biopsies 108 A. Österborg

Problems associated with erythropoietin receptor determination

109

As discussed in detail in another chapter in this book (Fandrey Chapter 3), researchers from different groups often provide conﬂicting results on whether erythropoietin stimulates proliferation in cell lines reported to express EPO-R (Table 1). Similarly, animal tumor models yield different results to in vitro experiments, with the majority of studies suggesting either no effect, a reduction in tumor size associated with exogenous erythropoietin, or synergism of erythropoietin with anticancer drugs resulting in enhanced cytotoxicity (Table 2). This has also been reviewed extensively in a recent publication by Sinclair et al (2007). The beneﬁcial effects of erythropoietin in tumor models in anemic animals are thought to relate, at least in part, to a reversal of hypoxia in tumor tissue following improvements in hemoglobin levels (Thews et al. 2001; Sigounas et al. 2004). However, two studies using EPO-R antagonists to inhibit endogenous erythropoietin signaling have suggested that erythropoietin may also regulate tumor growth in animal models (Yasuda et al. 2001 and 2003). In a related area of research, EPO-R expression has been noted in endothelial cells and various researchers have investigated whether erythropoietin may induce angiogenesis in tumors (Österborg et al. 2007). Whether this would result in beneﬁcial or negative effects on tumor control is difﬁcult to predict. An improvement in blood supply may reverse tumor hypoxia and increase sensitivity of tumors to oxygen-dependent cytotoxic drugs, while also improving delivery of the drugs to the tumor tissue. Conversely, improving oxygen and nutrient supply through increased blood ﬂow may promote a survival advantage in tumors. In a study of tumor cells derived from several common pediatric tumors, increased production and release of angiogenic growth factors, vascular endothelial growth factor and/or placenta growth factor was noted following exposure to high doses of erythropoietin 10 or 30 IU/ml for 16 hours (Batra et al. 2003). Another study which evaluated erythropoietin-induced signaling in animal models with xenografts of human stomach choriocarcinoma or melanoma found that administration of an EPO-R antagonist resulted in inhibition of angiogenesis and decreased survival of tumor cells (Yasuda et al. 2003). Conversely, another study evaluating erythropoietin treatment of rats with cancer xenografts showed no effect of erythropoietin treatment on tumor angiogenesis (Hardee et al. 2005). In addition, a study of human squamous cell and colorectal cancer xenografts in mice has suggested that erythropoietin-induced modulation of tumor vasculature can improve response to chemotherapy (Tóvári et al. 2005). In this study, although erythropoietin did not affect the microvessel density of the tumor xenografts, there was evidence of signiﬁcant vessel enlargement, suggesting that the increased proliferation index of endothelial cells in response to erythropoietin resulted in vessel dilation rather than providing a source of new blood vessels. Moreover, erythropoietin signiﬁcantly enhanced the antitumor efﬁcacy of

Cancer type

Erythropoietin dose

Head and neck cancer model

Murine breast cancer model

LaMontagne et al. 2006

Murine/rat models of colon, breast and head and neck cancers Rat mammary cancer model

Kjellen et al. 2006

Hardee et al. 2006

Hardee et al. 2005

Epoetin alfa or beta 2.5 μmg/kg every other day or darbepoetin alfa 7.5 μg/kg once weeklyb

400 IU/kg every 3 days

2000 IU/kg

2000 IU/kg three-times weeklya

Studies suggesting erythropoietin has no tumor-promoting effects

Reference

No difference in tumor growth or proliferation in erythropoietin versus control animals. No evidence that erythropoietin induced angiogenesis Despite evidence of erythropoietin-induced signaling and a cytoprotective effect in vitro, erythropoietin did not affect tumor growth delay after taxol treatment No effect of erythropoietin alone. Small but signiﬁcant increase in tumor growth with erythropoietin treatment following surgical trauma Addition of erythropoietin did not affect the efﬁcacy of paclitaxel therapy

Effect of erythropoietin

Table 2. Examples of preclinical studies examining effects of erythropoietin in animal tumor models

110 A. Österborg

Murine lung cancer model

Murine lung cancer model

Rat sarcoma model

Murine squamous cell and colorectal cancer models

Shannon et al. 2005

Sigounas et al. 2004

Thews et al. 2001

Tóvári et al. 2005

b

Erythropoietin treatment for several weeks induced complete tumor regression in 30–60% of mice. Erythropoietin prolonged survival and reduced mortality Erythropoietin had no effect on tumor growth alone but enhanced the efﬁcacy of radiotherapy in anemicmice Erythropoietin treatment, commencing 10 days before chemotherapy, resulted in a signiﬁcant reduction in tumor volume compared with chemotherapy alone Evidence of synergism and enhanced cytotoxicity when erythropoietin is added to cisplatin, mitomycin C or cyclophosphamide Chemotherapy-induced anemia reduced the cytotoxicity of cyclophosphamide, but erythropoietin treatment of anemic animals restored chemotherapy sensitivity Erythropoietin treatment increased proliferation index of tumor-associated endothelial cells. Erythropoietin alone had no effect on growth of tumor xenografts, but enhanced the cytotoxicity when given in combination with 5-ﬂuorouracil Dose chosen to increase hemoglobin by 1–2 g/dl per week

150 IU/kg three-times weekly

1,000 IU/kg three-times weekly

60 IU/kg × two injections

Darbepoetin alfa 10 μg/kg weekly

Darbepoetin alfa 30 μg/kg

30 IU daily

Considered equivalent to a human dose of 60,000–100,000 IU;

Squamous cell carcinoma model

Ning et al. 2005

a

Murine myeloma model

Mittelman et al. 2004

Studies suggesting erythropoietin has tumor growth inhibitory effects

Problems associated with erythropoietin receptor determination 111

112

A. Österborg

5-ﬂuorouracil in this study and the authors postulated that the increased tumor blood vessel surface in erythropoietin-treated animals resulted in improved delivery of chemotherapy to tumor cells (Tóvári et al. 2005). The growth-regulatory potential of erythropoietin in EPO-R expressing cancers and the impact of erythropoietin on tumor hypoxia, angiogenesis and anticancer therapy effectiveness are the subjects of much ongoing preclinical research. These studies are vital for improving our understanding of signaling pathways and the effects of erythropoietin in patients with cancer. Conﬂicting results from the literature and methodological limitations must be considered when evaluating preclinical studies. Ongoing, well-designed and controlled clinical studies will determine the safety of erythropoietin in anemic patients with cancer.

References 1. Acs G, Acs P, Beckwith SM, Pitts RL, Clements E, Wong K, Verma A (2001) Erythropoietin and erythropoietin receptor expression in human cancer. Cancer Res 61: 3561–3565 2. Acs G, Chen M, Xu X, Acs P, Verma A, Koch CJ (2004a) Autocrine erythropoietin signalling inhibits hypoxia-induced apoptosis in human breast carcinoma cells. Cancer Lett 214: 243–251 3. Acs G, Xu X, Chu C, Acs P, Verma A (2004b) Prognostic signiﬁcance of erythropoietin expression in human endometrial carcinoma. Cancer 100: 2376–2386 4. Acs G, Zhang PJ, McGrath CM, Acs P, McBroom J, Mohyeldin A, Liu S, Lu H, Verma A (2003) Hypoxia-inducible erythropoietin signaling in squamous dysplasia and squamous cell carcinoma of the uterine cervix and its potential role in cervical carcinogenesis and tumor progression. Am J Pathol 162: 1789–1806 5. Acs G, Zhang PJ, Rebbeck TR, Acs P, Verma A (2002) Immunohistochemical expression of erythropoietin and erythropoietin receptor in breast carcinoma. Cancer 95: 969–981 6. Amin K, Haroon ZA, Kim SJ, Li S, Rabbani ZN, Vollmer RT, Wang XF, Kelley MJ, Arcasoy MO (2005) Erythropoietin and erythropoietin receptor expression in early stage non-small cell lung cancer: prognostic signiﬁcance. Blood 106: 145b [Abstract 4258] 7. Arcasoy MO, Amin K, Chou S-C, Haroon ZA, Varia M, Raleigh JA (2005b) Erythropoietin and erythropoietin receptor expression in head and neck cancer: relationship to tumor hypoxia. Clin Cancer Res 11: 20–27 8. Arcasoy MO, Amin K, Chou S-C, Lininger R, Raleigh JA, Varia MA (2005c) The expression of erythropoietin and its receptor in breast cancer is associated with in vivo tumor hypoxia. Blood 106: 144b [Abstract 4256] 9. Arcasoy MO, Amin K, Karayal AF, Chou SC, Raleigh JA, Varia MA, Haroon ZA (2002) Functional signiﬁcance of erythropoietin receptor expression in breast cancer. Lab Invest 82: 911–918 10. Arcasoy MO, Amin K, Vollmer RT, Jiang X, Demark-Wahnefried W, Haroon ZA (2005a) Erythropoietin and erythropoietin receptor expression in human prostate cancer. Mod Pathol 18: 421–430

Problems associated with erythropoietin receptor determination

113

11. Batra S, Perelman N, Luck LR, Shimada H, Malik P (2003) Pediatric tumor cells express erythropoietin and a functional erythropoietin receptor that promotes angiogenesis and tumour cell survival. Lab Invest 83: 1477–1487 12. Belenkov AI, Shenouda G, Rizhevskaya E, Cournoyer D, Belzile JP, Souhami L, Devic S, Chow TY (2004) Erythropoietin induces cancer cell resistance to ionizing radiation and to cisplatin. Mol Cancer Ther 3: 1525–1532 13. Boogaerts M (2006) Pleiotropic effects of erythropoietin in neuronal and vascular systems. Curr Med Res Opin 22 [Suppl 4]: S15–S22 14. Brown WM, Maxwell P, Graham ANJ, Yakkundi A, Dunlop EA, Shi Z, Johnston PG, Lappin TRJ (2007) Erythropoietin receptor expression in nonsmall cell lung carcinoma: a question of antibody speciﬁcity. Stem Cells 25: 718–722 15. Carvalho G, Lefaucheur C, Cherbonnier C, Métivier D, Chapel A, Pallardy M, Bourgeade MF, Charpentier B, Hirsch F, Kroemer G (2005) Chemosensitization by erythropoietin through inhibition of the NF-κB rescue pathway. Oncogene 24: 737–745 16. Dagnon K, Pacary E, Commo F, Antoine M, Bernaudin M, Bernaudin JF, Callard P (2005) Expression of erythropoietin and erythropoietin receptor in non-small cell lung cancer. Clin Cancer Res 11: 993–999 17. D’Andrea AD, Zon LI (1990) Erythropoietin receptor: subunit structure and activation. J Clin Invest 86: 681–687 18. Della Ragione F, Cucciolla V, Borriello A, Oliva A (2007) Erythropoietin receptors on cancer cells: a still open question. J Clin Oncol 25: 1812–1813 19. Dillard DG, Venkatraman G, Cohen C, Delgaudio J, Gal AA, Mattox DE (2001) Immunolocalization of erythropoietin and erythropoietin receptor in vestibular schwannoma. Acta Otolaryngol 121: 149–152 20. Dunlop EA, Percy MJ, Boland MP, Maxwell AP, Lappin TR (2006) Induction of Signalling in Non-Erythroid cells by pharmacological levels of erythropoietin. Neurodegenerative Dis 3: 94–100 21. Elliott S, Busse L, Bass MB, Lu H, Sarosi I, Sinclair AM, Spahr C, Um M, Van G, Begley CG (2006) Anti-Epo receptor antibodies do not predict Epo receptor expression. Blood 107: 1892–1895 22. Farrell F, Lee A (2004) The erythropoietin receptor and its expression in tumor cells and other tissues. Oncologist 9: 18–30 23. Feldman L, Wang Y, Rhim JS, Bhattacharya N, Loda M, Sytkowski AJ (2006) Erythropoietin stimulates growth and STAT5 phosphorylation in human prostate epithelial and prostate cancer cells. Prostate 66: 135–145 24. Gewirtz DA, Di X, Walker TD, Sawyer ST (2006) Erythropoietin fails to interfere with the antiproliferative and cytotoxic effects of antitumor drugs. Clin Cancer Res 12: 2232–2238 25. Hardee ME, Kirkpatrick JP, Shan S, Snyder SA, Vujaskovic Z, Rabbani ZN, Dewhirst MW, Blackwell KL (2005) Human recombinant erythropoietin (rEpo) has no effect on tumour growth or angiogenesis. Br J Cancer 93: 1350– 1355 26. Hardee ME, Rabbani ZN, Arcasoy MO, Kirkpatrick JP, Vujaskovic Z, Dewhirst MW, Blackwell KL (2006) Erythropoietin inhibits apoptosis in breast cancer cells via an Akt-dependent pathway without modulating in vivo chemosensitivity. Mol Cancer Ther 5: 356–361

114

A. Österborg

27. Henke M, Mattern D, Pepe M, Bezay C, Weissenberger C, Werner M, Pajonk F (2006) Do erythropoietin receptors on cancer cells explain unexpected clinical ﬁndings? J Clin Oncol 24: 4708–4713 28. Jelkmann W, Laugsch M (2007) Problems in identifying functional receptors in cancer tissue. J Clin Oncol 25: 1627–1628 29. Jelkmann W, Wagner K (2004) Beneﬁcial and ominous aspects of the pleiotropic action of erythropoietin. Ann Hematol 83: 673–686 30. Khuri FR (2007) Weighing the hazards of erythropoiesis stimulation in patients with cancer. N Engl J Med 356: 2445–2448 31. Kjellen E, Sasaki Y, Kjellstrom J, Zackrisson B, Wennerberg J (2006) Recombinant erythropoietin beta enhances growth of xenografted human squamous cell carcinoma of the head and neck after surgical trauma. Acta Otolaryngol 126: 545–547. 32. Kokhaei P, Abdalla AO, Hansson L, Mikaelsson E, Kubbies M, Haselbeck A, Jernberg-Wiklund H, Mellstedt H, Österborg A (2007) Expression of erythropoietin receptor and in vitro functional effects of epoetins in B-cell malignancies. Clin Cancer Res 13: 3536–3544 33. Kumar SM, Acs G, Fang D, Herlyn M, Elder DE, Xu X (2005) Functional erythropoietin autocrine loop in melanoma. Am J Pathol 166: 823–830 34. Lacombe C, Mayeux P (1999) The molecular biology of erythropoietin. Nephrol Dial Transplant 14: 22–28 35. LaMontagne KR, Butler J, Marshall DJ, Tullai J, Gechtman Z, Hall C, Meshaw A, Farrell FX (2006) Recombinant epoetins do not stimulate tumor growth in erythropoietin receptor-positive breast carcinoma models. Mol Cancer Ther 5: 347–355 36. Liu WM, Powles T, Shamash J, Propper D, Oliver T, Joel S (2004) Effect of haematopoietic growth factors on cancer cell lines and their role in chemosensitivity. Oncogene 23: 981–990 37. McBroom JW, Acs G, Rose GS, Krivak TC, Mohyeldin A, Verma A (2005) Erythropoietin receptor function and expression in epithelial ovarian carcinoma. Gynecol Oncol 99: 571–577 38. Mittelman M, Zeidman A, Kanter P, Katz O, Oster H, Rund D, Neumann D (2004) Erythropoietin has an anti-myeloma effect – a hypothesis based on a clinical observation supported by animal studies. Eur J Haematol 72: 155–165 39. Ohigashi T, Yoshioka K, Fisher JW (1996) Autocrine regulation of erythropoietin gene expression in human hepatocellular carcinoma cells. Life Sci 58: 421–427 40. Österborg A, Aapro M, Cornes P, Haselbeck A, Hayward CR, Jelkmann W (2007) Preclinical studies of erythropoietin receptor expression in tumour cells: impact on clinical use of erythropoietic proteins to correct cancer-related anaemia. Eur J Cancer 43: 510–519 41. Selzer E, Wacheck V, Kodym R, Schlagbauer-Wadl H, Schlegel W, Pehamberger H, Jansen B (2000) Erythropoietin receptor expression in human melanoma cells. Melanoma Res 10: 421–426 42. Sigounas G, Sallah S, Sigounas VY (2004) Erythropoietin modulates the anticancer activity of chemotherapeutic drugs in a murine lung cancer model. Cancer Lett 214: 171–179 43. Sinclair A, Busse L, Rogers N, Arnold G, Hoey T, Sarosi I, Elliott S (2005) EPO receptor transcription is not elevated nor predictive of surface expression in human tumor cells. Proc Amer Assoc Cancer Res 46: [Abstract 5457]

Problems associated with erythropoietin receptor determination

115

44. Sinclair AM, Todd MD, Forsythe K, Knox SJ, Elliott S, Begley CG (2007) Expression and function of erythropoietin receptors in tumors. Implications for the use of erythropoiesis-stimulating agents in cancer patients. Cancer 110: 477–488. 45. Spivak JL (2005) The anaemia of cancer: death by a thousand cuts. Nat Rev Cancer 5: 543–555 46. Thews O, Kelleher DK, Vaupel P (2001) Erythropoietin restores the anemiainduced reduction in cyclophosphamide cytotoxicity in rat tumors. Cancer Res 61: 1358–1361 47. Tóvári J, Gilly R, Rásó E, Paku S, Bereczky B, Varga N, Vágó A, Tímár J (2005) Recombinant human erythropoietin alpha targets intratumoral blood vessels, improving chemotherapy in human xenograft models. Cancer Res 65: 7186–7193 48. Um M, Gross AW, Lodish HF (2007) A “classical” homodimeric erythropoietin receptor is essential for the antiapoptotic effects of erythropoietin on differentiated neuroblastoma SH-SY5Y and pheochromocytoma PC-12 cells. Cell Signal 19: 634–645 49. Westenfelder C, Baranowski RL (2000) Erythropoietin stimulates proliferation of human renal carcinoma cells. Kidney Int 58: 647–657 50. Westphal G, Niederberger E, Blum C, Wollman Y, Knoch TA, Rebel W, Debus J, Friedrich E (2002) Erythropoietin and G-CSF receptors in human tumor cells: expression and aspects regarding functionality. Tumori 88: 150–159 51. Yasuda Y, Fujita Y, Masuda S, Musha T, Ueda K, Tanaka H, Fujita H, Matsuo T, Nagao M, Sasaki R, Nakamura Y (2002) Erythropoietin is involved in growth and angiogenesis in malignant tumours of female reproductive organs. Carcinogenesis 23: 1797–1805 52. Yasuda Y, Fujita Y, Matsuo T, Koinuma S, Hara S, Tazaki A, Onozaki M, Hashimoto M, Musha T, Ogawa K, Fujita H, Nakamura Y, Shiozaki H, Utsumi H (2003) Erythropoietin regulates tumour growth of human malignancies. Carcinogenesis 24: 1021–1029 53. Yasuda Y, Musha T, Tanaka H, Fujita Y, Fujita H, Utsumi H, Matsuo T, Masuda S, Nagao M, Sasaki R, Nakamura Y (2001) Inhibition of erythropoietin signalling destroys xenografts of ovarian and uterine cancers in nude mice. Br J Cancer 84: 836–843 Correspondence: Anders Österborg, Department of Oncology (Radiumhemmet), Karolinska University Hospital, 171 76 Stockholm, Sweden, E-mail: anders. [email protected]

Chapter 5

Deﬁnition, classiﬁcation and characterization of anemia in cancer M. R. Nowrousian Department of Internal Medicine (Cancer Research), West German Cancer Center, University Hospital of Essen Medical School, Essen, Germany

Deﬁnition, diagnosis and grading of anemia Anemia is deﬁned as a condition of decreased red blood cell (RBC) mass, reﬂected in a decreased hemoglobin (Hb) and hematocrit (Hct) level. Signs traditionally used in the physical diagnosis of anemia are pallor of the conjunctivae, nail beds, face, palms, and palmar creases. Among these, only pallor of the conjunctivae, nail beds, and palms can be used in patients of any race. There is evidence suggesting that conjunctival pallor may be more sensitive than pallor of the nail beds and the palm, but conjunctival pallor as such is more frequently evident in association with Hb levels <9 g/dl and may be absent even in the presence of such a degree of anemia (Nardone et al. 1990; Sheth et al. 1997; Strobach et al. 1988). Conjunctival pallor, however, should always be a reason to determine Hb level. Hb level should also be determined in the absence of conjunctival pallor when patients are suffering from fatigue, lethargy, dizziness, headache, depression, cognitive impairment, and exertional dyspnea, which are frequent symptoms of chronic anemia with Hb levels between 8–12 g/dl. Hb levels below 8 g/dl are usually associated with markedly reduced exercise capacity, difﬁculty breathing at rest, and rapid or irregular heartbeat at rest. There is also an increased risk of angina pectoris, myocardial infarction, and transient ischemic events (Barrett-Lee et al. 2000; Cella 1997, 1998; Groopman and Itri 1999; Sobrero et al. 2001; Yellen et al. 1997) (see also Chapter 13). Hb level has been reported to be more precise in diagnosis of anemia than Hct level (Keen 1998). In healthy adults, the reference range for Hb varies, depending on sex, age and race. For white American men aged between 20 and 59 years and women aged between 20 and 50+ years, the lower limits of normal for Hb are 13.7 and 12.2 g/dl, respectively, and for men aged 60 years or above 13.2 g/dl. Black American men and women usually have lower Hb levels with a difference of approximately 0.5–1 g/dl to Hb values in white individuals (Beutler and Waalen 2006). According

118

M. R. Nowrousian

to the World Health Organization (WHO), however, Hb levels <13 g/dl in men and <12 g/dl in women are generally deﬁned as anemia (Blanc 1968). In cancer patients, various systems have been used to grade the severity of anemia. These systems mainly differ in their cut-off points referring to the mild or moderate grades of anemia, but they are almost identical regarding more severe or life-threatening grades of anemia (Groopman and Itri 1999) (Table 1). A simple system frequently used in clinical practice is to discriminate between Hb levels of >10 g/dl, 8–10 g/dl and <8 g/dl for characterizing anemia as mild, moderate and severe, respectively (Fig. 1).

Table 1. Grading systems for anemia Severity of anemia

WHO

EORTC

NCI, ECOG, CALGB, GOG

SWOG

Grade 0 Grade 1 (mild) Grade 2 (moderate) Grade 3 (severe) Grade 4 (lifethreatening)

≥11 g/dl 9.5–10.9 g/dl 8–9.4 g/dl 6.5–7.9 g/dl <6.5 g/dl

>12 g/dl 10–12 g/dl 8–9.9 g/dl 6.5–7.9 g/dl <6.5 g/dl

wnr 10 g/dl to wnr 8–9.9 g/dl 6.5–7.9 g/dl <6.5 g/dl

wnr 10 g/dl to wnr 8–9.9 g/dl 6.5–7.9 g/dl <6.5 g/dl

Values are hemoglobin levels. WHO = World Health Organization; EORTC = European Organization for Research and Treatment of Cancer; NCI = National Cancer Institute; ECOG = Eastern Cooperative Oncology Group; SWOG = Southwest Oncology Group; CALGB = Cancer and Leukemia Group B; GOG = Gynecologic Oncology Group; wnr = “within normal range” (12–15 g/dl for women, 14–16 g/dl for men.

Severity of anemia Mild

Moderate

12 g/dl 13 g/dl

10 g/dl (women) 10 g/dl (men) 10 g/dl

Severe

8 g/dl 8 g/dl

Hemoglobin Fig. 1. Classiﬁcation of anemia

Deﬁnition, classiﬁcation and characterization of anemia in cancer

119

Differential diagnosis of anemia Any mechanism that can cause anemia in other groups of patients can also cause anemia in cancer patients. It is, therefore, necessary to differentiate the type of anemia also in these patients before starting any kind of treatment (Fig. 2, Tables 2, 3). A simple and useful approach for classifying the type of anemia is the use of the reticulocyte count and the mean corpuscular volume (MCV) of red blood cells (RBCs). An adequate reticulocyte count related to the degree of anemia usually indicates blood loss or hemolysis, while an inadequate reticulocyte count usually indicates a hyporegenerative anemia as a result of a disturbed RBC production. In the latter case, anemia can be further categorized by using MCV (Fig. 2).

Anemia

Reticulocyte count inadequate related to the degree of anemia

MCV

MCV

MCV

decreased (microcytic anemia)

normal (normocytic anemia)

increased (macrocytic anemia)

Iron deficiency; thalassemia

Renal insufficiency; chronic disease; bone marrow failure

Vitamin B12 deficiency; folate deficiency

Reticulocyte count adequate related to the degree of anemia

Blood loss

Hemolysis

Fig. 2. Basic evaluations for differentiating anemia

120

M. R. Nowrousian

Table 2. Laboratory tests to evaluate anemia Investigation

Diagnostic value

Blood: Hemoglobin, hematocrit, RBC count MCV, MCH, smear (RBC morphology) WBC count, differential WBC count, platelet count Reticulocyte count Smear Direct antiglobulin test (DAT, Coombs test) Serum: Creatinine, protein electrophoresis LDH, haptoglobin, bilirubin, GPT, GOT Iron, ferritin, transferrin, transferrin saturation* Vitamin B12, methymalonic acid, folinic acid, homocysteine Urine: Blood, protein electrophoresis, urobilinogen Stool: Blood

Anemia and its severity Type of anemia Panmyelopathy, leukemia Hemolysis, bleeding, bone marrow regenerative activity Leukoerythroblastosis, hemolysis Immune hemolysis

Renal disease, multiple myeloma Hemolysis, liver disease Iron deﬁciency Vitamin B12 and folinic acid deﬁciency Blood loss, multiple myeloma, hemolysis Blood loss

RBC = red blood cells; WBC = white blood cells; MCV = mean cell volume; MCH = mean cell hemoglobin; GPT = glutamate-pyruvate transaminase; GOT = glutamineoxalacetic transaminase; LDH = lactate dehyrogenase. * Tranferrin saturation (%) = {iron (μg/dl)/transferrin (g/dl)} × 0.71 or {iron (μmol/l)/transferrin (g/l)} × 3.98.

Microcytic anemia (MCV <80 ﬂ) The most common cause of microcytic anemia is iron deﬁciency (ID), usually indicated by a decreased serum ferritin and typical changes in RBC on the peripheral blood smear including anisocytosis, poikilocytosis and in severe cases the presence of cigar-shaped RBC and elliptocytes. ID may also be associated with reactive thrombocytosis. If microcytic anemia is accompanied by a normal or even elevated serum ferritin level, a diagnosis of thalassemia or an acquired microcytosis should be considered, depending on whether the microcytosis is pre-existing or new, respectively. A further characteristic of the thalassemia trait is an increased RBC count despite microcytic anemia. Additional characteristics are polychromasia as a sign of reticulocytosis, basophilic stippling of RBC and target cells on the peripheral blood smear.

Deﬁnition, classiﬁcation and characterization of anemia in cancer

121

Table 3. Possible causes of anemia associated with cancer Direct effects of malignancy A. Blood loss 1. Exogenous: Head and neck, gastrointestinal, genitourinary, cervical and vaginal cancers 2. Intratumoral: Sarcomas, melanoma, hepatoma, ovarian cancer, adrenocortical tumors B. Bone marrow inﬁltration Leukemias, lymphomas, myeloma, carcinomas (breast, prostate, lung) C. Erythrophagocytosis Histiocytic neoplasms (medullary reticulosis, lymphomas, etc.) Indirect effects of malignancy A. Hemolysis 1. Warm antibodies: Chronic lymphocytic leukemia, other lymphomas, mucin-producing adenocarcinomas 2. Cold agglutinins: Waldenström’s macroglobulinemia, other lymphomas 3. Microangiopathy: Gastrointestinal carcinomas, lung cancer, breast cancer, prostate cancer, cytotoxic agents B. Bone marrow replacement Plasma cell dyscrasia (amyloid) C. Pure red cell aplasia Thymoma, chronic lymphocytic leukemia D. Cancer-related anemia E. Tumor therapy Myelosuppresion, renal impairment, hemolysis

Conditions associated with microcytosis other than ID or thalassemia include systemic diseases, such as rheumatoid arthritis, polymyalgia rheumatica, diabetes mellitus, connective tissue disease, chronic infection, Hodgkin lymphoma, Castleman’s disease, renal cell carcinoma and myeloﬁbrosis with myeloid metaplasia. Cancer-related anemia (CRA) is usually normocytic, but in a small proportion of patients, more frequently in patients with solid tumors than patients with hematological malignancies, it may be microcytic (Nowrousian et al. 1996) (see also Chapter 6). Normocytic anemia (MCV = 80–100 ﬂ) A normocytic anemia may be induced by nutritional deﬁciencies, although iron and vitamin B12/folate deﬁciencies usually produce microcytic and

122

M. R. Nowrousian

macrocytic anemias, respectively. Other types of anemia associated with normocytosis are anemia of renal insufﬁciency, anemia of chronic disease (e.g. rheumatoid arthritis, polymyalgia rheumatica, diabetes mellitus, congestive heart failure, connective tissue disease, protracted infection, cancer) (see also Chapter 6), and anemia due to a primary disorder of the bone marrow (Nowrousian et al. 1996; Tefferi 2003). Macrocytic anemia (MCV >100 ﬂ) Alcohol and the use of certain drugs, such as trimethoprim, zidovudine and cytotoxic agents including methotrexate, 5-ﬂuorouracil, and, particularly, hydroxyurea can produce macrocytosis. Chemotherapy-related macrocytosis is usually oval, and in case of hydroxyurea-related macrocytosis, MCV values are usually >110 ﬂ (Colon-Otero et al. 1992). An important step in differential diagnosis of macrocytic anemia is the evaluation of vitamin B12 and folate deﬁciencies. Serum folate level is usually low in folate deﬁciency, but it may be affected by recent dietary changes. An alternative may be the folate level in RBC, which retain the amount of folate they acquired at birth during their life. A more precise parameter, however, appears to be the level of serum homocysteine, which usually increases in folate deﬁciency due to an impaired folate-dependent conversion of homocysteine to methionine. A normal serum homocysteine level makes the diagnosis of folate deﬁciency highly unlikely (Tefferi 2003; Tefferi et al. 2006). In vitamin B12 deﬁciency, serum vitamin B12 levels are usually low, but they may also be spuriously low in pregnancy and elderly patients and patients with low white blood cell count. In such cases and cases with borderline serum levels of vitamin B12, the more sensitive and highly speciﬁc serum level of methylmalonic acid, which more accurately reﬂects tissue vitamin B12 status, should be used. A normal level of serum methylmalonic acid makes the diagnosis of vitamin B12 deﬁciency extremely unlikely; an increased level, however, is not speciﬁc to vitamin B12 deﬁciency, since it can also occur in renal insufﬁciency or as a result of an inborn metabolic disorder. Once vitamin B12 deﬁciency is conﬁrmed, the presence of intrinsic factor antibodies should be evaluated and, if given, a working diagnosis of pernicious anemia should be performed, including the classical Schilling test or a protein-bound vitamin B12 absorption test (Bates et al. 2003; Beck 1991; Russell 2001). If there is no drug exposure or nutritional deﬁciency explaining the macrocytosis, a primary bone marrow disease, e.g. myelodysplasia, aplastic anemia, and pure red cell aplasia is suspected, particularly, when MCV values are >110 ﬂ. In these cases, a bone marrow biopsy has to be done. In cases with MCV values of 100–110 ﬂ, additional causes of macrocytosis, such as liver disease (round macrocytes, target cells), reticulocytosis (oval macrocytes,

Deﬁnition, classiﬁcation and characterization of anemia in cancer

123

polychromasia) due to hemolysis, and the presence of monoclonal gammopathy should be considered before bone marrow biopsy (Horstman et al. 2005; Tefferi 2003; Van der Weyden and Campbell 1988).

Anemia in cancer Although any type of anemia that can occur in other groups of patients can also occur in patients with cancer, the development of anemia in these patients is frequently linked to their malignant disease, either as a direct or as indirect effect of the malignant process. According to the mechanisms involved, various types of anemia may be present and various types of treatment may be required. The purpose of this review is to describe the types of anemia that occur in association with cancer and the corresponding treatments (Table 4) (Frenkel et al. 1996). Table 4. Antineoplastic agents, which can cause hemolysis Immune hemolysis: Fludarabine 2-Chlorodeoxyadenosine (cladribine) Cisplatin, carboplatin, oxaliplatin Melphalan Methotrexate 6-Mercaptopurine Teniposide Interferon alfa Imatinib mesylate Campath-1H Microangiopathic hemolysis: Mitomycin C 5-Fluorouracil Cisplatin, carboplatin Doxorubicin, daunorubicin Cytarabine Etoposide Cyclophosphamide Hydroxyurea Tamoxifen

Methyl-CCNU Gemcitabine Vinblastine Bleomycin Mitoxantrone Dactinomycin Dacarbazine Deoxycoformycin

Oxidative hemolysis: Doxorubicin, pentostatin, ﬂuoropyrimidines Tamoxifen

124

M. R. Nowrousian

Anemia as a direct effect of cancer Blood loss An example for a direct connection between anemia and malignancy is an acute or chronic exogenous bleeding from the malignant disease itself or from inﬁltrated normal tissues around it. Such a bleeding usually occurs in carcinomas located in the head and neck region, gastrointestinal tract, urinary tract, and cervix or vagina. It is frequently occult and accompanied by signs of ID, particularly decreased transferrin saturation (TfS), which appears to be a more sensitive marker than serum ferritin in ID conditions associated with cancer, as found in patients with colorectal cancer (Beale et al. 2005). Further examples for blood loss as a cause of anemia are intratumoral bleedings in large tumor masses and intracavitary bleedings in carcinomatosis of pleura or peritoneum (Table 4). Bone marrow inﬁltration Depression of hematopoiesis through bone marrow inﬁltration by the malignant disease is another example for a direct effect of cancer as a cause of anemia (Table 3). It frequently occurs in hematological malignancies, but can also occur in solid tumors, particularly in carcinomas of breast, prostate and lung (Table 4). These tumors frequently induce a desmoid reaction of the bone marrow with increased ﬁbrosis. The latter may additionally reduce the marrow space and alter the sinusoidal matrix and the usual orderly release of mature blood cells. The primary clue to an inﬁltration of the bone marrow by the malignant disease is leukoerythroblastosis, which is characterized by the presence of immature red and white cells in circulation, identiﬁed on the blood smear. There are nucleated red cells of all stages of maturity with increased variation in red cell shape (poikilocytosis) and red cell distribution width (RDW). Immature granulocytes are also present, often associated with a mild leukocytosis. The platelet count may be increased, and there may be an increased variation in shape and size of platelets. Due to the ﬁbrotic changes of the bone marrow, it is often difﬁcult to obtain marrow aspirates in patients with solid tumors and marrow involvement. Bone marrow biopsy is, therefore, the method of choice, if such an involvement is suspected (Clifford 1966; Dunn et al. 1993; Makoni and Laber 2004; Mathew et al. 2000; Varma et al. 2000). Anemia as an indirect effect of cancer The mechanisms by which malignant diseases may indirectly cause anemia can be divided into those connected to substances or factors that are

Deﬁnition, classiﬁcation and characterization of anemia in cancer

125

produced by the malignant disease itself or the production of which is induced by the latter and those that are related to anticancer treatment. An example for the development of anemia as an indirect effect of cancer is replacement of the bone marrow by amyloid in multiple myeloma (Stone and Frenkel 1975). Further examples are pure red cell aplasia associated with thymoma or B-cell lymphoma, particularly chronic lymphocytic leukemia (CLL) (Alter et al. 1990; Dessypris 1991; Masaoka et al. 1989), and hemophagocytic syndrome associated with malignant diseases, especially gastric carcinoma, lymphoma or leukemia (Reiner and Spivak 1988). The former may result from activation of T-gamma cells within the bone marrow, but the underlying mechanisms of this process and the activation of macrophages in hemophagocytic syndrome are still unknown. As will be discussed below, the vast majority of anemias in cancer results from an indirect effect of malignancy on erythropoiesis or from the effect of anticancer treatment on this system.

Hemolysis One of the mechanisms of anemia that can be caused indirectly by tumor is hemolysis. It can be induced either immunologically by production of antibodies against RBC or mechanically by changes in microvessels resulting in platelet aggregation and, consequently, thrombotic destruction of erythrocytes. Laboratory evidences of hemolysis are increased reticulocyte count and RDW as well as increased serum level of lactate dehyrogenase (LDH) and indirect bilirubin and decreased level of haptoglobin. While immunemediated hemolysis usually takes place extravascularly, microangiopathic, infection-associated and chemical-induced hemolysis generally occurs intravascularly. Intravascular hemolysis has typically a more rapid and aggressive course than extravascular hemolysis (Gehrs and Friedberg 2002).

Immune hemolysis Immune hemolysis (IH) is a condition, in which antibodies bind to RBC surface antigens and initiate RBC destruction via the complement and the reticuloendothelial system. IH can be classiﬁed as autoimmune, alloimmune or drug-induced depending on the antigenic stimulus responsible for the immune response. Autoimmune hemolysis (AIH) is characterized by the production of antibodies directed against self RBCs. Alloimmune hemolysis requires exposure to allogenic RBCs, and the resulting antibodies show no reactivity toward autologous RBCs. The principal manifestations of RBC alloimmunization are hemolytic transfusion reactions and hemolytic disease of the newborn. Drug-induced antibodies can recognize either intrinsic RBC antigens or RBC-bound drugs. Antibodies that react with intrinsic RBC

126

M. R. Nowrousian

antigens are serologically indistinguishable form autoantibodies, while antibodies that react against RBC-bound drugs require the drug for hemolysis (Gehrs and Friedberg 2002). AIH may be mediated by warm antibodies or cold agglutinins. Occasionally, a combination of both can occur. Warm antibodies react most strongly near 37°C and exhibit decreased afﬁnity at lower temperatures. The cold agglutinins bind to RBCs most strongly near 0–4°C and typically show little afﬁnity at physiologic temperature. Warm antibodies are commonly seen in lymphoproliferative diseases, particularly in CLL and other types of low- or intermediate-grade malignant lymphomas. They can also occur in Hodgkin’s disease, but less frequently, and quite rarely in adenocarcinomas, predominantly in those that produce mucin, such as ovarian carcinoma. Cold agglutinins are often the cause of anemia in Waldenström’s macroglobulinemia, but can also occur in other types of lymphoma and in solid tumors. In AIH associated with malignant diseases, the symptoms of hemolysis may precede the evidence of the underlying malignancy for months or years. In patients with solid tumors, erythrocyte autoantibodies and immune hemolysis tend to occur with a large tumor mass and metastatic disease, and generally indicate a poor prognosis (Berentsen et al. 2006; Doll and Weiss 1985; Frenkel et al. 1996; Garratty 2005; Gehrs and Friedberg 2002; Rytting et al. 1996; Semple and Freedman 2005; Sokol et al. 1994; Wortman et al. 1979).

Warm antibodies The laboratory features of warm-antibody-mediated hemolysis are elevated MCV and polychromasia and macrocytosis from the reticulocytosis as well as nucleated RBCs on the blood smear. Mild to moderate indirect bilirubinemia and an increased LDH level are commonly present. Warm autoantibodies are generally polyclonal. They can be identiﬁed by the evidence of a positive direct antiglobulin test (DAT, Coombs’ test) with immunoglobulin G or complement or both found on the surface of erythrocytes. The proportion of positive Coombs’ tests in AIH is 83–95%. In suspicious cases with negative DAT result, elution of RBC antibodies can be used to clarify the presence of autoantibodies (Garratty 2005; Sachs et al. 2006). The RBCs coated with the antibody have a shortened survival due to an earlier sequestration in the spleen, and to a much lesser extent in the liver. After sequestration, the cells will be destroyed by macrophages. Treatment of warm-antibody-mediated immune hemolysis is directed at reduction of both antibody production and destruction of RBC. Corticosteroids, in a dose of 1–2 mg/kg per day, are the initial treatment of choice, in addition to the treatment of the underlying malignancy. In emergency situations, megadosages of 500–1,000 mg of methylprednisone have been reported to be useful at the initiation of treatment, but there are limited experiences

Deﬁnition, classiﬁcation and characterization of anemia in cancer

127

with this approach and there is a lack of controlled studies. Treatment of the underlying neoplasm, if effective, can help to control the hemolysis. It is of particular value in patients in whom the antibodies are produced by lymphoproliferative disease (Doll and Weiss 1985; Frenkel et al. 1996; Rytting et al. 1996). Patients treated with glucocorticoids gradually show some improvement within one week, and in 70–80% of cases, an improvement within three weeks. In new cases, the rate of complete remissions amounts to 15–20%. In responding cases, the dose of steroid should be gradually tapered as long as tolerated. In cases with complete remission, treatment can be ﬁnally discontinued, but more typically, a greater proportion of patients requires a maintenance dose. In severe or prolonged hemolysis, additional folic acid supplementation should be performed (Gehrs and Friedberg 2002). In some recent reports, rituximab, an anti-CD20-antibody, either as single agent or in combination with cytotoxic drugs or high-dose immunoglobulin, has been found to be effective in the treatment of AIH associated with lymphoproliferative diseases, such as CLL. Although the number of patients treated is small, the results indicate that in such cases this drug should be considered in the management of steroid refractory AIH anemia. An effective dose of rituximab appears to be 375 mg/m2/week for 4 weeks (Chemnitz et al. 2002; D’Arena et al. 2006; Gupta et al. 2002; Iannitto et al. 2002; Robak 2004; Seipelt et al. 2001; Trape et al. 2003; Zaja et al. 2003). Another monoclonal antibody possibly helpful in the treatment of refractory AIH in CLL is campath-1H or alemtuzumab, a humanized anti-CD52antibody. Experiences with this drug are encouraging but still limited to a small number of patients (Lundin et al. 2006; Robak 2004; Willis et al. 2001). The dosage of campath-1H applied in one study was 10 mg/d as intravenous infusion for 10 days (Willis et al. 2001). The indications for RBC transfusions in patients with AIH are not signiﬁcantly different than for similarly anemic patients without AIH, as long as appropriate compatibility procedures are performed to detect and identify RBC alloantibodies. The decision to transfuse should primarily depend on the patient’s need, particularly on the evaluation if anemia is life-threatening or if there is a high risk of cardiac or cerebrovascular ischemic events without RBC transfusions (Gehrs and Friedberg 2002).

Cold agglutinins The primary diagnostic clue to cold agglutinins is autoagglutination or “laking” of RBC noted in the blood smear. There may be a falsely elevated measurement of the mean cell volume (MCV) of erythrocytes due to their clumping during automated blood cell count analysis. The value, however, decreases markedly, after warming the blood toward 37°C, e.g. by holding the specimen tube in the hand for a few minutes, just before the blood is

128

M. R. Nowrousian

taken up by the automated counting instrument. The peripheral smear may show agglutination, polychromasia, anisocytosis, poikilocytosis, and occasionally spherocytosis. A mild elevation of indirect bilirubin and LDH in serum may also occur. The pathophysiology of cold agglutinin syndrome typically involves immunoglobulin M (IgM) autoantibodies and complement 3 (C3). The direct Combs test is usually positive, but only for C3: since IgM antibodies dissociate from the surface of erythrocytes and are not detectable. The sequestration of cold-agglutinin-coated erythrocytes occurs primarily in the liver, where the cells are removed by Kupffer cells of the reticuloendothelial system (Doll and Weiss 1985; Frenkel et al. 1996; Gehrs and Friedberg 2002; Rytting et al. 1996). Treatment of anemias caused by cold agglutinins is generally less effective. An exception are those associated with lymphoproliferative diseases, which may respond to alkylating agents or purine analogs, especially ﬂudarabine and 2-chlorodeoxyadenosine (Frenkel et al. 1996; Gehrs and Friedberg 2002; Petz 2001). Recently, several small prospective studies and some case reports have shown beneﬁcial effects of rituximab on cold agglutinin syndrome associated with lymphoproliferative as well as idiopathic diseases (Bauduer 2001; Cohen et al. 2001; Lee and Kueck 1998; Robak 2004; Siddiqui et al. 2003; Vassou et al. 2005; Webster et al. 2004). In a review of cases published, the rate of responses was found to be 89% and the rate of complete responses 44% (Robak 2004). Rituximab was usually given in a dose of 375 mg/m2/weekly for 4 weeks. RBC transfusions should be limited to patients with life-threatening anemia or those who are at risk of cardiac or cerebrovascular ischemic events. RBC transfusions can potentiate the hemolysis. In addition, cold agglutinins cause serologic difﬁculties during the blood bank work-up, and this can induce the release of the so-called least incompatible RBC units, which are associated with the risk of containing undetected alloantibodies. If required, the exogenous complement load can be reduced by the use of washing RBC units. The risk of additional, transfusion-related hemolysis can be reduced by using an in-line blood warmer at 37°C and by keeping the patient warm (Gehrs and Friedberg 2002). Patients with cold agglutinin syndrome should be advised to avoid cold exposure, when possible. If they go out in cold weather, they should clothe themselves well, including earmuffs, warm socks, and warm mittens to prevent acute hemolytic crises.

Microangiopathic hemolysis (MAH) MAH is a rare, but life-threatening, phenomenon in cancer. Prostate cancer and mucin-producing adenocarcinomas of ovary and gastrointestinal tract appear to be particularly predisposed to be associated with microangiopathy.

Deﬁnition, classiﬁcation and characterization of anemia in cancer

129

The underlying mechanisms are still unknown, but there is evidence suggesting that microangiopathy may be induced by procoagulant proteins that are released by tumor cells activating factor X and inducing a pattern of localized intravascular microthrombosis, similar to that seen in disseminated intravascular coagulopathy (DIC) (Moake 1997). A typical laboratory feature of MAH is the evidence of red cell fragmentation indicated by schistocytes in the blood smear (Fig. 2). The erythrocyte MCV is usually decreased and a graph of red cell MCV distribution shows two distinct red cell populations, one consisting of small cell fragments and the other of nearly normalsized cells. MAH is usually associated with anemia and thrombocytopenia. In contrast to DIC, consumptive coagulopathy is either absent or present only in a mild form (Frenkel et al. 1996; George 2006; Gordon and Kwaan 1997, 1999; Kwaan and Gordon 2001). Recognition of MAH may be difﬁcult because of the variety of symptoms and the lack of speciﬁc diagnostic criteria. The underlying systemic microvascular thrombosis may induce functional abnormalities in a number of organs and produce cerebral and myocardial infarctions and renal failure. Depending on its clinical feature, MAH may present as thrombotic thrombocytopenic pupura (TTP), predominantly occurring in adults and being associated with neurological symptoms, or as hemolytic uremic syndrome (HUS), more frequently occurring in children and being associated with renal failure. However, patients with MAH may have none of these abnormalities or both. TTP and HUS are often described as acute, but one fourth of patients with TTP may have symptoms for several weeks before diagnosis. The most common symptoms at presentation are abdominal pain, nausea, vomiting, and weakness. Approximately half of patients with TTP have severe neurologic abnormalities at presentation or during the course of disease, such as seizures and ﬂuctuating focal deﬁcits. On the other hand, many patients may have no or only minor neurologic abnormalities, such as transient confusion. Fever is relatively uncommon (Böhm et al. 2005; George 2006; McCarthy et al. 2004; Sadler et al. 2004; Vesely et al. 2003). Since the symptoms are nonspeciﬁc, they may be misdiagnosed, e.g., as gasteroenteritis, sepsis, or transient cerebral ischemia. The most important diagnostic clues are the laboratory ﬁnding of anemia and thrombocytopenia and signs of MAH indicated by the presence of fragmented or “split” red cells (fragmentocytes, schistocytes or schizocytes) (≥2 in a microscopic ﬁeld with a magniﬁcation of 100 or >1% of total RBCs) and polychromatic red cells on the peripheral smear (Fig. 3) (Burns et al. 2004). Increased serum levels of LDH and indirect bilirubin as well as a negative direct Coombs’ test further support the diagnosis. Although TTP and HUS share the microangiopathic process as the underlying cause of organ dysfunctions, they differ in the type and localization as well as pathogenesis of their microangiopathic thrombi. TTPassociated thrombi are platelet-rich and occur with decreasing severity in heart, pancreas, kidney, adrenal gland, and brain, while HUS-associated

130

M. R. Nowrousian

Fig. 3. Fragmented or “split” red cells (fragmentocytes, schistocytes, schizocytes) (arrows) on peripheral blood smear of a patient with lymphoblastic lymphoma and thrombotic thrombocytopenic purpura (TTP)

thrombi are ﬁbrin/red cell-rich and largely conﬁned to the kidney, often severe, and much less frequently pancreas, adrenal gland, brain, and heart (Hosler et al. 2003; Moake 2002). In addition, a subset of patients with TTP, but not those with HUS, shows a severe deﬁciency (<5% activity) of a metalloprotease, termed ADAMTS13, which cleaves the large von Willebrand factor multimers which are synthesized and secreted by endothelial cells (Allford et al. 2003; Tsai and Lian 1998). The ADAMTS13 deﬁciency observed in TTP appears to be predominantly due to the presence of an IgG antibody against this enzyme, which has been reported to occur in 44%–94% of cases with acquired TTP (Furlan et al. 1998; Sadler et al. 2004; Tsai and Lian 1998; Yomtovian et al. 2004). A lack of ADAMTS13 is considered to result in an increased concentration of von Willebrand factor multimers in plasma and thus a greater ability of these multimers to react with platelets and cause the disseminated platelet thrombin characteristic for TTP (Moake 2002). Despite these differences, the common and essential diagnostic criteria for both TTP and HUS are anemia, thrombocytopenia, and MAH. The presence of these symptoms in adults, with or without neurologic or renal abnormalities, should be regarded as signs of TTP and prompt consideration of treatment (George 2006; Sadler et al. 2004; Vesely et al. 2003; Veyradier and Meyer 2005; Zheng et al. 2004). The mortality rate of TTP is >90%, when it is untreated (Amorosi and Ultmann 1966). Plasma exchange (PE) using fresh frozen plasma is the primary treatment of MAH presenting as TTP. High-dose steroid does not appear to be of beneﬁcial effect as ﬁrst-line treatment and should not delay the use of PE

Deﬁnition, classiﬁcation and characterization of anemia in cancer

131

(Gurkan et al. 2005). In a randomized study, PE appeared to be more effective in prolonging survival in patients with TTP than plasma infusion (Rock et al. 1991). The effectiveness of PE in TTP has been attributed to replacement of ADAMTS13 activity (Moake 2002; Sadler et al. 2004), but PE also appears to be effective in patients without severe ADAMTS13 deﬁciency (Vesely et al. 2003). The American Association of Blood Banks, the American Society for Apheresis, and the British Committee for Standards in Hematology recommend daily PE with replacement of 1.0 to 1.5 times the predicted plasma volume of the patient as standard therapy for TTP (Allford et al. 2003; Smith et al. 2003). The British guidelines recommend that PE should be continued for a minimum of two days after the platelet count returns to >150,000/mm3. They also recommend the use of glucocorticoids for all patients with TTP (Allford et al. 2003). The mortality rate associated with PE is reported to be 2.4%, predominantly related to pulmonary complications and systemic infections (Howard et al. 2006; McMinn et al. 2003; Rizvi et al. 2000). The use of PE has decreased the mortality rate of TTP to <10–30%. Relapses, however, occur in more than one-third of responding cases and a subset of patients develop multiple relapses or persistent disease requiring repeated sessions of PE or additional treatments, respectively. Recent anecdotal experiences in patients with TTP associated with an ADAMTS 13 inhibitor show that in such cases the use of rituximab may be of value. The dosages of rituximab used were 375 mg/m2/wk for 2–8 weeks (Benetatos et al. 2006; Millward et al. 2005; Tsai and Shulman 2003; Yomtovian et al. 2004). Particularly important in the treatment of MAH are also measures of supportive care, including RBC transfusions and, if required, antihypertensive therapy, preferably with angiotensin-converting enzyme inhibitors. Anticonvulsive prophylaxis should also be performed. Patients with MAH rarely develop major bleedings despite severe thrombocytopenia. Thrombocyte transfusions should be avoided in these patients as long as possible, since such transfusions can aggravate the process of intravascular microthrombosis, particularly in the central nervous system. An important aspect of treatment of MAH in cancer patients is a successful therapy of the underlying malignancy (Kwaan and Gordon 2001; Kwaan and Soff 1997).

Cancer-related anemia Cancer-related anemia (CRA) is a term that is exclusively reserved for a type of anemia that occurs in cancer in the absence of blood loss, bone marrow inﬁltration, hemolysis, organ dysfunctions or nutritional deﬁciencies. CRA may be caused by activation of the immune and inﬂammatory system and substances that are released by this system, or possibly by the malignant disease itself or both (Nowrousian et al. 1996) (see also Chapter 6). CRA has

132

M. R. Nowrousian

many similarities with anemias that occur in chronic inﬂammatory disorders, such as rheumatoid arthritis, or chronic infectious diseases (e.g. tuberculosis, systemic fungal infections or acquired immunodeﬁciency syndrome) (Bertero and Caligaris-Cappio 1997; Cartwright 1966; Kreuzer et al. 1997; Lee 1983; Voulgari et al. 1999; Zucker 1985). Beside radiotherapy- or chemotherapyinduced anemia, CRA is the most frequent type of anemia in cancer with an incidence of up to 67% at presentation, depending on the type and stage of underlying malignancy (Barrett-Lee et al. 2000; Groopman and Itri 1999; Harrison et al. 2000; Ludwig et al. 2004; Skillings et al. 1993, 1995). To be able to diagnose CRA, other types of anemia have to be excluded. Obligatory investigations are those that allow exclusion of blood loss, hemolysis, renal and hepatic dysfunction, iron, folic acid and vitamin B12 deﬁciency (Table 2). If there is a suspicion of tumor inﬁltration of the bone marrow, a bone marrow biopsy should also be done. As a type of chronic anemia, CRA has many impacts on organ functions and tremendous impacts on patients’ well-being, physical activity, and quality of life. There is also evidence suggesting that it may have an impact on the outcome of anticancer treatment. These aspects of CRA and its treatment with recombinant human erythropoietin (rhEPO) are presented extensively in other chapters of this book (see Chapters 8–14, 28, 29).

Treatment-related anemia An important factor in inducing or aggravating anemia in cancer patients is anticancer treatment. Myelosuppression is the most important mechanism of this type of anemia, but reduced erythropoietin production related to nephrotoxicity of some drugs and, even if less frequent, hemolysis induced either by antibodies or microangiopathy may also play a role (Tables 1, 2).

Myelosuppression Many patients with cancer are anemic at diagnosis, and depending on the type and intensity of treatment, the proportion of anemic patients can increase considerably during radiotherapy and chemotherapy (Ludwig et al. 2004) (see also Chapter 7). In a study of solid tumor patients receiving radiotherapy, 32% of males and 57% of females were anemic at presentation, and by the end of radiotherapy, the proportion of anemic patients increased to 51% and 64%, respectively, in the two groups of patients. Anemia was most prevalent in patients with uterine-cervix tumors, increasing from 67% at the start of radiotherapy to 82% at its end. In patients with colorectal and lung cancer, the proportion of anemic patients was 53% at baseline, increasing to 67% and 63%, respectively, at the end of radiotherapy (Harrison et al. 2000).

Deﬁnition, classiﬁcation and characterization of anemia in cancer

133

In a study of patients with solid tumors or hematological malignancies receiving non-platinum chemotherapy, 37% of patients were anemic at baseline and 54% after the third cycle of chemotherapy (Coifﬁer et al. 2001). In another study, the proportion of anemic patients rose over the course of chemotherapy from 17% before the ﬁrst cycle to 38% by the sixth cycle. One third of patients required at least one blood transfusion during treatment, with more than 42% of the patients with lung and ovarian cancer requiring transfusions (Barrett-Lee et al. 2000). Symptoms prompting transfusions varied, but considerable proportions of patients complained of lethargy (51%), tiredness (42%), or breathlessness (33%). The myelotoxicity of chemotherapy, and thus the incidence of anemia, depends on the type and intensity of treatment. The incidence of grade 1 and 2 anemia for some new agents, such as docetaxel, topotecan, and gemcitabine ranges between 55 to 81%. The introduction of such agents and the widespread use of more aggressive and dose-dense regimens have resulted in a high incidence of anemia and the need for RBC transfusions in a considerable proportion of patients. In a recent, large prospective European survey, the frequency of anemia in cancer patients receiving chemotherapy was reported to be 75% (Ludwig et al. 2004). The highest incidence of anemia requiring transfusions has been found to occur in patients with lymphoma, lung cancer and gynecologic or genitourinary tumors, in whom the incidence may be as high as 50%–60% (Groopman and Itri 1999; Ludwig et al. 2004; Skillings et al. 1993, 1995) (see also Chapter 7).

Reduced EPO production A number of studies has shown that chemotherapy may have an impact on EPO production. Particularly, this appears to be the case for platinum-based regimens (Nowrousian 1998). Patients receiving this type of chemotherapy more frequently develop anemia and require RBC transfusions (Abels et al. 1991, 1994; Thatcher 1998). The development and severity of anemia appears to be not only related to the myelosuppressive effect of treatment, but also to a reduced EPO production related to platinum nephrotoxicity, although there are conﬂicting data regarding this point (Canaparo et al. 2000; Hasegawa and Tanaka 1992; Matsumoto et al. 1990; Miller et al. 1990; Ozguroglu et al. 2000; Wood and Hrushesky 1995). The mechanisms involved in renal toxicity of platinum are activation of mitogen-activated protein kinase (MAPK) and the inﬂammatory cascade, induction of cell cycle events and metabolic as well as molecular responses typical of the stress response (Arany and Saﬁrstein 2003; Ramesh and Reeves 2002; Schrier 2002). An important factor in the development and severity of platinum-induced anemia appears to be the cumulative dose of the drug. Other important factors are the initial Hb level and the degree of decrease in Hb after the ﬁrst

134

M. R. Nowrousian

cycle of treatment. Patients with an initially reduced Hb level of less than 12 g/dl and those with a fall in Hb of 1–2 g/dl after the ﬁrst cycle of chemotherapy are highly likely to develop anemia and require RBC transfusions (Nowrousian et al. 1996; Nowrousian 1998). In a study evaluating predictive factors for developing cisplatin-induced anemia, an ultraﬁlterable (UF) plasma platinum concentration of >50 ng/ml measured 16 h after drug administration was found to signiﬁcantly predict Hb loss (Pivot et al. 2000a-b). Ninety-one percent of patients with such a UF concentration developed signiﬁcant decrease in Hb compared with 18% of patients with a UF concentration of ≤50 ng/ml.

Hemolysis There is a number of anticancer drugs that can produce hemolysis, as a result of induction of either antibody production or microangiopathy (Table 4).

Immune hemolysis Drug-induced hemolysis is a relatively rare phenomenon (Arndt and Garratty 2005; Braathen and Stavem 1989; Doll and Weiss 1985; Gordon and Kwaan 1997). In cancer patients, however, some drugs, particularly cytotoxic agents, which are able to produce hemolysis are frequently used, such as cisplatin, carboplatin and oxaliplatin (Chen et al. 2004; Garuﬁ et al. 2000; Koutras et al. 2004; Marani et al. 1996; Noronha et al. 2005; Novaretti et al. 2003; Pujol et al. 2000; Sacchi et al. 1995; Taleghani et al. 2005; Tothova et al. 2002). The hemolysis associated with oxaliplatin is reported to be induced by drug-dependent antibodies with or without autoantibodies to RBC (Taleghani et al. 2005). Antibodies against thrombocytes and neutrophils may also occur in some cases resulting in an acute-onset thrombocytopenia, and hemorrhage beside hemolysis (Chen et al. 2004; Koutras et al. 2004; Taleghani et al. 2005). Another group of drugs, which are able to induce antibody production against RBC are purine analogs, particularly ﬂudarabine and cladribine (2-chlorodeoxyadenosine) (Aslan et al. 2006; Bay et al. 1997; Chasty et al. 1998; Dighiero 1996; Fleischman and Croy 1995; Gonzalez et al. 1998; Hamblin et al. 1998; Juliusson 1997; Myint et al. 1995; Orchard et al. 1998; Robak et al. 1997; Tertian et al. 1996; Tsiara et al. 1997; Vick et al. 1998). Another mechanism, which may play a role in the pathogenesis of AIH in patients receiving purine analogs is an increased interferon-γ production by lymphocytes and lymphoma cells after exposure to these drugs. Interferon-γ appears to be a a key cytokine in the development of AIH (Gamberale et al. 2006).

Deﬁnition, classiﬁcation and characterization of anemia in cancer

135

Immune hemolysis, even if associated with antineoplastic agents, can be identiﬁed by a positive direct Coombs test. Its treatment consists of withdrawing the responsible drug and the use of corticosteroids as described above. In refractory cases, treatment with rituximab may be of value, as observed in patients with ﬂudarabine-induced AIH (Swords et al. 2006; Telek et al. 2005). Recent experiences also show encouraging results with rituximab in AIH occurring after allogenic hematopoietic stem cell transplantation, which is usually poorly responsive to steroids. Such an AIH occurs in 6% of patients at one year post transplant, with a median time of 4 months, and is reported to be associated with a mortality rate of up to 52% (Petz 2005; Sokol et al. 2000, 2002).

Microangiopathic hemolysis (MAH) Another mechanism that may be associated with the use of anticancer drugs leading to hemolysis is microangiopathy. The list of agents that have been reported to induce such a process is relatively long (Table 4) (Canpolat et al. 1994; Fung et al. 1999; Leach et al. 1999; Nordstrom and Strang 1993; Saif and McGee 2005; Sakai et al. 1995; Susano et al. 1994; Walker et al. 1989; Watson et al. 1989; Wu et al. 1997). However, by far the most cases have been reported in association with the use of mitomycin C, predominantly presented as HUS after a cumulative dose of >60 mg (D’Elia et al. 1987; Doll and Weiss 1985; Gordon and Kwaan 1997, 1999; Hamner et al. 1983; Hanna et al. 1981; Kwaan and Gordon 2001; Lesesne et al. 1989; Verwey et al. 1987). There are also several reports indicating gemcitabine as an agent inducing microangiopathy, particularly after prolonged application. Gemcitabineinduced microangiopathy also appears to predominantly present as HUS (Blaise et al. 2005; Chopin et al. 2006; Dilhuydy et al. 2002; Dumontet et al. 2001; Flombaum et al. 1999; Fung et al. 1999; Humphreys et al. 2004; Lesesne et al. 1989; Lhotta et al. 1999; Müller et al. 2005; Nadir et al. 2004; Pﬁster 2005; Saif and McGee 2005; Venat-Bouvet et al. 2003). The criteria for the diagnosis of MAH associated with cancer chemotherapy are the same as described above. Microangiopathy, either manifested as TTP or HUS, should be considered in any cancer patient with or without chemotherapy who presents with worsening anemia and thrombocytopenia and other signs of MAH, with or without neurological changes or renal failure, and prompt therapy with PE or infusion or protein A immunoadsorption should be instituted. In chemotherapy-induced MAH, PE appears to be less effective than immunoadsorption using staphylococcus aureus protein A. High-dose corticosteroids (200 mg prednisone per day) may also be useful in this setting and should be included into treatment. Supportive measures as described above are also important (Gordon and Kwaan 1997,

136

M. R. Nowrousian

1999; Kwaan and Gordon 2001; Lesesne et al. 1989). In case of chemotherapy-induced MAH, the drug responsible has also to be stopped. A special type of treatment-related MAH is the so-called transplantation-associated microangiopathy (TAM), which occurs in 5% to 74% of patients with hematopoetic stem cell transplantation, predominantly of allogeneic type, within the ﬁrst 100 days of treatment. Risk factors for developing TAM appear to be high dosages of radio- and chemotherapy, infections, graft-versus-host disease, and the use of cyclosporine A or tacrolimus. There is no consensus on the most appropriate treatment for TAM. PE appears to be considerably less effective in this setting than in TTP, and many patients require additional therapies, such as protein A immunoadsorption, splenectomy, nitric oxide, deﬁbrotide, and immunosuppressive agents including steroids. There are some case reports indicating that the use of vincristine may be of value (Böhm et al. 2005; Daly et al. 2002; Kornacker et al. 2005; Llamas et al. 1997; Martinez et al. 2005; Mateos et al. 2006; PerkowskaPtasinska et al. 2006; Silva et al. 1991).

Oxidative hemolysis Some pharmacologic agents including anticancer drugs can induce metabolic changes in erythrocytes resulting in oxidative stress and hemolysis. Such an agent is, e.g., doxorubicin which is able to produce oxidative hemolysis in case of glucose-6-phosphate dehyrogenase deﬁciency (Doll 1983; Doll and Weiss 1985). Doxorubicin is known to generate reactive oxygen compounds and methemoglobin in normal RBC and to induce oxidant stress to glucose6-phosphate dehydrogenase (G6PD)–deﬁcient RBC (Barr et al. 1980; Henderson et al. 1978; Sagone and Burton 1979). Another agent reported to produce oxidative hemolysis is pentostatin, possibly as a result of a reduced adenosine triphosphate (ATP) level in RBC (Prentice et al. 1981; Siaw et al. 1980). Mild to moderate hemolysis as a cause of anemia may also occur during treatment with ﬂuoropyrimidines, such as 5-ﬂuorouracil, uracil plus tegafur, and capecitabine (Nikolic-Tomasevic et al. 2005; Sandvei et al. 1987; Zurita Saavedra et al. 2001). The mechanism of hemolysis associated with these agents has not yet been deﬁned. Oxidative stress and hemolysis does not appear to be restricted to cytotoxic agents, since they have also been observed in patients treated with tamoxifen as a result of disruption of RBC membrane structure (Cruz Silva et al. 2000). In certain malignant diseases, increased susceptibility to hemolysis may be an inherent property of RBC, as in childhood acute lymphoblastic leukemia, in which altered membrane characteristics of erythrocytes and induced oxidative stress indicated by elevated serum levels of nitric oxide and thiobarbituric acid reactive species were observed (Ghosh et al. 2005).

Deﬁnition, classiﬁcation and characterization of anemia in cancer

137

Conclusion In cancer patients, anemia can be caused by any mechanism that can also cause anemia in other groups of patients. Therefore, it has to be differentiated, even in these patients, before starting any kind of treatment. However, in cancer patients, anemia is frequently linked, either directly or indirectly, with the malignant disease or its treatment. Anemia as a direct effect of cancer is usually caused by blood loss through exogenous bleeding from the tumor or by bone marrow inﬁltration through the latter. Indirect effects inducing anemia may be hemolysis through warm antibodies or cold agglutinins or microangiopathy. The vast majority of anemias resulting from indirect effects of malignancy are either cancer-related or induced by its treatment. Cancer-related anemia is deﬁned as a type of anemia that occurs in the absence of blood loss, hemolysis, bone marrow inﬁltration, renal and hepatic dysfunction or iron, folic acid and vitamin B12 deﬁciency. It may be caused by activation of the immune and inﬂammatory system and substances that are released by this system and, probably, by the malignant disease itself. Treatment-related anemia is both in radiotherapy and chemotherapy mainly induced by myelosuppression, but in the case of chemotherapy, some drugs, such as cisplatin or carboplatin, may also reduce the production of erythropoietin. Further mechanisms of anemia induced by cytotoxic agents may be hemolysis related either to antibody production or microangiopathy.

References 1. Abels R, Gordon D, Nelson R, Krantz K, Ageeb M, Goon B, Liebow B (1991) Transfusion practice in advanced cancer patients. Blood 78 [Suppl 1]: 474a 2. Abels R, Larholt K, Nelson R, Young D (1994) Risk of transfusion in small cell lung cancer patients receiving chemotherapy. Blood 87 [Suppl 1]: 664a 3. Allford SL, Hunt BJ, Rose P, Machin SJ (2003) Guidelines on the diagnosis and management of the thrombotic microangiopathic haemolytic anaemias. Br J Haematol 120: 556–573 4. Alter R, Joshi SS, Verdirame JD, Weisenburger DD (1990) Pure red cell aplasia associated with B cell lymphoma: demonstration of bone marrow colony inhibition by serum immunoglobulin. Leuk Res 14: 279–286 5. Amorosi EL, Ultmann JE (1966) Thrombotic thrombocytopenic purpura: report of 16 cases and review of the literature. Medicine 45: 139–159 6. Arany I, Saﬁrstein RL (2003) Cisplatin nephrotoxicity. Semin Nephrol 23: 460–464 7. Arndt PA, Garratty G (2005) The changing spectrum of drug-induced immune hemolytic anemia. Semin Hematol 42: 137–144 8. Aslan DL, Peterson BA, Long-Tsai M, Eastlund T (2006) Early-onset autoimmune hemolytic anemia after cladribine therapy for Waldenstrom’s macroglobulinemia. Transfusion 46: 90–94

138

M. R. Nowrousian

9. Barr RD, Davidson AR, Jung LK, Mohan Pai KR (1980) Erythrocytotoxicity induced by cancer chemotherapeutic agents. In vitro studies of osmotic fragility and methaemoglobin generation. Scand J Haematol 25: 363–368 10. Barrett-Lee PJ, Bailey NP, O’Brien ME, Wager E (2000) Large-scale UK audit of blood transfusion requirements and anaemia in patients receiving cytotoxic chemotherapy. Br J Cancer 82: 93–97 11. Bates CJ, Schneede J, Mishra G, Prentice A, Mansoor MA (2003) Relationship between methylmalonic acid, homocysteine, vitamin B12 intake and status and socio-economic indices, in a subset of participants in the British National Diet and Nutrition Survey of people aged 65 y and over. Eur J Clin Nutr 57: 349–357 12. Bauduer F (2001) Rituximab: a very efﬁcient therapy in cold agglutinins and refractory autoimmune haemolytic anaemia associated with CD20-positive, lowgrade non-Hodgkin’s lymphoma. Br J Haematol 112: 1085–1086 13. Bay JO, Fouassier M, Beal D, Alcaraz L, Cure H, Chollet P, Plagne R, Travade P (1997) Autoimmune thrombocytopenia after six cycles of ﬂudarabine phosphate in a patient with chronic lymphocytic leukemia. Hematol Cell Ther 39: 209–212 14. Beale AL, Penney MD, Allison MC (2005) The prevalence of iron deﬁciency among patients presenting with colorectal cancer. Colorectal Dis 7: 398–402 15. Beck WS (1991) Diagnosis of megaloblastic anemia. Annu Rev Med 42: 311–322 16. Benetatos L, Vassou A, Bourantes KL (2006) Effectiveness of rituximab as prophylaxis in thrombotic thrombocytopenic purpura. Clin Lab Haematol 28: 288–289 17. Berentsen S, Ulvestad E, Langholm R, Beiske K, Hjorth-Hansen H, Ghanima W, Sorbo JH, Tjonnfjord GE (2006) Primary chronic cold agglutinin disease: a population based clinical study of 86 patients. Haematologica 91: 460–466 18. Bertero MT, Caligaris-Cappio F (1997) Anemia of chronic disorders in systemic autoimmune diseases. Haematologica 82: 375–381 19. Beutler E, Waalen J (2006) The deﬁnition of anemia: what is the lower limit of normal of the blood hemoglobin concentration? Blood 107: 1747–1750 20. Blaise S, Appeltants H, Carpentier PH, Debru JL (2005) Digital ischaemia and gemcitabine. Two new cases. J Mal Vasc 30: 53–57 21. Blanc B, Hallberg L, et al (1968) Nutritional anaemias. WHO Techn Rep Ser 405: 1–40 22. Bohm M, Betz C, Miesbach W, Krause M, von Auer C, Geiger H, Scharrer I (2005) The course of ADAMTS-13 activity and inhibitor titre in the treatment of thrombotic thrombocytopenic purpura with plasma exchange and vincristine. Br J Haematol 129: 644–652 23. Braathen LR, Stavem P (1989) Autoimmune haemolytic anaemia associated with interferon alfa-2a in a patient with mycosis fungoides. Br Med J 298: 1713 24. Burns ER, Lou Y, Pathak A (2004) Morphologic diagnosis of thrombotic thrombocytopenic purpura. Am J Hematol 75: 18–21 25. Canaparo R, Casale F, Muntoni E, Zara GP, Della Pepa C, Berno E, Pons N, Fornari G, Eandi M (2000) Plasma erythropoietin concentrations in patients receiving intensive platinum or nonplatinum chemotherapy. Br J Clin Pharmacol 50: 146–153 26. Canpolat C, Pearson P, Jaffe N (1994) Cisplatin-associated hemolytic uremic syndrome. Cancer 74: 3059–3062

Deﬁnition, classiﬁcation and characterization of anemia in cancer

139

27. Cartwright GE (1966) The anemia of chronic disorders. Semin Hematol 3: 351–375 28. Cella D (1997) The Functional Assessment of Cancer Therapy-Anemia (FACTAn) Scale: a new tool for the assessment of outcomes in cancer anemia and fatigue. Semin Hematol 34: 13–19 29. Cella D (1998) Factors inﬂuencing quality of life in cancer patients: anemia and fatigue. Semin Oncol 25: 43–46 30. Chasty RC, Myint H, Oscier DG, Orchard JA, Bussutil DP, Hamon MD, Prentice AG, Copplestone JA (1998) Autoimmune haemolysis in patients with BCLL treated with chlorodeoxyadenosine (CDA). Leuk Lymphoma 29: 391–398 31. Chemnitz J, Draube A, Diehl V, Wolf J (2002) Successful treatment of steroid and cyclophosphamide-resistant hemolysis in chronic lymphocytic leukemia with rituximab. Am J Hematol 69: 232–233 32. Chen VM, Thrift KM, Morel-Kopp MC, Jackson D, Ward CM, Flower RL (2004) An immediate hemolytic reaction induced by repeated administration of oxaliplatin. Transfusion 44: 838–843 33. Chopin N, Alexandre J, Chapron C, Moachon L, Goldwasser F (2006) Is thrombotic microangiopathy associated with antitumoral activity? Gynecol Oncol 101: 549–550 34. Clifford GO (1966) The clinical signiﬁcance of leukoerythroblastic anemia. Med Clin North Am 50: 779–790 35. Cohen Y, Polliack A, Zelig O, Goldfarb A (2001) Monotherapy with rituximab induces rapid remission of recurrent cold agglutinin-mediated hemolytic anemia in a patient with indolent lympho-plasmacytic lymphoma. Leuk Lymphoma 42: 1405–1408 36. Coifﬁer B, Guastalla, JP, Pujade-Lauraine E, Bastit P (2001) Predicting cancerassociated anaemia in patients receiving non-platinum chemotherapy: results of a retrospective survey. Eur J Cancer 37: 1617–1623 37. Colon-Otero G, Menke D, Hook CC (1992) A practical approach to the differential diagnosis and evaluation of the adult patient with macrocytic anemia. Med Clin North Am 76: 581–597 38. Cruz Silva MM, Madeira VM, Almeida LM, Custodio JB (2000) Hemolysis of human erythrocytes induced by tamoxifen is related to disruption of membrane structure. Biochim Biophys Acta 1464: 49–61 39. D’Arena G, Laurenti L, Capalbo S, D’Arco AM, De Filippi R, Marcacci G, Di Renzo N, Storti S, Califano C, Vigliotti ML, Tarnani M, Ferrara F, Pinto A (2006) Rituximab therapy for chronic lymphocytic leukemia-associated autoimmune hemolytic anemia. Am J Hematol 81: 598–602 40. D’Elia JA, Aslani M, Schermer S, Cloud L, Bothe, A, Dzik W (1987) Hemolyticuremic syndrome and acute renal failure in metastatic adenocarcinoma treated with mitomycin: case report and literature review. Ren Fail 10: 107–113 41. Daly AS, Xenocostas A, Lipton JH (2002) Transplantation-associated thrombotic microangiopathy: twenty-two years later. Bone Marrow Transplant 30: 709–715 42. Dessypris EN (1991) The biology of pure red cell aplasia. Semin Hematol 28: 275–284 43. Dighiero G (1996) Adverse and beneﬁcial immunological effects of purine nucleoside analogues. Hematol Cell Ther 38 [Suppl 2]: S75–S81

140

M. R. Nowrousian

44. Dilhuydy MS, Delclaux C, Pariente A, De Precigout V, Aparicio M (2002) Hemolytic-uremic syndrome complicating a long-term treatment with gemcitabine. Report of a case and review of the literature. Rev Med Interne 23: 189–192 45. Doll DC (1983) Oxidative haemolysis after administration of doxorubicin. Br Med J (Clin Res Ed) 287: 180–181 46. Doll DC, Weiss RB (1985) Hemolytic anemia associated with antineoplastic agents. Cancer Treat Rep 69: 777–782 47. Dumontet C, Morschhauser F, Solal-Celigny P, Bouaﬁa F, Bourgeois E, Thieblemont C, Leleu X, Hequet O, Salles G, Coifﬁer B (2001) Gemcitabine as a single agent in the treatment of relapsed or refractory low-grade non-Hodgkin’s lymphoma. Br J Haematol 113: 772–778 48. Dunn P, Shih LY, Liaw SJ, Sun CF (1993) Bone marrow necrosis in 38 adult cancer patients. J Formos Med Assoc 92: 1107–1110 49. Fleischman RA, Croy D (1995) Acute onset of severe autoimmune hemolytic anemia after treatment with 2-chlorodeoxyadenosine for chronic lymphocytic leukemia. Am J Hematol 48: 293 50. Flombaum CD, Mouradian JA, Casper ES, Erlandson RA, Benedetti F (1999) Thrombotic microangiopathy as a complication of long-term therapy with gemcitabine. Am J Kidney Dis 33: 555–562 51. Frenkel EP, Bick RL, Rutherford CJ (1996) Anemia of malignancy. Hematol Oncol Clin North Am 10: 861–873 52. Fung MC, Storniolo AM, Nguyen B, Arning M, Brookﬁeld W, Vigil J (1999) A review of hemolytic uremic syndrome in patients treated with gemcitabine therapy. Cancer 85: 2023–2032 53. Furlan M, Robles R, Galbusera M, Remuzzi G, Kyrle PA, Brenner B, Krause M, Scharrer I, Aumann V, Mittler U, Solenthaler M, Lammle B (1998) von Willebrand factor-cleaving protease in thrombotic thrombocytopenic purpura and the hemolytic-uremic syndrome. N Engl J Med 339: 1578–1584 54. Gamberale R, Fernandez-Calotti P, Sanchez-Avalos J, Alberto MF, Geffner J, Giordano M (2006) The effect of ﬂudarabine on interferon-gamma production by lymphoid cells from healthy donors and patients with B-cell chronic lymphocytic leukemia. Haematologica 91: 574–576 55. Garratty G (2005) Immune hemolytic anemia-a primer. Semin Hematol 42: 119–121 56. Garuﬁ C, Vaglio S, Brienza S, Conti L, D’Attino RM, Girelli G, Terzoli E (2000) Immunohemolytic anemia following oxaliplatin administration. (Letter.) Ann Oncol 11: 497 57. Gehrs BC, Friedberg RC (2002) Autoimmune hemolytic anemia. Am J Hematol 69: 258–271 58. George JN (2006) Clinical practice. Thrombotic thrombocytopenic purpura. N Engl J Med 354: 1927–1935 59. Ghosh S, Bandyopadhyay S, Bhattacharya DK, Mandal C (2005) Altered erythrocyte membrane characteristics during anemia in childhood acute lymphoblastic leukemia. Ann Hematol 84: 76–84 60. Gonzalez H, Leblond V, Azar N, Sutton L, Gabarre J, Binet JL, Vernant JP, Dighiero G (1998) Severe autoimmune hemolytic anemia in eight patients treated with ﬂudarabine. Hematol Cell Ther 40: 113–118

Deﬁnition, classiﬁcation and characterization of anemia in cancer

141

61. Gordon LI, Kwaan HC (1997) Cancer- and drug-associated thrombotic thrombocytopenic purpura and hemolytic uremic syndrome. Semin Hematol 34: 140–147 62. Gordon LI, Kwaan HC (1999) Thrombotic microangiopathy manifesting as thrombotic thrombocytopenic purpura/hemolytic uremic syndrome in the cancer patient. Semin Thromb Hemost 25: 217–221 63. Groopman JE, Itri LM (1999) Chemotherapy-induced anemia in adults: incidence and treatment. J Natl Cancer Inst 91: 1616–1634 64. Gupta N, Kavuru S, Patel D, Janson D, Driscoll N, Ahmed S, Rai KR (2002) Rituximab-based chemotherapy for steroid-refractory autoimmune hemolytic anemia of chronic lymphocytic leukemia. Leukemia 16: 2092–2095 65. Gurkan E, Baslamisli F, Guvenc B, Kilic NB, Unsal C, Karakoc E (2005) Thrombotic thrombocytopenic purpura in southern Turkey: a single-center experience of 29 cases. Clin Lab Haematol 27: 121–125 66. Hamblin TJ, Orchard JA, Myint H, Oscier DG (1998) Fludarabine and hemolytic anemia in chronic lymphocytic leukemia. J Clin Oncol 16: 3209–3210 67. Hamner RW, Verani R, Weinman EJ (1983) Mitomycin-associated renal failure. Case report and review. Arch Intern Med 143: 803–807 68. Hanna WT, Krauss S, Regester RF, Murphy WM (1981) Renal disease after mitomycin C therapy. Cancer 48: 2583–2588 69. Harrison LB, Shasha D, White C, Ramdeen B (2000) Radiotherapy-associated anemia: the scope of the problem. Oncologist 5 [Suppl 2]: 1–7 70. Hasegawa I, Tanaka K (1992) Serum erythropoietin levels in gynecologic cancer patients during cisplatin combination chemotherapy. Gynecol Oncol 46: 65–68 71. Henderson CA, Metz EN, Balcerzak SP, Sagone AL Jr (1978) Adriamycin and daunomycin generate reactive oxygen compounds in erythrocytes. Blood 52: 878–885 72. Horstman AL, Serck SL, Go RS (2005) Macrocytosis associated with monoclonal gammopathy. Eur J Haematol 75: 146–149 73. Hosler GA, Cusumano AM, Hutchins GM (2003) Thrombotic thrombocytopenic purpura and hemolytic uremic syndrome are distinct pathologic entities. A review of 56 autopsy cases. Arch Pathol Lab Med 127: 834–839 74. Howard MA, Williams LA, Terrell DR, Duvall D, Vesely SK, George JN (2006) Complications of plasma exchange in patients treated for clinically suspected thrombotic thrombocytopenic purpura-hemolytic uremic syndrome. Transfusion 46: 154–156 75. Humphreys BD, Sharman JP, Henderson JM, Clark JW, Marks PW, Rennke HG, Zhu AX, Magee CC (2004) Gemcitabine-associated thrombotic microangiopathy. Cancer 100: 2664–2670 76. Iannitto E, Ammatuna E, Marino C, Cirrincione S, Greco G, Mariani G (2002) Sustained response of refractory chronic lymphocytic leukemia in progression complicated by acute hemolitic anemia to anti-CD20 monoclonal antibody. Blood 99: 1096–1097 77. Juliusson G (1997) Complications in the treatment of CLL with purine analogues. Hematol Cell Ther 39 [Suppl 1]: S41–S44 78. Keen ML (1998) Hemoglobin and hematocrit: an analysis of clinical accuracy. Case study of the anemic patient. ANNA J 25: 83–86

142

M. R. Nowrousian

79. Kornacker M, Luft T, Ho AD, Schaefer HJ (2005) Thrombotic microangiopathy after combined autografting-allografting for multiple myeloma – report of three cases. Eur J Haematol 74: 250–253 80. Koutras AK, Makatsoris T, Paliogianni F, Kopsida G, Onyenadum A, Gogos CA, Mouzaki A, Kalofonos HP (2004) Oxaliplatin-induced acute-onset thrombocytopenia, hemorrhage and hemolysis. Oncology 67: 179–182 81. Kreuzer KA, Rockstroh JK, Jelkmann W, Theisen A, Spengler U, Sauerbruch T (1997) Inadequate erythropoietin response to anaemia in HIV patients: relationship to serum levels of tumour necrosis factor-alpha, interleukin-6 and their soluble receptors. Br J Haematol 96: 235–239 82. Kwaan HC, Gordon LI (2001) Thrombotic microangiopathy in the cancer patient. Acta Haematol 106: 52–56 83. Kwaan HC, Soff GA (1997) Management of thrombotic thrombocytopenic purpura and hemolytic uremic syndrome. Semin Hematol 34: 159–166 84. Leach JW, Pham T, Diamandidis D, George JN (1999) Thrombotic thrombocytopenic purpura-hemolytic uremic syndrome (TTP-HUS) following treatment with deoxycoformycin in a patient with cutaneous T-cell lymphoma (Sezary syndrome): A case report. Am J Hematol 61: 268–270 85. Lee EJ, Kueck B (1998) Rituxan in the treatment of cold agglutinin disease. Blood 92: 3490–3491 86. Lee GR (1983) The anemia of chronic disease. Semin Hematol 20: 61–80 87. Lesesne JB, Rothschild N, Erickson B, Korec S, Sisk R, Keller J, Arbus M, Woolley PV, Chiazze L, Schein PS, et al (1989) Cancer-associated hemolyticuremic syndrome: analysis of 85 cases from a national registry. J Clin Oncol 7: 781–789 88. Lhotta K, Kuhr T, Rumpelt HJ, Woll E, Thaler J, Konig P (1999) Thrombotic microangiopathy with renal failure in two patients undergoing gemcitabine chemotherapy. Am J Nephrol 19: 590–593 89. Llamas P, Romero R, Cabrera R, Sanjuan I, Fores R, Fernandez MN (1997) Management of thrombotic microangiopathy following allogeneic transplantation: what is the role of plasma exchange? Bone Marrow Transplant 20: 305–306 90. Ludwig H, Van Belle S, Barrett-Lee P, Birgegard G, Bokemeyer C, Gascon P, Kosmidis P, Krzakowski M, Nortier J, Olmi P, Schneider M, Schrijvers D (2004) The European Cancer Anaemia Survey (ECAS): a large, multinational, prospective survey deﬁning the prevalence, incidence, and treatment of anaemia in cancer patients. Eur J Cancer 40: 2293–2306 91. Lundin J, Karlsson C, Celsing F (2006) Alemtuzumab therapy for severe autoimmune hemolysis in a patient with B-cell chronic lymphocytic leukemia. Med Oncol 23: 137–139 92. Makoni SN, Laber DA (2004) Clinical spectrum of myelophthisis in cancer patients. Am J Hematol 76: 92–93 93. Marani TM, Trich MB, Armstrong KS, Ness PM, Smith J, Minniti C, Sandler SG (1996) Carboplatin-induced immune hemolytic anemia. Transfusion 36: 1016–1018 94. Martinez MT, Bucher C, Stussi G, Heim D, Buser A, Tsakiris DA, Tichelli A, Gratwohl A, Passweg JR (2005) Transplant-associated microangiopathy (TAM) in recipients of allogeneic hematopoietic stem cell transplants. Bone Marrow Transplant 36: 993–1000

Deﬁnition, classiﬁcation and characterization of anemia in cancer

143

95. Masaoka A, Hashimoto T, Shibata K, Yamakawa Y, Nakamae K, Iizuka M (1989) Thymomas associated with pure red cell aplasia. Histologic and followup studies. Cancer 64: 1872–1878 96. Mateos J, Perez-Simon JA, Caballero D, Castilla C, Lopez O, Perez E, Canizo C, Vazquez L, San Miguel JF (2006) Vincristine is an effective therapeutic approach for transplantation-associated thrombotic microangiopathy. Bone Marrow Transplant 37: 337–338 97. Mathew P, Fleming D, Adegboyega PA (2000) Myelophthisis as a solitary manifestation of failure from rectal carcinoma. A Batson phenomenon? Arch Pathol Lab Med 124: 1228–1230 98. Matsumoto T, Endoh K, Kamisango K, Akamatsu K, Koizumi K, Higuchi M, Imai N, Mitsui H, Kawaguchi T (1990) Effect of recombinant human erythropoietin on anticancer drug-induced anaemia. Br J Haematol 75: 463–468 99. McCarthy LJ, Dlott JS, Orazi A, Waxman D, Miraglia CC, Danielson CF (2004) Thrombotic thrombocytopenic purpura: yesterday, today, tomorrow. Ther Apher Dial 8: 80–86 100. McMinn JR Jr, Thomas IA, Terrell DR, Duvall D, Vesely SK, George JN (2003) Complications of plasma exchange in thrombotic thrombocytopenic purpurahemolytic uremic syndrome: a study of 78 additional patients. Transfusion 43: 415–416 101. Miller CB, Jones RJ, Piantadosi S, Abeloff MD, Spivak JL (1990) Decreased erythropoietin response in patients with the anemia of cancer. N Engl J Med 322: 1689–1692 102. Millward PM, Bandarenko N, Chang PP, Stagg KF, Afenyi-Annan A, Hay SN, Brecher ME (2005) Cardiogenic shock complicates successful treatment of refractory thrombotic thrombocytopenia purpura with rituximab. Transfusion 45: 1481–1486 103. Moake JL (1997) Studies on the pathophysiology of thrombotic thrombocytopenic purpura. Semin Hematol 34: 83–89 104. Moake JL (2002) Thrombotic microangiopathies. N Engl J Med 347: 589– 600 105. Müller S, Schütt P, Bojko P, Nowrousian MR, Hense J, Seeber S, Moritz T (2005) Hemolytic uremic syndrome following prolonged gemcitabine therapy: report of four cases from a single institution. Ann Hematol 84: 110–114 106. Myint H, Copplestone JA, Orchard J, Craig V, Curtis D, Prentice AG, Hamon MD, Oscier DG, Hamblin TJ (1995) Fludarabine-related autoimmune haemolytic anaemia in patients with chronic lymphocytic leukaemia. Br J Haematol 91: 341–344 107. Nadir Y, Hoffman R, Brenner B (2004) Drug-related thrombosis in hematologic malignancies. Rev Clin Exp Hematol, 8: E4 108. Nardone DA, Roth KM, Mazur DJ, McAfee JH (1990) Usefulness of physical examination in detecting the presence or absence of anemia. Arch Intern Med 150: 201–204 109. Nikolic-Tomasevic Z, Jelic S, Cassidy J, Filipovic-Ljeskovic I, Tomasevic Z (2005) Fluoropyrimidine therapy: hyperbilirubinemia as a consequence of hemolysis. Cancer Chemother Pharmacol 56: 594–602 110. Nordstrom B, Strang P (1993) Microangiopathic hemolytic anemias (MAHA) in cancer. A case report and review. Anticancer Res 13: 1845–1849

144

M. R. Nowrousian

111. Noronha V, Burtness B, Murren J, Duffy TP (2005) Oxaliplatin induces a delayed immune-mediated hemolytic anemia: a case report and review of the literature. Clin Colorectal Cancer 5: 283–286 112. Novaretti MC, Fonseca GH, Conchon M, Dorlhiac-Llacer PE, Chamone DA (2003) First case of immune-mediated haemolytic anaemia associated to imatinib mesylate. Eur J Haematol 71: 455–458 113. Nowrousian M, Kasper C, Oberhoff C, Essers U, Voigtmann R, Gallasch W, Quarder O (1996) Pathophysiology of cancer-related anemia. In: Smyth JF, Boogaerts MA, Ehmer BRM (eds) rhErythropoietin in cancer supportive treatment. Marcel Dekker, New York, pp 13–34 114. Nowrousian MR (1998) Recombinant human erythropoietin in the treatment of cancer-related or chemotherapy-induced anaemia in patients with solid tumours. Med Oncol 15 [Suppl 1]: S19–S28 115. Orchard J, Bolam S, Myint H, Oscier DG, Hamblin TJ (1998) In patients with lymphoid tumours recovering from the autoimmune complications of ﬂudarabine, relapse may be triggered by conventional chemotherapy. Br J Haematol 102: 1112–1113 116. Ozguroglu M, Arun B, Demir G, Demirelli F, Mandel NM, Buyukunal E, Serdengecti S, Berkarda B (2000) Serum erythropoietin level in anemic cancer patients. Med Oncol 17: 29–34 117. Perkowska-Ptasinska A, Sulikowska-Rowinska A, Pazik J, Komuda-Leszek E, Durlik M (2006) Thrombotic nephropathy and pulmonary hypertension following autologous bone marrow transplantation in a patient with acute lymphoblastic leukemia: case report. Transplant Proc 38: 295–296 118. Petz LD (2001) Treatment of autoimmune hemolytic anemias. Curr Opin Hematol 8: 411–416 119. Petz LD (2005) Immune hemolysis associated with transplantation. Semin Hematol 42: 145–155 120. Pﬁster C (2005) Hemolytic uremic syndrome following prolonged gemcitabine therapy: report of four cases from a single institution. Ann Hematol 84: 621–622 121. Pivot X, Guardiola E, Etienne M, Thyss A, Foa C, Otto J, Schneider M, Magne N, Bensadoun RJ, Renee N, Milano G (2000a) An analysis of potential factors allowing an individual prediction of cisplatin-induced anaemia. Eur J Cancer 36: 852–857 122. Pivot X, Marghali N, Etienne MC, Bensadoun RJ, Thyss A, Otto J, Francois E, Renee N, Lagrange JL, Schneider M, Milano G (2000b) A multivariate analysis for predicting cisplatin-induced delayed emesis. Oncol Rep 7: 515–519 123. Prentice HG, Russell NH, Lee N, Ganeshaguru K, Blacklock H, Piga A, Smyth JF, Hoffbrand AV (1981) Therapeutic selectivity of and prediction of response to 2’-deoxycoformycin in acute leukaemia. Lancet 2: 1250–1254 124. Pujol M, Fernandez F, Sancho JM, Ribera JM, Milla F, Feliu E (2000) Immune hemolytic anemia induced by 6-mercaptopurine. Transfusion 40: 75–76 125. Ramesh G, Reeves WB (2002) TNF-alpha mediates chemokine and cytokine expression and renal injury in cisplatin nephrotoxicity. J Clin Invest 110: 835–842 126. Reiner AP, Spivak JL (1988) Hematophagic histiocytosis. A report of 23 new patients and a review of the literature. Medicine (Baltimore) 67: 369–388

Deﬁnition, classiﬁcation and characterization of anemia in cancer

145

127. Rizvi MA, Vesely SK, George JN, Chandler L, Duvall D, Smith JW, Gilcher RO (2000) Complications of plasma exchange in 71 consecutive patients treated for clinically suspected thrombotic thrombocytopenic purpura-hemolytic-uremic syndrome. Transfusion 40: 896–901 128. Robak T (2004) Monoclonal antibodies in the treatment of autoimmune cytopenias. Eur J Haematol 72: 79–88 129. Robak T, Blasinska-Morawiec M, Krykowski E, Hellmann A, Konopka L (1997) Autoimmune haemolytic anaemia in patients with chronic lymphocytic leukaemia treated with 2-chlorodeoxyadenosine (cladribine). Eur J Haematol 58: 109–113 130. Rock GA, Shumak KH, Buskard NA, Blanchette VS, Kelton JG, Nair RC, Spasoff RA (1991) Comparison of plasma exchange with plasma infusion in the treatment of thrombotic thrombocytopenic purpura. Canadian Apheresis Study Group. N Engl J Med 325: 393–397 131. Russell MR (2001) Clinical implications of vitamin B 12 deﬁciency in the elderly. Nutrition in Clinical Care 4: 214–220 132. Rytting M, Worth L, Jaffe N (1996) Hemolytic disorders associated with cancer. Hematol Oncol Clin North Am 10: 365–376 133. Sacchi S, Kantarjian H, O’Brien S, Cohen PR, Pierce S, Talpaz M (1995) Immune-mediated and unusual complications during interferon alfa therapy in chronic myelogenous leukemia. J Clin Oncol 13: 2401–2407 134. Sachs UJ, Roder L, Santoso S, Bein G (2006) Does a negative direct antiglobulin test exclude warm autoimmune haemolytic anaemia? A prospective study of 504 cases. Br J Haematol 132: 655–656 135. Sadler JE, Moake JL, Miyata T, George JN (2004) Recent advances in thrombotic thrombocytopenic purpura. Hematology (Am Soc Hematol Educ Program) 407–423 136. Sagone AL Jr, Burton GM (1979) The effect of BCNU and adriamycin on normal and G6PD deﬁcient erythrocytes. Am J Hematol 7: 97–106 137. Saif MW, McGee PJ (2005) Hemolytic-uremic syndrome associated with gemcitabine: a case report and review of literature. JOP 6: 369–374 138. Sakai C, Takagi T, Wakatsuki S, Matsuzaki O (1995) Hemolytic-uremic syndrome due to deoxycoformycin: a report of the second case. Intern Med 34: 593–596 139. Sandvei P, Nordhagen R, Michaelsen TE, Wolthuis K (1987) Fluorouracil (5FU) induced acute immune haemolytic anaemia. Br J Haematol 65: 357–359 140. Schrier RW (2002) Cancer therapy and renal injury. J Clin Invest 110: 743–745 141. Seipelt G, Böhme A, Koschmieder S, Hoelzer D (2001) Effective treatment with rituximab in a patient with refractory prolymphocytoid transformed B-chronic lymphocytic leukemia and Evans syndrome. Ann Hematol 80: 170–173 142. Semple JW, Freedman J (2005) Autoimmune pathogenesis and autoimmune hemolytic anemia. Semin Hematol 42: 122–130 143. Sheth TN, Choudhry NK, Bowes M, Detsky AS (1997) The relation of conjunctival pallor to the presence of anemia. J Gen Intern Med 12: 102–106 144. Siaw MF, Mitchell BS, Koller CA, Coleman MS, Hutton JJ (1980) ATP depletion as a consequence of adenosine deaminase inhibition in man. Proc Natl Acad Sci U S A 77: 6157–6161

146

M. R. Nowrousian

145. Siddiqui K, Cahalane E, Keogan M, Hardiman O (2003) Chronic ataxic neuropathy with cold agglutinins: atypical phenotype and response to anti-CD20 antibodies. Neurology 61: 1307–1308 146. Silva VA, Frei-Lahr D, Brown RA, Herzig GP (1991) Plasma exchange and vincristine in the treatment of hemolytic uremic syndrome/thrombotic thrombocytopenic purpura associated with bone marrow transplantation. J Clin Apher 6: 16–20 147. Skillings JR, Rogers-Melamed I, Nabholtz JM, Sawka C, Gwadry-Sridhar F, Moquin JP, Rubinger M, Ganguly P, Burnell M, Shustik C, Dryer D, McLaughlin M, White D, Mertens W (1995) An epidemiological review of anaemia in cancer chemotherapy in Canada. Eur J Cancer 31A [Suppl 5]: S183 (Abstr 879) 148. Skillings JR, Sridhar FG, Wong C, Paddock L (1993) The frequency of red cell transfusion for anemia in patients receiving chemotherapy. A retrospective cohort study. Am J Clin Oncol 16: 22–25 149. Smith JW, Weinstein R, for the AHCKL (2003) Therapeutic apheresis: a summary of current indication categories endorsed by the AABB and the American Society for Apheresis. Transfusion 43: 820–822 150. Sobrero A, Puglisi F, Guglielmi A, Belvedere O, Aprile G, Ramello M, Grossi F (2001) Fatigue: a main component of anemia symptomatology. Semin Oncol 28 [Suppl 8]: 15–18 151. Sokol RJ, Booker DJ, Stamps R (1994) Erythrocyte autoantibodies, autoimmune haemolysis, and carcinoma. J Clin Pathol 47: 340–343 152. Sokol RJ, Booker DJ, Stamps R, Walewska R (2000) Cold haemagglutinin disease: clinical signiﬁcance of serum haemolysins. Clin Lab Haematol 22: 337–344 153. Sokol RJ, Stamps R, Booker DJ, Scott FM, Laidlaw ST, Vandenberghe EA, Barker HF (2002) Posttransplant immune-mediated hemolysis. Transfusion 42: 198–204 154. Stone MJ, Frenkel EP (1975) The clinical spectrum of light chain myeloma. A study of 35 patients with special reference to the occurrence of amyloidosis. Am J Med 58: 601–619 155. Strobach RS, Anderson SK, Doll DC, Ringenberg QS (1988) The value of the physical examination in the diagnosis of anemia. Correlation of the physical ﬁndings and the hemoglobin concentration. Arch Intern Med 148: 831–832 156. Susano R, Caminal L, Ferro J, Rubiales A, de Lera J, de Quiros JF (1994) [Microangiopathic hemolytic anemia associated with neoplasms: an analysis of 5 cases and a review of the literature]. Rev Clin Esp 194: 603–606 157. Swords R, Nolan A, Fay M, Quinn J, O’Donnell R, Murphy PT (2006) Treatment of refractory ﬂudarabine induced autoimmune haemolytic anaemia with the anti-CD20 monoclonal antibody rituximab. Clin Lab Haematol 28: 57–59 158. Taleghani BM, Meyer O, Fontana S, Ahrens N, Novak U, Borner MM, Salama A (2005) Oxaliplatin-induced immune pancytopenia. Transfusion 45: 704–708 159. Tefferi A (2003) Anemia in adults: a contemporary approach to diagnosis. Mayo Clin Proc 78: 1274–1280 160. Tefferi A, Dingli D, Li CY, Mesa RA (2006) Microcytosis in agnogenic myeloid metaplasia: prevalence and clinical correlates. Leuk Res 30: 677–680

Deﬁnition, classiﬁcation and characterization of anemia in cancer

147

161. Telek B, Batar P, Uvardy M (2005) [Successful combined treatment with rituximab and high dose immunoglobulin in a patient with chronic lymphocytic leukemia with ﬂudarabine-induced severe immune thrombocytopenia]. Orv Hetil 146: 1791–1793 162. Tertian G, Cartron J, Bayle C, Rudent A, Lambert T, Tchernia G (1996) Fatal intravascular autoimmune hemolytic anemia after ﬂudarabine treatment for chronic lymphocytic leukemia. Hematol Cell Ther 38: 359–360 163. Thatcher N (1998) Management of chemotherapy-induced anemia in solid tumors. Semin Oncol 25 [Suppl 7]: 23–26 164. Tothova E, Kafkova A, Stecova N, Fricova M, Guman T, Svorcova E (2002) Immune-mediated complications during interferon alpha therapy in chronic myelogenous leukemia. Neoplasma 49: 91–94 165. Trape G, Fianchi L, Lai M, Laurenti L, Piscitelli R, Leone G, Pagano L (2003) Rituximab chimeric anti-CD20 monoclonal antibody treatment for refractory hemolytic anemia in patients with lymphoproliferative disorders. Haematologica 88: 223–225 166. Tsai HM, Lian EC (1998) Antibodies to von Willebrand factor-cleaving protease in acute thrombotic thrombocytopenic purpura. N Engl J Med 339: 1585–1594 167. Tsai HM, Shulman K (2003) Rituximab induces remission of cerebral ischemia caused by thrombotic thrombocytopenic purpura. Eur J Haematol 70: 183–185 168. Tsiara S, Christou L, Konstantinidou P, Panteli A, Briasoulis E, Bourantas KL (1997) Severe autoimmune hemolytic anemia following ﬂudarabine therapy in a patient with chronic lymphocytic leukemia. Am J Hematol 54: 342 169. Van der Weyden MB, Campbell L (1988) Clinching the diagnosis: macrocytic anemia. Pathology 20: 353–357 170. Varma N, Vaiphei K, Varma S (2000) Angiosarcoma presenting with leucoerythroblastic anaemia bone marrow ﬁbrosis and massive splenomegaly. Br J Haematol 110: 503 171. Vassou A, Alymara V, Chaidos A, Bourantas KL (2005) Beneﬁcial effect of rituximab in combination with oral cyclophosphamide in primary chronic cold agglutinin disease. Int J Hematol 81: 421–423 172. Venat-Bouvet L, Ly K, Szelag JC, Martin J, Labourey JL, Genet D, TubianaMathieu N (2003) Thrombotic microangiopathy and digital necrosis: two unrecognized toxicities of gemcitabine. Anticancer Drugs 14: 829–832 173. Verwey J, de Vries J, Pinedo HM (1987) Mitomycin C-induced renal toxicity, a dose-dependent side effect? Eur J Cancer Clin Oncol 23: 195–199 174. Vesely SK, George JN, Lämmle B, Studt JD, Alberio L, El-Harake MA, Raskob GE (2003) ADAMTS13 activity in thrombotic thrombocytopenic purpurahemolytic uremic syndrome: relation to presenting features and clinical outcomes in a prospective cohort of 142 patients. Blood 102: 60–68 175. Veyradier A, Meyer D (2005) Thrombotic thrombocytopenic purpura and its diagnosis. J Thromb Haemost 3: 2420–2427 176. Vick DJ, Byrd JC, Beal CL, Chafﬁn DJ (1998) Mixed-type autoimmune hemolytic anemia following ﬂudarabine treatment in a patient with chronic lymphocytic leukemia/small cell lymphoma. Vox Sang 74: 122–126 177. Voulgari PV, Kolios G, Papadopoulos GK, Katsaraki A, Seferiadis K, Drosos AA (1999) Role of cytokines in the pathogenesis of anemia of chronic disease in rheumatoid arthritis. Clin Immunol 92: 153–160

148

M. R. Nowrousian: Deﬁnition, classiﬁcation and characterization

178. Walker RW, Rosenblum MK, Kempin SJ, Christian MC (1989) Carboplatinassociated thrombotic microangiopathic hemolytic anemia. Cancer 64: 1017–1020 179. Watson PR, Guthrie TH Jr, Caruana RJ (1989) Cisplatin-associated hemolyticuremic syndrome. Successful treatment with a staphylococcal protein A column. Cancer 64: 1400–1403 180. Webster D, Ritchie B, Mant MJ (2004) Prompt response to rituximab of severe hemolytic anemia with both cold and warm autoantibodies. Am J Hematol 75: 258–259 181. Willis F, Marsh JC, Bevan DH, Killick SB, Lucas G, Grifﬁths R, Ouwehand W, Hale G, Waldmann H, Gordon-Smith EC (2001) The effect of treatment with Campath-1H in patients with autoimmune cytopenias. Br J Haematol 114: 891–898 182. Wood PA, Hrushesky WJ (1995) Cisplatin-associated anemia: an erythropoietin deﬁciency syndrome. J Clin Invest 95: 1650–1659 183. Wortman J, Rosse W, Logue G (1979) Cold agglutinin autoimmune hemolytic anemia in nonhematologic malignancies. Am J Hematol 6: 275–283 184. Wu DC, Liu JM, Chen YM, Yang S, Liu SM, Chen LT, Whang-Peng J (1997) Mitomycin-C induced hemolytic uremic syndrome: a case report and literature review. Jpn J Clin Oncol 27: 115–118 185. Yellen SB, Cella DF, Webster K, Blendowski C, Kaplan E (1997) Measuring fatigue and other anemia-related symptoms with the Functional Assessment of Cancer Therapy (FACT) measurement system. J Pain Symptom Manage 13: 63–74 186. Yomtovian R, Niklinski W, Silver B, Sarode R, Tsai HM (2004) Rituximab for chronic recurring thrombotic thrombocytopenic purpura: a case report and review of the literature. Br J Haematol 124: 787–795 187. Zaja F, Vianelli N, Sperotto A, Patriarca F, Tani M, Marin L, Tiribelli M, Candoni A, Baccarani M, Fanin R (2003) Anti-CD20 therapy for chronic lymphocytic leukemia-associated autoimmune diseases. Leuk Lymphoma 44: 1951–1955 188. Zheng XL, Kaufman RM, Goodnough LT, Sadler JE (2004) Effect of plasma exchange on plasma ADAMTS13 metalloprotease activity, inhibitor level, and clinical outcome in patients with idiopathic and nonidiopathic thrombotic thrombocytopenic purpura. Blood 103: 4043–4049 189. Zucker S (1985) Anemia in cancer. Cancer Invest 3: 249–260 190. Zurita Saavedra AJ, Navarro Garcia M, Espanol I, Fernandez Ortega A (2001) UFT-induced haemolytic anaemia. Cancer Chemother Pharmacol 47: 280–281 Correspondence: Prof. Dr. M. R. Nowrousian, Department of Internal Medicine (Cancer Research), West German Cancer Center, University Hospital of Essen, Hufelandstrasse 55, 45122 Essen, Germany, E-mail: [email protected]

Chapter 6

Pathophysiology of anemia in cancer M. R. Nowrousian Department of Internal Medicine (Cancer Research), West German Cancer Center, University Hospital of Essen, Essen, Germany

Introduction Anemia is a frequent complication of cancer, and there are numerous causes that could have produced the anemia. In a considerable number of cases, however, there are no bone marrow inﬁltration, no signs of blood loss, hemolysis, renal, hepatic or endocrine disorders or nutritional deﬁciencies that could explain the anemia. This type of anemia, which can be called cancerrelated anemia (CRA), is exclusively a consequence of the presence of the malignant disease itself. It has many hematological and biochemical similarities with anemias that occur in chronic inﬂammatory diseases, such as rheumatoid arthritis, and chronic infectious diseases, such as tuberculosis, systemic fungal infections and acquired immunodeﬁciency syndrome (Cartwright 1966; Lee 1983; Zucker 1985; Kreuzer et al. 1997). This type of anemia, referred to as the anemia of chronic disease (ACD), accounts for 52% of anemias in patients without blood loss, hemolysis or hematologic malignancies. Cancer is the cause of anemia in 19% of cases of ACD (Cash and Sears 1989). CRA is often associated with advanced stages of disease, and usually worsens during chemotherapy or radiotherapy, leading to a high proportion of patients requiring red blood cell (RBC) transfusions (Skillings et al. 1993, 1995; Abels et al. 1991). Chemotherapy and radiotherapy themselves can also produce anemia (Skillings et al. 1993, 1995; Groopman et al. 1999; Barrett-Lee et al. 2000; Coifﬁer et al. 2001; Ludwig et al. 2004 and in this book). Experimental and clinical studies show that CRA may be the result of an activation of the immune and inﬂammatory system by the malignant disease. There is evidence suggesting that the presence of a tumor activates monocytes and macrophages, which, in turn, stimulate other inﬂammatory cells, such as stroma cells, natural killer cells, dendritic cells and cytotoxic lymphoid cells. The result is an increased production and secretion of inﬂammatory cytokines, such as interleukin-1 beta (IL-1βb), IL-6, tumor necrosis factor alfa (TNF-α), and interferons (IFNs), which are able to induce the development of anemia (Figs. 1, 2).

150

M. R. Nowrousian

Tumor tissue hypoxia

Monocytes Peripheral blood

Recruitment CSF-1, CCL2, CCL5, VEGF, etc

Tumor cells

Growth

Tumorassociated Signal transduction macrophages IL-6, ST, STC-1, etc Matrix remodeling Proliferation MMP-9, MMP-19, TNF-α etc etc

Invasion

Angiogenesis TNF-α, IL-1β, IL-6, IL-8, CSF-1, G-CSF, GM-CSF, VEGF, PDGF MIF, etc

Metastasis

Fig. 1. Interactions between tumor cells, monocytes and tumor-associated macrophages (TAMs) in hypoxic areas of tumors. CSF = colony-stimulating factor; CCL = monocyte chemoattractants; VEGF = vascular endothelial growth factor; IL = interleukin; TNF = tumor necrosis factor; MMP = matrix metalloproteinase; G-CSF = granulocyte colony-stimulating factor; GM-CSF = granulocyte-macrophage colonystimulating factor; PDGF = platelet-derived growth factor; MIF = macrophage migration inhibition factor

Monocytes and macrophages are essential cellular components of the innate immune system. They derive from myeloid progenitor cells in the bone marrow, which after being developed into promonocytes, are released into the peripheral blood. In circulation, they differentiate into monocytes and can migrate into almost all tissues of the body, where they further differentiate into resident macrophages. The primary function of resident macrophages is to protect the tissue from infection and injury. The differentiation and functional activities of these cells, however, vary considerably from tissue to tissue depending on speciﬁc inﬂuences currently present in the microenvironment. In the bone marrow, e.g. resident macrophages play an important role in supporting erythropoiesis. Late erythroid progenitor cells, CFU-E (colony-forming unit erythroid), normally adhere to a resident macrophage before they differentiate into erythroblasts in the presence of erythropoietin (EPO). In this cellular composition, known as erythroblastic island, the resident macrophage prevents the apoptosis of CFU-E and ery-

Pathophysiology of anemia in cancer

151

Macrophages IL-6

Neopterin IL-1

TNF-α

Stromal Cells

IFN-β

IFN-γ

T-Lymphocytes

Activated Immune System

Fig. 2. Cellular elements of the immune and inﬂammatory system and the respective cytokines postulated to be involved in the development of cancer-related anemia (CRA). TNF = tumor necrosis factor; IL = interleukin; IFN = interferon. Adapted from Nowrousian 2000 (reproduced with permission)

throblasts, most probably through interactions with adhesive molecules in the stroma, and also possibly acts as a source of iron (Sadahira and Mori 1999). In many tumors, inﬁltrating leukocytes, particularly macrophages, make up a considerable part of the tumor mass accounting for up to 70% of cells in some instances. This type of macrophages called “tumor-associated macrophages” (TAMs) almost entirely derive from peripheral blood monocytes recruited by a number of chemoattractants and cytokines, which are produced either by the tumor cells themselves or by stimulated stroma cells, such as ﬁbroblasts, endothelial cells and inﬁltrating leukocytes or by both (Fig. 1). After extravasation into the tumor, monocytes may develop into two different populations of TAMs, M1 or M2, depending on the degree of oxygenation within the tumor. M1 and M2 TAMs show considerable differences in their receptor expression, cytokine production and effector functions. M1 TAMs are potent effector cells, which are able to secrete a number of proinﬂammatory cytokines such as IL-12, and TNF-α and tumoricidal products such as reactive oxygen intermediates, and NO contributing to the death of tumor cells. They are also able to stimulate T-lymphocytes, natural killer cells, and dendritic cells to produce IFNs (Fig. 2). However, there are tumorderived cytokines, such as IL-4, IL-6, TGF-β, CSF-1, and p15E-related

152

M. R. Nowrousian

molecules, which can block the cytoxicity of macrophages. M2 TAMs have the capacity to tune inﬂammatory and adaptive immune responses, scavenge cellular debris, and promote tissue remodelling, but according to their secretory products, they predominantly promote tumor growth and spread by producing cytokines, which stimulate proliferation of tumor cells, suppress the immune response (TGF-β), remodel extracellular matrix, and induce angiogenesis by stimulating endothelial cells to proliferate (Fig. 1). M2 TAMs preferentially occur in hypoxic and necrotic areas of tumors, where they undergo fundamental changes in their phenotype including activation of hypoxiainducible transcription factors HIF1 and HIF2 and dramatic upregulation of genes, which encode the above-mentioned cytokines and other mitogenic, angiogenic, and prometastatic factors (Murdoch and Lewis 2005). With regard to the development and function of M2 TAMs, it has to be recognized that hypoxic areas are a hallmark of invasive human cancers, consistently present in tumors growing beyond 2 mm in diameter. In many types of tumors, including breast, prostate, cervix, uterine endometrium, liver, lung, bladder, kidney, brain and oral cancers, high densities of TAMs have been found to signiﬁcantly correlate with angiogenesis and inversely with prognosis (Bingle et al. 2002; Murdoch and Lewis 2005; Lamagna et al. 2006; Lee et al. 2006; Shih et al. 2006). The immunologic and inﬂammatory response associated with the presence of tumor cells and increased release of proinﬂammatory cytokines appear to be not only a locally restricted process but also a systemic phenomenon, which appears to play a major role in the development of anemia and other cancer-related symptoms, such as cachexia and fatigue (Nowrousian 1996, 2002, 2005; Kurzrock 2001; Argiles et al. 2003; Illman et al. 2005; Collado-Hidalgo et al. 2006). In patients with hematological malignancies and CRA, there are increased serum levels of neopterin and IFN-γ correlating signiﬁcantly with each other and inversely with those of hemoglobin (Hb) and iron. These correlations indicate a possible link between activation of macrophages and T-lymphocytes, respectively, as well as between activated immune and inﬂammatory system and anemia (Denz et al. 1990). The most relevant source of neopterin are monocytes and macrophages, when activated by IFN-γ (Fig. 2) (Melichar et al. 2006). Elevated concentrations of neopterin are present not only in patients with hematological malignancies (Denz et al. 1990; Murr et al. 1999; Reibnegger et al. 1991; Caenazzo et al. 1993), but also in those with solid tumors correlating with disease stage and predicting disease progression (Reibnegger et al. 1986, 1987; Denz et al. 1990; Weiss et al. 1993; Murr et al. 1998, 1999; Melichar et al. 2006). Biologically, neopterin appears to potentiate TNF-αinduced apoptosis and to be protective against free radical-induced tissue damage (Melichar et al. 2006). IFN-γ is a key cytokine in inducing and regulating inﬂammatory processes (Guillonneau et al. 2007; Zhang 2007). It is produced by

Pathophysiology of anemia in cancer

153

T-lymphocytes, natural killer cells, and dendritic cells after being stimulated by proinﬂammatory cytokines IL-1 and IL-12, which derive from monocytes and macrophages (Fig. 2) (Nowrousian 2000, 2002, 2005; Lamagna et al. 2006). In aplastic anemia, circulating and marrow T cells have been shown to contain increased levels of IFN-γ (Dufour et al. 2001; Sloand et al. 2002). Increased concentrations of IFN-γ have also been found in autoimmune diseases (Hooks et al. 1979), and in anemic children with solid tumors (Ek et al. 2005). In a phase II clinical trial in patients with colorectal cancer, one of the toxicities of repeated application of IFN-γ appeared to be anemia (Brown et al. 1991). IL-1 is a pleotropic cytokine with manifold roles in both physiological and pathological conditions. IL-1 is produced not only by monocytes and macrophages, but also by cells of various types of tumors, including head and neck, breast, lung, colon cancers and melanomas. IL-1 has been reported to promote tumor growth, probably by inducing pro-metastatic genes and stimulating stroma cells to produce angiogenic proteins and growth factors such as VEGF, IL-8, IL-6, TNF-α, and TGFβ (Elarai et al. 2006; Lewis et al. 2006). Increased serum concentrations of IL-1 have been observed in patients with rheumatoid arthritis (Eastgate et al. 1988; Maury et al. 1988; Voulgari et al. 1999) and patients with malignant diseases, such as chronic lymphocytic leukaemia and ovarian cancer (Hulkkonen et al. 2000; Maccio et al. 2005) correlating with the degree of anemia. Another cytokine with elevated levels in patients with malignancy and other chronic diseases is TNF-α. It is a monocyte/macrophage-derived cytokine with a central role in maintaining inﬂammation, immunity and host defence (Balkwill et al. 1987; Teppo and Maury 1987; Voulgari et al. 1999). TNF-α is also expressed by malignant cells, as shown in biopsies from ovarian and renal cell carcinoma, and it appears to be involved in pathogenesis and progression of malignant processes (Balkwill 2006; Szlosarek et al. 2006; Kulbe et al. 2007). In cancer patients, circulating levels of TNF-α have been reported to depend on the type and activity of the malignant disease. Patients with active disease and patients with ovarian or oat cell carcinoma have signiﬁcantly higher TNF-α levels than those with no evidence of disease or lymphoma, respectively (Balkwill et al. 1987). According to the results of some clinical and experimental studies, chronic exposure to TNF-α, either as single agent or in combination with IFN-γ produces anemia (Blick et al. 1987; Tracey et al. 1988; Moldawer et al. 1989; Johnson et al. 1989; Fiedler et al. 1991). In a phase I clinical study, patients treated with TNF-α developed anemia, irrespective of the dosage applied (Blick et al. 1987). In rats and mice, chronic administration of TNF-α induced a type of anemia that had many similarities with CRA (Tracey et al. 1988; Moldawer et al. 1989; Johnson et al. 1989). In pediatric patients with aplastic anemia, marrow T lymphocytes were reported to have increased concentrations of TNF-α (Dufour et al. 2001), and in patients with myelodysplasia, bone marrow cells

154

M. R. Nowrousian

were found to overexpress TNF-α, especially in patients with refractory anemia. In these patients, in addition, the degree of TNF-α expression signiﬁcantly correlated with the magnitude of anemia (Heiss et al. 2005; Stifter et al. 2005). Other groups of patients with a link between TNF-α activity and anemia are patients with chronic lymphocytic leukemia and patients with ovarian cancer. In the ﬁrst group of patients, those with anemia were found to have signiﬁcantly higher serum TNF-α levels than nonanemic patients (Capalbo et al. 2002; Ferrajoli et al. 2002), and in the second group, serum concentrations of TNF-α appeared to signiﬁcantly correlate with disease stage and the degree of anemia (Maccio et al. 2005). Recent clinical and experimental studies show that IL-6, a pleotropic cytokine produced by a broad spectrum of cells including monocytes and macrophages, may play a central role in the pathogenesis of CRA. IL-6 is an important cytokine in regulation of immune response, inﬂammation and hematopoiesis (Nishimoto and Kishimoto 2006; Scheller et al. 2006). Recent studies, however, show that IL-6 is also expressed by various types of malignant cells and is an important growth factor for a wide variety of tumors. IL-6 expression has been reported in cells or tissues of breast cancer (Knuepfer and Preiss 2006; Nicolini et al. 2006), colorectal caner (Chung et al. 2006), cholangiocarcinoma (Okada et al. 1994; Sugawara et al. 1998; Kobayashi et al. 2005; Meng et al. 2006; Yamagiva et al. 2006; Wehbe et al. 2006; Isomoto et al. 2007), renal cell carcinoma (Miki et al. 1989; Takenawa et al. 1991), malignant melanoma (Molnar et al. 2000), Hodgkin’s disease (Reynolds et al. 2002), B-cell chronic lymphocytic leukemia (Lahat et al. 1991), diffuse large-cell lymphoma (Kurzrock 1997), and multiple myeloma (Thabard et al. 1999, 2001; Barille et al. 2000; Lauta 2001, 2003; Bommert et al. 2006; Arnulf et al. 2007). In addition, in a number of these tumors, IL-6 has been identiﬁed as a potent growth factor (Chang et al. 2005; Haura et al. 2006; Hsu and Chung 2006; Yeh et al. 2006). Furthermore, increased serum or plasma concentrations of IL-6 have been measured in patients with a wide variety of tumors including those mentioned above together with ovarian cancer, non-small cell lung cancer, gastric cancer, prostate cancer, oral cancer, and bone sarcoma (Table 1) (Bataille et al. 1989; Berek et al. 1991; Blay et al. 1992; David et al. 1995; Gallo et al. 1995; Seymour et al. 1995; Brown et al. 1997; Jablonska et al. 1997; Goydos et al. 1998; Walther et al. 1998; Hulkkonen et al. 2000; Nishimura et al. 2000; Rokicka-Piotrowicz et al. 2000; Fayad et al. 2001; Lai et al. 2002; John et al. 2004; Songur et al. 2004; Tangkijvanich et al. 2004; Ashizawa et al. 2005; Huang et al. 2005; Kuku et al. 2005; Maccio et al. 2005; Nikiteas et al. 2005; Sifridaki et al. 2005; Zakrzewska and Omyla 2005; Asfandi et al. 2006; Bartsch et al. 2006; Kaminska et al. 2006; Casanovas et al. 2007). Moreover, increased concentrations of IL6 signiﬁcantly correlated with disease stage and activity and, if evaluated, negatively with survival. In studies, in which IL-6 was tested for its relation

Pathophysiology of anemia in cancer

155

Table 1. Cancers associated with abnormal IL-6 production Lymphomas: Hodgkin’s disease; non-Hodgkin’s lymphoma; chronic lymphocytic leukemia; multiple myeloma Solid tumors: Breast cancer; lung cancer; gastrointestinal cancers (squamous cell carcinoma of the oral cavity; gastric cancer; pancreatic cancer; cholangiocarcinoma; colorectal cancer); malignant melanoma; renal cell carcinoma; ovarian cancer, prostate cancer; bone sarcoma

to hemoglobin or hematocrit level, a signiﬁcant correlation was found between serum IL-6 activity and degree of anemia (Brown et al. 1997; Walther et al. 1998; Hulkkonen et al. 2000; Rokicka-Piotrowicz et al. 2000; Lai et al. 2002; Maccio et al. 2005; Casanovas et al. 2007). In one study including patients with ovarian cancer and evaluating the role of TNF-α, IL-1, and IL-6, IL-6 was found to be the only cytokine, which independently and signiﬁcantly predicted hemoglobin level. In addition, in case reports of patients with IL-6 producing tumors, anemia has been consistently found to be a prominent symptom (Ikeda et al. 1998; Matsumura et al. 2002; Kang et al. 2005). Based on these observations, increased expression and activity of IL-6 appears to be a common phenomenon in cancer and there is evidence suggesting that increased IL-6 concentrations are associated with anemia. As will be discussed below, a possible mechanism, through which IL-6 may contribute to the development of CRA is an upregulation of hepcidin expression. The red cell mass is normally determined by the life span of erythrocytes and the rate of their production. Anemia represents an imbalance between these two factors, and the relative importance of each of these factors will depend on the underlying condition that causes the anemia. In CRA, both factors appear to be involved (Fig. 2) (Nowrousian et al. 1996, 2002). The more important factor, however, seems to be the relative failure of the bone marrow to increase RBC production to compensate sufﬁciently for the shortened RBC survival. The following pathogenic mechanisms are postulated to be involved in this process: 1) impaired iron utilization, 2) suppressed erythroid progenitor cells, and 3) inadequate erythropoietin (EPO) production (Nowrousian 1996, 2000, 2002, 2005) (Fig. 3). A number of experimental and clinical data suggest that these mechanisms may be cytokine-mediated (Nowrousian et al. 1996, 2000, 2002). The purpose of this overview is to present the hematological and biochemical features of CRA and to give an up-to-date summary of its pathogenic mechanisms.

156

M. R. Nowrousian

Tumor Cells Erythrocytes Erythroblasts

AIS RCAS1 Fas-L/TRAIL

Erythrophagocytosis Dyserythropoiesis

TNF-α IFN-γ

Shortened Survival Neopterin TNF-α IFN-α,β

Anemia

Activated Immune System Macrophages

TNF-α IL-1 IFN-β,γ RCAS1

Reduced Erythropoietin Production

IL-6 Liver

TNF-α IL-1 IFN-γ α1-antiHepcidin trypsin

Suppressed Impaired BFU-E, CFU-E, Iron Erythroblasts Utilization

Fig. 3. Pathogenic mechanisms postulated to be involved in the development of cancer-related anemia (CRA). AIS = anemia-inducing substance; RCAS1 = receptorbinding cancer antigen expressed on SiSo cells; Fas-L/TRAIL = Fas-ligand/tumor necrosis factor-related apoptosis-inducing ligand; TNF = tumor necrosis factor; IFN = interferon; IL = interleukin; BFU-E = burst-forming unit erythroid; CFU-E = colony-forming unit erythroid. Adapted from Nowrousian et al. 1996 (reproduced with permission)

Hematological characteristics The hematological features of CRA have been described in a number of studies (Miller et al. 1956; Haurani et al. 1963; Hyman 1963; Cartwright 1966; Lee 1983; Zucker 1985; Cash and Sears 1989), but in these studies, either a small group of patients with cancer (Miller et al. 1956; Haurani et al. 1963; Hyman 1963) or groups of patients with various types of chronic diseases including cancer (Cartwright 1966; Cash and Sears 1989) were evaluated. In a recent study, however, the clinical features of CRA were investigated exclusively in a large group of patients with various types of malignant diseases (Nowrousian et al. 1996). A large group of patients with solid tumors, three groups of patients with hematological malignancies, but without hematopoietic stem cell disorders (multiple myeloma, malignant lymphomas, chronic lymphocytic leukaemia), and two groups of patients with such disorders (myelodysplastic syndromes, myeloproliferative diseases) were included. All

Pathophysiology of anemia in cancer

157

Table 2. Hematological characteristics of patients with various types of malignant diseases Malignancy

No. of patients

RBC ×106/μl

Hb g/dl

Hct %

MCV ﬂ

MCHC g/dl

Rc ×106/μl

MDS CMD CLL MM ML ST

62 22 36 94 61 124

2.9 3.0 2.9 2.7 2.9 3.3

9.3 8.7 9.3 9.2 9.0 9.6

27.4 26.1 28.1 27.0 28.0 29.6

95 88 96 96 94 89

34 34 33 34 33 33

13 27 20 27 24 48

MDS = myelodysplastic syndromes; CMD = chronic myeloproliferative diseases; CLL = chronic lymphocytic leukemia; MM = multiple myeloma; ML = malignant lymphomas; ST = solid tumors; RBC = red blood cells; Hb = hemoglobin; Hct = hematocrit; MCV = mean corpuscular volume; MCHC = mean corpuscular hemoglobin concentration; Rc = reticulocyte count, corrected for hemoglobin level. Data are shown as median. From Nowrousian et al. 1996 (reproduced with permission).

patients were anemic as deﬁned by a hemoglobin (Hb) concentration less than 14 g/dl for males and less than 12 g/dl for females. None of the patients had blood loss, hemolysis, vitamin B12 or folic acid deﬁciency, hepatic or endocrine disorders or iron deﬁciency as deﬁned by a serum ferritin less than 20 μg/l for males and 10 μg/l for females. All patients with solid tumors and the majority of patients with malignant lymphomas were without bone marrow inﬁltration. Three hundred and sixty-eight of the 401 evaluated patients had a creatinine value less than 1.5 mg/dl, and 33 a creatinine value of 1.5 mg/dl or higher. Eighteen of these 33 patients belonged to the group of patients with multiple myeloma. Eighty-three percent of the 401 patients had not received any kind of chemotherapy before the study, and 17% had previously received some form of chemotherapy, but not in the preceding 4 weeks. The hematological features of CRA observed in this study are shown in Table 2. In all patient groups, there was a low reticulocyte count related to the degree of anemia. Median Hb values were less than 10 g/dl in all groups, ranging from 8.7 to 9.6 g/dl between the groups. Patients with solid tumors were signiﬁcantly (p < 0.05) less anemic than those with hematological malignancies, except patients with myelodysplastic syndromes (MDS). The severity of anemia related to the underlying malignancy is demonstrated in Table 3. Fifty-two to 64% of patients with hematological malignancies had Hb values between 8 and 10 g/dl, and 19% to 23% Hb values less than 8 g/dl. The respective proportions of Hb values in patients with solid tumors were 49% and 9%. Regarding the severity of anemia, it has to be

158

M. R. Nowrousian

Table 3. Severity of anemia related to underlying malignancy Malignancy

Myelodysplastic syndromes Chronic myeloproliferative diseases Chronic lymphocytic leukemia Multiple myeloma Malignant lymphomas Solid tumors

Hemoglobin (g/dl) <8

8–10

>10

18.8 22.7 19.4 22.3 21.3 8.9

51.6 63.6 58.3 58.5 57.4 49.2

29.7 13.6 22.2 19.2 21.3 41.9

Data are shown as %. From Nowrousian et al. 1996 (reproduced with permission).

considered that the clinical symptoms of anemia not only depend on the level of Hb, but also on the age of patients, as well as their organ functions, particularly function of the cardiopulmonary system. In cancer patients, even mild (Hb 10–12 g/dl) to moderate (Hb 8–10 g/dl) anemia usually evokes symptoms, and severe (Hb < 8 g/dl) anemia may be life-threatening, since most of these patients are of advanced age, and many of them have comorbidities and reduced organ functions (Nowrousian et al. 1996; Ludwig and Fritz 1996; Ludwig and Nowrousian 2000). The median age of patients in this study was 62 years, ranging from 18 to 90 years. In all patient groups, the median values of MCV (mean corpuscular volume), MCHC (mean corpuscular Hb concentration), and MCH of erythrocytes were within the normal range, but there were differences between the groups regarding the distribution of MCV in each group (Table 4). A signiﬁcantly higher proportion of patients with solid tumors (13.9%) had a microcytic anemia (MCV < 80 ﬂ) compared to that of patients with other types of malignancy (0.0–4.5%), and a signiﬁcantly higher proportion of patients with MDS, chronic lymphocytic leukemia (CLL), multiple myeloma or malignant lymphomas (25.4–38.3%) had a macrocytic anemia (MCV > 100 ﬂ) than that of patients with solid tumors or chronic myeloproliferative diseases (13.6%, 15.6%) (p < 0.0002) (Nowrousian et al. 1996). The results of this study indicate that there are differences between anemias of various types of cancer regarding the severity and the proportions of microcytic and macrocytic subtypes of anemia. In most cases, however, CRA appears to be a hyporegenerative anemia with a reduced reticulocyte count related to the degree of anemia, an Hb value between 8 and 10 g/dl, and MCV, MCHC, and MCH values that are within the normal range. In some earlier studies, a mild to moderate degree of hypochromia of RBCs was reported in cancer patients without microcytosis probably indicating that in

Pathophysiology of anemia in cancer

159

Table 4. Red blood cell characteristics Malignancy

MDS CMD CLL MM ML ST

MCV (ﬂ)

MCH (pg)

MCHC (g/dl)

<80

80–100

>100

<26

26–34

>34

<32

32–36

>36

1.6 4.5 2.8 0.0 3.3 13.9

73.0 81.8 66.7 61.7 70.5 70.5

25.4 13.6 30.6 38.3 26.2 15.6

0.0 0.0 5.6 0.0 4.9 17.2

71.4 100.0 69.4 68.1 72.1 68.9

28.1 0.0 25.0 31.9 23.0 13.9

9.4 13.6 19.4 16.0 21.3 23.4

85.9 81.8 75.0 77.7 75.4 73.4

4.8 4.5 5.6 6.3 3.3 3.2

MDS = myelodysplastic syndromes; CMD = chronic myeloproliferative diseases; CLL = chronic lymphocytic leukemia; MM = multiple myeloma; ML = malignant lymphomas; ST = solid tumors. Data are shown as %. From Nowrousian et al. 1996 (reproduced with permission).

Table 5. Differential diagnosis of iron depletion and cancer-related anemia

Reticulocyte count MCV MCHC Serum ferritin Serum iron Transferrin saturation Serum transferrin receptor

Iron depletion

Cancer

reduced reduced reduced reduced reduced reduced increased

reduced normal normal normal – elevated reduced – normal reduced – normal normal

MCV = mean corpuscular volume; MCHC = mean corpuscular hemoglobin concentration. From Nowrousian 2000 (reproduced with permission).

this group of patients, in contrast to patients with iron deﬁciency anemia (IDA), hypochromia may precede the development of microcytosis (Cartwright 1966; Cash et al. 1989). In this regard, however, it has to be recognized that a part of patients with cancer, particularly those with gastrointentinal tumors, may have some degree of iron deﬁciency at diagnosis because of blood loss. In addition, iron deﬁciency in combination with ACD is not always easy to diagnose (Table 5) (Montecucco et al. 1986; Spell et al. 2004; Margetic et al. 2005; Gasche and Kulnigg 2006; Munoz et al. 2006; Thomas et al. 2006). However, in a recent study, anemia in patients with chronic inﬂammatory diseases and infection was found to be normochromic and normocytic, even if the mean MCH and MCV values were signiﬁcantly

160

M. R. Nowrousian

Table 6. Characteristics of anemia of chronic disease (ACD)* and iron deﬁciency anemia (IDA)

Hb (g/dl) MCV (ﬂ) MCH (pg) Iron (μmol/l) Ferritin (μg/l) Transferrin saturation (%) EPO (mU/ml) IL-6 (pg/ml) IL-10 (pg/ml) TNF-α (pg/ml) Pro-Hepcidin (ng/ml)

Control (n = 27)

ACD (n = 37)

IDA (n = 10)

15.1 92.1 31.0 19.4 113.5 27.9 10.4 2.2 1.1 1.6 110.0

10.7 87.9 29.3 5.5 589.0 13.6 16.6 23.4 5.9 3.2 154.0

10.0 78.4 23.5 7.1 12.9 9.3 37.7 1.4 1.1 1.3 74.6

* autoimmune diseases and infection. Hb = hemoglobin; MCV = mean corpuscular volume; MCH = mean corpuscular hemoglobin; EPO = erythropoietin; TNF = tumor necrosis factor; IL = interleukin. Adapted from Theurl et al. 2006 (reproduced with permission).

lower in this group of patients than in healthy control persons. In patients with IDA, in contrast, anemia was clearly hypochromic and microcytic (Table 6) (Theurl et al. 2006).

Shortened red blood cell survival In patients with chronic diseases, RBCs usually survive 60–90 days, compared with 120-day RBC survival in healthy persons (Zucker 1985, Mitlyng et al. 2006). In addition, when RBCs from healthy subjects are transfused into patients with advanced cancer, the life span of these cells is shortened (Hyman et al. 1956). There are clinical and experimental data indicating that this effect may be mediated by IL-1 and TNF-α. In patients with rheumatoid arthritis, the shortened RBC survival correlates with the levels of IL-1 (Salvazini et al. 1991), and repeated administration of TNF-α to rats produces dyserythropoiesis (Ulich et al. 1990) and anemia by decreasing RBC synthesis and reducing the life span of RBCs (Moldawer et al. 1989). TNF-α, in addition, is able to induce dyserythropoiesis and erythrophagocytosis in mice that carry experimental malaria (Clark and Chaudhri 1988). Such effects of TNF-α may be involved in reducing erythropoiesis and shortening RBC survival in patients with CRA (Fig. 3). The inhibitory effect of TNF-α on erythroid progenitor cells is reported to result from a prolonged cell cycle

Pathophysiology of anemia in cancer

161

duration due to a decreased level of cyclin-dependent kinase 6 (CDK 6) (Dai et al. 2003). Recently, a protein named anemia-inducing substance (AIS) has been identiﬁed in plasma of patients with advanced cancer that decreases the osmotic resistance of RBCs (Honda et al. 1995). This substance has also been found to be present in both the cytosol and the nuclear fraction of neoplastic cells. The underlying mechanism of the increased osmotic fragility of the RBCs induced by AIS appears to be inhibition of metabolism (glucose inﬂux, pyruvate kinase activity, and ATP concentration) of these cells. AIS seems to have not only an impact on the osmotic resistance of RBCs, but also to suppress the cellular immunity as measured by mitogen-stimulated lymphocyte proliferation. In addition, it appears to be speciﬁc to malignant diseases and not to be present in chronic inﬂammatory disorders (Ishiko et al. 1999). In a rabbit model, a sustained elimination of AIS appeared to be possible by cyclic perfusion of plasma, a method that might have the potential for clinical application in anemic cancer patients (Ishiko et al. 1999).

Impaired iron utilization The distinctive feature of ACD is a low level of serum iron, total iron-binding capacity, and transferrin saturation in the presence of adequate iron stores as measured by serum ferritin or bone marrow examination (Cartwright 1966; Lee 1983; Zucker 1985; Theurl et al. 2006). In iron deﬁciency anemia, in contrast, not only serum iron, but also iron stores are reduced (Tables 5, 6) (Nowrousian 2002; Theurl et al. 2006). In ACD, there is an inverse correlation between Hb and ferritin and markers of activated immune and inﬂammatory system such as IFN-γ and neopterin, indicating that activated cells from this system, particularly macrophages, may be involved in the alteration of iron metabolism and development of ACD (Cash and Sears 1989; Fuchs et al.1993). Human mononuclear phagocytes downregulate their transferrin receptor (TFR) expression and ferritin content, when they are exposed to IFN-γ [51]. The reduction in iron incorporation, however, is quantitatively lower than the reduction in TFR-expression and ferritin content. As a result, the ferritin remaining in these cells is approximately three times more saturated with incorporated iron than the ferritin in nonactivated cells (Byrd and Horwitz 1993). The ferritin in IFN-γ-activated monocytes, in addition, appears to be of the high-iron type, a type of ferritin that takes up iron rapidly and releases it slowly relative to the low-iron type (Byrd and Horwitz 1993; Mertz and Theil 1983). Furthermore, IFN-γ seems to increase the uptake of nontransferrin bound iron into monocytes and to downregulate the expression of ferroportin mRNA, a putative iron exporter, and thus to decrease the release of iron from these cells (Ludwiczek et al. 2003). Such quantitative and qualitative changes of iron metabolism in activated monocytes and

162

M. R. Nowrousian

Activated Immune System IL-6

TNF-α

IFN-γ

Mono Ferritin: nuc Concentration Iron Saturation lear High-Iron Type Pha Ferrogo portin cytes Iron

TNF, IL-1, IL-6

Downregulation of IFN-γ EPO Receptor

TNF-α

α1-Antitrypsin

Decreased TFR Expression

Impaired Binding of TF to TFR

Erythroid Progenitor Cells

Decreased TFR Expression

Transferrin-Iron Complex Hepcidin

Blood Iron Ferroportin

Duodenum (Enterocytes)

Fig. 4. Pathogenic mechanisms possibly involved in the impairment of iron utilization in cancer-related anemia (CRA). TNF = tumor necrosis factor; IFN = interferon; IL = interleukin; TF = transferrin; TFR = transferrin receptor. Adapted from Nowrousian 2000 (reproduced with permission)

macrophages may reduce the availability of iron and be involved in the development of ACD (Fig. 4) (Nowrousian et al. 1996, 2002). Iron metabolism can also be altered by TNF-α (Fig. 4). Rats and mice treated with TNF-α develop anemia and hypoferremia, and the hypoferremia is associated with a signiﬁcant decrease of iron release from macrophages and incorporation of iron into RBCs (Moldawer et al. 1989; Alvarez-Hernandez et al. 1989). In a clinical study, isolated limb perfusion with rhIFN-γ and rhTNF-α resulted in hypoferremia, ferritin production and decrease in circulating transferrin receptors in cancer patients (Feelders et al. 1998). In mice, systemic exposure to TNF-α produced hypoferremia through an increased iron deposition in the spleen and a reduced duodenal iron transfer, and in a human ex vivo small bowel culture system, TNF-α signiﬁcantly reduced both iron import as well as iron export by inducing the expression of a ferritin heavy chain and a decrease in concentrations of iron transporters DMT-1 (divalent metal transporter) and ferroportin (Sharma et al. 2005; Laftah et al. 2006). In both studies, TNF-α exerted its effects independent from hepcidin expression. On the other hand, there is evidence suggesting that TNF-α has some regulatory effects on the expression of hepcidin as well

Pathophysiology of anemia in cancer

163

A

800

C

25

IL-6 (pg/ml)

600

Fe (μ μmol/l)

other iron regulatory factors such as DMT1, IREG1, and TfR2 (Johnson et al. 2004; Dzikaite et al. 2006). In cultures of rat hepatocytes, TNF-α directly increased hepcidin gene (HAMP) mRNA and indirectly the synthesis of hepcidin by downregulating repulsive guidance molecule C (Rgmc) or upregulating IL-1β and IL, or both (Constante et al. 2007). Other cytokines that may be directly or indirectly involved in modulation of iron metabolism are IL-1 and IL-6. IL-1 is able to increase the production of ferritin (Rogers et al. 1991), which could act as a trap for iron that would otherwise be available for erythropoiesis (Means and Krantz 1992). IL-1 may also be involved in inducing the expression of the iron regulatory hormone, hepcidin, but the predominant cytokine involved in hepcidin synthesis appears to be IL-6 (Nemeth et al. 2003, 2004; Inamura et al. 2005; Lee et al. 2005). IL-6 treatment has been shown to stimulate the expression of hepcidin in isolated hepatocytes in vitro, and treatment of human subjects with IL-6 has been found to result in an increased urinary excretion of hepcidin and a decrease in serum iron (Fig. 5) (Nemeth et al. 2003, 2004). IL-6 directly regulates hepcidin expression through induction and subsequent promoter binding of signal transducer and activator of transcription 3 (STAT3) (Hershko 2006; Wrighting and Andrews 2006; Falzacappa et al. 2007).

20

400 200

15 10 P < 0.001

0

5

B

D

0.6 0.5

300

Fe/TIBC

Hepcidin (ng)/ creat (mg)

1000

100

30

P < 0.001

P < 0.001

P < 0.001

0.4 0.3 0.2 0.1

P < 0.001

0.0

0h

3h inf

2h after

24 h

0h

3h inf

2h after

Fig. 5. Effects of a 3-h-infusion of IL-6 (30 μg/h) on serum values of IL-6 (A), hepcidin (B), iron (C), and iron/total iron-binding capacity (Fe/TIBC) (D) in healthy individuals. IL = interleukin. From Nemeth et al. 2004 (reproduced with permission)

164

M. R. Nowrousian

Macrophage and hepatocyte iron release

Enterocytes iron release

Hepatocytes Hepcidin

Macrophage and hepatocyte iron release Decreased ferroportin Enterocytes iron release

Iron

Iron

Increased ferroportin

Inflammation, Malignancy (IL-6) Iron load Iron

Iron

Erythropoietic activity (anemia, hypoxia, EPO) Iron restriction

Fig. 6. Factors regulating hepcidin expression and the effects of increased or decreased hepcidin activity on ferroportin and the release of iron from enterocytes, hepatocytes and macrophages

Hepcidin is a newly identiﬁed cytokine that appears to play a major role in the development of ACD (Nicolas et al. 2002; Weinstein et al. 2002; Leong and Lönnerdal 2004; Nemeth and Ganz 2006; Ganz 2006, 2007). It is a 25-amino acid cationic peptide, predominantly expressed in the liver and identiﬁed as a part of the innate immune response. Hepcidin negatively regulates iron release by enterocytes, macrophages, and hepatocytes into circulation (Lou et al. 2005; Rivera et al. 2005; Andriopoulos and Pantopoulos 2006). Hepcidin expression is induced by iron load and inﬂammation and is suppressed by iron restriction and erythropoiesis (Fig. 6). Among the mechanisms involved in the induction and suppression of hepcidin, mainly those associated with inﬂammation are deﬁned. A key factor appears to be IL-6 (Nemeth et al. 2003, 2004; Papanikolaou et al. 2005; Ganz 2007). In anemia of inﬂammation, the production of hepcidin is considerably increased resulting in sequestration of iron in macrophages (Fig. 4) (Leong and Lönnerdal 2004; Theurl et al. 2006). Hypoxia and the use of EPO, in contrast, dramatically decrease hepatic hepcidin expression, most probably due to an increase in erythropoiesis and consequently iron requirement (Nicolas et al. 2002; Krijt et al. 2004, 2006; Kattamis et al. 2006; Leung et al. 2006; Pak et al. 2006; Vokurka et al. 2006; Wilkins et al. 2006). These ﬁndings show that in conditions with elevated hepcidin level, the use of EPO may not only have a stimulatory effect on erythropoiesis but also indirectly improve the availability of iron (Nicolas et al. 2002).

Pathophysiology of anemia in cancer

165

Hepcidin exerts its effect on iron metabolism by binding to ferroportin and inducing its internalization and lysosomal degradation. Other ironrelated proteins such as divalent metal transporter 1 (DMT1) and duodenal cytochrome b (Dcytb) may also be inhibited by hepcidin, but the more important factor appears to be ferroportin (Viatte et al. 2005). Ferroportin is a membrane protein of cells engaged in iron metabolism such as enterocytes, macrophages, and hepatocytes. It is an iron transporting protein, which is solely responsible for efﬂux of iron from these cells into plasma. Increased concentrations of hepcidin decrease the amount of ferroportin and thus the release of iron into circulation resulting in an accumulation of iron in the above-mentioned cells. Decreased concentrations of hepcidin, in contrast, increase the release of iron (Fig. 6) (Nemeth et al. 2004; Delaby et al. 2005; Rivera et al. 2005). There is little known about the role of hepcidin in CRA. No systematic evaluations have yet been performed and there are some phenotypic differences between CRA and hepcidin-induced anemia. Hepcidin, as observed in animal experiments and in patients with hepcidin-producing tumors, characteristically induces a microcytic and hypochromic anemia (Nicolas et al. 2002; Roy et al. 2003, 2007; Rivera et al. 2005). Anemia of inﬂammation, which is supposed to be at least in part due to increased hepcidin expression, also shows a trend toward a microcytic and hypochromic feature (Table 6) (Theurl et al. 2006). CRA, in contrast, is usually normocytic and normochromic indicating that pathogenic factors other than hepcidin may be predominantly or additionally involved (Nowrousian et al. 1996). On the other hand, there is indirect evidence suggesting that hepcidin may play a role in this type of anemia. As discussed above, a cytokine predominantly involved in inducing hepcidin expression is IL-6 (Fig. 6). Increased concentrations of IL-6, a possible surrogate for hepcidin, have been observed in a wide variety of cancers, depending on the activity and stage of disease. In addition, in a number of malignant diseases, including hematological malignancies as well as solid tumors, correlations have been found between the degree of anemia and the activity of IL-6 (Table 1). A prototype of tumors in which hepcidin could be a major pathogenic factor for developing anemia is multiple myeloma. IL-6 is a growth and proliferation factor for myeloma cells, which may be able to produce this cytokine by an autocrine mechanism (Fig. 7). The more important source of IL-6, however, are myeloma-stimulated stroma cells, such as macrophages, ﬁbroblasts, endothelial cells, and osteoblasts (Bataille et al. 1989; Thabard et al. 1999, 2001; Barille et al. 2000; Lauta 2001; Bommert et al. 2006; Karadag et al. 2006; Arnulf et al. 2007). As shown in vitro, stroma cells from patients with multiple myeloma appear to be primed to produce greater amounts of IL-6, even in the absence of myeloma cells (Arnulf et al. 2007). Multiple myeloma, however, does not represent an exception among various types of tumors, including hematological malignancies as well as solid tumors,

166

M. R. Nowrousian

Bone marrow

Myeloma cell Liver

IL-6Rα paracrine

IL-6 Stroma cells

autocrine

IL-6

-6 IL

Hepcidin

IL-6

Osteoclasts Suppressed erythroid progenitors

Bone

Impaired iron metabolism

Anemia

Fig. 7. Autocrine and paracrine secretion of IL-6 in multiple myeloma and its possible role in inducing anemia. IL = interleukin; IL-6R = IL-6 receptor

regarding the increased IL-6 activity and its possible role as a growthpromoting factor (Table 1). Increased IL-6 concentrations are found in many types of cancer and there is evidence suggesting that this cytokine may be involved both in the development and progression of tumor cells as well as in the pathogenesis of clinical symptoms, such as anemia, cachexia, and fatigue (Cozen et al. 2004; Barton 2005; Schubert et al. 2006; Jenkins et al. 2007). Interestingly, the use of rhEPO in anemic patients with multiple myeloma has been found to signiﬁcantly inhibit serum concentrations of IL-6 suggesting that this drug may reduce hepcidin expression not only by stimulating erythropoiesis but also by reducing IL-6 activity (Fig. 8) (Prutchi-Sagiv et al. 2006). In an animal model of cachexia in mice, the use of rhEPO has also been observed to reduce IL-6 production (Kanzaki et al. 2005), and in TNFα-treated human brain microvascular endothelial cells, rhEPO was found to downregulate the IL-6 gene (Avasarala and Konduru 2005). A further possible mechanism responsible for impaired iron metabolism in ACD could be an alteration of TFRs on erythroid cells. Erythroblasts of patients with ACD have a decreased number of TFRs, and TFRs on these cells have a lower afﬁnity to transferrin than those on erythroblasts from healthy individuals (Feelders et al. 1993). In the course of infections, malignancies, and immunologic disorders, IL-1, IL-6, and TNF increase the level of the acute phase protein α1-antitrypsin that is able to inhibit erythropoiesis

Pathophysiology of anemia in cancer

167

Serum IL-6 level 14

** 12

pg/mL

10 P<0.05

8

P<0.05

* 6 4 2 0 Healthy

Advanced MM

Advanced MM + EPO

Fig. 8. Increased serum level of IL-6 in patients with multiple myeloma (MM) compared with healthy individuals and the effect of treatment with recombinant human erythropoietin (rhEPO) on serum IL-6 concentration in MM patients. From Prutchi-Sagiv et al. 2006 (reproduced with permission)

by impairing transferrin binding to TFR and subsequent internalization of the TFR-transferrin complex (Fig. 4) (Graziadei et al. 1994). In patients with anemia of cancer or other chronic diseases, the initially normal or elevated levels of ferritin decrease when the patients are treated successfully with rhEPO (Ludwig et al. 1990; Oster et al. 1990; Vreugdenhil et al. 1992; Ludwig et al. 1994; Means 1995). In cancer patients, the behavior of serum ferritin during the early phase of treatment appears to be even predictive of response to rhEPO (Ludwig et al. 1994). These observations may suggest that pharmacologically high dosages of rhEPO are able to overcome the impairment of iron metabolism in a part of patients with CRA or ACD. A possible explanation may be that EPO decreases the expression of hepcidin and thus increases the release of iron from the reticuloendothelial system and, additionally, it upregulates the expression of ironregulatory protein and transferrin receptor in erythroid cells and thus the uptake of iron in these cells (Fig. 4) (Weiss et al. 1997).

Suppressed erythroid progenitor cells Another mechanism that could potentially contribute to the development of CRA is a suppressing effect of IFN-γ, IL-1, TNF-α and hepcidin on erythroid progenitor cells such as CFU-E. These cytokines are known to inhibit

168

M. R. Nowrousian

erythropoiesis in vitro as well as in vivo, and they are also known to do this in synergistic fashion or enhance each other’s expression (Broxmeyer et al. 1986; Balkwill et al. 1987; Roodman 1987; Roodman et al. 1987; Schooley et al. 1987; Johnson et al. 1988; Furmanski and Johnson 1990; Means et al. 1992; Means and Krantz 1993; Ulich et al. 1993; Means 1995; Sloand et al. 2002). The inhibitory effect of IL-1 requires the presence of T lymphocytes, and is mediated by IFN-γ, and the inhibitory effect of TNF-α requires the presence of marrow stroma cells, and is mediated by IFN-β (Fig. 2) (Means et al. 1992; Means and Krantz 1993; Means 1995). TNF-α also appears to have a direct impact on erythroid cells by potentiating their apoptosis (Tsushima et al. 1999). IFN-γ can bind to high afﬁnity receptors of these cells (Taniguchi et al. 1997) and induce their apoptosis by upregulation of caspases 1, 3, and 8 activity (Means and Krantz 1993; Dai and Krantz 1999). In addition, it is able to downregulate the expression of EPO receptors, a mechanism which may further potentiate the apoptosis of erythroid progenitor cells (Taniguchi et al. 1997). The inhibitory effect of IFN-γ on CFU-E in vitro can be overcome by very high, pharmacological concentrations of EPO (Means and Krantz 1991) and the suppressive effects of IL-1 and TNF-α on CFU-E in mice by administration of EPO (Johnson et al. 1989, 1990). In addition, hepcidin has ben reported to be suppressive to CFU-E only at reduced EPO concentrations (Dallalio et al. 200). These results indicate that the therapeutic effect of EPO observed in patients with CRA may be due in part to an overcoming of suppressive effects of these cytokines on erythroid cells. A factor recently identiﬁed to have apoptotic effects on erythroid progenitor cells is the receptor-binding cancer antigen expressed on SiSo cells, the so called RCAS1. It is a human cancer-associated antigen, which is expressed on the surface and in the cytoplasm of cells from various types of tumors including skin, breast, lung, gastric, pancreatic, hepatocellular, cholangiocellular, gallbladder, colorectal, uterine, and ovarian cancers as well as cancer of the oral cavity and Hodgkin’s disease (Izumi et al. 2001; Oshikiri et al. 2001; Oshima et al. 2001; Oizumi et al. 2002; Ikeguchi et al. 2003; Fukuda et al. 2004; Nakamura et al. 2004; Enjoji et al. 2005; Kato et al. 2005). RCAS1 is also expressed in activated monocytes and macrophages (Suehiro et al. 2001, 2005) In addition, it occurs in circulation and is used as a biomarker for a number of malignant diseases (Watanabe et al. 2003; Enjoji et al. 2005; Yamaguchi et al. 2005; Coban et al. 2006; Sonoda et al. 2006, 2007). It is reported to be apoptotic for T cells, natural killer (NK) cells, and colonyforming erythroid progenitor cells, which express putative receptors for RCAS1, particularly at early stages of their maturation. The apoptotic effect of RCAS1 on erythroid progenitor cells appears to be independent from activation of the death receptor Fas (Koury 2001; Matsushima et al. 2001; Suehiro et al. 2005). Considered these ﬁndings, increased concentrations of

Pathophysiology of anemia in cancer

169

RCAS1 in serum may be a contributing factor for developing anemia in cancer. In case of bone marrow inﬁltration, tumor cells may have a direct toxic effect on erythropoietic progenitor cells shortening the survival of these cells. In a study in patients with multiple myeloma, the severity of anemia was found to inversely correlate with the proliferative activity of myeloma cells (Fossa et al. 1999), and in other studies, highly proliferating myeloma cells were reported to have an up-regulated expression of the apoptogenic receptors Fas-L and TRAIL resulting in a suppression of erythropoiesis by inducing maturation arrest and apoptosis of early and intermediate stages of erythroblasts and by accumulation of early erythroblasts possibly resulting from an upregulated expression of the apoptogenic enzyme FLICE (caspase 8) and consequently a reduced activity of the maturation and survival factor GATA-1 in these cells (Fig. 9) (Silvestris et al. 2001, 2002). Erythroblasts both at early and intermediate stages of maturation were found to overexpress the

Erythroblasts Prebasophilic/basophilic (immature)

Polychromatophilic (semimature) GATA-1

GpA+ dim Fas+ FLICE+ DR4+, DR5+

Accumulation

Maturation arrest, apoptosis

Orthochromatic (mature) GATA-1

GpA+ interm DR4+, DR5+ Fas+ ICE+ Caspase-10+

Fas-L+ TRAIL+

Maturation arrest, apoptosis

GpA+ bright Fas-L+ TRAIL+ ICE+ Caspase-10+

Highly malignant myeloma cells

Fig. 9. Negative regulation of erythroblast maturation by Fas-L+/TRAIL+ highly malignant myeloma cells. GATA-1 = the transcription factor necessary for the terminal differentiation of the erythroid precursors; GpA = glycophorin A; Fas = Apo-1/ CD95, agonist receptor for Fas-L (Fas-ligand); DR4 and DR5 = agonist receptors for TRAIL (tumor necrosis factor-related apoptosis-inducing ligand); ICE = caspase 1; FLICE = caspase 8. Adapted from Silvestris et al. 2002 (reproduced with permission)

170

M. R. Nowrousian

Survival and differentiation

IFN-gg

TNF-a a

- Downregulation of EPO receptor - Upregulation of caspases 1,3 and 8 - Upregulation of Fas antigen (CD95)

Erythroid progenitor cells

- Expression of Bcl-XL and Bcl-2 - Activation of kinase 1 - Activation of ironregulatory protein and upregulation of TFR

Apoptosis

EPO

Fig. 10. Effects of TNF-a, IFN-g and EPO on erythroid progenitor cells

apoptogenic receptors Fas for Fas-L and DR4 and DR5 for TRAIL, but early erythroblasts appeared to be particularly sensitive to Fas-L and those at intermediate stage to TRAIL (Silvestris et al. 2002). The increased erythroblast expression of Fas, DR4, and DR5 may be induced by inﬂammatory cytokines such as TNF-α and IFN-γ (Dai et al. 1998; Tsushima et al. 1999). Other factors that could potentially increase the apoptotic effect of the Fas-L/TRAIL system on erythroid progenitor cells may be downregulation of EPO receptor in these cells by TNF-α and IFN-γ (Fig. 10) and the defective EPO response in anemic cancer patients, as will be discussed below. However, there is evidence suggesting that the expression of the Fas, and its ligand, FasL can be suppressed by EPO. In a study in mice, a single injection of EPO speciﬁcally inhibited early erythroblast Fas and FasL mRNA leading to a dramatic increase in the frequency of erythroblasts (Liu et al. 2006).

Inadequate erythropoietin production EPO, a glycoprotein hormone, produced in adults mainly by the kidney, is crucial to the regulation of RBC production. It promotes the survival, proliferation and differentiation of erythroid progenitor cells by binding to speciﬁc receptors on these cells and by repressing their apoptosis through Bcl-XL and Bcl-2 expression and promoting their differentiation by kinase-1 activation (Koury and Boundurant 1990; Silva et al. 1996; Gregory et al. 1999; De Maria et al. 1999; Nagata et al. 1999). Considering theses effects, EPO

Pathophysiology of anemia in cancer

171

Erythropoiesis and erythropoietin (EPO) Colony-forming cells

Pluripotent stem cells

BFU-E, CFU-E

Morphologically identifiable cells

ErythroEryth blasts

Reticulocytes

Mature cells

Erythrocytes

Cells responsive to EPO

Fig. 11. Erythroid progenitor cells and their relation to EPO. EPO stimulates the proliferation of BFU-E (burst-forming unit-erythroid) and CFU-E (colony-forming unit-erythroid) and postpones the apoptosis of these cells and erythroblasts. In addition, it upregulates the expression of transferrin receptor on erythroid progenitor cells. From Nowrousian 2000 (reproduced with permission)

may be regarded as an antagonist of inﬂammatory cytokines such as TNF-α and IFN-γ (Fig. 10). EPO also appears to inﬂuence the iron metabolism by activation of iron-regulatory protein and upregulation of transferrin receptor expression in erythroid precursors (Weiss et al. 1997) and by suppression of hepcidin expression in the liver (Nicolas et al. 2002; Leong and Lönnerdal 2004). The latter, however, has been reported to result from EPO-stimulated erythropoiesis and, consequently, increased demand for iron (Nicolas et al. 2002; Krijt et al. 2004, 2006; Kattamis et al. 2006; Leung et al. 2006; Pak et al. 2006; Vokurka et al. 2006; Wilkins et al. 2006). The erythroid cells responsive to EPO are the early progenitor cells BFU-E (burst-forming unit erythroid), the late progenitor cells CFU-E and erythroblasts (Fig. 11) (Nowrousian 2000, 2002, see also Chapters 1, 6 in this book). Serum EPO concentration is directly related to the rate of its production in the kidney. On the other hand, EPO is consumed by erythroid progenitor cells, and there is an inverse relationship between the mass of these cells in the bone marrow and serum EPO concentration (Cazzola et al. 1998). The production of EPO, however, is primarily regulated by tissue oxygenation. Hypoxia is the most powerful stimulus for EPO production with serum concentrations inversely correlating with hemoglobin levels (Caro et al. 1979). Based on an assumed serum half life of 5–9 h and a mean distribution volume of 0.07 l kg−1, as measured in pharmacokinetic studies using recombinant human EPO, the rate of endogenous EPO production is estimated to be

172

M. R. Nowrousian

normally about 2–4Ukg−1 24h−1 (Eckardt and Kurtz 2005). The normal reference range for EPO in serum is reported to be 2.5–20 mU/ml or 5.0–25 mU/ml, depending on the use of enzyme-like immunosorbant assay (ELISA) or chemiluminescence assay (CLIA), respectively (Marsden et al. 2006). In patients with CRA, erythroid progenitor cells respond normally to EPO (Dainiak et al. 1983), but there appears to be a disturbed EPO response to anemia. The latter has been the subject of many studies, but the results of earlier studies utilizing bioassays for analyzing EPO response are to be questioned because of the insensitivity of these assays (Firat and Banzon 1971; Ward et al. 1971; Zucker et al. 1974), and the results of recent studies utilizing sensitive immunoassays are in part contradictory (Schreuder et al. 1984; Cox et al. 1986; Miller et al. 1990; Nielsen et al. 1990; Ariad et al. 1992; Beguin et al. 1992; Cazzola et al. 1992; Urabe et al. 1992; Kettelhack et al. 1994; Beguin et al. 1996; Dowlati et al. 1997; Corazza et al. 1998; Ozguroglu et al. 2000; Kim et al. 2002). In nonuremic patients with multiple myeloma or patients with malignant lymphomas, EPO response was found to be adequate in two studies (Nielsen et al. 1990; Ariad et al. 1992), and to be reduced in six studies (Miller et al. 1990; Beguin et al. 1992; Urabe et al. 1992; Kostova and Siljanovski 2004, 2005; Shen et al. 2005). In patients with solid tumors, EPO response was reported to be normal in two studies (Schreuder et al. 1984; Corazza et al. 1998), and to be reduced in four studies (Cox et al. 1986; Cazzola et al. 1992; Kettelhack et al. 1994). The expected linear relation between EPO levels and Hb values was found to be absent in two studies (Cox et al. 1986; Ozguroglu et al. 2000). In some studies, either a small number of patients was evaluated (Ariad et al. 1992; Cazzola et al. 1992) or only mean EPO levels were used to compare EPO response in patients with cancer to that in control subjects (Cox et al. 1986; Kettelhack et al. 1994; Arslan et al. 2005) or EPO concentrations in anemic cancer patients were compared to EPO levels in nonanemic control subjects (Schreuder et al. 1984). An adequate analysis of EPO response, however, can only be performed when EPO level and concurrent Hb or hematocrit (Hc) value are correlated individually in each patient, and the result is compared with an appropriate EPO response to that degree of anemia (Miller et al. 1990; Beguin et al. 1992; Urabe et al. 1992; Nowrousian et al. 1996, 2000, 2002; Corazza et al. 1998; Ozguroglu et al. 2000; Capalbo et al. 2002; Kostova and Siljanovski 2004, 2005). In cancer patients, EPO production may additionally be altered by tumor or therapy-related renal dysfunction (Beguin et al. 1992; Wood and Hrushesky 1995) or chemotherapy (Wood and Hrushesky 1995; Schapira et al. 1990; Pohl et al. 1992; Cerruti et al. 1994; Sawabe et al. 1998; Lee et al. 2001; Ramesh et al. 2002; Schrier 2002; Johansson and Andreasson 2006), particularly agents that inhibit RNA synthesis (Jelkmann et al. 1994). It is, therefore, necessary that patients with renal dysfunction or recently given chemotherapy are investigated separately.

Pathophysiology of anemia in cancer

173

In a study concerning these factors, EPO response was evaluated in a large number of anemic cancer patients with various types of malignancy (Nowrousian et al. 1996). Serum EPO levels were determined using a solidphase enzyme immunoassay with a detection limit of 1 mU/ml, and a normal range of 4.4 + 2.9 (mean + SD) mU/ml, evaluated in 99 healthy adults. EPO values were log transformed and individually correlated to concurrent Hc level in each patient, and for that Hc level, the predicted log EPO level was calculated, based on the data from 102 subjects with a normal EPO response to anemia (Beguin et al. 1992). Since EPO levels are best evaluated in relation to the degree of anemia, and this relation is best expressed in the ratio of observed-to-predicted log EPO levels (O/P ratio), O/P ratios were determined in all patient groups to evaluate the appropriateness of EPO response in each group, and to compare EPO responses in patient groups with various types of malignancy (Nowrousian et al. 1996). In this study, a weak, but signiﬁcant inverse log linear correlation between EPO and Hc levels was observed in patients with solid tumors, multiple myeloma, malignant lymphomas, CLL or MDS. In patients with chronic myeloproliferative diseases, in contrast, EPO and Hc levels did not correlate signiﬁcantly. The median EPO levels observed in various patient groups and the corresponding median O/P ratios are shown in Table 7, and the distribution of O/P ratios in Fig. 12. The median O/P ratio in patients with MDS was markedly above 1 and signiﬁcantly higher than that in all other groups of patients. The median O/P ratio in patients with CLL was nearly 1 and signiﬁcantly higher than that in patients with malignant lymphomas, multiple myeloma or solid tumors. The median O/P ratio in patients with chronic

Table 7. Serum EPO levels and O/P ratios related to underlying malignancy Malignancy

MDS CMD CLL MM ML Solid tumors

Serum EPO mU/ml*

512 72 87 42 39 26

O/P ratio*

1.19 0.88 0.93 0.76 0.75 0.73

Distribution of O/P ratios <1.0

≥1.0

37.5% 59.1% 52.8% 79.8% 80.3% 86.3%

62.5% 40.9% 47.2% 20.2% 19.7% 13.7%

EPO = erythropoietin; O/P = ratio of observed to predicted log serum EPO levels; MDS = myelodysplastic syndromes; CMD = chronic myeloproliferative diseases; CLL = chronic lymphocytic leukemia; MM = multiple myeloma; ML = malignant lymphomas; ST = solid tumors. *Data are shown as median. From Nowrousian et al. 1996 (reproduced with permission).

174

M. R. Nowrousian

2,5

O/P Log EPO

2,0 1,5 1,0 0,5 0,0 CLL

CMD

MDS

ML

MM

ST

Fig. 12. Distribution of O/P ratios (ratio of observed to predicted log EPO levels) in anemic patients with various types of malignant disease. From Nowrousian et al. 1996 (reproduced with permission)

myeloproliferative disease was slightly below 1 and signiﬁcantly higher than that in patients with solid tumors. The median O/P ratio in patients with malignant lymphomas, multiple myeloma or solid tumors was markedly below 1, and there was no signiﬁcant difference between these groups of patients regarding their O/P ratios. In patients with multiple myeloma, the median O/P ratio was signiﬁcantly lower in those with creatinine values above 1.5 mg/dl than in those with creatinine values of 1.5 mg/dl or less (0.47 vs 0.78) (Fig. 13). The proportion of patients with O/P ratios less than 1 was 86% in patients with solid tumors, 80% in patients with multiple myeloma or malignant lymphomas, 59% in patients with chronic myeloproliferative diseases, 53% in patients with CLL, and 38% in patients with MDS (Table 7) (Nowrousian et al. 1996). EPO response in patients with MDS has also been evaluated in other studies indicating that the majority of these patients have a normal or increased EPO response (Jacobs et al. 1989; Bowen et al. 1990; Merchav et al. 1990; Aul et al. 1991; Bourantas et al. 1995). Considering the results of these studies, EPO response in patients with CRA appears to differ considerably depending on the type of underlying malignancy (Jacobs et al. 1989; Miller et al. 1990; Urabe et al. 1992; Bourantas et al. 1995; Nowrousian et al. 1996). Most patients with malignant lymphomas, multiple myeloma or solid tumors have a signiﬁcantly reduced EPO response related to the degree of anemia, similar to that observed in

Pathophysiology of anemia in cancer

O/P Log EPO

1,4

175

p = 0.0001

1,2 1,0 0,8 0,6 0,4 0,2 0,0 >1.5 mg/dl

≤1.5 mg/dl

Serum creatinine Fig. 13. Distribution of O/P ratios in anemic patients with multiple myeloma related to the serum creatinine level

patients with ACD (Baer et al. 1987; Spivak et al. 1989; Noé et al. 1994; Kostova and Siljanovski 2004, 2005; Shen et al. 2005). In these patients, the relative EPO deﬁciency may be a pathogenic factor that contributes to decreased erythropoiesis and development of anemia. In patients with multiple myeloma, the defective EPO response seems to worsen further when the creatinine level rises above 1.5 mg/dl (Nowrousian et al. 1996; Kostova and Siljanovski 2004). The impaired EPO response seen in CRA may be the result of suppressive effects of neopterin, TNF-α or IL-1-α or β on EPO-producing cells. These cytokines are able to inhibit the production of EPO in human hepatoblastoma cell cultures and/or in isolated perfused rat kidneys (Faquin et al. 1992; Jelkmann et al. 1992; Wolff and Jelkmann 1993; Noé et al. 1994; Vannucchi et al. 1994; Pagel et al. 1999; Capalbo et al. 2002). TNF-α has also been found to inhibit serum EPO levels in patients with advanced cancer who were receiving this drug (Braczkowski et al. 2001). The suppressive effects of IL-1 and TNF on EPO production of hepatoblastoma cells appear to occur at the level of the EPO mRNA (Faquin et al. 1992). In patients with multiple myeloma or Waldenström’s disease, suppression of EPO response may be additionally induced by hyperviscosity, since in these patients, EPO levels are inversely related to plasma viscosity, and at higher plasma viscosities, reduction in EPO levels is parallel to decreases in renal EPO mRNA (Singh et al. 1993).

176

M. R. Nowrousian

Summary Patients with cancer frequently suffer from anemia, and there are numerous causes that could have produced the anemia. In a considerable number of patients, however, no other explanation is found for the development of anemia except the malignant disease itself. Such a CRA shares many similarities with anemias that occur in other types of chronic diseases. In most cases, CRA is a hyporegenerative, normocytic and normochromic anemia characterized by a normal or reduced serum iron and transferrin saturation despite a normal or elevated ferritin level. Recent investigations show that CRA may be the result of an activation of the immune and inﬂammatory system, and certain cytokines such as IFNs, TNF-α, IL-1, and IL-6 could potentially be involved in its development. Concentrations of these cytokines are found to be increased in patients with cancer correlating with the degree of anemia. Another cytokine possibly involved may be hepcidin, which is induced by IL-6 and appears to play a major role in ACD. In CRA, the RBC survival is shortened, but the more important factor for the development of anemia is a relative failure of erythropoiesis to compensate sufﬁciently for the shortened RBC survival. The pathogenic mechanisms postulated to be involved in this process are: 1) impaired iron utilization, 2) suppression of erythroid progenitor cells, and 3) inadequate erythropoietin production. Most patients with CRA have a reduced EPO response related to the degree of anemia, and administration of rhEPO has been shown to correct the anemia in a part of patients. There are experimental and clinical studies indicating that pharmacological dosages of rhEPO may not only correct the relative EPO deﬁciency, but also overcome the suppression of erythroid progenitor cells and the impairment of iron mobilization.

References 1. Abels R, Gordon D, Nelson R, et al (1991) Transfusion practice in advanced cancer patients. Blood 78 [Suppl 1]: 474a 2. Alvarez-Hernandez X, Liceaga J, McKay IC, et al (1989) Induction of hypoferremia and modulation of macrophage iron metabolism by tumor necrosis factor. Lab Invest 61: 319–322 3. Argiles JM, Busquets S, Lopez-Soriano FJ (2003) Cytokines in the pathogenesis of cancer cachexia. Curr Opin Clin Nutr Metab Care 6: 401–406 4. Ariad S, Clifford D, Penfold G, et al (1992) Erythropoietin response in anaemic patients with multiple myeloma and other lymphoid malignancies inﬁltrating the bone marrow. Eur J Haematol 49: 59–62 5. Arslan M, Evrensel T, Kurt E, et al (2005) Comparison of the serum erythropoietin levels in chemotherapy-naive and cisplatin-treated cancer patients. Neoplasma 52: 43–45

Pathophysiology of anemia in cancer

177

6. Aul C, Arning M, Runde V, et al (1991) Serum erythropoietin concentrations in patients with myelodysplastic syndromes. Leuk Res 15: 571–575 7. Avasarala JR, Konduru SS (2005) Recombinant erythropoietin down-regulates IL-6 and CXCR4 genes in TNF-alpha-treated primary cultures of human microvascular endothelial cells implications for multiple sclerosis. J Mol Neurosci 25: 183–189 8. Baer AN, Dessypris EN, Goldwasser E, et al (1987) Blunted erythropoietin response to anaemia in rheumatoid arthritis. Br J Haematol 66: 559–564 9. Balkwill F, Osborne R, Burke F, et al (1987) Evidence for tumour necrosis factor/cachectin production in cancer. Lancet 2: 1229–1232 10. Balkwill F (2006) TNF-α in promotion and progression of cancer. Cancer Metastasis Rev 25: 409–416 11. Barrett-Lee PJ, Bailey NP, O’Brien ME, et al (2000) Large-scale UK audit of blood transfusion requirements and anaemia in patients recveiving cytotoxic chemotherapy. Br J Cancer 82: 93–97 12. Bataille R, Jourdan M, Zhang X-G, et al (1989) Serum levels of interleukin 6, a potent myeloma cell growth factor, as a reﬂect of disease severity in plasma cell dyscasias. J Clin Invest 84: 2008–2011 13. Beguin Y, Lampertz S, Bron D, et al (1996) Serum erythropoietin in chronic lymphocytic leukaemia. Br J Haematol 93: 154–156 14. Beguin Y, Yerna M, Loo M, et al (1992) Erythropoiesis in multiple myeloma: defective red cell production due to inappropriate erythropoietin production. Br J Haematol 82: 648–653 15. Bingle L, Brown NJ, Lewis CE (2002) The role of tumour-associated macrophages in tumour progression: implications for new anticancer therapies. J Pathol 196: 254–265 16. Blick M, Sherwin SA, Rosenblum M, et al (1987) Phase I study of recombinant tumor necrosis factor in cancer patients. Cancer Res 47: 2986–2989 17. Blood (1956) 11: 618–631 18. Bourantas K, Christou L, Tsiara St, et al (1995) Myelodysplastic syndromes: erythropoietin level and treatment with recombinant human erythropoietin. J Exp Clin Cancer Res 14: 205–210 19. Bowen DT, Jacobs A, Cotes PM, et al (1990) Serum erythropoietin and erythropoiesis in patients with myelodysplastic syndromes. Eur J Haematol 44: 30–32 20. Bowman SJ (2002) Hematological manifestations of rheumatoid arthritis. Scand J Rheumatol 31: 251–259 21. Braczkowski R, Romanowsky W, Danikiewicz A, et al (2001) Decrease of erythropoietin level by human recombinant tumour necrosis factor α (hrec TNFα) in patients with advanced cancer. J Biol Regul Homeost Agents 15: 366–369 22. Brown TD, Goodman PJ, et al (1991) Phase II trial of recombinant DNA gamma-interferon in advanced colorectal cancer: a Southwest Oncology Group study. J Immunother 10: 379–382 23. Broxmeyer HE, Williams DE, Lu L, Cooper S, et al (1986) The suppressive inﬂuences of human tumor necrosis factors on bone marrow hematopoietic progenitor cells from normal donors and patients with leukemia: synergism of tumor necrosis factor and interferon-γ. J Immunol 136: 4487–4495

178

M. R. Nowrousian

24. Byrd TF, Horwitz MA (1993) Regulation of transferrin receptor expression and ferritin content in human mononuclear phagocytes. J Clin Invest 91: 969–976 25. Caenazzo A, Pietrogrande F, Sgarabotto D, et al (1993) Serum neopterin levels in haematological malignancies. Haematologica 78: 225–229 26. Capalbo S, Battista C, Delia M, et al (2002) Evaluation of tumor necrosis factor-α and erythropoietin serum levels in B-cell chronic lymphocytic leukemia patients with anemia. Acta Haematol 108: 84–89 27. Cartwright GE (1966) The anemia of chronic disorders. Semin Hematol 3: 351–355 28. Cash JM, Sears DA (1989) The anemia of chronic disease: spectrum of associated diseases in a series of unselected hospitalized patients. Am J Med 87: 638–644 29. Cazzola M, Guarnone R, Cerani P, et al (1998) Red blood cell precursor mass as an independent determinant of serum erythropoietin level. Blood 91: 2139–2145 30. Cazzola M, Ponchio L, Beguin Y, et al (1992) Subcutaneous erythropoietin for treatment of refractory anemia in hematologic disorders: Results of a phase I/II clinical trial. Blood 79: 29–37 31. Cerruti A, Castello G, Balleari E, et al (1994) Serum erythropoietin increase in patients receiving adjuvant therapy with 5-ﬂuorouracil and leucovorin. Exp Hematol 22: 1261–1263 32. Clark IA, Chaudhri G (1988) Tumour necrosis factor may contribute to the anaemia of malaria by causing dyserythropoiesis and erythrophagocytosis. Br J Haematol 70: 99–103 33. Coifﬁer P, Guastalla JP, Pujade-Lauraine E, et al (2001) Predicting cancer-associated anaemia in patients receiving non-platinum chemotherapy: results of a retrospective survey. Eur J Cancer 37: 1617–1623 34. Collado-Hidalgo A, Bower JE, Ganz PA, et al (2006) Inﬂammatory biomarkers for persistent fatigue in breast cancer survivors. Clin Cancer Res 12: 2759– 2766 35. Constante M, Wang D, Raymond V, et al (2007) Repression of repulsive guidance molecule C during inﬂammation is independent of Hfe and involves tumor necrosis factor-α. Am J Pathol 170: 497–504 36. Corazza F, Beguin Y, Bergmann P, et al (1998) Anemia in children with cancer is associated with decreased erythropoietic activity and not with inadequate erythropoietin production. Blood 92: 1793–1798 37. Cox R, Musial T, Gyde OHB (1986) Reduced erythropoietin levels as a cause of anemia in patients with lung cancer. Eur J Cancer Clin Oncol 22: 511–514 38. Dai C, Chung IJ, Jiang S, et al (2003) Reduction of cell cycle progression in human erythroid progenitor cells treated with tumour necrosis factor alpha occurs with reduced CDK6 and is partially reversed by CDK6 transduction. Br J Haematol 121: 919–927 39. Dai C, Krantz SV (1999) Interferon-γ induces upregulation and activation of caspases 1, 3, and 8 to produce apoptosis in human erythroid progenitor cells. Blood 93: 3309–3316 40. Dainiak N, Kulkarni V, Howard D, et al (1983) Mechanisms of abnormal erythropoiesis in malignancy. Cancer 51: 1101–1106

Pathophysiology of anemia in cancer

179

41. Dallalio G, Law E, Means Jr RT (2006) Hepcidin inhibits in vitro erythroid colony formation at reduced erythropoietin concentrations. Blood 107: 2702– 2704 42. De Maria R, Testa U, Luchetti L, et al (1999) Apoptotic role of Fas/Fas ligand system in the regulation of erythropoiesis. Blood 93: 796–803 43. Denz H, Fuchs D, Huber H, et al (1990) Correlation between neopterin, interferon-gamma and haemoglobin in patients with haematological disorders. Eur J Haematol 44: 186–189 44. Dowlati A, R’Zik S, Fillet G, et al (1997) Anaemia of lung cancer is due to impaired erythroid marrow response to erythropoietin stimulation as well as relative inadequacy of erythropoietin production. Br J Haematol 97: 297–299 45. Dufour C, Corcione A, Svahn J, et al (2001) Interferon γ and tumour necrosis factor α are overexpressed in bone marrow T lymphocytes from paediatric patients with aplastic anaemia. Br J Haematol 115: 1023–1031 46. Dzikaite V, Holmström P, Stal P, et al (2006) Regulatory effects of tumor necrosis factor-α and interleukin-6 on HAMP expression in iron loaded rat hepatocytes. J Hep 44: 544–551 47. Eastgate JA, Symons JA, Wood NC, et al (1988) Correlation of plasma interleukin 1 levels with disease activity in rheumatoid arthritis. Lancet 2: 706–709 48. Eckardt K-U, Kurtz A (2005) Regulation of erythropoietin production. Eur J Clin Invest 35 [Suppl 3]: 13–19 49. Ek T, Mellander L, Abrahamsson J (2005) Interferon-γ and tumor necrosis factor α in relation to anemia and prognosis in childhood cancer. Acta Paediatrica 94: 435–437 50. Faquin WC, Schneider RJ, Goldberg MA (1992) Effect of inﬂammatory cytokines on hypoxia-induced erythropoietin production. Blood 79: 1987– 1994 51. Fayad L, Keating MJ, Reuben JM, et al (2001) Interleukin-6 and interleukin-10 levels in chronic lymphocytic leukemia: correlation with phenotypic characteristics and outcome. Blood 97: 256–263 52. Feelders RA, Vreugdenhil G, Eggermont AMM, et al (1998) Regulation of iron metabolism in the acute-phase response: interferon-γ and tumour necrosis factor-α induce hypoferraemia, ferritin production and a decrease in circulating transferrin receptors in cancer patients. Eur J Clin Invest 28: 520–527 53. Feelders RA, Vreugdenhil G, van Dijk JP, et al (1993) Decreased afﬁnity and number of transferrin receptors on erythroblasts in the anemia of rheumatoid arthritis. Am J Hematol 43: 200–204 54. Fiedler W, Zeller W, et al (1991) A phase II combination trial with recombinant human tumor necrosis factor and gamma interferon in patients with colorectal cancer. Klin Wochenschr 69: 261–268 55. Firat D, Banzon J (1971) Erythropoietic effect of plasma from patients with advanced cancer. Cancer Res 31: 1353–1359 56. Fossa A, Brandhorst D, Myklebust JH, et al (1999) Relation between S-phase fraction of myeloma cells and anemia in patients with multiple myeloma. Exp Hematol 27: 1621–1626 57. Fuchs D, Zangerle R, Artner-Dworzak E, et al (1993) Association between immune activation, changes of iron metabolism and anaemia in patients with HIV infection. Eur J Haematol 50: 90–94

180

M. R. Nowrousian

58. Furmanski P, Johnson CS (1990) Macrophage control of normal and leukemic erythropoiesis: identiﬁcation of the macrophage-derived erythroid suppression activity as interleukin-1 and the mediator of its in vivo action as tumor necrosis factor. Blood 75: 2328–2334 59. Ganz T (2007) Molecular control of iron transport. J Am Soc Nephrol 18: 394–400 60. Ganz T (2006) Molecular pathogenesis of anemia of chronic disease. Pediatr Blood Cancer 46: 554–557 61. Gasche C, Kulnigg S (2006) Intravenous iron in inﬂammatory bowel disease. Semin Hematol 43 [Suppl 6]: S18–S22 62. Graziadei I, Gaggl S, Kaserbacher R, et al (1994) The acute-phase protein α1-antitrypsin inhibits growth and proliferation of human early erythroid progenitor cells (burst-forming units-erythroid) and of human erythroleukemic cells (K562) in vitro by interfering with transferrin iron uptake. Blood 83: 260–268 63. Gregory T, Yu C, Ma A, et al (1999) GATA-1 and erythropoietin cooperate to promote erythroid cell survival by regulating bcl-xL expression. Blood 94: 87–96 64. Groopman JE, Itri LM (1999) Chemotherapy-induced anemia in adults: Incidence and treatment. J Natl Cancer Inst 91: 1616–1634 65. Guillonneau C, Hill M, Hubert FX, et al (2007) CD40lg treatment results in allograft acceptance mediated by CD8+CD45Rc low T cells, IFN-γ and indoleamine 2,3-dioxygenase. J Clin Invest 117: 1096–1106 66. Haurani FI, Young K, Tocantins LM (1963) Reutilization of iron in anemia complicating malignant neoplasms. Blood 22: 73–81 67. Honda K, Ishiko O, Tatsuta I, et al (1995) Anemia-inducing substance from plasma of patients with advanced malignant neoplasms. Cancer Res 55: 3623– 3628 68. Hooks JJ, Moutsopoulos HM, Geis SA, et al (1979) Immune interferon in the circulation of patients with autoimmune disease. N Engl J Med 301: 5–8 69. Hulkkonen J, Vilpo J, Vilpo L, et al (2000) Interleukin-1β, interleukin-1 receptor antagonist and interleukin-6 plasma levels and cytokine gene polymorphisms in chronic lymphocytic leukemia: correlation with prognostic parameters. Haematologica 85: 600–606 70. Hyman GA, Gellhorn A, Harvey JL Studies on the anemia of disseminated malignant neoplastic disease. II. Study of the life span of the erythrocyte. Blood 11: 618–631 71. Hyman GA (1963) Anemia in malignant neoplastic disease. J Chron Dis 16: 645–666 72. Ikeda T, Kawakami K, Fujita J, et al (1998) Thymic carcinoma associated with a high serum level of interleukin 6 diagnosed through the evaluation for asymptomatic elevation of acute-phase reactants. Intern Med 37: 414–416 73. Illmann J, Corringham R, Robinson D, et al (2005) Are inﬂammatory cytokines the common link between cancer-associated cachexia and depression? J Support Oncol 3: 37–50 74. Inamura J, Ikuta K, Jimbo J, et al (2005) Upregulation of hepcidin by interleukin-1beta in human hepatoma cell lines. Hepatol Res 33: 198–205 75. Ishiko O, Hirai K, Nishimura S, et al (1999) Elimination of anemia-inducing substance by cyclic plasma perfusion of tumor-bearing rabbits. Clin Cancer Res 5: 2660–2665

Pathophysiology of anemia in cancer

181

76. Jacobs A, Janowska-Wieczorek A, Caro J, et al (1989) Circulating erythropoietin in patients with myelodysplastic syndromes. Br J Haematol 73: 36–39 77. Jelkmann W, Pagel H, Wolff M, et al (1992) Monokines inhibiting erythropoietin production in human hepatoma cultures and in isolated perfused rat kidneys. Life Sci 50: 301–308 78. Jelkmann W, Wolff M, Fandrey J (1994) Inhibition of erythropoietin production by cytokines and chemotherapy may contribute to the anemia in malignant diseases. In: Vaupel P, et al (eds) Oxygen transport to tissue, Vol XV. Plenum Press, New York, Adv Exp Med Biol 345: 525–530 79. Johansson P, Andréasson B (2006) Hydroxyurea therapy increases plasma erythropoietin in patients with essential thrombocythaemia of polycythaemia vera. Clin Lab Haematol 28: 233–236 80. Johnson CS, Chang MJ, Furmanski P (1988) In vivo hematopoietic effects of tumor necrosis factor-α in normal and erythroleukemic mice: characterization and therapeutic applications. Blood 72: 1875–1883 81. Johnson CS, Cook CA, Furmanski P (1990) In vivo suppression of erythropoiesis by tumor necrosis factor-α (TNF-α): reversal with exogenous erythropoietin (EPO). Exp Hematol 18: 109–113 82. Johnson CS, Keckler DJ, Topper MI, et al (1989) In vivo hematopoietic effects of recombinant interleukin-1α in mice: stimulation of granulocytic, monocytic, megakaryocytic, and early erythroid progenitors, suppression of late-stage erythropoiesis, and reversal of erythroid suppression with erythropoietin. Blood 73: 678–683 83. Johnson D, Bayele H, Johnston K, et al (2004) Tumour necrosis factor alpha regulates iron transport and transporter expression in human intestinal epithelial cells. FEBS Lett 573: 195–201 84. Johnson RA, Waddelow TA, Caro J, et al (1989) Chronic exposure to tumor necrosis factor in vivo preferentially inhibits erythropoiesis in nude mice. Blood 74: 130–138 85. Kang JM, Lee WJ, Kim WB, et al (2005) Systemic inﬂammatory syndrome and hepatic inﬂammatory cell inﬁltration caused by an interleukin-6 producing pheochromocytoma. Endocrine J 52: 193–198 86. Kanzaki M, Soda K, Gin PTG, et al (2005) Erythropoietin attenuates cachectic events and decreases production of interleukin-6, a cachexia-inducing cytokine. Cytokine 32: 234–239 87. Karadag A, Zhou M, Croucher PI (2006) ADAM-9 (MDC-9/meltrin-γ), a member of the a disintegrin and metalloproteinase family, regulates myelomacell-induced interleukin-6 production in osteoblasts by direct interaction with the αv β5 integrin. Blood 107: 3271–3278 88. Kattamis A, Papassotiriou I, Palaiologou D, et al (2006) The effects of erythropoetic activity and iron burden on hepcidin expression in patients with thalassemia major. Haematologica 91: 809–812 89. Kettelhack C, Schöter D, Matthias D, et al (1994) Serum erythropoietin levels in patients with solid tumours. Eur J Cancer 30A: 1289–1291 90. Kim MH, Lee JH, Wu CW, et al (2002) Defective erythropoiesis in bone marrow is a mechanism of anemia in children with cancer. J Korean Med Sci 17: 337–340 91. Kostova G, Siljanovski N (2004) Inadequate erythropoietin production (epo) in patients with multiple myeloma. Prilozi 25: 53–66

182

M. R. Nowrousian

92. Kostova G, Siljanovski N (2005) Erythropoietin production in patients with malignant lymphoma. Prilozi 26: 157–168 93. Koury MJ, Bondurant MC (1990) Erythropoietin retards DNA breakdown and prevents programmed death in erythroid progenitor cells. Science 248: 378–381 94. Kreuzer KA, Rockstroh JK, Jelkmann W, et al (1997) Inadequate erythropoietin response to anaemia in HIV patients: relationship to serum levels of tumor necrosis factor-alpha, interleukin-6 and their soluble receptors. Br J Haematol 96: 235–239 95. Krijt J, Vokurka M, Chang KT, et al (2004) Expression of Rgmc, the murine ortholog of hemojuvelin gene, is modulated by development and inﬂammation, but not by iron status or erythropoietin. Blood 104: 4308–4310 96. Kulbe H, Thompson R, Wilson JL, et al (2007) The inﬂammatory cytokine tumor necrosis factor-α generates an autocrine tumor-promoting network in epithelial ovarian cancer cells. Cancer Res 67: 585–592 97. Kurzrock R (2001) The role of cytokines in cancer-related fatigue. Cancer 92: 1684–1688 98. Laftah A, Sharma N, Brookes M, et al (2006) Tumour necrosis factor alpha causes hypoferraemia and reduced intestinal iron absorption in mice. Biochem J 397: 61–67 99. Lai R, O’Brien S, Maushouri T, et al (2002) Prognostic value of plasma interleukin-6 levels in patients with chronic lymphocytic leukemia. Cancer 95: 1071– 1075 100. Lamagna C, Aurrand-Lions M, Imhof BA (2006) Dual role of macrophages in tumor growth and angiogenesis. J Leukoc Biol 80: 705–713 101. Lauta VM (2003) A review of the cytokine network in multiple myeloma. Cancer 97: 2440–2452 102. Lee CC, Liu KJ, Huang TS (2006) Tumor-associated macrophage: Its role in tumor angiogenesis. J Cancer Mol 2: 135–140 103. Lee GR (1983) The anemia of chronic disease. Semin Hematol 20: 61–80 104. Lee P, Peng H, Gelbart T, et al (2005) Regulation of hepcidin transcription by interleukin-1 and interleukin-6. Proc Natl Acad Sci 102: 1906–1910 105. Lee SJ, Known JH, Jung CW (2001) Erythropoietin response is inadequate in cancer patients receiving chemotherapy. Int J Hematol 74: 416–420 106. Leong WI, Lönnerdal B (2004) Hepcidin, the recently identiﬁed peptide that appears to regulate iron absorption. J Nutr 134: 1–4 107. Lewis C, Pollard JW (2006) Distinct role of macrophages in different tumor microenvironments. Cancer Res 66: 605–612 108. Liu Y, Pop R, Sadegh C, et al (2006) Suppression of Fas-FasL coexpression by erythropoietin mediates erythroblast expansion during the erythropoietic stress response in vivo. Blood 108: 123–133 109. Lou D-Q, Lesbordes J-C, Nicolas G, et al (2005) Iron- and inﬂammationinduced hepcidin gene expression in mice is not mediated by Kupffer cells in vivo. Hepatology 41: 1056–1064 110. Ludwiczek S, Aigner E, Theurl I, et al (2003) Cytokine-mediated regulation of iron transport in human monocytic cells. Blood 101: 4148–4154 111. Ludwig H, Fritz E, Kotzmann H, et al (1990) Erythropoietin treatment of anemia associated with multiple myeloma. N Engl J Med 322: 1693–1699

Pathophysiology of anemia in cancer

183

112. Ludwig H, Fritz E (1996) Incidence and clinical signiﬁcance of anemia in malignant diseases. In: Smyth JF, Boogaerts MA, Ehmer BRM (eds) rhErythropoietin in Cancer Supportive Treatment. Marcel Dekker, New York, pp 35–44 113. Ludwig H, Nowrousian MR (2000) Therapie der tumorbedingten und chemotherapieinduzierten Anämie mit rekombinantem humanen Erythropoetin. In: Nowrousian MR (ed) Supportive Therapie in der Onkologie. W. Zuckschwerdt Verlag, München, Bern, Wien, New York, pp 99–113 114. Ludwig H, Fritz E, Leitgeb C, et al (1994) Prediction of response to erythropoietin treatment in chronic anemia of cancer. Blood 84: 1056–1063 115. Ludwig H, van Belle S, Barrett-Lee P, et al (2004) The European Cancer Anaemia Survey (ECAS): A large, multinational, prospective survey deﬁning the prevalence, incidence, and treatment of anaemia in cancer patients. Eur J Cancer 40: 2293–2306 116. Macció A, Madeddu C, Massa D, et al (2005) H levels correlate with interleukin6 levels in patients with advanced untreated epithelial ovarian cancer: role of inﬂammation in cancer-related anemia. Blood 106: 362–367 117. Marsden JT, Day P, Ellis R, et al (2006) A sample distribution programme for erythropoietin. Clin Lab Haematol 28: 228–232 118. Margetic S, Topic E, Ruzic F, et al (2005) Soluble transferrin receptor and transferrin receptor-ferritin index in iron deﬁciency anemia and anemia in rheumatoid arthritis. Clin Chem Lab Med 43: 326–331 119. Matsumura N, Shiki H, Saito N, et al (2002) Interleukin-6-producing thymic squamous cell carcinoma associated with Castleman’s disease and nephrotic syndrome. Intern Med 41: 871–874 120. Maury CPJ, Andersson LC, Teppo AM, et al (1988) Mechanism of anaemia in rheumatoid arthritis: demonstration of raised interleukin 1β concentrations in anaemic patients and of interleukin 1 mediated suppression of normal erythropoiesis and proliferation of human erythroleukaemia (HEL) cells in vitro. Ann Rheum Dis 47: 972–978 121. Means RT, Dessypris EN, Krantz SB (1992) Inhibition of human erythroid colony-forming units by interleukin-1 is mediated by gamma interferon. J Cell Physiol 150: 59–64 122. Means RT, Krantz SB (1993) Inhibition of human erythroid colony-forming units by tumor necrosis factor requires beta interferon. J Clin Invest 91: 416– 419 123. Means RT, Krantz SB (1991) Inhibition of human erythroid colony-forming units by γ interferon can be corrected by recombinant human erythropoietin. Blood 78: 2564–2567 124. Means RT, Krantz SB (1992) Progress in understanding the pathogenesis of the anemia of chronic disease. Blood 80: 1639–1647 125. Means RT (1995) Pathogenesis of the anemia of chronic disease: a cytokine mediated anemia. Stem Cells 13: 32–37 126. Merchav S, Nielsen OJ, Rosenbaum H, et al (1990) In vitro studies of erythropoietin-dependent regulation of erythropoiesis in myelodysplastic syndromes. Leukemia 11: 771–774 127. Mertz JR, Theil EC (1983) Subunit dimers in sheep spleen apoferritin: The effect on iron storage. J Biol Chem 258: 1719–1726

184

M. R. Nowrousian

128. Melichar B, Solichova D, Freedman RS (2006) Neopterin as an indicator of immune activation and prognosis in patients with gynecological malignancies. Int J Gynecol Cancer 16: 240–252 129. Miller A, Chodos RB, Emerson CP, et al (1956) Studies of the anemia and iron metabolism in cancer. J Clin Invest 35: 1248–1262 130. Miller CB, Jones RJ, Piantadosi S, et al (1990) Decreased erythropoietin response in patients with the anemia of cancer. N Engl J Med 322: 1689–1692 131. Mitlyng BL, Singh JA, Furne JK, et al (2006) Use of breath carbon monoxide measurements to assess erythrocyte survival in subjects with chronic diseases. Am J Hematol 81: 432–438 132. Moldawer LL, Marano MA, Wei H, et al (1989) Cachectin/tumor necrosis factor-α alters red blood cell kinetics and induces anemia in vivo. FASEB J 3: 1637–1643 133. Montecucco C, Carnevale R, Cazzola M, et al (1986) Microcytic anemia in rheumatoid arthritis. Relationship with activity and duration of the disease and iron status. Haematologica 71: 383–387 134. Munoz M, Campos A, Garcia-Erce J (2006) Intravenous iron in colorectal cancer surgery. Semin Hematol 43 [Suppl 6]: S36–S38 135. Murdoch C, Lewis CE (2005) Macrophage migration and gene expression in response to tumor hypoxia. Int J Cancer 117: 701–708 136. Murr C, Berchtold J, Norer B, et al (1998) Neopterin as a prognostic parameter in patients with squamous-cell carcinomas of the oral cavity. Int J Cancer 79: 476–480 137. Murr C, Fuith LC, Widner B, et al (1999) Increased neopterin concentrations in patients with cancer: indicator of oxidative stress? Anticancer Res 19: 1721–1728 138. Nagata Y, Kiefer F, Watanabe T, et al (1999) Activation of hematopoietic progenitor kinase-1 by erythropoietin. Blood 93: 3347–3354 139. Naka T, Nishimoto N, Kishimoto T (2002) The paradigm of IL-6: from basic science to medicine. Arthritis Res 4[Suppl 3]: S233–S242 140. Nemeth E, Ganz T (2006) Hepcidin and iron-loading anemias. Haematologica 91: 727–732 141. Nemeth E, Rivera S, Gabayan V, et al (2004) IL-6 mediates hypoferremia of inﬂammation by inducing the synthesis of the iron regulatory hormone hepcidin. J Clin Invest 113: 1271–1276 142. Nemeth E, Tuttle M, Powelson J, et al (2004) Hepcidin regulates cellular iron efﬂux by binding to ferroportin and inducing its internalization. Science 306: 2090–2093 143. Nicolas G, Bennoun M, Porteu A, et al (2002) Severe iron deﬁciency anemia in transgenic mice expressing liver hepcidin. Proc Natl Acad Sci 99: 4596–4601 144. Nicolas G, Chauvet C, Viatte L, et al (2002) The gene encoding the iron regulatory peptide hepcidin is regulated by anemia, hypoxia and inﬂammation. J Clin Invest 110: 1037–1044 145. Nielsen OJ, Brandt M, Drivsholm A (1990) The secretory erythropoietin response in patients with multiple myeloma and Waldenström’s macroglobulinaemia. Scand J Clin Lab Invest 50: 697–703 146. Nishimoto N, Kishimoto T (2006) Interleukin 6: from bench to bedside. Nat Clin Pract 2: 619–626

Pathophysiology of anemia in cancer

185

147. Noé G, Schrezenmeier H, Rich IN, et al (1994) Circulating erythropoietin levels in pathophysiological conditions. In: Rich IN, Lappin TRJ (eds) Molecular, cellular and developmental biology of erythropoietin and erythropoiesis. Ann NY Acad Sci 718: 94–102 148. Nowrousian MR, Kasper C, Oberhoff C, et al (1996) Pathophysiology of cancerrelated anemia. In: Smyth JF, Boogaerts MA, Ehmer BRM (eds) rhErythropoietin in Cancer Supportive Treatment. Marcel Dekker, New York, pp 13–34 149. Nowrousian MR (2000) Pathophysiologie der tumorbedingten oder chemotherapieinduzierten Anämie. In: Nowrousian MR (ed) Supportive Therapie in der Onkologie. W. Zuckschwerdt Verlag, München, Bern, Wien, New York, pp 84–98 150. Nowrousian MR (2002) Pathophysiology of cancer-related anemia. In: Nowrousian MR (ed) Recombinant Human Erythropoietin (rhEPO) in Clinical Oncology – Scientiﬁc and Clinical Aspects of Anemia in Cancer. Springer, Wien, New York, pp 39–62 151. Nowrousian MR (2005) Pathophysiology of anemia in cancer. In: Bokemeyer C, Ludwig H (eds) Anaemia in cancer. Elsevier, Edinburgh London, New York, pp 25–45 152. Oster W, Herrmann F, Gamm H, et al (1990) Erythropoietin for the treatment of anemia of malignancy associated with neoplastic bone marrow inﬁltration. J Clin Oncol 8: 956–962 153. Ozguroglu M, Arun B, Demir G, et al (2000) Serum erythropoietin level in anemic cancer patients. Med Oncol 17: 29–34 154. Pagel H, Fandrey J, Schobersberger W, et al (1999) Effects of neopterin and 7,8-dihydroneopterin of hypoxia-induced renal erythropoietin production. Eur J Haemataol 62: 341–345 155. Papanikolaou G, Tzilanos M, Christakis J, et al (2005) Hepcidin in iron overload disorders. Blood 105: 4103–4105 156. Pohl C, Moter A, Hasenclever D, et al (1992) Inappropriate increase in plasma erythropoietin levels following PEB chemotherapy. Onkologie 15: 384–389 157. Prutchi-Sagiv S, Golishevsky N, Oster HS, et al (2006) Erythropoietin treatment in advanced multiple myeloma is associated with improved immunological functions: could it be beneﬁcial in early disease? Br J Haematol 135: 660–672 158. Ramesh G, Reeves WB (2002) TNF-α mediates chemokine and cytokine expression and renal injury in cisplatin nephrotoxicity. J Clin Invest 110: 835–842 159. Reibnegger G, Hetzel H, Fuchs D, et al (1987) Clinical signiﬁcance of neopterin for prognosis and follow-up in ovarian cancer. Cancer Res 47: 4977–4981 160. Reibnegger G, Krainer M, Herold M, et al (1991) Predictive value of interleukin-6 and neopterin in patients with multiple myeloma. Cancer Res 51: 6250–6253 161. Reibnegger GJ, Bichler AH, Dapunt O, et al (1986) Neopterin as a prognostic indicator in patients with carcinoma of the uterine cervix. Cancer Res 46: 950– 955 162. Rivera S, Liu L, Nemeth E, et al (2005) Hepcidin excess induces the sequestration of iron and exacerbates tumor-associated anemia. Blood 105: 1797–1802 163. Rogers J, Durmowicz G, Kasschau K, et al (1991) A motif within the 5′noncoding regions of hepatic acute phase mRNAs mediates ferritin translation by interleukin-1β and may contribute to the anemia of chronic disease. Blood 78 [Suppl 1]: 367a

186

M. R. Nowrousian

164. Roodman GD, Bird A, Hutzler D, et al (1987) Tumor necrosis factor-alpha and hematopoietic progenitors: effects of tumor necrosis factor on the growth of erythroid progenitors CFU-E und BFU-E and the hematopoietic cell lines K562, HL60, and HEL cells. Exp Hematol 15: 928–935 165. Roodman GD (1987) Mechanisms of erythroid suppression in the anemia of chronic disease. Blood Cells 13: 171–184 166. Sadahira Y, Mori M (1999) Role of the macrophage in erythropoiesis. Pathol Int 49: 841–848 167. Salvazini C, Casali B, Salvo D, et al (1991) The role of interleukin 1, erythropoietin, and red cell bound immunoglobulins in the anemia of rheumatoid arthritis. Clin Exp Rheumatol 9: 241–246 168. Sawabe Y, Kikuno K, Iseki T, et al (1998) Serum erythropoietin values and serum iron status during chemotherapy for leukemia. Eur J Haematol 60: 315–316 169. Schapira L, Antin JH, Ransil BJ, et al (1990) Serum erythropoietin levels in patients receiving intensive chemotherapy and radiotherapy. Blood 76: 2354– 2359 170. Scheller J, Ohnesorge N, Rose-John S (2006) Interleukin-6 trans-signalling in chronic inﬂammation and cancer. Scand J Immunol 63: 321–329 171. Schooley JC, Kullgren B, Allison AC (1987) Inhibition by interleukin-1 of the action of erythropoietin on erythroid precursors and its possible role in the pathogenesis of hypoplastic anaemias. Br J Haematol 67: 11–17 172. Schreuder WO, Ting WC, Smith S, et al (1984) Testosterone, erythropoietin and anaemia in patients with disseminated bronchial cancer. Br J Haematol 57: 521–526 173. Schrier RW (2002) Cancer therapy and renal injury. J Clin Invest 110: 743–745 174. Schubert C, Hong S, Natarajan L, et al (2007) The association between fatigue and inﬂammatory marker levels in cancer patients: A quantitative review. Brain Behav Immun 21: 413–427 175. Seymour JF, Talpaz M, Cabanillas F, et al (1995) Serum interleukin-6 levels correlate with prognosis in diffuse large-cell lymphoma. J Clin Oncol 13: 575–582 176. Sﬁridaki A, Miyakis S, Tsirakis G, et al (2005) Systemic levels of interleukin-6 and matrix metalloproteinase-9 in patients with multiple myeloma may be useful as prognostic indexes of bone disease. Clin Chem Lab Med 43: 934–938 177. Sharma N, Laftah A, Brookes M, et al (2005) A role for tumour necrosis factor alpha in human small bowel iron transport. Biochem J 390: 437–446 178. Shen T, Shi Y, Zhu J, et al (2005) Erythropoietin response to anemia is impaired in patients with hematologic malignancies. Blood 106: 146 (Abstr 3744) 179. Shih JY, Yuan A, Chen JJW, et al (2006) Tumor-associated macrophage: Its role in cancer invasion and meastasis. J Cancer Mol 2: 101–106 180. Silva M, Grillot D, Benito A, et al (1996) Erythropoietin can promote erythroid progenitor survival by repressing apoptosis through Bcl-XL and Bcl-2. Blood 88: 1576–1582 181. Silvestris F, Cafforio P, Tucci M, et al (2002) Negative regulation of erythroblast maturation by Fas-L+/TRAIL+ highly malignant plasma cells: a major pathogenic mechanism of anemia in multiple myeloma. Blood 99: 1305– 1313

Pathophysiology of anemia in cancer

187

182. Silvestris F, Tucci M, Cafforio P, et al (2001) Fas-L upregulation by highly malignant myeloma plasma cells: role in the pathogenesis of anemia and disease progression. Blood 97: 1155–1164 183. Singh A, Eckardt KU, Zimmermann A, et al (1993) Increased plasma viscosity as a reason for inappropriate erythropoietin formation. J Clin Invest 91: 251–256 184. Skillings JR, Rogers-Melamed I, Nabholtz JM, et al (1995) An epidemiological review of anaemia in cancer chemotherapy in Canada. Eur J Cancer 31A [Suppl 5]: S183 185. Skillings JR, Sridhar FG, Wong C, et al (1993) The frequency of red cell transfusion for anemia in patients receiving chemotherapy. A retrospective cohort study. Am J Clin Oncol 16: 22–25 186. Sloand E, Kim S, Maciejewski JP, et al (2002) Intracellular interferon–γ in circulating and marrow T cells detected by ﬂow cytometry and the response to immunosuppressive therapy in patients with aplastic anemia. Blood 100:1185– 1191 187. Spivak JL, Barnes DC, Fuchs E, et al (1989) Serum immunoreactive erythropoietin in HIV-infected patients. JAMA 261: 3104–3107 188. Spell DW, Jones DV Jr, Harper WF, et al (2004) Thr value of a complete blood count in predicting cancer of the colon. Cancer Detect Prev 28: 37–42 189. Stifter G, Heiss S, Gastl G, et al (2005) Over-expression of tumor necrosis factoralpha in bone marrow biopsies from patients with myelodysplastic syndromes: relationship to anemia and prognosis. Eur J Haematol 75: 485–491 190. Szlosarek P, Charles KA, Balkwill FR (2006) Tumor necrosis factor as a tumor promoter. Eur J Cancer 42: 745–750 191. Taniguchi S, Dai C-H, Price JO, et al (1997) Interferon gamma downregulates stem cell factor and erythropoietin receptors but not insulin-like growth factorI receptors in human erythroid colony-forming cells. Blood 90: 2244–2252 192. Taniguchi S, Dai C-H, Krantz SB (1997) Speciﬁc binding of interferon-gamma to high afﬁnity receptors on human erythroid colony-forming cells. Exp Hematol 25: 193–198 193. Teppo AM, Maury CPJ (1987) Radioimmunoassay of tumor necrosis factor in serum. Clin Chem 33: 2024–2027 194. Theurl I, Mattle V, Seifert M, et al (2006) Dysregulated monocyte iron homeostasis and erythropoietin formation in patients with anemia of chronic disease. Blood 107: 4142–4148 195. Thomas C, Kirschbaum A, Boehm D, et al (2006) The diagnostic plot: a concept for identifying different states of iron deﬁciency and monitoring the response to epoetin therapy. Med Oncol 23: 23–36 196. Tracey KJ, Wei H, Manogue KR, et al (1988) Cachectin/tumor necrosis factor induces cachexia, anemia, and inﬂammation. J Exp Med 167: 1211–1227 197. Tsushima H, Imaizumi Y, Imanishi D, et al (1999) Fas antigen (CD95) in pure erythroid cell line AS-E2 is induced by interferon-γ and tumor necrosis factorα and potentiates apoptotic death. Exp Hematol 27: 433–440 198. Ulich TR, Shin SS, del Castillo J (1993) Haematologic effects of TNF. Res Immunol 144: 347–354 199. Ulich TR, del Castillo J, Yin S (1990) Tumor necrosis factor exerts dosedependent effects on erythropoiesis and myelopoiesis in vivo. Exp Hematol 18: 311–315

188

M. R. Nowrousian: Pathophysiology of anemia in cancer

200. Urabe A, Mitani K, Yoshinago K, et al (1992) Serum erythropoietin titers in hematological malignancies and related diseases. Int J Cell Cloning 10: 333–337 201. Vannucchi AM, Grossi A, Rafanelli D, et al (1994) Inhibition of erythropoietin production in vitro by human interferon gamma. Br J Haematol 87: 18–23 202. Viatte L, Lesbordes-Brion JC, Lou DQ, et al (2005) Deregulation of proteins involved in iron metabolism in hepcidin-deﬁcient mice. Blood 105: 4861–4864 203. Vokurka M, Krijt J, Sulc K, Necas E (2006) Hepcidin mRNA levels in mouse liver respond to inhibition of erythropoiesis. Physiol Res 55: 667–674 204. Voulgari PV, Kolios G, Papadopoulos GK, et al (1999) Role of cytokines in the pathogenesis of anemia of chronic disease in rheumatoid arthritis. Clin Immunol Immunopathol 92: 153–160 205. Vreugdenhil G, Manger B, Nieuwenhuizen C, et al (1992) Iron stores and serum transferrin receptor levels during recombination human erythropoietin treatment of anemia in rheumatoid arthritis. Ann Hematol 65: 265–268 206. Ward HP, Kurnick JE, Pisarczyk MJ (1971) Serum level of erythropoietin in anemias associated with chronic infection, malignancy, and primary hematopoietic disease. J Clin Invest 50: 332–335 207. Weinstein DA, Roy CN, Fleming MD, et al (2002) Inappropriate expression of hepcidin is associated with iron refractory anemia: implications for the anemia of chronic disease. Blood 100: 3776–3781 208. Weiss G, Houston T, Kastner St, et al (1997) Regulation of cellular iron metabolism by erythropoietin: Activation of iron regulatory protein and upregulation of transferrin receptor expression in erythroid cells. Blood 89: 680–687 209. Weiss G, Kronberger P, Conrad F, et al (1993) Neopterin and prognosis in patients with adenocarcinoma of the colon. Cancer Res 53: 260–265 210. Wolff M, Jelkmann W (1993) Effects of chemotherapeutic and immunosuppressive drugs on the production of erythropoietin in human hepatoma cultures. Ann Hematol 66: 27–31 211. Wood PA, Hrushesky WJM (1995) Cisplatin-associated anemia: an erythropoietin deﬁciency syndrome. J Clin Invest 95: 1650–1659 212. Zhang J (2007) Yin and yang interplay of IFN-γ in inﬂammation and autoimmune disease. J Clin Invest 117: 871–873 213. Zucker S, Friedman S, Lysik RM (1974) Bone marrow erythropoiesis in the anemia of infection, inﬂammation, and malignancy. J Clin Invest 53: 1132–1138 214. Zucker S (1985) Anemia in cancer. Cancer Invest 3: 249–260 Correspondence: Prof. Dr. M. R. Nowrousian, Department of Internal Medicine (Cancer Research), West German Cancer Center, University Hospital of Essen, Hufelandstrasse 55, 45122 Essen, Germany, E-mail: [email protected]

Chapter 7

Prevalence and incidence of anemia and risk factors for anemia in patients with cancer H. Ludwig Department of Medicine I, Center for Oncology and Hematology, Vienna, Austria

Deﬁnition of anemia Anemia is deﬁned as a reduction in the total body red blood cell mass. As measurement of total body red cell (RBC) mass requires special radiolabeling techniques that usually are not amenable to general medical diagnostic workup, hemoglobin (Hb) or hematocrit levels are substituted for RBC mass determination. This is justiﬁable because of the body’s tendency to maintain normal total blood volume by dilution of the depleted RBC component with plasma. This adjustment results in decrease of the total blood Hb concentration, the RBC count, and the hematocrit. Therefore, a pragmatic deﬁnition of anemia is best described as a state with a reduction of these parameters and when the Hb is less than 12 g/dL or the hematocrit is less than 37 cL/L. Although normal range of Hb level differs between men (14– 18 g/dL) and women (12–16 g/dL), there are no uniform accepted lower limits of normal Hb concentration (Beutler and Waalen 2006). Most conducted studies deﬁne anemia in cancer patients as Hb <12 g/dL. This accords with the common toxicity criteria of the National Cancer Institute (NCI) (1999), and offers the advantage of combining and analyzing results obtained in patients of both sexes together and reduces complexity, albeit at the price of some inaccuracy.

Anemia in cancer The main focus of this article relates to the two most common types of anemia in patients with cancer, namely chronic anemia of cancer and cancer therapyinduced anemia or a combination of both. Cancer may be the cause of several types of anemia such as pure red cell anemia, hemolytic anemia, anemia due to blood loss or iron, hormone or vitamin deﬁciency and it can, of course, be associated with all other forms of anemia.

190

H. Ludwig

Prevalence of anemia in cancer Prevalence of anemia in patients with cancer depends on various factors. Higher age is a general risk factor that applies to all individuals even to randomly selected elderly people (Denny et al. 2006). In addition, in malignant diseases, the underlying type of cancer, stage of the disease, type and dose of cancer therapy and duration of treatment inﬂuence the risk of anemia. Several studies have been conducted to assess the prevalence of anemia in patients with various types of cancers. Many of these reports are retrospective chart reviews and based on a limited number of patients (Table 1). Studies involving more than 1,000 patients have been reported by few authors. Moullet et al. (1998) studied 1,077 adult lymphoma patients with different histologic subtypes treated between 1980 and 1995. Thirty-two percent of patients were found to be anemic and anemia was identiﬁed as adverse prognostic factor. When patients with and without bone marrow involvement were considered separately, anemia remained an adverse factor, but only in univariate analysis. In multivariate analysis, anemia remained an adverse prognostic factor only in patients with bone marrow inﬁltration. Barrett-Lee et al. (2000) conducted a retrospective audit showing that cancer patients receiving cytotoxic chemotherapy often become anemic and may require blood transfusions. This large-scale audit of patients with a variety of solid tumors receiving chemotherapy was conducted in 28 specialist centers throughout the United Kingdom in order to quantify the prevalence of anemia and the need for transfusions. Data from 2,719 patients with a mean age of 55 years receiving 3,206 courses of cytotoxic chemotherapy for tumors of the breast (878), ovary (856), lung (772) or testis (213) were analyzed. The mean proportion of patients with Hb <11 g/dL rose over the course of chemotherapy from 17% before the ﬁrst cycle, to 38% by the sixth, despite transfusion in 33% of patients. The proportion of patients requiring at least one blood transfusion varied from 19% for breast cancer to 43% for lung cancer. Sixteen percent of patients required more than one transfusion (7% with breast-22% lung cancer). Of the patients receiving transfusions, 25% required an inpatient admission and overnight stay. The most common symptoms reported at the time of transfusion were lethargy, tiredness and breathlessness. Cofﬁer et al. (2001) conducted a 2-year retrospective chart survey in 1,064 patients with colorectal, breast, lung or ovarian cancer, Hodgkin’s disease, or non-Hodgkin’s lymphoma in 24 centers in France to determine the prevalence of anemia (Hb ≤12 g/dL) and need for transfusion in patients who received nonplatinum-based chemotherapy for more than 3 cycles or 3 months. Baseline Hb levels documented anemia in 37.1% of patients (all tumor types). By cycle 3, the prevalence of anemia increased to 54.1% of patients and remained over 50% at cycle 4. At some time during chemotherapy 14.5% of patients were transfused. Predictive risk factors for anemia requiring transfusion included low baseline Hb, decrease in Hb during the

1,077

2,719

1,064

Barrett-Lee PJ et al., Br J Cancer 2000

Cofﬁer B et al., Eur J Cancer, 2001

Number of Patients

Moullet I et al., Ann Oncol 1998

Authors

Colorectal, breast, lung, ovarian

Breast (878), ovary (856), lung (772), testis (213)

NHL

Tumor Type

≤12 g/dL

Men and women <50 years: <12 g/dL, women >50 years: <11 g/dL <11 g dL−1

Deﬁnition of Anemia, Hb level

37.1%, after 3 cycles chemotherapy: 54.1%

17% before chemotherapy, 38% after 6 cycles

32%

Prevalence of Anemia

Table 1. Prevalence of anemia in patients with various tumor types and treatment conditions

Retrospective audit of patients’ charts. 33% of the patients received transfusions and 25% of these required inpatient admission and overnight stay 2 year retrospective chart review. 14.5% received transfusions, risk factors for transfusions: low baseline Hb, decrease of Hb after ﬁrst mo of chemotherapy, prior RBC transfusions, duration of chemotherapy

Retrospective chart review. Anemia was an adverse prognostic factor for PFS and OS

Comments Prevalence and incidence of anemia and risk factors 191

Number of Patients

15,367

2,002

4,994

2,360

Authors

Ludwig H et al., Eur J Cancer, 2004

Kosmidis P and Krzakowski M, Lung Cancer, 2005

Barrett-Lee P et al., Oncologist, 2005

Birgegard G et al., Haematologica, 2006

Table 1. Continued

Evaluable 2,316 MM: 704; NHL and HD: 1,612

Breast: 3,253, ovarian: 741, cervical: 289, other: 279

Breast, lung, head and neck, gastrointestinal, lymphoma/ myeloma, Lung cancer

Tumor Type

Prevalence of Anemia

39.3%

37.6%

Breast cancer: 30.4%, Gynecological cancer: 49.1% 52.5% Ever anemic during the survey: 72.9%

Deﬁnition of Anemia, Hb level <12 g/dL

<12 g/dL

<12 g/dL

<12 g/dL

Prospective epidemiologic survey with 6 months follow-up, subanalysis of ECAS study Anemia risk factors: platinum chemotherapy, female gender, low baseline Hb 62.3% of breast cancer and 81.4% of gynecologic cancer patients were anemic at least once during the 6 month survey, subanalysis of ECAS study Prevalence of anemia was 37.4% in Hodgkin’s disease, 49% in nonHodgkin’s lymphoma and 69.2% in MM, subanalysis of ECAS study

Prospective epidemiologic survey with 6 months follow-up. 67% of patients became anemic at least once during the survey

Comments

192 H. Ludwig

Various cancers

Colorectal cancer not speciﬁed

694

358

616

Seshadri T et al., Med J Aust, 2005 Sadahiro S et al., J Gastroenterol, 1998 Skillings JR et al., Cancer Prev Control, 1999

CLL, MM, NHL, HD

273

Steurer M et al., Wien Klin Wochschr, 2004

Prostate, breast, head & neck, colorectal, lung, uterine

574

Harrison L et al., Semin Oncol, 2001

41% at presentation, Highest frequency during radiotherapy in lung (77%) and uterine cancer (79%) Myeloma: 77.4%

35%

Men: 20.8%, women: 25.8% 28%

<12 g/dL

<12 g/dL <10 g/dL <10 g/dL

<12 g/dL

Multivariate analysis: age, tumor site and tumor size predictors of anemia 12% received transfusions Multivariate analysis: risk factors for transfusion: platinum, anthracycline, low baseline Hb, and disease stage

Retrospective chart review; patients were followed through four cycles of nonplatinum chemotherapy Prevalence of anemia increased for all malignancies after cycle 4, NHL: 35.1% at baseline to 73.7% HD: 21.9% at baseline to 54.5 Predictors for anemia: low baseline Hb, platinum chemotherapy

Retrospective chart review; 2 centers participated, prevalence of anemia increased by 16% in head and neck cancer and by 20% in lung cancer during radiotherapy

Prevalence and incidence of anemia and risk factors 193

194

H. Ludwig

ﬁrst month of chemotherapy, primary tumor site, prior blood transfusions and duration of chemotherapy. By early identiﬁcation of patients at the highest risk of developing anemia, interventions such as epoetin can be employed to reduce or eliminate the need for transfusions. The largest anemia survey reported is the European Cancer Anemia Survey (ECAS) by Ludwig et al. (2004). This was a prospective, observational survey conducted in 24 European countries with 748 participating cancer centers and enrolling 15,367 patients with cancer. Patients were followed for up to 6 data points or 6 months. Patient enrolment was started in January 2001 and continued until July 2001. The survey addressed several questions, namely prevalence, incidence and treatment of anemia in European cancer patients as they present in daily clinical practice. Of the 14,520 patients with Hb levels available at enrollment, 39.3% were anemic. Most patients presented with mild (Hb 10.0 to 11.9 g/dL) anemia (29.3%), while moderate anemia (Hb 8.0 to 9.9 g/dL) was seen in 8.7% and severe anemia (Hb <8 g/dL) in 1.3% only. Anemia was twice as prevalent in patients with lymphoma or myeloma and gynecologic cancers compared to patients with head and neck cancer (Fig. 1). Anemia was more common in patients with relapsing or persistent disease (48%) and in those receiving treatment at enrolment (41.1%) compared to patients with newly diagnosed disease (30.7%) or in clinical remission (31.3%). Importantly, anemia was found to correlate with poor World Health Organization (WHO) performance status (p < 0.001). When patients were followed for up to 6 months it was found that 67% of the patients surveyed had Hb levels <12 g/dL at least once during the observation period. The highest “ever anemic” rates were seen in patients with gynecologic 70 % Patients

60

49,1

50 37,6

40 24,9

30

29,2

52,5

38,9

30,4

20 10 ph G yn om a/ M ye lo m a

Ly m

G

Lu ng I-C ol or ec ta l

l

Br ea st

U ro ge ni ta

H ea d

an d

N ec k

0

Hb <8.0

Hb 8.0-9.9

Hb 10.0-11.9

Hb ≥12

Fig. 1. Prevalence of anemia and different Hb categories at baseline in patients with various cancers. Results are based on 14,288 cancer patients. (Modiﬁed from Ludwig et al., Eur. J. Cancer 2004)

Prevalence and incidence of anemia and risk factors

195

No Treatment

Radiotherapy

Combined Radio-Chemotherapy

Chemotherapy

O th er

H ea d/ N ec k G yn ae co l. Ly m ph /M ye lo U ro ge ni ta l

G I/C ol o

Lu ng

Br ea st

100 90 80 70 60 50 40 30 20 10 0

Fig. 2. Percentage of patients who were anemic (Hb <12 g/dL) at least once during the survey over 6 months. (Modiﬁed from Ludwig et al., Eur. J. Cancer 2004)

(81.4%) and with lung cancer (77.7%) (Fig. 2). ECAS is also the ﬁrst large study that evaluated the incidence of anemia in a cohort of 2,732 cancer patients who were not anemic at enrollment and received their ﬁrst cancer treatment during the survey period and had at least 2 cycles of chemotherapy (n = 2,101) or two data points after radiotherapy (n = 514). The overall incidence of anemia was 53.7%; 38.5% of patients had Hb levels between 10.0 and 11.9 g/dL, 13.8% had moderate anemia (Hb 8.0–9.9 g/dL), and 1.4% severe anemia (Hb <8 g/dL). Patients who received chemotherapy had a higher incidence of anemia (62.7%), compared with combined radiochemotherapy (41.9%) and radiotherapy only (19.5%) (Fig. 3). The incidence of anemia varied between patients with different types of cancer, was highest in patients with lung cancer (70.9%) followed by those with gynecological malignancies (64.6%) and increased steadily with the number of chemotherapy cycles being applied (cycle 1: 19.5%, cycle 2: 34.3%, cycle 3: 42.0% and 46.7% in cycles 4 and 5). Anemia treatment was given to 38.9% of patients who presented at least once during the survey with Hb <12 g/dL. Only a minority of patients with Hb <8.0 g/dL (0.9%) did not receive anemia treatment and anemia therapy was withheld in 12.9% who had Hb 8.0 and 9.9 g/dL. The trigger point was Hb 9.9 g/dL for erythropoietin therapy and Hb 8.6 g/dL for red cell transfusions. A subgroup of patients with lung cancer enrolled into the ECAS survey was studied by Kosmidis and Krzakowski (2005). Prevalence of anemia was 37.6% (753/2,002) at enrollment and varied according to treatment status.

196

H. Ludwig 100 90 75

80 % Patients

70 60

54

50

42

40 30

20

20 10

he m ot he ra py C

Pl at in um

C

ot he ra py N

on -P la tin um

R ad io /C he m

ad io th er ap y R

he m ot he ra py

0

Fig. 3. Incidence of anemia in patients who were nonanemic at baseline and followed up to 6 months in relation to treatment. (Modiﬁed from Ludwig et al., Eur. J. Cancer 2004)

Anemia was found in 50.0% of patients on concomitant chemotherapy/ radiotherapy, 39.0% on chemotherapy, 31.7% on radiotherapy, 38.6% on combination treatment, and 30.7%, on no treatment. Patients who were on platinum treatment at enrollment had a signiﬁcantly higher prevalence of anemia (50.1%) compared to those on nonplatinum chemotherapy (30.6%). During the 6-month follow-up, 83.3% of the patients on chemotherapy were anemic at least once during the survey. The prevalence of anemia increased progressively from 23.5% at cycle 1 to 77.3% at cycle 6 in patients on platinum-based chemotherapy and from 32.9% to 57.7% in patients on nonplatinum chemotherapy. Barrett-Lee et al. (2005) studied patients with female cancers enrolled into the ECAS survey in greater detail. They analyzed the data of 3,253 women with breast cancer and of 1,741 women with gynecologic cancer (ovarian, n = 1,173; cervical, n = 289; other, n = 279). Patients with breast cancer had a lower mean age (54.1 years; range, 21–93) than the gynecologic cancer patients (57.6 years; range, 18–89). Although disease status in terms of the proportion of patients with newly diagnosed, treated or relapsing/ persistent disease was similar between patients with breast and gynecologic cancer, more patients with gynecologic cancer (49.1%) than breast cancer (30.4%) were anemic (Hb <12 g/dL) at enrollment, and more gynecologic cancer (12.7%) than breast cancer (4.3%) patients had moderate to severe anemia (Hb ≤9.9 g/dL) at that time. Overall, 62.4% of breast cancer patients

Prevalence and incidence of anemia and risk factors

197

and 81.4% of gynecologic cancer patients were anemic at some time during the survey. Among the breast cancer patients those with persistent/recurrent disease were the most frequently anemic (41.7%), followed by newly diagnosed patients receiving chemotherapy (33.9%), patients in remission (22.6%), and newly diagnosed patients who had not received chemotherapy (20.4%). In the gynecologic cancer group anemia occurred most frequently in patients with persistent/recurrent disease (57.5%), followed by newly diagnosed patients receiving chemotherapy (52.4%), newly diagnosed patients not receiving chemotherapy (41%), and patients in remission (40.8%). Another subgroup of patients of the ECAS survey was analyzed in detail by Birgegard et al. (2006). They studied 1,612 patients with non-Hodgkin’s lymphoma (NHL) or Hodgkin’s disease (HD) and 704 patients with multiple myeloma (MM). At enrollment, 52.5% of patients with lymphoma or MM were anemic. Patients with MM were most frequently anemic (69.2%); 49.0% of patients with NHL were anemic, as were 37.4% of those with HD. When evaluated by malignancy, most (28.8%–39.5%) patients who were anemic had Hb levels of 10.0 to 11.9 g/dL. However, the prevalence of different Hb categories in patients with the individual diagnoses was not evenly distributed. Low Hb levels (≤9.9 g/dL) were much more common in patients with MM (29.7%) than in those with NHL (17.4%) or HD (8.6%). 72.9% experienced anemia at some time during ECAS. Among the 3 diagnostic groups, patients with MM were most frequently anemic (85.3%), followed by patients with NHL (77.9%) and those with HD (57.4%). The frequency of anemia was increased for patients 60 years of age and older compared with those under age 60. In multiple myeloma, Hb levels were <12 g/dL in 90% of patients aged 70 or older at least once during the survey (Fig. 4) and more than half (52.0%) of patients who were anemic at some time during the survey had Hb nadirs less than 10.0 g/dL. Displaying a pattern similar to that seen at enrollment, patients with MM were most likely to have Hb nadirs 100

90 85

90

79

80 %Patients

70

60

60 50 40 30 20 10 0 <40

40-59

60-69

≥70

Age, Years

Fig. 4. Percentage of patients with multiple myeloma “ever anemic” during the survey in relation to age. (Modiﬁed from Birgegard et al., Haematologica 2006)

198

H. Ludwig

≤9.9 g/dL compared to those with NHL (59.4% vs. 51.5%) and with HD (37.6%). Two hundred and thirteen chemotherapy patients fulﬁlled the criteria for inclusion in the chemotherapy incidence population. Of these patients, 55.4% became anemic. The incidence of anemia in chemotherapy patients was greater in the older age group, i.e. 48.9% in the group below 60 years of age became anemic during cancer therapy compared with 65.9% in the group aged 60 years and older. Steurer et al. (2004) conducted a retrospective chart review in 273 patients with chronic lymphocytic leukemia (CLL), MM, NHL, and HD, who were scheduled to receive cyclic chemotherapy. At baseline, prevalence of anemia was greatest in patients with MM (77.4%). Prevalence of anemia increased for all malignancies after cycle 4, with the largest increases noted for patients with NHL (from 35.1% at baseline to 73.7%) and HD (from 21.9% to 54.5%). Cyclic chemotherapy and prior anticancer treatment indicated an increased risk for developing anemia.

Risk factors for prediction of anemia Several authors investigated factors associated with increased risk for developing anemia (Table 2). Sadahiro et al. (1998) analyzed the relationship between clinical parameters and risk for anemia, which was deﬁned as Hb <10 g/dL. Univariate analysis showed that carcinoma of the cecum, ascending colon, and transverse colon, large-size carcinoma, invasion beyond the proper muscle layer, lymph node metastasis and clinical stage (Dukes’ B, C, and D) were factors associated with high incidence of anemia. Histological type did not affect the hematological ﬁndings. Multivariate analysis showed that age, tumor site, and tumor size were signiﬁcant factors related to anemia but not to iron deﬁciency. Depth of invasion, the presence or absence of lymph node metastasis, and Dukes’ classiﬁcation were not signiﬁcant factors. In another retrospective chart review, Skillings et al. (1995) studied 616 patients that were started on chemotherapy. Data collection ﬁnished 4 weeks after the end of the ﬁrst regimen or after a maximum follow-up period of 26 weeks. Seventy-two patients (12%) were transfused for anemia (reasons other than blood loss), and 28% of the subjects were anemic during treatment. Multivariate analysis revealed treatment with platinum compounds (odds ratio [OR]: 6.69), anthracyclines (OR: 3.56), and disease stage (OR: 1.62) as signiﬁcant risk factors. Ray-Coquard et al. (1999) developed a risk model for chemotherapyinduced severe anemia requiring RBC transfusions (SARRT) within 31 days after the administration of chemotherapy in a cohort of cancer patients. The risk model was developed on a series of 1,051 patients (analysis population) and tested in two subsequent patient cohorts comprising 797 and 295 patients,

2,070 Not anemic at enrollment (Hb ≥12 g/dL)

331

385

Barrett-Lee PJ et al., Oncology, 2006

Dranitsaris G et al., Lancet Oncol 2005

Sadahiro S et al., J Gastroenterol, 1998

2,360 enrolled

analysis population: 1,051 test population: 797 and 295 40

Ray-Coquard I et al., J Clin Oncol, 1999

Pivot X et al., Eur J Cancer, 2000 Birgegard G et al., Haematologica, 2006

Number of patients

Author

Table 2. Risk factors for anemia

Colorectal cancer

Breast; lung; head and neck; gynecologic cervix, ovary, uterus); gastrointestinal (GI)/colorectal; urogenital (prostate, urethral, male genital organs, bladder, kidney); lymphoma/myeloma (including CLL); leukemia (acute lymphocytic leukemia and acute myelogenous leukemia); and “other” Breast cancer receiving adjuvant chemotherapy

Breast, colorectal, ovary, head and neck, other gastrointestinal, lung, lymphoma, bone and soft tissue, germ cell tumors, other Head and neck cancer treated with cisplatin and 5-ﬂuorouracil Evaluable 2,316 MM: 704; NHL and HD: 1612

Patient population

Low pretreatment Hb, platelet count <200.000/μl, age >65 years, type of adjuvant chemotherapy, antibiotic prophylaxis Age, tumor site, and tumor size

Risk factors for severe anemia related red cell transfusion: d1 Hb <12 g/dL, performance status >1, and d1 lymphocyte count < or = 700/μL Ultraﬁlterable platinum concentration >50 ng/ml associated with anemia risk Low initial Hb (females Hb <12.7 g/dL, males initial Hb <13.3 g/dL, persistent/ recurrent disease, female gender, treatment with platinum chemotherapy. Enrollment Hb ≤12.9 g/dL females, Hb ≤13.5 g/dL males Lung/Gyn cancer vs. GI, colorectal cancer Other cancers vs. GI, colorectal cancer Platinum vs. nonplatinum treatment Female vs. male

Risk factors for anemia

Prevalence and incidence of anemia and risk factors 199

200

H. Ludwig

respectively. One hundred and seven (10%) of the analysis population experienced SARRT. In univariate analysis, only female sex, performance status greater than 1, Hb level less than 12 g/dL before chemotherapy on day 1 (d1), and d1 lymphocyte count ≤700/μL signiﬁcantly correlated with the risk of SARRT. Using logistic regression, d1 Hb level less than 12 g/dL (OR = 14.0; 95% conﬁdence interval [CI], 7–30), performance status greater than 1 (OR = 2.2; 95% CI, 1.4–3.5), and d1 lymphocyte count ≤700/μL (OR = 1.7; 95% CI, 1.1–2.6) were identiﬁed as independent risk factors for SARRT. These three factors were given arbitrary risk coefﬁcients of 3, 1, and 1 respectively, and a risk score for each individual patient was obtained by adding the coefﬁcients. The calculated probability of RBC transfusions was 30% for patients with a score > or =4, and 11%, 4%, and 1% in patients with a score of 2 or 3, 1, and 0, respectively. The authors concluded that patients at high risk for chemotherapy-induced SARRT might be appropriate candidates for prophylactic erythropoietin treatment. A small study aiming to identify predictors for cisplatin-induced anemia was reported by Pivot et al. (2000) in 40 patients with head and neck cancer. Patients received 3 cycles of a cisplatin-5-ﬂuorouracil regimen at three weekly intervals. Total and ultraﬁlterable plasma platinum concentration was measured 16 h after administration and 5-FU pharmacokinetics was measured in all patients. Thirty-eight percent of patients showed signiﬁcant decrease in Hb concentration (Hb ≥3 g/dL). Patient age, 5-FU area under the curve (AUC) (0–105 h) and total platinum concentration were unrelated to Hb loss. In contrast, ultraﬁlterable (UF) platinum concentration was signiﬁcantly correlated to Hb loss: the higher the UF platinum concentration, the greater the Hb loss (P = 0.015). A discriminant analysis allowed a cut-off value for UF platinum to be proposed to identify patients developing significant loss of Hb: 91% of patients exhibiting an UF platinum concentration above 50 ng/ml developed signiﬁcant decrease in Hb levels in contrast to 18% in the group of patients with an UF platinum concentration below 50 ng/ml (OR = 46; 95% CI, 4.7–446). The Australian cancer anemia survey was designed similarly as the ECAS study and revealed anemia in 35% of the patients with solid tumors or hematological malignancies at enrollment, with 78% of these 199 patients having mild anemia (Hb 10.0–11.9 g/dL): Independent predictors for anemia in chemotherapy patients were low baseline Hb level (OR, 5.4; 95% CI, 2.7–10.9) and use of platinum chemotherapeutic agents (OR, 4.8; 95% CI, 2.1–11.4) (P < 0.001) (Seshadri et al. 2005). Part of the main objective of the ECAS survey was to deﬁne risk factors for anemia in the entire patient population and in patients with speciﬁc tumor types. Barrett-Lee et al. (2006) studied 2,070 nonanemic patients (Hb >12 g/dL) at enrollment who received their ﬁrst chemotherapy during ECAS. Patients were divided randomly into split half (SH) 1 and SH2 (n = 1,035 each). The model was developed on SH1 using logistic regression to simul-

Prevalence and incidence of anemia and risk factors

201

taneously evaluate predictive factors, and was validated using SH2 and an additional similar subpopulation of 5,901 ECAS patients. The variables ultimately identiﬁed as signiﬁcant predictive factors for anemia were: lower initial Hb (< or =12.9 g/dL in females, and < or =13.4 g/dL in males); having lung or gynecologic cancer versus gastrointestinal (GI)/colorectal cancer; cancer at any other site versus GI/colorectal cancer; treatment with platinum chemotherapy, and female gender (Table 3). Each predictive factor for anemia was assigned an Anemia Risk Value that preserved the original predictive relationship of the variables by using the beta coefﬁcients. A lower initial Hb has an Anemia Risk Value of 13, treatment with platinum chemotherapy has an Anemia Risk Value of 7, being female has an Anemia Risk Value of 4. Adding together the Anemia Risk Values that correspond to the baseline characteristics of a patient yields the Total Anemia Risk Score for that patient. Patients with Total Anemia Risk Scores 1 to 15 were considered to have low risk for anemia. Although 45% of patients with these scores became anemic, this was signiﬁcantly lower than 71% of patients with scores 16 to 23 who became anemic and were considered at high risk, and 84% of patients with scores 24+ who became anemic and were considered to be at highest risk for anemia. Using the overall model, a breast cancer patient with baseline Hb 13.8 g/dL receiving adjuvant chemotherapy without platinum has a risk score of 12 and a low risk of anemia. A patient with prostate cancer, baseline Hb of 12.5 g/dL has a risk score of 25 and, hence, is at high risk for anemia. The risk factors for anemia were subsequently studied in more detail in a subgroup of lymphoma and myeloma patients enrolled into the ECAS survey by Birgegard et al. (2006). A population of 678 myeloma or lymphoma patients that presented with normal Hb (>12 g/dL) at enrollment was subjected to ≥2 cycles of chemotherapy and analyzed for potential predictors for the development of anemia. Four dichotomous variables were found to signiﬁcantly (P < 0.03) increase the risk for becoming anemic: low initial Hb (males Hb <12.7 g/dL, females Hb <13.3 g/dL, OR: 4), persistent/recurrent disease (OR: 1.5), female gender (OR: 3), and treatment with platinum chemotherapy (OR: 5.5). Platinum was administered to 114 patients with L/MM (6.3%); these included 82 of 889 patients with NHL (9.2%), 15 of 394 with HD (3.8%), and 17 of 519 with MM (3.3%). Similarly, risk factors for developing anemia in lung cancer patients have been described (Kosmidis and Krzakowski 2005). At enrollment, of 605 patients receiving platinum therapy, 50.1% were anemic versus 30.6% of 1,252 receiving nonplatinum regimens. During ECAS, 83.3% of lung cancer patients who received chemotherapy were anemic at some time, with the prevalence of anemia in platinum-treated patients increasing progressively from 23.5% at cycle 1 to 77.3% at cycle 6 (corresponding values for nonplatinum treated patients, 32.9% and 57.7%). Logistical analysis of ECAS

2.0–5.6 1.6–3.5 1.4–3.0 1.1–2.0

3.3 2.3 2.1 1.5

Beta Coefﬁcient 1.32

1.21 0.85 0.72 0.38

Wald P-Values <0.0001 <0.0001 <0.0001 0.0003 0.0158

12 8 7 4

13

Anemia Risk Values

AOR = adjusted odds ratio; CI = conﬁdence interval; GI = gastrointestinal; * Other malignancies excluding lung or gynecologic cancers. Risk categories: Low risk: 1–15, Intermediate risk: 16–24, high risk: >24.

2.7–5.2

3.8

Enrollment Hb: lowest 40% ≤12.9 females, ≤13.4 males Cancer type: Lung/Gyn vs. GI/Colorectal Other* vs. GI/Colorectal Treatment with platinum Gender; Female vs. Male

95% CI

AOR

Predictor Variable

Table 3. Variables found to be statistically signiﬁcant predictive factors for anemia using logistic regression modeling. (Modiﬁed from Barrett-Lee et al., 2006)

202 H. Ludwig

Prevalence and incidence of anemia and risk factors

203

data identiﬁed treatment with platinum, female sex, and low initial Hb level as signiﬁcant risk factors for anemia in lung cancer patients. Further, the ECAS data set was used to analyze the risk of anemia in patients with breast and gynecologic cancers by Barrett-Lee et al. (2005). The incidence of anemia in patients who received chemotherapy was determined from an “incidence population,” deﬁned as patients not anemic at enrollment or receiving anemia treatment who received their ﬁrst chemotherapy during the survey and underwent at least two consecutive chemotherapy cycles during the survey. The incidence of anemia was found to be 59.8% for patients with breast cancer and 74.8% for those with gynecologic cancer. The incidence of anemia increased with increasing cycles of chemotherapy, rising from 16.1% for breast cancer patients and 25.9% for gynecologic cancer patients at cycle 1 to 49.8% and 60.5%, respectively, at cycle 6 (Fig. 5). The times required for Hb level to decline to 12 g/dL or 11 g/dL were comparable for patients with breast cancer and gynecologic cancer, but Hb decline to 10 g/dL occurred more rapidly in patients with gynecologic cancer. Examination of the breast cancer group data (n = 298) showed a cumulative effect of chemotherapy on anemia development, with 28.8% of patients experiencing the ﬁrst occurrence of anemia during cycle 1, 26.2% experienced the ﬁrst occurrence during cycle 2 (cumulative rate, 53%), and 19.1% with the

Enrollment

Cycle 1

Cycle 2

Cycle 3

Cycle 4

Cycle 5

Cycle 6 0

10

20

30

40

50

% of Patients Hb 10.0-11.9

Hb 8.0-9.9

Hb <8

Fig. 5. Increasing incidence of anemia in patients with breast cancer with increasing numbers of cyclic chemotherapy. (Modiﬁed from Barrett-Lee et al., Oncologist 2005)

204

H. Ludwig

ﬁrst occurrence during cycle 3 (cumulative rate, 72.1%). Corresponding rates for gynecologic cancer patients were: cycle 1, 34.7%; cycle 2, 28.7% (cumulative rate, 63.4%), and cycle 3, 12.4% (cumulative rate, 75.7%). Dranitsaris et al. (2005) reviewed the medical records of 331 patients who had received adjuvant chemotherapy for breast cancer. They applied a multivariate logistic regression analysis to develop a risk scoring system ranging from 0 (low risk) to 50 (high risk), based on the ﬁnal regression variables. Results revealed that the risk of anemia increased as the pretreatment Hb concentration decreased and was reduced with successive chemotherapy cycles. Risk was also predicted by a platelet count of 200 × 10(9) cells/L or less before chemotherapy, age 65 years or older, type of adjuvant chemotherapy, and use of prophylactic antibiotics. ROC analysis had acceptable areas under the curve of 0.88 for the internal-validation sample and 0.84 for the external validation sample. A risk score of > or =24 to <25 before chemotherapy was identiﬁed as the optimum cut-off for maximum sensitivity (83.5%) and speciﬁcity (92.3%) of the prediction model. In conclusion, anemia is a frequent complication in patients with cancer and found in roughly 4 of 10 patients with cancer. Given the fact that the estimated prevalence of cancer in Europe amounts to more than 5 million cases (Pisani P et al. 1999) at least 2 million cancer patients should be affected by anemia in Europe today. Most of them will present with mild anemia (Hb 10–11.9 g/dL) only, about 10% will suffer from moderate to severe anemia. From a clinical viewpoint it is important to identify patients at risk for anemia, because early intervention with erythropoietic stimulating agents precludes development of severe anemia in a majority of these patients (Lyman and Glaspy 2006). The most important factors predicting a high risk of anemia are low baseline Hb, treatment with platinum drugs, female gender and high age in patients with most cancers.

Acknowledgement This study was supported by the Austrian Forum against cancer. Ms. Silvia Bakos assisted in proofreading and correcting the manuscript.

References 1. Barrett-Lee P, Bokemeyer C, Gascon P, Nortier JW, Schneider M, Schrijvers D, Van Belle S; ECAS Advisory Board and Participating Centers (2005) Management of cancer-related anemia in patients with breast or gynecologic cancer: new insights based on results from the European Cancer Anemia Survey. Oncologist 10: 743–757 2. Barrett-Lee PJ, Bailey NP, O’Brien ME, Wager E (2000) Large-scale UK audit of blood transfusion requirements and anaemia in patients receiving cytotoxic chemotherapy. Br J Cancer 82: 93–97

Prevalence and incidence of anemia and risk factors

205

3. Barrett-Lee PJ, Ludwig H, Birgegard G, Bokemeyer C, Gascon P, Kosmidis PA, Krzakowski M, Nortier JW, Kongable G, Schneider M, Schrijvers D, Van Belle SJ (2006) Independent risk factors for anemia in cancer patients receiving chemotherapy: results from the European Cancer Anaemia Survey. Oncology 70: 34–48 4. Beutler E, Waalen J (2006) The deﬁnition of anemia: What is the lower limit of normal of the blood hemoglobin concentration. Blood 107: 1747–1750 5. Birgegård G, Gascón P, Ludwig H (2006) Evaluation of anemia in patients with multiple myeloma and lymphoma as deﬁned by the European Cancer Anaemia Survey (ECAS). Haematologica (in press) 6. Common Toxicity Criteria, Version 2.0. Cancer Therapy Evaluation Program (1999) DCTD, NCI, NIH, DHHS. March 1998 7. Coifﬁer B, Guastalla JP, Pujade-Lauraine E, Bastit P; Anemia Study Group (2001) Predicting cancer-associated anaemia in patients receiving non-platinum chemotherapy: results of a retrospective survey. Eur J Cancer 37: 1617–1623 8. Denny SD, Kuchibhatla MN, Cohen HJ (2006) Impact of anemia on mortality, cognition, and function in community-dwelling elderly. Am J Med 119: 327– 334 9. Dranitsaris G, Clemons M, Verma S, Lau C, Vincent M (2005) Chemotherapyinduced anaemia during adjuvant treatment for breast cancer: development of a prediction model. Lancet Oncol 6: 856–863 10. Harrison L, Shasha D, Shiaova L, White C, Ramdeen B, Portenoy R (2001) Prevalence of anemia in cancer patients undergoing radiation therapy. Semin Oncol 28 [2 Suppl 8]: 54–59 11. Kosmidis P, Krzakowski M; The ECAS Investigators (2005) Anemia proﬁles in patients with lung cancer: what have we learned from the European Cancer Anaemia Survey (ECAS)? Lung Cancer 50: 401–412 12. Ludwig H, Van Belle S, Barrett-Lee P, Birgegard G, Bokemeyer C, Gascon P, Kosmidis P, Krzakowski M, Nortier J, Olmi P, Schneider M, Schrijvers D (2004) The European Cancer Anaemia Survey (ECAS): a large, multinational, prospective survey deﬁning the prevalence, incidence, and treatment of anaemia in cancer patients. Eur J Cancer 40: 2293–2306 13. Lyman GH, Glaspy J (2006) Are there clinical beneﬁts with early erythropoietic intervention for chemotherapy-induced anemia? A systematic review. Cancer 106: 223–233 14. Moullet I, Salles G, Ketterer N, Dumontet C, Bouaﬁa F, Neidhart-Berard EM, Thieblemont C, Felman P, Coifﬁer B (1998) Frequency and signiﬁcance of anemia in non-Hodgkin’s lymphoma patients. Ann Oncol 9: 1109–1115 15. Pisani P, Bray F, Parkin DM (1999) Estimates of the world-wide prevalence of cancer for 25 sites in the adult population. Int J Cancer 97: 72–81 16. Pivot X, Guardiola E, Etienne M, Thyss A, Foa C, Otto J, Schneider M, Magne N, Bensadoun RJ, Renee N, Milano (2000) An analysis of potential factors allowing an individual prediction of cisplatin-induced anaemia. Eur J Cancer 36: 852–857 17. Ray-Coquard I, Le Cesne A, Rubio MT, Mermet J, Maugard C, Ravaud A, Chevreau C, Sebban C, Bachelot T, Biron P, Blay JY (1999) Risk model for severe anemia requiring red blood cell transfusion after cytotoxic conventional chemotherapy regimens. The Elypse 1 Study Group. J Clin Oncol 17: 2840–2846

206

H. Ludwig: Prevalence and incidence of anemia and risk factors

18. Sadahiro S, Suzuki T, Tokunaga N, Mukai M, Tajima T, Makuuchi H, Saito T (1998) Anemia in patients with colorectal cancer. J Gastroenterol 33: 488–494 19. Seshadri T, Prince HM, Bell DR, Coughlin PB, James PP, Richardson GE, Chern B, Briggs P, Norman J, Olver IN, Karapetis C, Stewart J; Australian Cancer Anaemia Study Group (2005) The Australian Cancer Anaemia Survey: a snapshot of anaemia in adult patients with cancer. Med J Aust 182: 453–457 20. Skillings JR, Rogers-Melamed I, Nabholtz JM, et al (1995) An epidemiological review of anaemia in cancer chemotherapy in Canada. Eur J Cancer 31A [Suppl 5]: S183 (Abstr) 21. Skillings JR, Rogers-Melamed I, Nabholtz JM, Sawka C, Gwadry-Sridhar F, Moquin JP, Rubinger M, Ganguly P, Burnell M, Shustik C, Dryer D, McLaughlin M, White D (1999) An epidemiological review of red cell transfusions in cancer chemotherapy. Cancer Prev Control 3: 207–212 22. Steurer M, Wagner H, Gastl G (2004) Prevalence and management of anaemia in haematologic cancer patients receiving cyclic nonplatinum chemotherapy: results of a prospective national chart survey. Wien Klin Wochenschr 116: 367–372 Correspondence: Prof. Dr. Heinz Ludwig, Department of Medicine I, Center for Oncology and Hematology, Montleartstrasse 37, 1171 Vienna, Austria, E-mail: [email protected]

Chapter 8

Signiﬁcance of anemia in cancer chemotherapy M. R. Nowrousian Department of Internal Medicine (Cancer Research), West German Cancer Center, University Hospital of Essen, Essen, Germany

Introduction As a common symptom in cancer, anemia is often induced either directly or indirectly by the malignant disease itself or its treatment. The prevalence and incidence of anemia depend on one side on the type and stage of malignancy and on the other side on the type, schedule and intensity of treatment. Hematological malignancies, particularly myeloid disorders, are generally more often associated with anemia than solid tumors but, depending on the type of treatment, the latter may be associated with anemia and require red blood cell (RBC) transfusions in a similarly high proportion of patients (Skillings et al. 1993; Coifﬁer et al. 2001). Anemia and the need for transfusions are also more frequently associated with advanced stage of diseases than early stage (Durie and Salmon 1975; Binet et al. 1977; Moullet et al. 1998; BarrettLee et al. 2000). Preexisting anemia related to cancer or prior treatment usually worsens during radiotherapy and chemotherapy, and in a considerable number of patients, radiotherapy and chemotherapy as such produce anemia, primarily due to their myelosuppressive effect (Skillings et al. 1993, 1999; Sadahiro et al. 1998; Estrin et al. 1999; Groopman and Itri 1999; Moullet et al. 1998; Lammering et al. 1999; Barrett-Lee et al. 2000, 2006; Harrison et al. 2000; Coifﬁer et al. 2001; Ludwig et al. 2004; Ludwig in this book; Steurer et al. 2004; Kosmidis et al. 2005; Seshardi et al. 2005; Birgegard et al. 2006). Myelosuppression is one of the most frequent side effects of chemotherapy, and, depending on the type and intensity of treatment, repeated cycles of chemotherapy may have cumulative toxic effects on hematopoiesis. Although erythropoiesis appears to be less readily affected than granulocytopoiesis or thrombocytopoiesis, many patients develop anemia during treatment and a considerable number of patients requires RBC transfusions (Skillings et al. 1993; Estrin et al. 1999; Groopman and Itri 1999; Barrett-Lee et al. 2000; Coifﬁer et al. 2001; Crawford et al. 2002; Ludwig et al. 2004). The frequency and severity of anemia, however, vary from drug to drug and are dependent on the dosage, schedule and combination of drugs (Tables 1–3).

208

M. R. Nowrousian

Table 1. Frequency and severity of anemia associated with cytotoxic agents given as single drug Cytotoxic agent

Anemia (grade) I/II (%)

References

III/IV (%)

Previously untreated patients with solid tumors: Cisplatin 68 2–11 Carboplatin 66 0–26 Paclitaxel 23–93 0–13 Docetaxel 73–97 0–14 Gemcitabine 63–69 5 Methotrexate 25 3 Vinorelbine 50–77 1–21 Topotecan 31–90 4–32 Irinotecan 60 8 5-FU 50–54 5–11 5-FU, LV 6–53 2–5 UFT, LV 3–21 0

136, 228 143, 228 56, 80, 85, 184, 191, 220 35, 39, 87, 89, 90, 133, 183 9, 101 84 92, 93, 139, 200, 294 52, 53, 236, 256 227 111, 127, 136, 209 1, 151, 209 108, 232

Previously treated patients with solid tumors: Paclitaxel 18–90 2–64 Docetaxel 60–87 3–42 Vinorelbine 6–40 3–14 Topotecan 64–87 12–46 Irinotecan 49 10–62 Etoposide 31–56 7–13 Ifosfamide 19 5–32

63, 75, 76, 192, 238, 270, 277 89, 90, 146, 211, 221, 257, 281 58, 100, 139, 144 10, 38, 52, 53, 74, 154, 270 94, 226, 227, 287 133, 224 70, 262

5-FU = 5-ﬂuorouracil, LV = leucovorin, UFT = tegafur-uracil.

Some cytotoxic agents appear to produce more frequently and severely anemia (Table 1), related either to a more severe myelosuppression or additional mechanisms, such as impairment of renal function and erythropoietin production (Wood and Hrushesky 1995; Unami et al. 1996; Horiguchi et al. 2006). In recent years, cancer-related or chemotherapy-induced anemia has been shown to have enormous impact on physical well-being, functional capacity and quality of life of patients (Abels 1993; Henry and Abels 1994; Leitgeb et al. 1994; Cella 1997 and 2002; Glaspy et al. 1997; Daneryd et al. 1998; Demetri et al. 1998; Gabrilove et al. 2001; Littlewood et al. 2001; Crawford et al. 2002; Ossa et al. 2007). In addition, anemia has been found to be a negative prognostic factor for the outcome of treatment and survival of patients with various types of malignant diseases (Table 4). In this regard,

Signiﬁcance of anemia in cancer chemotherapy

209

Table 2. Frequency and severity of anemia associated with chemotherapy in patients with solid tumors* Chemotherapy regimen

Cancer

Anemia (grade)

References

I/II (%) III/IV (%) Cisplatin, paclitaxel

Lung, Ovarian

43–60

5–9

Cisplatin, etoposide

Lung

59–73

16–42

Cisplatin, etoposide, Ifos Cisplatin, Cy

Lung Ovarian

32–97

23–53 2–29

Cisplatin, gemcitabine

Lung, Urinary tract, Cervix Lung Lung

63

13–40

61

24 9

Cisplatin, vinorelbine Cisplatin, MMC, vinblastine Cisplatin, vinblastine Cisplatin, doxorubicin, Cy Cisplatin, 5-FU Cisplatin, paclitaxel, 5-FU Cisplatin, paclitaxel, Ifos Cisplatin, irinotecan Cisplatin, Ifos without vs with paclitaxel Cisplatin, docetaxel Cisplatin, vinorelbine Carboplatin, docetaxel Carboplatin, etoposide Carboplatin, Cy Carboplatin, Ifos Carboplatin, etoposide, Ifos Carboplatin, paclitaxel Carboplatin, paclitaxel, etoposide Carboplatin, 5-FU Epirubicin, 5-FU, Cy, every 3 vs 2 weeks Doxorubicin without vs with cisplatin Doxorubicin, paclitaxel Doxorubicin, Cy, vincristine Doxorubicin, Cy, 5-FU, MTX Doxorubicin, Ifos (6 vs 12 g/m2) Cy, topotecan

Lung Ovarian 6 Head and neck 55–74 Head and neck 35 Head and neck Lung, cervix 80 Cervix

5–12 12 30 14–45 11 vs 18

152 48 84, 85, 136, 206 135 13, 245 137, 155, 193, 261 25

Lung Lung Lung Lung Ovarian Cervix Lung

7 24 11 54 3–42 30 6–11

88 88 88 171, 248 7, 263 24 295

2–34 32–35

150, 159, 249 115

Lung, Ovarian Lung

39 41–98 77–78 10–59

13

177, 213, 215, 288 116, 166, 181, 182, 223, 248 81, 166 7, 8, 176–177, 263 6, 140, 153, 168, 169, 243 298 77

Head and neck 42 14 Breast 19 vs 36 <1 vs 3

84, 85 286

Endometrium

4, 22

278

Breast Lung

78–84 16–54

8–11

103 83

Breast

27

1

26

Soft tissue sarcoma Pediatric solid tumors

30 vs 17

9 vs 23

297

27

234

* previously untreated. Cy = cyclophosphamide, Ifos = ifosfamide, 5-FU = 5-ﬂuorouracil, MMC = mitomycin C, MTX = methotrexate.

210

M. R. Nowrousian

Table 3. Frequency and severity of anemia associated with chemotherapy in patients with aggressive non-Hodgkin’s lymphoma (NHL) or Hodgkin’s disease (HD)* Chemotherapy regimen

ProMACE, MOPP MACOP-B CHOP CHOP-21 vs CHOP-14 in elderly patients (61–75 yrs)1 CHOEP-21 vs CHOEP-14 in elderly patients (61–75 yrs)1 CIE MOPP ABVD ABVD vs Stanford V vs MOPPEBVCAD2 COPP/ABVD vs BEACOPP standard in elderly patients (66–75 yrs)3 Doxorubicin, etoposide, vinblastine CEBOPP/VIML COPP/ABVD vs BEACOPP standard vs BEACOPP dose-intensiﬁed4 BEACOPP standard, timeintensiﬁed (every 14 days)

Malignancy

NHL NHL NHL NHL

Anemia (grade) I/II (%)

III/IV (%)

63 55 49

9 10 17–79 13 vs 20

240 240 108, 175 210

29 vs 45

210

62 12 0 0 vs 15 vs 19

231 30 30 107

24 vs 44

15

NHL NHL HD HD HD

Reference No.

38 31 5

HD

HD

59

13

31

HD HD

62

16 5 vs 17 vs 66

196 62

HD

83

65

246

* previously untreated.1–4: randomized studies. ProMACE = procarbazine, methotrexateleucovorin, doxorubicin, cyclophosphamide, etoposide; MOPP = mechlorethamine, vincristine, procarbazine, prednisone; MACOP-B = methotrexate-leucovorin, doxorubicin, cyclophosphamide, vincristine, prednisone, bleomycin; CHOP = cyclophosphamide, doxorubicin, vincristine, prednisone; CHOP-21, CHOP-14 = CHOP given every 21 and 14 days, respectively; CHOEP = CHOP plus etoposide; CIE = cyclophosphamide, idarubicin, etoposide; ABVD = doxorubicin, bleomycin, vinblastine, dacarbazine; Stanford V = doxorubicin, vinblastine, mechlorethamine, etoposide, vincristine, bleomycin, prednisone; MOPPEBVCAD = MOPP plus lomustine, vindesine, melphalan, epirubicin, vinblastine, bleomycin; COPP = cyclophosphamide, vincristine, procarbazine, prednisone; CEBOPP = cyclophosphamide, epirubicin, bleomycin, vincristine, procarbazine, prednisone; VIML = etoposide, ifosfamide, methotrexateleucovorin; BEACOPP = bleomycin, etoposide, doxorubicin, cyclophosphamide, vincristine, procarbazine, prednisone.

Signiﬁcance of anemia in cancer chemotherapy

211

Table 4. Cancers in which anemia has been reported to be a negative prognostic factor Hodgkin’s disease Non-Hodgkin lymphoma Multiple myeloma Waldenström macroglobulinemia Chronic lymphocytic leukemia Hairy cell leukemia Chronic myelogenous leukemia Ewing’s sarcoma Head and neck cancer Lung cancer (small-cell and non-small-cell) Breast cancer Esophageal cancer Gastric cancer Colorectal cancer Ovarian cancer Endometrial carcinoma Cervical cancer Renal cell carcinoma Urothelial cancer Bladder cancer Prostate cancer

anemia may be considered as an expression of a more aggressive disease or a more intensive immunological and inﬂammatory reaction of the host, but there is evidence suggesting that it may independently inﬂuence the results of radiotherapy and chemotherapy (Hirst 1986; Bush 1986; Thews et al. 1998; Grogan et al. 1999; Silver et al. 1999; Dunst 2000; Harrison et al. 2000; Wagner et al. 2000; Glaser et al. 2001; Höckel and Vaupel 2001; Littlewood et al. 2001a, 2001b, Obermair et al. 2001, 2003; Shasha 2001; Thews et al. 2001; Münstedt et al. 2003, 2004; Santin et al. 2003; Chua et al. 2004; Hofheinz et al. 2004; Peters-Engl et al. 2005; Laurie et al. 2006). Considering these aspects of anemia in cancer and the fact that it could be treated successfully in a considerable number of patients by the use of recombinant human erythropoietin (rhEPO), it appears to be of particular clinical interest to have reliable data on the incidence, pathophysiology, predictive factors and clinical signiﬁcance of anemia in various groups of patients to be able to select those groups that could be appropriate candidates for a treatment with rhEPO and could beneﬁt from such a treatment. These aspects of anemia, especially in cancer chemotherapy, are the subjects of this review.

212

M. R. Nowrousian

Incidence The frequency and severity of chemotherapy-induced anemia depends on the age of patients, the type and stage of malignant disease, previous treatment with radiotherapy or chemotherapy, and the type, schedule and intensity of chemotherapy. Important factors for transfusion requirement are the degree of anemia, the age of patients and the functional state of the cardiovascular and pulmonary system. Considering these factors and the fact that in many studies the course of hemoglobin has not been followed appropriately, it is often difﬁcult to assess the impact of various types of chemotherapy on erythropoiesis and the incidence and severity of anemia in various groups of patients. Nevertheless, there is a number of retrospective as well as prospective investigations, including large-scale audits and surveys, evaluating the prevalence and incidence of anemia and the need for RBC transfusion in cancer patients including those who receive chemotherapy (Skillings et al. 1993, 1999; Sadahiro et al. 1998; Estrin et al. 1999; Groopman and Itri 1999; Moullet et al. 1998; Lammering et al. 1999; Barrett-Lee et al. 2000, 2006; Harrison et al. 2000; Coifﬁer et al. 2001; Ludwig et al. 2004; Steurer et al. 2004; Kosmidis et al. 2005; Seshardi et al. 2005; Birgegard et al. 2006). The largest survey published is the European Cancer Anemia Survey (ECAS), in which prevalence and incidence of anemia and the risk factors for developing anemia were evaluated in a great cohort of patients and their subgroups related to various types of malignancies and treatments (Ludwig et al. 2004; see also Ludwig in this book). In addition to these studies, there are prospective chemotherapy trials reporting on the incidence of anemia associated with the use of various chemotherapeutic agents as single drug or in combination regimens commonly used for the treatment of certain types of malignant diseases (Tables 1–3). A comprehensive review of such trials published in the literature between 1990 and 1998 revealed a generally high incidence of anemia in patients receiving chemotherapy, and patients with lung cancer or ovarian cancer were reported to develop more frequently severe anemia (grade 3/4, Hb level <8 g/dl) than other groups of patients with solid tumors (Groopman and Itri 1999). The incidence of severe anemia in these two groups of patients was reported to be as high as 54% and 42%, respectively. Even as single agents, most cytotoxic drugs are able to produce anemia in a high proportion of patients, but there are differences in the frequency of severe anemia induced by various types of drugs (Table 1). Some agents, such as cisplatin, carboplatin, paclitaxel, docetaxel, vinorelbine, topotecan, irinotecan, etoposide, and ifosfamide seem to be frequently associated with severe anemia, particularly in previously treated patients. The latter group of patients is usually suffering more severely from myelosuppressive effects of chemotherapy than patients without prior treatment (Table 1). The frequency of severe anemia also depends on the combination of drugs. Regimens containing cisplatin or carboplatin or one of these two drugs together with etopo-

Signiﬁcance of anemia in cancer chemotherapy

213

side, ifosfamide, paclitaxel, cyclophosphamide, gemcitabine, vinorelbine or irinotecan appear to be particularly associated with a high incidence of severe anemia (Table 2). Such regimens are widely used in the treatment of patients with solid tumors, particularly those with head and neck, lung, ovarian, cervical or testicular cancer, a factor that may explain the high incidence of anemia, particularly its severe form, in these groups of patients (Table 2) (Skillings 1993, 1995; Groopman and Itri 1999; Ludwig et al. 2004; see also Ludwig in this book). Among the patients receiving platinum-based chemotherapy, those with testicular cancer appear to be transfused less frequently because of a markedly lower mean age of these patients compared with other groups (Barrett-Lee et al. 2000). Anemia is also a common side effect of non-platinum-based chemotherapy. In a retrospective analysis of data of 1064 patients with solid tumors or hematological malignancies receiving non-platinum-based chemotherapy, the proportion of anemic patients increased from 37% before the ﬁrst cycle of treatment to 54% by the third cycle (Coifﬁer et al. 2001). Fifteen percent of patients required RBC transfusions at some time during treatment. The highest transfusion rate was found in patients with non-Hodgkin lymphoma (NHL) (33%) followed by patients with Hodgkin’s disease (20%), ovarian cancer (17%), breast cancer (8%), lung cancer (6%) or colorectal cancer (5%). Patients with NHL were mainly treated with regimens containing doxorubicin or epirubicin in combination with cyclophosphamide. Such regimens may be associated with a high incidence of anemia, including its severe form, as observed in studies using CHOP for treatment of aggressive NHL (Table 3). The proportion of severe anemia in these patients may be as high as 79% (Meyer et al. 1995). In a study of patients with breast cancer, adjuvant chemotherapy consisting of six cycles of epirubin, cyclophosphamide and 5ﬂuorouracil induced a mean hemoglobin decrease of 3.1 g/dl during treatment, and 52% of patients developed hemoglobin levels of ≤10 g/dl by the end of the sixth cycle (Del Mastro et al. 1997). Other important factors determining the frequency and severity of anemia in cancer chemotherapy are the schedule and intensity of treatment. In a study in patients with small-cell lung cancer, intravenous topotecan was found to produce grade 3/4 anemia in 31% of cases compared with 23% in patients with oral topotecan (Eckardt et al. 2007). In a randomized study investigating two different schedules of etoposide in combination with cisplatin in patients with lung cancer, the proportion of patients with severe anemia was 32%, if etoposide was given orally for 21 days (50 mg/m2/d), every 28 days for six cycles, compared with 55%, if it was given intravenously for three days (130 mg/m2/d), every 21 days for eight cycles (Miller et al. 1995). In a randomized study evaluating the importance of cisplatin dose in patients with ovarian cancer, 56% of patients who were treated with 50 mg/m2 cisplatin plus 750 mg/m2 cyclophosphamide developed anemia compared with 82% of patients who received 75 mg/m2 cisplatin and the same dose of

214

M. R. Nowrousian

cyclophosphamide per cycle for a maximum of six cycles (Kaye et al. 1992). The proportion of patients with severe anemia was 0% in the ﬁrst group of patients and 15% in the second. In a randomized study investigating the efﬁcacy of doxorubicin (60 mg/m2) combined with two different dosages of ifosfamide in patients with soft tissue sarcoma, the incidence of grade 3/4 anemia was 23% in patients who received 6 g/m2 ifosfamide compared with 58% in those receiving 12 g/m2 ifosfamide per cycle (Worden et al. 2005). Another example for the impact of the dose intensity on incidence of anemia is the use of BEACOPP in its standard or intensiﬁed version in patients with Hodgkin’s disease (Table 3). In a randomized study, 17% of patients who received 8 cycles of standard-dose BEACOPP developed grade 3/4 anemia compared with 66% of patients who were randomly assigned to receive an escalated dose of this drug combination (Diehl et al. 2003) (Table 3). However, standard-dose BEACOPP used in elderly patients may also be associated with a considerably higher incidence of severe anemia. In a study using this drug combination in Hodgkin’s lymphoma patients at an age of 66–75 years, the proportion of patients with grade 3/4 anemia was 41% compared with 24% in patients who received COPP/ABVD (Table 3) (Ballova et al. 2005). The risks of developing anemia and requiring red blood cell (RBC) transfusions increase not only with the age of patients or increased dose of the drugs but also with the addition of new myelotoxic agents or by shortening the time interval between chemotherapy cycles. In a study including younger patients (median age 31 years, range 17–63) with Hodgkin’s disease and using a time-intensiﬁed version of standard-dose BEACOPP (given every 14 days instead of 21 days), 83% of patients suffered from anemia during treatment and 65% of patients had grade 3/4 anemia (Sieber et al. 2003). Another example is a randomized study comparing the efﬁcacy of CHOP given every 21 or 14 days with or without etoposide in elderly patients (61–75 years) with aggressive non-Hodgkin’s lymphoma. The proportions of patients who developed grade 3/4 anemia during the 6 cycles of chemotherapy were 12.5% with CHOP-21, 19.5% with CHOP-14, 28.7% with CHOEP-21, and 45.1% with CHOEP-14 (p < .001) (Pfreundschuh et al. 2004). Interestingly, the proportions of patients who required RBC transfusions were 24.6%, 40.2%, 39.2%, and 64.3% (p < .001), respectively, indicating that in each treatment group, a part of patients with grade 1/2 anemia also had to be given RBC transfusions. The results of the above-mentioned studies clearly show that a high proportion of patients receiving chemotherapy is suffering from anemia during treatment, and, depending on the age of patients, the type, schedule, and intensity of treatment, a high percentage of patients requires RBC transfusion. The incidence of anemia and the need for RBC transfusion are particularly high in elderly patietns and patients who receive chemotherapeutic regimens containing platinum or anthracycline in combination with agents such as etoposide, ifosfamide, cyclophosphamide, paclitaxel, gem-

Signiﬁcance of anemia in cancer chemotherapy

215

citabine, vinorelbine or irinotecan. In these groups of patients, the incidence of anemia and its severe form may be as high as 97% and 79%, respectively (Tables 2, 3).

Pathophysiology Myelosuppression is the most frequent cause of anemia induced by the majority of cytotoxic agents (Daniel and Crawford 2006). Hemolysis, related either to antibody production or microangiopathy, may also be a causative mechanism, but it is relatively rare and more often associated with certain agents, such as ﬂudarabine or mitomycin C, respectively (Doll and Weiss 1985; Rytting et al. 1996; Hamblin et al. 1998). Drug-induced hemolysis, however, can also occur during treatment with other cytotoxic drugs and should be generally considered as a possible cause of anemia in patients receiving chemotherapy (cf. Nowrousian chapter 5 in this book). Cisplatin and carboplatin are known to be associated more frequently with anemia and the need for RBC transfusions than other cytotoxic agents, and anemia induced by these two drugs appears to be disproportionate to their effects on other blood cells (Rossof et al. 1972; Wiltshaw et al. 1976; Nowrousian and Schmidt 1982; Canetta et al. 1985; Matsumoto et al. 1990; Abels et al. 1991; Kaye et al. 1992; Okamoto et al. 1992; Abels et al. 1994; Wood and Hrushesky 1995; McLaren et al. 2000; Pivot et al. 2000; Hensley et al. 2001; Jiang et al. 2007). Animal experiments and clinical studies show that the more pronounced inhibition of erythropoiesis by cisplatin and carboplatin is not related to a more selected suppression of erythroid progenitor cells (Nowrousian and Schmidt 1982), but a deﬁcient EPO production (Wood and Hrushesky 1995; Unami et al. 1996), possibly resulting from a direct suppression or damage of EPO producing proximal renal tubular cells (Horiguchi et al. 2006). In a study of patients receiving platinum-based chemotherapy, the degree of anemia was found to correlate with the cumulative dose of cisplatin, a progressive decrease in renal function and a progressively defective EPO production (Table 5) (Wood and Hrushesky 1995). In rats, repeated application of cisplatin similarly resulted in a progressive renal dysfunction and anemia associated with a reduced EPO and reticulocyte response, but without effects on WBC counts. After treatment with rhEPO, reticulocytes and hematocrit responded adequately, indicating that there was only minimal erythroid precursor cell damage from cisplatin (Wood and Hrushesky 1995). The mechanism by which cisplatin or carboplatin suppress erythropoietin production may be a speciﬁc inhibition of hypoxia-induced signalling and expression of EPO mRNA, which occurs in vitro, when EPO producing cells, such as those from a human hepatoma cell line, are exposed to cisplatin. This effect was reported to be dose-dependent, but not associated with damage

216

M. R. Nowrousian

Table 5. Relationship between renal function, EPO production, hemoglobin level, and RBC transfusions in patients with ovarian or bladder cancer treated with platinum-based chemotherapy Chemotherapy cycles 1

2

3

4

5

6

7

8

9

10

Magnesium (meq/l) 1.7 1.7 1.5 1.3 1.4 1.4 1.3 1.3 1.4 1.4 CrCl (cm3/min/1.73 m2) 83.9 76.9 71.4 68.8 64.9 59.1 61.7 54.3 53.1 49.1 O/E EPO 80.1 65.2 60.8 60.9 53.2 66.6 54.4 57.4 62.7 58.3 Hemoglobin (g/dl) 12.3 11.8 11.3 10.9 10.7 11.3 11.1 10.7 10.1 10.8 Cumulative units RBC 0.2 0.4 0.8 1.6 2.5 3.9 5.0 5.9 6.9 8.4 EPO = erythropoietin, RBC = red blood cell, CrCl = creatinine clearance. O/E EPO = ratio of observed to expected serum EPO levels × 100. Values are means. Adapted from ref. (296). Table 6. Comparison of renal toxicities and anemia before and during chemotherapy and in long-term follow-up in patients treated with platinum-based chemotherapy Study entry

Completion of therapy

Long-term follow-up (4.2 yrs)

Renal function: CrCl (cm/min/1.73 m2) Serum Mg (meq/l) No. of patients Mg supplemented

82.1 1.8 0/10

46.8* 1.4* 5/10*

41.7* 1.4 0/10

Erythropoietic parameters: Hemoglobin (g/dl) Serum EPO (mU/ml) No. of patients transfused

12.6 52.2 0/10

10.5* 30.8* 7/10*

13.9 51.4 0/10

The results are given as means. Chemotherapy = cisplatin (total dose 460 mg/m2) plus doxorubicin. * signiﬁcantly (<0.05) different from the study entry values. CrCl = creatinine clearance. Mg = magnesium. EPO = erythropoietin. Adapted from ref. (296).

of cells (Horiguchi et al. 2000). In a long-term follow-up study of patients with cisplatin-induced anemia, Hb level and serum EPO concentration also recovered completely after cessation of treatment, although serum magnesium concentration remained permanently reduced as a sign of a sustained tubular cellular damage (Table 6) (Wood and Hrushesky 1995). Platinum-

Signiﬁcance of anemia in cancer chemotherapy

217

based chemotherapy does not appear to be associated with defective EPO production and consequently severe anemia, if renal dysfunction is prevented by hydration and forced diuresis (Canaparo et al. 2000). Another interesting observation is the protective effect of EPO against nephrotoxicity of various agents including cisplatin. In a rat model, the use of EPO has been found to prevent cisplatin-induced renal tubular damage and to enhance the recovery of tubular cell generation after cisplatin. EPO has also been found to correct cisplatin-induced anemia (Vaziri et al. 1994; Baldwin et al. 1998; Bagnis et al. 2001; Yalcin et al. 2003).

Predictive factors As shown in a number of studies, patients with a low level of hemoglobin (<12 g/dl) at the start of chemotherapy are particularly at risk of developing severe anemia and requiring RBC transfusions (Skillings et al. 1993; Abels et al. 1994; Heddens et al. 1998; Thatcher 1998; Ray-Coquard et al. 1999; Skillings et al. 1999; Hensley et al. 2001; Ludwig et al. 2004; see also Ludwig in this book; Dranitsaris et al. 2005). Similarly, patients with platinum-based chemotherapy more frequently develop anemia and need RBC transfusions (Abels et al. 1991, 1992, 1993; Groopman and Itis 1999; Skillings et al. 1999; Ludwig et al. 2004; see also Ludwig in this book). The frequency of anemia in these patients increases with the dosage of platinum and the number of additional cytotoxic drugs (Kaye et al. 1992; McLaren et al. 2000; Hensley et al. 2001). Cisplatin and carboplatin appear to be comparable in the rate of anemia they produce (Canetta et al. 1985), but in combination with certain drugs such as paclitaxel, carboplatin seems to produce more often severe anemia requiring RBC transfusions than cisplatin (Hensley et al. 2001). During treatment with platinum-based chemotherapy, the frequency of transfusion requirement can amount to 47%–100%, depending on the cumulative dose of platinum and patient-related as well as disease-related risk factors (Rossof et al. 1972; Wiltshaw and Kroner 1976; Okamoto et al. 1992; Gamucci et al. 1993; de Campos et al. 1995; Wood and Hrushesky 1995). At a dose of 60–80 mg/m2 of cisplatin or 300 mg/m2 of carboplatin per cycle, transfusion requirement usually occurs after 3–4 cycles. Predictive factors for developing anemia are advanced age, loss of body weight before treatment, advanced stage of disease and, particularly, a low primary level of Hb and a marked decrease in Hb level (1–2 g/dl) after the ﬁrst or second cycle of chemotherapy (Table 7) (Okamoto et al. 1992; Skillings et al. 1993; Thatcher 1994; de Campos et al. 1995; Wood and Hrushesky 1995; Heddens et al. 1998; Thatcher et al. 1998; Hensley et al. 2001). In a study including two cohorts of patients with ovarian cancer receiving platinum-based chemotherapy, each 10-year increase in age and each 1-g/dl decrease in baseline-Hb level increased the risk of RBC transfusion requirement by 66% and 65%, respectively

218

M. R. Nowrousian

Table 7. Platinum-induced anemia A. Characteristics and development: – Normocytic, normochromic, hyporegenerative – Independent development from the type of malignancy – Transfusion requirement rate can amount to 47% to 100%, depending on the cumulative dose of platinum and patient-related risk factors – At a dose of 60–80 mg/m2 of cisplatin or 300 mg/m2 of carboplatin per cycle, transfusion requirement usually occurs after 3–4 cycles B. Predictive factors: – Low level of Hb at the start of treatment – Decrease in Hb level (>1–2 g/dl) after the ﬁrst cycle of chemotherapy – Cumulative dose of platinum – Advanced age – Loss of body weight before chemotherapy – Advanced stage of malignant disease – Lack of disease response to chemotherapy – Plasma level of ultraﬁltrable platinum >50 ng, measured 24 hrs after drug administration From ref. (57), (99), (203), (214), (225), (272), (294), (296).

(Heddens et al. 1998). In another study including patients with small-cell lung cancer and a primary cisplatin-based chemotherapy given for six cycles, patients with a baseline Hb level of 11 g/dl appeared to have a 79% risk of requiring transfusion compared to 37% in patients with a baseline Hb level of 13 g/dl and 8% in patients with a baseline Hb level of 15 g/dl. In the last two groups of patients, however, the risk of transfusion requirement increased considerably, if there was a decrease in Hb level of at least 1–2 g/dl from baseline to the start of the second cycle of chemotherapy (Table 8) (Abels et al. 1994; Thatcher 1998). A powerful predictor of anemia in patients receiving cisplatin seems to be a high plasma concentration of ultraﬁlterable (UF) platinum measured by atomic absorption spectrophotometry the day after drug administration. Ninety-one percent of patients with UF platinum concentrations >50 ng/ml will develop anemia compared with 18% in those with UF concentrations ≤ 50 ng/ml (Table 7) (Pivot et al. 2000). Platinuminduced anemia is usually normocytic, normochromic and hyporegenerative (Wood and Hrushesky 1995). A low primary level of Hb (<12 g/dl) is also predictive of anemia and the need for RBC transfusions in patients with non-platinum-based chemotherapy (Carabantes et al. 1999; Ray-Coquard et al. 1999; Skillings et al. 1999; Ludwig et al. 2004; cf. Ludwig in this book). In a study of 381 patients with various types of malignant disease, mainly solid tumors, 8% of patients with a baseline Hb level of more than 12 g/dl required RBC transfusion compared

Signiﬁcance of anemia in cancer chemotherapy

219

Table 8. Decrease in Hb level during chemotherapy* as a risk factor for RBC transfusion Baseline Hb (g/dl)

Decrease in Hb (g/dl) from baseline to start of cycle 2

Risk of transfusion (%)

11

0 1 2 0 1 2 0 1 2

79 93 96 37 66 86 8 22 49

13

15

* mainly platinum-based. RBC = red blood cell. From ref. (5).

Table 9. Relationship between baseline hemoglobin level and the need for RBC transfusions during chemotherapy Baseline hemoglobin level (g/dl)

<8 8– <10 10–12 >12

Transfusion for anemia No

Yes (%)

0 14 70 227

8 (100) 17 (55) 25 (26) 20 (8)

RBC = red blood cell. Adapted from ref. (252).

with 26%, 55%, and 100% of those with a baseline Hb level of 10–12, 8-<10 and <8 g/dl, respectively (Table 9) (Skillings et al. 1993) In another study including patients with various types of chemotherapy and evaluating the predictive value of a number of pretherapeutic parameters, a low primary level of Hb, a reduced performance status and a reduced lymphocyte count were found to be the only independent variables predicting severe anemia requiring RBC transfusions within the ﬁrst 31 days of chemotherapy (Ray-Coquard et al. 1999). To obtain a risk model for developing severe anemia, these three parameters were given arbitrary coefﬁcients of 3, 1, and 1, respectively, and a risk score for each individual patient was calculated by

220

M. R. Nowrousian

adding the coefﬁcients. The resultant index score was tested and validated in two cohorts of patients receiving chemotherapy, and it was found to be highly predictive of developing severe anemia and requiring RBC transfusions. Using this model, the probability of RBC transfusion requirement appears to increase from 1% to 30% with increasing score from 0 to 4 or above. The risk of anemia in patients receiving adjuvant chemotherapy for breast cancer has also been reported to increase with decreasing pretreatment hemoglobin level. Other factors independently and signiﬁcantly predicting the development of anemia in this group of patients have been found to be age ≥ 65 years, platelet count ≤ 200 × 109 cells/l before treatment and the type of chemotherapy. Using these factors, a risk score for anemia has been developed for patients who receive CEF, CAF (C = cyclophosphamide, E = epirubicin, A = doxorubicin, F = ﬂuorouracil), CMF (M = methotrexate), AC or other types of non-platinum-based combination chemotherapy (Dranitsaris et al. 2005). According to the results of these studies, a low primary level of Hb (<12 g/dl), both in platinum-based and non-platinum-based chemotherapy, appears to be predictive of development of anemia in a high proportion of patients and requirement of RBC transfusion in 25%–100%, depending on the reduced level of Hb at the beginning of treatment. A rapid decrease in Hb level (1–2 g/dl) after the ﬁrst cycle of treatment also predicts anemia and need for RBC transfusions in a high proportion of patients, even in those with a primary normal level of Hb. Another signiﬁcant predictive factor is an age >60 years. In elderly patients, the risk of developing anemia, particularly its severe grades, is more than twice as high as in younger patients receiving the same chemotherapeutic regimens. Further factors increasing the risk of anemia are advanced stage of disease, reduced general condition and a lack of disease response to treatment.

Prognostic signiﬁcance The presence of anemia at diagnosis or during treatment is a negative prognostic factor for disease progression and survival in a number of malignant diseases (Table 4), and in some of them, anemia is included into their staging systems or prognostic models. In multiple myeloma, anemia is already present at diagnosis in more than 60% of patients (Kyle 1975; Mittelman 2003). In this disease and in chronic lymphocytic leukemia (CLL), anemia predicts shorter survival and is an important parameter in disease staging according to classiﬁcations proposed by Durie and Salmon (1975) for multiple myeloma and Rai et al. (1975) or Binet et al. (1977) for CLL. In multiple myeloma, anemia is associated with a greater myeloma cell mass and a higher level of myeloma protein (Durie and Salmon 1975). It is also directly related to the percentage of myeloma

Signiﬁcance of anemia in cancer chemotherapy

221

Table 10. Relationship between baseline hemoglobin (Hb) level and duration of plateau phase in multiple myeloma Hb (g/dl)

<8.5 8.5–9.9 10–11.9 >11.9

No. of patients

52 73 156 147

Duration of plateau (mo.) 0–2

3–11

12–23

>23

37% 30% 17% 10%

25% 26% 21% 18%

19% 21% 31% 26%

19% 23% 31% 46%

Signiﬁcant difference between the proportions of patients with Hb levels <8.5 or >12 g/dl and a plateau phase of at least 12 months (p < .0001). Adapted from ref. (202).

cells in the S phase and, thus, the proliferative activity of myeloma cells (Fossa et al. 1999). In a study including 432 patients with multiple myeloma, a close relation was found between the duration of plateau phase after chemotherapy and the survival of patients, and the former was found to correlate with Hb level and inversely with the degree of anemia (Table 10) (Oivanen et al. 1996). In advanced Hodgkin’s disease, anemia is consistently found to be a negative prognostic factor for the outcome of treatment and, as such, it is incorporated in prognostic models for this disease (Straus et al. 1990; Ferme et al. 1997; Hasenclever and Diehl 1998; Gobbi et al. 2001). In a study including 185 patients, the probability of survival at 5 years was 66% in patients with initially reduced hematocrit level (<34% in women and <38% in men) compared to 91% in patients with normal hematocrit (Straus et al. 1990). A large international study including 5141 patients identiﬁed anemia (Hb level <10.5 g/dl) as an independent factor predicting a poor outcome of chemotherapy (Table 11). Anemic patients appeared to have a 35% higher risk of experiencing disease progression than nonanemic patients (Hasenclever and Diehl 1998). In non-Hodgkin lymphoma, anemia is not included into the International Prognostic Index (IPI), because it was not tested, when this index was developed (The International Non-Hodgkin’s Lymphoma Prognostic Factors Project 1993; Coifﬁer et al. 2000). In a number of studies, however, anemia was found to be an important prognostic factor both in indolent (Cabanillas et al. 1978; Gallagher et al. 1986; Leonard et al. 1991; Moullet et al. 1998) as well as aggressive lymphoma (Coifﬁer et al. 1989; Cowan et al. 1989; Conlan 1991; Moullet et al. 1998; Zinzani et al. 2005). A study analyzing the data of

222

M. R. Nowrousian Table 11. Risk factors for disease progression in patients with Hodgkin’s lymphoma Factor

Relative risk*

Serum albumin < 40 g/dl Hemoglobin < 10.5 g/dl Sex, male Stage IV disease Age ≥ 45 yrs WBC count ≥ 15,000/mm3 Lymphocyte count < 600/mm3

1.49 1.35 1.35 1.26 1.39 1.41 1.38

* compared with patients without the risk factor. From ref. (120).

Table 12. Incidence and prognostic value of anemia in patients with various types of NHL Lymphoma (No. of patients)

Incidence of anemia

OS (p-value)

DFS (p-value)

SL/LP (127) MC (50) MZ (62) F (208) T-cell (104) DLC (426) HG (73)

37% 32% 19% 17% 28% 39% 31%

0.0026 0.0033 ns 0.02 0.013 <0.0001 0.0031

0.047 0.018 ns ns ns <0.0001 <0.0001

NHL = non-Hodgkin’s lymphoma; OS = overall survival; DFS = disease-free survival; SL/LP = small lymphocytic or lymphoplasmocytoid; MC = mantle cell; MZ = marginal zone; F = follicular; DLC = diffuse large-cell; HG = high-grade (24 lymphoblastic and 49 Burkitt); ns = not signiﬁcant. From ref. (187).

1077 patients with various histologic subtypes of NHL found that anemia (Hb level <11 g/dl in women and <12 g/dl in men) was present in 17% to 39% of patients at diagnosis, depending on disease subtype and stage, and that it was signiﬁcantly associated with shorter progression-free survival (PFS) in small lymphocytic or lymphoplasmacytoid, mantle cell, diffuse large-cell and highgrade lymphoma and with shorter overall survival (OS) in all histologic subgroups except marginal-zone lymphoma (Table 12) (Moullet et al. 1998). Multivariate analysis identiﬁed anemia as a negative prognostic factor for

Signiﬁcance of anemia in cancer chemotherapy

223

Fig. 1. Overall survival and progression-free survival in patients with non-Hodgkin’s lymphoma according to the presence or absence of anemia. From ref. 187 (reproduced with permission)

PFS and OS for the population of patients as a whole (Fig. 1) and in patients with bone marrow involvement, but not in those without bone marrow inﬁltration. Furthermore, the addition of anemia to the IPI led to an improvement of this index for PFS and OS (Moullet et al. 1998). In a recently published study of elderly patients with aggressive non-Hodgkin’s lymphoma, post-treatment hemoglobin levels appeared to be a strong independent predictive factor for survival (Zinzani et al. 2005). Other types of malignant lymphomas with anemia as a negative prognostic factor are Waldenström macroglobulinemia (Morel et al. 2000), hairy-cell leukemia (Frassoldati et al. 1994) and chronic myelocytic leukemia (Jootar et al. 1990).

224

M. R. Nowrousian

Also in patients with solid tumors, anemia predicts a worse outcome of treatment (Table 4) (Berry et al. 1979; Emrich et al. 1985; Bush 1986; Gospodarowicz et al. 1989; Dische 1991; Citterio et al. 1997; Gelb 1997; Wigren et al. 1997; Grau and Overgaard 1998; Lee et al. 1998; Obermair et al. 2000, 2001, 2003; Yasunaga et al. 1998; Grogan et al. 1999; Bacci et al. 2000; Dietz et al. 2000; Jazieh et al. 2000; Kumar 2000; Wagner et al. 2000; Oettle et al. 2001; Tammemagi et al. 2003; Van Belle and Cocquyt 2003). Particularly, the outcome of radiotherapy appears to be negatively inﬂuenced by anemia (Hirst 1986; Thews et al. 1998, 2001; Grogan et al. 1999; Dunst 2000; Harrison et al. 2000; Smaniotto et al. 2000; Henke 2001; Shasha 2001; Schäfer et al. 2003; Haugen et al. 2004; Macdonald and Hurman 2004; Münstedt et al. 2004; Lohynska et al. 2005; Pradier et al. 2005; van de Pol et al. 2006; Dietl et al. 2007; see also Dunst et al. and Vaupel et al. in this book). However, there are experimental and clinical studies indicating that anemia is also associated with a poor outcome of chemoradiotherapy (Wagner et al. 2000; Glaser et al. 2001; Obermair et al. 2001, 2003; Shasha 2001; Chua et al. 2004; Denis et al. 2004; Prosnitz et al. 2005; Choi et al. 2006; Laurie et al. 2006; Weissenberger et al. 2006) or chemotherapy alone (Silver et al. 1999; Bacci et al. 2000; Obermair et al. 2000; Sengelov et al 2000; Littlewood et al. 2001a; Gadducci et al. 2003, 2005; Münstedt et al. 2003; Van Belle and Cocquyt 2003; Shannon et al. 2005; Di Maio et al. 2006; Park et al. 2006). In a study of patients with ovarian cancer receiving chemotherapy or radiotherapy after debulking surgery, anemia (Hb < 12 g/dl) was found to be present in 26% of patients at diagnosis, and the probability of 5-year survival was signiﬁcantly lower in this group of patients than in those without anemia (39% versus 52%) (Obermair et al. 2000). In a study of patients with nonmetastatic Ewing’s sarcoma, anemia appeared to be an adverse independent prognostic factor for event-free survival (EFS) after chemotherapy followed by surgery or radiotherapy or both (Bacci et al. 2000). Patients with anemia had a probability of long-term EFS of 38% compared with 58% in patients without anemia. Of particular interest are studies, in which anemia not at baseline but during treatment or at its end appeared to be associated with a worse prognosis. The results of these studies indicate that treatment-induced or -aggravated anemia independent from the primary hemoglobin level inﬂuences the outcome. On the other hand, they also show that anemia at the start of treatment, even if considered as a surrogate for a greater aggressiveness of underlying malignancy or as a sign of a more intensive immunoglogical and inﬂammatory reaction of the host body, has the potential to be independently a factor which negatively inﬂuences the results. Associations between decreased hemoglobin levels during or at the end of treatment and worse prognosis have been reported for radiotherapy in endometrial and cervix carcinoma (Grogan et al. 1999; Santin et al. 2003; Münstedt et al. 2004), for chemoradiotherapy in nasopharyngeal cancer, cervical carcinoma, squamous

Signiﬁcance of anemia in cancer chemotherapy

225

cell carcinoma of esophagus, and small-cell lung cancer (Wagner et al. 2000; Obermair et al. 2001, 2003; Chua et al. 2004; Hofheinz et al. 2004; Laurie et al. 2006), and for chemotherapy in aggressive non-Hodgkin’s lymphoma, breast, ovarian and testicular cancer (Bokemeyer et al. 2002; Münstedt et al. 2003; Peters-Engl et al. 2005; Zinzani et al. 2005; Boehm et al. 2007) (Figs. 2, 3). A comprehensive review of 60 studies reporting on survival of cancer patients according to the presence or absence of anemia found that anemic

Fig. 2. Correlation of hemoglobin level at the end of chemoradiotherapy and recurrence-free survival of patients with advanced head and neck cancer. From ref. 289 (reproduced with permission)

Fig. 3. Survival according to Hb level (threshold 10 g/dl) during adjuvant chemotherapy in patients with primary breast cancer. log rank test. From ref. 208 (reproduced with permission)

226

M. R. Nowrousian Table 13. Increased mortality related to anemia in various types of cancer Cancer

Increase in mortality*

95%CI

Lung Prostate Head and neck Lymphoma Overall

19% 47% 75% 67% 65%

10–29% 21–78% 37–123% 30–113% 54–77%

* based on cumulative survival data from 60 papers stratiﬁed by hemoglobin levels or anemia. CI = conﬁdence interval. Adapted from ref. (34).

patients had an overall increased risk of death of 65% compared with nonanemic patients (Caro et al. 2001). Anemia was found to be associated with increased risk of death in patients with lung cancer, cervico-uterine cancer, head and neck cancer, prostate cancer, malignant lymphoma, and multiple myeloma. The increased risk was highest in patients with head and neck cancer or malignant lymphoma (Table 13). The presence of anemia and its association with inferior results of radiotherapy and chemotherapy may be linked to a more aggressive underlying malignancy, but, as mentioned above, there are experimental and clinical studies indicating that anemia as such may have an independent impact on the outcome of treatment. Such an impact may result from alterations of various organ functions (see also Nowrousian in this book), patient performance and the sensitivity of tumor cells to radiotherapy and chemotherapy. Anemic patients require signiﬁcantly more frequently inpatient hospital admission, emergency department visits and outpatient services during chemotherapy than nonanemic patients, indicating that anemia may be a factor which negatively inﬂuences the capability of patients to tolerate chemotherapy and thus reduces the effectiveness of treatment (Lyman et al. 2005). Anemia can also aggravate tumor hypoxia (Becker et al. 2000; Höckel and Vaupel 2001 and in this book), and experimental studies show that the latter can induce the selection of more resistant tumor cells, increase the metastatic potential of these cells (Young and Hill 1990), and reduce their sensitivity to irradiation and chemotherapeutic agents (Table 14) (Teicher et al. 1990, 1994, 1995; Tomida and Tsuruo 1999; Höckel and Vaupel 2001; Stüben et al. 2001, Wouters et al. 2007). Following mechanisms may be involved in this process: 1) tissue acidosis, 2) generation of stress proteins, 3) decrease in cytotoxicity, 4) inhibition of cell proliferation; and loss of apoptotic potential of tumor cells (Höckel and Vaupel 2001; see also Vaupel in this book).

Signiﬁcance of anemia in cancer chemotherapy

227

Table 14. Survival of experimental ﬁbrosarcoma cells (FSIIa) carried in mice after treatment with various cytotoxic agents related to oxygenation Cytotoxic drug

Dose (mg/kg)

Oxygenated cells

Hypoxic cells

Ratio

Cyclophosphamide BCNU Carboplatin X-ray Melphalan Cisplatin Thiotepa Procarbazine Antibiotics Actinomycin D Doxorubicin Vincristine Etoposide Bleomycin 5-Fluorouracil Mitomycin C

150 50 50 (10 Gy) 10 10 10 20

0.035 0.050 0.072 0.11 0.025 0.15 0.13 0.75

0.22 0.16 0.17 0.25 0.055 0.27 0.22 0.93

6.3 3.2 2.4 2.2 2.2 1.8 1.7 1.2

0.29 0.33 0.47 0.38 0.44 0.43 0.37

1.00 0.74 1.00 0.62 0.66 0.97 0.11

3.4 2.2 2.1 1.6 1.5 2.3 0.29

1 25 2 20 10 40 5

Mice were treated with single doses of each drug. Data are means of surviving fractions from three independent experiments. Except for mitomycin C, all cytotoxic agents tested were more effective against oxygenated cells than hypoxic cells. Adapted from ref. (269).

Anemia as a result of a reduced red blood cell mass may also have a negative impact on pharmacokinetics of chemotherapeutic agents. RBCs have been reported to play an important role in storage, transport and metabolism of particular cytotoxic drugs. Anthracyclines, ifosfamide and its metabolites, and topoisomerase I/II inhibitors are incorporated in erythrocytes and may be transported by these cells to the tumor tissue and mobilised by active or passive mechanisms (Highley et al. 1997; Ramanathan-Girish and Boroujerdi 2001; Schrijvers 2003). 6-mercaptopurine, methotrexate and aminotrexate are reported to accumulate in erythrocytes and the levels of the ﬁrst two metabolites in these cells have been found to reﬂect the intensity of treatment in children with acute lymphocytic leukemia (Cole et al. 2006; Halonen et al. 2006). As shown for oxaliplatin, platinum-derived cytotoxic agents are also bound to erythrocytes and transported by these cells (Luo et al. 1999). In an animal model of mice, a signiﬁcant correlation was found between concentrations of melphalan in erythrocytes and the tumor availability of this drug (Wildiers et al. 2002). Because of their potential ability to uptake,

228

M. R. Nowrousian

transport and deliver various types of bioactive agents including antineoplastic drugs, erythrocytes have increasingly become interesting objects to be evaluated as biological carriers in clinical oncology. There is evidence suggesting that encapsulation in erythrocytes may help to increase the efﬁcacy of chemotherapeutic agents and to reduce their toxicity (Millan et al. 2004; Skorokhod et al. 2004, 2007).

Conclusion Patients with cancer frequently develop anemia, due either to the malignant disease itself or to its treatment. Various factors, including the type and stage of malignancy and the type, schedule and intensity of chemotherapy, inﬂuence the prevalence and incidence of anemia. The frequency of transfusion requirement depends on the degree of anemia, and the age and general condition of patients, particularly their cardiopulmonary function. Some cytotoxic agents such as cisplatin, carboplatin, paclitaxel, docetaxel, vinorelbine, irinotecan, topotecan, etoposide, and ifosfamide, either as single agent or in combination, are frequently associated with anemia. The incidence of anemia and the need for RBC transfusions, however, are particularly high in patients receiving platinum or anthracycline-based regimens additionally containing agents such as etoposide, ifosfamide, cyclophosphamide, gemcitabine, vinorelbine, paclitaxel or irinotecan. Such regimens may be associated with anemia and the need for RBC transfusions in up to 97% and 79% of patients, respectively. In patients receiving platinum-based chemotherapy, the frequency of transfusion requirement can amount to 47% to 100%, depending on the cumulative dose of platinum and other risk factors, such as advanced age, loss of body weight before treatment, advanced stage of disease and, particularly, a low primary level of Hb (12 g/dl) or a rapid decrease in Hb level (1–2 g/dl) after the ﬁrst or second cycle of treatment, irrespective of the primary level of Hb. The causative mechanism of platinum-induced anemia is reported to be, beside myelosuppression, a defective EPO production resulting from drug-induced renal tubular damage. A low primary level of Hb (<12 g/dl) is also predictive of anemia and necessity of RBC transfusions in patients receiving non-platinum-based chemotherapy. Twenty-ﬁve to 100% of these patients may require RBC transfusions during chemotherapy, depending on the degree of Hb reduction at the start of treatment. Based on these data, patients with platinum or anthracycline-based chemotherapy, particularly elderly patients and those with a primarily low level of Hb or a rapid decrease in Hb level after the ﬁrst or second cycle of treatment, appear to have a high probability of developing anemia and requiring RBC transfusions during chemotherapy.

Signiﬁcance of anemia in cancer chemotherapy

229

Anemia has been shown to have major impacts on metabolic functions, exercise capacity and quality of life of cancer patients. In addition, anemia has been reported to be a negative prognostic factor for the outcome of chemotherapy with or without radiotherapy in a number of malignancies, such as multiple myeloma, Hodgkin’s disease, non-Hodgkin lymphoma, Ewing’s sarcoma, head and neck cancer, small-cell and non-small-cell lung cancer, esophagal cancer, cervical and ovarian cancer. There is also evidence suggesting that anemia induced or aggravated by treatment as such and not dependent on other factors signiﬁcantly determines the outcome. This may result from a reduced capability of anemic patients to tolerate treatment and a reduced sensitivity of tumor cells for chemotherapeutic agents associated with anemia. Considering all these aspects of anemia in cancer patients, it appears to be of particular clinical interest to treat anemia during chemotherapy.

References 1. Corfu-A Study Group (1995) Phase III randomized study of two ﬂuorouracil combinations with either interferon alfa-2a or leucovorin for advanced colorectal cancer. J Clin Oncol 13: 921–928 2. Abels R, Gordon D, Nelson R, et al (1991) Transfusion practice in advanced cancer patients. Blood 78 [Suppl 1]: 474a (Abstr) 3. Abels RI (1992) Use of recombinant human erythropoietin in the treatment of anemia in patients who have cancer. Semin Oncol 19 [Suppl 8]: 29–35 4. Abels R (1993) Erythropoietin for anaemia in cancer patients. Eur J Cancer 29A [Suppl 2]: S2–S8 5. Abels R, Larholt K, Nelson R, et al (1994) Risk of transfusion in small cell lung cancer patients receiving chemotherapy. Blood 84 [Suppl 1]: 664a (Abstr) 6. Abratt RP, Hacking DJ, Goedhals L, et al (1997) Weekly gemcitabine and monthly cisplatin for advanced non-small cell lung carcinoma. Semin Oncol 24 [Suppl 8]: S18–S23 7. Alberts DS, Green S, Hannigan EV, et al (1992) Improved therapeutic index of carboplatin plus cyclophosphamide versus cisplatin plus cyclophosphamide: ﬁnal report by the Southwest Oncology Group of a phase III randomized trial in stages III and IV ovarian cancer. J Clin Oncol 10: 706–717 8. Alberts DS, Liu PY, Hannigan EV, et al (1996) Intraperitoneal cisplatin plus intravenous cyclophosphamide versus intravenous cisplatin plus intravenous cyclophosphamide for stage III ovarian cancer. N Engl J Med 335: 1950–1955 9. Anderson H, Lund B, Bach F, et al (1994) Single-agent activity of weekly gemcitabine in advanced non-small-cell lung cancer: a phase II study. J Clin Oncol 12: 1821–1826 10. Ardizzoni A, Hansen H, Dombernowsky P, et al (1997) Topotecan, a new active drug in the second-line treatment of small-cell lung cancer: a phase II study in patients with refractory and sensitive disease. The European Organization for

230

11.

12.

13.

14. 15.

16.

17.

18.

19. 20. 21.

22. 23.

24. 25.

26.

M. R. Nowrousian Research and Treatment of Cancer Early Clinical Studies Group and New Drug Development Ofﬁce, and the Lung Cancer Cooperative Group. J Clin Oncol 15: 2090–2096 Bacci G, Ferrari S, Bertoni F, et al (2000) Prognostic factors in nonmetastatic Ewing’s sarcoma of bone treated with adjuvant chemotherapy: analysis of 359 patients at the Istituto Ortopedico Rizzoli. J Clin Oncol 18: 4–11 Bagnis C, Beauﬁls H, Jacquiaud C, et al (2001) Erythropoietin enhances recovery after cisplatin-induced acute renal failure in the rat. Nephrol Dial Transplant 16: 932–938 Bajorin DF, McCaffrey JA, Hilton S, et al (1998) Treatment of patients with transitional-cell carcinoma of the urothelial tract with ifosfamide, paclitaxel, and cisplatin: a phase II trial. J Clin Oncol 16: 2722–2727 Baldwin MD, Zhou XJ, Todd S, et al (1998) Erythropoietin ameliorates anemia of cisplatin induced acute renal failure. ASAIO J 44: 44–47 Ballova V, Rüffer JU, Haverkamp H, et al (2005) A prospectively randomized trial carried out by the German Hodgkin Study Group (GHSG) for elderly patients with advanced Hodgkin’s disease comparing BEACOPP baseline and COPP-ABVD (study HD9elderly). Ann Oncol 16: 124–131 Barrett-Lee PJ, Bailey NP, O’Brien ME, et al (2000) Large-scale UK audit of blood transfusion requirements and anaemia in patients receiving cytotoxic chemotherapy. Br J Cancer 82: 93–97 Barrett-Lee PJ, Ludwig H, Birgegard G, et al (2006) Independent risk factors for anemia in cancer patients receiving chemotherapy: results from the European Cancer Anaemia Survey. Oncology 70: 34–48 Becker A, Stadler P, Lavey RS, et al (2000) Severe anemia is associated with poor tumor oxygenation in head and neck squamous cell carcinomas. Int J Radiat Oncol Biol Phys 46: 459–466 Berry WR, Laszlo J, Cox E, et al (1979) Prognostic factors in metastatic and hormonally unresponsive carcinoma of the prostate. Cancer 44: 763–775 Binet JL, Lepoprier M, Dighiero G, et al (1977) A clinical staging system for chronic lymphocytic leukemia: prognostic signiﬁcance. Cancer 40: 855–864 Birgegard G, Gascón P, Ludwig H (2006) Evaluation of anaemia in patients with multiple myeloma and lymphoma: ﬁndings of the European Cancer Anaemia Survey (ECAS). Eur J Haematol 77: 378–386 Boehm DU, Lebrecht A, Schmidt M, et al (2007) Prognostic impact of haemoglobin levels in breast cancer. Anticancer Res 27: 1223–1226 Bokemeyer C, Oechsle K, Hartmann JT, et al (2002) Treatment-induced anaemia and its potential clinical impact in patients receiving sequential high dose chemotherapy for metastatic testicular cancer. Br J Cancer 87: 1066–1071 Blohmer JU, Paepke S, Böhmer D, et al (2001) [Adjuvant chemotherapy of cervix carcinoma–results of a phase II study]. Zentralbl Gynäkol 123: 286–291 Buda A, Fossati R, Colombo N, et al (2005) Randomized trial of neoadjuvant chemotherapy comparing paclitaxel, ifosfamide, and cisplatin with ifosfamide and cisplatin followed by radical surgery in patients with locally advanced squamous cell cervical carcinoma: the SNAP01 (Studio Neo-Adjuvante Portio) Italian Collaborative Study. J Clin Oncol 23: 4137–4145 Budd GT, Green S, O’Brayan RM, et al (1995) Short-course FAC-M versus 1 year of CMFVP in node-positive, hormone receptor-negative breast cancer: an Intergroup study. J Clin Oncol 13: 831–839

Signiﬁcance of anemia in cancer chemotherapy

231

27. Bush RS (1986) The signiﬁcance of anemia in clinical radiation therapy. Int J Radiat Oncol Biol Phys 12: 2047–2050 28. Cabanillas F, Burke JS, Smith TL, et al (1978) Factors predicting for response and survival in adults with advanced non-Hodgkin’s lymphoma. Arch Intern Med 138: 413–418 29. Canaparo R, Casale F, Muntoni E, et al (2000) Plasma erythropoietin concentrations in patients receiving intensive platinum or nonplatinum chemotherapy. Br J Clin Pharmacol 50: 146–153 30. Canellos GP, Anderson JR, Propert KJ, et al (1992) Chemotherapy of advanced Hodgkin’s disease with MOPP, ABVD, or MOPP alternating with ABVD. N Engl J Med 327: 1478–1484 31. Canellos GP, Petroni GR, Barcos M, et al (1995) Etoposide, vinblastine, and doxorubicin: an active regimen for the treatment of Hodgkin’s disease in relapse following MOPP. Cancer and Leukemia Group B. J Clin Oncol 13: 2005– 2011 32. Canetta R, Rozencweig M, Carter SK (1985) Carboplatin: the clinical spectrum to date. Cancer Treat Rev 12 [Suppl A]: 125–136 33. Carabantes FJ, Benavides M, Trujillo R, et al (1999) Epoetin alpha in the prevention of anemia in cancer patients undergoing platinum-based chemotherapy. A prospective randomised study. Proc Am Soc Clin Oncol 18: 596a (Abstr 2303) 34. Caro JJ, Salas M, Ward A, et al (2001) Anemia as an independent prognostic factor for survival in patients with cancer: a systemic, quantitative review. Cancer 91: 2214–2221 35. Catimel G, Verweij J, Mattijssen V, et al (1994) Docetaxel (Taxotere): an active drug for the treatment of patients with advanced squamous cell carcinoma of the head and neck. EORTC Early Clinical Trials Group. Ann Oncol 5: 533–537 36. Cella D (1997) The functional assessment of cancer therapy-anemia (FACT-An) scale: a new tool for the assessment of outcomes in cancer anemia and fatigue. Semin Hematol 34 [Suppl 2]: 13–19 37. Cella D (2002) The effects of anemia and anemia treatment on the quality of life of people with cancer. Oncology (Huntingt) 16 [Suppl 10]: 125–132 38. Chen S, Evans WK, Stys-Norman, et al (2007) Chemotherapy for relapsed small cell lung cancer: A systematic review and practice guideline. J Thorac Oncol 2: 348–354 39. Chevallier B, Fumoleau P, Kerbrat P, et al (1995) Docetaxel is a major cytotoxic drug for the treatment of advanced breast cancer: a phase II trial of the Clinical Screening Cooperative Group of the European Organization for Research and Treatment of Cancer. J Clin Oncol 13: 314–322 40. Choi YS, Yi CM, Sin J-I, et al (2006) Impact of hemoglobin on survival of cervical carcinoma patients treated with concurrent chemoradiotherapy is dependent on lymph node metastasis ﬁndings by magnetic resonance imaging. Int J Gynecol Cancer 16: 1846–1854 41. Chua DTT, Sham JST, Choy DTK (2004) Prognostic impact of hemoglobin levels on treatment outcome in patients with nasopharyngeal carcinoma treated with sequential chemoradiotherapy or radiotherapy alone. Cancer 101: 307–316 42. Citterio G, Bertuzzi A, Tresoldi M, et al (1997) Prognostic factors for survival in metastatic renal cell carcinoma: retrospective analysis from 109 consecutive patients. Eur Urol 31: 286–291

232

M. R. Nowrousian

43. Coifﬁer B, Gisselbrecht C, Herbrecht R, et al (1989) LNH-84 regimen: a multicenter study of intensive chemotherapy in 737 patients with aggressive malignant lymphoma. J Clin Oncol 7: 1018–1026 44. Coifﬁer B (2000) The impact and management of anaemia in haematological malignancies. Med Oncol 17 [Suppl 1]: S2–S10 45. Coifﬁer B, Guastalla JP, Pujade-Lauraine E, et al (2001) Predicting cancerassociated anaemia in patients receiving non-platinum chemotherapy: results of a retrospective survey. Eur J Cancer 37: 1617–1623 46. Cole PD, Alcaraz MJ, Smith AK (2006) Pharmacodynamic properties of methotrexate and AminotrexateTM during weekly therapy. Cancer Chemother Pharmacol 57: 826–834 47. Conlan MG, Armitage JO, Bast M, et al (1991) Clinical signiﬁcance of hematologic parameters in non-Hodgkin’s lymphoma at diagnosis. Cancer 67: 1389–1395 48. Conte PF, Bruzzone M, Chiara S, et al (1986) A randomized trial comparing cisplatin plus cyclophosphamide versus cisplatin, doxorubicin, and cyclophosphamide in advanced ovarian cancer. J Clin Oncol 4: 965–971 49. Cotterill SJ, Ahrens S, Paulussen M, et al (2000) Prognostic factors in Ewing’s tumor of bone: analysis of 975 patients from the European Intergroup Cooperative Ewing’s Sarcoma Study Group. J Clin Oncol 18: 3108–3114 50. Cowan RA, Jones M, Harris M, et al (1989) Prognostic factors in high and intermediate grade non-Hodgkin’s lymphoma. Br J Cancer 59: 276–282 51. Crawford J, Cella D, Cleeland C, et al (2002) Relationship between changes in hemoglobin level and quality of life during chemotherapy in anemic cancer patients receiving epoetin alfa therapy. Cancer 95: 888–895 52. Creemers GJ, Bolis G, Gore M, et al (1996a) Topotecan, an active drug in the second-line treatment of epithelial ovarian cancer: results of a large European phase II study. J Clin Oncol 14: 3056–3061 53. Creemers GJ, Gerrits CJ, Schellens JH, et al (1996b) Phase II and pharmacologic study of topotecan administered as a 21-day continuous infusion to patients with colorectal cancer. J Clin Oncol 14: 2540–2545 54. Daneryd P, Svanberg E, Körner U, et al (1998) Protection of metabolic and exercise capacity in unselected weight-losing cancer patients following treatment with recombinant erythropoietin: a randomized prospective study. Cancer Res 58: 5374–5379 55. Daniel D, Crawford J (2006) Myelotoxicity from chemotherapy. Semin Oncol 33: 74–85 56. Davidson NG (1996) Single-agent paclitaxel as ﬁrst-line treatment of metastatic breast cancer: the British experience. Semin Oncol 23 [Suppl 11]: 6–10 57. de Campos E, Radford J, Steward W, et al (1995) Clinical and in vitro effects of recombinant human erythropoietin in patients receiving intensive chemotherapy for small-cell lung cancer. J Clin Oncol 13: 1623–1631 58. Degardin M, Bonneterre J, Hecquet B, et al (1994) Vinorelbine (Navelbine) as a salvage treatment for advanced breast cancer. Ann Oncol 5: 423–426 59. Del Mastro L, Venturini M, Lionetto R, et al (1997) Randomized phase III trial evaluating the role of erythropoietin in the prevention of chemotherapyinduced anemia. J Clin Oncol 15: 2715–2721 60. Demetri GD, Kris M, Wade J, et al (1998) Quality-of-life beneﬁt in chemotherapy patients treated with epoetin alfa is independent of disease response or

Signiﬁcance of anemia in cancer chemotherapy

61.

62.

63.

64.

65. 66.

67. 68. 69.

70. 71.

72.

73.

74.

75.

76.

233

tumor type: results from a prospective community oncology study. Procrit Study Group. J Clin Oncol 16: 3412–3425 Denis F, Garaud P, Bardet E, et al (2004) Final results of the 94–01 French Head and Neck Oncology and Radiotherapy Group randomized trial comparing radiotherapy alone with concomitant radiochemotherapy in advanced-stage oropharynx carcinoma. J Clin Oncol 22: 69–76 Diehl V, Franklin J, Pfreundschuh M, et al (2003) Standard and increaseddose BEACOPP chemotherapy compared with COPP-ABVD for advanced Hodgkin’s disease. N Engl J Med 348: 2386–2395 Dieras V, Marty M, Tubiana N, et al (1995) Phase II randomized study of paclitaxel versus mitomycin in advanced breast cancer. Semin Oncol 22 [Suppl 8]: 33–39 Dietl B, Marienhagen J, Schafer C, et al (2007) The prognostic value of anaemia at different treatment times in patients with locally advanced head and neck cancer treated with surgery and postoperative radiotherapy. Clin Oncol (R Coll Radiol) 19: 228–233 Dietz A, Rudat V, Conradt C, et al (2000) [Prognostic value of hemoglobin level for primary radiochemotherapy of head-neck carcinomas.] HNO 48: 655–664 Di Maio M, Pisano C, Tambaro R, et al (2006) The prognostic role of prechemotherapy hemoglobin level in patients with ovarian cancer. Front Biosci 11: 1585–1590 Dische S (1991) Radiotherapy and anaemia–the clinical experience. Radiother Oncol 20 [Suppl 1]: 35–40 Doll DC, Weiss RB (1985) Hemolytic anemia associated with antineoplastic agents. Cancer Treat Rep 69: 777–782 Dranitsaris G, Clemons M, Verma S, et al (2005) Chemotherapy-induced anaemia during adjuvant treatment for breast cancer: development of a prediction model. Lancet Oncol 6: 856–863 Dorval T, Soussain C, Beuzeboc P, et al (1996) Ifosfamide seven-day infusion for recurrent and cisplatin refractory ovarian cancer. J Infus Chemother 6: 47–49 Dunst J (2000) Hemoglobin level and anemia in radiation oncology: prognostic impact and therapeutic implications. Semin Oncol 27 [Suppl 4]: 4–8; discussion 16–17 Durie BG, Salmon SE (1975) A clinical staging system for multiple myeloma. Correlation of measured myeloma cell mass with presenting clinical features, response to treatment, and survival. Cancer 36: 842–854 Dwilewicz-Trojaczek J, Sek S (1996) Prognostic factors in patients with lymphoplasmacytic lymphoplasmacytoid lymphoma in stage IV of clinical progession. Pol Tyg Lek 51: 266–268 Eckardt JR, von Pawel J, Pujol J-L, et al (2007) Phase III study of oral compared with intravenous topotecan as second-line therapy in small-cell lung cancer. J Clin Oncol 25: 2086–2092 Einzig AI, Wiernik PH, Sasloff J, et al (1992) Phase II study and long-term follow-up of patients treated with taxol for advanced ovarian adenocarcinoma. J Clin Oncol 10: 1748–1753 Eisenhauer EA, ten Bokkel Huinink WW, Swenerton KD, et al (1994) EuropeanCanadian randomized trial of paclitaxel in relapsed ovarian cancer: high-dose versus low-dose and long versus short infusion. J Clin Oncol 12: 2654–2666

234

M. R. Nowrousian

77. Ellis PA, Smith IE, Hardy JR, et al (1995) Symptom relief with MVP (mitomycin C, vinblastine and cisplatin) chemotherapy in advanced non-small-cell lung cancer. Br J Cancer 71: 366–370 78. Emrich LJ, Priore RL, Murphy GP, et al (1985) Prognostic factors in patients with advanced stage prostate cancer. Cancer Res 45: 5173–5179 79. Estrin JT, Schocket L, Kregenow R, et al (1999) A retrospective review of blood transfusions in cancer patients with anemia. Oncologist 4: 318– 324 80. Ettinger DS, Finkelstein DM, Sarma RP, et al (1995) Phase II study of paclitaxel in patients with extensive-disease small-cell lung cancer: an Eastern Cooperative Oncology Group study. J Clin Oncol 13: 1430–1435 81. Faylona EA, Loehrer PJ, Ansari R, et al (1995) Phase II study of daily oral etoposide plus ifosfamide plus cisplatin for previously treated recurrent small-cell lung cancer: a Hoosier Oncology Group trial. J Clin Oncol 13: 1209–1214 82. Ferme C, Bastion Y, Brice P, et al (1997) Prognosis of patients with advanced Hodgkin’s disease: evaluation of four prognostic models using 344 patients included in the Groupe d’Etudes des Lymphomes de l’Adulte study. Cancer 80: 1124–1233 83. Figueredo AT, Hryniuk WM, Strautmanis I, et al (1985) Co-trimoxazole prophylaxis during high-dose chemotherapy of small-cell lung cancer. J Clin Oncol 3: 54–64 84. Forastiere AA, Metch B, Schuller DE, et al (1992) Randomized comparison of cisplatin plus ﬂuorouracil and carboplatin plus ﬂuorouracil versus methotrexate in advanced squamous-cell carcinoma of the head and neck: a Southwest Oncology Group study. J Clin Oncol 10: 1245–1251 85. Forastiere AA, Neuberg D, Taylor SG, et al (1993) Phase II evaluation of taxol in advanced head and neck cancer: an Eastern Cooperative Oncology group trial. J Natl Cancer Inst Monogr 15: 181–184 86. Fossa A, Brandhorst D, Myklebust JH, et al (1999) Relation between S-phase fraction of myeloma cells and anemia in patients with multiple myeloma. Exp Hematol 27: 1621–1626 87. Fossella FV, Lee JS, Murphy WK, et al (1994) Phase II study of docetaxel for recurrent or metastatic non-small-cell lung cancer. J Clin Oncol 12: 1238–1244 88. Fossella F, Pereira JR, von Pawel J, et al (2003) Randomized, multinational, phase III study of docetaxel plus platinum combinations versus vinorelbine plus cisplatin for advanced non–small-cell lung cancer: The TAX326 Study Group. J Clin Oncol 21: 3016–3024 89. Francis P, Schneider J, Hann L, et al (1994a) Phase II trial of docetaxel in patients with platinum-refractory advanced ovarian cancer. J Clin Oncol 12: 2301–2308 90. Francis PA, Rigas JR, Kris MG, et al (1994b) Phase II trial of docetaxel in patients with stage III and IV non-small-cell lung cancer. J Clin Oncol 12: 1232–1237 91. Frassoldati A, Lamparelli T, Federico M, et al (1994) Hairy cell leukemia: a clinical review based on 725 cases of the Italian Cooperative Group for Hairy Cell leukemia (ICGHCL). Leuk Lymphoma 13: 307–316

Signiﬁcance of anemia in cancer chemotherapy

235

92. Fumoleau P, Delgado FM, Delozier T, et al (1993) Phase II trial of weekly intravenous vinorelbine in ﬁrst-line advanced breast cancer chemotherapy. J Clin Oncol 11: 1245–1252 93. Furuse K, Kubota K, Kawahara M, et al (1994) A phase II study of vinorelbine, a new derivative of vinca alkaloid, for previously untreated advanced non-small cell lung cancer. Japan Vinorelbine Lung Cancer Study Group. Lung Cancer 11: 385–391 94. Futatsuki K, Wakui A, Nakao I, et al (1994) [Late phase II study of irinotecan hydrochloride (CPT-11) in advanced gastric cancer. CPT-11 Gastrointestinal Cancer Study Group]. Gan To Kagaku Ryoho 21: 1033–1038 95. Gabrilove JL, Cleeland CS, Livingston RB, et al (2001) Clinical evaluation of once-weekly dosing of epoetin alfa in chemotherapy patients: Improvements in hemoglobin and quality of life are similar to three-times-weekly dosing. J Clin Oncol 19: 2875–2882 96. Gadducci A, Cosio S, Fanucchi A, et al (2003) Is pretreatment hemoglobin level a predictor of complete response to salvage chemotherapy for recurrent platinum-pretreated ovarian carcinoma? Eur J Gynaecol Oncol 24: 405–410 97. Gadducci A, Sartori E, Landoni F, et al (2005) Pre-chemotherapy hemoglobin levels and survival in patients with advanced epithelial ovarian cancer who received a ﬁrst-line taxane/platinum-based regimen: Results of a multicenter retrospective Italian study. Gynecol Oncol 98: 118–123 98. Gallagher GJ, Gregory WM, Jones AE, et al (1986) Follicular lymphoma: Prognostic factors for response and survival. J Clin Oncol 4: 1470–1480 99. Gamucci T, Thorel MF, Frasca AM, et al (1993) Erythropoietin for the prevention of anaemia in neoplastic patients treated with cisplatin. Eur J Cancer 29A [Suppl 2]: 13–14 100. Gasparini G, Caffo O, Barni S, et al (1994) Vinorelbine is an active antiproliferative agent in pretreated advanced breast cancer patients: a phase II study. J Clin Oncol 12: 2094–2101 101. Gatzemeier U, Shepherd FA, Le Chevalier T, et al (1996) Activity of gemcitabine in patients with non-small cell lung cancer: a multicentre, extended phase II study. Eur J Cancer 32A: 243–248 102. Gelb AB (1997) Renal cell carcinoma: current prognostic factors. Union Internationale Contre le Cancer (UICC) and the American Joint Committee on Cancer (AJCC). Cancer 80: 981–986 103. Gianni L, Munzone E, Capri G, et al (1995) Paclitaxel by 3-hour infusion in combination with bolus doxorubicin in women with untreated metastatic breast cancer: high antitumor efﬁcacy and cardiac effects in a dose-ﬁnding and sequence-ﬁnding study. J Clin Oncol 13: 2688–2699 104. Glaser CM, Millesi W, Kornek GV, et al (2001) Impact of hemoglobin level and use of recombinant erythropoietin on efﬁcacy of preoperative chemoradiation therapy for squamous cell carcinoma of the oral cavity and oropharynx. Int J Radiat Oncol Biol Phys 50: 705–715 105. Glaspy J, Bukowski R, Steinberg D, et al (1997) Impact of therapy with epoetin alfa on clinical outcomes in patients with nonmyeloid malignancies during cancer chemotherapy in community oncology practice. J Clin Oncol 15: 1218–1234 106. Gobbi PG, Zinzani PL, Broglia C, et al (2001) Comparison of prognostic models in patients with advanced Hodgkin disease. Cancer 91: 1467–1478

236

M. R. Nowrousian

107. Gobbi PG, Levis A, Chisesi T, et al (2005) ABVD versus modiﬁed Stanford V versus MOPPEBVCAD with optional and limited radiotherapy in intermediateand advanced-stage Hodgkin’s lymphoma: Final results of a multicenter randomized trial by the Intergruppo Italiano Linfomi. J Clin Oncol 23: 9198–9207 108. Gonzalez Baron M, Feliu J, Garcia Giron C, et al (1997) UFT modulated with leucovorin in advanced colorectal cancer: Oncopaz experience. Oncology 54: 24–29 109. Gordon LI, Harrington D, Andersen J, et al (1992) Comparison of a secondgeneration combination chemotherapeutic regimen (m-BACOD) with a standard regimen (CHOP) for advanced diffuse non-Hodgkin’s lymphoma. N Engl J Med 327: 1342–1349 110. Gospodarowicz MK, Hawkins NV, Rawlings GA, et al (1989) Radical radiotherapy for muscle invasive transitional cell carcinoma of the bladder: failure analysis. J Urol 142: 1448–1453 111. Grau C, Overgaard J (1998) Signiﬁcance of hemoglobin concentration for treatment outcome. In: Molls M, Vaupel P (eds) Blood perfusion and micro-environment of human tumors. Springer, Berlin Heidelberg New York, pp 101–102 112. Greco FA, Figlin R, York M, et al (1996) Phase III randomized study to compare interferon alfa-2a in combination with ﬂuorouracil versus ﬂuorouracil alone in patients with advanced colorectal cancer. J Clin Oncol 14: 2674–2681 113. Grogan M, Thomas GM, Melamed I, et al (1999) The importance of hemoglobin levels during radiotherapy for carcinoma of the cervix. Cancer 86: 1528–1536 114. Groopman JE, Itri LM (1999) Chemotherapy-induced anemia in adults: incidence and treatment. J Natl Cancer Inst 91: 1616–1634 115. Hainsworth JD, Gray JR, Stroup SL, et al (1997) Paclitaxel, carboplatin, and extended-schedule etoposide in the treatment of small-cell lung cancer: comparison of sequential phase II trials using different dose-intensities. J Clin Oncol 15: 3464–3470 116. Hainsworth JD, Levitan N, Wampler GL, et al (1995) Phase II randomized study of cisplatin plus etoposide phosphate or etoposide in the treatment of small-cell lung cancer. J Clin Oncol 13: 1436–1442 117. Halonen P, Mattila J, Makipernaa A, et al (2006) Erythrocyte concentrations of metabolites or cumulative doses of 6-mercaptopurine and methotrexate do not predict liver changes in children treated for acute lymphoblastic leukemia. Pediatr Blood Cancer 46: 762–766 118. Hamblin TJ, Orchard JA, Myint H, et al (1998) Fludarabine and hemolytic anemia in chronic lymphocytic leukemia. J Clin Oncol 16: 3209–3210 119. Harrison LB, Shasha D, White C, et al (2000) Radiotherapy-associated anemia: the scope of the problem. Oncologist 5: 1–7 120. Hasenclever D, Diehl V (1998) A prognostic score for advanced Hodgkin’s disease. International Prognostic Factors Project on Advanced Hodgkin’s Disease. N Engl J Med 339: 1506–1514 121. Haugen H, Magnusson B, Svensson M, et al (2004) Preradiotherapy hemoglobin level but not microvessel density predicts locoregional control and survival in laryngeal cancer treated with primary radical radiotherapy. Clin Cancer Res 10: 7941–7949 122. Heddens DK, Alberts DS, Garcia DJ, et al (1998) Factors associated with platinum-induced anemia in ovarian cancer patients in Southwest Oncology Group studies. Proc Am Soc Clin Oncol 17: 359a (Abstr 1387)

Signiﬁcance of anemia in cancer chemotherapy

237

123. Henke M (2001) Correction of cancer anemia – impact on disease course, prognosis and treatment efﬁcacy, particularly for patients undergoing radiotherapy. Onkologie 24: 450–454 124. Henry DH, Abels RI (1994) Recombinant human erythropoietin in the treatment of cancer and chemotherapy-induced anemia: Results of double-blind and open-label follow-up studies. Semin Oncol 21 [Suppl 3]: 21–28 125. Hensley ML, Lebeau D, Leon LF, et al (2001) Identiﬁcation of risk factors for requiring transfusion during front-line chemotherapy for ovarian cancer. Gynecol Oncol 81: 485–489 126. Highley MS, Schrijvers D, van Oosterom AT, et al (1997) Activated oxazaphosphorines are transported predominantly by erythrocytes. Ann Oncol 8: 1139–1144 127. Hill M, Norman A, Cunningham D, et al (1995) Impact of protracted venous infusion ﬂuorouracil with or without interferon alfa-2b on tumor response, survival, and quality of life in advanced colorectal cancer. J Clin Oncol 13: 2317–2323 128. Hirst DG (1986) Anemia: A problem or an opportunity in radiotherapy. Int J Radiat Oncol Biol Phys 12: 2009–2017 129. Höckel M, Vaupel P (2001) Tumor hypoxia: deﬁnitions and current clinical, biologic, and molecular aspects. J Natl Cancer Inst 93: 266–276 130. Hofheinz RD, Raab B, Mai S, et al (2004) Impact of chemoradiotherapy-induced anemia on survival in uniformly staged patients with locally advanced squamous cell carcinoma of the esophagus. Onkologie 27: 462–466 131. Horiguchi H, Kayama F, Oguma E, et al (2000) Cadmium and platinum suppression of erythropoietin production in cell culture: clinical implications. Blood 96: 3743–3747 132. Horiguchi H, Oguma E, Kayama F (2006) Cadmium and cisplatin damage erythropoietin-producing proximal renal tubular cells. Arch Toxicol 80: 680–686 133. Hoskins PJ, Swenerton KD (1994) Oral etoposide is active against platinumresistant epithelial ovarian cancer. J Clin Oncol 12: 60–63 134. Hudis CA, Seidman AD, Crown JP, et al (1996) Phase II and pharmacologic study of docetaxel as initial chemotherapy for metastatic breast cancer. J Clin Oncol 14: 58–65 135. Hussain M, Salwen W, Kucuk O, et al (1997) Paclitaxel, cisplatin, and 5ﬂuorouracil in patients with advanced or recurrent squamous cell carcinoma of the head and neck: a preliminary report. Semin Oncol 24 [Suppl 19]: S43–S45 136. Jacobs C, Lyman G, Velez-Garcia E, et al (1992) A phase III randomized study comparing cisplatin and ﬂuorouracil as single agents and in combination for advanced squamous cell carcinoma of the head and neck. J Clin Oncol 10: 257–263 137. Jagasia MH, Langer CJ, Johnson DH, et al (2001) Weekly irinotecan and cisplatin in advanced non-small cell lung cancer: a multicenter phase II study. Clin Cancer Res 7: 68–73 138. Jakobsen EB, Eickhoff JH, Andersen JP, et al (1994) Prognosis after nephrectomy for renal cell carcinoma. Scand J Urol Nephrol 28: 229–236 139. Jassem J, Karnicka-Mlodkowska H, van Pottelsberghe C, et al (1993) Phase II study of vinorelbine (Navelbine) in previously treated small cell lung cancer patients. EORTC Lung Cancer Cooperative Group. Eur J Cancer 29A: 1720–1722

238

M. R. Nowrousian

140. Jassem J, Krzakowski M, Roszkowski K, et al (2002) A phase II study of gemcitabine plus cisplatin in patients with advanced non-small cell lung cancer: clinical outcomes and quality of life. Lung Cancer 35: 73–79 141. Jazieh AR, Hussain M, Howington JA, et al (2000) Prognostic factors in patients with surgically resected stages I and II non-small cell lung cancer. Ann Thorac Surg 70: 1168–1171 142. Jiang J, Liang X, Zhou X, et al (2007) A meta-analysis of randomized controlled trials comparing carboplatin-based to cisplatin-based chemotherapy in advanced non-small cell lung cancer. Lung Cancer 57: 348–358 143. Jones A, Wiltshaw E, Harper P, et al (1992) A randomized study of high vs conventional-dose carboplatin for previously untreated ovarian cancer. Br J Cancer 65 [Suppl 16]: 15 (Abstr C8) 144. Jones S, Winer E, Vogel C, et al (1995) Randomized comparison of vinorelbine and melphalan in anthracycline-refractory advanced breast cancer. J Clin Oncol 13: 2567–2574 145. Jootar S, Chuncharunee S, Ongphiphaphadhanakul B, et al (1990) Multivariate analysis of prognostic factors in chronic myelogenous leukemia. J Med Assoc Thai 73: 662–669 146. Kavanagh JJ, Kudelka AP, de Leon CG, et al (1996) Phase II study of docetaxel in patients with epithelial ovarian carcinoma refractory to platinum. Clin Cancer Res 2: 837–842 147. Kaye SB, Lewis CR, Paul J, et al (1992) Randomised study of two doses of cisplatin with cyclophosphamide in epithelial ovarian cancer. Lancet 340: 329–333 148. Keating MJ (1999) Chronic lymphocytic leukemia. Semin Oncol 26 [Suppl 14]:107–114 149. Kosmidis P, Krzakowski M (2005) Anemia proﬁles in patients with lung cancer: What have we learned from the European Cancer Anaemia Survey (ECAS)? Lung Cancer 50: 401–412 150. Kosmidis P, Mylonakis N, Fountzilas G, et al (1997) Paclitaxel (175 mg/m2) plus carboplatin versus paclitaxel (225 mg/m2) plus carboplatin in non-small cell lung cancer: a randomized study. Semin Oncol 24 [Suppl 12]: S30–S33 151. Kosmidis PA, Tsavaris N, Skarlos D, et al (1996) Fluorouracil and leucovorin with or without interferon alfa-2b in advanced colorectal cancer: analysis of a prospective randomized phase III trial. Hellenic Cooperative Oncology Group. J Clin Oncol 14: 2682–2687 152. Kosty MP, Fleishman SB, Herndon JE, 2nd, et al (1994) Cisplatin, vinblastine, and hydrazine sulfate in advanced, non-small-cell lung cancer: a randomized placebo-controlled, double-blind phase III study of the Cancer and Leukemia Group B. J Clin Oncol 12: 1113–1120 153. Kroep JR, Peters GJ, van Moorsel CJA, et al (1999) Gemcitabine-cisplatin: a schedule-ﬁnding study. Ann Oncol 10: 1503–1510 154. Kudelka AP, Tresukosol D, Edwards CL, et al (1996) Phase II study of intravenous topotecan as a 5-day infusion for refractory epithelial ovarian carcinoma. J Clin Oncol 14: 1552–1557 155. Kudoh S, Fujiwara Y, Takada Y, et al (1998) Phase II study of irinotecan combined with cisplatin in patients with previously untreated small-cell lung cancer. West Japan Lung Cancer Group. J Clin Oncol 16: 1068–1074

Signiﬁcance of anemia in cancer chemotherapy

239

156. Kumar P (2000) Impact of anemia in patients with head and neck cancer. Oncologist 5: 13–18 157. Kyle RA (1975) Multiple myeloma: review of 869 cases. Mayo Clin Proc 50: 29–40 158. Lammering G, Carl UM, Pape H, et al (1999) [Changes in hemoglobin concentrations in combined radio- and chemotherapy in locally advanced ORL tumors.] Strahlenther Onkol 175: 559–562 159. Langer CJ, Leighton JC, Comis RL, et al (1995) Paclitaxel and carboplatin in combination in the treatment of advanced non-small-cell lung cancer: a phase II toxicity, response, and survival analysis. J Clin Oncol 13: 1860–1870 160. Laurie S, Jeyabalan N, Nicholas G, et al (2006) Association between anemia arising during therapy and outcomes of chemoradiation for limited small-cell lung cancer. J Thorac Oncol 1: 146–151 161. Lee WR, Berkey B, Marcial V, et al (1998) Anemia is associated with decreased survival and increased locoregional failure in patients with locally advanced head and neck carcinoma: a secondary analysis of RTOG 85–27. Int J Radiat Oncol Biol Phys 42: 1069–1075 162. Leitgeb C, Pecherstorfer M, Fritz E, et al (1994) Quality of life in chronic anemia of cancer during treatment with recombinant human erythropoietin. Cancer 73: 2535–2542 163. Leonard RC, Hayward RL, Prescott RJ, et al (1991) The identiﬁcation of discrete prognostic groups in low grade non-Hodgkin’s lymphoma. The Scotland and Newcastle Lymphoma Group Therapy Working Party. Ann Oncol 2: 655–662 164. Littlewood TJ, Bajetta E, Nortier JWR, et al for the Epoetin Alfa Study Group (2001a) Effects of epoetin alfa on hematologic parameters and quality of life in cancer patients receiving nonplatinum chemotherapy: Results of a randomized, double-blind, placebo-controlled trial. J Clin Oncol 19: 2865–2874 165. Littlewood TJ (2001b) The impact of hemoglobin levels on treatment outcomes in patients with cancer. Semin Oncol 28 [Suppl 8]: 49–53 166. Loehrer PJ Sr, Ansari R, Gonin R, et al (1995) Cisplatin plus etoposide with and without ifosfamide in extensive small-cell lung cancer: a Hoosier Oncology Group study. J Clin Oncol 13: 2594–2599 167. Lohynska R, Slavicek A, Bahanan A, et al (2005) Predictors of local failure in early laryngeal cancer. Neoplasma 52: 483–488 168. Lorusso V, Manzione L, De Vita F, et al (2000) Gemcitabine plus cisplatin for advanced transitional cell carcinoma of the urinary tract: a phase II mulitcenter trial. J Urol 164: 53–56 169. Lorvidhaya V, Kamnerdsupaphon P, Chitapanarux I, et al (2004) Cisplatin and gemcitabine in patients with metastatic cervical cancer. Gan To Kagaku Ryoho 31: 1057–1062 170. Ludwig H, van Belle S, Barrett-Lee P (2004) The European Cancer Anaemia Survey (ECAS): a large, multinational, prospective survey deﬁning the prevalence, incidence, and treatment of anaemia in cancer patients. Eur J Cancer 40: 2293–2306 171. Luikart SD, Goutsou M, Mitchell ED, et al (1993) Phase I/II trial of etoposide and carboplatin in extensive small-cell lung cancer. A report from the Cancer and Leukemia Group B. Am J Clin Oncol 16: 127–131

240

M. R. Nowrousian

172. Luo FR, Wyrick SD, Chaney SG (1999) Pharmacokinetics and biotransformations of oxaliplatin in comparison with ormaplatin following a single bolus intravenous injection in rats. Cancer Chemother Pharmacol 44: 19–28 173. Lyman GH, Berndt ER, Kallich JD, et al (2005) The economic burden of anemia in cancer patients receiving chemotherapy. Value in Health 8: 149–156 174. Macdonald G, Hurman DC (2004) Inﬂuence of anaemia in patients with head and neck cancer receiving adjuvant postoperative radiotherapy in the Grampian region. Clin Oncol (R Coll Radiol) 16: 63–70 175. Matsumoto T, Endoh K, Kamisango K, et al (1990) Effect of recombinant human erythropoietin on anticancer drug-induced anaemia. Br J Haematol 75: 463–468 176. McGuire WP, Hoskins WJ, Brady MF, et al (1995) Assessment of dose-intensive therapy in suboptimally debulked ovarian cancer: a Gynecologic Oncology Group study. J Clin Oncol 13: 1589–1599 177. McGuire WP, Hoskins WJ, Brady MF, et al (1996) Cyclophosphamide and cisplatin compared with paclitaxel and cisplatin in patients with stage III and stage IV ovarian cancer. N Engl J Med 334: 1–6 178. McLaren CE, Holcombe RF, Wong WM (2000) Patient-speciﬁc sequential analysis to monitor cisplatin-induced anemia. Proc Am Soc Clin Oncol 9: 640a (Abstr 2527) 179. Meyer RM, Browman GP, Samosh ML, et al (1995) Randomized phase II comparison of standard CHOP with weekly CHOP in elderly patients with nonHodgkin’s lymphoma. J Clin Oncol 13: 2386–2393 180. Millan CG, Marinero ML, Castaneda AZ (2004) Drug, enzyme and peptide delivery using erythrocytes as carriers. J Control Release 95: 27–49 181. Miller AA, Herndon JE 2nd, Hollis DR, et al (1995a) Schedule dependency of 21-day oral versus 3-day intravenous etoposide in combination with intravenous cisplatin in extensive-stage small-cell-lung cancer: a randomized phase III study of the Cancer and Leukemia Group B. J Clin Oncol 13: 1871–1879 182. Miller AA, Niell HB, Grifﬁn JP (1995b) Phase II study of prolonged oral etoposide in combination with intravenous cisplatin in advanced non-small cell lung cancer. Lung Cancer 12: 59–65 183. Miller VA, Rigas JR, Francis PA, et al (1995c) Phase II trial of a 75-mg/m2 dose of docetaxel with prednisone premedication for patients with advanced nonsmall cell lung cancer. Cancer 75: 968–972 184. Millward MJ, Bishop JF, Friedlander M, et al (1996) Phase II trial of a 3-hour infusion of paclitaxel in previously untreated patients with advanced nonsmall-cell lung cancer. J Clin Oncol 14: 142–148 185. Mittelman M (2003) The implications of anemia in multiple myeloma. Clin Lymphoma 4 [Suppl 1]: S23–29 186. Morel P, Monconduit M, Jacomy D, et al (2000) Prognostic factors in Waldenström macroglobulinemia: a report on 232 patients with the description of a new scoring system and its validation on 253 other patients. Blood 96: 852–858 187. Moullet I, Salles G, Ketterer N, et al (1998) Frequency and signiﬁcance of anemia in non-Hodgkin’s lymphoma patients. Ann Oncol 9: 1109–1115 188. Mounier N, Morel P, Haioun C, et al (1998) A multivariate analysis of the survival of patients with aggressive lymphoma. Cancer 82: 1952–1962

Signiﬁcance of anemia in cancer chemotherapy

241

189. Münstedt K, Kovacic M, Zygmunt M, et al (2003) Impact of hemoglobin levels before and during chemotherapy on survival of patients with ovarian cancer. Int J Oncol 23: 837–843 190. Münstedt K, Völzing M, von Georgi R (2004) Hemoglobin levels during radiation therapy and their inﬂuence on local control and survival of patients with endometrial carcinoma. Oncol Rep 11: 711–717 191. Murphy WK, Fossella FV, Winn RJ, et al (1993) Phase II study of taxol in patients with untreated advanced non-small-cell lung cancer. J Natl Cancer Inst 85: 384–388 192. Nabholtz JM, Gelmon K, Bontenbal M, et al (1996) Multicenter, randomized comparative study of two doses of paclitaxel in patients with metastatic breast cancer. J Clin Oncol 14: 1858–1867 193. Nagao K, Fukuoka M, Fujita A, et al (2000) [A phase II study of irinotecan combined with cisplatin in non-small cell lung cancer. CPT-11 Lung Cancer Study Group.] Gan To Kagaku Ryoho 27: 413–421 194. Nowrousian MR, Schmidt CG (1982) Effects of cisplatin on different haemopoietic progenitor cells in mice. Br J Cancer 46: 397–402 195. Nowrousian MR, Kasper C, Oberhoff C, et al (1996) Pathophysiology of cancerrelated anemia. In: Smyth JF, Boogaerts MA, Ehmer BRM (eds) rhErythropoietin in Cancer Supportive Treatment. Marcel Dekker, New York, pp 13–34 196. Nowrousian MR, Ernst S, Beling C, et al (2005) CEBOPP/VIML followed by radiotherapy for intermediate- and advanced-stage Hodgkin’s disease: A longterm analysis. Blood 106: 751a (Abstr 2675) 197. Obermair A, Petru E, Windbichler G, et al (2000) Signiﬁcance of pretreatment serum hemoglobin and survival in epithelial ovarian cancer. Oncol Rep 7: 639–644 198. Obermair A, Cheuk R, Horwood K, et al (2001) Impact of hemoglobin levels before and during concurrent chemoradiotherapy on the response of treatment in patients with cervical carcinoma. Cancer 92: 903–908 199. Obermair A, Cheuk R, Horwood K, et al (2003) Anemia before and during concurrent chemoradiotherapy in patients with cervical carcinoma: Effect on progression-free survival. Int J Gynecol Cancer 13: 633–639 200. O’Rourke M, Crawford J, Schiller J, et al (1993) Survival advantage for patients with stage IV NSCLC treated with single agent Navelbine© in a randomized controlled trial. Proc Am Soc Clin Oncol 12: 343 (Abstr) 201. Oettle H, Riess H, Raguse JD, et al (2001) Recombinant human erythropoietin in the treatment of head and neck tumor anaemia. Int J Oral Maxillofac Surg 30: 148–155 202. Oivanen TM (1996) Plateau phase in multiple myeloma: an analysis of longterm follow-up of 432 patients. Finnish Leukaemia Group. Br J Haematol 92: 834–839 203. Okamoto H, Saijo N, Shinkai T, et al (1992) Chemotherapy-induced anemia in patients with primary lung cancer. Ann Oncol 3: 819–824 204. Ossa DF, Briggs A, McIntosh E, et al (2007) Recombinant erythropoietin for chemotherapy-related anemia. Pharmacoeconomics 26: 223–237 205. Overgaard J, Hansen HS, Jorgensen K, et al (1986) Primary radiotherapy of larynx and pharynx carcinoma – An analysis of some factors inﬂuencing local control and survival. Int J Radiat Oncol Biol Phys 12: 515–521

242

M. R. Nowrousian

206. Paredes J, Hong WK, Felder TB, et al (1988) Prospective randomized trial of high-dose cisplatin and ﬂuorouracil infusion with or without sodium diethyldithiocarbamate in recurrent and/or metastatic squamous cell carcinoma of the head and neck. J Clin Oncol 6: 955–962 207. Park SH, Lee J, Lee SH, et al (2006) Anemia is the strongest prognostic factor for outcomes of 5-ﬂuorouracil-based ﬁrst-line chemotherapy in patients with advanced gastric cancer. Cancer Chemother Pharmacol 57: 91–96 208. Peters-Engl C, Cassik P, Schmidt I, et al (2005) Impact of haemoglobin levels during adjuvant chemotherapy on the survival of patients with primary breast cancer. Acta Oncol 44: 129–133 209. Petrelli N, Douglass HO Jr, Herrera L, et al (1989) The modulation of ﬂuorouracil with leucovorin in metastatic colorectal carcinoma: a prospective randomized phase III trial. Gastrointestinal Tumor Study Group. J Clin Oncol 7: 1419–1426 210. Pfreundschuh M, Trümper L, Kloess M, et al (2004) Two-weekly or 3-weekly CHOP chemotherapy with or without etoposide for the treatment of elderly patients with aggressive lymphomas: results of the NHL-B2 trial of the DSHNHL. Blood 104: 634–641 211. Piccart MJ, Gore M, Ten Bokkel Huinink W, et al (1995) Docetaxel: an active new drug for treatment of advanced epithelial ovarian cancer. J Natl Cancer Inst 87: 676–681 212. Pines A, Ben-Bassat I, Modan M, et al (1987) Survival and prognostic factors in chronic lymphocytic leukemia. Eur J Haematol 38: 123–130 213. Pirker R, Krajnik G, Zochbauer S, et al (1995) Paclitaxel/cisplatin in advanced non-small-cell lung cancer (NSCLC). Ann Oncol 6: 833–835 214. Pivot X, Guardiola E, Etienne M, et al (2000) An analysis of potential factors allowing an individual prediction of cisplatin-induced anaemia. Eur J Cancer 36: 852–857 215. Postmus PE, Giaccone G, Debruyne C, et al (1996) Results of the phase II EORTC study comparing paclitaxel/cisplatin with teniposide/cisplatin in patients with non-small cell lung cancer. EORTC Lung Cancer Cooperative Group. Semin Oncol 23 [Suppl 12]: 10–13 216. Pradier O, Lederer K, Hille A, et al (2005) Concurrent low-dose cisplatin and thoracic radiotherapy in patients with inoperable stage III non-small cell lung cancer: a phase II trial with special reference to the hemoglobin level as prognostic parameter . J Cancer Res Clin Oncol 131: 261–269 217. Prosnitz RG, Yao B, Farrell CL, et al (2005) Pretreatment anemia is correlated with the reduced effectiveness of radiation and concurrent chemotherapy in advanced head and neck cancer. Int J Radiat Oncol Biol Phys 15: 1087– 1095 218. Rai KR, Sawitsky A, Cronkite EP, et al (1975) Clinical staging of chronic lymphocytic leukemia. Blood 46: 219–234 219. Ramanathan-Girish S, Boroujerdi M (2001) Contradistinction between doxorubicin and epirubicin: in vitro interaction with blood components. J Pharm Pharmacol 53: 815–821 220. Ranson MR, Jayson G, Perkins S, et al (1997) Single-agent paclitaxel in advanced non-small cell lung cancer: single-center phase II study using a 3-hour administration schedule. Semin Oncol 24 [Suppl 12]: S6–S9

Signiﬁcance of anemia in cancer chemotherapy

243

221. Ravdin PM, Burris HA 3rd, Cook G, et al (1995) Phase II trial of docetaxel in advanced anthracycline-resistant or anthracenedione-resistant breast cancer. J Clin Oncol 13: 2879–2885 222. Ray-Coquard I, Le Cesne A, Rubio MT, et al (1999) Risk model for severe anemia requiring red blood cell transfusion after cytotoxic conventional chemotherapy regimens. The Elypse 1 Study Group. J Clin Oncol 17: 2840–2846 223. Robert F, Wheeler RH, Molthrop D, et al (1994) Phase 2 study of prolonged administration of oral etoposide in combination with weekly cisplatin in advanced non-small cell lung cancer. Am J Clin Oncol 17: 383–386 224. Rose PG, Blessing JA, Mayer AR, et al (1998) Prolonged oral etoposide as second-line therapy for platinum-resistant and platinum-sensitive ovarian carcinoma: a Gynecologic Oncology Group study. J Clin Oncol 16: 405–410 225. Rossof AH, Slayton RE, Perlia CP (1972) Preliminary clinical experience with cis diamminedichloroplatinum (II) (NSC 119875, CACP). Cancer 30: 1451–1456 226. Rothenberg ML, Eckardt JR, Kuhn JG, et al (1996) Phase II trial of irinotecan in patients with progressive or rapidly recurrent colorectal cancer. J Clin Oncol 14: 1128–1135 227. Rougier P, Bugat R, Douillard JY, et al (1997) Phase II study of irinotecan in the treatment of advanced colorectal cancer in chemotherapy-naive patients and patients pretreated with ﬂuorouracil-based chemotherapy. J Clin Oncol 15: 251–260 228. Rozencweig M, Martin A, Beltangady M, et al (1990) Randomized trials of carboplatin versus cisplatin in advanced ovarian cancer. In: Bunn PA, Canetta R, Ozols RF, Rozencweig M (eds) Carboplatin: current perspectives and future directions. Saunders, Philadelphia, pp 175–186 229. Rytting M, Worth L, Jaffe N (1996) Hemolytic disorders associated with cancer. Hematol Oncol Clin North Am 10: 365–376 230. Sadahiro S, Suzuki T, Tokunaga N, et al (1998) Anemia in patients with colorectal cancer. J Gastroenterol 33: 488–494 231. Salman H, Perez A, Sparano JA, et al (2003) Phase II trial of infusional cyclophosphamide, idarubicin and etoposide in poor prognosis non-Hodgkin’s lymphoma. Am J Clin Oncol 26: 338–343 232. Sanchiz F, Milla A (1994) Tegafur-uracil (UFT) plus folinic acid in advanced rectal cancer. Jpn J Clin Oncol 24: 322–326 233. Santin AD, Bellone S, Parrish RS, et al (2003) Inﬂuence of allogeneic blood transfusion on clinical outcome during radiotherapy for cancer of the uterine cervix. Gynecol Obstet Invest 56: 28–34 234. Saylors RL, 3rd, Stine KC, Sullivan J, et al (2001) Cyclophosphamide plus topotecan in children with recurrent or refractory solid tumors: a Pediatric Oncology Group phase II study. J Clin Oncol 19: 3463–3469 235. Schäfer U, Micke O, Müller SB, et al (2003) Hemoglobin as an indepent prognostic factor in the radiotherapy of head and neck tumors. Strahlenther Onkol 179: 527–534 236. Schiller JH, Kim K, Hutson P, et al (1996) Phase II study of topotecan in patients with extensive-stage small-cell carcinoma of the lung: an Eastern Cooperative Oncology Group trial. J Clin Oncol 14: 2345–2352 237. Schrijvers D (2003) Role of red blood cells in pharmacokinetics of chemotherapeutic agents. Clin Pharmacokinet 42: 779–791

244

M. R. Nowrousian

238. Seidman AD, Reichman BS, Crown JP, et al (1995) Paclitaxel as second and subsequent therapy for metastatic breast cancer: activity independent of prior anthracycline response. J Clin Oncol 13: 1152–1159 239. Sengelov L, Kamby C, Geersten P, et al (2000) Predictive factors of response to cisplatin-based chemotherapy and the relation of response to survival in patients with metastatic urothelial cancer. Cancer Chemother Pharmacol 46: 357–364 240. Seshadri T, Prince HM, Bell DR, et al (2005) The Australian Cancer Anaemia Survey: a snapshot of anaemia in adult patients with cancer. Med J Aust 182: 453–457 241. Sertoli, MR, Santini, G, Chisesi T, et al (1994) MACOP-B versus ProMACEMOPP in the treatment of advanced diffuse non-Hodgkin’s lymphoma: results of a prospective randomized trial by the non-Hodgkin’s Lymphoma Cooperative Study Group. J Clin Oncol 12: 1366–1374 242. Shannon AM, Bouchier-Hayes DJ, Condron CM, et al (2005) Correction of anaemia through the use of darbepoetin alfa improves chemotherapeutic outcome in a murine model of Lewis lung carcinoma. Br J Cancer 93: 224–232 243. Shasha D (2001) The negative impact of anemia on radiotherapy and chemoradiation outcomes. Semin Hematol 38 [Suppl 7]: 8–15 244. Shepherd FA, Cormier Y, Burkes R, et al (1997) Phase II trial of gemcitabine and weekly cisplatin for advanced non-small cell lung cancer. Semin Oncol 24 [Suppl 8]: S27–S30 245. Shin DM, Glisson BS, Khuri FR, et al (1998) Phase II trial of paclitaxel, ifosfamide, and cisplatin in patients with recurrent head and neck squamous cell carcinoma. J Clin Oncol 16: 1325–1330 246. Sieber M, Bredenfeld H, Josting A, et al (2003) 14-Day variant of the bleomycin, etoposide, doxorubicin, cyclophosphamide, vincristine, procarbazine, and prednisone regimen in advanced-stage Hodgkin’s lymphoma: Results of a pilot study of the German Hodgkin’s Lymphoma Study Group. J Clin Oncol 21: 1734–1739 247. Silver DF, Piver MS (1999) Effects of recombinant human erythropoietin on the antitumor effect of cisplatin in SCID mice bearing human ovarian cancer: A possible oxygen effect. Gynecol Oncol 73: 280–284 248. Skarlos DV, Samantas E, Kosmidis P, et al (1994) Randomized comparison of etoposide-cisplatin vs. etoposide-carboplatin and irradiation in small-cell lung cancer. A Hellenic Co-operative Oncology Group study. Ann Oncol 5: 601–607 249. Skarlos DV, Aravantinos G, Kosmidis P, et al (1997) Paclitaxel with carboplatin versus paclitaxel with carboplatin alternating with cisplatin as ﬁrst-line chemotherapy in advanced epithelial ovarian cancer: preliminary results of a Hellenic Cooperative Oncology Group study. Semin Oncol 24 [Suppl 15]: S57–S61 250. Skillings JR, Rogers-Melamed I, Nabholtz JM, et al (1995) An epidemiological review of anaemia in cancer chemotherapy in Canada. Eur J Cancer 31A [Suppl 5]: S183 (Abstr 879) 251. Skillings JR, Rogers-Melamed I, Nabholtz JM, et al (1999) An epidemiological review of red cell transfusions in cancer chemotherapy. Cancer Prev Control 3: 207–212 252. Skillings JR, Sridhar FG, Wong C, et al (1993) The frequency of red cell transfusion for anemia in patients receiving chemotherapy. A retrospective cohort study. Am J Clin Oncol 16: 22–25

Signiﬁcance of anemia in cancer chemotherapy

245

253. Skorokhod O, Garmaeva T, Vitvitsky V, et al (2004) Pharmacokinetics of erythrocyte-bound daunorubicin in patients with acute leukemia. Med Sci Monit 10: PI55–64 254. Skorokhod O, Kulikova EV, Galkina NM, et al (2007) Doxorubicin pharmacokinetics in lymphoma patients treated with doxorubicin-loaded erythrocytes. Haematologica 92: 570–571 255. Smaniotto D, Luzi S, Morganti AG, et al (2000) Prognostic signiﬁcance of anemia and role of erythropoietin in radiation therapy. Tumori 86: 17–23 256. Smith RE, Lew D, Rodriguez GI, et al (1996) Evaluation of topotecan in patients with recurrent or metastatic squamous cell carcinoma of the head and neck. A phase II Southwest Oncology Group study. Invest New Drugs 14: 403–407 257. Smyth JF, Smith IE, Sessa C, et al (1994) Activity of docetaxel (Taxotere) in small cell lung cancer. The Early Clinical Trials Group of the EORTC. Eur J Cancer 30 A: 1058–1060 258. Steurer M, Wagner H, Gastl G (2004) Prevalence and management of anaemia in haematologic cancer patients receiving cyclic nonplatinum chemotherapy: Results of a prospective national chart survey. Wien Klin Wochenschr 116: 367–372 259. Straus DJ, Gaynor JJ, Myers J, et al (1990) Prognostic factors among 185 adults with newly diagnosed advanced Hodgkin’s disease treated with alternating potentially noncross-resistant chemotherapy and intermediate-dose radiation therapy. J Clin Oncol 8: 1173–1186 260. Stüben G, Thews O, Pöttgen C, et al (2001) Recombinant human erythropoietin increases the radiosensitivity of xenografted human tumours in anaemic nude mice. J Cancer Res Clin Oncol 127: 346–350 261. Sugiyama T, Yakushiji M, Noda K, et al (2000) Phase II study of irinotecan and cisplatin as ﬁrst-line chemotherapy in advanced or recurrent cervical cancer. Oncology 58: 31–37 262. Sutton GP, Blessing JA, Homesley HD, et al (1989) Phase II trial of ifosfamide and mesna in advanced ovarian carcinoma: a Gynecologic Oncology Group study. J Clin Oncol 7: 1672–1676 263. Swenerton K, Jeffrey J, Stuart G, et al (1992) Cisplatin-cyclophosphamide versus carboplatin-cyclophosphamide in advanced ovarian cancer: a randomized phase III study of the National Cancer Institute of Canada Clinical Trials Group. J Clin Oncol 10: 718–726 264. Tammemagi CM, Neslund-Dudas C, Simoff M, et al (2003) Impact of comorbidity on lung cancer survival. Int J Cancer 103: 792–802 265. Tampellini M, Saini A, Alabiso I, et al (2006) The role of haemoglobin level in predicting the response to ﬁrst-line chemotherapy in advanced colorectal cancer patients. Br J Cancer 95: 13–20 266. Tamussino KF, Gücer F, Reich O, et al (2001) Pretreatment hemoglobin, platelet count, and prognosis in endometrial carcinoma. Int J Gynecol Cancer 11: 236–240 267. Teicher BA (1994) Hypoxia and drug resistance. Cancer Metastasis Rev 13: 139–168 268. Teicher BA (1995) Physiologic mechanisms of therapeutic resistance. Blood ﬂow and hypoxia. Hematol Oncol Clin North Am 9: 475–506

246

M. R. Nowrousian

269. Teicher BA, Holden SA, al-Achi A, et al (1990) Classiﬁcation of antineoplastic treatments by their differential toxicity toward putative oxygenated and hypoxic tumor subpopulations in vivo in the FSaIIC murine ﬁbrosarcoma. Cancer Res 50: 3339–3344 270. ten Bokkel Huinink W, Gore M, Carmichail J, et al (1997) Topotecan versus paclitaxel for the treatment of current epithelial ovarian cancer. J Clin Oncol 15: 2183–2193 271. ten Tije AJ, Verweij J, Loos WJ, et al (2003) Pharmacological effects of formulation vehicles: implications for cancer chemotherapy. Clin Pharmacokinet 42: 665–685 272. Thatcher N (1994) Controlled study of the efﬁcacy and safety of recombinant human erythropoietin (epoetin alfa; Cilag) in the prevention of anaemia in patients with small cell lung cancer receiving chemotherapy. Ann Oncol 5 [Suppl 8]: 89 273. Thatcher N (1998) Management of chemotherapy-induced anemia in solid tumors. Semin Oncol 25 [Suppl 7]: 23–26 274. The International Non-Hodgkin’s Lymphoma Prognostic Factors Project (1993) A predictive model for aggressive non-Hodgkin’s lymphoma. N Engl J Med 329: 987–994 275. Thews O, Koenig R, Kelleher DK, et al (1998) Enhanced radiosensitivity in experimental tumours following erythropoietin treatment of chemotherapyinduced anaemia. Br J Cancer 78: 752–756 276. Thews O, Kelleher DK, Vaupel P (2001) Erythropoietin restores the anemiainduced reduction in cyclophosphamide cytotoxicity in rat tumors. Cancer Res 61: 1358–1361 277. Thigpen JT, Blessing JA, Ball H, et al (1994) Phase II trial of paclitaxel in patients with progressive ovarian carcinoma after platinum-based chemotherapy: a Gynecologic Oncology Group study. J Clin Oncol 12: 1748–1753 278. Thigpen JT, Brady MF, Homesley HD, et al (2004) Phase III trial of doxorubicin with or without cisplatin in advanced endometrial carcinoma: a Gynecologic Oncology Group study. J Clin Oncol 22: 3902–3908 279. Tomida A, Tsuruo T (1999) Drug resistance mediated by cellular stress response to the microenvironment of solid tumors. Anticancer Drug Des 14: 169– 177 280. Unami A, Nishina N, Terai T, et al (1996) Effects of cisplatin on erythropoietin production in rats. J Toxicol Sci 21: 157–165 281. Valero V, Holmes FA, Walters RS, et al (1995) Phase II trial of docetaxel: a new, highly effective antineoplastic agent in the management of patients with anthracycline-resistant metastatic breast cancer. J Clin Oncol 13: 2886–2894 282. Van Belle SJP, Cocquyt V (2003) Impact of hemoglobin levels on the outcome of cancers treated with chemotherapy. Crit Rev Oncol Hematol 47: 1–11 283. van de Pol SM, Doornaert PA, de Bree R, et al (2006) The signiﬁcance of anemia in squamous cell head and neck cancer treated with surgery and postoperative radiotherapy. Oral Oncol 42: 131–138 284. Vaziri ND, Zhou XJ, Liao SY (1994) Erythropoietin enhances recovery from cisplatin-induced acute renal failure. Am J Physiol 266: F360–F366 285. Veach S, Waltzman R, McGuckin J, et al (1998) A retrospective analysis of transfusion requirements according to salvage regimen in patients with recurrent ovarian cancer. Proc Am Soc Clin Oncol 17: 376a (Abstr)

Signiﬁcance of anemia in cancer chemotherapy

247

286. Venturini M, Mastro LD, Aitini E, et al (2005) Dose-dense adjuvant chemotherapy in early breast cancer patients: results from randomized trial. J Natl Cancer Inst 97: 1724–1733 287. Verschraegen CF, Levy T, Kudelka AP, et al (1997) Phase II study of irinotecan in prior chemotherapy-treated squamous cell carcinoma of the cervix. J Clin Oncol 15: 625–631 288. von Pawel J, Wagner H, Niederle N, et al (1996) Paclitaxel and cisplatin in patients with non-small cell lung cancer: results of a phase II trial. Semin Oncol 23 [Suppl 12]: 7–9 289. Wagner W, Hermann R, Hartlapp J, et al (2000) Prognostic value of hemoglobin concentrations in patients with advanced head and neck cancer treated with combined radio-chemotherapy and surgery. Strahlenther Onkol 176: 73–80 290. Weber BL, Vogel C, Jones S, et al (1995) Intravenous vinorelbine as ﬁrst-line and second-line therapy in advanced breast cancer. J Clin Oncol 13: 2722–2730 291. Weissenberger C, Geissler M, Otto F, et al (2006) Anemia and long-term outcome in adjuvant and neoadjuvant radiochemotherapy of stage II and III rectal adenocarcinoma: the Freiburg experience (1989–2002). World J Gastroenterol 12: 1849–1858 292. Wigren T, Oksanen H, Kellokumpu-Lehtinen P (1997) A practical prognostic index for inoperable non-small-cell lung cancer. J Cancer Res Clin Oncol 123: 259–266 293. Wildiers H, Guetens G, de Boeck G, et al (2002) Melphalan availability in hypoxia-inducible factor-1alpha+/+ and factor-1alpha −/− tumors is independent of tumor vessel density and correlates with melphalan erythrocyte transport. Int J Cancer 99: 514–519 294. Wiltshaw E, Kroner T (1976) Phase II study of cis-dichlorodiammineplatinum (II) (NSC-119875) in advanced adenocarcinoma of the ovary. Cancer Treat Rep 60: 55–60 295. Wolff AC, Ettinger DS, Neuberg D, et al (1995) Phase II study of ifosfamide, carboplatin, and oral etoposide chemotherapy for extensive-disease small-cell lung cancer: an Eastern Cooperative Oncology Group pilot study. J Clin Oncol 13: 1615–1622 296. Wood PA, Hrushesky JM (1995) Cisplatin-associated anemia: an erythropoietin deﬁciency syndrome. J Clin Invest 95: 1650–1659 297. Worden FP, Taylor JMG, Biermann JS, et al (2005) Randomized phase II evaluation of 6 g/m2 of ifosfamide plus doxorubicin and granulocyte colonystimulating factor (G-CSF) compared with 12 g/m2 of ifosfamide plus doxorubicin and G-CSF in the treatment of poor-prognosis soft tissue sarcoma. J Clin Oncol 23: 105–112 298. Wouters A, Pauwels B, Lardon F, et al (2007) Review: implications of in vitro research on the effect of radiotherapy and chemotherapy under hypoxic conditions. The Oncologist 12: 690–712 299. Wozniak AJ, Crowley JJ, Balcerzak SP, et al (1998) Randomized trial comparing cisplatin with cisplatin plus vinorelbine in the treatment of advanced nonsmall-cell lung cancer: a Southwest Oncology Group study. J Clin Oncol 16: 2459–2465 300. Yalcin S, Müftüoglu S, Cetin E, et al (2003) Protection against cisplatin-induced nephrotoxicity by recombinant human erythropoietin. Med Oncol 20: 169–173

248

M. R. Nowrousian: Signiﬁcance of anemia in cancer chemotherapy

301. Yasunaga Y, Shin M, Miki T, et al (1998) Prognostic factors of renal cell carcinoma: a multivariate analysis. J Surg Oncol 68: 11–18 302. Young SD, Hill RP (1990) Effects of reoxygenation on cells from hypoxic regions of solid tumors: anticancer drug sensitivity and metastatic potential. J Natl Cancer Inst 82: 371–380 303. Zinzani PL, Tani M, Alinari L, et al (2005) Role of anemia in survival of patients with elderly aggressive non-Hodgkin’s lymphoma after chemotherapy. Leuk Lymphoma 46: 1449–1454 Correspondence: Prof. Dr. M. R. Nowrousian, Department of Internal Medicine (Cancer Research), West German Cancer Center, University Hospital of Essen, Hufelandstrasse 55, 45122 Essen, Germany, E-mail: [email protected]

Chapter 9

Incidence and impact of anemia in radiation oncology J. Dunst1 and M. Molls2 1

2

Department of Radiation Oncology, University of Luebeck, Germany Department of Radiation Oncology, Technical University Munich, Germany

Introduction About 45 years ago, Thomlinson and Gray (1955) demonstrated the presence of radioresistant hypoxic cells in experimental cancers. Their work stimulated numerous clinical and experimental investigations on hypoxia in tumors. In subsequent investigations, a marked impact of hemoglobin (Hb) levels on treatment outcome was found by Evans and Bergsjö in 1965. They found a stage-depenodent linear relationship between Hb levels and local control as well as survival after radiotherapy. Evans and Bergsjö interpreted their ﬁndings as a result of anemia-induced tumor hypoxia. Although numerous other studies have further supported an association between anemia and decreased local control and survival in radiotherapy patients, some controversies have remained and a major issue of current research is whether or not the wellknown association between anemia and survival reﬂects a causal relationship or represents only an epi-phenomemon (Fyles et al. 2000).

Frequency of anemia Laboratory ﬁndings and pathophysiology Anemia in cancer patients may result from the malignant disease itself, or can be caused by speciﬁc tumor therapies such as chemotherapy, radiotherapy or surgery or may result from independent concomitant diseases. A variety of mechanisms may cause or contribute to the presence and severity of anemia and should be taken into consideration for differential diagnosis (Table 1). Laboratory work-up reveals normal values of mean corpuscular volume (MCV) and mean corpuscular Hb concentration (MCHC) in most patients. Microcytic anemia (MCV < 80 ﬂ) occurs in about 14% of patients with solid cancers and anemia. The reticulocyte count is lowered indicating hypo-

250

J. Dunst and M. Molls

Table 1. Differential diagnosis of anemia in cancer patients Anemia caused by tumor • Tumor-related anemia (cytokine-mediated anemia comparable to anemia in chronic inﬂammatory diseases) • Acute blood loss due to tumor bleeding • Chronic blood loss due to subclinical bleeding • Bone marrow inﬁltration Anemia caused by cancer therapy • Hematological toxicity of chemotherapy and/or radiotherapy • Blood loss during surgery Patient-related causes of anemia others than tumor • Concomitant medical illness (e.g. renal failure, esophageal varicosis, rheumatoid arthritis) • Concomitant medication (e.g. NSAR) • Hereditary anemias

regenerative anemia. Serum iron levels, total iron binding capacity and transferrin saturation are often lowered whereas serum ferritin and total amount of marrow iron are normal. These ﬁndings support the assumption that cancer-related anemia does not result from an iron deﬁciency but from impaired mobilization of iron from storages (Nowrousian et al. 2006). A further characteristic ﬁnding is the relatively low serum levels of erythropoietin which are, with regard to the Hb levels, lower than in anemic patients with blood loss (Miller et al. 1990). This ﬁnding suggests that insufﬁcient endogeneous erythropoietin production is a major causative factor for anemia in patients with solid cancers (Nowrousian et al. 2006). The serum-concentrations of a variety of cytokines such as interleukin 6 or VEGF are increased in patients with solid cancers and anemia (Dunst et al. 2002). With regard to the mentioned laboratory ﬁndings, anemia in patients with solid cancers resembles the anemia of chronic inﬂammatory disease (ACD, e.g. in rheumatoid arthritis or Crohn’s disease).

Pretreatment anemia In contrast to hematological malignancies, severe anemia is relatively uncommon in patients with solid tumors who receive radiotherapy with curative intent. In summary, mild to moderate anemia, however, is frequently observed in patients undergoing radiotherapy. Moderate to severe anemia (Hb between 9 and 11 g/dl) is noticed in about 15%–20% of patients with head and neck and cervical cancers but was not observed in a smaller group of patients with esophageal and rectal cancers. Severe, transfusion-requiring

Incidence and impact of anemia in radiation oncology

251

Table 2. Frequency of anemia (Hb < 12 g/dl) prior to and during radiotherapy. Data from the Beth Israel Medical/St.Luke’s Roosevelt Hospital Center New York. Harrison et al. 2002 Tumor site

N

At baseline

During radiotherapy

Breast cancer Colorectal cancer Lung cancer Prostate cancer Uterine/cervical cancer Head and neck cancer

81 64 64 90 53 68

44% 44% 55% 9% 75% 16%

45% 63% 77% 26% 79% 32%

anemia (Hb < 9 g/dl) is normally observed only in a very minority of patients, less than 1% of all patients and 4% of patients with cervical cancers. In the ﬁrst publication that has highlighted the impact of Hb levels on treatment outcome, Evans and Bergsjö (1965) analysed 880 patients with stage I–IV cancer of the cervix. The frequency of patients with a pretreatment Hb <11 g/dl was 25%. Anemia (Hb <11 g/dl) increased with stage (13% in stage I, 21% in stage II, 34% in stage III, 73% in stage IV and 20% in patients treated postoperatively). Girinski et al. (1989) analysed 386 patients with cervical cancers from the Institute Gustave-Roussy with stage IIB (30%) or IIIB (70%). 34 patients of the whole group (9%) presented with a pretreatment Hb of less than 10 g/dl. 42 patients (10%) received transfusions prior to (6% also during) treatment and additional 56 patients (15%) were transfused during treatment. In an analysis by Grogan and coworkers (1999) in patients with cervical cancers, 35% of all patients had a pretreatment Hb <12 g/dl. A large survey on anemia in a general population of radiation therapy patients has recently been published by Harrison et al. (2002). They analyzed 574 randomly selected patients treated in a single institution in the period from December 1996 through June 1999. At presentation, 41% of patients had Hb levels <12 g/dl. By the end of radiotherapy, this percentage had increased to 54%. The prevalence of anemia was site dependent and most frequently observed in patients with cervical cancers (75%) (Table 2). In the analysis of our institutions, tumor site (head and neck vs. cervix vs. esophagus vs. rectum) had no signiﬁcant impact on average pretreatment Hb levels although the range was broader in patients with cancers of the head and neck and, especially, cervical cancers. Pronounced anemia with Hb levels < about 11 g/dl was mainly found in cervical and head and neck tumors. Tumor bleedings are probably not the most likely explanation for this ﬁnding because clinically relevant bleeding is rarely observed in most head and neck cancers and a prognostic impact of anemia has been demonstrated in early

252

J. Dunst and M. Molls

glottic cancers which do not bleed (Canaday et al. 1999). However, the clinical observation of a higher frequency of tumor bleedings in patients with cervical cancers may contribute to pretreatment anemia in this cancer site.

Radiation-induced anemia The bone marrow stem cells are highly radiosensitive. Even relatively low radiation doses may result in marked reduction of the progenitor cells within the irradiated marrow volume. Therefore, the risk of radiation-induced anemia in cancer patients mainly depends on the irradiated bone marrow volume. Total dose and fractionation seem to play a minor role as most patients with solid cancers and hematological malignancies receive total radiation doses far above the threshold for acute bone marrow toxicity. The amount of marrow in typical radiation ﬁelds mainly depends on the target volume and radiation technique. In most clinical situations, less than 10–20% of the bone marrow receives radiation doses above the threshold for marrow toxicity. The risk of anemia induced by radiotherapy alone is therefore relatively low because the non-irradiated marrow can fully compensate for the effects in the irradiated marrow parts. In the recent years, technical progress has allowed more complicated treatment planning on the basis of better imaging techniques and multiple ﬁeld arrangements. In general, these advanced methods of treatment planning and treatment delivery (e.g. 3D-conformal radiotherapy and IMRT) offer the chance to deliver higher doses and reduce acute toxicity (Ahmed et al. 2004). Reduction of toxicity mainly results from the use of multiple ﬁelds and smaller target volumes. However, it is currently not clear whether these techniques reduce bone marrow toxicity. If multiple ﬁelds are used, the amount of irradiated normal tissue volume increases although the mean dose and the absolute volume of normal tissue in the high dose volume decrease. Because of the low threshold of bone marrow toxicity, however, the amount of marrow volume which is irradiated with a dose above the threshold for bone marrow toxicity may increase. In previously untreated patients, radiotherapy as sole modality bears a low risk of treatment-related anemia except rare situations with extremely large radiation ﬁelds (e.g. total-nodal irradiation for Hodgkin’s disease). In solid tumors with relatively large ﬁelds (e.g. pelvic or pelvic plus paraaortic radiotherapy for gynecological malignancies or thoracic radiotherapy for lung cancer), a signiﬁcant drop of Hb during the whole course of radiotherapy is to be expected. However, the absolute difference between pretreatment Hb levels and Hb levels at the end of radiotherapy remains relatively small and seems to be in the range of about 1 g/dl for most clinical situations. In an analysis by van Acht et al. (1992), 40/216 patients with glottic cancers (19%) and 25/70 patients (36%) with supraglottic cancers had Hb levels below the normal range at the end of a curative radiation regimen.

Incidence and impact of anemia in radiation oncology

253

These ﬁgures are nearly identical to the frequency of reduced Hb levels prior to treatment and therefore further support that radiation therapy as sole modality has minimal impact on hematological toxicity especially if relatively small ﬁelds are used like in head and neck cancers.

Treatment-induced anemia in simultaneous radiochemotherapy protocols Combined modality regimens including concurrent chemotherapy during radiation treatment are increasingly used, especially in patients with head and neck, cervical and gastrointestinal cancers. The addition of chemotherapy markedly increases hematological toxicity. The risk of severe anemia remains low and transfusions are rarely required. However, there seems to be a marked increase in mild to moderate anemia. Data from Harrison et al. (2002) demonstrate that the prevalence of anemia increases signiﬁcantly during radiotherapy in the majority of patients (Table 2). In tumor sites with generally large radiation ﬁelds (e.g. pelvic cancers, lung cancers) or frequent use of concomitant chemotherapy (e.g. head and neck cancers), about 20% to 50% of the primarily non-anemic patients developed anemia during radiotherapy. In contrast, the frequency of anemia remains unaffected by radiotherapy in patients treated with limited radiation ﬁelds alone (e.g. in breast cancer). High frequencies of anemia have been reported in radiochemotherapy protocols with aggressive hematotoxic chemotherapy regimens (Wagner et al. 2000). Figure 1 shows the changes in Hb in 18 patients treated with

17

Hb at the end of RCT

16 15 14 13 12 11 10 9 8 8

9

10

11

12

13

14

15

16

17

Hb prior to RCT Fig. 1. Correlation between Hb levels prior to and at the end of therapy in patients treated with simultaneous radiochemotherapy for head and neck cancers. Technical University of Munich

254

J. Dunst and M. Molls

simultaneous radiochemotherapy for head and neck cancers in our institutions in a prospective protocol. The overall treatment lasted ﬁve weeks and chemotherapy was administered in weeks 1 and 4. The Hb levels decreased signiﬁcantly over the course of treatment and were already signiﬁcantly lower after the ﬁrst week. 15/18 (81%) experienced at least mild hematological toxicity with a nadir Hb < 12 g/dl. In this protocol, a signiﬁcant correlation between pretreatment Hb and nadir Hb was observed (Fig. 1). However, the most pronounced decrease in Hb was observed in patients with high pretreatment Hb values. The average decrease of Hb was 2.6 g/dl which was much higher than expected with radiotherapy alone.

Prognostic impact of anemia Pretreatment anemia and outcome after radiotherapy Anemia is a well-known prognostic factor for the outcome in patients undergoing deﬁnitive radiotherapy. About 35 years ago, Evans and Bergsjö (1965) demonstrated a highly signiﬁcant impact of the pretreatment Hb level on outcome in patients with cervical cancers. They analyzed the impact of Hb in 1295 patients treated in the Norwegian Radium Hospital in the period from 1940 through 1945 (N = 895) and in the period from 1956 through 1958 (N = 490). The 10-year overall survival ﬁgures for patients from the period 1940–1945 with a high (≥11 g/dl) versus low (<11 g/dl) pretreatment Hb were 52% versus 50% for 208 patients with stage I (not signiﬁcant), 43% versus 27% for 337 patients with stage II (signiﬁcant) and 28% versus 13% for 260 patients with stage III. They could further demonstrate a linear relationship between Hb levels and survival if the Hb dropped below a certain threshold. This threshold seemed to be dependent on stage (Fig. 2). In early stage I, a decrease in survival was noted if the pretreatment Hb was <10 g/dl. This critical threshold was about 11 g/dl in stage II and 12 g/dl in stage III tumors (Fig. 2). Furthermore, a linear relationship between Hb levels and local control has been reported by Overgaard (1988). He found a decrease in local control in laryngeal cancers if the Hb was <8 mmol/l (corresponding to 12 g/dl). This linear relationship below a certain threshold (which may be dependent of site and stage) strongly supports the hypothesis that there is a causal relation between Hb and response to radiotherapy. The prognostic impact of pretreatment Hb levels has been emphasized in numerous clinical investigations. A recent summary of the data in the radiotherapeutic literature (Grau and Overgaard 1998) revealed that the overwhelming number of studies looking on Hb levels and treatment outcome found a signiﬁcant impact of pretreatment Hb on local control and/or survival (Table 3). It should be noted that most of these data cover speciﬁc tumor sites where deﬁnitive radiotherapy has been considered as

Incidence and impact of anemia in radiation oncology

255

5-year survival 100% 80% Stage I 60% Stage II

40% 20%

Stage III 0%

≥100% 90-99 80-89 70-79 60-69 <60 Rel. Hb concentration (100% = 13.8g/dl)

Fig. 2. Impact of pretreatment Hb levels on 5-year survival in patients treated with deﬁnitive radiotherapy for cervical cancers. Data from the Norwegian Radium Hospital from the period 1940 through 1945. There seems to be a linear decrease of survival with decreasing Hb below a certain threshold. The threshold seemed to differ by stage. Modiﬁed from Evans and Bergsjø 1965

Table 3. Literature survey on the impact of pretreatment Hb levels on radiotherapeutic results (modiﬁed from Grau and Overgaard 1998). The vast majority of studies demonstrate a prognostic impact of hemoglobin levels on local control and/or survival. A smaller number of studies with smaller sample sizes did not found this association Tumor site

Number of positive studies/ Number of all studies

Number of patients in positive studies / number of all patients

Cervix Head and neck Bladder Prostate Total

18 / 22 11 / 17 6/7 4/5 39 / 51

6,946 / 7,458 4,646 / 6,213 1,781 / 1,846 1,109 / 1,174 14,482 / 16,691

standard treatment for cure, mainly head and neck, cervical and lung cancers. Some of the ancient studies have recently been criticized because of their retrospective nature with mostly univariate analyses of the data. However, there is a growing number of prospective studies with multivariate analysis of risk factors supporting an independent prognostic impact of pretreatment

256

J. Dunst and M. Molls

anemia in various cancer sites. In head and neck cancers, Lee et al. (1998) analyzed prognostic factors in 521 patients treated within the prospective randomized study RTOG 85-27. Low Hb levels (<14.5 g/dl for men and <13 g/dl for women) were associated with increased local failure rates (68% in anemic patients versus 52% in normemic patients, p = 0.0002) and reduced 5-year overall survival (22% versus 35%, p = 0.0016). Rudat et al. (1999) analyzed prognostic factors in 68 patients with locally advanced head and neck cancers treated in a prospective protocol with accelareted radiotherapy (with concomitant boost) plus simultaneous chemotherapy. Hemoglobin levels were independently associated with poorer overall survival and poorer local control. A large retrospective analysis of the University of Freiburg (Frommhold et al. 1998) demonstrated that anemia is the most important single prognostic factor for local control and survival. The prognostic impact was present for patients treated with deﬁnitive radiotherapy as well as in patients with postoperative radiotherapy. Fein et al. (1995) as well as Canaday et al. (1999) did also demonstrate the prognostic impact of Hb levels in patients with early glottic cancer. Patients with a normal Hb level prior to treatment (≥13 g/dl) had a signiﬁcantly better overall survival as compared to patients with pretreatment anemia (78% vs. 68%); in a multivariate analysis, Hb and age were the only prognostic factors. In a recent analysis of our institution, pretreatment Hb level was the most important single prognostic factor in cervical cancers treated with radiotherapy or radiochemotherapy. Hemoglobin was more predictive for cure than FIGO-stage in a multivariate analysis (Dunst et al. 2003). The signiﬁcantly lower 3-year overall survival in anemic patients (pretreatment Hb >11 g/dl) as compared to normemic patients was due to a higher frequency of local failures. There was no signiﬁcant difference with regard to systemic failures although anemic patients showed a trend towards a higher frequency of distant metastases (Table 4). According to these ﬁndings, Hb levels seem to be especially important for local control by radiotherapy. These data conﬁrm ﬁndings from Kapp et al. (1983) and Pedersen et al. (1995). In patients with bladder cancer, several studies have demonstrated a correlation between Hb levels and poor outcome after radiotherapy (Quilty and

Table 4. Impact of pretreatment anemia on local control, distant metastases rate and survival in 87 patients with locally advanced cervical cancers

Local failure rate Distant metastases 3-year overall survival

Pretreatment hemoglobin < 11 g/dl

Pretreatment hemoglobin > 11 g/dl

38% 31% 31% ± 13%

20% 23% 64% ± 6%

p < 0.01 n.s. p < 0.01

Incidence and impact of anemia in radiation oncology

257

Duncan 1986; Grau and Overgaard 1998). Moreover, a simple prognostic index has been proposed which allows an estimation of outcome after radiotherapy more accurately than the clinical T-category; this index includes the pretreatment Hb level (Hannisdal et al. 1993). In lung cancer, low Hb levels have also been found to be associated with poorer local control and survival in radiotherapy patients (Oehler et al. 1990; Werner-Wasik et al. 2000).

What is more important: pretreatment hemoglobin or hemoglobin at the end of treatment? Out of the variety of prospective and retrospective studies which have investigated the impact of anemia on outcome most of the data refer to pretreatment Hb levels. This can be explained by the fact that information on pretreatment factors is easily available even in retrospective analyses because a laboratory work-up is part of most staging procedures. There is less information on a possible impact of Hb levels during treatment. The current data, however, suggest that Hb levels during a course of fractionated radiotherapy or even Hb levels at the end of treatment may be at least as important or more crucial than pretreatment Hb levels. There are four recent studies in which both parameters (Hb levels prior to and during/ at the end of radiotherapy) have been analyzed. All four studies found a signiﬁcant impact of Hb levels during therapy. The Hb levels prior to treatment had either no or a lower impact. Van Acht and coworkers could demonstrate that Hb at the end of treatment had prognostic impact in patients with laryngeal cancers. Tarnawski et al. (1997) analyzed prognostic factors in patients with head and neck cancers. They found that the Hb levels at the end of radiotherapy in their analyses were the most powerful prognostic factor. The largest investigation has recently been published by Grogan et al. (1999). They analyzed 450 patients with cervical cancers treated in 5 major Canadian institutions with deﬁnitive radiotherapy for locally advanced cervical cancers. They distinguished in their analyses between four subgroups depending on the Hb levels prior to treatment (low versus high) and during or at the end of treatment (also low versus high). Patients with high Hb levels at the end of treatment had a better prognosis than patients with low Hb levels, irrespective whether or not they had presented with high (N = 228) or low pretreatment Hb levels (N = 25). In contrast, a low Hb at the end of treatment was present in 222 patients and was associated with a poorer survival, irrespective of whether or not the patients had had a high (N = 82) or low (N = 140) pretreatment Hb level. The average Hb level during radiotherapy was also a signiﬁcant prognostic factor. These data suggest that the Hb during radiotherapy or at the end of radiotherapy may have an independent prognostic impact and may be more important than the pretreatment Hb levels.

258

J. Dunst and M. Molls

The prognostic impact of pretreatment Hb levels would then be best explained by the correlation of Hb levels before and after treatment. The data are a strong argument in favor of correcting low Hb levels during treatment. The only randomized study addressing this question has demonstrated a small, but signiﬁcant impact on local control (Bush 1986). However, the design of this study has been criticized and the sample size was too small for to draw conclusions with regard to overall survival (Fyles et al. 2000). An impact of nadir Hb levels during radiation therapy for local control and survival in cervical cancers was also found by Logsdon and Eifel (1999). Another recent investigation has also found a signiﬁcant impact of the nadir Hb but not pretreatment Hb on the response on locally advanced head and neck cancers treated with preoperative radiochemotherapy. Wagner et al. (2000) treated 43 patients with an aggressive regimen of preoperative radiochemotherapy in a phase-II-study. None of the patients had anemia prior to the start of treatment because only patients ﬁt for surgery and ﬁt for radiochemotherapy were included in the protocol. The radiochemotherapy regimen which included ifosfamide and cisplatin produced profound anemia in a high proportion of patients at the end of treatment and the Hb level at the end of therapy was an independent and signiﬁcant prognostic factor for local control and survival, together with tumor size and histological response to radiochemotherapy. The hypothesis that the Hb level at the end of treatment represents the most critical parameter is in accordance with the prognostic impact of pretreatment Hb levels. In patients treated with radiotherapy alone, hematological toxicity is low and there is a strong correlation between Hb levels prior to and at the end of radiotherapy.

Explanations for the prognostic impact of anemia Anemia probably may impact on the course of disease and response to therapy in malignant cancers in different ways. In patients treated with deﬁnitive radiotherapy, a possible and likely explanation arises from the impact of anemia on tumor oxygenation. The efﬁcacy of radiotherapy decreases markedly if the irradiated tissue is hypoxic. Hypoxia has been demonstrated in a variety of tumors and seems to represent a general pathophysiological phenomenon. About 30% of the investigated tumors, irrespective of tumor type, site, histology or stage can be considered as hypoxic according to radiobiological deﬁnitions. The presence of tumor hypoxia has a marked and independent impact on prognosis. Recent clinical investigations have clearly demonstrated that low Hb levels decrease tumor oxygenation (Becker et al. 2000; Vaupel et al. 2002; Nordsmark et al. 2005). Becker et al. (2000) demonstrated that normal tissue oxygenation is independent from Hb levels whereas a Hb <11 g/dl was the strongest predictor of poor tumor oxygenation in head and neck cancers (Fig. 3). This aspect is discussed in further detail in another chapter of this book.

Tissue pO2 (mm Hg)

Incidence and impact of anemia in radiation oncology

259

Normal tissue (M. sternocleidom.)

50

* Tumor * = p<0.0001 0

Moderate/severe anemia (Hb <11 g/dL)

Mild anemia Hb f: 11.0–11.9 g/dL m: 11.0–12.9 g/dL

Normal Hb level

Fig. 3. Impact of Hb on tumor oxygenation (data from Becker et al. 2000). Hemoglobin levels >9 g/dl do not impact on normal tissue oxygenation due to compensatory mechanisms. However, low Hb levels <11 g/dl) have a signiﬁcant impact on tumor oxygenation

pVEGF (pg/ml) 80

40

0 10.0

11.0

12.0

13.0

14.0 Hb (g/dl)

Fig. 4. Increased serum levels of VEGF in anemic tumor patients. 89 patients with untreated head and neck and cervical cancers. Dunst et al. 1999. Dotted line = upper normal range

In a multivariate analysis, Hb level was, besides hypoxic tumor volume, the strongest predictor of failure and death in patients with advanced head and neck cancers (Stadler et al. 1999). The prognostic impact of anemia, therefore, probably results from an indirect effect on radiosensitivity via decreased tumor oxygenation. Further explanations include an independent action of anemia on the progression of malignant diseases. In an analysis of our institutions, anemic patients had different patterns of cytokine expression (Fig. 4). They were

260

J. Dunst and M. Molls

Intratumoral factors

Intratumoral hypoxia radioresistance

genomic instabiliy

Clonal selection

Anemia

angiogenesis

Malignant progression, metastases, radio/chemo resistance

Fig. 5. Hypothesis for the impact of anemia on tumor biology and response to therapy

found to have signiﬁcantly higher systemic levels of vascular endothelial growth factor (VEGF), the most important angiogenic cytokine (Dunst et al. 1999, 2002). This ﬁnding could explain the prognostic impact of anemia in patients with hematological malignancies or surgery as sole treatment (Hasenclever and Diehl 1998; Lutterbach and Guttenberger 2001). In summary, anemia seems to be a major cofactor for the development of an aggressive tumor biology and resistance to radio- and chemotherapy via its impact on tumor hypoxia (Fig. 5).

References 1. Ahmed RS, Kim RY, Duan J, Meleth S, De Los Santos JF, Fiveash JB (2004) IMRT dose escalation for positive paraaortic lymph nodes in patients with locally advanced cervical cancer while reducing dose to bone marrow and other organs at risk. Int J Radiat Oncol Biol Phys 60: 505–512 2. Becker A, Stadler P, Lavey R, Haensgen G, Kuhnt T, Lautenschlaeger C, Feldmann HJ, Molls M, Dunst J (2000) Severe anemia is associated with poor tumor oxygenation in head and neck squamous cell carcinoma. Int J Radiat Oncol Biol Phys 46 3. Bush RS (1986) The signiﬁcance of anemia in clinical radiation therapy. Int J Radiat Oncol Biol Phys 12: 2047–2050 4. Canaday DJ, Regine WF, Mohiuddin M, Zollinger W, Machtay M, Schultz D, Rudoltz MS (1999) Signiﬁcance of pretreatment hemoglobin levels in patients with T1 glottic cancer. Radiat Oncol Investig 7: 42–48 5. Dunst J, Hintner A, Haensgen G, Becker A (1999) Low hemoglobin is associated with increased serum levels of vascular endothelial growth factor (VEGF) in

Incidence and impact of anemia in radiation oncology

6.

7.

8. 9.

10.

11.

12.

13.

14.

15. 16.

17.

18.

19.

261

cancer patients. Does anemia stimulate angiogenesis? Strahlenther Onkol 175: 93–96 Dunst J, Becker A, Lautenschlaeger C, Markau S, Becker H, Fischer K, Haensgen G (2002) Anemia and elevated systemic levels of vascular endothelial growth factor. Strahlenther Onkol 178: 436–441 Dunst J, Kuhnt T, Pelz T, Strauss HG, Koelbl H, Haensgen G (2003) Anemia in cervical cancers: patterns of relapse and association with tumor hypoxia and angiogenesis. Int J Radiat Oncol Biol Phys 56: 778–787 Evans IC, Bergsjø P (1965) The inﬂuence of anemia on the results of radiotherapy in carcinoma of the cervix. Radiology 84: 709–717 Fein DA, Lee WR, Hanlon AL, Ridge JA, Langer CJ, Curran WJ Jr, Coia LR (1995) Pretreatment hemoglobin level inﬂuences local control and survival of T1-T2 squamous cell carcinomas of the glottic larynx. J Clin Oncol 13: 2077– 2083 Frommhold H, Guttenberger R, Henke M (1998) The impact of blood hemoglobin content on the outcome of radiotherapy. The Freiburg experience. Strahlenther Onkol 174 [Suppl IV]: 31–34 Fyles AW, Milosevic M, Pintilie M, Syed A, Hill RP (2000) Anemia, hypoxia and transfusion in patients with cervical cancer: a review. Radiother Oncol 57: 13–19 Girinksi T, Pejovic-Lenfant MH, Bourhis J, Campana F, Cosset JM, Petit C, Malaise EP, Haie C, Gerbaulet A, Chassagne D (1989) Prognostic value of hemoglobin concentrations and blood transfusions in advanced carcinoma of the cervix treated by radiation therapy: results of a retrospective study of 386 patients. Int J Radiat Oncol Biol Phys 16: 37–42 Grau C, Overgaard J (1998) Signiﬁcance of hemoglobin concentrations for treatment outcome. In: Molls M, Vaupel P (eds) Blood perfusion and microenvironment of human tumors. Implications for clinical radiooncology. Springer, Berlin, pp 101–112 Grogan M, Thomas GM, Melamed I, Wong FL (1999) The importance of maintaining high hemoglobin levels during radiation treatment of carcinoma of the cervix. Cancer 86: 1531–1536 Hannisdal E, Fossa SD, Host H (1993) Blood tests and prognosis in bladder carcinomas treated with deﬁnitive radiotherapy. Radiother Oncol 27: 117–122 Harrison L, Shasha D, Homel P (2002) Prevalence of anemia in cancer patients undergoing radiotherapy: prognostic signiﬁcance and treatment. Oncology 63 [Suppl 2]: 11–18 Hasenclever D, Diehl V (1998) A prognostic score for advanced Hodgkin’s disease: International prognostic factors project on advanced Hodgkin’s disease. N Engl J Med 339: 1506–1514 Kapp DS, Fisher D, Gutierrez E, Kohorn IE, Schwartz PE (1983) Pretreatment prognostic factors in carcinoma of the uterine cervix: a multivariate analysis of the effect of age, stage, histology, and blood counts on survival. Int J Radiat Oncol Biol Phys 9: 445–455 Lee WR, Berkey B, Marcial V, Fu KK, Cooper JS, Vikram B, Coia L, Rotman M, Ortiz H (1998) Anemia is associated with decreased survival and increased locoregional failure in patients with locally advanced head and neck carcinoma: a secondary analysis of RTOG 85-27. Int J Radiat Oncol Biol Phys 42: 1069–1075

262

J. Dunst and M. Molls

20. Lutterbach J, Guttenberger R (2001) Anemia is associated with decreased local control of surgically treated squamous cell carcinomas of the glottic larynx. Int J Radiat Oncol Biol Phys 48: 1345–1350 21. Logsdon MD, Eifel PJ (1999) FIGO IIIB squamous cell carcinoma of the cervix: an analysis of prognostic factors emphasizing the balance between external beam and intracavitary radiation therapy. Int J Radiat Oncol Biol Phys 43: 763–775 22. Miller CB, Jones RJ, Piantadosi S, Abeloff MD, Spivak JL (1990) Decreased erythropietin response in patients with the anemia of cancer. N Engl J Med 322: 1689–1692 23. Nordsmark M, Bentzen SM, Rudat V, Brizel D, Lartigau E, Stadler P, Becker A, Adam M, Molls M, Dunst J, Terris DJ, Overgaard J (2005) Prognostic value of tumor oxygenation in 397 head and neck tumors after primary radiation therapy. An international multicenter study. Radiother Oncol 77: 18–24 24. Nowrousian MR (2007) Pathophysiology of anemia in cancer. In: Nowrousian MR (ed) Recombinant human erythropoietin (rhEPO) in clinical oncology, 2nd edn. Springer, Wien New York, 149–188 25. Oehler W, Fischer J, Merkle K (1990) Beeinﬂusst der initiale Hämoglobinwert die Primärtumorreaktion ? Eine Untersuchung an 264 bestrahlten Bronchialkarzinomen. Radiobiol Radiother 31: 325–331 26. Overgaard J (1988) The inﬂuence of hemoglobin concentrations on the response to radiotherapy. Scand J Clin Lab Invest 48: 49–53 27. Pedersen D, Sogaard H, Overgaard J, Bentzen SM (1995) Prognostic value of pretreatment factors in patients with locally advanced carcinoma of the uterine cervix treated by radiotherapy alone. Acta Oncol 34: 787–795 28. Quilty PM, Duncan W (1986) The inﬂuence of the hemoglobin level on the regression and long term local control of transitional cell carcinoma of the bladder following photon irradiation. Int J Radiat Oncol Biol Phys 12: 1735–1742 29. Rudat V, Dietz A, Schramm O, Conradt C, Maier H, Flentje M, Wannenmacher M (1999) Prognostic impact of total tumor volume and hemoglobin concentration on the outcome of patients with advanced nead and neck cancer after concomitant boost radiochemotherapy. Radiother Oncol 53: 119–125 30. Stadler P, Becker A, Feldmann HJ, Dunst J, Molls M (1999) Inﬂuence of the hypoxic subvolume on the survival of patients with head and neck cancer. Int J Radiat Oncol Biol Phys 44: 749–754 31. Tarnawski R, Skladowski K, Maciejewski B (1997) Prognostic value of hemoglobin concentration in radiotherapy for cancer of the supraglottic larynx. Int J Radiat Oncol Biol Phys 38: 1007–1011 32. Thomlinson RH, Gray LH (1955) The histological structure of some human lung cancers and the possible implications for radiotherapy. Br J Cancer 9: 539–549 33. Van Acht MJ, Hermans J, Boks DE, Leer JW (1992) The prognostic value of hemoglobin and a decrease in hemoglobin during radiotherapy in laryngeal carcinoma. Radiother Oncol 23: 229–235 34. Vaupel P, Thews O, Mayer A, et al (2002) Oxygenation status of gynecological tumors: what is the optimal hemoglobin level? Strahlenther Onkol 35. Wagner W, Hermann R, Hartlapp J, Esser E, Christoph B, Müller M, Krech R, Koch O (2000) Prognostic value of hemoglobin concentrations in patients with advanced head and neck cancer treated with combined radio-chemotherapy and surgery. Strahlenther Onkol 176: 73–80

Incidence and impact of anemia in radiation oncology

263

36. Werner-Wasik M, Scott C, Cox JD, Sause WT, Byhardt RW, Asbell Russell A, Komaki R, Lee JS (2000) Recursive partitioning analysis of 1999 Radiation Therapy Oncology Group (RTOG) patients with locally-advanced non-small-cell lung cancer (LA-NSCLC): identiﬁcation of ﬁve groups with different survival. Int J Radiat Oncol Biol Phys 48: 1475–1482 Correspondence: Prof. Dr. Jürgen Dunst, Dept. of Radiation Oncology, University Clinic Schleswig-Holstein, Campus Lübeck, Ratzeburger Allee 160, 23538 Lübeck, Germany, E-mail: [email protected]

Chapter 10

Relationship between hemoglobin levels and tumor oxygenation* P. Vaupel 1, A. Mayer 1 and M. Höckel 2 1

2

Institute of Physiology and Pathophysiology, University of Mainz, Mainz, Germany Department of Obstetrics and Gynecology, University of Leipzig, Leipzig, Germany

Introduction Tissue hypoxia is a powerful and independent adverse prognostic factor in solid tumors. As outlined in Chapter 11, one major factor causing tumor hypoxia is a decreased O2 transport capacity of the blood resulting from tumor-associated and/or therapy-induced anemia, which is a frequent complication seen in cancer patients (Ludwig et al. 2004; Birgegard et al. 2005). In this chapter, current information compiled from experimental and clinical studies is presented which illustrates the relationship between tumor oxygenation and hemoglobin (Hb) levels. Additionally, the relevance of anemia and tumor hypoxia as negative prognostic factors is brieﬂy outlined.

Tumor hypoxia and anemia as adverse prognostic factors An adverse prognostic impact of tumor hypoxia in various tumor entities – including cancers of the uterine cervix, head and neck, and soft tissue sarcoma – has been repeatedly demonstrated (for reviews see Evans and Koch 2003; Vaupel and Mayer 2005). In cervical carcinomas, this impact on prognosis was found to be independent of treatment modality, being evident even in cases treated with surgery alone (Höckel et al. 1996). This ﬁnding speaks strongly in favor of a hypoxia-induced enhancement of tumor aggressiveness resulting in malignant progression. The role of hypoxia as a tumor promoter is indisputable (Spivak 2005). Even so, the possibility that the afore-mentioned causality may be reversed * This chapter is adapted from a review article published in Strahlenther. Onkol. 182: 63–71 (2006).

266

P. Vaupel et al.

also deserves consideration. If this were the case, then the inherently most malignant tumors would necessarily also be the most hypoxic (Coleman et al. 2002; Okunieff et al. 2003). On closer inspection, these two possibilities may not necessarily contradict each other, but may rather even be complementary. If hypoxia does indeed promote the malignant phenotype, then the most (inherently) malignant tumor cells would in turn be capable of generating a hypoxic environment. A combination of both scenarios would mean that hypoxia is the cause of increased aggressiveness since it promotes tumor progression, while at the same time being a consequence of aggressive malignant growth that leads to defective (“chaotic”) vascular morphology and function together with other alterations in the nonmalignant part of the tumor, thereby creating an environment which is adjusted to the pathophysiological demands of the tumor. Multivariate analyses have shown that hypoxia is a powerful prognostic factor in locally advanced cancers of the uterine cervix (Höckel et al. 1993, 1996, 1998; Fyles et al. 1998; Knocke et al. 1999; Sundfør et al. 2000; Höckel and Vaupel 2001a,b), in squamous cell carcinoma of the head and neck (Nordsmark et al. 1996, 2005; Brizel et al. 1997; Stadler et al. 1999; Nordsmark and Overgaard 2000, 2004) and in soft tissue sarcomas (Brizel et al. 1996; Nordsmark et al. 2001). This parameter, being independent of other prognostic factors which can be pretherapeutically assessed, such as clinical tumor size or stage may therefore become clinically useful (Pitson and Fyles 2001). The anemia prognostic signiﬁcance in patients with solid tumors has been documented in a series of clinical studies (for reviews see Caro et al. 2001; Harrison et al. 2002a,b; Nowrousian 2002; van Belle and Cocquyt 2003; Clarke and Pallister 2005; Österborg 2005). Some of these investigations suggest that Hb level during and at the end of radiation therapy are of prognostic signiﬁcance with respect to tumor recurrence and survival. Hb levels were signiﬁcant prognostic factors even after adjustment for other prognostic parameters such as tumor stage and histology (Tarnawski et al. 1997; Grogan et al. 1999). The mechanism(s) by which treatment efﬁcacy and survival are compromised by anemia are not fully understood, but may include an intensiﬁcation of tumor hypoxia, a more general compromise of the patients’ well-being (Dunst 2001; Littlewood 2001), poorer transport kinetics of small cytotoxic drugs (i.e. drug delivery) in oligocythemic states (Highley and de Bruin 1996; Schrijvers et al. 1999; Wildiers et al. 2002; Shannon et al. 2005) or remodeling of tumor microvessels (Tovari et al. 2005). Whether hemoglobin levels at the start of therapy (or at presentation, Rudat et al. 1999; Dietz et al. 2000; Lutterbach and Guttenberger 2000; Glaser et al. 2001; Hänsgen et al. 2001; Henke et al. 2004), at the nadir during therapy (Thomas 2001; Harrison et al. 2002b), at the peak during therapy (Lally et al. 2004), or at the end of (radio-) therapy (Wagner et al. 2000; Thomas 2001) are of prognostic value

Hemoglobin level and tumor oxygenation

267

in terms of better disease-free and overall survival is still being assessed in ongoing studies.

Relationship between hemoglobin level and tumor oxygenation: animal studies The results of several preclinical studies unequivocally showed tumor hypoxia in anemic rodents to be more pronounced than in nonanemic animals. Furthermore, there is clear evidence that correction of anemia can lead to an improvement in tumor oxygenation. The ﬁrst report describing the direct measurement of the oxygenation status in anemic vs. nonanemic animals was communicated by Lavey and McBride (1993). In this study (summarized later by Lavey 1998), a strong correlation between hematocrit and oxygenation within FSa murine ﬁbrosarcomas was presented. In a subsequent study by Terris and Minchinton (1994), a progressive drop of hematocrit values from 42% to 23% was associated with a decrease from 20 mmHg to 10 mmHg for mean pO2 values in murine squamous cell carcinomas. Kelleher et al. (1995, 1996, 1998, 1999) investigated the oxygenation status in a tumor-associated or chemotherapy-induced anemia model of the rat using O2 microsensors. In these studies, decreases in the hemoglobin concentrations of approx. 30% were induced. This moderate anemia resulted in a worsening of tumor oxygenation as reﬂected in a pronounced decrease in the median O2 partial pressure (median pO2 value) from 13 mmHg to approx. 1 mmHg (see Fig. 1) and a signiﬁcant increase in the size of the hypoxic fraction of pO2 values ≤2.5 mmHg (HF 2.5) from 21% to 76%. Worsening of the oxygenation status was also observed at Hb levels >16 g/dL. Kelleher et al. (1995) also investigated the effect of erythropoietin (rhEPO) and transfusions of fresh donor blood on tumor oxygenation in the preclinical setting. Both the administration of rhEPO over 14 days or an acute transfusion with red blood cells increased Hb levels in anemic rats. This rise was associated with a signiﬁcant improvement of the tumor oxygenation status in small tumors (<1.4 ml), although a full recovery of the oxygenation to levels found in nonanemic animals could not be achieved. In tumor-bearing mice, correction of anemia with darbepoietin, a longacting analogue of EPO, also improved tumor oxygenation as demonstrated using the exogenous hypoxia marker EF5 (Ning et al. 2005) or pimonidazole (Shannon et al. 2005). The question of whether a correction of anemia is associated with an increased sensitivity to radiation and O2-dependent chemotherapy was investigated in rodent models by Thews et al. (1998, 2001), Silver and Piver (1999), Stüben et al. (2001, 2003a,b), Pinel et al. (2004), Shannon et al. (2005), and Ning et al. (2005). All studies found that anemia correction (most probably

268

P. Vaupel et al.

Fig. 1. Association between Hb levels (cHb; 1 g/dL = 0.6206 mmol/L) and median pO2 values in experimental rat sarcomas (Kelleher et al. 1996, 1998), squamous cell carcinomas of the head and neck (Becker et al. 2000), of cervical and vulvar cancers (Vaupel et al. 2002b), and cancers of the breast (Vaupel et al. 2003). Values are means ± SEM. The line indicates the quadratic regression (2p = 0.002)

via improving tumor oxygenation) can play a pivotal role in increasing the therapeutic efﬁcacy of irradiation and O2-dependent chemotherapy. A study designed to examine the impact of anemia on the antitumor efﬁcacy of O2dependent photodynamic therapy (PDT) in a murine tumor model also showed that anemia can negatively inﬂuence the therapeutic effectiveness. Correspondingly, anemia correction could restore the antitumor effects of PDT (Golab et al. 2002).

Relationship between hemoglobin level and tumor oxygenation: clinical observations Breast cancer In the clinical setting, a direct correlation has been found between Hb levels and median intratumor pO2 values in breast cancer patients (Vaupel et al. 2002a, 2003). Median pO2 values correlated positively over a Hb range from 8.5 to 14.7 g/dL with an 5-fold increase (3–15 mmHg) in the median pO2 (see

Hemoglobin level and tumor oxygenation

269

Fig. 2. Median pO2 values in breast cancers (lower line) and within the normal breast tissue (upper line) as a function of pretreatment hemoglobin concentration (cHb). Values are means ± SEM; number of patients investigated in brackets. Adapted from Vaupel et al. (2003)

Fig. 2). In contrast, the pO2 values in normal breast tissue were substantially higher (52 mmHg), remaining constant irrespective of the hemoglobin level. This phenomenon indicates a physiological compensation in anemic patients most probably related to an increase in perfusion (and O2 extraction) in normal tissues (Vaupel et al. 2001).

Cancer of the uterine cervix Dunst and Molls (2002) described a linear relationship between hemoglobin levels and tumor oxygenation when hemoglobin concentrations were below the normal physiological range. In 67 cervical cancers, the median pO2 was signiﬁcantly lower in anemic patients as compared to nonanemic women. In a subsequent publication (Dunst et al. 2003), no signiﬁcant correlation between pretreatment Hb levels and pretreatment oxygenation parameters was found. These authors did, however, discover a signiﬁcant correlation between the tumor oxygenation status at approx. 20 Gy and the corresponding Hb levels at that time during treatment. The hypoxic fraction (pO2 <5 mmHg) strongly correlated with the Hb level, and a trend was observed for the median pO2.

270

P. Vaupel et al.

Analyzing data from 51 patients with primary cervical carcinoma, Knocke et al. (1999) were unable to conﬁrm a correlation between initial (pretreatment) hemoglobin levels (range: 8.4–17.0 g/dL) and median pO2 values or hypoxic fractions ≤2.5 mmHg and ≤5 mmHg. In another study involving 80 patients, there was again no correlation between oxygenation and the pretreatment hemoglobin levels (Fyles et al. 1998). Data reported thereafter by Fyles et al. (2000) relating Hb levels to the oxygenation status in cervix cancers showed a more complex relationship. Evaluating data from 91 patients, these authors reported that 7 of 8 patients (≈ 87%) with Hb levels ≤10 g/dL had hypoxic tumors, whereas only 33 of 69 patients (≈ 48%) with Hb levels between 10 and 14 g/dL presented with hypoxic tumors (HF 5 > 50%). Ten of 14 patients (≈ 71%) with hemoglobin concentrations >14 g/dL presented with hypoxic tumors, suggesting that there is an “optimal” Hb range for the oxygenation status of cervix cancers with a worsening below and above this “optimal” range. When the delivery of oxygen to solid tumors as a function of hemoglobin levels is modelled (Jung et al. 1984; Fyles et al. 2000) and the systemic effects of anemia are taken into account, then this relationship can in fact be predicted. The investigators attributed the poor oxygenation status of cervical cancers at higher Hb levels to the effect of carbon monoxide (CO) binding on O2 release from red blood cells in smokers. They concluded that the increased blood viscosity at high Hb concentrations was not associated with hypoxia (Fyles et al. 2000). Given this association between Hb levels and tumor hypoxia (deﬁned as HF 5 > 50%), it is, however, surprising that there is only a modest relationship between Hb levels and oxygenation status when HF 5 values are presented as a function of the initial hemoglobin concentrations. The reduction in O2 delivery at low hemoglobin levels may be higher in patients with sickle trait and intratumoral sickling, which can dramatically increase resistance to ﬂow and further limit the effect of anemia per se on tumor oxygenation (Milosevic et al. 2001). Analysis of the oxygenation status in cervix cancers as a function of pretreatment Hb levels (cHb) was performed by dividing patients into three groups based on “high” [above median cHb; median cHb of age-matched healthy women = 13.95 g/dL (Mertzlufft 1991)], “intermediate” cHb values (12 g/dL < cHb < 13.95 g/dL) and anemic patients [cHb < 12 g/dL (Thomas 1998)]. Using this classiﬁcation, a correlation between Hb levels and median pO2 values as shown in Fig. 3 was obtained (Vaupel et al. 2002b). In anemic patients, the median pO2 was 6 mmHg (cHb = 10.8 ± 0.2 g/dL). In tumors of anemic women, all median pO2 values were <16 mmHg. At a mean Hb level of 13.0 ± 0.1 g/dL, the median pO2 signiﬁcantly increased to 14.5 mmHg and declined thereafter to 6 mmHg in the group with the highest Hb values (cHb = 14.9 ± 0.2 g/dL). Although the O2 transport capacity of the blood must have been increased, in the latter group, no median pO2 values >20 mmHg were noted. The oxygenation status tended to worsen between the “intermediate”

Hemoglobin level and tumor oxygenation

271

Fig. 3. Correlation between pretreatment hemoglobin concentration (cHb) in venous blood and median pO2 values in cancers of the uterine cervix. Values are means ± SEM; numbers of tumors investigated are given in brackets. Adapted from Vaupel et al. (2002b)

and the “high” cHb concentrations as shown in Fig. 3. From this ﬁnding it can be concluded, that an optimal Hb level with regard to the oxygenation status of cancers of the uterine cervix has to be assumed at cHb values of between 12 and 14 g/dL. In this group, median pO2 values ranged from 1 to 44 mmHg. These data suggest that an optimal hematocrit or cHb range exists with regard to the median pO2. The rise in the median pO2 is based on an increase in the transport capacity of the blood with increases in cHb at values up to approx. 14 g/dL. At higher cHb values, this effect is counteracted by a substantial increase in the viscous resistance to ﬂow (i.e. a deterioration of the blood’s rheological properties) caused by a pronounced increase in blood viscosity which is further aggravated by a high vascular permeability (leaky blood vessel walls) which obligatorily leads to a hemoconcentration [9–15% of the plasma ﬂow extravasate during tumor passage (Vaupel 2004)]. The restricted O2 supply at higher cHb values is thus primarily caused by hyperviscosity in tortuous, elongated, dilated and functionally abnormal tumor microvessels, which counteracts and ﬁnally may overweigh the higher O2 transport capacity which might have been expected in this range. This decline in the oxygenation status at high cHb or Hct values is not tumorspeciﬁc. In healthy individuals, the maximum O2 availability is obtained at

272

P. Vaupel et al.

the physiologic Hct and cHb. Above Hct values of 40–45%, the systemic O2 transport capacity substantially declines (Zander 1999). Treatment of anemia can improve the oxygenation status of cervical cancers. In a casuistic observation, transfusion of red blood cells was followed by a rise in the median pO2 and a drastic fall in the hypoxic fraction (HF 2.5 and HF 5; Vaupel 1994). Clinical trials are consistent with this result (Sundfør et al. 1997). However, higher Hb concentrations after blood transfusion resulted in a substantially improved tumor oxygenation in only 50% of the patients. In this study, an increase of the initial Hb level from <9.0 g/dL to >11.5 g/dL after transfusion led to higher 50th percentiles, but unchanged 10th percentiles, suggesting that transfusion would not reduce tumor hypoxia extensively, although this may have been due to pO2 readings being obtained from necrotic tissue. In a review by Fyles et al. (2000), the authors conclude that RBC transfusion or erythropoietin treatment ameliorate the oxygenation in cervix cancers in only a proportion of anemic patients. Cancer of the vulva Using the same cHb grouping as described above, a correlation between Hb levels and median pO2 values in vulvar cancer as shown in Fig. 4 was obtained (Vaupel et al. 2002b). In principle, the pattern found in vulvar cancer was similar to that observed in cervical carcinoma. In anemic patients, the median pO2 in carcinomas of the vulva was 3 mmHg (cHb = 11.3 ± 0.5 g/dL). In this group, all median pO2 values were <9 mmHg. At a mean Hb level of 13.5 ± 0.5 g/dL (representing the “normal” range of cHb), the median pO2 was signiﬁcantly higher (19 mmHg) and declined at higher cHb values (14.8 ± 0.7 g/dL) to a median pO2 of 11 mmHg. In this latter group, no median pO2 values >17 mmHg were seen. From these data, Vaupel et al. (2002b) concluded that an optimal Hb level with regard to the oxygenation status should be expected at cHb values between 12 and 14 g/dL. Over this cHb range, the median pO2 values were between 2 and 58 mmHg. Analysis of the data published by Stone et al. (2005) show that nodenegative patients had higher pretreatment Hb levels (13.9 g/dL) than patients with nodal spread (12.0 g/dL). The higher Hb levels (13.9 g/dL) correlated with a median pO2 of 12 mmHg and a hypoxic fraction HF 5 of 16.5%, whereas the lower Hb concentration was associated with a median pO2 of 5 mmHg and a HF 5 of 52.5%. Head and neck cancer A series of studies on head and neck cancer patients could not substantiate a correlation between hemoglobin levels (or hematocrit) and the oxygena-

Hemoglobin level and tumor oxygenation

273

Fig. 4. Median pO2 values in primary and recurrent cancers of the vulva (lower curve) and within the normal subcutis (mons pubis, broken line) as a function of pretreatment hemoglobin concentration (cHb). Values are means ± SEM; number of patients investigated in brackets. Modiﬁed from Vaupel et al. (2002b) and complemented by data of Stone et al. (2005)

tion status (Terris and Dunphy 1994; Adam et al. 1999; Dietz et al. 2000; Rudat et al. 2000; Nordsmark and Overgaard 2004). In contrast, other investigations clearly suggest that Hb concentrations can impact tumor oxygenation. A signiﬁcant linear correlation was described for the oxygenation of primary tumors (not for metastases) and hemoglobin levels regarding the median pO2 and the hypoxic fraction ≤5 mmHg (Becker et al. 1998a; Molls et al. 1998). Similarly, Stadler et al. (1999) reported a linear correlation between Hb levels and hypoxic fractions (HF 2.5 and HF 5) or hypoxic subvolume. Brizel et al. (1999) also found a (weak) association between higher Hb levels and higher median pO2 values (p = 0.04). For patients with cHb < 13 g/dL, only 12% of the cancers investigated had a median pO2 > 10 mmHg. When cHb at presentation was >13 g/dL, 42% of the tumors exhibited pO2 values >10 mmHg. In another study by Rudat et al. (2001), patients with anemia (cHb ≤11 g/dL) showed a statistically signiﬁcant larger fraction of pO2 values ≤2.5 mmHg (HF 2.5 = 33.9%) compared to patients with mild anemia or normal hemoglobin concentration (HF 2.5 = 22.6%).

274

P. Vaupel et al.

In a further study (Clavo et al. 2003), the percentage of values ≤10 mmHg was 34% in tumors of nonanemic patients compared with 47% in anemic patients, and 69% in the most anemic patients (cHb < 11.5 g/dL). Using linear correlation analysis, a tendency was found for higher median tumor pO2 values to be associated with greater Hb concentrations. Data published by Becker et al. (1998b, 2000) corroborate with our observations on cancers of the uterine cervix (Vaupel et al. 2002b). The former study suggested a “decrease of tumor pO2 not only at low Hb levels but also at the upper end of the Hb scale”, although only a limited number of patients with squamous cell carcinoma of the head and neck showing “high” cHb levels could be assessed. A re-evaluation of the original data of Becker et al. (2000), in which the data were plotted following classiﬁcation of the data values into Hb ranges, indicates that a maximum tumor oxygenation is achieved at cHb values of between 13 and 14 g/dL. This cHb range represents an optimum in terms of the opposing effects of increasing the blood’s oxygen-carrying capacity and rising viscous resistance to ﬂow. A trend towards lower median oxygen partial pressures at hemoglobin concentrations above 15 g/dL is evident (Fig. 5, Vaupel and Mayer 2004; Vaupel et al. 2005). An international multicenter study on 356 head and neck cancers conﬁrmed this nonlinear relationship between Hb concentrations and tumor oxy-

Fig. 5. Correlation between pretreatment hemoglobin concentration (cHb) in venous blood and median pO2 values in cancers of the head and neck. Values are means ± SEM; numbers of tumors investigated are given in brackets. Adapted from Becker et al. (2000)

Hemoglobin level and tumor oxygenation

275

genation in agreement with previous reports on cervical and vulvar cancers (Nordsmark et al. 2005). In this latter study values of HF 2.5 and HF 5 were, however, not signiﬁcantly associated with Hb concentrations. In order to interpret this observation of a maximum tumor oxygenation in the “normal” Hb range with lower pO2 values above and below the physiological range, median pO2 values measured can be correlated with a calculated O2 transport index: O2 transport capacity / viscous resistance to ﬂow [with O2 transport capacity = cHb (g/dL) × Hüfner’s number; Hüfner’s number = 1.39 ml O2/g Hb]. This index clearly indicates that an optimal cHb range (or hematocrit range) exists with regard to the median pO2 (Vaupel et al. 2002b). In human cancers of the head and neck, casuistic observations also support the notion that treatment of anemia might improve tumor oxygenation as shown by pO2 measurements prior to and after transfusion in selected patients. Upon transfusion the median pO2 increased and the fraction of hypoxic readings decreased (Dunst and Molls 2002).

Conclusions Evidence has accumulated showing that up to 50–60% of locally advanced solid tumors may exhibit hypoxic and/or anoxic tissue areas. Tumor-associated or therapy-induced anemia are major pathogenetic factors that can contribute to the development of hypoxia (anemic hypoxia). Whereas in normal tissue this type of hypoxia can (primarily) be compensated by an increase in local blood ﬂow rate, locally advanced tumors (or at least larger tumor areas) cannot adequately counteract the restriction of O2 supply and thus the development of hypoxia, i.e. anemic hypoxia in tumors is a frequent complication in cancer patients. Whereas in some studies no correlation is seen between hemoglobin level and tumor oxygenation status, there is increasing evidence that low Hb levels are indeed associated with a poor tumor oxygenation, and that increasing Hb concentrations are correlated with higher pO2 values or lower hypoxic tissue fraction. This has been shown for the preclinical (experimental) and clinical setting. Clinical observations on cancers of the head and neck, of the uterine cervix and of the vulva are indicative of a nonlinear relationship between Hb level and tumor pO2 values (Fig. 6): maximum tumor oxygenation is observed between 13 and 15 g/dL with a worsening tumor oxygenation in anemia and at Hb levels above the median Hb concentrations of healthy, adult persons, providing gender-speciﬁc differences and age-related variations are taken into account. The model ﬁt of the quadratic regression was highly signiﬁcant with 2p = 0.001. Included in this nonmonotonous relationship were 277 squamous cell carcinomas of the head and neck (Becker et al. 2000), of the uterine cervix and of the vulva (Vaupel et al. 2002b).

276

P. Vaupel et al.

Fig. 6. Relationship between pretreatment hemoglobin levels (cHb) and the median pO2 in 277 patients with squamous cell carcinomas (SCC) of the head and neck (Becker et al. 2000), of the uterine cervix and of the vulva (Vaupel et al. 2002b). The line indicates the quadratic regression (2p = 0.001)

Addendum To allow a better comparison of the data published, the following conversions for Hb levels (cHb) and hematocrit values (Hct) were used: 1 g/dL = 0.6206 mmol Hb/L Hct (%) = 3 × cHb (g/dL)

[for cHb] [for conversion from Hct to cHb].

Acknowledgement We thank Dr. Debra Kelleher for valuable editorial help during preparation of this manuscript.

References 1. Adam MF, Gabalski EC, Bloch DA, Oehlert JW, Brown JM, Elsaid AA, Pinto HA, Terris DJ (1999) Tissue oxygen distribution in head and neck cancer patients. Head & Neck 21: 146–153 2. Becker A, Hänsgen G, Bloching M, Weigel C, Lautenschläger C, Dunst J (1998a) Oxygenation of squamous cell carcinoma of the head and neck: comparison of

Hemoglobin level and tumor oxygenation

3.

4.

5.

6.

7.

8.

9.

10. 11.

12. 13.

14. 15.

16.

17.

277

primary tumors, neck node metastases, and normal tissue. Int J Radiat Oncol Biol Phys 42: 35–41 Becker A, Stadler P, Hänsgen G, Feldmann HJ, Molls M, Dunst J (1998b) Hemoglobin level inﬂuences the oxygenation status of squamous cell carcinoma of the head and neck (SCCHN). Int J Radiat Oncol Biol Phys 42 [Suppl 1]: 233 Becker A, Stadler P, Lavey RS, Hänsgen G, Kuhnt T, Lautenschläger C, Feldmann HJ, Molls M, Dunst J (2000) Severe anemia is associated with poor tumor oxygenation in head and neck squamous cell carcinomas. Int J Radiat Oncol Biol Phys 46: 459–466 Birgegard G, Aapro MS, Bokemeyer C, Dicato M, Drings P, Hornedo J, Krzakowski M, Ludwig H, Pecorelli S, Schmoll H, Schneider M, Schrijvers D, Shasha D, van Belle S (2005) Cancer-related anemia: Pathogenesis, prevalence and treatment. Oncology 68 [Suppl 1]: 3–11 Brizel DM, Scully SP, Harrelson JM, Layﬁeld LJ, Bean JM, Prosnitz LR, Dewhirst MW (1996) Tumor oxygenation predicts for the likelihood of distant metastases in human soft tissue sarcoma. Cancer Res 56: 941–943 Brizel DM, Sibley GS, Prosnitz LR, Scher RL, Dewhirst MW (1997) Tumor hypoxia adversely affects the prognosis of carcinoma of the head and neck. Int J Radiat Oncol Biol Phys 38: 285–289 Brizel DM, Dodge RK, Clough RW, Dewhirst MW (1999) Oxygenation of head and neck cancer: changes during radiotherapy and impact on treatment outcome. Radiother Oncol 53: 113–117 Caro JJ, Salas M, Ward A, Goss G (2001) Anemia as an independent prognostic factor for survival in patients with cancer. A systemic quantitative review. Cancer 91: 2214–2221 Clarke H, Pallister CJ (2005) The impact of anaemia on outcome in cancer. Clin Lab Haem 27: 1–13 Clavo B, Pérez JL, López L, Suárez G, Lloret M, Morera J, Macías D, Martínez JC, Santana M, Hernández MA, Robaina F, Günderoth M (2003) Inﬂuence of haemoglobin concentration and peripheral muscle pO2 on tumour oxygenation in advanced head and neck tumours. Radiother Oncol 66: 71–74 Coleman CN, Mitchell JB, Camphausen K (2002) Tumor hypoxia: chicken, egg, or a piece of the farm? J Clin Oncol 20: 610–615 Dietz A, Rudat V, Conradt C, Vanselow B, Wollensack P, Staar S, Eckel H, Volling P, Schröder M, Wannenmacher M, Müller RP, Weidauer H (2000) Prognostischer Stellenwert des Hämoglobinwertes vor primärer Radiochemotherapie von KopfHals-Karzinomen. HNO 48: 655–664 Dunst J (2001) The use of epoietin alfa to increase and maintain hemoglobin levels during radiotherapy. Semin Oncol 28 [Suppl 8]: 42–48 Dunst J, Molls M (2002) Relationship between anemia and hypoxia. In: Nowrousian MR (ed) Recombinant human erythropoietin (rhEPO) in clinical oncology. Springer, Wien New York, pp 117–125 Dunst J, Kuhnt T, Strauss HG, Krause U, Pelz T, Koelbl H, Haensgen G (2003) Anemia in cervical cancers: Impact on survival, patterns of relapse, and association with hypoxia and angiogenesis. Int J Radiat Oncol Biol Phys 56: 778– 787 Evans SM, Koch CJ (2003) Prognostic signiﬁcance of tumor oxygenation in humans. Cancer Lett 195: 1–16

278

P. Vaupel et al.

18. Fyles AW, Milosevic M, Wong R, Kavanagh M-C, Pintilie M, Sun A, Chapman W, Levin W, Manchul L, Keane TJ, Hill RP (1998) Oxygenation predicts radiation response and survival in patients with cervix cancer. Radiother Oncol 48: 149–156 19. Fyles AW, Milosevic M, Pintilie M, Syed A, Hill RP (2000) Anemia, hypoxia and transfusion in patients with cervix cancer: a review. Radiother Oncol 57: 13–19 20. Glaser C, Millesi W, Kornek GV, Lang S, Schull B, Watzinger F, Selzer E, Lavey RS (2001) Impact of hemoglobin level and use of recombinant erythropoietin on efﬁcacy of preoperative chemoradiation therapy for squamous cell carcinoma of the oral cavity and oropharynx. Int J Radiat Oncol Biol Phys 50: 705–715 21. Golab J, Olszewska D, Mróz P, Kozar K, Kaminski R, Jalili A, Jakóbisiak M (2002) Erythropoietin restores the antitumor effectiveness of photodynamic therapy in mice with chemotherapy-induced anemia. Clin Cancer Res 8: 1265–1270 22. Grogan M, Thomas GM, Melamed I, Wong FLW, Pearcey RG, Joseph PK, Portelance L, Crook J, Jones KD (1999) The importance of hemoglobin levels during radiotherapy for carcinoma of the cervix. Cancer 86: 1528–1536 23. Hänsgen G, Krause U, Becker A, Stadler P, Lautenschlaeger C, Wohlrab W, Rath FW, Molls M, Dunst J (2001) Tumor hypoxia, p53, and prognosis in cervical cancers. Int J Radiat Oncol Biol Phys 50: 865–872 24. Harrison LB, Shasha D, Homel P (2002a) Prevalence of anemia in cancer patients undergoing radiotherapy: Prognostic signiﬁcance and treatment. Oncology 63 [Suppl 2]: 11–18 25. Harrison LB, Chadha M, Hill RJ, Hu K, Shasha D (2002b) Impact of tumor hypoxia and anemia on radiation therapy outcomes. Oncologist 7: 492–508 26. Henke M, Sindlinger F, Ikenberg H, Gerds T, Schumacher M (2004) Blood hemoglobin level and treatment outcome of early breast cancer. Strahlenther Onkol 180: 45–51 27. Highley MS, de Bruin EA (1996) Erythrocytes and the transport of drugs and endogenous compounds. Pharm Res 13: 186–195 28. Höckel M, Vaupel P (2001a) Biological consequences of tumor hypoxia. Semin Oncol 28 [Suppl 8]: 36–41 29. Höckel M, Vaupel P (2001b) Prognostic signiﬁcance of tissue hypoxia in cervical cancer. CME J Gynecol Oncol 6: 216–225 30. Höckel M, Knoop C, Schlenger K, Vorndran B, Baussmann E, Mitze M, Knapstein PG, Vaupel P (1993) Intratumoral pO2 predicts survival in advanced cancer of the uterine cervix. Radiother Oncol 26: 45–50 31. Höckel M, Schlenger K, Aral B, Mitze M, Schäffer U, Vaupel P (1996) Association between tumor hypoxia and malignant progression in advanced cancer of the uterine cervix. Cancer Res 56: 4509–4515 32. Höckel M, Schlenger K, Höckel S, Aral B, Schäffer U, Vaupel P (1998) Tumor hypoxia in pelvic recurrences of cervical cancer. Int J Cancer 79: 365–369 33. Jung C, Müller-Klieser W, Vaupel P (1984) Tumor blood ﬂow and O2 availability during hemodilution. Adv Exp Med Biol 180: 281–291 34. Kelleher DK, Matthiensen U, Thews O, Vaupel P (1995) Tumor oxygenation in anemic rats: Effects of erythropoietin treatment versus red blood cell transfusion. Acta Oncol 34: 379–384

Hemoglobin level and tumor oxygenation

279

35. Kelleher DK, Matthiensen U, Thews O, Vaupel P (1996) Blood ﬂow, oxygenation, and bioenergetic status of tumors following erythropoietin treatment in normal and anemic rats. Cancer Res 56: 4728–4734 36. Kelleher DK, Thews O, Vaupel P (1998) Can erythropoietin improve tumor oxygenation? Strahlenther Onkol 174 [Suppl IV]: 20–23 37. Kelleher DK, Thews O, Vaupel P (1999) Modulation of tumor oxygenation and radiosensitivity by erythropoietin. In: Vaupel P, Kelleher DK (eds) Tumor hypoxia. Pathophysiology, clinical signiﬁcance and therapeutic perspectives. Wissenschaftliche Verlagsgesellschaft, Stuttgart, pp 83–90 38. Knocke TH, Weitmann H-D, Feldmann H-J, Selzer E, Pötter R (1999) Intratumoral pO2-measurements as predictive assay in the treatment of carcinoma of the uterine cervix. Radiother Oncol 53: 99–104 39. Lally BE, Colasanto JM, Fischer JJ, Knisely JPS (2004) Is there an optimal hemoglobin level for patients with glioblastoma multiforme? Cancer J 10: 391– 396 40. Lavey RS (1998) Clinical trial experience using erythropoietin during radiation therapy. Strahlenther Onkol 174: 24–30 41. Lavey RS, McBride WH (1993) Inﬂuence of hematocrit on tumor oxygenation and sensitivity to radiation. Eur J Cancer 29A [Suppl 6]: S216, Abstr 1210 42. Littlewood TJ (2001) The impact of hemoglobin levels on treatment outcomes in patients with cancer. Semin Oncol 28 [Suppl 8]: 49–53 43. Ludwig H, van Belle S, Barrett-Lee P, Birgegard G, Bokemeyer C, Gascon P, Kosmidis P, Krzakowski M, Nortier J, Olmi P, Schneider M, Schrijvers D (2004) The European Cancer Anaemia Survey (ECAS): A large, multinational, prospective survey deﬁning the prevalence, incidence, and treatment of anaemia in cancer patients. Eur J Cancer 40: 2293–2306 44. Lutterbach J, Guttenberger R (2000) Anemia is associated with decreased local control of surgically treated squamous cell carcinomas of the glottic larynx. Int J Radiat Oncol Biol Phys 48: 1345–1350 45. Mertzlufft F (1991) Normal values for hemoglobin concentration. In: Zander R, Mertzlufft F (eds) The oxygenation status of arterial blood. Karger, Basel New York, pp 162–166 46. Milosevic M, Quirt I, Levin W, Fyles A, Manchul L, Chapman W (2001) Intratumoral sickling in a patient with cervix cancer and sickle trait: Effect on blood ﬂow and oxygenation. Gynecol Oncol 83: 428–431 47. Molls M, Stadler P, Becker A, Feldmann HJ, Dunst J (1998) Relevance of oxygen in radiation oncology. Mechanisms of action, correlation to low hemoglobin levels. Strahlenther Onkol 174 [Suppl IV]: 13–16 48. Ning S, Hartley C, Molineux G, Knox SJ (2005) Darbepoietin alfa potentiates the efﬁcacy of radiation therapy in mice with corrected or uncorrected anemia. Cancer Res 65: 284–290 49. Nordsmark M, Overgaard J (2000) A conﬁrmatory prognostic study on oxygenation status and loco-regional control in advanced head and neck squamous cell carcinoma treated by radiation therapy. Radiother Oncol 57: 39–43 50. Nordsmark M, Overgaard J (2004) Tumor hypoxia is independent of hemoglobin and prognostic for loco-regional tumor control after primary radiotherapy in advanced head and neck cancer. Acta Oncol 43: 396–403

280

P. Vaupel et al.

51. Nordsmark M, Overgaard M, Overgaard J (1996) Pretreatment oxygenation predicts radiation response in advanced squamous cell carcinoma of the head and neck. Radiother Oncol 41: 31–39 52. Nordsmark M, Alsner J, Keller J, Nielsen OS, Jensen OM, Horsman MR, Overgaard J (2001) Hypoxia in human soft tissue sarcomas: Adverse impact on survival and no association with p53 mutations. Br J Cancer 84: 1070–1075 53. Nordsmark M, Bentzen SM, Rudat V, Brizel D, Lartigau E, Stadler P, Becker A, Adam M, Molls M, Dunst J, Terris DJ, Overgaard J (2005) Prognostic value of tumor oxygenation in 397 head and neck tumors after primary radiation therapy. An international multi-center study. Radiother Oncol 77: 18–24 54. Nowrousian MR (2002) Prevalence, pathophysiology, predictive factors, and prognostic signiﬁcance of anemia in cancer chemotherapy. In: Nowrousian MR (ed) Recombinant human erythropoietin (rhEPO) in clinical oncology. Springer, Wien New York, pp 63–100 55. Okunieff P, Ding I, Vaupel P, Höckel M (2003) Evidence for and against hypoxia as the primary cause of tumor aggressiveness. Adv Exp Med Biol 510: 69–75 56. Österborg A (2005) Anaemia in patients with cancer: Association to prognosis and prediction of response to erythropoietic agents. In: Bokemeyer C, Ludwig H (eds) Anaemia in Cancer, 2nd edn. Elsevier, Edinburgh London New York, pp 75–89 57. Pinel S, Barberi-Heyob M, Cohen-Jonathan E, Merlin JL, Delmas C, Plenat F, Chastagner P (2004) Erythropoietin-induced reduction of hypoxia before and during fractionated irradiation contributes to improvement of radioresponse in human glioma xenografts. Int J Radiat Oncol Biol Phys 59: 250–259 58. Pitson G, Fyles A (2001) Uterine cervix cancer. In: Gospodarowicz MK, Henson DE, Hutter RVP, et al (eds) Prognostic factors in cancer, 2nd edn. Wiley-Liss, New York, pp 501–513 59. Rudat V, Dietz A, Schramm O, Conradt C, Maier H, Flentje M, Wannenmacher M (1999) Prognostic impact of total tumor volume and hemoglobin concentration on the outcome of patients with advanced head and neck cancer after concomitant boost radiochemotherapy. Radiother Oncol 53: 119–125 60. Rudat V, Vanselow B, Wollensack P, Bettscheider C, Osman-Ahmet S, Eble MJ, Dietz A (2000) Repeatability and prognostic impact of the pretreatment pO2 histography in patients with advanced head and neck cancer. Radiother Oncol 57: 31–37 61. Rudat V, Stadler P, Becker A, Vanselow B, Dietz A, Wannenmacher M, Molls M, Dunst J, Feldmann HJ (2001) Predictive value of the tumor oxygenation by means of pO2 histography in patients with advanced head and neck cancer. Strahlenther Onkol 177: 462–468 62. Schrijvers D, Highley M, de Bruin E, van Oosterom AT, Vermorken JB (1999) Role of red blood cells in pharmacokinetics of chemotherapeutic agents. AntiCancer Drugs 10: 147–153 63. Shannon AM, Bouchier-Hayes DJ, Condron CM, Toomey D (2005) Correction of anaemia through the use of darbepoetin alfa improves chemotherapeutic outcome in a murine model of Lewis lung carcinoma. Br J Cancer 93: 224–232 64. Silver DF, Piver MS (1999) Effects of recombinant human erythropoietin on the antitumor effect of cisplatin in SCID mice bearing human ovarian cancer: a possible oxygen effect. Gynecol Oncol 73: 280–284

Hemoglobin level and tumor oxygenation

281

65. Spivak JL (2005) The anaemia of cancer: Death by a thousand cuts. Nature Rev Cancer 5: 543–555 66. Stadler P, Becker A, Feldmann HJ, Hänsgen G, Dunst J, Würschmidt, Molls M (1999) Inﬂuence of the hypoxic subvolume on the survival of patients with head and neck cancer. Int J Radiat Oncol Biol 44: 749–754 67. Stone JE, Parker R, Gilks CB, Stanbridge EJ, Liao SY, Aquino-Parsons C (2005) Intratumoral oxygenation of invasive squamous cell carcinoma of the vulva is not correlated with regional lymph node metastasis. Eur J Gynaecol Oncol 26: 31–35 68. Stüben G, Thews O, Pöttgen C, Knühmann K, Vaupel P, Stuschke M (2001) Recombinant human erythropoietin increases the radiosensitivity of xenografted human tumours in anaemic nude mice. J Cancer Res Clin Oncol 127: 346–350 69. Stüben G, Pöttgen C, Knühmann K, Schmidt K, Stuschke M, Thews O, Vaupel P (2003a) Erythropoietin restores the anemia-induced reduction in radiosensitivity of experimental human tumors in nude mice. Int J Radiat Oncol Biol Phys 55: 1358–1362 70. Stüben G, Thews O, Pöttgen C, Knühmann K, Sack H, Stuschke M, Vaupel P (2003b) Impact of anemia prevention by recombinant human erythropoietin on the sensitivity of xenografted glioblastomas to fractionated irradiation. Strahlenther Onkol 179: 620–625 71. Sundfør K, Lyng H, Kongsgard UL, Trope C, Rofstad EK (1997) Polarographic measurement of pO2 in cervix carcinoma. Gynecol Oncol 64: 230–236 72. Sundfør K, Lyng H, Trope CG, Rofstad EK (2000) Treatment outcome in advanced squamous cell carcinoma of the uterine cervix: relationships to pretreatment tumor oxygenation and vascularization. Radiother Oncol 54: 101–107 73. Tarnawski R, Skladowski K, Maciejewski B (1997) Prognostic value of hemoglobin concentration in radiotherapy for cancer of supraglottic larynx. Int J Radiat Oncol Biol Phys 38: 1007–1011 74. Terris DJ, Dunphy EP (1994) Oxygen tension measurements of head and neck cancers. Arch Otolaryngol Head Neck Surg 120: 283–287 75. Terris DJ, Minchinton AI (1994) Computerized histographic characterization of changes in tissue pO2 induced by erythropoietin. Adv Exp Med Biol 361: 613–618 76. Thews O, Koenig R, Kelleher DK, Kutzner J, Vaupel P (1998) Enhanced radiosensitivity in experimental tumours following erythropoietin treatment of chemotherapy-induced anaemia. Br J Cancer 78: 752–756 77. Thews O, Kelleher DK, Vaupel P (2001) Erythropoietin restores the anemiainduced reduction in cyclophosphamide cytotoxicity in rat tumors. Cancer Res 61: 1358–1361 78. Thomas L (1998) Hämoglobine. In: Thomas L (Hrsg) Labor und Diagnose, 5. Auﬂ. TH Books Verlagsgesellschaft, Frankfurt, S 487–491 79. Thomas G (2001) The effect of hemoglobin level on radiotherapy outcomes: The Canadian experience. Semin Oncol 28 [Suppl 8]: 60–65 80. Tovari J, Gilly R, Raso E, Paku S, Bereczky B, Varga N, Vago A, Timar J (2005) Recombinant human erythropoietin alpha targets intratumoral blood vessels, improving chemotherapy in human xenograft models. Cancer Res 65: 7186–7193 81. Van Belle S J-P, Cocquyt V (2003) Impact of haemoglobin levels on the outcome of cancers treated with chemotherapy. Crit Rev Oncol/Hematol 47: 1–11 82. Vaupel P (1994) Blood ﬂow, oxygenation, tissue pH distribution and bioenergetic status of tumors. Berlin, Ernst Schering Research Foundation, Lecture 23

282

P. Vaupel et al.: Hemoglobin level and tumor oxygenation

83. Vaupel P (2004) Tumor microenvironmental physiology and its implications for radiation oncology. Semin Radiat Oncol 14: 198–206 84. Vaupel P, Mayer A (2004) Erythropoietin to treat anaemia in patients with head and neck cancer. Lancet 363: 992 85. Vaupel P, Mayer A (2005) Effects of anaemia and hypoxia on tumour biology. In: Bokemeyer C, Ludwig H (eds) Anaemia in cancer, 2nd edn. Elsevier, Edinburgh London New York, pp 47–66 86. Vaupel P, Thews O, Hoeckel M (2001) Treatment resistance of solid tumors. Med Oncol 18: 243–259 87. Vaupel P, Briest S, Höckel M (2002a) Hypoxia in breast cancer: Pathogenesis, characterization and biological/therapeutic implications. Wien Med Wochenschr 152: 334–342 88. Vaupel P, Thews O, Mayer A, Höckel S, Höckel M (2002b) Oxygenation status of gynecologic tumors: What is the optimal hemoglobin level? Strahlenther Onkol 178: 727–731 89. Vaupel P, Mayer A, Briest S, Höckel M (2003) Oxygenation gain factor: A novel parameter characterizing the association between hemoglobin level and the oxygenation status of breast cancers. Cancer Res 63: 7634–7637 90. Vaupel P, Dunst J, Engert A, Fandrey J, Feyer P, Freund M, Jelkmann W (2005) Effects of recombinant human erythropoietin (rHuEPO) on tumor control in patients with cancer-induced anemia. Onkologie 28: 216–221 91. Wagner W, Hermann R, Hartlapp J, Esser E, Christoph B, Müller MK, Krech R, Koch O (2000) Prognostic value of hemoglobin concentrations in patients with advanced head and neck cancer treated with combined radio-chemotherapy and surgery. Strahlenther Onkol 176: 73–80 92. Wildiers H, Guetens G, de Boeck G, Landuyt W, Verbeken E, Highley M, de Bruin EA, Oosterom AT (2002) Melphalan availability in hypoxia-inducible factor-1α+/+ and factor-1α−/− tumors is independent of tumor vessel density and correlates with melphalan erythrocyte transport. Int J Cancer 99: 514–519 93. Zander R (1999) Optimaler Hämatokrit 30%: Abschied von einer Illusion. Infusionsther Transfusionsmed 26: 186–190 Correspondence: Prof. Dr. Peter Vaupel, Institute of Physiology and Pathophysiology, University of Mainz, Duesbergweg 6, 55099 Mainz, Germany, E-mail: [email protected]. Dr. Arnulf Mayer, Institute of Physiology and Pathophysiology, University of Mainz, Duesbergweg 6, 55099 Mainz, Germany, E-mail: [email protected]. Prof. Dr. Dr. Michael Höckel, Department of Obstetrics and Gynecology, University of Leipzig, Philipp-Rosenthal-Strasse 55, 04103 Leipzig, Germany, E-mail: [email protected].

Chapter 11

Tumor hypoxia and therapeutic resistance* P. Vaupel 1 and M. Höckel 2 1 2

Institute of Physiology and Pathophysiology, University of Mainz, Mainz, Germany Department of Obstetrics and Gynecology, University of Leipzig, Leipzig, Germany

Introduction For many years, the identiﬁcation of tumor hypoxia, its systematic characterization and the assessment of its clinical relevance were not possible due to the lack of methods suitable for the routine measurement of intratumoral oxygen tensions in patients. In the late 1980s, a novel and clinically applicable standardized procedure was established enabling the determination of tumor oxygenation in accessible primary tumors, local recurrences, and metastatic lesions in patients using a computerized polarographic needle electrode system (Höckel et al. 1991; Vaupel et al. 1991). Within a relatively short period of time, the signiﬁcance of tumor oxygenation for therapy outcome became evident in numerous experimental and clinical studies (for a review see Vaupel and Kelleher 1999). In this chapter, current knowledge on the oxygenation status of tumors and the occurrence of hypoxia in solid malignancies has been compiled. The mechanisms causing tumor hypoxia are discussed, and emphasis is given to the signiﬁcance of tumor hypoxia for tumor therapy and long-term prognosis of patients. All data presented here are derived from clinical studies on the pretreatment oxygenation status of cancers of the uterine cervix in conscious patients. Besides head and neck cancer, this has been the most extensively studied tumor type over the last 15 years with about 700 patients involved worldwide so far.

Deﬁnition of hypoxia In pathophysiology hypoxia is deﬁned as a state of reduced O2 availability or decreased O2 partial pressures (O2 tensions, pO2 values) below critical thresholds thus limiting characteristic cellular or organ functions. In contrast to * Supported by a grant from the Deutsche Krebshilfe (106758).

284

P. Vaupel and M. Höckel

normal tissues, malignant tumors obviously have no “physiologic” function. Thus, tumor hypoxia cannot be deﬁned by functional deﬁcits, although areas of necrosis, which are often found in tumors on microscopic examination, indicate the loss of vital cellular functions. In this chapter, the term hypoxia is used to describe critical O2 levels below which clinical, biological and/or molecular effects are progressively observed (e.g. treatment resistance, ATP depletion, binding of hypoxic markers, slowing of proliferation rate, genome and proteome changes). In this discussion of hypoxic thresholds, it is important to note that, for any functional parameter, a sharp threshold between hypoxia and normoxia does not exist and should not be expected. This chapter deals with the problem of hypoxia as a whole, encompassing mild, moderate and severe hypoxia (divisions that are not well deﬁned). Approaches in which oxygen effects have been characterized under in vitro conditions by using half-maximum values (e.g. in ionizing radiation) have proven useful in some instances, such as when the radiosensitivities of different cell lines under identical boundary conditions have been compared. However, use of half-maximum values in a more general discussion of hypoxia is not very informative because these values do not give the O2 levels at which hypoxia starts and, thus, becomes a biological problem (for a review see Höckel and Vaupel 2001a). Unfortunately, in an increasing number of reports on tumor oxygenation, the term hypoxia has been used in a somewhat careless manner without due consideration of the clear deﬁnitions for certain (experimental) conditions and scientiﬁc questions. Therefore, different researchers and clinicians discussing the problem of tumor hypoxia have used the term in different ways which has led to pitfalls. Anoxia describes the (patho-)physiological state, where no O2 is detected in the tissue (pO2 = 0 mmHg).

Pathogenesis of tumor hypoxia Using a polarographic technique (pO2 histography system, Helzel Medical Systems, Kaltenkirchen, Germany), our investigations carried out between 1987 and 2005 demonstrated that the presence of hypoxic tissue areas (i.e. areas with pO2 values ≤2.5 mmHg) is a characteristic pathophysiological property of locally advanced solid tumors and such areas have been found in a wide range of human malignancies: breast cancer (for reviews see Vaupel and Höckel 1999, 2000), cancer of the uterine cervix (for reviews see Höckel and Vaupel 2001b; Vaupel and Höckel 2001), head and neck cancer (for reviews see Vaupel 1997a, 2001), rectal cancer (Kallinowski and Buhr 1995; Mattern et al. 1996), prostate cancer (Movsas et al. 1999, 2000; Parker et al. 2004), pancreatic cancer (Koong et al. 2000), brain tumors (Rampling et al. 1994; Collingridge et al. 1999; Evans et al. 2004), soft tissue sarcomas (Brizel et al. 1996; Nordsmark et al. 1996, 1997), malignant melanoma (Lartigau

Tumor hypoxia and therapeutic resistance

285

et al. 1997), Non-Hodgkin’s lymphomas (Powell et al. 1999), and renal cell carcinomas (Lawrentschuk et al. 2005). The hypoxic (or anoxic) areas arise as a result of an imbalance between the supply and consumption of oxygen. Whereas in normal tissues or organs the O2 supply meets the metabolic requirements, in locally advanced solid tumors, the O2 consumption rate of neoplastic as well as stromal cells may outweigh an insufﬁcient oxygen supply and result in the development of tissue areas with very low O2 levels. Tumor hypoxia preferentially results from an inadequate perfusion due to severe structural and functional abnormalities of the tumor microcirculation (for reviews see Vaupel et al. 1989; Vaupel 1994, 2004). Hypoxic (micro-)regions are heterogeneously distributed within the tumor mass and may be located adjacent to regions with O2 tensions (pO2) corresponding to those found in the normal tissue neighboring the neoplastic lesion. Perfusionlimited O2 delivery leads to ischemic hypoxia which is often transient. For this reason, this type of hypoxia is also called “acute” hypoxia, a term which does not take into account the mechanisms underlying this condition (Vaupel et al. 2004; Vaupel and Mayer 2005). Hypoxia in tumors can also be caused by an increase in diffusion distances, so that cells far away (>70 μm) from the nutritive blood vessel receive less oxygen than needed. This condition is termed diffusion-limited hypoxia, also known as “chronic” hypoxia. In addition to enlarged diffusion distances, an adverse diffusion geometry (e.g. concurrent vs. countercurrent tumor microvessels) can also cause hypoxia (Vaupel and Harrison 2004). Tumor-associated or therapy-induced anemia can lead to a reduced O2 transport capacity of the blood, further contributing to the development of hypoxia (anemic hypoxia). This type of hypoxia is intensiﬁed especially in tumors or tumor areas that have low perfusion rates. A similar condition can be caused by carboxyhemoglobin (HbCO) formation in heavy smokers leading to a functional anemia, because hemoglobin blocked by carbon monoxide (CO) can no longer transport oxygen. Very often, tumor microvessels are perfused (at least transiently) by plasma only. In this situation, hypoxemic hypoxia around those vessels develops very rapidly because only a few tumor cells at the arterial end can be supplied adequately under these conditions. Similarly, hypoxia can rapidly develop in liver tumors which are preferentially supplied by the portal vein. There is abundant evidence for the existence of a substantial heterogeneity in the development and extent of tumor hypoxia due to pronounced intratumor (and intertumor) variabilities in vascularity and perfusion rates (for reviews see Vaupel et al. 1989; Vaupel 1994, 2004).

Oxygenation status of primary carcinomas of the uterine cervix Current knowledge on the oxygenation status of cervix cancers greatly refers to pretherapeutic data obtained in pre- and postmenopausal, conscious

286

P. Vaupel and M. Höckel 100 90

cumulative frequency [%]

80 70 60 50 40

normal cervix (n=7)

30

SCC (n=109) 20

AC (n=15) recurrences (n=46)

10 0 0

10

20

30

40

50

60

70

80

90

100

tissue pO2 [mmHg]

Fig. 1. Cumulative frequency distributions of measured tissue pO2 values in the normal cervix of nulliparous women, in primary squamous cell cancers (SCC) and in adenocarcinomas (AC) of stages Ib – IVa, and in local recurrences of cervical cancers. n = number of tumors investigated

women. Mean and median O2 tensions (pO2) obtained from >10,000 measurements in 141 primary carcinomas of the uterine cervix were, on average, distinctly lower than in normal tissues. Oxygen tensions measured in the normal cervix of nulliparous women revealed a median pO2 of 42 mmHg, whereas in locally advanced cancers of the cervix (stages FIGO Ib-IVa), the median pO2 was 10 mmHg (see Fig. 1). When tumors of different clinical sizes are compared, there is no evidence of a correlation between the maximum tumor diameter and the median pO2, the fraction of pO2 values ≤2.5 mmHg or the fraction of pO2 values ≤5 mmHg. In addition, there is no characteristic topological distribution of O2 tensions within cervix cancers (i.e. as a function of the measurement site; e.g. tumor periphery vs. tumor center). The oxygenation status and the extent of pretherapeutically measured hypoxic tissue areas are independent of the FIGO stages, histological types and histological grades. Similarly, there was no association between the oxygenation patterns and parity, menopausal status, smoking habits, or a series of other clinically relevant parameters. In cervical cancers the median pO2 values tended to rise with increasing hemoglobin concentrations from 10 to 13 g/dl. At hemoglobin levels >14 g/dl, a worsening of the tumor oxygenation was apparent (Vaupel et al. 2002).

Tumor hypoxia and therapeutic resistance

287

About 60% of locally advanced carcinomas of the uterine cervix (FIGO stages Ib-III) exhibited hypoxic (pO2 ≤ 2.5 mmHg) and/or anoxic (pO2 = 0 mmHg) tissue areas, which are heterogeneously distributed within the tumor mass. All tumors of stages FIGO IVa were hypoxic, i.e. they consistently exhibited pO2 values ≤2.5 mmHg. From our systematic studies on the oxygenation status of locally advanced solid tumors there was clear evidence that tumor-to-tumor variability in the oxygenation status was signiﬁcantly greater than intratumor variability, both for squamous cell carcinomas and for adenocarcinomas of the uterine cervix (Höckel et al. 1999; Höckel and Vaupel 2001b; Vaupel and Höckel 2001). In order to clarify whether the pathological tumor stage (pT), rather than the FIGO stage may have an impact on the oxygenation status, in a subgroup of 65 patients treated with primary surgery (with curative intent), pathological tumor staging was performed based on histopathological investigation of the surgical specimens following radical hysterectomy or exenteration and lymph node dissection. This procedure identiﬁed a median maximum (histological) tumor diameter of 40 mm. In tumors with a maximum extension < 40 mm (n = 37), the median pO2 was 11 mmHg which was signiﬁcantly higher than the respective pO2 value in tumors with a maximum diameter > 40 mm (n = 28, median pO2 = 5 mmHg; p < 0.05). Median pO2 values in stage pT1b tumors were signiﬁcantly higher (18 mmHg) than in pT2b lesions (5 mmHg, p < 0.05). Hypoxic fractions were slightly lower in pT1b (bulky) tumors than in pT2b malignancies. From these data it is concluded that only a detailed histopathological tumor staging using surgical specimens enables detection of stage- and size-related differences in the oxygenation status of primary cancers of the uterine cervix. Clinical tumor dimensions and FIGO staging are not accurate enough for the estimation and characterization of the tumor oxygenation (Höckel and Vaupel 2001b). In an earlier study, Sundfor et al. (1998) pointed out that adenocarcinomas (AC) of the uterine cervix were signiﬁcantly better oxygenated than squamous cell carcinomas (SCC). In our studies, however, pO2 values were comparable in both histologies: median pO2 = 11 mmHg in SCC vs. 12 mmHg in AC, and mean pO2 = 16 mmHg in SCC vs. 18 mmHg in AC (see Fig. 1). There were only slight differences in the fraction of pO2 values ≤2.5 mmHg (25% in SCC vs. 16% in AC), in the fraction of pO2 values ≤5 mmHg (38% in SCC vs. 32% in AC) and in the percentage of patients with tumor pO2 values ≤2.5 mmHg (60% in SCC vs. 66% in AC). Better stage-for-stage prognosis for SCC than for AC of the uterine cervix thus cannot be explained by a substantially different oxygenation status of the two histologies.

Oxygenation status of locally recurrent cancers of the uterine cervix Measurements of intratumoral oxygenation in pelvic relapses of cervical carcinomas with a similar methodological approach as was developed for the

288

P. Vaupel and M. Höckel

primary disease showed a pronounced shift to more hypoxic oxygenation proﬁles in the recurrent tumors as compared to the primary lesions (Höckel et al. 1998). Median pO2 values in 46 pelvic recurrences of squamous cell carcinomas were signiﬁcantly lower than the median pO2 values of 95 primary tumors of comparable sizes (4 mmHg vs. 11 mmHg, p < 0.001; see Fig. 1). In locally recurrent tumors, no signiﬁcant differences in the oxygenation status between SCC and AC were observed. For both histologies the median pO2 was lower, and the hypoxic fraction with pO2 values ≤2.5 mmHg was higher in recurrent tumors than in the primaries. The percentage of patients with pO2 values ≤2.5 mmHg in recurrent tumors was 77% (SCC, n = 46) and 87% (AC, n = 14). An analysis of intergroup differences in tumor oxygenation indicated that the greater the extent of hypoxia in primary tumors, the higher the probability of local recurrence of cervix cancers. Summarizing available data on pretreatment tumor oxygenation of locally advanced cancers of the uterine cervix (Höckel et al. 1991, 1998; Höckel and Vaupel 2001b; Vaupel and Höckel 2001) there is evidence that 1. Oxygenation in tumors is heterogeneous and compromised as compared to normal tissues. 2. Tumor oxygenation is not regulated according to the metabolic demand as is the case in normal tissues. 3. Causative factors for the development of hypoxia are limitations in perfusion and diffusion as well as tumor-associated anemia. 4. On average, the median pO2 values in primary cancers of the uterine cervix are lower than those in the normal cervix. 5. Many cervical cancers contain hypoxic tissue areas (≈ 60% in SCC, ≈ 66% in AC). 6. There is no characteristic topological distribution of O2 tensions within cervix cancers. 7. Tumor-to-tumor variability in oxygenation is greater than intratumor variability. 8. Tumor oxygenation is independent of various patient demographics (e.g. age, menopausal status, parity). 9. Anemia (in approx. 30% of patients at diagnosis) considerably contributes to the development of hypoxia, especially in low-ﬂow tumor areas. 10. In cervix cancers of moderately/severely anemic patients hypoxic areas are more frequently found than in non-anemic patients. 11. Tumor oxygenation and the extent of hypoxia are independent of clinical size, FIGO stage, histological type (SCC vs. AC), grade and lymph node status.

Tumor hypoxia and therapeutic resistance

289

12. Tumor oxygenation is weakly dependent on the pathological tumor stage (pT stage). 13. Local recurrences of cervix cancers have a higher hypoxic fraction than the primary tumors.

Tumor hypoxia and treatment resistance General aspects Hypoxia protects tumor cells from damage by nonsurgical anticancer therapies which are directly or indirectly O2-dependent (or both; for reviews see Moulder and Rockwell 1987; Durand 1991, 1994; Tannock et al. 2005; Teicher 1993, 1994, 1995; Vaupel 1997b; Chaplin et al. 2000; Hall and Giaccia 2006; Höckel and Vaupel 2001a; Vaupel et al. 2001; see Table 1). Hypoxia can lead to therapeutic resistance most commonly through (i) direct effects via deprivation of molecular O2 and thus reduced generation of free radicals which some chemotherapeutic agents (e.g. the antibiotics bleomycin and doxorubicin; Erlichman 1992) and photodynamic therapy require to be maximally cytotoxic. Sparsely ionizing radiation (X- and γ-rays) needs O2 for “ﬁxation” of DNA damages (Fig. 2); (ii) indirect effects via hypoxia-mediated modulation (stimulation or inhibition) of gene expression (see Fig. 2) and posttranscriptional or posttranslational effects resulting in changes of the proteome and leading, inter alia, to • modulation of proliferation kinetics, cell cycle position and the number of tumor cells accumulating in Go-phase (e.g. the vinca alkaloids and methotrexate exhibit cell-cycle-phase speciﬁcity; Chabner et al. 1996); • quantitative changes in cellular metabolism (e.g. intensiﬁed glycolysis in hypoxic tumors with tissue acidosis which in turn can have an impact on cellular activation, intracellular accumulation, membrane transport of drugs), increased enzyme activities; elevated intracellular concentrations of glutathione (GSH) and associated nucleophilic thiols that can compete with the target DNA for alkylation (see Table 1); • increased transcription of membrane transporters (e.g. GLUT-1 facilitating the efﬂux of vinblastine; Vera et al. 1991), DNA repair enzymes, autocrine and paracrine growth factors (e.g. TGF-β), proteins involved in cell detachment and tumor invasiveness, and resistance-related proteins. Many hypoxia-inducible genes are controlled by the transcription factors HIF-1, nuclear factor κB (NFκB), and activator protein-1 (AP-1; Koong et al. 1994; Dachs and Tozer 2000; Laderoute et al. 2002).

proliferation kinetics, cell cycle effects elevated levels of glutathione and DNA repair enzymes

Indirect effects via proteome changes

intratumor pharmacokinetics (impaired and uneven delivery, large diffusion distances)

* Anemia acts as a factor worsening tumor hypoxia.

Secondary indirect effects via a chaotic angiogenesis

cellular uptake and activation

Secondary indirect effects via an intensiﬁed glycolysis & extracellular acidosis

cell cycle effects, repair processes

loss of apoptosis and of differentiation, clonal heterogeneity, proliferation of resistant clonal variants, malignant progression

Indirect effects via genome changes & clonal selection

increased transcription of resistancerelated proteins

upregulation of telomerase increased transcription of membrane transporters (GLUT-1, MDR-1)

reduced “ﬁxation” of DNA damage in hypoxia deprivation of molecular O2 leads to reduced generation of free radicals and thereby to less DNA damage

Direct effects

Mechanisms involved

Erlichman 1992

antibiotics (bleomycin, doxorubicin) etoposide photodynamic therapy

chemotherapeutic drugs, (passive) immunotherapy

doxorubicin, vinblastine, vincristine (via decreased cytotoxicity at pHe < 7) X- and γ-rays

all treatments

vinca alkaloids, methotrexate X- and γ-rays alkylating agents, platinum compounds, bleomycin X- and γ-rays telomerase inhibitors vinblastine (via a facilitated efﬂux) anthracyclines e.g., methotrexate (via an increased activity of dihydrofolate reductase)

Hall and Giaccia 2006

X- and γ-rays

Vaupel 1997b

Durand 1991, 1994

Song et al. 1993

Höckel and Vaupel 2001a

Comerford et al. 2002 Rice et al. 1986

Hall and Giaccia 2006 Nishi et al. 2004, Anderson et al. 2006 Vera et al. 1991

Chabner et al. 1996 Hall and Giaccia 2006 Chabner et al. 1996, Zeller 1995

Shannon et al. 2003 Henderson and Fingar 1987

References

Treatment affected

Table 1. Tumor hypoxia and acquired treatment resistance (selection of mechanisms)* 290 P. Vaupel and M. Höckel

Tumor hypoxia and therapeutic resistance

291

Fig. 2. Schematic representation of major pathogenetic factors causing tumor hypoxia and of the pivotal role of hypoxia in the development of therapeutic resistance via direct and indirect mechanisms. RT = radiotherapy, CT = chemotherapy

292

P. Vaupel and M. Höckel

In addition to hypoxia, other epigenetic microenvironmental factors (e.g. acidosis, glucose depletion, lactate accumulation) may also be involved in the mechanisms described above. (For more details on hypoxia-mediated proteome changes see Rice et al. 1986; Laderoute et al. 1992; Ausserer et al. 1994; Graeber et al. 1994; Sanna and Rofstad 1994; Giaccia 1996; Mattern et al. 1996; Raleigh 1996; Brown and Giaccia 1998; Sutherland 1998; Semenza 2000a, 2000b; Höckel and Vaupel 2001a.) (iii) indirect effects via (progressive) genome changes and clonal selection (Vaupel 2004). Increasing resistance towards nonsurgical therapy concomitant with primary tumor growth can also be driven by transient or persistent genomic changes and clonal selection (often associated with subsequent clonal dominance) due to a hypoxia-related strong selection pressure (see Fig. 2). Hypoxia promotes genomic instability (through point mutations, gene ampliﬁcation, and chromosomal rearrangements), thus increasing the number of genetic variants and thereby promoting clonal and intrinsic tumor cell heterogeneity. Emancipative proliferation of resistant clonal variants in a “survival of the ﬁttest” scenario and malignant progression are the ﬁnal results (see Table 1). Hypoxia-mediated clonal selection of tumor cells with persistent genomic changes can lead, inter alia, to a loss of differentiation and of apoptosis which can stabilize or further aggravate tumor hypoxia and which in turn again promotes malignant progression (Vaupel 2004). Thus, hypoxia is involved in a vicious circle that is regarded as a fundamental biologic mechanism of malignant disease (for reviews see Höckel and Vaupel 2001a; Vaupel et al. 2004). Other consequences of hypoxia-induced malignant progression are an increased locoregional spread and enhanced metastasis (Höckel et al. 1996, 1998). (For more details on hypoxia-mediated genome changes and expansion of aggressive tumor subclones see Young et al. 1988; Stoler et al. 1992; Cheng and Loeb 1993; Stackpole et al. 1994; Russo et al. 1995; Giaccia 1996; Graeber et al. 1996; Reynolds et al. 1996; Kim et al. 1997; Höckel et al. 1999; Höckel and Vaupel 2001a.)

Tumor hypoxia as an obstacle in radiotherapy Tumor hypoxia may present a severe problem for radiation therapy (X- and γ-radiation), because radiosensitivity is progressively limited when the O2 partial pressure in a tumor is less than 25–30 mmHg. Hypoxia-associated resistance to photon radiotherapy is multifactorial. Molecular oxygen “ﬁxes” (i.e. makes permanent) DNA damage produced by oxygen free radicals, which arise after the interaction of radiation with intracellular water (Hall

Tumor hypoxia and therapeutic resistance

293

and Giaccia 2006). Thus, because of this so-called “oxygen-enhancement effect”, the radiation dose required to achieve the same biologic effect is three times higher in the absence of oxygen than in the presence of normal levels of oxygen (Gray et al. 1953). Evidence suggests that hypoxia-induced proteome and genome changes (see Table 1) may also have a substantial impact on radioresistance by increasing the levels of heat shock proteins and repair enzymes or by increasing the number of cells in a tumor with diminished apoptotic potential or increased proliferation potential of selected clones, both of which have been linked to radioresistance (for a recent review see Höckel and Vaupel 2001a). Numerous clinical studies report an impaired radiocurability of anemic patients, most probably due to hypoxia-related radioresistance (Evans and Bergsjö 1965; Bush 1986; Frommhold et al. 1998; Henke et al. 1999; Grau and Overgaard 2000; Kumar 2000; Harrison et al. 2002; Dunst 2004; Harrison and Blackwell 2004). A signiﬁcant inﬂuence of hemoglobin level on the outcome of radiotherapy has been convincingly documented for carcinomas of the uterine cervix, head and neck, bladder and bronchus (for a review see Grau and Overgaard 2000). Carbon monoxide (CO) in tobacco smoke strongly binds to hemoglobin and thus decreases the amount of “effective” hemoglobin. Furthermore, CO increases the hemoglobin afﬁnity for O2. The sum of these effects is a significant increase in tumor hypoxia and in radioresistance resulting in a poorer treatment outcome after primary radiotherapy (for a review see Grau and Overgaard 2000). In a tumor-associated and chemotherapy-induced anemia model, experiments under well-deﬁned boundary conditions have clearly shown that anemia can be abrogated by treatment with erythropoietin (rhEPO). While anemic animals showed substantial radioresistance, prevention of anemia by rhEPO resulted in a signiﬁcant increase in tumor radiosensitivity, almost achieving the sensitivity found in non-anemic animals (Thews et al. 1998; Kelleher et al. 1999). Furthermore, preclinical studies have shown that the use of rhEPO can improve the efﬁcacy of radiation in xenografted human tumors (Stüben et al. 2001, 2003a, 2003b). First clinical trial experiences using erythropoietin to prevent or correct anemia found statistically signiﬁcant improvement in locoregional tumor control following radiation therapy (Henke et al. 1999; Lavey 1999). However, thereafter, a study using EPO with the objective of improving cancer control and survival of patients with head and neck cancer who underwent curative radiotherapy diverge from the results obtained earlier (Henke et al. 2003). Conﬂicting results have also been published concerning the relationship between the occurrence of hypoxia, the expression of HIF-1α, GLUT-1, CA IX and VEGF, vascular density, and blood ﬂow. Assuming that vascular density, poor blood ﬂow and high expression of HIF-1α, GLUT-1, CA IX and VEGF might be indicators (surrogate parameters) of tumor hypoxia, these

294

P. Vaupel and M. Höckel

parameters may be used as prognostic factors in conventional radiotherapy. There are preliminary data available indicating that patients with carcinomas of the cervix exhibiting low perfusion rates (Mayr et al. 1996), high VEGF expression (Loncaster et al. 2000), overexpression of HIF-1α (Birner et al. 2000) and GLUT-1 (Airley et al. 2001), and low vascular density (Awwad et al. 1986; Siracká et al. 1988; Révész et al. 1989) show poor disease control following radiotherapy. Subsequently published data, however, suggest that there is no clear cut correlation between vascular density and treatment outcome in advanced squamous cell carcinoma of the uterine cervix in patients treated with radiotherapy (Sundfor et al. 2000). Therefore, additional studies are urgently needed to provide a clear and more detailed picture of the interrelationship between tumor vascularization and radiosensitivity.

Tumor hypoxia, an adverse parameter in chemotherapy Besides restricted delivery and uneven distribution (due to poor and heterogeneous blood ﬂow) as well as reduced diffusional ﬂux (due to enlarged diffusion distances), oxygen-dependency has been documented for a number of cytotoxic drugs (e.g. cyclophosphamide, carboplatin, and doxorubicin) under in vitro and in vivo conditions (Teicher et al. 1981, 1990; Teicher 1994, 1995). However, these investigations have been qualitative, and clear hypoxic thresholds for O2-dependent anticancer agents are still not available, although they presumably exist for each agent. Thus, additional research is necessary to provide quantitative data on hypoxia-induced chemoresistance, although this information may be difﬁcult to obtain under in vivo conditions. Multiple (direct and indirect) mechanisms are probably also involved in the hypoxia-induced resistance to chemotherapeutic agents, including a reduced generation of free radicals (e.g. bleomycin, anthracyclines), the increased production of nucleophilic substances as glutathione, that can compete with the target DNA for alkylation (e.g. in the acquired resistance to alkylating agents), an increased activity of DNA repair enzymes (e.g. alkylating agents, platinum compounds; Chabner et al. 1996), an inhibition of cell proliferation, and tissue acidosis which is often observed in hypoxic tumors with a high glycolytic rate (Durand 1991, 1994). Furthermore, hypoxic stress proteins, the loss of apoptotic potential and multi-drug resistance proteins can impart resistance to certain chemotherapeutic drugs (Sakata et al. 1991; Hickman et al. 1994; Shannon et al. 2003). As already described above, anemic animals showed a substantial radioresistance and prevention of anemia by rhEPO could abrogate the loss of radiosensitivity. In another study, Thews et al. (2001) showed similar effects with cyclophosphamide-based chemotherapy. In these experiments, rhEPO could restore the anemia-induced reduction in cyclophosphamide cytotoxic-

Tumor hypoxia and therapeutic resistance

295

ity in rat tumors, most probably as a result of a better O2 supply to the tumor tissue. Silver and Piver (1999) also found in an animal study that the efﬁcacy of cisplatin treatment on xenotransplanted ovarian cancer was increased when mice were simultaneously treated with rhEPO. The authors also attributed the improved chemosensitivity to a better oxygenation status of the tumor as a result of a higher hemoglobin concentration. A further study by Shannon et al. (2005) also showed that correction of anemia with darbepoetin improved the outcome with chemotherapy (cisplatin + gemcitabine) in murine Lewis lung carcinomas. While these authors were able to demonstrate a reduction in tumor hypoxia [also described earlier by Kelleher et al. (1996) with rhEPO] upon darbepoetin treatment which might have been responsible for the increase in chemosensitivity, a further ﬁnding, namely an increase in the delivery of cytotoxic drugs in the darbepoetin group, may also contribute to better outcome observed. Other reports indicate that RBCs may transport important quantities of various anticancer agents. It is possible that RBCs are more important in drug delivery than currently assumed. Pretreatment elevation and/or maintenance of Hb levels are therefore essential, irrespective of the way in which this goal is achieved (Wildiers et al. 2002).

Tumor hypoxia as a barrier for other nonsurgical anticancer therapies Photodynamic therapy-mediated cell death requires the presence of oxygen, a photosensitizing drug, and light of the appropriate wavelength, both in vitro and in vivo (for a review see Freitas and Baronzio 1991). However, reports vary greatly on the extent to which photodynamic therapy with hematoporphyrin derivatives is dependent on oxygen (Moan and Sommer 1985; Henderson and Fingar 1987). Cells were not killed under anoxic conditions. The critical threshold below which progressively reduced cell death was observed varied between 15 and 35 mmHg (Mitchell et al. 1985; Henderson and Fingar 1987; Chapman et al. 1991), probably because of reduced production of singlet oxygen species (1O2) and different sensitivities to the treatment in different cell lines. Considering the reduced effectiveness of photodynamic agents at lower O2 partial pressures, the rapid induction of tumor hypoxia by photodynamic therapy itself – either as a consequence of a photodynamic therapy-induced decrease in blood ﬂow or as a result of oxygen consumption by the photodynamic therapy process itself – has to be considered under in vivo conditions, since it may mean that this therapy is self-limiting (Chapman et al. 1991). Photodynamic therapy involving prodrugs, such as aminolevulinic acid (ALA), may be further limited because conversion of the prodrug to the active photosensitizer appears to be less effective under hypoxic conditions. Finally, tumor hypoxia can dramatically modulate the effectiveness of certain (passive) immunotherapy using cytokines (interferon-γ and tumor necrosis factor-α) and alter interleukin-2-induced activation of lymphokine-

296

P. Vaupel and M. Höckel

activated killer (LAK) cells (reviewed by Chaplin et al. 2000). The potency of treatment started to decrease at O2 partial pressures of less than approximately 35 mmHg.

Tumor hypoxia as an adverse prognostic factor Hypoxia is a pivotal (patho-)physiological mechanism that can increase therapeutic resistance. In addition, O2 depletion in solid tumors can impact longterm prognosis, independent of the mode of primary treatment (e.g. radical surgery vs. radiotherapy of locally advanced cervical cancers; Höckel et al. 1993, 1996, 1998; Höckel and Vaupel 2001b). This ﬁnding led to the hypothesis that tumor hypoxia is associated with malignant progression in advanced cancer of the uterine cervix (see Fig. 3) and that hypoxia may not only counteract O2-dependent forms of therapy but may also advance tumor progression per se. The prognostic relevance of hypoxia for carcinomas of the uterine cervix treated either with radiation or with radical surgery has been conﬁrmed by other groups although inter-institutional variations in absolute oxygenation data have become evident (Fyles et al. 1998; Knocke et al. 1999; Sundfor et al. 2000). Multivariate analysis has shown that hypoxia is also a powerful prognostic factor in squamous cell carcinomas of the head and neck (Nordsmark et al. 1996, 2005; Brizel et al. 1997; Stadler et al. 1999; Nordsmark and Overgaard 2000, 2004) and in soft tissue sarcomas (Brizel et al. 1996; Nordsmark et al. 2001; Evans and Koch 2003). This parameter is independent of other prognostic factors which can be assessed pretherapeutically such as tumor size or stage and thus may become clinically useful.

Summary Heterogeneously distributed hypoxic areas are a characteristic property of locally advanced solid tumors. Hypoxia results from an imbalance between the supply and consumption of oxygen (O2). Major pathogenetic mechanisms for the emergence of hypoxia are (i) structural and functional abnormalities in the tumor microvasculature, (ii) an adverse diffusion geometry, and (iii) tumor-related and therapy-induced anemia leading to a reduced O2 transport capacity of the blood. There is pronounced intertumor variability in the extent of hypoxia, which is independent of clinical size, stage, histopathological type and grade. Local recurrences have a higher hypoxic fraction than primary tumors. Hypoxia is intensiﬁed in anemic patients, especially in tumor (areas) with low perfusion rates. Tumor hypoxia is a therapeutic problem as it makes solid tumors resistant to sparsely ionizing radiation, some forms of chemotherapy, and photodynamic

Tumor hypoxia and therapeutic resistance

297

Fig. 3. Schematic representation of the paramount role of hypoxia and other factors of the hostile microenvironment in solid tumors (e.g. acidosis, glucose depletion, lactate accumulation) in the development of an aggressive tumor cell phenotype and in malignant progression. Hypoxia is involved in a deadly roller coaster of two interwoven vicious circles that are regarded as fundamental biological mechanisms of the malignant disease

therapy. However, besides more direct mechanisms involved in the development of therapeutic resistance, there are, in addition, indirect machineries that can cause barriers to therapies. These include hypoxia-mediated alterations in gene expression, proteome and genome changes, and clonal selection. These, in turn, can drive subsequent events that are known to further increase resistance to therapy, in addition to critically affecting long-term prognosis.

References 1. Airley R, Loncaster J, Davidson S, Bromley M, Roberts S, Patterson A, Hunter R, Stratford I, West C (2001) Glucose transporter Glut-1 expression correlates

298

2.

3.

4.

5.

6.

7.

8. 9. 10.

11.

12.

13. 14.

15.

16.

17.

P. Vaupel and M. Höckel with tumor hypoxia and predicts metastasis-free survival in advanced carcinoma of the cervix. Clin Cancer Res 7: 928–934 Anderson CJ, Hoare SF, Ashcroft M, Bilsland AE, Keith WN (2006) Hypoxic regulation of telomerase gene expression by transcriptional and posttranscriptional mechanisms. Oncogene 25: 61–69 Ausserer WA, Bourrat-Floeck B, Green CJ, Laderoute KR, Sutherland RM (1994) Regulation of c-jun expression during hypoxic and low-glucose stress. Mol Cell Biol 14: 5032–5042 Awwad HK, Naggar M, Mocktar N, Barsoum M (1986) Intercapillary distance measurement as an indicator of hypoxia in carcinoma of the cervix uteri. Int J Radiat Oncol Biol Phys 12: 1329–1333 Birner P, Schindl M, Obermair A, Plank C, Breitenecker G, Oberhuber G (2000) Overexpression of hypoxia-inducible factor 1α is a marker for an unfavorable prognosis in early-stage invasive cervical cancer. Cancer Res 60: 4693– 4696 Brizel DM, Scully SP, Harrelson JM, Layﬁeld LJ, Bean JM, Prosnitz LR, Dewhirst MW (1996) Tumor oxygenation predicts for the likelihood of distant metastases in human soft tissue sarcoma. Cancer Res 56: 941–943 Brizel DM, Sibley GS, Prosnitz LR, Scher RL, Dewhirst MW (1997) Tumor hypoxia adversely affects the prognosis of carcinoma of the head and neck. Int J Radiat Oncol Biol Phys 38: 285–289 Brown JM, Giaccia AJ (1998) The unique physiology of solid tumors: opportunities (and problems) for cancer therapy. Cancer Res 58: 1408–1416 Bush RS (1986) The signiﬁcance of anemia in clinical radiation therapy. Int J Radiat Oncol Biol Phys 12: 2047–2050 Chabner B, Allegra CJ, Curt GA, Calabresi P (1996) Antineoplastic agents. In: Goodman & Gilman’s The pharmacological basis of therapeutics, 9th edn. McGraw-Hill, New York, pp 1233–1287 Chaplin DJ, Horsman MR, Trotter MJ, Siemann DW (2000) Therapeutic significance of microenvironmental factors. In: Molls M, Vaupel P (eds) Blood perfusion and microenvironment of human tumors. Implications for clinical radiooncology. Springer, Berlin Heidelberg New York, pp 133–143 Chapman JD, Stobbe CC, Arnﬁeld MR, Santus R, Lee L, McPhee MS (1991) Oxygen dependency of tumor cell killing in vitro by light-activated Photofrin II. Radiat Res 126: 73–79 Cheng KC, Loeb LA (1993) Genomic instability and tumor progression: mechanistic considerations. Adv Cancer Res 60: 121–156 Collingridge DR, Piepmeier JM, Rockwell S, Knisely JPS (1999) Polarographic measurements of oxygen tension in human glioma and surrounding peritumoural brain tissue. Radiother Oncol 53: 127–131 Comerford KM, Wallace TJ, Karhausen J, Louis NA, Montalto MC, Colgan SP (2002) Hypoxia-inducible factor-1-dependent regulation of the multidrug resistance (MDR1) gene. Cancer Res 62: 3387–3394 Dachs GU, Tozer GM (2000) Hypoxia modulated gene expression: angiogenesis, metastasis and therapeutic exploitation. Eur J Cancer 36: 1649– 1660 Dunst J (2004) Low hemoglobin levels: inﬂuence on tumor biology and radiotherapy treatment outcome. Eur J Cancer [Suppl] 2: 3–10

Tumor hypoxia and therapeutic resistance

299

18. Durand RE (1991) Keynote address: The inﬂuence of microenvironmental factors on the activity of radiation and drugs. Int J Radiat Oncol Biol Phys 20: 253–258 19. Durand RE (1994) The inﬂuence of microenvironmental factors during cancer therapy. In Vivo 8: 691–702 20. Erlichman C (1992) Pharmacology of anticancer drugs. In: Tannock IF, Hill RP (eds) The basic science of oncology, 2nd edn. McGraw-Hill, New York, pp 317–337 21. Evans IC, Bergsjø P (1965) The inﬂuence of anemia on the results of radiotherapy in carcinoma of the cervix. Radiology 84: 709–717 22. Evans SM, Koch CJ (2003) Prognostic signiﬁcance of tumor oxygenation in humans. Cancer Letters 195: 1–16 23. Evans SM, Judy KD, Dunphy I, Jenkins WT, Nelson PT, Collins R, Wileyto EP, Jenkins K, Hahn SM, Stevens CW, Judkins AR, Phillips P, Geoerger B, Koch CJ (2004) Comparative measurements of hypoxia in human brain tumors using needle electrodes and EF5 binding. Cancer Res 64: 1886–1892 24. Freitas I, Baronzio GF (1991) Tumor hypoxia, reoxygenation and oxygenation strategies: possible role in photodynamic therapy. J Photochem Photobiol B: Biol 11: 3–30 25. Frommhold H, Guttenberger R, Henke M (1998) The impact of blood hemoglobin content on the outcome of radiotherapy. Strahlenther Onkol 174: 31– 34 26. Fyles AW, Milosevic M, Wong R, Kavanagh M-C, Pintilie M, Sun A, Chapman W, Levin W, Manchul L, Keane TJ, Hill RP (1998) Oxygenation predicts radiation response and survival in patients with cervix cancer. Radiother Oncol 48: 149–156 27. Giaccia AJ (1996) Hypoxic stress proteins: survival of the ﬁttest. Semin Radiat Oncol 6: 46–58 28. Graeber TG, Osmanian C, Jacks T, Housman DE, Koch CJ, Lowe SW, Giaccia AJ (1996) Hypoxia-mediated selection of cells with diminished apoptotic potential in solid tumours. Nature 379: 88–91 29. Graeber TG, Peterson JF, Tsai M, Monica K, Fornace AJ, Giaccia AJ (1994) Hypoxia induces accumulation of p53 protein, but activation of a G1-phase checkpoint by low-oxygen conditions is independent of p53 status. Molecular Cell Biol 14: 6264–6277 30. Grau C, Overgaard J (2000) Signiﬁcance of hemoglobin concentration for treatment outcome. In: Molls M, Vaupel P (eds) Blood perfusion and microenvironment of human tumors. Implications for clinical radiooncology. Springer, Berlin Heidelberg New York, pp 101–112 31. Gray LH, Conger AD, Ebert M, Hornsey S, Scott OCA (1953) The concentration of oxygen dissolved in tissues at the time of irradiation as a factor in radiotherapy. Br J Radiol 26: 638–648 32. Hall EJ, Giaccia A (2006) Radiobiology for the radiologist, 6th edn. Lippincott, Philadelphia 33. Harrison L, Blackwell K ( 2004) Hypoxia and anemia: Factors in decreased sensitivity to radiation therapy and chemotherapy? Oncologist 9: 31–40 34. Harrison LB, Chadha M, Hill RJ, Hu K, Shasha D (2002) Impact of tumor hypoxia and anemia on radiation therapy outcomes. Oncologist 7: 492–508

300

P. Vaupel and M. Höckel

35. Henderson BW, Fingar VH (1987) Relationship of tumor hypoxia and response to photodynamic treatment in an experimental mouse tumor. Cancer Res 47: 3110–3114 36. Henke M, Momm F, Guttenberger R (1999) Erythropoietin for patients undergoing radiotherapy: The Freiburg experience. In: Vaupel P, Kelleher DK (eds) Tumor hypoxia. Pathophysiology, clinical signiﬁcance and therapeutical perspectives. Wissenschaftliche Verlagsgesellschaft, Stuttgart, pp 91–97 37. Henke M, Laszig R, Rübe C, Schäfer U, Haase KD, Schilcher B, Mose S, Beer KT, Burger U, Dougherty C, Frommhold H (2003) Erythropoietin to treat head and neck cancer patients with anaemia undergoing radiotherapy: Randomised, double-blind, placebo-controlled trial. Lancet 362: 1255–1260 38. Hickman JA, Potten CS, Merritt AJ, Fisher TC (1994) Apoptosis and cancer chemotherapy. Phil Trans R Soc B 345: 319–325 39. Höckel M, Vaupel P (2001a) Tumor hypoxia: Deﬁnitions and current clinical, biological and molecular aspects. J Natl Cancer Inst 93: 266–276 40. Höckel M, Vaupel P (2001b) Prognostic signiﬁcance of tissue hypoxia in cervical cancer. CME J Gynecol Oncol 6: 216–225 41. Höckel M, Knoop C, Schlenger K, Vorndran B, Baussmann E, Mitze M, Knapstein PG, Vaupel P (1993) Intratumoral pO2 predicts survival in advanced cancer of the uterine cervix. Radiother Oncol 26: 45–50 42. Höckel M, Schlenger K, Aral B, Mitze M, Schäffer U, Vaupel P (1996) Association between tumor hypoxia and malignant progression in advanced cancer of the uterine cervix. Cancer Res 56: 4509–4515 43. Höckel M, Schlenger K, Höckel S, Aral B, Schäffer U, Vaupel P (1998) Tumor hypoxia in pelvic recurrences of cervical cancer. Int J Cancer 79: 365– 369 44. Höckel M, Schlenger K, Höckel S, Vaupel P (1999) Hypoxic cervical cancers with low apoptotic index are highly aggressive. Cancer Res 59: 4525–4528 45. Hoeckel M, Schlenger K, Knoop C, Vaupel P (1991) Oxygenation of carcinomas of the uterine cervix: Evaluation of computerized O2 tension measurements. Cancer Res 51: 6098–6102 46. Kallinowski F, Buhr HJ (1995) Tissue oxygenation of primary, metastatic and xenografted rectal cancers. In: Vaupel P, Kelleher DK, Günderoth M (eds) Tumor oxygenation. Fischer, Stuttgart, pp 205–209 47. Kelleher DK, Matthiensen U, Thews O, Vaupel P (1996) Blood ﬂow, oxygenation, and bioenergetic status of tumors after erythropoietin treatment in normal and anemic rats. Cancer Res 56: 4728–4734 48. Kelleher DK, Thews O, Vaupel P (1999) Modulation of tumor oxygenation and radiosensitivity by erythropoietin. In: Vaupel P, Kelleher DK (eds) Tumor hypoxia. Pathophysiology, clinical signiﬁcance and therapeutic perspectives. Wissenschaftliche Verlagsgesellschaft, Stuttgart, pp 83–90 49. Kim CY, Tsai MH, Osmanian C, Graeber TG, Lee JE, Giffard RG, DiPaolo JA, Peehl DM, Giaccia AJ (1997) Selection of human cervical epithelial cells that possess reduced apoptotic potential to low-oxygen conditions. Cancer Res 57: 4200–4204 50. Knocke TH, Weitmann H-D, Feldmann H-J, Selzer E, Pötter R (1999) Intratumoral pO2-measurements as predictive assay in the treatment of carcinoma of the uterine cervix. Radiother Oncol 53: 99–104

Tumor hypoxia and therapeutic resistance

301

51. Koong AC, Chen EY, Giaccia AJ (1994) Hypoxia causes the activation of nuclear factor κB through the phosphorylation of IκBα on tyrosine residues. Cancer Res 54: 1425–1430 52. Koong AC, Mehta VK, Le QT, Fisher GA, Terris DJ, Brown JM, Bastidas AJ, Vierra M (2000) Pancreatic tumors show high levels of hypoxia. Int J Radiat Oncol Biol Phys 48: 919–922 53. Kumar P (2000) Tumor hypoxia and anemia: Impact on the efﬁcacy of radiation therapy. Sem Hematol 37: 4–8 54. Laderoute KR, Grant TD, Murphy BJ, Sutherland RM (1992) Enhanced epidermal growth factor receptor synthesis in human squamous carcinoma cells exposed to low levels of oxygen. Int J Cancer 52: 428–432 55. Laderoute KR, Calaoagan JM, Gustafson-Brown C, Knapp AM, Li G-C, Mendonca HL, Ryan HE, Wang Z, Johnson RS (2002) The response of c-Jun/ AP-1 to chronic hypoxia is hypoxia-inducible factor 1α dependent. Mol Cell Biol 22: 2515–2523 56. Lartigau E, Randrianarivelo H, Avril M-F, Margulis A, Spatz A, Eschwege F, Guichard M (1997) Intratumoral oxygen tension in metastatic melanoma. Melanoma Res 7: 400–406 57. Lavey RS (1999) Clinical trial experience using erythropoietin during radiation therapy. In: Vaupel P, Kelleher DK (eds) Tumor hypoxia. Pathophysiology, clinical signiﬁcance and therapeutic perspectives. Wissenschaftliche Verlagsgesellschaft, Stuttgart, pp 99–105 58. Lawrentschuk N, Poon AMT, Foo SS, Johns Putra LG, Murone C, Davis ID, Bolton DM, Scott AM (2005) Assessing regional hypoxia in human renal tumours using 18F-ﬂuoromisonidazole positron emission tomography. Br J Urol Internat 96: 540–546 59. Loncaster JA, Cooper RA, Logue JP, Davidson SE, Hunter RD, West CML (2000) Vascular endothelial growth factor (VEGF) expression is a prognostic factor for radiotherapy outcome in advanced carcinoma of the cervix. Br J Cancer 83: 620–625 60. Mattern J, Kallinowski F, Herfarth C, Volm M (1996) Association of resistancerelated protein expression with poor vascularization and low levels of oxygen in human rectal cancer. Int J Cancer 67: 20–23 61. Mayr NA, Yuh WTC, Magnotta VA, Ehrhardt JC, Wheeler JA, Sorosky JI, Davis CS, Wen B-C, Martin DD, Pelsang RE, Buller RE, Oberley LW, Mellenberg DE, Hussey DH (1996) Tumor perfusion studies using fast magnetic resonance imaging technique in advanced cervical cancer: a new noninvasive predictive assay. Int J Radiat Oncol Biol Phys 36: 623–633 62. Mitchell JB, McPherson S, De Graff W, Gamson J, Zabell A, Russo A (1985) Oxygen dependence of hematoporphyrin derivative-induced photo-inactivation of Chinese hamster cells. Cancer Res 45: 2008–2011 63. Moan J, Sommer S (1985) Oxygen dependence of the photosensitizing effect of hematoporphyrin derivative in NHIK 3025 cells. Cancer Res 45: 1608– 1610 64. Moulder JE, Rockwell S (1987) Tumor hypoxia: its impact on cancer therapy. Cancer Metast Rev 5: 313–341 65. Movsas B, Chapman JD, Greenberg RE, Hanlon AL, Horwitz EM, Pinover WH, Stobbe C, Hanks GE (2000) Increasing levels of hypoxia in prostate carcinoma

302

66.

67.

68.

69.

70.

71.

72.

73.

74.

75.

76. 77.

78.

79. 80.

P. Vaupel and M. Höckel correlate signiﬁcantly with increasing clinical stage and patient age. Cancer 89: 2018–2024 Movsas B, Chapman JD, Horwitz EM, Pinover WH, Greenberg RE, Hanlon AL, Iyer R, Hanks GE (1999) Hypoxic regions exist in human prostate carcinoma. Urology 53: 11–18 Nishi H, Nakada T, Kyo S, Inoue M, Shay JW, Isaka K (2004) Hypoxia-inducible factor 1 mediates upregulation of telomerase (hTERT). Mol Cell Biol 24: 6076–6083 Nordsmark M, Overgaard J (2000) A conﬁrmatory prognostic study on oxygenation status and loco-regional control in advanced head and neck squamous cell carcinoma treated by radiation therapy. Radiother Oncol 57: 39–43 Nordsmark M, Overgaard J (2004) Tumor hypoxia is independent of hemoglobin and prognostic for loco-regional tumor control after primary radiotherapy in advanced head and neck cancer. Acta Oncol 43: 396–403 Nordsmark M, Hoyer M, Keller J, Nielsen OS, Jensen OM, Overgaard J (1996) The relationship between tumor oxygenation and cell proliferation in human soft tissue sarcomas. Int J Radiat Oncol Biol Phys 35: 701–708 Nordsmark M, Keller J, Nielsen OS, Lundorf E, Overgaard J (1997) Tumour oxygenation assessed by polarographic needle electrodes and bioenergetic status measured by 31P magnetic resonance spectroscopy in human soft tissue tumours. Acta Oncol 36: 565–571 Nordsmark M, Alsner J, Keller J, Nielsen OS, Jensen OM, Horsman MR, Overgaard J (2001) Hypoxia in human sof tissue sarcomas: adverse impact on survival and no association with p53 mutations. Br J Cancer 84: 1070–1075 Nordsmark M, Bentzen SM, Rudat V, Brizel D, Lartigau E, Stadler P, Becker A, Adam M, Molls M, Dunst J, Terris DJ, Overgaard J (2005) Prognostic value of tumor oxygenation in 397 head and neck tumors after primary radiation therapy. An international multi-center study. Radiother Oncol 77: 18–24 Parker C, Milosevic M, Toi A, Sweet J, Panzarella T, Bristow R, Catton C, Catton P, Crook J, Gospodarowicz M, McLean M, Warde P, Hill RP (2004) Polarographic electrode study of tumor oxygenation in clinically localized prostate cancer. Int J Radiat Oncol Biol Phys 58: 750–757 Powell MEB, Collingridge DR, Saunders MI, Hoskin PJ, Hill SA, Chaplin DJ (1999) Improvement in human tumour oxygenation with carbogen of varying carbon dioxide concentrations. Radiother Oncol 50: 167–171 Raleigh JA (ed) (1996) Hypoxia and its clinical signiﬁcance. Semin Radiat Oncol 6: 1–70 Rampling R, Cruickshank G, Lewis AD, Fitzsimmons SA, Workman P (1994) Direct measurement of pO2 distribution and bioreductive enzymes in human malignant brain tumors. Int J Radiat Oncol Biol Phys 29: 427–432 Révész L, Siracká E, Siracky´ J, Delides G, Pavlaki K (1989) Variation of vascular density within and between tumors of the uterine cervix and its predictive value for radiotherapy. Int J Radiat Oncol Biol Phys 16: 1161–1163 Reynolds TY, Rockwell S, Glazer PM (1996) Genetic instability induced by the tumor microenvironment. Cancer Res 56: 5754–5757 Rice GC, Hoy C, Schimke RT (1986) Transient hypoxia enhances the frequency of dihydrofolate reductase gene ampliﬁcation in Chinese hamster ovary cells. Proc Natl Acad Sci USA 83: 5978–5982

Tumor hypoxia and therapeutic resistance

303

81. Russo CA, Weber TK, Volpe CM, Stoler DL, Petrelli NJ, Rodriguez-Bigas M, Burhans WC, Anderson GR (1995) An anoxia inducible endonuclease and enhanced DNA breakage as contributors to genomic instability in cancer. Cancer Res 55: 1122–1128 82. Sakata K, Kwok TT, Murphy BJ, Laderoute KR, Gordon GR, Sutherland RM (1991) Hypoxia-induced drug resistance: comparison to P-glycoproteinassociated drug resistance. Br J Cancer 64: 809–814 83. Sanna K, Rofstad EK (1994) Hypoxia-induced resistance to doxorubicin and methotrexate in human melanoma cell lines in vitro. Int J Cancer 58: 258– 262 84. Semenza GL (2000a) Hypoxia, clonal selection, and the role of HIF-1 in tumor progression. Crit Rev Biochem Molec Biol 35: 71–103 85. Semenza GL (2000b) HIF-1: mediator of physiological and pathophysiological response to hypoxia. J Appl Physiol 88: 1474–1480 86. Shannon AM, Bouchier-Hayes DJ, Condron CM, Toomey D (2003) Tumour hypoxia, chemotherapeutic resistance and hypoxia-related therapies. Cancer Treat Rev 29: 297–307 87. Shannon AM, Bouchier-Hayes DJ, Condron CM, Toomey D (2005) Correction of anaemia through the use of darbepoetin alfa improves chemotherapeutic outcome in a murine model of Lewis lung carcinoma. Br J Cancer 93: 224–232 88. Silver DF, Piver MS (1999) Effects of recombinant human erythropoietin on the antitumor effect of cisplatin in SCID mice bearing human ovarian cancer: a possible oxygen effect. Gynecol Oncol 73: 280–284 89. Siracká E, Révész L, Kovác R, Siracky´ J (1988) Vascular density in carcinoma of the uterine cervix and its predictive value for radiotherapy. Int J Cancer 41: 819–822 90. Song CW, Lyons JC, Luo Y (1993) Intracellular pH in solid tumors: Inﬂuence on therapeutic response. In: Teicher BA (ed) Drug resistance in oncology. Marcel Dekker, New York, pp 25–51 91. Stackpole CW, Groszek L, Kalbag SS (1994) Benign-to-malignant B16 melanoma progression induced in two stages in vitro by exposure to hypoxia. J Natl Cancer Inst 86: 361–367 92. Stadler P, Becker A, Feldmann HJ, Hänsgen G, Dunst J, Würschmidt, Molls M (1999) Inﬂuence of the hypoxic subvolume on the survival of patients with head and neck cancer. Int J Radiat Oncol Biol 44: 749–754 93. Stoler DL, Anderson GR, Russo CA, Spina AM, Beerman TA (1992) Anoxiainducible endonuclease activity as a potential basis of the genomic instability of cancer cells. Cancer Res 52: 4372–4378 94. Stüben G, Thews O, Pöttgen C, Knühmann K, Vaupel P, Stuschke M (2001) Recombinant human erythropoietin increases the radiosensitivity of xenografted human tumours in anaemic nude mice. J Cancer Res Clin Oncol 127: 346–350 95. Stüben G, Thews O, Pöttgen C, Knühmann K, Sack H, Stuschke M, Vaupel P (2003a) Impact of anemia prevention by recombinant human erythropoietin on the sensitivity of xenografted glioblastomas to fractionated irradiation. Strahlenther Onkol 179: 620–625 96. Stüben G, Pöttgen C, Knühmann K, Schmidt K, Stuschke M, Thews O, Vaupel P (2003b) Erythropoietin restores the anemia-induced reduction in radiosensi-

304

97.

98.

99. 100. 101. 102. 103. 104.

105.

106.

107.

108.

109. 110.

111.

112. 113. 114.

P. Vaupel and M. Höckel tivity of experimental human tumors in nude mice. Int J Radiat Oncol Biol Phys 55: 1358–1362 Sundfor K, Lyng H, Rofstad EK (1998) Oxygen tension and vascular density in adenocarcinoma and squamous cell carcinoma of the uterine cervix. Acta Oncol 37: 665–670 Sundfor K, Lyng H, Trope CG, Rofstad EK (2000) Treatment outcome in advanced squamous cell carcinoma of the uterine cervix: relationships to pretreatment tumor oxygenation and vascularization. Radiother Oncol 54: 101– 107 Sutherland RM (1998) Tumor hypoxia and gene expression. Implications for malignant progression and therapy. Acta Oncol 37: 567–574 Tannock IF, Hill RP, Bristow RG, Harrington (eds) (2005) The basic science of oncology, 4th edn. McGraw-Hill, New York Teicher BA (ed)(1993) Drug resistance in oncology. Marcel Dekker, New York Teicher BA (1994) Hypoxia and drug resistance. Cancer Metast 13: 139–168 Teicher BA (1995) Physiologic mechanisms of therapeutic resistance. Hematol/ Oncol Clinics North America 9: 475–506 Teicher BA, Lazo JS, Sartorelli AC (1981) Classiﬁcation of antineoplastic agents by their selective toxicities toward oxygenated and hypoxic tumor cells. Cancer Res 41: 73–81 Teicher BA, Holden SA, Al-Achi A, Herman TS (1990) Classiﬁcation of antineoplastic treatments by their differential toxicity toward putative oxygenated and hypoxic tumor subpopulations in vivo in the FSaII murine ﬁbrosarcoma. Cancer Res 50: 3339–3344 Thews O, Kelleher DK, Vaupel P (2001) Erythropoietin restores the anemiainduced reduction in cyclophosphamide cytotoxicity in rat tumors. Cancer Res 61: 1358–1361 Thews O, Koenig R, Kelleher DK, Kutzner J, Vaupel P (1998) Enhanced radiosensitivity in experimental tumours following erythropoietin treatment of chemotherapy-induced anaemia. Br J Cancer 78: 752–756 Vaupel P (1994) Blood ﬂow, oxygenation, tissue pH distribution and bioenergetic status of tumors. Ernst Schering Research Foundation, Lecture 23, Berlin Vaupel P (1997a) Blood ﬂow and oxygenation status of head and neck carcinomas. Adv Exp Med Biol 428: 89–95 Vaupel P (1997b) The inﬂuence of tumor blood ﬂow and microenvironmental factors on the efﬁcacy of radiation, drugs and localized hyperthermia. Klin Pädiatr 209: 243–249 Vaupel P (2001) Durchblutung und Oxygenierungsstatus von Kopf-HalsTumoren. In: Böttcher HD, Wendt TG, Henke M (Hrsg) Klinik des Rezidivtumors im Kopf-Hals-Bereich. Grundlagen – Diagnostik – Therapie. Zuckschwerdt, München Bern Wien New York, S 7–23 Vaupel P (2004) The role of hypoxia-induced factors in tumor progression. Oncologist 9: 10–17 Vaupel P (2004) Tumor microenvironmental physiology and its implications for radiation oncology. Semin Radiat Oncol 14: 198–206 Vaupel P, Harrison L (2004) Tumor hypoxia: Causative factors, compensatory mechanisms, and cellular response. Oncologist 9: 4–9

Tumor hypoxia and therapeutic resistance

305

115. Vaupel P, Höckel M (1999) Oxygenation status of breast cancer: The Mainz experience. In: Vaupel P, Kelleher DK (eds) Tumor hypoxia. Pathophysiology, clinical signiﬁcance and therapeutic perspectives. Wissenschaftliche Verlagsgesellschaft, Stuttgart, pp 1–11 116. Vaupel P, Höckel M (2000) Blood supply, oxygenation status and metabolic micromilieu of breast cancers: Characterization and therapeutic relevance. Int J Oncol 17: 869–879 117. Vaupel P, Höckel M (2001) Hypoxie beim Zervixkarzinom: Pathogenese, Charakterisierung und biologische/klinische Konsequenzen. Zentralbl Gynäkol 123: 192–197 118. Vaupel P, Kelleher DK (eds) (1999) Tumor hypoxia: pathophysiology, clinical signiﬁcance and therapeutic perspectives. Wissenschaftliche Verlagsgesellschaft, Stuttgart 119. Vaupel P, Mayer A (2005) Effects of anaemia and hypoxia on tumour biology. In: Bokemeyer C, Ludwig H (eds) Anaemia in cancer, 2nd edn. Elsevier, Edingburgh London New York, pp 47–66 120. Vaupel P, Kallinowski F, Okunieff P (1989) Blood ﬂow, oxygen and nutrient supply, and metabolic microenvironment of human tumors: A review. Cancer Res 49: 6449–6465 121. Vaupel P, Thews O, Hoeckel M (2001) Treatment resistance of solid tumors: role of hypoxia and anemia. Med Oncol 18: 243–259 122. Vaupel P, Mayer A, Höckel M (2004) Tumor hypoxia and malignant progression. Methods Enzymol 381: 335–354 123. Vaupel P, Schlenger K, Knoop C, Hoeckel M (1991) Oxygenation of human tumors: Evaluation of tissue oxygen distribution in breast cancers by computerized O2 tension measurements. Cancer Res 51: 3316–3322 124. Vaupel P, Thews O, Mayer A, Höckel S, Höckel M (2002) Oxygenation status of gynecologic tumors: What is the optimal hemoglobin level? Strahlenther Onkol 12: 727–731 125. Vera JC, Castillo GR, Rosen OM (1991) A possible role for a mammalian facilitative hexose transporter in the development of resistance to drugs. Mol Cell Biol 11: 3407–3418 126. Young SD, Marshall RS, Hill RP (1988) Hypoxia induces DNA overreplication and enhances metastatic potential of murine tumour cells. Proc Natl Acad Sci USA 85: 9533–9537 127. Wildiers H, Guetens G, de Boeck G, Landuyt W, Verbeken E, Highley M, de Brunn EA, van Oostrom AT (2002) Melphalan availability in hypoxia-inducible factor-1α+/+ and factor-1α−/− tumors is independent of tumor vessel density and correlates with melphalan erythrocyte transport. Int J Cancer 99: 514–519 128. Zeller WJ (1995) Bleomycin. In: Zeller WJ, zur Hausen H (Hrsg) Onkologie: Grundlagen, Diagnostik, Therapie, Entwicklungen. ecomed, Landsberg, pp IV3.12, 1–7 Correspondence: Prof. Dr. Peter Vaupel, Institute of Physiology and Pathophysiology, University of Mainz, Duesbergweg 6, 55099 Mainz, Germany, E-mail: [email protected]. Prof. Dr. Dr. Michael Höckel, Department of Obstetrics and Gynecology, University of Leipzig, Philipp-Rosenthal-Strasse 55, 04103 Leipzig, Germany, E-mail: [email protected].

Chapter 12

Symptoms of anemia R. Pirker Department of Medicine I, Medical University of Vienna, Vienna, Austria

Introduction Anemia is a frequent diagnosis in medicine. Anemia can be present as a disease by itself or can be the consequence of another disease. Anemia is particularly common in patients with cancer (Groopman and Itri 1999; Ludwig et al. 2004; Pirker et al. 2005). The incidence and severity of cancer–related anemia depend on cancer type, tumor stage, duration of the cancer, patient age, type as well as intensity of anticancer therapy, and on bone marrow reserve (Groopman and Itri 1999; Ludwig and Strasser 2001; Ludwig et al. 2004; Pirker et al. 2005). Cancer-related anemia can be caused by myelosuppressive chemotherapy and radiotherapy, tumor inﬁltration of the bone marrow, relative deﬁciency of erythropoietin, inappropriate response of the bone marrow, functional iron deﬁciency, nutritional deﬁciencies, bleeding, hemolysis, cytokines, anemia of chronic disease, and other factors (Cotran et al. 1999; Groopman and Itri 1999). The anemia of chronic disease is the most common type of non-treatment-induced anemia in patients with cancer (DeRienzo and Saleem 1990; Krantz 1994) and is caused by mainly an impaired marrow response to erythropoietin and partly also by a relative inadequacy of erythropoietin production (Dowlati 1997). Normal erythropoiesis is regulated by erythropoietin and other cytokines (Fisher 2003). Erythropoietin is synthesized primarily in the kidneys and its synthesis is increased by hypoxia (Fisher 2003). Erythropoietin binds to the erythropoietin receptor on immature erythroid cells in the bone marrow. After binding erythropoietin, the erythropoietin receptor dimerizes and activates speciﬁc intracellular kinases which subsequently affect transcription factors (Fisher 2003). This then results in survival, proliferation and maturation of erythroid cells (Fisher 2003). A lack of erythropoietin leads to the development of normochromic normocytic anemia. Anemia is deﬁned as a condition of decreased red blood cell mass resulting in decreased levels of hematocrit and hemoglobin (Hb). Serum Hb levels are usually used to deﬁne anemia. According to the World Health

308

R. Pirker

Organization, Hb levels below 12 g/dl in women and below 13 g/dl in men are deﬁned as anemia (World Health Organization 1968). Based on Hb levels, the severity of anemia is classiﬁed as mild, moderate, severe and lifethreatening. In clinical trials, anemia is usually deﬁned as a Hb level below 12 g/dl and the hematotoxicity is classiﬁed according to either WHO criteria, NCI Common Toxicity Criteria or other grading systems. Red blood cells bind and deliver oxygen to tissues. Anemia leads to an imbalance between the supply and the consumption of oxygen in tissues. Respiration and circulation are the other two factors relevant for oxygen supply to tissues. Each of these three factors is able to compensate the others to a certain extent.

Compensatory mechanisms In the presence of anemia, the body has developed several compensatory mechanisms in order to guarantee the function of vital organs such as heart, lung, kidneys and brain. Adjustments to anemia mainly are achieved through the cardiovascular system and through changes in the hemoglobin-oxygen dissociation curve. The heart rate increases and this leads to an increased cardiac output which improves the blood ﬂow to the organs. Additional vasoconstriction of peripheral vessels distributes the blood ﬂow towards vital organs. The respiratory rate increases in order to improve oxygenation of the blood. Red blood cells of anemic patients built more 2,3-biphosphogycerate which decreases the oxygen afﬁnity of Hb and facilitates oxygen delivery to tissues. Finally, erythropoietin levels increase and lead to enhanced production of red blood cells. These compensatory mechanisms guarantee sufﬁcient blood ﬂow to the vital organs and are also responsible for some of the symptoms experienced by anemic patients.

Symptoms and quality of life The clinical manifestations of anemia depend on the reduction in the oxygencarrying capacity of the blood, the degree of change in total blood volume, compensatory mechanisms, and the underlying disease leading to anemia. Because anemia affects all organs and tissues, a wide range of both symptoms and signs can be seen in patients with anemia (Table 1). Patients complain about shortness of breath, palpitations and decreased work tolerance. Other symptoms are dizziness, vertigo, faintness, headache, pallor, angina pectoris, tachycardia, depression, loss of libido and fatigue. Fatigue is of particular importance in patients with cancer. Although the etiology of cancer-related fatigue is multifactorial, anemia as a cause of fatigue is of major clinical relevance because of its high prevalence among cancer

Symptoms of anemia

309

Table 1. Symptoms and signs of anemia (examples) General symptoms

Cardiovascular symptoms

Pulmonary symptoms Cerebral symptoms

Kidneys Gastrointestinal symptoms

Skin

Weakness Tiredness Fatigue Tachycardia Palpitations Dyspnea Angina pectoris Myocardial infarction Hypotension Shortness of breath Tachypnoe Dizziness Vertigo Headache Depression Anxiety Proteinurea Edema Glossitis Ulcers Malabsorption Pallor Cold skin and cold intolerance

patients and the possibilities for its treatment (see below). Pallor is a frequent sign of anemia and often noted by the patient’s relatives. The severity of symptoms depends on the absolute Hb level, the rapidity of onset of anemia, compensatory mechanisms, co-morbidity and other factors (Ludwig and Strasser 2001). In patients with chronic and slowly developing anemia, symptoms may develop only after Hb levels have dropped to low values. Co-morbidity, in particular pre-existent cardiovascular or pulmonary disease, may exaggerate the symptoms of anemia. Thus elderly patients who frequently suffer from co-morbidities will develop symptoms earlier and with greater severity than young patients. Patients without comorbidity and only slowly developing anemia will adjust to the situation and develop symptoms as well as seek medical attention only when severe anemia has developed. This phenomenon is particularly seen in patients with anemia due to iron deﬁciency or vitamin B12 deﬁciency. In the case of acute blood loss, the symptoms are those of hypovolemia and, dependent on the degree of blood loss, range from mild hypotension to severe shock.

310

R. Pirker

Cardiovascular system Anemia has a major impact on the cardiovascular system. Patients complain about palpitations, tachycardia and shortness of breath. In case of mild anemia, these symptoms might be experienced only during exercise. In case of more severe anemia and/or in patients with pre-existent arterial disease, anemia may trigger angina pectoris, intermittent claudication and even myocardial infarction. Severe and prolonged anemia may be associated with signs of cardiac failure, such as edema, and may lead to left-ventricular hypertrophy. Heart murmurs are a common sign in anemic patients. Arrhythmias and other electrocardiographic changes might occur. Patients with anemia also have a tendency towards low blood pressure.

Lung Dyspnea is one of the most important symptoms of anemia and its severity depends on the degree of anemia. Dyspnea initially occurs only during exertion or excitement but in case of more severe anemia also during rest. As part of a compensatory mechanism, both rate and depth of respiration increase. Thus tachypnoe is an important sign and indicates how the lungs try to compensate the insufﬁcient oxygenation of the blood by increasing the oxygen supply through increased ventilation. In case of additional left cardiac failure, pulmonary edema will occur and aggravate pulmonary symptoms.

Brain Decreased oxygen supply to the brain results in a broad spectrum of symptoms. Typical symptoms are dizziness, vertigo and headache. Anemia often results in a lack of concentration, impairs cognitive functions and is often associated with depression. In severe anemia and particularly in patients with pre-existent cerebrovascular disease, anemia may trigger cerebral infarction.

Kidney Kidney function can be affected by anemia. In case of mild anemia, sufﬁcient blood ﬂow can be maintained due to peripheral vasoconstriction. In case of more severe anemia, kidney function decreases and patients may develop signs of renal insufﬁciency, particularly edema and proteinuria. Because endogenous erythropoietin is primarily synthesized in the kidneys, impaired renal function may lead to insufﬁcient production of erythropoietin, thereby further worsening anemia.

Symptoms of anemia

311

Gastrointestinal tract Symptoms include those of the underlying disease such as ulcers or carcinomas or might be caused by anemia itself. Glossitis is a common symptom, seen both in patients with pernicious anemia and in those with irondeﬁciency. In patients with leukemias or in patients undergoing anticancer therapy, anemia might be associated with ulcerative lesions in the upper gastrointestinal tract. Anemia may lead to nausea, anorexia and malabsorption.

Sexual function Anemia might lead to loss of libido and impotence. Women may suffer from irregular menstrual cycles, amenorrhea and menorrhagia.

Skin Pallor of both skin and mucous membranes is a typical sign of anemia and can best be seen in the conjunctivae, mucous membranes of the mouth and pharynx, and in the skin of the palms. However, pallor can be caused by factors other than anemia such as cutaneous vasoconstriction, altered pigmentation, ﬂuid content of the subcutaneous tissue, and myxedema. Pallor is occasionally also seen in otherwise healthy persons. Pallor can be masked by cyanosis, jaundice, racial skin pigmentation, and vasodilatation. The skin is cold and patients complain about cold intolerance. Loss of both elasticity and tone of the skin, early graying of hair, and impaired nail growth and wound healing might also occur. Pallor may be associated with other signs of anemia and from these associations might indicate the cause of anemia. Pallor associated with icterus may indicate hemolytic anemia and pallor associated with petechiae may be a sign of acute leukemia. Patients with anemia due to iron-deﬁciency may suffer from impaired growth of the nails.

Quality of life Anemia has a detrimental effect on the overall quality of life (Cella 1997; Vogelzang et al. 1997; Yellen et al. 1997; Portenoy and Itri 1999; Curt et al. 2000; Stone et al. 2000; Cella et al. 2003). Quality of life is made up of physical, functional, emotional, social and spiritual well-being (Portenoy and Itri 1999; Curt et al. 2000; Stone et al. 2000). Anemia particularly reduces physical, functional and emotional domains (Cella 1997; Yellen et al. 1997; Cella et al. 2003). Changes in Hb levels correlate with changes in the quality of life.

312

R. Pirker

In patients with cancer, anemia is an important etiological factor in the pathogenesis of cancer-related fatigue which is among the most common symptoms experienced by cancer patients and has the greatest impact on their overall quality of life (Cella 1997; Vogelzang et al. 1997; Yellen et al. 1997; Groopman and Itri 1999; Portenoy and Itri 1999; Curt et al. 2000; Stone et al. 2000; Cella et al. 2003; Ludwig et al. 2004). The incidence and severity of fatigue depend on the characteristics of the patient, malignancy and type as well as intensity of treatment (Portenoy and Itri 1999; Curt et al. 2000; Stone et al. 2000). The etiology of fatigue is multifactorial with anemia being one of the most important etiological factors. The degree of fatigue correlates with Hb levels and successful treatment of anemia usually reduces fatigue. Other etiological factors of fatigue are underlying cancer and its treatment, infection, malnutrition, dehydration, metabolic disturbances, endocrine disorders, organ failure, immobility, pain, cytokines, anxiety and depression and drugs. Symptoms of cancer-related fatigue are summarized in Table 2. For diagnosis of cancer-related fatigue ﬁve or more of these symptoms have to be present daily for at least 2 weeks in the absence of any psychiatric illness. Typical complaints of patients with cancer-related fatigue are generalized weakness, easy tiring, diminished concentration, sleep disturbances including non-restorative sleep, memory problems, and a need to struggle to overcome fatigue. Fatigue is often associated with anxiety and depression (Stone et al. 2000). Most importantly, fatigue reduces the patient’s ability to perform normal daily activities and impairs social, occupational and other areas of functioning (Cella 1997; Yellen et al. 1997). Several multidimensional quality of life instruments have been developed. They allow a detailed evaluation of anemia and its impact on quality

Table 2. Cancer-related fatigue Presence of at least 5 of the following symptoms: Generalized weakness Decreased motivation Decreased concentration Sleep disturbances Non-restorative sleep Memory problems Frustration Problems with daily activities Need to struggle to overcome fatigue Prolonged malaise after exertion Daily for 2 weeks

Symptoms of anemia

313

of life. Widely used instruments include the Linear Analog Scale Assessment (LASA) and the Functional Assessment of Cancer Therapy-Anemia (FACTAnemia) scales (Cella 1997; Yellen et al. 1997). The FACT-Fatigue (FACTF) consists of the Functional Assessment of Cancer Therapy-General (FACT-G) plus 13 fatigue items (the Fatigue Subscale), and the FACTAnemia (FACT-An) consists of the FACT-F plus seven items evaluating other concerns related to anemia, but unrelated to fatigue (Cella 1997; Yellen et al. 1997). These questionnaires have been validated but are currently mainly used only in the context of clinical trials. The FACT-An, FACT-F and the Fatigue Subscale allow to discriminate patients based on Hb level and Eastern Cooperative Oncology Group (ECOG) performance status (Cella et al. 2003). Patients with Hb levels greater than 12 g/dL experienced less fatigue, fewer non-fatigue anemia symptoms, better physical well-being, better functional well-being, and higher general quality of life (Cella et al. 2003). Because of the importance of anemia for cancer-related fatigue, prevention or treatment of anemia plays a major role in the management of patients with cancer (Seidenfeld et al. 2001; Rizzo et al. 2002; Vansteenkiste et al. 2002).

Anemia and clinical outcome Anemia is a prognostic factor for shorter survival and/or a predictive factor for worse outcome of anticancer therapy in several cancers (Caro et al. 2001). Anemia can reduce acceptance or tolerance of anticancer therapy. Whether correction of anemia improves outcome of anticancer therapy including survival is currently a matter of debate (Bohlius et al. 2006).

References 1. Cotran RS, Kumar V, Collins T (1999) Red cell and bleeding disorders. In: Cotran RS, Kumar V, Collins T (eds.) Robbins pathologic basis of disease. WB Saunders Company, Philadelphia pp. 601–43 2. Bohlius J, Wilson J, Seidenfeld J, et al (2006) Recombinant human erythropoietins and cancer patients: updated meta-analysis of 57 studies including 9353 patients. J Natl Cancer Inst 98: 708–14 3. Caro JJ, Salas M, Ward A, et al (2001) Anemia as an independent prognostic factor for survival in patients with cancer: a systemic, quantitative review. Cancer 91: 2214–21 4. Cella D (1997) The Functional Assessment of Cancer Therapy-Anemia (FACTAn) Scale: a new tool for the assessment of outcomes in cancer anemia and fatigue. Semin Hematol 34 [Suppl 2]: 13–9

314

R. Pirker

5. Cella D, Zagari MJ, Vandoros C, et al (2003) Epoetin alfa treatment results in clinically signiﬁcant improvements in quality of life in anemic cancer patients when referenced to the general population. J Clin Oncol 21: 366–73 6. Curt GA, Breitbart W, Cella D, et al (2000) Impact of cancer-related fatigue on the lives of patients: new ﬁndings from the Fatigue Coalition. Oncologist 5: 353–60 7. DeRienzo DP, Saleem A (1990) Anemia of chronic disease: a review of pathogenesis. Tex Med 86: 80–3 8. Dowlati A, R’Zik S, Fillet G, et al (1997) Anaemia of lung cancer is due to impaired erythroid marrow response to erythropoietin stimulation as well as relative inadequacy of erythropoietin production. Br J Haematol 97: 297–9 9. Fisher JW (2003) Erythropoietin: physiology and pharmacology update. Exp Biol Med 228: 1–14 10. Groopman JE, Itri LM (1999) Chemotherapy-induced anemia in adults: incidence and treatment. J Natl Cancer Inst 91: 1616–34 11. Krantz SB (1994) Pathogenesis and treatment of the anemia of chronic disease. Am J Med Sci 307: 353–9 12. Ludwig H, Strasser K (2001) Symptomatology of anemia. Semin Oncol 28 [Suppl 8]: 7–14 13. Ludwig H, Van Belle S, Barrett-Lee P, Birgegard G, Bokemeyer C, Gascon P, Kosmidis P, Krzakowski M, Nortier J, Olmi P, Schneider M, Schrijvers D (2004) The European Cancer Anaemia Survey (ECAS): a large, multinational, prospective survey deﬁning the prevalence, incidence, and treatment of anaemia in cancer patients. Eur J Cancer 40: 2293–306 14. Pirker R, Wiesenberger K, Minar W (2005) Cancer-related anemia: clinical relevance and treatment strategies. Am J Cancer 4: 233–45 15. Portenoy RK, Itri LM (1999) Cancer-related fatigue: guidelines for evaluation and management. Oncologist 4: 1–10 16. Rizzo JD, Lichtin AE, Woolf SH, et al (2002) Use of epoetin in patients with cancer: evidence-based clinical practice guidelines of the American Society of Clinical Oncology and the American Society of Hematology. J Clin Oncol 20: 4083–107 17. Seidenfeld J, Piper M, Flamm C, et al (2001) Epoetin treatment of anemia associated with cancer therapy: a systematic review and meta-analysis of controlled clinical trials. J Natl Cancer Inst 93: 1204–14 17. Stone P, Richards M, A’Hern R, et al (2000) A study to investigate the prevalence, severity and correlates of fatigue among patients with cancer in comparison with a control group of volunteers without cancer. Ann Oncol 11: 561–7 19. Vansteenkiste J, Pirker R, Massuti B, et al (2002) Double-blind, placebocontrolled, randomized phase III trial of darbepoetin alfa in lung cancer patients receiving chemotherapy. J Natl Cancer Inst 94: 1211–20 20. Vogelzang NJ, Breitbart W, Cella D, et al (1997) Patient, caregiver, and oncologist perceptions of cancer-related fatigue: results of a tripart assessment survey. The Fatigue Coalition. Semin Hematol 34 [Suppl 2]: 4–12 21. World Health Organization (1968) Nutritional anemias. Reports of a WHO scientiﬁc group. Geneva

Symptoms of anemia

315

22. Yellen SB, Cella DF, Webster K, et al (1997) Measuring fatigue and other anemiarelated symptoms with the Functional Assessment of Cancer Therapy (FACT) measurement system. J Pain Symptom Manage 13: 63–74 Correspondence: Robert Pirker, MD, Department of Medicine I, Medical University of Vienna, Währinger Gürtel 18-20, 1090 Vienna, Austria, E-mail: [email protected]

Chapter 13

Impact of anemia and red blood cell transfusion on organ function M. R. Nowrousian Department of Internal Medicine (Cancer Research), West German Cancer Center, University Hospital of Essen, Essen, Germany

Introduction The immediate effect of anemia is a decreased oxygen-carrying capacity of the blood and, consequently, a decreased supply of oxygen to various organs. The resultant tissue hypoxia evokes a number of compensatory mechanisms and a series of physiological and metabolic abnormalities that are responsible for the clinical signs and symptoms of anemia and cause limitations in physical and mental well-being of patients and their quality of life (QOL). Recent clinical studies show that anemia has much greater impacts on tissue and organ functions than previously thought, and that chronic anemia may be a risk factor not only for morbidity, but also for mortality. In this overview, the results of these studies will be discussed.

Compensatory mechanisms in anemia Oxygen delivery to tissue depends on the following factors: 1) level of hemoglobin (Hb) in the peripheral blood, 2) degree of saturation of Hb with oxygen, 3) Hb-oxygen dissociation curve, and 4) tension of oxygen in the tissue. When the Hb level decreases, certain compensatory mechanisms, such as changes in the Hb-oxygen dissociation curve and cardiac output, occur to maintain the oxygen delivery to the tissue (Fig. 1).

Hemoglobin-oxygen dissociation curve The ﬁrst adaptive mechanism in response to anemia is the shift of the Hboxygen dissociation curve to the right. It results from an increased concentration of organic phosphates, especially 2,3-diphosphoglycerate (2,3-DPG) within the red blood cells (RBCs) and leads to a decreased afﬁnity of Hb for

318

M. R. Nowrousian

Anemia (decreased RBC mass) Tissue hypoxia Peripheral vasodilatation Decreased peripheral vascular resistance Hypoviscosity

Shift of hemoglobinoxygen dissociation curve to the right

Maintenance of oxygen delivery to tissue

Increased cardiac output (rate + volume)

Cardiac remodeling: Left ventricular hypertrophy and dilatation, cardiac failure Fig. 1. Cardiovascular consequences of chronic anemia

oxygen. The consequence is an increased dissociation and delivery of oxygen at the tissue level without the necessity of limiting oxygen pressure. The negative effect of the decreased RBC mass on oxygen pressure, on the other hand, can be offset by selective shunting of blood from nonvital or less oxygen-sensitive donor areas to oxygen-sensitive organs, such as myocardium, brain, and muscles. Donor areas in moderate acute anemia are the mesenteric and iliac beds and in chronic anemia cutaneous tissues and kidney (Table 1) (Varat et al. 1972; Zander 1978; Rossi 1994; Schleiffenbaum 1994; Habler and Messmer 1997; Hébert et al. 2004). In conditions of rest and at Hb levels above 10 g/dl, the shift in Hb-oxygen dissociation curve and shunting of blood are usually sufﬁcient to compensate for tissue hypoxia, but in nonresting conditions and at Hb levels of 10 g/dl or less, additional mechanisms, particularly those of the cardiovascular system are required to meet the oxygen need of tissues (Varat et al. 1972; Metivier et al. 2000).

Cardiovascular system The most important mechanisms to compensate for the decreased oxygencarrying capacity of the blood are those of the cardiovascular system, known

Impact of anemia and red blood cell transfusion on organ function

319

Table 1. Arteriovenous difference in oxygen tension within various organs Organ

Arteriovenous difference in oxygen tension (mL dl−1)

Kidney Liver Brain (cortex) Myocardium Total body

1.5 4.0 9.0 12.0 5.0

Adapted from Zander 1978; and Habler and Messmer 1997.

as the high-output hyperkinetic circulatory response. They include: 1) decrease in peripheral vascular resistance, and 2) increase in stroke volume and heart rate (cardiac index) (Fig. 1) (Varat et al. 1972; Rosner and Grünewald 1997).

Peripheral vascular resistance Decreased peripheral vascular resistance is, beside the shift in the Hb-oxygen dissociation curve, another early adaptive mechanism in response to anemia. It results from the reduced blood viscosity due to reduced RBC count and the arterial dilatation as a consequence of tissue hypoxia and locally increased lactate production, as well as accumulation of vasoactive substances such as adenosine and bradykinin and decreased inhibition of basal endothelium-derived relaxing factor activity (Varat et al. 1972; Anand et al. 1993; Metivier et al. 2000). The decreased peripheral vascular resistance is usually associated with a decreased afterload and a low blood pressure. The latter may be a stimulus for an increased production of noradrenalin, renin, aldosterone, and atrial natriuretic peptide, which have been reported to occur in patients with chronic severe anemia and edema. Beside cardiac failure, the low blood pressure may be a causative mechanism of salt and water retention in these patients (Anand et al. 1993). In chronic anemia, there is also an increased preload as a result of a decreased venous return resistance due to reduced red blood cell count and blood viscosity (Metivier et al. 2000). The target of the decreased peripheral vasodilatation is an increased blood ﬂow to the periphery. Together with the reduced blood viscosity, related to the decreased RBC mass, peripheral vasodilatation and decreased

320

M. R. Nowrousian

peripheral vascular resistance may be the most important factors in increasing cardiac output (Fig. 1) (Varat et al. 1972).

Cardiac output Increased cardiac output is the most important compensatory mechanism in response to anemia. Indeed, cardiac output inversely correlates with Hb or hematocrit level. The two possibly most important mechanisms involved in increased cardiac output are: 1) reduced blood viscosity, and 2) increased sympathetic stimulation of the cardiovascular effectors. While blood viscosity exerts its effects predominantly on preload and afterload as major determinants of cardiac output, sympathetic stimulation primarily increases heart rate and contractility, as well as venomotor tone. In chronic anemia, as well as other normovolemic anemias, as opposed to hypovolemic anemia, the effects of blood viscosity appear to predominate (Metivier et al. 2000; Hébert et al. 2004). In severe chronic anemia, cardiac output at rest is almost always elevated, generally in a roughly linear fashion with increasing severity of anemia. Exceptions occur in patients with underlying heart disease, and patients with hyperviscosity caused by a high serum globulin concentration, as in patients with Waldenström’s macroglobulinemia or multiple myeloma (Varat et al. 1972). The function of the increased cardiac output is to decrease the fraction of oxygen that has to be extracted by the tissue. Due to the adaptive effect of the Hb-oxygen dissociation curve, cardiac output at rest generally increases, when Hb level falls below 7 g/dl. The more reliable indicator of the altered hemodynamics is, therefore, the excessive response of cardiac output to exercise, which is typically greater than normal and frequently present, even in patients with Hb levels as high as 10 g/dl (Varat et al. 1972). In contrast to normal subjects and patients with acute anemia, the increased cardiac output in patients with chronic anemia principally results from increased cardiac stroke volume and not from increased heart rate, a phenomenon that explains the frequently missing tachycardia in these patients. The increased cardiac output, however, is not always completely compensatory, since total oxygen transport, which is a product of cardiac output und arterial oxygen content, is somewhat less than normal (Varat et al. 1972). In a randomized study in patients undergoing open heart surgery, postoperative anemia was found to be associated with lactacidosis as a result of a considerably reduced delivery of oxygen. In this study, the incidence and severity of lactacidosis appeared to be lower in patients who received rhEPO compared to those who received placebo, despite the signiﬁcantly higher amounts of RBC transfusions that were given to these patients (Sowade

Impact of anemia and red blood cell transfusion on organ function

321

et al. 1997). In chronic anemia, tissue oxygenation has been found to be in the lower normal range, but to improve signiﬁcantly after treatment of anemia (Nonnast-Daniel et al. 1990).

Symptoms and clinical ﬁndings of anemia Anemia is usually tolerated much better, if the Hb level has decreased slowly, as in patients with chronic diseases, than rapidly, as in those with severe acute bleeding or hemolysis. The clinical signs and symptoms of anemia, however, do not only depend on the degree and the rapidity of changes in RBC mass, but also on the age of patients and their organ functions, particularly on the capacity of the cardiopulmonary system to compensate for the reduced RBC mass. In cancer patients, even mild to moderate anemia can evoke symptoms, and severe anemia may be life-threatening, since most of these patients are of advanced age (Nowrousian et al. 1996) and many of them have decreased organ functions, related either to the malignant disease itself or its treatment or comorbidities, which these patients are frequently suffering from (Table 2).

Table 2. Factors potentially aggravating symptoms and signs of anemia in cancer patients Patient-related factors: Advanced age (median > 60 yrs) Multimorbidity (particularly reduced cardiopulmonary function) Malignancy-related factors: Tumor invasion Organ obliteration (e.g. lung cancer) External compression (e.g. pleural effusion) Organ insufﬁciency (e.g. myeloma kidney) Expansion of plasma volume (e.g. in multiple myeloma or Waldenström’s disease) Hemolysis (e.g. in CLL) Therapy-related factors: Cardiotoxicity (e.g. anthracyclines) Pulmonary toxicity (e.g. bleomycin, BCNU) Renal toxicity (e.g. cisplatinum) Infections (e.g. fever, pneumonia) Bleeding (e.g. surgical interventions, coagulopathy)

322

M. R. Nowrousian

Symptoms Signs and symptoms of anemia generally refer either to the compensatory mechanisms of the cardiovascular system or to the decreased RBC mass and oxygen delivery to the tissue. Palpitation and pounding pulse are signs of increased cardiac output, while pallor, postural hypotension, vertigo, and syncope mainly relate to the decreased RBC mass and decreased peripheral vascular resistance. Dyspnea, headache, tinnitus, sleeping disorders, lethargy, depression, limitations in cognitive function, transient cerebral ischemia, angina pectoris, weight loss, reduced exercise capacity and fatigue are predominantly related to the decreased organ and tissue oxygenation (Butt and Cella in this book, Chowdhury et al. in this book, Pirker in this book) (Fig. 2). In chronic anemia, fatigue, weakness, and reduced physical, as well as mental activities, are the most common complaints (Table 3) (Cella 1997; Macdougall 1998; Cella and Bron 1999; Barrett-Lee et al. 2000; Glaus and Müller 2000; Butt and Cella in this book; Chowdhury et al. in this book; Pirker in this book). They result from a number of physiological and metabolic abnormalities, which are the consequences of prolonged cardiac stress and tissue hypoxia. Unlike pain, symptoms of anemia, particularly fatigue, may be easily overlooked or not appropriately recognized by clinicians, but from the view-point of patients, they appear to be more debilitating than pain

Anemia Tissue hypoxia

RBC transfusions

CardioCNS vascular system

Metabolic functions

Sexual organs

Immune system

Iron overload

Vasodilatation, Increased cardiac workload

Decreased circulation and metabolism

Decreased anabolic metabolism, insulin resistance, hyperlipidemia, K +disbalance

Decreased function of ovary and testis

T cell suppression, decreased B cell function

Low blood pressure, LV hypertrophy and dilatation, cardiac failure, dyspnea, decreased exercise capacity, edema, fatigue

Headache, tinnitus, vertigo, syncope, lethargy, depression, difficulty in sleeping and concentration, decreased cognitive function, fatigue

Hemosiderosis, decreased phagocytic and killing activity of neutrophils

Lack of appetite, anorexia, muscle wasting, decreased exercise capacity, edema, fatigue

Menstrual abnormalities, erectile problems, loss of libido and fertility

Increased susceptibility to infections

Fig. 2. Impact of anemia on organ functions

Impact of anemia and red blood cell transfusion on organ function

323

Table 3. Symptoms and clinical ﬁndings in anemia Symptoms: Cardiovascular system: Palpitation; dyspnea on exertion, angina pectoris; decreased exercise capacity; fatigue CNS: Headache; tinnitus; vertigo; syncope; lethargy; depression; difﬁculty in concentration; decreased cognitive function; sleeping disorders; transient cerebral ischemia Metabolic functions: Lack of appetite; weight loss; muscle wasting; edema; decreased exercise capacity; fatigue Sexual organs: Males: Erectile problems, loss of libido and fertility Females: Menstrual abnormalities, loss of libido and fertility Clinical ﬁndings: Pallor of the skin, nailbeds and mucous membranes; tachycardia:usually present in acute anemia, but not in chronic anemias; pounding pulse; low blood pressure; heart enlargement with leftward displacement of apical impulse; ﬁrst and second heart sounds louder than normal; systolic bruits over the carotid and subclavian arteries; cervical venous hum with diastolic accentuation; edema Chest roentgenogram: cardiomegaly Echocardiogram: increased left volume index

(Vogelzang et al. 1997). As will be discussed below, not only fatigue, but also cardiac complications, inadequate nutrient intake and anorexia, muscle wasting, decreased exercise capacity, sexual dysfunctions and a number of other physical and mental complaints are symptoms of anemia (Figs. 1, 2, Table 3). Due to the belief that such symptoms are related to the malignant disease or its treatment and have to be endured, many patients do not talk about the limitations in their physical well-being and QOL with their physicians (Vogelzang et al. 1997). The dimension of these limitations often becomes evident after anemia has been treated with rhEPO and sufﬁcient and sustained increases in Hb level have been achieved. Another common complaint is dyspnea on exertion, which is the result of lactate acidosis that occurs earlier and at lower levels of exercise in patients with anemia than in nonanemic subjects (Varat et al. 1972; Schleiffenbaum 1994; Barrett-Lee et al. 2000). Palpitation and edema can also occur. The latter is the consequence of salt and water retention, which relates either to cardiac failure or increased neurohormonal activity or both (Anand et al.

324

M. R. Nowrousian

1993). In elderly patients, mental confusion may be a prominent symptom, particularly, when anemia is severe. Dyspnea at rest, orthopnea, distended neck veins, hepatomegaly, and extensive edema can also occur, but usually as a result of prolonged anemia or in association with additional heart disease. They indicate cardiac failure and circulatory congestion. Angina pectoris and myocardial infarction may also occur, but usually in connection with additional coronary artery disease (Varat et al. 1972; Sabatine et al. 2005).

Clinical ﬁndings A characteristic ﬁnding in anemia is the pallor of the skin, nailbeds, and mucous membranes. Tachycardia is usually present in acute anemia, but often absent in chronic anemia. The arterial pulse can be pounding, reﬂecting a somewhat widened pulse pressure. There can be an enlargement of the heart with a leftward displacement of the apical impulse. The ﬁrst and second heart sounds are usually louder than normal, and a systolic ejection murmur can often be heard in the second left intercostal space, along the lower left sternal border or at the apex. There may be loud systolic bruits over the carotid and subclavian arteries. A common ﬁnding is a cervical venous hum as a continuous murmur with diastolic accentuation, best heard over the neck, in the supraclavicular area and, occasionally, over the upper precordium, when the patient is in sitting position and his chin is turned to the contralateral side. Occasionally, there may be slight ankle edema (Varat et al. 1972). Severe edema can occur in case of chronic severe anemia (Table 3) (Anand et al. 1993). There are no characteristic electrocardiographic changes in anemia. The chest roentgenogram may reveal cardiomegaly, particularly in elderly patients and patients with severe and prolonged anemia. In these patients, echocardiography usually shows an increased left ventricular (LV) volume index as a result of LV dilatation (Varat et al. 1972; Foley et al. 2000).

Impact of anemia on organ function Clinical studies in patients with end-stage renal disease (ESRD) and patients with cancer have shown that anemia has a broad spectrum of impacts on various organs, and the sequelae of anemia are much greater than recognized previously (Figs. 1, 2) (Table 4). An important factor in evaluating the adverse effects of anemia on organ function has been the introduction of rhEPO, which allows to discriminate between the impacts of anemia and those from other debilitating factors, such as uremia or cancer.

Impact of anemia and red blood cell transfusion on organ function

325

Table 4. Impact of anemia on organ functions Cardiovascular system: Vasodilatation due to locally increased lactate production and accumulation of vasoactive substances; decreased peripheral vascular resistance; increased cardiac output; cardiac remodeling; cardiac failure; arterial remodeling; decreased hemostatic function CNS: Decreased circulation and metabolism Metabolic functions: Decreased utilization of ingested protein; decreased metabolism of amino acids and alpha-keto acid; insulin resistance; hyperlipidemia; increased oxidative stress; impairment of potassium (K+) regulation, probably related to depressed Na +K +-ATPase activity or exercise activity or both Sexual organs: Males: Decreased levels of total and free testosterone; increased levels of LH and FSH, but decreased LH/FSH ratio; decreased ejaculate volume with partial or complete azoospermia; decreased fertility Females: Absence of preovulatory peak of LH and estradiol; menstrual abnormalities including amenorrhea, decreased fertility Immune system: Decreased T cell and B cell function; decreased phagocytic function and intracellular killing activity of neutrophils due to iron overload resulting from repeated RBC transfusions

Cardiovascular system Chronic anemia has a number of deleterious effects on the heart and is a major risk factor for developing left ventricular hypertrophy (LVH) and dilatation, systolic and diastolic dysfunctions and, ﬁnally, cardiac failure (Macdougall et al. 1990; Foley et al. 1996, 2000a, b; Kausz et al. 2000; Levin and Foley 2000; Rigato et al. 2003; Das et al. 2005; Li et al. 2005; Oliva et al. 2005; Sandgren et al. 2005; Srivastava et al. 2006). The pathogenic mechanism of the LVH is primarily an increased cardiac workload in response to the decreased oxygen-carrying capacity of the blood and the resulting tissue hypoxia (Fig. 1). Chronic anemia usually leads to eccentric LVH (enlargement of the ventricular chamber plus increase in wall thickness), in contrast to concentric LVH (increase in ventricular wall thickness at the normal chamber radius), which is often the result of a sustained arterial hypertension or pressure overload (Varat 1972; Grossman et al. 1975; Levin and Foley 2000). Eccentric LVH, also called cardiac remodeling, increases the hemodynamic stresses on the walls of the heart and depresses its mechanical performance. It may also induce mitral regurgitation. It usually precedes clinical

326

M. R. Nowrousian

symptoms by months or even years and may be persistent despite symptom control because of permanent damages to the cardiac myocytes with myocardial ﬁbrosis (Hamblin 2005). Another effect of increased cardiac output in chronic anemia may be arterial remodeling of central elastic-type arteries, such as the aorta and the common carotid arteries. Arterial remodeling is characterized by arterial enlargement and compensatory thickening of the arterial intima-media. This may in turn aggravate the development of LVH through arterial stiffening with an abnormal increase in systolic pressure (Metivier et al. 2000). Cardiac remodeling has not yet been systematically evaluated in cancer patients, but it appears to be a frequent phenomenon in patients with refractory anemia due to myelodysplastic syndrome. In a study using echocardiographic tests, 62% of these patients had cardiac remodeling. The proportion of patients with cardiac remodeling was 92% in those who were transfusiondependent and 48% in those who were not (Oliva et al. 2005). In this study, Hb independently indicated cardiac hypertrophy with each unit increase predicting a 49% reduction in the risk of cardiac remodeling. The signiﬁcant inverse correlation between Hb level and LVH has also been evaluated in hemodialysis patients with chronic anemia and cardiomyopathy. In these patients, each 1 g/dl fall in Hb was associated with an increase in the cavity volume index of 8 ml/m2 (Foley et al. 2000). In hemodialysis patients, there is of course a number of other factors that can also contribute to the development of cardiovascular complications (Sarnak and Levey 2000), but the main factors underlying the development of cardiomyopathy are anemia and arterial hypertension. In a study evaluating the relationship between Hb level and cardiomyopathy, each 1 g/dl decrease in Hb level was found to be associated with the presence of LV dilatation on repeat echocardiogram and the development of de novo and recurrent cardiac failure. Furthermore, the decrease in Hb level was found to be associated with an increase in mortality, independent from blood pressure or other risk factors (Foley et al. 1996). Other mortality studies in patients with chronic renal failure also show that anemia generally increases the risk of death in these patients and that a decrease of Hb level below 11 g/dl is associated with an 18% to 40% increased risk of hospitalization and death (Collins et al. 2000, 2001; Eckardt 2005; Robinson et al. 2005; Portoles et al. 2007). Since cardiomyopathy is an early event, as well as a major cause of death in these patients, it is of particular interest that treatment of anemia with rhEPO can result in a regression of the LVH, and prevention of LV dilatation correlated with increments in Hb level (Besarab et al. 1998; Levin and Foley, 2000). Not only in patients with ESRD, but also in those with congestive heart failure (CHF), which is frequently associated with anemia, the latter has been found to be associated with an increased risk of death and hospitalization (Silverberg et al. 2000, 2001, 2006, and in this book; Anand et al. 2004; De Maria et al. 2005; Komajda et al. 2006). In addition, there are prospectively

Impact of anemia and red blood cell transfusion on organ function

327

randomized studies in these patients showing that treatment with rhEPO not only improves cardiac function, but also renal function and exercise capacity and decreases the use of diuretics and the need for hospitalization (Silverberg et al. 2000, 2001, 2006 and in this book; Palazzuoli et al. 2006; Mitchell 2007; Ponikowski et al. 2007; Tang 2007).

CNS In patients with cancer and in those with chronic renal failure, anemia has been shown to be associated with a signiﬁcant decrease in QOL, and treatment of anemia with erythropoietic stimulating proteins (ESPs), such as epoetin alfa or beta or darbepoetin, has been found to improve QOL (Abels 1992; Levin 1992; Leitgeb et al. 1994; Revicki et al. 1995; Glaspy et al. 1997; Demetri et al. 1998; Macdougall 1998; Cella and Bron 1999; Glaus and Müller 2000; Moreno et al. 2000; Gabrilove et al. 2001; Littlewood et al. 2001, 2003, 2006; Fallowﬁeld et al. 2002; Österborg et al. 2002; Vansteenkiste et al. 2002; Boogaerts et al. 2003; Cella et al. 2003; Hedenus et al. 2003; Iconomou et al. 2003; Jacobsen et al. 2004; Jones et al. 2004; Chang et al. 2005; Cortesi et al. 2005; O’Shaughnessy et al. 2005; Savonije et al. 2006; Lefebvre et al. 2006; Massa et al. 2006; Straus et al. 2006; Ossa et al. 2007; Butt and Cella in this book; Lyman and Glaspy in this book). Improvements in QOL, on the other hand, have been reported to be in part related to improvements of cognitive function and fatigue (Stivelman 2000; O’Shaughnessy et al. 2005; Massa et al. 2006). In patients with chronic renal failure, several studies using various neurophysiological and neuropsychological tests have shown that the beneﬁcial effect achieved by the treatment of anemia is independent from the effect of dialysis (Grimm et al. 1990; Marsh et al. 1991; Sagales et al. 1993; Pickett et al. 1999; Stivelman 2000). Among the neurophysiological tests used in these studies, the cognitive event-related potential (ERP) has been found to be particularly useful in evaluating the cognitive function (Pickett et al. 1999; Stivelman 2000). In this test, similar to other neurophysiological tests, electrical impulses from the brain are measured in the presence or absence of speciﬁc stimuli. Of particular interest are the latency and the amplitude of a positive wave, which occur approximately 300 milliseconds after the onset of target stimuli (P300). The latency of P300 is related to the time required for cognitive processing and the amplitude of this wave is related to the allocation of processing resources to a task. Diminished amplitudes typically indicate reduced attention (Pickett et al. 1999). In studies using this test, treatment of anemia with rhEPO has been shown to signiﬁcantly improve the latency or the amplitude of P300 or both (Stivelman 2000; Singh et al. 2006). In other studies, the use of rhEPO has been reported to produce a signiﬁcant decrease in electroencephalogram slowing and to be associated with improvements in the circulation and metabolism of the brain (Metry et al. 1999; Pickett et al. 1999).

328

M. R. Nowrousian

Another effect of treating anemia with rhEPO has been found to be improvement in sleeping disorders such as periodic limb movements in sleep, arousals from sleep and sleep fragmentation (Benz et al. 1999). These studies indicate that chronic anemia has major impacts on CNS as an organ with a particularly high sensitivity to hypoxia (Table 1). Furthermore, they show that treatment of anemia with rhEPO signiﬁcantly improves brain and cognitive function, an effect, which may be substantially involved in the improvement of QOL both in patients with cancer, as well as in those with ESRD. In these two groups of patients, the use of rhEPO may also have a direct effect on the brain tissue as a substitute for the defective endogenous EPO production, which is usually present in these patients. Recent studies show that EPO and its receptor are expressed in various organs including neuronal tissue, and that EPO plays a critical role in neuronal survival after hypoxic and other metabolic injuries and in inﬂammatory processes (Table 5) (Sakanaka et al. 1998; Brines et al. 2000; Chin et al. 2000; Cerami et al. 2001; Dame et al. 2001; Siren et al. 2001; Jelkmann and Wagner 2004; Santhanam and Katusic 2006; Jelkmann 2007; Ghezzi and Mengozzi 2007; Ehrenreich and

Table 5. Effects of EPO on various tissues and organ systems Erythropoietic system: Survival, proliferation and differentiation of erythroid progenitor cells (BFU-E, CFU-E, erythroblasts) Expression of transferrin receptor and activation of iron regulatory protein in erythroid progenitor cells Vascular system: Promotion of vascular endothelial cells and maturation of new vessels Immune system: Increase in immunoglobulin production by B cells, independent as well as dependent on an effect on T cells Increase in expression of complement receptor type 1 on the surface of erythrocytes Enhancement of the effect of G-CSF on granulopoiesis T cell-mediated antitumor effect in myeloma model CNS: Protection and recovery of neuronal tissue from hypoxic damage and other toxic injuries by inhibition of apoptosis Myoblasts: EPO stimulates proliferation and differentiation of myoblasts and may be potentially involved in muscle development and repair BFU-E = burst-forming unit erythroid; CFU-E = colony-forming unit erythroid; G-CSF = granulocyte colony-stimulating factor.

Impact of anemia and red blood cell transfusion on organ function

329

Bartels in this book). The mechanism underlying this effect appears to be the inhibition of neuronal apoptosis by the activation of a variety of genes and their consecutive protein production (Dame et al. 2001; Siren et al. 2001; Ehrenreich and Bartels in this book). The neuroprotective potential of EPO found in these studies may be of interest not only in improving cognitive function in anemic patients with cancer or ESRD, but also in preventing or treating cognitive side effects of cranial irradiation, chemotherapy, and certain cytokines, such as interferon-α in cancer patients (Abayomi 1996; Valentine et al. 1998; Brezden et al. 2000; Ehrenreich and Bartels in this book).

Metabolic functions and exercise capacity Malnutrition, wasting, and decreased exercise capacity are common ﬁndings both in patients with cancer and in those with chronic renal failure (Barany et al. 1991; Clyne and Jogestrand 1992; Daneryd et al. 1998). In the latter group of patients, a number of metabolic abnormalities has been observed and was previously thought to be mainly related to uremia. The reversing effect of treatment with rhEPO, however, has indicated that the more important factor responsible for such abnormalities and for the decreased exercise capacity is anemia. Decreased muscle strength, approximately half that of healthy subjects, is a frequent ﬁnding in patients with chronic renal failure and one of the principal limiting factors in exercise capacity of these patients (KettnerMelsheimer et al. 1987; Diesel et al. 1990; Fagher et al. 1994). A mechanism that appears to be involved in muscle wasting seems to be a decreased anabolic metabolism as a result of an increased proteinolysis and a decreased utilization of ingested protein and metabolism of amino acids and alpha-keto acid (Barany et al. 1991; Riedel et al. 2000). Other metabolic abnormalities observed in anemic patients with chronic renal failure are insulin resistance and hyperlipidemia with elevated levels of triglycerides, low-density lipoprotein, cholesterol, and apolipoprotein B. Insulin resistance, characterized by increased fasting serum insulin levels with normal fasting glucose values and increased insulin levels in response to oral or intravenous glucose, may be a contributing factor to hyperlipidemia, since insulin abnormalities may impair the activity of lipoprotein lipase (Chan et al. 1981; Chan 1990). As in children with chronic anemia, such as those with thalassemia major, who often receive RBC transfusions and are iron-overloaded, insulin resistance in patients with chronic renal failure was previously thought to be secondary to iron toxicity (Merkel et al. 1988). Treatment with rhEPO, however, has shown that regression of this abnormality and hyperlipidemia is related to a greater degree to correction of anemia rather than iron overload (Mak 1996, 1998). Both in patients with or without iron overload, the use of rhEPO reverses the insulin resistance and lipid abnormalities (Mak 1996, 1998; Spaia et al.

330

M. R. Nowrousian

2000; Igaki et al. 2004; Kadiroglu et al. 2005, 2006; Sit et al. 2005), probably resulting from increased plasma concentrations of lipoprotein lipase and hepatic triglyceride lipase, which occur after treatment with rhEPO (Goto et al. 1999). In addition, the use of rhEPO reduces oxidative stress in hemodialysis patients, as indicated by a decrease in plasma concentrations of lipid peroxidation products such as malondialdehyde and 4-hydroxynonenal (Sommerburg et al. 2000). Correction of anemia has also been shown to reduce plasma concentration of leptin, a hormone that regulates food intake and energy balance in humans (Considine et al. 1996). Leptin concentration is usually increased in patients with chronic renal failure and has also been found to be increased in patients with hematological malignancies including multiple myeloma, malignant lymphoma and acute leukemia (Alexandrakis et al. 2004; Pamuk et al. 2006). In the ﬁrst group of patients, increased leptin levels were postulated to be responsible for the lack of appetite (Kokot et al. 1998; Nishizawa et al. 1998; Stenvinkel 1998; Drueke et al. 1999), but recent studies indicate a paradoxically inverse association between increased serum leptin and markers of nutritional status (Bossola et al. 2004; Hung et al. 2005; Axelsson et al. 2005; Nasri 2006). Treatment with rhEPO, however, improves appetite and exercise capacity in anemic patients with renal failure, and patients with underweight will gain in weight (Barany et al. 1991; Clyne and Jogestrand 1992). Improvement in appetite has also been observed in anemic cancer patients treated with rhEPO (Leitgeb et al. 1994). In addition, in a prospective randomized study in patients with cancer and progressive cachexia, the use of rhEPO was found to prevent the development of anemia associated with a signiﬁcantly more preserved maximum exercise capacity, based on more effective ventilation and whole-body respiratory gas exchange, and a signiﬁcantly preserved metabolic efﬁciency, expressed as oxygen uptake per watt produced. In other words, treatment with rhEPO protected not only Hb, but also weight, metabolic function, and exercise capacity of becoming decreased (Table 6) (Daneryd et al. 1998; Lundholm and Daneryd in this book). Another metabolic abnormality in patients with ESRD that has been reported to improve after treatment with rhEPO is an impairment of potassium (K+) regulation, which may be the result of either a reduced exercise capacity or depressed Na+K+-ATPase activity that has been found to be associated with low endogenous EPO levels. Such an impaired K+ regulation may be a factor that contributes to limitations in physical performance of patients with chronic anemia (McMahon et al. 1999).

Sexual function The impact of anemia on sexual function has not yet been clearly deﬁned. In patients with chronic renal failure, however, there are study results indicat-

Impact of anemia and red blood cell transfusion on organ function

331

Table 6. Protection of metabolic and exercise capacity in cancer patients with progressive cachexia randomized to receive or not to receive rhEPO Study groups

Hb (g/dl) Body weight (kg) Oxygen uptake-R (ml/min) Oxygen uptake-M (ml/min) Carbon dioxide production-M (ml/min) Respiration (liter/min) Exercise capacity-M (W)

Control rhEPO Control rhEPO Control rhEPO Control rhEPO Control rhEPO Control rhEPO Control rhEPO

Months 0

10–30

12.1 11.9 67.1 67.3 222 229 1,154 1,288 1,071 1,262 37 41 72 86

10.8 13.0 55.6 71.1 188 224 790 1,366 688 1,330 23 42 51 107

<0.0001 <0.05 <0.005 <0.01 <0.009 <0.03 <0.0001

Hb = hemoglobin; R = resting condition; M = maximum exercise; W = the maximal mechanical power. Adapted from Daneryd et al. 1998; see also Lundholm et al. in this book.

ing that anemia may be in part responsible for the frequently observed sexual dysfunctions in these patients. Common disturbances are erectile problems in men, menstrual abnormalities including amenorrhea in women, and decreased libido and fertility in both sexes (Palmer 1999, 2003; Anantharaman and Schmidt 2007). They appear to be primarily organic in origin and to be related not only to uremia, but also to anemia and other comorbidities that frequently occur in patients with chronic renal failure. Disturbances in the hypothalamic-pituitary-gonadal axis have been reported to be already present before the need for dialysis. The sexual dysfunctions, however, rarely normalize with the initiation of hemodialysis or peritoneal dialysis but often continue to progress. In men, the volume of the ejaculate is decreased, with partial or complete azoospermia and a decreased percentage of mobile sperms. Plasma levels of total and free testosterone are reduced, although the binding capacity and concentration of sex-binding protein are normal. Further ﬁndings are increased concentrations of the pituitary gonadotropin luteinizing hormone (LH) and follicle-stimulating hormone (FSH), but there is typically a decreased LH/FSH ratio as a result of a more variable degree of increase in FSH concentration. In women, the preovulatory peak in plasma concentrations of LH and estradiol is frequently

332

M. R. Nowrousian

absent, and in both sexes, there is usually an increased plasma concentration of prolactin (Palmer 1999, 2003; Anantharaman and Schmidt 2007). In many patients with chronic renal failure, treatment of anemia with rhEPO improves the sexual function, along with the improvements in QOL, such as decreased fatigue, increased well-being and exercise capacity (Nissenson 1989; Canadian Erythropoietin Study Group 1990; Dowling 2007). There are, however, controversial results regarding the effects of treatment on hormones that regulate the sexual function (Lawrence et al. 1997). Therefore, it is unclear, whether the improved sexual function is a direct effect of rhEPO or an effect of anemia correction with improvements in the physical and psychological condition of patients, or both. However, there are studies showing that treatment of anemia with rhEPO may normalize the pituitary gonadal axis, lowering FSH, LH levels and prolactin and raising serum testosterone (Schaefer et al. 1994; Kokot et al. 1995; Tokgoz et al. 2001; Palmer 2003; Anantharaman and Schmidt 2007). In addition, in an in vitro study, EPO has been found to bind to EPO receptors on Leydig cells and to activate these cells (Yamazaki et al. 2004). Irrespective of these open questions, the fact is that chronic anemia related to a defective endogenous EPO production is able to induce sexual dysfunctions, and that treatment of anemia with rhEPO can reverse this sequel in a proportion of patients.

Immune system and other defence mechanisms Whether anemia itself has an impact on the immune system and other defense mechanisms is not clear. There are, however, two factors that may play a role in the susceptibility to infections in anemic patients with cancer or ESRD: 1) defective endogenous EPO production, 2) allogeneic RBC transfusions. Insufﬁcient production of endogenous EPO is the main causative mechanism of anemia in patients with ESRD and an important pathogenic mechanism of anemia in patients with cancer (Nowrousian et al. 1996; Eckardt 2000; Macdougall 2001). The former group of patients is known to be generally susceptible to infections, possibly related to an impaired immune function, as indicated by a decreased mitogen-induced proliferative activity and interleukin-2 production of T cells, an increased ratio of T helper cells to T suppressor cells, and a decreased antibody production by B cells (Senneseal et al. 1991; Singh et al. 1992; Birmingham et al. 1996; ShurtzSwirski et al. 1996; Bryl et al. 1998, 1999). Treatment of these patients with rhEPO has been shown to improve the immune function in terms of a signiﬁcant enhancement of mitogen driven T cell proliferation (Singh et al. 1992; Shurtz-Swirski et al. 1996) and a signiﬁcant increase in antibody production against T cell dependent antigens, such as hepatitis B vaccine or tetanus toxoid (Senneseal et al. 1991; Birmingham et al. 1996). No improvement has been observed regarding the production of antibodies against T cell

Impact of anemia and red blood cell transfusion on organ function

333

independent antigens, such as pneumococcal capsular polysaccharide (Birmingham et al. 1996). Further effects observed after treatment with rhEPO are a decrease in the total T cell count and an increase in the ratio of T helper cells to T suppressor cells (Pfaffel et al. 1988; Grimm et al. 1990; Senneseal et al. 1991). The mechanisms by which EPO may develop its effects on the immune response are still unknown. On the other hand, in vitro studies have shown that EPO increases the production of immunoglobulins by B cells, independent as well as dependent of an effect on T cells (Paczek et al. 1990; Kimata et al. 1991). Furthermore, EPO appears to increase the expression of complement receptor type 1 on the surface of erythrocytes (Hébert et al. 1994), a mechanism, which could result in an increased capacity to process circulating immune complexes in a way that they could serve as immunologic signals (Birmingham et al. 1996). In addition, EPO seems to have stimulatory effects on hematopoietic progenitor cells (Stockenhuber et al. 1990) and to enhance synergistically the effects of granulocyte colony-stimulating factor (G-CSF) on granulocytic progenitors (Souza et al. 1986; Negrin et al. 1996; Hellström-Lindberg et al. 1998). These mechanisms could possibly modulate antibody response by promoting antigen-presenting cells other than T cells. In a randomized study in patients with ovarian cancer who received chemotherapy, EPO was found to increase the hematologic effects of G-CSF in terms of a signiﬁcant decrease in severity of neutropenia and frequency of life-threatening infections, as well as a signiﬁcant increase in mobilization and collection of peripheral blood CD34+ hematopoietic progenitor cells (Pierelli et al. 1999). In a rat model, EPO has been found to enhance the healing of colonic anastomoses by increasing the number of ﬁbroblasts and accelerating the maturation of new vessels (Fatouros et al. 1999). Another mechanism, by which EPO could also inﬂuence the defense mechanisms, is the induction of transferrin receptor expression on the surface of erythroid progenitor cells and thus an increased uptake of iron in these cells (Weiss et al. 1997). The result is an increased mobilization of iron from iron stores, as indicated by a decrease of serum ferritin concentration in anemic patients with cancer or ESRD during treatment with rhEPO (Fig. 2) (Ludwig et al. 1990; Oberhoff et al. 1998; Eschbach and Adamson 1999). Although the role of iron in susceptibility to infection has not yet been clearly deﬁned (Hershko et al. 1988; Eschbach and Adamson 1999), the regulatory effect of EPO on iron utilization may be of protective value, at least in anemic patients with iron overload. In a recent study in mice, treatment with EPO was found to enhance immune responses consisting of increased immunoglobulin production and enhanced response to the clinically relevant hepatitis B antigen (Katz et al. 2007). In another study using two myeloma models of nonanemic mice, rhEPO was found to induce tumor regression and antitumor immune responses, mediated by T cells (Mittelman et al. 2001). The animals treated

334

M. R. Nowrousian

with rhEPO showed a marked decrease in mortality and a signiﬁcant increase in survival as compared with control animals. In a clinical study of anemic patients with multiple myeloma, treatment with rhEPO has been reported to prolong survival and to improve immunological abnormalities including normalization of the CD4 : CD8 cell ratio, enhanced T cell phytohemagglutininmediated activation and proliferation potential, T cell expression of the costimulatory CD28 and inhibitory CTLA-4 molecules, as well as reduced interleukin-6 serum values to normal levels. These abnormalities were found to be present, even in patients with early disease stages (Prutchi-Sagiv et al. 2006; Baz et al. 2007). An antitumor effect of rhEPO was also observed in a small group of patients with heavily pretreated, refractory or relapsed CLL, in whom treatment with rhEPO and GM-CSF not only corrected anemia, but also reduced the frequency of infections and led to a reduction of lymphocytosis, lymphadenopathy, and organomegaly with a progression-free time of at least 10 months (Russo et al. 1999). These observations indicate that EPO, beside its activity on the erythropoietic system, may have an immunemediated antitumor effect, which could be lacking in part in anemic cancer patients with defective endogenous EPO production (Table 4). As mentioned above, another risk factor for infectious complications in anemic cancer patients may be allogenic RBC transfusions. They not only are able to transfer infectious agents, but also to induce immunosuppression making the patients more susceptible to various types of infections.

Risks and limitations of RBC transfusion Although RBC transfusions have been used for many decades, there is no prospective randomized trial evaluating their beneﬁt and, particularly, their cost-effectiveness, and they have never been independently tested as a pharmaceutical would be to receive approval. However, a great deal of information is available on their risks and limitations. RBC transfusions are known to be associated with a broad spectrum of possible adverse effects including febrile, allergic and hemolytic reactions, immunomodulation and transmission of infections (Table 7). Transfusion services use a variety of donor screening interventions, including serologic assays, nucleic acid assays, and geographic exclusions to reduce the risk of infection and, indeed, such a risk has been considerably decreased with regard to major pathogens, such as hepatitis C and human immunodeﬁciency virus. However, the risk of infections with emerging or re-emerging pathogens, such as plasmodia, babesia, trypanosome cruzi, west Nile virus, dengue virus, parvovirus B19, human herpes virus-8, avian ﬂu virus, and prions is still a subject of concern (Alter et al. 2007). In addition, RBC transfusions are associated with the low but known risk of bacterial contamination and may substantially increase the risk of nosocomial infections, possibly due to their immunosuppressive effects

Impact of anemia and red blood cell transfusion on organ function

335

Table 7. Incidences of potential risks of allogenic blood transfusions Risk factor Mistransfusion

Infections (viral)

Infections Infections (bacterial) Immunological

Incidence Acute hemolytic reaction (fatalities) Delayed hemolytic reaction (fatalities) Anaphylaxis HIV Hepatitis A Hepatitis B Hepatitis C Cytomegalovirus (CMV) Epstein-Barr virus (EBV) Other viruses Plasmodium, Babesia, trypanosoma, prions Yersinia enterocolica, Serratia marcescens, Pseudomonas, enterobacteria (RBCs) Transfusion-related lung injury (TRALI) Alloimmunization (RBCs) Immunosuppression Allergic reaction, urticaria Fever, chills (RBCs) Post-transfusion purpura Graft-versus-host disease (GVHD)

1 : 6000–1 : 33,000 (1 : 1.000.000) 1 : 2000–1 : 11,000 (rare) 1 : 20.000–1 : 47.000 1 : 200.000–1 : 2.500.000 1 : 1.000.000 1 : 31.000–147.000 1 : 28.000–1 : 288.000 1 : 10–1 : 30 1 : 200 rare, unknown rare, unknown 1 : 1.000–1 : 10.000

1 : 300–1 : 5.000 1 : 100 1:1 1 : 30–1 : 100 1 : 100–1 : 200 unkown rare, unknown

Adapted from Mercuriali and Inghilleri 2002, Pape 2007)

(Edna and Bjerkeset 1998; Andrew et al. 2005; Brecher and Hay 2005; Mark et al. 2005; Shorr and Jackson 2005; Bilgin et al. 2007). Furthermore, there is a number of noninfectious complications, such as mistransfusion and, particularly, various types of immunological reactions, which more frequently occur and may be associated with serious hazards (Table 7) (Bux et al. 2007; Reed et al. 2007; Stroncek 2007). Other abnormalities are metabolic changes, such as acidosis, hyperkalemia, and hypocalcemia, which may occur after massive transfusions (Schmitt and Götz 1988). The aim of RBC transfusion is to increase the oxygen-carrying capacity of the blood and the oxygen delivery to tissue. There are, however, a number of factors that may tremendously limit these effects. RBCs are usually stored at 4°C before they are transfused. During storage, they undergo a number of metabolic, biochemical, and molecular changes, called “storage lesion”, which have enormous impact on their functional ability and also in part

336

M. R. Nowrousian

mediate the adverse effects associated with RBC transfusions (Table 8). These changes may result in irreversible damages of RBCs and ultimately limit the storage period. During storage, RBCs frequently change their shape, evolving from biconcave discs to malformed spheroechinocytes. These changes result from a number of processes including a depletion of intercellular adenosine triphosphate and 2,3-diphosphoglycerate (2,3-DPG), vesiculation and loss of membrane phospholipids, oxidation of membrane proteins, peroxidation of membrane lipids and, ﬁnally, loss of deformability. The latter, together with the loss of the biconcave disc shape reduces the ability of RBCs with a diameter of 8 μm to navigate the microcirculation with capillary diameters between 3–8 μm. The storage changes also increase the interaction between RBCs and vascular endothelial cells and thus further compromise the microvascular ﬂow. The combination of these factors may have an enormous negative effect on tissue oxygenation, which may be particularly deteriorating in patients with an already reduced organ perfusion and function, such as patients with sepsis and septic shock (Offner 2004; Högmann and Meryman 2006; Raghavan et al. 2006; Hébert et al. 2007; Klein et al. 2007). An important factor affecting the functional ability of RBCs is the decrease in their intracellular concentration of 2,3-DPG during storage (Table 8). The decrease starts to occur within the ﬁrst 2 hours of storage and is completed after 2 weeks. It signiﬁcantly increases the afﬁnity of Hb to oxygen. As a result, stored RBCs initially after transfusion have a reduced ability of more than 50% to ofﬂoad oxygen. They take up oxygen much more intensively than they release it. Oxygen may be taken up from circulating Table 8. Red blood cell “storage lesion” Storage effects

Consequences

Decreased 2,3-diphosphoglycerate

Increased oxygen afﬁnity and decreased oxygen unloading by hemoglobin Erythrocyte shape changes Increased osmotic fragility Decreased deformability Decreased erythrocyte viability

ATP depletion

Microvesiculation and loss of lipid membrane Lipid peroxidation Bioactive substance generation: histamine, lipids, cytokines (IL-1, IL-8, TNF, etc.)

Cellular injury and death Febrile transfusion reactions Neutrophil priming/endothelial activation Cellular injury Transfusion-related acute lung injury Multiple organ failure (?)

IL = interleukin; TNF = tumor necrosis factor. Adapted from Offner 2004.

Impact of anemia and red blood cell transfusion on organ function

337

plasma, other RBCs, and even from myoglobin. It has been shown that following transfusion of DPG-depleted RBCs, systemic DPG levels, as well as the oxygen-Hb dissociation curve values, signiﬁcantly fall and then regenerate at a very variable rate taking up 24 h to several days (Offner 2004; Spiess 2004; Raghavan and Marik 2005; Högmann et al. 2006). This may explain the result of a prospective study of patients after cardiac surgery, which did not show any improvement in oxygen delivery to tissue after transfusion of 1–2 units of stored RBCs (Suttner et al. 2004). Other time-dependent changes that occur during storage are a progressive fall in pH, an increase in plasma potassium concentration and release of free Hb from lysed RBCs. RBCs largely use anaerobic glycolysis to maintain their cellular integrity. As a result, the blood rapidly becomes acidotic and the pH value may decrease to a very low level by day 28 of storage. In addition, there is a leak of potassium from RBCs into the stored blood resulting in concentrations as high as ≥78 mmol/l. During storage, there is also a generation of bioactive substances, such as histamine, complement, lipids, fragments of cellular membranes, soluble human leukocyte antigen class I, and cytokines, such as IL-1, IL-6, IL-8, bactericidal permeability-increasing protein, and tumor necrosis factor, which mainly originate from co-stored white blood cells (WBC) and play an important role in transfusion-induced adverse effects and immunomodulation (TRIM) (Stack et al. 1995; Mynster et al. 1998; Fransen et al. 1999; Zallen et al. 2000; Raghavan and Marik 2005). The integrity of stored RBCs also appears to be affected by the amount of co-stored WBC. Increased hemolysis, microvesiculation, and potassium leakage from RBCs have been reported to occur with increasing levels of WBC contamination (Högmann et al. 1999a, b, 2006; Raghavan and Marik 2005). An important factor determining the viability of stored RBCs is the storage duration. According to the current viability standard, RBC units should have a viability of at least 75% at their maximum storage time. Based on this recommendation, a patient who receives 4 units of RBCs approaching outdate may have received one full unit of dead cells, which are not only functionally ineffective, but also have to be eliminated by the RES and undoubtedly jeopardize the function of this system. It has been calculated that based on differences in the remaining life span of the RBCs, the proportion of cells becoming nonviable will increase by approximately 1 percent per day following collection (Högman and Meryman 2006). Of particular note is the fact that nonviable RBCs will be removed from the circulation within 24 hours after transfusion. In a study of patients with cervix cancer receiving radiotherapy, RBC transfusions were successful in raising and maintaining an Hb level > 11 g/dl only in 18.5% of patients (Kapp et al. 2002) indicating that, despite an intensiﬁed approach, sufﬁcient and sustained levels of Hb can hardly be achieved using RBC transfusions, most probably related to the reduced viability and shortened survival of the stored RBCs.

338

M. R. Nowrousian

Another factor that determines the functional value of RBC units is their Hb content. It may considerably vary depending on donors’ Hb concentration, the volume of blood collected, and substantial losses that occur during the ﬁnal preparation of the unit, particularly due to buffy-coat collection and leukoﬁltration, and during storage. Considering this and the fact that at the time of transfusion up to 25% of RBCs may be nonviable and thus functionally ineffective, the clinically useful Hb content of a unit may be as little as 36 g, which is considerably less than the 56–74 g usually present in fresh units of 450 ml blood obtained from donors with the Hb concentrations of 12.5–16.5 g/dl (Högman and Meryman 2006). A number of retrospective studies has demonstrated an association between RBC transfusions and adverse clinical outcomes, such as mortality, pneumonia and other serious infections, and length of hospitalization in many patient populations including patients with multiple trauma, critically ill patients, and patients undergoing cardiac surgery. Of particularly negative effect appears to be a prolonged storage time because of the above-described morphological and functional changes of RBCs and substances and cytokines that increasingly occur during storage deteriorating the microcirculation and, thus, organ perfusion and oxygen delivery and inducing inﬂammatory processes and immunosuppression after transfusion. Immunosuppression appears to be a part of a broad spectrum of immunological reactions, which may be triggered by allogenic RBC transfusions. Allogenic blood transfusions may have primarily two opposite effects on the immune system of the recipient. They can induce immune activation as well as tolerance and suppression. Clinical syndromes associated with immune activation comprise a variety of transfusion reactions, transfusion-associated graft-versus-host disease (TAGVHD), transfusion-related acute lung injury (TRALI), alloimmunization and, possibly, the development of various autoimmune diseases. Syndromes associated with tolerance induction and immunosuppression include increased predisposition to nosocomial and postoperative infections, cancer recurrence, microchimerism, and enhanced survival of renal, hepatic, cardiac, pancreatic, and skin allografts. Immunization is reﬂected by HLA antibodies, activated T cells, increased number of HLA-DR-activated lymphocytes, and lymphocyte blastogenesis, while immunosuppression is suggested by a shift from Th1 to Th2 immune response, decreased natural killer cell activity, decreased lymphocyte response to mitogens, decreased cytotoxic T cell number and anergy to intradermal antigens, and reversal in the CD4/CD8 ratio (Lubin 2005; Raghavan and Marik 2005). A number of clinical studies, including prospective randomized trials, have evaluated the effect of allogenic transfusion on cancer recurrence. The results of these studies are controversial (Jensen et al. 2005; Amato and Pescatori 2006; Rinker et al. 2007). In a recent meta-analysis, however, pooled data from prospective randomized trials of perioperative blood transfusions in patients with colorectal cancer yielded an odds ratio of 1.42

Impact of anemia and red blood cell transfusion on organ function

339

(95% CI, 1.20 to 1.67) against transfused patients. In addition, stratiﬁed metaanalysis conﬁrmed these ﬁndings, also when patients were stratiﬁed by site and stage of their disease. Furthermore, the analyses involving transfusionrelated factors showed that recurrences were more likely with transfusions, independent of their timing and the type of blood product and that there was a dose-dependent relationship, with three or more units of transfused blood almost doubling the risk observed with one or two units. Although some heterogeneity was observed limiting the results of these analyses, the authors recommended carefully restricted indications for the use of perioperative transfusions (Amato and Pescatori 2006). In two recent studies of patients with oral cavity and oropharyngeal squamous cell carcinoma, perioperative transfusions of 3 or more units were also found to be associated with a worse prognosis (Taniguchi and Okura 2003; Szakmany et al. 2006). In another study including 249 patients with pancreaticoduodenectomy for exocrine tumor of the pancreas, allogenic blood transfusions were observed to be generally associated with a signiﬁcantly shortened survival (Yeh et al. 2007). The authors recommended the avoidance of allogenic transfusions in these patients, when possible. Another risk factor associated with RBC transfusion is iron overload, which is usually the result of a decreased proliferative activity of erythroid marrow and the need for repetitive RBC transfusions to alleviate the symptoms of anemia (Porter 2005). While the consequences of iron overload have been evaluated extensively in patients with chronic renal failure and patients with myelodysplasia (Jensen et al. 1996; Eschbach and Adamson 1999; Greenberg 2006; Takatoku et al. 2007), they are less known in patients with cancer (Harrison et al. 1996; Kelekis et al. 1996; McKay et al. 1996; Emy et al. 1997; Lichtman et al. 1999; Barton and Bertoli 2000; Miceli et al. 2006). Many of these patients, however, particularly those with hematological malignancies, have excessive serum ferritin levels (Nowrousian et al. 1996) indicating impairments of iron utilization and, possibly, iron overload due to repeated RBC transfusions. In healthy adults, serum ferritin concentrations range between 15 and 300 μg/l, and values less than 15 μg/l indicate iron deﬁciency. In anemia of chronic diseases (ACD), including cancer, serum ferritin concentrations may increase up to 50 μg/l, because of an increased production of ferritin as an acute-phase protein and as storage protein for the upregulated sequestration of iron. However, serum ferritin concentrations greater than 300 μg/l in persons with or without ACD are a sign of excessive amounts of iron deposited in tissues. In normal individuals, 1 μg/l ferritin relates to 8–12 mg of storage iron. This ratio, however, is not constant and decreases with increasing iron stores and vice versa. The amount of storage iron in normal adults is approximately 1g and the storage limit of macrophages for accumulation of iron 4–5 g. This limit is usually exceeded, when serum ferritin levels are greater than 500 μg/l. In patients with chronic anemia, increased

340

M. R. Nowrousian

iron stores may result from repeated RBC transfusions, which contain on average 1 mg iron per milliliter or 200–250 mg of iron per unit packed RBCs. The consequence may be a secondary iron overload, also referred to as hemosiderosis, characterized by an accumulation of excess iron, primarily in the reticuloendothelial system (RES) of the liver, marrow, and spleen, but also in other predisposed organs, such as the heart, endocrine glands and joints. Ferrokinetic studies have shown that iron turnover is inﬂuenced by serum iron concentration and the percentage of transferrin saturation. After absorption from the gastrointestinal tract, iron is usually bound to transferrin and transported to either the erythroid marrow, RES or other tissues. After intravenous application as iron dextran, ferric gluconate or ferric saccharate, the metal is ﬁrst processed by the RES and then transported to either the erythroid marrow or other tissues. If serum iron concentration is greater than 150 μg/dl and the percentage of transferrin saturation greater than 60%, iron transport will be shifted from the erythroid marrow to other tissues (Bottomley 1998; Eschbach and Adamson 1999; Kaltwasser and Gottschalk 1999; Weinberg 1999). One of the consequences of iron overload is hepatomegaly, resulting from deposition of excess iron in liver parenchymal cells as well as Kupffer cells. Cirrhosis may also develop, but primarily in patients with a history of hepatitis B or C. Iron overload can also be associated with proximal myopathy and muscle weakness, but exclusively in patients with one or more of the hemochromatosis alleles HLA3, B7, or B14. Pancreatic ﬁbrosis and cardiac failure, as well as generation of free radicals, can also occur, but they have not played a signiﬁcant role in hemodialysis patients with transfusion-related hemosiderosis (Schaefer et al. 1981; Bottomley 1998; Eschbach and Adamson 1999). An impact of iron overload, which could be of particular concern in patients with cancer, is the increased risk of infections. In a study of 367 patients with multiple myeloma receiving autologous stem cell transplantation, increased pre-transplant bone marrow iron stores independently predicted severe infection after transplantation (Miceli et al. 2006). In other studies including patients with myelodysplasia, acute myeloid leukemia and other types of hematological malignancies receiving autologous or allogenic stem cell transplantations, iron overload was found to be a risk factor not only for infections but also for other types of complications and increased mortality (Butt and Clark 2003; Altes et al. 2004; Armand et al. 2007). An association between iron overload and infections has also been reported in patients with ESRD. In these patients, ferritin levels higher than 500 μg/l or 1,000 μg/l were found to be associated with a signiﬁcantly higher incidence of bacterial infections (Seifert et al. 1987; Tielemans et al. 1989; Boelaert et al. 1990a, b; Hoen et al. 1995). A mechanism by which such an increased risk of infection could be explained is a decreased phagocytic function of polymorph nuclear leukocytes and a decreased intracellular killing activity and oxidative burst of neutrophils, which have been observed in patients with iron over-

Impact of anemia and red blood cell transfusion on organ function

341

load (Flament et al. 1986; Cantinieaux et al. 1988; Boelaert et al. 1990; Vanholder et al. 1993; Patruta et al. 1998; Patruta and Horl 1999; Bullen et al. 2005; Khan et al. 2007). In anemic patients with iron overload, treatment with rhEPO may be of value not only in increasing Hb level, but also in improving iron utilization by increasing expression of transferrin receptor (TFR) and promoting the binding of TF to TFR and, thus, increasing iron uptake through erythroid progenitor cells (Nowrousian in this book). In a study in patients with myelodysplasia and iron overload, the use of rhEPO potentiated the effect of iron chelation therapy (Cermak et al. 2006).

Interaction between anemia and other pathophysiologic or disabling processes The impact of anemia on organ functions may be considerably aggravated, when additional pathophysiologic processes or disabling factors affecting the adaptive mechanisms are present. Particularly, pre-existing heart, lung, and cerebrovascular diseases may tremendously increase the risk of deleterious effects of anemia. Further factors possibly affecting the compensatory mechanisms are age, severity of underlying disease, and therapeutic interventions (Table 2). The magnitude of Hb decrease tolerated by the heart depends not only upon the degree of anemia but also the conditions under which the heart has to work and to compensate for the reduced oxygen-carrying capacity of the blood. Unfortunately, there are no adequate studies to answer these questions. The results of studies evaluating the threshold of Hb at which cardiac output begins to rise vary considerably, ranging between 7 to 12 g/dl, and the results of a study showing that oxygenation may still be adequate with Hb levels as low as 5 g/dl are questionable, since this study was carried out in healthy volunteers at rest and for a limited period of time. The results of this and other studies with similar design can not be generalized and transferred to chronic anemia in individuals with more advanced age, increased oxygen consumption, decreased ability to increase cardiac output and decreased capacity to enhance blood ﬂow to speciﬁc organs because of arterial stenosis or pathophysiological changes in the microcirculation (Herbert and Szick 2001; Hébert et al. 2004). The heart, and more speciﬁcally the left ventricle, is particularly susceptible to the adverse effects of anemia because of its central role as a compensatory organ and because of its extraordinary sensitivity to hypoxia (Table 1, Fig. 1). Pre-existing myocardial diseases may not only interfere with cardiac remodeling in response to anemia, but they can also accelerate the development of an acute or chronic heart failure. This relationship has been impressively shown in patients with CHF and mild to moderate anemia who experienced signiﬁcant improvements in their cardiac and renal function as

342

M. R. Nowrousian

well as exercise capacity after treatment with rhEPO (Silverberg et al. 2000, 2001a, b, 2006 and in this book; Palazzuoli et al. 2006; Mitchell 2007; Ponikowski et al. 2007; Tang 2007). Pre-existing coronary diseases are likewise a major risk factor for developing serious complications, since they increase the probability of developing angina pectoris and myocardial infarction. The myocardium usually extracts 60%–70% of all oxygen delivered to the coronary arteries. A further increase in oxygen delivery can therefore only be achieved by increase in coronary artery blood ﬂow (Hébert et al. 2004, 2007). The latter, however, may be substantially restricted in case of coronary artery diseases. In a study including 39922 patients with acute coronary syndromes, the likelihood of cardiovascular death, myocardial infarction, or recurrent ischemia was found to signiﬁcantly increase in patients with non-ST-elevation, when the Hb fell below 11 g/dl (Sabatine et al. 2005). Another factor which may also restrict the coronary artery ﬂow to the left ventricle is tachycardia. The perfusion of the left ventricle predominantly occurs during the diastolic period and increased heart rate usually shortens this period (Herbert and Szick 2001). Increased heart rate, on the other hand, is an adaptive process, which frequently occurs in cancer patients, e.g. in association with fever, bleeding or hemolysis. In addition, increased heart rate may be required during exercise, when increased stroke volume alone is not sufﬁcient to compensate for tissue hypoxia (Metivier et al. 2000; Herbert and Szick 2001; Hébert et al. 2004; Eckardt 2005). An interesting but often neglected and underestimated impact of anemia is its effect on hemostatic function. Experimental and clinical studies show a negative relationship between the hematocrit and the bleeding time (BT) (Hardy 2004). In a study in healthy volunteers, an acute 15% reduction in hematocrit induced a 60% increase in BT, while a 32% decrease in platelet count did not show any effect (Valeri et al. 2001). Furthermore, treatment of anemia in thrombocytopenic patients has been reported to shorten the BT (Escolar et al. 1988). The interaction between anemia and the BT is of particular interest for patients who are at risk of developing thrombocytopenia, e.g. patients who receive chemotherapy or radiotherapy, and patients who are thrombocytopenic and have to undergo surgical interventions. There is experimental evidence indicating that hematocrit levels as high as 35% may be required to sustain hemostasis (Hardy 2004). The mechanisms of interaction between anemia and hemostasis may be related to the reduced number of RBCs and, consequently, decreased rheological and activating effects of these cells on platelets. RBCs have their maximal ﬂow at the center of vessels pushing the platelets towards the periphery and, thus, optimizing their contact and interaction with injured endothelium. RBCs also have some proaggregatory properties by activating the platelets and modulating their biochemical and functional responsiveness (Hardy 2004). Anemia may also interact with respiratory disorders and aggravates their effects. In a study of 683 patients with chronic obstructive pulmonary disease

Impact of anemia and red blood cell transfusion on organ function

343

(COPD), anemia was found to be associated with a shortened median survival (49 versus 74 months) and it also independently predicted dyspnea and reduced functional capacity (Cote et al. 2007). Cerebral blood ﬂow as well as oxygen extraction have been shown to increase with the severity of anemia. On the other hand, the cerebral blood ﬂow can only increase within the context of a generally increased cardiac output and a shift of blood ﬂow from other organs to the CNS (Hébert et al. 2004). Furthermore, cerebral blood oxygen extraction can only be sufﬁciently increased, if no restricting cerebrovascular processes are present. In young and healthy adults, acute anemia with Hb levels of 5–6 g/dl has been reported to produce signiﬁcant cognitive dysfunctions, which resolve after increasing Hb level (Weiskopf et al. 2000). Chronic anemia, in contrast, appears to induce cognitive abnormalities already at Hb levels of <12 g/dl, as observed in patients with cancer (Jacobsen et al. 2004; Mancuso et al. 2006), myelodysplasia (Clavio et al. 2004), or ESRD (Stivelman 2000; Singh et al. 2006), and even in adolescents and young adults with iron deﬁciency (Sen and Kanani 2006; Murray-Kolb and Beard 2007; Petranovic et al. 2007). Cancer patients have been reported to be predisposed to develop long-term cognitive dysfunctions (Heﬂin et al. 2005). A group of patients, who appear to be particularly susceptible to cognitive abnormalities are patients, who receive chemotherapy (Jacobsen et al. 2004; O’Shaughnessy et al. 2005; Mancuso et al. 2006; Massa et al. 2006). In these patients, cognitive changes have been reported to occur both during as well as after completion of treatment. A number of mechanisms has been postulated to be responsible for this phenomenon, but anemia appears to be a major candidate. In a study of 77 patients with various types of malignant diseases receiving chemotherapy, a decline in Hb to a ﬁnal level of ≤12 g/dl was observed in 49 patients, and in these patients, greater declines in Hb were signiﬁcantly related to greater increases in fatigue duration and disruptiveness and more negative changes in cognitive performance (Fig. 3) (Jacobsen et al. 2004). In another study of 42 elderly patients with lung cancer, a signiﬁcant correlation was found between Hb level and functional and cognitive capacity during chemotherapy (Mancuso et al. 2006). The relationship between Hb level and cognitive function has also been evaluated in anemic patients with ESRD who received rhEPO. In these patients, signiﬁcant improvements in cognitive changes and electroencephalographic (EEG) measures of event-related potentials (ERPs) were observed in association with increase in Hb level within a range from 7 g/dl to normal value (Brown et al. 1991; Marsh et al. 1991; Sagales et al. 1993; Temple et al. 1995; Pickett et al. 1999; Singh et al. 2006). One of the most common and particularly troubling symptoms in cancer patients is fatigue (Cella et al. 2003; Butt and Cella in this book). Fatigue is a multifactorial process which may result from the malignant disease itself, antineoplastic therapies, and/or a broad spectrum of physical and psychologic comorbidities. In case of anemia, however, fatigue is generally accepted to

344

M. R. Nowrousian

Fig. 3. Changes in cognitive performance based on degree of hemoglobin decline among patients with cancer who received chemotherapy and experienced a decline in hemoglobin to a ﬁnal value ≤ 12 g/dl (n = 49). Greater decreases in hemoglobin were signiﬁcantly related to greater increase in fatigue duration and disruptiveness and more negative changes in performance on three cognitive tasks: Trails A = the ability to follow a simple number sequence; Trails B = the ability to follow a more complex sequence of alternating numbers and letters; VR = visual reproduction. Adapted from Jacobsen et al. 2004; with permission

be the cardinal symptom. There appears to be a close relationship between anemia, cognitive dysfunction and fatigue. Cognitive dysfunction and fatigue, on the other hand, have considerable negative effects on physical, functional, emotional, social and overall well-being of patients and thus on their QOL. Clinical studies with rhEPO show that increase in Hb level is able to reverse these effects (Cella 1998; Cella et al. 2004; Butt and Cella in this book; Glaus and Müller 2000; Sobrero et al. 2001; Jacobsen et al. 2004; Jones et al. 2004; Cleeland et al. 2005; Charu and Ben-Jacob 2005; Nieboer et al. 2005; O’Shaughnessy et al. 2005). As mentioned above, shunting the blood ﬂow from other organs to the heart and the brain is an important compensatory mechanism in chronic

Impact of anemia and red blood cell transfusion on organ function

345

anemia. The consequence of this redistribution of the blood, however, may be critical for certain organs such as the kidneys and the bowel, particularly when they are affected by additional pathophysiological processes or are challenged to increase their capacity. This is often the case in patients with cancer, since they are frequently suffering from malnutrition due to diseaseor treatment-related complications including vomiting, emesis and diarrhea and have to receive nephrotoxic drugs such as cisplatin and amphotericin B together with large volumes of ﬂuid. Particularly, renal function may be affected by anemia, as shown in patients with chronic heart failure. In these patients, anemia has been reported to be a risk factor for increased mortality, and treatment of anemia with rhEPO has been shown to signiﬁcantly improve kidney function and reduce the use of cardiac drugs and diuretics and the frequency of hospitalization (Silverberg et al. 2006 and in this book). Many patients with cancer are elderly people with decreased or maximally stressed functional capacity of organs, which have to compensate for anemia, such as the heart, or are particularly affected by anemia, such as the heart and the brain (Fig. 1, Tables 1, 3). In addition, these and other organs, such as the bone marrow and kidney, are often severely challenged during cancer treatment. Furthermore, elderly people are more frequently affected by processes which themselves are able to induce the development of anemia, such as nutritional deﬁciencies, hemorrhage and chronic diseases associated with a reduced endogenous EPO production (Balducci 2003, 2007; Beghe et al. 2004; Dharmarajan et al. 2005; Guralnik et al. 2005). On the other hand, anemia increases the risk of adverse effects of drugs, since it decreases the percentages of drugs bound to red blood cells and increases the concentration of free drugs in the circulation. It also causes tissue hypoxia and thus an increased susceptibility of tissues to diagnostic and therapeutic interventions, as well as to toxic effects of drugs. In younger individuals, the pharmacokinetic effects of anemia may be buffered in part by other tissues, particularly the muscles, but in older persons, this mechanism may be lacking or reduced due to the frequently present underweight up to cachexia (Balducci 2003, 2007). Other factors aggravating the toxicity of drugs in older persons are changes in pharmacokinetics due to decreased elimination of metabolized drugs in the liver, particularly those eliminated by the cytochrome enzyme system, and decreased elimination of drugs by the kidney, predominantly due to a fall in the glomerular ﬁltration rate (Cusack 2004). Based on these pathophysiological and functional impairments and a generally reduced bone marrow capacity, elderly patients with cancer are at a particularly high risk for developing more frequently anemia during chemotherapy and to suffer from severe symptoms of anemia. The latter include physical disability, functional dependence, cognitive dysfunctions up to dementia, fatigue, congestive heart failure, and coronary artery disease. In elderly people, anemia is an independent risk factor for morbidity and mortality, and it has also been shown to have signiﬁcant impact on health care requirements and health care

346

M. R. Nowrousian

expenditure (Aapro et al. 2002; Lipschitz 2003; Robinson 2003; Penninx et al. 2004, 2005; Culleton et al. 2006; Denny et al. 2006; Maraldi et al. 2006; Zamboni et al. 2006). Considering these aspects of anemia in older persons and the additional burdens, which are associated with cancer and its treatment, it appears to be a priority to treat anemia in elderly cancer patients to improve quality of life and to prevent severe and life-threatening complications. Many patients with cancer, particularly those with advanced age, are additionally suffering from a variety of disorders, such as diabetes mellitus, renal failure and rheumatoid diseases. There is evidence suggesting that anemia may aggravate the clinical symptoms of these comorbidities. In patients with diabetes mellitus, anemia has been reported to signiﬁcantly contribute to impairments of physical performance, cognitive function, sleep, appetite, sexual function, social activities and employment. In addition, it appears to increase the risk of vascular complications including retinopathy, neuropathy, impaired wound healing and macrovascular disease. Furthermore, a reduced Hb level, even within the normal range, has been found to predict progressive renal dysfunction, hospitalization, and premature death (Thomas et al. 2006). In patients with rheumatoid arthritis, treatment of anemia with rhEPO has been shown to reduce disease activity and fatigue and to improve muscle strength and vitality (Peeters et al. 1999; Kaltwasser et al. 2001).

Optimal level of hemoglobin in cancer patients In human beings, the Hb level is dependent on sex and age. Women usually have a lower Hb level than men, and there is a decrease in Hb level with increasing age (Nilsson-Ehle et al. 2000; Rushton et al. 2001). In men, Hb level normally ranges between 13.3–17.7 g/dl and in women between 11.7–15.7 g/dl (Williams 1988; Fairbanks and Tefferi 2000). Based on a deﬁnition of the World Health Organization, anemia is present, when Hb level is <13 g/dl in men and <12 g/dl in women (World Health Organization 1968). Clinical and epidemiological studies show that anemia is generally associated with increased mortality, even in elderly patients in the general population (Izaks et al. 1999; Collins et al. 2000; Caro et al. 2001). As mentioned above, the two most important compensatory mechanisms for anemia are increased oxygen extraction from Hb molecules and increased cardiac output. The ﬁrst mechanism, mediated by increase in red blood cell (RBC) 2,3-DPG content, seems to be of a comparatively low overall effect, since it probably compensates for no more than 0.5 g/dl reduction in Hb level (Mitchell and Pegrum 1971; Rossi 1994). In addition, the effect of this mechanism is even more limited in tissues and organs with a high primary oxygen extraction, such as the heart and the brain. Furthermore, increased oxygen

Impact of anemia and red blood cell transfusion on organ function

347

extraction may be associated with a reduced tissue oxygen tension, particularly in the veins, leading to a shunt diffusion of oxygen between arterial and venous vessels running in parallel. The consequence can be a considerable desaturation of Hb before RBC enter the capillary bed. The second and more important compensatory mechanism, increased cardiac output consisting of increased heart rate and stroke volume, results from increased vasodilatation, reduced peripheral resistance and blood viscosity, and increased sympathetic activity. This mechanism, however, is not always completely compensatory, since total oxygen transport, which is a product of cardiac output und arterial oxygen content, is somewhat less than normal (Varat et al. 1972). In chronic anemia, in addition, increased cardiac output, which is predominantly due to increased stroke volume, may be associated with considerable consequences for the heart (Harnett et al. 1995; Foley et al. 1996, 2000b; Levin and Foley 2000; O’Riordan and Foley 2000). Furthermore, there are studies indicating that despite the above-mentioned compensatory mechanisms, oxygen delivery to many organs rapidly decreases with decreasing Hb levels below 13 g/dl (Erslev et al. 1989; Metry et al. 1999). In current clinical practice, many patients with cancer and anemia either do not receive any type of treatment or only RBC transfusions, usually, when severe clinical symptoms are present or Hb level has decreased below 8 g/dl (Rossi 1994; Estrin et al. 1999; Murphy et al. 2001; Ludwig et al. 2004). This approach not only does not recognize the enormous impact of anemia on organ function, exercise capacity and QOL of patients, but also induces the risk of developing severe anemia and becoming RBC transfusion-dependent and being increasingly exposed to their adverse effects. In a meta-analysis of data from ﬁve randomized double-blind placebo-controlled studies including 1010 cancer patients receiving chemotherapy, prior transfusions were found to be associated with an increased risk for subsequent transfusions with or without rhEPO (Couture et al. 2005). On the other hand, RBC transfusions are of limited value in achieving appropriate and, particularly, sustained increase in Hb level and in improving oxygen transport and delivery. These limitations relate to the relatively short life of the RBCs transfused, because of their natural age and quantitative as well as qualitative changes they experience during storage. The result is the necessity of repeated transfusions in 2–3 weeks intervals to achieve a limited increase in Hb level for a short period of time. Repeated transfusions not only increase the risk of adverse effects but also induce considerable ﬂuctuations in Hb level with major impacts on quality of life and, possibly, a negative effect on the outcome of anticancer treatment. In a study of patients with myelodysplastic syndromes, transfusion-related ﬂuctuations in Hb were reported to negatively correlate with quality of life and to be associated with a greater degree of fatigue (Caocci et al. 2007). In a prospective, randomized, placebo-controlled study of anemic cancer patients receiving chemotherapy, Hb level and quality of life were found to be signiﬁcantly inferior in patients receiving placebo

348

M. R. Nowrousian

compared to those who received rhEPO, although the ﬁrst group of patients had received twice as much RBC transfusions as the second group (Littlewood et al. 2001). Fluctuations in Hb level may also induce cyclic changes in tumor oxygenation and, thus, a selection of tumor cells with a greater metastatic potential and greater resistance against radiotherapy and chemotherapy (Cairns et al. 2001, 2004; Weinmann et al. 2004; Rofstad et al. 2007). RBC transfusions may be of value in conditions, in which an urgent increase in Hb level is required, but they do not represent an adequate treatment of anemia in cancer patients, particularly at the threshold of Hb at which they are usually used. The restricted use of RBC transfusions may be understandable, because of the associated risks, limited efﬁcacy and supply but this can not be an argument for not treating anemia until Hb level is below 8 g/dl, particularly, when an effective and well-tolerated therapy such as rhEPO is available. There is a large number of studies, including prospective randomized trials and meta-analyses indicating that anemic cancer patients are already tremendously suffering from anemia symptoms long before such low levels of Hb are present. These studies, in addition, indicate epoetin alfa, epoetin beta and darbepoetin as drugs, which can be sufﬁciently and safely used to treat anemia in these patients. In many of these studies, treatment with rhEPO was started at Hb levels of <11 g/dl and target Hb levels around 13 g/dl were used. Such an approach appears to be acceptable regarding the increasing impact of anemia with decreasing Hb level below 11 g/dl and the improvements which can be achieved, when Hb level increases from 11 g/dl to 12–13 g/dl. In patients with myelodysplastic syndromes and ESRD, cardiac remodeling has been shown to occur with Hb levels below 11 g/dl (Hamblin 2005; Oliva et al. 2005) and, in cancer patients receiving chemotherapy, a decrease of Hb to ≤12 g/dl has been reported to be associated with a signiﬁcant decrease in cognitive function and increase in fatigue. Furthermore, Hb levels below 12 g/dl appear to have signiﬁcant negative effects on metabolic functions and exercise capacity in cancer patients. In a prospective randomized study of weight-losing cancer patients, prophylactic use of rhEPO prevented the development of anemia and protected the weight and the metabolic and exercise capacity of becoming decreased (Daneryd et al. 1998). Maximum exercise capacity, ventilation, whole-body oxygen uptake, and carbon dioxide production during exercise were signiﬁcantly higher in patients receiving rhEPO than in control patients. In addition, at the end of the observation period, there was a difference in the mean body weight between the two groups of 15.5 kg (71.1 versus 55.6) and a difference in the mean Hb level of 2.2 g/dl (13 versus 10.8) favoring patients who received rhEPO (Table 6). The mean baseline body weights (67.3 versus 67.1) and the mean baseline Hb levels (11.9 versus 12.1) were almost identical in the two groups. Of particular interest are also the results of a combined analysis of data from two large

Impact of anemia and red blood cell transfusion on organ function

349

community-based studies including 4,382 anemic patients with cancer who received chemotherapy and rhEPO. This analysis showed a direct relationship between increases in Hb level in a range from 8–14 g/dl and improvements in quality of life. The data, in addition, showed that the largest incremental gain in quality of life occurred with increase of Hb from 11 g/dl to 12 g/dl. Further increases continued to yield additional gains, but at a decreasing rate (Fig. 4) (Crawford et al. 2002). In another study including 1071 patients with multiple myeloma, a positive correlation was found between fatigue-related quality of life and Hb level. Improvements in FACTAn scores in women and men appeared with Hb increase up to sex-speciﬁc normal values (Fig. 5) (Palumbo et al. 2005). Hb levels of 12 g/dl– <14 g/dl in women and 13 g/dl– <15 g/dl in men may also be optimal for tumor oxygenation with regard to the effect of radiotherapy and also possibly chemotherapy (Vaupel et al. 2007 and in this book). Considering these data, the optimal level of Hb for starting treatment of anemia in cancer patients appears to be ≤11 g/dl and the optimal target Hb level around 12 g/dl. These levels are also recommended by the guidelines of the EORTC for treating anemia in cancer patients with erythropoietic proteins (Bokemeyer et al. 2004, 2007).

Fig. 4. Longitudinal and incremental analysis of correlation between hemoglobin and quality of life (QOL) in cancer patients receiving chemotherapy. Successively positive changes in LASA overall QOL scores continued with increases in hemoglobin levels to 14 g/dl, where the estimated QOL was 15.2 mm higher than at a hemoglobin level below 7.5 g/dl. Incremental increases in hemoglobin, however, rose until a hemoglobin level of 12 g/dl was reached. Beyond this level, subsequent incremental changes in hemoglobin continued to yield additional gains in QOL, but at a decreasing rate. LASA = Linear Analog Scale Assessment. Adapted from Crawford et al. 2002; with permission

350

M. R. Nowrousian

Fig. 5. Correlation between fatigue and hemoglobin level in patients with multiple myeloma: results of a cross-sectional study (n = 1071). FACT-An scores in women and men increased as hemoglobin increased until about 12 g/dl and 14 g/dl, respectively. FACT-An = Functional Assessment of Cancer Therapy – Anemia. Adapted from Palumbo et al. 2005; with permission

Conclusion Physical and metabolic abnormalities resulting from anemia have been evaluated extensively in patients with ESRD and, in part, in patients with cancer. In the former group of patients, there are other factors, such as arterial hypertension and uremia, that can also contribute to the development of cardiac complications and metabolic abnormalities, but the results of studies using rhEPO indicate that anemia is an independent causative factor in developing cardiomyopathy, decreased brain and cognitive function, metabolic abnormalities, sexual dysfunctions, decreased exercise capacity and QOL. In cancer patients, anemia has also been found to be associated with cognitive dysfunctions and fatigue, decreased metabolic and exercise capacity and QOL including physical, psychological, mental and social well-being. In addition, treatment of anemia with rhEPO has been demonstrated to reverse these abnormalities, independent of the response of the underlying malignancy to antitumor treatment. A comparison of the results of studies in these two groups of patients clearly identiﬁes anemia as a factor that, independent of its underlying mechanisms, affects a wide spectrum of organ functions and consequently the QOL. Anemia may also aggravate symptoms of comorbidities, which frequently occur in these patients. In addition, in patients with ESRD, anemia is a risk factor not only for increased morbidity but also

Impact of anemia and red blood cell transfusion on organ function

351

increased mortality, and in cancer patients, it usually predicts worse outcome of radiotherapy and chemotherapy. Traditionally, RBC transfusions have been used to treat anemia in cancer patients, but RBC transfusions are associated with a broad spectrum of risks and limitations and can only temporarily increase Hb level. In current clinical practice, in addition, physicians usually start to treat anemia, when severe clinical symptoms are present or Hb level has decreased below 8 g/dl. This strategy may be explained on one side by the tendency to avoid RBC transfusions as long as possible and, on the other side, by the idea that many patients with chronic anemia may tolerate such a low level of Hb before they develop major clinical symptoms. Many studies, both in cancer patients and in patients with ESRD, however, have shown that physical well-being and QOL signiﬁcantly improve after treatment of anemia, even at Hb levels between 8–12 g/dl. In addition, many physiological and metabolic dysfunctions have been found to reverse with increasing Hb level to 12–13 g/dl. These studies clearly indicate that anemia, even if mild or moderate, has major impact on tissue and organ functions, and that the adaptive mechanisms available are not able to sufﬁciently compensate for the reduced oxygen transport capacity and oxygen delivery to the tissue. In cancer patients, an Hb level of around 12 g/dl may also be of value in improving tumor oxygenations and, thus, the outcome of radiotherapy and chemotherapy.

References 1. Aapro MS, Cella D, Zagari M (2002) Age, anemia, and fatigue. Semin Oncol 29 [Suppl 8]: 55–59 2. Abayomi OK (1996) Pathogenesis of irradiation-induced cognitive dysfunction. Acta Oncol 35 [Suppl 8]: 659–663 3. Abels RI (1992) Use of recombinant human erythropoietin in the treatment of anemia in patients who have cancer. Semin Oncol 19 [Suppl 8]: 29–35 4. Alexandrakis MG, Passam FH, Sﬁridaki A, et al (2004) Serum levels of leptin in multiple myeloma patients and its relation to angiogenic and inﬂammatory cytokines. Int J Biol Markers 19: 52–57 5. Alter HJ, Stramer SL, Roger YD (2007) Emerging infectious diseases that threaten the blood supply. Semin Hematol 44: 32–41 6. Altes A, Remacha AF, Sarda P, et al (2004) Frequent severe iron overload after stem cell transplantation and its possible association with invasive aspergillosis. Bone Marrow Transplant 34: 505–509 7. Amato A, Pescatori M (2006) Perioperative blood transfusions for the recurrence of colorectal cancer. Cochrane Database Syst Rev: CD005033 8. Anand I, McMurray JJ, Whitmore J, et al (2004) Anemia and its relationship to clinical outcome in heart failure. Circulation 110: 149–154 9. Anand IS, Chandrashekhar Y, Ferrari R, et al (1993) Pathogenesis of oedema in chronic severe anaemia: studies of body water and sodium, renal function, haemodynamic variables, and plasma hormones. Br Heart J 70: 357–362

352

M. R. Nowrousian

10. Anantharaman P, Schmidt RJ (2007) Sexual function in chronic kidney disease. Adv Chronic Kidney Dis 14: 119–125 11. Armand P, Kim HT, Cutler CS, et al (2007) Prognostic impact of elevated pretransplantation serum ferritin in patients undergoing myeloablative stem cell transplantation. Blood 109: 4586–4588 12. Axelsson J, Qureshi AR, Heimburger O, et al (2005) Body fat mass and serum leptin levels inﬂuence epoetin sensitivity in patients with ESRD. Am J Kidney Dis 46: 628–634 13. Balducci L (2003) Anemia, cancer, and aging. Cancer Control 10: 478–486 14. Balducci L (2007) Cancer-related anemia: special considerations in the elderly. Oncology (Williston Park) 21: 81–86, 90; discussion 90, 100,104 15. Barany P, Pettersson E, Ahlberg M, et al (1991) Nutritional assessment in anemic hemodialysis patients treated with recombinant human erythropoietin. Clin Nephrol 35: 270–279 16. Barrett-Lee PJ, Bailey NP, O’Brien ME, et al (2000). Large-scale UK audit of blood transfusion requirements and anemia in patients receiving cytotoxic chemotherapy. Br J Cancer 82: 93–97 17. Barton JC, Bertoli LF (2000) Transfusion iron overload in adults with acute leukemia: manifestations and therapy. Am J Med Sci 319: 73–78 18. Baz R, Walker E, Choueiri TK, et al (2007) Recombinant human erythropoietin is associated with increased overall survival in patients with multiple myeloma. Acta Haematol 117: 162–167 19. Beghe C, Wilson A, Ershler WB (2004) Prevalence and outcomes of anemia in geriatrics: a systematic review of the literature. Am J Med 116 [Suppl 7A]: 3S–10S 20. Benz RL, Pressman MR, Hovick ET, et al (1999) A preliminary study of the effects of correction of anemia with recombinant human erythropoietin therapy on sleep, sleep disorders, and daytime sleepiness in hemodialysis patients (The SLEEPO study). Am J Kidney Dis 34: 1089–1095 21. Besarab A, Bolton WK, Browne JK, et al (1998) The effects of normal as compared with low hematocrit values in patients with cardiac disease who are receiving hemodialysis and epoetin. N Engl J Med 339: 584–590 22. Bilgin YM, van de Watering LM, Eijsman J (2007) Is increased mortality associated with post-operative infections after leukocytes containing red blood cell transfusions in cardiac surgery? An extended analysis. Transfus Med 17: 304–311 23. Birmingham DJ, Shen XP, Hartman JA, et al (1996) Effect of chronic human recombinant erythropoietin therapy on antibody responses to immunization in chronic hemodialysis patients. Kidney Int 50: 543–549 24. Boelaert JR, Cantinieaux BF, Hariga CF, et al (1990a) Recombinant erythropoietin reverses polymorphonuclear granulocyte dysfunction in iron-overloaded dialysis patients. Nephrol Dial Transplant 5: 504–517 25. Boelaert JR, Daneels RF, Schurgers ML, et al (1990b) Iron overload in haemodialysis patients increases the risk of bacteraemia: a prospective study. Nephrol Dial Transplant 5: 130–134 26. Bokemeyer C, Aapro MS, Courdi A, et al (2004) EORTC guidelines for the use of erythropoietic proteins in anaemic patients with cancer. Eur J Cancer 40: 2201–2216

Impact of anemia and red blood cell transfusion on organ function

353

27. Bokemeyer C, Aapro MS, Courdi A, et al (2007) EORTC guidelines for the use of erythropoietic proteins in anaemic patients with cancer: 2006 update. Eur J Cancer 43: 258–270 28. Boogaerts M, Coifﬁer B, Kainz C (2003) Impact of epoetin beta on quality of life in patients with malignant disease. Br J Cancer 88: 988–995 29. Bossola M, Muscaritoli M, Valenza V, et al (2004) Anorexia and serum leptin levels in hemodialysis patients. Nephron Clin Pract 97: c76–82 30. Bottomley SS (1998) Secondary iron overload disorders. Semin Hematol 35: 77–86 31. Brecher ME, Hay SN (2005) Bacterial contamination of blood components. Clin Microbiol Rev 18: 195–204 32. Brezden CB, Phillips KA, Abdolell M, et al (2000) Cognitive function in breast cancer patients receiving adjuvant chemotherapy. J Clin Oncol 18: 2695–2701 33. Brines ML, Ghezzi P, Keenan S, et al (2000) Erythropoietin crosses the bloodbrain barrier to protect against experimental brain injury. Proc Natl Acad Sci U S A 97: 10526–10531 34. Brown WS, Marsh JT, Wolcott D, et al (1991) Cognitive function, mood and P3 latency: effects of the amelioration of anemia in dialysis patients. Neuropsychologia 29: 35–45 35. Bryl E, Mysliwska J, Debska-Slizien A, et al (1998) The inﬂuence of recombinant human erythropoietin on tumor necrosis factor alpha and interleukin-10 production by whole blood cell cultures in hemodialysis patients. Artif Organs 22: 177–181 36. Bryl E, Mysliwska J, Debska-Slizien A, et al (1999) Recombinant human erythropoietin stimulates production of interleukin 2 by whole blood cell cultures of hemodialysis patients. Artif Organs 23: 809–816 37. Bullen JJ, Rogers HJ, Spalding PB, et al (2005) Iron and infection: the heart of the matter. FEMS Immunol Med Microbiol 43: 325–330 38. Butt NM, Clark RE (2003) Autografting as a risk factor for persisting iron overload in long-term survivors of acute myeloid leukaemia. Bone Marrow Transplant 32: 909–913 39. Bux J, Sachs UJ (2007) The pathogenesis of transfusion-related acute lung injury (TRALI). Br J Haematol 136: 788–799 40. Cairns RA, Hill RP (2004) Acute hypoxia enhances spontaneous lymph node metastasis in an orthotopic murine model of human cervical carcinoma. Cancer Res 64: 2054–2061 41. Cairns RA, Kalliomaki T, Hill RP (2001) Acute (cyclic) hypoxia enhances spontaneous metastasis of KHT murine tumors. Cancer Res 61: 8903–8908 42. Canadian Erythropoietin Study Group (1990) Association between recombinant human erythropoietin and quality of life and exercise capacity of patients receiving haemodialysis. Br Med J 300: 573–578 43. Cantinieaux B, Boelaert J, Hariga C, et al (1988) Impaired neutrophil defense against Yersinia enterocolitica in patients with iron overload who are undergoing dialysis. J Lab Clin Med 111: 524–528 44. Caocci G, Baccoli R, Ledda A, et al (2007) A mathematical model for the evaluation of amplitude of hemoglobin ﬂuctuations in elderly anemic patients affected by myelodysplastic syndromes: correlation with quality of life and fatigue. Leuk Res 31: 249–252

354

M. R. Nowrousian

45. Caro JJ, Salas M, Ward A, et al (2001) Anemia as an independent prognostic factor for survival in patients with cancer: a systemic, quantitative review. Cancer 91: 2214–2221 46. Cella D (1997) The Functional Assessment of Cancer Therapy-Anemia (FACTAn) Scale: a new tool for the assessment of outcomes in cancer anemia and fatigue. Semin Hematol 34 [Suppl 2]: 13–19 47. Cella D, Bron D (1999) The effect of epoetin alfa on quality of life in anemic cancer patients. Cancer Pract 7: 177–182 48. Cella D, Kallich J, McDermott A, et al (2004) The longitudinal relationship of hemoglobin, fatigue and quality of life in anemic cancer patients: results from ﬁve randomized clinical trials. Ann Oncol 15: 979–986 49. Cella D, Zagari MJ, Vandoros C, et al (2003) Epoetin alfa treatment results in clinically signiﬁcant improvements in quality of life in anemic cancer patients when referenced to the general population. J Clin Oncol 21: 366–373 50. Cella D (1998) Factors inﬂuencing quality of life in cancer patients: anemia and fatigue. Semin Oncol 25 [Suppl 7]: 43–46 51. Cerami A, Brines ML, Ghezzi P, et al (2001) Effects of epoetin alfa on the central nervous system. Semin Oncol 28 [Suppl 8]: 66–70 52. Cermak J (2006) Erythropoietin administration may potentiate mobilization of storage iron in patients on oral iron chelation therapy. Hemoglobin 30: 105–112 53. Chan MK (1990) Lipid metabolism in renal failure. Clin Biochem 23: 61–65 54. Chan MK, Varghese Z, Moorhead JF (1981) Lipid abnormalities in uremia, dialysis, and transplantation. Kidney Int 19: 625–637 55. Chang J, Couture F, Young S, et al (2005) Weekly epoetin alfa maintains hemoglobin, improves quality of life, and reduces transfusion in breast cancer patients receiving chemotherapy. J Clin Oncol 23: 2597–2605 56. Charu V, Saidman B, Ben-Jacob A, et al (2005) Improvements in fatigue are associated with early treatment with darbepoetin alfa every 3 weeks in anemic patients receiving chemotherapy. J Support Oncol 3 [Suppl 1]: 14–15 57. Chin K, Yu X, Beleslin-Cokic B, et al (2000) Production and processing of erythropoietin receptor transcripts in brain. Brain Res Mol Brain Res 81: 29–42 58. Clavio M, Nobili F, Balleari E, et al (2004) Quality of life and brain function following high-dose recombinant human erythropoietin in low-risk myelodysplastic syndromes: a preliminary report. Eur J Haematol 72: 113–120 59. Cleeland C, Rossi G, Gabrilove J (2005) Assessing symptom burden in breast cancer patients treated with darbepoetin alfa 200 μg every two weeks using the M.D. Anderson symptom inventory. J Support Oncol 3 [Suppl 1]: 34–35 60. Clyne N, Jogestrand T (1992) Effect of erythropoietin treatment on physical exercise capacity and on renal function in predialytic uremic patients. Nephron 60: 390–396 61. Collins AJ, Li S, St Peter W, et al (2001) Death, hospitalization, and economic associations among incident hemodialysis patients with hematocrit values of 36 to 39%. J Am Soc Nephrol 12: 2465–2473 62. Collins AJ, Ma JZ, Ebben J (2000) Impact of hematocrit on morbidity and mortality. Semin Nephrol 20: 345–349 63. Considine RV, Sinha MK, Heiman ML, et al (1996) Serum immunoreactiveleptin concentrations in normal-weight and obese humans. N Engl J Med 334: 292–295

Impact of anemia and red blood cell transfusion on organ function

355

64. Cortesi E, Gascon P, Henry D, et al (2005) Standard of care for cancer-related anemia: improving hemoglobin levels and quality of life. Oncology 68 [Suppl 1]: 22–32 65. Cote C, Zilberberg MD, Mody SH, et al (2007) Haemoglobin level and its clinical impact in a cohort of patients with COPD. Eur Respir J 29: 923–929 66. Couture F, Turner AR, Melosky B, et al (2005) Prior red blood cell transfusions in cancer patients increase the risk of subsequent transfusions with or without recombinant human erythropoietin management. Oncologist 10: 63–71 67. Crawford J, Cella D, Cleeland CS, et al (2002) Relationship between changes in hemoglobin level and quality of life during chemotherapy in anemic cancer patients receiving epoetin alfa therapy. Cancer 95: 888–895 68. Culleton BF, Manns BJ, Zhang J, et al (2006) Impact of anemia on hospitalization and mortality in older adults. Blood 107: 3841–3846 69. Cusack BJ (2004) Pharmacokinetics in older persons. Am J Geriatr Pharmacother 2: 274–302 70. Dame C, Juul SE, Christensen RD (2001) The biology of erythropoietin in the central nervous system and its neurotrophic and neuroprotective potential. Biol Neonate 79: 228–235 71. Daneryd P, Svanberg E, Körner U, et al (1998) Protection of metabolic and exercise capacity in unselected weight-losing cancer patients following treatment with recombinant erythropoietin: a randomized prospective study. Cancer Res 58: 5374–5379 72. Das SR, Dries DL, Drazner MH, et al (2005) Relation of lower hematocrit to progression from asymptomatic left ventricular dysfunction to symptomatic heart failure (from the Studies of Left Ventricular Dysfunction Prevention trial). Am J Cardiol 96: 827–831 73. De Maria R, Ripamonti V, Sandri R, et al (2005) The negative prognostic synergism of anemia and heart disease in female nursing home residents. Am J Cardiol 96: 1460–1462 74. Demetri GD, Kris M, Wade J, et al (1998) Quality-of-life beneﬁt in chemotherapy patients treated with epoetin alfa is independent of disease response or tumor type: results from a prospective community oncology study. Procrit Study Group. J Clin Oncol 16: 3412–3425 75. Denny SD, Kuchibhatla MN, Cohen HJ (2006) Impact of anemia on mortality, cognition, and function in community-dwelling elderly. Am J Med 119: 327– 334 76. Dharmarajan TS, Pais W, Norkus EP (2005) Does anemia matter? Anemia, morbidity, and mortality in older adults: need for greater recognition. Geriatrics 60: 22–27, 29 77. Diesel W, Noakes TD, Swanepoel C, et al (1990) Isokinetic muscle strength predicts maximum exercise tolerance in renal patients on chronic hemodialysis. Am J Kidney Dis 16: 109–114 78. Drueke TB, Eckardt KU, Frei U, et al (1999) Does early anemia correction prevent complications of chronic renal failure? Clin Nephrol 51: 1–11 79. Eckardt KU (2000) Pathophysiology of renal anemia. Clin Nephrol 53 [1 Suppl 1]: S2–S8 80. Eckardt KU (2005) Managing a fateful alliance: anaemia and cardiovascular outcomes. Nephrol Dial Transplant 20 [Suppl 6]: vi16–20

356

M. R. Nowrousian

81. Edna TH, Bjerkeset T (1998) Association between transfusion of stored blood and infective bacterial complications after resection for colorectal cancer. Eur J Surg 164: 449–456 82. Emy PY, Levin TL, Sheth SS, et al (1997) Iron overload in reticuloendothelial systems of pediatric oncology patients who have undergone transfusions: MR observations. AJR Am J Roentgenol 168: 1011–1015 83. Erslev AJ, Caro J, Schuster SJ (1989) Is there an optimal hemoglobin level? Transfus Med Rev 3: 237–242 84. Eschbach JW, Adamson JW (1999) Iron overload in renal failure patients: changes since the introduction of erythropoietin therapy. Kidney Int [Suppl] 69: S35–S43. 85. Escolar G, Garrido M, Mazzara R, et al (1988) Experimental basis for the use of red ell transfusion in the management of anemic thrombocytopenic patients. Transfusion 28: 406–411 86. Estrin JT, Schocket L, Kregenow R, et al (1999) A retrospective review of blood transfusion in cancer patients with anemia. Oncologist 4: 318–324 87. Fagher B, Thysell H, Monti M (1994) Effect of erythropoietin on muscle metabolic rate, as measured by direct microcalorimetry, and ATP in hemodialysis patients. Nephron 67: 167–171 88. Fairbanks VF, Tefferi A (2000) Normal ranges for packed cell volume and hemoglobin concentration in adults: relevance to “apparent polycythemia”. Eur J Haematol 65: 285–296 89. Fallowﬁeld L, Gagnon D, Zagari M, et al (2002) Multivariate regression analyses of data from a randomised, double-blind, placebo-controlled study conﬁrm quality of life beneﬁt of epoetin alfa in patients receiving non-platinum chemotherapy. Br J Cancer 87: 1341–1353 90. Fatouros MS, Vekinis G, Bourantas KL, et al (1999) Inﬂuence of growth factors erythropoietin and granulocyte macrophage colony-stimulating factor on mechanical strength and healing of colonic anastomosis in rats. Eur J Surg 165: 986–992 91. Flament J, Goldman M, Waterlot Y, et al (1986) Impairment of phagocyte oxidative metabolism in hemodialyzed patients with iron overload. Clin Nephrol 25: 227–230 92. Foley RN, Parfrey PS, Harnett JD, et al (1996) The impact of anemia on cardiomyopathy, morbidity, and mortality in end-stage renal disease. Am J Kidney Dis 28: 53–61 93. Foley RN, Parfrey PS, Kent GM, et al (2000) Serial change in echocardiographic parameters and cardiac failure in end-stage renal disease. J Am Soc Nephrol 11: 912–916 94. Fransen E, Maessen J, Dentener M, et al (1999) Impact of blood transfusions on inﬂammatory mediator release in patients undergoing cardiac surgery. Chest 116: 1233–1239 95. Gabrilove JL, Cleeland CS, Livingston RB, et al (2001) Clinical evaluation of once-weekly dosing of epoetin alfa in chemotherapy patients: improvements in hemoglobin and quality of life are similar to three-times-weekly dosing. J Clin Oncol 19: 2875–2882 96. Ghezzi P, Mengozzi M (2007) Activities of erythropoietin on tumors: an immunological perspective. Eur J Immunol 37: 1427–1430

Impact of anemia and red blood cell transfusion on organ function

357

97. Glaspy J, Bukowski R, Steinberg D, et al (1997) Impact of therapy with epoetin alfa on clinical outcomes in patients with nonmyeloid malignancies during cancer chemotherapy in community oncology practice. Procrit Study Group. J Clin Oncol 15: 1218–1234 98. Glaus A, Müller S (2000) [Hemoglobin and fatigue in cancer patients: inseparable twins?] Schweiz Med Wochenschr 130: 471–477 99. Goto T, Saika H, Takahashi T, et al (1999) Erythropoietin supplement increases plasma lipoprotein lipase and hepatic triglyceride lipase levels in hemodialysis patients. Kidney Int [Suppl] 71: S213-S215 100. Greenberg PL (2006) Myelodysplastic syndromes: iron overload consequences and current chelating therapies. J Natl Compr Canc Netw 4: 91–96 101. Grimm G, Stockenhuber F, Schneeweiss B, et al (1990) Improvement of brain function in hemodialysis patients treated with erythropoietin. Kidney Int 38: 480–486 102. Grossman W, Jones D, McLaurin LP (1975) Wall stress and patterns of hypertrophy in the human left ventricle. J Clin Invest 56: 56–64 103. Guralnik JM, Ershler WB, Schrier SL, et al (2005) Anemia in the elderly: a public health crisis in hematology. Hematology Am Soc Hematol Educ Program: 528–532 104. Habler OP, Messmer KF (1997) The physiology of oxygen transport. Transfus Sci 18: 425–435 105. Hamblin TJ (2005) The management of anemia in the myelodysplastic syndrome. Leuk Res 29: 1101–1102 106. Hardy JF (2004) Current status of transfusion triggers for red blood cell concentrates. Transfusion and Apheresis Science 31: 55–66 107. Harnett JD, Kent GM, Foley RN, et al (1995) Cardiac function and hematocrit level. Am J Kidney Dis 25 [Suppl]: S3-S7 108. Harrison LB, Chadha M, Hill RJ, et al (2002) Impact of tumor hypoxia and anemia on radiation therapy outcomes. Oncologist 7: 492–508 109. Hébert PC, Van der Linden P, Biro G, et al (2004) Physiologic aspects of anemia. Crit Care Clin 20: 187–212 110. Hébert PC, Tinmouth A, Corwin HL (2007) Controversies in RBC transfusion in the critically ill. Chest 131: 1583–1590 111. Hedenus M, Adriansson M, San Miguel J, et al (2003) Efﬁcacy and safety of darbepoetin alfa in anaemic patients with lymphoproliferative malignancies: a randomized, double-blind, placebo-controlled study. Br J Haematol 122: 394–403 112. Heﬂin LH, Meyerowitz BE, Hall P, et al (2005) Cancer as a risk factor for longterm cognitive deﬁcits and dementia. J Natl Cancer Inst 97: 854–856 113. Hellström-Lindberg E, Ahlgren T, Beguin Y, et al (1998) Treatment of anemia in myelodysplastic syndromes with granulocyte colony-stimulating factor plus erythropoietin: results from a randomized phase II study and long-term followup of 71 patients. Blood 92: 68–75 114. Herbert [Hébert] PC, Szick S (2001) The anemic patient in the ICU: How much does the heart tolerate? Anasthesiol Intensivmed Notfallmed Schmerzther 36 [Suppl 2]: S94–S100 115. Hershko C, Peto TE, Weatherall DJ (1988) Iron and infection. Br Med J (Clin Res Ed) 296: 660–664

358

M. R. Nowrousian

116. Hoen B, Kessler M, Hestin D, et al (1995) Risk factors for bacterial infections in chronic haemodialysis adult patients: a multicentre prospective survey. Nephrol Dial Transplant 10: 377–381 117. Högman CF, Knutson F, Loof H (1999a) Storage of whole blood before separation: the effect of temperature on red cell 2,3 DPG and the accumulation of lactate. Transfusion 39: 492–497 118. Högman CF, Meryman HT (2006) Red blood cells intended for transfusion: quality criteria revisited. Transfusion 46: 137–142 119. Högman CF, Meryman HT (1999b) Storage parameters affecting red blood cell survival and function after transfusion. Transfus Med Rev 13: 275–296 120. Hung SC, Tung TY, Yang CS, et al (2005) High-calorie supplementation increases serum leptin levels and improves response to rHuEPO in long-term hemodialysis patients. Am J Kidney Dis 45: 1073–1083 121. Iconomou G, Koutras A, Rigopoulos A, et al (2003) Effect of recombinant human erythropoietin on quality of life in cancer patients receiving chemotherapy: results of a randomized, controlled trial. J Pain Symptom Manage 25: 512–518 122. Igaki N, Takashima M, Ohyama M, et al (2004) The beneﬁcial effect of effective control of anemia on hyperinsulinemia and hypoxemia in a hemodialysis patient with corrected transposition of the great arteries. Clin Exp Nephrol 8: 163–167 123. Izaks GJ, Westendorp RG, Knook DL (1999) The deﬁnition of anemia in older persons. JAMA 281: 1714–1717 124. Jacobsen PB, Garland LL, Booth-Jones M, et al (2004) Relationship of hemoglobin levels to fatigue and cognitive functioning among cancer patients receiving chemotherapy. J Pain Symptom Manage 28: 7–18 125. Jelkmann W, Wagner K (2004) Beneﬁcial and ominous aspects of the pleiotropic action of erythropoietin. Ann Hematol 83: 673–686 126. Jelkmann W (2007) Erythropoietin after a century of research: younger than ever. Eur J Haematol 78: 183–205 127. Jensen LS, Puho E, Pedersen L, et al (2005) Long-term survival after colorectal surgery associated with buffy-coat-poor and leucocyte-depleted blood transfusion: a follow-up study. Lancet 365: 681–682 128. Jensen PD, Heickendorff L, Pedersen B, et al (1996) The effect of iron chelation on haemopoiesis in MDS patients with transfusional iron overload. Br J Haematol 94: 288–299 129. Jones M, Schenkel B, Just J, et al (2004) Epoetin alfa improves quality of life in patients with cancer: results of metaanalysis. Cancer 101: 1720–1732 130. Kaltwasser JP, Gottschalk R (1999) Erythropoietin and iron. Kidney Int [Suppl] 69: S49–S56 131. Kaltwasser JP, Kessler U, Gottschalk R, et al (2001) Effect of recombinant human erythropoietin and intravenous iron on anemia and disease activity in rheumatoid arthritis. J Rheumatol 28: 2430–2436 132. Kapp KS, Poschauko J, Geyer E, et al (2002) Evaluation of the effect of routine packed red blood cell transfusion in anemic cervix cancer patients treated with radical radiotherapy. Int J Radiat Oncol Biol Phys 54: 58–66 133. Katz O, Gil L, Lifshitz L, et al (2007) Erythropoietin enhances immune responses in mice. Eur J Immunol 37: 1584–1593

Impact of anemia and red blood cell transfusion on organ function

359

134. Kausz AT, Obrador GT, Pereira BJ (2000) Anemia management in patients with chronic renal insufﬁciency. Am J Kidney Dis 36 [Suppl]: S39–S51 135. Kelekis NL, Semelka RC, Jeon HJ, et al (1996) Dark ring sign: ﬁnding in patients with fungal liver lesions and transfusional hemosiderosis undergoing treatment with antifungal antibiotics. Magn Reson Imaging 14: 615–618 136. Kettner-Melsheimer A, Weiss M, Huber W (1987) Physical work capacity in chronic renal disease. Int J Artif Organs 10: 23–30 137. Khan FA, Fisher MA, Khakoo RA (2007) Association of hemochromatosis with infectious diseases: expanding spectrum. Int J Infect Dis (in press) 138. Kimata H, Yoshida A, Ishioka C, et al (1991) Human recombinant erythropoietin directly stimulates B cell immunoglobulin production and proliferation in serum-free medium. Clin Exp Immunol 85: 151–156 139. Klein HG, Spahn DR, Carson JL (2007) Red blood cell transfusion in clinical practice. Lancet 370: 415–426 140. Kokot F, Wiecek A, Mesjasz J, et al (1998) Inﬂuence of long-term recombinant human erythropoietin (rHuEpo) therapy on plasma leptin and neuropeptide Y concentration in haemodialysed uraemic patients. Nephrol Dial Transplant 13: 1200–1205 141. Kokot F, Wiecek A, Schmidt-Gayk H, et al (1995) Function of endocrine organs in hemodialyzed patients of long-term erythropoietin therapy. Artif Organs 19: 428–435 142. Komajda M, Anker SD, Charlesworth A, et al (2006) The impact of new onset anaemia on morbidity and mortality in chronic heart failure: results from COMET. Eur Heart J 27: 1440–1446 143. Lawrence IG, Price DE, Howlett TA, et al (1997) Erythropoietin and sexual dysfunction. Nephrol Dial Transplant 12: 741–747 144. Lefebvre P, Vekeman F, Sarokhan B, et al (2006) Relationship between hemoglobin level and quality of life in anemic patients with chronic kidney disease receiving epoetin alfa. Curr Med Res Opin 22: 1929–1937 145. Leitgeb C, Pecherstorfer M, Fritz E, et al (1994) Quality of life in chronic anemia of cancer during treatment with recombinant human erythropoietin. Cancer 73: 2535–2542 146. Levin NW (1992) Quality of life and hematocrit level. Am J Kidney Dis 20: 16–20 147. Li S, Foley RN, Collins AJ (2005) Anemia and cardiovascular disease, hospitalization, end stage renal disease, and death in older patients with chronic kidney disease. Int Urol Nephrol 37: 395–402 148. Lichtman SM, Attivissimo L, Goldman IS, et al (1999) Secondary hemochromatosis as a long-term complication of the treatment of hematologic malignancies. Am J Hematol 61: 262–264 149. Lipschitz D (2003) Medical and functional consequences of anemia in the elderly. J Am Geriatr Soc 51: S10–S13 150. Littlewood TJ, Bajetta E, Nortier JWR, et al (2001) Effects of epoetin alfa on hematologic parameters and quality of life in cancer patients receiving nonplatinum chemotherapy: Results of a randomized, double-blind, placebocontrolled trial. J Clin Oncol 19: 2865–2874 151. Littlewood TJ, Kallich JD, San Miguel J, et al (2006) Efﬁcacy of darbepoetin alfa in alleviating fatigue and the effect of fatigue on quality of life in anemic

360

152.

153. 154. 155.

156. 157. 158. 159. 160. 161.

162.

163. 164.

165.

166.

167.

168.

M. R. Nowrousian patients with lymphoproliferative malignancies. J Pain Symptom Manage 31: 317–325 Littlewood TJ, Nortier J, Rapoport B, et al (2003) Epoetin alfa corrects anemia and improves quality of life in patients with hematologic malignancies receiving non-platinum chemotherapy. Hematol Oncol 21: 169–180 Lubin NLC (2005) Transfusion safety: Where are we today? Ann NY Acad Sci 1054: 325–341 Ludwig H, Fritz E, Kotzmann H, et al (1990) Erythropoietin treatment of anemia associated with multiple myeloma. N Engl J Med 322: 1693–1699 Ludwig H, Van Belle S, Barrett-Lee P, et al (2004) The European Cancer Anaemia Survey (ECAS): a large, multinational, prospective survey deﬁning the prevalence, incidence, and treatment of anaemia in cancer patients. Eur J Cancer 40: 2293–2306 Macdougall IC (1998) Quality of life and anemia: the nephrology experience. Semin Oncol 25 [Suppl 7]: 39–42 Macdougall IC (2001) Should the hematocrit be normalized in dialysis and in pre-ESRD patients? Blood Purif 19: 157–167 Macdougall IC, Lewis NP, Saunders MJ, et al (1990) Long-term cardiorespiratory effects of amelioration of renal anaemia by erythropoietin. Lancet 335: 489–493 Mak RH (1996) Correction of anemia by erythropoietin reverses insulin resistance and hyperinsulinemia in uremia. Am J Physiol 270: F839–F844 Mak RH (1998) Metabolic effects of erythropoietin in patients on peritoneal dialysis. Pediatr Nephrol 12: 660–665 Mancuso A, Migliorino M, De Santis S, et al (2006) Correlation between anemia and functional/cognitive capacity in elderly lung cancer patients treated with chemotherapy. Ann Oncol 17: 146–150 Maraldi C, Volpato S, Cesari M, et al (2006) Anemia and recovery from disability in activities of daily living in hospitalized older persons. J Am Geriatr Soc 54: 632–636 Marsh JT, Brown WS, Wolcott D, et al (1991) rHuEPO treatment improves brain and cognitive function of anemic dialysis patients. Kidney Int 39: 155–163 Massa E, Madeddu C, Lusso MR, et al (2006) Evaluation of the effectiveness of treatment with erythropoietin on anemia, cognitive functioning and functions studied by comprehensive geriatric assessment in elderly cancer patients with anemia related to cancer chemotherapy. Crit Rev Oncol Hematol 57: 175–182 McKay PJ, Murphy JA, Cameron S, et al (1996) Iron overload and liver dysfunction after allogeneic or autologous bone marrow transplantation. Bone Marrow Transplant 17: 63–66 McMahon LP, McKenna MJ, Sangkabutra T, et al (1999) Physical performance and associated electrolyte changes after haemoglobin normalization: a comparative study in haemodialysis patients. Nephrol Dial Transplant 14: 1182–1187 Mercuriali F, Inghilleri G (2002) Red blood cell transfusion, risk and limiations. In: Nowrousian MR (ed) Recombinant human erythropoietin in clinical oncology – Scientiﬁc and clinical aspects of anemia in cancer. Springer, WienNewYork, pp 185–201 Merkel PA, Simonson DC, Amiel SA, et al (1988) Insulin resistance and hyperinsulinemia in patients with thalassemia major treated by hypertransfusion. N Engl J Med 318: 809–814

Impact of anemia and red blood cell transfusion on organ function

361

169. Metivier F, Marchais SJ, Guerin AP, et al (2000) Pathophysiology of anaemia: focus on the heart and blood vessels. Nephrol Dial Transplant 15 [Suppl 3]: 14–18 170. Metry G, Wikstrom B, Valind S, et al (1999) Effect of normalization of hematocrit on brain circulation and metabolism in hemodialysis patients. J Am Soc Nephrol 10: 854–863 171. Miceli MH, Dong L, Grazziutti ML, et al (2006) Iron overload is a major risk factor for severe infection after autologous stem cell transplantation: a study of 367 myeloma patients. Bone Marrow Transplant 37: 857–864 172. Mitchell JE (2007) Emerging role of anemia in heart failure. Am J Cardiol 99: 15D–20D 173. Mitchell TR, Pegrum GD (1971) The oxygen afﬁnity of haemoglobin in chronic renal failure. Br J Haematol 21: 463–472 174. Mittelman M, Neumann D, Peled A, et al (2001) Erythropoietin induces tumor regression and antitumor immune responses in murine myeloma models. Proc Natl Acad Sci U S A 98: 5181–5186 175. Moreno F, Sanz-Guajardo D, Lopez-Gomez JM, et al (2000) Increasing the hematocrit has a beneﬁcial effect on quality of life and is safe in selected hemodialysis patients. Spanish Cooperative Renal Patients Quality of Life Study Group of the Spanish Society of Nephrology. J Am Soc Nephrol 11: 335–342 176. Murphy MF, Wallington TB, Kelsey P, et al (2001) Guidelines for the clinical use of red cell transfusions. Br J Haematol 113: 24–31 177. Murray-Kolb LE, Beard JL (2007) Iron treatment normalizes cognitive functioning in young women. Am J Clin Nutr 85: 778–787 178. Mynster T, Dybkjoer E, Kronborg G, et al (1998) Immunomodulating effect of blood transfusion: is storage time important? Vox Sang 74: 176–181 179. Nasri H (2006) Association of serum leptin with anemia in maintenance hemodialysis patients. Saudi J Kidney Dis Transpl 17: 521–525 180. Negrin RS, Stein R, Doherty K, et al (1996) Maintenance treatment of the anemia of myelodysplastic syndromes with recombinant human granulocyte colony-stimulating factor and erythropoietin: evidence for in vivo synergy. Blood 87: 4076–4081 181. Nieboer P, Buijs C, Rodenhuis S, et al (2005) Fatigue and relating factors in high-risk breast cancer patients treated with adjuvant standard or high-dose chemotherapy: a longitudinal study. J Clin Oncol 23: 8296–8304 182. Nilsson-Ehle H, Jagenburg R, Landahl S, et al (2000) Blood haemoglobin declines in the elderly: implications for reference intervals from age 70 to 88. Eur J Haematol 65: 297–305 183. Nishizawa Y, Shoji T, Tanaka S, et al (1998) Plasma leptin level and its relationship with body composition in hemodialysis patients. Am J Kidney Dis 31: 655–661 184. Nissenson AR (1989) Recombinant human erythropoietin: impact on brain and cognitive function, exercise tolerance, sexual potency, and quality of life. Semin Nephrol 9: 25–31 185. Nonnast-Daniel B, Deschodt G, Brunkhorst R, et al (1990) Long-term effects of treatment with recombinant human erythropoietin on haemodynamics and tissue oxygenation in patients with renal anaemia. Nephrol Dial Transplant 5: 444–448

362

M. R. Nowrousian

186. Nowrousian MR, Kasper C, Oberhoff C, et al (1996) Pathophysiology of cancer-related anemia. In: Smyth JF, Boogaerts MA, Ehmer BRM (eds) rhErythropoietin in cancer supportive treatment. Marcel Dekker, New York, pp 13–34 187. Oberhoff C, Neri B, Amadori D, et al (1998) Recombinant human erythropoietin in the treatment of chemotherapy- induced anemia and prevention of transfusion requirement associated with solid tumors: a randomized, controlled study. Ann Oncol 9: 255–260 188. Offner PJ (2004) Age of blood: does it make a difference? Crit Care 8 [Suppl 2]: S24–S26 189. Oliva EN, Dimitrov BD, Benedetto F, et al (2005) Hemoglobin level threshold for cardiac remodeling and quality of life in myelodysplastic syndrome. Leuk Res 29: 1217–1219 190. O’Riordan E, Foley RN (2000) Effects of anaemia on cardiovascular status. Nephrol Dial Transplant 15: 19–22 191. O’Shaughnessy JA, Vukelja SJ, Holmes FA, et al (2005) Feasibility of quantifying the effects of epoetin alfa therapy on cognitive function in women with breast cancer undergoing adjuvant or neoadjuvant chemotherapy. Clin Breast Cancer 5: 439–446 192. Ossa DF, Briggs A, McIntosh E, et al (2007) Recombinant erythropoietin for chemotherapy-related anaemia: economic value and health-related quality-oflife assessment using direct utility elicitation and discrete choice experiment methods. Pharmacoeconomics 25: 223–237 193. Österborg A, Brandberg Y, Molostova V, et al (2002) Randomized, doubleblind, placebo-controlled trial of recombinant human erythropoietin, epoetin beta, in hematologic malignancies. J Clin Oncol 20: 2486–2494 194. Paczek L, Schaefer RM, Heidland A (1990) Improved function of B lymphocytes in dialysis patients treated by recombinant human erythropoietin. Contrib Nephrol 87: 36–41 195. Palazzuoli A, Silverberg D, Iovine F, et al (2006) Erythropoietin improves anemia exercise tolerance and renal function and reduces B-type natriuretic peptide and hospitalization in patients with heart failure and anemia. Am Heart J 152: 1096, 1099–1015 196. Palmer BF (1999) Sexual dysfunction in uremia. J Am Soc Nephrol 10: 1381–1388 197. Palmer BF (2003) Sexual dysfunction in men and women with chronic kidney disease and end-stage kidney disease. Adv Ren Replace Ther 10: 48–60 198. Palumbo A, Petrucci MT, Lauta VM, et al (2005) Correlation between fatigue and hemoglobin level in multiple myeloma patients: results of a cross-sectional study. Haematologica 90: 858–860 199. Pamuk GE, Demir M, Harmandar F, et al (2006) Leptin and resistin levels in serum of patients with hematologic malignancies: correlation with clinical characteristics. Exp Oncol 28: 241–244 200. Pape A (2007) Alternatives to allogeneic blood transfusions. Best Pract Res Clin Anaesthesiol 21: 221–239 201. Patruta SI, Edlinger R, Sunder-Plassmann G, et al (1998) Neutrophil impairment associated with iron therapy in hemodialysis patients with functional iron deﬁciency. J Am Soc Nephrol 9: 655–663

Impact of anemia and red blood cell transfusion on organ function

363

202. Patruta SI, Horl WH (1999) Iron and infection. Kidney Int [Suppl] 69: S125–S130 203. Peeters HR, Jongen-Lavrencic M, Bakker CH, et al (1999) Recombinant human erythropoietin improves health-related quality of life in patients with rheumatoid arthritis and anaemia of chronic disease; utility measures correlate strongly with disease activity measures. Rheumatol Int 18: 201–206 204. Penninx BW, Pahor M, Cesari M, et al (2004) Anemia is associated with disability and decreased physical performance and muscle strength in the elderly. J Am Geriatr Soc 52: 719–724 205. Penninx BW, Pluijm SM, Lips P, et al (2005) Late-life anemia is associated with increased risk of recurrent falls. J Am Geriatr Soc 53: 2106–2111 206. Petranovic D, Batinac T, Petranovic D, et al (2007) Iron deﬁciency anaemia inﬂuences cognitive functions. Med Hypotheses (in print) 207. Pfafﬂ W, Gross HJ, Neumeier D, et al (1988) Lymphocyte subsets and delayed cutaneous hypersensitivity in hemodialysis patients receiving recombinant human erythropoietin. Contrib Nephrol 66: 195–204 208. Pickett JL, Theberge DC, Brown WS, et al (1999) Normalizing hematocrit in dialysis patients improves brain function. Am J Kidney Dis 33: 1122–1130 209. Pierelli L, Perillo A, Greggi S, et al (1999) Erythropoietin addition to granulocyte colony-stimulating factor abrogates life-threatening neutropenia and increases peripheral-blood progenitor-cell mobilization after epirubicin, paclitaxel, and cisplatin combination chemotherapy: results of a randomized comparison. J Clin Oncol 17: 1288–1295 210. Ponikowski P, Anker SD, Szachniewicz J, et al (2007) Effect of darbepoetin alfa on exercise tolerance in anemic patients with symptomatic chronic heart failure: a randomized, double-blind, placebo-controlled trial. J Am Coll Cardiol 49: 753–762 211. Porter JB (2005) Monitoring and treatment of iron overload: state of the art and new approaches. Semin Hematol 42 [Suppl 1]: S14–S18 212. Portoles J, Lopez-Gomez JM, Aljama P (2007) A prospective multicentre study of the role of anaemia as a risk factor in haemodialysis patients: the MAR study. Nephrol Dial Transplant 22: 500–507 213. Prutchi-Sagiv S, Golishevsky N, Oster HS, et al (2006) Erythropoietin treatment in advanced multiple myeloma is associated with improved immunological functions: could it be beneﬁcial in early disease? Br J Haematol 135: 660–672 214. Raghavan M, Marik PE (2005) Anemia, allogenic blood transfusion, and immunomodulation in the critically ill. Chest 127: 295–307 215. Reed W, Lee TH, Norris PJ, et al (2007) Transfusion-associated microchimerism: a new complication of blood transfusions in severely injured patients. Semin Hematol 44: 24–31 216. Revicki DA, Brown RE, Feeny DH, et al (1995) Health-related quality of life associated with recombinant human erythropoietin therapy for predialysis chronic renal disease patients. Am J Kidney Dis 25: 548–554 217. Riedel E, Nundel M, Wendel G, et al (2000) Amino acid and alpha-keto acid metabolism depends on oxygen availability in chronic hemodialysis patients. Clin Nephrol 53 [Suppl 1]: S56–S60 218. Rigatto C, Foley R, Jeffery J, et al (2003) Electrocardiographic left ventricular hypertrophy in renal transplant recipients: prognostic value and impact of blood pressure and anemia. J Am Soc Nephrol 14: 462–468

364

M. R. Nowrousian

219. Rinker BD, Bowling JT, Vasconez HC (2007) Blood transfusion and risk of metastatic disease or recurrence in patients undergoing immediate TRAM ﬂap breast reconstruction: a clinical study and meta-analysis. Plast Reconstr Surg 119: 2001–2007 220. Robinson B (2003) Cost of anemia in the elderly. J Am Geriatr Soc 51: S14–S17 221. Robinson BM, Joffe MM, Berns JS, et al (2005) Anemia and mortality in hemodialysis patients: accounting for morbidity and treatment variables updated over time. Kidney Int 68: 2323–2330 222. Rofstad EK, Galappathi K, Mathiesen B, et al (2007) Fluctuating and diffusionlimited hypoxia in hypoxia-induced metastasis. Clin Cancer Res 13: 1971– 1978 223. Rosner F, Grünewald HW (1997) The patient with anemia. Med Update for Psychiatrists 2: 177–180 224. Rossi EC (1994) Red cell transfusion therapy in chronic anemia. Hematol Oncol Clin North Am 8: 1045–1052 225. Rushton DH, Dover R, Sainsbury AW, et al (2001) Why should women have lower reference limits for haemoglobin and ferritin concentrations than men? Br Med J 322: 1355–1357 226. Russo F, Guadagni S, Mattera G, et al (1999) Combination of granulocytemacrophage colony-stimulating factor (GM-CSF) and erythropoietin (EPO) for the treatment of advanced non-responsive chronic lymphocytic leukemia. Eur J Haematol 63: 325–331 227. Sabatine MS, Morrow DA, Giugliano RP, et al (2005) Association of hemoglobin levels with clinical outcomes in acute coronary syndromes. Circulation 111: 2042–2049 228. Sagales T, Gimeno V, Planella MJ, et al (1993) Effects of rHuEPO on Q-EEG and event-related potentials in chronic renal failure. Kidney Int 44: 1109– 1115 229. Sakanaka M, Wen TC, Matsuda S, et al (1998) In vivo evidence that erythropoietin protects neurons from ischemic damage. Proc Natl Acad Sci U S A 95: 4635–4640 230. Sandgren PE, Murray AM, Herzog CA, et al (2005) Anemia and new-onset congestive heart failure in the general Medicare population. J Card Fail 11: 99–105 231. Santhanam AV, Katusic ZS (2006) Erythropoietin and cerebral vascular protection: role of nitric oxide. Acta Pharmacol Sin 27: 1389–1394 232. Savonije JH, van Groeningen CJ, Wormhoudt LW, et al (2006) Early intervention with epoetin alfa during platinum-based chemotherapy: an analysis of the results of a multicenter, randomized, controlled trial based on initial hemoglobin level. Oncologist 11: 206–216 233. Schaefer F, van Kaick B, Veldhuis JD, et al (1994) Changes in the kinetics and biopotency of luteinizing hormone in hemodialyzed men during treatment with recombinant human erythropoietin. J Am Soc Nephrol 5: 1208–1215 234. Schafer AI, Cheron RG, Dluhy R, et al (1981) Clinical consequences of acquired transfusional iron overload in adults. N Engl J Med 304: 319–324 235. Schleiffenbaum B (1994) Verminderte Sauerstofftransport-Kapazität als Ursache von Dyspnoe. Schweiz Med Wochenschr 124: 1177–1182 236. Schmitt HJ, Götz E (1988) Metabolische Störungen durch Bluttransfusionen. Infusionstherapie 15: 254–260

Impact of anemia and red blood cell transfusion on organ function

365

237. Seifert A, von Herrath D, Schaefer K (1987) Iron overload, but not treatment with desferrioxamine favours the development of septicemia in patients on maintenance hemodialysis. Q J Med 65: 1015–1024 238. Sen A, Kanani SJ (2006) Deleterious functional impact of anemia on young adolescent school girls. Indian Pediatr 43: 219–226 239. Sennesael JJ, van der Niepen P, Verbeelen DL (1991) Treatment with recombinant human erythropoietin increases antibody titers after hepatitis B vaccination in dialysis patients. Kidney Int 40: 121–128 240. Shorr AF, Jackson WL (2005) Transfusion practice and nosocomial infection: assessing the evidence. Curr Opin Crit Care 11: 468–472 241. Shurtz-Swirski R, Kristal B, Shkolnik T, et al (1996) Short-term effect of erythropoietin on T-cell mitogenic proliferation in chronic renal failure patients. Nephron 72: 27–29 242. Silverberg DS, Wexler D, Blum M, et al (2000) The use of subcutaneous erythropoietin and intravenous iron for the treatment of the anemia of severe, resistant congestive heart failure improves cardiac and renal function and functional cardiac class, and markedly reduces hospitalizations. J Am Coll Cardiol 35: 1737–1744 243. Silverberg DS, Iaina A, Wexler D, et al (2001a) The pathological consequences of anemia. Clin Lab Haematol 23: 1–6 244. Silverberg DS, Wexler D, Sheps D, et al (2001b) The effect of correction of mild anemia in severe, resistant congestive heart failure using subcutaneous erythropoietin and intravenous iron: a randomized controlled study. J Am Coll Cardiol 37: 1775–1780 245. Silverberg DS, Wexler D, Iaina A, et al (2006) The interaction between heart failure and other heart diseases, renal failure, and anemia. Semin Nephrol 26: 296–306 246. Singh AB, Singh M, Palekar S, et al (1992) The effects of recombinant human erythropoietin on the cell mediated immune response of renal failure patients. J Med 23: 289–302 247. Singh NP, Sahni V, Wadhwa A, et al (2006) Effect of improvement in anemia on electroneurophysiological markers (P300) of cognitive dysfunction in chronic kidney disease. Hemodial Int 10: 267–273 248. Siren AL, Fratelli M, Brines M, et al (2001) Erythropoietin prevents neuronal apoptosis after cerebral ischemia and metabolic stress. Proc Natl Acad Sci USA 98: 4044–4049 249. Sit D, Kadiroglu AK, Yilmaz ME, et al (2005) The prevalence of insulin resistance and its relationship between anemia, secondary hyperparathyroidism, inﬂammation, and cardiac parameters in chronic hemodialysis patients. Ren Fail 27: 403–407 250. Sobrero A, Puglisi F, Guglielmi A, et al (2001) Fatigue: a main component of anemia symptomatology. Semin Oncol 28 [Suppl 8]: 15–18 251. Sommerburg O, Grune T, Hampl H, et al (2000) Does treatment of renal anemia with recombinant erythropoietin inﬂuence oxidative stress in hemodialysis patients? Clin Nephrol 53 [1 Suppl 1]: S23–S29 252. Souza LM, Boone TC, Gabrilove J, et al (1986) Recombinant human granulocyte colony-stimulating factor: effects on normal and leukemic myeloid cells. Science 232: 61–65

366

M. R. Nowrousian

253. Sowade O, Gross J, Sowade B, et al (1997) Evaluation of oxygen availability with oxygen status algorithm in patients undergoing open heart surgery treated with epoetin beta. J Lab Clin Med 129: 97–105 254. Spaia S, Pangalos M, Askepidis N, et al (2000) Effect of short-term rHuEPO treatment on insulin resistance in haemodialysis patients. Nephron 84: 320–325 255. Spiess BD (2004) Risks of transfusion: outcome focus. Transfusion 44: 4S–14S 256. Srivastava PM, Thomas MC, Calaﬁore P, et al (2006) Diastolic dysfunction is associated with anaemia in patients with type II diabetes. Clin Sci (Lond) 10: 109–116 257. Stack G, Baril L, Napychank P, et al (1995) Cytokine generation in stored, white cell-reduced, and bacterially contaminated units of red cells. Transfusion 35: 199–203 258. Stivelman JC (2000) Beneﬁts of anaemia treatment on cognitive function. Nephrol Dial Transplant 15: 29–35 259. Stockenhuber F, Kurz RW, Geissler K, et al (1990) Recombinant human erythropoietin activates a broad spectrum of progenitor cells. Kidney Int 37: 150–156 260. Straus DJ, Testa MA, Sarokhan BJ, et al (2006) Quality-of-life and health beneﬁts of early treatment of mild anemia: a randomized trial of epoetin alfa in patients receiving chemotherapy for hematologic malignancies. Cancer 107: 1909–1917 261. Stroncek DF (2007) Pulmonary transfusion reactions. Semin Hematol 44: 2–14 262. Suttner S, Piper SN, Kumle K, et al (2004) Allogeneic red blood cell transfusion compared with 100% oxygen ventilation on systemic oxygen transport and skeletal muscle oxygen tension after cardiac surgery. Anesth Analg 99: 2–11 263. Szakmany T, Dodd M, Dempsey GA, et al (2006) The inﬂuence of allogenic blood transfusion in patients having free-ﬂap primary surgery for oral and oropharyngeal squamous cell carcinoma. Br J Cancer 94: 647–653 264. Takatoku M, Uchiyama T, Okamoto S, et al (2007) Retrospective nationwide survey of Japanese patients with transfusion-dependent MDS and aplastic anemia highlights the negative impact of iron overload on morbidity/mortality. Eur J Haematol 78: 487–494 265. Tang WH (2007) Anemia in heart failure: current evidence and challenges. Rev Cardiovasc Med 8: 78–86 266. Taniguchi Y, Okura M (2003) Prognostic signiﬁcance of perioperative blood transfusion in oral cavity squamous cell carcinoma. Head Neck 25: 931–936 267. Temple RM, Deary IJ, Winney RJ (1995) Recombinant erythropoietin improves cognitive function in patients maintained on chronic ambulatory peritoneal dialysis. Nephrol Dial Transplant 10: 1733–1738 268. Thomas MC, Cooper ME, Rossing K, et al (2006) Anemia in diabetes: is there a rationale to treat? Diabetologia 49: 1151–1157 269. Tielemans CL, Lenclud CM, Wens R, et al (1989) Critical role of iron overload in the increased susceptibility of haemodialysis patients to bacterial infections. Beneﬁcial effects of desferrioxamine. Nephrol Dial Transplant 4: 883–887 270. Tokgoz B, Utas C, Dogukan A, et al (2001) Effects of long-term erythropoietin therapy on the hypothalamo-pituitary-testicular axis in male CAPD patients. Perit Dial Int 21: 448–454 271. Valentine AD, Meyers CA, Kling MA, et al (1998) Mood and cognitive side effects of interferon-alpha therapy. Semin Oncol 25 [Suppl 1]: 39–47

Impact of anemia and red blood cell transfusion on organ function

367

272. Valeri CR, Cassidy G, Pivacek LE, et al (2001) Anemia-induced increase in the bleeding time: implications for treatment of nonsurgical blood loss. Transfusion 41: 977–983 273. Vanholder R, Van Biesen W, Ringoir S (1993) Contributing factors to the inhibition of phagocytosis in hemodialyzed patients. Kidney Int 44: 208–214 274. Vansteenkiste J, Pirker R, Massuti B, et al (2002) Double-blind, placebo-controlled, randomized phase III trial of darbepoetin alfa in lung cancer patients receiving chemotherapy. J Natl Cancer Inst 94: 1211–1220 275. Varat MA, Adolph RJ, Fowler NO (1972) Cardiovascular effects of anemia. Am Heart J 83: 415–426 276. Vaupel P, Mayer A, Hockel M (2006) Impact of hemoglobin levels on tumor oxygenation: the higher, the better? Strahlenther Onkol 182: 63–71 277. Vaupel P, Mayer A (2007) Hypoxia in cancer: signiﬁcance and impact on clinical outcome. Cancer Metastasis Rev 26: 225–239 278. Vogelzang NJ, Breitbart W, Cella D, et al (1997) Patient, caregiver, and oncologist perceptions of cancer-related fatigue: results of a tripart assessment survey. The Fatigue Coalition. Semin Hematol 34 [Suppl 2]: 4–12 279. Weinberg ED (1999) Iron therapy and cancer. Kidney Int [Suppl] 69: S131–S134 280. Weinmann M, Jendrossek V, Guner D, et al (2004) Cyclic exposure to hypoxia and reoxygenation selects for tumor cells with defects in mitochondrial apoptotic pathways. FASEB J 18: 1906–1908 281. Weiskopf RB, Kramer JH, Viele M, et al (2000) Acute severe isovolemic anemia impairs cognitive function and memory in humans. Anesthesiology 92: 1646–1652 282. Weiss G, Houston T, Kastner S, et al (1997) Regulation of cellular iron metabolism by erythropoietin: activation of iron-regulatory protein and upregulation of transferrin receptor expression in erythroid cells. Blood 89: 680–687 283. Williams WJ (1988) Examination of the blood. Hematology. McGraw-Hill Book, New York, pp 9–21 284. World Health Organization (1968) Nutritional Anemias. Report of a WHO scientiﬁc group. Geneva 285. Yamazaki T, Kanzaki M, Kamidono S, et al (2004) Effect of erythropoietin on Leydig cell is associated with the activation of Stat5 pathway. Mol Cell Endocrinol 213: 193–198 286. Yeh JJ, Gonen M, Tomlinson JS, et al (2007) Effect of blood transfusion on outcome after pancreaticoduodenectomy for exocrine tumor of the pancreas. Brit J Surg 94: 466–472 287. Zallen G, Moore EE, Ciesla DJ, et al (2000) Stored red blood cells selectively activate human neutrophils to release IL-8 and secretory PLA2. Shock 13: 29–33 288. Zamboni V, Cesari M, Zuccala G, et al (2006) Anemia and cognitive performance in hospitalized older patients: results from the GIFA study. Int J Geriatr Psychiatry 21: 529–534 289. Zander R (1978) [The oxygen transport capacity of blood replacement ﬂuids compared with other infusion ﬂuids.] Klin Wochenschr 56: 567–573 Correspondence: Prof. Dr. M. R. Nowrousian, Department of Internal Medicine (Cancer Research), West German Cancer Center, University Hospital of Essen, Hufelandstrasse CB 55, 45122 Essen, Germany, E-mail: [email protected]

Chapter 14

Relationship of hemoglobin, fatigue, and quality of life in anemic cancer patients Z. Butt and D. Cella Center of Outcomes, Research and Education, Evanston Northwestern Healthcare, Evanston, IL, USA

Introduction Anemia is an abnormal reduction in red blood cells (RBC) caused by difﬁculties in RBC production, recycling, and/or regulation. The resultant decrease in hemoglobin (Hb) leads to a reduction in the oxygen-carrying capacity of blood. In turn, this leads to patient weakness, pallor, dyspnea, and most commonly, fatigue. Given their somewhat nonspeciﬁc nature, the symptoms of anemia are often difﬁcult to attribute directly to the anemia itself. However, the impact of low Hb can be far-reaching (Beghe et al. 2004; Pujade-Lauraine and Gascon 2004). Important heath-related outcomes, such as quality of life (QOL) can be enhanced with proper evaluation and treatment of fatigue and other anemia-related symptoms. While it is well recognized that fatigue is a common symptom among cancer patients, it remains a relatively poorly understood phenomenon (Cella et al. 2002b; Lipman and Lawrence 2004). Assessment and treatment of cancer-related fatigue is complicated by its multifactorial nature, involving both physical and psychological components. Depending on the patient population and means of measuring fatigue, prevalence estimates for fatigue among cancer patients are generally quite high, ranging from 60 to over 90% (Wagner and Cella 2004). Patients may describe their experience of fatigue in terms of being exhausted, tired, weak, or slowed. In an effort to operationalize the fatigue construct/syndrome, a multidisciplinary group of clinicians, researchers, and patient advocates developed diagnostic criteria for cancer-related fatigue (Cella et al. 1998). These criteria take into account both fatigue presence and severity. Cases identiﬁed by these criteria tend to experience quite signiﬁcant fatigue; compared to patients identiﬁed with cancer-related fatigue using previous criteria, their fatigue is more severe and functionally impairing (Sadler et al. 2002). Recent ﬁndings suggest that the assessment and treatment of cancerrelated fatigue may be impacted by patient-oncologist miscommunication

370

Z. Butt and D. Cella

and/or inconsistent beliefs and attitudes. According to a survey conducted by Curt et al. (2000), approximately 75% of patients undergoing chemotherapy reported a “general feeling of debilitating tiredness or loss of energy” at least a few days a month, and 30% experienced fatigue on a daily basis. This same sample of patients reported that their caregivers used an average of 4.5 sick/vacation days per month during or immediately after treatment as a direct result of fatigue. Vogelzang et al. (1997) interviewed over 400 cancer patients, more than half of whom reported that fatigue had a more signiﬁcant impact on their lives than cancer-related pain. Seventy-seven percent of patients who discussed fatigue with their oncologists reported that the physicians told them that fatigue was something to endure (Vogelzang et al. 1997). Interestingly, when oncologists were surveyed regarding the need to treat cancer-related fatigue, 80% believed that the symptom is suboptimally treated by their peers. Cancer-related fatigue is clearly an important symptom for patients and clinicians to address. Anemia is one of many potential causes of cancer-related fatigue, which in turn may adversely impact patients’ QOL.

Measurement of anemia-related quality of life For the anemic cancer patient, there may be multiple physical, functional, social, and emotional obstacles to overcome related to the disease as well as its treatment. For example, while fatigue is a common experience for anemic patients undergoing chemotherapy, its etiology may be multifactorial. In addition to being a byproduct of the anemia, fatigue may be brought about by mood disturbance, sleep disturbance, infection, pain, and/or tumor burden. Self-report instruments are the most common – and perhaps the most appropriate – way to assess the subjective impact of fatigue on the anemic patients’ QOL. Some of these instruments are as brief as a single screening question, while others are more comprehensive.

Single-item screens In some cases, a single-item screening instrument may help identify cases of signiﬁcant cancer-related fatigue. Based on the recommendations of the National Comprehensive Cancer Network (Mock et al. 2000), Wagner et al. (2003) asked 482 ambulatory outpatients to rate the severity of their fatigue over the previous 3 days from 0 to 10 (most severe). Between 93–98% of patients in the cut score range of 3–5 reported that some relief in their fatigue would signiﬁcantly improve their lives. A subset of 80 of these patients were also administered a longer measure of fatigue, with which the single item correlated modestly (r = 0.47). Based on established criteria for identifying cases of signiﬁcant fatigue, an optimal cutoff of 5 on the one-item screening instru-

Hemoglobin, fatigue, and quality of life

371

ment was suggested to identify cases for follow-up. While potentially promising, further research on such screening tools is necessary to determine whether their use leads to improved outcomes for cancer patients.

Fatigue questionnaires Single-item assessments, while potentially useful for screening purposes, have limited psychometric validity and offer only a limited understanding of a patient’s fatigue experience. Wu and McSweeney (2001) recently reviewed a number of self-report instruments that have been validated to measure fatigue in patients with cancer. Among the instruments mentioned are the Revised Piper Fatigue Scale (Piper et al. 1998), the Cancer Fatigue Scale (Okuyama et al. 2000), and the Multidimensional Fatigue Inventory (Smets et al. 1995). Many of these questionnaires were developed for the express purpose of assessing cancer-related fatigue and most have good to adequate psychometric properties (Wu and McSweeney 2001). However, there are unanswered questions related to the appropriate use of self-reports of fatigue in clinical decision-making (Jacobsen 2004). Single-item and other brief screening instruments can provide useful information on fatigue, but necessarily capture limited QOL information from the anemic cancer patient. We know that anemia may impact patients’ QOL by reducing exercise tolerance, ability to work, social interaction, and ability to participate in leisure activities (Ludwig and Strasser 2001). Ideally, a QOL assessment would measure fatigue as well as those other domains thought to be affected by anemia (Beghe et al. 2004; Pujade-Lauraine and Gascon 2004). Two of the most commonly used instruments to assess anemiarelated QOL are the Cancer Linear Analog Scale (CLAS) and the Functional Assessment of Chronic Illness Therapy – Anemia Scale (FACIT-An).

Multidimensional QOL measures The Cancer Linear Analog Scale (CLAS) consists of three independent component scales, which assess energy, ability to do daily activities, and general quality of life. (The CLAS is also known as the Linear Analog Scale Assessment, or LASA.) Each of the three CLAS scales are rated along a 100 mm line: the left extreme represents the worst possible score and the right extreme represents the best possible score. Patients mark each symptom scale along the 100 mm line; the location of these marks indicates the magnitude of their response. The validity and reliability of the linear analog approach to assessing QOL has been favorably reviewed (Gough et al. 1983; McCormack et al. 1988). Furthermore, the speciﬁc set of three CLAS items mentioned above has been used to assess QOL in a number of clinical trials

372

Z. Butt and D. Cella

of epoetin alfa in anemic cancer patients (Jones et al. 2004). While the CLAS is easy to administer, it is somewhat unclear how patients and clinicians should interpret scores or changes in the scores. In a recent report, de Boer et al. (2004) presented ﬁndings from a small, homogeneous oncology sample that suggested that a visual analog scale may be responsive to important clinical changes; further research is necessary to evaluate such instruments. Depending on the assessment purpose or setting, it may be useful to include a validated measure of QOL with scores that are indexed against important clinical changes in the anemic cancer patient (Jacobsen 2004).

Functional Assessment of Chronic Illness Therapy – Anemia (FACIT-An) The Functional Assessment of Chronic Illness Therapy (FACIT) measurement system takes a modular approach in the assessment of QOL. Given its increasing validation and application outside of the cancer experience, its name was changed from its original Functional Assessment of Cancer Therapy (FACT). At the core of the system is the Functional Assessment of Cancer Therapy – General (FACT-G), a 27-item general QOL measure that measures physical well-being (PWB), social/family well-being (SFWB), emotional well-being (EWB), and functional well-being (FWB). Additional subscales are added to this core to develop symptom- or population-speciﬁc QOL instruments. A panel of oncology patients and experts was interviewed to create a subscale tailored speciﬁcally to symptoms and limitations associated with anemia. Based on these panel interviews, a 13-item fatigue subscale was created. An additional seven non-fatigue, anemia-related items were also written. Together these items comprise the 20-item anemia subscale. The fatigue subscale can be summed with PWB and FWB to form the Trial Outcome Index-Fatigue (TOI-F). A similar scale can be created using the anemia subscale (TOI-An). The FACIT-Fatigue (or FACIT-F) consists of the FACT-G plus the fatigue subscale; similarly, the FACIT-Anemia (or FACITAn) consists of FACT-G and the anemia subscale. The 47-item FACIT-An was designed to assess fatigue and anemia-related concerns in people with chronic illness. [See Appendix A for a copy of the FACIT-An.] The original FACIT-An validation sample consisted of 50 (27 females) previously untested cancer patients with varying hemoglobin levels. These patients were all receiving treatment; however, patients were excluded if they 1) were receiving cytotoxic chemotherapy or surgery either within the past week or expected within one week; 2) were currently receiving radiation therapy; 3) received either a blood transfusion or growth factor injection 3 days prior to data collection; 4) were anticipating transfusion or a growth factor injection within one week after baseline; 5) had brain metastasis; or 6) were pregnant. The ﬁnal sample consisted of 49 patients at baseline, with 44 of these patients providing evaluable retest data (Yellen et al. 1997).

Hemoglobin, fatigue, and quality of life

373

Results indicated that patients’ responses to the FACT-F and FACT-An were both stable (test-retest rs = 0.87) and internally consistent (coefﬁcient alpha range = 0.95–0.96). Furthermore, as expected, the FACT-F and FACTAn were signiﬁcantly correlated with other known measures of fatigue (magnitudes of correlation coefﬁcients = 0.74 to 0.77). Divergent validity was demonstrated by the strong correlation with a measure of vigor in the predicted direction (magnitudes of correlation coefﬁcients = 0.65 to 0.66) and a lack of relationship between the scales and a measure of social desirability (Yellen et al. 1997). The FACT-F and FACT-An were also found to discriminate patients successfully based on Hb level and patient-rated performance status (Cella 1997). When patients were divided into groups by Hb levels, patients with Hb greater than 13 g/dL reported signiﬁcantly less fatigue, fewer non-fatigue anemia-related symptoms, and better overall quality of life than patients with Hb < 11 g/dL (Cella 1997; Yellen et al. 1997).

Relationship between hemoglobin, fatigue, and quality of life Anemia may impact several aspects of an individual’s QOL including decreased capacity to exercise, increased weakness, lightheadedness, and dizziness (Cella 1997). In addition, anemia subjectively impacts important aspects of individuals’ lives related to their ability to work, socialize, and enjoy leisure pursuits (Cella 1997; Yellen et al. 1997). While our focus in this chapter is on the association of anemia, fatigue and QOL in patients with cancer, it is also important to note that there is evidence of a link between these variables in the general population. In order to appreciate differences between the fatigue experiences of these populations, Cella et al. (2002b) administered the 13-item Fatigue Subscale of the FACT to three groups: 1,010 individuals in the general US population, 113 non-anemic cancer patients, and 2,369 anemic cancer patients. Most participants (84.4%) in the general population described their health as excellent, good, or very good, whereas 15.6% reported fair or poor health on the 12-item short form of the Medical Outcomes Survey (MOS SF-12, Ware et al. 1995). Efforts to ensure that the general population sample was representative of the US population were successful. The nonanemic cancer patients were pooled from two datasets that included both Hb and FACT QOL scores. The anemic cancer patients had recently been entered into an open-label trial of recombinant human erythropoietin (rhEPO) for treatment of chemotherapy-related anemia. The general population group reported less fatigue (higher QOL scores) than both of the cancer groups. As expected, nonanemic cancer patients reported statistically signiﬁcant less fatigue than their anemic counterparts. The mean, median, and range of scores obtained from each group are shown in Table 1. When gender, age, and Hb values were entered into a regression

374

Z. Butt and D. Cella

Table 1. Comparing fatigue in the general population with fatigue in the context of cancer FACTFatigue Subscale

Group General Population (G) n = 1,010

Non-anemic Cancer Patients (N) n = 113

Anemic Cancer Patients (A) n = 2,292a

F (post-hoc)

Mean (SD) Median Range

43.6 (9.4) 47 2–52

40.0 (9.8) 42 9–52

23.9 (12.6) 23 0–52

1,071.8b, G > N > A – –

Notes: (1) Adapted from Cella, Lai, Chang, Peterman and Slavin (2002b). (2) Higher scores indicate better quality-of-life/less fatigue. a Of 2,369 anemic cancer patients, 97% completed the baseline questionnaire. b p < 0.001.

model to predict fatigue in the anemic cancer patient sample, Hb emerged as the only signiﬁcant predictor (standardized β = 0.275, p < 0.001), accounting for approximately 8% of the variance in fatigue scores. As discussed earlier, one of the most important symptoms reported by patients with cancer is fatigue (Wagner and Cella 2004). Although not the sole cause of fatigue in cancer, anemia was deﬁnitely an important contributing factor in this sample. So while fatigue is a common symptom among the general population, it is more severe in patients with cancer. As a follow-up to, Cella et al. (2003) administered the FACT-An to a nationally representative sample of 1,400 people using an Internet survey panel to allow meaningful comparisons to cancer clinical trial data. The sample is fully described in the original publication, but brieﬂy, 1,400 people were randomly drawn from more than 100,000 individuals enrolled in an Internet-based survey panel, representing a demographically balanced sample of the United States population. Respondents answered questionnaires in their home. To make the FACT-An suitable for administration to a general population, six items referencing treatment or illness (1 from PWB, 2 from SFWB, 2 from EWB, 1 from FWB) were eliminated and these missing values were derived. Of the 1,400 participants who were selected to participate in the survey, 1,068 (76.2%) provided evaluable data. Data from the general population were compared to the QOL data of a randomized, double-blind clinical trial of epoetin alfa in anemic cancer patients (n = 375). Results indicated that the FACT-An was successful in discriminating respondents in the general population with self-reported histories of anemia and cancer vs. those without. Cancer patients reported lower QOL than the general population,

Hemoglobin, fatigue, and quality of life

375

but treatment with epoetin alfa resulted in statistically, as well as clinically meaningful QOL improvements (Cella et al. 2003b). We now know that patients with low Hb values (i.e. <12 g/dL) report more fatigue, more nonfatigue anemia symptoms, worse physical well-being, worse functional well-being, and poorer general QOL on FACIT-based instruments than patients with Hb values ≥12 g/dL (Cella 1997; Yellen et al. 1997). These data provide some interesting, preliminary information on the relationship between anemia, fatigue and QOL. Recently, Jacobsen et al. (2004) examined the longitudinal relationship of Hb changes to changes in fatigue in a mixed group of cancer patients. All participants were scheduled for four cycles of chemotherapy. Participants in the study were 77 patients who completed a baseline and follow-up assessment. Follow-up assessments were conducted prior to the start of the fourth cycle of chemotherapy, but within one week of completion of the third cycle. In addition to a self-report measure of fatigue, the Fatigue Symptom Inventory (Hann et al. 2000), patients completed blood work and a cognitive assessment at each assessment. On the basis of regression analysis, changes in Hb from baseline to follow-up assessment were related to changes in fatigue disruptiveness. Speciﬁcally, declines in Hb resulted in increases in fatigue disruptiveness. Of note, in the full sample, there was no association between changes in Hb and fatigue severity and number of days in the past week that patients experienced fatigue. In a subset of 49 patients whose Hb values dropped below 12 g/dL, regression analyses suggested that greater declines in Hb were signiﬁcantly related to increases in fatigue disruptiveness and number of days in the past week the patient was fatigued. However, there was no association between Hb and fatigue severity in this low-hemoglobin subgroup.

Clinically signiﬁcant changes in quality of life While there has been an increasing appreciation for patient-reported outcomes, such as QOL, clinicians may ﬁnd some difﬁculty in interpreting information gleaned from a QOL evaluation. One way to increase the clinical utility of scores on questionnaires like the FACT-An is by exploring crosssectional differences and longitudinal changes scores. Speciﬁcally, anchoring change scores to familiar clinical events that are related to patient well-being can increase the utility of QOL scores. Simply describing statistically/probabilistically signiﬁcant changes (i.e. p < 0.05) may not convey meaningful information to patients or clinicians. Indeed, there has been a growing movement to demonstrate the clinical signiﬁcance of changes in QOL scores. As described by Jaeschke et al. (1989), the minimal clinically important difference (MCID) in QOL score is “the smallest difference in score in the domain of interest which patients perceive as beneﬁcial and which would

376

Z. Butt and D. Cella

mandate, in the absence of troublesome side effects and excessive cost, a change in the patient’s management.”Among the methods used to derive these minimal clinically important differences are anchor-based and distributionbased techniques (Guyatt et al. 2002; Movsas 2003). Brieﬂy, anchor-based methods map QOL difference scores onto cross-sectional (i.e. between groups) or longitudinal (i.e. one group over time) differences in clinical outcomes. On the other hand, distribution-based methods focus on the distribution of QOL scores within a study, given by such indices as the standard deviation and/or standard error of measurement (Lydick and Epstein 1993). It may be possible to increase the practical use of QOL assessment by linking changes in QOL scores to clinically important and meaningful differences. Cella et al. (2002a) used both anchor- and distribution-based criteria to determine clinically signiﬁcant differences on the FACT-F and FACT-An instruments. QOL was assessed in three samples: (1) 50 mixed-diagnosis cancer patients currently receiving treatment; (2) 131 mixed-diagnosis cancer patients in an observational study of fatigue and QOL during chemotherapy; and (3) 2,402 mixed-diagnosis cancer patients in an open-label, nonrandomized, community-based clinical trial evaluating a treatment for anemia in cancer patients. [Note: Portions of the QOL data from samples 2 and 3 were also reported on in Cella et al. 2002b discussed above.] Across all samples, the FACT scales were internally consistent (coefﬁcient αs ≥ 0.85). In addition to FACT-derived subscales, patient demographic and medical data were available. Of particular interest in the patients medical data were baseline and follow-up Hb values as well as patient- and physician-rated performance status ratings (PSR). For the purposes of the present discussion, we present data from the Hb analyses as well as tables summarizing ﬁndings from both Hb and PSR analyses. To begin with, FACT scores were cross-sectionally compared across Hb level. First, patients were divided into three groups, based on their Hb values. Samples 1 and 2 were trichotomized by Hb into <10 g/dL, 10–11.9 g/dL, and ≥12 g/dL, while patients in sample 3, who were participating in a clinical trial of the effects of epoetin alfa on anemia, were divided into <8 g/dL, 8–9.9 g/dL, and 10–11 g/dL. These hemoglobin categories may be thought of as clinically distinguishable groups against which to gauge QOL differences. By investigating mean differences and effect sizes of adjacent clinical categories, Cella et al. (2002a) were able to estimate anchor-based clinically important differences on the FACT. Tables 2 and 3 show the results of these analyses. A similar strategy was used to index cross-sample differences in baseline PSR. Cella et al. (2002a) also quantiﬁed clinically important differences in QOL by comparing FACT change scores to changes in clinical status over time. Speciﬁcally, in sample 2, changes in FACT subscales were compared to changes in Hb from baseline to the 6- and the 12-month assessments. Results of these analyses can be seen in Table 4. In sample 3, changes in FACT subscales were indexed against changes in Hb from baseline to trial completion

33.8 (13.0) 79.2 (15.5) 131.3 (24.5) 68.5 (19.0) 86.8 (22.3)

76.1 (16.8) 124.8 (33.4) 64.7 (23.5) 82.6 (28.6)

(2) 10–11.99 g/dL (n = 13)

30.8 (14.9)

(1) <10 g/dL (n = 9)

Hemoglobin Level

90.7 (11.5) 152.1 (20.9) 85.1 (15.7) 106.4 (18.5)

40.2 (8.4)

(3) ≥12 g/dL (n = 27)

Notes: (1) Adapted from Cella, Eton, Lai, Peterman and Merkel (2002a). (2) TOI = Trial Outcome Index; a p < 0.05; b p < 0.01.

Fatigue Subscale [mean (SD)] FACT-G FACT-Anemia TOI-Fatigue TOI-Anemia

FACT Scale or Aggregate

Table 2. Mean FACT scores by hemoglobin level at baseline (sample 1)

5.4b, 5.8b, 6.2b, 6.0b,

3> 3> 3> 3>

1, 1, 1, 1,

3.8a, 3 > 1 2 2 2 2

F (post-hoc)

3.1, 6.5, 3.8, 4.2,

11.5 20.8 16.6 19.6

3.0, 6.4

Mean Differences

0.22, 0.25, 0.19, 0.18,

0.82 0.80 0.83 0.83

0.29, 0.62

Effect Sizes

Adjacent Category

Hemoglobin, fatigue, and quality of life 377

37.0 (10.1) 84.2 (13.9) 78.9 (19.1)

78.1 (19.8) 70.1 (28.4)

(2) 10–11.99 g/dL (n = 31)

32.9 (14.2)

(1) <10 g/dL (n = 14)

Hemoglobin Level

83.5 (14.1) 82.7 (18.9)

40.3 (10.2)

(3) ≥12 g/dL (n = 86)

Notes: (1) Adapted from Cella, Eton, Lai, Peterman, and Merkel (2002a). (2) TOI = Trial Outcome Index; a p < 0.05; b ns.

Fatigue Subscale [mean (SD)] FACT-G TOI-Fatigue

FACT Scale or Aggregate

Table 3. Mean FACT scores by hemoglobin level at baseline (sample 2)

0.38, 0.30 0.41, −0.05 0.83, 0.19

6.1, −0.7 8.8, 3.8 0.9b 2.5b

Effect Sizes

4.1, 3.3

Mean Differences

Adjacent Category

3.4a, 3 > 1

F (post-hoc)

378 Z. Butt and D. Cella

−3.7 (10.5) n = 26 −3.8 (6.6) n = 11 −1.7 (11.4) n = 26 −6.4 (7.9) n = 11 −6.5 (19.1) n = 26 −9.8 (12.2) n = 11

0.7 (13.9) n = 54 1.7 (11.2) n = 45

−0.2 (20.8) n = 54 4.7 (14.9) n = 45

Decreasedb

−0.5 (11.6) n = 54 3.6 (9.2) n = 45

Increased or no changea

Change in Hb

3.0, p < 0.01

14.5

6.2

8.1

2.2, p < 0.05

1.3, ns

2.3

7.4

2.5, p < 0.05

0.7, ns

3.2

Mean Differences

1.2, ns

t

Note: (1) Adapted from Cella, Eton, Lai, Peterman and Merkel (2002a). a Follow-up Hb equals or exceeds baseline Hb. b Follow up Hb falls below baseline Hb.

Baseline to 12 mos

ΔTOI-Fatigue Baseline to 6 mos

Baseline to 12 mos

ΔFACT-G Baseline to 6 mos

Baseline to 12 mos

ΔFatigue Subscale Baseline to 6 mos

FACT Scale or Aggregate

Table 4. FACT change scores by change in hemoglobin (Hb) level from baseline to follow-up (sample 2)

0.71

0.31

0.55

0.16

0.68

0.29

Effect Sizes

Hemoglobin, fatigue, and quality of life 379

380

Z. Butt and D. Cella

[mean (±SD) time = 89 (±37) days.] Results from these comparisons can be seen in Table 5. Furthermore, Table 6 summarizes the mean differences and respective effect sizes from the anchor-based analyses. The summary table includes MCID data indexed against changes in hemoglobin, patient-rated PSR, and physician-rated PSR. In addition to anchor-based methods, Cella et al. (2002a) also derived distribution-based measures of clinical signiﬁcance. The ﬁrst of these standards were 1/2 standard deviation (SD) changes for the FACT scores of interest, chosen to approximate Cohen’s (1977) criteria for a medium effect size. The other selected standard was the standard error of measurement (SEM), which frequently corresponds to a minimally important difference (Wyrwich et al. 1999). Table 7 summarizes distribution-speciﬁc criteria for clinical signiﬁcance for several FACT scales. Using anchor-based criteria to determine MCID, observed differences were required to exceed an effect size threshold of 0.20, or “small” in Cohen’s (1977) designation. The distribution-based criteria selected for this study, the 1 /2 SD and the less sample-dependent SEM, offered another estimate of clinically signiﬁcant change. Recent work by Wyrwich et al. (1999a, b) has suggested that the SEM and anchor-based estimates of clinically important difference may approximate one another. Based on the clinical anchors (e.g. Hb, PSR) used to determine clinically important differences on the FACT, it would be difﬁcult to consider differences of less than 2.7 on the Fatigue Subscale, 3.1 on the FACT-G, 6.5 on the FACT-An, 4.8 on the TOI-Fatigue, and 5.8 on the TOI-Anemia to be clinically signiﬁcant. Rounding to the whole number above each MCID, recommended thresholds for each scale are: Fatigue Subscale = 3; FACT-G = 4; FACT-An = 7; TOI-F = 5; and TOI-An = 6. These conservative estimates may be used with more traditional clinical trial endpoints to determine treatment efﬁcacy. However, it is important to note that QOL MICD may be best represented by a range of values, dependent on the clinical anchors of interest. Patient-reported QOL may have meaningfully changed when the magnitude of change exceeds these FACT MCID criteria.

Anemia and the older cancer patient Anemia is common in the elderly, and the incidence increases with age. A recent analysis of U.S. national data using the National Health and Nutrition Examination Survey (1988–1994) suggests that the rate of anemia is 10–11% for individuals over the age of 65 and greater than 20% for individuals over the age of 85 (Smith 2000; Guralnik et al. 2004). Given that both anemia and cancer are common among the elderly, the question of age-associated QOL changes is important to consider. Although relatively common in older adults, anemia should not be considered a normal consequence of aging (Vu and Dugas 2004). The majority

6.6 (13.7) n = 1,011 4.9 (15.9) n = 1,003 13.2 (30.9) n = 979 10.6 (23.3) n = 989 12.3 (26.7) n = 988

1.7 (11.2) n = 303 −1.6 (14.9) n = 299 0.1 (26.4) n = 295 1.3 (19.6) n = 297 1.2 (22.3) n = 297

−4.3 (12.7) n = 64 −5.6 (14.7) n = 62 −11.6 (28.3) n = 61 −9.2 (20.6) n = 63 −10.8 (23.3) n = 63 40.9d, I > U > W

38.7d, I > U > W

37.2d, I > U > W

29.4d, I > U, W

32.8d, I > U > W

F (post-hoc)

Worsened (W)c

Improved (I)a

Unchanged (U)b

Adjacent Category

Change in Hb

Note: (1) Adapted from Cella, Eton, Lai, Peterman and Merkel (2002a). a Completion Hb exceeded baseline Hb by ≥1 g/dL. Mean change in Hb = 2.9. b Completion Hb within 1 g/dL of baseline Hb. Mean change in Hb = 0.1. c Completion Hb falls below baseline Hb by ≥1 g/dL. Mean change in Hb = −1.8. d p < 0.0001.

ΔTOI-Anemia

ΔTOI-Fatigue

ΔFACT-Anemia

ΔFACT-G

ΔFatigue Subscale

FACT Scale or Aggregate

11.1, 12.0

9.3, 10.5

11.7, 13.1

4.0, 6.5

4.9, 6.0

Mean Differences

Table 5. FACT change scores by change in hemoglobin (Hb) level from baseline to completion (sample 3)

0.44, 0.47

0.42, 0.48

0.39, 0.44

0.25, 0.41

0.39, 0.48

Effect Sizes

Hemoglobin, fatigue, and quality of life 381

382

Z. Butt and D. Cella

Table 6. Summary of mean differences and effect sizes for anchor-based analyses FACT Scale or Aggregate

Fatigue Subscale FACT-G FACT-Anemia TOI-Fatigue TOI-Anemia

Cross-sectional Analyses

Longitudinal Analyses

Mean Difference Range

Effect Size Range

Mean Difference Range

Effect Size Range

2.7–15.0 3.1–18.1 6.5–28.3 4.8–26.6 5.8–30.1

0.21–1.38 0.22–1.23 0.22–1.09 0.22–1.31 0.23–1.28

3.2–8.8 4.0–9.9 8.7–17.4 6.2–14.5 8.0–13.9

0.29–0.81 0.25–0.67 0.29–0.59 0.31–0.71 0.31–0.55

Notes: (1) Adapted from Cella, Eton, Lai, Peterman and Merkel (2002a) (2) Only those changes corresponding to a “small” effect size (ES ≥ 0.20) are included in this summary table. Effects of this magnitude approximate changes that are minimally clinically important.

Table 7. Summary of distribution-based criteria of clinical signiﬁcance FACT Scale or Aggregate

Fatigue Subscalea FACT-Ga FACT-Anemiab TOI-Fatiguea TOI-Anemiab

Criterion 1

/2 SD

SEM

5.8 7.9 14.8 10.5 12.9

2.6 5.1 6.6 4.2 5.3

Note: (1) Adapted from Cella, Eton, Lai, Peterman and Merkel (2002a). a Criterion values are averages across three separate samples. b Criterion values are averages across two separate samples.

of elderly individuals maintains normal RBC, Hb and hematocrit levels, and for most of those older adults with anemia, an underlying cause is found (Garry et al. 1983). Although percentages vary somewhat (Smith 2000; Guralnik et al. 2004), causes of anemia in community-dwelling elderly fall into the following categories: (1) nutrient deﬁciency [33%]; (2) chronic inﬂammation/renal disease [33%]; and (3) unexplained [33%]. Anemia of chronic

Hemoglobin, fatigue, and quality of life

383

disease is associated with cancer, congestive heart failure, rheumatoid arthritis, stroke, chronic obstructive lung disease and other conditions that affect the elderly (Thomas 2004). Recent studies suggest that anemic older adults are at greater risk for mortality than their nonanemic counterparts (Izaks et al. 1999; Wu et al. 2001). There appears to be a difference between the general population and cancer patients in the way that increasing age impacts fatigue and, more generally, anemia-related quality of life. We recently reanalyzed the data from a nationally representative sample of people who completed the FACIT-An as part of an internet survey panel (Cella et al. 2003b). We divided this general population sample into 3 age groups (i.e. 18–45, 46–65, 66+) and tested for age differences in FACT-An subscale scores. Speciﬁcally, we tested for crosssectional differences in PWB, SFWB, EWB, FWB, Fatigue subscale and Anemia subscale (Fatigue subscale + other anemia-related concerns) scores. The main effect of age was signiﬁcant (F(12, 2132) = 5.97, p < 0.0001). Followup univariate tests suggested that the 20-item Anemia subscale was the only one of the FACT-An subscales to demonstrate signiﬁcant age-related differences (F(2, 1070) = 5.74, p = 0.003). The oldest group obtained statistically signiﬁcantly worse QOL scores (M = 58.69) compared to those in the middle group (M = 61.95) and the youngest group (M = 63.10; Bonferroni-adjusted tests). Data from this representative sample of adults suggest that anemiarelated QOL (indexed by fatigue and other anemia-related symptoms) declines with age in the general population (see Fig. 1). Based on these ﬁndings and extant research on the anemia of aging, one might expect that age would be a risk factor for increased fatigue and/or decreased QOL in the older cancer patient. However, this does not seem to be the case. For example, in their comparison of cancer patients with control participants, Cella et al. (2002b) found that people older than 50 years in the general population reported more fatigue than their younger counterparts on the 13item FACT-Fatigue Subscale, but found no such age effect in the sample of anemic cancer patients. Similar analyses in other cancer clinical trials have failed to ﬁnd a substantial effect of age on fatigue- and/or anemia-related quality of life in cancer patients (e.g. Thome et al. 2004; for a review, see Aapro et al. 2002). Older cancer patients do not seem to experience fatigue more acutely than their younger counterparts. It is also possible that younger patients feel the effects of treatment more severely. Alternately, as suggested by Curt’s survey (Curt et al. 2000), older cancer patients may also underreport their experience of fatigue. However plausible these explanations may be, they do not adequately address what it is about the cancer experience that diminishes the impact of age on anemia-related quality of life seen in the general population. There remains a need to determine the most reliable and valid way to assess the general and anemia-related QOL in older adults, especially the oldest old. Doing so will help us determine the medical and psychological correlates of QOL in the aging cancer patient.

384

Z. Butt and D. Cella

FACIT-An Anemia subscale scores

64.00

63.00

62.00

61.00

60.00

59.00

58.00 18-45

46-65

66 +

Age Fig. 1. Anemia-related quality of life declines with age in the general US population

Mechanisms and treatments How to assess and treat anemia in the cancer patient is complicated by the fact that the anemia may be due to the disease, its treatment, or diseaseunrelated factors. For example, among patients with malignancies, anemia may be directly related to the tumor due to blood loss or hemolysis (Bron et al. 2001). Metastases to the bone marrow may impair production of hematopoietic growth factors or induce production of cytokines that dampen erythropoiesis (Mercadante et al. 2000). Even in the absence of bone metastases, the tumor itself may induce anemia via decreased erythropoietin (Miller et al. 1990). Furthermore, the myelosuppressive properties of chemotherapy and radiotherapy may also result in anemia (Groopman and Itri 1999). However, anemia in the cancer patient may also be caused by comorbid nutritional deﬁciencies or other chronic diseases. For a minority of patients in the general population, there is no clear etiology for their anemia (Guralnik et al. 2004). Regardless of etiology, there is a growing body of research that demonstrates the association between anemia and QOL. In part, anemia may impact patients’ QOL by reducing exercise tolerance, ability to work, social inter-

Hemoglobin, fatigue, and quality of life

385

action, mood, and ability to participate in leisure activities (Cella 1997; Ludwig and Strasser 2001). If low hemoglobin values result in diminished QOL, it follows that patients’ QOL will improve by increasing hemoglobin concentrations. And for a number of patients, this is exactly what happens. The most appropriate treatment for anemia will be determined by its etiology. For example, for patients with anemia caused by vitamin B12 (folate) deﬁciency, correction of the nutritional deﬁciency would naturally be considered as the ﬁrst treatment for the anemia. In other anemic patients, rhEPO therapy may be indicated. Erythropoietin therapy increases hemoglobin concentration and QOL in cancer patients with anemia, as well as other patients with chronic illnesses (Thomas 2004). In their recent meta-analysis of 23 cancer trials, Jones et al. (2004) demonstrated that epoetin alfa signiﬁcantly improved CLAS scores by 20–25%, FACT-Fatigue scores by 17% and FACT-Anemia scores by 12%. Cella et al. (2004) examined data from ﬁve randomized clinical trials and determined that patients whose Hb improved at least 2 g/dL reported commensurate increases in FACT-Fatigue subscale scores, even after controlling for potential confounding variables. Clinically meaningful increases in fatigue scores (i.e. ≥3 points increase) were also signiﬁcantly associated with moderate increases in physical, emotional, and functional well-being (as measured by the FACT), even after controlling for potential confoundings. We are beginning to understand the biological mode of action behind these improved outcomes, and there may be many potential mechanisms of change (Blackwell et al. 2004). While some patients beneﬁt considerably from epoetin therapy, not all patients respond to the treatment (Cella et al. 2003a). Glaspy and colleagues recently reanalyzed data from two large, communitybased trials of epoetin alfa and found that the best predictors of meaningful response to treatment (deﬁned as ≥2 g/dL increase in Hb) were lack of transfusions immediately before or during study and an increase of ≥1 g/dL Hb within the ﬁrst 4 weeks of the study. The chapter by Yip and Harper in this volume addresses the issue of rhEPO treatment and QOL in much greater detail.

Summary Anemia among patients with cancer is common (Mercadante et al. 2000; Demetri 2001; Glaspy 2001; Pujade-Lauraine and 2004). The reduced hemoglobin concentration in anemia leads to a reduction in blood oxygenation which in turn results in weakness and fatigue, among other symptoms. While anemia often results in fatigue, cancer-related fatigue is not always due to low hemoglobin (Wu et al. 2001; Jacobsen 2004; Lipman and Lawrence 2004; Wagner and Cella 2004). Regardless of etiology, research has demonstrated that anemia and fatigue adversely impact QOL in patients with cancer

386

Z. Butt and D. Cella

(Yellen et al. 1997; Yount et al. 2002; Cella et al. 2003a, b; Cella et al. 2004). Although increasing age seems to be a risk factor for decreased anemiarelated QOL in the general population, this association does not seem to hold in cancer patients (Aapro et al. 2002). Reliable and valid measures of QOL are necessary to best assess changes in QOL over the course of illness and the changes associated with anemia treatment. Within the FACIT system of QOL measurement, subscales of the FACIT-Anemia have been shown to differentiate cancer patients from the general population (Cella et al. 2002b; Cella et al. 2003b). The FACITAnemia has also been validated against cross-sectional and longitudinal differences in important clinical markers. The establishment of these meaningful clinically important differences will likely prove useful in future clinical trials (Cella et al. 2002a).

References 1. Aapro MS, Cella D, Zagari M (2002) Age, anemia, and fatigue. Semin Oncol 29: 55–59 2. Beghe C, Wilson A, Ershler WB (2004) Prevalence and outcomes of anemia in geriatrics: a systematic review of the literature. Am J Med 116 [Suppl 7A]: 3S–10S 3. Blackwell K, Gascon P, Sigounas G, Jolliffe L (2004) rHuEPO and improved treatment outcomes: potential modes of action. Oncologist 9 [Suppl 5]: 41–47 4. Bron D, Meuleman N, Mascaux C (2001) Biological basis of anemia. Semin Oncol 28 [Suppl 8]: 1–6 5. Cella D (1997) The Functional Assessment of Cancer Threapy-Anemia (FACTAn) Scale: A new tool for the assessment of outcomes in cancer anemia and fatigue. Semin Hematol 34 [Suppl 2]: 13–19 6. Cella D, Dobrez D, Glaspy J (2003a) Control of cancer-related anemia with erythropoietic agents: a review of evidence for improved quality of life and clinical outcomes. Ann Oncol 14: 511–519 7. Cella D, Eton DT, Lai JS, Peterman AH, Merkel DE (2002a) Combining anchor and distribution-based methods to derive minimal clinically important differences on the Functional Assessment of Cancer Therapy (FACT) anemia and fatigue scales. J Pain Symptom Manage 24: 547–561 8. Cella D, Kallich J, McDermott A, Xu X (2004) The longitudinal relationship of hemoglobin, fatigue and quality of life in anemic cancer patients: results from ﬁve randomized clinical trials. Ann Oncol 15: 979–986 9. Cella D, Lai JS, Chang CH, Peterman A, Slavin M (2002b) Fatigue in cancer patients compared with fatigue in the general United States population. Cancer 94: 528–538 10. Cella D, Peterman A, Passik S, Jacobsen P, Breitbart W (1998) Progress toward guidelines for the management of fatigue. Oncology (Huntingt) 12: 369–377 11. Cella D, Zagari MJ, Vandoros C, Gagnon DD, Hurtz HJ, Nortier JW (2003b) Epoetin alfa treatment results in clinically signiﬁcant improvements in quality of life in anemic cancer patients when referenced to the general population. J Clin Oncol 21: 366–373

Hemoglobin, fatigue, and quality of life

387

12. Cohen J (1977) Statistical power analysis for the behavioral sciences. Academic Press, New York 13. Curt GA, Breitbart W, Cella D, Groopman JE, Horning SJ, Itri LM, Johnson DH, Miaskowski C, Scherr SL, Portonoy RK, Vogelzang JF (2000) Impact of cancerrelated fatigue on the lives of patients: new ﬁndings from the Fatigue Coalition. Oncologist 5: 353–360 14. de Boer AG, van Lanschot JJ, Stalmeier PF, van Sandick JW, Hulscher JB, De Haes JC, Sprangers MA (2004) Is a single-item visual analogue scale as valid, reliable and responsive as multi-item scales in measuring quality of life? Qual Life Res 13: 311–320 15. Demetri GD (2001) Anaemia and its functional consequences in cancer patients: current challenges in management and prospects for improving therapy. Br J Cancer 84 [Suppl 1]: 31–37 16. Garry PJ, Goodwin JS, Hunt WC (1983) Iron status and anemia in the elderly: new ﬁndings and a review of previous studies. J Am Geriatr Soc 31: 389–399 17. Glaspy J (2001) Anemia and fatigue in cancer patients. Cancer 92: 1719–1724 18. Gough IR, Furnival CM, Schilder L, Grove W (1983) Assessment of the quality of life of patients with advanced cancer. Eur J Cancer Clin Oncol 19: 1161–1165 19. Groopman JE, Itri LM (1999) Chemotherapy-induced anemia in adults: incidence and treatment. J Natl Cancer Inst 91: 1616–1634 20. Guralnik JM, Eisenstaedt RS, Ferrucci L, Klein HG, Woodman RC (2004) The prevalence of anemia in persons age 65 and older in the United States: evidence for a high rate of unexplained anemia. Blood 104: 2263–2268 21. Guyatt GH, Osoba D, Wu AW, Wyrwich KW, Norman GR (2002) Methods to explain the clinical signiﬁcance of health status measures. Mayo Clin Proc 77: 371–383 22. Hann DM, Denniston MM, Baker F (2000) Measurement of fatigue in cancer patients: further validation of the Fatigue Symptom Inventory. Qual Life Res 9: 847–854 23. Izaks GJ, Westendorp RG, Knook DL (1999) The deﬁnition of anemia in older persons. JAMA 281: 1714–1717 24. Jacobsen PB (2004) Assessment of fatigue in cancer patients. J Natl Cancer Inst Monogr 32: 93–97 25. Jacobsen PB, Garland LL, Booth-Jones M, Donovan KA, Thors CL, Winters E, Grendys E (2004) Relationship of hemoglobin levels to fatigue and cognitive functioning among cancer patients receiving chemotherapy. J Pain Symptom Manage 28: 7–18 26. Jaeschke R, Singer J, Guyatt GH (1989) Measurement of health status: Ascertaining the minimal clinically important difference. Control Clin Trials 10: 407–415 27. Jones M, Schenkel B, Just J, Fallowﬁeld L (2004) Epoetin alfa improves quality of life in patients with cancer: results of metaanalysis. Cancer 101: 1720–1732 28. Lipman AJ, Lawrence DP (2004) The management of fatigue in cancer patients. Oncology (Huntingt) 18: 1527–1534 29. Ludwig H, Strasser K (2001) Symptomatology of anemia. Semin Oncol 28 [Suppl 8]: 7–14 30. Lydick E, Epstein RS (1993) Interpretation of quality of life changes. Qual Life Res 2: 221–226

388

Z. Butt and D. Cella

31. McCormack HM, Horne DJ, Sheather S (1988) Clinical applications of visual analogue scales: a critical review. Psychol Med 18: 1007–1019 32. Mercadante S, Gebbia V, Marrazzo A, Filosto S (2000) Anaemia in cancer: pathophysiology and treatment. Cancer Treat Rev 26: 303–311 33. Miller CB, Jones RJ, Piantadosi S, Abeloff MD, Spivak JL (1990) Decreased erythropoietin response in patients with the anemia of cancer. N Engl J Med 322: 1689–1692 34. Mock V, Atkinson A, Barsevick A, Cella D, Cimprich B, Cleeland C, Donnelly J, Eisenberger MA, Escalante C, Hinds P, Jacobsen PB, Kaldor P, Knight SJ, Peterman A, Piper BF, Rugo H, Sabbatini P, Stahl C (2000) NCCN Practice Guidelines for Cancer-Related Fatigue. Oncology (Huntingt) 14: 151–161 35. Movsas B (2003) Quality of life in oncology trials: a clinical guide. Semin Radiat Oncol 13: 235–247 36. Okuyama T, Akechi T, Kugaya A, Okamura H, Shima Y, Maruguchi M, Hosaka T, Uchitomi Y (2000) Development and validation of the cancer fatigue scale: a brief, three-dimensional, self-rating scale for assessment of fatigue in cancer patients. J Pain Symptom Manage 19: 5–14 37. Piper BF, Dibble SL, Dodd MJ, Weiss MC, Slaughter RE, Paul SM (1998) The revised Piper Fatigue Scale: psychometric evaluation in women with breast cancer. Oncol Nurs Forum 25: 677–684 38. Pujade-Lauraine E, Gascon P (2004) The burden of anaemia in patients with cancer. Oncology 67 [Suppl 1]: 1–4 39. Sadler IJ, Jacobsen PB, Booth-Jones M, Belanger H, Weitzner MA, Fields KK (2002) Preliminary evaluation of a clinical syndrome approach to assessing cancer-related fatigue. J Pain Symptom Manage 23: 406–416 40. Smets EM, Garssen B, Bonke B, De Haes JC (1995) The Multidimensional Fatigue Inventory (MFI): psychometric qualities of an instrument to assess fatigue. J Psychosom Res 39: 315–325 41. Smith DL (2000) Anemia in the elderly. Am Fam Physician 62: 1565–1572 42. Thomas DR (2004) Anemia and quality of life: unrecognized and undertreated. J Gerontol A Biol Sci Med Sci 59: 238–241 43. Thome B, Dykes AK, Hallberg IR (2004) Quality of life in old people with and without cancer. Qual Life Res 13: 1067–1080 44. Vogelzang NJ, Breitbart W, Cella D, Curt GA, Groopman JE, Horning SJ, Itri LM, Johnson DH, Scherr SL, Portenoy RK (1997) Patient, caregiver, and oncologist perceptions of cancer-related fatigue: results of a tripart assessment survey. The Fatigue Coalition. Semin Hematol 34 [Suppl 2]: 4–12 45. Vu TTM, Dugas M (2004) The three sides of anemia in the elderly. The Canadian Journal of CME: July, 91–94 (in press) 46. Wagner LI, Cella D (2004) Fatigue and cancer: causes, prevalence and treatment approaches. Br J Cancer 91: 822–828 47. Wagner LI, Cella D, Cashy J, Peterman AH, Paice JA, Marks BA, Shevrin DH, Muir JC, Straus J, von Roenn J (2003) Single item screening to detect clinically signiﬁcant fatigue, pain, distress, and anorexia in ambulatory oncology practice. Proc Am Soc Clin Oncol 22: 527 (Abstr 2122) 48. Ware JE, Kosinski M, Keller SD (1995) SF-12: How to score the SF-12 physical and mental summary scales. The Health Institute, New England Medical Center, Boston

Hemoglobin, fatigue, and quality of life

389

49. Wu HS, McSweeney M (2001) Measurement of fatigue in people with cancer. Onc Nurs Forum 28: 1371–1384 50. Wu WC, Rathore SS, Wang Y, Radford MJ, Krumholz HM (2001) Blood transfusion in elderly patients with acute myocardial infarction. N Engl J Med 345: 1230–1236 51. Wyrwich KW, Nienaber NA, Tierney WM, Wolinsky FD (1999a) Linking clinical relevance and statistical signiﬁcance in evaluating intra-individual changes in health-related quality of life. Med Care 37: 469–478 52. Wyrwich KW, Tierney WM, Wolinsky FD (1999b) Further evidence supporting an SEM-based criterion for identifying meaningful intra-individual changes in health-related quality of life. J Clin Epidemiol 52: 861–873 53. Yellen SB, Cella DF, Webster K, Blendowski C, Kaplan E (1997) Measuring fatigue and other anemia-related symptoms with the Functional Assessment of Cancer Therapy (FACT) measurement system. J Pain Symptom Manage 13: 63–74 54. Yount S, Lai JS, Cella D (2002) Methods and progress in assessing the quality of life effects of supportive care with erythropoietin therapy. Curr Opin Hematol 9: 234–240 Correspondence: Zeeshan Butt, Center on Outcomes, Research and Education, Evanston Northwestern Healthcare, 1000 University Place, Suite 100, Evanston, Illinois 60201, USA, E-mail: [email protected]

Appendix A FACT-An (Version 4) Below is a list of statements that other people with your illness have said are important. By circling one (1) number per line, please indicate how true each statement has been for you during the past 7 days. PHYSICAL WELL-BEING

Not A little Some- Quite Very at all bit what a bit much

GP1

I have a lack of energy ....................

0

1

2

3

4

GP2

I have nausea ....................................

0

1

2

3

4

GP3

Because of my physical condition, I have trouble meeting the needs of my family..................................................

0

1

2

3

4

GP4

I have pain.........................................

0

1

2

3

4

GP5

I am bothered by side effects of treatment.......................................

0

1

2

3

4

GP6

I feel ill...............................................

0

1

2

3

4

GP7

I am forced to spend time in bed ......................................................

0

1

2

3

4

390

Z. Butt and D. Cella SOCIAL/FAMILY WELL-BEING

GS1 GS2 GS3 GS4 GS5 GS6 Q1

GS7

Not A little Some- Quite Very at all bit what a bit much I feel close to my friends ................. 0 1 2 3 4 I get emotional support from my family ........................................... 0 1 2 3 4 I get support from my friends .......... 0 1 2 3 4 My family has accepted my illness ...... 0 1 2 3 4 I am satisﬁed with family communication about my illness ....... 0 1 2 3 4 I feel close to my partner (or the person who is my main support)...... 0 1 2 3 4 Regardless of your current level of sexual activity, please answer the following question. If you prefer not to answer it, please check this box and go to the next section. I am satisﬁed with my sex life ........ 0 1 2 3 4

By circling one (1) number per line, please indicate how true each statement has been for you during the past 7 days. EMOTIONAL WELL-BEING GE1 GE2 GE3 GE4 GE5 GE6

Not A little Some- Quite Very at all bit what a bit much I feel sad ............................................ 0 1 2 3 4 I am satisﬁed with how I am coping with my illness ..................... 0 1 2 3 4 I am losing hope in the ﬁght against my illness ............................. 0 1 2 3 4 I feel nervous .................................... 0 1 2 3 4 I worry about dying.......................... 0 1 2 3 4 I worry that my condition will get worse..................................... 0 1 2 3 4 FUNCTIONAL WELL-BEING

GF1 GF2 GF3 GF4 GF5 GF6 GF7

I am able to work (include work at home) .................................. My work (include work at home) is fulﬁlling .............................. I am able to enjoy life...................... I have accepted my illness............... I am sleeping well............................. I am enjoying the things I usually do for fun .............................. I am content with the quality of my life right now..........................

Not A little Some- Quite Very at all bit what a bit much 0

1

2

3

4

0 0 0 0

1 1 1 1

2 2 2 2

3 3 3 3

4 4 4 4

0

1

2

3

4

0

1

2

3

4

Hemoglobin, fatigue, and quality of life

391

FACT-An (Version 4) By circling one (1) number per line, please indicate how true each statement has been for you during the past 7 days. ADDITIONAL CONCERNS

Not A little Some- Quite Very at all bit what a bit much

HI7

*I feel fatigued ..................................

0

1

2

3

4

HI12

*I feel weak all over.........................

0

1

2

3

4

An1

*I feel listless (“washed out”).........

0

1

2

3

4

An2

*I feel tired ........................................

0

1

2

3

4

An3

*I have trouble starting things because I am tired...........................

0

1

2

3

4

An4

*I have trouble ﬁnishing things because I am tired...........................

0

1

2

3

4

An5

*I have energy...................................

0

1

2

3

4

An6

I have trouble walking ...................

0

1

2

3

4

An7

*I am able to do my usual activities ...........................................

0

1

2

3

4

An8

*I need to sleep during the day ....................................................

0

1

2

3

4

An9

I feel lightheaded (dizzy)...............

0

1

2

3

4

An10

I get headaches ...............................

0

1

2

3

4

B1

I have been short of breath...........

0

1

2

3

4

An11

I have pain in my chest..................

0

1

2

3

4

An12

*I am too tired to eat.......................

0

1

2

3

4

BL4

I am interested in sex.....................

0

1

2

3

4

An13

I am motivated to do my usual activities ...........................................

0

1

2

3

4

An14

*I need help doing my usual activities ...........................................

0

1

2

3

4

An15

*I am frustrated by being too tired to do the things I want to do..................................................

0

1

2

3

4

*I have to limit my social activity because I am tired .............

0

1

2

3

4

An16

* These items comprise the 13-item fatigue scale.

Chapter 15

When to use red blood cell transfusions in cancer patients with solid tumours? J. K. Jacob and P. J. Barrett-Lee Velindre Cancer Center, Velindre NHS Trust, Whitchurch, Cardiff, Wales, UK

Introduction A major population of cancer patients experience anaemia during the course of their disease process and by treatments such as chemotherapy, radiotherapy and also whilst on supportive therapies (Groopman and Itri 1999). Red Blood Transfusion (RBT) is the most commonly used form of correction of anaemia worldwide and in the United Kingdom (U.K) alone over 3 million blood product transfusions are carried out each year (McClelland et al. 2007). RBT can be broadly classiﬁed as being homologous transfusions (transfusions using the stored blood of donors) and autologous transfusions (transfusions using patient’s own stored blood in a predetermined manner). In the U.K the service relies on volunteers for blood components, who give their services free. RBT is used in the correction of symptomatic anaemia of chronic disease including cancer, in patients with haematological disorders, for prevention in asymptomatic patients and also during surgical procedures or acute bleeding. RBT is associated with adverse events such as a risk of infection and disease transmission. Peri-operative blood transfusions have also been associated with recurrence of colo-rectal cancer (Amato et al. 2006). The decision to give RBT to cancer patients is based in practice on a combination of clinical symptoms and haemoglobin (Hb) values, and there is variation across different centres. This review will examine the role of RBT in cancer patients with solid tumours. It should be noted that there are virtually no randomised trials of the use of RBT in this setting, but retrospective evidence will be presented and reviewed.

Why use transfusions in cancer patients? Asymptomatic patients – prevention of anaemia in solid tumours The role of RBT in anaemia prevention and maintenance of Hb levels within the normal range is based on the premise that there are survival beneﬁts in

394

J. K. Jacob and P. J. Barrett-Lee

certain site speciﬁc cancers. Some studies have shown a correlation between anaemia correction and local control of the primary tumour. In patients receiving radical radiotherapy for head and neck cancer it has been shown that low Hb levels are associated with a statistically signiﬁcant reduction in survival and an increase in loco-regional failure. (Dally et al. 2003; Lee et al. 1998). The effects of pre-treatment anaemia on tumour recurrence and survival in patients treated with primary radiotherapy for early squamous cell carcinoma of the larynx was also investigated in a retrospective analysis of 117 patients with previously untreated T1N0M0 and T2N0M0 stage tumours. The two- and ﬁve-year loco-regional control estimates for anaemic patients were 58 per cent and 53 per cent respectively, whilst patients with normal haemoglobin levels (patients were considered anaemic if their pre-treatment Hb levels were below 13 g/dl in males and 11.5 g/dl in females) had two and ﬁveyear local-regional control rates of 90 per cent and 81 per cent respectively (P = 0.002). Multivariate Cox regression analysis showed pre-treatment Hb level signiﬁcantly inﬂuenced recurrence-free survival (P = 0.0094). Patients with a low Hb level prior to radiation therapy suffered higher levels of localregional failure (Grant et al. 1999). In a further study, haemoglobin levels in patients receiving radiotherapy for carcinoma of the cervix were highly predictive of treatment outcome. Blood transfusion appeared to overcome the negative prognostic effects of low initial Hb levels and of the average weekly value of Hb in this retrospective study of 605 patients (Grogan et al. 1999). The median follow-up of patients in this study was 41 months (range, 0–92 months). The presenting Hb level, Average Weekly Nadir Haemoglobin (AWNH) during radiotherapy and blood transfusion requirements were correlated with local control, disease free survival, and overall survival on univariate analysis. The AWNH remained signiﬁcant for these parameters on multivariate analysis, whereas Hb at presentation and blood transfusion requirements did not. The 5-year survival was 74% for patients with an average weekly haemoglobin nadir of 12 g/dL, 52% for patients with AWNH levels 11–11.9 g/dL inclusive, and 45% for patients with AWNH levels 11 g/dL (P < 0.0001). At each Hb level, patients who were transfused and maintained at a speciﬁc Hb level had a survival rate that was not signiﬁcantly different from patients whose Hb was at that level spontaneously. There was a signiﬁcant reduction in both pelvic and distant recurrence (P < 0.0001 and P < 0.0006, respectively) in patients whose AWNH level during radiotherapy was ≥12 g/dL compared with <12g/dL. A reduction in the rate of distant recurrence was observed in patients with and without pelvic recurrence whose Hb levels were maintained at ≥12 g/dL. The survival beneﬁt associated with maintaining hemoglobin at ≥12 g/dL was due to both improved pelvic control and a reduction in distant metastatic disease, which was independent of the effects on pelvic

When to use red blood cell transfusions in cancer patients

395

control rates. The magnitude of the difference in survival rates between patients with low hemoglobin levels during radiotherapy and those with high levels was approximately 24% at 5 years. The observed beneﬁt exceeded that demonstrated in any interventional studies comparing new therapies with standard radiotherapy. The only therapeutic intervention that has shown a beneﬁt when used in conjunction with radiotherapy is concurrent chemotherapy. The effects of anaemia were also investigated in stage III Non Small Cell Lung Cancer (NSCLC) patients undergoing combined modality therapy, in three sequential trials of concurrent weekly paclitaxel, carboplatin and radiation therapy. The Hb levels of 115 patients were studied before treatment and weekly during radiotherapy; on univariate log-rank analysis the presenting Hb, average Hb and minimum Hb during treatment were not statistically signiﬁcantly predictive of survival. However in a Coxregression model, declining Hb during chemo-radiation had a statistically signiﬁcant deleterious impact on survival. Overall the analyses revealed that a decline in Hb during chemo-radiation for stage III NSCLC had a statistically signiﬁcant correlation with poor overall survival (MacRae et al. 2002). There is a need for a general consensus in the role of maintaining normal Hb levels whilst treating asymptomatic cancer patients with solid tumours but the lack of current guidelines reﬂects the lack of randomised controlled trials in the ﬁeld. However these studies in NSCLC, head and neck cancer and cervical cancer strongly suggest that maintenance of Hb above 12 g/dL levels is important for local control and survival.

Symptomatic patients receiving chemotherapy for solid tumours Most cancer physicians in Europe appear to adhere to the concept of a transfusion “trigger” based on a minimal Hb level or haematocrit value at which RBT is initiated. A large-scale audit (Barrett-Lee et al. 2000) of patients with a variety of solid tumours receiving chemotherapy at 28 specialist centres throughout the UK reported on 2719 patients receiving 3206 courses of cytotoxic chemotherapy for tumours of the breast (878), ovary (856), lung (772) or testis (213) with a mean age was 55 years (range 16–87). Overall, 33% of patients required at least one blood transfusion during a course of chemotherapy but the proportion varied from 19% for breast cancer to 43% for lung cancer. Sixteen per cent of patients required more than one transfusion (7% for breast, 22% for lung). The mean proportion of patients with Hb < 11 g /dl rose over the course of chemotherapy from 17% before the ﬁrst cycle, to 38% by the sixth, despite transfusion in 33% of patients (Fig. 1).

396

J. K. Jacob and P. J. Barrett-Lee 40

12.6

Hb<11

35

12.4

Mean Hb

12.2 12

25

11.8 20 11.6 15

Mean Hb

% Hb<11

30

11.4

10

11.2

5

11

0

10.8 1

2

3

4

5

6

Cycle

Fig. 1. Graph showing the mean haemoglobin concentrations and the proportion of anaemic patients by each chemotherapy cycle in patients receiving chemotherapy (all sites.). From ref. [Barrett-Lee et al. Large scale UK audit of blood transfusion requirements and anaemia in patients receiving cytotoxic chemotherapy. Br J Cancer 82(1): 97–7]

The transfusion “trigger” (i.e. the mean Hb concentration at which a transfusion was given) was 10.7 g /dl for all patients (range 7.2–16.1) in the ﬁrst cycle but this fell to 9.9 g dl–1 (range 6.3–14.0) by the sixth cycle. The mean transfusion triggers in the lung cancer group were signiﬁcantly higher during both the ﬁrst and sixth chemotherapy cycles than those for other cancers (11 vs. 10.2 g dl–1 (P = 0.008) in cycle 1 and 10.4 vs. 9.7 (P < 0.001) in cycle 6). The ﬁrst cycle transfusion trigger for patients with testicular cancer was signiﬁcantly lower (9.6 g dl–1) than that for the other groups (P = 0.009). In the lung cancer group no difference in the mean transfusion triggers was seen in relation to the sex of the patient (10.4 for women vs. 10.5 for men, P = 0.3). One interpretation of these differences is that lung cancer patients had more symptomatology at higher Hb levels and this may explain the observation that the trigger for transfusion among patients with lung cancer was higher than in other groups (i.e. the threshold for treating anaemia was lower . In comparison, testicular cancer patients had the lowest trigger Hb concentrations which may indicate that clinicians are more reluctant to give transfusions to younger patients with a good chance of recovery than to patients with a shorter life expectancy, unless their symptoms are severe. The European Cancer Anaemia Survey (ECAS) was a prospective, epidemiological, observational survey (Ludwig et al. 2004) conducted in 24

When to use red blood cell transfusions in cancer patients

397

European countries to document the prevalence, incidence, evolution, severity and management of anaemia in a large, representative population of European cancer patients. Results were reported for data from 748 cancer centres in 24 European countries with over 1000 physicians participating. For 14,520 patients with haemoglobin levels available at enrolment, 39.3% were anaemic (the deﬁnition of anaemia for ECAS was haemoglobin <12.0 g/dL based on toxicity grading criteria from the National Cancer Institute (NCI) and the European Organisation for Research and Treatment of Cancer (EORTC)). Most patients (29.3%) had mild anaemia, with Hb levels between 10.0 and 11.9 g/dL; moderate anaemia was recorded for 8.7% (Hb 8.0–9.9 g/dL), and severe anaemia for 1.3% of patients (Hb <8.0 g/dL). Amongst the patients who were ever anaemic (n = 9118), 61.1% did not receive any treatment for their anaemia. Amongst the patients who were not treated, 47.2% had Hb levels between 10.0 and 11.9 g/dL; 12.9% who were not treated had Hb levels between 8.0 and 9.9 g/dL, and 0.9% had Hb levels < 8.0 g/dL. Amongst the patients who received RBT, in over half the treatment was not initiated until Hb was <9.0 g/dL (Table 1). The concept of transfusion trigger is currently based on local practices and guidelines orientated towards correction of chronic diseases other than cancer. Once again there is a lack of consistency across centres and cancer sites reflecting a lack of agreed guidelines nationally or locally.

Table 1. Haemoglobin levels in patients receiving blood transfusions versus no treatment for cancer patients receiving chemotherapy only (ECAS data) Haemoglobin level (g/dL)

<9.0 9.0–9.9 10.0–10.9 11.0–11.9 ≥12.0 Total

Transfusion

No treatment

n

%

n

%

562 316 130 32 27 1067

52.7 29.6 12.2 3 2.5 100.0

275 682 1330 1590 2000 5877

4.7 11.6 22.6 27.1 34.0 100.0

Adapted from ref. (Ludwig H, Van Belle S, Barrett-Lee P, Birgegard G, Bokemeyer C, Gascon P, Kosmidis P, Krzakowski M, Nortier J, Olmi P, Schneider M, Schrijvers D (2004) The European Cancer Anaemia Survey (ECAS): a large, multinational, prospective survey deﬁning the prevalence, incidence, and treatment of anaemia in cancer patients. Eur J Cancer 40(15): 2293–306).

398

J. K. Jacob and P. J. Barrett-Lee

Symptomatic patients with metastatic solid tumour The place of blood transfusion in anaemic terminally ill cancer patients is even less well established. There are no blood transfusion guidelines available and transfusion decisions are greatly inﬂuenced by the personal views of the medical team. A mail survey (Leibovitz et al. 2003) of 500 physicians (from several specialties) and nurses revealed that a general consensus was that blood transfusions, as a rule, should not be withheld from terminal cancer patients. There emerged a short list of agreed suggestions for blood transfusion–namely, Hb level < or =7 g/dL, active bleeding (acute and/or occult), functional deterioration of the patient, presence of anaemia resulting from chemotherapy, anginal symptoms, dyspnoea, and worsening congestive heart failure. The agreed suggestions for transfusions in terminally ill cancer patients may serve as a reasonable physician standard for this complex clinical, medico-legal, and emotional issue.

Oncology patients at the end of life Terminally ill cancer patients in many cases end up having transfusions based on a single measured Hb value. The role of blood transfusion at this stage has to be a decision based on the patho-physiology of the anaemia i.e. anaemia of chronic disease versus active bleeding, the quality of life of the patient and the moral and ethical issues at the period of time (Booth et al. 2003). In this setting the beneﬁts of RBT were studied in ninety-seven anaemic patients from eight palliative care centres using patient directed visual analogue scales performed before and on two occasions after RBT (2 days post transfusion and at 2 weeks post transfusion). The impact of RBT was assessed with respect to dyspnoea, weakness and overall sense of well-being. The mean VAS score of strength (the authors terminology) showed a signiﬁcant rise from 3.0 at pre RBT to 5.4 at two days post RBT (P > 0.001) and 4.9 at 14 days post RBT (P > 0.001). The mean VAS score for breathing (the authors terminology) a signiﬁcant initial improvement from 5.7 pre RBT to 7.3 at two days post RBT (P > 0.001) and was 6.7 at 14 days post RBT (P = 0.058). Mean scores for overall sense of well being also showed improvement from 4.2 pre RBT (P > 0.001) to 5.9 at two days (P > 0.001) and 5.8 at 14 days post RBT (P = 0.003). In these patients the results indicated that a signiﬁcant proportion showed improvement in all three parameters. Those whose main indication for transfusion was weakness showed a particular beneﬁt. The study concluded that transfusion offers symptom relief and improvement in well-being in patients with advanced malignant disease. It should be considered as a worthwhile option in palliative treatment of weakness, dyspnoea and impaired overall

When to use red blood cell transfusions in cancer patients

399

sense of well being, when associated with anaemia (Gleeson and Spencer 1995). In a further study (Monti et al. 1996) 246 terminally ill cancer patients admitted to a palliative care unit were studied when they were transfused in the presence of low Hb levels (< or =8 g/dL) and/or severe fatigue or dyspnoea. Pre-transfusion performance status, cognitive function, dyspnoea, and fatigue at rest (evaluated by a four-point scale), complete blood count, serum albumin, and C-reactive protein were determined. The day after transfusion, subjective well-being was recorded as “yes/no” improvement in comparison with the pre-transfusion day. Improved subjective well-being after blood transfusion was reported in 51.4%, without signiﬁcant relationship to pretransfusion Hb levels or performance status. The inﬂuence of blood transfusion on subjective well-being was not related to the severity of dyspnoea or fatigue. Twenty-one patients (60%), including seven with subjective improvement, died during the same hospitalisation, a median of 49 days after transfusion. Pre-transfusion Hb level did not differ signiﬁcantly in patients who beneﬁted and did not beneﬁt from transfusion, whereas time before death was signiﬁcantly (P < 0.001) shorter in patients who did not beneﬁt. In the discharged patients (40%), the median interval between transfusion and discharge was 13 days and the frequency of subjective improvement in well-being was 78.6%. The inﬂuence of blood transfusion on subjective wellbeing was not related to the severity of dyspnoea or fatigue. Pre-transfusion Hb level did not differ signiﬁcantly in patients who beneﬁted and did not beneﬁt from transfusion. Clearly these studies suggest that there are benefits for RBT in terminally ill patients but the defined patient population who might benefit is not clear. Therefore, more prospective studies are required in this setting.

When should transfusions be used in cancer patients? There are no randomised control studies or any structured guidelines specifically relating to RBT in cancer patients. Most guidelines refer to the use of RBT based on Hb values and the use in chronic disease states. ASCO/ASH guidelines (Douglas Rizzo et al. 2002) recommend the use of epoetin as a treatment option for patients with chemotherapy-associated anaemia and a Hb concentration that has declined to a level 10 g/dL. RBT is also a treatment option depending on the severity of anaemia or clinical circumstances. Clinical Practice Guidelines on the Use of Blood Components published by the National Health and Medical Research Council (Australia) (http://www.nhmrc.gov.au/publications/synopses/_files/cp78.pdf) has recommended:

400

J. K. Jacob and P. J. Barrett-Lee

1. Use of red blood cells is likely to be inappropriate when Hb > 10g/dL unless there are speciﬁc indications. If red blood cells are given at this haemoglobin level, reasons should be well documented. 2. Use of red blood cells may be appropriate when Hb is in the range 7–10 g/dL. In such cases, the decision to transfuse should be supported by the need to relieve clinical signs and symptoms and prevent signiﬁcant morbidity and mortality. 3. Use of red blood cells is likely to be appropriate when Hb <7 g/dL. In some patients who are asymptomatic and/or where speciﬁc therapy is available, lower threshold levels may be acceptable. The United Kingdom Blood Services guidelines (McClelland et al.2007) mention that the quality of life of patients with malignancy-associated anaemia may be improved by regular allogeneic RBT. RBT is often a mainstay of supportive therapy in malignant conditions predominantly associated with marrow failure (such as myelodysplasia, myeloﬁbrosis and aplastic anaemia) or extensive marrow inﬁltration (such as chronic lymphocytic leukaemia). Speciﬁc haematinic supplementation may be of beneﬁt in any patient in whom vitamin deﬁciency has been identiﬁed. Iron therapy is often poorly tolerated. EORTC guidelines (Fig. 2) recommendations offered for anaemia management in adult cancer patients with solid tumours or haematological malignancies pertaining to the use of the use of RBT are: 1. Causes of anaemia other than cancer or its treatment should be evaluated. Iron deﬁciency, nutritional defects, bleeding, or haemolysis should be corrected prior to erythropoietic protein therapy. Functional iron deﬁciency should be addressed with intravenous iron. 2. Patients whose Hb level is below 9 g/dL should be evaluated for need of transfusions, in addition to erythropoietic proteins. 3. The target Hb concentration should be 12 to 13 g/dL. 4. The use of erytropoetin stimulating factors should be considered as shown in Fig. 2. 5. The two major goals of erythropoietic protein therapy are improvement of quality of life and prevention of transfusions. 6. The use of erythropoietic proteins with the aim of improving survival or response to treatment should be evaluated in clinical trials in anaemic patients, as there is insufﬁcient evidence to support this use in practice. Furthermore, there is evidence of the lack of long term efﬁcacy of RBT. Thus data from three Canadian trials, with a combined total of 665 patients receiving epoetin alfa treatments for their cancer- and chemotherapy-induced anaemia, showed that pre-transfusion was the most signiﬁcant baseline predictor of subsequent transfusion, with patients that were pre-transfused

When to use red blood cell transfusions in cancer patients

401

Fig. 2. Suggested dosing algorithm for erythropoietic proteins in patients with cancer with anaemia due to cancer or its treatment. Abbreviation: ESP, erythropoiesis stimulating factor. From ref. [Bokemeyer et al. (2007) EORTC guidelines for the use of erythropoietic proteins in anaemic patients with cancer. Eur J Cancer 43: 258–270]

having a signiﬁcantly greater likelihood of being transfused than their transfusion-naive counterparts. Furthermore, and corroborating previous ﬁndings, baseline Hb level was found to be a signiﬁcant predictor of transfusion, with patients who were treated at a baseline Hb level <10 g/dl having a higher chance of being transfused than patients in whom epoetin alfa was initiated at baseline Hb levels of 10–11 g/dl. In addition, when the total units transfused in patients receiving epoetin alfa at different baseline Hb levels were analysed, >85% of the units of blood transfused were received by patients with baseline Hb levels <10 g/dl (Fig. 3). These data suggest that patients requiring and or receiving blood transfusions are at greater risk of requiring further transfusions and may be better treated with alternative therapies.

Conclusion The correction of anaemia in cancer patients with solid tumours appears to have a signiﬁcant role in the local control and overall survival of patients with certain site speciﬁc tumours receiving radiotherapy. In patients

402

J. K. Jacob and P. J. Barrett-Lee

Fig. 3. Relative risk of subsequent transfusion was calculated as a function of baseline Hb strata relative to baseline Hb strata >11 g/dl. A Relative risk of subsequent transfusion calculated in patients transfused after day 28 of epoetin alfa treatment. Patients initiating epoetin alfa therapy at a baseline Hb <10 g/dL had a RR of 2.65 (95% CI, 1.54–4.56) of receiving subsequent transfusions compared to patients initiating therapy earlier (Hb 10–11 g/dL). B Relative risk of subsequent transfusion calculated in patients transfused from baseline to end of study. Patients initiating epoetin alfa therapy at a baseline Hb <10 g/dL had a RR of 2.29 (95% CI, 1.54–3.42) of receiving subsequent transfusions compared to patients initiating therapy earlier (Hb 10–11 g/dL). Abbreviations: CI, conﬁdence interval; Hb, hemoglobin; RR, Relative risk. From ref. [Quirt et al. (2006) Patients previously transfused or treated with epoetin alfa at low baseline hemoglobin are at higher risk for subsequent transfusion: an integrated analysis of the canadian experience. Oncologist 11(1): 73–82]

receiving chemotherapy, patients with metastases and in terminally ill oncology patients, the beneﬁts are less clear. RBT ensures a rapid correction of anaemia; however the rationale for regular transfusions to maintain a clinically optimum Hb level has not yet been tested within a randomised trial in comparison with other modalities of anaemia correction such as intravenous iron therapy, erythropoietin stimulating factors and even newer alternatives to blood in preventing tissue hypoxia (Table 2). There is a need for a better understanding of the concept of transfusion trigger and to determine the

When to use red blood cell transfusions in cancer patients

403

Table 2. Comparison of epoetin and RBT Parameter

Recombinant Human Erythropoietins

Transfusion

Rise in Haemoglobin Speed of effect Safety

Long lasting 4–6 weeks Undesirable effects rare and generally minor Subcutaneous injection weekly £100–£250 /week

Transient (2–4 weeks) Immediate Risk of transfusion reaction and infection Intravenous infusion every 4 weeks £500–£750/ transfusion

Inconvenience Acquisition costs

From ref (http://www.palliativedrugs.com/ May 2002 Newsletter. Hot topicsAnaemia).

outcomes and beneﬁts of planned correction when the Hb levels fall below a speciﬁc nadir. Currently at most centres in Europe, RBT is based on local practices. The concept of transfusion trigger and its impact in the management of anaemia related to the cancer of solid tumours is not well understood. Guidelines for correction of anaemia with RBT need to be more speciﬁc encompassing the role of other alternatives in maintaining a normal Hb level (the deﬁnition of which is also varied) in speciﬁc subpopulations of cancer patients based on disease status, site of tumour and proposed treatment. The EORTC guidelines of 2007 appear to be a step in the right direction but more research and a clearer consensus is needed for such guidelines speciﬁc to RBT in the future.

References 1. Amato A, Pescatori M (2006) Peri-operative blood transfusions for the recurrence of colorectal cancer. Cochrane Database Syst Rev 25(1): CD005033 2. Barrett-Lee PJ, Bailey NP, O’Brien ME, Wager E (2000) Large-scale UK audit of blood transfusion requirements and anaemia in patients receiving cytotoxic chemotherapy. Br J Cancer 82(1): 93–7 3. Bokemeyer C, Aapro MS, Courdi A, Foubertd H. Linke A, Osterborg L, Repetto, Soubeyran P (2007) EORTC guidelines for the use of erythropoietic proteins in anaemic patients with cancer: 2006 update in anaemic patients with cancer. Eur J Cancer 43: 258–270 4. Booth S, Bruera E (2003) Palliative care consultations in haemato-oncology. Oxford University Press, Oxford 5. Daly T, Poulsen MG, Denham JW, Peters LJ, Lamb DS, Krawitz H, Hamilton C, Keller J, Tripcony L, Walker Q (2003) The effect of anaemia on efﬁcacy and

404

6.

7.

8. 9.

10.

11. 12. 13.

14.

15.

16. 17.

18.

19.

J. K. Jacob and P. J. Barrett-Lee normal tissue toxicity following radiotherapy for locally advanced squamous cell carcinoma of the head and neck. Radiother Oncol 68(2): 113–122 Engert A (2000) Recombinant human erythropoietin as an alternative to blood transfusion in cancer-related anaemia. Disease Management and Health Outcomes 8: 259–272. Lee WR, Berkey B, Marcial V, Fu KK, Cooper JS, Vikram B, Coia LR, Rotman M, Ortiz H (1998) Anemia is associated with decreased survival and increased loco-regional failure in patients with locally advanced head and neck carcinoma: a secondary analysis of RTOG 85-27. Int J Radiat Oncol Biol Phys 1;42(5): 1069–1075 Gleeson C, Spencer D (1995) Blood transfusion and its beneﬁts in palliative care. Palliat Med 9(4): 307–313 Grant DG, Hussain A, Hurman D (1999) Pre-treatment anaemia alters outcome in early squamous cell carcinoma of the larynx treated by radical radiotherapy. J Laryngol Otol 113(9): 829–833 Grogan M, Thomas GM, Melamed I, Wong FL, Pearcey RG, Joseph PK, Portelance L, Crook J, Jones KD (1999) The importance of hemoglobin levels during radiotherapy for carcinoma of the cervix. Cancer 15; 86(8): 1528–1536 Groopman JE, Itri LM (1999) Chemotherapy-induced anaemia in adults: incidence and treatment. J Natl Cancer Inst 91: 1616–1634. http://www.nhmrc.gov.au/publications/synopses/_ﬁles/cp78.pdf Leibovitz A, Baumoehl Y, Walach N, Kaplun V, Sigler E, Balan S, Habot B (2004) Medical staff attitudes: views and positions regarding blood transfusion to terminally ill cancer patients. Am J Clin Oncol 27(5): 542–546 Ludwig H, Van Belle S, Barrett-Lee P, Birgegard G, Bokemeyer C, Gascon P, Kosmidis P, Krzakowski M, Nortier J, Olmi P, Schneider M, Schrijvers D (2004) The European Cancer Anaemia Survey (ECAS): a large, multinational, prospective survey deﬁning the prevalence, incidence, and treatment of anaemia in cancer patients. Eur J Cancer 40(15): 2293–2306 MacRae R, Shyr Y, Johnson D, Choy H (2002) Declining haemoglobin during chemo-radiotherapy for locally advanced non-small cell lung cancer is signiﬁcant. Radiother Oncol 64(1): 37–40 McClelland DBL, et al (2007) Handbook of transfusion medicine, 4th edn. TSO publishers, www.tsoshop.co.uk Monti M, Castellani L, Berlusconi A, Cunietti E (1996) Use of red blood cell transfusions in terminally ill cancer patients admitted to a palliative care unit. J Pain and Symptom Management 12(1): 18–22 Quirt I, Kovacs M, Couture F, Turner AR, Noble M, Burkes R, Dolan S, Plante RK, Lau CY, Chang J (2006) Patients previously transfused or treated with epoetin alfa at low baseline hemoglobin are at higher risk for subsequent transfusion: an integrated analysis of the Canadian experience. Oncologist 11(1): 73–82 Rizzo JD, Lichtin AE, Woolf SH, Seidenfeld J, Bennett CL, Cella D, Djulbegovic B, Goode MJ, Jakubowski AA, Lee SJ, Miller CB, Rarick MU, Regan DH, Browman GP, Gordon MS (2002) Use of epoetin in patients with cancer: evidence-based clinical practice guidelines of the American Society of Clinical Oncology and the American Society of Hematology. J Clin Oncol 20(19): 4083–4107

When to use red blood cell transfusions in cancer patients

405

20. Twycross R, Wilcock A (Editors in Chief) (2007) Palliative care formulary, 3rd edn. Palliativedrugs.com Ltd, Nottingham, UK 21. Williamson LM, Lowe S, Love EM, Cohen H, Soldan K, McClelland DBL, Skacel P, Barbara JAJ (1999) Serious hazards of transfusion (SHOT) initiative: analysis of the ﬁrst two annual reports. Br Med J 319: 16–19 Correspondence: Velindre Cancer Centre, Velindre NHS Trust, Whitchurch, Cardiff, Wales CF14 2TL, UK, E-mail: [email protected]

Chapter 16

Pharmacology, pharmacokinetics and safety of recombinant human erythropoietin preparations W. Jelkmann Institute of Physiology, University of Luebeck, Germany

Introduction Hematocrit, blood hemoglobin concentration and red cell mass are regulated by the glycoprotein hormone erythropoietin (EPO). The degree of tissue oxygenation is the primary determinant of the expression of the EPO gene in the kidney. Other production sites contribute little to circulating EPO in humans after birth, albeit the liver is the main site of EPO mRNA expression in fetuses. EPO inhibits the programmed cell death (apoptosis) of the erythrocytic progenitors in the bone marrow, and it stimulates their proliferation and differentiation. Lack of EPO results in normochromic normocytic anemia. The physical power of anemic persons is reduced. Anemia can lead to fatigue, pallor, shortness of breath, tachycardia and angina pectoris due to tissue hypoxia. In severe cases allogeneic red cell transfusion may be required, although there is no clear threshold hemoglobin concentration for intervention. The decision to transfuse is usually based on the severity of the clinical symptoms of hypoxia, and on the patient’s age and co-morbidity. The transfusion of allogeneic blood components is associated with the risk of immunologic reactions and of bacterial or viral infection. In addition, repeated red blood cell transfusions can lead to iron overload. Therefore, the introduction of recombinant human erythropoietin (rhEPO) as an anti-anemic drug for stimulation of erythropoiesis has represented a major therapeutic step in medicine. This chapter provides information on: (a) the genetic engineering and the structure of the convential rhEPO preparations and their novel analogues, (b) the molecular aspects of their action, (c) the pharmacokinetics and safety of the recombinant erythropoietic drugs and (d) future therapeutic developments.

Biochemical characterization of recombinant DNA-derived EPO preparations Recombinant DNA technology allows for the large scale manufacture of rhEPO preparations. Because EPO is a complex glycoprotein, cultures of

408

W. Jelkmann

mammalian cells transfected with the EPO gene linked to an expression vector (“recombinant DNA”) are most suitable for production of the drugs (Hayakawa et al. 1992; Inoue et al. 1995). Mammalian cells can synthesize complex carbohydrate side chains that are similar to the human glycans and capped by terminal sialic acid residues, whereas bacteria are unable to glycosylate proteins. Chinese hamster ovary (CHO) cell cultures are preferably engineered by pharmaceutical companies, because EPO gene ampliﬁcation can be achieved in CHO cell clones deﬁcient in the dihydrofolate reductase gene, which allows for co-selection in the presence of methotrexate (Inoue et al. 1995). Similar to the isolation of other DNA-derived proteins for therapy, recombinant erythropoiesis stimulating factors (ESF) ESF are puriﬁed by a number of chromatographic procedures to ensure that the drugs are free from xenogenic proteins, oncogenes, pyrogens and microorganisms. RhEPO derived from cells transfected with the authentic human EPO gene has been named “Epoetin”. The peptide core of 165 amino acids of epoetin is identical with that of endogenous human EPO (Recny et al. 1987). In addition, epoetin and endogenous EPO are alike with respect to their 2 bisulﬁde bridges and their 4 glycosylation sites, as well as in their secondary structure. The molecules possess 3 tetra-antennary N-linked (Asn 24, 38 and 83) side chains and 1 small O-linked (Ser 126) acidic oligosaccharide side chain (Fig. 1). RhEPO and the native hormone exhibit minor differences in their sugar moieties. Endogenous circulating EPO presents with several glycosylation isoforms (Wide and Bengtsson 1990). In particular, there is micro-

Fig. 1. Scheme illustrating the sugar attachments of epoetin (similar to native EPO), the hyperglycosylated darbepoetin (2 additional glycans) and the continuous erythropoiesis receptor activator CERA containing PEG (polyethylene glycol attached to rhEPO)

Pharmacology, pharmacokinetics and safety

409

heterogeneity of the sugar chain in position Asn 24 (Rahbek-Nielsen H. et al. 1997). The carbohydrates, which amount to 40% of the molecular mass of the glycoprotein (30 kDa), increase its integrity. In addition, they are essential for the in vivo biological activity of EPO. The survival of EPO in circulation depends mainly on the presence of terminal sialic acid residues. Asialo-glycoproteins are rapidly cleared via galactose-receptors of hepatocytes (Weikert et al. 1999). EPO can contain up to 14 sialic acid residues per molecule. RhEPO appears to be more completely glycosylated in the beginning compared to endogenous EPO, as the speciﬁc in vivo biological activity of rhEPO (about 200,000 IU/mg peptide) is greater than that of puriﬁed human urinary EPO (70,000 IU/mg peptide). In addition, the half-life of intravenously injected rhEPO is longer (4–11 h) than that of native EPO (about 2 h) (Embury et al. 1984). Two brands of rhEPO, namely epoetin alfa and epoetin beta, have been marketed for several years for treatment of anemias due to EPO deﬁciency in renal failure, inﬂammatory diseases and malignancies, and for stimulation of erythropoiesis in autologous blood collection programs. Like the endogenous hormone, the epoetins are microheterogenous with respect to their carbohydrate side chains. This statement has been conﬁrmed by mass spectrometry and NMR spectroscopy (Rush et al. 1995). Structural differences between epoetin alfa and epoetin beta have been investigated by isoelectric focusing technique, lectin-binding studies, in vivo and in vitro bioassays, and immunoassays (Storring et al. 1998). Epoetin alfa is more homogenous and possesses less basic isoforms than epoetin beta. In addition, there are less Nglycans with non-sialylated outer Galβ1-4GlcNAc moieties, N-glycans with repeated Galβ1-4GlcNAc sequences, tetra-antennary and 2,6-branched triantennary N-glycans, and Galβ1-3GalNAc containing O-glycans in epoetin alfa compared to epoetin beta. The in vivo-/in vitro-bioactivity of epoetin alfa was slightly lower than that of Epoetin beta when tested in murine systems (Storring et al. 1998). In humans, the plasma half-life of epoetin alfa was measured to be shorter than that of epoetin beta in one study (Halstenson et al. 1991) but not in another (Stockenhuber et al. 1991). Interestingly, the alleged difference in bioactivity of the two brands cannot be related to the degree of sialylation, because epoetin alfa is moderately more sialylated than epoetin beta. Apart from the use of the established epoetins alfa and beta, clinical trials with another rhEPO product (epoetin omega; produced in baby hamster kidney cells, BHK) were carried out in India (Acharya et al. 1995), Slovenia (Bren et al. 2002) and Macedonia (Sikole et al. 2002). The trials have shown that epoetin omega is effective in correcting the anemia in hemodialysis patients. Epoetin omega exhibits differences in the structure of its glycans compared to the CHO cell-derived epoetins alfa and beta (Skibeli et al. 2001). Differing from the CHO cell-derived epoetins, only 60% of the rhEPO produced by BHK cells is O-glycosylated at the serine residue in position

410

W. Jelkmann

126. In addition, it contains a phosphorylated oligomannosidic side chain at one of the N-glycans (Nimtz et al. 1995). Several strategies have been explored to develop recombinant ESF with prolonged survival in circulation. First, dimers and trimers of authentic rhEPO were produced which are in vivo more efﬁcient in stimulating erythropoiesis than the monomers (Sytkowski et al. 1998). However, it is not expected that these products will be licensed for use in humans within the next couple of years. In contrast, darbepoetin alfa (initially termed Novel Erythropoiesis Stimulating Protein, “NESP”) has been approved as an EPOanalogue with prolonged survival in the circulation (MacDougall 2000; Zamboni and Stewart 2002; Cases 2003; Cvetkovic and Goa 2003). Based on the assumption that increasing the sialic acid containing carbohydrate portion of EPO would increase its plasma half-life, additional N-linked carbohydrate chains were added to the molecule by site-directed mutagenesis (Egrie and Browne 2001; Egrie et al. 2003; Elliott et al. 2003). The amino acid sequence of darbepoetin alfa differs from that of human EPO at 5 positions (Ala30Asn, His32Thr, Pro87Val, Trp88Asn, and Pro90Thr) having allowed for additional oligosaccharide attachments at Asn 30 and Asn 88 (Fig. 1). The carbohydrate portion of darbepoetin alfa amounts to 51% resulting in an increased molecular mass of 37 kDa. Furthermore, darbepoetin alfa is characterized by a lower isoelectric point when compared to human urinary EPO or the epoetins. Darbepoetin alfa can contain up to 22 sialic acid residues. In contrast to native human EPO and rhEPO, which are traditionally calibrated in international units (IU), concentrations and doses of darbepoetin alfa are expressed in μg (1 μg is equivalent to about 200 IU rhEPO, based on polypeptide core mass). Pharmacodynamic studies in patients with chronic kidney disease (Locatelli et al. 2001) and in cancer patients undergoing chemotherapy (Glaspy et al. 2002) have conﬁrmed that doses of 100 IU epoetin and 0.5 μg darbepoetin are equally effective in stimulating red cell production.

Action of EPO and its analogues Single intravenous injections of rhEPO (10–1,000 IU/kg body weight) cause dose-proportional increases in reticulocyte counts in healthy volunteers (Flaharty et al. 1990). Similar effects have been demonstrated in patients with chronic renal failure undergoing hemodialysis or continuous ambulatory peritoneal dialysis (CAPD). In the long term, the administration of rhEPO produces an increase in the blood hemoglobin concentration and hematocrit levels (Eschbach et al. 1987; Lim et al. 1989). In previously anemic patients, the improvement of red cell status results in a number of beneﬁcial hemodynamic effects, including a reduction in resting heart rate, cardiac index, left ventricular and diastolic diameter and left ventricular mass. In addition, the

Pharmacology, pharmacokinetics and safety

411

cognitive and psychosomatic functions of the brain improve with the correction of anemia. Details of the pharmacodynamic effects of rhEPO therapy in renal failure patients have been summarized previously (Jelkmann and Gross 1989; Adamson and Eschbach 1990; Jelkmann 1992; Pagel et al. 1992; Dunn and Markham 1996). It is reasonable to assume that the cellular mechanisms of the stimulation of red cell production by recombinant ESF and endogenous EPO are similar. Erythrocytes are the progeny of lineage-speciﬁc progenitors originating from a small pool of multipotent myeloid stem cells (colony-forming unit producing granulocytes, erythrocytes, monocytes and megakaryocytes; CFU-GEMM). Within the various stages of proliferation and differentiation in erythrocytic development, colony-forming units-erythroid (CFU-E) and proerythroblasts are the main targets of EPO. These cells express the greatest number of EPO receptors in their membranes and are dependent on EPO for their survival (Krantz 1991; Koury and Bondurant 1992). In fact, they undergo apoptosis in the absence of EPO (Koury and Bondurant 1990; De Maria et al. 1999). Thus, when the concentration of circulating EPO is low, most CFU-E will die, whereas under conditions of high EPO concentrations a large number of CFU-E will proliferate. The EPO receptor is one of the members of the cytokine receptor superfamily I that show structural similarities in both extracellular and intracellular domains. The active human EPO receptor of hemopoietic cells forms homodimers (Fig. 2). The two subunits of the mature EPO receptor are transmembrane glycoproteins (66 kDa) of 484-amino acids, including a single membrane spanning domain between amino acids 251 and 272. One EPO molecule binds to a high-afﬁnity site of one of the receptor subunits and to a low-afﬁnity site of the other (Philo et al. 1996). The receptor binding afﬁnity is inversely correlated to the degree of sialylation of the individual EPO molecules (Darling et al. 2002; Elliott et al. 2004). The cytoplasmic portion of the EPO receptor subunits consists of two domains. While the membrane proximal domain is required for mitogenesis, the membrane distal domain can exert an inhibitory inﬂuence. The proximal domain of the receptor couples to the cytoplasmic protein tyrosine kinase Janus kinase 2 (JAK2). On EPO binding the homodimers are tightened and the receptor-associated JAK2 is activated. In turn, JAK2, the EPO receptor and several cytosolic proteins are tyrosine-phosphorylated. EPO-induced proliferation and differentiation is inhibited by the receptor-associated tyrosine phosphatase SHP-1 (Klingmuller et al. 1995; Tauchi et al. 1995; Yi et al. 1995). The role of the different signaling pathways paved by EPO (primarily the JAK2/STAT and rasMAP kinase pathways) in cellular survival, proliferation and differentiation are only partly understood. Recent studies have suggested that signaling through the Grb2-associated binder 1 (Gab1) is sufﬁcient to induce proliferation of erythrocytic progenitors, whereas Gab2 activation is required for their differentiation (van den Akker et al. 2004). On EPO binding, the

412

W. Jelkmann

Fig. 2. Scheme of EPO signalling, involving autophosphorylation of JAK2 (Janus kinase 2), phosphorylation of the EPO receptor, homodimerization of STAT5 (signal transducer and activator of transcription 5), activation of PI-3K (phosphatidylinositol-3-kinase), phosphorylation of the adapter protein SHC (SrC-homology and collagen) to form a complex with GRB (growth factor receptor binding protein), SOS (son of sevenless) and the G-protein Ras, and sequential activation of the serinekinase RAF, MEK (syn. MAPKK) and MAPK (mitogen activated protein kinase). The signalling cascade results in survival, proliferation and differentiation of erythrocytic progenitors. The EPO/EPO-receptor complex is internalized and degraded. In addition, the action of EPO is terminated by HCP (hemopoietic cell phosphatase) which catalizes the dephosphorylation of JAK2. Reproduced with permission from Jelkmann W (2004) Molecular biology of erythropoietin. Intern Med 43: 649–659

EPO/EPO-receptor complex is internalized and undergoes proteasomal and lysosomal degradation (Yen et al. 2000; Walrafen et al. 2005). The action of EPO is not restricted absolutely to cells of the erythrocytic lineage. To some extent, rhEPO may also stimulate the growth of megakaryocytic progenitors (Dessypris et al. 1987). Furthermore, recent advances in analytical techniques have led to the demonstration of EPO receptor expression and signaling in a variety of nonhemopoietic cells and organs, including the brain, cardiovascular tissues (endothelium, vascular smooth muscle, cardiomyocytes), the liver, gastrointestinal tissues, pancreatic islands, the kidney, the testis and the female reproductive organs (for references see Moritz et al. 1997; Masuda et al. 1999; Juul 2000; Sasaki et al. 2000). It is assumed that EPO has neurotrophic and neuroprotective (Masuda et al. 1999; Marti et al. 2000; Cerami et al. 2002; Jumbe 2002; Chong et al. 2003), vascular (Masuda et al. 1999; Smith et al. 2003) and cardioprotective (Parsa et al. 2003;

Pharmacology, pharmacokinetics and safety

413

Smith et al. 2003) functions. The therapeutic value of the administration of rhEPO to humans with hypoxic brain injury has already been demonstrated (Ehrenreich et al. 2002). In nonhemopoietic cells the EPO receptor may form heterooligomers with the common cytokine receptor β-subunit (Brines et al. 2004). Thus, mutated EPO derivatives have been designed that are without erythropoietic potential but can still confer neuroprotection (Leist et al. 2004). Whether the expression of EPO receptors on tumor cells is of pharmacological relevance is a matter of present debate (Farrell and Lee 2004; Jelkmann and Wagner 2004).

Pharmacokinetics The pharmacokinetic properties of rhEPO have been well studied in adult healthy volunteers and in patients with renal failure. No such studies have been reported for patients with tumors, autoimmune diseases or AIDS. Speciﬁc observations made in pediatric patients have been described by Widness et al. (1996). RhEPO, administered i.v. at single doses of 50 IU/kg, is eliminated in normal adults or uremic patients at a ﬁrst order kinetic rate following the rapid distribution phase. Calculations of the volume of distribution have ranged from 0.03 to 0.09 l/kg body weight (Cotes et al. 1989; MacDougall et al. 1989; Flaharty et al. 1990; McMahon et al. 1990; Salmonson et al. 1990a, b; Brockmöller et al. 1992), which exceeds the plasma volume. Hence, as a rule of thumb, peak plasma EPO concentrations (IU/l) following i.v. administration can be roughly estimated by multiplying the dose (IU/kg) with the factor 20. The terminal plasma elimination half-life of circulating rhEPO (t 1/2 β) has been measured in the range 4 to 11 h (Table 1). It has been reported that the half-life of i.v. administered rhEPO is approximately 20% shorter in normal subjects than in renal failure patients (Jensen et al. 1994). However, a pharmacokinetic study of rhEPO in patients grouped according to the degree of renal failure revealed small differences between the various groups (Kindler et al. 1989). In addition, direct measurements of 125I-labeled EPO disappearance have shown that there is no difference in the rate of EPO degradation between normal and uremic human subjects (Coles et al. 1992). Following s.c. administration of rhEPO peak plasma levels are achieved after about 12 to 18 h, with bioavailability amounting to about 30% (Table 1). Peak plasma concentrations are approximately 1/20 of the initial values measured after i.v. administration. However, due to the slow absorption increased plasma rhEPO levels are sustained on s.c. application, thereby allowing for up to 30% lower requirements with s.c. versus i.v. administration of the drug (Zachee 1995). Results of high-dose rhEPO therapy studies in healthy volunteers indicate that the absorption rate of rhEPO is independent of dose in the range from 300 to 2,400 IU/kg (single injection), while clearance is dose-

414

W. Jelkmann

Table 1. Synopsis of pharmacokinetics of single doses of rhEPO after intravenous (i.v.) or subcutaneous (s.c.) administration in normal adult persons (NP) and renal failure patients on hemodialysis (RF)* Condition Bioavailability tmax (%) (h)

t 1/2 b (h)

Vd Cl References (ml/kg) (ml/h per kg)

i.v. NP

100

0

4–11

40–90

4–15

RF

100

0

5–10

66–75

11

s.c. NP

36–39

9–29

24–79

–

–

RF

23–49

10–18 9–22

–

–

(Flaharty et al. 1990) (McMahon et al. 1990) (Salmonson et al. 1990a) (Halstenson et al. 1991) (Jensen et al. 1994) (MacDougall et al. 1989) (Cotes et al. 1989) (Salmonson et al. 1990b) (Stockenhuber et al. 1991) (Brockmöller et al. 1992) (Jensen et al. 1994) (Salmonson et al. 1990a) (McMahon et al. 1990) (Halstenson et al. 1991) (Jensen et al. 1994) (Cheung et al. 1998) (MacDougall et al. 1989) (Nielsen 1990) (Stockenhuber et al. 1991) (Brockmöller et al. 1992) (Kampf et al. 1992) (Jensen et al. 1994)

tmax: time to peak plasma concentration, t 1/2 β: elimination half-life in circulation, Vd: volume of distribution; Cl: total clearance. * As a rule of thumb, peak plasma EPO concentrations (U/l) at tmax can be roughly estimated by multiplying the administered dose (IU/kg) with a factor of 20 on i.v., and a factor of 1 on s.c. injection (f.e. 2000 U/l on i.v. versus 100 U/l on s.c. administration of 100 IU/kg).

dependent in that it decreases with increasing dose (Cheung et al. 1998). Up to 1,800 IU/kg the resulting reticulocytosis is proportional to dose, while there is saturation of response beyond the 1,800 IU/kg dose (Cheung et al. 1998). In addition, the data show that repeated s.c. administrations of portions of the total dose are more effective in stimulating erythropoiesis than single bolus doses at the same total amount of rhEPO (Cheung et al. 1998). Note that the intraperitoneal route of rhEPO administration is not recommended due to the low bioavailability (MacDougall et al. 1989; Lui et al. 1990).

Pharmacology, pharmacokinetics and safety

415

Some investigators have proposed that the elimination half-life of rhEPO decreases in renal failure patients during long term therapy (Lim et al. 1989; Brockmöller et al. 1992). However, other investigators have not observed major changes in the pharmacokinetics of rhEPO following repeated administrations in healthy volunteers (McMahon et al. 1990) or renal failure patients (Cotes et al. 1989; Salmonson et al. 1990b; Kampf et al. 1992). The pharmacokinetic properties of darbepoetin alfa were ﬁrst studied in detail in iron-replete dialysed patients with chronic kidney disease and blood hemoglobin concentration <100 g/l (MacDougall et al. 1999). The mean terminal half-life after i.v. administration of darbepoetin alfa (0.5 μg/kg body weight) was 3-fold longer than that of epoetin alfa (100 IU/kg body weight) in this study (25.3 vs. 8.5 h). The area under the serum concentration-time curve was signiﬁcantly greater for darbepoetin alfa (291 vs. 132 ng·h per ml), and clearance was signiﬁcantly lower (1.6 vs. 4.0 ml/h per kg). The volume of distribution was similar for darbepoetin alfa and epoetin alfa (0.052 vs. 0.049 l/kg). Following s.c. injection, plasma levels of darbepoetin alfa increased rather slowly, reaching a peak at around 54 h and mean maximum concentrations that were approximately 10% of that obtained after the equivalent i.v. dose (MacDougall et al. 1999). The half-life of darbepoetin alfa after s.c. administration ranged from 33.5 to 68.0 h (mean 48.8 h). Bioavailability was approximately 37% by the s.c. route. At 168 h after the s.c. injection, the level of circulating darbepoetin alfa was still signiﬁcantly above baseline (MacDougall et al. 1999). The prolonged survival of darbepoetin alfa has been conﬁrmed in patients with non-myeloid malignancies undergoing multiple cycles of chemotherapy (Glaspy et al. 2002; Vansteenkiste et al. 2002; Kotasek et al. 2003). Heatherington et al. (2001) studied the pharmacokinetics in anemic patients with lymphoproliferative malignancies and solid tumors receiving single s.c. doses of darbepoetin alfa (2.25 μg/kg) immediately before chemotherapy. The mean terminal half-life of darbepoetin alfa was 32.6 h with a clearance rate of 3.7 ml/h per kg. The pharmacokinetics were not altered substantially after multiple dosing when determined at the beginning of the third cycle of chemotherapy (Heatherington et al. 2001). Thus, both patients with renal or with nonrenal anemia require less frequent dosing (every 1–3 weeks) with darbepoetin alfa compared with rhEPO. Moreover, a randomized trial has shown that darbepoetin alfa (6.75 μg/kg) is similarly effective when administered once every 3 weeks to anemic cancer patients undergoing chemotherapy on either synchronous (day 1) or asynchronous schedule (day 15) relative to the start of chemotherapy (Glaspy et al. 2005). The bone marrow, the liver and the kidneys have been considered as sites of the degradation of EPO (Fig. 3). Recently, evidence has been summarized suggesting that catabolism by EPO receptor-mediated uptake and degradation in erythropoietic tissues accounts for most of the loss of EPO from the circulation (Jelkmann 2002). Thus, the plasma EPO level in patients with

416

W. Jelkmann

Fig. 3. Alleged sites and mechanisms of the elimination of EPO from circulation. EPO receptor-mediated uptake and degradation in erythropoietic tissues is thought to be most important. Reproduced with permission from Jelkmann W (2002) The enigma of the metabolic fate of circulating erythropoietin (Epo) in view of the pharmacokinetics of the recombinant drugs rhEPO and NESP. Eur J Haematol 69: 265–274

aplastic anemia is much higher than in patients with thalassemia intermedia at the same hemoglobin concentration (Cazzola et al. 1998). It has been earlier proposed that the greater the erythrocytic activity and the number of myeloid red blood cell precursors, the lower the plasma EPO level will be (Jelkmann and Wiedemann 1990). This concept is in line with results indicating nonlinearity of low dose EPO pharmacokinetics (Veng-Pedersen et al. 1995; Yoon et al. 1997; Veng-Pedersen et al. 1999), suggesting that EPO clearance from plasma is associated with a transient saturation of EPO receptors (Chapel et al. 2001). It would be of interest to study the relationship between ESF dosing requirements and chemotherapy inhibition of the proliferation of erythrocytic progenitors expressing EPO receptors.

Dosing The primary goals of ESF therapy are to maintain the hemoglobin concentration above the transfusion trigger, to reduce fatigue and to increase exer-

Pharmacology, pharmacokinetics and safety

417

cise tolerance (Glaspy et al. 1997; Demetri et al. 1998; Nowrousian 1998). Eventually, the quality of life may improve (Littlewood et al. 2003). In patients with malignancies who are under chemotherapy the dosage of epoetin or darbepoetin alfa should be titrated to maintain a hemoglobin concentration of 120 g/l (Rizzo et al. 2002). In general, 10% to 30% of the tumor patients still have to be transfused despite treatment with recombinant ESF (Barosi et al. 1998). In addition, the rhEPO doses (starting dose 450 IU/kg epoetin or 2.25 μg/kg darbepoetin alfa per week) required for prevention of anemia are about 3-fold higher than those in renal failure patients. The combination of a low base-line endogenous serum EPO concentration, a low observed/predicted (O/P) ratio of serum log[EPO] values, and serum ferritin <400 μg/l is considered as a predictor for a positive response (Beguin et al. 1993; Cazzola et al. 1995), which is deﬁned by an increase in blood hemoglobin ≥10 g/l or in reticulocyte count ≥40 × 109/l after 4 weeks of therapy (Cazzola et al. 1997). Anemia correction is not merely a palliative intervention in patients with solid tumors, because anemia (Hb < 120 g/l) is a negative prognostic factor for successful antitumor chemo- and radiotherapy as well as for disease-free survival (Glaspy 1997; Lee 1998; Thomas 2001; Vaupel et al. 2003). Hypoxia induces the transcription of several genes that encode proteins which favour tumor cell survival and tumor growth (Sutherland 1998; Brown 2000). DNA damage by sparsely ionizing radiation (γ-radiation and X-ray) is caused by reactive O2 species and thus depends on the availability of O2. Likewise, the cytotoxicity of various chemotherapeutics is greater in normoxia than in hypoxia (Teicher et al. 1990). On the other hand, hemoglobin levels above 140 g/l are associated with a disturbed intratumoral blood ﬂow (Vaupel and Mayer 2004). Accordingly, clinical trials aimed at elevating the hemoglobin concentration into the normal range (or even above the normal range) by rhEPO therapy showed a negative outcome with respect to tumor progression and survival rates in breast cancer patients under chemotherapy (Leyland-Jones and BEST Investigators and Study Group 2003) and in patients with head and neck cancers undergoing radiotherapy (Henke et al. 2003).

Safety Recombinant ESFs are contraindicated in patients with known hypersensitivity to mammalian cell-derived products. The maximum amount of ESF that can be safely administered as a single dose is not known, since acute toxic effects of EPO have never been reported. Chronic treatment with ESFs may result in erythrocytosis if the hematocrit is not carefully monitored. There is no evidence for a carcinogenic action of EPO. ESFs do not appear to promote tumor growth, although EPO receptor molecules are expressed by tumor cells (Farrell and Lee 2004; Jelkmann and Wagner 2004). This issue needs to

418

W. Jelkmann

be further studied, particularly with a view to myeloid malignancies. In pregnancy, ESF should be administered very cautiously, because the risks for the fetus have not been evaluated in humans. To the best of present knowledge, EPO has no mutagenic potential. Since it is known that proteins can be excreted in human milk, and EPO can be absorbed on oral supply by newborns, caution should be exercised when the drug is administered to nursing women. The question of immunogenicity is important when protein drugs are repeatedly administered to patients. Because native human EPO and rhEPO are almost identical in structure, rhEPO is only very weakly immunogenic. Anti-EPO antibody formation due to epoetin or darbepoetin alfa therapy has not been reported for cancer patients, so far (Bokemeyer et al. 2004). In contrast, in patients with chronic kidney disease about 200 cases have been uncovered of pure red cell aplasia (PRCA) due to the production of neutralizing anti-EPO speciﬁc antibodies (Casadevall et al. 2002; Bennett et al. 2004; Smalling et al. 2004). Most of these cases occurred in patients with chronic kidney disease between 2001 and 2003 and were mainly associated with the subcutaneous use of the epoetin alpha (EPREX®/ERYPO®), marketed outside the USA (Casadevall et al. 2004). The increased immunogenicity was probably caused by leaches from uncoated rubber stoppers in pre-ﬁlled syringes containing polysorbate 80 instead of human serum albumin as stabilizer. The diagnostic criteria for anti-EPO-antibody-induced PRCA include a rapid fall of blood hemoglobin (1 g/l and day), reticulocyte count <10 × 109/l, the demonstration of neutralizing anti-EPO antibodies in serum, and the absence of erythroblasts from an otherwise normal bone marrow (Casadewall et al. 2004). In case of PRCA, erythropoietic products have to be discontinued. Immunosuppressive treatment usually leads to the disappearance of anti-EPO antibodies. Note, too, that major advances have been made recently in adopting procedures to ensure appropriate galenic formulations, storage, handling and administration of erythropoietic drugs to reduce the incidence of antibody-mediated PRCA (Bennett et al. 2004). In particular, the use of ﬂuoro-resin coated stoppers has greatly contributed to the decreased incidence of PRCA (Boven et al. 2005). In renal failure patients treated with ESF the most common unwanted effect is an increase in arterial blood pressure. Hypertension develops mainly in persons on hemodialysis who have a positive family history of high blood pressure (Ishimitsu et al. 1993). Importantly, patients with nonrenal anemia do not develop hypertension on EPO therapy. Also, when hematocrit is raised by EPO treatment from 0.45 to 0.50 in healthy men, their arterial blood pressure values remain unchanged at rest. Only on submaximal exercise, the systolic blood pressure is higher than before rhEPO treatment (Berglund and Ekblom 1991). The concept that arterial blood pressure is not simply correlated with hematocrit is in accord with studies of mice transgenic for the EPO gene. Arterial blood pressure is not elevated in such animals although their

Pharmacology, pharmacokinetics and safety

419

hematocrits are raised to about 0.80 (Ruschitzka et al. 2000; Wagner et al. 2001). In renal failure patients on dialysis slight increases of circulating potassium, phosphorus, creatinine and blood urea nitrogen were observed in earlier clinical trials. However, although these changes were statistically signiﬁcant, they are clinically less important, as all of the parameters remain usually in the normal range. Clearly, care needs to be taken with respect to a strict compliance to medication, diet and regular dialysis. Sowade et al. (1998) have summarized the results of controlled studies of the anticipated safety proﬁle of rhEPO in various indications, including renal anemia, anemia of prematurity, cancer, AIDS and elective surgery with autologous blood transfusion. Importantly, the risk of thromboembolic events did not appear to be increased in any of the indications. On the other hand, defects in hemostasis are occasionally corrected in the course of rhEPO treatment. The prolonged bleeding time in patients with chronic renal failure or in other cases of anemia may actually be shortened as hematocrit rises on rhEPO therapy. During earlier clinical trials with rhEPO, modest statistical increases were seen in platelets (and leukocytes) counts. Renal patients on hemodialysis may require intensiﬁed anticoagulation with heparin to prevent clotting in the dialysis system.

Perspectives In many countries, recombinant erythropoietic agents have been approved for treatment of nonrenal anemias such as for those associated with cancer, myelodysplastic syndromes, bone marrow transplantation, autoimmune diseases and AIDS (Henry et al. 2004). Contrasting the high response rate in renal anemia, ESF resistance (hemoglobin increase <10 g/l in 4 weeks) is relatively often seen in these diseases (Barosi 1994). In patients with solid tumors the justiﬁcation of the administration of ESF is a medical and pharmacoeconomic issue of current debate (Nowrousian 1998). Since the drugs are costly, they must be administered most economically. Dosage and administration rules need to be clearly deﬁned (Rizzo et al. 2002). Attempts have been partially successful to determine predictors of the response to rhEPO (Beguin et al. 1993; Cazzola et al; 1995; Fjornes et al. 1998). Some evidence for a positive response is provided by low base-line endogenous EPO levels relative to the blood hemoglobin concentration, low serum ferritin concentrations (<400 μg/l) and moderate renal failure (increased serum creatinine). The pharmacokinetic, pharmacodynamic and antigenic properties of counterfeit rhEPO preparations that are expected to come to market in the near future will have to be characterized because the present epoetins will no longer be protected by patent rights. RhEPO with full in vivo biological activity can be puriﬁed from the culture supernatant of various genetically

420

W. Jelkmann

engineered mammalian cells such as COS-1 African green monkey kidney cells (Jacobs et al. 1985), BHK baby hamster kidney cells (Broudy et al. 1988; Tsuda et al. 1988; Hayakawa et al. 1992; Nimtz et al. 1993) or C127 mouse mammary cells (Hayakawa et al. 1992). Interestingly, the speciﬁc activities of rhEPO preparations from CHO, BHK-21 and C127 cultures are not identical but were found to range from 1.8 × 105 to 1.0 × 105 units per mg protein (Hayakawa et al. 1992). It remains to be investigated whether these differences in potency are due to differences in the velocity of clearance from circulation related to the structure of the oligosaccharide chains. Strategies are being developed to improve the cell culture conditions for optimal glycosylation of recombinant therapeutic glycoproteins during large-scale production (Yang and Butler 2002). Furthermore, genetic engineering of host cells by stable transfection with genes encoding human glycosyltransferases may enable it to obtain products with human-like tissue-speciﬁc appropriate glycans (Grabenhorst et al. 1999). On the other hand, even human cells transfected with the human EPO gene may express N-glycans with unusual characteristics, as has been demonstrated by an analysis of rhEPO from the human lymphoblastoid cell line, RPMI 1788 (Cointe et al. 2000). Another pharmacological approach has been the search for small EPO mimetics that bind to the EPO receptor and induce intracellular signaling (Barbone et al. 1999). Previously, a family has been described of cyclic 2 kDa peptides of about 20 amino acids (without sequence homology to EPO) which stimulate the proliferation of erythrocytic progenitors in vitro and in vivo (Livnah et al. 1996; Wrighton et al. 1996). Distinct nonpeptide organic molecules can also mimic the action of EPO, at least in vitro (Qureshi et al. 1999). The idea is to design small erythropoietic molecules that are orally active. The third pharmacological approach is to develop compounds that augment the action of EPO, thereby allowing for a reduction of the doses of recombinant erythropoietic agents to be administered. To this end, drugs are under investigation that inhibit the activity of the hematopoietic cell phosphatase, SHP-1 (Barbone et al. 1999). Finally, EPO gene transfer could become an alternative to the administration of rhEPO in anemic patients (Naffakh and Danos 1996). Animal studies have shown that it is possible, by transferring the gene for EPO, to induce translation and secretion of the hormone in therapeutic levels. It will be important to succeed in controlling the expression of the EPO gene to avoid overproduction and to allow the plasma concentration to be adapted to the blood hemoglobin concentration. In an attempt to overcome this problem Rinsch et al. (1997) have developed cell lines genetically engineered to release human EPO as a function of the oxygen tension. In their study a vector containing the human EPO cDNA driven by the hypoxia- responsive phosphoglycerate kinase promoter was used for transfection. Undoubtedly, however, more extensive studies are required to demonstrate medical and economic beneﬁts of a transfer of the EPO gene compared with injections of

Pharmacology, pharmacokinetics and safety

421

pure rhEPO (Naffakh and Danos 1996). Treatment of macaques with adenoassociated virus vectors expressing EPO caused autoimmune anemia in some of the animals (Gao et al. 2004). In view of the relationships between receptor afﬁnity, clearance, efﬁcacy and dosing requirements of recombinant ESFs pharmaceutical research focuses on the development of long-acting rhEPO analogues. With respect to darbepoetin alfa, this goal has been achieved by hyperglycosylation through site-directed mutagenesis (Elliott et al. 2003). Despite the introduction of 5 extraneous amino acids, the drug does not appear to be more immunogenic than the conventional epoetins. Alternatively, the half-life of circulating proteins can be prolonged by the attachment of large polyethylene glycol (PEG) molecules (Abuchowski et al. 1977a, b). In addition, the coupling to PEG can greatly reduce the immunogenicity of proteins (Abuchowski et al. 1977a, b). It has been shown that pegylation of growth hormone (Clark et al. 1996) or interferon-α (Grace et al. 2005) leads to a strong reduction in receptor-binding potential and clearance rate. Similarly, CHO cell-derived rhEPO coupled to PEG is very slowly cleared from circulation (MacDougall et al. 2003). Pegylated rhEPO has been termed Continuous Erythropoiesis Receptor Activator (CERA, Fig. 1), although the mechanism underlying the prolonged action of the novel drug still needs to be identiﬁed in more clarity. CERA contains a single methoxy-polyethylene glycol polymer of approximately 30 kD integrated via succidinimidyl butanoic acid. The median half-life of i.v. injected CERA has been reported to be 7-fold longer than that of epoetin beta in dogs, but only 2-fold longer in rats (MacDougall et al. 2003). Phase I and phase II studies in healthy humans (Dougherty et al. 2003) respectively patients with multiple myeloma (Dmoszynska et al. 2003) have shown that CERA stimulates erythropoiesis in a dose-dependent way.

References 1. Abuchowski A, McCoy JR, Palczuk NC, van Es T, Davis FF (1977b) Effect of covalent attachment of polyethylene glycol on immunogenicity and circulating life of bovine liver catalase. J Biol Chem 252: 3582–3586 2. Abuchowski A, van Es T, Palczuk NC, Davis FF (1977a) Alteration of immunological properties of bovine serum albumin by covalent attachment of polyethylene glycol. J Biol Chem 252: 3578–3581 3. Acharya VN, Sinha DK, Almeida AF, Pathare AV (1995) Effect of low dose recombinant human omega erythropoietin (rHuEPO) on anaemia in patients on hemodialysis. J Assoc Physicians India 43: 539–542 4. Adamson JW, Eschbach JW (1990) Treatment of the anemia of chronic renal failure with recombinant human erythropoietin. Annu Rev Med 41: 349–360 5. Barbone FP, Johnson DL, Farrell FX, Collins A, Middleton SA, McMahon FJ, Tullai J, Jolliffe LK (1999) New epoetin molecules and novel therapeutic approaches. Nephrol Dial Transplant 14: 80–84

422

W. Jelkmann

6. Barosi G (1994) Inadequate erythropoietin response to anemia: deﬁnition and clinical relevance. Ann Hematol 68: 215–223 7. Barosi G, Marchetti M, Liberato NL (1998) Cost-effectiveness of recombinant human erythropoietin in the prevention of chemotherapy-induced anaemia. Br J Cancer 78: 781–787 8. Beguin Y, Loo M, R’Zik S, Sautois B, Lejeune F, Rorive G, Fillet G (1993) Early prediction of response to recombinant human erythropoietin in patients with the anemia of renal failure by serum transferrin receptor and ﬁbrinogen. Blood 82: 2010–2016 9. Bennett CL, Luminari S, Nissenson AR, Tallman MS, Klinge SA, McWilliams N, McKoy JM, Kim B, Lyons EA, Triﬁlio SM, Raisch DW, Evens AM, Kuzel TM, Schumock GT, Belknap SM, Locatelli F, Rossert J, Casadevall N (2004) Pure red-cell aplasia and epoetin therapy. N Engl J Med 351: 1403–1408 10. Berglund B, Ekblom B (1991) Effect of recombinant human erythropoietin treatment on blood pressure and some haematological parameters in healthy men. J Intern Med 229: 125–130 11. Bokemeyer C, Aapro MS, Courdi A, Foubert J, Link H, Österborg A, Repetto L, Soubeyran P (2004) EORTC guidelines for the use of erythropoietic proteins in anaemic patients with cancer. Eur J Cancer 40: 2201–2216 12. Boven K, Stryker S, Knight J, Thomas A, van Regenmortel M, Kemeny DM, Power D, Rossert J, Casadevall N (2005) The increased incidence of pure red cell aplasia with an Eprex formulation in uncoated rubber stopper syringes. Kidney Int 67: 2346–2353 13. Bren A, Kandus A, Varl J, Buturovic J, Ponikvar R, Kveder R, Primozic S, Ivanovich P (2002) A comparison between epoetin omega and epoetin alfa in the correction of anemia in hemodialysis patients: a prospective, controlled crossover study. Artif Organs 26: 91–97 14. Brines M, Grasso G, Fiordaliso F, Sfacteria A, Ghezzi P, Fratelli M, Latini R, Xie QW, Smart J, Su-Rick CJ, Pobre E, Diaz D, Gomez D, Hand C, Coleman T, Cerami A (2004) Erythropoietin mediates tissue protection through an erythropoietin and common beta-subunit heteroreceptor. Proc Natl Acad Sci USA 101: 14907–14912 15. Brockmöller J, Köchling J, Weber W, Looby M, Roots I, Neumayer H-H (1992) The pharmacokinetics and pharmacodynamics of recombinant human erythropoietin in haemodialysis patients. Br J Clin Pharmacol 34: 499–508 16. Broudy VC, Tait JF, Powell JS (1988) Recombinant human erythropoietin: puriﬁcation and analysis of carbohydrate linkage. Arch Biochem Biophys 265: 329–336 17. Brown JM (2000) Exploiting the hypoxic cancer cell: mechanisms and therapeutic strategies. Mol Med Today 6: 157–162 18. Casadevall N, Nataf J, Viron B, Klota A, Kiladjian JJ, Martin-Dupont P, Michaud P, Papo T, Ugo V, Teyssandier I, Varet B, Mayeux P (2002) Pure red-cell aplasia and antierythropoietin antibodies in patients treated with recombinant erythropoietin. N Engl J Med 346: 469–475 19. Cases A (2003) Darbepoetin alfa: a novel erythropoiesis-stimulating protein. Drugs of Today 39: 477–495 20. Cazzola M, Guarnone R, Cerani P, Centenara E, Rovati A, Beguin Y (1998) Red blood cell precursor mass as an independent determinant of serum erythropoietin level. Blood 91: 2139–2145

Pharmacology, pharmacokinetics and safety

423

21. Cazzola M, Mercuriali F, Brugnara C (1997) Use of recombinant human erythropoietin outside the setting of uremia. Blood 89: 4248–4267 22. Cazzola M, Messinger D, Battistel V, Bron D, Cimino R, Enller ZL, Essers U, Greil R, Grossi A, Jager G, LeMevel A, Najman A, Silingardi V, Spriano M, van Hoof A, Ehmer B (1995) Recombinant human erythropoietin in the anemia associated with multiple myeloma or non-Hodgkin’s lymphoma: dose ﬁnding and identiﬁcation of predictors of response. Blood 86: 4446–4453 23. Cerami A, Brines M, Ghezzi P, Cerami C, Itri LM (2002) Neuroprotective properties of epoetin alfa. Nephrol Dial Transplant 17: 8–12 24. Chapel SH, Veng-Pedersen P, Schmidt RL, Widness JA (2001) Receptor-based model accounts for phlebotomy-induced changes in erythropoietin pharmacokinetics. Exp Hematol 29: 425–431 25. Cheung WK, Goon BL, Guilfoyle MC, Wacholtz MC (1998) Pharmacokinetics and pharmacodynamics of recombinant human erythropoietin after single and multiple subcutaneous doses to healthy subjects. Clin Pharmacol Ther 64: 412–423 26. Chong ZZ, Kang JQ, Maiese K (2003) Erythropoietin: cytoprotection in vascular and neuronal cells. Curr Drug Targets Cardiovasc Haematol Disord 3: 141–154 27. Clark R, Olson K, Fuh G, Marian M, Mortensen D, Teshima G, Chang S, Chu H, Mukku V, Canova-Davis E, Somers T, Cronin M, Winkler M, Wells JA (1996) Long-acting growth hormones produced by conjugation with polyethylene glycol. J Biol Chem 271: 21969–21977 28. Cointe D, Beliard R, Jorieux S, Leroy Y, Glacet A, Verbert A, Bourel D, Chirat F (2000) Unusual N-glycosylation of a recombinant human erythropoietin expressed in a human lymphoblastoid cell line does not alter its biological properties. Glycobiology 10: 511–519 29. Coles GA, Liberek T, Davies ME, Robinson M, Jones J, Thomas G, Davies M, MacDougall IC, Williams JD (1992) Estimation of erythropoietin secretion rate in normal and uremic subjects. Am J Physiol 263: F939-F944 30. Cotes PM, Pippard MJ, Reid CD, Winearls CG, Oliver DO, Royston JP (1989) Characterization of the anaemia of chronic renal failure and the mode of its correction by a preparation of human erythropoietin (r-HuEPO). An investigation of the pharmacokinetics of intravenous erythropoietin and its effects on erythrokinetics. Q J Med 70: 113–137 31. Cvetkovic RS, Goa KL (2003) Darbepoetin alfa: in patients with chemotherapyrelated anaemia. Drugs 63: 1067–1074 32. Darling RJ, Kuchibhotla U, Glaesner W, Micanovic R, Witcher DR, Beals JM (2002) Glycosylation of erythropoietin affects receptor binding kinetics: role of electrostatic interactions. Biochemistry 41: 14524–14531 33. De Maria R, Zeuner A, Eramo A, Domenichelli C, Bonci D, Grignani F, Srinivasula SM, Alnemri ES, Testa U, Peschle C (1999) Negative regulation of erythropoiesis by caspase-mediated cleavage of GATA-1. Nature 401: 489– 493 34. Demetri GD, Kris M, Wade J, Degos L, Cella D (1998) Quality-of-life beneﬁt in chemotherapy patients treated with epoetin alfa is independent of disease response or tumor type: results from a prospective community oncology study. Procrit Study Group. J Clin Oncol 16: 3412–3425

424

W. Jelkmann

35. Dessypris EN, Gleaton JH, Armstrong OL (1987) Effect of human recombinant erythropoietin on human marrow megakaryocyte colony formation in vitro. Br J Haematol 65: 265–269 36. Dmoszynska A, Kloczko J, Rokicka M, Hellmann A, Spicka I, Henry D (2003) CERA (Continuous erythropoiesis receptor activator) produces a dose-related response in patients with multiple myeloma: an exploratory phase I-II doseescalation study. Blood 102: 503a (Abstr 1830) 37. Dougherty FC, Reigner B, Jordan P, Pannier A (2003) CERA (Continuous erythropoiesis receptor activator) demonstrates dose-dependent activity and is well tolerated in phase I multiple ascending dose studies. Blood 102: 713 38. Dunn CJ, Markham A (1996) Epoetin beta. A review of its pharmacological properties and clinical use in the management of anaemia associated with chronic renal failure. Drugs 51: 299–318 39. Egrie JC, Browne JK (2001) Development and characterization of novel erythropoiesis stimulating protein (NESP). Br J Cancer 84: 3–10 40. Egrie JC, Dwyer E, Browne JK, Hitz A, Lykos MA (2003) Darbepoetin alfa has a longer circulating half-life and greater in vivo potency than recombinant human erythropoietin. Exp Hematol 31: 290–299 41. Ehrenreich H, Hasselblatt M, Dembrowski C, Cepek L, Lewczuk P, Stiefel M, Rustenbeck H-H, Breiter N, Jacob S, Knerlich F, Bohn M, Poser W, Rüther E, Kochen M, Gefeller O, Gleiter C, Wessel TC, De Ryck M, Itri L, Prange H, Cerami A, Brines M, Sirén A-L (2002) Erythropoietin therapy for acute stroke is both safe and beneﬁcial. Mol Med 8: 495–505 42. Elliott S, Egrie J, Browne J, Lorenzini T, Busse L, Rogers N, Ponting I (2004) Control of rHuEPO biological activity: the role of carbohydrate. Exp Hematol 32: 1146–1155 43. Elliott S, Lorenzini T, Asher S, Aoki K, Brankow D, Buck L, Busse L, Chang D, Fuller J, Grant J, Hernday N, Hokum M, Hu S, Knudten A, Levin N, Komorowski R, Martin F, Navarro R, Osslund T, Rogers G, Rogers N, Trail G, Egrie J (2003) Enhancement of therapeutic protein in vivo activities through glycoengineering. Nat Biotechnol 21: 414–421 44. Embury SH, Garcia JF, Mohandas N, Pennathur DR, Clark MR (1984) Effects of oxygen inhalation on endogenous erythropoietin kinetics, erythropoiesis, and properties of blood cells in sickle-cell anemia. N Engl J Med 311: 291– 295 45. Eschbach JW, Egrie JC, Downing MR, Browne JK, Adamson JW (1987) Correction of the anemia of end-stage renal disease with recombinant human erythropoietin. Results of a combined phase I and II clinical trial. N Engl J Med 316: 73–78 46. Farrell F, Lee A (2004) The erythropoietin receptor and its expression in tumor cells and other tissues. Oncologist 9 [Suppl 5]: 18–30 47. Fjornes T, Wiedemann GJ, Sack K, Jelkmann W (1998) Serum erythropoietin and creatinine concentrations as predictive factors for response to recombinant human erythropoietin treatment in anaemic tumour patients on chemotherapy. Oncol Rep 5: 81–86 48. Flaharty KK, Caro J, Erslev A, Whalen JJ, Morris EM, Bjornsson TD, Vlasses PH (1990) Pharmacokinetics and erythropoietic response to human recombinant erythropoietin in healthy men. Clin Pharmacol Ther 47: 557–564

Pharmacology, pharmacokinetics and safety

425

49. Gao G, Lebherz C, Weiner DJ, Grant R, Calcedo R, McCullough B, Bagg A, Zhang Y, Wilson JM (2004) Erythropoietin gene therapy leads to autoimmune anemia in macaques. Blood 103: 3300–3302 50. Glaspy J (1997) The impact of epoetin alfa on quality of life during cancer chemotherapy: a fresh look at an old problem. Semin Hematol 34 [Suppl 2]: 20–26 51. Glaspy J, Bukowski R, Steinberg D, Taylor C, Tchekmedyian S, Vadhan RS (1997) Impact of therapy with epoetin alfa on clinical outcomes in patients with nonmyeloid malignancies during cancer chemotherapy in community oncology practice. J Clin Oncol 15: 1218–1234 52. Glaspy J, Henry D, Patel R, Tchekmedyian S, Applebaum S, Berdeaux D, Lloyd R, Berg R, Austin M, Rossi G (2005) The effects of chemotherapy on endogenous erythropoietin levels and the pharmacokinetics and erythropoietic response of darbepoetin alfa: a randomised clinical trial of synchronous versus asynchronous dosing of darbepoetin alfa. Eur J Cancer (in press) 53. Glaspy JA, Jadeja JS, Justice G, Kessler J, Richards D, Schwartzberg L, Tchekmedyian NS, Armstrong S, O’Byrne J, Rossi G, Colowick AB (2002) Darbepoetin alfa given every 1 or 2 weeks alleviates anaemia associated with cancer chemotherapy. Br J Cancer 87: 268–276 54. Grabenhorst E, Schlenke P, Pohl S, Nimtz M, Conradt HS (1999) Genetic engineering of recombinant glycoproteins and the glycosylation pathway in mammalian host cells. Glycoconjugate J 16: 81–97 55. Grace MJ, Lee S, Bradshaw S, Chapman J, Spond J, Cox S, Delorenzo M, Brassard D, Wylie D, Cannon-Carlson S, Cullen C, Indelicato S, Voloch M, Bordens R (2005) Site of pegylation and polyethylene glycol molecule size attenuate interferon-alpha antiviral and antiproliferative activities through the JAK/STAT signaling pathway. J Biol Chem 280: 6327–6336 56. Halstenson CE, Macres M, Katz SA, Schnieders JR, Watanabe M, Sobota JT, Abraham PA (1991) Comparative pharmacokinetics and pharmacodynamics of epoetin alfa and epoetin beta. Clin Pharmacol Ther 50: 702–712 57. Hayakawa T, Wada M, Mizuno K, Abe S, Miyashita M, Ueda M (1992) In vivo biological activities of recombinant human erythropoietin analogues produced by CHO cells, BHK cells and C127 cells. Biologicals 20: 253–257 58. Heatherington AC, Schuller J, Mercer AJ (2001) Pharmacokinetics of novel erythropoiesis stimulating protein (NESP) in cancer patients: preliminary report. Br J Cancer 84 [Suppl 1]: 11–16 59. Henke M, Laszing R, Rübe C, Schäfer U, Haase K-D, Schilcher B, Mose S, Beer KT, Burger U, Dougherty C, Frommhold H (2003) Erythropoietin to treat head and neck cancer patients with anaemia undergoing radiotherapy: randomised, double-blind, placebo-controlled trial. Lancet 362: 1255–1260 60. Henry DH, Bowers P, Romano MT, Provenzano R (2004) Epoetin alfa. Clinical evolution of a pleiotropic cytokine. Arch Intern Med 164: 262–276 61. Inoue N, Takeuchi M, Ohashi H, Suzuki T (1995) The production of recombinant human erythropoietin. Biotechnol Annu Rev 1: 297–313 62. Ishimitsu T, Tsukada H, Ogawa Y, Numabe A, Yagi S (1993) Genetic predisposition to hypertension facilitates blood pressure elevation in hemodialysis patients treated with erythropoietin. Am J Med 94: 401–406 63. Jacobs K, Shoemaker C, Rudersdorf R, Neill SD, Kaufman RJ, Mufson A, Seehra J, Jones SS, Hewick R, Fritsch EF, Kawakita M, Shimizu T, Miyake T

426

64.

65. 66. 67. 68.

69.

70. 71. 72.

73.

74.

75.

76. 77. 78. 79. 80.

W. Jelkmann (1985) Isolation and characterization of genomic and cDNA clones of human erythropoietin. Nature 313: 806–810 Jelkmann W (2002) The enigma of the metabolic fate of circulating erythropoietin (Epo) in view of the pharmacokinetics of the recombinant drugs rhEpo and NESP. Eur J Haematol 69: 265–274 Jelkmann W, Gross A (1989) Erythropoietin. Springer, Berlin Heidelberg New York Jelkmann W (1992) Erythropoietin: structure, control of production, and function. Physiol Rev 72: 449–489 Jelkmann W, Wagner K (2004) Beneﬁcial and ominous aspects of the pleiotropic action of erythropoietin. Ann Hematol 83: 673–686 Jelkmann W, Wiedemann G (1990) Serum erythropoietin level: relationships to blood hemoglobin concentration and erythrocytic activity of the bone marrow. Klin Wochenschr 68: 403–407 Jensen JD, Madsen JK, Jensen LW, Pedersen EB (1994) Reduced production, absorption, and elimination of erythropoietin in uremia compared with healthy volunteers. J Am Soc Nephrol 5: 177–185 Jumbe NL (2002) Erythropoietic agents as neurotherapeutic agents: what barriers exist? Oncology (Hungtingt) 16: 91–107 Juul SE (2000) Nonerythropoietic roles of erythropoietin in the fetus and neonate. Clin Perinatol 27: 527–541 Kampf D, Eckardt KU, Fischer HC, Schmalisch C, Ehmer B, Schostak M (1992) Pharmacokinetics of recombinant human erythropoietin in dialysis patients after single and multiple subcutaneous administrations. Nephron 61: 393– 398 Kindler J, Eckardt KU, Ehmer B, Jandeleit K, Kurtz A, Schreiber A, Scigalla P, Sieberth HG (1989) Single-dose pharmacokinetics of recombinant human erythropoietin in patients with various degrees of renal failure. Nephrol Dial Transplant 4: 345–349 Klingmuller U, Lorenz U, Cantley LC, Neel BG, Lodish HF (1995) Speciﬁc recruitment of SH-PTP1 to the erythropoietin receptor causes inactivation of JAK2 and termination of proliferative signals. Cell 80: 729–738 Kotasek D, Steger G, Faught W, Underhill C, Poulsen E, Colowick AB, Rossi G, Mackey J (2003) Darbepoetin alfa administered every 3 weeks alleviates anaemia in patients with solid tumours receiving chemotherapy; results of a double-blind, placebo-controlled, randomised study. Eur J Cancer 39: 2026– 2034 Koury MJ, Bondurant MC (1990) Erythropoietin retards DNA breakdown and prevents programmed death in erythroid progenitor cells. Science 248: 378–381 Koury MJ, Bondurant MC (1992) The molecular mechanism of erythropoietin action. Eur J Biochem 210: 649–663 Krantz SB (1991) Erythropoietin. Blood 77: 419–434 Lee JS (1998) The use of erythropoietin in radiation oncology. Cancer Control 5 [Suppl]: 33–39 Leist M, Ghezzi P, Grasso G, Bianchi R, Villa P, Fratelli M, Savino C, Bianchi M, Nielsen J, Gerwien J, Kallunki P, Larsen AK, Helboe L, Christensen S, Pedersen LO, Nielsen M, Torup L, Sager T, Sfacteria A, Erbayraktar S, Erbayraktar Z, Gokmen N, Yilmaz O, Cerami-Hand C, Xie QW, Coleman

Pharmacology, pharmacokinetics and safety

81.

82.

83.

84.

85.

86.

87. 88.

89.

90.

91.

92.

93.

94. 95.

427

T, Cerami A, Brines M (2004) Derivatives of erythropoietin that are tissue protective but not erythropoietic. Science 305: 239–242 Leyland-Jones B, BEST Investigators and Study Group (2003) Breast cancer trial with erythropoietin terminated unexpectedly. Lancet Oncol 4: 459– 460 Lim VS, DeGowin RL, Zavala D, Kirchner PT, Abels R, Perry P, Fangman J (1989) Recombinant human erythropoietin treatment in pre-dialysis patients. A double-blind placebo-controlled trial. Ann Intern Med 110: 108–114 Littlewood TJ, Nortier J, Rapoport B, Pawlicki M, de Wasch G, Vercammen E, Schuette W, Wils J, Freund M (2003) Epoetin alfa corrects anemia and improves quality of life in patients with hematologic malignancies receiving non-platinum chemotherapy. Hematol Oncol 21: 169–180 Livnah O, Stura EA, Johnson DL, Middleton SA, Mulcahy LS, Wrighton NC, Dower WJ, Jolliffe LK, Wilson IA (1996) Functional mimicry of a protein hormone by a peptide agonist: the EPO receptor complex at 2.8 A. Science 273: 464–471 Locatelli F, Olivares J, Walker R, Wilkie M, Jenkins B, Dewey C, Gray SJ (2001) Novel erythropoiesis stimulating protein for treatment of anemia in chronic renal insufﬁciency. Kidney Int 60: 741–747 Lui SF, Chung WW, Leung CB, Chan K, Lai KN (1990) Pharmacokinetics and pharmacodynamics of subcutaneous and intraperitoneal administration of recombinant human erythropoietin in patients on continuous ambulatory peritoneal dialysis. Clin Nephrol 33: 47–51 MacDougall IC (2000) Novel erythropoiesis stimulating protein. Semin Nephrol 20: 375–381 MacDougall IC, Bailon P, Tare N, Pahlke W, Pill J, Brandt M (2003) CERA (Continuous Erythropoiesis Receptor Activator) for the treatment of renal anemia: an innovative agent with unique receptor binding characteristics and prolonged serum half-life. J Am Soc Nephrol 14: 769A MacDougall IC, Gray SJ, Elston O, Breen C, Jenkins B, Browne J, Egrie J (1999) Pharmacokinetics of novel erythropoiesis stimulating protein compared with epoetin alfa in dialysis patients. J Am Soc Nephrol 10: 2392–2395 MacDougall IC, Roberts DE, Neubert P, Dharmasena AD, Coles GA, Williams JD (1989) Pharmacokinetics of recombinant human erythropoietin in patients on continuous ambulatory peritoneal dialysis. Lancet 1: 425–427 Marti HH, Bernaudin M, Petit E, Bauer C (2000) Neuroprotection and angiogenesis: Dual role of erythropoietin in brain ischemia. News Physiol Sci 15: 225–229 Masuda S, Nagao M, Sasaki R (1999) Erythropoietic, neurotrophic, and angiogenic functions of erythropoietin and regulation of erythropoietin production. Int J Hematol 70: 1–6 McMahon FG, Vargas R, Ryan M, Jain AK, Abels RI, Perry B, Smith IL (1990) Pharmacokinetics and effects of recombinant human erythropoietin after intravenous and subcutaneous injections in healthy volunteers. Blood 76: 1718–1722 Moritz KM, Lim GB, Wintour EM (1997) Developmental regulation of erythropoietin and erythropoiesis. Am J Physiol 273: R1829-R1844 Naffakh N, Danos O (1996) Gene transfer for erythropoiesis enhancement. Mol Med Today 2: 343–348

428

W. Jelkmann

96. Nielsen OJ (1990) Pharmacokinetics of recombinant human erythropoietin in chronic haemodialysis patients. Pharmacol Toxicol 66: 83–86 97. Nimtz M, Martin W, Wray V, Kloppel KD, Augustin J, Conradt HS (1993) Structures of sialylated oligosaccharides of human erythropoietin expressed in recombinant BHK-21 cells. Eur J Biochem 213: 39–56 98. Nimtz M, Wray V, Rudiger A, Conradt HS (1995) Identiﬁcation and structural characterization of a mannose-6-phosphate containing oligomannosidic Nglycan from human erythropoietin secreted by recombinant BHK-21 cells. FEBS Lett 365: 203–208 99. Nowrousian MR (1998) Recombinant human erythropoietin in the treatment of cancer-related or chemotherapy-induced anaemia in patients with solid tumours. Med Oncol 15 [Suppl 1]: S19-S28 100. Pagel H, Weiss C, Jelkmann W (1992) Pathophysiology and pharmacology of erythropoietin. Springer, Berlin Heidelberg 101. Parsa CJ, Matsumoto A, Kim J, Riel RU, Pascal LS, Walton GB, Thompson RB, Petrofski JA, Annex BH, Stamler JS, Koch WJ (2003) A novel protective effect of erythropoietin in the infarcted heart. J Clin Invest 112: 999–1007 102. Philo JS, Aoki KH, Arakawa T, Narhi LO, Wen J (1996) Dimerization of the extracellular domain of the erythropoietin (EPO) receptor by EPO: one highafﬁnity and one low-afﬁnity interaction. Biochemistry 35: 1681–1691 103. Qureshi SA, Kim RM, Konteatis Z, Biazzo DE, Motamedi H, Rodrigues R, Boice JA, Calaycay JR, Bednarek MA, Grifﬁn P, Gao YD, Chapman K, Mark DF (1999) Mimicry of erythropoietin by a nonpeptide molecule. Proc Natl Acad Sci USA 96: 12156–12161 104. Rahbek-Nielsen H, Roepstorff P, Reischl H, Wozny M, Koll H, Haselbeck A (1997) Glycopeptide proﬁling of human urinary erythropoietin by matrixassisted laser desorption/ionization mass spectrometry. J Mass Spectrometry 32: 948–958 105. Recny MA, Scoble HA, Kim Y (1987) Structural characterization of natural human urinary and recombinant DNA-derived erythropoietin. Identiﬁcation of des-arginine 166 erythropoietin. J Biol Chem 262: 17156–17163 106. Rinsch C, Regulier E, Deglon N, Dalle B, Beuzard Y, Aebischer P (1997) A gene therapy approach to regulated delivery of erythropoietin as a function of oxygen tension. Hum Gene Ther 8: 1881–1889 107. Rizzo JD, Lichtin AE, Woolf SH, Seidenfeld J, Bennett CL, Cella D, Djulbegovic B, Goode MJ, Jakubowski AA, Lee SJ, Miller CB, Rarick MU, Regan DH, Gordon MS, Gordon MS (2002) Use of epoetin in patients with cancer: evidence-based clinical practice guidelines of the American Society of Clinical Oncology and the American Society of Hematology. Blood 100: 2303–2320 108. Ruschitzka FT, Wenger RH, Stallmach T, Quaschning T, de Wit C, Wagner K, Kelm M, Noll G, Rülicke T, Shaw S, Lindberg RLP, Rodenwald B, Lutz H, Bauer C, Lüscher TF, Gassmann M (2000) Nitric oxide prevents cardiovascular disease and determines survival in polyglobulic mice overexpressing erythropoietin. Proc Natl Acad Sci USA 97: 11609–11613 109. Rush RS, Derby PL, Smith DM, Merry C, Rogers G, Rohde MF, Katta V (1995) Microheterogeneity of erythropoietin carbohydrate structure. Anal Chem 67: 1442–1452

Pharmacology, pharmacokinetics and safety

429

110. Salmonson T, Danielson BG, Grahnen A, Wikstrom B (1990b) Pharmacokinetics of intravenous recombinant human erythropoietin in patients with chronic renal failure. J Intern Med 228: 53–57 111. Salmonson T, Danielson BG, Wikström B (1990a) The pharmacokinetics of recombinant human erythropoietin after intravenous and subcutaneous administration to healthy subjects. Br J Clin Pharmacol 29: 709–713 112. Sasaki R, Masuda S, Nagao M (2000) Erythropoietin: multiple physiological functions and regulation of biosynthesis. Biosci Biotechnol Biochem 64: 1775–1793 113. Sikole A, Spasovski G, Zaﬁrov D, Polenakovic M (2002) Epoetin omega for treatment of anemia in maintenance hemodialysis patients. Clin Nephrol 57: 237–245 114. Skibeli V, Nissen-Lie G, Torjesen P (2001) Sugar proﬁling proves that human serum erythropoietin differs from recombinant human erythropoietin. Blood 98: 3626–3634 115. Smalling R, Foote M, Molineux G, Swanson SJ, Elliott S (2004) Drug-induced and antibody-mediated pure red cell aplasia: A review of literature and current knowledge. Biotechnol Annu Rev 10: 237–250 116. Smith KJ, Bleyer AJ, Little WC, Sane DC (2003) The cardiovascular effects of erythropoietin. Cardiovasc Res 59: 538–548 117. Sowade B, Sowade O, Möcks J, Franke W, Warnke H (1998) The safety of treatment with recombinant human erythropoietin in clinical use: a review of controlled studies. Int J Mol Med 1: 303–314 118. Stockenhuber F, Loibl U, Gottsauner-Wolf M, Jahn C, Manker W, Meisl TF, Balcke P (1991) Pharmacokinetics and dose response after intravenous and subcutaneous administration of recombinant erythropoietin in patients on regular haemodialysis treatment or continuous ambulatory peritoneal dialysis. Nephron 59: 399–402 119. Storring PL, Tiplady RJ, Gaines DR, Stenning BE, Lamikanra A, Rafferty B, Lee J (1998) Epoetin alfa and beta differ in their erythropoietin isoform compositions and biological properties. Br J Haematol 100: 79–89 120. Sutherland RM (1998) Tumor hypoxia and gene expression - implications for malignant progression and therapy. Acta Oncol 37: 567–574 121. Sytkowski AJ, Lunn ED, Davis KL, Feldman L, Siekman S (1998) Human erythropoietin dimers with markedly enhanced in vivo activity. Proc Natl Acad Sci USA 95: 1184–1188 122. Tauchi T, Feng GS, Shen R, Hoatlin M, Bagby GC Jr, Kabat D, Lu L, Broxmeyer HE (1995) Involvement of SH2-containing phosphotyrosine phosphatase Syp in erythropoietin receptor signal transduction pathways. J Biol Chem 270: 5631–5635 123. Teicher BA, Holden SA, al-Achi A, Herman TS (1990) Classiﬁcation of antineoplastic treatments by their differential toxicity toward putative oxygenated and hypoxic tumor subpopulations in vivo in the FSaIIC murine ﬁbrosarcoma. Cancer Res 50: 3339–3344 124. Thomas G (2001) The effect of hemoglobin level on radiotherapy outcomes: the Canadian experience. Semin Oncol 28 [Suppl 8]: 60–65 125. Tsuda E, Goto M, Murakami A, Akai K, Ueda M, Kawanishi G, Takahashi N, Sasaki R, Chiba H, Ishihara H, et al (1988) Comparative structural study of

430

126.

127.

128. 129.

130.

131.

132.

133.

134.

135. 136.

137.

138. 139.

W. Jelkmann N-linked oligosaccharides of urinary and recombinant erythropoietins. Biochemistry 27: 5646–5654 van den Akker E, van Dijk T, Parren-van Amelsvoort M, Grossmann KS, Schaeper U, Toney-Earley K, Waltz SE, Lowenberg B, von Lindern M (2004) Tyrosine kinase receptor RON functions downstream of the erythropoietin receptor to induce expansion of erythroid progenitors. Blood 103: 4457–4465 Vansteenkiste J, Pirker R, Massuti B, Barata F, Font A, Fiegl M, Siena S, Gateley J, Tomita D, Colowick AB, Musil J (2002) Double-blind, placebo-controlled, randomized phase III trial of darbepoetin alfa in lung cancer patients receiving chemotherapy. J Natl Cancer Inst 94: 1211–1220 Vaupel P, Mayer A (2004) Erythropoietin to treat anaemia in patients with head and neck cancer. The Lancet 363: 992 Vaupel P, Mayer A, Briest S, Höckel M (2003) Oxygenation gain factor: a novel parameter characterizing the association between hemoglobin level and the oxygenation status of breast cancers. Cancer Res 63: 7634–7637 Veng-Pedersen P, Widness JA, Pereira LM, Peters C, Schmidt RL, Lowe LS (1995) Kinetic evaluation of nonlinear drug elimination by a disposition decomposition analysis. Application to the analysis of the nonlinear elimination kinetics of erythropoietin in adult humans. J Pharm Sci 84: 760–767 Veng-Pedersen P, Widness JA, Pereira LM, Schmidt RL, Lowe LS (1999) A comparison of nonlinear pharmacokinetics of erythropoietin in sheep and humans. Biopharm Drug Dispos 20: 217–223 Wagner K, Katschinski DM, Hasegawa J, Schumacher D, Meller B, Gembruch U, Schramm U, Jelkmann W, Gassmann M, Fandrey J (2001) Chronic inborn erythrocytosis leads to cardiac dysfunction and premature death in mice overexpressing erythropoietin. Blood 97: 536–542 Walrafen P, Verdier F, Kadri Z, Chretien S, Lacombe C, Mayeux P (2005) Both proteasomes and lysosomes degrade the activated erythropoietin receptor. Blood 105: 600–608 Weikert S, Papac D, Briggs J, Cowfer D, Tom S, Gawlitzek M, Lofgren J, Mehta S, Chisholm V, Modi N, Eppler S, Carroll K, Chamow S, Peers D, Berman P, Krummen L (1999) Engineering Chinese hamster ovary cells to maximize sialic acid content of recombinant glycoproteins. Nature Biotechnol 17: 1116– 1121 Wide L, Bengtsson C (1990) Molecular charge heterogeneity of human serum erythropoietin. Br J Haematol 76: 121–127 Widness JA, Veng-Pedersen P, Peters C, Pereira LM, Schmidt RL, Lowe LS (1996) Erythropoietin pharmacokinetics in premature infants: developmental, nonlinearity, and treatment effects. J Appl Physiol 80: 140–148 Wrighton NC, Farrell FX, Chang R, Kashyap AK, Barbone FP, Mulcahy LS, Johnson DL, Barrett RW, Jolliffe LK, Dower WJ (1996) Small peptides as potent mimetics of the protein hormone erythropoietin. Science 273: 458–464 Yang M, Butler M (2002) Effects of ammonia and glucosamine on the heterogeneity of erythropoietin glycoforms. Biotechnol Prog 18: 129–138 Yen CH, Yang YC, Ruscetti SK, Kirken RA, Dai RM, Li CC (2000) Involvement of the ubiquitin-proteasome pathway in the degradation of nontyrosine kinase-type cytokine receptors of IL-9, IL-2, and erythropoietin. J Immunol 165: 6372–6380

Pharmacology, pharmacokinetics and safety

431

140. Yi T, Zhang J, Miura O, Ihle JN (1995) Hematopoietic cell phosphatase associates with erythropoietin (Epo) receptor after Epo-induced receptor tyrosine phosphorylation: identiﬁcation of potential binding sites. Blood 85: 87–95 141. Yoon WH, Park SJ, Kim IC, Lee MG (1997) Pharmacokinetics of recombinant human erythropoietin in rabbits and 3/4 nephrectomized rats. Res Commun Mol Pathol Pharmacol 96: 227–240 142. Zachee P (1995) Controversies in selection of epoetin dosages. Issues and answers. Drugs 49: 536–547 143. Zamboni WC, Stewart CE (2002) An overview of the pharmacokinetic disposition of darbepoetin alfa. Pharmacotherapy 22: 133S–140S Correspondence: Wolfgang Jelkmann, M.D., Director, Professor of Physiology, Institute of Physiology, Medical University of Luebeck, Ratzeburger Allee 160, 23538 Luebeck, Germany, E-mail: [email protected]

Chapter 17

Epoetin treatment of anemia associated with multiple myeloma and non-Hodgkin’s lymphoma A. Österborg Departments of Oncology and Hematology, Karolinska University Hospital, Stockholm, Sweden

Introduction Anemia is observed in most patients with multiple myeloma (MM), nonHodgkin’s lymphoma (NHL) and chronic lymphocytic leukemia (CLL) and may already be manifested at the time of diagnosis. In patients with MM, hemoglobin values < 10.0 g/dL were found in 49% and severe anemia (hemoglobin < 7.5 g/dL) in 19% of the patients already at the time of diagnosis (MRC working party 1980). Patients with NHL have been reported to have an anemia rate of approximately 40% (Coifﬁer et al. 1999). In a recent European survey, anemia (Hb < 12 g/dL) was found in 52% of patients with MM or NHL at enrollment and observed in 73% of those patients who were followed for up to 6 months (Ludwig et al. 2004). Anemia usually (but not always, especially in MM) normalizes in patients who achieve complete remission after chemotherapy. However, it persists in patients who are unresponsive to treatment, and recurs in those with relapsing disease. Anemia is a general ﬁnding in later phases of the disease, when toxicity of long-term treatment, impairment of renal function, and heavy tumor load contribute to its induction and aggravation. The cause of anemia in MM, NHL and CLL is multifactorial. Among them, anemia of chronic disease (ACD) is a major cause. It is characterized by blunted production of endogenous erythropoietin (EPO), reduced erythrocyte life-span, poor iron re-utilization and suppressed erythropoiesis due to secretion of inﬂammatory cytokines such as interleukin-1, TNF- α and interleukin-6 (Denz et al. 1990; Means 1995; Maccio et al. 2005) (see also chapter 5 in this book). The inhibitory effect of these cytokines on erythropoiesis has been shown in vitro to be partially overcome by recombinant human erythropoietin (rhEPO, epoetin) (Means and Crantz 1991). In vivo, inﬂammatory cytokines were reported to correlate with rhEPO hyporesponsiveness in hemodialysis patients (Kalantar-Zadeh et al. 2003). IL-1 and TNF also suppress the endogenous EPO synthesis (Faquinet et al. 1992). Inade-

434

A. Österborg

quate EPO production seems to be of pathogenic importance, especially in patients with MM (Musto 1998) where it is found also in many patients with normal creatinine levels (Beguin et al. 1992). Another contributing factor to the patients’ blunted EPO response may be the increased plasma viscosity due to a high level of M-component in the plasma of some patients with MM (Singh et al. 1993). The impact of displacement of erythropoiesis in the bone marrow by tumor cell inﬁltration has probably been overestimated in the past because normal blood counts may sometimes be seen even in patients with heavy myeloma and lymphoma cell inﬁltration of the bone marrow. Anemia may also be induced by chemotherapy and radiotherapy. A variety of cytostatic drugs as well as irradiation directly impair erythropoietic precursor cells in the bone marrow and, thus, inhibit their proliferation. Notably, also low-intensive regimens such as oral melphalan and chlorambucil may induce anemia, probably because of their potent stem cell killing capacity.

Treatment of anemia: blood transfusions For decades, only markedly decreased hemoglobin levels (below 8.0– 8.5 g/dL) were the indication for treatment of anemia. The immediate effect of blood transfusions is crucial when a patient is suffering from severe anemia. However, the resulting increase in hemoglobin levels is transient, i.e. returning to baseline within a short time. Even if the risk of infection due to allogeneic red blood cell (RBC) transfusions has been reduced, there are other negative aspects of transfusions that need to be taken into account in patients with advanced malignancies. Allogeneic RBC transfusions may induce or worsen immunosuppression, which was associated with a signiﬁcant risk of infectious complications (George and Morella 1986; Heiss et al. 1993). RBC transfusions may also possibly inhibit endogenous erythropoietin production resulting in further impairment of erythropoiesis and, as a consequence, an even higher dependence on allogeneic transfusions (Stockman 1996). Repeated RBC transfusions are associated with a highly variable, unstable hemoglobin concentration which often includes periods of overt anemic symptoms. The ﬂuctuating hematocrit may negatively inﬂuence the physiological compensatory mechanisms, such as increase in cardiac output and red cell 2,3-diphosphoglycerate.

Treatment of anemia: epoetin The introduction of epoetin represents a therapeutic alternative to blood transfusions by stabilization of the hemoglobin concentration at a level not usually reached by the current transfusion policy. In addition, maintaining an optimal quality of life (QOL) has become as important as intensive attempts

Epoetin treatment of anemia associated with MM and NHL

435

to control the tumor. This is particularly important when treating patients with non-curable tumors. The changing treatment paradigm is a consequence of a better understanding of the clinical importance of anemia, in particular with regard to QOL of the individual patient. Treatment also appears to be safe when applied according to the current label (Bohlius et al. 2005; Österborg 2005). Considering the relative effectiveness and good tolerability of epoetin in lymphoproliferative malignancies (see below), the main obstacle is treatment cost (Marchetti 2004). However, the ﬁnancial cost of RBC transfusions is also not insigniﬁcant. Already in 1991, the calculated cost of a single allogeneic blood transfusion in the United States was around US$ 450; almost half of that amount was attributable to indirect costs such as travelling, working time at the hospital etc. (Denton 1994). Since then, rapidly increasing costs for blood products have been a general phenomenon in all health care systems (Cremieux 2000). In contrast, the individual costs for epoetin therapy may be going in the opposite direction, resulting from a better understanding of how to select patients for epoetin treatment and individual titration of the lowest effective maintenance dose (Österborg 1998), as well as the emerging price competition between different epoetin preparations, which is partly driven by drug patent expiry.

The use of epoetin in MM, NHL and CLL The development of epoetin provided a new therapeutic option for patients with symptomatic anemia. Beneﬁcial effects on erythropoiesis have been shown both in patients with chronic renal failure and in patients with solid tumors as well as with hematological malignancies. The results from the major studies in patients with lymphoproliferative malignancies are summarized below.

Pilot and phase II studies The ﬁrst pilot study was performed by Ludwig et al. (1990), who observed an increase in the hemoglobin concentration by at least 2g/dL in 11 of 13MM patients. Symptoms of anemia subsided and no adverse side effects were reported. These results were conﬁrmed in a later study by Barlogie and Beck (1993), where 21 of 28MM patients responded with a rise in the hemoglobin concentration. The effect was attributable to epoetin and not to regression of the bone marrow tumor cell inﬁltration or change in dose intensity of concomitant chemotherapy. Oster et al. (1990) reported that escalating doses of epoetin increased levels of hemoglobin in a small cohort of patients with low-grade NHL. These results were mainly conﬁrmed by later phase II studies. A slightly lower response (48%) was reported by Bessho et al. (1994). Musto et al. (1997) conducted a phase II trial on 37 patients with advanced, transfusion-

436

A. Österborg

dependent and chemotherapy-refractory MM and 35% of these poorprognosis patients with advanced MM responded with a rise in hemoglobin concentration and elimination of the need for further transfusions. Mittelman et al. (1997) reported a signiﬁcant rise in hemoglobin concentration in 12/17 chemotherapy-treated MM patients. Taken together, the response rate in these early trials, usually deﬁned as a rise in hemoglobin by ≥2 g/dl, was 61% among totally 142 treated patients.

Randomized trials It is important to remember that factors other than epoetin, such as concomitant chemotherapy and regression or progression of bone marrow tumor cell inﬁltration, may also inﬂuence the hemoglobin concentration and transfusion need. Therefore, several randomized studies have been performed in order to verify and extend the results obtained from the phase II trials. The results are summarized in Table 1. Garton et al. (1995) conducted a prospective, randomized, placebocontrolled trial on epoetin in a small cohort (n = 25) of anemic MM patients. After 12 weeks, nonresponders in the placebo arm were switched to epoetin for 6 weeks (cross-over design). Twenty patients were evaluable for response. Six out of 10 patients (60%) who received epoetin achieved a normal level of hemoglobin concentration, whereas no responses were observed among 10 patients in the control arm. In the open-label phase, 4 of these patients responded to subsequent epoetin treatment. In a study by Silvestris et al. (1995), 54 patients received epoetin (n = 30) or no treatment (n = 24). The total treatment period was as long as 24 weeks. Seventy-eight percent of epoetin-treated patients responded, but the difference compared to controls was statistically signiﬁcant only in those who had not received previous RBC transfusions. Cazzola et al. (1995) conducted a dose-ﬁnding, randomized trial in 146 anemic (Hb > 11 g/dL) but non-transfusion-dependent patients with MM (n = 84) or NHL (n = 62). Four different dose levels of epoetin beta (1,000 U/day, n = 31; 2,000 U/day, n = 29; 5,000 U/day, n = 31; and 10,000 U/day, n = 26) were compared with an untreated control group (n = 29). Epoetin was administered 7 days per week for 8 weeks. Response was deﬁned as an increase in hemoglobin level ≥2 g/dl. The cumulative response rate was signiﬁcantly higher on 5,000 U/day (61%) and 10,000 U/day (62%) but not on 2,000 U/day (31%). However, patients with a normal platelet count achieved an acceptable response rate (50%) also on the low dose (2,000 U/day) of epoetin. 1,000 U/day of epoetin was, however, clearly inferior, as the results were not different from that of the untreated control group. Taken together, the results of this study indicate that the optimum daily dose of rhEPO was 5,000 U (corresponding approximately to 30,000 U per week). Importantly, patients with

Epoetin treatment of anemia associated with MM and NHL

437

Table 1. Randomized trials on epoetin in anemic patients with MM, low-grade NHL and CLL Reference

Tumor type and study design

Garton et al. 1995

MM, placebocontrolled, 2-arm

Silvestris et al. 1995

Response

Predictor for response

20

epoetin arm: 6/10(60%)a control arm: 0/10(0%)a

Not analysed

MM, open, 2-arm

54

epoetin arm: 21/27(78%)b,c control arm: number not givenb,c

Not analysed

Österborg et al. 1996

MM + NHL, open, 3-arm

121

epoetin groups: 60%d control group: 24%d

serum EPO conc., platelet count

Cazzola et al. 1995

MM + NHL, open, 5-arm

146

10,000 U: 62%b 5,000 U: 61%b 2,000 U: 31%b 1,000 U: 6%b control group: 7%b

serum EPO conc., platelet count

Dammacco et al. 2001

MM, placebocontrolled, 2-arm

132

Signiﬁcant rise of hemoglobin conc. and reduced transfusions in epoetin group

Not analysed

Rose et al. 1994

CLL, placebocontrolled, 2-arm

221

epoetin arm: 30%a controlgroup: 5%a

Not analysed

Hedenus et al. 2003

MM + NHL placebocontrolled 2-arm

344

epoetin: 60% placebo: 18%

Serum EPO conc.

Österborg et al. 2002

MM + NHL + CLL placebocontrolled 2-arm

343

epoetin: 67% placebo: 27%

platelet count

a

No. of patients

normalization of hematocrit (≥38%). increase in hemoglobin conc. by ≥2 g/dl. c the difference between control and epoetin was signiﬁcant for transfusiondependent patients only. d elimination of transfusion need plus increase in hemoglobin conc. by >2 g/dl. b

438

A. Österborg

a well-preserved bone marrow function, reﬂected by a normal platelet count, may be treated with lower doses. Österborg et al. (1996) conducted a randomized three-arm study on 121 anemic, transfusion-dependent patients with MM (n = 65) or low-grade NHL (n = 56). Patients were randomized to receive epoetin beta 10,000 U/day, 7 days/wk (ﬁxed-dose group, n = 38), or 2,000 U/day for 8 wk, followed by stepwise escalation of the epoetin dose in nonresponders (titration dose group, n = 44), or to no epoetin treatment (open control group, n = 39). The total treatment time was 24 wk and response was deﬁned as elimination of the transfusion need in combination with an increase in the hemoglobin concentration by ≥2 g/dl. At the end of the study, 60% of the patients in each epoetin group and 24% in the control group had fulﬁlled the response criteria (p < 0.05, log rank test). No emerging differences were observed between the epoetin treatment groups and the control group during the ﬁrst 4 weeks, high-lighting that the onset of the effect of epoetin is relatively slow with at least a 4-week lag time period. Thereafter, the cumulative response rate curves separated step by step between the epoetin treatment groups and the control group. Only 14 percent of the patients in the titration group responded to epoetin therapy at the ﬁrst dose level (2,000 U/day). After stepwise escalation to 5,000 and 10,000 units daily, the cumulative response rate increased to 42% and 60%, respectively. The ﬁrst dose level (2,000 U/day) used in the titration group was thus clearly inferior. Furthermore, there was a tendency towards a more striking effect in MM than in NHL patients and in patients on cheHmotherapy compared with those who did not receive concomitant cytotoxic treatment; the differences were, however, statistically not signiﬁcant. This study further emphasized the importance to include a control group when evaluating the effect of epoetin, as other reasons such as obtaining a remission in the bone marrow as well as cessation of chemotherapy may result in a rise in the hemoglobin concentration; the proportion of patients fulﬁlling the strict response criteria was, in the control group, as high as 24%, despite the advanced stage of disease in most patients. The results of these trials were further conﬁrmed in a randomized placebo-controlled study on 145 anemic MM patients on chemotherapy. A signiﬁcant rise in the hemoglobin concentration and reduction of the proportion of the patients who required transfusions was observed in the epoetin alfa group compared to patients given placebo (Dammacco et al. 2001). In a preliminary report by Rose et al. (1994), 221 anemic patients with CLL were enrolled into a randomized, double-blind, placebo-controlled study. One-hundred and ﬁfty U/kg of epoetin alfa thrice weekly resulted in a signiﬁcant improvement in mean hematocrit values in the treated group versus the control group (plus 5.7% vs plus 1.5%, respectively). Fifty percent versus 15% respectively, showed a ≥6% increase in hematocrit. Although never ﬁnally published, this trial indicated that epoetin may be effective also in anemic patients with CLL.

Epoetin treatment of anemia associated with MM and NHL

439

A subsequent large study investigated the effect of epoetin beta on erythropoiesis, transfusion need and QOL in 349 transfusion-dependent patients with inadequate endogenous EPO levels (Österborg et al. 2002). Of these, 106 patients had NHL, 126 had CLL and 117 had MM; all patients received either 150 U/kg epoetin beta or placebo subcutaneously 3 times weekly for 16 weeks. The response rate was 67% in the epoetin group versus 27% in the placebo group (p < 0.001).After 12 and 16 weeks of treatment, QOL as determined by the Functional Assessment of Cancer Therapy-Anemia (FACT-An) scale (Cella 1997), was signiﬁcantly improved in the epoetin group compared with placebo (p < .05); this improvement correlated with an increase in Hb concentration (<2 g/dL). Cazzola et al. (2003) studied the efﬁcacy of once-weekly 30,000 U of epoetin beta in comparison to the traditional dosing of 10,000 U 3 times weekly. Two hundred and forty-one anemic patients with MM, low-grade NHL or CLL, and defective endogenous EPO production (serum EPO values ≤ 100 U/L) were randomized and treated for 16 weeks. The primary efﬁcacy criterion, the time-adjusted area under the Hb time curve from week 5 to week 16, was comparable between the once-weekly and 3-times-weekly groups (difference = 0.20 g/dL (90% conﬁdence interval = 0.052–0.11)), as were the response rates of 72% and 75%, respectively. Baseline serum EPO was predictive of response: the lower the serum EPO, the higher the likelihood of response (p = 0.002). The authors concluded that administering epoetin once weekly is equally effective as t.i.w. dosing. Darbepoetin alfa is hyperglycosylated epoetin with signiﬁcantly increased serum half-life but reduced receptor afﬁnity, allowing increase of dosing intervals compared with native epoetin (Österborg et al. 2004). In a dose-ﬁnding randomized phase II study, 2.25 μg/kg/wk was identiﬁed as a suitable starting dose (Hedenus et al. 2002). A randomized, double-blind, placebo-controlled study evaluated the efﬁcacy and safety of darbepoetin alfa in 344 anemic patients with lymphoproliferative malignancies (Hedenus et al. 2003). Patients received darbepoetin alfa 2.25 μg/kg or placebo subcutaneously, once weekly for 12 weeks. The dose was to be doubled in case of inadequate response after 4 weeks (increase in Hb > 1.0 g/dL). An Hb response (increase in Hb ≥ 2.0 g/dL) was noted in 60% of patients in the darbepoetin alfa group compared to 18% in the placebo group (p < 0.001). Responsiveness was higher in patients with lower baseline EPO levels. Darbepoetin alfa also resulted in higher mean changes in Hb than placebo from baseline to the last value during the treatment phase (1.80 g/dL versus 0.19 g/dL) and after 12 weeks of treatment (2.66 g/dL versus 0.69 g/dL). A signiﬁcantly lower percentage of patients in the darbepoetin alfa group received red blood cell transfusions than in the placebo group (p < 0.001). The efﬁcacy of darbepoetin alfa was consistent for patients with NHL or MM. Improvements in QOL (FACT-Fatigue subscale) were also observed. Taken together, these randomized, controlled multicenter studies showed that MM and NHL/CLL patients treated with epoetin had a statistically sig-

440

A. Österborg

niﬁcantly higher rate of patients responding with hemoglobin increment and transfusion independence than patients allocated to the control group. The difference could be attributed to the epoetin administration, since [a] there were no differences between the groups with regard to reduction or progression of the underlying malignant disease during the study period, and [b] the intensity of concomitant chemotherapy was similar in the groups. However, not more than 50–60% of all patients responded to epoetin (compared with placebo), which highlights the need for easy-to-use algorithms for prediction of response (see below) and the need for further research to understand and eliminate the underlying mechanisms for lack of response to erythropoietic agents.

Optimal use of epoetin in lymphoproliferative diseases What is the optimum dose of epoetin? To reduce the drug costs, it is important to identify the optimal dose of epoetin. As discussed above, a high weekly dose of epoetin seems to be only marginally better than a standard dose in transfusion-dependent patients (Österborg et al. 1996) and not at all in nontransfused, anemic patients (Cazzola et al. 1995). In these trials, doses as low as 7,000 or 14,000 U/week of epoetin were however, clearly suboptimal if the patients had a suppressed bone marrow function. Thus, a starting dose of 150 U/kg (10,000 U) 3 times/wk of epoetin alfa or beta, or 30,000 U once weekly, may be recommended even though largely overlapping conﬁdence intervals were observed in the dose-ﬁnding trial on darbepoetin alfa (Hedenus et al. 2002). For darbepoetin, 2.25 μg/kg once weekly has been recommended. For patients with a well-preserved bone marrow function, as reﬂected by a normal platelet count, lower epoetin doses may be considered, as the response rate for such patients may still be as high as 50% (Cazzola et al. 1995). It is important to conduct further dose modiﬁcations on an individual basis, in order to maintain the hemoglobin concentration between 11 and 12 g/dL. Importantly, individual titration of the lowest effective epoetin dose may result in a very low weekly maintenance dose (and thus low drug costs) in some but not all responding patients (Österborg et al. 1998). Whether the dose shall be doubled in nonresponding patients is questionable; randomized trials in this setting are still lacking and indirect evidence suggests only a modest additional effect on the response rate (Bokemeyer et al. 2004).

Iron supplementation Administration of epoetin will temporarily lead to iron-deﬁcient erythropoiesis, diminished erythroid marrow response and, thereby, potentially a

Epoetin treatment of anemia associated with MM and NHL

441

nonoptimal effect of the drug (Cazzola et al. 1997). The most frequent phenomenon is development of functional iron deﬁciency, whereby iron stores are at normal but iron supply from the stores to the erythroid system cannot meet the need during the ﬁrst weeks of epoetin therapy (Finch and Huebers 1982). Functional iron deﬁciency may, thus, result in a poor efﬁcacy of epoetin (Glaspy 1999). In order to optimize epoetin therapy, iron supplementation has, therefore, been increasingly recommended within the ﬁrst 4 to 6 weeks of epoetin treatment in all patients, possibly except those with increased serum iron and transferring saturation. Intravenous administration is recommended; randomized trials in patients with ongoing chemotherapy showed that oral iron was clearly less effective than i.v. iron supplementation (Auerbach et al. 2004; Henry et al. 2004).

Quality of life (QOL) The impact of epoetin on QOL has been studied only in some of the trials. Österborg et al. (2002) conducted the most comprehensive study in lymphoproliferative diseases and applied various subscale questionnaires (anemia, fatigue, general) to analyze changes in QOL in epoetin-treated and untreated patients. Both groups showed a tendency for an increase in scores, but a higher beneﬁt was found in the patients on epoetin. The difference in improvement in the treated patients in relation to the control group reached levels of statistical signiﬁcance for the FACT-An and FACT-G score, and this effect became apparent only after 12 weeks of therapy. This phenomenon was due to the substantial proportion of nonresponders in the epoetin group (23%) and the remarkable percentage of responders (27%) in the placebo group. Surprisingly, the FACT-F (fatigue) subscale did not discern several signiﬁcant differences between the two groups: this ﬁnding was recently conﬁrmed also in patients with mainly solid tumors (Witzig et al. 2005). The comparison of the gain in QOL between responders and nonresponders, independent of their treatment group, yielded a highly signiﬁcant improvement in all parameters in the patients with an Hb response >2 g/dL already after a few weeks of treatment. Recently, a target Hb of 12 g/dL was recommended, as the highest improvement in QOL score was observed when Hb increased from 11.0 g/dL to 12.0 g/dL (Crawford 2002). However, analysis of individual data from patients with MM, NHL and CLL revealed that an Hb increase of at least 2 g/dL, irrespective of the ﬁnal Hb concentration, may also be important to improve QOL (Österborg et al. 2003). The FACT-F scale was used by Hedenus et al. (2003) in a similar patient population on either darbepoetin or placebo. Patients on darbepoetin showed a greater improvement in their score compared with placebo, regardless of their level of fatigue at baseline. Patients with the highest fatigue level at baseline reported the biggest gain at the end of treatment. The change in Hb levels correlated with the change in FACT-F over the pretreatment period.

442

A. Österborg

Predictors of response Identiﬁcation of prognostic factors would be a valuable tool in clinical decision-making regarding epoetin therapy. Several models have been published that allow prediction of response to epoetin treatment in individual patients. All of them are based on two fundamental criteria, namely, [1] whether prediction can be done at baseline and [2] whether ﬁrst signs of therapeutic beneﬁts can be detected during the early phase of epoetin treatment. Cazzola et al. (1996) proposed a predictive model of the ﬁrst category for patients with MM and NHL: A relative endogenous EPO deﬁciency was quantiﬁed by the O/P-ratio, where O stands for the observed serum EPO level of the patient and P signiﬁes the hypothetical elevated EPO level that would be expected from his/her degree of anemia. O/P-ratios greater than 0.9 (i.e. absence of relative EPO deﬁciency) predict a highly probable failure to respond to EPO treatment (Cazzola et al. 1995; Österborg et al. 1996). This is further supported by studies on darbepoetin alfa, in which responsiveness was higher in patients with lower serum EPO levels (Hedenus et al. 2003). Around 75% of MM and NHL patients respond if EPO deﬁciency is present, in contrast to only 20–25% in those with a normal O/P ratio (Österborg 1996). As for the second type of prediction model, i.e. whether early signs of response can be detected after 1–2 weeks of therapy (Ludwig et al. 1994; Beguin et al. 2002; Cazzola et al. 2003), data on reliability and clinical value are still missing in speciﬁed diagnoses such as MM, NHL and CLL. The usefulness of algorithms in unselected groups of patients has recently been questioned (Littlewood et al. 2003). In addition, the complexity of most algorithms may hinder penetration into a general use in routine health care. A simple method to be used clinically is to restrict epoetin usage in hematological malignancies to patients with a relative EPO deﬁciency (sEPO levels <200 U/ml if hemoglobin is >9.0 g/dl) and with a platelet count of <100 × 109/L, both of which are independent prognostic factors for response to rhEPO treatment (Cazzola et al. 1995; Österborg et al. 1996, 2002; Hedenus et al. 2003).

Safety aspect The safety of epoetin therapy has recently been discussed due to unexpected inferior outcome in the epoetin group in two randomized trials on patients with head and neck cancer receiving radiotherapy (Henke et al. 2003) and metastatic breast cancer (Leyland-Jones 2003). These safety issues were discussed at an FDA hearing (Luksenburg et al. 2004). To date, there are, however, no data indicating that patients with lymphoproliferative malignancies may have an inferior outcome on epoetin therapy, neither with regard to risk of tumor progression or survival (Bohlius et al. 2005; Österborg et al.

Epoetin treatment of anemia associated with MM and NHL

443

1.0

Epoetin beta Control

0.6

Treatment end

Survival

0.8

0.4

0.2

p=0.76

0.0 0

20

40

60

80

100

120

140

160

180

Weeks from treatment start Fig. 1. Survival of patients with lymphoproliferative malignancies during and after treatment with epoetin beta (n = 170) or placebo ( n = 173), intent-to-treat population

2005). Recently, we conducted an analysis of long-term survival in patients with lymphoproliferative diseases treated with epoetin-beta or placebo in a randomized, double-blind large-scale study in which patients with transfusion-dependent anaemia received epoetin-beta 150 IU/kg or placebo three times weekly for 16 weeks (Österborg et al. 2005). The intention-to-treat population consisted of 343 patients (epoetin-beta, n = 170; placebo, n = 173). The two patient groups were comparable with regard to their demographic or clinical characteristics and prognostic factors. Kaplan-Meier curves for survival were similar in both groups (Fig. 1) and a log-rank test indicated no signiﬁcant difference in survival (P = 0.76). This long-term follow up indicated that epoetin-beta has no signiﬁcant effect on survival compared to placebo in anaemic patients with lymphoproliferative malignancies. A nonsigniﬁcantly increased relative risk (1.58) for thromboembolic events has been described in anemic cancer patients treated with epoetin (Bohlius et al. 2005). It is, therefore, recommended to follow clinical guidelines to avoid very high Hb levels during epoetin treatment (Bokemeyer et al. 2004); a target Hb of ∼12 g/dL is recommended.

Conclusions In the past, blood transfusions have formed the mainstay of therapy for the treatment of anemia in patients with lymphoproliferative malignancies. Treat-

444

A. Österborg

ment with epoetin represents an alternative to blood transfusion for alleviating the symptoms of anemia in these patients, by stabilization of the hemoglobin concentration and at a level not usually reached by the current transfusion policy. The results of randomized controlled clinical trials have conﬁrmed the safety and therapeutic utility of these agents in patients with lymphoproliferative diseases having symptomatic anemia, with effectively increased Hb levels in 50–70% of patients. Identiﬁcation of prognostic factors (i.e. the endogenous serum erythropoietin concentration and the platelet count) and the possibility of reducing the maintenance dose of epoetin, result in a better cost-beneﬁt relationship.

References 1. Auerbach M, Ballard H, Trout JR, McIlwain M, Ackerman A, Bahrain H, Balan S, Barker L, Rana J (2004) Intravenous iron optimizes the response to recombinant human erythropoietin in cancer patients with chemotherapy-related anemia: a multicenter, open-label, randomized trial. J Clin Oncol 22: 1301–1307 2. Barlogie B, Beck T (1993) Recombinant human erythropoietin and the anemia of multiple myeloma. Stem Cells 11: 88–94 3. Beguin Y (2002) Prediction of response and other improvements on the limitations of recombinant human erythropoietin therapy in anemic cancer patients. Haematologica 87: 1209–1221 4. Beguin Y, Yerna M, Loo M, Weber M, Fillet G (1992) Erythropoiesis in multiple myeloma: defective red cell production due to inappropriate erythropoietin production. Br J Haematol 82: 648–653 5. Bessho M, Hirashima K, Tsuchiya J (1994) Improvement of anemia in patients with multiple myeloma by recombinant erythropoietin: A multicenter study. Exp Hematol 22: 705 (abstract) 6. Bohlius J, Langensiepen S, Schwarzer G, Seidenfeld J, Piper M, Bennett C, Engert A (2005) Recombinant human erythropoietin and overall survival in cancer patients: results of a comprehensive meta-analysis. J Natl Cancer Inst 97: 489– 498 7. Bokemeyer C, Aapro MS, Courdi A, Foubert J, Link H, Osterborg A, Repetto L, Soubeyran P (2004) EORTC guidelines for the use of erythropoietic proteins in anaemic patients with cancer. Eur J Cancer 40: 2201–2216 8. Cazzola M, Beguin Y, Kloczko J, Spicka I, Coifﬁer B (2003) Once-weekly epoetin beta is highly effective in treating anaemic patients with lymphoproliferative malignancy and defective endogenous erythropoietin production. Br J Haematol 122: 386–393 9. Cazzola M, Mercuriali F, Brugnara C (1997) Use of recombinant human erythropoietin outside the setting of uremia. Blood 89: 4248–4267 10. Cazzola M, Messinger D, Battistel V, Bron D, Cimino R, Enller-Ziegler L, Essers U, Greil R, Grossi A, Jager G, LeMevel A, Najman A, Silingardi V, Spriano M, van Hoof A, Ehmer B (1995) Recombinant human erythropoietin in the anemia associated with multiple myeloma or non-Hodgkin’s lymphoma: dose ﬁnding and identiﬁcation of predictors of response. Blood 86: 4446–4453

Epoetin treatment of anemia associated with MM and NHL

445

11. Cella D (1997) The Functional Assessment of Cancer Therapy-Anemia (FACTAn) Scale: a new tool for the assessment of outcomes in cancer anemia and fatigue. Semin Hematol 34 [Suppl 2]: 13–19 12. Cofﬁer B (1999) Anemia associated with non-platinum chemotherapy (CT) for Hodgkin’s lymphoma (HL) or non-Hodgkin’s lymphoma (NHL). Eur J Cancer 35 [Suppl 4]: s331 (abstract) 13. Crawford J, Cella D, Cleeland CS, Cremieux PY, Demetri GD, Sarokhan BJ, Slavin MB, Glaspy JA (2002) Relationship between changes in hemoglobin level and quality of life during chemotherapy in anemic cancer patients receiving epoetin alfa therapy. Cancer 95: 888–895 14. Cremieux PY, Barrett B, Anderson K, Slavin MB (2000) Cost of outpatient blood transfusion in cancer patients. J Clin Oncol 18: 2755–2761 15. Dammacco F, Castoldi G, Rodjer S (2001) Efﬁcacy of epoetin alfa in the treatment of anaemia of multiple myeloma. Br J Haematol 113: 172–179 16. Denton TA, Diamond GA, Matloff JM, Gray RJ (1994) Anemia therapy: individual beneﬁt and societal cost. Semin Oncol 21 [Suppl 3]: 29–35 17. Denz H, Fuchs D, Huber H, Nachbaur D, Reibnegger G, Thaler J, Werner ER, Wachter H (1990) Correlation between neopterin, interferon-gamma and haemoglobin in patients with haematological disorders. Eur J Haematol 44: 186–189 18. Faquin WC, Schneider TJ, Goldberg MA (1992) Effect of inﬂammatory cytokines on hypoxia-induced erythropoietin production. Blood 79: 1987–1994 19. Finch CA, Huebers H (1982) Perspectives in iron metabolism. N Engl J Med 306: 1520–1528 20. Garton JP, Gertz MA, Witzig TE, Greipp PR, Lust JA, Schroeder G, Kyle RA (1995) Epoetin alfa for the treatment of the anemia of multiple myeloma. A prospective, randomized, placebo-controlled, double-blind trial. Arch Intern Med 155: 2069–2074 21. George CD, Morello PJ (1986) Immunologic effects of blood transfusion upon renal transplantation, tumor operations, and bacterial infections. Am J Surg 152: 329–337. 22. Glaspy J, Cavill I (1999) Role of iron in optimizing responses of anemic cancer patients to erythropoietin. Oncology (Huntingt) 1; 461–473; discussion 477–8, 483–488 23. Hedenus M, Adriansson M, San Miguel J, Kramer MH, Schipperus MR, Juvonen E, Taylor K, Belch A, Altes A, Martinelli G, Watson D, Matcham J, Rossi G, Littlewood TJ (2003) Efﬁcacy and safety of darbepoetin alfa in anaemic patients with lymphoproliferative malignancies: a randomized, double-blind, placebocontrolled study. Br J Haematol 122: 394–403 24. Hedenus M, Hansen S, Taylor K, Arthur C, Emmerich B, Dewey C, Watson D, Rossi G, Osterborg A (2002) Randomized, dose-ﬁnding study of darbepoetin alfa in anaemic patients with lymphoproliferative malignancies. Br J Haematol 119: 79–86 25. Heiss MM, Mempel W, Jauch KW, Delanoff C, Mayer G, Mempel M, Eissner HJ, Schildberg FW (1993) Beneﬁcial effect of autologous blood transfusion on infectious complications after colorectal cancer surgery. Lancet 342: 1328–1333 26. Henke M, Laszig R, Rube C, Schafer U, Haase KD, Schilcher B, Mose S, Beer KT, Burger U, Dougherty C, Frommhold H (2003) Erythropoietin to treat head

446

27.

28.

29. 30.

31.

32.

33.

34.

35.

36.

37. 38. 39. 40.

A. Österborg and neck cancer patients with anaemia undergoing radiotherapy: randomised, double-blind, placebo-controlled trial. Lancet 362: 1255–1260 Henry D, Dahl N, Auerbach M, Tcheckmedysan S, Laufman L (2004) Intravenous ferric gluconate (FG) for increasing response to epoetin (EPO) in patients with anemia of cancer chemotherapy – results from a multicenter, randomized trial. Blood 104: 10b (abstract) Kalantar-Zadeh K, McAllister CJ, Lehn RS, Lee GH, Nissenson AR, Kopple JD (2003) Effect of malnutrition-inﬂammation complex syndrome on EPO hyporesponsiveness in maintenance hemodialysis patients. Am J Kidney Dis 42: 761– 773 Leyland-Jones B (2003) Breast cancer trial with erythropoietin terminated unexpectedly. Lancet Oncol 4: 459–460 Littlewood TJ, Zagari M, Pallister C, Perkins A (2003) Baseline and early treatment factors are not clinically useful for predicting individual response to erythropoietin in anemic cancer patients. Oncologist 8: 99–107 Ludwig H, Fritz E, Kotzmann H, Hocker P, Gisslinger H, Barnas U (1990) Erythropoietin treatment of anemia associated with multiple myeloma. N Engl J Med 322: 1693–1699 Ludwig H, Leitgeb C, Fritz E, Krainer M, Kuhrer I, Kornek G, Sagaster P, Weissmann A (1993) Erythropoietin treatment of chronic anemia of cancer. Eur J Cancer 29A [Suppl 2]: 8–12 Ludwig H, Fritz E, Leitgeb C, Pecherstorfer M, Samonigg H, Schuster J (1994) Prediction of response to erythropoietin treatment in chronic anemia of cancer. Blood 84: 1056–1063 Ludwig H, Van Belle S, Barrett-Lee P, Birgegard G, Bokemeyer C, Gascon P, Kosmidis P, Krzakowski M, Nortier J, Olmi P, Schneider M, Schrijvers D (2004) The European Cancer Anaemia Survey (ECAS): a large, multinational, prospective survey deﬁning the prevalence, incidence, and treatment of anaemia in cancer patients. Eur J Cancer 40: 2293–2306 Luksemburg H, Weir A, Wager R (2004) Safety conserns associated with Aranesp (darbepoetin alfa) Amgen, Inc. and Procrit (epoetin alfa) Ortho Biotech, L.P., for the Treatment of Anemia Associated with Cancer Chemotherapy [monograph on the internet], FDA Brieﬁng Document, May 4 2004, Oncologic Drugs Advisory Committee [cited 2005 Jan 3]. Available fromm http://www.fda.gov/ ohrms/dockets/ac/04/brieﬁng/4037B2_04_FDA-Aranesp-Procrit.htm. Maccio A, Madeddu C, Massa D, Mudu MC, Lusso M, Gramignano G, Serpe R, Melis G, Mantovani G (2005) Hemoglobin levels correlate with interleukin-6 levels in patients with advanced untreated epithelial ovarian cancer: Role of inﬂammation in cancer-related anemia. Blood, Epub ahead of print Marchetti M, Barosi G (2004) Clinical and economic impact of epoetins in cancer care. Pharmacoeconomics 22: 1029–1045 Maxwell MB (1984) When the cancer patient becomes anemic. Cancer Nurs 7: 321–326 Means RT Jr (1995) Pathogenesis of the anemia of chronic disease: a cytokinemediated anemia. Stem Cells 13: 32–37 Means RT Jr, Krantz SB (1991) Inhibition of human erythroid colony-forming units by gamma interferon can be corrected by recombinant human erythropoietin. Blood 78: 2564–2567

Epoetin treatment of anemia associated with MM and NHL

447

41. Medical Research Council’s Working Party on Leukaemia in Adults (1980) Prognostic features in the third MRC myelomatosis trial. Br J Cancer 42: 831– 840 42. Mittelman M, Zeidman A, Fradin Z, Magazanik A, Lewinski UH, Cohen A (1997) Recombinant human erythropoietin in the treatment of multiple myeloma-associated anemia. Acta Haematol 98: 204–210 43. Musto P, Falcone A, D’Arena G, Scalzulli PR, Matera R, Minervini MM, Lombardi GF, Modoni S, Longo A, Carotenuto M (1997) Clinical results of recombinant erythropoietin in transfusion-dependent patients with refractory multiple myeloma: role of cytokines and monitoring of erythropoiesis. Eur J Haematol 58: 314–319 44. Oster W, Herrmann F, Gamm H, Zeile G, Lindemann A, Muller G, Brune T, Kraemer HP, Mertelsmann R (1990) Erythropoietin for the treatment of anemia of malignancy associated with neoplastic bone marrow inﬁltration. J Clin Oncol 8: 956–962 45. Osterborg A (2004) New erythropoietic proteins: rationale and clinical data. Semin Oncol 31 [Suppl 8]: 12–18 46. Osterborg A, Boogaerts MA, Cimino R, Essers U, Holowiecki J, Juliusson G, Jager G, Najman A, Peest D (1996) Recombinant human erythropoietin in transfusion-dependent anemic patients with multiple myeloma and nonHodgkin’s lymphoma–a randomized multicenter study. The European Study Group of Erythropoietin (Epoetin Beta) Treatment in Multiple Myeloma and Non-Hodgkin’s Lymphoma. Blood 87: 2675–2682 47. Osterborg A, Brandberg Y (2003) Relationship between changes in hemoglobin level and quality of life during chemotherapy in anemic cancer patients receiving epoetin alfa therapy. Cancer 973125–6; author reply 3126– 3127 48. Osterborg A, Brandberg Y, Hedenus M (2005) Impact of epoetin-beta on survival of patients with lymphoproliferative malignancies: long-term follow up of a large randomized study. Br J Haematol 129: 206–209 49. Osterborg A, Brandberg Y, Molostova V, Iosava G, Abdulkadyrov K, Hedenus M, Messinger D (2002) Randomized, double-blind, placebo-controlled trial of recombinant human erythropoietin, epoetin Beta, in hematologic malignancies. J Clin Oncol 20: 2486–2494 50. Osterborg A (1998) Recombinant human erythropoietin (rHuEPO) therapy in patients with cancer-related anaemia: What have we learned? Med Oncol 15 [Suppl 1]: 47–49 51. Rose E, Rai K, Revicki D, Brown R, Reblando J (1994) Clinical and health status assessments in anemic chronic lymphocytic leukemia (CLL) patients trated with epoetin alfa (EPO). Blood 84 [Suppl 1]: 526 (abstract) 52. Silvestris F, Romito A, Fanelli P, Vacca A, Dammacco F (1995) Long-term therapy with recombinant human erythropoietin (rHu-EPO) in progressing multiple myeloma. Ann Hematol 70: 7013–7318 53. Singh A, Eckardt KU, Zimmermann A, Gotz KH, Hamann M, Ratcliffe PJ, Kurtz A, Reinhart WH (1993) Increased plasma viscosity as a reason for inappropriate erythropoietin formation. J Clin Invest 91: 251–256 54. Stockman JA, 3rd (1986) Anemia of prematurity. Current concepts in the issue of when to transfuse. Pediatr Clin North Am 33: 111–128

448 A. Österborg: Epoetin treatment of anemia associated with MM and NHL 55. Witzig TE, Silberstein PT, Loprinzi CL, Sloan JA, Novotny PJ, Maillard JA, Rowland KM, Alberts SR, Krook JE, Levitt R, Morton RF (2005) Phase III, randomized, double-blind study of epoetin alfa compared with placebo in anemic patients receiving chemotherapy. J Clin Oncol 23: 2606–2617 Correspondence: Anders Österborg, Department of Oncology (Radiumhemmet), Karolinska University Hospital, 171 76 Stockholm, Sweden, E-mail: anders. [email protected]

Chapter 18

rhEPO in anemic patients with solid tumors and chemotherapy – efﬁcacy and safety M. R. Nowrousian Department of Internal Medicine (Cancer Research), West German Cancer Center, University Hospital of Essen Essen, Germany

Introduction Patients with cancer frequently develop anemia, related either to the malignant disease itself or its treatment. The frequency of anemia, however, depends on the type and stage of malignant disease and the type and intensity of treatment. Important factors for transfusion requirement are the degree of anemia, the age of patients and the presence of additional disorders, particularly those of the cardiovascular and pulmonary system. In a retrospective study, 18% of patients with various types of solid tumors receiving chemotherapy required red blood cell (RBC) transfusions, and among them, those with lung cancer had the highest rate (34%) (Skillings et al. 1993, 1995). In another study including 2,719 patients with solid tumors receiving chemotherapy, 33% required at least one blood transfusion during the course of chemotherapy and 16% multiple transfusions (Barrett-Lee et al. 2000). The mean proportion of patients with hemoglobin (Hb) levels less than 11 g/dl rose from 17% before the ﬁrst cycle, to 38% by the sixth cycle of chemotherapy, despite transfusions in 33% of patients. In a large European cancer anemia survey (ECAS) including 8,470 patients on chemotherapy, anemia (Hb < 12 g/dl) was reported to occur in 62 to 88% of patients, depending on the type of underlying malignancy and the treatment used (Ludwig et al. 2004, Ludwig, Chapter 7 in this book). The highest proportions of patients with anemia were observed in patients with lung cancer (83.3%) and gynecological malignancies (88.3%). In a comprehensive review of chemotherapy trials published in the literature, patients with lung cancer or ovarian cancer were also found to develop more frequently severe anemia (grade 3 to 4, Hb level < 8 g/dl) than patients with other types of solid tumors. Severe anemia was reported to occur in up to 54% in the ﬁrst group of patients and in up to 42% in the second (Groopman and Itri 1999). In a study of patients with advanced malignancy, the proportion of patients requiring RBC transfusions was 38% in those not receiving chemotherapy and 49% and 69% in those

450

M. R. Nowrousian

receiving non-platinum-based or platinum-based chemotherapy, respectively (Abels et al. 1991). The type of chemotherapy has also been reported to determine the frequency of anemia and its severity as well as the need for RBC transfusions in other studies. The incidence of anemia and the need for RBC transfusions appear to be particularly high in elderly patients receiving chemotherapy and patients who receive chemotherapeutic regimens containing platinum or anthracycline in combination with agents such as etoposide, ifosfamide, cyclophosphamide, paclitaxel, gemcitabine, vinorelbine or irinotecan (Nowrousian, Chapter 8 in this book). As shown in a number of studies and extensively discussed in another chapter of this book (Nowrousian, Chapter 13), RBC transfusions only temporarily reduce the sequelae of anemia and are, beside the inconvenience for patients, associated with the risk of alloimmunization, allergic and febrile reactions, and transmission of bacterial, parasitic and viral infections (Table 1).

Table 1. Comparison between RBC transfusions and rhEPO RBC transfusions

rhEPO

Advantages:

Immediate effect, beneﬁcial effect in patients not responding to rhEPO

Disadvantages:

Transient beneﬁt, allergic reactions, alloimmunization, transmission of viral, bacterial and parasitic infections, acute and delayed hemolytic reactions, transfusionrelated lung injury and graftversus-host disease, suppression of cellular immunity with the risk of infections and possibly tumor growth, iron overload, volume expansion, highly time-consuming, inconvenient, patients’ aversion, limited supply, loss of efﬁcacy in alloimmunized patients, little effect on QOL

Sustained physiological increase in hemoglobin level, well tolerated, signiﬁcant improvement in metabolic functions, exercise capacity, physical and mental well-being and QOL Response rate 50–75%, time to response > 4 weeks, no reliable predictors of response

RBC = red blood cells, rhEPO = recombinant human erythropoietin, QOL = quality of life.

rhEPO in anemic patients with solid tumors and chemotherapy

451

Table 2. Indications for RBC transfusion and rhEPO related to baseline hemoglobin (Hb) level and the need of rapid increase in Hb Anemia

Treatment

Hemoglobin <9 g/dl Hemoglobin 9–11 g/dl, but a rapid increase in hemoglobin required Hemoglobin 9–11 g/dl and no need for a rapid increase in hemoglobin

RBC transfusions + rhEPO RBC transfusions + rhEPO rhEPO

RBC = red blood cell, rhEPO = recombinant human erythropoietin.

In addition, allogeneic transfusions downregulate the cellular immunity, and there are, even if controversial, experimental and clinical study results suggesting that this may favor the development of certain types of cancer or cancer recurrence (Blumberg and Heal 1996; Blumberg 1997; Jacob and Barrett-Lee, Chapter 15 in this book; Nowrousian, Chapter 13 in this book). RBC transfusions may be of value in conditions, in which an urgent increase in Hb is required (Table 2), but they do not represent a sufﬁcient treatment of chronic anemia because of their functional limitations, short-lived effect, and risks (Nowrousian, Chapter 13 in this book). Anemia has not only major negative impacts on organ function and quality of life (QOL) of patients (Butt and Cella, Chapter 14 in this book; Chowdhury et al., Chapter 29 in this book; Nowrousian, Chapter 13 in this book) but is also associated with a poor outcome of treatment in various types of malignancies (Oivanen 1996; Ferme et al. 1997; Hasenclever and Diehl 1998; Moullet et al. 1998; Bacci et al. 2000; Caro et al. 2001; Nowrousian, Chapter 8 in this book). In this regard, it may be merely considered as an expression of a more aggressive disease, but there is evidence suggesting that it may independently have an impact on the outcome of treatment. Such an impact may result from alterations of various organ functions, patient compliance and the sensitivity of tumor cells to chemotherapy. Anemia can aggravate tumor hypoxia (Becker et al. 2000), and experimental studies show that the latter can increase the metastatic potential of tumor cells (Young and Hill 1990) and reduce the sensitivity of these cells to irradiation and a broad spectrum of chemotherapeutic agents (Teicher et al. 1990; Teicher 1994, 1995; Tomida and Tsuruo 1999; Höckel and Vaupel 2001; Airley and Mobasheri 2007; Thews et al. 2007; Vaupel et al., Chapter 10 in this book; Wouters et al. 2007). In cell culture, hypoxic conditions have been reported to reduce the sensitivity of malignant cells to a broad spectrum of cytotoxic agents, including methotrexate, 5-ﬂuorouracil, doxorubicin, etoposide, gemcitabine, actinomycin D, carboplatin, bleomycin, and paclitaxel (Teicher et al.

452

M. R. Nowrousian

1990; Teicher 1994, 1995; Wouters et al. 2007). The underlying mechanisms of hypoxia-induced drug resistance may be 1) tissue acidosis, 2) generation of stress proteins, 3) decrease in cytotoxicity, 4) inhibition of cell proliferation, and 5) loss of apoptotic potential of tumor cells (Airley and Mobasheri 2007; Dunst, Chapter 9 in this book; Höckel and Vaupel 2001; Thews et al. 2007; Vaupel et al., Chapter 11 in this book). The different mode of action of cytotoxic agents and the potential ways to induce drug resistance, however, suggest that under hypoxic conditions universally active mechanisms or coactivation of several pathways are involved in inducing drug resistance (Wouters et al. 2007). Considering the impact of anemia on QOL, its possible impact on the outcome of chemotherapy and the problems associated with RBC transfusions, it is reasonable to treat anemia more consistently in patients with cancer and to look for alternatives to RBC transfusions. In recent years, many studies have revealed new insights into the pathogenesis of cancer-related and chemotherapy-induced anemia (Nowrousian et al. 1996; Nowrousian 1998 and Chapter 6 in this book), and since the introduction of recombinant human erythropoietin (rhEPO), it has become possible to treat these types of anemia in a considerable proportion of patients more effectively than by RBC transfusion. The purpose of this overview is to give an up-to-date summary of the results of these studies in patients with solid tumors and chemotherapy-induced anemia.

Treatment of anemia with erythropoiesis-stimulating agents (ESAs) Pharmacological properties of ESAs The currently available ESAs for treating cancer-related and chemotherapyinduced anemia are the two recombinant forms of endogenous human EPO, epoetin alfa and epoetin beta, and a genetically engineered analog of epoetin alfa named darbepoetin alfa. The latter contains an increased number of carbohydrate chains and sialic acid residues resulting in a two- to threefold longer plasma half-life but a fourfold lower binding afﬁnity to EPO receptor than epoetin alfa or epoetin beta (Table 3) (Egrie and Browne 2001; Gascon 2005; Glaspy 2005; Vansteenkiste and Wauters 2005; Asif et al. 2006a,b; Ludwig 2006a; Vansteenkiste 2006; Jelkmann, Chapter 16 in this book). Another difference resulting from hyperglycosylation and increased sialic acid content is a prolonged time to maximal serum concentration of darbepoetin alfa after subcutaneous (sc) administration compared with epoetin alfa and epoetin beta. The two latter ESAs are very similar in their pharmacokinetic and pharmacodynamic properties, but epoetin beta has been reported to have a greater volume of distribution, a prolonged elimination after intravenous administration and a delayed absorption after subcutaneous application than epoetin alfa (Halstenson et al. 1991; Gascon 2005;

150 U/kg sc, increasing to 300 U/kg sc tiw 450 U/kg sc, increasing to 900 U/kg sc QW 30,000 U (total dose) sc, increasing to 60,000 U sc QW

165 amino acids 3 N-linked carbohydrate chains, 1 O-linked carbohydrate chain up to 14 sialic acid residues 40% carbohydrate MW = 30.4 kDa 13–28 h 15 ± 7 h (healthy volunteers) 23–42% 4,146.7 ± 988.8 mU h/ml (healthy volunteers)*

40,000 U (total dose) sc, increasing to 60,000 U sc QW

150 U/kg sc, increasing to 300 U/kg sc tiw 450 U/kg sc QW

165 amino acids 3 N-linked carbohydrate chains, 1 O-linked carbohydrate chain up to 14 sialic acid residues 40% carbohydrate MW = 30.4 kDa 16–19 h 15 ± 8 h (healthy volunteers) 20–30% 3,933.2 ± 891.4 mU h/ml (/healthy volunteers)*

Epoetin alfa

2.25 μg/kg sc, increasing to 4.5 μg/kg sc QW 500 μg (total dose) sc Q3W

165 amino acids (with 5 substitutions) 3 N-linked carbohydrate chains, 1 O-linked carbohydrate chain up to 22 sialic acid residues 51% carbohydrate MW = 37.1 kDa 33–48 h 86.1 ± 22.8 (cancer patients) 37% (dialysis patients) 291.0 ± 7.6 ng h/ml (dialysis patients)**

Darbepoetin alfa

MW: molecular weight; t1/2: elimination half-life; tmax: time to maximal serum concentration; AUC: area under the serum concentration-time curve; * AUC (0-inﬁnity); ** AUC (0–96). sc: subcutaneous; U: units; tiw: three times a week; QW: once a week; Q3W: once every three weeks. Adapted from Macdougall et al. 1999b, Heatherington et al. 2001, Morreale et al. 2004, Gascon 2005, and Ludwig 2006b.

Dosage regimens

t1/2 (sc) tmax (sc) Bioavailability AUC (sc)

Structure

Epoetin beta

Table 3. Structure, pharmacokinetics, pharmacodynamics and dosage regimens of the three commercially available erythropoiesisstimulating agents (ESAs) rhEPO in anemic patients with solid tumors and chemotherapy 453

454

M. R. Nowrousian

Ludwig 2006a; Jelkmann, Chapter 16 in this book). Whether these differences are of clinical relevance is unclear.

Dose-response relation, response rate Since the early 1990s, a number of studies, both randomized and nonrandomized, has been performed to evaluate the efﬁcacy and safety of rhEPO (epoetin alfa or epoetin beta) in the treatment of anemia related to solid tumors and chemotherapy (Table 4) (Platanias et al. 1991; Miller et al. 1992; Abels 1993; Cascinu et al. 1993, 1994a,b, 1995; Tsukuda et al. 1993; Henry and Abels 1994; Ludwig et al. 1995; Sevelda et al. 1996; Voigtmann et al. 1996; Glaspy et al. 1997; Kasper et al. 1997; Pawlicki et al. 1997; Demetri et al. 1998; Glimelius et al. 1998; Oberhoff et al. 1998; Gabrilove et al. 2001; Littlewood et al. 2001; Chang et al. 2005; Witzig et al. 2005; Aapro et al. 2006a,b). In a part of these studies, only patients with solid tumors (Miller et al. 1992; Cascinu et al. 1993; 1994a,b, 1995; Sevelda et al. 1996; Voigtmann et al. 1996; Glimelius et al. 1998; Oberhoff et al. 1998; Chang et al. 2005; Aapro et al. 2006a,b), and in another part, both patients with solid tumors or hematological malignancies were evaluated (Platanias et al. 1991; Abels 1993; Tsukuda et al. 1993; Henry and Abels 1994; Ludwig et al. 1995; Voigtmann et al. 1996; Glaspy et al. 1997; Kasper et al. 1997; Pawlicki et al. 1997; Demetri et al. 1998; Gabrilove et al. 2001; Littlewood et al. 2001; Witzig et al. 2005). In most of these studies, anemia was considered to be induced by chemotherapy, but in a large, randomized, double-blind, and placebo-controlled trial, the therapeutic effect of rhEPO was also tested in patients not receiving chemotherapy (Abels 1993; Henry and Abels 1994). In this study, the response rate of anemia was 40% and the mean dose of rhEPO at response 382 U/kg/wk. There was a reduced transfusion requirement in rhEPO-treated patients, but the difference to placebo was statistically not signiﬁcant. This and the lower response rate of anemia to rhEPO compared with that observed in two parallel studies including patients receiving chemotherapy (40% vs. 56% and 58%, respectively) (Abels 1993; Henry and Abels 1994) may have been related to a lower dose of rhEPO (100 U/kg sc vs 150 U/kg sc three times a week) used. In two studies using an rhEPO dose of 150 U/kg sc three times a week (tiw) and in one study using an equally effective dose (2.25 μg/kg, once a week) (Glaspy et al. 2001) of darbepoetin alfa in anemic cancer patients not receiving chemotherapy, the response rates were 62%, 48%, and 67%, respectively (Ludwig et al. 1995; Quirt et al. 2001; Smith et al. 2001). In some studies, the relation between the dose of rhEPO and the response rate of anemia was evaluated indicating an increasing rate of response with increasing dose of rhEPO (Platanias et al. 1991; Miller et al. 1992; Tsukuda et al. 1993; Cascinu et al. 1994a; Glimelius et al. 1998). In a study using an rhEPO dose of 25 or 50 U/kg sc ﬁve times a week (ﬁw), the

No. of patients

17/13

11/10

20 12/14

100 19 144 107 20/20 68 42

48

1,047 99 116 1,224

100

189 2,964

218 131

Cancer, chemotherapy

ST/HM, Ch

ST, ChP

ST, ChP ST, HM, Ch/ChP

ST, ChP ST, ChP(CarboPl) ST/HM, Ch ST/HM, ChP ST, ChP ST/HM, Ch/ChP ST, Ch/ChP

ST/HM, Ch/ChP

ST/HM, Ch/ChP ST/HM, Ch ST/HM, ChP ST, Ch/ChP

ST, Ch

ST, Ch/ChP ST/HM, Ch?

ST/HM, Ch? ST, Ch

rhEPO 25,50,100/200,300 U/kg, iv ﬁw rhEPO 25,50/100, 200 U/kg, sc ﬁw rhEPO 50–100 U/kg, sc tiw rhEPO 3,000/6,000 U (total) iv tiw rhEPO 100 U/kg, sc tiw rhEPO 75/100/150 U/kg, sc tiw rhEPO 150 U/kg, sc tiw rhEPO 150 U/kg, sc tiw rhEPO 100 U/kg, sc tiw rhEPO 150–300 U/kg, sc tiw rhEPO 2,000–5,000 U (total), sc daily rhEPO 2,000–5,000 U (total), sc daily rhEPO 150–300 U/kg, sc tiw rhEPO 150 U/kg, sc tiw rhEPO 150 U/kg, sc tiw rhEPO 10,000–20,000 U (total), sc tiw rhEPO 2,000/10,000 U (total), sc tiw rhEPO 5,000 U (total), sc daily rhEPO 40,000–60,000 U (total), sc QW rhEPO 150–300 U/kg, sc tiw rhEPO 150–300 U/kg, sc tiw

Dose of rhEPO or DE

Hb, QOL Hb, transfusions, QOL

Hb, transfusions Hb, transfusions, QOL

Hb, QOL

Hb, transfusions, QOL Hb, transfusions, QOL Hb, transfusions, QOL Hb, transfusions, QOL

Hb, transfusions

Hb, transfusions Hb Hc, transfusions, QOL Hc, transfusions, QOL Hb Hb, transfusions, QOL Hb, transfusions

Hb Hb

Hb

Hb

Treatment targets

63 66

38 68

30/73

53 63 70 64

48

82 11/68/84 56 58 75/85 52 62

75 58/79

45/70

24/85

Quirt et al. 2001 Littlewood et al. 2001

Oberhoff et al. 1998 Gabrilove et al. 2001

Glimelius et al. 1998

Glaspy et al. 1997 Pawlicki et al. 1997 Pawlicki et al. 1997 Demetri et al. 1998

Kasper et al. 1997

Cascinu et al. 1994 Cascinu et al. 1994 Henry and Abels 1994 Henry and Abels 1994 Cascinu et al. 1995 Ludwig et al. 1995 Voigtmann et al. 1996

Cascinu et al. 1993 Tsukuda et al. 1993

Miller et al. 1992

Platanias et al. 1991

% Hb response Reference

Table 4. Dose-response relation and proportion of responses of chemotherapy-induced anemia to epoetin alfa or beta (rhEPO) or darbepoetin (DE) in patients with solid tumors

rhEPO in anemic patients with solid tumors and chemotherapy 455

175

255 166

231 299

69

40 365 2,880

13/36/35/59/29/30/14

156

32/17/46/28/35/40

1,493 352 353 ?

ST, Ch

ST,ChP ST/HM, Ch/ChP

ST, Ch** ST, Ch

ST/HM, Ch/ChP

ST, Ch/ChP ST, Ch/ChP ST/HM, Ch/ChP

ST, Ch?

ST, ChP

ST, Ch?

ST/HM, Ch/ChP ST/HM, Ch/ChP ST/HM, Ch/ChP ST/HM, Ch/ChP

DE 4.5, 6.75, 9, 12, 13.5, 15 μg/kg, sc Q3W DE 300–500 μg (total), sc Q3W DE 2.25 μg/kg, sc QW DE 500 μg/kg (total), sc Q3W DE 500 μg/kg (total), sc Q3W

DE 0.5, 1.5, 2.25, 4.5, 6, 8 μg/kg, sc QW DE 2.25–4.5 μg/kg, sc QW

rhEPO 40,000–60,000 U (total), sc QW rhEPO 150–300 U/kg, sc tiw rhEPO 40,000–60,000 U (total), sc QW rhEPO 30,000 U (total), sc QW rhEPO 40,000–60,000 U (total), sc QW rhEPO 9,000/18,000/36,000 U (total), sc QW rhEPO 30,000 U (total), sc QW rhEPO 30,000 U (total), sc QW rhEPO 30,000 U (total), sc QW

Dose of rhEPO or DE

Hb, transfusions, QOL Hb, transfusions Hb, transfusions Hb, transfusions

Hb, transfusions, QOL

Hb, transfusions, QOL

Hb, QoL

Hb, transfusions Hb, transfusions Hb, transfusions, QOL

Hb, transfusions, QOL

Hb, survival Hb, transfusions, QOL

Hb, QOL Hb, transfusions, QOL

Hb, transfusions, QOL

Treatment targets

79 84 77 54

62 at 12 μg/kg

66

76 at 4.5 μg/kg

48 61 55

41/67/78

68 55

62 72.2

52

Boccia et al. 2006 Canon et al. 2006 Canon et al. 2006 Verhulst et al. 2007

Vansteenkiste et al. 2002 Kotasek et al. 2003

Pirker et al. 2006 Spaeth et al. 2007 Ray-Coquard et al. 2007 Glaspy et al. 2002

Morishima et al. 2006

Aapro et al. 2006 Chu et al. 2006

Ordonez et al. 2005 Witzig et al. 2005

Chang et al. 2005

% Hb response Reference

* deﬁned as a signiﬁcant increase in Hb or Hc with or without transfusion decrease or independency; ST = solid tumors; HM = hematological malignancies; ChP = platinum-based chemotherapy; Ch = non-platinum-based chemotherapy; ? = no details available; QOL = quality of life; ** = published as abstract; tiw = 3 times a week; ﬁw = 5 times a week; QW = once a week; Q3W = once every 3 weeks.

No. of patients

Cancer, chemotherapy

Table 4. Continued

456 M. R. Nowrousian

rhEPO in anemic patients with solid tumors and chemotherapy

457

response rate was 45% compared to 70% when an rhEPO dose of 100 or 200 U/kg sc ﬁw was applied (Miller et al. 1992). In another study using rhEPO dosages of 75, 100, and 150 U/kg sc tiw, the response rates were 11%, 68%, and 84%, respectively (Cascinu et al. 1994a). In two studies in which rhEPO was given intravenously (IV), dosages of 25 to 100 U/kg ﬁw or 200 and 300 U/kg ﬁw produced response rates of 24% and 85% and dosages of 3,000 U tiw or 6,000 U tiw response rates of 58% and 79%, respectively (Platanias et al. 1991; Tsukuda et al. 1993). In all these studies, small groups of patients were treated and the only criterion for response was an increase in Hb (1 g/dl or more from baseline or an increase above 10 g/dl). The results, however, clearly show a close relation between the dose of rhEPO and the response of anemia. Such a relation was also observed in a study including a larger group of patients treated with rhEPO dosages of either 2,000 U/kg or 10,000 U/kg (approximately 150 U/kg) sc tiw (Glimelius et al. 1998). The response rates achieved were 30% and 73%, respectively. A similar doseresponse relationship was also observed in anemic patients with multiple myeloma or non-Hodgkin’s lymphoma who received sc dosages of 2,000 U, 5,000 U or 10,000 U of rhEPO daily. In this study, the response rates were 31%, 61% and 62%, respectively, indicating that dosages >5,000 U sc daily (almost comparable with 10,000 U sc tiw) did not further improve efﬁcacy (Cazzola et al. 1995). Based on the results of these studies, the most appropriate dose schedule of rhEPO for the start of treatment both in patients with solid tumors and in those with hematological malignancies, appears to be 150 U/kg tiw or a ﬁxed dose of 10,000 U sc tiw. An equally effective dose schedule, which is more comfortable for patients, is the use of a total dose of 30,000 U sc once a week. The efﬁcacy of ﬁxed weekly dosages of rhEPO has been proven both in randomized and nonrandomized studies in anemic patients with solid tumors or hematological malignancies receiving dosages of 30,000 or 40,000Usc once a week (Table 4) (Gabrilove et al. 2001; Cazzola et al. 2003; Chang et al. 2005; Witzig et al. 2005; Aapro et al. 2006a,b; Chu et al. 2006; Morishima et al. 2006; Pirker et al. 2006; Razzouk et al. 2006; Abdelrazik and Fouda 2007; Voravud et al. 2007). rhEPO may also be effectively applied in greater dosing intervals or more intensiﬁed induction schedules, but such approaches are still investigational and have not yet been approved for clinical practice (Argianello et al. 2004; Santini et al. 2005; Henry et al. 2006; Steensma et al. 2006). rhEPO should preferably be administered subcutaneously, since the sc route of application can reduce the dosage required by up to 52%, as indicated in patients with renal anemia (41) (Zachee 1995; McClellan et al. 2001). In addition, it is more convenient to patients than the IV route. rhEPO dosages of 100–150 U/kg or a ﬁxed dose of 10,000 U sc tiw or a total dose of 30,000 or 40,000 U sc once a week were used in various studies to treat chemotherapy-induced anemia (Abels 1993; Cascinu et al. 1994a,b, 1995; Henry and Abels 1994; Ludwig et al. 1995; Sevelda et al. 1996; Glaspy et al. 1997; Pawlicki et al. 1997; Demetri et al. 1998; Glimelius et al. 1998;

458

M. R. Nowrousian

Oberhoff et al. 1998; Gabrilove et al. 2001; Littlewood et al. 2001; Chang et al. 2005; Witzig et al. 2005; Aapro et al. 2006a,b). In these studies, either patients with platinum-based or non-platinum-based chemotherapy or both were evaluated and the response was frequently deﬁned as an increase in Hb ≥2 g/dl or hematocrit ≥6% from baseline and independence from transfusions or decrease in transfusion requirement. The response rates observed ranged between 52 and 73% (Table 4). The variation in the response rate may have been related in part to the heterogeneity of patient groups treated (type of malignancy, disease stage, intensity of chemotherapy, etc.) and in part to various dosages and lengths of treatment with rhEPO. The results, however, show no marked difference between the response rates of anemia to rhEPO in patients receiving platinum-based or non-platinum-based chemotherapy. In a large study, however, there was a more rapid response of anemia to rhEPO in patients with platinum-based chemotherapy than in those with other types of chemotherapy, but the overall response rate was similar in the two groups (Abels 1993; Henry and Abels 1994). In another study, the reduction in transfusion requirement achieved by rhEPO appeared to be more pronounced in patients with platinum-based chemotherapy (Oberhoff et al. 1998). However, in a combined analysis of data from two large communitybased clinical trials, treatment with rhEPO was found to be similarly effective in signiﬁcantly reducing transfusion requirements, increasing Hb level, and improving QOL, regardless of whether they received platinum- or nonplatinum-based chemotherapy (Glaspy et al. 2002). The dose-response relationship of darbepoetin alfa in patients with solid tumors and chemotherapy-induced anemia was evaluated in a dose- and schedule-ﬁnding study using various sc dosages of 0.5–4.5 μg/kg weekly, 6.75 μg/kg every 3 weeks, and 6.75 μg/kg or 10.0 μg/kg every 4 weeks and in two phase II trials using dosages ranging from 0.5–8.0 μg/kg once a week and 4.5–15 μg/kg once every 3 weeks, respectively (Glaspy et al. 2002; Kotasek et al. 2003; Smith et al. 2003). The results of these studies indicated that the drug, depending on the dosages used, could be effectively given in various time intervals up to 4 weeks and that at weekly dosing intervals, there was an increasing response rate with increasing doses up to 4.5 μg/kg sc resulting in an Hb response in 76% of patients at this dose level (Glaspy et al. 2002). Higher dosages did not further increase efﬁcacy. At dosing intervals every 3 weeks, Hb response increased with increasing dose up to 12 μg/kg sc reaching a level of 62%. No further increase was observed with dosages >12 μg/kg sc. In a phase III study of patients with lung cancer receiving chemotherapy, a weekly dose of 2.25 μg/kg sc resulted in a hematological response of 66%. In this study, the dose was doubled after 6 weeks if there was no increase in Hb level of ≥1 g/dl from baseline (Vansteenkiste et al. 2002). In another study including both patients with solid tumors and hematological malignancies receiving chemotherapy, a ﬁxed total dose of 500 μg sc once every 3 weeks appeared to be as effective as a weekly dose of 2.25 μg/kg sc (Canon et al.

rhEPO in anemic patients with solid tumors and chemotherapy

459

2006). In some studies, a ﬁxed total dose of 300 μg sc or 325 μg sc once every 2 or 3 weeks, in one study synchronized with chemotherapy, has also been observed to be effective as induction and/or maintenance dose for treating chemotherapy-induced anemia (Hesketh et al. 2004; Boccia et al. 2006; Charu et al. 2007; Rearden et al. 2007). In a trial using synchronous and asynchronous dosing of darbepoetin alfa every 3 weeks, the drug showed the same effect regardless of timing of administration with respect to chemotherapy (Glaspy et al. 2005). A number of studies has also shown signiﬁcant improvements in functional capacity and QOL of patients treated with rhEPO or darbepoetin alfa (Henry and Abels 1994; Ludwig et al. 1995; Glaspy et al. 1997; Pawlicki et al. 1997; Demetri et al. 1998; Glimelius et al. 1998; Gabrilove et al. 2001; Littlewood et al. 2001; Glaspy et al. 2002; Vansteenkiste et al. 2002; Kotasek et al. 2003; Chang et al. 2005; Witzig et al. 2005; Chu et al. 2006; Morishima et al. 2006; Butt and Cella, Chapter 14 in this book; Chowdhury et al., Chapter 29 in this book; Lyman and Glaspy, Chapter 19 in this book; Nowrousian, Chapter 13 in this book). In addition, these effects have been found to signiﬁcantly correlate with incremental changes in Hb level (Glaspy et al. 1997; Demetri et al. 1998; Gabrilove et al. 2001), most evident with an increase in Hb level from 11 to 12 g/dl (range 11 to 13 g/dl) (Crawford et al. 2002). Beside the clinical ﬁndings mentioned above, there are experimental results indicating that treatment with ESAs may also have a protective effect against certain toxicities of chemotherapy. In cellular models and animal experiments, the use of ESAs has been observed to protect against cardiotoxicity of doxorubicin, neurotoxicity of vincristine and platinum and renal toxicity of the latter (Orhan et al. 2004; Vesey et al. 2004; Kim and Backx 2005; Ping and Murat 2005; Sharples et al. 2005; Bianchi et al. 2006, 2007). The clinical relevance of these ﬁndings, however, remains to be evaluated.

Randomized studies Treatment of anemia The efﬁcacy of rhEPO and darbepoetin alfa in the treatment of chemotherapyinduced anemia were evaluated in a number of prospective, randomized trials (James et al. 1992; Abels 1993; Case et al. 1993; Cascinu et al. 1994a,b; Henry and Abels 1994; Porter et al. 1996; Sevelda et al. 1996; Oberhoff et al. 1998; Littlewood et al. 2001; Vansteenkiste et al. 2002; Kotasek et al. 2003; Chang et al. 2005; Witzig et al. 2005; Aapro et al. 2006a,b; Canon et al. 2006) (Table 5). In these trials, patients with various types of malignant diseases and various types of chemotherapy, either platinum-based or non-platinumbased, were treated. The results consistently indicate that treatment with rhEPO or darbepoetin alfa signiﬁcantly increases Hb level and reduces the

21 153 125 153 100 20 189 192 354

344

231

314

198

386 705

Ovarian, ChP ST/HM, Ch ST/HM, ChP ST/HM, ? ST, ChP Sarcoma, Ch ST, Ch/ChP ST, Ch ST, Ch

ST/HM, Ch/ChP

ST, Ch**

ST, ChP

ST, Ch?

ST/HM, Ch? ST, Ch/ChP

Hb, survival Hb, transfusions, QOL Hb, transfusions, QOL

Hb, transfusions Hb, transfusions, QOL

DE 2.25–4.5 μg/kg, sc QW DE 4.5–15 μg/kg, sc Q3W

DE 300–500 μg (total), sc Q3W DE 2.25 μg/kg QW vs 500 μg (total),sc Q3W

Hb, transfusions, QOL

Hb, transfusions Hc, transfusions, QOL Hc, transfusions, QOL Hc, transfusions, QOL Hb, transfusions Transfusions Transfusions Hb, transfusions, OOL Hb, transfusions, QOL

Treatment targets

rhEPO 30,000 U (total), sc QW

rhEPO 300 U/kg, sc tiw rhEPO 150 U/kg, sc tiw rhEPO 150 U/kg, sc tiw rhEPO 150 U/kg, sc tiw rhEPO 100 U/kg, sc tiw rhEPO 150 U/kg, sc tiw rhEPO 5,000 U (total), sc daily rhEPO 150–300 U/kg, sc tiw rhEPO 40,000–60,000 U (total), sc QW rhEPO 40,000–60,000 U (total), sc QW

Dose of rhEPO or DE

IS, DS, IS depending on DE dose and Hb increase IS, DS Comparable efﬁcacy of the two schedules

IS, DS, IS

IS, DS, IS depending on Hb increase IS, comparable

IS, D IS, DS, IS IS, DS, IS IS, D, IS IS, DS DS DS IS, DS, IS IS, DS, IS

Results*

Taylor et al. 2005 Canon et al. 2006

Vansteenkiste et al. 2002 Kotasek et al. 2003

Aapro et al. 2006

Witzig et al. 2005

James et al. 1992 Abels 1993 Abels 1993 Case et al. 1993 Cascinu et al. 1994b Porter et al. 1996 Oberhoff et al. 1998 Littlewood et al. 2001 Chang et al. 2005

References

ChP = platinum-based chemotherapy; Ch = non-platinum-based chemotherapy; * = in the rhEPO- or DE-treated group compared to that in the control group; ST = solid tumors; HM = hematological malignancies; IS = increased signiﬁcantly; DS = decreased signiﬁcantly; ? = no details available; ** = published as abstract, QOL = quality of life.

No. of pts

Cancer, chemotherapy

Table 5. Randomized studies using rhEPO or darbepoetin (DE) in patients with solid tumors and chemotherapy-induced anemia

460 M. R. Nowrousian

rhEPO in anemic patients with solid tumors and chemotherapy

461

need for RBC transfusions. The risk of requiring RBC transfusions decreases by approximately 50%. In addition, the results indicate that these agents are effective both in patients with platinum-based and non-platinum-based chemotherapy. Furthermore, they show that treatment of anemia with these drugs signiﬁcantly improves functional capacity and QOL of patients. In one study using epoetin alfa, a trend to improved survival was also observed, although the study was primarily not powered for evaluating survival as an end point. The estimated hazard ratio was 1.309 (p = 0.052) in favor of treating anemia with rhEPO, suggesting that the risk of dying was approximately 31% higher for patients who were treated with placebo. When examined by tumor type, a similar pattern favoring treatment with rhEPO was observed both in patients with solid tumors and hematological malignancies (Littlewood et al. 2001). In another study including anemic patients with lung cancer receiving chemotherapy, treatment with darbepoetin alfa was associated with a trend towards improved progression-free survival, particularly in patients with a baseline Hb level ≥10 g/dl (Vansteenkiste et al. 2002, 2004).

Prevention of anemia, early intervention The efﬁcacy of rhEPO in preventing chemotherapy-induced anemia in patients with solid tumors was evaluated in several prospective, randomized studies, mainly in patients with platinum-based chemotherapy (Table 6) (Gamucci et al. 1993; de Campos et al. 1995; Wurnig et al. 1996; Del Mastro et al. 1997; ten Bokkel Huinink 1998; Dunphy et al. 1999; Thatcher et al. 1999; Bamias et al. 2003; Grote et al. 2005; Moebus et al. 2007). In three studies, however, patients with breast cancer and non-platinum-based chemotherapy were included (Del Mastro et al. 1997; Leyland-Jones et al. 2005; Moebus et al. 2007). In all studies, patients who received rhEPO showed a signiﬁcantly higher Hb level or a signiﬁcantly reduced RBC transfusion requirement or both at the end of treatment compared with patients not receiving rhEPO. These results indicate that the concomitant use of rhEPO and chemotherapy can prevent or reduce the development of anemia and the need for RBC transfusions in patients with solid tumors, and that this effect can be achieved both in patients with platinum-based and non-platinumbased chemotherapy. In a randomized study in weight-losing cancer patients not receiving chemotherapy, treatment with rhEPO was found to prevent the development of anemia and to protect from decrease in metabolic function with further loss of weight and decrease in exercise capacity (Daneryd et al. 1998). In this study, there was no impact of treatment with rhEPO on survival. A similar result was also observed in a randomized study of treatment with or without rhEPO of mainly nonanemic patients with small-cell lung cancer receiving chemotherapy (Grote et al. 2005). rhEPO and placebotreated patients had comparable median overall survival and overall

120

30 130 140 939

224

1,284

Ovarian, ChP

ST, ChP SCLC, ChP ST, ChP Breast, Ch

SCLC, ChP

Breast, Ch

rhEPO 150 U/kg tiw

rhEPO 150 U/kg tiw

rhEPO 150–450 U/kg tiw rhEPO 150–300 U/kg tiw rhEPO 10,000 U (total) tiw rhEPO 40,000 U (total) QW

rhEPO 150 U/kg tiw rhEPO 150–300 U/kg tiw rhEPO 600 U/kg iv, 2x/week rhEPO 150 U/kg tiw rhEPO 12,000–30,000 U (total) QW rhEPO 150–300 U/kg tiw

Dose sc

Transfusions Hb, transfusions Hb, transfusions Hb, transfusions, progression-free survival, survival Hb, transfusions, survival and overall mortality Hb, transfusions, disease-free survival and survival

Hb Hb, transfusions Transfusions Hb, transfusions Hb, metabolic and exercise capacity Transfusions

Response criteria

R SH, SR SH, SR SH, BS, comparable, SR SH, SR, comparable, comparable SH, SR, comparable, comparable

SR

SH SH, SR SR SH, R SH

Results*

Moebus et al. 2007

Grote et al. 2005

ten Bokkel Huinink et al. 1998 Dunphy et al. 1999 Thatcher et al. 1999 Bamias et al. 2003 Leyland-Jones et al. 2005

Gamucci et al. 1993 de Campos et al. 1995 Wurnig et al. 1996 Del Mastro et al. 1997 Daneryd et al. 1998

Reference

ChP = platinum-based chemotherapy; Ch = chemotherapy without cisplatin or carboplatin; * = in the rhEPO-treated group compared to that in the control group; SH = signiﬁcantly higher; SR = signiﬁcantly reduced; BS = reduced with borderline signiﬁcance; ST = solid tumors; SCLC = small-cell lung cancer; tiw = three times per week; QW = once a week; **mainly gastrointestinal tumors without chemotherapy.

38 36 29 62 108

No. of pts

ST, ChP SCLC, ChP Bone tumors, ChP Breast, Ch ST**

Cancer, chemotherapy

Table 6. Randomized studies using rhEPO for preventing chemotherapy-induced anemia in patients with solid tumors

462 M. R. Nowrousian

rhEPO in anemic patients with solid tumors and chemotherapy

463

mortality. In a recent study of mainly nonanemic breast cancer patients receiving chemotherapy, however, the use of rhEPO was observed to be associated with a reduced probability of survival (Leyland-Jones et al. 2005). Negative results have also been observed in two studies of patients with head and neck cancer who received radiotherapy at relatively high baseline and target levels of Hb (Henke et al. 2003, 2006; Overgaard et al. 2007). Although the results of these studies have been critically reviewed because of a number of methodological problems (Vaupel et al. 2005, 2006; Crawford 2007), the use of rhEPO or darbepoetin alfa in nonanemic cancer patients can not be recommended (Bokemeyer et al. 2004, 2007; Rizzo et al. 2007; EORTC Guidelines 2007). An efﬁcient and safe alternative may be an early intervention in terms of starting treatment at Hb levels between >10 and <12 g/dl (Savonije et al. 2005, 2006; Lyman and Glaspy 2006; Straus et al. 2006; Crawford et al. 2007). Such an early intervention is the subject of a separate chapter in this book (Lyman and Glaspy, Chapter 19).

Patient selection, prediction of response Epoetin alfa, epoetin beta and darbepoetin alfa are effective and well-tolerated drugs in treating chemotherapy-induced anemia in cancer patients, but treatment with these drugs is expensive and the response rate is variable, ranging from 50 to 75%. The mechanisms involved in the development of resistance to ESAs are still not clearly deﬁned. Possible factors are nutritional impairments including folate or vitamin B12 deﬁciency, hemolysis, blood loss, infection and, particularly, excessive release of inﬂammatory cytokines, such as tumor necrosis factor, interleukin (IL) 1, IL-6 and interferon gamma associated with cancer and functional iron deﬁciency during treatment. In anemia of end-stage renal disease, inﬂammatory conditions and functional iron deﬁciency are the most relevant factors for resistance of anemia to ESAs (Macdougall and Cooper 2002; Stenvinkel 2003; Macdougall 2004; Johnson et al. 2007). As will be discussed below, functional iron deﬁciency may also play a major role in erythropoietin resistance of anemia in cancer. In cancer patients, a number of clinical and laboratory parameters was evaluated to predict the development of anemia, the need for RBC transfusions during chemotherapy and the response of anemia to ESAs. Such factors may help to improve the results and to increase the cost-effectiveness of treatment by identifying patients who are likely to beneﬁt from treatment and patients who have a high probability of response of anemia to ESAs. Patient selection In several studies, various pretreatment parameters, such as advanced age, reduced performance status, loss of body weight, low baseline Hb level, low

464

M. R. Nowrousian

lymphocyte count, advanced stage of malignant disease and platinum-based chemotherapy, have been reported to predict the development of anemia and the need for red blood cell (RBC) transfusion. Among these factors, however, the only one which has consistently been found to be of predictive value is a low baseline Hb level or a rapid decrease in Hb after the ﬁrst or second cycle of chemotherapy (Skillings et al. 1993; Abels et al. 1994; Thatcher 1998; Carabantes et al. 1999; Ray-Coquard et al. 1999; Hensley et al. 2001; BarrettLee et al. 2006; Ludwig, Chapter 7 in this book). In a large prospective survey of 2,070 nonanemic cancer patients who received their ﬁrst chemotherapy during the survey and underwent at least two cycles of chemotherapy, initial Hb levels of ≤12.9 g/dl in women and ≤13.4 g/dl in men appeared to predict anemia during treatment. Other risk factors were the type of malignant disease, particularly lung or gynecologic cancer, platinum-based chemotherapy and female gender. Based on these parameters, a risk model for identifying patients who were at the highest risk of developing anemia during chemotherapy was developed (Barrett-Lee et al. 2006; Ludwig, Chapter 7 in this book). This and other risk models proposed may be useful to identify patients who should be monitored more carefully during chemotherapy for the development of anemia and for being treated with ESAs before debilitating or even more severe symptoms of anemia and the need for RBC transfusions occur. Prior transfusions in cancer patients receiving chemotherapy have been found to increase the risk of subsequent transfusions with or without treatment with ESAs (Couture et al. 2005; Quirt et al. 2006). An important aspect of treating chemotherapy-induced anemia with ESAs is the threshold of Hb level at which treatment should be started. The most relevant factor determining this threshold is the beneﬁt of treatment related to the degree of anemia and its symptoms. Symptoms of anemia, however, may considerably vary from patient to patient for the same degree of anemia depending on the age of patients and functional capacity of compensatory mechanisms such as those of the cardiovascular and pulmonary system. In cancer patients, even mild to moderate anemia has been shown to be associated with signiﬁcantly reduced metabolic and exercise capacity, cognitive dysfunctions, fatigue, and decreased QOL (Daneryd et al. 1998; Jacobson et al. 2004; Crawford et al. 2002; Butt and Cella, Chapter 14 in this book; Chowdhury et al., Chapter 29 in this book). In a retrospective analysis of data from 4,382 anemic cancer patients receiving chemotherapy, a signiﬁcant but not linear relationship was found between Hb increase in a range from 8 to 14 g/dl and improvements in QOL following treatment with rhEPO. Of particular interest is the observation that there was not only an improvement in QOL with increasing Hb level, but also an incremental beneﬁt in QOL with each 1-g/dl Hb increase, with the highest incremental beneﬁt occurring with an Hb increase from 11 to 12 g/dl. Beyond an Hb level of 12 g/dl, subsequent increases in Hb continued to yield additional gains in QOL, but at a

rhEPO in anemic patients with solid tumors and chemotherapy

465

Anemia exclude causes* other than cancer and chemotherapy

ESA

after 4 weeks Hb increase < 1 g/dl

IV iron if there are signs of functional or real iron deficiency

after 4 weeks

Hb increase ≥ 1 g/dl

Hb increase < 1 g/dl

Continue treatment and adapt the dose to maintain Hb at 12 g/dl

Stop treatment

Fig. 1. Treatment algorithm using increase in hemoglobin (Hb) level after 4–8 weeks of treatment as predictive factor of response to erythropoiesis-stimulating agents (ESAs) in anemic cancer patients receiving chemotherapy. *bleeding, hemolysis, nutritional defects such as vitamin B12 and folic acid deﬁciency

decreasing rate (Crawford et al. 2002). Hb levels have also been shown to have a great impact on the outcome of anticancer treatment. Both Hb levels of 11 g/dl or below and Hb levels of 13 g/dl or above have been found to be associated with decreased tumor oxygenation and, consequently, reduced sensitivity of tumor cells to radiotherapy and chemotherapy (Vaupel et al. 2005, 2006). Based on the results of these studies, an appropriate Hb level for initiating treatment with ESAs appears to be ≤11 g/dl and an appropriate Hb level as target 12 g/dl (Nowrousian, Chapter 13 in this book) (Fig. 1). Such thresholds of Hb are also subjects of recommendations for the use of ESAs in chemotherapy-induced anemia by the European Organisation for Research and Treatment of Cancer (EORTC) and the National Comprehensive Cancer Network (NCCN) (Table 7) (Bokemeyer et al. 2007; NCCN 2007).

12 g/dl Increase in Hb, decrease in RBC transfusions Discontinue after 6–8 weeks, if no response

Target Hb Treatment targets

9–11 g/dl based on anemia-related symptoms; <11.9 g/dl to prevent further Hb decline and according to individual factors: baseline Hb, type, intensity and duration of further planned treatment 12 g/dl Decrease in RBC transfusions, improvements of quality of life Discontinue after 4–8 weeks if no response; IV iron in case of absolute or functional iron deﬁciency

EORTC3)

* Limited cardiopulmonary reserve, underlying coronary artery disease or symptomatic angina, or substantially reduced exercise capacity, energy, or ability to carry out activities of daily living. Hb = hemoglobin; RBC = red blood cell. 1) Rizzo et al. 2007; 2) NCCN Guidelines 2007; 3) EORTC Guidelines: The 2007 position.

Cessation of treatment

≤11 g/dl anemia-related symptoms or risk factors

≤10 g/dl; >10–<12 g/dl, declining Hb and certain clinical circumstances*

Hb level for initiating treatment

12 g/dl Maintaining Hb level at 11–12 g/dl Discontinue if no response after 8–12 weeks, even after dose escalation; iron supplementation if ferritin <100 μg/l, transferrin saturation <20%

NCCN2)

ASCO/ASH1)

Recommendations

Table 7. American Society of Clinical Oncology, American Society of Hematology (ASCO/ASH), National Comprehensive Cancer Network (NCCN), and European Organisation for Research and Treatment of Cancer (EORTC) Guidelines for treating chemotherapy-induced anemia with erythropoiesis-stimulating agents

466 M. R. Nowrousian

rhEPO in anemic patients with solid tumors and chemotherapy

467

Prediction of response Baseline parameters In several studies, various parameters were tested for their predictive value for response of cancer-related and chemotherapy-induced anemia to rhEPO (Ludwig et al. 1990, 1993, 1994; Platanias et al. 1991; Cascinu et al. 1994b; Cazzola et al. 1995; Henry et al. 1995; Cazzola et al. 1996; Österborg et al. 1996; Voigtmann et al. 1996). In one of these studies including both patients with solid tumors and patients with hematological malignancies, a large number of variables, including baseline EPO level and inﬂammatory markers, such as interleukin (IL)-1β, tumor necrosis factor (TNF)-α, TNF-β, IL-6 and interferon-γ was analyzed at the start of treatment, and none of these parameters was found to have sufﬁcient prognostic power (Ludwig et al. 1994). Baseline EPO level was also tested in some studies exclusively evaluating patients with solid tumors, and the results of these studies generally showed no correlation or only a poor correlation between this parameter and the response of anemia to rhEPO (Platanias et al. 1991; Henry et al. 1995; Voigtmann et al. 1996; Glimelius et al. 1998). In studies including patients receiving chemotherapy, the failure of baseline EPO level to predict response to rhEPO might have been related to changes in serum EPO concentrations that occur during treatment (Birgegard et al. 1989; Schapira et al. 1990; Hasegawa and Tanaka 1992; Sawabe et al. 1996). In a large study of patients not receiving chemotherapy, however, this parameter was also found to correlate poorly with response to rhEPO (Henry et al. 1995). By contrast to the studies of patients with solid tumors, those exclusively including patients with hematological malignancies showed an inverse correlation between baseline EPO level or the calculated ratio of baseline EPO level to predicted level of EPO for the same degree of anemia (O/P ratio) and the response of anemia to rhEPO. Baseline EPO levels <100 mU/ml or O/P ratios <0.9 were found to predict signiﬁcantly higher response rates (Cazzola et al. 1995; Österborg et al. 1996; Voigtmann et al. 1996; Beguin and Van Straelen, Chapter 21 in this book).

Early predictors In some studies, various parameters were tested for their predictive value as early indicators of response of anemia to rhEPO (Ludwig et al. 1993, 1994; Henry et al. 1995; Cazzola et al. 1996). In one of them, serum EPO level, change in Hb, and serum ferritin level after 2 weeks of treatment were found to be of predictive value, but the best predictor appeared to be a combination of an EPO level of <100 mU/ml and an increase in Hb level of ≥0.5 g/dl, predicting a response rate of 95% (Ludwig et al. 1994). A serum ferritin level

468

M. R. Nowrousian

of <400 ng/ml alternatively predicted response with a probability of 72%. In another study, a combination of an Hb increase of ≥0.5 g/dl and an increase in reticulocyte count of ≥40,000 cells/μl, both after 2 weeks of treatment, predicted a response rate of 91%, if the patients were not receiving chemotherapy (Henry et al. 1995). In patients receiving chemotherapy, the same parameters were of prognostic value not after 2 but after 4 weeks of treatment, predicting a response rate of 84%. In a community-based study including a large number of patients receiving chemotherapy, 75% of responders were found to have had an increase in Hb level of ≥1 g/dl from baseline to week 4 of treatment (Glaspy et al. 1997). In a study using a combination of baseline EPO level and change in soluble transferrin receptor (sTFR) level after 2 weeks of treatment, a response rate of 96% was found in patients with a baseline EPO level of <100 mU/ml and a 2-week increment of sTFR level of >25% (Cazzola et al. 1996). In a recent study of anemic (Hb < 11 g/dl) cancer patients receiving chemotherapy and darbepoetin alfa, multivariate analysis of endogenous serum EPO (<100 mU/ml) and O/P ratio of EPO (<0.9) at baseline, increase in sTFR in serum (>25%) after 2 weeks, increase in reticulocyte count (>40,000/ml) after 2 and 4 weeks, and increase in Hb (>1 g/dl) after 4 weeks identiﬁed the combination of sTFR after 2 weeks and Hb after 4 weeks as predictive as the combination of all ﬁve factors tested (Steinmetz et al. 2007). The single most predictive factor for response to ESAs, however, was the increase in Hb (>1 g/dl) after 4 weeks. Patients with such an increase in Hb had an 11-fold higher chance of being responders than those without. In another recent study including anemic (Hb < 10.5 g/dl) patients with multiple myeloma and malignant lymphomas treated with rhEPO (30,000 U sc weekly for 6 weeks), the proportion of baseline hypochromic red blood cells (<5%) either alone or in combination with an increase in absolute reticulocyte count (>50,000/ml) or RETICS-Hct (absolute number of reticulocytes × MCV of reticulocytes) between baseline and week 2 of treatment was found to be a reliable parameter for predicting response to ESAs with positive predictive values of 79%, 95%, and 92%, and speciﬁcities of 50%, 93%, and 86%, respectively (Katodritou et al. 2007). Most of the studies mentioned above were not carried out exclusively either in patients with solid tumors or hematological malignancies, and some of them also included patients with myelodysplastic syndromes. In addition, there are many differences between these studies regarding the usage and intensity of chemotherapy, dosage and length of treatment with rhEPO, and deﬁnition of response. Furthermore, some of the parameters used may be considerably inﬂuenced or affected in their predictive value by therapeutic factors such as RBC transfusions or chemotherapy. For instance, an increase in Hb can only be used as predictive factor, when the patients do not require RBC transfusions during treatment. Reticulocyte count and serum EPO level may be less reliable in patients receiving chemotherapy, since they undergo signiﬁcant changes during treatment (Birgegard et al. 1989; Schapira et al.

rhEPO in anemic patients with solid tumors and chemotherapy

469

1990; Hasegawa and Tanaka 1992; Sawabe et al. 1996). In addition, an increase in reticulocyte count after application of rhEPO may be caused by an increased release of these cells from the bone marrow into the blood rather than a real increase in erythropoiesis (Major et al. 1994). Furthermore, an increase in sTFR may indicate an increased erythropoiesis, but it may also indicate iron deﬁciency or both (Beguin and Van Straelen, Chapter 21 in this book). Irrespective of these uncertainties, the various algorithms proposed in these studies have not yet been validated prospectively in sufﬁciently large clinical trials. In addition, in a retrospective analysis of data pooled from four randomized trials of 604 patients, none of these algorithms was found to approach the 80–90% levels of sensitivity and speciﬁcity that are regarded as appropriate for clinically useful predictive tests (Littlewood et al. 2003). Similarly disappointing results were also found in another retrospective analysis of pooled data of 1,010 anemic cancer patients treated with rhEPO (Anon 2005). Considering these controversies, no reliable factor appears to be available today for predicting response of anemia to ESAs in cancer patients. Baseline EPO level appears to be of predictive value in patients with myelodysplastic syndromes (Hellström-Lindberg, Chapter 20 in this book) and in patients with hematological malignancies (Cazzola et al. 1995; Österborg et al. 1996; Voigtmann et al. 1996), but not in patients with solid tumors (Platanias et al. 1991; Henry et al. 1995; Voigtmann et al. 1996; Glimelius et al. 1998). Even in patients with hematological malignancies, the low speciﬁcity of baseline EPO level makes its use as a predictive parameter questionable. The low speciﬁcity is also a major disadvantage of all other currently available parameters and the algorithms proposed (Ludwig et al. 1994; Henry et al. 1995; Cazzola et al. 1996). Today, the only parameter generally used to predict response of anemia to rhEPO in cancer patients receiving chemotherapy is an increase in Hb after 4–6 weeks of treatment. Patients with an Hb increase of ≥1 g/dl from baseline have a probability of response of up to 84% (Glaspy et al. 1997; Quirt et al. 2001; NCCN 2007; Bokemeyer et al. 2007; Rizzo et al. 2007; Steinmetz et al. 2007).

Iron supplementation In healthy individuals, there is an average 4 g of total iron in the body, with 3 g in circulating RBC mass and approximately 1 g in reticuloendothelial (RE) stores. The size of the RE storage pool is usually measured by serum ferritin, but the latter is an acute-phase protein that increases independently of iron stores in clinical states associated with inﬂammation, infection or malignancy (Hazard and Drysdale 1977; Konijn and Hershko 1977; Ali et al. 1982; Cook 1982; Adamson et al. 1999; Kaltwasser and Gottschalk 1999; Goodnough et al. 2000; Goodnough 2006, and Chapter 26 in this book). In such conditions, serum ferritin levels may be as high as 25–50 μg/l in spite of

470

M. R. Nowrousian

iron depletion. Serum ferritin levels >100 μg/l, however, are almost invariably associated with some amounts of iron stores and are inconsistent with true iron deﬁciency (Cook 1982; Adamson 1994). On the other hand, for synthesizing every 1 g/dl of Hb, the erythropoietic system requires 150 mg iron from iron stores, corresponding to 20 μg/l ferritin. Thus, patients with a baseline ferritin level of <100 μg/l are highly likely to develop iron deﬁciency during treatment with rhEPO, if an increase in Hb level of 3–5 g/dl should be achieved (Kaltwasser and Gottschalk 1999; Macdougall 1999). Transferrin saturation (TFS), a test that measures the amount of iron immediately available to the erythroid marrow, can also be inﬂuenced by chronic diseases, but persistent levels of TFS <18% are invariably associated with iron-deﬁcient erythropoiesis and levels >50% with iron overload (Adamson et al. 1999). The simplest and most effective parameters to evaluate the adequacy of iron delivery to erythroid tissue are the percentage of hypochromic RBCs or the Hb content of reticulocytes (CHr). Iron-deﬁcient erythropoiesis is present when >10% of RBCs are hypochromic (<28 g/dl) or the mean CHr is <26 pg. The diagnostic sensitivity and speciﬁcity of CHr is reported to be 100% and 80%, respectively, and thus higher than the other parameters mentioned above (Fishbane et al. 1997; Brugnara 1998; Adamson et al. 1999; Cullen et al. 1999; Schaefer and Schaefer 1999). In a study of hemodialysis patients on maintenance rhEPO, however, iron depletion, deﬁned as a response of Hb level to IV iron, appeared to be best detected by the percentage of hypochromic RBCs, used at a threshold of >6% (Tessitore et al. 2001).

Functional and real iron deﬁciency During treatment with ESAs, two types of iron deﬁciency, real or functional, may develop. Real iron deﬁciency may result from chronic administration of ESAs with a progressive shift of iron from body stores to the erythron, leading to an exhaustion of iron in the stores indicated by low levels of ferritin (<100 μg/l). Functional or relative iron deﬁciency may occur when there is a need for a greater amount of iron to support Hb synthesis than can be released from body stores. In this case, body iron stores, as indicated by serum ferritin level, are usually normal or even elevated, but iron supply to erythroid marrow is inadequate, either due to a high stimulation of erythropoiesis through the use of ESAs or due to a limitation of iron release induced by excessive production of immunological or inﬂammatory cytokines, such as tumor necrosis factor α, interleukin-1, interferon γ and the recently identiﬁed iron regulatory protein hepcidin, or a combination of the two processes (Nowrousian et al. 1996 and Chapter 6 in this book; Kaltwasser and Gottschalk 1999; Beguin and Van Straelen, Chapter 21 in this book; Goodnough et al. 2000; Goodnough 2006 and Chapter 26 in this book).

rhEPO in anemic patients with solid tumors and chemotherapy

471

Functional iron deﬁciency is the most common cause of suboptimal response to ESAs in anemic patients with chronic renal failure, and the treatment of choice is IV iron that enhances the response and reduces the dose of ESAs by 19–70% and thus the cost of treatment. Oral iron is, beside its inconvenience for patients, not able to keep pace with the demand on iron during treatment with ESAs (Macdougall et al. 1996; Nissenson 1997; Macdougall 1999; Macdougall et al. 1999; Ahsan 2000; Besarab et al. 2000; Vogel 2000; Hudson and Comstock 2001; Kato et al. 2001; Silverberg et al. 2001; Beguin and Van Straelen, Chapter 21 in this book; Goodnough et al. 2000; Goodnough 2006 and Chapter 26 in this book). Functional iron deﬁciency has also been reported in a high proportion of anemic patients with rheumatoid arthritis (RA) who were treated with rhEPO, and in these patients, IV iron was found to be signiﬁcantly effective in inducing response of anemia to rhEPO and, additionally, it was found to be safe (Arndt et al. 2005). Of particular interest is the observation that the use of rhEPO and IV iron not only corrected anemia, improved physical well-being and reduced fatigue, but also signiﬁcantly improved the RA activity index. Functional iron deﬁciency is also a hallmark of anemia in cancer, and many patients have already signs of restricted availability of iron to the erythron at the beginning of treatment with rhEPO or develop such signs during treatment. In a study of cancer patients with chemotherapy-induced anemia, 10% of patients had baseline TFS levels <20% and 94% of these patients had serum ferritin levels >100 ng/ml. In another study including patients with anemia of lymphoproliferative disease without chemotherapy, 39% of patients had baseline TFS levels <20% despite proven iron deposits in the bone marrow and 77% of these patients had serum ferritin levels >100 μg/l. The proportion of patients with serum TFS levels <20% increased to 87% during treatment with rhEPO, when the patients did not receive IV iron. In addition, in this study, 32% of patients had baseline serum ferritin levels <100 μg/l indicating possible real iron deﬁciency despite proven iron deposits in the bone marrow.

Intravenous iron in cancer patients In cancer patients, some concerns exist regarding the use of IV iron. They include possible stimulation of tumor growth and inhibition of cellular defense mechanisms by weakening phagocytic activity of polymorphonuclear cells and cell-mediated immunity. Another concern is iron overload that could occur as a result of cytokine-mediated abnormal accumulation of iron in the RES, and ﬁnally, there is concern about possible severe side effects of IV iron, such as anaphylactic or anaphylactoid reactions (Weinberg 1996; Burns and Pomposelli 1999; Weinberg 1999; Weiss 1999; Weiss and Gordeuk 2005; Maynor and Brophy 2007).

472

M. R. Nowrousian

Table 8. Changes in variables of iron metabolism from baseline to ﬁnal value in a randomized study evaluating the effect of rhEPO in patients with chemotherapyinduced anemia Control (n = 88)

Iron (μmol/l) Ferritin (μg/l) Transferrin saturation (%)

rhEPO (n = 101)

Baseline

Final value

Baseline

Final value

14.3 264 24.9

14.0 324 23.9

13.2 208 24.9

10.2 148 19.4

rhEPO = recombinant human erythropoietin. Data are shown as median. From Oberhoff et al. 2000.

Although laboratory studies suggest that iron may favor neoplastic cell growth and epidemiological studies show a possible link between increased body iron stores and increased risk of cancer and infection (Weinberg 1996, 1999; Weiss 1999; Weiss and Gordeuk 2005; Maynor and Brophy 2007), the relevance of these ﬁndings for the clinical use of IV iron together with ESAs is questionable (Cavill 2003; Macdougall 2006), particularly in cancer patients who usually need short-term treatments and relatively low total doses of iron. In these patients, serum ferritin and TFS usually decrease during treatment with ESAs, predominantly during the ﬁrst 4–8 weeks, indicating an increased iron mobilization from iron stores and its utilization in the erythroid marrow (Ludwig et al. 1990; Oberhoff et al. 2000; Hedenus et al. 2007; Henry et al. 2007). In addition, the response of erythropoiesis appears to depend on the availability of iron (Heiss et al. 1996; Henry et al. 1998; Henry 1998; Ballard et al. 1999; Auerbach et al. 2004; Lerchenmüller et al. 2006; Vandebroek et al. 2006; Hedenus et al. 2007; Henry et al. 2007). In a study of patients with chemotherapy-induced anemia receiving rhEPO, the median value of serum ferritin level decreased from 208 to 148 μg/l during treatment and the median value of TFS decreased from 25 to 19%, suggesting a condition of iron deﬁciency, either functional or real, in approximately 50% of patients (Table 8) (Oberhoff et al. 2000). A response rate of 38% was achieved. In this study, treatment with rhEPO would have been more effective or would have required lower doses of rhEPO, if the patients had concomitantly received IV iron. This assumption is supported by the results of a study of 148 anemic (baseline Hb ≤ 10.5 g/dl) cancer patients receiving chemotherapy who were randomized to receive, in addition to rhEPO (40,000 U sc, weekly), no iron, oral iron (FeSO4 325 mg bid) or IV iron, with the latter being given either as a total-dose infusion at a single setting or as weekly IV boluses of 100 mg iron dextran to a total dose calculated by the following formula: Dose (ml) =

rhEPO in anemic patients with solid tumors and chemotherapy

473

0.0442 (desired Hb − observed Hb) × lean body weight + (0.26 × lean body weight) (Auerbach et al. 2004). After 6 weeks of treatment, signiﬁcantly better results were obtained in the two IV iron groups than in the groups of patients with oral iron or no iron. The mean increases in Hb level in the four groups of patients were 2.4 g/dl, 2.5 g/dl, 1.5 g/dl, and 0.9 g/dl, respectively, and the rates of hematopoietic responses, deﬁned as an increase in Hb ≥ 2 g/dl from baseline or achievement of an Hb level ≥12 g/dl, were 68%, 68%, 36%, and 25%. Improvements in activity and overall QOL were also signiﬁcantly greater in the IV iron groups than the other two groups. In this study, the differences in Hb response did not appear to be dependent on baseline TFS at a cutoff point of 15%. The latter, however, might have been relatively low for a more precise discrimination between patients with or without iron-restricted erythropoiesis. Adverse events, possibly related to IV iron, occurred in 7–8% of patients and included grade 1 or 2 delayed arthralgia/myalgia syndrome, fatigue and shortness of breath. Hypersensitivity reaction comprising chest/back pain, nausea, vomiting, ﬂushing, and hypotension was observed only in one patient at a test dose of iron dextran and precluded further therapy. This effect, however, subsequently completely resolved. The superiority of IV iron has also been shown in four other prospective randomized trials, two already published as full papers (Hedenus et al. 2007; Henry et al. 2007) and two as abstracts (Lerchenmüller et al. 2006; Vandebroek et al. 2006). One of these trials included 187 patients with chemotherapy-induced anemia (Hb < 11 g/dl; serum ferritin >100 ng/ml or TFS >15%) who were treated with rhEPO (40,000 U sc weekly) and randomized to 8 weeks of IV iron (125 mg ferric gluconate weekly), oral iron (325 mg ferrous sulphate three times daily), or no iron (Henry et al. 2007). Mean increase in Hb and Hb response, deﬁned as an Hb increase of ≥2 g/dl, were 2.4 g/dl and 73%, respectively, in the IV iron group and, thus, signiﬁcantly greater than in the other two groups of patients with mean increases of Hb of 1.6 g/dl and 1.5 g/dl and Hb response rates of 46% and 41%. Among the subgroup of patients with a baseline TFS <20%, those who received rhEPO together with IV iron had a response rate of 81%, which was signiﬁcantly greater than the response rates of 37% in patients with oral iron and 27% in patients with no iron. The corresponding response rates in patients with a baseline TFS ≥20% were 68%, 52%, and 48%. Baseline TFS, however, like baseline ferritin and baseline CHr, did not appear to have a linear relationship to mean Hb change from baseline and to be a signiﬁcant predictor of response. Serum ferritin level showed a signiﬁcantly different and opposed behavior during treatment with rhEPO in patients with IV iron and those with oral iron or no iron. In the ﬁrst group of patients, there was a mean increase in ferritin level of 343.7 ng/ml from baseline to end point, while the other two groups experienced a mean ferritin decrease of 13.9 ng/ml and 95.8 ng/ml, respectively. The highest ferritin levels observed during treatment, however, were 3,586 ng/ml in the IV-iron group, 6,186 ng/ml in the oral-iron group, and 3,830 in the

474

M. R. Nowrousian

no-iron group. Furthermore, no patient who received IV iron was withdrawn from the study as a result of a TFS value increasing to ≥50%. The degree of myelosuppression and the rate of infections, as well as the frequency of adverse effects including asthenia, nausea, vomiting, constipation, diarrhea, and the rate of serious adverse events (SAEs) and survival were comparable in the three groups of patients. In the IV iron group, only one SAE consisting of angina was considered to be possibly related to the study drug. In a randomized trial of 67 patients with anemia (Hb 9–11 g/dl) of lymphoproliferative diseases not requiring chemotherapy and RBC transfusions, the use of rhEPO (30,000 U sc once a week for 16 weeks) was also found to be signiﬁcantly more effective when it was given together with IV iron than without. The mean increases in Hb were 2.91 g/dl and 1.50 g/dl in the two groups of patients, respectively, and the proportions of patients with Hb response, deﬁned as an Hb increase ≥2 g/dl, 93% and 53%. In addition, patients with IV iron needed a median of 6 weeks to achieve an Hb response, compared with 12 weeks in patients with no iron. Furthermore, from week 5 of treatment, there was an increasing difference in mean weekly rhEPO dose in favor of patients with IV iron, reaching signiﬁcance at week 13 and accounting for an average rhEPO dosage approximately 25% lower than in patients with no iron. In both groups of patients, serum ferritin level rapidly decreased at week 1 of treatment but, thereafter, it almost doubled from baseline (128 μg/l) during the study in the IV-iron group, while it continued to decrease to a mean of 112 mg/l (baseline 130 mg/l) in the non-iron group. The mean TFS was approximately 30% (baseline 21%) during treatment period in the ﬁrst group of patients and 20% (baseline 22%) in the second. In addition, in the IV-iron group, all patients with TFS <20% at baseline or during treatment achieved an Hb response, compared with only 54% in the non-iron group. The mean serum levels of sTFR and the mean reticulocyte counts rapidly increased in both groups after 1 week of treatment and remained elevated throughout the study, but without signiﬁcant differences between the two groups. The most commonly observed adverse events in this study were upper respiratory infections, skeletal or back pain, and vertigo, distributed evenly between the two groups of patients. The proportions of severe AEs were also comparable in the two groups, including four deaths unrelated to the study drug, all occurring in the non-iron group. Of interest are also the results of an observational study including 41 anemic (Hb < 10.5 g/dl) patients with multiple myeloma or malignant lymphoma who were treated with rhEPO (10,000 U sc three times a week or 30,000 U sc once a week). After 6 weeks of treatment, hematological response, deﬁned as Hb increase ≥2 g/dl from baseline, was achieved in 66% of patients. Twelve out of 14 patients without response subsequently received IV iron (200 mg iron sucrose once a week) for 4 weeks and 83% of these patients responded, resulting in a total response rate of 90% (Katodritou et al. 2007).

rhEPO in anemic patients with solid tumors and chemotherapy

475

Considering the results of these studies, IV iron supplementation appears to be one of the key points in optimizing treatment with ESAs, not only in patients with anemia of chronic renal failure but also in patients with anemia of chronic diseases including cancer (Auerbach et al. 2007). Many patients with anemia of cancer may have iron-depleted or iron-restricted erythropoiesis before treatment with ESAs and a considerable number may develop functional iron deﬁciency during treatment, resulting in hyporesponsiveness of erythropoiesis and, consequently, response failure, longer time to response or higher doses of ESAs required to achieve response. In these patients, sufﬁcient iron supply could not only improve the results and distinguish between patients with responding or nonresponding anemias to ESAs, but also can reduce the cost of treatment (Ford et al. 1998; Oberhoff et al. 2000; Hedenus et al. 2007; Henry et al. 2007). Of course, there are open questions regarding the long-term use of IV iron, but cancer patients with chemotherapy-induced anemia receiving ESAs usually need short-term IV iron supplementation with a relatively low total dose, which is comparable with the total dose of iron conferred by 4–5 units of red blood cells to these patients. Each unit of RBCs usually contains 200 mg iron (Ludwig 2006a; Gascon, Chapter 27 in this book).

Intravenous iron administration The currently available iron formulations for IV iron therapy are iron dextran (INFeD® and DexFerrum®), iron gluconate (Ferrlecit®, Schein Pharm Corp, Florham Park, NJ), and iron sucrose (saccharate) (Venofer®, Vifor International, Inc., St. Gallen, Switzerland) (Table 9). Although the three formulations considerably vary in their pharmacology, the mechanism of iron distribution in the body is the same for all. After IV administration, the ironcarbohydrate complex is separated by the RES and iron is gradually released into the circulation, where it binds to transferrin and is transported to the liver, spleen and the bone marrow. In the bone marrow, iron binds to the receptor site of erythroid progenitor cells and is used for Hb synthesis. The absolute reticulocyte count can increase within 7 days and Hb response may occur within 1–2 weeks of iron dextran administration. Since iron gluconate and iron sucrose are more readily available to erythropoiesis than iron dextran, Hb increase may more rapidly occur with these two drugs and has been observed ﬁrst after 1 week of iron sucrose. Iron dextran has the advantage of being able to be infused for the patient’s total iron need in one administration (total-dose infusion). This iron formulation, however, may be associated with clinically signiﬁcant dosedependent and dose-independent adverse effects, including life-threatening anaphylactic reactions, presumably due to preformed antibodies against dextran (Table 9). These reactions have led to some reluctance to use IV

No IV injection, IV infusion 0.9% sodium chloride 100 mL 0.9% NS IV over 15 min 48 h (concentration of 0.5–2 mg/ml)

100 mg 100 mg over 2–5 min <3,000 mg

Yes IV infusion 0.9% sodium chloride Dilute dose in 250–1,000 mL of 0.9% NS infused over 1–6 h Not reported

Dosing IV injection Maximal dose for one infusion Total dose infusion Routes Diluent Infusion

34–60,000 Da None

Abbreviations: IV, intravenous; NS, normal saline. * Low-molecular-weight iron dextran, ** high-molecular-weight iron dextran. Adapted from Silverstein and Rodgers 2004, Ludwig 2006a, and Auerbach et al. 2007.

Molecular weight Preservative

165,000 Da*, 265,000 Da** None

100 mg 100 mg over 5 min <400 mg

In case of total-dose

Premedication

Stability

Physician discretion 25-mg IV slow push

Required 25-mg IV slow IV push No

20 mg/min Not to exceed 20 mg/min

50 mg/mL (2-mL vial) Not to exceed 50 mg/min

Iron sucrose

Concentration IV injection (maximum rate) Test dose Test dose

Iron dextran

Table 9. IV Administration of iron products

289–440,000 Da Benzyl alcohol

No IV injection, IV infusion 0.9% sodium chloride 125 mg in 100 mL of NS IV over 1 h Not reported

125 mg 125 mg over 10 min <125 mg

Physician discretion 25-mg slow push or 25 mg in 50 ml of NS IV over 60 min No

12.5 mg/mL (5-mL ampule) Not to exceed 12.5 mg/min

Ferric gluconate

476 M. R. Nowrousian

rhEPO in anemic patients with solid tumors and chemotherapy

477

iron in general. The low-molecular-weight iron dextran (INFeD®, North and South America, CosmoFer®, Europe and Asia) has been reported to be associated with a lower incidence of adverse effects (AEs) than the high-molecular-weight iron dextran (DexFerrum®). All new patients started on iron dextran, however, should receive a 25-mg test dose and be monitored for adverse effects for 1 h after the test dose before continuing administration. Uneventful test doses do not exclude adverse reactions after the ﬁrst and subsequent doses. Iron dextran should be administered IV and the rate of injection or infusion should not exceed 50 mg/min. The maximum dose for one infusion is <3,000 mg (Burns and Pomposelli 1999; Macdougall 1999; Bailie et al. 2000; Fishbane and Kowalski 2000; Van Wyck et al. 2000; Kosch et al. 2001; Silverstein and Rodgers 2004; Ludwig 2006a; Auerbach et al. 2007). Iron gluconate and iron sucrose are associated with much lower risk of allergic reaction than iron dextran. Iron gluconate, however, has to be given in low doses over a prolonged injection or infusion time to avoid anaphylactoid reactions that may occur as a result of a rapid dissociation of iron, leading to an increased production of ionized free iron in circulation. The rate of administration should not exceed 12.5 mg/min. A standard dose of 125 mg may be administered by IV injection over 10 minutes, but maximum doses of 250 mg have to be applied as infusion over at least 1 h to be safe. The likelihood of “free iron” reactions is considerably lower with iron sucrose. Recommended doses of this drug are very safe with minimal risk of anaphylactic or anaphylactoid reactions. AEs are uncommon and not life-threatening. The rate of administration, however, should not exceed 20 mg/min. The maximum dose for one infusion is <400 mg (Burns and Pomposelli 1999; Macdougall 1999; Bailie et al. 2000; Fishbane and Kowalski 2000; Van Wyck et al. 2000; Kosch et al. 2001; Silverstein and Rodgers 2004; Auerbach et al. 2007). The most common adverse effects of IV iron therapy include hypotension, brachycardia, chest pain, nausea, vomiting, diarrhea, abdominal pain, headache, fever, allergic reaction, pruritus, malaise, arthralgias, myalgias and back pain. The rates of AEs associated with iron dextran, iron gluconate, and iron sucrose have been reported to be up to 50%, 35%, and 36%, respectively, but according to a recently published report on data from the United States Food and Drug Administration (FDA), the frequency of AEs associated with IV iron has decreased and the rates are currently extremely low (Chertow et al. 2006). Anaphylactic reactions can occur in 0.6–0.7% of patients receiving high-molecular-weight iron dextran, but are rare with iron gluconate and extremely rare with iron sucrose (Table 9). Fatal hypersensitivity reactions have not been reported with these two drugs (Silverstein and Rodgers 2004; Auerbach et al. 2007). In general, the incidence of AEs associated with IV iron largely depends on the dose and velocity of injection or infusion. The slower the drugs are injected or infused

478

M. R. Nowrousian

the lower is the incidence of adverse effects, particularly acute adverse effects. An issue of concern regarding IV iron in anemic cancer patients receiving ESAs may be its administration together with anthracyclines, such as doxorubicin (Gascon, Chapter 27 in this book). Oxidative stress and production of free radials have been postulated to play a major role in cardiotoxicity of anthracyclines, and free radicals have been reported to increase their toxic effects when they react with iron. In addition, anthracylines have been reported to inhibit the release of iron from ferritin (Kwok and Richardson 2003; Minotti et al. 2004; Xu et al. 2005; Elliott 2006). There is, however, some controversy regarding the role of iron in toxic mechanisms of anthracyclines and, in a recent study, iron did not appear to be involved in the oxidative stress-mediated cytotoxicity of doxorubicin (Kaiserova et al. 2006). An appropriate strategy for IV iron supplementation in patients with real or functional iron deﬁciency may be 100 mg iron sucrose or 125 mg iron gluconate weekly or 200 mg iron sucrose or 250 mg iron gluconate every other week as slow infusion for the ﬁrst 4–8 weeks followed by doseschedule adaptation to keep TFS between 25–35%. Patients receiving IV iron should be monitored closely for clinical and laboratory evidence of iron toxicity or overload. Serum ferritin level and TFS should be measured weekly, and treatment should be stopped, if ferritin level is >1,000 μg/l or TFS is >50%, possibly indicating iron overload (Kaltwasser and Gottschalk 1999; Macdougall 1999; Silverstein and Rodgers 2004; Auerbach et al. 2007). Ascorbic acid (vitamin C) is known to be involved in iron absorption and metabolism (Bothwell et al. 1964; Bothwell 1968; Lipschitz et al. 1971). Now, there are studies using this drug intravenously to improve iron mobilization and utilization in anemic, iron overloaded hemodialysis patients not responding to rhEPO (Gastaldello et al. 1995; Tarng and Huang 1998; Tarng et al. 1999; Giancaspro et al. 2000; Gibbs 2000; Melendez 2000; Sezer et al. 2002; Sturm et al. 2005; Jacobs 2006). The use of ascorbic acid in these patients may be of particular interest, since many of them have ascorbic acid deﬁciency due to its loss during dialysis and lack of dietary intake (Ponka and Kuhlback 1983; Gibbs 2000). Ascorbic acid, on the other hand, is an antioxidant and a reducing agent that appears to potentiate the mobilization of iron from inert tissue stores and to facilitate its incorporation into protoporphyrin (Bothwell et al. 1964; Bothwell 1968; Gibbs 2000; Tarng et al. 2001). The use of this agent could, therefore, be generally of interest in patients with functional iron deﬁciency and high ferritin level not responding to rhEPO, especially in those with pathologically restricted iron mobilization, such as patients with anemia of cancer or other chronic diseases. In these patient groups, however, the efﬁcacy and safety of IV ascorbic acid remain to be evaluated.

rhEPO in anemic patients with solid tumors and chemotherapy

479

Safety of ESAs Epoetin alfa, epoetin beta, and darbepoetin alfa are generally well tolerated when they are used according to prescribing information. Side effects are rare, usually mild and seldom a reason for discontinuation of treatment (Sowade et al. 1998). In prospective randomized trials using these drugs for treating chemotherapy-induced anemia, side effects such as anorexia, asthenia, fatigue, fever, ﬂulike syndrome, skin rush, facial ﬂushing, constipation, cough, diarrhea, dizziness, dyspnoe, edema, nausea, vomiting, and bone pain were reported to occur, but in most cases with a similar frequency both in ESA- and placebo-treated groups, indicating that they were not speciﬁcally related to treatment with ESAs (Miller et al. 1992; Abels 1993; Cascinu et al. 1993, 1994a,b, 1995; Case et al. 1993; Gamucci et al. 1993; Tsukuda et al. 1993; Cazzola et al. 1995; de Campos et al. 1995; Österborg et al. 1996; Voigtmann et al. 1996; Del Mastro et al. 1997; Glaspy et al. 1997; Pawlicki et al. 1997; Sowade et al. 1998; ten Bokkel Huinink 1998; Vansteenkiste et al. 2002; Kotasek et al. 2003; Glaspy et al. 2005). Pain at the injection site has been reported to occur and to be more frequently associated with the use of epoetin alfa and darbepoetin alfa than epoetin beta (Ludwig 2006a). Hypertension as a result of an excessive increase in Hb can rarely occur in patients treated with ESAs, but mainly in patients with renal anemia (Abels 1993; Case et al. 1993). Blood pressure, however, should also be monitored carefully in anemic cancer patients when they are treated with ESAs, particularly when there is a history of hypertension or cardiovascular disease.

Pure red cell aplasia (PRCA) A serious AE of treatment with ESAs may be the development of an antibody-mediated PRCA, but this AE is very rare and has not been reported in patients with chemotherapy-induced anemia. Between 2001 and 2003, an increased incidence (approximately 200 global cases) of PRCA was observed in patients with renal anemia, mainly in association with sc use of epoetin alfa (EPREX®/ERYPO®) outside the USA, most probably caused by leaks from uncoated rubber stoppers in preﬁlled syringes containing polysorbate 80 instead of human serum albumin as stabilizer. Since 2003, however, the frequency of PRCA has declined due to increased awareness and, particularly, due to elimination of product-related immunogenic factors (Boven et al. 2005; Jelkmann, Chapter 16 in this book). The diagnostic criteria for anti-ESA antibody-induced PRCA include a rapid fall of blood Hb (1 g/l and day), reticulocytes <10 × 109/l, the demonstration of neutralizing anti-EPO antibodies in serum, and the absence of erythroblasts from an otherwise normal bone marrow (Casadevall et al. 2004, 2005). In case of PRCA,

480

M. R. Nowrousian

ESAs have to be discontinued and the patients should not be retreated with these drugs. There are some clinical reports suggesting that immunosuppressive therapy may be of value in reconstituting erythropoiesis (Bennett et al. 2005).

Thromboembolic events (TEE) A trend toward an increased incidence of TEE was already reported in some nonrandomized and randomized studies evaluating the efﬁcacy and safety of ESAs in anemic cancer patients receiving chemotherapy (Wun et al. 2003; Rosenzweig et al. 2004; Food and Drug Administration 2006; Bokemeyer et al. 2007; Rizzo et al. 2007; Westin et al. 2007; Glaspy, Chapter 30 in this book). In a recent meta-analysis including 57 randomized trials and 6769 patients, a 1.67-fold increased risk of TEE was reported for patients who were treated with ESAs compared with control patients (Bohlius et al. 2005, 2006). This meta-analysis, however, comprised studies of various groups of patients including those with and without anemia and patients with chemotherapy, radiotherapy or without any kind of cancer treatment. In another metaanalysis including 9 randomized studies and evaluating individual data from a total of 1413 patients, the rate of TEE was slightly higher in patients treated with rhEPO compared with control patients (5.9% vs 4.2% of patients), but the rate of thromboembolism-related mortality was identical in both groups (1.1%). A comprehensive meta-analysis reviewing the data from 40 prospective randomized and nonrandomized trials of 21,378 patients with chemotherapy-induced anemia showed a slight but not statistically signiﬁcant difference of TEE between patients with ESAs or without (5.2% vs 3.1%) (Ross et al. 2006). In spite of these controversial results, a slightly but signiﬁcantly increased risk of TEE appears to be present in patients treated with ESAs and should be accordingly recognized, particularly in patients with predispositions for developing TEE, such as patients with a history of thromboses, surgery, inactivity, and immobilization, and patients who are treated with thrombogenic drugs, such as thalidomide or lenalidomide (Rizzo et al. 2007). To minimize the risk of TEE during treatment with ESAs, the dose should be adapted to avoid an Hb increase >1 g/dl in 2 weeks and to maintain the Hb level around 12 g/dl. The latter can be achieved by increasing intervals of dosing and/or titration of the lowest effective maintenance dose individually in each patient (EORTC Guidelines 2007). According to the recommendations of the Food and Drug Administration (FDA), the dose of rhEPO or darbepoetin alfa should be reduced by 25% and 40%, respectively, when there is an increase in Hb > 1 g/dl in 2 weeks or Hb levels are approaching 12 g/dl. In cases with Hb levels >12 g/dl, treatment should be withheld until Hb is around 11 and restarted with a dose at 25% and 40% below the previous dose of rhEPO or darbepoetin alfa, respectively (Rizzo et al. 2007).

rhEPO in anemic patients with solid tumors and chemotherapy

481

Treatment with ESAs should be discontinued 4 weeks after the end of chemotherapy (EORTC 2007).

EPO receptor (EPO-R) expression on tumor cells A number of studies has reported on EPO-R expression on the surface of various types of malignant cell lines or tumor biopsy specimens and, in some studies, exposure to EPO has been reported to stimulate proliferation of tumor cells. The results of these studies, however, are subjects of controversial discussions, because of the methodological problems associated with determination of EPO receptors and their functionality and because of artiﬁcially high EPO concentrations, which were mostly used to evaluate the proliferative effects of EPO on tumor cells in vitro (Table 10, Addendum 1 in this book) (Fandrey, Chapter 3 in this book; Jelkmann and Laugsch 2007; Österborg et al. 2007 and Chapter 4 in this book; Perrotta 2007; Sinclair et al. 2007; Sizer 2007). In addition, there are in-vitro studies and, particularly, animal experiments with contradictory results almost exclusively showing a positive effect of treatment with ESAs on the outcome of radiotherapy or chemotherapy under anemic or nonanemic conditions, even in tumors expressing such EPO receptors (Table 10) (Österborg, Chapter 4 in this book).

Effects of treatment with ESAs on survival Another issue of controversial discussion is the impact of treating cancer- and chemotherapy-induced anemia with ESAs on survival. As already mentioned, in two studies using epoetin alfa and darbepoetin alfa in patients with hematological malignancies and patients with lung cancer, respectively, there was a trend toward improved survival and progression-free survival in patients who were treated with ESAs (Table 5) (Littlewood et al. 2001; Vansteenkiste et al. 2002, 2004). Contrary to the results of these and other studies showing a positive effect and numerous studies showing no effect of treatment of anemia with ESAs on survival, ﬁve recent studies reported on a negative impact (Henke et al. 2003; Leyland-Jones et al. 2005; Glaspy et al. 2007; Overgaard et al. 2007; Wright et al. 2007). Three of these studies were carried out in mainly nonanemic patients, one in patients with breast cancer and the other two in patients with head and neck cancer (Henke et al. 2003; Leyland-Jones et al. 2005; Overgaard et al. 2007). In the last two studies, relatively high target Hb levels were used and, in one of them, the dosage of rhEPO was twice as high as usually given in anemic patients receiving radiotherapy or chemotherapy resulting in a rapid and excessive increase in Hb within a few weeks (Henke et al. 2003, 2006). Two further studies were carried

Breast cancer Various Various

Cell lines, biopsies

Cell lines, biopsies Cell lines

Cell lines, biopsies

Cell lines

Cell line

Cell lines, biopsies

Cell lines, biopsies Cell lines

Cell lines

Cell lines

Westenfelder et al. 2000 Acs et al. 2001 Westphal et al. 2002 Batra et al. 2003

Acs et al. 2004

Tullai et al. 2004

Mohyeldin et al. 2005 Kokhaei et al. 2007 Acs et al. 2003 Liu et al. 2004

Uchida et al. 2004

Belenkov et al. 2004 Carvalho et al. 2005

Cell lines

Puriﬁed cells

A wide range of human solid tumors Renal cell carcinoma

Cell lines (n = 22)

Berdel et al. 1991

Cellular model for CML with blastic crisis Human malignant glioma, cervical cancer RCC, myelomonocytic leukemia

Human lymphod malignancies Human cervical cancer Various cancer types

Human colorectal carcinoma Head and neck cancer

Breast cancer

Tumor type

Cellular model

Study

4,000, 8,000

30,000

10–10,000

25,000–200,000 10,000

100,000

200,000

5,000

200,000

10,000–30,000

250,000 5,000

500–100,000

100–100,000

rhEPO concentration (mU/ml) used

rhEPO did not induce proliferation or activation of cells rhEPO inhibited cytotoxicity of cisplatin rhEPO did not affect the sensitivity of cells to cisplatin rhEPO inhibited imatinib-induced apoptosis and induced erythroid differenttiation of cells rhEPO induced resistance to ionizing radiation and to cisplatin rhEPO induced chemosensitization of cells through inhibition of NF-kB rescue pathway

rhEPO increased proliferation rate of cells by 25% rhEPO did not induce proliferation in any of the cell lines tested rhEPO increased antiapptotic gene expression and production of angiogenic growth factors Autocine EPO signaling inhibited hypoxiainduced apoptosis rhEPO did not activate the JAK/STAT or the ERK1/2 pathwasy in cells tested rhEPO promoted invasiveness of cells

rhEPO stimulated proliferation of cells

rhEPO did not stimulate clonal cell growth

ESA, results

Table 10. Cellular models and animal experiments evaluating the effects of ESAs on chemosensitivity and radiosensitivity of tumor cells

482 M. R. Nowrousian

Anemic mice Anemic rats Anemic/nonanemic mice Anemic nude mice

Ning et al. 2005

Human tumors Experimental tumors Xenografted glioblastomas Experimental human tumors Murine model of Lewis lung carcinoma Murine colon-26 adenocarcinoma Murine squamous cell carcinoma

Anemic mice Anemic mice

Thews et al. 1998 Silver and Piver 1999 Stüben et al. 2001 Thews et al. 2001 Stüben et al. 2003a Stüben et al. 2003b Sigounas et al. 2004 Golab et al. 2002

Anemic/nonanemic mice

Anemic mice

Anemic mice

Experimental tumors Human ovarian cancer

Animal experiment

Study

Tumor type

Cell lines

Palumbo et al. 2007

Human pleural malignant mesothelioma

Breast cancer, F-MEL leukemic cells Breast cancer cells expressing EPO-R

Cell lines

Cell lines

Ovarian cancer

Cell line

McBroom et al. 2005 Gewirtz et al. 2006 LaMontagne et al. 2006

Melanoma

Cell lines

Kumar et al. 2005

30 μg/kg i.p. every 2 weeks

60 U/kg two injections 1,000 U/kg sc daily

750 U/kg sc tiw

1,000 U/kg sc tiw 1,000 U/kg sc tiw 750 U/kg sc tiw

1,000 U/kg sc tiw 20 U sc daily

ESA dose

1,000–10,000

100–10,000

10,000

25,000–200,000

10,000 and 100,000

rhEPO synergistically enhanced cytotoxicity of cisplatin, mitomycin C, and cyclophosphamide rhEPO restored antitumor effectiveness of photodynamic therapy DE potentiated the efﬁcacy of radiotherapy in mice with corrected or uncorrected anemia

rhEPO improved radiosensitivity rhEPO improved response to cyclophosphamide rhEPO prevented anemia and partially restored radiosensitivity of tumors rhEPO restored radiosensitivity of tumors

rhEPO improved radiosensitivity rhEPO improved response to cisplatin

ESA, results

rhEPO did not stimulate the proliferation of cells and their sensitivity to cytotoxic agents* rhEPO did not stimulate migration and proliferation of cells, or activation of mitogenactivated protein kinase rhEPO differently affected proliferation and sensitivity of cells to cisplatin and pemetrexed

rhEPO increased resistance to hypoxia and dacarbazine rhEPO decreased sensitivity to cisplatin

rhEPO in anemic patients with solid tumors and chemotherapy 483

Nonanemic mice

Shannon et al. 2005 Hardee et al. 2006

Nonanemic mice

Nonanemic mice

Lövey et al. 2007

Mittelman et al. 2001 Katz et al. 2005 Murine myeloma and leukemia models

Murine colorectal cancer model Human epidermoid carcinoma Murine myeloma

Murine mammary adenocarcinoma Xenografted breast cancer expressing EPO-R

Human squamous cell and colorectal carcinoma Murine model of Lewis lung carcinoma Murine mammary adenocarcinoma

Tumor type

10–100 U sc daily of tiw 30 U sc daily

150 U/kg i.p. tiw

2.5 μg/kg rhEPO sc every other day or 7.5 mg/kg DE sc weekly 286 U sc tiw

3 μg/kg sc tiw

2,000 U/kg sc tiw

10 μg/kg i.p. weekly

150 U/kg i.p. tiw

ESA dose

rhEPO induced tumor regression and improved antitumor immune response rhEPO induced tumor mass reduction

rhEPO modulated the effects of radiotherapy on tumor microvessels rhEPO increased the efﬁcacy of irradiation

rhEPO improved results of chemotherapy with 5-ﬂuorouracil DE did not stimulate tumor growth but improved chemotherapeutic outcome rhEPO inhibited apoptosis of cells in vitro but did not affect tumor growth and its delay after taxol in vivo DE did not signiﬁcantly alter tumor growth and radioresponsiveness rhEPO did not stimulate tumor growth or resistance to paclitaxel

ESA, results

ESAs = erythropoiesis-stimulating agents; rhEPO = recombinant human erythropoietin (epoetin alfa or beta); DE darbepoetin alfa; EPO-R = EPO receptor; sc = subcutaneous; i.p. = intraperitoneal; tiw = 3 times a week.

Nonanemic mice

Nonanemic rats

Nonanemic mice

Nonanemic rats

Ceelen et al. 2007

Kirkpatrick et al. 2006 LaMontagne et al. 2006

Nonanemic mice

Tovari et al. 2005

Nonanemic rats

Animal experiment

Study

Table 10. Continued

484 M. R. Nowrousian

Patients

PR, rhEPO 1,000 U vs 5,000 U sc tiw PR, rhEPO vs placeb PR, rhEPO vs SC PR, rhEPO vs placebo PR, rhEPO vs placebo PR, rhEPO vs SC PR, rhEPO vs SC PR, rhEPO vs placebo Meta-analysis of R trials Meta-analysis of R trials

Breast cancer, Ch, hormonal therapy HM, Ch? Ovarian cancer, ChP SCLC, ChP Breast cancer, Ch Breast cancer, Ch Breast cancer, Ch NSCLC, in part Ch Various, Ch, RT, RCT, no therapy HM, Ch

n = 119, anemic

ST, HM, Ch/Chp ST, HM, Ch/Chp

n = 21,378, anemic

n = 2,141, anemic

Ross et al. 2006

Ludwig et al. 2007*

Meta-analysis of R and NR trials Meta-analysis of R trials

Meta-analysis of R trials

Comparable PFS and OS in in patients with DE or placebo Comparable OS, but signiﬁcantly reduced risk of disease progression favoring rhEPO No signiﬁcant impact of treatment with rhEPO or DE on survival Comparable risks of PFS and death between in patients with DE or placebo

Signiﬁcantly decreased OS dysfavoring rhEPO No association between treatment with rhEPO or DE and survival

Signiﬁcantly decreased OS dysfavoring rhEPO, but not comparable TDP and PFS Comparable PFS and OS Comparable DFS and OS

Signiﬁcantly higher Hb and tumor response favoring 5,000 U rhEPO sc tiw Comparable OS Trend toward improved PFS favoring rhEPO, comparable OS Comparable OS and mortality

Comparable tumor progression rates and death rates Trend toward improved survival favoring rhEPO, both in patients with ST or HM Trend toward improved PFS favoring DE

Survival outcome

ST = solid tumors; HM = hematological malignancies; PR = prospective randomized; rhEPO = recombinant human erythropoietin (epoetin alfa or beta); DE = darbepoetin alfa; U = international units; sc = subcutaneous; Ch = non-platinum-based chemotherapy; ChP = platinum-based chemotherapy; RT = radiotherapy; CRT = chemoradiotherapy; CRR = complete remission rate; SC = standard care; R = randomized; NR nonrandomized; OS = overall survival; DFS = disease-free survival; PFS = progression-free survival; RFS = relapse-free survival; Hb = hemoglobin. * meta-analyses including individual patient data.

ST, HM, Ch/Chp

n = 1,413, anemic

Aapro et al. 2006a*

PR, DE vs placebo

SCLC, NSCLC, Ch/ChP

n = 320, anemic

PR, rhEPO vs placebo

PR, rhEPO vs SC

Study design

ST, HM, Ch

HM, Ch

Type of cancer, treatment

n = 375, anemic

Österborg et al. 2005 n = 343, anemic Reed et al. 2005 n = 120, anemic/ nonanemic Grote et al. 2005 n = 224, anemic/ nonanemic Leyland-Jones et al. n = 939, mainly 2005 nonanemic Aapro et al. 2007 n = 463, anemic Moebus et al. 2007 n = 658, anemic/ nonanemic Wright et al. 2007 n = 70, anemic Bohlius et al. 2005, n = 3,287, 9,353, 2006 anemic/ nonanemic Hedenus et al. 2005* n = 314, anemic

Littlewood et al. 2001 Vansteenkiste et al. 2002 Larsson et al. 2004

Österborg et al. 1996 n = 121, anemic

Study

Table 11. Clinical studies reporting on survival in cancer patients receiving chemotherapy and rhEPO or DE

rhEPO in anemic patients with solid tumors and chemotherapy 485

486

M. R. Nowrousian

Fig. 2. Deaths during studies comparing epoetin (EPO) or darbepoetin (DARB) with control (placebo or standard care) in patients with chemotherapy-induced anemia. All-cause mortality during the study period was 10.3% among rhEPO patients and 10.6% among control patients, for an OR of 0.86 (95% CI, 0.58–1.28). For darbepoetin alfa, all-cause mortality was 7.3% compared with 6.8% for control, for an OR of 1.26 (95% CI, 0.74–2.14). These results did not reach statistical signiﬁcance in either case. From Ross et al. 2006, reprinted with permission

out in anemic patients with non-small-cell lung cancer and patients with various types of nonmyeloid malignancies, who should not have received any kind of anticancer treatment. However, in one of them, a part of patients received non-platinum-based chemotherapy after randomization. All these studies were performed in settings other than currently approved for the use of ESAs in cancer patients and the results became subject of critical reviews, because of a number of methodological problems (Blumberg and Heal 2004; Freidlin and Korn 2004; Gemici 2004; Haddad and Posner 2004; Janecka 2004; Kaanders and van der Kogel 2004; Leyland-Jones and Mahmud 2004; Vaupel and Mayer 2004; Vaupel et al. 2005, 2006; Crawford 2006, 2007; Crawford et al. 2007; Jelkmann and Laugsch 2007; Ragione et al. 2007; Sizer 2007), particularly, an imbalanced distribution of patient- and disease-related factors determining survival between patients receiving ESAs and control patients

rhEPO in anemic patients with solid tumors and chemotherapy

487

A

Progression-Free Survival (%)

100 80 60 40 20 0 0

3

Placebo n = 159 Darbepoetin alla n = 155

114 116

6 9 12 15 Months From First Dose

18

21

24

7 13

4 6

1 5

0 1

6 9 12 15 Months From First Dose

18

21

24

86 94

16 16

3 7

0 1

50 63

26 35

14 16

Overall Survival (%)

100

B

80 60 40 20 0 0

3

Placebo n = 159 Darbepoetin alla n = 155

127 133

58 74

44 49

30 34

Fig. 3. Disease progression-free survival (PFS) and overall survival (OS) analysis from a randomized, double-blind, placebo-controlled trial in anemic patients with lung cancer treated with darbepoetin alfa. The median duration of PFS was 5.1 months (95% CI, 4.1 to 6.9 months) and 4.4 months (95% CI, 3.7 to 5.3 months) for darbepoetin alfa and placebo (HR, 0.79; 95% CI, 0.62 to 1.00; P = .051; upper panel). The median OS time was 10.4 months (95% CI, 8.8 to 12.0 months) and 7.8 months (95% CI, 6.6 to 9.0 months; HR, 0.77; 95% CI, 0.59 to 1.01; P = .060; lower panel). From Hedenus et al. 2005, reprinted with permission

in at least three studies (Henke et al. 2003; Leyland-Jones et al. 2005; Wright et al. 2007). Several comprehensive systematic reviews of the literature including meta-analyses of individual patient data from prospective randomized trials using rhEPO or darbepoetin alfa in patients with chemotherapy-induced anemia have not shown any impact of treatment with these agents on survival, neither in patients with solid tumors nor in those with hematological malignancies (Table 11) (Figs. 2–5) (Hedenus et al. 2005; Aapro et al. 2006a, 2007; Bohlius et al. 2005, 2006; Ross et al. 2006; Ludwig et al. 2007). In one of these meta-analyses, treatment with rhEPO was found to signiﬁcantly reduce the risk of a rapidly progressive disease (Fig. 5) (Aapro

488

M. R. Nowrousian

Progression-Free Survival (%)

100

A

80 60 40 20 0 0

3

6

9 12 15 18 21 Months From First Dose

24

27

30

33

61 60

51 52

43 37 51 37

21 21

9 12 15 18 21 Months From First Dose

24

27

30

33

Placebo n = 169 160 146 135 132 124 116 113 102 93 Darbepoetin alla n = 175 162 150 136 126 122 117 101 89 87

79 69

50 39

B

Overall Survival (%)

Placebo n = 169 150 119 104 Darbepoetin alla n = 175 146 125 109

95 100

83 85

69 72

100 80 60 40 20 0 0

3

6

Fig. 4. Disease progression-free survival (PFS) and overall survival (OS) analysis from a randomized, double-blind, placebo-controlled trial in anemic patients with lymphoproliferative malignancies treated with or without darbepoetin alfa. The median duration of PFS was 14.2 months (95% CI, 12.2 to 17.5 months) and 15.9 months (95% CI, 13.1 to 19.0 months) for darbepoetin alfa and placebo (HR, 1.03; 95% CI, 0.80 to 1.32; p = .803; upper panel). The median OS time was 30.4 months (95% CI, 22.7 months to not assessable) for darbepoetin alfa and 36.6 months (95% CI, 30.2 months to not assessable) for placebo (HR, 1.26; 95% CI, 0.92 to 1.71; p = .152; lower panel). From Hedenus et al. 2005, reprinted with permission

et al. 2006a, 2007). In addition, in a recent prospective randomized study exclusively designed to address the issue of survival, no difference was found between anemic patients with metastatic breast cancer receiving chemotherapy with or without rhEPO (Aapro et al. 2007). Similar results were also observed in two other prospective randomized trials of patients with breast cancer receiving adjuvant chemotherapy and patients with small-cell lung cancer receiving platinum-based chemotherapy with or without rhEPO or darbepoetin alfa, respectively (Moebus et al. 2007; Pirker et al. 2007). From these studies, it can be concluded that in clinically and statistically welldesigned studies, ESA therapy has no adverse effects on survival.

rhEPO in anemic patients with solid tumors and chemotherapy A

489

1.0 0.9 0.8 Overall survival

0.7 0.6 0.5 0.4 Treatment group

0.3

Control Epoetin Beta

0.2 0.1 0.0 0

3

6

9

12

15

18

4 3

0 2

0 1

0 0

9

12

15

18

0 2

0 1

0 0

Months since treatment N at risk Control 613 Epoetin Beta 800 B

388 509

60 137

1.0 Time to disease progression

0.9 0.8 0.7 0.6 0.5 0.4 Treatment group

0.3

Control Epoetin Beta

0.2 0.1 0.0 0

3

6

Months since treatment N at risk Control 613 Epoetin Beta 800

335 450

56 121

3 3

Fig. 5. Overall survival (OS) and time to progression (TTP) from a meta-analysis of nine randomized clinical trials in anemic patients with solid tumors or lymphoproliferative diseases treated with or without epoetin beta. Kaplan–Meier analysis showed no relevant difference between epoetin beta and control, with respective event rates of 10.0 and 9.5% and an overall hazard ratio (HR) of 0.97 (95% CI 0.69, 1.36; logrank, p = 0.87). Kaplan–Meier analysis also showed a reduced risk of progression among patients treated with epoetin beta (HR 0.78, 95% CI 0.62, 0.99; log-rank test, p = 0.042). From Aapro et al. 2006, reprinted with permission

490

M. R. Nowrousian

Conclusions There are many studies indicating that ESAs can be effectively and safely used in the treatment of cancer- and chemotherapy-induced anemia in solid tumor patients, both in patients receiving platinum-based and non-platinum-based chemotherapy. Considering the results of dose-ﬁnding studies and trials dealing with the pharmacodynamics of the drug, an appropriate dose schedule for the start of treatment appears to be 150 U/kg sc or a ﬁxed dose of 10,000 U sc tiw or a ﬁxed total dose of 30,000 U sc once a week for rhEPO and 2.25 μg/kg sc once a week or a ﬁxed total dose of 500 μg sc once every 3 weeks for darbepoetin alfa. The subcutaneous route of administration should be preferred, since it allows a dose reduction of up to 52% and is more comfortable for patients than the IV route. A key point in improving the results of treatment with ESAs appears to be iron supplementation, either from the beginning of treatment or subsequently in case of functional or real iron deﬁciency. The results of clinical studies show that the use of ESAs in anemic cancer patients receiving chemotherapy signiﬁcantly increases Hb level, avoids or reduces the need for RBC transfusions and improves exercise capacity, cognitive functions, and QOL. Appropriate patients for treatment are those with baseline Hb levels of ≤11 g/dl and clinical symptoms of anemia, and patients with declining Hb < 12 g/dl further receiving chemotherapy, particularly when they have limited cardiopulmonary reserve, coronary artery disease or angina, substantially reduced exercise capacity, energy, or ability to carry out activities of daily living. Target Hb level should be about 12 g/dl (Table 7). Treatment with ESAs is generally well tolerated and adverse effects are rare. The slight but signiﬁcant increase in TEE should be recognized in patients who are at risk for such events, e.g. patients with a history of thromboses, surgery, prolonged periods of immobilization or limited activity and patients who are treated with thrombogenic agents. The risk of TEE can be minimized by avoiding increases of >1 g/dl within 2 weeks and by adapting the dose individually in each patient to maintain the Hb level around 12 g/dl.

References 1. Aapro M, Coifﬁer B, Dunst J, et al (2006a) Effect of treatment with epoetin beta on short-term tumour progression and survival in anaemic patients with cancer: a meta-analysis. Br J Cancer 95: 1467–1473 2. Aapro M, Barnadas A, Leonard RC, et al (2006b) Effects of epoetin beta treatment in patients with metastatic breast cancer receiving chemotherapy. Results of the BRAVE trial. Breast Cancer Res 100 [Suppl 1]: Abstr 6095 3. Aapro M, Leonhard RC, Barnadas A, et al (2007) Effect of once weekly epoetin beta on survival in patients with metastatic breast cancer receiving anthracycline- and/or taxane-based chemotherapy – results of the BRAVE study. J Clin Oncol (in press)

rhEPO in anemic patients with solid tumors and chemotherapy

491

4. Abdelrazik N, Fouda M (2007) Once weekly recombinant human erythropoietin treatment for cancer-induced anemia in children with acute lymphoblastic leukemia receiving maintenance chemotherapy: A randomized case-controlled study. Hematology (in press) 5. Abels R (1993) Erythropoietin for anaemia in cancer patients. Eur J Cancer 29A [Suppl 2]: S2–8 6. Abels R, Gordon D, Nelson R, et al (1991) Transfusion practice in advanced cancer patients. Blood 78 [Suppl 1]: 474a 7. Abels R, Larholt K, Nelson R, et al (1994) Risk of transfusion in small cell lung cancer patients receiving chemotherapy. Blood 87 [Suppl]: 664a (Abstr 2642) 8. Acs G, Acs P, Beckwith SM, et al (2001) Erythropoietin and erythropoietin receptor expression in human cancer. Cancer Res 61: 3561–3565 9. Acs G, Chen M, Xu X, et al (2004) Autocrine erythropoietin signalling inhibits hypoxia-induced apoptosis in human breast carcinoma cells. Cancer Lett 214: 243–251 10. Acs G, Zhang PJ, McGrath CM, et al (2003) Hypoxia-inducible erythropoietin signalling in squamous dysplasia and squamous cell carcinoma of the uterine cervix and its potential role in cervical carcinogenesis and tumor progression. Am J Pathol 162: 1789–1806 11. Adamson JW (1994) The relationship of erythropoietin and iron metabolism to red blood cell production in humans. Semin Oncol 21: 9–15 12. Adamson JW, Cavill I, Fishbane S, et al (1999) A consensus on current issues and controversies in iron management of patients with chronic renal failure. Seminars in Dialysis 12: 182–194 13. Ahsan N (2000) Infusion of total dose iron versus oral iron supplementation in ambulatory peritoneal dialysis patients: a prospective, cross-over trial. Adv Perit Dial 16: 80–84 14. Airley RE, Mobasheri A (2007) Hypoxic regulation of glucose transport, anaerobic metabolism and angiogenesis in cancer: Novel pathways and targets for anticancer therapeutics. Chemotherapy 53: 233–256 15. Ali M, Rigolosi R, Fayemi AO, et al (1982) Failure of serum ferritin levels to predict bone-marrow iron content after intravenous iron-dextran therapy. Lancet 1: 652–655 16. Anon (2005) Pre-treatment erythropoietin levels cannot be used to predict individual response to epoetin alfa in anaemic cancer patients. Cur Med Res Opin 21 [Suppl 2]: S19–S21 17. Ariganello O, Mancuso A, Di Molfetta M, et al (2004) A new induction schedule of epoetin alfa 40,000 IU in anemic patients with advanced lung cancer. Lung Cancer 46 (1): 119–124 18. Arndt U, Kaltwasser JP, Gottschalk R (2005) Correction of iron-deﬁcient eryhtropoiesis in the treatment of anemia of chronic disease with recombinant human erythropoetin. Ann Hematol 84: 159–166 19. Asif M, Siddiqu A, Keating GM (2006a) Spotlight on darbepoetin alfa in the treatment of anemia in patients with cancer receiving chemotherapy. Biodrugs 20: 321–323 20. Asif M, Siddiqu A, Keating GM (2006b) Darbepoetin alfa. Drugs 66: 997–1012 21. Auerbach M, Ballard H, Trout JR, et al (2004) Intravenous iron optimizes the response to recombinant human erythropoietin in cancer patients with

492

22. 23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

M. R. Nowrousian chemotherapy-related anemia: a multicenter, open-label, randomized trial. J Clin Oncol 22: 1301–1307 Auerbach M, Ballard H, Glaspy J (2007) Clinical update: intravenous iron for anaemia. Lancet 369: 1502–1504 Bacci G, Ferrari S, Bertoni F, et al (2000) Prognostic factors in nonmetastatic Ewing’s sarcoma of bone treated with adjuvant chemotherapy: analysis of 359 patients at the Istituto Ortopedico Rizzoli. J Clin Oncol 18: 4–11 Bailie GR, Johnson CA, Mason NA (2000) Parenteral iron use in the management of anemia in end-stage renal disease patients. Am J Kidney Dis 35: 1–12 Ballard H, Rana J, Ackerman A, et al (1999) Total dose infusion (TDI) of iron dextran (ID) optimizes erythropoietin (EPO) responsiveness in the anemia of cancer (CA). Proc Am Soc Clin Oncol 18: 581a (Abstr) Bamias A, Aravantinos G, Kalofonos C, et al (2003) Prevention of anemia in patients with solid tumors receiving platinum-based chemotherapy by recombinant human erythropoietin (rHuEpo): A prospective, open label, randomized trial by the Hellenic Cooperative Oncology Group. Oncology 64: 102–110 Barrett-Lee PJ, Bailey NP, O’Brien MER, et al (2000) Large-scale UK audit of blood transfusion requirements and anaemia in patients receiving cytotoxic chemotherapy. Br J Cancer 82: 93–97 Barrett-Lee PJ, Ludwig H, Birgegard G, et al (2006) Independent risk factors for anemia in cancer patients receiving chemotherapy: Results from the European Cancer Anaemia Survey. Oncology 70: 34–48 Batra S, Perelman N, Luck LR, et al (2003) Pediatric tumor cells express erythropoietin and a functional erythropoietin receptor that promotes angiogenesis and tumour cell survival. Lab Invest 83: 1477–1487 Becker A, Stadler P, Lavey RS, et al (2000) Severe anemia is associated with poor tumor oxygenation in head and neck squamous cell carcinomas. Int J Radiat Oncol Biol Phys 46: 459–466 Belenkov AI, Shenouda G, Rizhevskaya E, et al (2004) Erythropoietin induces cancer cell resistance to ionizing radiation and to cisplatin. Mol Cancer Ther 3: 1525–1532 Bennett CL, Cournoyer D, Carson KR, et al (2005) Long-term outcome of individuals with pure red cell aplasia and antierythropoietin antibodies in patients treated with recombinant epoetin: a follow-up report from the Research on Adverse Drug Events and Reports (RADAR (Project). Blood 106: 3343–3347 Berdel WE, Oberberg D, Reuﬁ B, et al (1991) Studies on the role of recombinant human erythropoietin in the growth regulation of human nonhematopoietic tumor cells in vitro. Ann Hematol 63: 5–8 Besarab A, Amin N, Ahsan M, et al (2000) Optimization of epoetin therapy with intravenous iron therapy in hemodialysis patients. J Am Soc Nephrol 11: 530–538 Bianchi R, Brines M, Lauria G, et al (2006) Protective effect of erythropoietin and its carbamylated derivative in experimental cisplatin peripheral neurotoxicity. Clin Cancer Res 12: 2607–2612 Bianchi R, Gilardini A, Rodriguez-Menendez V, et al (2007) Cisplatin-induced peripheral neuropathy: neuroprotection by erythropoietin without affecting tumour growth. Eur J Cancer 43: 710–717

rhEPO in anemic patients with solid tumors and chemotherapy

493

37. Birgegard G, Wide L, Simonsson B (1989) Marked erythropoietin increase before fall in Hb after treatment with cytostatic drugs suggests mechanism other than anaemia for stimulation. Br J Haematol 72: 462–466 38. Blumberg N, Heal JM (1996) The transfusion immunomodulation theory: The Th1/Th2 paradigm and an analogy with pregnancy as a unifying mechanism. Semin Hematol 33: 329–340 39. Blumberg N (1997) Allogeneic transfusion and infection: Economic and clinical implications. Semin Hematol 34 [Suppl 2]: 34–40 40. Blumberg N, Heal JM (2004) Erythropoietin to treat anaemia in patients with head and neck cancer. Lancet 363: 80–81 41. Boccia R, Malik IA, Raja V, et al (2006) Darbepoetin alfa administered every three weeks is effective for the treatment of chemotherapy-induced anemia. Oncologist 11: 409–417 42. Bohlius J, Langensiepen S, Schwarzer G, et al (2005) Recombinant human erythropoietin and overall survival in cancer patients: results of a comprehensive meta-analysis. J Natl Cancer Inst 97: 489–498 43. Bohlius J, Wilson J, Seidenfeld J, et al (2006) Recombinant human erythropoietins and cancer patients: updated meta-analysis of 57 studies including 9353 patients. J Natl Cancer Inst 98: 708–714 44. Bokemeyer C, Aapro MS, Courdi A, et al (2004) EORTC guidelines for the use of erythropoietic proteins in anaemic patients with cancer. Eur J Cancer 40: 2201–2216 45. Bokemeyer C, Aapro MS, Courdi A, et al (2007) EORTC guidelines for the use of erythropoietic proteins in anaemia patients with cancer: 2006 update. Eur J Cancer 43: 258–270 46. Bothwell TH, Bradlow BA, Jacobs P, et al (1964) Iron metabolism in scurvy with special reference to erythropoiesis. Br J Haematol 10: 50–58 47. Bothwell TH (1968) The control of iron absorption. Br J Haematol 14: 453– 456 48. Boven K, Stryker S, Knight J, et al (2005) The increased incidence of pure red cell aplasia with an Eprex formulation in uncoated rubber stopper syringes. Kidney Int 67: 2346–2353 49. Brugnara C (1998) Use of reticulocyte cellular indices in the diagnosis and treatment of hematological disorders. Int J Clin Lab Res 28: 1–11 50. Burns DL, Pomposelli JJ (1999) Toxicity of parenteral iron dextran therapy. Kidney Int 69 [Suppl]: S119–124 51. Canon JL, Vansteenkiste J, Bodoky G, et al (2006) Randomized, double-blind, active-controlled trial of every-3-week darbepoetin alfa for the treatment of chemotherapy-induced anemia. J Natl Cancer Inst 98: 273–284 52. Carabantes FJ, Benavides M, Trujillo R, et al (1999) Epoetin alpha in the prevention of anemia in cancer patients undergoing platinum-based chemotherapy. A prospective randomised study. Proc Am Soc Clin Oncol 18: 596a (Abstr 2303) 53. Caro JJ, Slas M, Ward A, et al (2001) Anemia as an independent prognostic factor for survival in patients with cancer. Cancer 91: 2214–2221 54. Carvalho G, Lefaucheur C, Cherbonnier C, et al (2005) Chemosensitization by erythropoietin through inhibition of the NF-κb rescue pathway. Oncogene 24: 737–745

494

M. R. Nowrousian

55. Casadevall N, Cournoyer D, Marsh J, et al (2004) Recommendations on haematological criteria for the diagnosis of epoetin-induced pure red cell aplasia. Eur J Haematol 73: 389–396 56. Casadevall N, Eckardt KU, Rossert J (2005) Epoetin-induced autoimmune pure red cell aplasia. J Am Soc Nephrol 16 [Suppl 1]: S67–S69 57. Cascinu S, Fedeli A, Fedeli SL, et al (1993) Cisplatin-associated anaemia treated with subcutaneous erythropoietin. A pilot study. Br J Cancer 67: 156–158 58. Cascinu S, Fedeli A, Del Ferro E, et al (1994a) Carboplatin associated anemia treated with subcutaneous erythropoietin. A pilot study. Oncol Rep 1: 169–172 59. Cascinu S, Fedeli A, Del Ferro E, et al (1994b) Recombinant human erythropoietin treatment in cisplatin-associated anemia: a randomized, double-blind trial with placebo. J Clin Oncol 12: 1058–1062 60. Cascinu S, Del Ferro E, Fedeli A, et al (1995) Recombinant human erythropoietin treatment in elderly cancer patients with cisplatin-associated anemia. Oncology 52: 422–426 61. Case DCJr, Bukowski RM, Carey RW, et al (1993) Recombinant human erythropoietin therapy for anemic cancer patients on combination chemotherapy. J Natl Cancer Inst 85: 801–806 62 Cavill I (2003) Intravenous iron as adjuvant therapy: a two-edged sword? Nephrol Dial Transplant 18 [Suppl 8]: viii24–viii28 63. Cazzola M, Messinger D, Battistel V, et al (1995) Recombinant human erythropoietin in the anemia associated with multiple myeloma or non-Hodgkin’s lymphoma: dose ﬁnding and identiﬁcation of predictors of response. Blood 86: 4446–4453 64. Cazzola M, Ponchio L, Pedrotti C, et al (1996) Prediction of response to recombinant human erythropoietin (rHuEpo) in anemia of malignancy. Haematologica 81: 434–441 65. Cazzola M, Beguin Y, Kloczko J, et al (2003) Once-weekly epoetin beta is highly effective in treating anaemia patients with lymphoproliferative erythropoietin production. Br J Haematol 122: 386–393 66. Ceelen W, Boterberg T, Smeets P, et al (2007) Recombinant human erythropoietin α modulates the effects of radiotherapy on colorectal cancer microvessels. Br J Cancer 96: 692–700 67. Chang J, Couture F, Young S, et al (2005) Weekly epoetin alfa maintains hemoglobin, improves quality of life, and reduces transfusion in breast cancer patients receiving chemotherapy. J Clin Oncol 23: 2597–2605 68. Charu V, Saidman B, Ben-Jacob A, et al (2007) A randomized, open-label, multicenter trial of immediate versus delayed intervention with darbepoetin alfa for chemotherapy-induced anemia. Oncologist 12: 1253–1263 69. Chertow GM, Mason PD, Vaage-Nilsen (2006) Update on adverse drug events associated with parenteral iron. Nephrol Dial Transplant 21: 378–382 70. Chu E, Einhorn LH, Lefebvre P (2006) Clinical beneﬁts of once-weekly epoetin alfa in anemic patients with colorectal cancer receiving chemotherapy. J Support Oncol 4: 243–250 71. Cook JD (1982) Clinical evaluation of iron deﬁciency. Semin Hematol 19: 6–18 72. Couture F, Turner AR, Melosky B, et al (2005) Prior red blood cell transfusion in cancer patients increase the risk of subsequent transfusions with or without recombinant human erythropoietin management. Oncologist 10: 63–71

rhEPO in anemic patients with solid tumors and chemotherapy

495

73. Crawford J, Cella D, Cleeland CS, et al (2002) Relationship between changes in hemoglobin level and quality of life during chemotherapy in anemic cancer patients receiving epoetin alfa therapy. Cancer 95: 888–895 74. Crawford J (2006) Erythropoietin-stimulating protein support and survival. Oncology (Huntingt) 20 [8 Suppl 6]: 39–43 75. Crawford J (2007) Erythropoietin: High proﬁle, high scrutiny. J Clin Oncol 25: 1021–1023 76. Crawford J, Robert F, Perry MC, et al (2007) A randomized trial comparing immediate versus delayed treatment of anemia with once-weekly epoetin alfa in patients with non-small lung cancer scheduled to receive ﬁrst-line chemotherapy. J Thorac Oncol 2: 210–220 77. Cullen P, Soffker J, Hopﬂ M, et al (1999) Hypochromic red cells and reticulocyte haemoglobin content as markers of iron-deﬁcient erythropoiesis in patients undergoing chronic haemodialysis. Nephrol Dial Transplant 14: 659–665 78. Daneryd P, Svanberg E, Körner U, et al (1998) Protection of metabolic and exercise capacity in unselected weight-losing cancer patients following treatment with recombinant erythropoietin: a randomized prospective study. Cancer Res 58: 5374–5379 79. de Campos E, Radford J, Steward W, et al (1995) Clinical and in vitro effects of recombinant human erythropoietin in patients receiving intensive chemotherapy for small-cell lung cancer. J Clin Oncol 13: 1623–1631 80. Del Mastro L, Venturini M, Lionetto R, et al (1997) Randomized phase III trial evaluating the role of erythropoietin in the prevention of chemotherapyinduced anemia. J Clin Oncol 15: 2715–2721 81. Demetri GD, Kris M, Wade J, et al (1998) Quality-of-life beneﬁt in chemotherapy patients treated with epoetin alfa is independent of disease response or tumor type: results from a prospective community oncology study. Procrit Study Group. J Clin Oncol 16: 3412–3425 82. Dunphy FR, Harrison BR, Dunleavy TL, et al (1999) Erythropoietin reduces anemia and transfusions. A randomised trial with or without erythropoietin during chemotherapy. Cancer 86: 1362–1367 83. Egrie JC, Browne JK (2001) Development and characterisation of novel erythropoiesis stimulating protein (NESP) Br J Cancer 84 [Suppl 1]: 3–10 84. Elliot P (2006) Pathogenesis of cardiotoxicity induced by anthracyclines. Semin Oncol 33 [Suppl 8]: S2–S7 85. EORTC Guidelines: Erythropoiesis-stimulating agents – ESAs: The 2007 position of the EORTC working party (2007). www.cancerworld.org/CancerWorld/ eventDetail.aspx?id_sito=1&id_stato=1&id=1795 86. Fermé C, Bastion Y, Brice P, et al (1997) Prognosis of patients with advanced Hodgkin’s disease: evaluation of four prognostic models using 344 patients included in the Groupe d’Etudes des Lymphomes de l’Adulte study. Cancer 80: 1124–1133 87. Fishbane S, Galgano C, Langley RC Jr, et al (1997) Reticulocyte hemoglobin content in the evaluation of iron status of hemodialysis patients. Kidney Int 52: 217–222 88. Fishbane S, Kowalski EA (2000) The comparative safety of intravenous iron dextran, iron saccharate, and sodium ferric gluconate. Semin Dial 13: 381– 384

496

M. R. Nowrousian

89. Food and Drug Administration. Department of Health and Human Services/Oncology Drugs Advirsory Committee May 4 2004 meeting notes [online]. Available from URL: http://www.fda.gov/ohrms/dockets/ac/cder04.html [Accessed 2006 May 7] 90. Ford P, Henry D, Mason B (1998) Iron status in anemic cancer patients. Proc Am Soc Clin Oncol 17: 58a 91. Freidlin B, Korn EL (2004) Erythropoietin to treat anaemia in patients with head and neck cancer. Lancet 363: 81 92. Gabrilove JL, Cleeland CS, Livingston RB, et al (2001) Clinical evaluation of once-weekly dosing of epoetin alfa in chemotherapy patients: Improvements in hemoglobin and quality of life are similar to three-times-weekly dosing. J Clin Oncol 19: 2875–2882 93. Gamucci T, Thorel MF, Frasca AM, et al (1993) Erythropoietin for the prevention of anaemia in neoplastic patients treated with cisplatin. Eur J Cancer 29A [Suppl 2]: S13–14 94. Gascón P (2005) Evaluating erythropoietic agents for the treatment of anaemia in the oncology setting. Eur J Cancer 41: 2601–2612 95. Gastaldello K, Vereerstraeten A, Nzame-Nze T, et al (1995) Resistance to erythropoietin in iron-overloaded haemodialysis patients can be overcome by ascorbic acid administration. Nephrol Dial Transplant 10: 44–47 96. Gemici C (2004) Erythropoietin to treat anaemia in patients with head and neck cancer. Lancet 363: 992–993 97. Gewirtz DA, Xu D, Teneille DW, et al (2006) Erythropoietin fails to interfere with the antiproliferative and cytotoxic effects of antitumor drugs. Clin Cancer Res 12: 2232–2238 98. Giancaspro V, Nuzziello M, Pallotta G, et al (2000) Intravenous ascorbic acid in hemodialysis patients with functional iron deﬁciency: a clinical trial. J Nephrol 13: 444–449 99. Gibbs MA (2000) Ascorbic acid use in hyporesponders to epoetin alfa. Nephrol Nurs J 27: 413–415 100. Glaspy J, Bukowski R, Steinberg D, et al (1997) Impact of therapy with epoetin alfa on clinical outcomes in patients with nonmyeloid malignancies during cancer chemotherapy in community oncology practice. Procrit Study Group. J Clin Oncol 15: 1218–1234 101. Glaspy J, Jadeja JS, Justice G, et al (2001) A dose-ﬁnding and safety study of novel erythropoiesis stimulating protein (NESP) for the treatment of anaemia in patients receiving multicycle chemotherapy. Br J Cancer 84 [Suppl 1]: 17–23 102. Glaspy JA, Jadeja JS, Justice G, et al (2002) Darbepoetin alfa given every I or 2 weeks alleviates anaemia associated with cancer chemotherapy. Br J Cancer 87: 268–276 103. Glaspy J (2005) Phase II clinical trials with darbepoetin: implications for clinicians. Best Pract Res Clin Haematol 18: 407–416 104. Glaspy J, Henry D, Patel R, et al (2005) Effects of chemotherapy on endogenous erythropoietin levels and the pharmacokinetics and erythropoietic response of darbepoetin alfa: A randomised clinical trial of synchronous versus asynchronous dosing of darbepoetin alfa. Eur J Cancer 41: 1136–1145 105. Glaspy J, Smith R, Aapro M, et al (2007) Results from a phase III, randomized, double-blind, placebo-controlled study of darbepoetin alfa for the treatment of

rhEPO in anemic patients with solid tumors and chemotherapy

106.

107.

108. 109. 110. 111.

112. 113.

114.

115. 116.

117. 118.

119.

120.

121.

122.

497

anemia in patients not receiving chemotherapy or radiotherapy. 2007 AACR Annual Meeting: (Abstr) Glimelius B, Linne T, Hoffman K, et al (1998) Epoetin beta in the treatment of anemia in patients with advanced gastrointestinal cancer. J Clin Oncol 16: 434–440 Golab J, Olszewska D, Mróz P, et al (2002) Erythropoietin restores the antitumor effectiveness of photodynamic therapy in mice with chemotherapy-induced anemia. Clin Cancer Res 8: 1265–1270 Goodnough LT, Skikne B, Brugnara C (2000) Erythropoietin, iron, and erythropoiesis. Blood 96: 823–833 Goodnough LT (2006) The relevance of iron in erythropoietin-stimulated erythropoiesis. Semin Hematol 43 [Suppl 6]: S3–S8 Groopman JE, Itri LM (1999) Chemotherapy-induced anemia in adults: incidence and treatment. J Natl Cancer Inst 91: 1616–1634 Grote T, Yeilding AL, Castillo R, et al (2005) Efﬁcacy and safety analysis of epoetin alfa in patients with small-cell lung cancer: A randomized, double-blind, placebo-controlled trial. J Clin Oncol 23: 9377–9386 Haddad R, Posner M (2004) Erythropoietin to treat anaemia in patients with head and neck cancer. Lancet 363: 79–80 Halstenson CE, Macres M, Katz SA, et al (1991) Comparative pharmacokinetics and pharmacodynamics of epoetin alfa and epoetin beta. Clin Pharmacol Ther 50: 702–712 Hardee ME, Rabbani ZN, Arcasoy MO, et al (2006) Erythropoietin inhibits apoptosis in breast cancer cells via an Akt-dependent pathway without modulating in vivo chemosensitivity. Mol Cancer Ther 5: 356–361 Hasegawa I, Tanaka K (1992) Serum erythropoietin levels in gynecologic cancer patients during cisplatin combination chemotherapy. Gynecol Oncol 46: 65–68 Hasenclever D, Diehl V (1998) A prognostic score for advanced Hodgkin’s disease. International Prognostic Factors Project on Advanced Hodgkin’s Disease. N Engl J Med 339: 1506–1514 Hazard JT, Drysdale JW (1977) Ferritinaemia in cancer. Nature 265: 755–756 Heatherington AC, Schuller J, Mercer AJ (2001) Pharmacokinetics of novel erythropoiesis stimulating protein (NESP) in cancer patients: preliminary report. Br J Cancer 84 [Suppl 1]: 11–16 Hedenus M, Vansteenkiste J, Kotasek D, et al (2005) Darbepoetin alfa for the treatment of chemotherapy-induced anemia: disease progression and survival analysis from four randomized, double-blind, placebo-controlled trials. J Clin Oncol 23: 6941–6948 Hedenus M, Birgegard G, Näsman P, et al (2007) Addition of intravenous iron to epoetin beta increases hemoglobin response and decreases epoetin dose requirement in anemic patients with lymphoproliferative malignancies: a randomized multicenter study. Leukemia 21: 627–632 Heiss MM, Tarabichi A, Delanoff C, et al (1996) Perisurgical erythropoietin application in anemic patients with colorectal cancer: A double-blind randomized study. Surgery 119: 523–527 Henke M, Laszig R, Rübe C, et al (2003) Erythropoietin to treat head and neck cancer patients with anaemia undergoing radiotherapy: randomised, doubleblind, placebo-controlled trial. Lancet 362: 1255–1260

498

M. R. Nowrousian

123. Henke M, Mattern D, Pepe M, et al (2006) Do erythropoietin receptors on cancer cells explain unexpected clinical ﬁndings? J Clin Oncol 24: 4708–4713 124. Henry DH, Abels RI (1994) Recombinant human erythropoietin in the treatment of cancer and chemotherapy-induced anemia: Results of double-blind and open-label follow-up studies. Semin Oncol 21 [Suppl 3]: 21–28 125. Henry D, Abels R, Larholt K (1995) Prediction of response to recombinant human erythropoietin (r-HuEPO/Epoetin-a) therapy in cancer patients. Blood 85: 1676–1678 126. Henry D, Mason B, Staddon A, et al (1998) Efﬁcacy of INFeD® plus recombinant human erythropoietin (RHUEPO) in treating anemia in cancer patients with suboptimal response to RHUEPO: a pilot study. Proc Am Soc Clin Oncol 17: 59a (Abstr) 127. Henry DH (1998) Supplemental iron: a key to optimizing the response of cancer-related anemia to rHuEPO? Oncologist 3: 275–278 128. Henry DH, Gordan LN, Charu V, et al (2006) Randomized, open-label comparison of epoetin alfa extended dosing (80,000 U Q2W) vs weekly dosing (40,000 U QW) in patients with chemotherapy-induced anemia. Curr Med Res Opin 22: 1403–1413 129. Henry DH, Dahl NV, Auerbach M (2007) Intravenous ferric gluconate signiﬁcantly improves response to epoetin alfa versus oral iron or no iron in anemic patients with cancer receiving chemotherapy. Oncologist 12: 231–242 130. Hensley ML, Lebeau D, Leon LF, et al (2001) Identiﬁcation of risk factors for requiring transfusion during front-line chemotherapy for ovarian cancer. Gynecol Oncol 81: 485–489 131. Hesketh PJ, Arena F, Patel D, et al (2004) A randomized controlled trial of darbepoetin alfa administered as a ﬁxed or weight-based dose using a front-loading schedule in patients with anemia who have nonmyeloid malignancies. Cancer 100: 859–868 132. Höckel M, Vaupel P (2001) Tumor hypoxia: deﬁnitions and current clinical, biologic, and molecular aspects. J Natl Cancer Inst 93: 266–276 133. Hudson JQ, Comstock TJ (2001) Considerations for optimal iron use for anemia due to chronic kidney disease. Clin Ther 23: 1637–1671 134. Jacobs C (2006) Iron metabolism pre and post the erythreopoietin era. Nephrol Ther [Suppl 5]: S313–S320 135. Jacobsen PB, Garland LL, Booth-Jones M, et al (2004) Relationship of hemoglobin levels to fatigue and cognitive functioning among cancer patients receiving chemotherapy. J Pain Symptom Manage 28: 7–18 136. James RD, Wilkinson PM, Belli F, et al (1992) Recombinant human erythropoietin in patients with ovarian carcinoma and anaemia secondary to cisplatin and carboplatin chemotherapy: preliminary results. Acta Haematol 87: 12–15 137. Janecka IP (2004) Erythropoietin to treat anaemia in patients with head and neck cancer. Lancet 363: 993–994 138. Jelkmann W, Laugsch M (2007) Problems in identifying functional erythropoietin receptors in cancer tissue. J Clin Oncol 25: 1627–1628 139. Johnson DW, Pollock CA, Macdougall IC (2007) Erythropoiesis-stimulating agent hyporesponsiveness. Nephrology 12: 321–330 140. Kaanders JHAM, van der Kogel AJ (2004) Erythropoietin to treat anaemia in patients with head and neck cancer. Lancet 363: 78–79

rhEPO in anemic patients with solid tumors and chemotherapy

499

141. Kaiserová H, den Hartog GI, Simunek T, et al (2006) Iron is not involved in oxidative stress-mediated cytotoxcity of doxorubicin and bleomycin. Br J Pharmacol 149: 920–930 142. Kaltwasser JP, Gottschalk R (1999) Erythropoietin and iron. Kidney Int [Suppl] 69: S49–56 143. Kasper C, Terhaar A, Fossa A, et al (1997) Recombinant human erythropoietin in the treatment of cancer-related anaemia. Eur J Haematol 58: 251–256 144. Kato A, Hamada M, Suzuki T, et al (2001) Effect of weekly or successive iron supplementation on erythropoietin doses in patients receiving hemodialysis. Nephron 89: 110–112 145. Katodritou, E, Terpos E, Zervas K, et al (2007) Hypochromic erythrocytes (%): a reliable marker for recognizing iron-restricted erythropoiesis and predicting response to erythropoietin in anemic patients with myeloma and lymphoma. Ann Hematol 86: 369–376 146. Katz O, Barzilay E, Skaat A, et al (2005) Erythropoietin induced tumour mass reduction in murine lymphoproliferative models. Acta Haematol 114: 177– 179 147. Kim KH, Backx P (2005) The protective role of erythropoietin against doxorubicin-induced cardiomyopathy. J Clin Oncol 23 [16 Suppl]: 850s (Abstr) 148. Kirkpatrick JP, Hardee ME, Snyder SA, et al (2006) The effect of darbepoetin alfa on growth, oxygenation and radioresponsiveness of a breast adenocarcinoma. Radiat Res 165: 192–201 149. Kokhaei P, Abdalla AO, Hansson L, et al (2007) Expression of erythropoietin receptor and in vitro functional effects of epoetins in B-cell malignancies. Clin Cancer Res 13: 3536–3544 150. Konijn AM, Hershko C (1977) Ferritin synthesis in inﬂammation. I. Pathogenesis of impaired iron release. Br J Haematol 37: 7–16 151. Kosch M, Bahner U, Bettger H, et al (2001) A randomized, controlled parallelgroup trial on efﬁcacy and safety of iron sucrose (Venofer) vs iron gluconate (Ferrlecit) in haemodialysis patients treated with rHuEpo. Nephrol Dial Transplant 16: 1239–1244 152. Kotasek D, Steger G, Faught W, et al (2003) Darbepoetin alfa administered every 3 weeks alleviates anaemia in patients with solid tumours receiving chemotherapy; results of a double-blind, placebo-controlled, randomised study. Eur J Cancer 39: 2026–2034 153. Kumar SM, Acs G, Fang D, et al (2005) Functional erythropoietin autocrine loop in melanoma. Am J Pathol 166: 823–830 154. Kwok JC, Richardson DR (2003) Anthracyclines induce accumulation of iron in ferritin in myocardial and neoplastic cells: Inhibition of the ferritin iron mobilization pathway. Mol Pharmacol 63: 849–861 155. LaMontagne KR, Butler J, Marshall DJ, et al (2006) Recombinant epoetins do not stimulate tumor growth in erythropoietin receptor-positive breast carcinoma models. Mol Cancer Ther 5: 347–355 156. Larsson AM, Landberg G, Pahlman S, et al (2004) Erythropoietin enhances response to treatment in patients with advanced breast cancer. Acta Oncol 43: 594–597 157. Lerchenmueller Ch, Husseini F, Gaede B, et al (2006) Intravenous (IV) iron supplementation in patients with chemotherapy-induced anemia (CIA) receiv-

500

158. 159.

160. 161.

162.

163. 164. 165. 166.

167. 168.

169.

170. 171. 172.

173.

174.

M. R. Nowrousian ing darbepoetin alfa every 3 weeks (Q3W): iron parameters in a randomized controlled trial. Blood 108: 445 (Abstr 1552) Leyland-Jones B, Mahmud S (2004) Erythropoietin to treat anaemia in patients with head and neck cancer. Lancet 363: 80 Leyland-Jones B, Semiglazov V, Pawlicki M, et al (2005) Maintaining normal hemoglobin levels with epoetin alfa in mainly nonanemic patients with metastatic breast cancer receiving ﬁrst-line chemotherapy: A survival study. J Clin Oncol 23: 5960–5972 Lipschitz DA, Bothwell TH, Seftel HC, et al (1971) The role of ascorbic acid in the metabolism of storage iron. Br J Haematol 20: 155–163 Littlewood TJ, Bajetta E, Nortier JWR, et al (2001) Effects of epoetin alfa on hematologic parameters and quality of life in cancer patients receiving nonplatinum chemotherapy: results of a randomized, double-blind, placebocontrolled trial. J Clin Oncol 19: 2865–2874 Littlewood TJ, Zagari M, Pallister C, et al (2003) Baseline and early treatment factors are not clinically useful for predicting individual response to erythropoietin in anemic cancer patients. Oncology (Huntingt) 8: 99–107 Liu WM, Powles T, Shamash J, et al (2004) Effect of haemopoietic growth factors on cancer cell lines and their role in chemosensitivity. Oncogene 23: 981–990 Lövey J, Kenessey I, Erzsébet R, et al (2007) [Human recombinant erythropoietin-α increases the efﬁcacy of irradiation in preclinical model]. Hung Oncol 51: 53–61 Ludwig H, Fritz E, Kotzmann H, et al (1990) Erythropoietin treatment of anemia associated with multiple myeloma. N Engl J Med 322: 1693–1699 Ludwig H, Fritz E, Leitgeb C, et al (1993) Erythropoietin treatment for chronic anemia of selected hematological malignancies and solid tumors. Ann Oncol 4: 161–167 Ludwig H, Fritz E, Leitgeb C, et al (1994) Prediction of response to erythropoietin treatment in chronic anemia of cancer. Blood 84: 1056–1063 Ludwig H, Sundal E, Pecherstorfer M, et al (1995) Recombinant human erythropoietin for the correction of cancer associated anemia with and without concomitant cytotoxic chemotherapy. Cancer 76: 2319–2329 Ludwig H, Van Belle S, Barrett-Lee P, et al (2004) The European Cancer Anaemia Survey (ECAS): A large, multinational, prospective survey deﬁning the prevalence, incidence, and treatment of anaemia in cancer patients. Eur J Cancer 40: 2293–2306 Ludwig H (2006a) Iron metabolism and iron supplementation in anemia of cancer. Semin Hematol 43 [Suppl 6]: S13–S17 Ludwig H (2006b) Epoetin beta in oncology: examining the current evidence. Future Oncol 2: 21–38 Ludwig H, Crawford J, Österborg A, et al (2007) Patient-level integrated analysis of data from 6 randomized, double-blind, placebo-controlled trials of darbepoetin alfa (DA) in patients (pts) with chemotherapy-induced anemia (CIA). Eur J Cancer 42 [Suppl 5]: (Abstr 1104) Lyman GH, Glaspy J (2006) Are there clinical beneﬁts with early erythropoietic intervention for chemotherapy-induced anemia? A systematic review. Cancer 106: 223–233 Macdougall IC, Tucker B, Thompson J, et al (1996) A randomized controlled study of iron supplementation in patients treated with erythropoietin. Kidney Int 50: 1694–1699

rhEPO in anemic patients with solid tumors and chemotherapy

501

175. Macdougall IC (1999) Strategies for iron supplementation: oral versus intravenous. Kidney Int [Suppl] 69: S61–66 176. Macdougall IC, Chandler G, Elston O, et al (1999a) Beneﬁcial effects of adopting an aggressive intravenous iron policy in a hemodialysis unit. Am J Kidney Dis 34: S40–46 177. Macdougall IC, Gray ST, Elston O, et al (1999b) Pharmacokinetics of novel erythropoiesis stimulating protein compared with epoetin alfa in dialysis patients. J Am Soc Nephrol 10: 2392–2395 178. Macdougall IC, Cooper AC (2002) Erythropoietin resistance: the role of inﬂammation and pro-inﬂammatory cytokines. Nephrol Dial Transplant 17 [Suppl 11]: 39–43 177. Macdougall IC (2004) Could anti-inﬂammatory cytokine therapy improve poor treatment outcomes in dialysis patients? Nephrol Dial Transplant 19 [Suppl 5]: v73–v78 178. Macdougall IC (2006) Experience with intravenous iron in nephrology. Semin Hematol 43 [Suppl 6]: S9–S12 179. Major A, Bauer C, Breymann C, et al (1994) rh-Erythropoietin stimulates immature reticulocyte release in man. Br J Haematol 87: 605–608 180. Maynor L, Brophy DF (2007) Risk of infection with intravenous iron therapy. Ann Pharmacother 41: 1476–1480 181. McBroom JW, Acs G, Rose GS, et al (2005) Erythropoietin receptor function and expression in epithelial ovarian carcinoma. Gynecol Oncol 99: 571–577 182. McClellan WM, Frankenﬁeld DL, Wish JB, et al (2001) Subcutaneous erythropoietin results in lower dose and equivalent hematocrit levels among adult hemodialysis patients: results from the 1998 End-Stage Renal Disease Core Indicators Project. Am J Kidney Dis 37: E36 183. Melendez O (2000) Intravenous vitamin C for erythropoietin resistance. Semin Dial 13: 335–336 184. Miller CB, Platanias LC, Mills SR, et al (1992) Phase I-II trial of erythropoietin in the treatment of cisplatin-associated anemia. J Natl Cancer Inst 84: 98– 103 185. Minotti G, Menna P, Salvatorelli E, et al (2004) Anthracyclines: Molecular advances and pharmacologic developments in antitumor activity and cardiotoxicity. Pharmacol Rev 56: 185–229 186. Mittelman M, Neumann D, Peled A, et al (2001) Erythropoietin induces tumor regression and antitumor immune responses in murine myeloma models. Proc Natl Acad Sci USA 98: 5181–5186 187. Moebus V, Lueck H, Thomssen C, et al (2007) The impact of epoetin-alpha on anemia, red blood cell (RBC) transfusions, and survival in breast cancer patients (pts) treated with dose-dense sequential chemotherapy: Mature results of an AGO phase III study (ETC trial). J Clin Oncol 25 [Suppl]: 20s (Abstr) 188. Mohyeldin A, Lu H, Dalgard C, et al (2005) Erythropoietin signaling promotes invasiveness of human head and neck squamous cell carcinoma. Neoplasia 7: 537–543 189. Morishima Y, Ogura M, Yoneda S, et al (2006) Once-weekly epoetin-beta improves hemoglobin levels in cancer patients with chemotherapy-induced anemia: A randomized, double-blind, dose-ﬁnding study. Jpn J Clin Oncol 36: 655–661 190. Morreale A, Plowman B, DeLattre M, et al (2004) Clinical and economic comparison of epoetin alfa and darbepoetin alfa. Curr Med Res Opin 20: 381–395

502

M. R. Nowrousian

191. Moullet I, Salles G, Ketterer N, et al (1998) Frequency and signiﬁcance of anemia in non-Hodgkin’s lymphoma patients. Ann Oncol 9: 1109–1115 192. NCCN Cancer- and treatment-related anemia (2007). www.nccn.org/ professionals/physician:gls/PDF/anemia.pdf 193. Ning S, Hartley C, Molineux G, et al (2005) Darbepoietin alfa potentiates the efﬁcacy of radiation therapy in mice with corrected or uncorrected anemia. Cancer Res 65: 284–290 194. Nissenson AR (1997) Achieving target hematocrit in dialysis patients: new concepts in iron management. Am J Kidney Dis 30: 907–911 195. Nowrousian MR, Kasper C, Oberhoff C, et al (1996) Pathophysiology of cancer-related anemia. In: Smyth JF, Boogaerts MA, Ehmer BRM (eds) rhErythropoietin in cancer supportive treatment. Marcel Dekker, New York, pp 13–34 196. Nowrousian MR (1998) Recombinant human erythropoietin in the treatment of cancer-related or chemotherapy-induced anaemia in patients with solid tumours. Medical Oncol 15 [Suppl 1]: 19–28 197. Oberhoff C, Neri B, Amadori D, et al (1998) Recombinant human erythropoietin in the treatment of chemotherapy-induced anemia and prevention of transfusion requirement associated with solid tumors: A randomised controlled study. Ann Oncol 9: 255–260 198. Oberhoff C, Krumreich B, Petry KU, et al (2000) Effekt von rekombinantem humanen Erythropoetin auf den Transfusionsbedarf und die Hämoglobinkonzentration bei Patienten mit soliden Tumoren und chemotherapieinduzierter Anämie. Tumordiagn u Ther 21: 15–25 199. Oivanen TM (1996) Plateau phase in multiple myeloma: an analysis of longterm follow-up of 432 patients. Finnish Leukaemia Group. Br J Haematol 92: 834–839 200. Ordonez A, Gonzalez-Baron M, Isla D, et al (2005) Epoetin beta treatment to prevent anemia in solid tumor patients receiving platinum-based chemotherapy. J Clin Oncol 23 [Suppl]: 8230 (Abstr) 201. Orhan B, Yalcin S, Nurlu G, et al (2004) Erythropoietin against cisplatin-induced peripheral neurotoxicity in rats. Med Oncol 21: 197–203 202. Österborg A, Boogaerts MA, Cimino R, et al (1996) Recombinant human erythropoietin in transfusion-dependent anemic patients with multiple myeloma and non-Hodgkin’s lymphoma–a randomized multicenter study. The European Study Group of Erythropoietin (Epoetin Beta) Treatment in Multiple Myeloma and Non-Hodgkin’s Lymphoma. Blood 87: 2675–2682 203. Österborg A, Brandberg Y, Hedenus M (2005) Impact of epoetin-β on survival of patients with lymphoproliferative malignancies: long-term follow up of a large randomized study. Br J Haematol 129: 206–209 204. Österborg A, Aapro M, Cornes P, et al (2007) Preclinical studies of erythropoietin receptor expression in tumour cells: impact on clinical use of erythropoietic proteins to correct cancer-related anaemia. Eur J Cancer 43: 510–519 205. Overgaard J, Hoff C, Sand Hansen H, et al (2007) Randomized study of the importance of novel erythropoiesis stimulating protein (Aranesp®) for the effect of radiotherapy in patients with primary squamous cell carcinoma of the head and neck (HNSCC) – the Danish Head and Neck Cancer Group DAHANCA 10 randomized trial. Eur J Cancer 5 [Suppl 6]: 7 (Abstr 6LB#)

rhEPO in anemic patients with solid tumors and chemotherapy

503

206. Palumbo C, Battisti S, Carbone D, et al (2007) Recombinant erythropoietin differently affects proliferation of mesothelioma cells but not sensitivity to cisplatin and pemetrexed. Cancer Chemother Pharmacol (in press) 207. Pawlicki M, Jassem J, Bosze P, et al (1997) A multicenter study of recombinant human erythropoietin (epoetin alpha) in the management of anemia in cancer patients receiving chemotherapy. Anticancer Drugs 8: 949–957 208. Ping F, Murat O (2005) Erythropoietin protects cardiomyocytes not only during hypoxic injury but also against doxorubicin-induced apoptosis: a novel effect mediated by phosphatidylinositol 3-kinase pathway. Blood 106: (Abstr 3142) 209. Pirker R, Lehnert M, Minar W (2006) Once-weekly epoetin beta (30,000 IU) in anemic patients with lung cancer receiving chemotherapy. Lung Cancer 55: 89–94 210. Pirker R, Ramlau R, Schuette W, et al (2007) A phase 3 randomized, doubleblind, placebo-controlled study of patients with previously untreated extensivestage small cell lung cancer (SCLC) treated with platinum plus etoposide chemotherapy with or without darbepoetin alfa. J Thorac Oncol 2 [Suppl 4]: S433 (Abstr PD6–3-6) 211. Platanias LC, Miller CB, Mick R, et al (1991) Treatment of chemotherapyinduced anemia with recombinant human erythropoietin in cancer patients. J Clin Oncol 9: 2021–2026 212. Ponka A, Kuhlback B (1983) Serum ascorbic acid in patients undergoing chronic hemodialysis. Acta Med Scand 213: 305–307 213. Porter JC, Leahey A, Polise K, et al (1996) Recombinant human erythropoietin reduces the need for erythrocyte and platelet transfusions in pediatric patients with sarcoma: a randomized, double-blind, placebo-controlled trial. J Pediatr 129: 656–660 214. Quirt I, Robeson C, Lau CY, et al (2001) Epoetin alfa therapy increases hemoglobin levels and improves quality of life in patients with cancer-related anemia who are not receiving chemotherapy and patients with anemia who are receiving chemotherapy. J Clin Oncol 19: 4126–4134 215. Quirt I, Kovacs M, Couture F, et al (2006) Patients previously transfused or treated with epoetin alfa at low baseline hemoglobin are at higher risk for subsequent transfusion: an integrated analysis of the Canadian experience. Oncologist 11: 73–82 216. Ragione FD, Cucciolla V, Borriello A, et al (2007) Erythropoietin receptors in cancer cells: a still open question. J Clin Oncol 25: 1812–1813 217. Ray-Coquard I, Le Cesne A, Rubio MT, et al (1999) Risk model for severe anemia requiring red blood cell transfusion after cytotoxic conventional chemotherapy regimens. The Elypse 1 Study Group. J Clin Oncol 17: 2840–2846 218. Ray-Coquard I, Audhuy B, Facon T, et al (2007) Epoetin beta therapy in anemic patients with solid tumor or non myeloid hematological malignancies receiving chemotherapy: results of a large prospective cohort study. Eur J Cancer 5 [Suppl 4]: 154 (Abstr 1140) 219. Razzouk BI, Hord JD, Hockenberry M (2006) Double-blind, placebo-controlled study of quality of life, hematologic end points, and safety of weekly epoetin alfa in children with cancer receiving myelosuppressive chemotherapy. J Clin Oncol 24: 3583–3589

504

M. R. Nowrousian

220. Rearden T, Schwartzberg L, Mirtsching B, et al (2007) Evaluation of extended dosing intervals versus weekly dosing of darbepoetin alfa (DA): A phase 2 study in cancer patients (pts) with chemotherapy-induced anemia (CIA). Eur J Cancer 5 [Suppl 4]: 154 Abstr 1141 221. Reed N, Chan S, Hayward C, et al (2005) Impact of epoetin beta on the survival of anemic patients with ovarian cancer receiving platinum-based chemotherapy. J Clin Oncol 23 [Suppl]: 479s, (Abstr 5102) 222. Rizzo JD, Somerﬁeld MR, Hagerty KL, et al (2007) American Society of Clinical Oncology/American Society of Hematology 2007 Clinical Practice Guideline Update on the Use of Epoetin und Darbepoetin. J Clin Oncol 25: 1–17 223. Rosenzweig MQ, Bender CM, Lucke JP, et al (2004) The decision to prematurely terminate a trial of R-HuEPO due to thrombotic events. J Pain Symptom Manage 27: 185–190 224. Ross SD, Allen JE, Henry DH, et al (2006) Clinical beneﬁts and risks associated with epoetin and darbepoetin in patients with chemotherapy-induced anemia: A systematic review of the literature. Clin Ther 28: 801–831 225. Santini D, Vincenzi B, La Cesa A, et al (2005) A new dose-intense epoetin alfa regimen effective in anemic cancer patients receiving chemotherapy: an openlabel, non randomized, pilot study. Anticancer Res 25: 669–674 226. Savonije JH, van Groeningen CJ, van Bochove A, et al (2005) Effects of early intervention with epoetin alfa on transfusion requirement, hemoglobin level and survival during platinum-based chemotherapy: results of a multicenter randomised controlled trial. Eur J Cancer 41: 1560–1569 227. Savonije JH, van Groeningen CJ, Wormhoudt LW, et al (2006) Early intervention with epoetin alfa during platinum-based chemotherapy: an analysis of quality-of-life results of a multicenter, randomized, controlled trial compared with population normative data. Oncologist 11: 197–205 228. Sawabe Y, Kikuno K, Iseki T, et al (1996) Changes in serum erythropoietin and the reticulocyte count during chemotherapy for leukemias. Eur J Haematol 57: 384–388 229. Schaefer RM, Schaefer L (1999) Hypochromic red blood cells and reticulocytes. Kidney Int [Suppl] 69: S44–48 230. Schapira L, Antin JH, Ransil BJ, et al (1990) Serum erythropoietin levels in patients receiving intensive chemotherapy and radiotherapy. Blood 76: 2354–2359 231. Sevelda P, Kurz C, Marth C, et al (1996) Prospective randomized placebo controlled trial of erythropoietin (Erypo) in patients with chronic tumor anemia and gynecological cancer. Proc Am Soc Clin Oncol 15: 287 (Abstr) 232. Sezer S, Özdemir FN, Yakupoglu U, et al (2002) Intravenous ascorbic acid administration for erythropoietin-hyporesponsive anemia in iron loaded hemodialysis patients. Artiﬁcial Organs 26: 366–370 233. Shannon AM, Bouchier-Hayes DJ, Condron CM, et al (2005) Correction of anaemia through the use of darbepoetin alfa improves chemotherapeutic outcome in a murine model of Lewis lung carcinoma. Br J Cancer 93: 224–232 234. Sharples EJ, Thiemermann C, Yaqoob M (2005) Mechanisms of disease: cell death in acute renal failure and emerging evidence for a protective role of erythropoietin. Nat Clin Pract Nephrol 1: 87–97 235. Sigounas G, Sallah S, Sigounas VY (2004) Erythropoietin modulates the anticancer activity of chemotherapeutic drugs in a murine lung cancer model. Cancer Lett 214: 171–179

rhEPO in anemic patients with solid tumors and chemotherapy

505

236. Silver DF, Piver MS (1999) Effects of recombinant human erythropoietin on the antitumor effect of cisplatin in SCID mice bearing human ovarian cancer: A possible oxygen effect. Gynecol Oncol 73: 280–284 237. Silverberg DS, Blum M, Agbaria Z, et al (2001) The effect of i.v. iron alone or in combination with low-dose erythropoietin in the rapid correction of anemia of chronic renal failure in the predialysis period. Clin Nephrol 55: 212–219 238. Silverstein SB, Rodgers GM (2004) Parenteral iron therapy options. Am J Hematol 76: 74–78 239. Sinclair AM, Todd MD, Forsythe K, et al (2007) Expression and function of erythropoietin receptors in tumors. Cancer 110: 477–488 240. Sizer KC (2007) Heat shock protein 70, erythropoietin, and cancer. J Clin Oncol 25: 4326 (Letter) 241. Skillings JR, Sridhar FG, Wong C, et al (1993) The frequency of red cell transfusion for anemia in patients receiving chemotherapy. A retrospective cohort study. Am J Clin Oncol 16: 22–25 242. Skillings JR, Rogers-Melamed I, Nabholtz JM, et al (1995) An epidemiological review of anaemia in cancer chemotherapy in Canada. Eur J Cancer 31A [Suppl 5]: S183 243. Smith RE, Jaiyesimi IA, Meza LA, et al (2001) Novel erythropoiesis stimulating protein (NESP) for the treatment of anemia of chronic disease associated with cancer. Br J Haematol 84: 24–30 244. Smith RE Jr, Tchekmedyian NS, Chan D, et al (2003) A dose- and scheduleﬁnding study of darbepoetin alpha for the treatment of chronic anaemia of cancer. Br J Cancer 88: 1851–1858 245. Sowade B, Sowade O, Mocks J, et al (1998) The safety of treatment with recombinant human erythropoietin in clinical use: a review of controlled studies. Int J Mol Med 1: 303–314 246. Spaeth D, Quoix E, Moullet I, et al (2007) Epoetin beta once-weekly (QW) treatment in anaemia patients with solid tumour receiving chemotherapy. Eur J Cancer 5 [Suppl]: 149 (Abstr 1126) 247. Steensma DP, Molina R, Sloan JA, et al (2006) Phase III Study of two different dosing schedules of erythropoietin in anemic patients with cancer. J Clin Oncol 24: 1079–1089 248. Steinmetz T, Hellmich M, Neise M, et al (2007) Prediction of the responsiveness to treatment with erythropoiesis-stimalting factors: a prospective clinical study in patients with solid tumors. The Oncologist 12: 748–755 249. Stenvinkel P (2003) Anaemia and inﬂammation: what are the implications for the nephrologist? Nephrol Dial Transplant 18 [Suppl 8]: viii17–viii22 250. Straus DJ, Testa MA, Sarokhan BJ, et al (2006) Quality-of-life and health beneﬁts of early treatment of mild anemia: a randomized trial of epoetin alfa in patients receiving chemotherapy for hematologic malignancies. Cancer 107: 1909–1917 251. Strauss H, Haensgen G, Dunst J, et al (2005) Effects of anaemia correction with epoetin beta in patients with advanced cervical cancer and radiochemotherapy. J Clin Oncol 23 [1 Suppl]: 5121 (Abstr) 252. Stüben G, Thews O, Pöttgen C, et al (2001) Recombinant human erythropoietin increases the radiosensitivity of xenografted human tumours in anaemic nude mice. J Cancer Res Clin Oncol 127: 346–350

506

M. R. Nowrousian

253. Stüben G, Thews O, Pöttgen C, et al (2003a) Impact of anemia prevention by recombinant human erythropoietin on the sensitivity of xenografted glioblastomas to fractionated irradiation. Strahlenther Onkol 179: 620–625 254. Stüben G, Pöttgen C, Knühmann K, et al (2003b) Erythropoietin restores the anemia-induced reduction in radiosensitivity of experimental human tumors in nude mice. Int J Radiat Oncol Biol Phys 55: 1358–1362 255. Sturm B, Laggner H, Ternes N, et al (2005) Intravenous iron preparations and ascorbic acid: Effects on chelatable and bioavailable iron. Kidney Int 67: 1161–1170 255. Tarng DC, Huang TP (1998) A parallel, comparative study of intravenous iron versus intravenous ascorbic acid for erythropoietin-hyporesponsive anaemia in haemodialysis patients with iron overload. Nephrol Dial Transplant 13: 2867–2872 256. Tarng DC, Wei YH, Huang TP, et al (1999) Intravenous ascorbic acid as an adjuvant therapy for recombinant erythropoietin in hemodialysis patients with hyperferritinemia. Kidney Int 55: 2477–2486 257. Tarng DC, Huang TP, Wei YH (2001) Erythropoietin and iron: the role of ascorbic acid. Nephrol Dial Transplant 16: 35–39 258. Taylor K, Ganly P, Charu V, et al (2005) Randomized, double-blind, placebocontrolled study of darbepoetin alfa every 3 weeks for the treatment of chemotherapy-induced anemia. Blood 106: (Abstr 3556) 260. Teicher BA, Holden SA, al-Achi A, et al (1990) Classiﬁcation of antineoplastic treatments by their differential toxicity toward putative oxygenated and hypoxic tumor subpopulations in vivo in the FSaIIC murine ﬁbrosarcoma. Cancer Res 50: 3339–3344 261. Teicher BA (1994) Hypoxia and drug resistance. Cancer Metastasis Rev 13: 139–168 262. Teicher BA (1995) Physiologic mechanisms of therapeutic resistance. Blood ﬂow and hypoxia. Hematol Oncol Clin North Am 9: 475–506 263. ten Bokkel Huinink WW, de Swart CA, van Toorn DW, et al (1998) Controlled multicenter study of the inﬂuence of subcutaneous recombinant hum erythropoietin on anaemia and transfusion dependency in patients with ovarian carcinoma treated with platinum-based chemotherapy. Med Oncol 15: 174–182 264. Tessitore N, Solero GP, Lippi G, et al (2001) The role of iron status markers in predicting response to intravenous iron in haemodialysis patients on maintenance erythropoietin. Nephrol Dial Transplant 16: 1416–1423 265. Thatcher N (1998) Management of chemotherapy-induced anemia in solid tumors. Semin Oncol 25 [Suppl 7]: 23–26 266. Thatcher N, De Campos ES, Bell DR, et al (1999) Epoetin alpha prevents anaemia and reduces transfusion requirements in patients undergoing primarily platinum-based chemotherapy for small cell lung cancer. Br J Cancer 80: 396–402 267. Thews O, Koenig R, Kelleher DK, et al (1998) Enhanced radiosensitivity in experimental tumours following erythropoietin treatment of chemotherapyinduced anaemia. Br J Cancer 78: 752–756 268. Thews O, Kelleher K, Vaupel P (2001) Erythropoietin restores the anemiainduced reduction in cyclophosphamide cytotoxicity in rat tumors. Cancer Res 61: 1358–1361

rhEPO in anemic patients with solid tumors and chemotherapy

507

269. Thews O, Gassner B, Kelleher DK, et al (2007) Impact of hypoxic and acidic extracellular conditions on cytotoxicity of chemotherapeutic drugs. Adv Exp Med Biol 599: 155–161 270. Tomida A, Tsuruo T (1999) Drug resistance mediated by cellular stress response to the microenvironment of solid tumors. Anticancer Drug Des 14: 169–177 271. Tóvári J, Gilly R, Rásó E, et al (2005) Recombinant human erythropoietin α targets intratumoral blood vessels, improving chemotherapy in human xenograft models. Cancer Res 65: 7186–7193 272. Tsukuda M, Mochimatsu I, Nagahara T, et al (1993) Clinical application of recombinant human erythropoietin for treatments in patients with head and neck cancer. Cancer Immunol Immunother 36: 52–56 273. Tullai J, Chitikila C, Gechtman Z, et al (2004) Erythropoietin did not activate the JAK/STAT or the ERK 1/2 pathway in human colorectal cell line, HT-29. Blood 104: 4221 (Abstr) 274. Uchida M, Watanabe T, Kunitama M, et al (2004) Erythropoietin overcomes imatinib-induced apoptosis and induces erythroid differentiation in TF-1/bcr-abl cells. Stem Cells 22: 609–616 275. Van Wyck DB, Cavallo G, Spinowitz BS, et al (2000) Safety and efﬁcacy of iron sucrose in patients sensitive to iron dextran: North American clinical trial. Am J Kidney Dis 36: 88–97 276. Vandebroek A, Gaede B, Altintas S, et al (2006) A randomized open-label study of darbepoetin alfa administered every 3 weeks with or without parenteral iron in anemic subjects with nonmyeloid malignancies receiving chemotherapy. J Clin Oncol 24 [Suppl]: 496s (Abstr 8612) 277. Vansteenkiste J, Pirker R, Massuti B, et al (2002) Double-blind, placebo-controlled, randomized phase III trial of darbepoetin alfa in lung cancer patients receiving chemotherapy. J Natl Cancer Inst 94: 1211–1220 278. Vansteenkiste J, Tomita D, Rossi G, et al (2004) Darbepoetin alfa in lung cancer patients on chemotherapy: a retrospective comparison of outcomes in patients with mild versus moderate-to-servere anaemia at baseline. Support Care Cancer 12: 253–262 279. Vansteenkiste J, Wauters I (2005) The use of darbepoetin alfa for the treatment of chemotherapy-induced anaemia. Expert Opin Pharmacother 6: 429–440 280. Vansteenkiste J (2006) Every 3 weeks dosing with darbepoetin alfa: A new paradigm in anaemia management. Cancer Treat Rev 32 [Suppl 2]: S11–S15 281. Vaupel P, Mayer A (2004) Erythropoietin to treat anaemia in patients with head and neck cancer. Lancet 363: 992 282. Vaupel P, Dunst J, Engert A, et al (2005) Effects of recombinant human erythropoetin (rHuEPO) on tumor control in patients with cancer-induced anemia. Onkologie 28: 216–221 283. Vaupel P, Mayer A, Höckel M (2006) Impact of hemoglobin levels on tumor oxygenation: the higher, the better? Strahlenther Onkol 186: 63–71 284. Verhulst D, Pat K, Anrys B, et al (2007) A prospective observation study of treatment of chemotherapy-induced anaemia with darbepoetin alfa every 3 weeks: the OASIS (Observational Aranesp® Survey to Investigate the q3w Schedule) study. Eur J Cancer 5 [Suppl 4]: 158 (Abstr 1155) 285. Vesey DA, Cheung C, Pat B, et al (2004) Erythropietin protects against ischaemic acute renal injury. Nephrol Dial Transplant 19: 348–355

508

M. R. Nowrousian: rhEPO in anemic patients

286. Vogel S (2000) Beneﬁts of early utilization of intravenous iron. Nephrol Nurs J 27: 61–65 287. Voigtmann R, Nowrousian MR, Essers U, et al (1996) rhEPO in the treatment of anemia associated with malignancy. In: Smyth JF, Boogaerts MA, Ehmer BRM (eds) rhErythropoietin in cancer supportive treatment. Marcel Dekker, New York, pp 113–127 288. Voravud N, Sriuranpong V, Suwanrusme H (2007) Antianemic effect of once weekly regimen of epoetin alfa 40,000 units in anemic cancer patients receiving chemotherapy. J Med Assoc Thai 90: 1082–1088 289. Weinberg ED (1996) The role of iron in cancer. Eur J Cancer Prev 5: 19–36 290. Weinberg ED (1999) Iron therapy and cancer. Kidney Int [Suppl] 69: S131–134 291. Weiss G (1999) Iron and anemia of chronic disease. Kidney Int [Suppl] 69: S12–17 292. Weiss G, Gordeuk VR (2005) Beneﬁts and risks of iron therapy for chronic anaemias. Eur J Clin Invest 35 [Suppl 3]: 36–45 293. Westin SN, Skinner EN, Funk MJ, et al (2007) Incidence of symptomatic deep venous thrombosis with epoetin alfa or darbepoetin alfa treatment of anemia in patients with ovarian or primary peritoneal cancer. Gynecol Oncol 105: 414–417 294. Westenfelder C, Baranowski RL (2000) Erythropoietin stimulates proliferation of human renal carcinoma cells. Kidney Int 58: 647–657 295. Westphal G, Niederberger E, Blum C, et al (2002) Erythropietin and G-CSF receptors in human tumor cells: expression and aspects regarding functionality. Tumori 88: 150–159 296. Witzig TE, Silberstein PT, Loprinzi CL, et al (2005) Phase III, randomized, double-blind study of epoetin alfa compared with placebo in anemic patients receiving chemotherapy. J Clin Oncol 23: 2606–2617 297. Wouters A, Pauwels B, Lardon F, et al (2007) Review: Implications of in vitro research on the effect of radiotherapy and chemotherapy under hypoxic conditions. The Oncologist 12: 690–712 298. Wright JR, Ung CY, Julian JA, et al (2007) Randomized, double-blind, placebocontrolled trial of erythropoetin in non-small-cell lung cancer with diseaserelated anemia. J Clin Oncol 25: 1027–1032 299. Wun T, Law L, Harvey D, Sieracki B, et al (2003) Increased incidence of symptomatic venous thrombosis in patients with cervical carcinoma treated with concurrent chemotherapy, radiation, and erythropoietin. Cancer 98: 1514–1520 300. Wurnig C, Windhager R, Schwarmeis E, et al (1996) Prevention of chemotherapyinduced anemia by the use of erythropoietin in patients with primary malignant bone tumors (a double-blind, randomized, phase III study). Transfusion 36: 155–159 301. Xu X, Persson HL, Richardson DR (2005) Molecular pharmacology of the interaction and anthracyclines with iron. Mol Pharmacol 68: 261–271 302. Young SD, Hill RP (1990) Effects of reoxygenation on cells from hypoxic regions of solid tumors: anticancer drug sensitivity and metastatic potential. J Natl Cancer Inst 82: 371–80 303. Zachee P (1995) Controversies in selection of epoetin dosages. Issues and answers. Drugs 49: 536–547 Correspondence: Prof. Dr. M. R. Nowrousian, Department of Internal Medicine (Cancer Research), West German Cancer Center, University Hospital of Essen, Hufelandstrasse 55, 45122 Essen, Germany, E-mail: [email protected]

Chapter 19

Early intervention with recombinant human erythropoietin for chemotherapy-induced anemia G. H. Lyman1 and J. Glaspy2 1 Division of Medical Oncology, Duke University School of Medicine, Duke Center for Clinical Health Policy Research and the Duke Comprehensive Cancer Center, Durham, North Carolina; USA 2 Division of Hematology-Oncology, University of California Los Angeles School of Medicine, Los Angeles, California, USA

Introduction Chemotherapy-induced anemia Chemotherapy-induced anemia (CIA) is a frequent complication of modern myelosuppressive chemotherapy (Groopman and Itri 1999). Complications of CIA include fatigue and its associated physical, emotional, psychological, and social effects (Curt 2000). The effects of CIA are perhaps most noticeable in the treatment of elderly cancer patients at increased risk of mortality, functional dependence, cognitive disturbances, falls, and cerebrovascular and cardiovascular effects (Balducci et al. 2000). The incidence of severe anemia (hemoglobin <8 g/dL) has been reported to range from 50% to 60% in patients with certain malignancies receiving chemotherapy while the total incidence of anemia including grade 1 (hemoglobin = 10 g/dL to <12 g/dL) and 2 (hemoglobin = 8 g/dL to hemoglobin <10 g/dL) is high across most major solid tumors (Groopman and Itri 1999; CTEP 1998; FDA 2004). Clinical practice guidelines Several professional organizations have generated guidelines for the use of the erythropoietic stimulating agents (ESAs) in patients receiving cancer chemotherapy (Table 1). Guidelines from the American Society of Clinical Oncology (ASCO) and the American Society of Hematology (ASH) recommend use of ESAs in patients receiving chemotherapy with hemoglobin <10 g/dL (Rizzo et al. 2002). These recommendations were based on a systematic review of the literature from 1985 to 1999 on the effectiveness of ESAs for the treatment of CIA in 22 randomized controlled trials (RCTs)

510

G. H. Lyman and J. Glaspy

Table 1. Chemotherapy-induced anemia guideline recommendations Guideline

ASH/ASCO

EORTC

NCCN

Year of Recommendations Treatment Initiation at Hemoglobin Level Target Hemoglobin Range Basis of Recommendation

2002, 2007

2006

2006

<10 g/dL

<11 g/dL

<11 g/dL

12 g/dL or near to 12 g/dL ∼50% relative risk reduction in RBC transfusions

12–13 g/dL

12 g/dL

∼50% relative risk reduction in RBC transfusions QOL improvement

∼50% relative risk reduction in RBC transfusions QOL improvement

(Seidenfeld et al. 2001). However, no ﬁrm recommendation was made on the use of ESAs in patients with mild anemia due to the limited data available at the time (Del Mastro et al. 1997; ten Bokkel Huinink et al. 1998; Thatcher et al. 1999). On the other hand, subsequent guidelines for chemotherapy induced anemia have recommended consideration of an ESA when the hemoglobin level falls below 11 g/dL (Bokemeyer et al. 2007; NCCN 2007). All major anemia guidelines have been consistent in recommending a target hemoglobin with ESAs near 12 g/dL.

Clinical considerations Randomized controlled trials have clearly demonstrated the clinical beneﬁt of the ESA’s including a reduction in the magnitude of hemoglobin decrease, a reduction in transfusion requirements and improvement in health-related quality of life. Guideline recommendations, therefore, have been based largely on the demonstrated efﬁcacy of the ESAs in reducing transfusion requirements among patients with CIA (Table 1). Unfortunately, there are very limited data from controlled clinical trials on the impact of the ESAs on cancer patient survival or health-related quality of life (HRQOL). Retrospective analyses and more recent clinical trials have suggested that anemia correction as a result of ESAs is associated with improvement in HRQOL (Littlewood et al. 2001; Fairclough et al. 2003; Fallowﬁeld et al. 2002; Cella et al. 2004). The impact of mild anemia on patient fatigue and HRQOL remains a area of some controversy and wide variation in practice and reimbursement in the United States. A longitudinal analysis of the relationship

Early intervention with erythropoietin

511

between incremental changes in hemoglobin and quality of life scores revealed that the greatest improvement in quality of life occurs when hemoglobin levels increase from 11 g/dL to 12 g/dL (Crawford et al. 2002). Consistent with this observation, Straus reported that patients who were treated early (hemoglobin ≥ 10 g/dL but ≤ 12 g/dL) had signiﬁcantly higher scores in various quality of life measures, spent fewer days in bed (52.2% vs 3.1%, respectively; P = .017), and experienced fewer days of restricted activity (41.6% vs 12.2%, respectively; P = .042), compared with patients who received later intervention (Straus et al. 2006).

Economic considerations While the economic impact of the ESAs has received little study, it is intrinsically linked to costs associated with anemia in patients receiving chemotherapy. Few studies have assessed the economic burden associated with anemia in patients with malignancy. A retrospective study of the impact of anemia within six months of cancer diagnosis estimated average total expenditures among cancer patients developing anemia of $US 63,694 compared to $US 28,955 in nonanemic patients (P < .0001) (Lyman et al. 2005). The greatest cost differential between the groups was reported for inpatient services as patients with anemia had about twice as many hospital admissions as nonanemic patients. In a similar study, the incremental economic burden of medical care and short-term disability associated with anemia in patients with cancer receiving chemotherapy was evaluated up to 6 months following diagnosis (Berndt et al. 2005). Clearly, indirect and out of pocket expenses represent a substantial additional costs associated with anemia (Lyman 2005). Previous studies have suggested that the recombinant ESAs are preferred by clinicians and patients and are reasonably cost-effective compared to transfusion therapy, particularly when patient preferences or utilities are considered.

Study rationale While the rationale is generally understood, the overall merits of initiating ESA early for mild anemia remain a controversial topic (Fig. 1). The clinical beneﬁts associated with treating mild anemia were ﬁrst suggested by two studies published prior to 1999, that were included as part of the metaanalysis used to develop the ASH/ASCO guidelines (Seidenfeld et al. 2001). Therefore, systematic review was undertaken to ﬁrst determine whether there is evidence to suggest a clinical beneﬁt for initiating ESAs in cancer patients with hemoglobin ≥ 10 g/dL versus no treatment? Secondly, this study addressed whether there is a comparative clinical beneﬁt associated with

512

A

G. H. Lyman and J. Glaspy

Study

Treated n/N

Control n/N

Relative Risk [95% CI]

Bamias 2003

11 / 72

24 / 72

0.46 [0.24, 0.86]

Chang 2003

15 / 175

40 / 175

0.38 [0.22, 0.65]

Del Mastro 1997

0 / 31

2 / 31

0.20 [0.01, 4.00]

Savonije 2004

76 / 211

68 / 104

0.55 [0.44, 0.69]

ten Bokkel Huinink 1998

2 / 45

13 / 33

0.11 [0.03, 0.47]

Thatcher 1999

19 / 42

26 / 44

0.77 [0.51, 1.16]

Vansteenkiste 2002

42/156

82/158

0.52 [0.38, 0.70]

Combined

165/732

255/617

0.50 [0.43, 0.59] 0.1 0.2 0.5 1

2

P<.0001

5

Favors Epo Favors Control

B Study

Treated n/N

Control n/N

Relative Risk [95% CI]

Bamias 2003

12 / 72

33 / 72

0.36 [0.20, 0.65]

Del Mastro 1997

0 / 31

16 / 31

0.03 [0.00, 0.48]

ten Bokkel Huinink 1998

6/34

12 / 24

0.35 [0.15, 0.81]

Thatcher 1999

20 / 42

29 / 44

0.72 [0.49, 1.06]

Combined

38/179

90/171

0.40 [0.19, 0.83] 0.1 0.2 0.5 1

2

P=.0147

5

Favors Epo Favors Control

Fig. 1. Transfusion risk (A) and risk of Hb < 10 g/dL* (B) associated with chemotherapyinduced anemia for epoetin alfa and control groups in RCTs examining the efﬁcacy of rHuEPO versus no treatment. The risk of Hb < 10 g/dL was evaluated over the course of the study with the exception of the ten Bokkel Huinink study, which was evaluated during cycle 6 of chemotherapy. For the ten Bokkel Huinink and Thatcher studies that reported results from both 150 U/kg and 300 U/kg dose cohorts, data from the lower dose cohort was used as both manuscripts suggest that 150 U/kg is an appropriate dose for efﬁcacy. The point labeled “combined” is the pooled estimate of response; the center of the diamond represents the relative risk estimate; while the lateral points of the diamonds represent the 95% CI. * Hb ≤ 10 g/dL for the Del Mastro study

Early intervention with erythropoietin

513

treating mild anemia versus waiting until hemoglobin falls below 10 g/dL before intervening (Lyman and Glaspy 2006).

Methods Search and study selection A literature search of Medline was undertaken for the period from 1997 to April 2006 using the Boolean search string “erythropoietin AND cancer AND anemia NOT myelodysplastic”. References from identiﬁed articles were also reviewed for additional citations along with a search of presentations delivered at ASCO and ASH conferences from 1999 through 2006. Studies were required to be (1) RCTs in patients with CIA evaluating the ability of ESAs to prevent the onset or worsening of CIA and (2) a baseline hemoglobin concentration ≥ 10 g/dL was required for eligibility and one of the following endpoints were reported: incidence of transfusion, changes in hemoglobin, hemoglobin response, or HRQOL. Studies reporting planned subset analyses based on baseline hemoglobin, or retrospective analyses of clinical trial data, were included if study outcomes were otherwise satisﬁed. For purposes of analysis, studies were separated into those that evaluated the effects of treating mild anemia compared with no erythropoietic intervention, and those that assessed the clinical beneﬁts of treating mild anemia versus waiting until hemoglobin <10 g/dL before treating.

Statistical methods Study-by-study heterogeneity was estimated based on the Q statistic and an inconsistency index (I2) was calculated as an estimate of the proportion of variation in estimates due to heterogeneity, rather than between study random variation (Higgings et al. 2003). The relative risks (RR) for transfusion and hemoglobin <10 g/dL using the method of Mantel and Haenszel were estimated (Mantel 1964) and the standardized mean difference in hemoglobin change was estimated by the method of Cohen (Fleiss 1993). Fixed effects models were utilized to estimate summary measures when no signiﬁcant heterogeneity was found across studies while random effects models were utilized to estimate summary measures when signiﬁcant heterogeneity was observed for outcomes across reporting studies. Hypothesis testing on summary effect estimates was based on a z-statistic with estimates of standard error and 95% CIs. Results are presented as Forest plots where appropriate, with effect estimates and 95% CIs presented for each individual study, and a summary measure and CI presented for all studies combined.

514

G. H. Lyman and J. Glaspy

Results Of 69 references identiﬁed, fourteen met the criteria for inclusion in this systematic review (Del Mastro et al. 1997; ten Bokkel Huinink et al. 1998; Thatcher et al. 1999; Crawford et al. 2002; Straus et al. 2006; Kunikane et al. 2001; Littlewood et al. 2003; Bamias et al. 2003; Vansteenkiste et al. 2004; Vansteenkiste et al. 2002; Chang et al. 2005; Crawford et al. 2007; Rearden et al. 2004; Savonije et al. 2006). Early intervention versus no treatment Transfusion outcome Table 2 summarizes the transfusion outcomes of all seven prospective studies for this outcome. All six studies compared epoetin alfa versus no treatment or best supportive care (Del Mastro et al. 1997; ten Bokkel Huinink et al. 1998; Thatcher et al. 1999; Bamias et al. 2003; Chang et al. 2005; Savonije et al. 2006; Savonije et al. 2005), while one study was a double-blind placebocontrolled trial of darbepoetin alfa (Vansteenkiste et al. 2002). Transfusion rates signiﬁcantly decreased with erythropoietic therapy in all but one study (Del Mastro et al. 1997). The proportion of patients requiring transfusion across trials was 22.5% (95% CI: 19.6% to 25.7%) in patients randomized to receive rHuEPO and 41.3% (37.5% to 45.3%) in controls. The overall RR for transfusion with rHuEPO in the seven studies was 0.50 [0.43, 0.59]; P < .0001 (Fig. 1A). Hemoglobin response Hemoglobin levels improved with erythropoietic treatment in all seven trials (Table 3). The proportion of patients experiencing hemoglobin <10 g/dL across the four reporting RCTs was 21.2% (15.9% to 27.8%) with rHuEPO and 52.6% (45.2% to 60.0%) among controls. Due to the presence of signiﬁcant heterogeneity, a random effects model was used to estimate the RR of a hemoglobin decline to <10 g/dL. The RR of hemoglobin <10 g/dL with rHuEPO in these studies was 0.40 (0.19, 0.83; P = .0147) (Fig. 1B). Early intervention versus delayed treatment Transfusion outcome Three RCTs directly comparing early versus delayed intervention were identiﬁed in our review (Straus et al. 2006; Crawford et al. 2007; Rearden et al.

150 U/kg TIW (EPO 150) 300 U/kg TIW (EPO 300) No treatment (control) 45 (EPO 150) 42 (EPO 300) 33 (control) Median (range) 12.0 (11.3–12.6; EPO 150) 11.6 (10.5–12.2; EPO 300) 11.8 (10.6–12.5; control) 4% (EPO 150) 14% (EPO 300) 39% (control)

150 U/kg TIW (EPO) No treatment (control)

0% (EPO) 6.55 (control)

Mean (SD) 13.0 (0.7; EPO) 13.1 (0.6; control)

31 (EPO) 31 (control)

Ovarian <13 g/dL

Breast ≥12 g/dL

ten Bokkel Huinink199846

150 U/kg TIW (EPO 150) 300 U/kg TIW (EPO 300) No treatment (control) 42 (EPO 150) 44 (EPO 300) 44 (control) Median (range) 13.7 (10.7–16.1; EPO 150) 13.6 (10.9–17.0 EPO 300) 13.4 (10.9–16.4; control) 45%* (EPO 150) 20%*** (EPO 300) 59% (control)

SCLC ≥10.5 g/dL

Thatcher 199947

9%*** (EPO) 23% (control)

Mean 11.2 (0.9; EPO)a 11.3 (0.8; control)

175 (EPO) 175 (control)

40,000 U QW (EPO) Best Supportive Care (control)

Breast ≤12 g/dL

Chang 20037

15%* (EPO) 33% (control)

Mean (95% Cl) 11.5 (11.1–11.9; EPO) 11.5 (11.2–11.8; control)

72 (EPO) 72 (control)

10,000 U TIW (EPO) No treatment (control)

Solid tumor ≤13 g/dL

Bamias 20032

36%*** (EPO) 65% (control)

Mean (SD) 10.7 (0.1; EPO) 10.8 (0.1; control)

211 (EPO) 104 (control)

10,000 U TIW (EPO) Best Supportive Care (control)

Solid tumor ≤12.1 g/dL

Savonije 200441,42

a

Type of error measurement (SD or SE) was not reported. * P < .05 vs control. ** P < .01 vs control. *** P < .001 vs control. Hb = hemoglobin. EPO = epoetin alfa; DA = darbepoetin alfa.

Transfusion Incidence

Patient Sample Size (n) Baseline Hb (g/dL)

Tumor Type Hb Entry Criterion Dosing Regimen

Del Mastro 199712

27%*** (DA) 52% (placebo)

Mean (SD) 10.3 (1.1; DA) 9.9 (1.0; placebo)

156 (DA) 158 (placebo)

2.25 mcg/kg QW (DA) Placebo

Lung ≤11 g/dL

Vansteenkiste 200248

Table 2. Randomized controlled trials to evaluate early erythropoietic intervention in patients with CIA: effect on transfusion incidence (from Lyman and Glaspy, 2007)34

Early intervention with erythropoietin 515

NR

NR

0%*** (EPO) 52% (control)

NR

↓ 0.8 (95% Cl: 0.3 to 1.4) (EPO) ↓ 3.05 (2.6 to 3.5) (control)

% Patients with Hb < 10 g/dLa % Patients with Hb > 13 g/dLc % Patients with Hb increase ≥ 2 g/dLd Mean change in Hb (g/dL) NR

NR

Median (range) 13.7 (10.7–16.1; EPO 150) 13.6 (10.9–17.0 EPO 300) 13.4 (10.9–16.4; control) 48%* (EPO 150) 39%** (EPO 300) 66% (control) NR

Thatcher 199947

* P < .05 vs control; ** P < .01 vs control; *** P < .001 vs control. a Hb ≤ 10 g/dL in Del Mastro study. b Proportion of patients with Hb < 10 g/dL during cycle 6 of chemotherapy. c % patients with Hb > 12 g/dL in the Chang study. d % patients with >2 g/dL increase in Hb in the savonije study. Hb = hemoglobin. EPO = epoetin alfa; DA = darbepoetin alfa.

NR

Median (range) 12.0 (11.3–12.6; EPO 150) 11.6 (10.5–12.2; EPO 300) 11.8 (10.6–12.5; control) 17.6%* (EPO 150) 29.6%* (EPO 300) 50% (control)b NR

Mean (SD) 13.0 (0.7; EPO) 13.1 (0.6; control)

ten Bokkel Huinink 199846

Baseline Hb (g/dL)

Del Mastro 199712

NR

NR

52%*** (EPO) 5.1% (control)

NR

11.2 (0.9; EPO)a 11.3 (0.8; control)

Chang 20037

NR

20.8%*** (EPO) 2.7% (control)

47.2%*** (EPO) 11.1% (control)

16.6%*** (EPO) 45.8% (control)

Mean (95% Cl) 11.5 (11.1–11.9; EPO) 11.5 (11.2–11.8; control)

Bamias 20032

↑ 1.6*** (EPO) ↓ 0.4 (control)

69.8%*** (EPO) 31% (control)

NR

NR

Mean (SD) 10.7 (1.0) g/dL (EPO) 10.8 (1.0) g/dL (control)

Savonije 200441,42

NR

66%*** (DA) 24% (placebo)

NR

NR

Mean (SD) 10.3 (1.1; DA) 9.9 (1.0; placebo)

Vansteenkiste 200248

Table 3. Summary of hemoglobin outcomes from early intervention studies of erythropoietic agnets in patients with CIA [from Lyman and Glaspy, 2007]34

516 G. H. Lyman and J. Glaspy

Early intervention with erythropoietin

517

2004). In addition, two previously reported phase 3 studies (Littlewood et al. 2001; Vansteenkiste et al. 2002) included a priori planned subset analyses of the relationship between baseline hemoglobin concentration and transfusion outcomes (Littlewood and Bajetta 2002; Vansteenkiste et al. 2004). Both studies speciﬁed analyses of outcomes by baseline hemoglobin and demonstrated a substantial reduction in transfusion requirements among patients treated early (Table 4). Across all ﬁve studies, the proportion of patients requiring transfusion was 14.3% (11.4% to 17.7%) in those randomized to receive early treatment compared to 25.6% (22.2% to 29.2%) among those receiving delayed treatment. The RR for transfusion for early treatment was 0.55 (0.42, 0.73; P = .0001; Fig. 2A).

Hemoglobin response The proportion of patients with CIA experiencing hemoglobin <10 g/dL in the two reporting RCTs of early versus late ESAs was 23.4% (18.1% to 29.7%) with early treatment and 54.1% (47.3% to 60.8%) with delayed treatment (Crawford et al. 2007; Rearden et al. 2004). (Table 5) The RR for hemoglobin <10 g/dL with early treatment was 0.44 (0.33, 0.57; P < .0001) (Fig. 2B).

HRQOL outcomes Several studied evaluated the effect of treatment initiation timing and HRQOL (Table 6). Crawford et al. (2002) conducted an analysis of two previously reported studies of rHuEPO (Glaspy et al. 1997; Demetri et al. 1998) to determine the change in HRQOL associated with a 1-g/dL increase in hemoglobin level. This analysis demonstrated a signiﬁcant correlation between high hemoglobin levels and high LASA and FACTAnemia scores (r = 0.25 and 0.29, respectively; P < .01). Analysis of the relationship between incremental changes in hemoglobin and HRQOL scores revealed the greatest improvement in HRQOL occurring when hemoglobin level increased from 11 g/dL to 12 g/dL. Straus et al. (2006) published results of an open-label RCT comparing rHuEPO when administered early to patients who had mild anemia or after patients became moderately anemic (hemoglobin <9 g/dL). Patients who were treated early had signiﬁcantly higher scores in various HRQOL assessments, spent fewer days in bed (52.2% vs 3.1%; P = .017), and experienced fewer days of restricted activity (41.6% vs 12.2%, P = .042) than those with delayed intervention.

Hb = hemoglobin.

Baseline Hb (g/dL) Mean (SD) # Patients Treated Transfusion Incidence

Hb Entry Criteria 11.2 (SE 0.7) 26 26%

135 18%

<9 g/dL

Late (n = 134)

≥10 g/dL but ≤12 g/dL 11.1 (SE 0.7)

Straus 200345 rHuEPO QW Early (n = 135)

106 12%

≥11 g/dL but <15 g/dL 13.1 (1.0)

Crawford 20039 rHuEPO QW Early (n = 106)

48 21%

13 (1.2)

≤10 g/dL

Late (n = 105)

99 17%

≥10.5 g/dL but ≤12.0 g/dL 11.1 (0.7)

64 26%

11.2 (0.6)

≤10 g/dL

Rearden 200437 Darbepoetin alfa Q3W Early Late (n = 99) (n = 102)

Table 4. Randomized trials of erythropoietic agents to evaluate early versus late intervention: effect on transfusion incidence [from Lyman and Glaspy 2007]34

518 G. H. Lyman and J. Glaspy

Early intervention with erythropoietin

519

A Study

Treated n/N

Control n/N

Relative Risk [95% CI]

Crawford 2003

13 / 106

22 / 105

0.59 [0.31, 1.10]

Littlewood 2002

3 / 42

59 / 209

0.25 [0.08, 0.77]

Rearden 2004

14 / 99

22 / 102

0.66 [0.36, 1.21]

Straus 2003

24 / 135

35 / 134

0.68 [0.43, 1.08]

Vansteenkiste 2004

15/ 102

14 / 45

0.47 [0.25, 0.89]

Combined

69 / 484

152 / 595

0.55 [0.42, 0.73] 0.1 0.2

0.5 1

2

P < . 0001

5

Favors Early Favors Late

B Study

Treated n/N

Control n/N

Relative Risk [95% CI]

Craw ford 2003

19 / 106

46 / 105

0.41 [0.26, 0.65]

Rearden 2004

29 / 99

66 / 102

0.45 [0.32, 0.63]

Combined

48 / 205

112 / 207

0.44 [0.33, 0.57] 0.1 0.2 0.5

1

2

P < .0001

5

Favors Early Favors Late

Fig. 2. Transfusion risk (A) and risk of Hb < 10 g/dL* (B) associated with chemotherapy-induced anemia for early and late erythropoietic intervention groups in RCTs that directly compared the outcomes among patients treated with epoetin alfa or darbepoetin alfa early (≥10 g/dL) compared with those treated late (Hb < 10 g/dL). The point labeled ‘combined’ is the pooled estimate of response; the center of the diamond represents the relative risk estimate; the lateral points of the diamond represent the 95% CI. Estimates for the Rearden study are based on K-M proportions. * Hb ≤ 10 g/dL for the Crawford study

Discussion Summary The systematic review presented here demonstrates a clinical beneﬁt associated with early initiation of ESAs at hemoglobin >10 g/dL. Compared with placebo or no treatment, erythropoietic intervention at hemoglobin >10 g/dL

18%*** −0.2 g/dLc

11.2 (SE 0.7) NR −0.2 (SE 0.09)

NR

1.2 g/dL*** (SE 0.12)

≥11 g/dL but <15 g/dL 13.1 (1.0)

<9 g/dL

≥10 g/dL but ≤12 g/dL 11.1 (SE 0.7)

−1.4 g/dLc

44%

13 (1.2)

≤10 g/dL

Late (n = 105)

29%*** (95% CL 19, 38) 1.0 g/dLd

≥10.5 g/dL but ≤12.0 g/dL 11.1 (0.7)

Early (n = 99)

66% (95% CL 55, 75) 0.5 g/dLd

11.2 (0.6)

≤10 g/dL

Late (n = 102)

Rearden 200437 Darbepoetin alfa Q3W

*** P < .001 vs late intervention. a Rearden et al. reported Kaplan-Meier estimates for this endpoint. b Hb ≤ 10 g/dL in Crawford study. c Estimated from mean Hb values of epoetin alfa (early intervention) and uncensored control (late intervention) groups at week 16 provided in Slide 24 (Fig. 3) of their presentation. d Estimated from mean Hb values at week 19 provided in Slide 14 of their presentation. Hb = hemoglobin.

Baseline Hb (g/dL) Mean (SD) % Patients with Hb <10 g/dLa,b Mean change in Hb (g/dL) from baseline

Hb Entry Criteria

Early (n = 106)

Late (n = 134)

Crawford 20039 rHuEPO QW

Early (n = 135)

Straus 200345 rHuEPO QW

Table 5. Summary of hemoglobin outcomes associated with early versus late erythropoietic intervention in patients with CIA [from Lyman and Glaspy, 2007]34

520 G. H. Lyman and J. Glaspy

Study Design

Retrospective analysis of 2 nonrandomized, single-arm, open-label, communitybased trials13,18

Retrospective analysis of data from a nonrandomized, open-label, communitybased trial

Erythropoietic Agent

rHuEPO

rHuEPO

Breast cancer receiving anthracycline (±taxane) chemotherapy

Nonmyeloid malignancies

Study Population

1597

4382

Study Sample Size (n)

Not reported Baseline Hb: 10 g/dL to 14 g/dL

Signiﬁcant positive correlation Crawford between Hb levels and LASA 20028 overall HRQOL scores in both trials (r = 0.25 and 0.29, respectively; P < .01). Similar correlation between Hb and FACT-Anemia scores was observed (r = 0.27; P < .01). A nonlinear relationship was observed between Hb and HRQOL up to Hb = 12 g/dL. Beyond this, subsequent incremental increases in Hb corresponded with smaller magnitude of HRQOL improvement. Signiﬁcant positive correlation Galow observed between Hb and 200319 HRQOL in all domains tested (P < .0001 in each domain). The greatest increases in HRQOL scores occurred when Hb increased from 11.0 to 12 g/dL. A clinically meaningful improvement in overall HRQOL corresponded to rise in Hb from 10 to 12 g/dL.

Not deﬁned13 ≤11 g/dL18

Reference

HRQOL Results

Hb Inclusion Criterion

Table 6. Summary of HRQOL outcomes from early intervention studies of erythropoietic agents in patients with CIA [Lyman and Glasp 2007]34

Early intervention with erythropoietin 521

Study Design

Randomized, open-label, multicenter trial

Randomized, open-label, multicenter trial

Erythropoietic Agent

rHuEPO

Darbepoetin alfa

Table 6. Continued

Nonmyeloid malignancies

Hematologic malignancies

Study Population

Hb Inclusion Criterion

135: early Early intervention; intervention: 134: late ≥10 g/dL intervention and ≤ 12 g/dL Late intervention: <9 g/dL 99: early Early interventio; intervention: 102: late ≥10.5 g/dL intervention but ≤12 g/dL Late intervention: ≤10 g/dL

Study Sample Size (n)

Straus 200345

Reference

Mean change in FACT-Fatigue Rearden subscale score was higher in 200437 early vs late intervention group at the end of the test period (week 13; 1.5 vs −0.8, respectively) and comparable for both arms at the end of the treatment period (week 22; 1.5 vs 1.8, respectively)

Signiﬁcantly higher baselineadjusted scores in early vs late group were observed in all HRQOL domains evaluated (P = .005 to P = .024)

HRQOL Results

522 G. H. Lyman and J. Glaspy

Early intervention with erythropoietin

523

signiﬁcantly reduced transfusion requirements and hemoglobin decline below 10 g/dL. In addition, reductions in transfusion requirements and hemoglobin decline below 10 g/dL following early intervention compared to delayed intervention were observed in both prospective studies (Straus et al. 2006; Crawford et al. 2007; Rearden et al. 2004) and planned subset analyses (Vansteenkiste et al. 2004; Littlewood and Bajetta 2002), Studies of the impact of ESAs on HRQOL (Crawford et al., 2002; Straus et al. 2006; Rearden et al. 2004; Gralow and Williams 2003) have demonstrated that early intervention reduces the symptoms of anemia signiﬁcantly improving HRQOL and productivity, compared to delayed intervention.

Safety concerns Current studies To date, most data support the safety and efﬁcacy of the ESAs in patients with CIA when used in accordance with current guidelines. However, safety concerns have recently been raised concerning an increased risk of mortality observed in clinical trials in selected clinical settings. Survival outcomes were not reported in most of the trials included in this review. Although terminated early due to slow accrual and not powered for comparing response and survival outcomes, Crawford et al. reported no signiﬁcant differences in tumor response (30% vs 22%) or overall survival (median survival 11.5 vs 8.3 months) between early intervention with epoetin alfa compared to delayed intervention, respectively, in patients with non-small cell lung cancer (Crawford et al. 2007). Figure 3 presents the Kaplan-Meier survival estimates for immediate vs delayed intervention in this trial (P = .401).

Additional studies Data from two clinical trials not meeting inclusion criteria for the review reported here have suggest that patients treated with an ESA in certain clinical settings have decreased survival compared with patients receiving placebo (Henke et al. 2003; Leyland-Jones 2003). The BEST trial in breast cancer patients, reported an increased mortality for those receiving epoetin alfa primarily as a result of a higher incidence of disease progression and an increase in the number of thrombotic and vascular events (Leyland-Jones 2003). In the second study of head-and-neck cancer patients receiving epoetin beta and concurrent radiotherapy, a shorter median duration of locoregional progression-free survival was observed in the ESA treated group (Henke et al. 2003). There are a number of limitations associated with these studies which have prevented the drawing of ﬁrm conclusions (Kaanders and van der

524

G. H. Lyman and J. Glaspy

Fig. 3. Kaplan-Meier plot of overall cumulative survival of patients with stage IIIB/IV non-small cell lung cancer receiving systemic chemotherapy randomized to receive immediate weekly epoetin alfa or delayed epoetin alfa only for hemoglobin ≤10 g/dL [logrank test, P = .401]

Kogel 2004; Leyland-Jones and Salaheddin 2004; Haddad 2004). In the breast cancer study, important prognostic factors for survival were not assessed or documented while the study was ongoing. For the Henke study, an imbalance in baseline prognostic factors between the treatment and control groups may have affected the outcome. It is also of note that the observed adverse outcomes may have arisen in part from the initiation of treatment in non-anemic patients (hemoglobin > 12 g/dL) (Leyland-Jones 2003) as well as allowing hemoglobin levels to rise to 15 g/dL during the study period (Henke et al. 2003). Regardless of the limitations or interpretation of these studies, concerns regarding the safety of ESA therapy should not be ignored, particularly with respect to the normalization of hemoglobin concentration and the rapidity of hemoglobin increase.

Meta-analyses Three systematic reviews and formal meta-analyses have been recently reported including one from the Cochrane Collaboration, another from the

Early intervention with erythropoietin

525

Overall Mortality Start Hemoglobin

0.1

0.2

0.5

1

2

5

10

Effect [95% CI] P Value

< 10 (n=20)

.976

.814

1.168 .788

10-12 (n=8)

.969

.749

1.255 .812

> 12 (n=8)

1.286 1.027 1.611 .028

unclear (n=7)

1.611 1.061 2.445 .025

Combined

1.095

Favors ESA

.973

1.233 .132

Favors No ESA

Fig. 4. Forest plot of subgroup analysis conducted of RCTs included in the updated Cochrane meta-analysis based on study starting hemoglobin (Bohlius et al. 2006)

Agency of Healthcare Research and Quality (AHRQ) and a third by Ross et al. (Bohlius et al. 2006; Ross et al. 2006; Seidenfeld et al. 2000). While therapeutic efﬁcacy is conﬁrmed in these overviews, an increased risk of venous thromboembolism in patients receiving ESAs was observed with a RR from the Cochrane review for all cancer patients of 1.67 [95%CI: 1.35–2.06]. In the Cochrane review, three trials out of the 42 with survival data reported a signiﬁcant increase in mortality for an overall odds ratio of 1.08 [0.99–1.18] (Bohlius et al. 2006). When stratiﬁed by the hemoglobin level at the initiation of treatment, the odds ratios for mortality were 1.01[0.89–1.15], 0.98[0.92–1.10] and 1.27[1.05–1.54] for hemoglobin of < 10; 10–12 and >12 g/dl, respectively (Fig. 4). The AHRQ meta-analysis also demonstrated that increased mortality was predominately in those studies targeting hemoglobin levels well above that advocated by current guidelines (Table 7). It remains prudent, therefore, to limit ESA therapy to patients with demonstrable anemia, i.e., hemoglobin <12 g/dL, targeting hemoglobin levels consistent with current guidelines. Further data on the safety and efﬁcacy of early ESA intervention in patients with mild chemotherapy-induced anemia are needed and should soon be available. Potential role for risk models One potential strategy to balance the demonstrated beneﬁts of early intervention with ESAs with potential safety concerns includes better delineation of risk factors for developing CIA and patients at greatest risk and most likely to beneﬁt from intervention. Risk factors for anemia were recently studied in a prospective study of 3,640 patients with cancers of the breast, lung, colon

526

G. H. Lyman and J. Glaspy

Table 7. Mortality and target hemoglobin stopping level Target stop hemoglobin (g/dL)

Hazard Ratio

95% Conﬁdence Interval

≤12 >12–≤13 >13–≤14 >14–≤15 >15–≤16

Not estimable 0.91 1.16 1.03 1.67

Not available 0.47–1.78 1.00–1.35 0.90–1.19 1.13–2.48

US Agency for Healthcare Research and Quality.

and ovary or malignant lymphoma receiving a new chemotherapy regimen at 110 randomly selected U.S. practices. Independent risk factors in multivariate analysis for hemoglobin <10 g/dL included female gender; poor performance status; a history of vascular disease, advanced cancer stage; lung cancer or Hodgkin’s disease, baseline blood counts, poor renal function and certain chemotherapy agents, most notably the anthracyclines, carboplatin, gemcitabine and the topoisomerase inhibitors (Lyman et al. 2006). Future investigations should explore the potential beneﬁt of early ESA intervention among patients with a high risk for clinically signiﬁcant anemia while receiving cancer chemotherapy.

Conclusions The results of the analysis presented here along with recent safety concerns clearly deﬁne a need for additional clinical trials of the ESAs in patients with CIA to which include careful monitoring of clinical efﬁcacy, safety and HRQOL along with long term followup for disease progression and survival. Early intervention with an ESA in patients at increased risk for clinically signiﬁcant anemia offers the opportunity to reduce transfusion requirements and other complications associated with CIA. Such patients must be carefully monitored, however, with target hemoglobin levels consistent with current guidelines. Further data on the safety and efﬁcacy of early ESA intervention in patients with mild chemotherapy-induced anemia are needed and should soon be available.

References 1. Balducci L, Hardy CL, Lyman GH (2000) Hemopoietic reserve in the older cancer patient: clinical and economic considerations. Cancer Control 7(6): 539–547

Early intervention with erythropoietin

527

2. Bamias A, Aravantinos G, Kalofonos C, et al (2003) Prevention of anemia in patients with solid tumors receiving platinum-based chemotherapy by recombinant human erythropoietin (rHuEPO): A prospective, open label, randomized trial by the Hellenic Cooperative Oncology Group. Oncology 64(2): 102–110 3. Berndt E, Crown W, Kallich J, Long S, Song X, Lyman GH (2005) The impact of anaemia and its treatment on employee disability and medical costs. Pharmacoeconomics 23(2): 183–192 4. Bohlius J, Wilson J, Seidenfeld J, et al (2006) Recombinant human erythropoietins and cancer patients: updated meta-analysis of 57 studies including 9353 patients. J Natl Cancer Inst 98(10): 708–714 5. Bokemeyer C, Aapro MS, Courdi A, et al (2007) EORTC guidelines for the use of erythropoietic proteins in anaemic patients with cancer: 2006 update. Eur J Cancer 43(2): 258–270 6. Cella D, Kallich J, McDermott A, Xu X (2004) The longitudinal relationship of hemoglobin, fatigue and quality of life in anemic cancer patients: results from ﬁve randomized clinical trials. Ann Oncol 15(6): 979–986 7. Chang J, Couture F, Young S, McWatters KL, Lau CY (2005) Weekly epoetin alfa maintains hemoglobin, improves quality of life, and reduces transfusion in breast cancer patients receiving chemotherapy. J Clin Oncol 23(12): 2597– 2605 8. Crawford J, Cella D, Cleeland CS, et al (2002) Relationship between changes in hemoglobin level and quality of life during chemotherapy in anemic cancer patients receiving epoetin alfa therapy. Cancer 95(4): 888–895 9. Crawford J, Robert F, Perry MC, Belani C, Williams D (2007) A randomized trial comparing immediate versus delayed treatment of anemia with once-weekly epoetin alfa in patients with non-small cell lung cancer scheduled to receive ﬁrstline chemotherapy. J Thorac Oncol 2(3): 210–220 10. CTEP. Cancer Therapy Evaluation Program (CTEP) (1998) Common Toxicity Criteria, Version 2.0. DCTD, NCI, HIH, DHHS. March 1998. Available at: http://www.fda.gov/cder/cancer/toxicityframe.htm. Accessed February 2004 11. Curt GA (2000) Impact of fatigue on quality of life in oncology patients. Semin Hematol 37 [4:Suppl 6]: 14–17 12. Del Mastro L, Venturini M, Lionetto R, et al (1997) Randomized phase III trial evaluating the role of erythropoietin in the prevention of chemotherapy-induced anemia. J Clin Oncol 15(7): 2715–2721 13. Demetri GD, Kris M, Wade J, Degos L, Cella D (1998) Quality-of-life beneﬁt in chemotherapy patients treated with epoetin alfa is independent of disease response or tumor type: Results from a prospective community oncology study. J Clin Oncol 16(10): 3412–3425 14. Fairclough D, Gagnon D, Zagari M, Marschner N, Dicato M (2003) Evaluation of quality of life in a clinical trial with nonrandom dropout: The effect of epoetin alfa in anemic cancer patients. Quality of Life Research 12: 1013–1027 15. Fallowﬁeld L, Gagnon D, Zagari M, et al (2002) Multivariate regression analyses of data from a randomised, double-blind, placebo-controlled study conﬁrm quality of life beneﬁt of epoetin alfa in patients receiving non-platinum chemotherapy. Br J Cancer 87(12): 1341–1353 16. FDA. Food and Drug Administration/Center for Drug Evaluation and Research Oncology Tools page. World Health Organization Toxicity Criteria. Food and

528

17. 18.

19.

20.

21. 22.

23. 24. 25.

26. 27. 28.

29.

30.

31. 32.

G. H. Lyman and J. Glaspy Drug Administration Web site. Available at: http://www.fda.gov/cder/cancer/ toxicityframe.htm. Accessed February 2004 Fleiss J (1993) The statistical basis of meta-analysis. Statistical Methods in Medical Research 2: 121–145 Glaspy J, Bukowski R, Steinberg D, Taylor C, Tchekmedyian S, Vadhan-Raj S (1997) Impact of therapy with epoetin alfa on clinical outcomes in patients with nonmyeloid malignancies during cancer chemotherapy in community oncology practice. J Clin Oncol 15(3): 1218–1234 Gralow JR HC, Williams D (2003) Incremental improvements in QOL related to changes in hemoglobin in breast cancer (BC) patients (pts) receiving adjuvant chemotherapy (ADJCT) and QW epoetin alfa therapy [abstract]. Blood 102: Abstract 4392 Groopman JE, Itri LM (1999) Chemotherapy-induced anemia in adults: Incidence and treatment. J Natl Cancer Inst 10/6/1999 [erratum appears in J Natl Cancer Inst 2000 Mar 15;92(6):497]. 91(19): 1616–1634 Haddad R, MP (2004) Erythropoetin to treat anaemia in patients with head and neck cancer [letter: comment]. Lancet 363: 79 Henke M, Laszig R, Rube C, et al (2003) Erythropoietin to treat head and neck cancer patients with anaemia undergoing radiotherapy: randomised, doubleblind, placebo-controlled trial. Lancet 362(9392): 1255–1260 Higgins JT S, Deeks, J, Altman G (2003) Measuring inconsistency in metaanalyses. Br Med J 327 Kaanders JHAM, van der Kogel AJ (2004) Erythropoetin to treat anaemia in patients with head and neck cancer [letter: comment]. Lancet 363: 79–80 Kunikane H, Watanabe K, Fukuoka M, et al (2001) Double-blind randomized control trial of the effect of recombinant human erythropoietin on chemotherapy-induced anemia in patients with non-small cell lung cancer. Int J Clin Oncol 6(6): 296–301 Leyland-Jones B (2003) Breast cancer trial with erythropoietin terminated unexpectedly. Lancet Oncol 4(8): 459–460 Leyland-Jones B, Salaheddin M (2004) Erythropoetin to treat anaemia in patients with head and neck cancer [letter: comment]. Lancet 363: 80 Littlewood T, Bajetta E (2002) Early administration of epoetin alfa optimizes anemia management with respect to hematologic and quality of life outcomes in anemia cancer patients undergoing chemotherapy [abstract]. Blood 100: 18b. Abstract 3524 Littlewood TJ, Bajetta E, Nortier JW, Vercammen E, Rapoport B, and the Epoetin Alfa Study Group. (2001) Effects of epoetin alfa on hematologic parameters and quality of life in cancer patients receiving nonplatinum chemotherapy: Results of a randomized, double-blind, placebo-controlled trial. J Clin Oncol 19(11): 2865–2874 Littlewood TJ, Nortier J, Rapoport B, et al (2003) Epoetin alfa corrects anemia and improves quality of life in patients with hematologic malignancies receiving non-platinum chemotherapy. Hematological Oncology 21(4): 169–180 Lyman GH (2005) Time is money for both the healthy and the sick. Med Care 43(7): 637–639 Lyman GH, Berndt ER, Kallich JD, et al (2005) The economic burden of anemia in cancer patients receiving chemotherapy. Value Health 8(2): 149–156

Early intervention with erythropoietin

529

33. Lyman GH, Dale DC, Kuderer NM, et al (2006) Prospective validation of a predictive model for early anemia in patients receiving cancer chemotherapy. Paper presented at: Blood, 2006; 48th Annual Meeting of the American Society of Hematology 34. Lyman GH, Glaspy J (2006) Are there clinical beneﬁts with early erythropoietic intervention for chemotherapy-induced anemia? A systematic review. Cancer 106(1): 223–233 35. Mantel N (1964) Chi-square tests with one degree of freedom: Extensions of the Mantel-Haenszel procedure. J Am Stat Assoc 58: 163–170 36. NCCN. Cancer and Treatment Related Anemia. (2007) http://www.nccn.org/professionals/physician_gls/PDF/anemia.pdf 37. Rearden T, Charu V, Saidman B, et al (2004) Results of a randomized study of every-three week dosing (Q3W) of darbepoetin alfa for chemotherapyinduced anemia (CIA) [presentation]. 2004 Annual Meeting. Available at http://www.asco.org/ac/1,1003,_12-002643-00_18-0026-00_19-002479,00.asp. Accessed July 2004 38. Rizzo JD, Lichtin AE, Woolf SH, et al (2002) Use of epoetin in patients with cancer: Evidence-based clinical practice guidelines of the American Society of Clinical Oncology and the American Society of Hematology. Blood 100(7): 2303–2320 39. Ross SD, Allen IE, Henry DH, Seaman C, Sercus B, Goodnough LT (2006) Clinical beneﬁts and risks associated with epoetin and darbepoetin in patients with chemotherapy-induced anemia: a systematic review of the literature. Clin Ther 28(6): 801–831 40. Sabbatini P (2004) Clinical practice guidelines in oncology: cancer and treatmentrelated anemia. National Comprehensive Cancer Network. 01/14/2004 41. Savonije JH, van Groeningen CJ, van Bochove A, et al (2005) Effects of early intervention with epoetin alfa on transfusion requirement, hemoglobin level and survival during platinum-based chemotherapy: Results of a multicenter randomised controlled trial. Eur J Cancer 41(11): 1560–1569 42. Savonije JH, van Groeningen CJ, Wormhoudt LW, Giaccone G (2006) Early intervention with epoetin alfa during platinum-based chemotherapy: an analysis of the results of a multicenter, randomized, controlled trial based on initial hemoglobin level. Oncologist 11(2): 206–216 43. Seidenfeld J, Piper M, Flamm C, et al (2000) Comparative effectiveness of epoetin and darbepoetin for managing anemia in patients undergoing cancer treatment. Comparative Effectiveness. Effectiveness Review no. 3. (Prepared by Blue Cross and Blue Shield Association Evaluation Center Evidence-based Practice Center under contract no. 290-02-0026 44. Seidenfeld J, Piper M, Flamm C, et al (2001) Epoetin treatment of anemia associated with cancer therapy: A systematic review and meta-analysis of controlled clinical trials. J Natl Cancer Inst 93(16): 1204–1214 45. Straus DJ, Testa MA, Sarokhan BJ, et al (2006) Quality-of-life and health beneﬁts of early treatment of mild anemia: a randomized trial of epoetin alfa in patients receiving chemotherapy for hematologic malignancies. Cancer 107(8): 1909–1917 46. ten Bokkel Huinink WW, de Swart CAM, van Toorn DW, et al (1998) Controlled multicentre study of the inﬂuence of subcutaneous recombinant human erythro-

530

G. H. Lyman and J. Glaspy: Early intervention with erythropoietin

poietin on anaemia and transfusion dependency in patients with ovarian carcinoma treated with platinum-based chemotherapy. Medical Oncol 15: 174–182 47. Thatcher N, De Campos ES, Bell DR, et al (1999) Epoetin α prevents anaemia and reduces transfusion requirements in patients undergoing primarily platinumbased chemotherapy for small cell lung cancer. Br J Cancer 80(3–4): 396–402 48. Vansteenkiste J, Pirker R, Massuti B, et al (2002) Double-blind, placebocontrolled, randomized phase III trial of darbepoetin alfa in lung cancer patients receiving chemotherapy. J Natl Cancer Inst 94(16): 1211–1220 49. Vansteenkiste J, Tomita D, Rossi G, Pirker R (2004) Darbepoetin alfa in lung cancer patients on chemotherapy: A retrospective comparison of outcomes in patients with mild verses moderate-to-severe anaemia at baseline. Supportive Care in Cancer 12: 253–262 Correspondence: Gary H. Lyman MD MPH FRCP (Edin), Duke University Medical Center, Box 3645, Duke Comprehensive Cancer Center, Durham, North Carolina 27710, USA, E-mail: [email protected]

Chapter 20

Recombinant human erythropoietin (rhEPO) therapy in myelodysplasia E. Hellström-Lindberg Karolinska Institutet, Department of Medicine, Division of Hematology, Karolinska University Hospital Huddinge, Stockholm, Sweden

Clinical presentation of MDS Myelodysplastic syndrome (MDS) constitutes a group of malignant hematopoietic stem cell disorders characterized by ineffective hematopoiesis, and a signiﬁcant risk of progression to acute myeloid leukemia (AML). MDS is generally a disease of the elderly, with a median age of 70 years. The incidence is 4–5/100,000/year, similar to that of acute leukemia. MDS is generally idiopathic, but can also result from hematopoietic stem cell injury due to cytotoxic chemotherapy, radiation, or genetic predisposition. In low-risk MDS, the main cause of cytopenia is increased apoptosis of hemopoietic progenitors, while in high-risk MDS marrow apoptosis is accompanied by additional genetic and epigenetic events, leading to expansion of immature cells, and eventually transformation to AML. Signs and symptoms of MDS relate to hematopoietic failure, manifesting in anemia, thrombocytopenia or leukopenia. The anemia is often severe, leading to regular transfusion need and reduced quality of life.

Diagnosis and prognosis MDS is classiﬁed largely on the basis of cellular morphology, i.e. the percent of myeloblasts in the bone marrow and blood, the type and degree of dysplasia, and the presence of ringed sideroblasts. The recent reclassiﬁcation by the World Health Organization (WHO) is based on the previous FrenchAmerican-British (FAB) classiﬁcation (Bennett et al. 1982; WHO 2001), but allows distinction between pure refractory anemia (RA) and refractory anemia with ringed sideroblasts (RARS) and RA/RARS with multi-lineage dysplasia (Fig. 1). Moreover, it divides RA with excess of blasts (RAEB) into RAEB-1 and RAEB-2, and transfers patients with an isolated deletion of 5q and <5% blasts, into a separate category, the 5q- syndrome. The previous

532

E. Hellström-Lindberg

Fig. 1. FAB versus WHO classiﬁcation of primary MDS

Table 1. International Prognostic Scoring System (IPSS) (Greenberg et al. 1997) Score value Prognostic variable Bone marrow blasts (%) Karyotype* Cytopenias

0 <5 Good 0/1

0.5 5–10 Intermediate 2/3

1.0 – Poor

1.5 11–20

2.0 21–30

* Good, normal, -Y, del(5q), del(20q); Poor, complex (≥3 abnormalities) or chromosome 7 anomalies; Intermediate, other abnormalities. IPSS risk groups: Low (score 0), Intermediate-1 (INT-1, score 0.5–1.0), Intermediate2 (INT-2, score 1.5–2.0), High (score ≥ 2.5)

FAB group chronic myelomonocytic leukemia (CMML) is transferred into a new main category, mixed myelodysplastic/myeloproliferative disorders. The WHO classiﬁcation does not address the problem of classifying patients with severe ﬁbrosis leading to uncertain blast counts, or patients with markedly reduced cellularity (hypoplastic MDS). And, importantly, apart from the 5qsyndrome, it is still based on morphological ﬁnal common pathways, such as dysplasia and blast increase, rather than on speciﬁc biological and molecular ﬁndings. Compared to some other hematological malignancies, therapeutic development has been rather slow in MDS, and not until the last years have new innovative therapeutic approaches emerged. The International Prognostic Scoring System (IPSS), using cytogenetic proﬁle, bone marrow blast percentage, and number of cytopenias, is currently

Recombinant human erythropoietin therapy in myelodysplasia

533

the best available tool for estimating the probability for survival and leukemic transformation of untreated patients with MDS. Based on these variables, patients are assigned to one of four risk groups: (Table 1) (Greenberg et al. 1997). As a general rule, patients with <10% bone marrow blasts have a longer survival and lower rate of transformation to AML than those with higher blast counts.

Management of MDS While allogeneic stem cell transplantation may offer a possibility for cure, probably less than 15% of patients with MDS are eligible for this treatment due to age, co-morbid conditions, or resistant high-risk disease. The main therapeutic approach for patients with low-risk MDS is supportive care, aiming to alleviate the effects of the underlying cytopenia, and in particular anemia.

Blood transfusions and iron overload The most important part of supportive care in MDS is transfusion therapy. 80–90% of MDS patients will receive a transfusion at some point during their clinical course. It has recently been reported that a need for transfusion therapy has a negative impact on long-term outcome, irrespective of the MDS subgroup, but it is not known whether this is due to the transfusions as such, or to the fact that transfusion need implies a more severe MDS (Malcovati et al. 2005). The optimal target hemoglobin level differs between patients, and it is important to individualize the management of each patient. Regular transfusion therapy will eventually lead to iron overload, and a need for chelation therapy, if the patient otherwise has a reasonable prognosis. If iron overload can be controlled with deferoxamine or new oral chelators, many MDS patients may survive for years by means of chronic transfusion therapy (Greenberg 2006).

Immunomodulation Several phase II studies have shown that treatment with antithymoglobulin (ATG) may be effective in subsets of patients with low-risk MDS. Responders to this treatment are mainly younger (<60 years), with WHO RA or RCMD, normal cytogenetics, short history of transfusion dependency and possibly HLA DR15 phenotype (Modlldrem et al. 1997; Saunthararajah et al. 2002). Cyclosporine A has similar, but less pronounced effects. Thalidomide may also induce erythroid responses in a small proportion of low-risk MDS, but has generally problematic side effects, which limits its use (Bouscary et al. 2005).

534

E. Hellström-Lindberg

Lenalidomide, one of the new immunomodulating drugs (IMIDS) is a thalidomide structural analogue. Lenalidomide has shown remarkable efﬁcacy in low-risk MDS with a deletion of 5q (List et al. 2005). Two thirds of patients with this MDS subtype have obtained a complete erythroid response, and also a complete cytogenetic remission. Lenalidomide is at present investigated in a series of follow-up studies to establish its role in MDS.

Erythropoietin in MDS The process of erythropoiesis in the bone marrow takes 14–17 days and is highly dependent on a number of hemopoietic growth factors (Fig. 2). Erythropoietin (EPO) alone, or in combination with other colony-stimulating factors (CSFs), has therefore been investigated as a potential approach to correct the anemia associated with MDS. Treatment with erythropoietin as monotherapy may induce erythroid responses in 16–30% of patients with low-risk MDS (Hellström-Lindberg 2005). One placebo-controlled randomized study demonstrated an overall positive effect in the EPO-treated cohort, however, at the subgroup level, only patients with FAB RA, and patients without transfusion need showed a signiﬁcant response to treatment (Italian Cooperative Study Group for rHuEpo in Myelodysplastic Syndromes 1998). Several smaller phase II studies have indicated that patients with low pretreatment serum EPO levels (<150–200 U/l) are particularly likely to beneﬁt (Hellström-Lindberg 2005). A meta-analysis of 205 patients from 17 studies revealed an overall response rate of 16%, and showed that patients with no transfusion requirement and MDS other than RARS were more likely to beneﬁt from treatment than other patient types (Hellström-Lindberg 1995). There is no consistent information as to the optimal duration of treatment necessary to assess a response

Fig. 2. Pro-erythroid growth factors and erythropoiesis BFU-E, burst-forming unit-erythroid; CFU-E, colony forming unit-erythroid; Epo, erythropoietin; GF, growth factor

Recombinant human erythropoietin therapy in myelodysplasia

535

to treatment. While the majority of studies have treated patients for 12–16 weeks, a few trials have reported that additional responses may be observed if treatment is pursued for as long as 6 months (Terpos et al. 2002). The effect of the new long-acting darbepoetin-alfa has been assessed in some recent trials, and it has been found to be at least as effective as erythropoietin (Mannone et al. 2006; Stasi et al. 2005).

Erythropoietin in combination with other growth factors The overall response rate to EPO alone in patients with MDS can be improved by the addition of granulocyte colony-stimulating factor (G-CSF) or granulocyte/macrophage colony-stimulating factor (GM-CSF). A comparative meta-analysis reported a better effect of the combination of EPO and G-CSF, which also has less side effects than the combination of EPO and GM-CSF (Kasper et al. 2002). The effects of EPO and G-CSF have been assessed in a number of phase II trials, and in two randomized phase III trials (Hellström-Lindberg 2005). Response rates vary between studies, but are clearly higher than for EPO alone, usually around 40–50%. It has been shown that the addition of G-CSF can induce responses in patients not responding to EPO alone, and that withdrawal of G-CSF in patients responding to the combination may cause relapse of anemia (Negrin et al. 1996; HellströmLindberg et al. 1998; Remacha et al. 1999). One randomized study comparing EPO + G-CSF treatment vs supportive care showed a signiﬁcant effect on anemia in the treated cohort (Casadevall et al. 2004). Recently, a small randomized trial compared the effect of EPO vs EPO + G-CSF during 8 weeks, and reported a signiﬁcantly better erythroid response rate in patients treated with the combination (Balleari et al. 2006). The biological effects of EPO and G-CSF in MDS erythropoiesis have been assessed in a series of recent publications (Tehranchi et al. 2003; Tehranchi et al. 2005a; Tehranchi et al. 2005b). The ineffective erythropoiesis is associated with a profound erythroid apoptosis, which in turn is mediated through a constant pro-apoptotic release of cytochrome c from the mitochondrial inter-membrane space to the cytosol. In vitro as well as in vivo treatment with G-CSF signiﬁcantly inhibits the release of cytochrome c, and thus mitochondria-mediated apoptosis. This pattern is more pronounced in RARS, a subgroup that also responds particularly well to the combined treatment (Howe et al. 2004). EPO + G-CSF – long-term outcome and decision-making In unselected patients, the response rate to the combination is below 50% and considering the high cost for the treatment and the fact that certain

536

E. Hellström-Lindberg

Table 2. Validated predictive model for treatment of the anemia of MDS with EPO+G-CSF 4 Predictive variable Transfusion need (U/month) Serum-EPO (U/l)

Score <2 U/m <500

0 0

Score ≥2 U/m ≥500

1 1

Total score 0, response rate 74%; 1, response rate 23%, 2 response rate 7%. Complete erythroid response = Stable hemoglobin level >115 g/l. Partial erythroid response = Increase of hemoglobin >15 g/l, or stopped transfusion need.

patients could beneﬁt from other therapies, a structured decision-making process is important. A decision-making model for the use of EPO + G-CSF in MDS was developed from a large phase II trial, and then validated in a prospective study (Hellström-Lindberg et al. 2003). In this model, patients with serum EPO levels ≤500 U/l and a transfusion need of <2 units/month are identiﬁed as having a high probability of response to treatment while patients with EPO levels of >500 U/l and ≥2 units/month are likely to have a very poor response (Table 2). Recently, the Nordic cohort of 129 EPO+ G-CSF treated patients underwent a long-term follow-up, and the outcome was compared with untreated individuals from the IPSS cohort (Jadersten et al. 2005). The median duration of response was 23 months, 29 months in patients achieving a normalized hemoglobin level, and 12 months in those obtaining a partial response (Table 2 for response criteria). There were patients responding for up to 10 years, and only 1 out of 20 patients who responded for more than 2 years developed leukemia. Treatment did not affect survival or the rate of leukemic transformation. Importantly, patients having a low probability of response according to the decision model, had a very poor outcome, and a high risk for leukemic transformation, further underlining that this group should be given alternative treatment.

Conclusion Erythropoiesis-stimulating growth factors have a well-established role in the treatment of low-risk MDS, but should be given only after taking predictive variables into account. Moreover, new biologically targeted treatment alternatives are likely to add to the list of options, and may in fact precede the use of growth factors in certain subsets of patients. Table 3 gives a suggestion for a treatment algorithm for low-risk MDS.

Recombinant human erythropoietin therapy in myelodysplasia

537

Table 3. Algorithm for treatment of low-risk MDS (<10% marrow blasts, IPSS Low+INT-1) 1. 2. 3. 4. 5. 6.

High quality transfusion and chelation therapy Evaluate WHO RA / RCMD <60 years for antithymoglobulin (ATG) treatment Lenalidomide for IPSS Low-INT-1 patients with del 5q Evaluate for curative approaches (selected young patients with IPSS INT-1) rhEPO/Darbepoetin ± G-CSF according to decision model Low probability or no response to above: supportive care or treatment within clinical trials

References 1. A randomized double-blind placebo-controlled study with subcutaneous recombinant human erythropoietin in patients with low-risk myelodysplastic syndromes. Italian Cooperative Study Group for rHuEpo in Myelodysplastic Syndromes. Br J Haematol 1998; 103: 1070–1074 2. Balleari E, Rossi E, Clavio M, Congiu A, Gobbi M, Grosso M, Secondo V, Spriano M, Timitilli S, Ghio R (2006) Erythropoietin plus granulocyte colonystimulating factor is better than erythropoietin alone to treat anemia in low-risk myelodysplastic syndromes: results from a randomized single-centre study. Ann Hematol 85: 174–180 3. Bennett JM, Catovsky D, Daniel MT, Flandrin G, Galton DA, Gralnik HR, Sultan C (1982) Proposals for the classiﬁcation of the myelodysplastic syndromes. Br J Haematol 51: 189–199 4. Bouscary D, Legros L, Tulliez M, Dubois S, Mahe B, Beyne-Rauzy O, Quarre MC, Vassilief D, Varet B, Aouba A, Gardembas M, Giraudier S, Guerci A, Rousselot P, Gaillard F, Moreau A, Rousselet MC, Ifrah N, Fenaux P, Dreyfus F; For The Groupe Francais des Myelodysplasies (GFM) (2005) A non-randomised dose-escalating phase II study of thalidomide for the treatment of patients with low-risk myelodysplastic syndromes: the Thal-SMD-2000 trial of the Groupe Francais des Myelodysplasies. Br J Haematol 131: 609–618 5. Casadevall N, Durieux P, Dubois S, Hemery F, Lepage E, Quarre MC, Damaj G, Giraudier S, Guerci A, Laurent G, Dombret H, Chomienne C, Ribrag V, Stamatoullas A, Marie JP, Vekhoff A, Maloisel F, Navarro R, Dreyfus F, Fenaux P (2004) Health, economic, and quality-of-life effects of erythropoietin and granulocyte colony-stimulating factor for the treatment of myelodysplastic syndromes: a randomized, controlled trial. Blood 104: 321–327 6. Greenberg P, Cox C, LeBeau MM, Fenaux P, Morel P, Sanz G, Sanz M, Vallespi T, Hamblin T, Oscier D, Ohyashiki K, Toyama K, Aul C, Mufti C, Bennett J (1997) International scoring system for evaluating prognosis in myelodysplastic syndromes. Blood 89: 2079–2088 7. Greenberg PL (2006) Myelodysplastic syndromes: iron overload consequences and current chelating therapies. J Natl Compr Canc Netw 4: 91–96

538

E. Hellström-Lindberg

8. Hellström-Lindberg E, Ahlgren T, Beguin Y, Carlsson M, Carneskog J, Dahl IM, Dybedal I, Grimfors G, Kanter-Lewensohn L, Linder O, Luthman M, Löfvenberg E, Nilsson-Ehle H, Samuelsson J, Tangen JM, Winqvist I, Öberg G, Österborg A, Öst A (1998) Treatment of anemia in myelodysplastic syndromes with granulocyte colony-stimulating factor plus erythropoietin: results from a randomized phase II study and long-term follow-up of 71 patients. Blood 92: 68–75 9. Hellström-Lindberg E (1995) Efﬁcacy of erythropoietin in the myelodysplastic syndromes: a meta-analysis of 205 patients from 17 studies. Br J Haematol 89: 67–71 10. Hellström-Lindberg E, Gulbrandsen N, Lindberg G, Ahlgren T, Dahl IMX, Dybedal I, Grimfors G, Hesse-Sundin E, Hjorth M, Kanter-Lewensohn L, Linder O, Luthman M, Löfvenberg E, Öberg G, Porwit-MacDonald A, Radlund A, Samuelsson J, Tangen JM, Winquist I, Wisloff F (2003) A validated decision model for treating the anaemia of myelodysplastic syndromes with erythropoietin + granulocyte colony-stimulating factor: signiﬁcant effects on quality of life. Br J Haematol 120: 1037–1046 11. Hellström-Lindberg E (2005) Update on supportive care and new therapies: immunomodulatory drugs, growth factors and epigenetic-acting agents. Hematology (Am Soc Hematol Educ Program) 161–166 12. Howe RB, Porwit-MacDonald A, Wanat R, Tehranchi T, Hellström-Lindberg E (2004) The WHO classiﬁcation of MDS does make a difference. Blood 103: 3265–3270 13. Jadersten M, Montgomery SM, Dybedal I, Porwit-MacDonald A, HellströmLindberg E (2005) Long-term outcome of treatment of anemia in MDS with erythropoietin and G-CSF. Blood 106: 803–811 14. Kasper C, Zahner J, Sayer HG (2002) Recombinant human erythropoietin in combined treatment with granulocyte- or granulocyte-macrophage colonystimulating factor in patients with myelodysplastic syndromes. J Cancer Res Clin Oncol 128: 497–502 15. List A, Kurtin S, Roe DJ, Buresh A, Mahadevan D, Fuchs D, Rimsza L, Heaton R, Knight R, Zeldis JB (2005) Efﬁcacy of lenalidomide in myelodysplastic syndromes. N Engl J Med 352: 549–557 16. Malcovati L, Porta MG, Pascutto C, Invernizzi R, Boni M, Travaglino E, Passamonti F, Arcaini L, Mafﬁoli M, Bernasconi P, Lazzarino M, Cazzola M (2005) Prognostic factors and life expectancy in myelodysplastic syndromes classiﬁed according to WHO criteria: a basis for clinical decision making. J Clin Oncol 23: 7594–7603 17. Mannone L, Gardin C, Quarre MC, Bernard JF, Vassilieff D, Ades L, Park S, Vaultier S, Hamza F, Beyne-rauzy MO, Cheze S, Giraudier S, Agape P, Legros L, Voillat L, Dreyfus F, Fenaux P; Groupe Francais des Myelodysplasies (2006) High-dose darbepoetin alpha in the treatment of anaemia of lower risk myelodysplastic syndrome results of a phase II study. Br J Haematol 133: 513–519 18. Molldrem JJ, Caples M, Mavroudis D, Plante M, Young NS, Barrett J (1997) Antithymocyte globulin for patients with myelodysplastic syndrome. Br J Haematol 99: 699–705 19. Negrin RS, Stein R, Doherty K, Cornwell J, Vardiman J, Krantz S, Greenberg PL (1996) Maintenance treatment of the anemia of myelodysplastic syndromes with

Recombinant human erythropoietin therapy in myelodysplasia

20.

21.

22.

23.

24.

25.

26.

27.

539

recombinant human granulocyte colony-stimulating factor and erythropoietin: evidence for in vivo synergy. Blood 87: 4076–4081 Remacha AF, Arrizabalaga B, Villegas A, Manteiga R, Calvo T, Julia A, Fernandez Fuertes I, Gonzales FA, Font L, Junca J, del Arco A, Malcorra JJ, Equiza EP, de Mendiguren BP, Romero M (1999) Erythropoietin plus granulocyte colony-stimulating factor in the treatment of myelodysplastic syndromes. Identiﬁcation of a subgroup of responders. The Spanish Erythropathology Group. Haematologica 84: 1058–1064 Saunthararajah Y, Nakamura R, Nam JM, Robyn J, Loberiza F, Maciejewski JP, Simonis T, Molldrem J, Young NS, Barrett AJ (2002) HLA-DR15 (DR2) is overrepresented in myelodysplastic syndrome and aplastic anemia and predicts a response to immunosuppression in myelodysplastic syndrome. Blood 100: 1570– 1574 Stasi R, Abruzzese E, Lanzetta G, Terzoli E, Amadori S (2005) Darbepoetin alfa for the treatment of anemic patients with low- and intermediate-1-risk myelodysplastic syndromes. Ann Oncol 16: 1921–1927 Tehranchi R, Fadeel B, Forsblom AM, Christensson B, Samuelsson J, Zhivotovsky B, Hellström-Lindberg E (2003) Granulocyte colony-stimulating factor inhibits spontaneous cytochrome c release and mitochondria-dependent apoptosis of myelodysplastic syndrome hematopoietic progenitors. Blood 101: 1080–1086 Tehranchi R, Invernizzi R, Grandien A, Zhivotovsky B, Fadeel B, Forsblom AM, Travaglino E, Samuelsson J, Hast R, Nilsson L, Cazzola M, Wibom R, HellströmLindberg E (2005a) Aberrant mitochondrial iron distribution and maturation arrest characterize early erythroid precursors in low-risk myelodysplastic syndromes. Blood 106: 247–253 Tehranchi R, Fadeel B, Schmidt-Mende J, Forsblom AM, Emanuelsson E, Jadersten M, Christensson B, Hast R, Howe RB, Samuelsson J, Zhivotovsky B, Hellström-Lindberg E (2005b) Antiapoptotic role of growth factors in the myelodysplastic syndromes: concordance between in vitro and in vivo observations. Clin Cancer Res 11: 6291–6299 Terpos E, Mougiou A, Kouraklis A, Chatzivassili A, Michalis E, Giannakoulas N, Manioudaki E, Lazaridou A, Bakaloudi V, Protopappa M, Liapi D, Grouzi E, Parharidou A, Symeonidis A, Kokkini G, Laoutaris NP, Vaipoulos G, Anagnostopulos NI, Christakis JI, Meletis J, Bourantas KL, Zoumbos NC, Yataganas X, Viniou NA (2002) Prolonged administration of erythropoietin increases erythroid response rate in myelodysplastic syndromes: a phase II trial in 281 patients. Br J Haematol 118: 174–180 WHO (2001) Classiﬁcation of tumours of haematopoietic and lymphoid tissues. IARC Press, Lyon

Correspondence: Eva Hellström-Lindberg, MD, PhD, Karolinska Institutet, Department of Medicine, Division of Hematology, Karolinska University Hospital, Huddinge, Ihn 86 Stockholm, Sweden, E-mail: Eva.Hellströ[email protected]

Chapter 21

Prediction of response to rhEPO in the anemia of cancer Y. Beguin and G. Van Straelen Y.B. and G.V. are respectively Research Director and Télévie Research Assistant of the National Fund for Scientiﬁc Research (FNRS, Belgium), Department of Medicine, Division of Hematology, University of Liège, Liège, Belgium

Abstract The anemia of cancer can be effectively treated with erythropoietic proteins, such as recombinant human erythropoietin (rHuEPO) alfa or beta or darbepoietin alfa in about 60% of the patients. However, the response rate varies according to treatment modalities as well as the response criteria used. A number of disease- or chemotherapy-related factors determine the probability of response. Several speciﬁc mechanisms of anemia, such as hemolysis, splenomegaly, bleeding, hemodilution, or ineffective erythropoiesis can seriously interfere with response. However, the type of tumor, in particular hematologic versus nonhematologic, is not critical, except in situations of major marrow involvement and limited residual hematopoiesis. Stem cell damage by previous therapy, reﬂected by low platelet counts or high transfusion needs, will impair response. In addition, marrow suppression by current intensive chemotherapy will also have a negative impact. Besides its intensity, the type of chemotherapy may not be critical, and patients undergoing platinum-based chemotherapy or nonplatinum regimens respond similarly. Complications such as infections, bleeding or nutritional deﬁciencies may have a major negative impact on outcome. An important response-limiting factor is functional iron deﬁciency, i.e. an imbalance between iron needs in the erythropoietic marrow and iron supply, which depends on the level of iron stores and its rate of mobilization. Therefore, intravenous iron supplements should be given when serum ferritin is below 100 ng/mL, reﬂecting the absence of iron stores, or when the transferrin saturation is below 20% or the percentage of hypochromic red cells above 10%, indicating functional iron deﬁciency even in the presence of adequate storage iron (normal or increased ferritin). Because up to 40% of the patients will not respond to rhEPO, it is of great importance to develop models that could help predict response to rhEPO and thus select the most appropriate cancer patients for this therapy. Most studies in patients with myeloma or lymphoma have indicated that patients with a low baseline serum EPO level will respond better, but

542

Y. Beguin and G. Van Straelen

this is not so consistent in patients with solid tumors. Also of considerable interest are early changes of erythropoietic parameters after 2 to 4 weeks of treatment, including increments of serum transferrin receptor (sTfR), reticulocytes and hemoglobin. Combination of baseline serum EPO and the 2-week increment of sTfR or hemoglobin may provide the best prediction of response. Distinct predictive models should be used in patients with myelodysplastic syndromes. In these patients, the best models are different when rhEPO is used alone, with low baseline serum EPO, MDS other than RAS and transfusion independence predicting good response, or rhEPO together with rhG-CSF, with low baseline serum EPO and low transfusion needs predicting good response.

Introduction Patients with solid tumors or hematological malignancies often develop anemia at diagnosis or in the course of the disease (Beguin 1996; Knight et al. 2004; Ludwig et al. 2004), particularly when given chemotherapy (Groopman et al. 1999; Ludwig et al. 2004). Many studies have shown that erythropoietin therapy, using recombinant human erythropoietin (rhEPO) alfa or beta or darbepoietin alfa, can ameliorate the anemia associated with cancer and chemotherapy, reduce the need for transfusions and improve quality of life (Bohlius et al. 2005). Clinical practice guidelines have been published to provide guidance on the use of rhEPO (Rizzo et al. 2002; Bokemeyer et al. 2004). However, as many as 30–50% of the patients do not respond even to very high doses of rhEPO. A number of factors may interfere with response to rhEPO in cancer patients, reﬂecting differences in disease- and treatmentrelated factors, but also large variations in dose, frequency and route of administration, duration of therapy and the response criteria used (Beguin 2001; Beguin 2002a; Beguin 2002b). It is therefore important to be able to recognize and correct conditions adversely affecting response to rhEPO, in particular functional iron deﬁciency. When no such particular condition can be identiﬁed, it would also be of great interest to have at one’s disposal predictive algorithms of response. Thereby patients can be selected on the basis of their probability to achieve a good response to treatment, and prolonged ineffective use of an expensive medication can be avoided in those patients with a low probability of response. In this paper, we will review factors potentially affecting response to rhEPO (Table 1) and comment on the use of predictive algorithms.

Factors inﬂuencing response to EPO Criteria of response Before analyzing factors potentially affecting response to rhEPO, it is critical to deﬁne response criteria. Trials employing more favorable inclusion

x x x x x x x

x x

Factors relating to the patient • Age • Sex x x

x

x

• Duration • Type of rhEPO

Factors relating to the disease • Type of cancer • Marrow inﬁltration • Mechanisms of anemia Hemolysis Bleeding Hypersplenism Marrow necrosis or ﬁbrosis Hemophagocytosis Folate, B12, iron deﬁciency • Transfusion dependence

x

No

x

x

Yes

Factor inﬂuences response signiﬁcantly

• Route • Frequency

Factors relating to rhEPO treatment • Dose

Factor

Table 1. Factors potentially limiting response to rhEPO

Possibly best in myeloma Unless massive (acute leukemia)

rhEPO: 450 U/kg/wk Darbepoetin: 2.25 μg/kg/wk SC > IV rhEPO: t.i.w. or weekly Darbepoietin: weekly or q 2–3 wks Needs at least 2 months

Comments

Prediction of response to rhEPO 543

x x

Functional ID • Caused by ACD • Induced by rhEPO therapy

x

x

x x x

Inﬂammation • Complications Infection Inﬂammatory disorders • Surgery

Baseline parameters • Low platelet count • Neutrophil count • Hb, sTfR, reticulocytes • Creatinine • Transferrin saturation, ferritin • Cytokines • Serum EPO > 100–200 mU/ml (or O/P ratio > 0.9)

x x

Yes

x x x x x

x

No

Factor inﬂuences response signiﬁcantly

Factors relating to chemotherapy • Type of chemotherapy Platinum vs non-platinum Intensity of chemotherapy • Previous stem cell damage

Factor

Table 1. Continued

Hematological malignancies only?

A major cause of treatment failure

Bleeding + inﬂammation

A major cause of treatment failure

Not effective if intensiﬁed chemotherapy Low platelet count, transfusion-dependence

Comments

544 Y. Beguin and G. Van Straelen

Prediction of response to rhEPO

545

Table 2. Response criteria Cancer • Hb response: – Hb increment ≥2 g/dL (Hct increment ≥6%) without transfusion • Hematopoietic response: either criteria – Hb increment ≥2 g/dL (Hct increment ≥6%) without transfusion – Hb ≥12 g/dL without transfusion MDS 1. Nordic and French groups • Complete response: – Hb ≥11.5 g/dL without transfusion • Partial response: either criteria – Hb increment ≥1.5 g/dL without transfusion – Hb stable without transfusion in previously transfused patient 2. US, Italian, Spanish and German groups • Good or major response: either criteria – Hb increment ≥2 g/dL without transfusion – No transfusion requirement in previously transfused patient • Minor or partial response: either criteria – Hb increment 1–2 g/dL without transfusion – 50% reduction in transfusion requirement

criteria and less stringent deﬁnitions of response are very likely to report better outcome. Therefore, uniform response criteria should be proposed for transfused and untransfused, severely or not severely anemic cancer patients. However, these criteria should necessarily be partly different when rhEPO is used for the prevention or the treatment of anemia (Table 2). Prevention means that rhEPO is used in a nonanemic patient to avert the occurrence of anemia after chemotherapy or other interventions. Treatment signiﬁes that rhEPO is given to reverse an anemia that is already present. When treating anemia, a hemoglobin (Hb) response is deﬁned by an increase in Hb by at least 2 g/dL (or increase in hematocrit (Hct) by at least 6 percentage points) from baseline without transfusion. However, when the entry Hb is 10–11 g/dL and the target Hb value 12–13 g/dL, such Hb response may not be achievable before reaching the target Hb and entering maintenance with lower rhEPO doses. Therefore, hematopoietic response can be deﬁned as either an Hb response or an increase in Hb to >12 g/dL without transfusion. When rhEPO is given to prevent anemia, response criteria are less universally deﬁned. A working deﬁnition of response could be a drop of Hb by less than 2 g/dL (Hct by less than 6 percentage points) without need for transfusion. For MDS patients, some groups deﬁne a complete response (CR) as a Hb value >11.5 g/d/ and a partial response (PR) as an increase of Hb by at

546

Y. Beguin and G. Van Straelen

least 1.5 g/dL or transfusion independence in combination with stable Hb in previously transfused patients. Others deﬁne a good or major response (GR) as an increase in Hb by at least 2 g/d/ or transfusion independence in previously transfused patients, and a minor or partial response by an increase of Hb by 1–2 g/dL or 50% reduction in transfusion requirements. The latter definition has been proposed as a standardized deﬁnition of response, but has yet to be widely applied (Hellstrom-Lindberg 2003, see chapter 20).

Treatment schedules (Table 1) There is a clear dose-response effect with rhEPO and most studies in cancer patients have used doses of rhEPO in the range of 300–900 U/kg/wk, well above those given to renal failure patients. For instance, treatment of anemia in patients with advanced gastrointestinal cancer was much more successful with 10,000 U compared with 2,000 U TIW (Glimelius et al. 1998). In patients with metastatic breast cancer, 5,000 U TIW was more effective than 1,000 U TIW (Olsson et al. 2002). In patients with hematological malignancies, a dose of 2,000 U/d was associated with much lower response rates than 10,000 U daily (Osterborg et al. 1996). Daily doses of 5,000 U were more effective than lower doses and daily doses of 10,000 U did not bring about further improvement in anemic patients with myeloma or lymphoma (Cazzola et al. 1995). Similarly, darbepoetin alfa displays a dose-response relationship in the range of 0.5–4.5 μg/kg/wk in patients with lymphoid malignancies (Hedenus et al. 2002) or miscellaneous nonmyeloid malignancies (Smith Jr. et al. 2003), or in the range of 4.5–12 μg/kg/3 wks (but not beyond) in patients with solid tumors (Kotasek et al. 2003) receiving chemotherapy. However, a frontloading regimen of double-dose weekly darbepoietin alfa administered for 4 wks, followed by lower maintenance dose, decreased time to response, but overall response rates remained similar (Glaspy et al. 2003). Finally, weekly administration of a ﬁxed or weight-based dose of darbepoietin alfa using a front-loading schedule result in similar hematopoietic responses (Hesketh et al. 2004). The more convenient subcutaneous route of administration has been shown to ensure more favorable pharmacokinetics (Macdougall et al. 1989) that translates into higher efﬁcacy in renal failure patients (Paganini et al. 1995). There is no known difference in the efﬁcacy and safety proﬁle of epoetin alfa (Janssen-Cilag) or epoetin beta (Roche) (Deicher et al. 2004). Although there is a lot of commercial fuss about comparisons involving data from observational studies, a recent study combining three randomized trials has shown that 40,000 U/wk rhEpo alfa and 200 μg/wk darbepoietin alfa can achieve similar Hb responses and impact on transfusions (Schwartzberg et al. 2004). Most trials administered rhEPO thrice weekly, a schedule demonstrated to be more efﬁcient than daily injections in normal subjects

Prediction of response to rhEPO

547

(Breymann et al. 1996). Although once-weekly dosing of 40,000 U epoetin alpha has been shown to increase Hb, decrease transfusions and improve quality of life in a fashion analogous to what is obtained with 10,000 U thriceweekly administration (Gabrilove et al. 2001), it is only with epoetin beta that it has been formally proven that the same total dose of 30,000 IU given subcutaneously once weekly was at least as effective as 10,000 t.i.w. administration in anemic patients with lymphoproliferative malignancies (Cazzola et al. 2003). Novel long-acting rhEPO molecules may also considerably prolong exposure to the active drug and thus improve the efﬁcacy of therapy with fewer injections. One of these molecules is the “Novel Erythropoiesis-Stimulating Protein” (NESP) or darbepoetin alfa, which is produced by changing several amino acids in the rhEPO molecule in order to add additional carbohydrates and to prolong half-life by a factor of 3 while maintaining the same mechanism of action through receptor binding (Syed et al. 1998). Another longacting rhEPO molecule named “Continuous Erythropoietin Receptor Activator” (CERA) incorporates a large polymer chain and is characterized by less tight binding to and different uptake by EPO receptor, resulting in a more potent erythropoietic activity and a considerably extended half-life. Once-every-3-wks schedules of CERA are currently tested in clinical trials in cancer patients (Dmoszynska et al. 2004; Dougherty et al. 2004; Österborg et al. 2004; Macdougall 2005). The duration of treatment is of critical importance. In large clinical trials, whereas there is no signiﬁcant difference in the rate of transfusions between placebo and rhEPO-treated patients during the ﬁrst month of therapy, the difference becomes highly signiﬁcant during the second and third months of treatment (Abels 1992; Glaspy et al. 1997). In one of these trials also, the efﬁcacy of rhEPO appeared to be lower in cancer patients not treated with chemotherapy, because rhEPO was given for a shorter duration (and at a lower dose) (Abels 1992). This is because expansion of the erythropoietic marrow in response to rhEPO is very gradual and achieves maximum activity only after several weeks (Beguin et al. 1995). The response rate can thus be further improved when patients are treated for 6 months or more (Henry et al. 1994). In order to maximize “time with response”, it would be desirable to achieve a faster response. Whether this can be achieved without increasing costs by providing higher doses of rhEPO for a short period of time (front-loading concept) followed by lower maintenance doses remains to be demonstrated (Glaspy et al. 2003; Hesketh et al. 2004).

Disease-associated factors (Table 1) A number of mechanisms can be involved in the pathogenesis of anemia associated with cancer (Beguin 1996; Spivak 2005; Weiss et al. 2005, see chapter 6) and therefore interfere with response to rhEPO in individual

548

Y. Beguin and G. Van Straelen

patients. Red cell loss may result from hypersplenism, blood losses consecutive to hemorrhage or iatrogenic phlebotomy, and autoimmune or microangiopathic hemolysis. Chronic or acute bleeding is a frequent complication of cancer, particularly in thrombocytopenic patients. Red cell production may be diminished by bone marrow inﬁltration, marrow necrosis, hemophagocytosis, myeloﬁbrosis, deﬁciency of erythropoietic cofactors (folic acid, vitamin B12, iron), or infections. These mechanisms of anemia are much more prevalent in hematologic malignancies, but it is always important to identify them because speciﬁc therapeutic intervention can be effective. However, cancerassociated anemia is often delineated by the more general features of the so-called “anemia of chronic disorders” (ACD). ACD is a cytokine-driven condition characterized by inadequate production of EPO, inhibition of the proliferation of erythroid progenitor cells in the bone marrow and disturbances of iron utilization (Beguin 1996; Spivak 2005; Weiss et al. 2005). Other factors have been examined. Age and sex have not been reported to inﬂuence response. Except when there is major invasion by cancer cells and limited residual normal hematopoiesis, marrow involvement by the tumor does not appear to limit the efﬁcacy of rhEPO (Oster et al. 1990; Abels 1992; Musto et al. 1997; Littlewood et al. 2003), although this was found in one small study (Lastiri et al. 2002). The type of tumor has generally not inﬂuenced the response rate, provided that no other speciﬁc mechanism of anemia is at work. Patients with multiple myeloma or low-grade lymphoma apparently have similar response rates (Cazzola et al. 1995; Osterborg et al. 1996; Osterborg et al. 2002; Hedenus et al. 2003), except in one study where myeloma patients responded better (Hedenus et al. 2002). Although there were no apparent differences between hematologic and nonhematologic malignancies in the largest studies published (Abels 1992; Glaspy et al. 1997; Demetri et al. 1998; Glaspy et al. 2002), there has been a suggestion that patients with breast or colon cancer (Ludwig et al. 1993a), but not those with squamous cell carcinoma (Ludwig et al. 1993b), may respond less well than patients with myeloma, but this was not conﬁrmed in larger studies (Demetri et al. 1998; Glaspy et al. 2002).

Chemotherapy-related factors (Table 1) Anemia in cancer patients is often caused or aggravated by therapy with antineoplastic agents. In particular, treatment with platinum, but not with other chemotherapeutic agents, has been associated with impairment of EPO production (Wood et al. 1995). Patients who have been heavily pretreated with chemotherapy usually experience severe stem cell damage that could interfere with response to rhEPO (Musto et al. 1997), but other studies have not made the same observation (Demetri et al. 1998). As a matter of fact, the poorer response obtained in patients with lower platelet counts probably just indicates that (Cazzola et al. 1995; Osterborg et al. 1996; Osterborg et al.

Prediction of response to rhEPO

549

2002). In addition, lower doses of rhEPO may be effective in patients with excellent platelet counts (Cazzola et al. 1995). Furthermore, transfusionindependent patients are more likely to respond than patients previously receiving transfusions (Demetri et al. 1998; Glaspy et al. 2002; Hedenus et al. 2002; Lastiri et al. 2002; Osterborg et al. 2002). Indeed, a meta-analysis of ﬁve randomized, double-blind, placebo-controlled trials in cancer patients receiving chemotherapy and epoetin alfa showed that being transfusiondependent before rhEPO therapy was the most signiﬁcant predictor for subsequent transfusions (Couture et al. 2005). For patients treated concomitantly with chemotherapy, it is likely that response to rhEPO could be impaired in proportion to the intensity of the chemotherapy being administered. However, there is very little data in the literature to obtain evidence for such an effect. The negative impact of chemotherapy has been very well illustrated in animal studies in which rhEPO was much more “efﬁcient” when it was started before the administration of 5-FU, because it could then increase the Hct better while myelosuppression was not occurring yet (Matsumoto et al. 1990). Patients receiving chemotherapy of moderate intensity respond as well as those not receiving concomitant chemotherapy (Abels 1992; Cazzola et al. 1995; Osterborg et al. 1996; Quirt et al. 2001). It is probable that more intensive chemotherapy regimens would be associated with lower response rates. In particular, rhEPO therapy is not capable to stimulate erythropoiesis in the early period following intensiﬁed chemotherapy with autologous stem cell transplantation (Link et al. 1994). Finally, responders to chemotherapy may also beneﬁt more from rhEPO therapy than nonresponders (Bamias et al. 2003). Multicenter studies have shown the same Hb response (speed and magnitude) in patients receiving platinum-based vs other forms of chemotherapy (Pawlicki et al. 1997; Demetri et al. 1998). Retrospective analyses of two large phase IV community-based studies (Glaspy et al. 2002), as well as of four prospective randomized trials (Littlewood et al. 2003) conﬁrmed that the pattern of response was quite similar with these two forms of chemotherapy. In a large study (Abels 1992), patients receiving platinum-based chemotherapy responded more rapidly than those receiving other combinations but the overall response rate was similar in the two groups, whereas in other small studies patients receiving platinum chemotherapy responded a little better (Oberhoff et al. 1998; Lastiri et al. 2002). However, dose intensity of the two forms of chemotherapy was not assessed and it is therefore impossible to compare the degrees of myelosuppression induced by chemotherapy and, thus, the capacity of rhEPO to overcome it.

Functional iron deﬁciency Iron requirements for erythropoiesis, and in particular EPO-stimulated erythropoiesis, have been increasingly recognized and red cell production may be

550

Y. Beguin and G. Van Straelen

Table 3. Laboratory ﬁndings in functional ID • Normal or increased ferritin • Laboratory signs of iron deﬁcient erythropoiesis: Serum iron <60 μg/dl Transferrin saturation <20% Hypochromic RBC >10% CHr <26 pg Soluble transferrin receptor >7 mg/L Erythrocyte protoporphyrin >70 μg/dl

limited by inadequate iron supply (Goodnough et al. 2000; Cavill 2002; Eschbach 2005, see chapter 26) (Table 3). Absolute iron deﬁciency (ID) is deﬁned by the exhaustion of iron stores in macrophages and hepatocytes (Provan 1999). It is characterized by serum ferritin values decreased below the normal range. However, ID erythropoiesis and anemia may occur even if iron stores are not exhausted or are even elevated (Brugnara et al. 1993; Brugnara et al. 1994). This is called functional ID, deﬁned as an imbalance between iron needs in the bone marrow and iron supply by macrophages (Fig. 1). This can be encountered in two different situations, corresponding to either decreased iron supply or increased iron needs. Iron release by macrophage is impaired typically by infection, inﬂammation or cancer (Beguin 1996; Spivak 2005; Weiss et al. 2005, see chapter 6). Iron needs are increased when marrow erythroid activity is stimulated, the best example being provided by rhEPO therapy (Brugnara et al. 1993; Brugnara et al. 1994). Macrophage iron, originating from normal or even elevated stores as well as from Hb-iron recycled when senescent RBC are phagocytosed, may not be mobilized sufﬁciently rapidly to match iron needs for the production of new red cells. Functional ID is a major factor limiting the efﬁcacy of erythropoietic agents (Fig. 1). It can occur before rhEPO therapy is started, either because iron stores are absent (absolute iron deﬁciency) or because storage iron release is impaired, a typical feature of ACD (Fillet et al. 1989). It can also develop in the course of rhEPO therapy when iron stores become progressively exhausted or more frequently when the increased iron needs of an expanding erythroid marrow cannot be matched by sufﬁcient mobilization of often enlarged iron stores. Indeed, the vast majority of renal failure patients treated with rhEPO develop functional iron deﬁciency that limits seriously their erythropoietic response (Macdougall 1999). Similar observations have been made in patients receiving rhEPO to facilitate an aggressive program of autologous blood donation (Goodnough et al. 1998, see chapter 26). Although this has not been speciﬁcally examined in cancer patients treated with erythropoietic agents, there is every reason to believe that its prevalence is very high in this setting as well.

Prediction of response to rhEPO Macrophages (ferritin)

551

Plasma (transferrin)

Marrow (hemoglobin) 7

1

Normal

3

4

5

6

2

ACD (cancer…)

rhEPO

ACD + rhEPO Fig. 1. Iron metabolism in conditions illustrating functional iron deﬁciency. A. Normal: when senescent red cells are phagocytosed (1) by macrophages, iron is recycled into a transit pool (2); part is stored as ferritin (hatched area) (3) and the rest is released (4) to plasma transferrin (5); iron is then taken up (6) by the erythroid marrow (7) to produce normal red cells. Iron supply (4) by storage cells matches iron demand (6) by the erythroid marrow and transferrin remains adequately (20–40%) saturated (black ﬁlling) by iron (5). B. Anemia of chronic disorder (ACD): iron release by macrophages is blocked and more iron is stored as ferritin within these cells. Iron supply can no longer match iron demand by the erythroid marrow: transferrin saturation decreases (<20%), the erythroid marrow becomes functionally iron deﬁcient and new red cells are hypochromic. C. Treatment with rhEPO: the erythroid marrow expands upon intense stimulation by rhEPO. Its increased demand for iron cannot be matched by storage iron release: transferrin saturation decreases (<20%), the erythroid marrow becomes functionally iron deﬁcient and new red cells are hypochromic. d. ACD treated with rhEPO: impaired iron supply and increased iron demand combine to decrease transferrin saturation and cause functional ID

A number of biological tools can be used to assess the adequacy of iron supply to the bone marrow (Table 3) (Kaltwasser et al. 1999; Goodnough et al. 2000). Serum ferritin levels are directly proportional to storage iron in macrophages and hepatocytes (Worwood 1990). However, numerous conditions, including hepatic cytolysis, inﬂammation, and renal failure, are associated with falsely elevated serum ferritin levels. Indeed cutoff values for absolute ID can be as high as 40–120 μg/L instead of the classical 12 μg/L in situations such as renal failure or cancer. On the other hand, serum iron saturates serum transferrin to a certain degree and transferrin saturation is a

552

Y. Beguin and G. Van Straelen

reﬂection of the equilibrium between iron supply and iron usage (Fig. 1) (Ponka et al. 1998). A soluble form of the transferrin receptor (sTfR) circulates in the plasma in proportion to the total body mass of cellular TfR (Beguin 2003). It is therefore largely inﬂuenced by the level of erythropoietic activity (through changes in the number of erythroblasts) and to a lesser extent by iron stores (through regulation of the number of TfR per cell) (Beguin 2003). However, as the impact of erythropoiesis quantitatively predominates, sTfR cannot be used as a marker of iron deﬁciency during rhEPO treatment (Beguin 2003). The percentage of hypochromic red cells increases over the upper limit of 5% when erythroblasts are deprived of adequate iron supply (Brugnara 1998). However, as reticulocytes are 20% bigger than mature red cells for the same amount of Hb, they are also hypochromic and high reticulocytosis should be taken into account in the interpretation of the percent of hypochromic red cells (Bovy et al. 1999; Bovy et al. 2004; Bovy et al. 2005). The Hb content of reticulocytes (CHr) will also decrease below 26 pg when the marrow does not receive enough iron to match its requirements (Brugnara 1998; Mast et al. 2002). As reticulocytes have a much shorter life span than red cells, the CHr will change much more rapidly than the percent of hypochromic red cells following the recent onset of iron deﬁcient erythropoiesis (Brugnara et al. 1993; Brugnara et al. 1994). Because there is some concern that tumor cells may need iron for optimal growth (Weinberg 1996), routine iron supplementation of all cancer patients receiving erythropoietic agents is not recommended. However, this should be balanced with the fact that transfusion of one RBC unit also provides a large amount (200 mg) of iron. Iron supplements should be given when absolute or functional ID is suspected (Fig. 2). The experience in iron-replete dialysis patients has clearly indicated that oral iron supplementation is not effective (Macdougall et al. 1996) but that IV iron substantially improves response when rhEPO therapy is instituted (Macdougall et al. 1996) and allows considerable (∼40%) reduction in rhEPO dose requirements during maintenance (Ahsan 2000; Besarab et al. 2000; Johnson et al. 2001). In predialysis patients (Aggarwal et al. 2003) or in patients undergoing preoperative stimulation of erythropoiesis (Rohling et al. 2000) as well, the IV route of iron supplementation has proved superior to the oral route, but too low doses were not effective (Olijhoek et al. 2001; Stoves et al. 2001). Iron usage has not been energetically pursued in clinical trials of erythropoietic agents in cancer patients and was generally left to the discretion of the individual investigator (Glaspy et al. 1999). In addition, iron has usually been given orally, a method proved to be of little efﬁcacy in other settings and presumably even less effective in cancer patients because of impaired iron absorption, another characteristic of the ACD (Beguin 1996; Spivak 2005; Weiss et al. 2005). The safety and efﬁcacy of IV iron to correct functional ID and improve anemia has been well documented in rheumatoid arthritis during rhEPO therapy (Vreugdenhil et al. 1992) or after failure of

Prediction of response to rhEPO

553

rhEPO + IV iron

IV iron RBC iron Fig. 2. Correction of functional iron deﬁciency by intravenous iron. The plain arrows represent recycling of red blood cell iron, as described in the third panel of Fig. 3 “rhEPO therapy” in which the expansion of erythroid marrow by rhEPO causes functional ID. The additional iron provided by intravenous iron products (dotted line) is also ﬁrst taken up by macrophages that process it to release iron from the iron-glycan complex. Iron is then available for release by macrophages to plasma transferrin. Adding up to iron recycled from phagocytosed red cells, this allows for correction of transferrin saturation (hatched area) and provision of sufﬁcient iron for erythropoiesis. The erythroid marrow can further expand without limitation by the iron supply

oral iron in juvenile chronic arthritis, another form of ACD (Martini et al. 1994). A recent study in cancer patients undergoing rhEPO therapy has demonstrated that systematic IV iron supply clearly improved the early response to EPO and that oral iron was not effective (Auerbach et al. 2004). Despite many questions left unanswered (Beguin 2005), this study should be hailed as the ﬁrst paper to address the question of iron supply during rhEPO therapy in cancer patients and should open the way for better designed clinical trials. The safety of systematic IV iron supplementation has been demonstrated in renal failure patients under treatment with erythropoietic agents (Agarwal et al. 2002; Fishbane 2003; Afzali et al. 2004; Aronoff 2004; Van Wyck 2004). There are three major forms of IV iron on the market (iron dextran, iron gluconate and iron sucrose or saccharate) (Danielson 2004). All three compounds are primarily taken up by macrophages from where iron is released to plasma transferrin that transports it to the erythroid marrow (Fig. 2). Iron is very slowly released over a period of weeks from iron dextran complexes, allowing to inject very large doses of iron in a single infusion (“total dose infusion”), but much more rapidly from iron sucrose and iron gluconate complexes, so that their maximum tolerated doses are approximately 500 and 250 mg iron, respectively. On the other hand, iron dextran has been associated with rare but potentially fatal anaphylactic reactions (Bailie et al. 2005). The safety proﬁle of iron sucrose (Yee et al. 2002) or iron gluconate (Fishbane et al. 2001) makes them the preferred IV compounds (Drueke et al. 1997).

554

Y. Beguin and G. Van Straelen

Iron sucrose has the advantage of allowing higher iron doses to be given at once, because iron gluconate at comparable doses would be associated with more toxicity due to free iron release (Van Wyck 2004). Because of limited data in cancer patients, recommendations can only be based on guidelines that have been published to provide treatment schedules with IV iron in renal failure patients (NKF-K/DOQi 2001b; Locatelli et al. 2004). An initial weekly loading dose of 100–300 mg IV iron could be recommended during the correction phase of the anemia, particularly, if there are signs of real or functional ID, while much lower doses (if any) are necessary during the maintenance phase. Iron supplementation should be targeted to maintain hypochromic red cells below 2.5%, CHr above 26 pg and transferrin saturation at 25–40%. To avoid toxicity from iron excess, IV iron should be withheld when transferrin saturation is above 50% and/or serum ferritin greater than 1,000 ng/mL, although the latter may not hold true in cancer patients. Oral iron should only be given to patients in whom IV supplements are not feasible or who do not tolerate them. Iron should be withheld in case of active sepsis (Aronoff 2004). Concomitant administration with chemotherapy, particularly anthracyclines, should be avoided, because transiently elevated transferrin saturation may enhance the toxicity of treatment (Link et al. 1996).

Inﬂammation Besides functional ID, numerous potential causes of resistance to rhEPO have been identiﬁed in renal failure patients, but most of them are encountered only in small numbers of patients (Macdougall 2001; NKF-K/DoQi 2001b). However, inﬂammation has increasingly been recognized as the major cause of resistance to treatment (usually deﬁned as higher rhEPO doses required to maintain similar target Hb) and efforts have been put into the identiﬁcation of potential mechanisms (Macdougall et al. 2002; Del Vecchio et al. 2005). Infections have been shown to cause hyporesponsiveness to rhEPO (Danielson et al. 1995). Elevated baseline ﬁbrinogen has been shown to predict failure of rhEPO therapy in hemodialysis patients (Beguin et al. 1993b). The weekly rhEPO dose was 80% higher in patients with Creactive protein (CRP) >20 mg/L compared to those with lower values (Barany et al. 1997). Even intermittent CRP elevations could reﬂect chronic inﬂammation impairing response to EPO (Sezer et al. 2004). Indeed, in welldialyzed and iron-replete patients, the acute-phase response, as represented by low serum albumin (a measure of both nutrition and inﬂammation) or high CRP in the context of high ferritin and low transferrin, was the most important predictor of rhEPO resistance (Gunnell et al. 1999). Malnutrition could also be a consequence of chronic inﬂammation (Del Vecchio et al. 2005). Induction of tumor necrosis factor alfa (TNF-α) (Takemasa et al. 2000;

Prediction of response to rhEPO

555

Macdougall et al. 2002; Cooper et al. 2003b), Interleukin (IL)-10 (Macdougall et al. 2002; Cooper et al. 2003a; Cooper et al. 2003b), Interferon gamma (IFN-γ) (Macdougall et al. 2002; Cooper et al. 2003b), IL-12 (Macdougall et al. 2002), IL-13 (Cooper et al. 2003b), soluble TNF receptor p80 (Kato et al. 2001), decreased CD28 expression on T lymphocytes (Macdougall et al. 2002; Cooper et al. 2003a), but not soluble IL-2 receptor (sIL-2R) (Cooper et al. 2003a) or IL-4 (Cooper et al. 2003b), have been suggested as potential mechanisms. Findings have been less consistent with IL-6 (Kato et al. 2001; Cooper et al. 2003a). At the molecular level, the EPO signaling pathway in early erythroid progenitors, the so-called burst-forming unit erythroid (BFUE), has been found to be attenuated in EPO-resistant patients as a result of dephosphorylation of STAT5 via upregulation of SHP-1 protein phosphatase activity (Akagi et al. 2004). These relatively EPO-resistant patients have a greater disease burden and in particular experience more infections (Kausz et al. 2005). Changing from conventional to ultrapure dialysis could reduce inﬂammation and hence decrease rhEPO dose requirements (Hsu et al. 2004). Following nephrectomy of a failed kidney transplant, patients demonstrated decreased inﬂammation and improved response to rhEPO (LopezGomez et al. 2004). Treatment with the phosphodiesterase inhibitor pentoxyphilline could reduce T-cell expression of inhibitory cytokines and restore response to rhEPO (Cooper et al. 2004; Macdougall 2004). Despite this wealth of evidence in renal failure patients that inﬂammation is a major factor impairing response to rhEPO, very little is known on the impact of inﬂammation on response to rhEPO therapy in cancer patients. Any source of inﬂammation, be it related to surgery, trauma, infection or concomitant disorders should interfere with response to rhEPO. Surgery is often followed by a transient loss of response to rhEPO, not only because it may be complicated by signiﬁcant blood losses, but also because post-operative erythropoiesis is limited by the inﬂammatory effect of surgery on iron metabolism that impairs iron reutilization (Biesma et al. 1995). Infections occur frequently in cancer patients receiving chemotherapy. This should slow or totally prevent response at the beginning of rhEPO therapy as well as abrogate response when the target Hb is being maintained with lower doses, requiring higher doses to be started again.

Predictive models Because response rates vary considerably among patients treated similarly and clinical efﬁcacy cannot be assessed before weeks of treatment, identiﬁcation of early predictors of response would be of major interest. The use of such prognostic factors of response could help provide the beneﬁts of rhEPO therapy to as many anemic cancer patients as possible while avoiding prolonged ineffective use of an expensive medication.

556

Y. Beguin and G. Van Straelen

Renal failure

Probability of response (%)

A predictive algorithm of response to rhEPO has ﬁrst been proposed in the setting of the anemia associated with renal failure (Fig. 3) (Beguin et al. 1993b). The best prediction by baseline parameters only was obtained with pretreatment soluble transferrin receptor (sTfR) and ﬁbrinogen. Serum sTfR represents a quantitative measure of erythropoietic activity (Huebers et al. 1990) and is also increased when functional iron deﬁciency develops (Skikne et al. 1990). Soluble TfR is of particular interest in assessing response to rhEPO, including in patients with chronic renal failure (Beguin et al. 1987; Eschbach et al. 1992; Barosi et al. 1993; Beguin et al. 1993b; Beguin et al. 1994; Beguin et al. 1995; Ahluwalia et al. 1997; Mizuguchi et al. 1997; Lorenzo et al. 2001; Nakanishi et al. 2001; Bovy et al. 2004). There was a 100% response rate when both sTfR and ﬁbrinogen were low, versus only 29% when there were both high, and 67% when one was low and the other high (Beguin et al. 1993b). The value of baseline sTfR was conﬁrmed in another study (Ahluwalia et al. 1997). Indeed, higher sTfR levels in renal failure patients not receiving rhEPO therapy are often associated with functional iron deﬁciency (Fernandez-Rodriguez et al. 1999; Matsuda et al. 2002; Gupta et al. 2003). Changes of sTfR after 2 weeks of treatment were also predictive (Beguin et al. 1993b). When the 2-week sTfR increment was ≥20%, the response rate was 96%. When sTfR increment was <20%, the response rate was 100% when baseline sTfR was low and ﬁbrinogen normal, 12% when baseline ﬁbrinogen was elevated and 62% when baseline ﬁbrinogen was normal but baseline sTfR high. In other studies as well, early sTfR increments correlate with later Hb responses (Beguin et al. 1995; Mizuguchi et al.

140 120

96%

100%

62%

12%

> 20

< 20 N < 3.5

< 20 N > 3.5

< 20 High

100 80 60 40 20 0

2-wk sTfR increment Baseline fibrinogen Baseline sTfR

(%) (μg/L)

Fig. 3. Prediction of response to EPO in the anemia of renal failure by baseline sTfR (an indicator of functional ID), baseline ﬁbrinogen (a parameter of inﬂammation) and the 2-wk sTfR increment (a marker of increasing erythropoietic activity) (Beguin et al. 1993b)

Prediction of response to rhEPO

557

1997). The same is true for detecting an imminent Hb response to an increased rhEPO dose, which could be predicted with perfect speciﬁcity after 1 week based on a 20% increment of sTfR over baseline, but whose sensitivity could be increased when combined with changes in reticulocyte counts (Ahluwalia et al. 1997). Good (low baseline and large increment) and poor (high baseline and small increment) responders will thus have similar sTfR levels during treatment, so that such absolute levels cannot distinguish them (Hou et al. 1998). These prognostic factors illustrate the importance of the early erythropoietic response (changes of sTfR levels), subclinical inﬂammation (ﬁbrinogen) and functional iron deﬁciency (baseline sTfR).

Baseline parameters Serum EPO Theoretically, cancer patients with a defect in their capacity to produce EPO would be more likely to respond to rhEPO than those with adequate serum EPO levels for their degree of anemia (Canadian Erythropoietin Study Group 1991). As EPO levels must be interpreted in relation to the degree of anemia, the ratio of observed-to-predicted EPO levels (O/P ratio) represents a better assessment of the adequacy of EPO production (Beguin et al. 1993a). Based on regression equations obtained in reference subjects, predicted log (EPO) values can be derived for each Hct, and O/P ratios of observed-topredicted values can be calculated (95% conﬁdence limits 0.80–1.20) (Beguin et al. 1993a). In patients with hematologic malignancies, it has been observed that low baseline serum EPO levels (Ludwig et al. 1994) or decreased O/P ratio (Cazzola et al. 1992; Musto et al. 1997) were associated with a signiﬁcantly higher probability of response to rhEPO. This has been conﬁrmed in large multicenter trials in patients with multiple myeloma or non-Hodgkin’s lymphoma (Hedenus et al. 2003; Cazzola et al. 1995; Osterborg et al. 1996), as well as in a combined analysis of several trials incorporating both patients with solid tumors and hematological malignancies (Littlewood et al. 2003). An O/P ratio <0.9 was found to be associated with high response rates, whereas patients with an O/P ratio >0.9 rarely beneﬁted from therapy (Cazzola et al. 1995; Cazzola et al. 1996; Osterborg et al. 1996). Even in patients with relative EPO deﬁciency (serum EPO <100 U/L), the lower the serum EPO, the higher the likelihood of response (Cazzola et al. 2003). In addition, in a single study that did not show a correlation between baseline EPO and Hb response, an association was still found between serum EPO and the later risk of receiving a blood transfusion despite darbepoietin alfa treatment (Hedenus et al. 2002). Nevertheless, studies in patients with solid tumors have failed to conﬁrm such a consistent predictive value of baseline EPO even when EPO deﬁciency was demonstrated in part of the patients

558

Y. Beguin and G. Van Straelen

(Platanias et al. 1991; Abels 1992; Ponchio et al. 1992; Cascinu et al. 1994; Glaspy et al. 1997; Demetri et al. 1998; Glimelius et al. 1998; Oberhoff et al. 1998; Glaspy et al. 2002; Gonzalez-Baron et al. 2002; Lastiri et al. 2002). However, a study aiming at preventing anemia in patients with ovarian carcinoma undergoing platinum-based chemotherapy showed a trend for lower transfusion needs in those with an O/P ratio <0.8 (ten Bokkel Huinink et al. 1998). In another trial in patients with solid tumors undergoing platinumbased chemotherapy, baseline O/P ratio <0.9 was strongly associated with response (Bamias et al. 2003). Other studies in children (Leon et al. 1998) or adults (Fjornes et al. 1998) with solid tumors found baseline EPO to be highly predictive of response (Leon et al. 1998). In addition, a small study in patients with a variety of solid tumors suggested that the ratio of baseline EPO/ corrected reticulocyte count could provide some predictive information (Charuruks et al. 2001). Of importance, in patients treated with chemotherapy, serum EPO should be evaluated just prior to chemotherapy for its interpretation to be valid (Fig. 4). Indeed, without any change in Hct, serum EPO may be inappropriately elevated in the 2 weeks after chemotherapy compared to pre-chemotherapy values, most probably because myelosuppression then decreases EPO utilization by target cells (Beguin et al. 1991). Therefore, it cannot be excluded that the failure to predict response may just be related to an inadequate timing of serum EPO sampling (Glimelius et al. 1998; Osterborg et al. 2002), but most studies do not report when serum EPO was evaluated relative to chemotherapy.

Values expressed as % of baseline Chemo

300

Chemo

200

200

150

sTfR Hb

Epo 100

100

0

50 -5

0

5

10

15

20

25

Days post-chemotherapy Fig. 4. Changes in serum EPO, Hb and sTfR after a chemotherapy cycle. Chemotherapy transiently causes an increase in serum EPO levels that is disproportionate to the degree of anemia

Prediction of response to rhEPO

559

Other parameters Other baseline parameters have been examined as possible predictors of response. Pretreatment Hct is of course an important factor when rhEPO is given for the prevention of anemia, but is no longer helpful when it is given after anemia is well established (Ludwig et al. 1994; Demetri et al. 1998; Glaspy et al. 2002; Gonzalez-Baron et al. 2002; Littlewood et al. 2003), even if a single study found an association between low baseline Hb and the risk of being transfused on study (Hedenus et al. 2002). Other measurements of erythropoietic activity, such as the reticulocyte count or sTfR levels were not predictive of response (Ludwig et al. 1994; Ludwig et al. 1994; Musto et al. 1997; Gonzalez-Baron et al. 2002; Littlewood et al. 2003). Patients with more BFU-E in the peripheral blood apparently respond better (Musto et al. 1997). Platelet counts below 50–100 × 109/L are usually (Cazzola et al. 1995; Osterborg et al. 1996; Osterborg et al. 2002), but not always (Ludwig et al. 1994), associated with poorer responses. Similarly, low neutrophil counts may be associated with a lesser probability of response (Cazzola et al. 1995), but this has not been found in most studies (Ludwig et al. 1994; Osterborg et al. 1996; Osterborg et al. 2002; Littlewood et al. 2003). Serum creatinine (Cazzola et al. 1995; Osterborg et al. 1996), except in one small study (Fjornes et al. 1998), is not predictive of response. Baseline ferritin below 400 ng/mL has been found to predict for good response in one study (Littlewood et al. 2003) but not other studies (Ludwig et al. 1994; Gonzalez-Baron et al. 2002), while high transferrin saturation has not predicted failure in most studies (Ludwig et al. 1994; Cazzola et al. 1995; Osterborg et al. 1996; Gonzalez-Baron et al. 2002; Osterborg et al. 2002), except one (Littlewood et al. 2003). Only large doses of rhEPO can overcome the strong inhibition of erythropoiesis induced by such cytokines as IL-1, TNF-α and IFN-γ. The results of a study evaluating the predictive values of serum levels of these cytokines were disappointing (Ludwig et al. 1994). This was conﬁrmed by others (Lastiri et al. 2002). This is not entirely surprising since serum levels of these cytokines may not be relevant, whereas local intramedullary levels may be much more important but are very difﬁcult to evaluate. However, others have observed that response was more likely in patients with low serum levels of TNF-α or IL-1 (Musto et al. 1997).

Early changes in erythropoietic parameters Early changes in parameters of erythropoietic activity observed after 2 weeks of treatment could be more informative. A rapid elevation of Hb levels by 0.3–1.0 g/dL after 2–4 weeks often predicted a good probability of later response (Ludwig et al. 1994; Cazzola et al. 1995; Henry et al. 1995; Glaspy et al. 1997; Demetri et al. 1998; Glimelius et al. 1998; Quirt et al. 2001; Glaspy et al. 2002; Gonzalez-Baron et al. 2002; Littlewood et al. 2003). An Hb

560

Y. Beguin and G. Van Straelen

increment of 0.5 g/dL after 4 wks was conﬁrmed by a number of different statistical techniques to provide the best predictive power among several hematological and iron parameters (Gonzalez-Baron et al. 2002). An increase of reticulocyte counts by ≥40,000/μl from baseline to week 2 or 4 appeared to be predictive of response but its discriminative power was weak (Ludwig et al. 1994; Henry et al. 1995; Gonzalez-Baron et al. 2002; Littlewood et al. 2003) and it was not found in other studies (Lastiri et al. 2002). In several studies, hematologic response to rhEPO was strongly associated with early increases (20–25% over baseline) of sTfR levels after 1–2 weeks of treatment (Cazzola et al. 1992; Ponchio et al. 1992; Ludwig et al. 1994; Cazzola et al. 1996; Beshara et al. 1997; Musto et al. 1997; Beguin 1998a; Beguin 1998b). A study presenting the most comprehensive analysis found that increases of Hb, sTfR and reticulocytes, as well as decreases of serum EPO, ferritin, iron, CRP or neopterin after 2 weeks were all correlated with response to rhEPO (Ludwig et al. 1994). Apart from changes in Hb and CRP, changes in sTfR levels was the single best predictor of response among 63 parameters examined in univariate analysis at baseline and after 2 weeks, although it provided information redundant of the Hb changes. In another large series of patients, increase in Hb and reticulocytes as well as lower absolute values for ferritin and/or transferrin saturation after 2–4 wks were all associated with higher response rates to rhEPO (Littlewood et al. 2003). More recently, Hagberg at al. observed that healthy volunteers treated with rhEPO showed a more distinct increase in β-globin mRNA levels than in Hb, sTfR or reticulocytes, but this remains to be studied in cancer patients (Hagberg et al. 2003).

Predictive algorithms based on early changes Various models have sought to combine the predictive power of several parameters. In a study including similar numbers of patients with solid tumors or hematologic malignancies, if after 2 weeks of therapy EPO was >100 mU/ml and Hb had not increased by at least 0.5 g/dL, there was a 94% probability of unresponsiveness; otherwise response was likely in 80% of the patients (Ludwig et al. 1994) (Fig. 5). If serum EPO was <100 mU/ml and Hb had increased by ≥0.5 g/dL, the probability of responses was 100%; otherwise the probability of failure was 62%. However, 34 out of the 80 patients evaluated did not fall into any of these two categories and thus prediction was valid only in a little more than half of the patients. The predictive value of a decrease in serum EPO levels may have two explanations. Endogenous serum EPO could decrease as the Hct rose in responders, but the magnitude of the Hct changes by 2 weeks seemed to be too small for that. On the other hand, EPO could be utilized by an expanding erythroid marrow or conversely accumulate in nonresponders, but it cannot be excluded that these later patients were receiving more intensive chemotherapy than others and thus

Pr obabilit y of response (%)

Prediction of response to rhEPO

561

140 120

100% n=15

6% n=31

< 100 > 0.5

> 100 < 0.5

70% n=30

100 80 60 40 20 0

2-wk serum EPO Epo 2-wk Hb increment

<100 >100 (mU/ml) <0.5 >0.5 (g/dl)

Fig. 5. Prediction of response to rhEPO in the anemia of cancer by the week-2 absolute serum EPO level and the 2-wk Hb increment (Ludwig et al. 1994)

be more likely to have inappropriate increases of endogenous serum EPO values (Beguin et al. 1991). Alternatively, a serum ferritin value of ≥400 ng/mL after 2 weeks predicted for failure in 88% of the cases, whereas levels <400 ng/mL predicted for success in 72% of the cases. However, the speciﬁc cutpoint of 400 ng/mL cannot be extrapolated to other patients because it depends so much on the previous transfusion history. In a subset of patients from a large multicenter study, some prediction of response could be derived from changes observed in reticulocytes and Hb from baseline to week 2 of therapy (Abels 1992; Henry et al. 1995) (Fig. 6). Among patients not receiving chemotherapy (Fig. 6A), the response rate was poor when the 2-week increment of Hb level was <0.5 g/dL, but it was excellent when the Hb level or reticulocyte count increased by ≥0.5 g/dL or ≥40,000/μL, respectively. The predictive power of these parameters was much less substantial when the hemoglobin increased by ≥0.5 g/dL but the reticulocyte elevation was smaller. Adequate prediction of response could not be provided on the basis of Hb and reticulocyte changes in patients receiving concomitant chemotherapy (Fig. 6B). Although some improvement in forecast could be obtained in patients increasing their Hb by ≥1 g/dL after 4 weeks of treatment, predicting response on the basis of the response itself may appear to be trivial.

Predictive algorithms based on a combination of baseline parameters and early changes A combination of baseline parameters and early changes observed after 2 weeks of rhEPO may provide another useful approach. Among evaluable

562

Y. Beguin and G. Van Straelen

Probability of response (%)

A

140 120 100

91% n=11

36% n=14

17% n=6

4% n=23

80 60 40 20 0

2-wk Hb increment 2-wk retic increment

> 0.5

< 0.5

> 40

< 40

> 40

< 40

67% n=21

66% n=29

50% n=20

48% n=62

(g/dL) (x103/μL)

Probability of response (%)

B 140 120 100 80 60 40 20 0

2-wk Hb increment 2-wk retic increment

> 0.5 > 40

< 40

< 0.5 > 40

< 40

(g/dL) (x103/μL)

Fig. 6. Prediction of response to EPO in the anemia of cancer by the 2-wk Hb and reticulocyte increments (Henry et al. 1995). A good prediction can be obtained in patients not receiving chemotherapy (A) but not in those receiving chemotherapy (B)

patients treated in a large multicenter study, the failure rate was almost 90% when baseline serum O/P Epo was higher than 0.9 or when serum O/P Epo was less than 0.9 but the Hb increment by week 2 was <0.3 g/dL (Cazzola et al. 1995) (Fig. 7). On the other hand, the success rate was around 90% when baseline serum O/P Epo was less than 0.9 and Hb increased by ≥0.3 g/d. Similar ﬁndings were obtained in a smaller study in children with solid tumors: an O/P ratio <1.0 at baseline and an Hb increment >0.5 gr/dL after 2 weeks were associated with higher response rates (Leon et al. 1998). In another large single center study, the combined use of baseline serum EPO and the 2-week increment of sTfR proved to be very powerful (Cazzola et al. 1996) (Fig. 8). Only 18% of patients with a baseline serum EPO greater

Probability of response (%)

Prediction of response to rhEPO

563

140 120

13% n=8

0% n=6

88% n=34

100 80 60 40 20 0

Baseline O/P EPO 2-wk Hb increment

> 0.9

< 0.9 < 0.3

> 0.3

(g/dL)

Probability of response (%)

Fig. 7. Prediction of response to rhEPO in the anemia associated with lymphoma or multiple myeloma by the baseline observed/predicted serum EPO ratio and the 2-wk Hb increment (Cazzola et al. 1995)

140 120

18% n=17

29% n=7

96% n=24

100 80 60 40 20 0

Baseline serum EPO 2-wk sTfR increment

> 100

< 100 < 25

> 25

(mU/mL) (%)

Fig. 8. Prediction of response to rhEPO in the anemia of cancer by the baseline serum EPO level and the 2-wk sTfR increment (Cazzola, 1996 18252 /id)

than 100 mU/ml responded to treatment, and only 29% responded when the baseline serum EPO was <100 mU/mL but the 2-week sTfR increment was less than 25%. On the other hand, the response rate was 96% among patients with a low baseline serum EPO and a substantial sTfR elevation. In a large series of patients combined from four different randomized trials, increase in Hb and reticulocytes as well as lower absolute values for ferritin and/or transferrin saturation after 2–4 wks were all associated with higher response rates to rhEPO (Littlewood et al. 2003). In this study, two-factor analysis, combining Hb increment after 4 wks >1 g/dL with either baseline EPO <100 U/L or ferritin <400 ng/mL or transferrin saturation <40%, yielded response rates of about 90% in the good vs 45% in the poor prognostic groups, respectively.

564

Y. Beguin and G. Van Straelen

Three-factor analysis did not add any speciﬁcity or sensitivity to the predictive algorithm.

Applicability of predictive factors There are a number of theoretical reasons why some or all of these parameters could not be of value in certain situations. While evaluation of endogenous EPO production may be relevant in various forms of anemia, it is of no interest in subjects in whom the aim of rhEPO therapy is to prevent an anemia that is not yet present, in those in whom better tumor oxygenation before radiotherapy is sought, or in disorders characterized by universal EPO deﬁciency. Even among cancer patients, whereas low baseline serum EPO levels or decreased observed-to-predicted EPO levels (O/P ratio) were associated with a signiﬁcantly higher probability of response in patients with hematologic malignancies, this was usually not the case in patients with solid tumors. On the other hand, Hb increments after 2 weeks of treatment may be of value in steady state patients, but are of little help in transfused patients and in those in whom rhEPO is intended to prevent the occurrence of severe anemia but cannot avert some decrease in Hb induced by phlebotomy or myelosuppressive treatments. Finally, changes in parameters directly reﬂecting erythropoietic activity, i.e. reticulocyte counts and sTfR, may be the most appropriate. However, changes in reticulocyte counts may simply reﬂect output of shift reticulocytes and not true expansion of erythropoiesis, and often have not been found to be a good indicator of response (Beguin et al. 1993b; Ludwig et al. 1994). Although sTfR levels represent the best quantitative measurement of total erythropoietic activity, they may also increase secondary to functional ID (Huebers et al. 1990). In addition, particularly in patients treated with chemotherapy, the timing of the evaluation of these parameters relative to chemotherapy may be critical for their interpretation. For instance, measuring serum EPO after chemotherapy may yield elevated levels compared to pre-chemotherapy values, without any change in Hct (Beguin et al. 1991) (Fig. 4). Finally, the most effective predictive algorithm may also vary according to the dose of rhEPO used (Cazzola et al. 1995). For instance, front-loading regimens with high doses of rhEPO may well yield signiﬁcantly faster responses (Glaspy et al. 2003), which in turn could prove more useful to identify poor responders earlier, allowing physicians to decide to stop treatment after 2–4 wks when no evidence of stimulated erythropoiesis is detected.

Myelodysplastic syndromes (MDS) The response rate to rhEPO therapy is much less in MDS patients compared to patients with other forms of cancer (Hellstrom-Lindberg 2003, see chapter

Prediction of response to rhEPO

565

Table 4. Prediction of good response to rhEPO in MDS • Treatment with rhEPO alone: – Transfusion-independence – Serum EPO <200 mU/mL – RA or RAEB I • Treatment with rhEPO + rhG-CSF: – Transfusion needs <2 U/mo – Serum EPO <500 mU/mL – RA or RAEB I or RA-S

20). Hence, it would be even more critical to identify possible predictive factors of response in this particular patient population (Table 4). These patients can be treated with rhEPO alone or in combination with low-dose recombinant human granulocyte colony-stimulating factor (rhG-CSF). Although no study has compared the two approaches, the combination of rhEPO and rhG-CSF may be associated with better response rates (40–50%) compared to treatment with rhEPO alone (20–25%) (Hellstrom-Lindberg 2003, see chapter 20). Moreover, the addition of rhG-CSF has been associated with response in rhEPO-resistant patients, while rhG-CSF withdrawal has caused loss of response. In series of patients treated with rhEPO alone, age (Stein et al. 1991; Rose et al. 1995; Stasi et al. 1997a; Terpos et al. 2002), gender (Stein et al. 1991; Rose et al. 1995; Stasi et al. 1997a; Terpos et al. 2002), type of MDS (Stein et al. 1991), cytogenetics (Stasi et al. 1997a), time since diagnosis (Stein et al. 1991; Rose et al. 1995; Stasi et al. 1997a; Terpos et al. 2002), time since initiation of transfusions (Stein et al. 1991), transfusion requirements (Rose et al. 1995; Stasi et al. 1997a; Terpos et al. 2002; Terpos et al. 2002) or baseline Hct/Hb (Rose et al. 1995; Stasi et al. 1997a; Terpos et al. 2002), reticulocytes (Stasi et al. 1997a; Terpos et al. 2002), platelets (Terpos et al. 2002) or neutrophils (Terpos et al. 2002) have usually not been found to be associated with response. Others have obtained better responses in patients with refractory anemia (RA) compared to those with RA with ring sideroblasts (RAS) (Rose et al. 1995; ICSG 1998; Terpos et al. 2002) or RA with excess of blasts (RAEB) (Rose et al. 1995; ICSG 1998; Terpos et al. 2002; Wallvik et al. 2002), particularly those with RAEB-II (Terpos et al. 2002), patients with normal cytogenetics (Terpos et al. 2002; Wallvik et al. 2002) or no need for transfusions (Wallvik et al. 2002). Although some small series have not found baseline serum EPO levels to be useful predictors of response (Stein et al. 1991; Stasi et al. 1997b), most studies have found that responders had lower serum EPO levels at baseline compared to nonresponders (Depaoli et al. 1993; Rose et al. 1995; Stasi et al. 1997a; ICSG 1998; Terpos et al. 2002; Wallvik et al. 2002). Cutoff values of 50 (Wallvik et al. 2002), 100 (Rose

566

Y. Beguin and G. Van Straelen

et al. 1995), 150 mU/mL (Terpos et al. 2002), or 200 mU/mL (HellstromLindberg 1995; ICSG 1998) have been identiﬁed as the most signiﬁcant, although the use of the O/P ratio could improve separation into responders and nonresponders (Musto et al. 1995). Increased erythroid cellularity in bone marrow or elevated BFU-E in peripheral blood were also found to be highly predictive of response (Stasi et al. 1997a). Erythroid responses were mostly obtained in patients with the glutathione S-transferase (GST) M1 null genotype and not the GST T1 null genotype (Tsabouri et al. 2004). Poor responders have been found to have higher serum TNF-α levels with (Musto et al. 1994a) or without (Stasi et al. 1997b) elevated IL-1 levels compared to responders. Responders may be characterized by sTfR levels inadequately low for the degree of anemia (Musto et al. 1994a), but such evidence of intrinsic marrow hypoproliferation is not predictive of response to rhEPO (Bucalossi et al. 1996; Hellstrom-Lindberg et al. 1998). Soluble TfR levels have been shown to increase in response to rhEPO, but early elevation of sTfR levels may overly predict later Hb response because in some patients this may only represent stimulation of ineffective erythropoiesis (Adamson et al. 1992; Cazzola et al. 1992; Ghio et al. 1993; Musto et al. 1994b). Although nonresponders usually do not show modiﬁcations while responders increase sTfR levels (Musto et al. 1994b; Hellstrom-Lindberg et al. 1998; ICSG 1998), patients with increased sTfR without concomitant elevation of high ﬂuorescence reticulocytes also do not show signiﬁcant Hb response (Musto et al. 1994b). An early meta-analysis of 205 patients from 17 studies identiﬁed RAS vs other types of MDS, transfusion-dependence vs no need for transfusion, and serum EPO levels above 200 mU/mL vs lower levels, as predictors of poor response to rhEPO (Hellstrom-Lindberg 1995, see chapter 20) (Table 4). Patients with no transfusion requirements and MDS other than RAS had a response rate ≥50%, irrespective of their serum EPO levels. Patients with RAS had a 33% response rate when serum EPO was <200 mU/mL, but the others (RAS with transfusion needs or serum EPO >200 mU/mL, MDS other than RAS with transfusion needs and serum EPO >200 mU/mL showed response rates lower than 10%. It should be emphasized that prolonged duration of treatment may be associated with improved response rates (Terpos et al. 2002). Most reports of patients treated with a combination of rhEPO and rhGCSF have also failed to identify age (Negrin et al. 1993; Negrin et al. 1996; Mantovani et al. 2000), gender (Remacha et al. 1999; Mantovani et al. 2000; Hellstrom-Lindberg et al. 2003), type of MDS (Negrin et al. 1993; Negrin et al. 1996; Hellstrom-Lindberg et al. 1997; Remacha et al. 1999; Mantovani et al. 2000; Hellstrom-Lindberg et al. 2003), time since diagnosis (Negrin et al. 1993; Negrin et al. 1996; Hellstrom-Lindberg et al. 1997; Remacha et al. 1999; Mantovani et al. 2000), prior transfusion requirements (Negrin et al. 1993; Negrin et al. 1996; Mantovani et al. 2000), bone marrow erythroid cellularity (Negrin et al. 1996; Hellstrom-Lindberg et al. 1997), IPSS score

Prediction of response to rhEPO

567

(Hellstrom-Lindberg et al. 2003), baseline neutrophils (Negrin et al. 1993; Negrin et al. 1996; Hellstrom-Lindberg et al. 1997; Mantovani et al. 2000), platelets (Negrin et al. 1993; Negrin et al. 1996; Remacha et al. 1999; Mantovani et al. 2000) or reticulocytes (Negrin et al. 1993; Mantovani et al. 2000) as predictors of response. However, higher reticulocytes (Negrin et al. 1996), Hb (Mantovani et al. 2000), platelets (Hellstrom-Lindberg et al. 1997) or neutrophils (Remacha et al. 1999), as well as transfusion needs (Remacha et al. 1999; Hellstrom-Lindberg et al. 2003), have been found predictive in some studies. Favorable cytogenetics have sometimes (Negrin et al. 1996), but not always (Negrin et al. 1993; Hellstrom-Lindberg et al. 2003), been associated with better responses. The limited experience in RAEB-T and CMML has been very disappointing. Again patients responding well had lower serum EPO levels prior to initiation of rhEPO therapy (Vannucchi et al. 1993; Negrin et al. 1996; Hellstrom-Lindberg et al. 1997; Mantovani et al. 2000; Hellstrom-Lindberg et al. 2003), and cutoff values of 100 mU/mL (HellstromLindberg et al. 1997), 250 mU/mL (Mantovani et al. 2000) or 500 mU/mL (Negrin et al. 1996; Hellstrom-Lindberg et al. 1997) offered the best discriminative power. Few studies have tested the combination of rhEPO and rh granulocyte/macrophage (GM)-CSF, but only in small groups of patients. Only lower serum EPO (Thompson et al. 2000), higher erythroid cellularity (Stasi et al. 1999) and lower serum TNF-α levels (Stasi et al. 1999) were found associated with better response. No predictor of response was identiﬁed in a small series of patients treated with rhEPO and all transretinoic acid (ATRA) (Stasi et al. 2002). In an analysis of patients included in previous American and Scandinavian studies (Hellstrom-Lindberg et al. 1997), a predictive model based on serum EPO levels and transfusion requirements was proposed (Table 4). Using baseline serum EPO as a ternary variable (<100 mU/mL = score +2; 100–500 mU/mL = score +1; >500 mU/mL = score −3) and transfusion needs as a binary variable (<2 U/month = score +2; ≥2 U/month = score −2), three groups were separated: one group (score > +1) with a high probability of response (74%), one (score ±1) intermediate group (23%) and one group (score < −1) with poor response (7%). This model was later validated in other series of patients (Remacha et al. 1999; HellstromLindberg et al. 2003), but others obtained similar response rates in patients with scores > +1 or ±1 (Mantovani et al. 2000).

Conclusion Several algorithms have been proposed for patients with the anemia of cancer. Their sensitivity (how well the algorithm identiﬁes all those who will respond) and speciﬁcity (how well the algorithm excludes all those who will fail), and thus their overall efﬁcacy, vary considerably. In the study conducted by Ludwig (Ludwig et al. 1994), when one tries primarily to identify

568

Y. Beguin and G. Van Straelen Transfusion -free patient

Baseline serum EPO < 200 mU/mL

rhEPO Yes

2 -w k Hb increase > 0.3 gr/dL

No

rhEPO Yes

No

Stop rhEPO therapy

No

Transfusion -dependent patient

Baseline serum EPO < 200 mU/mL

No Yes

rhEPO

2 -w k sTfR increase > 20%

Yes

rhEPO

Fig. 9. Practical use of algorithms for prediction of response in cancer patients treated with rhEPO, based on baseline endogenous EPO level and an early (2-wk) indicator of increased erythropoietic activity. The ﬁrst step (baseline Epo) could be omitted in solid tumor patients. The only difference between untransfused and transfused patients is that the 2-wk Hb increment cannot be used in transfusiondependent patients and must be replaced by the 2-wk sTfR increment

nonresponders instead of responders, sensitivity and overall accuracy can be increased from 42% and 70% to 76% and 86%, respectively. Overall accuracy is not improved by doing so in the study conducted by Henry in patients receiving chemotherapy, because enhanced sensitivity (54%) is compensated by diminished speciﬁcity (52%). The positive predictive value (probability of response in those predicted to respond) of the algorithms is usually better then their negative predictive value (probability of failure in those predicted to fail). The best algorithms appear to be those combining an assessment of the adequacy of endogenous EPO production (at least in hematologic malignancies) together with some early indicators of erythropoietic marrow response (changes in Hb or sTfR). The following attitude can be proposed in practice (Fig. 9): (1) Baseline serum EPO should be measured at baseline in patients with hematologic malignancies: treatment should not be initiated if endogenous serum EPO is above 100 mU/mL (or 200 mU/mL in severely anemic patients) or the O/P ratio is >0.9.

Prediction of response to rhEPO

569

(2) Erythropoietic response should be assessed after 2 weeks: – In untransfused patients, if the Hb has increased by at least 0.3–0.5 g/dL, continue treatment; otherwise stop treatment or consider doubling the dose (although there is no published evidence that this will work (Bokemeyer et al. 2004)) and deﬁnitively discontinue EPO after 2 additional weeks if Hb has not increased by at least 0.3 g/dL. – In transfused patients, if sTfR has increased by at least 20%, continue treatment; otherwise stop treatment or consider doubling the dose (although there is no published evidence that this will work (Bokemeyer et al. 2004)) and deﬁnitively discontinue rhEPO after 2 additional weeks if sTfR has not increased by at least 20%. (3) It is critical that all preventable causes of failure are identiﬁed prospectively and corrected, or else no predictive model will be valid. In particular this includes vigorous iron supply and energetic treatment of intercurrent complications such as infections and bleeding. (4) MDS patients could be treated with rhEPO alone when they have RA or RAEB with serum EPO <200 mU/mL and no transfusion need, or with a combination of rhEPO and rhG-CSF if they have RA, RAS or RAEB with serum EPO <500 mU/mL and transfusion needs lower than 2 U/month (Hellstrom-Lindberg 2003, see also chapter 20) (Table 4). Failure to detect increases in sTfR after 2 weeks should justify discontinuation of rhEPO therapy.

References 1. Abels RI (1992) Use of recombinant human erythropoietin in the treatment of anemia in patients who have cancer. Semin Oncol 19 [Suppl 8]: 29–35 2. Adamson JW, Schuster M, Allen S, Haley NR (1992) Effectiveness of recombinant human erythropoietin therapy in myelodysplastic syndromes. Acta Haematol 87 [Suppl 1]: 20–24 3. Afzali B, Goldsmith DJ (2004) Intravenous iron therapy in renal failure: friend and foe? J Nephrol 17: 487–495 4. Agarwal R, Warnock D (2002) Issues related to iron replacement in chronic kidney disease. Semin Nephrol 22: 479–487 5. Aggarwal HK, Nand N, Singh S, Singh M, Hemant, Kaushik G (2003) Comparison of oral versus intravenous iron therapy in predialysis patients of chronic renal failure receiving recombinant human erythropoietin. J Assoc Physicians India 51: 170–174 6. Ahluwalia N, Skikne BS, Savin V, Chonko A (1997) Markers of masked iron deﬁciency and effectiveness of EPO therapy in chronic renal failure. Am J Kidney Dis 30: 532–541 7. Ahsan N (2000) Infusion of total dose iron versus oral iron supplementation in ambulatory peritoneal dialysis patients: a prospective, cross-over trial. Adv Perit Dial 16: 80–84

570

Y. Beguin and G. Van Straelen

8. Akagi S, Ichikawa H, Okada T, Sarai A, Sugimoto T, Morimoto H, Kihara T, Yano A, Nakao K, Nagake Y, Wada J, Makino H (2004) The critical role of SRC homology domain 2-containing tyrosine phosphatase-1 in recombinant human erythropoietin hyporesponsive anemia in chronic hemodialysis patients. J Am Soc Nephrol 15: 3215–3224 9. Aronoff GR (2004) Safety of intravenous iron in clinical practice: implications for anemia management protocols. J Am Soc Nephrol 15 [Suppl 2]: S99–106 10. Auerbach M, Ballard H, Trout JR, McIlwain M, Ackerman A, Bahrain H, Balan S, Barker L, Rana J (2004) Intravenous iron optimizes the response to recombinant human erythropoietin in cancer patients with chemotherapy-related anemia: a multicenter, open-label, randomized trial. J Clin Oncol 22: 1301–1307 11. Bailie GR, Clark JA, Lane CE, Lane PL (2005) Hypersensitivity reactions and deaths associated with intravenous iron preparations. Nephrol Dial Transplant 20: 1443–1449 12. Bamias A, Aravantinos G, Kalofonos C, Timotheadou N, Siafaka V, Vlahou I, Janinis D, Pectasides D, Pavlidis N, Fountzilas G (2003) Prevention of anemia in patients with solid tumors receiving platinum-based chemotherapy by recombinant human Erythropoietin (rHuEpo): a prospective, open label, randomized trial by the Hellenic Cooperative Oncology Group. Oncology 64: 102–110 13. Barany P, Divino Filho JC, Bergstrom J (1997) High C-reactive protein is a strong predictor of resistance to erythropoietin in hemodialysis patients. Am J Kidney Dis 29: 565–568 14. Barosi G, Merlo C, Palestra P, Liberato NL, Guarnone R, Di Dio F, Piazza V, Salvadeo A (1993) Variations in erythropoiesis and serum ferritin during erythropoietin therapy for anaemia of end-stage renal disease. Acta Haematol 90: 13–18 15. Beguin Y (1996) Erythropoietin and the anemia of cancer. Acta Clin Belg 51: 36–52 16. Beguin Y (1998a) Prediction of response to optimize outcome of treatment with erythropoietin. Semin Oncol 25 [Suppl 7]: 27–34 17. Beguin Y (1998b) Prediction of response to treatment with recombinant human erythropoietin in anaemia associated with cancer. Med Oncol 15 [Suppl 1]: 38–46 18. Beguin Y (2001) Prediction of response to rHuEpo. In: Bokemeyer C, Ludwig H (eds) Anaemia of cancer. Elsevier Science, Amsterdam, pp 153–165 19. Beguin Y (2002a) Prediction of response and other improvements on the limitations of recombinant human erythropoietin therapy in anemic cancer patients. Haematologica 87: 1209–1221 20. Beguin Y (2002b) Predictive factors for response of anemia to recombinant human erythropoietin. In: Nowrousian MR (ed) Recombinant human erythropoietin (rhEpo) in clinical oncology. Springer, Berlin Heidelberg, pp 263–286 21. Beguin Y (2003) Soluble transferrin receptor for the evaluation of erythropoiesis and iron status. Clin Chim Acta 329: 9–22 22. Beguin Y (2005) Intravenous iron and recombinant human erythropoietin in cancer patients. J Clin Oncol 23: 651–652 23. Beguin Y, Clemons G, Pootrakul P, Fillet G (1993a) Quantitative assessment of erythropoiesis and functional classiﬁcation of anemia based on measurements of serum transferrin receptor and erythropoietin. Blood 81: 1067–1076

Prediction of response to rhEPO

571

24. Beguin Y, Clemons GK, Oris R, Fillet G (1991) Circulating erythropoietin levels after bone marrow transplantation: Inappropriate response to anemia in allogeneic transplants. Blood 77: 868–873 25. Beguin Y, Huebers H, Pootrakul P, Haley R, Eschbach JW, Adamson JW, Finch CA (1987) Plasma transferrin receptor levels as a monitor of erythropoiesis in man: correlation with ferrokinetics and stimulation by recombinant human erythropoietin (rHuEpo). Blood 70: 51a 26. Beguin Y, Loo M, R’Zik S, Sautois B, Lejeune F, Rorive G, Fillet G (1993b) Early prediction of response to recombinant human erythropoietin in patients with the anemia of renal failure by serum transferrin receptor and ﬁbrinogen. Blood 82: 2010–2016 27. Beguin Y, Loo M, R’Zik S, Sautois B, Lejeune F, Rorive G, Fillet G (1994) Effect of recombinant human erythropoietin on platelets in patients with anemia of renal failure: correlation of platelet count with erythropoietic activity and iron parameters. Eur J Haematol 53: 265–270 28. Beguin Y, Loo M, R’Zik S, Sautois B, Lejeune F, Rorive G, Fillet G (1995) Quantitative assessment of erythropoiesis in haemodialysis patients demonstrates gradual expansion of erythroblasts during constant treatment with recombinant human erythropoietin. Br J Haematol 89: 17–23 29. Besarab A, Amin N, Ahsan M, Vogel SE, Zazuwa G, Frinak S, Zazra JJ, Anandan JV, Gupta A (2000) Optimization of epoetin therapy with intravenous iron therapy in hemodialysis patients. J Am Soc Nephrol 11: 530– 538 30. Beshara S, Letocha H, Linde T, Wikstrom B, Sandhagen B, Nilsson S, Danielson BG (1997) Anemia associated with advanced prostatic adenocarcinoma: effects of recombinant human erythropoietin. Prostate 31: 153–160 31. Biesma DH, van de Wiel A, Beguin Y, Kraaijenhagen RJ, Marx JJ (1995) Postoperative erythropoiesis is limited by the inﬂammatory effect of surgery on iron metabolism. Eur J Clin Invest 25: 383–389 32. Bohlius J, Langensiepen S, Schwarzer G, Seidenfeld J, Piper M, Bennett C, Engert A (2005) Recombinant human erythropoietin and overall survival in cancer patients: results of a comprehensive meta-analysis. J Natl Cancer Inst 97: 489–498 33. Bokemeyer C, Aapro MS, Courdi A, Foubert J, Link H, Osterborg A, Repetto L, Soubeyran P (2004) EORTC guidelines for the use of erythropoietic proteins in anaemic patients with cancer. Eur J Cancer 40: 2201–2216 34. Bovy C, Gothot A, Krzesinski JM, Beguin Y (2005) Mature erythrocyte indices: new markers of iron availability. Haematologica 90: 549–551 35. Bovy C, Krzesinski JM, Gothot A, Beguin Y (2004) Impact of erythropoietic activity on red cell parameters in chronic renal failure patients. Haematologica 89: 748–749 36. Bovy C, Tsobo C, Crapanzano L, Rorive G, Beguin Y, Albert A, Paulus JM (1999) Factors determining the percentage of hypochromic red blood cells in hemodialysis patients. Kidney Int 56: 1113–1119 37. Breymann C, Bauer C, Major A, Zimmermann R, Gautschi K, Huch A, Huch R (1996) Optimal timing of repeated rh-erythropoietin administration improves its effectiveness in stimulating erythropoiesis in healthy volunteers. Br J Haematol 92: 295–301

572

Y. Beguin and G. Van Straelen

38. Brugnara C (1998) Use of reticulocyte cellular indices in the diagnosis and treatment of hematological disorders. Int J Clin Lab Res 28: 1–11 39. Brugnara C, Chambers LA, Malynn E, Goldberg MA, Kruskall MS (1993) Red blood cell regeneration induced by subcutaneous recombinant erythropoietin: iron-deﬁcient erythropoiesis in iron-replete subjects. Blood 81: 956–964 40. Brugnara C, Colella GM, Cremins J, Langley RC, Jr., Schneider TJ, Rutherford CJ, Goldberg MA (1994) Effects of subcutaneous recombinant human erythropoietin in normal subjects: development of decreased reticulocyte hemoglobin content and iron-deﬁcient erythropoiesis. J Lab Clin Med 123: 660–667 41. Bucalossi A, Marotta G, Bigazzi C, Galieni P, Vessihelli F, Falbo R, Dispensa E (1996) Use of a functional classiﬁcation of anemia in myelodysplastic syndromes to identify subgroups of patients responsive to recombinant human-erythropoietin therapy. Eur J Haematol 56: 106–108 42. Canadian Erythropoietin Study Group (1991) Effect of recombinant human erythropoietin therapy on blood pressure in hemodialysis patients. Am J Nephrol 11: 23–26 43. Cascinu S, Fedeli A, Del Ferro E, Luzi Fedeli S, Catalano G (1994) Recombinant human erythropoietin treatment in cisplatin-associated anemia: a randomized, double-blind trial with placebo. J Clin Oncol 12: 1058–1062 44. Cavill I (2002) Erythropoiesis and iron. Best Pract Res Clin Haematol 15: 399– 409 45. Cazzola M, Beguin Y, Kloczko J, Spicka I, Coifﬁer B (2003) Once-weekly epoetin beta is highly effective in treating anaemic patients with lymphoproliferative malignancy and defective endogenous erythropoietin production. Br J Haematol 122: 386–393 46. Cazzola M, Messinger D, Battistel V, Bron D, Cimino R, Enller-Ziegler L, Essers U, Greil R, Grossi A, Jager G, LeMevel A, Najman A, Silingardi V, Spriano M, van Hoof A, Ehmer B (1995) Recombinant human erythropoietin in the anemia associated with multiple myeloma or non-Hodgkin’s lymphoma: dose ﬁnding and identiﬁcation of predictors of response. Blood 86: 4446–4453 47. Cazzola M, Ponchio L, Beguin Y, Rosti V, Bergamaschi G, Liberato NL, Fregoni V, Nalli G, Barosi G, Ascari E (1992) Subcutaneous erythropoietin for treatment of refractory anemia in hematologic disorders. Results of a phase I/II clinical trial. Blood 79: 29–37 48. Cazzola M, Ponchio L, Pedrotti C, Farina G, Cerani P, Lucotti C, Novella A, Rovati A, Bergamaschi G, Beguin Y (1996) Prediction of response to recombinant human erythropoietin (rHuEpo) in anemia of malignancy. Haematologica 81: 434–441 49. Charuruks N, Voravud N, Limpanasithikul W (2001) Ratio of baseline erythropoietin (EPO) level and corrected reticulocyte count as an indicator for a favourable response to recombinant human erythropoietin (rhEPO) therapy in anaemic cancer patients. J Clin Lab Anal 15: 260–266 50. Cooper A, Mikhail A, Lethbridge MW, Kemeny DM, Macdougall IC (2004) Pentoxifylline improves hemoglobin levels in patients with erythropoietin-resistant anemia in renal failure. J Am Soc Nephrol 15: 1877–1882 51. Cooper AC, Breen CP, Vyas B, Ochola J, Kemeny DM, Macdougall IC (2003a) Poor response to recombinant erythropoietin is associated with loss of Tlymphocyte CD28 expression and altered interleukin-10 production. Nephrol Dial Transplant 18: 133–140

Prediction of response to rhEPO

573

52. Cooper AC, Mikhail A, Lethbridge MW, Kemeny DM, Macdougall IC (2003b) Increased expression of erythropoiesis inhibiting cytokines (IFN-gamma, TNFalpha, IL-10, and IL-13) by T cells in patients exhibiting a poor response to erythropoietin therapy. J Am Soc Nephrol 14: 1776–1784 53. Couture F, Turner AR, Melosky B, Xiu L, Plante RK, Lau CY, Quirt I (2005) Prior red blood cell transfusions in cancer patients increase the risk of subsequent transfusions with or without recombinant human erythropoietin management. Oncologist 10: 63–71 54. Danielson B, Beguin Y, Bommer J, Cannata B, De Broe ME, Drueke TB, Ittel TH, Jolliffe LK, Krantz SB, Nielsen OJ, Pippard MJ, Rosenlof K, Stockenhuber B, Winearls C (1995) R-HuEPO hyporesponsiveness – who and why? Nephrol Dial Transplant 10 [Suppl 2]: 69–73 55. Danielson BG (2004) Structure, chemistry, and pharmacokinetics of intravenous iron agents. J Am Soc Nephrol 15 [Suppl 2]: S93–S98 56. Deicher R, Horl WH (2004) Differentiating factors between erythropoiesisstimulating agents: a guide to selection for anaemia of chronic kidney disease. Drugs 64: 499–509 57. Del Vecchio L, Pozzoni P, Andrulli S, Locatelli F (2005) Inﬂammation and resistance to treatment with recombinant human erythropoietin. J Ren Nutr 15: 137–141 58. Demetri GD, Kris M, Wade J, Degos L, Cella D (1998) Quality-of-life beneﬁt in chemotherapy patients treated with epoetin alfa is independent of disease response or tumor type: results from a prospective community oncology study. Procrit Study Group. J Clin Oncol 16: 3412–3425 59. Depaoli L, Levis A, Isabella N, Ficara F, Priotto C, Lista P, Foa R (1993) Serum erythropoietin level and marrow erythroid inﬁltration predict response to recombinant human erythropoietin in myelodysplastic syndromes. Haematologica 78: 118–122 60. Dmoszynska A, Kloczko J, Rokicka M, Hellmann A, Spicka I (2004) Efﬁcacy and pharmacokinetics of CERA (continuous erythropoietin receptor activator) in patients with multiple myeloma: An exploratory phase I–II dose escalation study. Proc Am Soc Clin Oncol 23: 568 (abstract 6552) 61. Dougherty FC, Reigner B, Jordan P, Pannier A (2004) CERA (continuous erythropoietin receptor activator): Dose-response, pharmacokinetics and tolerability in phase I multiple ascending dose studies. Proc Am Soc Clin Oncol 23: 603 (abstract 6692) 62. Drueke TB, Barany P, Cazzola M, Eschbach JW, Grutzmacher P, Kaltwasser JP, Macdougall IC, Pippard MJ, Shaldon S, van Wyck D (1997) Management of iron deﬁciency in renal anemia: guidelines for the optimal therapeutic approach in erythropoietin-treated patients. Clin Nephrol 48: 1–8 63. Eschbach JW (2005) Iron requirements in erythropoietin therapy. Best Pract Res Clin Haematol 18: 347–361 64. Eschbach JW, Haley NR, Egrie JC, Adamson JW (1992) A comparison of the responses to recombinant human erythropoietin in normal and uremic subjects. Kidney Int 42: 407–416 65. Fernandez-Rodriguez AM, Guindeo-Casasus MC, Molero-Labarta T, Dominguez-Cabrera C, Hortal C, Perez-Borges P, Vega-Diaz N, SaavedraSantana P, Palop-Cubillo L (1999) Diagnosis of iron deﬁciency in chronic renal failure. Am J Kidney Dis 34: 508–513

574

Y. Beguin and G. Van Straelen

66. Fillet G, Beguin Y, Baldelli L (1989) Model of reticuloendothelial iron metabolism in humans: abnormal behavior in idiopathic hemochromatosis and in inﬂammation. Blood 74: 844–851 67. Fishbane S (2003) Safety in iron management. Am J Kidney Dis 41: 18–26 68. Fishbane S, Wagner J (2001) Sodium ferric gluconate complex in the treatment of iron deﬁciency for patients on dialysis. Am J Kidney Dis 37: 879–883 69. Fjornes T, Wiedemann GJ, Sack K, Jelkmann W (1998) Serum erythropoietin and creatinine concentrations as predictive factors for response to recombinant human erythropoietin treatment in anaemic tumour patients on chemotherapy. Oncol Rep 5: 81–86 70. Gabrilove JL, Cleeland CS, Livingston RB, Sarokhan B, Winer E, Einhorn LH (2001) Clinical evaluation of once-weekly dosing of epoetin alfa in chemotherapy patients: improvements in hemoglobin and quality of life are similar to three-times-weekly dosing. J Clin Oncol 19: 2875–2882 71. Ghio R, Balleari E, Ballestrero A, Gatti AM, Mareni C, Massa G, Patrone F, Sessarego M, Timitilli S (1993) Subcutaneous recombinant human erythropoietin for the treatment of anemia in myelodysplastic syndromes. Acta Haematol 90: 58–64 72. Glaspy J, Bukowski R, Steinberg D, Taylor C, Tchekmedyian S, Vadhan-Raj S (1997) Impact of therapy with epoetin alfa on clinical outcomes in patients with nonmyeloid malignancies during cancer chemotherapy in community oncology practice, for the Procrit Study Group. J Clin Oncol 15: 1218–1234 73. Glaspy J, Cavill I (1999) Role of iron in optimizing responses of anemic cancer patients to erythropoietin. Oncology (Huntingt) 13: 461–473 74. Glaspy J, Degos L, Dicato M, Demetri GD (2002) Comparable efﬁcacy of epoetin alfa for anemic cancer patients receiving platinum- and nonplatinumbased chemotherapy: a retrospective subanalysis of two large, community-based trials. Oncologist 7: 126–135 75. Glaspy JA, Jadeja JS, Justice G, Fleishman A, Rossi G, Colowick AB (2003) A randomized, active-control, pilot trial of front-loaded dosing regimens of darbepoetin-alfa for the treatment of patients with anemia during chemotherapy for malignant disease. Cancer 97: 1312–1320 76. Glimelius B, Linne T, Hoffman K, Larsson L, Svensson JH, Nasman P, Svensson B, Helmers C (1998) Epoetin beta in the treatment of anemia in patients with advanced gastrointestinal cancer. J Clin Oncol 16: 434–440 77. Gonzalez-Baron M, Ordonez A, Franquesa R, Constenla M, Montalar J, Gili F, Camps C, Sancho JF, Perez-Cachot P (2002) Response predicting factors to recombinant human erythropoietin in cancer patients undergoing platinumbased chemotherapy. Cancer 95: 2408–2413 78. Goodnough LT, Marcus RE (1998) Erythropoiesis in patients stimulated with erythropoietin: the relevance of storage iron. Vox Sang 75: 128–133 79. Goodnough LT, Skikne B, Brugnara C (2000) Erythropoietin, iron, and erythropoiesis. Blood 96: 823–833 80. Groopman JE, Itri LM (1999) Chemotherapy-induced anemia in adults: incidence and treatment. J Natl Cancer Inst 91: 1616–1634 81. Gunnell J, Yeun JY, Depner TA, Kaysen GA (1999) Acute-phase response predicts erythropoietin resistance in hemodialysis and peritoneal dialysis patients. Am J Kidney Dis 33: 63–72

Prediction of response to rhEPO

575

82. Gupta M, Kannan M, Gupta S, Saxena R (2003) Contribution of iron deﬁciency to anemia in chronic renal failure. Indian J Pathol Microbiol 46: 563–564 83. Hagberg A, Barbany G, Landegren U, Birgegard G (2003) Beta-globin mRNA increases rapidly during erythropoietin treatment. Scand J Clin Lab Invest 63: 239–245 84. Hedenus M, Adriansson M, San Miguel J, Kramer MH, Schipperus MR, Juvonen E, Taylor K, Belch A, Altes A, Martinelli G, Watson D, Matcham J, Rossi G, Littlewood TJ (2003) Efﬁcacy and safety of darbepoetin alfa in anaemic patients with lymphoproliferative malignancies: a randomized, double-blind, placebo-controlled study. Br J Haematol 122: 394–403 85. Hedenus M, Hansen S, Taylor K, Arthur C, Emmerich B, Dewey C, Watson D, Rossi G, Osterborg A (2002) Randomized, dose-ﬁnding study of darbepoetin alfa in anaemic patients with lymphoproliferative malignancies. Br J Haematol 119: 79–86 86. Hellstrom-Lindberg E (1995) Efﬁcacy of erythropoietin in the myelodysplastic syndromes: a meta-analysis of 205 patients from 17 studies. Br J Haematol 89: 67–71 87. Hellstrom-Lindberg E (2003) Approach to anemia associated with myelodysplastic syndromes. Curr Hematol Rep 2: 122–129 88. Hellstrom-Lindberg E, Ahlgren T, Beguin Y, Carlsson M, Carneskog J, Dahl IM, Dybedal I, Grimfors G, Kanter-Lewensohn L, Linder O, Luthman M, Lofvenberg E, Nilsson-Ehle H, Samuelsson J, Tangen JM, Winqvist I, Oberg G, Osterborg A, Ost A (1998) Treatment of anemia in myelodysplastic syndromes with granulocyte colony-stimulating factor plus erythropoietin: results from a randomized phase II study and long-term follow-up of 71 patients. Blood 92: 68–75 89. Hellstrom-Lindberg E, Gulbrandsen N, Lindberg G, Ahlgren T, Dahl IM, Dybedal I, Grimfors G, Hesse-Sundin E, Hjorth M, Kanter-Lewensohn L, Linder O, Luthman M, Lofvenberg E, Oberg G, Porwit-MacDonald A, Radlund A, Samuelsson J, Tangen JM, Winquist I, Wisloff F (2003) A validated decision model for treating the anaemia of myelodysplastic syndromes with erythropoietin + granulocyte colony-stimulating factor: signiﬁcant effects on quality of life. Br J Haematol 120: 1037–1046 90. Hellstrom-Lindberg E, Negrin R, Stein R, Krantz S, Lindberg G, Vardiman J, Ost A, Greenberg P (1997) Erythroid response to treatment with G-CSF plus erythropoietin for the anaemia of patients with myelodysplastic syndromes: proposal for a predictive model. Br J Haematol 99: 344–351 91. Henry D, Abels R, Larholt K (1995) Prediction of response to recombinant human erythropoietin (r-huepo/epoetin-alpha) therapy in cancer patients [letter]. Blood 85: 1676–1678 92. Henry DH, Abels RI (1994) Recombinant human erythropoietin in the treatment of cancer and chemotherapy-induced anemia: Results of double-blind and open-label follow-up studies. Semin Hematol 21 [Suppl 3]: 21–28 93. Hesketh PJ, Arena F, Patel D, Austin M, D’Avirro P, Rossi G, Colowick A, Schwartzberg L, Bertoli LF, Cole JT, Demetri G, Dessypris E, Dobbs T, Eisenberg P, Fleischman R, Hall J, Hoffman PC, Laber DA, Leonard J, Lester EP, McCachren S, McMeekin S, Meza L, Miller DS, Nand S, Oliff I, Paroly W, Pawl L, Perez A, Raftopoulos H, Rigas J, Rowland K, Scullin DC, Jr., Tezcan H, Waples J, Ward J, Yee LK (2004) A randomized controlled trial of darbepoetin

576

94.

95.

96.

97.

98.

99. 100.

101.

102. 103.

104.

105.

106.

Y. Beguin and G. Van Straelen alfa administered as a ﬁxed or weight-based dose using a front-loading schedule in patients with anemia who have nonmyeloid malignancies. Cancer 100: 859–868 Hou CC, Wu SC, Wu SC, Yang WC, Huang TP, Ng YY (1998) Serum transferrin receptor concentration is not indicative of erythropoietic activity in chronic hemodialysis patients with poor response to recombinant human erythropoietin. Chung Hua I Hsueh Tsa Chih 61: 456–462 Hsu PY, Lin CL, Yu CC, Chien CC, Hsiau TG, Sun TH, Huang LM, Yang CW (2004) Ultrapure dialysate improves iron utilization and erythropoietin response in chronic hemodialysis patients – a prospective cross-over study. J Nephrol 17: 693–700 Huebers HA, Beguin Y, Pootrakul P, Einspahr D, Finch CA (1990) Intact transferrin receptors in human plasma and their relation to erythropoiesis. Blood 75: 102–107 ICSG (1998) A randomized double-blind placebo-controlled study with subcutaneous recombinant human erythropoietin in patients with low-risk myelodysplastic syndromes. Italian Cooperative Study Group for rHuEpo in Myelodysplastic Syndromes. Br J Haematol 103: 1070–1074 Johnson DW, Herzig KA, Gissane R, Campbell SB, Hawley CM, Isbel NM (2001) A prospective crossover trial comparing intermittent intravenous and continuous oral iron supplements in peritoneal dialysis patients. Nephrol Dial Transplant 16: 1879–1884 Kaltwasser JP, Gottschalk R (1999) Erythropoietin and iron. Kidney Int [Suppl] 69: S49–S56 Kato A, Odamaki M, Takita T, Furuhashi M, Maruyama Y, Hishida A (2001) High blood soluble receptor p80 for tumour necrosis factor-alpha is associated with erythropoietin resistance in haemodialysis patients. Nephrol Dial Transplant 16: 1838–1844 Kausz AT, Solid C, Pereira BJ, Collins AJ, St Peter W (2005) Intractable anemia among hemodialysis patients: a sign of suboptimal management or a marker of disease? Am J Kidney Dis 45: 136–147 Knight K, Wade S, Balducci L (2004) Prevalence and outcomes of anemia in cancer: a systematic review of the literature. Am J Med 116 [Suppl 7A]: 11S–26S Kotasek D, Steger G, Faught W, Underhill C, Poulsen E, Colowick AB, Rossi G, Mackey J (2003) Darbepoetin alfa administered every 3 weeks alleviates anaemia in patients with solid tumours receiving chemotherapy; results of a double-blind, placebo-controlled, randomised study. Eur J Cancer 39: 2026–2034 Lastiri JM, Specterman SR, Rendo P, Pallotta MG, Varela MS, Goldstein S (2002) Predictive response variables to recombinant human erythropoietin treatment in patients with anemia and cancer. Medicina (B Aires) 62: 41–47 Leon MP, Jimenez MM, Barona ZP, Riol DM, Castro PL, Sierrasesumaga AL (1998) Recombinant human erythropoietin in anemia associated with pediatric cancer: study of the identiﬁcation of predictors of response. An Esp Pediatr 49: 17–22 Link G, Tirosh R, Pinson A, Hershko C (1996) Role of iron in the potentiation of anthracycline cardiotoxicity: identiﬁcation of heart cell mitochondria as a major site of iron-anthracycline interaction. J Lab Clin Med 127: 272–278

Prediction of response to rhEPO

577

107. Link H, Boogaerts MA, Fauser AA, Slavin S, Reiffers J, Gorin NC, Carella AM, Mandelli F, Burdach S, Ferrant A, Linkesch W, Tura S, Bacigalupo A, Schindel F, Heinrichs H (1994) A controlled trial of recombinant human erythropoietin after bone marrow transplantation. Blood 84: 3327–3325 108. Littlewood TJ, Zagari M, Pallister C, Perkins A (2003) Baseline and early treatment factors are not clinically useful for predicting individual response to erythropoietin in anemic cancer patients. Oncologist 8: 99–107 109. Locatelli F, Aljama P, Barany P, Canaud B, Carrera F, Eckardt KU, Horl WH, Macdougal IC, MacLeod A, Wiecek A, Cameron S (2004) Revised European best practice guidelines for the management of anaemia in patients with chronic renal failure. Nephrol Dial Transplant 19 [Suppl 2]: ii1–47 110. Lopez-Gomez JM, Perez-Flores I, Jofre R, Carretero D, Rodriguez-Benitez P, Villaverde M, Perez-Garcia R, Nassar GM, Niembro E, Ayus JC (2004) Presence of a failed kidney transplant in patients who are on hemodialysis is associated with chronic inﬂammatory state and erythropoietin resistance. J Am Soc Nephrol 15: 2494–2501 111. Lorenzo JD, Rodriguez MM, Martin SS, Romo JM (2001) Assessment of erythropoiesis activity during hemodialysis therapy by soluble transferrin receptor levels and ferrokinetic measurements. Am J Kidney Dis 37: 550–556 112. Ludwig H, Fritz E, Leitgeb C, Krainer M, Kuhrer I, Sagaster P, Umek H (1993a) Erythropoietin treatment for chronic anemia of selected hematological malignancies and solid tumors. Ann Oncol 4: 161–167 113. Ludwig H, Fritz E, Leitgeb C, Pecherstorfer M, Samonigg L, Schuster J (1994) Prediction of response to erythropoietin treatment in chronic anemia of cancer. Blood 84: 1056–1063 114. Ludwig H, Pecherstorfer M, Leitgeb C, Fritz E (1993b) Recombinant human erythropoietin for the treatment of chronic anemia in multiple myeloma and squamous cell carcinoma. Stem Cells 11: 348–355 115. Ludwig H, Van Belle S, Barrett-Lee P, Birgegard G, Bokemeyer C, Gascon P, Kosmidis P, Krzakowski M, Nortier J, Olmi P, Schneider M, Schrijvers D (2004) The European Cancer Anaemia Survey (ECAS): a large, multinational, prospective survey deﬁning the prevalence, incidence, and treatment of anaemia in cancer patients. Eur J Cancer 40: 2293–2306 116. Macdougall IC (1999) Strategies for iron supplementation: oral versus intravenous. Kidney Int [Suppl 69]: S61–S66 117. Macdougall IC (2001) Hyporesponsiveness to anemia therapy–what are we doing wrong? Perit Dial Int 21 [Suppl 3]: S225–S230 118. Macdougall IC (2004) Could anti-inﬂammatory cytokine therapy improve poor treatment outcomes in dialysis patients? Nephrol Dial Transplant 19 [Suppl 5]: V73–V78 119. Macdougall IC, Cooper AC (2002) Erythropoietin resistance: the role of inﬂammation and pro-inﬂammatory cytokines. Nephrol Dial Transplant 17 [Suppl 11]: 39–43 120. Macdougall IC, Roberts DE, Neubert P, Dharmasena AD, Coles GA, Williams JD (1989) Pharmacokinetics of recombinant human erythropoietin in patients on continuous ambulatory peritoneal dialysis. Lancet 1: 425–427

578

Y. Beguin and G. Van Straelen

121. Macdougall IC (2005) CERA (continuous erythropoietin receptor activator): a new erythropoiesis-stimulating agent for the treatment of anemia. Curr Hematol Rep 4: 436–440 122. Macdougall IC, Tucker B, Thompson J, Tomson CR, Baker LR, Raine AE (1996) A randomized controlled study of iron supplementation in patients treated with erythropoietin. Kidney Int 50: 1694–1699 123. Mantovani L, Lentini G, Hentschel B, Wickramanayake PD, Loefﬂer M, Diehl V, Tesch H (2000) Treatment of anaemia in myelodysplastic syndromes with prolonged administration of recombinant human granulocyte colony-stimulating factor and erythropoietin. Br J Haematol 109: 367–375 124. Martini A, Ravelli A, Di Fuccia G, Rosti V, Cazzola M, Barosi G (1994) Intravenous iron therapy for severe anaemia in systemic-onset juvenile chronic arthritis. Lancet 344: 1052–1054 125. Mast AE, Blinder MA, Lu Q, Flax S, Dietzen DJ (2002) Clinical utility of the reticulocyte hemoglobin content in the diagnosis of iron deﬁciency. Blood 99: 1489–1491 126. Matsuda A, Bessho M, Mori S, Takeuchi T, Abe T, Yawata Y, Mori H, Omine M, Nakamura Y, Furusawa S, Maeda T, Haginosita S, Hirasawa Y, Kinugasa E, Akizawa T, Kawakami N, Nagata A, Hirashima K (2002) Diagnostic signiﬁcance of serum soluble transferrin receptors in various anemic diseases: the ﬁrst multi-institutional joint study in Japan. Haematologia (Budap ) 32: 225– 238 127. Matsumoto T, Endoh K, Kamisango K, Akamatsu K, Koizumi K, Higuchi M, Imai N, Mitsui H, Kawaguchi T (1990) Effect of recombinant human erythropoietin on anticancer drug-induced anaemia. Br J Haematol 75: 463–468 128. Mizuguchi T, Minakuchi J, Kawahara K, Kawashima S, Hikawa A, Nomata Y (1997) Serum transferrin receptor level as an index of the response to erythropoietin therapy for anemia in pre-dialysis patients with chronic renal failure. Nippon Jinzo Gakkai Shi 39: 497–502 129. Musto P, Falcone A, D’Arena G, Scalzulli PR, Matera R, Minervini MM, Lombardi GF, Modoni S, Longo A, Carotenuto M (1997) Clinical results of recombinant erythropoietin in transfusion-dependent patients with refractory multiple myeloma: role of cytokines and monitoring of erythropoiesis. Eur J Haematol 58: 314–319 130. Musto P, Matera R, Minervini MM, Checchia-de Ambrosio C, Bodenizza C, Falcone A, Carotenuto M (1994a) Low serum levels of tumor necrosis factor and interleukin-1 beta in myelodysplastic syndromes responsive to recombinant erythropoietin. Haematologica 79: 265–268 131. Musto P, Modoni S, Alicino G, Savino A, Longo A, Bodenizza C, Falcone A, D’Arena G, Scalzulli P, Perla G, et al (1994b) Modiﬁcations of erythropoiesis in myelodysplastic syndromes treated with recombinant erythropoietin as evaluated by soluble transferrin receptor, high ﬂuorescence reticulocytes and hypochromic erythrocytes. Haematologica 79: 493–499 132. Musto P, Scalzulli PR, Carotenuto M (1995) Recombinant erythropoietin for myelodysplastic syndromes [letter]. Br J Haematol 91: 256–258 133. Nakanishi T, Moriguchi R, Itahana R, Otaki Y, Inoue T, Takamitsu Y (2001) Acute effect of human recombinant erythropoietin administration on soluble transferrin receptor. Nephron 89: 119–120

Prediction of response to rhEPO

579

134. Negrin RS, Stein R, Doherty K, Cornwell J, Vardiman J, Krantz S, Greenberg PL (1996) Maintenance treatment of the anemia of myelodysplastic syndromes with recombinant human granulocyte colony-stimulating factor and erythropoietin: evidence for in vivo synergy. Blood 87: 4076–4081 135. Negrin RS, Stein R, Vardiman J, Doherty K, Cornwell J, Krantz S (1993) Treatment of the anemia of myelodysplastic syndromes using recombinant human granulocyte colony-stimulating factor in combination with erythropoietin. Blood 82: 737–743 136. NKF-K/DOQI (2001b) IV. Clinical Practice Guidelines for Anemia of Chronic Kidney Disease: update 2000. Am J Kidney Dis 37: S182–S238 137. Oberhoff C, Neri B, Amadori D, Petry KU, Gamucci T, Rebmann U, Nowrousian MR, Voigtmann R, Monfardini S, Armand JP, Herrmann R, Netter-Pinon J, Tubiana-Mathieu N, Zwierzina H (1998) Recombinant human erythropoietin in the treatment of chemotherapy-induced anemia and prevention of transfusion requirement associated with solid tumors: a randomized, controlled study. Ann Oncol 9: 255–260 138. Olijhoek G, Megens JG, Musto P, Nogarin L, Gassmann-Mayer C, Vercammen E, Hayes-Licitra SA (2001) Role of oral versus IV iron supplementation in the erythropoietic response to rHuEPO: a randomized, placebo-controlled trial. Transfusion 41: 957–963 139. Olsson AM, Svensson JH, Sundstrom J, Bergstrom S, Edekling T, Carlsson G, Hansen J, Svensson B, Albertsson M (2002) Erythropoietin treatment in metastatic breast cancer–effects on Hb, quality of life and need for transfusion. Acta Oncol 41: 517–524 140. Oster W, Herrmann F, Gamm H, Zeile G, Lindemann A, Muller G, Brune T, Kraemer HP, Mertelsmann R (1990) Erythropoietin for the treatment of anemia of malignancy associated with neoplastic bone marrow inﬁltration. J Clin Oncol 8: 956–962 141. Osterborg A, Boogaerts MA, Cimino R, Essers U, Holowiecki J, Juliusson G, Jager G, Najman A, Peest D (1996) Recombinant human erythropoietin in transfusion-dependent anemic patients with multiple myeloma and nonHodgkin’s lymphoma. A randomized multicenter study. Blood 87: 2675–2682 142. Osterborg A, Brandberg Y, Molostova V, Iosava G, Abdulkadyrov K, Hedenus M, Messinger D (2002) Randomized, double-blind, placebo-controlled trial of recombinant human erythropoietin, epoetin Beta, in hematologic malignancies. J Clin Oncol 20: 2486–2494 143. Osterborg A, Hellmann A, Couban S (2004) CERA (continuous erythropoietin receptor activator): Dose-response trial of subcutaneous (sc) administration once every three weeks (q3w) to patients with non-Hodgkin’s lymphoma and anemia receiving chemotherapy. Blood 104: 142b (abstract 4225) 144. Paganini EP, Eschbach JW, Lazarus JM, Van Stone JC, Gimenez LF, Graber SE, Egrie JC, Okamoto DM, Goodkin DA (1995) Intravenous versus subcutaneous dosing of epoetin alfa in hemodialysis patients. Am J Kidney Dis 26: 331–340 145. Pawlicki M, Jassem J, Bosze P, Lotan C, Kurteva GP, Siddiqui M, Kosmidis PA, Rigatos GA, Kansu E, Durkovic P, Aziz Z, Al Idrissi H, Roth A, Cozma G (1997) A multicenter study of recombinant human erythropoietin (epoetin alpha) in the management of anemia in cancer patients receiving chemotherapy. Anticancer Drugs 8: 949–957

580

Y. Beguin and G. Van Straelen

146. Platanias LC, Miller CB, Mick R, Hart RD, Ozer H, McEvilly JM, Jones RJ, Ratain MJ (1991) Treatment of chemotherapy-induced anemia with recombinant human erythropoietin in cancer patients. J Clin Oncol 9: 2021–2026 147. Ponchio L, Beguin Y, Farina G, Pedrazzoli P, Pedrotti C, Poggi G, Rosti V, Bergamaschi G, Battistel V, Cazzola M (1992) Evaluation of erythroid marrow response to recombinant human erythropoietin in patients with cancer anaemia. Haematologica 77: 494–501 148. Ponka P, Beaumont C, Richardson DR (1998) Function and regulation of transferrin and ferritin. Semin Hematol 35: 35–54 149. Provan D (1999) Mechanisms and management of iron deﬁciency anaemia. Br J Haematol 105 [Suppl 1]: 19–26 150. Quirt I, Robeson C, Lau CY, Kovacs M, Burdette-Radoux S, Dolan S, Tang SC, McKenzie M, Couture F (2001) Epoetin alfa therapy increases hemoglobin levels and improves quality of life in patients with cancer-related anemia who are not receiving chemotherapy and patients with anemia who are receiving chemotherapy. J Clin Oncol 19: 4126–4134 151. Remacha AF, Arrizabalaga B, Villegas A, Manteiga R, Calvo T, Julia A, Fernandez F, I, Gonzalez FA, Font L, Junca J, del Arco A, Malcorra JJ, Equiza EP, de Mendiguren BP, Romero M (1999) Erythropoietin plus granulocyte colonystimulating factor in the treatment of myelodysplastic syndromes. Identiﬁcation of a subgroup of responders. The Spanish Erythropathology Group. Haematologica 84: 1058–1064 152. Rizzo JD, Lichtin AE, Woolf SH, Seidenfeld J, Bennett CL, Cella D, Djulbegovic B, Goode MJ, Jakubowski AA, Lee SJ, Miller CB, Rarick MU, Regan DH, Browman GP, Gordon MS (2002) Use of epoetin in patients with cancer: evidence-based clinical practice guidelines of the American Society of Clinical Oncology and the American Society of Hematology. Blood 100: 2303–2320 153. Rohling RG, Zimmermann AP, Breymann C (2000) Intravenous versus oral iron supplementation for preoperative stimulation of hemoglobin synthesis using recombinant human erythropoietin. J Hematother Stem Cell Res 9: 497–500 154. Rose EH, Abels RI, Nelson RA, McCullough DM, Lessin L (1995) The use of r-huepo in the treatment of anaemia related to myelodysplasia (mds). Br J Haematol 89: 831–837 155. Schwartzberg LS, Yee LK, Senecal FM, Charu V, Tomita D, Wallace J, Rossi G (2004) A randomized comparison of every-2-week darbepoetin alfa and weekly epoetin alfa for the treatment of chemotherapy-induced anemia in patients with breast, lung, or gynecologic cancer. Oncologist 9: 696–707 156. Sezer S, Kulah E, Ozdemir FN, Tutal E, Arat Z, Haberal M (2004) Clinical consequences of intermittent elevation of C-reactive protein levels in hemodialysis patients. Transplant Proc 36: 38–40 157. Skikne BS, Flowers CH, Cook JD (1990) Serum transferrin receptor: a quantitative measure of tissue iron deﬁciency. Blood 75: 1870–1876 158. Smith RE, Jr., Tchekmedyian NS, Chan D, Meza LA, Northfelt DW, Patel R, Austin M, Colowick AB, Rossi G, Glaspy J (2003) A dose- and schedule-ﬁnding study of darbepoetin alpha for the treatment of chronic anaemia of cancer. Br J Cancer 88: 1851–1858 159. Spivak JL (2005) The anaemia of cancer: death by a thousand cuts. Nat Rev Cancer 5: 543–555

Prediction of response to rhEPO

581

160. Stasi R, Brunetti M, Bussa S, Conforti M, Di Giulio C, Crescenzi A, Terzoli E, Vecchione A, Pagano A (1997a) Response to recombinant human erythropoietin in patients with myelodysplastic syndromes. Clin Cancer Res 3: 733– 739 161. Stasi R, Brunetti M, Bussa S, Conforti M, Martin LS, La Presa M, Bianchi M, Parma A, Pagano A (1997b) Serum levels of tumour necrosis factor-alpha predict response to recombinant human erythropoietin in patients with myelodysplastic syndrome. Clin Lab Haematol 19: 197–201 162. Stasi R, Brunetti M, Terzoli E, Amadori S (2002) Sustained response to recombinant human erythropoietin and intermittent all-trans retinoic acid in patients with myelodysplastic syndromes. Blood 99: 1578–1584 163. Stasi R, Pagano A, Terzoli E, Amadori S (1999) Recombinant human granulocyte-macrophage colony-stimulating factor plus erythropoietin for the treatment of cytopenias in patients with myelodysplastic syndromes. Br J Haematol 105: 141–148 164. Stein RS, Abels RI, Krantz SB (1991) Pharmacologic doses of recombinant human erythropoietin in the treatment of myelodysplastic syndromes. Blood 78: 1658–1663 165. Stoves J, Inglis H, Newstead CG (2001) A randomized study of oral vs intravenous iron supplementation in patients with progressive renal insufﬁciency treated with erythropoietin. Nephrol Dial Transplant 16: 967–974 166. Syed RS, Reid SW, Li C, Cheetham JC, Aoki KH, Liu B, Zhan H, Osslund TD, Chirino AJ, Zhang J, Finer-Moore J, Elliott S, Sitney K, Katz BA, Matthews DJ, Wendoloski JJ, Egrie J, Stroud RM (1998) Efﬁciency of signalling through cytokine receptors depends critically on receptor orientation. Nature 395: 511–516 167. Takemasa A, Yorioka N, Ueda C, Amimoto D, Taniguchi Y, Yamakido M (2000) Stimulation of tumour necrosis factor-alpha production by recombinant human erythropoietin may contribute to failure of therapy. Scand J Urol Nephrol 34: 131–135 168. ten Bokkel Huinink WW, de Swart CA, van Toorn DW, Morack G, Breed WP, Hillen HF, van der Hoeven JJ, Reed NS, Fairlamb DJ, Chan SY, Godfrey KA, Kristensen GB, van Tinteren H, Ehmer B (1998) Controlled multicentre study of the inﬂuence of subcutaneous recombinant human erythropoietin on anaemia and transfusion dependency in patients with ovarian carcinoma treated with platinum-based chemotherapy. Med Oncol 15: 174–182 169. Terpos E, Mougiou A, Kouraklis A, Chatzivassili A, Michalis E, Giannakoulas N, Manioudaki E, Lazaridou A, Bakaloudi V, Protopappa M, Liapi D, Grouzi E, Parharidou A, Symeonidis A, Kokkini G, Laoutaris NP, Vaipoulos G, Anagnostopoulos NI, Christakis JI, Meletis J, Bourantas KL, Zoumbos NC, Yataganas X, Viniou NA (2002) Prolonged administration of erythropoietin increases erythroid response rate in myelodysplastic syndromes: a phase II trial in 281 patients. Br J Haematol 118: 174–180 170. Thompson JA, Gilliland DG, Prchal JT, Bennett JM, Larholt K, Nelson RA, Rose EH, Dugan MH (2000) Effect of recombinant human erythropoietin combined with granulocyte/ macrophage colony-stimulating factor in the treatment of patients with myelodysplastic syndrome. GM/EPO MDS Study Group. Blood 95: 1175–1179

582

Y. Beguin and G. Van Straelen: Prediction of response to rhEPO

171. Tsabouri SE, Georgiou I, Katsaraki A, Bourantas KL (2004) Treatment of MDS patients with recombinant human erythropoietin and the role of GSTs. J Exp Clin Cancer Res 23: 417–424 172. Van Wyck DB (2004) Labile iron: manifestations and clinical implications. J Am Soc Nephrol 15 [Suppl 2]: S107–S111 173. Vannucchi AM, Grossi A, Bosi A, Rafanelli D, Statello M, Guidi S, Rossi-Ferrini P (1993) Effects of cyclosporin A on erythropoietin production by the human Hep3B hepatoma cell line. Blood 82: 978–984 174. Vreugdenhil G, Manger B, Nieuwenhuizen C, Feelders RA, van Eijk HG, Swaak AJ (1992) Iron stores and serum transferrin receptor levels during recombinant human erythropoietin treatment of anemia in rheumatoid arthritis. Ann Hematol 65: 265–268 175. Wallvik J, Stenke L, Bernell P, Nordahl G, Hippe E, Hast R (2002) Serum erythropoietin (EPO) levels correlate with survival and independently predict response to EPO treatment in patients with myelodysplastic syndromes. Eur J Haematol 68: 180–185 176. Weinberg ED (1996) The role of iron in cancer. Eur J Cancer Prev 5: 19–36 177. Weiss G, Goodnough LT (2005) Anemia of chronic disease. N Engl J Med 352: 1011–1023 178. Wood PA, Hrushesky WJ (1995) Cisplatin-associated anemia: an erythropoietin deﬁciency syndrome. J Clin Invest 95: 1650–1659 179. Worwood M (1990) Ferritin. Blood Rev 4: 259–269 180. Yee J, Besarab A (2002) Iron sucrose: the oldest iron therapy becomes new. Am J Kidney Dis 40: 1111–1121 Correspondence: Yves Beguin, MD, University of Liège, Department of Hematology, CHU Sart-Tilman, 4000 Liège, Belgium, E-mail: [email protected]

Chapter 22

rhEPO in hematopoietic stem cell transplantation G. Van Straelen and Y. Beguin GV is Télévie Research Assistant and YB Research Director of the National Fund for Scientiﬁc Research (FNRS, Belgium) Department of Medicine, Division of Hematology, University of Liège, Liège, Belgium Center for Cellular and Molecular Therapy (Centre de Thérapie Cellulaire et Moléculaire, CTCM), University of Liège, Liège, Belgium

Introduction Erythropoietin (Epo) is the critical regulatory factor of erythropoiesis, and recombinant human erythropoietin (rhEPO) has become a well-established treatment for chronic renal failure patients. In these subjects, endogenous serum Epo levels are very low (Erslev 1991) and the administration of rhEPO, restoring adequate levels of the hormone, permits correction of the anemia (Eschbach et al. 1987). In patients with normal kidney function, serum Epo levels increase exponentially when an anemia develops (Erslev 1991). After high-dose chemotherapy, serum Epo levels ﬁrst rapidly increase to disproportionately high levels for 1–3 weeks, with peak values usually observed in the ﬁrst week after the conditioning regimen (Birgegard et al. 1989; Abedi et al. 1990; Schapira et al. 1990; Beguin et al. 1991; Bosi et al. 1991a; Bosi et al. 1991b; Grace et al. 1991; Lazarus et al. 1992; Miller et al. 1992a; Beguin et al. 1993; Davies et al. 1995; Beguin et al. 1998). However, after allogeneic hematopoietic stem cell transplantation (HSCT) the Epo response to anemia then generally becomes impaired, resulting in inappropriately low levels of Epo for the degree of anemia and prolonged anemia (Ireland et al. 1990; Schapira et al. 1990; Beguin et al. 1991; Bosi et al. 1991a; Miller et al. 1992a; Beguin et al. 1993). This is speciﬁc for allogeneic transplant because serum Epo levels remain adequate throughout the posttransplant course in recipients of an autologous marrow or peripheral blood stem cell transplant (Ireland et al. 1990; Schapira et al. 1990; Beguin et al. 1991; Bosi et al. 1991b; Lazarus et al. 1992; Davies et al. 1995; Beguin et al. This work was supported in part by grants from the National Fund for Scientiﬁc Research (FNRS, Belgium).

584

G. Van Straelen and Y. Beguin

1993; Beguin et al. 1998). These studies thus indicate that after auto-HSCT, the development of erythropoiesis is determined by the overall marrow proliferative activity and that erythropoietin plays only a facilitating role. On the other hand, after allo-HSCT, erythropoiesis depends on Epo production, which remains inadequate for prolonged periods of time. This pathophysiology of endogenous erythropoietin production after HSCT suggests that the two forms of transplantation may respond differently to rhEPO. Indeed, one could expect better responses to rhEPO therapy after allogeneic HSCT, which is characterized by prolonged, severe Epo deﬁciency differences, compared with autologous HSCT that is associated with adequate Epo levels. Indeed, the early clinical experience with rhEPO therapy after HSCT has shown such differences. However, more recent studies have demonstrated that it is possible to optimize the timing of rhEPO administration in HSCT, resulting in higher efﬁcacy. We review here the clinical experience with rhEPO therapy in the setting of HSCT. After brieﬂy describing the pathophysiology of Epo production after HSCT, we will successively address the use of rhEPO after allogeneic HSCT, its use after autologous HSCT, comparative studies after autologous and allogeneic transplantation, rhEPO therapy as preparation to HSCT, HSC mobilization with rhEPO and administration of rhEPO to HSCT donors. Finally, we will brieﬂy comment on the speciﬁc tolerance to rhEPO in HSCT.

Serum Epo levels after HSCT Successive phases of changes in serum Epo levels are observed after HSCT, i.e. universal peak immediately after the conditioning regimen, a progressive return to normal with erythroid recovery and a variable defect in endogenous Epo in the following months (Table 1). Elevated serum Epo levels are observed transiently after intensive conditioning regimens without concomitant changes in hemoglobin, and this is observed in the context of autologous or allogeneic transplantation of

Table 1. Factors potentially affecting endogenous erythropoietin production after HSCT Inhibiting factors

Stimulating factors

Allogeneic donor Acute GVHD Ciclosporin A Amphotericin B Cytomegalovirus infection

Conditioning regimen Liver toxicity Anemia Interstitial pneumonia

rhEPO in HSCT

585

marrow or PBSC alike (Birgegard et al. 1989; Abedi et al. 1990; Schapira et al. 1990; Beguin et al. 1991; Bosi et al. 1991a; Grace et al. 1991; Lazarus et al. 1992; Miller et al. 1992a; Beguin et al. 1993; Davies et al. 1995; Bosi et al. 1991b; Beguin et al. 1998;). This is also independent of the type of chemotherapy or the use of TBI. The peak Epo values are usually observed 0–7 days after transplant, at the time of the nadir of erythropoietic activity. As Epo exerts its action on target cells after binding to a speciﬁc Epo receptor (Youssouﬁan et al. 1993), severe myelosuppression following the conditioning regimen could disrupt the usual Epo degradation by Epo receptorbearing cells and provoke a surge of serum Epo concentration through prolonged Epo life span (Bowen et al. 1990; Cazzola et al. 1998). However, transplant recipients conditioned with a much milder nonmyeloablative regimen (NMSCT) still experience a surge in serum Epo levels that is quite similar to that observed after myeloablative conditioning (Baron et al. 2002b), but the reason for this observation remains unclear. With marrow recovery after transplantation, Epo levels progressively return to an appropriate range and the duration of this correction phase inversely correlates with the speed of engraftment (Beguin et al. 1998). In addition, patients with particularly intense erythropoietic activity even exhibit somewhat decreased Epo levels during their recovery phase (Beguin et al. 1998; Baron et al. 2003b). This is consistent with increased Epo consumption by an expanding pool of erythroid precursors. After marrow recovery, endogenous Epo remains appropriate for the degree of anemia in autologous HSCT recipients (Ireland et al. 1990; Schapira et al. 1990; Beguin et al. 1991; Bosi et al. 1991b; Lazarus et al. 1992; Beguin et al. 1993; Davies et al. 1995; Beguin et al. 1998) but rapidly becomes inadequately low in patients receiving an allogeneic transplant (Ireland et al. 1990; Schapira et al. 1990; Beguin et al. 1991; Bosi et al. 1991a; Miller et al. 1992a; Beguin et al. 1993). Therefore, the development of erythropoiesis after autologous transplantation is limited by the availability of Epo-receptor bearing erythroid precursors rather than the supply of Epo, whereas after allogeneic transplantation erythropoietic recovery is impaired because Epo levels remain inadequate for prolonged periods of time. An obvious difference between autologous and allogeneic transplantation is the use of ciclosporin, which has been shown to affect renal function and directly inhibit Epo production (Vannucchi et al. 1991; Vannucchi et al. 1993; Vannucchi et al. 1994a). Indeed, the defect in Epo production is much more pronounced in patients receiving ciclosporin than in those undergoing T-cell depletion as prophylaxis against graft-versus-host disease (GVHD) (Abedi et al. 1990). However, there is no direct relationship between ciclosporin blood levels and the degree of impairment of Epo production (Ireland et al. 1990; Beguin et al. 1991; Miller et al. 1992a; Beguin et al. 1993). Furthermore, NMSCT recipients, who have high ciclosporin concentration, do not show the same defect in endogenous Epo levels (Baron et al. 2002b).

586

G. Van Straelen and Y. Beguin

A number of other factors may contribute to these differences. Epo levels inappropriately low for the degree of anemia are observed after amphotericin B administration (Lin et al. 1990), but this has not been a clear ﬁnding in HSCT recipients (Schapira et al. 1990; Miller et al. 1992a). Acute GVHD has often (Ireland et al. 1990; Beguin et al. 1991; Miller et al. 1992a; Beguin et al. 1993), but not always (Grace et al. 1991), been associated with a further depression of serum Epo levels. As the incidence and severity of acute GVHD are lower after NMSCT compared to conventional transplant (Baron et al. 2002a), this may partly explain appropriate levels of serum Epo after NMSCT (Baron et al. 2002b). On the other hand, chronic GVHD has not been found to signiﬁcantly affect serum Epo levels (Beguin et al. 1991; Miller et al. 1992a; Beguin et al. 1993). CMV reactivation or clinical infection have also been found to cause a further depression of Epo production (Ireland et al. 1990; Beguin et al. 1991). Both GVHD (Ferrara et al. 2003) and CMV infection (Lacey et al. 2004) are associated with excessive production of a number of cytokines. It has been reported that IL-1α, IL-1β, TNF-α, IFN-γ, and TGF-β inhibited, whereas IL-6 stimulated Epo production (Faquin et al. 1992; Chuncharunee et al. 1993; Vannucchi et al. 1994b; Nieken et al. 1995; Frede et al. 1997). Others have found inhibition of Epo production with IL-1 and TNF but not for TGF-β, IFN-γ, or IL-6 (Jelkmann et al. 1990; Fandrey et al. 1991; Jelkmann et al. 1992; Miller et al. 1992b). IL-1, TNF-α and IL-6 also blocked hypoxia-induced Epo formation by the isolated rat kidney (Jelkmann et al. 1992). All these observations highlight the inhibitory properties of various cytokines on Epo production and suggest that such mechanisms may be operative after allogeneic transplantation.

Treatment with rhEPO after HSCT RhEPO after allogeneic transplantation In 1991, Heyll (1991) reported the ﬁrst use of rhEPO in a transplant patient. A 31-yr-old man underwent allogeneic BMT for CML in chronic phase from an HLA-matched sister. There was a major ABO incompatibility between the donor and recipient. Whereas engraftment of granulocytes and platelets was adequate, anemia persisted with continuing requirements for RBC transfusions 8 months after BMT. Anemia was due to pure red cell aplasia in the presence of markedly increased endogenous Epo levels. After rhEPO was given at a dose of 4000 I.U. daily, reticulocytes began to increase and hemoglobin exceeded 10 g/dl. Two weeks later, therapy was stopped but hemoglobin levels remained around 10 g/dl with no further need for transfusions. This report was followed by others. Paltiel et al. (1993) described the response to rhEPO at a dose of 50, then 25 U/kg/d for a total of 100 days in a patient with pure red cell aplasia and elevated endogenous Epo levels after ABOincompatible BMT. This resulted in prompt reticulocytosis, conversion to

rhEPO in HSCT

587

donor type blood group and correction of the anemia. Taniguchi et al. (1993) reported another patient who underwent major ABO-incompatible BMT and developed pure red cell aplasia. Despite elevated serum Epo concentration, after treating her with 9000 U/day of rhEPO, the hemoglobin concentration increased and was maintained despite cessation of therapy 50 days later. There was a simultaneous occurrence of a drop in agglutinin titers and a possible explanation was in vivo absorption by newly produced erythroid cells. In another case reported by Ohashi et al. (1994), pure red cell aplasia after major ABO-incompatible BMT was resistant to treatment with I.V. γ-globulins, prednisolone or erythropoietin but ﬁnally responded to a combination of erythropoietin (given for 3 weeks) and methylprednisolone. Blood group converted to donor type with hemoglobin correction. Santamaria et al. (1997) reported a patient with continuing transfusion-dependent pure red cell aplasia despite conversion to donor type ABO blood group and titers of anti-donor isohemagglutinin being undetectable. Although endogenous serum Epo was high, treatment with a 2-wk course rhEPO resulted in rapid improvement and maintenance of normal hemoglobin thereafter. Fujisawa et al. (1996) reported 2 patients with pure red cell aplasia after major ABOincompatible BMT. One patient responded well to rhEPO with appearance of donor type RBC and correction of anemia, while the other patient did not respond. Ustun et al. (1999) reported a patient with myelodysplastic syndrome who developed aregenerative anemia after ABO-incompatible allogeneic PBSCT. A ﬁrst course of high-dose methylprednisolone and rhEPO did not result in any improvement of anemia in the early period after allogeneic PBSCT. However, following plasma exchange a second course with higher dose of rhEPO and methylprednisolone associated with danazol was successful. Martelli et al. (1994) also reported the successful correction of pure red cell aplasia with an 8-wk course of rhEPO after an autologous peripheral blood autologous BMT with adequate endogenous Epo levels. There is also a report of two patients with osteomyeloﬁbrosis and prolonged anemia after PBSCT that responded well after being switched from erythropoietin to darbepoietin alpha (Nguyen et al. 2003). Delayed immune hemolysis is another possible complication of major ABO-incompatible bone marrow transplant. Instead of increasing immunosuppression, Lopez et al. (1994) gave erythropoietin and observed an increase in reticulocytes sufﬁcient to maintain hemoglobin despite persistent hemolysis. A dose of 125 U/kg/day was initially required but rhEPO could ﬁnally be discontinued with no further hemolytic episode. However, the most frequent cause of prolonged anemia after allogeneic BMT is a marked impairment of Epo production in response to anemia. Locatelli et al. (1992) reported 2 children with aplastic anemia undergoing allogeneic BMT with partially matched family donors who, after initial erythropoietic recovery, became severely anemic in the presence of inappropriately low serum Epo levels. Both were treated with rhEPO at an initial dosage of 50 U/kg 3 times weekly and experienced a prompt reticulocyte and

588

G. Van Straelen and Y. Beguin

hemoglobin response. A hemoglobin level around 12 g/dl was maintained with a single weekly SC administration of rhEPO. Fujimori et al. (1998) treated 9 patients with late-onset anemia caused by GVHD, CMV infection or impaired Epo secretion more than 50 days after allogeneic BMT. Very short courses (median 3 weeks) of rhEPO were given in most instances and 6/9 patients responded well, including 2 achieving Hb levels ≥13 g/dL. The 3 patients with GVHD did not respond. Miller et al. (1994) conducted an open trial of rhEPO in 18 allo-BMT recipients and compared them to 50 historical controls. RhEPO was given IV at the dose of 500–1,000 U/kg/wk in 5 doses between day 1 and 28, followed by 450–900 U/kg/wk in 3 doses between days 29 and 70. Although the median time to a corrected reticulocyte count >0132% was signiﬁcantly reduced from 31 to 19 days, there was no effect on the number of RBC transfusions given between days 5 and 82 (16 vs 19). Interestingly, median time to 1,000 leukocytes was less in the rhEPO patients (14 vs 17, p = 0.03) but the median time to 500 granulocytes was not different. Vannucchi et al. (1992) treated 8 allogeneic BMT patients with 1,050 U/kg/wk IV rhEPO given in 3 daily IV injections from day 1 to day 30. There was no difference in neutrophil or platelet engraftment nor in the number of platelet transfusions. Erythroid engraftment was faster in patients receiving rhEPO, as illustrated by a shorter time to high ﬂuorescent reticulocyte count ≥10,000/μl or to an Hct ≥35%, as well as by higher total and high ﬂuorescent reticulocyte counts at day 21. As a consequence, the number of RBC transfusions was sharply reduced from 11 to 4 units per patient. Link et al. (1993) administered 1,050 U/kg/wk of rhEPO by continuous IV infusion to 19 allo-BMT recipients until the Hct exceeded 35% for one week without transfusion. Erythrocyte engraftment was much faster and this translated into higher reticulocyte counts at day 22 (p = 0.042) or day 29 (p = 0.065), as well as into a shorter time to transfusion independence (17 vs 24 days, p = 0.015). As compared to the 43 historical controls, the number of RBC transfusions was reduced from 10 ± 7 U to 7 ± 4 U between days 0 and 30 and this was at the limit of statistical signiﬁcance. There were no data available for total RBC requirements. Locatelli et al. (1993) carried out a pilot study of rhEPO to accelerate erythroid engraftment in 15 children receiving an allogeneic transplant. Erythropoietin was given daily IV at the dose of 525 U/kg/wk from day 1 to day 30. As compared to 16 historical controls, median times to 500 neutrophils or 50,000 platelets were not affected. However, the number of platelet transfusions administered between days 0 and 30, as well as the total number of platelet transfusions, were signiﬁcantly less in the rhEPO patients. Engraftment of the erythroid lineage was consistently accelerated and this translated into day 30 reticulocyte counts, as well as day 15 or day 30 sTfR levels that were about twice as high as in controls. This accelerated erythroid recovery and resulted into a signiﬁcant reduction in the number of RBC transfusions administered.

rhEPO in HSCT

589

Vannucchi et al. (1997) conducted a small randomized placebo-controlled trial in which rhEPO or placebo was administered at enormous doses (3,500 U/kg/wk) by continuous infusion for 30 days to 20 patients undergoing an allogeneic BMT. Treatment with erythropoietin produced signs of accelerated erythropoiesis with higher reticulocyte and sTfR values, resulting in a two-fold reduction in RBC transfusion requirements in the ﬁrst 30 days posttransplant. No signiﬁcant effect was noted on granulopoiesis, platelet recovery or the number of platelet transfusions. Steegmann et al. (1992) randomized 28 allogeneic BMT patients between rhEPO given daily IV for a total dose of 1,050 U/kg/wk between days 0 and 30 or no treatment. Among 24 evaluable patients, 13 received rhEPO and 11 no treatment. Neutrophil engraftment was not affected by therapy but the median time to 25,000 platelets was signiﬁcantly reduced by about one third and the median number of platelet transfusions transfused in the ﬁrst 30 days was greatly diminished. Erythroid engraftment, as measured by median time to 0.5% or 2% reticulocytes, was considerably accelerated, with the consequence that the number of RBC transfusions was reduced by two-thirds. Biggs et al. (1995) conducted a prospective double-blind placebocontrolled trial in 91 recipients of an allogeneic transplant (Table 2). Erythropoietin was given IV 3 times a week for a total of 900 U/kg/wk from day 1 to day 42. Neutrophil and platelet engraftments were the same between 48 rhEPO patients and 43 placebo patients. After day 14, reticulocyte counts and hemoglobin levels were signiﬁcantly higher in the rhEPO group. Despite this enhancement of erythroid activity, RBC transfusion needs within the ﬁrst 6 weeks after transplantation were not improved (6 ± 5 vs 7 ± 5 units) and there was also no effect on platelet transfusion requirements during the same time period (11 ± 9 vs 11 ± 7 transfusions). In multivariate analysis rhEPO use was associated with only an 18% reduction in RBC transfusion requirements.

Table 2. Treatment with rhEPO (300 U/kg tiw IV days 0–42) after allo-BMT (Biggs et al. 1995)

Days to neutrophils >500/μl Days to neutrophils >1,000/μl Days to platelets >50,000/μl Platelet transfusions (days 0–42) Hb (g/dl) at day 28 Reticulocytes (×103/μl) at day 19 RBC transfusions (days 0–42)

rhEPO (N = 48)

Placebo (N = 43)

P value

20 ± 4 25 ± 7 29 ± 6 11 ± 9 10.1 31.8 6 ± 5 (*)

20 ± 5 23 ± 5 26 ± 5 11 ± 7 11.3 9.6 7±5

NS NS NS NS 0.003 0.008 NS

* Multivariate analysis: 82% of control group (P = 0.022).

590

G. Van Straelen and Y. Beguin

Table 3. Treatment with rhEPO (200 U/kg/d IV days 1–28, 200 U/kg biw IV days 29–56) after allo-BMT (Klaesson et al. 1994b; Klaesson et al. 1994a)

Days to WBC >200/μl Days to neutrophils >500/μl Platelet transfusions (days 0–60) Day of last platelet transfusion Reticulocytes (%) at month 1 Reticulocytes (%) at month 2 Days to reticulocytes >2% Hct (%) at month 1 Hct (%) at month 2 RBC transfusions (days 0–60) Day of last RBC transfusions Days to Hb > 7 g/dl without transfusion Number of patients with Hb > 12 g/dl

rhEPO (N = 22)

Placebo (N = 23)

P value

16 20 16 ± 12 24 ± 13 1.0 ± 0.9 2.7 ± 1.3 15 ± 5 26 ± 3 30 ± 4 5±5 17 ± 14 14 7

17 20 17 ± 15 29 ± 17 1.1 ± 2.6 2.9 ± 2.7 17 ± 5 24 ± 2 26 ± 3 10 ± 9 30 ± 22 24 0

NS NS NS NS NS NS NS 0.03 0.01 0.04 0.03 0.03 0.004

In the study reported by Klaesson et al. (1994a,b), patients undergoing allogeneic BMT were randomized to IV placebo (n = 25) or rhEPO (n = 25) (Table 3). Erythropoietin was ﬁrst given daily for a total dose of 1,400 U/kg/wk for 4 weeks and then twice a week for a total dose of 400 U/kg/wk for an additional 4 weeks. Neutrophil and platelet engraftments were not different in the 2 groups and there was no difference in number of platelet transfusions administered. Although there was no difference in reticulocyte counts after 4 or 8 weeks or for the median time to 2% reticulocytes, hematocrit was signiﬁcantly higher in the rhEPO group after 4 weeks, as well as after 8 weeks. Consequently, transfusion independence was achieved sooner when rhEPO was given and the number of RBC transfusions received within 60 days was diminished by half. Seven of the treated patients compared with none of the controls reached a hemoglobin ≥12 g/dl during the study period. There is only one pilot trial of combined administration of rhEPO and G-CSF in patients undergoing allogeneic bone marrow transplantation (Locatelli et al. 1994b). Thirteen children in group 1 received 75 U/kg/day rhEPO given IV from day 1 to day 30 plus G-CSF 5 μg/kg/day given IV from day 5 to a WBC count ≥5,000/μl. Group 2 consisted of 15 children receiving rhEPO alone and group 3 of 16 historical controls receiving no rhEPO. Patients receiving G-CSF had a faster neutrophil engraftment than the two other groups. The median time to 30,000 or 50,000 platelets was signiﬁcantly shorter in patients receiving rhEPO with or without G-CSF as compared to historical controls and the number of platelet transfusions was reduced by more than half. The two groups of patients receiving rhEPO showed an accel-

rhEPO in HSCT

591

erated erythropoietic recovery as assessed by reticulocyte counts at day 30 or sTfR levels at day 15 or day 30. Although their rate of erythroid engraftment appeared to be similar, there was an apparent synergistic effect of rhEPO and G-CSF on the number of red cell transfusions, which were two times less in patients receiving combined treatment as compared to those receiving rhEPO alone, the latter having also less transfusion needs than historical controls. These trials of rhEPO therapy after allogeneic HSCT have not taken the pathophysiology of erythropoiesis into account. All studies to date have administered very high doses (usually >1,000 U/kg/wk) of rhEPO starting on day 1 and continuing for 1–2 months or until erythroid engraftment, and thus the cost was prohibitive. Pilot trials showed accelerated erythropoiesis with increased reticulocyte, sTfR and/or hematocrit values, as well as a reduction in RBC transfusions compared to historical controls (Steegmann et al. 1992; Link et al. 1993; Locatelli et al. 1993; Locatelli et al. 1994a,b; Vannucchi et al. 1997). Some even reported an impact on platelet engraftment and/or platelet transfusions(Steegmann et al. 1992; Locatelli et al. 1994a,b). However, larger placebo-controlled studies with rhEPO doses 900–1,400 U/kg/wk conﬁrmed the potential for accelerating red cell but not platelet recovery (Klaesson et al. 1994a; Link et al. 1994). A reduction in RBC transfusions was observed in some studies, but only between day 20 and 40 (not overall), and particularly in patients with severe acute GVHD, in the largest trial. Therefore, soaking patients with huge doses of rhEPO at a time when the erythroid marrow has not developed enough erythroid precursors to respond and when many intercurrent complications such as organ toxicity, infection, acute GVHD, and bleeding may blunt response may not be the best way to use rhEPO after an allogeneic transplant. Therefore, we looked for a more physiological approach. Baron (Baron et al. 2002c) enrolled 34 recipients of an allogeneic HSCT in 3 consecutive trials to determine the optimal utilization of rhEPO therapy in this setting. RhEPO was started on day 1 in 7 patients at a dose of 1,400 IU/kg/week in the ﬁrst trial, between day 56 and 1,440 in 13 other patients at a dose of 500 U/kg/week in a second trial, and ﬁnally on day 35 in another group of 14 patients at a dose of 500 U/kg/week in a third trial. In the ﬁrst trial (only PBSCT recipients), compared to a historical group of 10 patients, erythroid recovery (median time to 1% reticulocytes: 12 vs 27 days) and RBC transfusion independence (21 vs 40 days) were signiﬁcantly faster in the study group but the transfusion requirements were not reduced. In the second trial, responses to rhEPO were high with transfusions signiﬁcantly reduced already in the ﬁrst month of rhEPO therapy, transfusion independence achieved in one week in 12 of 13 patients (92%) and 2 g/dL Hb increments achieved after a median of 6 weeks. In the third trial that involved only PBSCT recipients, transfusion independence was achieved after one week in 13 of 14 patients (92%), 2 g/dL Hb increment after 3 wks and normal Hb values after 8 wks.

592

G. Van Straelen and Y. Beguin

Median number of transfusions per month decreased from 3 at baseline to 0 in months 1, 2 and 3 of rhEPO therapy. The anemia after allogeneic HSCT is thus exquisitely sensitive to rhEPO. The beneﬁt is minimal when it is given early post-transplant, as used in all trials to date. However, the rate of major response is >90% when rhEPO is started after day 35. These results were then conﬁrmed in another study published by Baron et al. (Baron et al. 2003a) who enrolled 13 recipients of an allogeneic PBSCT in a trial of rhEPO therapy given at a once-weekly dose of 500 U/kg/week starting on day 30 post-transplant and compared them to an untreated group control of 10 patients. This again demonstrated that rhEPO therapy is very effective when started one month after transplantation, with an overall probability of achieving an Hb level ≥13 g/dL of 91% versus 14% in controls. After 2 weeks of treatment, transfusion independence was achieved in 12/13 (92%) patients vs 5/10 (50%) patients in the control group (p = 0.05). RBC transfusions were no longer required between day 50 and 150 post-transplant in 11/13 patients in the rhEPO group vs 3/10 patients in the control group (p = 0.0131). The rhEPO group was then compared to a historical group of patients receiving rhEPO at the same total dose and starting also on day 30, but administered in three injections. There was no signiﬁcant difference between the two groups in terms of erythropoietic activity or RBC transfusion requirements (Fig. 1). Our data set the stage for a more rational use of rhEPO after allogeneic HSCT and should renew interest in erythropoietin therapy after the relative disappointment with previous trials that targeted initial erythroid recovery rather than the more physiologically appropriate period that follows engraftment.

RhEPO after autologous transplantation The results of treatment with rhEPO after autologous BMT have remained much more disappointing until recently. In the study published by Ayash et al. (1994), 10 patients with solid tumors undergoing autologous BMT received iron supplementation along with daily IV injections of rhEPO for a total dose of 1,400 U/kg/wk for 4 weeks. As compared to 37 historical controls, median times to 500 neutrophils or to 20,000 platelets were not shortened, but median time to an Hct ≥30% was reduced by half (24 vs about 57 days, p = 0.001). Eight out of 10 rhEPO vs 20 out of 37 historical controls achieved an Hct of 30% within 32 days but this difference did not reach statistical signiﬁcance. The number of RBC transfusions was the same (9 vs 9 units) in the two groups. In the study published by Filip et al. (1999), 11 patients with breast cancer undergoing 5 cycles of chemotherapy supported by PBPC infusions were given G-CSF and rhEPO. Compared with 12 historical controls receiving the same treatment plan but without growth factors, the duration of neutropenia

rhEPO in HSCT

593

Fig. 1. (A) Endogenous erythropoietin production after allogenic peripheral blood hematopoietic stem cell transplantation (HSCT), as assessed by observed to predicted (O/P) erythropoietin ratios (Mean + SEM). The mean value in normal donors is also shown (open circle). (B) Kaplan-Meier plots of time to Hb > 13 g/dL. (C–D) Hb (C) and serum transferrin receptor (sTfR) (D) from day of transplantation in rhEPO group (rhEPO once weekly), historical group (rhEPO at the same dose thrice weekly), and control group (no rhEPO). p values are given for comparisons of the rhEPO group with the control group: (*) <0.05; (**) <0.01; (***) <0.001. From Baron et al. (2003a) with permission

was shorter, the severity of thrombocytopenia was lower, and the number of platelet and red cell transfusions was smaller. However, the need for PBPC support for an epirubicin-cyclophosphamide regimen is questionable and in this context, rhEPO treatment is more analogous to the setting of the anemia of cancer and chemotherapy than to that of transplantation. Miller et al. (1994) conducted a randomized, double-blind trial of rhEPO vs placebo in 50 patients undergoing auto-BMT with purged marrow for acute myelogenous leukemia or non-Hodgkin’s lymphoma. Twenty-six patients received rhEPO IV daily for a total dose of 1,400 U/kg/wk for 4 weeks followed by 3 injections per week for a total dose of 600 U/kg/wk from day 29 to day 50. Median times to 500 neutrophils or 50,000 platelets were the same in the two groups and, consequently, there was no difference in the number of platelet transfusions required in the ﬁrst 50 days after the transplant. Although there was no difference in the median time to reach a

594

G. Van Straelen and Y. Beguin

corrected reticulocyte count ≥2%, the Hb level at day 50 was higher in the rhEPO group (10.5 vs 9.5 g/dl, p = 0.02), but this did not translate into reduced RBC requirements (13 vs 15 U). Because of the poor results obtained with rhEPO alone in auto-BMT recipients, several trials were conducted using a combination of rhEPO and a myeloid growth factor, either G-CSF or GM-CSF. Pene et al. (1993) gave rhEPO at the dose of 150 U/kg/day IV and GM-CSF at the dose of 10 μg/kg/d IV to 18 autologous BMT patients. RhEPO was given until Hb reached 12 g/dl and GM-CSF until the neutrophils reached 500/μl. As compared to 6 concomitant controls receiving GM-CSF and placebo and 65 historical controls receiving GM-CSF alone, there was no difference for the median time to 500 neutrophils (13 vs 18 vs 19 days), median time to 50,000 platelets (31 vs 28 vs 26 days) or the median time to an Hb ≥ 12 g/dl (12 vs 12 vs ? days). As a consequence, there was no difference between rhEPO and placebo for the number of platelet (26 vs 28 transfusions) or RBC (10 vs 12 units) transfusions. Pedrazzini (1993) compared 6 patients receiving a combination of IV rhEPO (300 U/kg/day) and GM-CSF (250 μg/m2/day), given from day 10 to the day on which a neutrophil count of 500/μl was achieved, with 7 control patients. There was no difference for the median times to 500 neutrophils (39 vs 31 days), to 20,000 platelets (45 vs 46 days) or to 10 g/dl of hemoglobin (61 vs 75 days). Consequently, the number of platelet transfusions was not different nor was the number of RBC transfusions between day 0 and 35 (10 ± 3 vs 10 ± 7) or after day 35 (5 ± 7 vs 8 ± 12). Pierelli et al. (1996) gave rhEPO (150 U/kg SC every 48 hours, days 1–11) and G-CSF (5 μg/kg/d SC, days 1–12) to 15 patients with breast or ovarian carcinoma undergoing intensive chemotherapy followed by peripheral blood stem cell support. Compared to 8 historical controls, neutrophil and platelet engraftments were accelerated and the number of platelet transfusions was reduced, but there was no effect on erythroid recovery. The same group (Benedetti et al. 1997) later extended their observations with the same protocol, adding 2 further controls and 15 patients receiving the same schedule of rhEPO but with GM-CSF (5 μg/kg/d, days 1–12). The use of either growth factor combination reduced the duration of neutropenia similarly. With 10 instead of 8 controls, the effect on platelet recovery and platelet transfusions was no longer signiﬁcant. There was no impact of either growth factor combination on RBC transfusion requirements. Olivieri et al. (2004) conducted a trial combining rhEPO and G-CSF in 32 multiple myeloma and lymphoma patients undergoing 39 cycles of autologous HSCT. Starting on day 1, G-CSF was administrated until neutrophil engraftment, while rhEPO was given at the dose of 10,000 U/d for 3 weeks starting on day 1. Compared to a historical control group, neutrophil and platelet reconstitution was signiﬁcantly faster in patients receiving the combination of G-CSF and rhEPO, but only by a few days to 500 neutrophils or 20,000 platelets/μL. The median duration of neutropenia was also signiﬁcantly shorter by 2 days, as were the numbers of days with fever or on anti-

rhEPO in HSCT

595

biotics. Finally, platelet (1 vs 2) and RBC (0 vs 2) transfusion requirements were almost abolished in the treated patients. As the duration of hospitalization was decreased to a median of 9.5 days (4–27) compared with 22 (15–43) in the control group, the mean estimated cost of the transplant procedure with Epo + G-CSF was 18,394 Euros in patients treated with Epo+G-CSF, compared to 23,988 Euros in the control group. However, the major problem of this retrospective analysis is that G-CSF was only started on day 5 in the control group, compared to day 1 in the G-CSF group. This could delay neutrophil engraftment, facilitate infections and hence have deleterious effects on transfusion needs and duration of hospitalization. Chao et al. (1994) randomized 35 patients with Hodgkin’s or nonHodgkin’s lymphoma to receive rhEPO (600 U/kg/wk, IV from day 1 to day 30) or placebo in addition to G-CSF given at the dose of 10 μg/kg/day IV from day 1 until a neutrophil count of 500/μL (Table 4). rhEPO was begun 3 weeks before administration of high-dose therapy but was held during the week of the preparatory regimen. There was no difference between the 18 rhEPO patients and the 17 placebo patients for the times to 500 neutrophils (12 vs 10 days) or 20,000 platelets (22 vs 20 days), number of platelet (10 vs 5 transfusions) or RBC (8 vs 6 units) transfusions between day 0 and day 30. Vannucchi et al. (1996) conducted an open randomized pilot study using the combination of rhEPO (150 U/kg/day IV from day 1 to day 21) ANC and G-CSF (5 μg/kg/day SC from day 1 to ANC recovery) in 30 patients suffering from malignant disorders of the lymphoid lineage and undergoing autoBMT. Ten patient each received G-CSF and rhEPO, G-CSF alone or no growth factor. Neutrophil engraftment was faster in patients receiving GCSF. No effect on platelet engraftment or platelet transfusion requirements was observed. Median time to 30 × 109/L reticulocytes was signiﬁcantly reduced with rhEPO but hematocrit recovery to ≥30% was not faster. In addition, RBC transfusion needs were not reduced by rhEPO, although both groups receiving G-CSF had one-third less transfusions than the control group. Perillo et al. (2002) also examined the impact of the addition of low-dose Il-2 to the combination of G-CSF + EPO, all given from day 1 through to day Table 4. Combined treatment with rhEPO (600 U/kg tiw IV days 1–30) and G-CSF after auto-BMT (Chao et al. 1994)

Days to neutrophils >500/μl Days to platelets >20,000/μl Platelet transfusions (days 0–30) RBC transfusions (days 0–30) Median values.

G-CSF + rhEPO (N = 18)

G-CSF + placebo (N = 17)

P Value

12 22 10 8

10 20 5 6

NS NS NS NS

596

G. Van Straelen and Y. Beguin

12 post-transplant. To this end, two consecutive series of breast or ovarian cancer patients undergoing autologous PBSCT were compared, the ﬁrst 17 patients receiving G-CSF plus rhEPO (150 U/kg/dose every 48 hours) and the last 15 the same combination plus IL-2. Hematopoietic and post-transplant clinical courses were comparable. In particular, transfusion requirements and duration of hospitalization were similar. The addition of IL-2 to the combination of G-CSF + EPO produced various effects on immune reconstitution but had no speciﬁc impact on erythropoiesis. All these trials of rhEPO therapy after autologous HSCT have not taken the pathophysiology of erythropoiesis into account. All studies have administered very high doses of rhEPO starting on day 1 and continuing for 1–2 months or until erythroid engraftment and have shown no advantage to rhEPO therapy. Therefore, soaking patients with huge doses of rhEPO at a time when the erythroid marrow has not developed enough erythroid precursors to respond and when endogenous Epo production is appropriate or even excessive for the degree of anemia may not be the best way to use rhEPO after transplantation. A more physiological alternative could be the administration of rhEPO starting one month after transplantation. Baron et al. (2003b) enrolled 41 consecutive patients with lymphoma or myeloma in a trial of rhEPO therapy given at a dose of 500 U/kg/wk starting on day 30 post-transplant. Compared with a historical control group of 45 consecutive patients, Hb levels were signiﬁcantly higher in the rhEPO group from day 42 through day 150 post-transplant. Eight of 45 patients in the control group versus 0 of 41 patients in the rhEPO group required RBC transfusions after day 30 (p = 0.0059) (Figs. 2 and 3).

Fig. 2. Endogenous serum Epo levels, as assessed by observed to predicted (O/P) Epo ratios (mean ± SE) in 45 patients after autologous peripheral blood hematopoietic stem cell transplantation (HSCT). Mean ± SE values of 31 normal donors are also shown (open circle) (From Baron et al. (2003b) with permission)

rhEPO in HSCT

597

Fig. 3. Evolution of serum transferrin receptor (sTfR) levels (A), Hb levels (B), and reticulocyte counts (C) after autologous hematopoetic stem cell transplantation (HSCT) in 41 patients receiving rhEPO and 44 control patients without rhEPO. p values are given for comparisons between rhEPO group and control group: *, <0.05; **, <0.01; ***, <0.001. (From Baron et al. (2003b) with permission)

598

G. Van Straelen and Y. Beguin

The efﬁcacy of treatment with rhEPO started one month after autologous HSCT was then evaluated in a small prospective randomized trial including 10 patients with and 10 patients without rhEPO (Vanstraelen et al. 2005). After 3 weeks, Hb and serum transferring receptor levels were signiﬁcantly (p < 0.0001) higher in patients who received rhEPO compared with patients who did not. Hb response (+2 g/dL) was 100% versus 28% (p > 0.0001) and Hb correction (≥13 g/dL) 70% versus 10% (p = 0.0238) in the two groups of patients, respectively. The rational use of rhEPO after autologous HSCT, as proposed here, should therefore renew interest in erythropoietin therapy. The results are impressive and suggest that autologous HSCT, similarly to allogeneic HSCT, is associated with the best response rate to rhEPO outside the setting of uremia. These trials now justify the development of prospective, randomized trials that should investigate clinical endpoints, such as transfusion requirements and quality of life, as well as cost-effectiveness.

Comparative effects of rhEPO in autologous and allogeneic transplantation The generally negative results obtained in patients undergoing autologous BMT were conﬁrmed in studies comparing the effects of rhEPO in autologous versus allogeneic transplants. Locatelli et al. (1994a) carried out a pilot study on the use of rhEPO in children undergoing allogeneic or purged autologous BMT for acute leukemia. RhEPO was given IV daily for a total of 725 U/kg/wk from day 1 to day 30 post-transplant. In auto-BMT patients, there was no difference between 10 rhEPO patients and 10 historical controls for neutrophil, platelet or erythroid engraftment. There was no apparent stimulation of reticulocyte output, no increase in sTfR at day 15 or at day 30, and the numbers of platelet or RBC transfusions were not reduced. By contrast, rhEPO had a signiﬁcant impact on erythroid engraftment in 10 alloBMT compared to 15 historical controls. Although the reticulocyte count was not different between the 2 groups at day 15, sTfR levels were already signiﬁcantly elevated. This was conﬁrmed at day 30 when the reticulocyte count was also higher than in the historical controls. As a result of faster erythroid engraftment, the number of red cell transfusions was reduced by half. Interestingly, although the median time to 500 neutrophils or to 50,000 platelets was not signiﬁcantly affected by rhEPO therapy, the number of platelet transfusions was also cut by half. Another comparison of the effect of rhEPO in autologous vs allogeneic transplantation was conducted by Link et al. (1994) in a prospective, randomized, double-blind, placebo-controlled multicenter trial (Table 5). After allo-BMT, 106 patients received rhEPO and 109 received placebo. After auto-BMT, there were 57 patients in each arm. Patients received rhEPO by continuous IV infusion for a total dose of 1,050 U/kg/wk until day 41 or until independence from erythrocyte transfusions was achieved with a stable Hb

* not clearly deﬁned.

Days to neutrophils >500/μl Days to platelet independence (*) Platelet transfusions Reticulocytes (×103/μl) at day 35 Days to RBC independence (*) RBC transfusions (days 0–20) RBC transfusions (days 21–41) RBC transfusions (days 42–100) RBC transfusions (total) – 23 – ≈160 19 7±5 1±3 2±6 8±6

rhEPO (N = 106) – 21 – ≈45 27 6±4 3±4 5 ± 10 9±7

Placebo (N = 109)

Allogeneic BMT

NS NS NS 0.05 0.03 NS 0.04 NS NS

P value

– 24 – ≈25 24 5±4 2±3 2±4 8±6

rhEPO (N = 57)

– 22 – ≈20 27 5±3 3±3 3±9 7±5

Placebo (N = 57)

Autologous BMT

Table 5. Comparative effect of rhEPO (150 U/kg/d IV days 1–42) in auto-BMT and allo-BMT (Link et al. 1994)

NS NS NS NS NS NS NS NS NS

P value rhEPO in HSCT 599

600

G. Van Straelen and Y. Beguin

> 9 g/dl. Neither in allogeneic nor in autologous BMT was there any effect on the median times to 500 neutrophils or to platelet independence, and the number of platelet transfusions remained unchanged. In auto-BMT, there was no difference in the regeneration of reticulocytes, time to transfusion independence, and the number of RBC transfusions. After allo-BMT, reticulocyte counts were higher with rhEPO from day 21 to day 42 and median time to transfusion independence was signiﬁcantly shortened by about onethird. Whereas the mean number of RBC transfusions was not different for the periods going from day 0 to day 20 or from day 42 to day 100, there was a signiﬁcant reduction between days 21 and 41. After day 20, rhEPO significantly reduced transfusion needs in these patient groups. The most spectacular effect was in patients with grade 3–4 GVHD in whom transfusion requirement decreased from 18 ± 9 to 9 ± 7 units. In allo-BMT, a multivariate analysis of time to erythrocyte transfusion independence showed that only treatment with rhEPO signiﬁcantly reduced this time interval, while age, diagnosis, bleeding events, CMV infection, GVHD, T-cell depletion, ABO blood group incompatibility and HLA match or mismatch had no impact. A multivariate analysis of factors inﬂuencing the number of RBC transfusions from day 0 to 41 disclosed an impact only for bleeding events, severe GVHD, and major ABO blood group incompatibility, while treatment with rhEPO or placebo, age, diagnosis, CMV infection, T-cell depletion and HLA match or mismatch had no relevant inﬂuence. When the analysis was performed for the period from day 21 to 41, only rhEPO had a signiﬁcant impact on transfusion needs, particularly in patients with severe GVHD, major ABO incompatibility or hemorrhage. In auto-BMT, only bone marrow purging signiﬁcantly increased the time to transfusion independence, and rhEPO treatment, age, diagnosis, bleeding event or CMV infection had no inﬂuence. Several factors increased the number of transfusions: age older than 35 years, AML diagnosis, and bone marrow purging, but not bleeding events or CMV infection.

RhEPO before transplantation As rhEPO therapy early after an autologous transplant does not appear to provide signiﬁcant beneﬁts, an alternative could be to raise the hemoglobin level before the transplant so that the risk of transfusion could be lowered. The success of such a strategy has ﬁrst been reported by Estrin et al. (1997) in a Jehovah’s Witness undergoing high-dose chemotherapy with PBPC transplantation for non-Hodgkin’s lymphoma. De la Serna et al. (1999) reported a patient with acute promyelocytic leukemia scheduled for allogeneic transplantation, who had liver disease due to severe iron overload, previous drug toxicity and hepatitis C infection. In order to decrease the risk of severe liver toxicity, iron depletion was planned prior to transplantation. The patient was

rhEPO in HSCT

601

given low-dose rhEPO so as to facilitate 34 phlebotomies carried out over 9 months. This treatment markedly reduced iron stores to normal, normalized liver function tests, and effectively permitted to perform the transplant without signiﬁcant liver toxicity. This concept was extended by Ponchio et al. (2000) who treated 10 breast cancer patients with rhEPO 10,000 U tiw for 8 weeks before high-dose chemotherapy with PBPC transplantation. The hemoglobin increased by an average of 4 g/dL and normalized in 8/10 patients, resulting in levels 2.3 g/dL higher than in 25 historical controls before high-dose chemotherapy. RBC transfusion needs after the transplant were signiﬁcantly lower in the treatment group, with only 1 patient requiring a single 2-units transfusion. However, it should be emphasized that transfusion requirements in the historical groups were rather low and that this strategy did not appear to be cost-effective. These results were conﬁrmed by Hunault-Berger et al. (2005) who treated 11 consecutive patients with hematological diseases with rhEPO given during chemotherapy before 15 courses of HDT and autologous PBSCT. RhEPO was administered SC at a dose of 10,000 UI tiw for a maximum of 12 weeks before transplantation. Immediately before transplantation, mean Hb levels were signiﬁcantly higher (p = 0.002) in rhEPOtreated patients (12.9 ± 2.2 g/dL) than in 17 historical controls undergoing 22 courses of HDT and autologous PBSCT (11.4 ± 1.5 g/dL). RBC transfusion requirements fell from 95% (21/22) in historical controls to 26% (4/15) in rhEPO-treated patients (p = 0.00001). Baron et al. (2003c) examined a similar strategy, but after a ﬁrst procedure of autologous transplantation. RhEPO was administered to 11 multiple myeloma patients undergoing tandem autologous PBSC transplantation, with the aim of avoiding RBC transfusions in the second HSCT procedure. RhEPO was not given prior to the ﬁrst transplant, so that patients served as their own internal controls. Patients were scheduled to start rhEPO (500 U/kg/wk) on day 30 after the ﬁrst PBSCT. Median time to an Hb increment >2 g/dL was 4 weeks and 8 out of 11 patients achieved an Hb of 13 g/dL after a median of 8 weeks. The Hb level increased from 9.6 ± 1.0 g/dL at baseline to 13.9 ± 1.4 g/dL on day 100 (p < 0.001), compared to a change from 11.1 ± 0.9 to 11.6 ± 1.9 g/dL in a historical control group. Hb values after the 2nd transplant remained higher throughout the ﬁrst 2 wks but on day 28 reached levels identical to those of the ﬁrst transplant. Ten of 11 patients required RBC transfusions for the ﬁrst PBSCT versus 1 of 11 for the second transplant (p < 0.001). RBC and platelet requirements were 1.7 ± 1.3 and 1.0 ± 1.1 for the ﬁrst versus 0.1 ± 0.3 (p = 0.003) and 0.5 ± 0.7 (NS) for the second procedure, respectively. In conclusion, rhEPO treatment before high-dose chemotherapy appears as the most effective strategy to facilitate the performance of an autologous transplant without RBC transfusions.

602

G. Van Straelen and Y. Beguin

RhEPO to mobilize PBSC Several investigators have studied the usefulness of rhEPO to mobilize hematopoietic progenitor cells into the peripheral blood for the purpose of PBSC collection. Kessinger et al. (1995) administered rhEPO 200 U/kg/day SC to 12 patients with relapsed malignancies. Steady-state collections were compared with aphereses started after 4 days of Epo administration and continued to a total of 6.5 × 108 MNC/kg. After a median of 8 (range 5–14) aphereses the total progenitor cell doses collected were still low. Although there was no signiﬁcant change in the percentage of CD34+ cells, there was a signiﬁcant increase in BFU-E and CFU-GM, which culminated on the 5th day of Epo administration. Nine patients received Epo-mobilized PBSC after high-dose therapy and the median times to 500 neutrophils and last platelet transfusion were 16 and 24 days, respectively. No details are given on the high-dose chemotherapy regimen used. The same group (O’Kane-Murphy et al. 1994) evaluated the CD34 content of the Epo-mobilized PBSC products and found no correlation with the number of colony-forming cells. CD34+ cells were increased in some patients but others showed elevated CD34+ cell counts already before rhEPO administration. Pettengell et al. (1994) gave rhEPO at a dose of 300–450 U/kg SC twice weekly for 2 weeks to 11 patients with untreated lymphoma. The total number of peripheral blood colonies increased 5-fold, including a 7-fold increase in CFU-GM, a 4-fold increase in BFU-E and a 2-fold increase in primitive BFU-E. While there was no change in CFU-MK or CFU-MIX numbers, the number of CD34 + cells increased also signiﬁcantly. Maximal progenitor cell release was observed at days 5–8 but there was no return to baseline within the 15 days of study. Unfortunately, no change in long-term culture-initiating cells could be demonstrated. In the bone marrow, the only signiﬁcant rise was in erythroid progenitors. Although these studies demonstrate that rhEPO can somewhat mobilize progenitor cells into the peripheral blood of steady-state patients, this effect is much less pronounced thanthe one obtained with other agents such as G-CSF. However, the combination of rhEPO and G-CSF could prove to have synergistic effects. Ferrari et al. (1999) documented CD34+ progenitor cell mobilization with topotecan followed by the combination of G-CSF (5 μg/kg/d) and rhEPO (10,000 IU/d) starting 24 h after chemotherapy in 10 patients with small-cell lung cancer. Filip et al. (1997) administered G-CSF (5 μg/kg/d) and rhEPO (250 IU/kg/d) to 11 consecutive breast carcinoma patients after priming with epirubicin and cyclophosphamide chemotherapy. The results show a signiﬁcant mobilization of CD34+ cells, CFU-GM and BFU-E. Joshi et al. (2000) studied the functional and phenotypic properties of PBSC collected from 15 cancer patients mobilized with G-CSF (10 μg/kg/d) + rhEPO (300 IU/kg/d). Only the ﬁrst 4 apheresis collections from each patient were evaluated for this study. The results show that the combination of EPO + G-CSF not only mobilized hematopoietic precursor cells but also

rhEPO in HSCT

603

increased the number of myeloid cells, B-cells and NK cells in the peripheral blood. Pierelli et al. (1994) tested the hypothesis in a comparison of PBSC mobilization after chemotherapy with either G-CSF alone or G-CSF + rhEPO. Although there were only 5 patients with ovarian carcinoma in each group, those receiving rhEPO at a dose of 150 U/kg SC every 48 h in addition to G-CSF 5 μg/kg/day SC for 13 days, had a signiﬁcant increment of circulating CFU-GM and BFU-E whose peak value also occurred earlier. In addition, although there was no difference in neutrophil or platelet nadir, those patients receiving rhEPO had a signiﬁcantly higher hematocrit nadir, which could facilitate PBSC collection without transfusion. Olivieri et al. (1995) conducted a retrospective study in 34 patients to assess the effectiveness of the combination of rhEPO (50 U/kg/d) + G-CSF (5 μg/kg/d) (n = 16) starting 24 h after priming chemotherapy, compared to the results obtained by G-CSF alone (n = 18). The duration of post-priming neutropenia was similar in the 2 groups. The combination of Epo and G-CSF was more effective than G-CSF alone, with a median of 1.9-fold increase for circulating MNC, 4.0-fold for CFU-GM, 4.7-fold for BFU-E and 2.8-fold for CD34+ cells. The differences were statistically signiﬁcant both for mobilization and collection. Waller et al. (1999) then carried out a prospective randomized trial in 32 patients with newly diagnosed stage II-IV breast cancer. Mobilization and harvest of PBSC followed cycle 2 of VIP-E chemotherapy. Sixteen patients were randomized to G-CSF (5 μg/kg) + rhEPO (150 IU/kg) and 14 patients to G-CSF alone, starting day 1 after chemotherapy. There was no signiﬁcant difference with regard to CD34+ cell kinetics in peripheral blood or MNC, CD34+ cells, BFU-E and CFU-GM in apheresis products. Transplantation of >1 × 106 CD34+ cells/kg after high-dose chemotherapy resulted in similar hematological recovery of the two groups. In another prospective randomized trial, Pierelli et al. (1999) included 50 ovarian cancer patients to mobilize PBSC after a ﬁrst course of ETP chemotherapy using either G-CSF (5 μg/kg/d) or G-CSF plus rhEPO (150 IU/kg every 48 h) from day 2 to day 13. The addition of rhEPO to G-CSF increased PBSC mobilization (p = 0.0009) and collection (p = 0.0026). Patients receiving G-CSF alone required a signiﬁcantly (p = 0.0076) higher number of leukaphereses to obtain the planned minimum dose of CD34+ cells. However, the number of CFU-GM and BFU-E was not different. Fortytwo patients underwent transplants after high-dose chemotherapy with the CEM regimen. The ﬁrst 18 patients received reinfusions of a ﬁxed dose of 4 × 106/kg of G-CSF (9 patients) or G-CSF + EPO (9 patients) mobilized CD34+ cells to allow formal comparison of the in vivo functional properties of the collected PBSC. Trilineage hematopoietic recovery was identical. The subsequent 24 patients were given reinfusions of the entire G-CSF (12 patients) or G-CSF + EPO (12 patients) mobilized products to evaluate the in vivo hematopoietic potential of the entire graft. Despite higher CD34+ cell

604

G. Van Straelen and Y. Beguin

dose in the G-CSF + EPO arm, neutrophil and erythroid recovery (including number of RBC transfusions) were similar, although platelet engraftment occurred one day earlier and number of platelet transfusions was reduced from 1.4 to 0.5 per patient. Perillo et al. (2001) conducted a randomized comparison of the mobilizing capacity of G-CSF + EPO versus sequential GM-/G-CSF + EPO following a ﬁrst cycle of ETP chemotherapy (epirubicin, paclitaxel and cisplatin) in ovarian cancer patients. Twenty patients received G-CSF (5 μg/kg/d) from day 2 to day 13 and 20 patients received GM-CSF (5 μg/kg/d) from day 2 to day 6 followed by G-CSF (5 μg/kg/d) from day 7 to day 13. RhEPO was administered every other day from day 2 to day 13 at a dose of 150 IU/kg in both arms. Although CD34+ cell mobilization into the peripheral blood was comparable, GM-/G-CSF+EPO patients had signiﬁcantly higher CD34+ cell yields because of higher CD34+ cell collection efﬁciency. However, numbers of CFU-GM, BFU-E and LTC-IC collected were identical in the two arms. Identical doses of PBSC mobilized by GM-/G-CSF+EPO or G-CSF+EPO resulted in comparable hematopoietic recovery after reinfusion in patients treated with identical high-dose chemotherapy. Therefore, although one randomized trial suggests synergistic effects of the combination of G-CSF and Epo for mobilization of autologous PBSC, others do not and this strategy has not been shown to result in signiﬁcant acceleration of engraftment after autologous transplantation. Furthermore, a note of caution was derived from the only study that investigated PBSC mobilization in normal donors for allogeneic use. Sautois et al. (2001) administered rhEPO (600 IU/kg) twice weekly with IV iron (200 mg) supplementation for 3 weeks to normal donors and collected RBC units. G-CSF was added in the last 5 days and PBSC were collected and infused into their respective recipients. The administration of rhEPO alone for 2.5 weeks was not associated with any signiﬁcant increase of circulating WBC, CD34+ cells or progenitor cell numbers over baseline. Compared to 10 donor/recipient pairs receiving G-CSF alone for PBSC mobilization, the cumulative yields of NC and CFU-GM were signiﬁcantly lower in the study group while those of BFU-E, CFU-Mix and CD34+ cells were similar. However, neutrophil and platelet engraftment were similar, while erythroid recovery was signiﬁcantly accelerated in the study group because rhEPO was then administered also post-transplant. These results do not support synergistic effects of G-CSF and Epo for PBSC mobilization in normal donors, and may even suggest deleterious effects.

Administration of rhEPO to donors Another interesting use of rhEPO in the setting of HSCT could be its administration to normal donors. This approach could beneﬁt both the donor and

rhEPO in HSCT

605

the recipient. In the donor, this could preclude the need for autologous blood collection and, more importantly, enable donors in whom autologous blood cannot be obtained to avoid homologous blood exposure. This would be most beneﬁcial to children or anemic donors. This approach would also beneﬁt the recipient by priming the allograft to increase the number of Epo-responsive progenitors and by increasing blood donation by the donor for subsequent use by the recipient. York et al. (1992) selected 2 children and 8 adults who either weighed less than 30 kg and did not have a unit of autologous blood stored or had a unit of blood stored but were anemic or were projected to have an operative blood loss exceeding 15 ml/kg. These donors received 9–22 daily SC injections of 100 U/kg of rhEPO as well as oral iron. Between baseline and preoperative hematocrit, there was a 16% increase in the rhEPO group as compared to a 4% decrement in the control group. Between baseline and postoperative hematocrit, there was a signiﬁcantly lower decrement in the rhEPO group than in controls (4% vs 26%). After marrow harvest, rhEPO could be continued safely at a dose of 150 U/kg SC three times a week for 2 weeks or until the hematocrit reached 40%. Akiyama (Akiyama et al. 1994) gave 100 U/kg SC rhEPO 3 times a week for 3 weeks to 4 bone marrow donors who had less than 13 g/dl of Hb 3 weeks before marrow donation. As compared to control donors, preoperative and postoperative Hb levels were signiﬁcantly higher but the proportion of erythroid cells in the harvested marrow was unchanged. Martinez et al. (1998) administered rhEPO to 11 healthy bone marrow donors weighing less than 30 kg. Three weeks before harvesting, the donors received 100 IU/kg/d rhEPO and oral iron supplementation. Hematocrit before harvesting increased by 10.6 ± 1.3% above baseline value. Six children with subnormal Hb after harvesting received rhEPO 150 IU/kg tiw for two weeks. No patient required transfusion during or after bone marrow harvest. On day 15 after bone marrow collection, all donors but one had a Hct value = baseline value. Mitus et al. (1994) exploited the fact that donor blood becomes autologous to the recipient graft by virtue of the transplant. Donors received oral iron and SC rhEPO at 300 U/kg ﬁve days per week beginning 3 weeks before transplant and continuing for 2 weeks afterwards. One unit of red blood cells was obtained twice weekly if hematocrit was ≥33%. A median of 6 units (range 4 to 11 units) were collected. Five of the 11 donors received one or 2 units of autologous blood and none required homologous blood. The number of MNC obtained at marrow harvest was greater in the rhEPO-treated group. Recipients received rhEPO at 200 U/kg IV daily for 28 days and transfusions were administered at the discretion of the physicians. There was no difference in neutrophil and platelet engraftment nor in platelet transfusions between rhEPO-treated and control recipients. Although the time to reticulocytes ≥10,000/μl was shorter with the use of rhEPO, the mean total number of RBC transfusions was the same as in controls. However, rhEPO patients

606

G. Van Straelen and Y. Beguin

received only 5 ± 6 RBC units vs 8 ± 6 in controls (n.s.). Six of 11 rhEPOtreated recipients received only donor-derived blood, compared with 0 of 11 in the control group. However, taking into account platelet transfusions, total exposure to homologous blood was not diminished by rhEPO as administered in this study. Sautois et al. (2001) took this approach further, this time with PBSC donors rather than marrow donors. RhEPO (600 IU/kg) was administered twice weekly with IV iron (200 mg) supplementation for 3 weeks before and 3 weeks after the transplant to 8 ABO-compatible donors. A red blood cell (RBC) unit was collected at each visit, so that up to 12 RBC units per donor were available for use in their recipient. These recipients were treated with 200 IU/kg/d IV rhEPO until they reached an unsupported hematocrit ≥30%. Ninety-ﬁve percent of planned RBC units were collected in the donors without signiﬁcant Hct decrease because of a 4-fold increase in RBC production. Erythroid recovery was signiﬁcantly accelerated in the study group and RBC transfusion independence was achieved after a median of 21 (7–29) days in the study group vs 40 (21–110) days in the control group (p = 0.0007). The total numbers of RBC units transfused were not signiﬁcantly different between the 2 groups, but the number of unrelated RBC units transfused was dramatically lower (2 vs 15, p = 0.026) in the study group. However, some RBC units were not used and the difﬁcult logistics and high cost of the procedure precludes widespread use of this strategy.

Side effects of rhEPO administration in HSCT None of the studies performed in the setting of HSCT reported an increased incidence of side effects, such as hypertension or other cardiovascular events. In particular, there was no evidence of “stem cell steal” by which the stimulation of the erythroid compartment would be detrimental to the engraftment or maintenance of cells in other lineages. However, few studies were large enough to fully address the short-term and long-term safety of erythropoietin therapy after HSCT. Van den Bent (van den Bent et al. 1999) reported a bone marrow transplant patient who developed visual hallucination episodes in the context of hypertension after administration of rhEPO, with full recovery after discontinuation. T2-weighted magnetic resonance imaging showed increased signal in the occipital white matter of both hemispheres, in the periventricular white matter and in the cerebellar hemisphere. These characteristic lesions were typical of posterior leukoencephalopathy syndrome. Mann et al. (1996) reported the effect of erythropoietin administration on immunity in murine bone marrow chimeras. Half of the mice received 3 injections per week of Epo (12 UI), whereas the others received vehicle for 7 weeks starting 48 hours after transplantation. There was no signiﬁcant alter-

rhEPO in HSCT

607

ation in lymphocyte numbers, although a nonsigniﬁcant shift in lymphocytes toward T cell predominance was observed. RhEPO administration resulted in enhanced cell proliferation in response to T and B cell mitogens, although no alteration in cytotoxicity or natural killer cell activity was observed. It is unknown whether these observations would be relevant in human transplantation. Summary Anemia is almost universal after allogeneic and autologous HSCT. After allogeneic, but less so after autologous transplantation, this is in large part due to defective endogenous EPO production that starts about one month after transplantation. In this group of patients, the use of rhEPO from transplantation to engraftment has been shown to accelerate the recovery of erythropoiesis and to shorten the time to transfusion independence and to reduce RBC transfusions in some but not all studies. In patients receiving autologous HSCT, such a beneﬁt has not been shown. Recent studies, however, indicate that in this group of patients as well as in those receiving allogeneic HSCT, the results could markedly improve when treatment with rhEPO would only start after initial engraftment (i.e. after day 30), when endogenous EPO production becomes defective, and not immediately after the transplant. Recent studies also show that in the autologous setting, treatment with rhEPO before HD chemotherapy may be an effective strategy to facilitate the performance of a HSCT without RBC transfusions. The role of rhEPO in stem cell mobilization has not yet been conclusively deﬁned.

References 1. Abedi MR, Backman L, Bostrom L, Lindback B, Ringden O (1990) Markedly increased serum erythropoietin levels following conditioning for allogeneic bone marrow transplantation. Bone Marrow Transplant 6: 121–126 2. Akiyama H, Tanikawa S, Takamoto S, Sakamaki H, Sasaki T, Onozawa Y (1994) Recombinant human erythropoietin (EPO) administration to marrow donors. Blood 84 [Suppl 1]: 732a (Abstract) 3. Ayash LJ, Elias A, Hunt M, Demetri G, Wheeler C, Tepler I, Schwartz G, Mazanet R, Reich E, McCauley M, Antman K, Anderson KC (1994) Recombinant human erythropoietin for the treatment of the anaemia associated with autologous bone marrow transplantation. Br J Haematol 87: 153–161 4. Baron F, Beguin Y (2002a) Nonmyeloablative allogeneic hematopoietic stem cell transplantation. J Hematother Stem Cell Res 11: 243–263 5. Baron F, Fillet G, Beguin Y (2002b) Erythropoiesis after nonmyeloablative stemcell transplantation is not impaired by inadequate erythropoietin production as observed after conventional allogeneic transplantation. Transplantation 74: 1692–1696

608

G. Van Straelen and Y. Beguin

6. Baron F, Frere P, Beguin Y (2003a) Once weekly recombinant human erythropoietin therapy is very efﬁcient after allogeneic peripheral blood stem cell transplantation when started soon after engraftment. Haematologica 88: 718–720 7. Baron F, Frere P, Fillet G, Beguin Y (2003b) Recombinant human erythropoietin therapy is very effective after an autologous peripheral blood stem cell transplant when started soon after engraftment. Clin Cancer Res 9: 5566–5572 8. Baron F, Frere P, Fillet G, Beguin Y (2003c) Tandem high-dose therapy (HDT) for multiple myeloma: recombinant human erythropoietin therapy given between ﬁrst and second HDT allows second peripheral blood stem cell transplantation without red blood cell transfusion. Br J Haematol 123: 103–105 9. Vanstraelen G, Baron F, Frere P, Hafraoui K, Fillet G, Beguin Y (2005) Efﬁcacy of recombinant human erythropoietin therapy started one month after autologous peripheral blood stem cell transplantation. Haematologica 90: 1269–1270 10. Baron F, Sautois B, Baudoux E, Matus G, Fillet G, Beguin Y (2002c) Optimization of recombinant human erythropoietin therapy after allogeneic hematopoietic stem cell transplantation. Exp Hematol 30: 546–554 11. Beguin Y, Baron F, Fillet G (1998) Inﬂuence of marrow erythropoietic activity on serum erythropoietin levels after autologous hematopoietic stem cell transplantation. Haematologica 83: 1076–1081 12. Beguin Y, Clemons GK, Oris R, Fillet G (1991) Circulating erythropoietin levels after bone marrow transplantation: Inappropriate response to anemia in allogeneic transplants. Blood 77: 868–873 13. Beguin Y, Oris R, Fillet G (1993) Dynamics of erythropoietic recovery after bone marrow transplantation: role of marrow proliferative capacity and erythropoietin production in autologous versus allogeneic transplants. Bone Marrow Transplant 11: 285–292 14. Benedetti PP, Pierelli L, Scambia G, Foddai ML, Salerno MG, Menichella G, Vittori M, Maneschi F, Caracussi U, Seraﬁni R, Leone G, Mancuso S (1997) Highdose carboplatin, etoposide and melphalan (CEM) with peripheral blood progenitor cell support as late intensiﬁcation for high-risk cancer: non-haematological, haematological toxicities and role of growth factor administration. Br J Cancer 75: 1205–1212 15. Biggs JC, Atkinson KA, Booker V, Concannon A, Dart GW, Dodds A, Downs K, Szer J, Turner J, Worthington R (1995) Prospective randomised double-blind trial of the in vivo use of recombinant human erythropoietin in bone marrow transplantation from HLA-identical sibling donors. the Australian Bone Marrow Transplant Study Group. Bone Marrow Transplant 15: 129–134 16. Birgegard G, Wide L, Simonsson B (1989) Marked erythropoietin increase before fall in Hb after treatment with cytostatic drugs suggests mechanism other than anaemia for stimulation. Br J Haematol 72: 462–466 17. Bosi A, Vannucchi AM, Grossi A, Guidi S, Saccardi R, Rafanelli D, Longo G, Ferrini PR (1991a) Inadequate erythropoietin production in allogeneic bone marrow transplant patients. Haematologica 76: 280–284 18. Bosi A, Vannucchi AM, Grossi A, Guidi S, Vannucchi L, Saccardi R, Bernabei PA, Longo G, Rafanelli D, Rossi Ferrini P (1991b) Serum erythropoietin levels in patients undergoing autologous bone marrow transplantation. Bone Marrow Transplant 7: 421–425 19. Bowen DT, Janowska-Wieczorek A (1990) Serum erythropoietin following cytostatic therapy [letter]. Br J Haematol 74: 372–373

rhEPO in HSCT

609

20. Cazzola M, Guarnone R, Cerani P, Centenara E, Rovati A, Beguin Y (1998) Red blood cell precursor mass as an independent determinant of serum erythropoietin level. Blood 91: 2139–2145 21. Chao NJ, Schriber JR, Long GD, Negrin RS, Catolico M, Brown BW, Miller LL, Blume KG (1994) A randomized study of erythropoietin and granulocyte colonystimulating factor (G-CSF) versus placebo and G-CSF for patients with Hodgkin’s and non-Hodgkin’s lymphoma undergoing autologous bone marrow transplantation. Blood 83: 2823–2828 22. Chuncharunee S, Carter CD, Studtmann KE, Caro J, Coffey RJ, Dessypris EN (1993) Chronic administration of transforming growth factor-beta suppresses erythropoietin-dependent erythropoiesis and induces tumour necrosis factor in vivo. Br J Haematol 84: 374–380 23. Davies SV, Fegan CD, Kendall R, Beguin Y, Cavill I (1995) Serum erythropoietin during autologous bone marrow transplantation: relationship to measures of erythroid activity. Clin Lab Haematol 17: 139–144 24. de la Serna J, Bornstein R, Garcia-Bueno MJ, Lahuerta-Palacios JJ (1999) Iron depletion by phlebotomy with recombinant erythropoietin prior to allogeneic transplantation to prevent liver toxicity. Bone Marrow Transplant 23: 95–97 25. Erslev AJ (1991) Erythropoietin titers in health and disease. Semin Hematol 28 [Suppl 3]: 2–7 26. Eschbach JW, Egrie JC, Downing MR, Browne JK, Adamson JW (1987) Correction of the anemia of end-stage renal disease with recombinant human erythropoietin. N Engl J Med 316: 73–78 27. Estrin JT, Ford PA, Henry DH, Stradden AP, Mason BA (1997) Erythropoietin permits high-dose chemotherapy with peripheral blood stem-cell transplant for a Jehovah’s Witness [letter]. Am J Hematol 55: 51–52 28. Fandrey J, Jelkmann WE (1991) Interleukin-1 and tumor necrosis factor-alpha inhibit erythropoietin production in vitro. Ann NY Acad Sci 628: 250–255 29. Faquin WC, Schneider TJ, Goldberg MA (1992) Effect of inﬂammatory cytokines on hypoxia-induced erythropoietin production. Blood 79: 1987–1994 30. Ferrara JL, Cooke KR, Teshima T (2003) The pathophysiology of acute graftversus-host disease. Int J Hematol 78: 181–187 31. Ferrari S, Danova M, Porta C, Comolli G, Brugnatelli S, Pugliese P, Riccardi A, Ascari E (1999) Circulating progenitor cell release and functional characterization after topotecan plus G-CSF and erythropoietin in small cell lung cancer patients. Int J Oncol 15: 811–815 32. Filip S, Vanasek J, Blaha M, Mericka P, Vavrova J, Podzimek K (1999) The increase of the rate of hemopoietic recovery and clinical beneﬁt of the erythropoietin (EPO) and granulocyte colony-stimulating factor (G-CSF) with peripheral blood progenitor cells (PBPC) after intensive cyclic chemotherapy in high-risk breast cancer patients. Neoplasma 46: 166–172 33. Filip S, Vanasek J, Blaha M, Vavrova J (1997) Circulation of progenitor cells after intensive chemotherapy followed by combination G-CSF and EPO in breast carcinoma. Neoplasma 44: 212–218 34. Frede S, Fandrey J, Pagel H, Hellwig T, Jelkmann W (1997) Erythropoietin gene expression is suppressed after lipopolysaccharide or interleukin-1 beta injections in rats. Am J Physiol 273: R1067–R1071 35. Fujimori Y, Kanamaru A, Saheki K, Mori A, Takatsuka H, Wada H, Okada M, Tamura S, Okamoto T, Takemoto Y, Kakishita E, Nagai K (1998) Recombinant

610

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

47.

48.

G. Van Straelen and Y. Beguin human erythropoietin for late-onset anemia after allogeneic bone marrow transplantation. Int J Hematol 67: 131–136 Fujisawa S, Maruta A, Sakai R, Taguchi J, Tomita N, Ogawa K, Kodama F, Takahashi K, Shibayama S, Kobayashi S, Ikuta K, Okubo T (1996) Pure red cell aplasia after major ABO-incompatible bone marrow transplantation: two case reports of treatment with recombinant human erythropoietin. Transpl Int 9: 506–508 Grace RJ, Kendall RG, Chapman C, Hartley AE, Barnard DL, Norfolk DR (1991) Changes in serum erythropoietin levels during allogeneic bone marrow transplantation. Eur J Haematol 47: 81–85 Heyll A, Aul C, Runde V, Arning M, Schneider W, Wernet P (1991) Treatment of pure red cell aplasia after major ABO-incompatible bone marrow transplantation with recombinant erythropoietin [letter]. Blood 77: 906 Hunault-Berger M, Tanguy-Schmidt A, Rachieru P, Levy V, Truchan-Graczyk M, Francois S, Gardembas-Pain M, Dib M, Foussard C, Piard N, Godon A, SolalCeligny P, Ifrah N (2005) rHuEpo before high-dose therapy allows autologous peripheral stem-cell transplantation without red blood cell transfusion: a pilot study. Bone Marrow Transplant 35: 903–907 Ireland RM, Atkinson K, Concannon A, Dodds A, Downs K, Biggs JC (1990) Serum erythropoietin changes in autologous and allogeneic bone marrow transplant patients. Br J Haematol 76: 128–134 Jelkmann W, Pagel H, Wolff M, Fandrey J (1992) Monokines inhibiting erythropoietin production in human hepatoma cultures and in isolated perfused rat kidneys. Life Sci 50: 301–308 Jelkmann W, Wolff M, Fandrey J (1990) Modulation of the production of erythropoietin by cytokines: in vitro studies and their clinical implications. Contrib Nephrol 87: 68–77 Joshi SS, Miller K, Jackson JD, Warkentin P, Kessinger A (2000) Immunological properties of mononuclear cells from blood stem cell harvests following mobilization with erythropoietin + G-CSF in cancer patients. Cytotherapy 2: 15–24 Kessinger A, Bishop MR, Jackson JD, O’Kane-Murphy B, Vose JM, Bierman PJ, Reed EC, Warkentin PI, Armitage JO, Sharp JG (1995) Erythropoietin for mobilization of circulating progenitor cells in patients with previously treated relapsed malignancies. Exp Hematol 23: 609–612 Klaesson S, Ringden O, Ljungman P, Lonnqvist B, Wennberg L (1994a) Reduced blood transfusions requirements after allogeneic bone marrow transplantation: results of a randomised, double-blind study with high-dose erythropoietin. Bone Marrow Transplant 13: 397–402 Klaesson S, Ringden O, Ljungman P, Lonnqvist B, Wennberg L (1994b) Treatment with erythropoietin after allogeneic bone marrow transplantation: a randomized, double-blind study. Transplant Proc 26: 1827–1828 Lacey SF, Diamond DJ, Zaia JA (2004) Assessment of cellular immunity to human cytomegalovirus in recipients of allogeneic stem cell transplants. Biol Blood Marrow Transplant 10: 433–447 Lazarus HM, Goodnough LT, Goldwasser E, Long G, Arnold JL, Strohl KP (1992) Serum erythropoietin levels and blood component therapy after autologous bone marrow transplantation: implications for erythropoietin therapy in this setting. Bone Marrow Transplant 10: 71–75

rhEPO in HSCT

611

49. Lin AC, Goldwasser E, Bernard EM, Chapman SW (1990) Amphotericin B blunts erythropoietin response to anemia. J Infect Dis 161: 348–351 50. Link H, Boogaerts MA, Fauser AA, Slavin S, Reiffers J, Gorin NC, Carella AM, Mandelli F, Burdach S, Ferrant A, Linkesch W, Tura S, Bacigalupo A, Schindel F, Heinrichs H (1994) A controlled trial of recombinant human erythropoietin after bone marrow transplantation. Blood 84: 3327–3325 51. Link H, Brune T, Hubner G, Diedrich H, Freund M, Stoll M, Peest D, Ebell W, Bettoni C, Oster W, Nicolay U, Heinrichs H (1993) Effect of recombinant human erythropoietin after allogenic bone marrow transplantation. Ann Hematol 67: 169–173 52. Locatelli F, Pedrazzoli P, Barosi G, Zecca M, Porta F, Liberato L, Gambarana D, Nespoli L, Cazzola M (1992) Recombinant human erythropoietin is effective in correcting erythropoietin-deﬁcient anaemia after allogeneic bone marrow transplantation. Br J Haematol 80: 545–549 53. Locatelli F, Zecca M, Beguin Y, Giorgiani G, Ponchio L, De Stefano P, Cazzola M (1993) Accelerated erythroid repopulation with no stem-cell competition effect in children treated with recombinant human erythropoietin after allogeneic bone marrow transplantation. Br J Haematol 84: 752–754 54. Locatelli F, Zecca M, Pedrazzoli P, Prete L, Quaglini S, Comoli P, De Stefano P, Beguin Y, Robustelli della Cuna G, Severi F, Cazzola M (1994a) Use of recombinant human erythropoietin after bone marrow transplantation in pediatric patients with acute leukemia: effect on erythroid repopulation in autologous versus allogeneic transplants. Bone Marrow Transplant 13: 403–410 55. Locatelli F, Zecca M, Ponchio L, Beguin Y, Giorgiani G, Maccario R, Bonetti F, De Stephano P, Cazzola M (1994b) Pilot trial of combined administration of erythropoietin and granulocyte colony-stimulating factor to children undergoing allogeneic bone marrow transplantation. Bone Marrow Transplant 14: 929– 935 56. Lopez J, Steegmann JL, Perez G, Otero MJ, Berberana M, Camara R, Lamana M, Fernandez Villalta MJ, Fernandez-Ranada JM (1994) Erythropoietin in the treatment of delayed immune hemolysis of a major ABO-incompatible bone marrow transplant. Am J Hematol 45: 237–239 57. Mann RA, Jetzt AE, Singh M, Singh AB (1996) The effect of erythropoietin administration on murine bone marrow chimeras. Immunol Lett 49: 15–20 58. Martelli M, Ponchio L, Beguin Y, Meloni G, Mandelli F, Cazzola M (1994) Pure red cell aplasia following peripheral stem cell transplantation: complete response to a short course of high-dose recombinant human erythropoietin. Haematologica 79: 456–459 59. Martinez AM, Sastre A, Munoz A, Badell I, Maldonado MS, Cubells J (1998) Recombinant human erythropoietin (rh-Epo) administration to normal child bone marrow donors. Bone Marrow Transplant 22: 137–138 60. Miller CB, Jones RJ, Zahurak ML, Piantadosi S, Burns WH, Santos GW, Spivak JL (1992a) Impaired erythropoietin response to anemia after bone marrow transplantation. Blood 80: 2677–2682 61. Miller CB, Mills S (1994) Erythropoietin after bone marrow transplantation. Hematol Oncol Clin North Am 8: 975–992 62. Miller KL, Carlino JA, Ogawa Y, Avis PD, Carroll KG (1992b) Alterations in erythropoiesis in TGF-beta 1-treated mice. Exp Hematol 20: 951–956

612

G. Van Straelen and Y. Beguin

63. Mitus AJ, Antin JH, Rutherford CJ, McGarigle CJ, Goldberg MA (1994) Use of recombinant human erythropoietin in allogeneic bone marrow transplant donor/ recipient pairs. Blood 83: 1952–1957 64. Nguyen VA, Fauser AA, Basara N, Kiehl M (2003) Erythropoietic recovery during treatment with darbepoietin-alpha after impaired rHuEPO response to anemia in two patients with osteomyeloﬁbrosis after peripheral blood stem cell transplantation. Hematol J 4: 456–458 65. Nieken J, Mulder NH, Buter J, Vellenga E, Limburg PC, Piers DA, de Vries EG (1995) Recombinant human interleukin-6 induces a rapid and reversible anemia in cancer patients. Blood 86: 900–905 66. O’Kane-Murphy B, Jackson JD, Kuszynski C, Costas P, Wang PN, Warkentin PI, Kessinger A (1994) CD34 analysis in erythropoietin mobilized peripheral blood stem cells. Prog Clin Biol Res 389: 371–376 67. Ohashi K, Akiyama H, Takamoto S, Tanikawa S, Sakamaki H, Onozawa Y (1994) Treatment of pure red cell aplasia after major ABO-incompatible bone marrow transplantation resistant to erythropoietin. Bone Marrow Transplant 13: 335– 336 68. Olivieri A, Ofﬁdani M, Cantori I, Ciniero L, Ombrosi L, Masia MC, Brunori M, Montroni M, Leoni P (1995) Addition of erythropoietin to granulocyte colonystimulating factor after priming chemotherapy enhances hemopoietic progenitor mobilization. Bone Marrow Transplant 16: 765–770 69. Olivieri A, Scortechini I, Capelli D, Montanari M, Lucesole M, Gini G, Troiani M, Ofﬁdani M, Poloni A, Masia MC, Raggetti GM, Leoni P (2004) Combined administration of alpha-erythropoietin and ﬁlgrastim can improve the outcome and cost balance of autologous stem cell transplantation in patients with lymphoproliferative disorders. Bone Marrow Transplant 34: 693–702 70. Paltiel O, Cournoyer D, Rybka W (1993) Pure red cell aplasia following ABOincompatible bone marrow transplantation: response to erythropoietin. Transfusion 33: 418–421 71. Pedrazzini A (1993) Erythropoietin and GM-CSF following autologous bone marrow transplantation. Eur J Cancer 29 [Suppl 2]: S15–S17 72. Pene R, Appelbaum FR, Fisher L, Lilleby K, Nemunaitis J, Storb R, Buckner CD (1993) Use of granulocyte-macrophage colony-stimulating factor and erythropoietin in combination after autologous marrow transplantation. Bone Marrow Transplant 11: 219–222 73. Perillo A, Pierelli L, Battaglia A, Salerno MG, Rutella S, Cortesi E, Fattorossi A, De Rosa L, Ferrau F, Lalle M, Leone G, Mancuso S, Scambia G (2002) Administration of low-dose interleukin-2 plus G-CSF/EPO early after autologous PBSC transplantation: effects on immune recovery and NK activity in a prospective study in women with breast and ovarian cancer. Bone Marrow Transplant 30: 571–578 74. Perillo A, Pierelli L, Scambia G, Seraﬁni R, Paladini U, Salerno MG, Bonanno G, Fattorossi A, Leone G, Mancuso S, Menichella G (2001) Peripheral blood progenitor cell collection after epirubicin, paclitaxel, and cisplatin combination chemotherapy using EPO-based cytokine regimens: a randomized comparison of G-CSF and sequential GM-/G-CSF. Transfusion 41: 674–680 75. Pettengell R, Woll PJ, Chang J, Coutinho L, Testa NG, Crowther D (1994) Effects of erythropoietin on mobilisation of haemopoietic progenitor cells. Bone Marrow Transplant 14: 125–130

rhEPO in HSCT

613

76. Pierelli L, Menichella G, Scambia G, Teoﬁli L, Iovino S, Seraﬁni R, Benedetti Panici P, Salerno G, Rumi C, Zini G, d’Onofrio G, Leone G, Mancuso S, Bizzi B (1994) In vitro and in vivo effects of recombinant human erythropoietin plus recombinant human G-CSF on human haemopoietic progenitor cells. Bone Marrow Transplant 14: 23–30 77. Pierelli L, Perillo A, Greggi S, Salerno G, Panici PB, Menichella G, Fattorossi A, Leone G, Mancuso S, Scambia G (1999) Erythropoietin addition to granulocyte colony-stimulating factor abrogates life-threatening neutropenia and increases peripheral-blood progenitor-cell mobilization after epirubicin, paclitaxel, and cisplatin combination chemotherapy: results of a randomized comparison. J Clin Oncol 17: 1288–1295 78. Pierelli L, Scambia G, Menichella G, d’Onofrio G, Salerno G, Panici PB, Foddai ML, Vittori M, Lai M, Ciarli M, Puglia G, Mancuso S, Bizzi B (1996) The combination of erythropoietin and granulocyte colony-stimulating factor increases the rate of haemopoietic recovery with clinical beneﬁt after peripheral blood progenitor cell transplantation. Br J Haematol 92: 287–294 79. Ponchio L, Zambelli A, De Stefano A, Robustelli Della Cuna FS, Perotti C, Pedrazzoli P (2000) Transfusion requirement can be abolished by epoietin-α and autologous platelet predeposit in patients receiving high dose chemotherapy with stem cell support [letter]. Haematologica 85: 219–220 80. Santamaria A, Sureda A, Martino R, Domingo-Albos A, Muniz-Diaz E, Brunet S (1997) Successful treatment of pure red cell aplasia after major ABO-incompatible T cell-depleted bone marrow transplantation with erythropoietin. Bone Marrow Transplant 20: 1105–1107 81. Sautois B, Baudoux E, Salmon JP, Michaux S, Schaaf-Lafontaine N, Pereira M, Paulus JM, Fillet G, Beguin Y (2001) Administration of erythropoietin and granulocyte colony-stimulating factor in donor/recipient pairs to collect peripheral blood progenitor cells (PBPC) and red blood cell units for use in the recipient after allogeneic PBPC transplantation. Haematologica 86: 1209– 1218 82. Schapira L, Antin JH, Ransil BJ, Antman KH, Eder JP, McGarigle CJ, Goldberg MA (1990) Serum erythropoietin levels in patients receiving intensive chemotherapy and radiotherapy. Blood 76: 2354–2359 83. Steegmann JL, Lopez J, Otero MJ, Lamana ML, de la Camara R, Berberana M, Diaz A, Fernandez-Ranada JM (1992) Erythropoietin treatment in allogeneic BMT accelerates erythroid reconstitution: results of a prospective controlled randomized trial. Bone Marrow Transplant 10: 541–546 84. Taniguchi S, Yamasaki K, Shibuya T, Asayama R, Harada M, Niho Y (1993) Recombinant human erythropoietin for long-term persistent anemia after major ABO-incompatible bone marrow transplantation [letter]. Bone Marrow Transplant 12: 423 85. Ustun C, Celebi H, Arat M, Ozcan M, Dilek I, Gurman G, Demirer T, Ilhan O, Keskin R, Koc H (1999) Treatment of aregeneratoric anemia following an ABOincompatible allogeneic peripheral blood stem cell transplantation: a case report. Ther Apher 3: 275–277 86. van den Bent MJ, Bos GM, Sillevis Smitt PA, Cornelissen JJ (1999) Erythropoietin induced visual hallucinations after bone marrow transplantation [letter]. J Neurol 246: 614–616

614

G. Van Straelen and Y. Beguin: rhEPO in HSCT

87. Vannucchi AM, Bosi A, Grossi A, Guidi S, Saccardi R, Lombardini L, Rossi Ferrini P (1992) Stimulation of erythroid engraftment by recombinant human erythropoietin in ABO-compatible, HLA-identical, allogeneic bone marrow transplant patients. Leukemia 6: 215–219 88. Vannucchi AM, Bosi A, Grossi A, Guidi S, Saccardi R, Rossi-Ferrini P (1994a) Down-modulation of serum erythropoietin levels following cyclosporin A infusion [letter]. Bone Marrow Transplant 13: 497–498 89. Vannucchi AM, Bosi A, Ieri A, Guidi S, Saccardi R, Lombardini L, Linari S, Laszlo D, Longo G, Rossi-Ferrini P (1996) Combination therapy with G-CSF and erythropoietin after autologous bone marrow transplantation for lymphoid malignancies: a randomized trial. Bone Marrow Transplant 17: 527–531 90. Vannucchi AM, Bosi A, Linari S, Guidi S, Longo G, Lombardini L, Mariani MP, Saccardi R, Laszlo D, Rossi FP (1997) High doses of recombinant human erythropoietin fail to accelerate platelet reconstitution in allogeneic bone marrow transplantation. Results of a pilot study. Haematologica 82: 53–56 91. Vannucchi AM, Grossi A, Bosi A, Rafanelli D, Guidi S, Saccardi R, Alterini R, Ferrini PR (1991) Impaired erythropoietin production in mice treated with cyclosporin A. Blood 78: 1615–1618 92. Vannucchi AM, Grossi A, Bosi A, Rafanelli D, Statello M, Guidi S, Rossi-Ferrini P (1993) Effects of cyclosporin A on erythropoietin production by the human Hep3B hepatoma cell line. Blood 82: 978–984 93. Vannucchi AM, Grossi A, Rafanelli D, Statello M, Cinotti S, Rossi-Ferrini P (1994b) Inhibition of erythropoietin production in vitro by human interferon gamma. Br J Haematol 87: 18–23 94. Waller CF, von Lintig F, Daskalakis A, Musahl V, Lange W (1999) Mobilization of peripheral blood progenitor cells in patients with breast cancer: a prospective randomized trial comparing rhG-CSF with the combination of rhG-CSF plus rhEpo after VIP-E chemotherapy. Bone Marrow Transplant 24: 19–24 95. York A, Clift RA, Sanders JE, Buckner CD (1992) Recombinant human erythropoietin (rh-Epo) administration to normal marrow donors. Bone Marrow Transplant 10: 415–417 96. Youssouﬁan H, Longmore G, Neumann D, Yoshimura A, Lodish HF (1993) Structure, function, and activation of the erythropoietin receptor. Blood 81: 2223–2236 Correspondence: Yves Beguin, MD, University of Liège, Department of Hematology, CHU Sart-Tilman, 4000 Liège, Belgium, E-mail: [email protected]

Chapter 23

Treatment of anemia with rhEPO in radiation oncology J. Dunst Department of Radiation Oncology, University of Luebeck, Germany

Introduction Cancer-related anemia is a complex disease (for review see: Birgegaard et al. 2005; Nowrousian 2007) and is a well-known prognostic factor in patients treated with radiotherapy (for review see: Grau and Overgaard 1998; Caro et al. 2001; Knight et al. 2004; Dunst and Molls 2007). Its impact on local control and survival is mainly explained by the fact that anemia may worsen tumor oxygenation and thereby decreases the radiation sensitivity of a tumor (so-called oxygen effect). Improving tumor tissue oxygenation by avoiding or correcting anemia during radiotherapy is therefore an attractive approach to increase radiation efﬁcacy. Tumor hypoxia, on the other hand, is a complex phenomenom which may be caused by a variety of mechanisms of which anemia is only one.

Hemoglobin (Hb) level and tumor oxygenation Several studies in animal tumors have established an association between low Hb levels and impaired tumor oxygenation. Recent clinical investigations have supported the experimental ﬁndings. Becker et al. (2000) investigated 133 patients with head and neck squamous cell cancers, and an Hb level <11 g/dl was signiﬁcantly associated with poor tumor oxygenation but had no impact on normal tissue oxygenation. Anemia (Hb <11 g/dl) was the strongest predictor of tumor hypoxia. There seems to exist a relatively small range of Hb where tumor oxygenation is optimal. Data from Vaupel and coworkers suggest that the optimal Hb range with regard to tumor oxygenation is between 11 g/dl and 15 g/dl (Vaupel et al. 2002). From a pathophysiological point of view, it is not astonishing that tumor oxygenation is more sensitive to changes in Hb than normal tissue oxygenation for two main reasons (MuellerKlieser et al. 1981; Vaupel et al. 1989). Firstly, normal tissue can compensate for a decrease in Hb over a broad range, mainly by increasing ﬂow. The chaotic vessel network in tumors, however, is characterized by torturous vessels with

616

J. Dunst

varying diameter. These vessels do not have a muscle wall and cannot respond to pharmacological or vegetative stimuli. Thus, a decrease in Hb has more pronounced changes with regard to tissue oxygenation in tumors than in normal tissue. Secondly, tumor vessels are leakier than normal vessels with subsequent ﬂow impairments especially if the hematocrit is high. Other studies have not demonstrated a signiﬁcant impact of Hb levels and tumor oxygenation or only a marginal correlation (Hoeckel et al. 1996; Nordsmark et al. 1996; Nordsmark et al. 2005). The explanation of these discrepancies probably results from the low proportion of anemic patients in the mentioned analyses. There is indeed no association between Hb and tumor oxygenation in patients with Hb levels in the normal range and tumor hypoxia, on the other hand, may be present in about 30% to 50% of tumors in non-anemic patients (Becker et al. 2000). As anemia is a major cause of tumor hypoxia, it is reasonable to assume that treating or preventing anemia during radiotherapy, via avoidance of tumor hypoxia, may improve radiation response.

Prevention of anemia by reduction of bone marrow irradiation The hematological toxicity of radiotherapy depends mainly on the amount of bone marrow which is exposed to therapeutic doses. If pelvic, abdominal or mediastinal lymph nodes are irradiated, large amounts of bone marrow lie nearby the target volume and a signiﬁcant radiation exposure of bone marrow is often unavoidable. Recently, several clinical research groups have investigated whether improved radiotherapy techniques using IMRT (intensity-modulated radiotherapy) can be used to decrease marrow irradiation in patients treated with concurrent chemoradiation regimens for pelvic cancers. Early results demonstrate that a signiﬁcant reduction of marrow irradiation is feasible even in treatment protocols with radiation dose escalation. With these techniques, the volume of bone marrow which is exposed to more than 40 Gy can be reduced by about 50% as compared to standard 4-ﬁeld-box-techniques. So far, however, the clinical relevance of these techniques has not been prospectively evaluated (Ahmed et al. 2004).

Red blood cell transfusions Impact of transfusions on survival in cancer patients Transfusions have traditionally been used to treat anemia in cancer patients. The question whether transfusions might improve or impair survival has been debated over long periods (for review see Dellinger and Anaya 2004). Some

Treatment of anemia with rhEPO in radiation oncology

617

retrospective studies have reported decreased survival ﬁgures in patients with gastrointestinal cancers receiving transfusions during surgery; the interpretation of these data is difﬁcult because the prognostic differences might result from the underling anemia, speciﬁc therapy-related factors (blood loss during surgery, indicating locally advanced or not curatively resectable disease) or the transfusion itself (e.g. by immunosuppressive effects of donated blood). A negative impact of donation of whole blood cannot be excluded from retrospective analyses. Recent multivariate analyses, however, have not conﬁrmed a negative impact of the use of leukocyte-depleted packed erythrocytes on survival (Dellinger and Anaya 2004). There is, on the other hand, evidence that the donation of allogenic blood may (probably transiently) alter immune function parameters (Blumberg and Heal 1996; Santin et al. 2002). The clinical relevance of these ﬁndings in patients with solid cancers remains to be determined. There is some evidence from retrospective studies that transfusions may improve survival in cervical cancers (Table 1). Grogan et al. (1999) found a better survival in patients with Hb levels of 12 g/dl during radiotherapy as compared to patients with lower Hb levels. There was no difference between patients who had spontaneously high Hb levels and patients who had high Hb levels after transfusions. The data therefore suggest that improving Hb levels may have an impact on prognosis. A further retrospective analysis by Kapp et al. (2002) demonstrated excellent survival in cervical cancer patients whose Hb levels were >11 g/dl during the whole course of radiotherapy. The prognosis was signiﬁcantly worse in patients who had Hb levels <11 g/dl at least once (measured weekly) during radiotherapy. The authors demonstrated, in a multivariate model, a beneﬁcial effect of successful transfusion (with keeping the Hb >11 g/dl) after correcting for stage and lymph node involvement. Tumor-related anemia was signiﬁcantly associated with stage, tumor size and lymph node involvement and this association was stronger in patients who did not sufﬁciently respond to transfusions. In a subgroup of anemic patients whose anemia was not tumor-related but resulted from other medical reasons (e.g. renal insufﬁciency or esophageal varicosis), anemia and success of transfusions were not related to outcome and none of these anemic patients failed with regard to cervical cancer. The data highlight three major clinical issues. Firstly, keeping the Hb over a certain threshold seems to improve radiation response (pelvic control). Secondly, however, it is difﬁcult to maintain Hb levels within the desired range with transfusions. Thirdly, the authors assume that tumorrelated anemia and anemia caused by other concomitant illnesses have different impact on prognosis. So far, one randomized study from the Princess Margaret Hospital investigated the effect of transfusions in patients with locally advanced cervical cancers stages FIGO IIB and III (Bush 1986). 132 patients were randomized either in a transfusion group (N = 66) or a non-transfusion group (N = 66).

618

J. Dunst

Table 1. Clinical studies investigating the effect of transfusions in patients with cervical cancer Authors

Type and design of study

Results

Bush RS 1986

Phase-III-randomized study. 132 patients with stage II/III cervical cancer, deﬁnitive radiotherapy with external beam irradiation plus brachytherapy. Endpoint local control. Control arm: transfusions, if Hb drops below 10 g/dl Experimental arm: transfusions, if hb drops below 12.5 g/dl. Retrospective investigation. 475 patients with cervical cancer treated with deﬁnitive radiotherapy.

Decreased local failure rate (12/66 versus 20/66) in experimental transfusion group. Lowest failure rate (6/38) in subgroup with transfusions in experimental arm.

Grogan et al. 1999

Kapp et al. 2002

Retrospective investigation. 204 patients with stage IB-IV cervical cancer treated with deﬁnitive radiotherapy. Transfusions regularly used with the objective to keep the Hb level during radiotherapy above 11 g/dl.

High prognostic impact of average hb-levels during radiotherapy. Best prognosis with high Hb levels (≥12 g/dl) during XRT. No difference in survival between patients with spontaneously high Hb levels or high Hb levels after transfusions. Signiﬁcant impact of Hb levels during XRT on pelvic control and disease-free survival (71% vs. 26%). Successful transfusion (Hb permanently >11 g/dl during XRT) only in 10/54 patients (19%). Successfully transfused patients had identical prognosis as compared to non-anemic patients without transfusion.

In the transfusion group, the study design recommended transfusion if the Hb dropped <12.5 g/dl. In the non-transfusion group, patients were not transfused if the Hb dropped <12.5 g/dl; however, transfusions were administered in case of a Hb <10 g/dl. Patients with Hb levels above the mentioned thresholds did not receive transfusions in both groups. A subset of patients in both groups was transfused (38 in the transfusion and 25 in the non-transfusion group). The endpoint of the study was the local failure rate and the study was closed after a signiﬁcant difference in the local failure rate between the trans-

Treatment of anemia with rhEPO in radiation oncology

619

fused subgroups had been observed (6/38 versus 11/25 failures, p = 0.015). The difference in the whole group of patients, however, was not signiﬁcantly different (12/66 versus 20/66, p = 0.07). This study, from a today’s point of view, was not well planned and has therefore been criticized because of its design, the missing stratiﬁcation for major prognostic factors such as tumor size, and the resulting imbalance between the treatment groups (Fyles et al. 2000). Nevertheless, the study demonstrated an advantage in local control (the primary endpoint) for the experimental group in which the Hb level was kept >12.5 g/dl by transfusions.

Efﬁcacy of transfusions The life span of erythrocytes is reduced in cancer patients (see chapter 6). This holds also true for transfused erythrocytes. Therefore, the effect of a transfusion on the Hb level is transient. There are few data in the literature regarding the efﬁcacy of transfusions in terms of keeping the Hb above a certain threshold. Kapp et al. (2002) have recently in a retrospective investigation analysed the efﬁcacy of transfusions in patients treated with radiotherapy for cervical cancers. Although it was the strategy to transfuse anemic patients with the objective to keep the Hb >11 g/dl, only 19% of 54 transfused patients achieved this desired Hb range; the prognosis of this successfully transfused subgroup with regard to pelvic control and survival was identical to primarily non-anemic, non-transfused patients. However, the majority of patients were not successfully transfused suggesting that it is difﬁcult to achieve a long-lasting stabilization or increase in Hb levels in severely anemic cancer patients.

Risks and side effects of transfusions Possible risks of transfusions mainly depend on the type of donated blood (for review see Dellinger and Anaya 2004). The risk of transmitting infectious agents (mainly hepatitis-virus or HIV) has been reduced in the past two decades and is in the range of less than 1 case per 1.5 million units blood for HIV and about 1/220,000 units for hepatitis B. The risk of a fatal haemolytic transfusion reaction is higher and estimated to be about 1 in 100,000. Immunosuppressive effects of donated blood have been described. The clinical relevance mainly depends on the degree of leukocyte depletion prior to transfusion. The use of buffy-coat-free blood reduces the number of white blood cells (WBC) by 75% down to 109 leukocytes/unit. Leukocyte ﬁlters can further reduce the number of WBC to 106/unit. Meta-analyses from retrospective studies suggest that transfusion of such leukocyte-depleted blood carries no signiﬁcant immunological risk or risk of cancer progression. There

620

J. Dunst

is no clear evidence that other blood preparations are comparably harmless (Dellinger and Anaya 2004).

Recombinant human erythropoietin (rhEPO) Anemia prevention in patients undergoing radiotherapy rhEPO has been successfully used to prevent or treat anemia in cancer patients undergoing curative radiotherapy. A variety of single arm phase-II and one randomized studies have investigated the efﬁcacy of rhEPO (Table 2). All studies demonstrated very similar results. With weekly rhEPO doses of 450 to 1,000 IU/kg, an average increase in Hb levels of 0.4 to 0.7 g/dl was achieved. This weekly increment in Hb is slightly higher than that observed in chemotherapy patients, probably because the degree of anemia is lower in radiotherapy patients as compared to patients treated with chemotherapy and also because of the lower hematological toxicity of radiotherapy regimens. The highest weekly Hb increment was observed in a study by Henke et al. (2003) but the Hb levels at entry were the highest in this study and patients were treated with high rhEPO doses. The largest study has recently recruited 777 anemic patients with nonmyeloid malignancies who were treated with radiation therapy plus concomitant or sequential chemotherapy (Shasha et al. 2003). RhEPO was administered once weekly in a dosage of 40,000 IU subcutaneously and was escalated to 60,000 IU per week in case of no response (Hb increase ≤1 g/dl) after 4 weeks. Among 442 evaluable patients, 74% responded to treatment; response was deﬁned as either an increase in Hb of ≥2 g/dl from baseline or an increase of Hb to >12 g/dl (the target hb-level). The average increase in Hb over the treatment period was 1.9 g/dl. Treatment with rhEPO signiﬁcantly improved overall quality of life. The data of the prospective single arm studies have been conﬁrmed in a randomized phase-II-study (Sweeney et al. 1998). 48 patients with lung, breast, prostate or cervix cancer were treated either with radiation ± chemotherapy or additional rhEPO. rhEPO was administered in a dosage of 5 times weekly 200 U/kg plus oral iron supplementation. Patients in the rhEPO arm achieved a mean weekly increment in Hb levels of 0.41 g/dl compared to a decrease of 0.073 g/dl in the control arm. The mean Hb at the end of treatment was 13.6 g/dl in the rhEPO arm versus 11.0 g/dl in the control patients (p = 0.0012). Moreover, the decline in platelet counts was signiﬁcantly lower in patients treated with rhEPO (11.2% vs. 26.3%). Efﬁcacy of rhEPO versus transfusions for anemia prevention Treating and preventing anemia can be achieved by rhEPO and/or transfusions. Both modalities are required in the daily clinical practice (Table 3). The

Patients with thoracic and head and neck tumors, standard radiotherapy N = 40 Patients with cervical cancer, Hb < 12 g/dl, standard pelvic radiotherapy, N = 20 additional cisplatin chemotherapy in 14 pats. Randomized phase-II-trial of RT ± rhEPO in patients with lung, breast, cervix and prostate cancers, N = 48 Patients with head and neck tumors, standard radiotherapy N = 38 Pelvic irradiation Concurrent cisplatin N = 53

3 × 300 U/kg per week s.c. in week 1, then 2 × 150 U/kg + iron 325 mg tid po 5 × 200 U/kg per week s.c. in week 1/2, then 3 × 200 U/kg + iron 325 mg tid po

Lavey et al. 1993

5 × 200 U/kg per week s.c., reduce to 50% if target-hb reached + iron 325 mg tid po

3 × 150 (300) U/kg per week i.v. (s.c.) + 800 mg Fe-Gluconate iv

3 × 10,000 U per week s.c. + oral iron

Sweeney et al. 1998

Henke et al. 1999

Lavey et al. 2004

Dusenbery et al. 1994

Radiotherapy/ No. of patients

rhEPO dose iron supplementation

Author

10.4

12.1

11.4

11.9

–

Mean Hb at start of study (g/dl)

11.8

14.6

+ Epo: 13.6 No Epo: 11.0

13.2

15.1

Mean Hb at end of study (g/dl)

0.7

+ Epo: 0.41 no Epo: −0.07

0.5

0.6

Average Hb increment per week (g/dl)

Table 2. Dosage of rhEPO and hematological response in prospective phase-II-studies in radiotherapy patients

15 g/dl (female 14 g/dl) (72%) 12.5 g/dl (40%)

15 g/dl (female 14 g/dl) (42%)

14 g/dl

15 g/dl (65%)

Target Hb (% target Hb reached)

Treatment of anemia with rhEPO in radiation oncology 621

622

J. Dunst

Table 3. Advantages and disadvantages of rhEPO versus transfusions in patients treated with deﬁnitive radiation therapy Transfusions

rhEPO

Advantages: • rapid effect on Hb

Advantages: • steady increase of Hb in majority of patients • convenient, especially in outpatients Major disadvantages and risks • delayed effect • uncertain response (no response in a minority of patients) • slightly increased risk of deep vein thrombosis

Major disadvantages and risks • short duration of effect • repeated transfusions necessary if stabilization of Hb over a certain threshold is desired • time-consuming procedure • possible immunological effects and risks

clinical data with rhEPO clearly demonstrate that rhEPO has several advantages over transfusions, mainly the long-lasting effect with a stable course of Hb levels. In contrast, transfusions have the advantage of a rapid effect but this effect is transient and it seems to be difﬁcult to achieve a long-lasting stabilization of Hb with transfusions alone. From a clinical point of view, the logical consequence should be that, depending on the individual situation and the desired effect, both modalities are required and should be used.

Use of rhEPO as radiosensitizer: experimental data The prognostic impact of anemia in radiotherapy patients has led to the hypothesis that rhEPO might act as an indirect radiosensitizer by improving tumor oxygenation and thereby increasing the cytotoxic efﬁcacy of radiation in tumors. The positive impact of correcting low Hb levels on tumor oxygenation is well proven in animal models. Kelleher and coworkers clearly demonstrated that induction of anemia led to impaired tumor oxygenation in tumor bearing mice with a reduced median pO2 and increased hypoxic fraction with pO2 values <5 mm Hg (Kelleher et al. 1995). The effect of anemia was at least partly reversible by correcting Hb levels with the use of EPO; the median pO2 increased and the hypoxic fraction decreased. There are, however, no systematic studies on the impact of transfusions or EPO on tumor oxygenation in human tumors in vivo due to ethical problems associated with repeated invasive pO2-measurements in humans. Some casuistic observations with transfusions suggest that tumor oxygenation may be partly improved (Sundfor et al. 1998; Dunst et al. 1999).

Treatment of anemia with rhEPO in radiation oncology

623

A variety of experimental studies demonstrated that anemia decreases radiation sensitivity and sensitivity to cyclophosphamide (a highly oxygendependent cytotoxic drug, oxygen effect comparable to radiation). Radiation sensitivity was determined in normal animal versus anemic animals versus animals with additional treatment with rhEPO and full or nearly full correction of anemia. Anemia was associated with decreased radiation response and correcting Hb by EPO signiﬁcantly increased the sensitivity to singledose and fractionated irradiation and to cyclophosphamide. All experimental studies in animal tumors demonstrated similar results (Table 4). The data suggest that anemia directly reduces radiosensitivity and that the sensitivity

Table 4. Experimental studies investigating the effect of anemia and anti-anemic treatment with rhEPO on tumor oxygenation and radio/chemosensitivity Authors

Experimental question and design

Results

Kelleher et al. 1995

Oxygenation of experimental tumors in normemic/anemic animals. Impact of rhEPO or transfusions on tumor oxygenation rhEPO in anemic, tumor-bearing animals, single dose irradiation with 10 Gy, growth delay

Increased hypocia due to anemia; partly restoration of tumor oxygenation after correction of anemia

Thews et al. 1998 Thews et al. 2001 Stueben et al. 2001

Stueben et al. 2003

Pinel et al. 2004

Effect of rhEPO in anemic, tumor-bearing animals, treatment with cyclophosphamide, growth delay Effect of rhEPO in anemic nudemice, treatment of xenografted tumors with fractionated irradiation, impact on growth delay and local control Effect of rhEPO in anemic nudemice, treatment of xenografted glioblastomas with fractionated irradiation, impact on growth delay and local control Effect of rhEPO in anemic nude-mice, treatment of two xenografted glioblastomas with fractionated irradiation, impact on tumor oxygenation, vascular density and radiosensitivity

Decreased growth delay in anemic animals, nearly full restoration of normal radiosensitivity after treatment with rhEPO Decreased growth delay in anemic animals, full restoration of sensitivity to cyclophosphamide after treatment with rhEPO Decreased radiosensitivity in anemic animals, restoration of radiosensitivity after treatment with rhEPO Decreased radiosensitivity in anemic animals, restoration of radiosensitivity after treatment with rhEPO Decrease in viable hypoxic tumor cell fraction (by about 20%), no effect on vascular density, signiﬁcant radiation enhancement (enhancement ratio 1.21 and 1.54)

624

J. Dunst

to radiation can fully or at least partly be restored by rhEPO and correction of Hb levels.

Use of rhEPO as radiosensitizer: clinical data The well-known prognostic impact of Hb levels during radiotherapy on locoregional control and survival, the efﬁcacy of rhEPO to improve Hb levels and the animal data demonstrating increased radiation sensitivity after correcting anemia with rhEPO have been the scientiﬁc basis for several studies in which rhEPO has been used to prevent or correct anemia with the objective of improving local control by radiotherapy. In a non-randomized phase-II-study at the university of Vienna (Glaser et al. 2001), 191 patients with squamous cell cancers of the oral cavity and oropharynx were treated with preoperative chemoradiotherapy followed by surgery in the period from 1989 through 1998. Radiation was administered in conventional fractionation over ﬁve weeks to a total dose of 50 Gy. Chemotherapy consisted of mitomycin C (15 mg/m2 on day 1) plus continuous infusional 5-FU (750 mg/m2 per day on days 1–5). This regimen was associated with marked hematological toxicity and a low Hb level during and after radiochemotherapy was identiﬁed as a prognostic factor. Therefore, patients with low pretreatment Hb at the beginning of treatment received additional rhEPO (epoietin alfa 10,000 IU/kg 3–6 times weekly) since 1996. In a multivariate analysis, Hb levels and the use of rhEPO were signiﬁcant and independent prognostic factors for pathohistological response to treatment and locoregional tumor control. Patients with a pretreatment Hb levels <14.5 g/dl had a signiﬁcant better survival if treated with rhEPO compared to patients not receiving rhEPO. In a recent non-randomized phase-II-study at the University of Hamburg, 60 patients with esophageal cancers were treated with radiochemotherapy (Rades et al. 2005). Radiation was given in conventional fractionation up to 45–50.4 Gy. Two to four courses of chemotherapy were administered prior to and during radiotherapy with continuous infusional 5FU (1,000 mg/m2 daily on days 1–5 of each course) plus cisplatin (75 mg/m2 on day 1 of each course). 30 patients received additional rhEPO (150 IU/kg 3 times weekly) in case of Hb levels <13 g/dl plus additional oral iron supplementation if the ferritin level was <100 ng/ml or transferring saturation <20%. Both groups of patients with and without rhEPO were balanced with regard to established prognostic factors as well as pretreament Hb levels. Treatment with rhEPO was effective in terms of keeping the Hb level in the optimal range. 67% of patients with rhEPO had Hb levels in the range of 12 g/dl to 14 g/dl during radiotherapy compared to only 10% of patients who did not receive rhEPO. The univariate analysis demonstrated a signiﬁcant advantage in local control in patients treated with rhEPO (p = 0.012) and a

Treatment of anemia with rhEPO in radiation oncology

625

trend towars improved overall survival (p = 0.08). In a multivariate analysis, the use of rhEPO (p = 0.007) and ECOG performance status (p = 0.027) were the only independent prognostic factors for local control. More recently, several randomized studies have tested this hypothesis in head and neck and pelvic cancers. In contrast to the former studies, in which rhEPO was used as a supportive drug these studies aimed at improving local control and survival (Table 5). The ﬁrst and largest study by Henke and coworkers (ENHANCE-study, Henke et al. 2003) recruited patients with squamous cell cancers of the head and neck who were treated either with postoperative or deﬁnitive radiotherapy. Radiotherapy was administered in conventional fractionation up to total doses of 60 to 70 Gy. The study was double-blinded and placebo-controlled and patients in the experimental arm received 3 times weekly 300 IU/kg epoietin beta. The Hb levels at entry were relatively high with an average of 11.8 g/dl and 11.7 g/dl, respectively, in both arms. Treatment with epoietin started two weeks prior to radiotherapy and was continued during treatment unless a target Hb of 16 g/dl was reached. Additional iron supplementation with intravenous iron was mandatory depending on laboratory investigations of iron stores. Due to the high Hb levels at entry and the high dosage of rhEPO (twice as high as the standard dose), a rapid and steep increase of Hb was observed in the rhEPO group; the average weekly increment in the ﬁrst three weeks of EPO treatment was 1.5 g/dl. This was the highest Hb increment that ever has been observed in a prospective trial. In this study, locoregional progression-free survival was superior in the placebo arm as compared to the rhEPO arm (Table 5). The negative effect of EPO was more pronounced in patients with inoperable tumors or unradical resection and there was no negative effect in patients treated with postoperative irradiation after R0-resection. This study has been criticized for various reasons, for example imbalances in prognostic factors between treatment groups, discrepancies in outcome between oropharyngeal and hypoharynx cancers and inclusion of non-anemic patients. The major criticism from a radiation biology viewpoint is the fact that Hb levels might have been overcorrected (Dunst 2004): The course of Hb during radiotherapy was favorable even in the placebo group with a slight increase in the mean Hb from 11.7 g/dl to 12.5 g/dl. Thus, the placebo group had Hb levels during radiotherapy in the radiobiologially optimal range. In contrast, the mean Hb in the rhEPo group increased from 11.8 g/dl to 15.6 g/dl suggesting that a substantial percentage of patients had Hb levels beyond a radiobiologically optimal range. Moreover, a possible negative impact of the extremely rapid Hb increase in the ﬁrst weeks of treatment cannot be ruled out. Two other randomized studies in head and neck cancers and three randomized studies in cervical cancers (one of these also included patients with bladder cancers) have recently been completed; four of the studies have not yet been fully published (Table 5). Standard therapy was concurrent

H&N SCC, def. RT or postop. RT after R0/R1/R2-resection Hb < 12/13 g/dl* at entry Non-metastatic H&N stage IV or stage III (hypopharynx, base of tongue), Hb <16 g/dl at entry

Non-resectable nonmetastatic H&N SCC, Hb 9–12.5/ 13.5 g/dl* at entry

Henke et al. 2003 ENHANCEstudy

Machtay et al. 2004 RTOG 99-03

Cervix cancer after radical hysterectomy PT1b2/pT2 or pN1 or risk factors

No EPO: N = 26 rhEpo: N = 28 No EPO: N = 38 rhEpo: N = 33 No EPO: N = 128 rhEpo: N = 128

No EPO: N = 68 rhEpo: N = 67

No EPO: N = 43 rhEpo: N = 47

Placebo: N = 171 rhEpo: N = 180

N

4 courses carboplatin/ ifosfamide followed by pelvic irradiation with 50 Gy

XRT in standard fractionation, total dose 60 Gy (R0/R1-resection) or 70 Gy (R2-resection, no surgery) 2 × 1.5 Gy days 1–5, HU 2 × 500 mg p.o./d days 0–5 Paclitaxel 100 mg/m2 day 1 CI-5-FU 600 mg/m2/d days 1–5 repeat XRT/ chemo cycles in weeks 3, 5, 7, 9; total XRT dose75 Gy Def XRT 66–72 Gy, later modiﬁcation of protocol allowed for concurrent chemotherapy with cisplatin or moderate platin/paclitaxel Def. XRT in conv. fractionation, intracavitary brachytherapy for cervix cancers, weekly carboplatin 90 mg/m2 XRT (external beam plus brachy) + weekly cisplatin 40 mg/m2

Standard therapy

rhEPO 10,000 IU sc ﬁve times weekly rhEPO beta 150 IU/Kg sc three times weekly rhEPO alpha 10,000 IU sc three times weekly

rhEPO 40,000 IU sc one weekly

rhEpo alpha 40,000 IU sc one weekly over 14 weeks

rhEPO beta 300 IU/Kg sc three times weekly

Epo-therapy

Trend towards reduced recurrence rate in epo-arm (11% vs. 22%, p = 0.06)

No signiﬁcant effect of progression-free survival

Locoregional progression free survival at 1 ear: No epo: 65% epo: 60% p = 0.65 No signiﬁcant differences in progression-free or overall survival

Locoregional progression free survival signiﬁcantly better in placebo group compared to rhEPO, HR 1.69, p = 0.007 Progression free survival at 36 months**: no epo: ∼58% epo: ∼68% p = 0.35

Results

Abbreviations: H&N = head and neck cancer, SCC = squamous cell cancer, Hb = Hb level, * different upper limits refer to values in female/male patients. ** ﬁgures not given in text and estimated from Kaplan-Meier-plots in publication *** randomised phase-II-trial.

Blohmer et al. 2002

Cervical cancer IBIIIB, blader cancer B2-C, hb at baseline 10–13 g/dl Strauss et al. Cervix cancer stage unpublished IIB-IVA, hb MARCH-study 9–13 g/dl

Throuvalas et al. 2004***

Rosen et al. 2003 Univ. of Chicago

Inclusion criteria

Author

Table 5. Results of randomized phase-III-trials with rhEPO in addition to radiotherapy or simultaneous radiochemotherapy using survival or locoregional control as endpoint

626 J. Dunst

Treatment of anemia with rhEPO in radiation oncology

627

radiation and chemotherapy in three studies and radiation alone or radiation plus cisplatin chemotherapy in the fourth study. Although all ﬁve studies demonstrated a signiﬁcant effect of rhEPO on Hb levels, there was no signiﬁcant impact on locoregional tumor control or survival. Three studies recruited patients with inoperable head and neck or cervix cancer (Rosen et al. 2003; Machtay et al. 2004; Strauss et al. in press); all studies showed Kaplan-Meier curves which are nearly identical in the rhEPO and control arms. In the fourth randomized phase-II-study, patients with pelvic cancers (cervical cancers stage FIGO IB-IIIB or muscle-invasive bladder cancers) were recruited and treated with radiation and weekly carboplatin; no significant difference between both treatment arms has been reported so far (Throuvalas et al. 2004). The ﬁfth study has been performed in patients after radical hysterectomy for cervical cancer (Blohmer et al. 2002). This study recruited patients with locally advanced cervix cancer who received sequential adjuvant treatment with 4 courses of chemotherapy (ifosfamide and carboplatin) followed by standard pelvic radiotherapy (50 Gy). In contrast to the other studies (in which patients received deﬁnitive radiotherapy or radiochemotherapy), this study was performed in the adjuvant setting. Patients in the rhEPO had a signiﬁcant improvement in Hb levels and a signiﬁcant reduction in transfusion need. The interim analysis showed a trend towards improved recurrence-free survival in patients with rhEPO. In summary, all of the randomized studies have failed to prove the study hypothesis of improved radiation response by rhEPO. There was, however, on the other hand no detrimental effect of erythropoietin on tumor progression or survival. For to explain these obvious discrepancies between the nonrandomized clinical studies on the one hand (which support the experimental animal studies) and the randomized studies on the other hand, it is helpful to analyze the study designs in detail. In the non-randomized studies from Vienna and Hamburg, patients were treated with an aggressive radiochemotherapy regimen which produced marked hematological toxicity. Patients with rhEPO were selected on the basis of their pretreatment Hb levels or their individual course of Hb. As a result of this selection, a high frequency of patients without rhEPO had low Hb levels below the optimal range (<12–11 g/dl) whereas the majority of patients with rhEPO maintained Hb levels in the optimal range. In contrast, the randomized studies included patients with higher Hb levels and the treatment regimens were less toxic with regard to hematological toxicity. Furthermore, the target Hb was high (14–16 g/dl); this has probably resulted in unphysiologically high Hb levels in a remarkable number of patients. Moreover, ﬁndings in head and neck cancers suggests that the negative effect of Hb levels on local control and survival is less pronounced in patients treated with radiochemotherapy compared to radiation alone. Thus, the optimal Hb range may be broader in patients treated with concurrent radiochemotherapy as compared to

628

J. Dunst

radiation alone. The negative results of the randomized clinical trials could therefore be related to a small difference in patients who were or were not in the optimal Hb range during radiotherapy. These studies therefore do not rule out that rhEPO may improve radiation response in selected subgroups.

Role of EPO receptors on tumor cells EPO receptors have been demonstrated in a variety of cell lines and tissues besides hematopoietic precursor cells (see chapter 3, 4). These receptors have probably a physiological function and seem to play a role in protecting some tissues (e.g. glial cells) from hypoxia-induced apoptosis (Cerami 2001; Cerami et al. 2002). The functional properties of EPO receptors on tumor cells have been controversially discussed. Some studies have observed decreased apoptosis and increased resistance to radiation by rhEPO (Westenfelder and Baranowski 2000; Westphal et al. 2002; Batra et al. 2003; Belenkov et al. 2004). Neuroectodermal tumor cells have been demonstrated to upregulate the expression of EPO receptors and the secretion of EPO under hypoxia and it is assumed that cells thereby try to escape hypoxia-induced cell death via autocrine stimulation of hypoxia-induced EPO receptors (Batra et al. 2003). Recent studies have conﬁrmed that the expression of EPO receptors may be upregulated in human tumors in response to hypoxia (Selzer et al. 2000; Acs et al. 2003; Arcasoy et al. 2005). EPO may serve as an autocrine stimulus against apoptosis in hypoxic tumor models suggesting that the EPO-EPO-receptor system is an evolutionary preserved mechanism for the regulation of apoptosis in response to hypoxia and environmetal stress. This may have consequences for the use of rhEPO in a variety of clinical settings. The clinical relevance of EPO receptors on tumor cells is so far questionable. In animal models, the use of rhEPO alone (in the control arms of studies) has not promoted tumor growth (Kelleher et al. 1996; Thews et al. 1998; Thews et al. 2001; Stueben et al. 2001, 2003; Pinel et al. 2004). In clinical studies in humans, a meta-analysis has not found any evidence for enhanced tumor progression or decreased response to chemotherapy by the use of rhEPO (Table 6); in these studies, rhEPO had been administered as a supportive drug with the objective to improve quality of life and reduce the need for transfusions. Recently, Henke and colleagues have retrospectively analysed the impact of EPO-receptor (ERO-R) expression in head and neck cancers treated with rh-EPO within the randomised ENHANCE-study in which (as mentioned above) patients were treated with radiotherapy and placebo or epoetin beta. The authors stained pretreatment tissue samples from squamous cell cancers and measured the immunohistological EPO-Rexpression (Henke et al. 2006). 104 out of 154 tumor samples were positive

Treatment of anemia with rhEPO in radiation oncology

629

Table 6. Meta-analysis of randomized studies with rhEPO, publication date in 2001 or earlier. Signiﬁcant improvement of Hb and reduction in transfusion requirements. No effect on tumor response and a trend towards improved survival with rhEPO. Bohlius JF et al. 2003

Hb response Transfusion requirements: Solid cancers Hematological malignancies Tumor response Mortality

N patients

N trials

Rel. risk (95%-CI) rhEPO vs. control

Signiﬁcance

1,150

14

3.60 (3.07–4.23)

p < 0.0001

1,094 966

13 5

0.49 (0.41–0.59) 0.77 (0.67–0.87)

p < 0.0001 p < 0.001

1,150 1,624

7 8

1.08 (0.84–1.38) 0.80 (0.65–1.00)

n.s. p = 0.05

for EPO-R. In patients treated with radiotherapy and placebo, EPO-Rpositive tumors had a better locoregional progression-free survival than EPO-R-negative tumors (relative risk 0.70). Patients in the experimental arm of the study received epoetin beta and in this subgroup, EPO-R-positive tumors had a worse locoregional progression-free survival (relative risk 1.4). Epoetin beta treatment had no effect on outcome in patients with EPOR-negative tumors (relative risk 0.98, p = 0.95), but was associated with an increased risk for locoregional progression in EPO-R-positive tumors (relative risk 2.0, p = 0.003). In the multivariate analysis, however, EPO-Rexpression was statistically not signiﬁcant. The data are difﬁcult to interpret because of the differential prognostic impact of EPO-R-expression in the placebo versus the epoetin-beta arm of this subgroup of the trial. There are no data in how far EPO-R-expression is correlated to other adverse or favourable prognostic factors. Moreover, the majority of patients in the ENHANCE-trial was not severely anemic and the impact of hb-levels on outcome remains unclear. In summary, there is currently no evidence that rhEPO increases the risk of tumor progression if used according to standard guidelines. The role of EPO-receptor expression in tumors requires further investigation in clinical and experimental studies.

Summary and recommendations RhEPO has a clear role for the treatment of anemia in cancer patients. If it is administered according to the current guidelines, a variety of beneﬁcial effects can be achieved (Boogaerts et al. 2003; Boogaerts et al. 2005). The

630

J. Dunst

risk of side effects is relatively low. RhEPO is more effective than transfusions in keeping the Hb level during therapy continuously over a certain threshold. Experimental data suggest that preventing or correcting anemia during radiotherapy may improve local control by radiotherapy, probably because improved tumor tissue oxygenation with increased radiation sensitivity. However, there is currently no evidence that rhEPO, in a clinical setting, improves radiation response. A potential beneﬁcial effect of rhEPO is probably restricted to the subset of patients who experience moderate to severe anemia during radiotherapy. The current data do not support the use of rhEPO as radiosensitizer and the hypothesis of improving radiation response by rhEPO requires further clinical investigation. RhEPO has proven efﬁcacy in anemic radiotherapy patients and improves the spectrum of supportive drugs in radiation oncology. Outside clinical studies, however, rhEPO should only be used according to standard guidelines even in radiotherapy patients (Rizzo et al. 2002; Bokemeyer et al. 2004).

References 1. Acs G, Zhang PJ, McGrath CM, et al (2003) Hypoxia-inducible erythropoietin signaling in squamous dysplasia and squamous cell carcinoma of the uterine cervix and its potential role in cervical carcinogenesis and tumor progression. Am J Pathol 162: 1789–1806 2. Ahmed RS, Kim RY, Duan J, Meleth S, De Los Santos JF, Fiveash JB (2004) IMRT dose escalation for positive paraaortic lymph nodes in patients with locally advanced cervical cancer while reducing dose to bone marrow and other organs at risk. Int J Radiat Oncol Biol Phys 60: 505–512 3. Arcasoy MO, Amin K, Chou SC, Haroon ZA, Varia M, Raleigh JA (2005) Erythropoietin and erythropoietin receptor expression in head and neck cancer: relationship to tumor hypoxia. Clin Cancer Res 11: 20–27 4. Batra S, Perelman N, Lori R, et al (2003) Pediatric tumor cells express erythropoietin and a functional erythropoietin receptor that promotes angiogenesis and tumor cell survival. Lab Invest 83: 1477–1487 5. Becker A, Stadler P, Lavey R, Haensgen G, Kuhnt T, Lautenschlaeger C, Feldmann HJ, Molls M, Dunst J (2000) Severe anemia is associated with poor tumor oxygenation in head and neck squamous cell carcinoma. Int J Radiat Oncol Biol Phys 46: 459–466 6. Belenkov AI, Shenouda G, Rizhevskaya E, Cournoyer D, Belzile JP, Souhami L, Devic S, Chow TY (2004) Erythropoietin induces cancer cell resistance to ionizing radiation and cisplatin. Mol Cancer Ther 3: 1525–1532 7. Birgegard G, Aapro MS, Bokemeyer C, Dicato M, Drings P, Hornedo J, Krzakowski M, Ludwig L, Pecorelli S, Schmoll H, Schneider M, Schrijvers D, Shasha D, Van Belle S (2005) Cancer-related anemia: pathogenesis, prevalence and treatment. Oncology 68 [Suppl 1]: 3–11

Treatment of anemia with rhEPO in radiation oncology

631

8. Blohmer JU, von Minckwitz G, Paepke, et al (2002) Sequential adjuvant chemoradiotherapy with vs. without erythropoietin for patients with high-risk cervical cancer – second analysis of a prospective, randomized, open and controlled AGOand NOGGO-intergroup study. Proc ASCO 21: 206a 9. Blumberg N, Heal JM (1996) The transfusion modulation theory: the th1/th2 paradigm and an analogy with pregnancy as a unifying mechanism. Semin Hematol 33: 329–340 10. Bohlius JF, Langensiepen S, Schwarzer G, Bennet CL, Engert A (2003) Does erythropoietin improve survival in the treatment of patients with malignant diseases? Results of a comprehensive meta-analysis. Blood 102, [Suppl] Abstract #709 11. Bokemeyer C, Aapro MS, Courdi A, Foubert J, Link H, Osterborg A, Repetto L, Soubeyran P (2004) EORTC guidelines for the use of erythropoietic proteins in anaemic patients with cancer. Eur J Cancer 40: 2201–2216 12. Boogaerts M, Coifﬁer B, Kainz C and the Epoietin beta QoL Working Group (2003) Impact of epoietin beta on quality of life in patients with malignant disease. Br J Cancer 88: 988–995 13. Boogaerts M, Mittelman M, Vaupel P (2005) Beyond anaemia management: evolving role of erythropoietin therapy in neurological disorders, multiple myeloma and tumour hypoxia models. Oncology 69 [Suppl 2]: 22–30 14. Bush RS (1986) The signiﬁcance of anemia in clinical radiation therapy. Int J Radiat Oncol Biol Phys 12: 2047–2050 15. Caro JJ, Salas M, Ward A, Goss G (2001) Anemia as an independent prognostic factor for survival in patients with cancer: a systematic, quantitative review. Cancer 91: 2214–2221 C 16. Cerami A (2001) Beyond erythropoiesis: novel applications for recombinant erythropoietin. Semin Hematol 38 [Suppl 7]: 33–39 17. Cerami A, Brins M, Ghezzi P, Cerami C, Itri LM (2002) Neuroprotective properties of epoietin alfa. Nephrol Dial Transplant 17 [Suppl 1]: 8–12 18. Dellinger EP, Anaya DA (2004) Infectious and immunological consequences of blood transfusions. Critical Care 8 [Suppl 2]: S18–S23 19. Dunst J, Feldmann HJ, Becker A, Stadler P, Haensgen G, Molls M (1999) Oxygenation of human tumors: the Halle/Munich experience. In: Vaupel P, Kelleher D (eds) Tumor hypoxia. Wissenschaftliche Verlagsgesellschaft mbH, Stuttgart 20. Dunst J (2004) Erythropoietin and radiotherapy: a dangerous combination? Strahlenther Onkol 180: 133–135 21. Dunst J, Molls M (2006) Incidence and impact of anemia in radiation oncology. In: Nowrousian MR (ed) Recombinant human erythropoietin (rhEPO) in clinical oncology. Springer, Wien New York 22. Dusenbery KE, McGuie WA, Holt PJ, et al (1994) Erythropoietin increases Hb during radiotherapy for cervical cancer. Int J Radiat Oncol Biol Phys 29: 1079–1084 23. Fyles AW, Milosevic M, Pintilie M, Syed A, Hill RP (2000) Anemia, hypoxia and transfusion in patients with cervical cancer: a review. Radiother Oncol 57: 13–19 24. Glaser CM, Millesi W, Kornek GV, et al (2001) Impact of Hb level and use of recombinant erythropoietin on efﬁcacy of preoperative chemoradiation therapy for squamous cell carcinoma of the oral cavity and oropharynx. Int J Radiat Oncol Biol Phys 50: 705–715

632

J. Dunst

25. Grau C, Overgaard J (1998) Signiﬁcance of Hb concentrations for treatment outcome. In: Molls M, Vaupel P (eds) Blood perfusion and microenvironment of human tumors. Implications for clinical radiooncology. Springer, Berlin, pp 101–112 26. Grogan M, Thomas GM, Melamed I, Wong FL (1999) The importance of maintaining high Hb levels during radiation treatment of carcinoma of the cervix. Cancer 86: 1531–1536 27. Henke M, Guttenberger R, Barke A, Pajonk F, Pötter R, Frommhold H (1999) Erythropoietin for patients undergoing radiotherapy: a pilot study. Radiother Oncol 50: 185–190 28. Henke M, Laszig R, Rübe C, Schaefer U, Haase KD, Schilcher D, Mose S, Beer KT, Burger U, Dougherty C, Frommhold H (2003) Erythropoietin to treat head and neck cancer patients with aneamia undergoing radiotherapy: randomized, double-blind, placebo-controlled trial. Lancet 362: 1255–1260 28a. Henke M, Mattern D, Pepe M, Bézay C, Weissenberger C, Werner M, Pajonk F (2006) Do erythropoietin receptors on cancer cells explain unexpected clinical ﬁndings? J Clin Oncol 24: 4708–4713 29. Höckel M, Schlenger K, Aral B, Mitze M, Schaffer U, Vaupel P (1996) Association between tumor hypoxia and malignant progression in advanced cancer of the uterine cervix. Cancer Res 56: 4509–4515 30. Kapp KS, Poschauko J, Geyer E, Berghold A, Oechs AC, Petru E, Lahousen M, Kapp DS (2002) Evaluation of the effect of routine packed blood cell transfusion in anemic cervix cancer patients treated with radical radiotherapy. Int J Radiat Oncol Biol Phys 54: 58–66 31. Kelleher DK, Matthiensen U, Thews O, Vaupel P (1995) Tumor oxygenation in anemic rats: effects of erythropoietin treatment versus red blood cell transfusion. Acta Oncol 34: 379–384 32. Kelleher DK, Matthiesen U, Thews O, Vaupel P (1996) Blood ﬂow, oxygenation and bioenergetic status of tumors after erythropoietin treatment in normal and anemic rats. Cancer Res 56: 4728–4734 33. Knight K, Wade S, Balducci L (2004) Prevalence and outcomes of anemia in cancer: a systematic review of the literature. Am J Med 116: 11S–26S 34. Lavey RS, Dempsey WH (1993) Erythropoietin increases Hb in cancer patients during radiation therapy. Int J Radiat Oncol Biol Phys 27: 1147–1152 35. Lavey RS, Liu PY, Greer BF, Robinson WRD, Chang PC, Wynn RB, Conrad ME, Jiang C, Markman M, Alberts DS (2004) Recombinant human erythropoietin as an adjunct to radiation therapy and cisplatin for stage IIB-IVA carcinoma of the cervix: a Southwest Oncology Group study. Gyn Oncol 95: 145–151 36. Leyland-Jones B, Semiglazov V, Pawlicki M, Pienkowski T, Tjulandin S, Manikhas G, Makhson A, Roth A, Dodwell D, Baselga J, Biakhov M, Valuckas K, Voznyi E, Liu X, Vercammen E (2005) Maintaining normal Hb levels with epoetin alfa in mainly nonanemic patients with metastatic breast cancer receiving ﬁrst-line chemotherapy: a survival study. J Clin Oncol 23: 5960–5972 37. Leyland-Jones B, BEST Investigators and Study Group (2003) Breast cancer trial with erythropoietin terminated unexpectedly. Lancet Oncol 5: 206–207 38. Machtay M, Pajak T, Suntharalingam M, Hershock D, Stripp DC, Cmelak A, Shenouda G, Schulsinger A, Fu K (2004) Deﬁnitive radiotherapy ± erythropoi-

Treatment of anemia with rhEPO in radiation oncology

39.

40.

41.

42.

43.

44.

45.

46.

47.

48.

49.

50. 51.

633

etin for squamous cell carcinoma of the head and neck: preliminary report of RTOG 99-03. Int J Radiat Oncol Biol Phys 60 [Suppl 1]: 132 Mueller-Klieser W, Vaupel P, Manz R, Schmidseder R (1981) Intracapillary oxyHb saturation in malignant tumors in humans. Int J Radiat Oncol Biol Phys 7: 1397–1404 Ning S, Hartley C, Molineux G, Knox SJ (2005) Darbepoietin alfa potentiates the efﬁcacy of radiation therapy in mice with corrected or uncorrected anemia. Cancer Res 65: 284–290 Nordsmark M, Overgaard M, Overgaard J (1996) Pretreatment oxygenation predicts radiation response in advanced squamous cell carcinoma of the head and neck. Radiother Oncol 41: 31–39 Nordsmark M, Bentzen SM, Rudat V, Brizel D, Lartigau E, Stadler P, Becker A, Adam M, Molls M, Dunst J, Terris DJ, Overgaard J (2005) Prognostic value of tumor oxygenation in 397 head and neck tumors after primary radiation therapy. An international multicenter study. Radiother Oncol 77: 18–24 Nowrousian MR (2006) Pathophysiology of cancer-related anemia. In: Nowrousian MR (ed) Recombinant human erythropoietin (rhEPO) in clinical oncology. Springer, Wien New York Pinel S, Barberi-Heyob M, Cohen-Jonathan E, Merlin JL, Delmas C, Plenat F, Chastagner P (2004) Erythropoietin-induced reduction of hypoxia before and during fractionated irradiation contributes to improvement of radioresponse in human glioma xenografts. Int J Radiat Oncol Biol Phys 59: 250–259 Rades D, Schild SE, Yekebas EF, Job H, Schwarz R, Rudat V (2005) Epoetinalpha during radiotherapy for stage III esophageal carcinoma. Cancer 103: 2274–2279 Rizzo JD, Lichtin AE, Woolf SH, et al (2002) Use of epoetin in patients with cancer: evidence-based clinical practice guidelines of the American Society of Clinical Oncology and the American Society of Hematology. J Clin Oncol 20: 4083–4107 Rosen FR, Haraf DJ, Kies MS, Stenson K, Portugal L, List MA, Brockstein BE, Mittal BB, Rademaker AW, Witt ME, Pelzer H, Weichselbaum RR, Vokes EE (2003) Multicenter randomized phase II study of paclitaxel (1hour infusion), ﬂuorouracil, hydroxyurea, and concomitant twice daily radiation with or without erythropoietin for advanced head and neck cancer. Clin Cancer Res 9: 1689– 1697 Santin AD, Bellone S, Palmieri M, et al (2002) Effect of blood transfusions during radiotherapy on the immune fanction of patients with cancer of the uterine cervix: role of interleukin-10. Int J Radiat Oncol Biol Phys 54: 1345–1355 Shasha D, George MJ, Harrison LB (2003) Once-weekly dosing of epoetin-alfa increases Hb and improves quality of life in anemic cancer patients receiving radiation therapy either concomitantly or sequentially with chemotherapy. Cancer 98: 1072–1079 Selzer E, Wachek V, Kodym R, et al (2000) Erythropoietin receptor expression in human melanoma cells. Melanoma Res 10: 421–426 Strauss HG, Haensgen G, Dunst J, Hayward CRW, Burger HU, Koelbl H (2006) Effects of anaemia correction with epoetin beta in patients receiving radiochemotherapy for advanced cervical cancer. Clin Cancer Res (in press)

634

J. Dunst: Treatment of anemia with rhEPO

52. Stueben G, Thews O, Poettgen C, et al (2001) Recombinant human erythropoietin increases the radiosensitivity of xenografted human tumours in anaemic nude mice. J Cancer Res Clin Oncol 127: 346–350 53. Stueben G, Thews O, Poetttgen C, et al (2003) Impact of anemia prevention by recombinant erythropoietin on the sensitivity of xenografted glioblastomas to fractionated irradiation. Strahlenther Onkol 179: 620–625 54. Sundfor K, Lyng H, Rofstad EK (1998) Tumor hypoxia and vascular density as predictors of metastasis in squamous cell carcinoma of the uterine cervix. Br J Cancer 78: 822–827 55. Sweeney PJ, Nicolae D, Ignacio L, Chen L, Roach M, Wara W, Marcus KC, Vijayakumar S (1998) Effect of subcutaneous recombinant human erythropoietin in cancer patients reciving radiotherapy: ﬁnal report of a randomized, openlabelled phase II trial. Br J Cancer 77: 1996–2002 56. Thews O, König R, Kelleher DK, et al (1998) Enhanced radiosensitivity in experimental tumours following erythropoietin treatment of chemotherapy-induced anaemia. Br J Cancer 78: 752–766 57. Thews O, Kelleher DK, Vaupel P (2001) Erythropoietin restores the anemiainduced reduction in cyclophosphamide cytotoxicity in rat tumors. Cancer Res 61: 1358–61 58. Throuvalas N, Antonadou D, Lavey R, Bouﬁ M, Malamos N (2004) Final results of a randomized phase II study evaluating the role of erythropoietin during radiochemotherapy for pelvic tumors. Int J Radiat Oncol Biol Phys 60 [Suppl 1]: 300 59. Vaupel P, Kallinowski F, Okunieff P (1989) Blood ﬂow, oxygen and nutrient supply, and metabolic environment of human tumors: a review. Cancer Res 49: 6449–6465 60. Vaupel P, Thews O, Mayer A, et al (2002) Oxygenation status of gynecological tumors: what is the optimal Hb level? Strahlenther Onkol 178: 727–731 61. Westenfelder C, Baranowski RL (2000) Erythropoietin stimulates proliferation of human renal carcinoma cells. Kidney Int 58: 647–657 62. Westphal G, Niederberger L, Blum C, Wollman Y, Knoch TA, Rebel W, Debus J, Friedrich E (2002) Erythropoietin and G-CSF receptors in human tumor cells: expression and aspects regarding functionality. Tumori 88: 150–159 Correspondence: Prof. Dr. Jürgen Dunst, Department of Radiation Oncology, University Clinic Schleswig-Holstein, Campus Lübeck, Ratzeburger Allee 160, 23538 Lübeck, Germany, E-mail: [email protected]

Chapter 24

Recombinant human erythropoietin in pediatric oncology C. Hastings and J. Feusner 1

Department of Pediatric Hematology and Oncology, Children’s Hospital and Research Center of Oakland, Oakland, CA, USA 2 Director, Pediatric Oncology, Department of Hematology and Oncology, Children’s Hospital and Research Center of Oakland, Oakland, CA, USA

Introduction Cancer- and treatment-related anemia affects many children diagnosed with a variety of malignancies. The incidence of anemia in children with solid tumors or Hodgkin’s disease at the time of diagnosis has been reported to be 51–74% (Hockenberry et al. 2002). Several pediatric studies report a high incidence of anemia (over 50%) requiring intervention with transfusion in children receiving therapy for Wilms’ tumor, leukemia, and osteosarcoma (Borsi et al. 1995; Green et al. 1998; Nachman et al. 1998). Although baseline hemoglobin values differ by age, gender, and type of cancer, a recent European pediatric survey reported over 80% of children receiving chemotherapy were anemic (Michon 2002). It is clearly a common ﬁnding at diagnosis and during the course of treatment for childhood cancer. Historically, treatment for children with cancer and anemia has consisted of blood transfusions. Transfusion, however, can be associated with potential short and long-term sequelae, such as hemolytic or febrile reactions, transmission of infectious agents, and iron overload. The introduction of recombinant human erythropoietin (rhEPO) introduced a new option for the treatment of cancer associated anemia. The purpose of this review is to discuss the unique features of anemia in children and to summarize the current clinical data on the role of rhEPO in the treatment or prevention of cancer-related anemia in pediatrics.

Anemia Pathophysiology Anemia occurs as a result of blood loss, or impaired production or accelerated destruction of red cells. Cancer related anemia is multifactorial and

636

C. Hastings and J. Feusner

often has both acute and chronic components (MacMillan and Freedman 1998; Groopman and Itri 1999; Sobrero et al. 2001; Hockenberry et al. 2002). Impaired production results from bone marrow replacement by tumor cells, suppression of erythropoiesis related to chemotherapy or radiation treatment, iron deﬁciency and impaired use of iron stores, and inadequate production of erythropoietin (Cazzola 2000; Stasi et al. 2003). Blood loss may be due to hemorrhage facilitated by concomitant thrombocytopenia, mucositis/enteritis, or repetitive blood sampling. Hemolysis can be due to infection, side effects of chemotherapy, or present as part of the underlying disease. Less frequent causes of anemia in children with cancer include infections, which may have a signiﬁcant impact on blood counts and suppress marrow production of red blood cells. Infection with parvovirus B19, the cause of erythema infectiosum, has been reported in children with acute leukemia and following bone marrow transplantation (Koch et al. 1990; Rao et al. 1990; Corbett et al. 1995; Cohen et al. 1997). Persistence of infection secondary to other viruses, such as cytomegalovirus (CMV) and Epstein-Barr virus (EBV) have also been implicated in chronic anemia or pancytopenia (Alpert and Fleisher 1984; Adachi et al. 1995; Almeida-Porada and Ascensao 1996; Kaptan et al. 2001; Hagihara et al. 2003). The severity of anemia in children is inﬂuenced by several factors including the diagnosis, degree of metastatic disease or bone marrow inﬁltration, and intensity and length of the chemotherapy regimen (MacMillan and Freedman 1998). Prolonged treatment with chemotherapy may lead to progressive impairment of red cell production resulting in an extended period of hematologic recovery. Currently, there is a trend towards more intensive treatment protocols utilizing stem cell rescue in high risk patients. These intensive therapies are likely to result in a higher incidence or greater degree of anemia as compared to conventional treatments. Platinum based therapies may further exacerbate anemia secondary to the toxic effects on renal cells (Lelieveld et al. 1984; Rothmann et al. 1985).

Signs and symptoms of anemia in children At the presentation of leukemia or a malignancy with signiﬁcant marrow involvement, the child often is anemic. The anemia is usually normochromic and normocytic with a low reticulocyte count. A slow decline in erythrocyte production typically leads to an insidious onset with relatively few signs or symptoms such as pallor and gradual onset of fatigue. These signs and symptoms of anemia are related to the degree of reduction in oxygen carrying capacity of the blood, change in blood volume, rate at which these changes occur, and the ability of the cardiovascular and hematopoietic systems to compensate. One study found that anemic pediatric patients function in a

Recombinant human erythropoietin in pediatric oncology

637

state of relative oxygen deﬁcit but are able to adapt physiologically to function in this state (Grant et al. 2003). The clinical signs of anemia are often related to its severity, rapidity of onset, and underlying malignancy. General signs and symptoms of anemia in childhood can include poor feeding, loss of appetite, headaches, dizziness, fatigue, vertigo, tinnitus, dyspnea, irritability, faintness, inactivity, loss of concentration, change in behavior, and poor school performance (Cunningham 2003; Mock and Olsen 2003). It is uncertain what risk anemia poses to the child or adolescent with cancer or at which level of hemoglobin it is best to offer an intervention. Children are potentially at risk for acute and long-term effects of severe or moderate anemia. The acute effects are usually evident in the history and physical examination of the patient. Patients may experience transient cognitive dysfunction including decreased mental alertness, poor concentration and memory problems. The long-term effects of chronic anemia in young patients with cancer are not known but may include neuro-cognitive sequelae (Hockenberry-Eaton and Hinds 2000). Children receiving immune suppressive therapies for extended periods may experience chronic anemia with little opportunity between cycles of chemotherapy or radiation to fully recover. Young children may have an adverse impact on growth and development. Fatigue is a commonly unrecognized and under-treated complication of anemia (Cazzola 2000; Mock and Olsen 2003). An increased collective recognition of fatigue, and the possible implications on quality of life in cancer patients, has now turned its attention to children and adolescents (Hockenberry-Eaton and Hinds 2000). Fatigue is frequently present at diagnosis. Both disease- and treatment-related factors are contributory to its development (Sobrero et al. 2001). In adults undergoing cancer treatment, fatigue was identiﬁed as inﬂuencing their life more than pain (Vogelzang et al. 1997). Fatigue affected their ability to participate in social activities, maintain interpersonal relationships, and carry out typical cognitive tasks. It also had a remarkably negative impact on employment and ﬁnancial status. Cancer patients are often tired at rest and may have a decreased capability to carry out activities of daily living with slow physical recovery from tasks. Concentration may be impaired and patients may have a reduced efﬁciency in responding to stimuli. Chronic pathological fatigue may result in asthenia. Chronic fatigue is often not alleviated by rest and can itself become a very stressful condition (Stasi et al. 2003). Maintaining a higher hemoglobin concentration during chemotherapy results in a better quality of life and possibly survival in adults (Fanucchi et al. 1997; Estrin et al. 1999; Hockenberry-Eaton and Hinds 2000; Caro et al. 2001; Mock and Olsen 2003; Van-Steenkiste 2003; Knight et al. 2004). Quality of life measures typically take into consideration effect of fatigue on physical, functional, emotional, and social well-being of the patient (Cella 1997). General scores for quality of life, fatigue, and sensations of physical and

638

C. Hastings and J. Feusner

functional well-being are signiﬁcantly higher amongst patients with hemoglobin levels above 12 g/dL in these studies. Other interventions for the treatment of fatigue include exercise, optimal nutrition, and psychosocial interventions (Mock and Olsen 2003). Contributory factors of fatigue in children and adolescents with cancer include loss of muscle mass, medications, immobility, altered sleep patterns, anxiety, and depression (HockenberryEaton and Hinds 2000). Maintaining a higher hemoglobin (9 to 11 g/dL) in these children and adolescents for an extended period of time might ameliorate the symptoms of anemia and maintain optimal quality of life for children. Thus, it seems the tradition of leaving anemia untreated even at moderate levels may compromise the patient’s overall well-being. These clinical manifestations of anemia may compound the consequences of the malignancy and its treatment, adding to the deterioration in the quality of life for these young patients, as well as complicating effective therapy. Therefore, assessment of the impact of fatigue on quality of life should be considered when deciding on treatment for anemia or a transfusion threshold for an individual patient (Bosanquet and Tolley 2003).

Erythropoietins Erythropoietin is a 14–39 kDalton sialoglycoprotein primarily produced in the cortical region of the kidney. Its gene is located on chromosome 7. EPO production is intimately linked to hypoxia, and its receptors are found primarily on erythroblasts and CFU-E cells. The main effect of EPO is induction of proliferation and terminal differentiation of red cell progenitors. In anemic adult cancer patients, rhEPO has been well tolerated and effective in reducing red cell transfusion requirements, increasing hemoglobin levels, and improving overall quality of life (Abels 1993; Glaspy 1997; Demetri et al. 1998; Gabrilove et al. 2001; Yount et al. 2002; Littlewood et al. 2003). Though the beneﬁts of rhEPO treatment in adult anemic cancer patients is well proven, is rhEPO a treatment option for cancer or chemotherapy associated anemia in children? An evidence-based review of published reports of rhEPO use in children with cancer was conducted and is summarized in Table 1. Six reports are of randomized controlled trials (Bennetts et al. 1995; Porter et al. 1996; Csaki et al. 1998; Ragni et al. 1998; Cappelli et al. 2002; Buyukpamukcu et al. 2002), and of these only 3 have been published as peer reviewed manuscripts (Porter et al. 1996; Csaki et al. 1998; Buyukpamukcu et al. 2002). Five additional open phase I/II single institution trials are also summarized (Beck and Beck 1995; Bolonaki et al. 1996; Leon et al. 1998; Kronberger et al. 2002; Yilmaz et al. 2004). Many variabilities were evident in the analyzed reports, so no attempt at meta-analysis was made. Hence, the studies demonstrate the efﬁcacy and

Na

111

Razzouk et al.

600 U/kg weekly 16 weeks

17

Büyükpamukçu et al.

HM + ST

22b

Ragni et al.

150 U/Kg tiw IV/SC escalating dose to 300 U/kg (median dose 198 U/kg) 16-week course 150 U/kg tiw SC 16 weeks 150 U/kg tiw SC 8-week course

Various

10

Porter et al.

150 U/kg tiw IV/SC 4 months of chemotherapy

rhEPO/Dose Timing//Duration

ST

ST (Sarcoma)

19

HM (ALL)

Diagnosis

Bennetts et al.

Randomized, controlled

Study

Placebo n = 111

No rhEPO n = 17

No rhEPO n = 60b

Placebo n = 10

No rhEPO n = 18

Control

Table 1. Overall assessment of rhEPO treatment in children with cancer

Mean Hb nadir with no rhEPO and before rhEPO treatment signiﬁcantly decreased vs. rhEPO. Mean Hb increase in rhEPO group, not in control group (p = 0.027). More patients in control group required transfusion (p = 0.008). Increase in Hb favoring rhEPO group (p = 0.012). RhEpo group was transfusion free more than control group (36% vs 23%, p = 0.038)

No efﬁcacy difference overall in number or amount of RBC transfusions, but signiﬁcant difference vs. no rhEPO in low-risk ALL group. Signiﬁcant decrease in median number of RBC units transfused, median RBC amount transfused, and median number of platelet transfusions for patients who received rhEPO vs. placebo.

Results Recombinant human erythropoietin in pediatric oncology 639

Na

(232 total in study)

8

13

15

Study

Henze et al.

Csáki et al.

Cappelli et al.

Open Beck

Table 1. Continued

HM + ST

ST

ST

Various

Diagnosis

25 U/kg to 100 U/kg QD × 14 days Initial dose IV, then SC

150 U/kg tiw SC for 16 weeks

600 U/kg or 900 U/kg Weekly for 20 weeks 150 U/kg tiw SC 12 weeks or 3 chemotherapy cycles (median dose 155 U/kg)

rhEPO/Dose Timing//Duration

–

No rhEPO n = 13

No rhEPO n=7

No rhEPO

Control

No effect of rhEPO, maximum tolerated dose not attained.

Overall transfusion rates similar: rhEPO vs controls (62% vs 69%, [p = .32]). However, transfusion rate in ALL rhEPO less (66% vs 89% in ALL controls [p = 0.02]). Increase in Hb in rhEPO group by week 8 higher than in control group (p < 0.05). Response rate higher in rhEPO group (Hb increase >2 g/dL by week 12) compared to control group, p < 0.05). Trend towards reduction in transfusions in group (p = NS). Reduction in mean number of transfusions per chemotherapy cycle (0.21 versus 0.81 in rhEPO and control groups, respectively [p = 0.0008]). Mean Hb nadir lower in control group (p < 0.0001).

Results

640 C. Hastings and J. Feusner

37

41

Kronberger et al.

Yilmaz et al.

53.5% of patients had response to rhEPO. RBC transfusion requirements decreased by 33% with rhEPO. No signiﬁcant effect seen on platelet transfusion requirement. 72% of patients had response to rhEPO. Hb increased from 9.8 to 12.4 g/dL in rhEPO group and from 9.5 to 9.6 gm/dL in controls. RBC transfusions given in 15% of the rhEPO group vs 96% in the control group. 62.2% of treatment group received RBC transfusion vs 89.2% in control group. Platelet transfusion requirement signiﬁcantly lower in treatment group. Mean Hb level in rhEPO group higher than control by second month (p < .05). RBC transfusion requirement decreased 27% for all rhEPO treated patients. HM* includes acute myelogenous leukemia RBC = red blood cell

No rhEPO n = 30 HM + ST

No rhEPO n = 37

No rhEPO (patients are own controls) No rhEPO (historical controls)

HM = hematologic malignancies ST = solid tumors

150 U/kg tiw Hgb ≥ 12 SC/IV or 300 tiw ≤16 28-week course Randomized 150 U/kg or 250 U/kg tiw SC 12-week course

150 U/kg 5 days/week SC 12-week course

150, 250, 400 U/kg tiw SC 8 weeks per dose group

ALL, acute lymphocytic leukemia; RBC, red blood cells; Hb, hemoglobin; tiw, three times a week; IV, intravenous; SC, subcutaneous

b

HM* + ST

ST

ST

HM + ST

Patients who received rhEPO. Courses, not patients.

25

Leon et al.

a

15

Bolonaki et al.

Recombinant human erythropoietin in pediatric oncology 641

642

C. Hastings and J. Feusner

safety proﬁles of rhEPO in a variety of circumstances and treatment regimens. The types of malignancies in the eleven studies reviewed include acute leukemia (Beck and Beck 1995), acute lymphocytic leukemia (ALL) (Bennetts et al. 1995; Bolonaki et al. 1996; Yilmaz et al. 2004), acute myelogenous leukemia (Yilmaz et al. 2004), non-Hodgkin lymphoma (NHL) (Beck and Beck 1995; Bolonaki et al. 1996; Yilmaz et al. 2004), central nervous system (CNS) tumors (Bolonaki et al. 1996; Leon et al. 1998; Buyukpamukcu et al. 2002; Cappelli et al. 2002; Yilmaz et al. 2004) and several types of non-CNS solid tumors (Beck and Beck 1995; Bolonaki et al. 1996; Porter et al. 1996; Csaki et al. 1998; Buyukpamukcu et al. 2002; Cappelli et al. 2002; Kronberger et al. 2002; Yilmaz et al. 2004). Hemoglobin (Hb) level required to be eligible for rhEPO treatment ranged from “before anemia” (Kronberger et al. 2002) to <10.5 g/dL (Beck and Beck 1995; Leon et al. 1998), <10.0 g/dL (Buyukpamukcu et al. 2002; Yilmaz et al. 2004) or <7.5 g/dL (Beck and Beck 1995). In one study, patients were required to have Hb levels under the third percentile for sex and age (Bolonaki et al. 1996) and three reports did not state a speciﬁc Hb level required for rhEPO (Bennetts et al. 1995; Ragni et al. 1998; Cappelli et al. 2002). Dosage of rhEPO administered varied from 25 to 400 IU/kg/dose, although most trials initiated treatment at 150 IU/kg/dose. The schedule of administration was daily (Beck and Beck 1995) or three times per week (tiw) (Beck and Beck 1995; Bennetts et al. 1995; Bolonaki et al. 1996; Porter et al. 1996; Ragni et al. 1998; Buyukpamukcu et al. 2002; Kronberger et al. 2002; Yilmaz et al. 2004). The duration of treatment varied between 14 days (Beck and Beck 1995) and 11 months (Kronberger et al. 2002). Routes of administration were subcutaneous (Bolonaki et al. 1996; Csaki et al. 1998; Leon et al. 1998; Ragni et al. 1998; Buyukpamukcu et al. 2002; Cappelli et al. 2002; Yilmaz et al. 2004), subcutaneous or intravenous (Beck and Beck 1995; Bennetts et al. 1995; Porter et al. 1996), or were not stated (Kronberger et al. 2002). The studies were also screened for data reporting on the inﬂuence of iron supplementation, especially in patients with iron deﬁciency that would be expected to impact on rhEPO treatment. Iron supplementation was given in eight studies (Bolonaki et al. 1996; Porter et al. 1996; Csaki et al. 1998; Leon et al. 1998; Ragni et al. 1998; Cappelli et al. 2002; Kronberger et al. 2002), and was not permitted in one study (Buyukpamukcu et al. 2002). Use was not speciﬁcally indicated in two studies (Bennetts et al. 1995; Beck and Beck 1995). The Hb level that triggered red blood cell (RBC) transfusion varied between <9 g/dL (Bennetts et al. 1995) and <6 g/dL (Beck and Beck 1995). In four studies, manifestation of anemia symptoms could also trigger transfusion (Beck and Beck 1995; Porter et al. 1996; Csaki et al. 1998; Yilmaz et al. 2004). Bolonaki et al. administered RBC transfusions when Hb levels were <7 g/dL or a febrile infection was observed (Bolonaki et al. 1996). Yilmaz et al. transfused patients when Hb levels were <7 g/dL or the patient was symptomatic of anemia (Yilmaz et al. 2004). Büyükpamuçku et al.

Recombinant human erythropoietin in pediatric oncology

643

decided on a Hb threshold of transfusion below 6.0 g/dL (Buyukpamukcu et al. 2002), Kronberger et al. used a cutoff of 6.5 g/dL for red cell transfusion (Kronberger et al. 2002), and Csáki et al. reported using a Hb less than 8.0 g/dL (Csaki et al. 1998). Four reports did not state any guidelines for the use of RBC transfusions (Ragni et al. 1998; Buyukpamukcu et al. 2002; Cappelli et al. 2002; Kronberger et al. 2002). Overall assessment of rhEPO effects Table 1 summarizes the overall rhEPO experience in the studies assessed. Eight studies reported a beneﬁt with rhEPO treatment (Porter et al. 1996; Csaki et al. 1998; Leon et al. 1998; Ragni et al. 1998; Buyukpamukcu et al. 2002; Cappelli et al. 2002; Kronberger et al. 2002); ﬁve of these trials were randomized and controlled, and differences between treatment groups were signiﬁcant (Porter et al. 1996; Csaki et al. 1998; Ragni et al. 1998; Buyukpamukcu et al. 2002; Cappelli et al. 2002). One randomized, controlled trial did not report signiﬁcant differences in overall response to rhEPO in patients with ALL, although treatment in rhEPO was signiﬁcantly more effective in patients with low-risk ALL (Bennetts et al. 1995). One open study demonstrated increased Hb levels and reticulocyte counts during rhEPO administration, but no statistics are reported (Bolonaki et al. 1996). Another open study assessed the safety and efﬁcacy of rhEPO, and a response to rhEPO, deﬁned as an increase in Hb of ≥2 g/dL, was achieved. Overall, an increase in Hb and a reduction in transfusion requirements was demonstrated (Leon et al. 1998). Yet another study concluded that rhEPO was not effective (Beck and Beck 1995). The eleven clinical trials are reviewed here in more detail: six randomized, controlled trials (Bennetts et al. 1995; Porter et al. 1996; Csaki et al. 1998; Ragni et al. 1998; Buyukpamukcu et al. 2002; Cappelli et al. 2002) and ﬁve open, non-comparative trials (Beck and Beck 1995; Bolonaki et al. 1996; Porter et al. 1996; Leon et al. 1998; Kronberger et al. 2002). Additionally, two studies investigating darbopoietin in children are also summarized. rhEPO in children with ALL Bennetts et al. conducted a randomized controlled study in which thirty-seven children (18 girls, 19 boys) with newly diagnosed ALL were randomized to receive rhEPO (epoetin alfa) 150 IU/kg/dose tiw intravenously or subcutaneously (n = 19; median age 3.5 years), or no rhEPO (n = 18; median age 3.9 years) during three courses of chemotherapy (Bennetts et al. 1995). There was equal distribution between groups for patients with average and low-risk ALL. Transfusions were considered for both groups if Hb levels decreased to <7.5 g/dL.

644

C. Hastings and J. Feusner

Total amounts of packed RBCs (pRBCs) administered were 27 ± 18 cc/kg in the rhEPO group vs. 35 ± 5 cc/kg in the control group (p = 0.11). Per-patient amounts of RBC (pRBC) administered were 2.21 ± 1.58 for the rhEPO group vs. 3.06 ± 1.69 for the control group (p = 0.39). In a subanalysis of the lowrisk ALL group (number of patients not stated), there was a signiﬁcantly lower volume of pRBC transfused in the rhEPO group (16.8 ± 12.7 cc/kg) vs. the control group (69.5 ± 36.1 cc/kg) (p = 0.02). No increase in toxicity was reported in the patients who received rhEPO. Iron deﬁciency developed in 5/19 (26.3%) patients who received rhEPO, and 3/18 (16.7%) patients in the control group (p = NS). The investigators concluded that rhEPO was safe in children with average- and low-risk ALL undergoing induction, consolidation, and delayed chemotherapy intensiﬁcation, and of signiﬁcant beneﬁt in the children with low-risk ALL.

Erythropoietin in children with solid tumors Porter et al. conducted a randomized, controlled study in children with malignant sarcomas investigating the effects of escalating doses of rhEPO (epoetin alfa) (Porter et al. 1996). Patients were eligible for this study if they had a Hb level below 10.5 g/dL, and their anemia was not related to blood loss, hemolysis, or vitamin deﬁciency. Twenty-four patients (13 girls, 11 boys; median age 14 years) were enrolled and randomized to receive either rhEPO for 16 weeks during chemotherapy or placebo. Administration of study drug was 150 IU/kg intravenously or subcutaneously three times weekly. If after 4 weeks of treatment, the patient required blood transfusion or did not maintain a Hb level above 11.5 g/dL, the dosage was increased by 50 U/kg. The rhEPO dose was adjusted every 4 weeks to maintain the Hb above 11.5 g/dL, to a maximum rhEPO dose of 300 U/kg. Study drug was withheld if Hb level exceeded 16.5 g/dL until the level decreased to 11.5 g/dL. Ferrous sulfate 6 mg/kg/day in divided doses was administered, but discontinued if serum ferritin concentration exceeded 1,000 ng/ml. Twenty patients, 10 in each group, were evaluable. All but one required dose escalation, and the median rhEPO dose administered during this study was 198 U/kg tiw. Nine patients who received rhEPO and 10 who received placebo required RBC transfusion. The median number of units transfused in the rhEPO group was signiﬁcantly fewer: 4.5 compared to 13 (p = 0.01). Furthermore, the median amount transfused was signiﬁcantly less for patients in the rhEPO group compared to the placebo group (23 ml/kg versus 80 ml/kg, p = 0.02). Mean Hb level at the time of transfusion was 8.1 g/dL in both groups. Three patients in the rhEPO group received platelet transfusions vs. nine patients in the placebo group. The median number of platelet units transfused was signiﬁcantly fewer for patients who received rhEPO compared with

Recombinant human erythropoietin in pediatric oncology

645

patients who received placebo (0 versus 4, p = 0.005). At the end of the double-blind portion of the study, the patients continued to receive rhEPO, which was also administered to seven patients from the placebo group. Eighty-seven percent of patients who continued to receive rhEPO had further decreases in RBC transfusion requirements. rhEPO was well tolerated, and the investigators concluded that use of rhEPO (epoetin alfa) in children deserves further study. Büyükpamukçu et al. (Buyukpamukcu et al. 2002) conducted an 8 week randomized trial to evaluate safety and efﬁcacy of rhEPO (epoetin alfa) in the treatment of chemotherapy-induced anemia in children with solid tumors, both receiving platinum- and non-platinum-based therapies. The 17 study patients received rhEPO 150 IU/kg thrice weekly subcutaneously, while the 17 control patients received standard of care only. Hemoglobin levels had to drop below 10 g/dL prior to study entry. No iron supplementation was given. Transfusions were administered if the Hb values dropped below 6 g/dL. The intensity of therapies was compared between control and study groups by evaluating absolute neutrophil counts and platelet counts periodically; there was no signiﬁcant difference between the groups. EPO levels were measured prior to study entry and though no difference between the groups was apparent at that time, the EPO levels at the end of study were signiﬁcantly lower in the control group. Mean Hb levels rose in study patients from a mean of 8.5 g/dL to 10.21 g/dL after 2 months, while control patients remained stable (8.48 g/dL to 8.41 g/dL) (p = 0.027). The number of transfusions between the groups during the 8-week study period was statistically signiﬁcant with only 1 transfusion in the study group and 8 in the control group (p = 0.008). One patient developed hypertension after 2 weeks of treatment with rhEPO. The blood pressure returned to a normal range after the drug was discontinued. In general, rhEPO was well tolerated and led to an increase in Hb levels and minimized RBC transfusions for children with solid tumors receiving chemotherapy (and some radiation therapy). Ragni et al. (1998) evaluated the efﬁcacy of rhEPO (epoetin alfa) in children who received chemotherapy for a variety of tumor types, and also explored the effects of rhEPO on Hb recovery following chemotherapy in a randomized comparative study. Data are reported in terms of courses of chemotherapy, not numbers of patients. The control group received no rhEPO. The study group received rhEPO 150 IU/kg/dose subcutaneously three times weekly for 16 weeks, with iron supplementation after a mean of 4.7 courses of chemotherapy. Twenty-two courses of chemotherapy plus rhEPO were evaluated. The mean Hb nadir was 10.36 g/dL (range: 7.7–13.8 g/dL). The Hb decreased to less than 9 g/dL during four courses (18.2%) and the mean time to Hb recovery was 3.5 days (range: 3–5 days). Sixty courses of chemotherapy were analyzed in the control group. The mean Hb nadir in patients who did not receive

646

C. Hastings and J. Feusner

rhEPO was 8.7 g/dL (range: 5.5–13.5 g/dL), much lower than in the rhEPO treated patients (p < 0.05). In 36 courses (60%), Hb levels decreased to <9 g/dL, and the mean time to Hb recovery was 7.3 days (range: 3–23 days). Data was further analyzed with respect to Hb nadirs and transfusion requirements during rhEPO administration versus these parameters for the mean 4.7 courses of chemotherapy that were administered before rhEPO was initiated in these same patients. Mean Hb nadirs before rhEPO was initiated (8.7 g/dL; range: 6.9–10.5 g/dL) were shown to be signiﬁcantly (p < 0.05) lower than Hb nadirs during treatment. Hemoglobin levels decreased to less than 9 g/dL in 16 (53.3%) courses of chemotherapy. Red cell and platelet transfusion requirements were signiﬁcantly less in the rhEPO treated patients than the nontreated control group (18% versus 70%, p < 0.05). rhEPO was well tolerated, and the investigators concluded that rhEPO seemed to be effective in reducing chemotherapy-induced anemia and decreasing transfusion needs in this group of pediatric cancer patients. Csáki et al. conducted a prospective, randomized trial of rhEPO in children with solid tumors undergoing cyclic combination chemotherapy (Csaki et al. 1998). Of 15 evaluable patients, eight were randomized to the rhEPO arm and received 150 IU/kg/dose three times a week for a minimum of 12 weeks or 3 cycles of chemotherapy. All rhEPO-treated patients received iron supplementation for the duration of the therapy (ferrous sulfate 50–300 mg/day orally depending on body weight). The dose of rhEPO remained the same for Hb levels between 11 and 13 g/dL. If the Hb level dropped below 11 g/dL, the individual dose could be increased by 50 IU/kg/dose. When the Hb level exceeded 14 g/dL, the dose could be withdrawn. Actual doses of rhEPO given ranged between 123 and 230 IU/kg, with an average dose of 161 IU/kg/patient and a median of 155 IU/kg/patient. Transfusions were given whenever the Hb concentration fell below 8 g/dL or at any Hb level if the patient displayed symptoms of hypoxia. The patients treated with rhEPO compared to the control patients had signiﬁcantly higher Hb concentrations by week 8 (13.11 ± 1.13 g/dL versus 11.06 ± 1.35 g/dL, p < 0.05). There was a statistically signiﬁcant difference in pre-cycle Hb levels between cycles 3 and 4 between control and rhEPO treated groups (p < 0.05). Response to rhEPO was deﬁned as an increase of he patient’s initial Hb level by 1 g/dL or by 2 g/dL by week 12, unrelated to the administration of a transfusion. Response was seen in 2 of the 8 study patients by week 4, and 6 of the 8 patients by week 12. In the control group, only 1 of 7 showed a spontaneous increase in Hb by week 12. The response rate was signiﬁcantly higher in the rhEPO group as compared with the control group by week 12 (p < 0.05). However, transfusion requirements were not signiﬁcantly different between the two groups. Serum iron and ferritin levels were lower in the rhEPO group during the entire study period despite supplementation in the treated group and none in the control group. Inter-

Recombinant human erythropoietin in pediatric oncology

647

estingly, the authors also comment on a trend toward an increased performance status in the treated group, deﬁned as a lesser mean decrease in body weight as compared to the control group. Cappelli et al. randomized 26 children receiving therapy for solid tumors, including medulloblastoma (Cappelli et al. 2002). The 13 patients in the treatment group received rhEPO (epoetin alfa) at a dose of 150 IU/kg subcutaneously three times a week for 16 weeks. Oral iron was given if the serum ferritin level dropped below 100 ng/ml. All patients (13 control and 13 treated) received the same chemotherapy and growth factors except for rhEPO. Signiﬁcant differences in the number of RBC transfusions per cycle and total number of transfusions were found in this study. The control group received a total of 34 blood transfusions compared to 9 in the treated group, and the mean number of transfusions per cycle was 0.21 versus 0.81 in the treated and control groups, respectively (p = 0.0008). Additionally, the mean Hb nadir was signiﬁcantly different between the treated (9.64 g/dL) and control (8.26 g/dL) groups (p < 0.0008). No signiﬁcant difference was observed in the number of platelet transfusions between the two groups. No adverse effects of rhEPO were observed during the entire course of therapy.

Results of open-labeled, non-randomized trials of rhEPO in children Five reports of open studies were analyzed (Beck and Beck 1995; Bolonaki et al. 1996; Leon et al. 1998; Kronberger et al. 2002; Yilmaz et al. 2004); three included patients with hematologic and solid tumors (Beck and Beck 1995; Bolonaki et al. 1996; Yilmaz et al. 2004) and two focused on children with solid tumors (Leon et al. 1998; Kronberger et al. 2002). The publication by Beck was the only study that did not conclude that administration of rhEPO was beneﬁcial (Beck and Beck 1995). This phase I/II trial administered 18 courses of rhEPO (epoetin alfa) to 15 patients with hematologic or solid tumors, and Hb levels less than 7.5 g/dL. Patients were divided into cohorts of three, and rhEPO was administered at doses of 25, 50, 70, 80, 90, and 100 IU/kg. The ﬁrst dose was administered intravenously with subsequent doses administered subcutaneously daily for 14 days. Response criteria were: increased Hb to greater than 11.0 g/dL, necessity for blood transfusion within 24 hours if the Hb decreased to less than 6.0 g/dL or symptoms of anemia developed, or in the event of a major adverse event. No adverse events other than local burning sensation at the site of rhEPO injections were reported by Beck, and Hb levels did not increase to >11.0 g/dL (Beck and Beck 1995). The Hb level increased or was stable in seven courses (39%) and no transfusions were given. Transfusions were given in 11 courses (61%) with no evidence of a dose-response relationship with rhEPO.

648

C. Hastings and J. Feusner

The investigators concluded that rhEPO was safe, but ineffective in these children, probably due to a mechanism of transient primary resistance. Subsequent trials, however, have shown that children who receive weekly doses of at least 450 IU/kg per week for at least 8 weeks do respond to rhEPO, beginning by week 4 (Varan et al. 1999). In the Beck study (Beck and Beck 1995), only 12 children received at least this much rhEPO, and treatment was continued for only 14 days. In addition, baseline Hb levels for study inclusion were very low (7.5 g/dL), only 1.5 g/dL greater than the protocolmandated transfusion trigger of 6.0 g/dL. In a second open study conducted by Bolonaki et al., 15 children were enrolled in a clinical trial that compared change in Hb level and transfusion requirements during rhEPO (epoetin alfa) (Bolonaki et al. 1996). The patients served as their own controls during a similar phase of chemotherapy when they did not receive rhEPO. Study eligibility included a Hb level under the third percentile for sex and age. Patients had a variety of malignancies including hematologic malignancies (n = 6) and solid tumors (n = 9). rhEPO 150 IU/kg/dose subcutaneously three times weekly was administered for 8 weeks; dosage was increased to 250 IU/kg if the Hb failed to improve, and again to 400 IU/kg, if no response occurred at the 250 IU/kg dose. Oral iron 5mg/kg/day was administered concomitantly. Response was deﬁned as an increase in the Hb level to above the tenth percentile following 8 weeks of treatment. Blood transfusions were given for Hb < 7.0 g/dL or a febrile infection. No differences in response were noted between patients with hematologic and solid tumors. Eight of the 15 children (three hematologic and ﬁve solid tumors; 53.3%) responded to rhEPO after 8 weeks of treatment with 150 IU/kg/dose. Hemoglobin increases were noted during the ﬁrst 2 weeks of rhEPO treatment, with increases most pronounced after 6 weeks (hematologic cases) or 8 weeks (solid tumor cases) of administration. These Hb changes are shown in Fig. 1. These results include four patients who received cisplatin for a CNS tumor. Seven children required an increased dose of rhEPO to 250 IU/kg before the Hb level increased, and three children, all with hematologic malignancies, required 400 IU/kg to increase the Hb level. All patients required RBC transfusion prior to therapy, and this requirement decreased to 66% in children with either tumor type. Baseline serum EPO levels showed that children who responded had serum EPO levels signiﬁcantly lower (p < 0.008) than patients who did not respond. No adverse events were noted in association with the administration of rhEPO, and the treatment was well tolerated. The investigators concluded that the efﬁcacy of rhEPO may be especially prominent in patients with low serum EPO levels before therapy, and an increase in the Hb level within 2 weeks of starting therapy. Leon et al. conducted an open study in 25 children under the age of 18 with solid tumors (Leon et al. 1998). Patients were diagnosed with Ewing

Recombinant human erythropoietin in pediatric oncology HEMATOLOGIC MALIGNANCIES

649 (a)

15 CONTROLS rhEPO ADMINISTRATION

14

Hb (gr/dL)

13 12 11 10 9 8 7

0

2

4

6

8

WEEKS SOLID TUMORS

(b)

15 CONTROLS rhEPO ADMINISTRATION

14

Hb (gr/dL)

13 12 11 10 9 8 7

0

2

4

6

8

WEEKS

Fig. 1. (a) Hemoglobin (Hb) mean value (±SE) of children with hematologic malignancies and (b) solid tumors before and after rhEPO administration (Adopted from Bolonaki I et al. (1996) Treatment with recombinant human erythropoietin in clindren with malignancies. Pediatr Hematol Oncol 13(2): 111–21)

sarcoma (36%), osteosarcoma (32%), CNS tumors (16%), Hodgkin lymphoma (8%), rhabdomyosarcoma (4%), and tumors of unknown site (4%). The children received cyclic combination chemotherapy for a total of at least 5 days every 3–4 weeks. Further inclusion criteria were: life expectancy for at least 3 months, clinical stability for at least 1 month, Hb level <10.5 g/dL, and stool negative for occult blood. Patients were compared to matched historical control patients (n = 25). Children in the rhEPO group were admin-

650

C. Hastings and J. Feusner

istered 150 IU/kg/dose of rhEPO, subcutaneously, ﬁve times a week for 12 consecutive weeks. Oral iron supplementation was administered if serum ferritin fell to <100 ng/ml and/or transferrin saturation fell to <20%. RBC transfusions were performed if the Hb levels fell to <9 g/dL and/or symptoms of anemia developed. In the rhEPO group, 18 patients (72%) responded to the treatment. Response was deﬁned as an increase in the Hb of ≥2 g/dL without requiring RBC transfusions. Most responders achieved the increase in the Hb between the 4th and 8th week of therapy. Only four patients (16%) in the rhEPO group required RBC transfusions during the course of the study, compared to 24 (96%) in the control group. The mean baseline values of Karnofsky index were 70.8% in the rhEPO group vs. 71.2% in the control group. A statistically signiﬁcant (p < 0.05) improvement in Karnofsky index between start and end of the study was observed in the rhEPO group, whereas no such improvement was observed in the control group. Fourteen (56%) of patients in the rhEPO group required iron supplementation. Overall, rhEPO was well tolerated and no severe adverse events were reported. The authors concluded that rhEPO is effective and safe in increasing Hb levels and reducing the need for RBC transfusions. Kronberger et al. (2002) conducted a single-center, case-control, openlabel safety and efﬁcacy study of rhEPO (epoetin alfa) in pediatric cancer patients with solid tumors treated with both cisplatin and non-cisplatin containing regimens. Thirty-seven patients received either 150 IU/kg three times a week (if Hb ≥ 12 g/dL) or 300 IU/kg three times a week (if Hb ≤ 12 g/dL) for 28 weeks. The medication was administered either intravenously or subcutaneously. There were 37 control patients who did not receive rhEPO during the course of their therapy. Ten patients were reported in a previous pilot study (Kronberger et al. 1995), and 27 more patients were enrolled subsequently. Patients also were administered iron 5 mg/kg/day while on rhEPO. The study and control patients were comparable with respect to baseline Hb values and variability of the reticulocyte, platelet, and leukocyte counts during the study period. RBC transfusion requirements were signiﬁcantly lower in the treated group compared to the control group (62.2% versus 89.2%, p = 0.007). Interestingly, the platelet transfusion requirements in the study group were also signiﬁcantly lower as compared to the control group (35.1% versus 64.9%, p = 0.010). The authors note that by the median time on study (15 weeks) 57% of rhEPO patients had not yet received a ﬁrst platelet transfusion, versus only 16% of the control patients. Mean Hb levels at baseline and end of study were 10.8 g/dL and 10.1 g/dL, respectively, in the treated group, and 11.1 g/dL and 9.3 g/dL, respectively, in the control group. There was no signiﬁcant difference in mean Hb levels between the cisplatin and non-cisplatin treated patients, in both treatment and control groups. Overall, rhEPO was well tolerated, with one patient developing myocarditis and two with deep

Recombinant human erythropoietin in pediatric oncology

651

venous thromboses, both believed to be related to the underlying neoplasm (osteogenic sarcoma). In a more recently reported study, Yilmaz et al. (Yilmaz et al. 2004) evaluated 41 children with either hematologic malignancies (n = 27) or solid tumors (n = 14). There were 30 age-matched control patients (19 with hematologic malignancy and 11 with solid tumors). Patients were entered on study when their Hb levels dropped to 10 g/dL or lower. Exclusion criteria included symptomatic anemia, hypertension, gastrointestinal bleeding, or deﬁciencies in folate, iron, or vitamin B12. The study patients were randomized between 150 U/kg/dose versus 250 U/kg/dose of rhEPO (epoetin alfa) to be given subcutaneously three times a week for 12 weeks. If patients had a ferritin level of less than 400 ng/dL, they were given iron, either orally or intravenously. Patients were transfused red cells if the Hb level dropped below 7 g/dL. Hemoglobin levels and transfusion requirements during the ﬁrst 3 months prior to starting were compared to the same parameters during the 3 months on study. The Hb levels of the whole treatment group increased monthly after the start of rhEPO, whereas the trend in the control group was a diminishing mean Hb level. Mean Hb levels of the rhEPO-treated patients were found to be signiﬁcantly higher than controls at the second (p = 0.008) and third months (p = 0.0001). Similarly, by the third month of treatment, rhEPO patients required fewer red cell transfusions (p = 0.001 for hematologic malignancies and p = 0.01 for solid tumors). None of the 20 patients who received the higher dose of 250 U/kg of rhEPO required transfusion with RBC beyond the second month of treatment. At the 150 U/kg dose level, only 3 of 13 patients with hematologic malignancies and none of the solid tumor patients (n = 8) required RBC transfusions after the second month of treatment. It was noted that the patients with hematologic malignancies who had a Hb less than 8 g/dL seemed to respond better to the higher rhEPO dose, but the number of patients was small, precluding analysis for statistical signiﬁcance. In this study, a beneﬁt was seen in the patients receiving rhEPO for both hematologic malignancies and solid tumors, with respect to increased Hb levels and decreased transfusion requirements, as compared to a control group. The authors concluded rhEPO was safe and effective. There may be an advantage to a higher dosing of rhEPO in patients with hematologic malignancies, but the number of patients studied was very small.

Summary of data of randomized, controlled and open-labeled trials of erythropoietin These reports have included a mixture of patients with solid tumors or hematologic malignancies (predominantly ALL), and most commonly utilized rhEPO at 150 IU/kg three times per week, administered subcutaneously.

652

C. Hastings and J. Feusner

Variability was noted in dosing, with one study (Porter et al. 1996) allowing an escalation for non-responders at four weeks, and one study (Yilmaz et al. 2004) randomizing between 150 IU/kg and 250 IU/kg. Yet another study allowed for dose variability related to the Hb level (Csaki et al. 1998). Two of the peer reviewed trials showed a reduction in RBC transfusions for the rhEPO treated patients (4.5 median versus 13 in one trial; 5.9% pts transfused versus 47.1% in the other) (Porter et al. 1996; Buyukpamukcu et al. 2002). In addition, the latter trial (Buyukpamukcu et al. 2002) showed a signiﬁcant increase in Hb in rhEPO treated patients from study onset to study end, 2 months later: 1.7 g/dL increase versus 0.07 g/dL decrease (p = 0.027). The deﬁciencies in the published data (only 6 properly randomized and controlled trials; and variable transfusion triggers, iron use, and routes of rhEPO administration) led to the design and recent completion of two very large prospective randomized trials – one in the US and the other in Europe. They both evaluated the effect of larger doses of rhEPO (600 to 900 IU/kg) given once weekly. These have not been published yet, but have been presented in abstract form (Henze et al. 2002; Razzouk et al. 2004) and conﬁrm a beneﬁt to rhEPO use, and show that the novel schedule of 600–900 IU/kg given once weekly is safe. In the US trial (Razzouk et al. 2004) involving 222 children (111 each for patients and controls), there was a signiﬁcant reduction in RBC cell transfusions in the treated patients, from 69.4% of the control patients after the ﬁrst 4 weeks of study to 51.4% of the rhEPO-treated patients (p = 0.008). The European study (Henze et al. 2002) showed a reduction in transfusions for patients with ALL (89% versus 66%; p = 0.02), but no signiﬁcant difference in Hb from baseline to last value at 20 weeks (1.7 g/dL increase in rhEpo patients versus 1.3 g/dL increase in the controls). Curiously, they did not ﬁnd a difference in transfusion rate in the non-ALL patients (n = 147 patients plus controls), but did ﬁnd an increase in Hb (data not shown) – just the opposite of what they found for the ALL subgroup. And, ﬁnally, there was a quality of life component to the US trial, which demonstrated some beneﬁt, but only in those patients who achieved at least a 2 g/dL increase in Hb after 4 weeks of rhEpo. The results of this study have recently been published (Razzouk et al. 2006).

Special circumstances Jehovah’s witnesses The religious beliefs and practices of Jehovah’s Witnesses forbid the ingestion of blood per the literal interpretation of the bible. Accepting blood and blood products in any form or any method, such as intravenously, are

Recombinant human erythropoietin in pediatric oncology

653

traditionally-prohibited. This prohibition can prove to be challenging in a situation in which a child may receive life-saving therapies that lead to severe anemia. This situation can compromise the child’s quality of life and/or put the child at risk for preventable morbidity or mortality. The courts have upheld the rights of adult Jehovah’s Witnesses to refuse life-saving blood transfusions. However, the rights of parents to refuse this therapy for their child has been challenged by the legal system. When the medical caregivers feel the child’s life is in danger, a court order has frequently been issued to administer the life-saving therapy. Many families have found this acceptable. Pediatric cancer patients belonging to Jehovah’s Witnesses have been successfully treated with chemotherapy. A restricted transfusion regimen and the broad application of hematopoietic supportive care measures may also reduce transfusion needs (Rothmann et al. 1985; Tenenbaum et al. 2004). This may be limited by the diagnosis and intensity of therapy. Stimulation of erythropoiesis with agents such as rhEPO should be considered and instituted if medically applicable (Johnson et al. 1991). Also, consideration should be made to establishing a lower transfusion threshold and minimizing blood draws, dependent, of course, on the clinical status. While treating any patient with a malignancy involves a balance between treating the disease and avoiding treatment toxicities, this balance becomes much more tenuous for patients who refuse blood products (Aguilera 1993; Cothren et al. 2002; Penson and Amrein 2004).

Radiation Anemia may inﬂuence response to radiation. Tissue oxygenation is thought to play an important role in the effectiveness of radiation in causing tumor cell death (Stuben et al. 2003). Hypoxia has long been felt to be a clinically signiﬁcant barrier to cure with radiation (Knight et al. 2004). A beneﬁt has been seen with higher Hb levels and outcomes in patients with cancer of the head, neck, larynx, and cervix (Dische et al. 1983; Knight et al. 2004). In a randomized, controlled study, rhEPO was shown to possibly inﬂuence local control in adult head and neck cancer patients who experienced a rapid rise in their Hb levels (Henke et al. 1999). However, there was not a non-therapy control group in this study and the results comparing differing dose levels were not statistically signiﬁcant. Limited data is available in children. One brief report evaluated the effect of Hb on radiotherapy response in children with medulloblastoma and did not ﬁnd a correlation to local control or survival (Chow et al. 1999). There is current interest in performing randomized trials of rhEPO to determine if conclusive evidence exists for such an effect and if this will improve outcomes in those undergoing radiotherapy for treatment of childhood cancer.

654

C. Hastings and J. Feusner

End of life management of anemia Patients receiving palliative care may develop anemia related to chronic disease, marrow invasion, blood loss, or toxicity of therapy. Anemia is quite prevalent in the palliative care setting in adults (Dunn et al. 2003). A recent report on end of life issues in children found that patients and families reported fatigue as the most common symptom contributing to suffering (Wolfe et al. 2000). The etiology of fatigue is multifactorial and includes natural progression of the disease, depression, poor nutritional status, and anemia. Certainly no effective treatment is available for some of these factors, but anemia can be alleviated by transfusion or possibly erythropoietin. Though ample data is available to demonstrate the beneﬁts of transfusion in actively treated oncology patients, little is known on the beneﬁts of transfusion or treatment with rhEPO in the palliative setting. Children with advanced cancer are likely to be transfusion dependent and may beneﬁt from higher Hb levels with improved emotional, physical, and cognitive well-being. However, anemia is not thought to be painful in the dying child and its treatment may prove to cause difﬁculty in the last few days when a more important goal may be to minimize intervention or prolongation of life. rhEPO may provide a means to offer less invasive therapy and increase time outside the hospital setting.

Conclusions RhEPO appears to be a safe and effective treatment for anemia in pediatric cancer patients. Randomized, controlled clinical trials and open trials have shown that the use of rhEPO results in increased Hb levels and decreased transfusion requirements in some patient subgroups. However, these observations are based on limited clinical data. More clinical trials in children with cancer are needed, and several are currently underway. At the current time, critical analysis of the best available data supports use of rhEPO for solid tumor patients in a non-transplant setting. Although the increments of Hb gained are modest, there is good evidence for avoidance of some RBC transfusions, and for the ﬁrst time, a suggestion of an improvement in quality of life (Razzouk et al. 2004). An interesting study of predictors of need for RBC transfusions in pediatric solid tumor patients was recently published (Albitar et al. 1995). The investigators determined via multivariate analyses that performance status, Hb <12 g/dL and absolute lymphocyte count <70 l/uL pre treatment all predicted need for transfusion within 31 days of completing a chemotherapy cycle. If this ﬁnding is conﬁrmed by others it might help focus the use of rhEPO to those who need it the most.

Recombinant human erythropoietin in pediatric oncology

655

The role of rhEPO in pediatric ALL is not settled. The two largest trials appear to have completely opposite ﬁndings, which needs to be explained. Two possible reasons for these differing ﬁndings are the route of rhEPO administration, and the use (or not) of supplemental iron. The published literature implies an equipotency of rhEPO whether given intravenously or subcutaneously, but this has not been studied scientiﬁcally and, in fact, some evidence exists in the renal dialysis literature that the subcutaneous route is nearly twice as efﬁcient (Glaspy and Cavill 1999; Knauss et al. 2002). If rhEPO is to be best utilized, serum iron and TIBC should be monitored to detect relative iron deﬁciency that may limit the efﬁcacy of the cytokine (VanSteenkiste 2003). RhEPO has not been evaluated extensively in AML, due to concerns about possible stimulation of leukemic cells (Takeshita et al. 2000). A newer rhEPO congener, darbepoietin, has been synthesized. It has two additional carbohydrate binding sites and gives the compound a longer halflife, by 2-3-fold over standard rhEPO. A number of studies in adults suggest it has similar action as rhEPO, at less frequent dosing intervals (Hedenus et al. 2002; Kotasek et al. 2003; Leyland-Jones 2003). A recent study was conducted in children receiving multicycle chemotherapy for nonmyeloid malignancies (Blumer et al. 2005). This phase I, multicenter, open-label, uncontrolled trial was designed to assess the pharmacokinetic proﬁle and tolerability of darbopoietin alfa administered once weekly at a dose of 2.25 μg/kg subcutaneously. Sixteen patients were enrolled on the study, all of whom were being treated for solid tumors. The drug was well tolerated with the most common adverse effect being localized discomfort at the injection site. The pharmacokinetic proﬁle indicated no signiﬁcant accumulation of darbopoetin alfa and a long terminal half-life due to the slow absorption and clearance phases of the drug. The mean Hb increased from 9.4 g/dL at study entry to 10.7 g/dL after 8 weeks of therapy. A ﬁnal note should be made of two very recent trials of rhEPO in adults that demonstrated an increased rate of adverse events in the rhEPO treated patients. One was a trial in women with metastatic breast cancer that was stopped early due to an increased mortality rate in the patients receiving rhEPO (Leyland-Jones 2003); the second was a trial in head and neck cancer patients with an increase in disease recurrence in the rhEPO treated patients (Henke et al. 2003). In the ﬁrst study, mainly nonanemic cancer patients were treated with rhEPO to maintain Hb levels between 12–14 g/dL and in the second study, target Hb levels to be achieved with rhEPO were ≥14 g/dL in women and ≥15 g/dL in men. As a result, the majority of patients treated with rhEPO in both studies achieved Hb levels higher than recommended by evidence-based guidelines of the American Society of Clinical Oncology and the American Society of Hematology (12 g/dL) (Rizzo et al. 2002). In a recent meta-analysis of 27 randomized studies in anemic adult patients, in which 19 studies were included in the overall survival analysis, there was no evidence for shortened survival in rhEPO-treated subjects (Bohlius et al. 2005). In

656

C. Hastings and J. Feusner

pediatric cancer patients, there has been no report of increased relapse rates in any study, but the number of reported children treated with rhEPO is relatively small. For the time being, it is prudent not to treat oncology patients to Hb levels >12 g/dL, until the outcome of further studies in adults is determined.

Issues for further study Most clinical investigators who have evaluated the use of rhEPO in children with cancer have concluded that more data are necessary. There are numerous issues to be addressed in future clinical trials with children with cancer and anemia including optimal dosing, frequency and route of administration, effect of Hb level achieved on radiation efﬁcacy, and quality of life issues. The traditional dosing regimen of rhEPO is three times weekly, but a study in adult cancer patients (Gabrilove et al. 2001) and one in pediatric patients (Razzouk et al. 2006) reported that a once-per-week administration effectively increased ﬁnal Hb levels. It will be interesting to investigate if such a schedule, or even less frequent, is practical in children. In addition, the route of administration may affect dosing compliance. Another important issue for future studies is the administration of iron supplementation. Should iron be given routinely to all patients or only if indicated because of iron insufﬁciency; and what is the best route of administration? A major emphasis on rhEPO in adult patients is the improvement of quality of life (QOL) (Glaspy 1997; Demetri et al. 1998; Gabrilove et al. 2001; Littlewood et al. 2001; Demetri et al. 2002). To date, there are minimal data concerning QOL in children. This is an important outcome to address. Just as adults, children also suffer fatigue, impaired mental function, respiratory distress, and cardiac decompensation as a result of anemia. Minimizing these symptoms and signs, and maintaining optimum QOL for children with cancer is of great importance. Their mental and emotional, as well as physical, wellbeing must be sustained.

References 1. Abels R (1993) Erythropoietin for anaemia in cancer patients. Eur J Cancer 29A [Suppl 2]: S2–S8 2. Adachi N, et al (1995) Fatal cytomegalovirus myocarditis in a seronegative ALL patient. Acta Paediatr Jpn 37(2): 211–216 3. Aguilera P (1993) [Blood transfusions in Jehovah’s witnesses]. Rev Med Chil 121(4): 447–451 4. Albitar S, et al (1995) Subcutaneous versus intravenous administration of erythropoietin improves its efﬁciency for the treatment of anaemia in haemodialysis patients. Nephrol Dial Transplant 10 [Suppl 6]: 40–43

Recombinant human erythropoietin in pediatric oncology

657

5. Almeida-Porada GD, Ascensao JL (1996) Cytomegalovirus as a cause of pancytopenia. Leuk Lymphoma 21(3–4): 217–223 6. Alpert G, Fleisher GR (1984) Complications of infection with Epstein-Barr virus during childhood: a study of children admitted to the hospital. Pediatr Infect Dis 3(4): 304–307 7. Beck MN, Beck D (1995) Recombinant erythropoietin in acute chemotherapyinduced anemia of children with cancer. Med Pediatr Oncol 25(1): 17–21 8. Bennetts G, Bertolone S, Bray G, et al (1995) Erythropoietin reduces volume of red cell transfusions required in some subsets of children with acute lymphocytic leukemia [abstract]. Blood 86 [Suppl 1]: 853a 9. Blumer J, BS, Adamson PC, et al (2005) Darbopoetin alfo for the treatment of anemia in Pediatric patients with non myeloid malignancies receiving chemotherapy: A phase 1, open-label pharmacokinetic study. In American Society of Pediatric Hematology and Oncology. Washington D.C 10. Bohlius J, Langensiepen S, Schwarzer G, et al (2005) Recombinant human erythropoietin and overall survival in cancer patients: results of a comprehensive meta-analysis. J Natl Cancer Inst 97: 489–498 11. Bolonaki I, et al (1996) Treatment with recombinant human erythropoietin in children with malignancies. Pediatr Hematol Oncol 13(2): 111–121 12. Borsi JD, Ferencz T, Csaki C (1995) Tranfusion requirements of children with cancer and the use of recombinant erythropoietin for the prevention and treatment of cytostatics induced anemia in children. Can J Infect Dis 6 [Suppl C]: 235C (abstract) 13. Bosanquet N, Tolley K (2003) Treatment of anaemia in cancer patients: implications for supportive care in the National Health Service Cancer Plan. Curr Med Res Opin 19(7): 643–650 14. Buyukpamukcu M, et al (2002) Is epoetin alfa a treatment option for chemotherapyrelated anemia in children? Med Pediatr Oncol 39(4): 455–458 15. Cappelli C, Ragni G, Clerico A (2002) Recombinant human erythropoietin in pediatric oncology. In: Nowrousian M (ed) Recombinant human erythropoietin (rhEPO) in clinical oncology – scientiﬁc and clinical aspects of anemia in Cancer. Springer, Wien New York, pp 313–323 16. Caro JJ, et al (2001) Anemia as an independent prognostic factor for survival in patients with cancer: a systemic, quantitative review. Cancer 91(12): 2214–2221 17. Cazzola M (2000) Mechanisms of anaemia in patients with malignancy: implications for the clinical use of recombinant human erythropoietin. Med Oncol 17 [Suppl 1]: S11–S16 18. Cella D (1997) The Functional Assessment of Cancer Therapy-Anemia (FACTAn) Scale: a new tool for the assessment of outcomes in cancer anemia and fatigue. Semin Hematol 34 [3 Suppl 2]: 13–19 19. Chow E, et al (1999) Effect of hemoglobin on radiotherapy response in children with medulloblastoma: should patients with a low hemoglobin be transfused? Med Pediatr Oncol 32(5): 395–397 20. Cohen BJ, et al (1997) Chronic anemia due to parvovirus B19 infection in a bone marrow transplant patient after platelet transfusion. Transfusion 37(9): 947–952 21. Corbett TJ, et al (1995) Successful treatment of parvovirus B19 infection and red cell aplasia occurring after an allogeneic bone marrow transplant. Bone Marrow Transplant 16(5): 711–713

658

C. Hastings and J. Feusner

22. Cothren C, et al (2002) Blood substitute and erythropoietin therapy in a severely injured Jehovah’s witness. N Engl J Med 346(14): 1097–1098 23. Csaki C, et al (1998) Recombinant human erythropoietin in the prevention of chemotherapy-induced anaemia in children with malignant solid tumours. Eur J Cancer 34(3): 364–367 24. Cunningham RS (2003) Anemia in the oncology patient: cognitive function and cancer. Cancer Nurs 26 [6 Suppl]: 38S–42S 25. Demetri GD, et al (1998) Quality-of-life beneﬁt in chemotherapy patients treated with epoetin alfa is independent of disease response or tumor type: results from a prospective community oncology study. Procrit Study Group. J Clin Oncol 16(10): 3412–3425 26. Demetri GD, et al (2002) Beneﬁts of epoetin alfa in anemic breast cancer patients receiving chemotherapy. Clin Breast Cancer 3(1): 45–51 27. Dische S, et al (1983) Carcinoma of the cervix–anaemia, radiotherapy and hyperbaric oxygen. Br J Radiol 56(664): 251–255 28. Dunn A, Carter J, Carter H (2003) Anemia at the end of life: prevalence, significance, and causes in patients receiving palliative care. J Pain Symptom Manage 26(6): 1132–1139 29. Estrin JT, et al (1999) A retrospective review of blood transfusions in cancer patients with anemia. Oncologist 4(4): 318–324 30. Fanucchi M, et al (1997) Effects of polyethylene glycol-conjugated recombinant human megakaryocyte growth and development factor on platelet counts after chemotherapy for lung cancer. N Engl J Med 336(6): 404–409 31. Gabrilove JL (2001) Hematologic malignancies: an opportunity for targeted drug therapy. Oncologist 6 [Suppl 5]: 1–3 32. Gabrilove JL, et al (2001) Clinical evaluation of once-weekly dosing of epoetin alfa in chemotherapy patients: improvements in hemoglobin and quality of life are similar to three-times-weekly dosing. J Clin Oncol 19(11): 2875–2882 33. Glaspy J (1997) The impact of epoetin alfa on quality of life during cancer chemotherapy: a fresh look at an old problem. Semin Hematol 34 [3 Suppl 2]: 20–26 34. Glaspy J, Cavill I (1999) Role of iron in optimizing responses of anemic cancer patients to erythropoietin. Oncology (Huntingt) 13(4): 461–473; discussion 477–478, 483–488 35. Grant MJ, Huether SE, Witte MK (2003) Effect of red blood cell transfusion on oxygen consumption in the anemic pediatric patient. Pediatr Crit Care Med 4(4): 459–464 36. Green DM, et al (1998) Comparison between single-dose and divided-dose administration of dactinomycin and doxorubicin for patients with Wilms’ tumor: a report from the National Wilms’ Tumor Study Group. J Clin Oncol 16(1): 237–245 37. Groopman JE, Itri LM (1999) Chemotherapy-induced anemia in adults: incidence and treatment. J Natl Cancer Inst 91(19): 1616–1634 38. Hagihara M, et al (2003) Clinical effects of infusing anti-Epstein-Barr virus (EBV)-speciﬁc cytotoxic T-lymphocytes into patients with severe chronic active EBV infection. Int J Hematol 78(1): 62–68 39. Hedenus M, et al (2002) Randomized, dose-ﬁnding study of darbepoetin alfa in anaemic patients with lymphoproliferative malignancies. Br J Haematol 119(1): 79–86

Recombinant human erythropoietin in pediatric oncology

659

40. Henke M, et al (1999) Erythropoietin for patients undergoing radiotherapy: a pilot study. Radiother Oncol 50(2): 185–190 41. Henke M, et al (2003) Erythropoietin to treat head and neck cancer patients with anaemia undergoing radiotherapy: randomised, double-blind, placebo-controlled trial. Lancet 362(9392): 1255–1260 42. Henze G, Michon J, Morland B, et al (2002) Phase III randomized study: efﬁcacy of epoetin alfa in reducing blood transfusions in newly diagnosed pediatric cancer patients receiving chemotherapy. Proc Am Soc Clin Oncol. 21: 387a 43. Hockenberry MJ, et al (2002) Incidence of anemia in children with solid tumors or Hodgkin disease. J Pediatr Hematol Oncol 24(1): 35–37 44. Hockenberry-Eaton M, Hinds PS (2000) Fatigue in children and adolescents with cancer: evolution of a program of study. Semin Oncol Nurs 16(4): 261–272; discussion 272–278 45. Johnson PW, et al (1991) The use of erythropoietin in a Jehovah’s Witness undergoing major surgery and chemotherapy. Br J Cancer 63(3): 476 46. Kaptan K, et al (2001) Successful treatment of severe aplastic anemia associated with human parvovirus B19 and Epstein-Barr virus in a healthy subject with alloBMT. Am J Hematol 67(4): 252–255 47. Knauss MD, Walton T, Macon EJ (2002) Switching from i.v. to s.c. epoetin in hemodialysis patients. Am J Health Syst Pharm 59(18): 1783–1784 48. Knight K, Wade S, Balducci L (2004) Prevalence and outcomes of anemia in cancer: a systematic review of the literature. Am J Med 116 [Suppl 7A]: 11S–26S 49. Koch WC, et al (1990) Manifestations and treatment of human parvovirus B19 infection in immunocompromised patients. J Pediatr 116(3): 355–359 50. Kotasek D, et al (2003) Darbepoetin alfa administered every 3 weeks alleviates anaemia in patients with solid tumours receiving chemotherapy; results of a double-blind, placebo-controlled, randomised study. Eur J Cancer 39(14): 2026–2034 51. Kronberger M, et al (2002) Reduction in transfusion requirements with early epoetin alfa treatment in pediatric patients with solid tumors: a case-control study. Pediatr Hematol Oncol 19(2): 95–105 52. Kronberger M, KB, Zoubek A, et al (1995) Prevention of anemia with r-HuEPO in children with Ewing’s or osteogenic sarcoma. Proc Am Soc Clin Oncol 14: 702 53. Lelieveld P, et al (1984) Preclinical studies on toxicity, antitumour activity and pharmacokinetics of cisplatin and three recently developed derivatives. Eur J Cancer Clin Oncol 20(8): 1087–1104 54. Leon P, et al (1998) Recombinant human erythropoietin for the treatment of anemia in children with solid malignant tumors. Med Pediatr Oncol 30(2): 110–116 55. Leyland-Jones B (2003) Breast cancer trial with erythropoietin terminated unexpectedly. Lancet Oncol 4(8): 459–460 56. Littlewood TJ, et al (2001) Effects of epoetin alfa on hematologic parameters and quality of life in cancer patients receiving nonplatinum chemotherapy: results of a randomized, double-blind, placebo-controlled trial. J Clin Oncol 19(11): 2865–2874 57. Littlewood TJ, et al (2003) Epoetin alfa corrects anemia and improves quality of life in patients with hematologic malignancies receiving non-platinum chemotherapy. Hematol Oncol 21(4): 169–180

660

C. Hastings and J. Feusner

58. MacMillan ML, Freedman MH (1998) Recombinant human erythropoietin in children with cancer. J Pediatr Hematol Oncol 20(3): 187–189 59. Michon J (2002) Incidence of anemia in pediatric cancer patients in Europe: results of a large, international survey. Med Pediatr Oncol 39(4): 448–450 60. Mock V, Olsen M (2003) Current management of fatigue and anemia in patients with cancer. Semin Oncol Nurs 19 [4 Suppl 2]: 36–41 61. Nachman J, et al (1998) Response of children with high-risk acute lymphoblastic leukemia treated with and without cranial irradiation: a report from the Children’s Cancer Group. J Clin Oncol 16(3): 920–930 62. Penson RT, Amrein PC (2004) Faith and freedom: leukemia in Jehovah Witness minors. Onkologie 27(2): 126–128 63. Porter JC, et al (1996) Recombinant human erythropoietin reduces the need for erythrocyte and platelet transfusions in pediatric patients with sarcoma: a randomized, double-blind, placebo-controlled trial. J Pediatr 129(5): 656–660 64. Ragni G, Clerico A, Sordi A, et al (1998) Recombinant human erythropoietin (rHuEpo) in children with cancer: A randomized study [abstract]. Med Pediatr Oncol 31: 274 65. Rao SP, Miller ST, Cohen BJ (1990) Severe anemia due to B19 parvovirus infection in children with acute leukemia in remission. Am J Pediatr Hematol Oncol 12(2): 194–197 66. Razzouk BI, Hord JD, Hockenberry M, et al (2006) Double-blind, placebocontrolled study of quality of life, hematologic end points, and safety of weekly epoetin alfa in children with cancer receiving myelosuppressive chemotherapy. J Clin Oncol 24: 3583–3589 67. Rizzo JD, Lichtin AE, Woolf SH et al (2002) Use of epoetin in patients with cancer: evidence-based clinical practice guidelines of the American Society of Clinical Oncology and the American Society of Hematology. Blood 100: 2303–2320 68. Rothmann SA, et al (1985) Effect of cis-diamminedichloroplatinum on erythropoietin production and hematopoietic progenitor cells. Int J Cell Cloning 3(6): 415–423 69. Sobrero A, et al (2001) Fatigue: a main component of anemia symptomatology. Semin Oncol 28 [2 Suppl 8]: 15–18 70. Stasi R, et al (2003) Cancer-related fatigue: evolving concepts in evaluation and treatment. Cancer 98(9): 1786–1801 71. Stuben G, et al (2003) Impact of anemia prevention by recombinant human erythropoietin on the sensitivity of xenografted glioblastomas to fractionated irradiation. Strahlenther Onkol 179(9): 620–625 72. Takeshita A, et al (2000) Quantitative expression of erythropoietin receptor (EPO-R) on acute leukaemia cells: relationships between the amount of EPO-R and CD phenotypes, in vitro proliferative response, the amount of other cytokine receptors and clinical prognosis. Japan Adult Leukaemia Study Group. Br J Haematol 108(1): 55–63 73. Tenenbaum T, et al (2004) Oncological management of pediatric cancer patients belonging to Jehovah’s Witnesses: a two-institutional experience report. Onkologie 27(2): 131–137 74. Van-Steenkiste J (2003) Pharmacotherapy of chemotherapy-induced anaemia. Expert Opin Pharmacother 4(12): 2221–2227

Recombinant human erythropoietin in pediatric oncology

661

75. Varan A, et al (1999) Recombinant human erythropoietin treatment for chemotherapy-related anemia in children. Pediatrics 103(2): E16 76. Vogelzang NJ, et al (1997) Patient, caregiver, and oncologist perceptions of cancer-related fatigue: results of a tripart assessment survey. The Fatigue Coalition. Semin Hematol 34 [3 Suppl 2]: 4–12 77. Wolfe J, et al (2000) Symptoms and suffering at the end of life in children with cancer. N Engl J Med 342(5): 326–333 78. Yilmaz D, et al (2004) A single institutional experience: is epoetin alpha effective in anemic children with cancer? Pediatr Hematol Oncol 21(1): 1–8 79. Yount S, Lai JS, Cella D (2002) Methods and progress in assessing the quality of life effects of supportive care with erythropoietin therapy. Curr Opin Hematol 9(3): 234–240 Correspondence: Caroline Hastings, MD, Director of Fellowship, Pediatric Hematology and Oncology, Children’s Hospital and Research Center Oakland, 747 52nd Street, Oakland CA 94609, USA, E-mail: [email protected]

Chapter 25

rhEPO in surgical oncology M. J. Fontaine and L. T. Goodnough Department of Pathology, Stanford University, Stanford, California, USA

Introduction Anemia is common in cancer patients and is often considered merely secondary to the underlying malignancy. Anemia associated with cancer varies widely with stage of disease and/or treatment and in general will develop slowly with hemoglobin values ranging from 8 to 10 g/dL (Mercuriali and Inghilleri 2002). But because anemia has an independent impact on length and quality of life (QOL) (Nissenson et al. 2003), anemia in cancer patients has been more closely scrutinized in order to improve its evaluation and treatment. Blood transfusion appears to overcome the negative prognostic effects of low presenting hemoglobin levels (Grogan et al. 1999), but is often associated with a higher risk of infection. Blood transfusions remain particularly helpful in the context of either severe anemia (Hb < 8.0 g/dL) or lifethreatening anemia (Hb < 6.5 g/dL), particularly when the condition is aggravated by bleeding as a result of a surgical intervention. This chapter, focused on the use of human recombinant erythropoietin (rhEPO) in surgical cancer patients, will ﬁrst attempt to review the transfusion requirements of cancer patients undergoing surgery and review the risks associated with allogeneic blood transfusion in these patients. Secondly, we will discuss the alternative use of rhEPO to treat perioperative anemia in cancer patients.

Transfusion requirements in cancer patients undergoing surgery Surgical intervention in cancer patients may worsen the anemia to a critical point where RBC transfusions are required. As deﬁned by Mercuriali et al., surgical transfusion requirements depend on several factors as follows: 1) the lowest level of hematocrit tolerated by the patient’s cardiovascular condition, 2) the volume of perioperative blood loss, itself dependent on the accuracy of surgical hemostasis and on coagulation monitoring (Mercuriali and Inghilleri 2002).

664

M. J. Fontaine and L. T. Goodnough

A “mathematical approach” (Mercuriali and Inghilleri 1996) to predict the transfusion need of an individual patient candidate to a speciﬁc surgical operation has been recently proposed. This is based on the two parameters that affect transfusion requirement, i.e. the perioperative blood loss (predicted loss) on the one hand and the quantity of blood that a speciﬁc patient, for his/her physiological and medical status, can tolerate to lose before blood transfusion support is indicated (tolerated loss). The volume of blood the patient probably will lose in the perioperative period can be calculated through a constantly updated analysis of the real perioperative blood loss that occurred in each patient undergoing a speciﬁc surgical operation performed by a speciﬁc surgical team. This can be obtained performing a retrospective analysis of the patients operated during the last 6–12 months. The surgical RBCs loss occurring in each patient is given by the circulating RBCs volume reduction from presurgery to a properly determined postoperative time, plus the volume of RBCs transfused during this period (Table 1). The volume of RBCs that a patient can tolerate to lose can be calculated knowing the baseline RBCs mass and the minimal Hct value that the patient, according to his clinical condition, can tolerate in the postoperative period (Table 1). The treatment of anemia in surgical cancer patients may vary from replacement of intravascular volume with crystalloids for patients with anemia resulting from acute blood loss and, if symptoms persist despite

Table 1. Mathematical formulas to deﬁne the perioperative RBCs loss, the tolerated blood loss and the perioperative transfusion need Perioperative RBC loss (L of RBCs) = Circulating RBC volume (C-RBCs-V) reduction (from presurgery to postoperative day 5) plus the RBC volume transfused; (C-RBCs-Vpresurgery − C-RBCs-Vday 5 postop) + volume of RBC transfused, C-RBC-V (in L of RBC) = Predicted Blood Volume (PBV) × Hct; PBV = female = 0.3561 × height + 0.0338 × weight (kg) + 0.1833 male = 0.3669 × height (m)Ÿ3 + 0.03219 × weight (kg) + 0.6041 Consequently: Perioperative RBC loss = PBV (Hctpresurgery − Hctday 5 postop) + liters of RBC transfused Tolerated Blood Loss = Volume of RBCs loss to reach an accepted minimal Hct value Tolerated Blood Loss (L of RBC) = (PBV × Hctbaseline) − (PBV × Hctmin. accepted) being: Hctminimal acceptable = minimal Hct value compatible with the clinical condition of the patient. Transfusion needs = Predicted RBC loss − tolerated RBC loss consequently = Predicted RBCs loss − (PBV × Hctbaseline − PBV × Hctmin. accepted)

rhEPO in surgical oncology

665

volume replacement therapy, patients should receive RBC transfusion (Groopman et al. 1999). The percentage of surgical cancer patients requiring RBC transfusion ranges between 45 to 84% (Blumberg and Heal 1994), as described mostly in patients with gastrointestinal malignancies, compared to 20 to 50% for nonsurgical cancer patients (Abels et al. 1991). Mercuriali has also observed that 30% of the patients undergoing cancer-related surgery had a low baseline hemoglobin below 11 g/dL (Mercuriali and Inghilleri 2002); thus, preoperative autologous blood donation (PABD) is often not feasible when hemoglobin is less than 11 g/dL, according to the standards from the American Association of Blood Banks (AABB 2005). Furthermore, perioperative blood salvage is contraindicated during cancer surgical resection due to contaminating tumor cells in the salvaged blood which reinfused could promote metastatic disease. Lastly, acute normovolemic hemodilution is only indicated in patients with optimal cardiovascular conditions and preoperative hematocrit values higher than 42–45%, which condition is rarely encountered in surgical cancer patients (Gillon et al. 1996). Therefore, the likelihood of allogeneic RBC transfusion during the perioperative management of surgical cancer patients is higher than for other surgical patients.

Risks associated with allogeneic blood transfusion in cancer patients Whether allogeneic blood exposure causes clinically signiﬁcant immune suppression in cancer patients remains a subject of debate. A number of observational, retrospective reports has described an association between exposure to allogeneic blood and either earlier recurrences of malignancy or increased rates of postoperative infection (Bordin et al. 1994). Only a few prospective studies of this issue have been performed in order to clarify the potential immunomodulatory effects of allogeneic blood transfusion (Vamvakas 1996; Vamvakas et al. 2001). A study of 120 patients undergoing curative resection of colorectal carcinoma failed to demonstrate a difference in relapse-free survival time or a difference in the prevalence of serious postoperative infection between patients randomized to allogeneic or autologous transfusion (Heiss et al. 1994). In another study of 423 patients, there was no difference in relapse-free survival time or in infectious complications when comparing allogeneic versus autologous red cell transfusions (Busch et al. 1993). Houbiers et al. compared transfusion of leukocyte-depleted (a three log10 reduction) components with buffy coat-depleted components (a one log10 reduction) and found no difference in cancer recurrence risk after colorectal surgery (Houbiers et al. 1994). More recently, a risk-factor multivariate analysis for the development of intrahepatic recurrence after hepatectomy for hepatocellular carcinoma (HCC) showed that young age, solitary and large HCC, high hepatitis activity, and large amount of intraoperative blood loss and blood transfusion were high risk factors for recurrence of HCC (Hanazaki et al.

666

M. J. Fontaine and L. T. Goodnough

2005). In a parallel study the inﬂuence of perioperative blood transfusion on survival and recurrence after curative hepatic resection for HCC was evaluated; in 210 patients, for whom a curative hepatectomy was performed and for whom 57% received perioperative blood transfusion, a multivariate analysis revealed that perioperative blood transfusion was an independent predictor for recurrence in patients with low serum albumin level (<3.5 g/dL) (Hanazaki et al. 2005). In a retrospective analysis of the time of tumor recurrence in 123 patients undergoing esophagectomy for thoracic esophageal cancer, there was a signiﬁcant increase in disease-free survival prior to recurrence in 18 patients who received autologous blood compared to 23 patients who received allogeneic blood while undergoing surgical esophagectomy; the clinicopathological factors that inﬂuenced prognosis were similar in the two groups (Motoyama et al. 2004). However, Vamvakas (1996) examined the available evidence from the observational cohort studies that investigated the hypothesis that allogeneic blood transfusion provokes cancer recurrence and/or postoperative infection. They concluded that the randomized control trials provided no indication that perioperative allogeneic blood transfusion causes an increase in cancer recurrence and/or postoperative infections (Busch et al. 1993; Houbiers et al. 1994). The rate of transfusion-transmitted viral diseases is the lowest it has been in the history of blood transfusion (Goodnough et al. 2003). The residual risk of infectious diseases such as viral hepatitis and HIV transmission through blood transfusion has been reduced mostly since nucleic acid testing has become the viral screening method of choice in most blood centers (Jackson et al. 2003). On the other hand, transfusion-associated bacterial sepsis remains the second most frequently reported cause of transfusion-related fatalities in the United States; the risk of bacterial contamination of blood products remains particularly high for platelet products, which are stored at room temperature and, therefore, vulnerable to bacterial growth. According to the latest reports an estimated one in 1,000–3,000 platelet units are contaminated with bacteria (AABB 2005). To reduce this risk, the American Association of Blood Banks (AABB) implemented a new standard on March 1, 2004 requiring accredited blood banks and transfusion services to take measures to detect and limit bacterial contamination in all platelet components. The direct impact of this measure on the safety of allogeneic transfusions remains to be fully determined, but despite the implementation of the AABB standard, false-negative results after platelet bacterial testing have recently been reported with at least two fatal cases of transfusion-associated sepsis in platelet recipients in 2004 (Arendt et al. 2005). One of the most serious outcomes of blood transfusion is now considered to be transfusion-related acute lung injury (TRALI), reported as the most common cause of fatality related to a blood transfusion (Holness et al. 2004). TRALI often remains unrecognized clinically and, more importantly, the pathophysiology of TRALI is only partially understood (Silliman et al.

rhEPO in surgical oncology

667

2005). But it is hypothesized that TRALI reaction may be the result of two cumulative events: the ﬁrst event is linked to the patient such as underlying sepsis, hematologic disease, and/or postsurgical status, and the second event is related to the transfusion of potential neutrophil primers such as inﬂammatory cytokines, active lipids and/or alloantibodies (Silliman et al. 2003). Preventive measures such as donor deferral and/or disqualiﬁcation for plasma donations are usually implemented depending on the results of the neutrophil and HLA antibody screen on the associated donors. But the lack of speciﬁcity of these preventive measures targeting blood donors has also encouraged clinicians to adhere to blood component utilization guidelines and to minimize inappropriate use (Kleinman et al. 2004).

Perioperative anemia as a risk factor for poor outcome in surgical cancer patients Anemia has been associated with poor prognosis and management of anemia can improve patient outcomes (Nissenson et al. 2003). An association has been observed between anemia and disease progression among patients undergoing radiation therapy (RT) particularly in those with cervical carcinoma or with squamous cell carcinoma of the head and neck (Harrison et al. 2000). Two thirds of women with cervical carcinoma are anemic at baseline, and 82% are anemic during RT (Harrison et al. 2000). Correlations between anemia, tumor tissue oxygenation, local recurrence, and survival have also been demonstrated in a study by Grogan et al., in which average weekly nadir hemoglobin (Hb) (AWNH) is highly predictive of outcome for patients treated with RT for carcinoma of the cervix (Grogan et al. 1999). In cases of head and neck cancer, 75% of patients undergoing combined chemotherapy and radiotherapy become anemic (with hemoglobin levels <8 g/dL) and anemia has been associated with worse local regional control and survival rates (Lee et al. 1998). The highest incidence of anemia that requires red blood cell (RBC) transfusion occurs in those patients with lymphomas, lung tumors, and gynecologic (ovarian) or genitourinary tumors requiring surgical ablation (Groopman and Itri 1999). Martin-Loeches recently reported on the prognostic implications of anemia in the outcome of patients with uterine cervix carcinoma and showed that the inﬂuence of hemoglobin is equally important as the volume of the tumor itself on the long-term survival (Martin-Loeches et al. 2003). Additionally, several recent multivariate analyses conﬁrm that long-term survival of surgical cancer patients is signiﬁcantly affected by preoperative anemia (Van Halteren et al. 2004; Shen et al. 2005; Van de Pol et al. 2005). But the high incidence of preoperative and postoperative anemia in surgical cancer patients coincides with an increase in blood utilization, which is itself associated with an increased risk for perioperative infection and adverse outcome (Dunne et al. 2002).

668

M. J. Fontaine and L. T. Goodnough

Therefore, the use of a strict transfusion strategy could help in reducing overall blood transfusion. But an alternative to transfusion, pharmacologic management of preoperative anemia with rhEPO, may stimulate erythropoiesis and help reduce the need for allogeneic blood transfusion (Monk 2004).

Management of anemia in surgical cancer patients Erythropoietin is currently approved and prescribed as rhEPO for the treatment of anemia associated with chronic renal failure and HIV infection. rhEPO is also approved for cancer patients to help reduce chemotherapyrelated anemia and to reduce allogeneic blood transfusion in cancer patients requiring surgical ablation. Cancer patients with advanced disease may require chemotherapy and/or radiation prior to surgical ablation. In a recent phase I trial in previously untreated patients with advanced head and neck carcinoma, patients received up to three cycles of paclitaxel and carboplatin with (n = 14) or without (n = 22) epoetin alfa before radiation therapy or surgery (Dunphy et al. 1997). Patients treated with epoetin alfa experienced a mean hemoglobin decrease of 0.5 g/dL during preoperative chemotherapy versus a decrease of 3.3 g of hemoglobin/dL in patients who did not receive epoetin alfa (P < .0001). In addition, fewer patients treated with epoetin alfa received RBC transfusions during preoperative chemotherapy (0% versus 18%). The results of these trials suggest that epoetin alfa can prevent chemotherapy-induced anemia and can reduce the need for RBC transfusions when administered concomitantly with chemotherapy regimens that produce a high incidence of anemia. These patients may beneﬁt from rhEPO treatment to prevent severe anemia and to decrease perioperative red cell transfusions. rhEPO treatment is considered as an effective strategy in reducing allogeneic blood transfusion requirement mostly in patients who are candidates for cardiac and orthopedic surgery. rhEPO is usually administered in combination with PABD to enhance collection of autologous blood in advance of elective surgery by reversing existing anemia and attenuating phlebotomyinduced decreases in the hemoglobin level (Goodnough et al. 1994). In many clinical studies rhEPO was effective in stimulating erythropoiesis, with a subsequent increase in the volume of RBCs produced during the course of treatment and in the number of units predeposited (Goodnough et al. 1994; Price et al. 1996; Mercuriali and Inghilleri 2002). Furthermore patients with low RBC mass and facing operations with high transfusion requirement have beneﬁted from rhEPO administration which conditioned them for autologous blood donation and signiﬁcantly reduced perioperative allogeneic blood transfusion (Mercuriali et al. 1993; Blumberg et al. 1994). Autologous blood transfusion (ABT) combined with the use of rhEPO has been investigated

rhEPO in surgical oncology

669

by Shinozuka et al. Their study combined ABT with rhEPO in 46 patients undergoing hepatectomy for hepatocarcinoma and showed a decreased exposure to allogeneic blood when compared with the control group of patients who had not predonated blood (Shinozuka et al. 2000). The efﬁcacy of rhEPO alone has been evaluated for patients undergoing major oncology ablative surgery and for whom preoperative autologous blood donation is not always feasible. These studies included patients with anemia or other disorders precluding the donation of autologous blood due to the limited preoperative time for surgery and due to individuals unwilling to participate in an autologous blood donation program. The protocols of rhEPO administration differed signiﬁcantly in these studies with the duration of treatment varying from 1 to 3 weeks and the dose interval varying from daily to weekly (De Andrade et al. 1996; Laupacis and Fergusson 1998; Stowell et al. 1999). Altogether, however, these studies showed a reduction in allogeneic transfusion rate in treated patients compared with controls. Mercuriali instituted a protocol as follows: subcutaneous (SC) administration of 100 IU/kg rhEPO daily beginning 4 days before surgery up to the second day following surgery and one bolus of 200 IU/kg rhEPO administered intravenously in association with iron (600 to 1,000 mg, according to baseline iron reserve levels). The treatment has been shown to be effective in producing a 2 to 7 points increase in hematocrit (Hct) value before surgery, with average increase in circulatory RBC mass of some 100 mL in the same period (Mercuriali et al. 1999). Mercuriali suggested that rhEPO administration together with IV iron during a preoperative period of 4–5 days is able to stimulate erythropoiesis signiﬁcantly, expand the circulatory red cell mass, and reduce the transfusion requirement in patients who, for clinical or logistic reasons (heart surgery and cancer patients) are not able to deposit autologous units prior to elective surgery (Mercuriali and Inghilleri 2002). The efﬁcacy of perisurgical rhEPO alone in reducing allogeneic transfusion requirements has been conﬁrmed by Qvist et al. in a large prospective study in anemic cancer patients (Hb < 9.5 g/dL) undergoing colorectal surgery (Qvist et al. 1999). In this study 38 patients received 300 IU/kg on day 4 before surgery and 150 IU/kg daily for the following 7 days; 43 patients received placebo. In addition, all the patients received a daily administration of 200 mg of oral iron for 4 days before surgery. Patients in the rhEPO group received a mean of 0.3 units of allogeneic blood compared to 1.6 units in the control group (Qvist et al. 1999). The use of erythropoietin (rhEPO) for the treatment of anemia associated with urological malignancies has been reviewed by Albers et al. (Albers et al. 2001). For patients with renal cell carcinoma anemia is rare, thus rhEPO treatment is usually not recommended. rhEPO treatment may be beneﬁcial to patients undergoing radical prostatectomy and/or presenting with advanced, hormone-refractory disease and similarly to patients who have to undergo radical cystectomy for bladder cancer (Albers et al. 2001).

670

M. J. Fontaine and L. T. Goodnough

Chun et al. demonstrated in a randomized controlled study on 120 patients, candidates for radical retropubic prostatectomy, that perisurgical rhEPO (600 IU/kg on day -14 and -7, preoperatively) is safe, well tolerated, and equally effective as preoperative autologous blood donation in reducing allogeneic blood transfusion requirements (Chun et al. 1997). Three recent studies have reported the efﬁcacy and safety of rhEPO administered perioperatively in patients with gastrointestinal cancer (Kosmadakis et al. 2003), with colorectal cancer (Christodoulakis et al. 2005), and with head and neck cancer (Scott et al. 2002). Kosmadakis showed that the use of rhEPO (300 IU/kg/day) perioperatively results in reduced transfusion needs, in improved hematopoiesis with signiﬁcantly increased hemoglobin, hematocrit, and reticulocyte count, and in signiﬁcantly better 1-year survival rate in treated patients compared to placebo group. Christodoulakis showed similar signiﬁcant results after treatment with the use of rhEPO (300 IU/kg/day) perioperatively, noting that a group of patients treated with 150 IU/kg did not improve their hematologic proﬁle signiﬁcantly, with no significant reduction in transfusion needs when compared to controls. Finally, Scott et al. used a three time dose a 600 IU/kg of rhEPO before surgery and showed a signiﬁcant improvement in hematologic parameters with decreased transfusion requirements in the treated group. In all three studies, iron supplementation was given to patients receiving rhEPO.

Safety of rhEPO in surgical cancer patients In cancer surgery, immunosuppression of allogeneic blood may be detrimental with increased cancer recurrence and shortened survival. There is also limited time available for an autologous donation program; so as not to delay surgery, collection has to be performed more rapidly. rhEPO therapy during autologous donation increases RBC production and facilitates the collection of autologous blood (Mercuriali et al. 1993). Other reports have indicated that surgical procedures and rhEPO therapy may suppress endogenous erythropoietin production in response to anemia (Levine et al. 1991; Tasaki et al. 1992). This may be a clinically important disadvantage of rhEPO therapy for postoperative hemoglobin recovery. But Hyllner et al. designed an intensive precollection program where three units of whole blood were collected, with or without erythropoietin therapy, in only 2 weeks before radical hysterectomy (Hyllner et al. 2005). The aim of this study was to determine whether rhEPO therapy and/or an aggressive donation schedule alter perioperative erythropoietin concentrations and whether rhEPO therapy leads to the release of the pro-inﬂammatory cytokines IL-6 and IL-8 in women undergoing radical hysterectomy and pelvic lymphadenectomy due to cervical carcinoma (Hyllner et al. 2005). Three units of whole blood were collected from each patient during the 2-

rhEPO in surgical oncology

671

weektime before the operation and patients were randomized to either a control group with no rhEPO therapy or to receive 10,000 IU of rhEPO (epoetin beta) daily for 10 days. There was no signiﬁcant difference between the two groups in postoperative serum-EPO concentrations on day 1, 5, and 35 of rhEPO treatment. Furthermore, there was no evidence of IL-6 or IL8 release during rhEPO therapy with serum cytokine levels remaining within reference values preoperatively. There was a peak in IL-6 and IL-8 concentrations in both groups one hour after the operation but there was no significant difference between the two groups. This peak was explained by the surgical tissue trauma (Ellstrom et al. 1996). These ﬁndings conﬁrmed a study by Goodnough et al. who found no evidence of an inhibitory effect of rhEPO therapy on postoperative endogenous erythropoietin levels (Goodnough et al. 1994). One of the most important key points conditioning the rhEPO effectiveness is the availability of iron for hemoglobin synthesis and, consequently, the amount and route of administration of combined iron supplementation. The importance of iron supplementation during rhEPO administration was ﬁrst demonstrated in patients with chronic renal failure (CRF), in whom a suboptimal response to rhEPO was associated with insufﬁcient iron availability (Anastassiades et al. 1993; Macdougall et al. 1996). The results of these studies suggest that the oral administration of iron in patients with CRF is not sufﬁcient to deliver an adequate amount of iron to optimize rhEPOstimulated erythropoiesis. Moreover, many studies have been published in which iron is given intravenously. In a recent study, Macdougall et al. compared three patient groups receiving rhEPO because of anemia of CRF: the ﬁrst received 250 mg of IV iron twice a week, the second received 200 mg of elemental iron daily, and the third received no iron supplement (Macdougall et al. 1996). At week 16, patients receiving IV iron had a higher hemoglobin level than the other two groups and their serum ferritin level was also higher. There was no signiﬁcant difference in hemoglobin level between the group on oral iron and the non-supplemented group. Furthermore, Macdougall showed that in the group on IV iron, the dose of erythropoietin could be reduced. This has been conﬁrmed by several studies that show the superiority of intravenous iron over oral iron in correcting both absolute iron deﬁciency and functional deﬁciency and enabling a reduction in the dose of erythropoietin required to maintain a stable hemoglobin level (SunderPlassmann and Horl 1995; Silverberg et al. 1996; Taylor et al. 1996). In a recent prospective randomized trial, IV iron has also been shown to be superior to no iron or oral iron in anemic cancer patients receiving rhEPO (Auerbach et al. 2004). Side effects attributed to rhEPO treatment have been described and they vary in severity. Patients presenting with uremia may develop severe hypertension due to the rapid correction of the anemia with increase in blood viscosity and reverse in vascular hypotonia (McEvoy 1996). Also, the increase

672

M. J. Fontaine and L. T. Goodnough

in viscosity may enhance the risk of thrombosis (De Andrade et al. 1996). In a study by Goldberg et al., in which two different doses of rhEPO were given (600 and 300 IU/kg weekly), the incidence of thrombosis was 5% in the group receiving 600 IU/kg and 0% in the 300 IU/kg group (Goldberg et al. 1996). These results suggest that the increase of viscosity due to the accelerated expansion of circulating RBC mass could be responsible for this event (Mercuriali and Inghilleri 2002). Concern was recently aroused by the identiﬁcation of 191 epoetinassociated cases of pure red-cell aplasia between 1998 and 2004, as compared with only 3 such cases between 1988 and 1998 (Bennett et al. 2004). The estimated exposure-adjusted incidence was 18 cases per 100,000 patient-years for the formulation of epoetin alfa in Eprex (Janssen-Cilag) without human serum albumin, 6 cases per 100,000 patient-years for the Eprex formulation with serum albumin, 1 case per 100,000 patient-years for epoetin beta, and 0.2 case per 100,000 patient-years for the formulation of epoetin alfa in Epogen (Amgen). After procedures were adopted to ensure appropriate storage, handling, and administration of Eprex to patients with chronic kidney disease, the exposure-adjusted incidence decreased by 83 percent worldwide. According to Yasuda et al. Epo contributes to the growth, viability and angiogenesis of most malignant tumors through Epo receptor and tyrosine phosphorylation of STAT5 (Yasuda et al. 2003). The production of erythropoietin receptors by cancer cells appears to be regulated by hypoxia, and in clinical cancer specimens the highest levels of erythropoietin receptors were associated with neoangiogenesis, tumor hypoxia, and inﬁltrating tumors (Acs et al. 2001). Epo-receptor expression in almost all cancer cell lines raises the question of its potential activation by rhEPO, which needs to be investigated experimentally particularly in transplanted models of cancer cell lines. In a number of trials using various human cancer cell lines, however, no stimulatory effect of EPO has been observed (Rosti et al. 1993; Selzer et al. 2000; Westphal et al. 2002; Liu et al. 2004). The role of EPO-receptor expression on tumor cells is the subject of a separate chapter in this book.

Conclusion The positive short-term effects of therapy with erythropoietic agents on the correction of anemia and avoidance of blood transfusions are well documented. Few data are available on possible effects on the course of underlying disease, particularly since epoietin can exert additional biologic effects, including interference with the signal-transduction cascade of cytokines. Our current review conﬁrms that preoperative anemia in cancer patients needs to be corrected before surgery in order to improve patient outcome and survival. Recent reports (Dunne et al. 2002) clearly show that preoper-

rhEPO in surgical oncology

673

ative anemia and blood transfusions are associated with higher risks of infection and higher mortality rates. Perioperative administration of rhEPO in three recent clinical trials (Scott et al. 2002; Kosmadakis et al. 2003; Christodoulakis et al. 2005) is shown to remain efﬁcacious and safe for cancer patients suffering from signiﬁcant anemia and requiring ablative surgery. Therefore, rhEPO should still be considered as a safe adjunct treatment of perioperative anemia in order to reduce exposure to allogeneic blood during ablative surgery.

References 1. Abels RI, Larholt KM, Krantz KD, et al (1991) Recombinant human erythropoietin (r-HuEPO) for the treatment of the anemia of cancer. In: Murphy MJ (ed) Blood cell growth factors: their present and future in hematology and oncology. Alpha Med Press, Dayton, OH, pp 121–141 2. Acs G, Acs P, Beckwith SM, Pitts RL, Clements E, Wong K, et al (2001) Erythropoietin and erythropoietin receptor expression in human cancer. Cancer Res 61: 3561–3565 3. Albers P, Heicappell R, Schwaibold H, Wolff J (2001) Erythropoietin in urologic oncology. Eur Urol 39: 1–8 4. American Association of Blood Banks (AABB) (2005) Standards for blood banks and transfusion services, 23rd edn. AABB, Bethesda, Maryland 5. Anastassiades EG, Howarth D, Howarth J, Shanks D, Waters HM, Hyde K, et al (1993) Monitoring of iron requirements in renal patients on erythropoietin. Nephrol Dial Transplant 8: 846–853 6. Arendt A, Carmean J, Koch E, et al (2005) Fatal bacterial infections associated with platelet transfusion. MMWR Weekly 54: 168–170 7. Auerbach M, Ballard H, Trout JR, et al (2004) Intravenous iron optimizes the response to recombinant human erythropoietin in cancer patients with chemotherapy-related anemia: a multicenter, open-label. Randomized trial. J Clin Oncol 22: 1301–1307 8. Bennett CL, Luminari S, Nissenson AR, Tallman MS, Klinge SA, McWilliams N, et al (2004) Pure red-cell aplasia and epoetin therapy. N Engl J Med 351: 1403–1408 9. Blumberg N, Heal JM (1994) Effect of transfusion on immune function – Cancer recurrence and infection. Arch Pathol Lab Med 118: 371–379 10. Bordin JO, Heddle NM, Blajchman MA (1994) Biologic effects of leukocytes present in transfused cellular blood products. Blood 84: 1703–1721 11. Busch OR, Hop WC, van Papendrecht H, Marquet RL, Jeekel J (1993) Blood transfusions and prognosis in colorectal cancer. N Engl J Med 328: 1372– 1376 12. Christodoulakis M, Tsiftsis DD (2005) Preoperative epoetin alfa in colorectal surgery: a randomized, controlled study. Ann Surg Oncol 12: 718–725 13. Chun TY, Martin S, Lepor H (1997) Preoperative recombinant human erythropoietin injection versus preoperative autologous blood donation in patients undergoing radical retropubic prostatectomy. Urology 50: 727–732

674

M. J. Fontaine and L. T. Goodnough

14. De Andrade JR, Jove M, Landon G (1996) Baseline hemoglobin as a predictor of risk of transfusion and response to epoetin alfa in orthopedic surgery patients. Am J Orthop 25: 533–542 15. Dunne JR, Gannon CJ, Osborn TM, Taylor MD, Malone DL, Napolitano LM (2002) Perioperative anemia: an independent risk factor for infection, mortality, and resource utilization in surgery. J Surg Res 102: 237–244 16. Dunne JR, Gannon CJ, Osborn TM, Taylor MD, Malone DL, Napolitano LM (2002) Preoperative anemia in colon cancer: assessment of risk factors. Am Surg 68: 582–587 17. Dunphy FR, Dunleavy TL, Harrison BR, Boyd JH, Varvares MA, Dunphy CH, et al (1997) Erythropoietin reduces anemia and transfusions after chemotherapy with paclitaxel and carboplatin. Cancer 79: 1623–1628 18. Ellstrom M, Bengtsson A, Tylman M, Haeger M, Olsson JH, Hahlin M (1996) Evaluation of tissue trauma after laparoscopic and abdominal hysterectomy: measurements of neutrophil activation and release of interleukin-6, cortisol, and C-reactive protein. J Am Coll Surg 182: 423–430 19. Gillon J, Thomas MJG, Desmond MJ (1996) Acute normovolemic hemodilution. Transfusion 36: 640–643 20. Goldberg MA, McCutchen JW, Jove M (1996) A safety and efﬁcacy comparison study of two dosing regimen of epoetin alfa in patients undergoing major orthopedic surgery. Am J Orthop 25: 544–552 21. Goodnough LT, Price TH, Parvin CA, Friedman KD, Vogler WR, Khan N, et al (1994) Erythropoietin response to anaemia is not altered by surgery or recombinant human erythropoietin therapy. Br J Haematol 87: 695–699 22. Goodnough LT, Shander A, Brecher ME (2003) Transfusion medicine: looking to the future. Lancet 361: 161–169 23. Grogan M, Thomas GM, Melamed I, Wong FL, Pearcey RG, Joseph PK, et al (1999) The importance of hemoglobin levels during radiotherapy for carcinoma of the cervix. Cancer 86: 1528–1536 24. Groopman JE, Itri LM (1999) Chemotherapy-induced anemia in adults: incidence and treatment. J Natl Cancer Inst 91: 1616–1634 25. Hanazaki K, Kajikawa S, Shimozawa N, Matsushita A, Machida T, Shimada K, et al (2005) Perioperative blood transfusion and survival following curative hepatic resection for hepatocellular carcinoma. Hepatogastroenterology 52: 524–529 26. Hanazaki K, Matsushita A, Nakagawa K, Misawa R, Amano J (2005) Risk factors of intrahepatic recurrence after curative resection of hepatocellular carcinoma. Hepatogastroenterology 52: 580–586 27. Harrison LB, Shasha D, White C, Ramdeen B (2000) Radiotherapy-associated anemia: The scope of the problem. Oncologist 5: 1–7 28. Heiss MM, Mempel W, Delanoff C, Jauch KW, Gabka C, Mempel M, et al (1994) Blood transfusion-modulated tumor recurrence: ﬁrst results of a randomized study of autologous versus allogeneic blood transfusion in colorectal cancer surgery. J Clin Oncol 12: 1859–1867 29. Holness L, Knippen MA, Simmons L, Lachenbruch PA (2004) Fatalities caused by TRALI. Transfus Med Rev 18: 184–188 30. Houbiers JG, van de Watering LM, Hermans J, Verwey PJM, Bijnen AB, Pahlplatz P, et al (1994) Randomized controlled trial comparing transfusion of

rhEPO in surgical oncology

31.

32.

33.

34.

35.

36.

37.

38.

39.

40.

41. 42.

43.

44.

45. 46.

675

leukocyte-depleted or buffy-coat-depleted blood in surgery for colorectal cancer. Lancet 344: 573–578 Hyllner M, Avall A, Bengtson JP, Bengtsson A (2005) IL-6 and IL-8 response to erythropoietin therapy in radical hysterectomy. Acta Anaesthesiol Scand 49: 47–51 Jackson BR, Busch MP, Stramer SL, AuBuchon JP (2003) The cost-effectiveness of NAT for HIV, HCV, and HBV in whole-blood donations. Transfusion 43: 721–729 Kleinman S, Caulﬁeld T, Chan P, Davenport R, McFarland J, McPhedran S, et al (2004) Toward an understanding of transfusion-related acute lung injury: statement of a consensus panel. Transfusion 44: 1774–1789 Kosmadakis N, Messaris E, Maris A, Katsaragakis S, Leandros E, Konstadoulakis M, Androulakis G (2003) Perioperative erythropoietin administration in patients with gastrointestinal tract cancer. Ann Surg 237: 417–421 Laupacis A, Fergusson D (1998) Erythropoietin to minimize perioperative blood transfusion: a systematic review of randomized trials. The International Study of Peri-operative Transfusion (ISPOT). Transfus Med 8: 309–317 Lee WR, Berkey B, Marcial V, Fu KK, Cooper JS, Vikram B, et al (1998) Anemia is associated with decreased survival and increased locoregional failure in patients with locally advanced head and neck carcinoma: a secondary analysis of RTOG 85–27. Int J Radiat Oncol Biol Phys 42: 1069–1075 Levine EA, Rosen AL, Sehgal LR, Gould SA, Egrie JC, Moss GS (1991) Erythropoietin deﬁciency after coronary artery bypass procedures. Ann Thorac Surg 51: 764–766 Liu WM, Powles T, Shamash J, Propper D, Oliver T, Joel S (2004) Effect of haemopietic growth factors on cancer cell lines and their role in chemosensitivity. Oncogene 23: 981–990 Macdougall IC, Tucker B, Thompson J, Tomson CR, Baker LR, Raine AE (1996) A randomized controlled study of iron supplementation in patients treated with erythropoietin. Kidney Int 50: 1694–1699 Martin-Loeches M, Orti RM, Asins E, Llixiona J (2003) The prognosis implications of anaemia in the outcome of patients with early stages of uterine cervix carcinoma. Arch Gynecol Obstet 267: 121–125 McEvoy GK (1996) AHFS Drug information 96.American Society of Health System Pharmacists, Bethesda, MD Mercuriali F, Inghilleri G (2002) rHuEPO in surgical oncology. In: Nowrousian MR (ed) Recombinant human erythropoietin (rhEPO). Springer, Wien New York, pp 325–346 Mercuriali F, Inghilleri G, Bifﬁ E (1999) Short term low dose perisurgical recombinant human erythropoietin in trauma patients. Haematologica 84: (EHA-4 Abstract book): 217 Mercuriali F, Zanella A, Barosi G, Inghilleri G, Bifﬁ E, Vinci A, et al (1993) Use of erythropoietin to increase the volume of autologous blood donated by orthopedic patients. Transfusion 33: 55–60 Monk TG (2004) Preoperative recombinant human erythropoietin in anemic surgical patients. Crit Care 8: S45–S48 Motoyama S, Okuyama M, Kitamura M, Saito R, Kamata S, Murata K, et al (2004) Use of autologous instead of allogeneic blood transfusion during

676

47. 48.

49.

50.

51.

52.

53.

54.

55. 56.

57. 58.

59. 60.

61.

M. J. Fontaine and L. T. Goodnough esophagectomy prolongs disease-free survival among patients with recurrent esophageal cancer. J Surg Oncol 87: 26–31 Nissenson AR, Goodnough LT, Dubois RW (2003) Anemia: not just an innocent bystander? Arch Intern Med 163: 1400–1404 Price TH, Goodnough LT, Vogler WR (1996) The effect of recombinant human erythropoietin on the efﬁcacy of autologous blood donation in patients with low hematocrits: a multicente, randomized, double blind, controlled study. Transfusion 36: 29–36 Qvist N, Sci D, Boesby S, et al (1999) Recombinant human erythropoietin and hemoglobin concentration at operation during the postoperative period: reduced need for blood transfusions in patients undergoing colorectal surgery. Prospective double-blind placebo-controlled study. World J Surg 23: 30–35 Rosti V, Pedrazzoli P, Ponchio L, Zibera C, Novella A, Lucotti C, Della GR, Cazzola M (1993) Effect of recombinant human erythropoietin on hematopoietic and non-hematopoietic malignant cell growth in vitro. Haematologica 78: 208–212 Scott SN, Boeve TJ, McCulloch TM, Fitzpatrick KA, Karnell LH (2002) The effects of epoetin alfa on transfusion requirements in head and neck cancer patients: a prospective, randomized, placebo-controlled study. Laryngoscope 112: 1221–1229 Selzer E, Wacheck V, Kodym R, Schlagbauer-Wadl H, Schlegel W, Pehamberger H, Jansen B (2000) Erythropoietin receptor expression in human melanoma cells. Melanoma Res 10: 421–426 Shen JG, Cheong JH, Hyung WJ, Kim J, Choi SH, Noh SH (2005) Pretreatment anemia is associated with poorer survival in patients with stage I and II gastric cancer. J Surg Oncol 91: 126–130 Shinozuka N, Koyama I, Arai T, Numajiri Y, Watanabe T, Nagashima N, et al (2000) Autologous blood transfusion in patients with hepatocellular carcinoma undergoing hepatectomy. Am J Surg 179: 42–45 Silliman CC, Ambruso DR, Boshkov LK (2005) Transfusion-related acute lung injury. Blood 105: 2266–2273 Silliman CC, Boshkov LK, Mehdizadehkashi Z, Elzi DJ, Dickey WO, Podlosky L, et al (2003) Transfusion-related acute lung injury: epidemiology and a prospective analysis of etiologic factors. Blood 101: 454–462 Silverberg DS, Blum M, Peer G, Kaplan E, Iaina A (1996) Intravenous ferric saccharate as an iron supplement in dialysis patients. Nephron 72: 413–417 Stowell CP, Chandler H, Jove M, et al (1999) An open-label randomized study to compare the safety and efﬁcacy of perioperative epoetin alfa with preoperative autologous blood donation in total joint arthroplasty. Orthopedics 22: S105–S112 Sunder-Plassmann G, Hörl WH (1995) Importance of iron supply for erythropoietin therapy. Nephrol Dial Transplant 10: 2070–2076 Tasaki T, Ohto H, Hashimoto C, Abe R, Saitoh A, Kikuchi S (1992) Recombinant human erythropoietin for autologous blood donation: effects on perioperative red-blood-cell and serum erythropoietin production. Lancet 339: 773–775 Taylor JE, Peat N, Porter C, Morgan AG (1996) Regular low-dose intravenous iron therapy improves response to erythropoietin in haemodialysis patients. Nephrol Dial Transplant 11: 1079–1083

rhEPO in surgical oncology

677

62. Vamvakas EC (1996) Transfusion-associated cancer recurrence and postoperative infection: Meta-analysis of randomized, controlled clinical trials. Transfusion 36: 175–186 63. Vamvakas EC, Blajchman MA (2001) Deleterious clinical effects of transfusionassociated immunomodulation: fact or ﬁction? Blood 97: 1180–1195 64. van de Pol SM, Doornaert PA, de Bree R, Leemans CR, Slotman BJ, Langendijk JA (2005) The signiﬁcance of anemia in squamous cell head and neck cancer treated with surgery and postoperative radiotherapy. Oral Oncol (Epub ahead of print) 65. van Halteren HK, Houterman S, Verheij CD, Lemmens VE, Coebergh JW (2004) Anaemia prior to operation is related with poorer long-term survival in patients with operable rectal cancer. Eur J Surg Oncol 30: 628–632 66. Westphal G, Braun K, Debus J (2002) Detection and quantiﬁcation of the soluble form of the human erythropoietin receptor (sEpoR) in the growth medium of tumor cell lines and in the plasma of blood samples. Clin Exp Med 2: 45–52 67. Yasuda Y, Fujita Y, Matsuo T, Koinuma S, Hara S, Tazaki A, et al (2003) Erythropoietin regulates tumour growth of human malignancies. Carcinogenesis 24: 1021–1029 Correspondence: Magali J. Fontaine, M.D., Ph.D., Assistant Professor of Pathology, Associate Director of Transfusion Services, 300 Pasteur Drive H1402, Stanford CA 94305, California, E-mail: [email protected]

Chapter 26

Erythropoiesis, iron metabolism and iron supplementation during erythropoietin therapy L. T. Goodnough Departments of Pathology and Medicine Stanford University, Stanford, CA, USA

Introduction Knowledge gained regarding the relationship between erythropoietin, iron, and erythropoiesis in patients with anemia undergoing recombinant human erythropoietin therapy (EPO) has implications for patient management. During EPO therapy, iron restricted erythropoiesis is evident even in the presence of storage iron and oral iron supplementation. Intravenous iron therapy in renal dialysis patients undergoing EPO therapy can produce hematological responses in patients with serum ferritin levels in the normal range, indicating that traditional biochemical markers of storage iron in patients with anemia of chronic disease have limitations in assessment of iron status. Newer measurements of erythrocyte and reticulocyte indices using automated counters show promise in the evaluation of iron restricted erythropoiesis. The availability of safer intravenous iron preparations allows for carefully controlled studies of their value in patients undergoing EPO therapy. This chapter will provide an overview of these emerging issues, along with future directions for clinical research.

Historical context Clinical settings have served as “natural experiments” that have furthered our understanding of the relationship between erythropoietin, iron, and the erythropoietic response to anemia in man. Nearly 20 years ago, Finch summarized the knowledge gained primarily from studies of normal individuals, patients with hereditary hemolytic anemias, and patients with hemochromatosis [41]. Under conditions of basal erythropoiesis in normal subjects, plasma iron turnover (as an index of marrow erythropoietic response) is little affected, whether transferrin saturation ranges from very low to very high levels. In contrast, the erythropoietic response in individuals with congenital

680

L. T. Goodnough

hemolytic anemia, in whom erythropoiesis is chronically raised up to six-fold over basal levels [31], is affected (and limited) by serum iron levels and by transferrin saturation [71]. Patients with hemochromatosis who underwent serial phlebotomy were observed to mount erythropoietic responses of up to eight-fold over basal rates, attributed to the maintenance of very high serum iron and transferrin saturation levels in these patients [32], whereas normal individuals were shown to have difﬁculty providing sufﬁcient iron to support rates of erythropoiesis greater than three times basal rates [28]. These observations led Finch to identify a “relative iron deﬁciency” state, also known as “functional iron deﬁciency”, deﬁned as circumstances when increased erythron iron requirements exceed the available supply of iron [42]. Insights gained over the last 20 years regarding the relationship between erythropoietin, iron, and erythropoiesis in patients with anemia [62], along with implications for clinical management of oncology patients with anemia, will be reviewed.

Erythropoiesis mediated by endogenous erythropoietin The practice of autologous blood donation in patients scheduled for elective surgery is a natural experiment in blood loss anemia. Patients undergoing autologous blood phlebotomy may donate a unit (450 ± 45 ml) of blood as often as twice weekly, until 72 hours before surgery [85]. Under routine conditions, patients usually donate once weekly [52]. Oral iron supplements are routinely prescribed. This iatrogenic blood loss is accompanied by a response in endogenous erythropoietin levels that, while increased signiﬁcantly over basal levels, remain within the range of normal (4–26 mu/mL) [75]. The erythropoietic response that occurs under these conditions is modest [62,85]. A summary of selected prospective, controlled trials [56,59,61,73,74,113] of patients undergoing phlebotomy is presented in Table 1. Calculated estimates of red blood cell (RBC) volume expansion (erythropoiesis in excess of basal rates) were determined [17]. Two hundred twenty to 351 mL (11% to 19% RBC expansion [73,74] or the equivalent to one to 1.75 blood units [51]) are produced in excess of basal erythropoiesis, deﬁning the efﬁcacy of this blood conservation practice. For patients subjected to more aggressive (up to two units weekly) phlebotomy, the endogenous erythropoietin response is more substantial [56,59,61,113]. In one clinical trial [61], a linear-logarithmic relationship was demonstrated between change in hemoglobin level and erythropoietin response [58], predicted previously by phlebotomy experiments in normal subjects [21]. Erythropoietin-mediated erythropoiesis in this setting is 397 to 568mL (19% to 26% RBC expansion [56,59,61,113], or the equivalent of two to three blood units [51]).

Erythropoiesis, iron metabolism and iron supplementation

681

Table 1. Endogenous erythropoietin-mediated erythropoiesis Patients (n)

Baseline RBC (mL)

Requested/ Donated Units

RBC (mL) Donated

RBC (mL) Produced

RBC (mL) Expansion (%)

Iron Therapy

Ref

“Standard Phlebotomy” 108 1884 3 22 1936 3 45 1991 3 41 1918 3

2.7 2.8 2.9 2.9

522 590 621 603

351 220 331 315

19% 11% 17% 16%

PO None PO PO + IV

73 74 74 74

“Aggressive Phlebotomy” 30 2075 ≥3 30 2024 ≥3 30 2057 ≥3 24 2157 6 23 2257 6

3.0 3.1 2.9 4.1 4.6

540 558 522 683 757

397 473 436 568 440

19% 23% 21% 26% 19%

None PO IV PO PO

113 113 113 59,61 56

Data expressed as means. PO = Oral. IV = Intravenous. Modiﬁed, from Goodnough et al. (Ref. 62).

Erythropoiesis mediated by erythropoietin therapy Clinical trials have demonstrated a dose-response relationship between erythropoietin and red blood cell expansion [56]. A study of “very low”-dose EPO therapy in autologous blood donors found that 400 u/kg administered over a two-week interval resulted in clinically signiﬁcant erythropoiesis [99]. Table 2 details red cell volume expansion in 134 patients treated with EPO therapy during aggressive blood phlebotomy [19,56,59,61,87], ranging from 358 to 1764 mL (28% to 79% RBC expansion) over 25–35 days, or the equivalent of two to nine blood units [51]. The range in response (erythropoiesis) to dose (erythropoietin) is not related to patient gender or age [57,63], suggesting that patient-speciﬁc factors such as accompanying chronic disease, iron-restricted erythropoiesis, or other factors that normally cause the wide distribution of the hemoglobin level, account for the variability in erythropoietic response to erythropoietin. Studies in patients with the anemia of chronic disease (osteoarthritis [84,88,97] or rheumatoid arthritis [53,86]) are summarized in Table 3. Red cell volume expansion ranged from 157 to 353 mL (11% to 24%) for endogenous erythropoietin-mediated erythropoiesis and 268 to 673 mL (21% to 40%) with erythropoietin therapy. These erythropoietic responses are

682

L. T. Goodnough

Table 2. Erythropoiesis during blood loss and erythropoietin (EPO) therapy Patients (n/sex)

Baseline RBC (mL)

Total EPO Dose (U/kg)

Units Donated

RBC (mL) Donated

RBC (mL) Produced

RBC (mL) Expansion (%)

Iron Therapy

Ref

(10/F) (24) (10/F) (26) (11/F) (12/M) (23) (18) (1/M)

1285 1949 1293 2032 1796 2296 2049 2019 2241

900 SQ 900 IV 1800 SQ 1800 IV 3600 IV 3600 IV 3600 IV 3600 IV 4200 IV

3.4 5.2 4.3 5.5 4.9 5.9 5.4 5.6 8

435 864 526 917 809 1097 970 972 1600

358 621 474 644 701 1102 911 856 1764

28 32 37 32 39 48 45 42 79

IV PO IV PO PO PO PO PO Hemachromatosis

87 56 87 56 59,61 59,61 59,61 56 19

Data expressed as means. PO = Oral. IV = Intravenous. Modiﬁed, from Goodnough et al. (Ref. 62).

indistinguishable from patients with anemia from blood loss alone, noted in Tables 1 and 2. An additional study of 17 patients with inﬂammatory bowel disease treated with EPO and oral iron therapy demonstrated a similar response, with an estimated 20% increase in red cell volume when compared to placebo-treated patients [100].

Iron restricted erythropoiesis and iron therapy Erythropoiesis in response to aggressive autologous phlebotomy via endogenous erythropoietin has been estimated to increase by up to threefold [54,56]. No apparent relationship exists between basal iron stores and this magnitude of erythropoiesis, suggesting that under conditions of moderate erythropoiesis, serum iron and transferrin saturation for erythron requirements are adequately maintained by storage iron [56,59,61,113]. Little or no beneﬁt to oral iron supplementation was found in two studies [15,113], whereas a third study [74] found some beneﬁt (Table 1). Intravenous iron supplementation was not found to be of value in enhancing erythropoiesis under these conditions [74,113]. With enhanced erythropoiesis during EPO therapy, iron-restricted erythropoiesis occurs even in patients with measurable storage iron (Fig. 1). Despite an eightfold increase in gastrointestinal iron absorption [104], serum ferritin and transferrin saturation levels decline up to 50% with EPO therapy [68]. A fourfold increase in erythropoietic activity is accompanied by declin-

Erythropoiesis, iron metabolism and iron supplementation

683

Table 3. Erythropoietin (EPO) and erythropoiesis in patients with anemia* of chronic disease Patients Units RBC RBC (mL) Iron Rx (n) Donated Produced Expansion (mL) (%) I. Osteoarthritis 1. Placebo Placebo EPO (1800 U/kg IV)** EPO (1800 U/kg IV) EPO (3600 U/kg IV) EPO (3600 U/kg IV) 2. Placebo EPO (3600 U/kg IV) 3. Placebo Placebo EPO (1200 U/kg SQ) EPO (1200 U/kg SQ) II. Rheumatoid Arthritis Placebo EPO (3600 U/kg IV) EPO (1800 U/kg IV) EPO (800 U/kg SQ)

Ref

6 3 10 9 8 12 77 75 26 26 26 26

2.6 3.3 3.7 5.2 4.0 5.0 3.0 4.5 None None None None

157 220 268 560 289 515 353 673 4 18 219 220

11 18 21 43 22 40 24 44 0.3 1 14 15

PO 88 PO + IV PO PO + IV PO PO + IV PO 97 PO PO 84** IV PO IV

6 4 11 11

2.3 4.8 2.6 2.5

271 624 291 337

25 37 27 27

PO PO IV IV

53 86

* With measurable storage iron, ** EPO is total dosage of EPO administered, *** Perisurgical therapy without autologous phlebotomy. Data expressed as means. Modiﬁed, from Goodnough et al. (Ref. 62).

ing reticulocyte counts and the appearance of hypochromic red cells by the second week of erythropoietin therapy [16,19]. In a study of escalating (400%) EPO dose administered to patients undergoing aggressive phlebotomy, the marrow erythropoietic index increased from 2.9-fold (with endogenous erythropoietin stimulation) to 3.6-fold over basal rates of erythropoiesis, representing only a 58% increase in erythropoiesis (Fig. 2). The superior erythropoietic response in a patient with hemochromatosis further suggests iron-restricted erythropoiesis in patients treated with EPO (Table 2) [19].

Anemia of chronic renal failure or chronic disease The success of EPO therapy in correcting the anemia of chronic renal failure has led to substantial clinical experience in iron metabolism and

684

L. T. Goodnough

TOTAL RED BLOOD CELL VOLUME (ML/KG)

25 20 15 10 5 0

0

400

800

1200

1600

STORAGE IRON (MG)

Fig. 1. The relationship between initial storage iron (mg) and red blood cell volume expansion (ml/kg) in patients undergoing aggressive phlebotomy with erythropoietin (EPO) therapy. Linear regression analysis demonstrated a signiﬁcant correlation (r = 0.6, p = 0.02). Reprinted with permission [54]

(marrow erythropoietic index) 3.6 4

3.1 2.9

3

2.3

2

1

0 Placebo

EPO 150 U

EPO 300 U

EPO 600 U

Fig. 2. The erythropoietic response, as reﬂected in the bone marrow erythropoietic index, in four cohorts of autologous donors treated with placebo or escalating doses of EPO therapy. Erythropoietic response (ml/kg/day) was estimated for each treatment group, according to the formula: bone marrow erythropoietic index = [[RBC expansion] + [baseline RBC production]] = [baseline RBC production]. Reprinted with permission [62]

Erythropoiesis, iron metabolism and iron supplementation

685

erythropoiesis in this setting [37,43]. Hypo-responsiveness to EPO therapy is a common phenomenon [83,109] due to a variety of co-morbid conditions, particularly aluminum toxicity and iron deﬁciency. Anemic patients undergoing dialysis may show suboptimal response to oral iron therapy for several reasons. During EPO therapy, absorption of iron increases up to ﬁvefold [104]. However, external iron losses, including hemodialysis and blood testing, exceed gastrointestinal iron absorption [43]. Poor compliance due to gastrointestinal symptoms is problematic, and signiﬁcantly reduced iron absorption may occur with some newer iron formulations [117]. Iron-restricted erythropoiesis is evident by clinical responses to ascorbate supplementation, thought to facilitate the release of iron from reticuloendothelial stores and increased iron utilization by erythrons [109], as well as the success of intravenous iron therapy in reducing EPO dosage [43]. Intravenous iron administration has become standard therapy in renal dialysis patients undergoing EPO therapy [3]. Patients treated with intravenous iron (100 mg twice weekly) achieved a 46% reduction in EPO dosage required to maintain hematocrit levels between 30% and 34%, compared to patients supplemented with oral iron [43]. In a study [103] of chronic renal failure (non-dialysis) patients, two-thirds of patients who were unresponsive to oral iron responded to weekly intravenous iron therapy. Improved erythropoiesis occurred despite initial serum ferritin levels as high as 400 μg/L [110], indicating that biochemical markers of storage iron in these patients are not helpful in evaluating iron-restricted erythropoiesis. The effect of intravenous iron therapy in other patients with the anemia of chronic disease undergoing EPO therapy is shown in Table 3. Patients with osteoarthritis and measurable storage iron doubled their red cell expansion, from a range of 21% to 22% (with oral iron) to 40% to 43% with intravenous iron [88]. Intravenous iron therapy in iron-deﬁcient patients with inﬂammatory bowel disease also result in improved responses to EPO therapy [47], compared to a similar patient group who received oral iron supplementation [100]. The clinical response to intravenous iron may be attributed to the effect of EPO therapy on iron mobilization from the reticuloendothelial system (RES) into red cell precursors [116]. The risk/beneﬁt proﬁle of intravenous iron is controversial in anemic renal dialysis patients [38,91] as well as in patients with anemia of chronic disease [114]. Intravenous iron can allow up to a ﬁvefold erythropoietic response to signiﬁcant blood loss anemia in normal individuals [65,71]. A greater rate of hemoglobin production is probably not possible unless red marrow expands into yellow marrow space, as is seen in hereditary anemias [31,65]. One limitation to intravenous iron therapy in patients not undergoing EPO therapy may be that much of the administered iron is transported into the RES as storage iron, where it is less readily available for erythropoiesis [12]. For irondeﬁcient patients, 50% of intravenous iron is incorporated into hemoglobin

686

L. T. Goodnough

within 3–4 weeks [118], whereas for patients with anemia of chronic disease or renal failure, intravenous iron is less rapidly mobilized from the RES [9]. The value of intravenous iron administration in patients receiving EPO therapy outside the setting of renal dialysis is currently not established. In one clinical trial [88], signiﬁcantly greater erythropoietic responses were seen with intravenous iron therapy compared to patients supplemented with oral iron only (Table 3). However, a subsequent study found no difference in red cell production between oral iron and intravenous iron therapy in patients before orthopedic surgery [84]. Another study found that intravenous iron supplementation was not accompanied by a corresponding erythropoietic response to increasing doses of EPO therapy; a twofold increase in EPO dose was associated with only a 32% increase in red cell production [87], similar to the dose-response relationship utilizing oral iron supplementation [56]. Intravenous iron administered to normal subjects treated with EPO abolished the marked reduction in serum ferritin and increased the reticulocyte hemoglobin content (a measure in g/L of the hemoglobin contained in all reticulocytes); however, the total number of reticulocytes generated over eight days after therapy was not affected [80]. Finally, perisurgical exposure to allogeneic blood is not different for autologous blood donors with or without measurable storage iron, regardless of oral [54,97] or intravenous iron [84] administration. The current status of intravenous iron therapy in patients with anemia is summarized in Table 4.

Table 4. Intravenous iron therapy for anemia Beneﬁcial

No Beneﬁt

Evolving indications

I.

I. Autologous blood donation in patients withor without iron deﬁciency [74,113]

I.

Anemia of renal failure, with or without EPO therapy [43,91,103] II. Patients with ongoing blood loss [68,82] IV. Jehovah’s Witness patients with iron deﬁciency [7] and/or blood loss [82]

Blood loss, iron deﬁciency and EPO therapy [54,97] III. Anemia of chronic disease and EPO therapy [47,53,100, 115] II. Perisurgical anemia, with or without EPO therapy [68,84]

Absolute iron deﬁciency is deﬁned as ferritin <200 μg/L and/or iron saturation <20%, or relative iron deﬁciency (ferritin <400 μg/L in dialysis patients receiving erythropoietin therapy, ref. 91) or the presence of >10% hypochromic erythrocytes and/or reticulocytes. (References) Modiﬁed, from Goodnough et al. (Ref. 62).

Erythropoiesis, iron metabolism and iron supplementation

687

Laboratory evaluation of iron metabolism Iron, transferrin and transferrin saturation The diagnosis of iron deﬁciency is traditionally based on a combination of parameters, including iron metabolism and hematological indices. Technical and biological issues limit the usefulness of these assays in the clinical setting [23,36,112], and the value of iron, transferrin, and transferrin saturation is limited to uncomplicated iron deﬁciency. Transferrin saturation falls below 16% only when iron stores are exhausted, in contrast to EPO therapyinduced erythropoiesis, in which iron saturation falls even in the presence of storage iron [27]. Detection of iron-restricted erythropoiesis during EPO therapy [27,60,67] therefore poses additional challenges.

Ferritin Ferritin is widely used as a marker of iron storage [29,77], with a cutoff of 15 μg/L indicating absent iron stores in normal individuals [94]. However, one study found that 25% of women with no stainable bone marrow iron had serum ferritin levels above the 15 μg/L cutoff [64]. A level of 30 μg/L [98] to 40 μg/L [81] for anemic patients is therefore desirable in order to provide optimal diagnostic efﬁciency (positive predictive values of 92% to 98%, respectively), even without clinical evidence of infection or inﬂammation. For patients with anemia of chronic disease, however, ferritin levels are normal or increased (Table 5) [115]. Subjects treated with EPO exhibit a rapid decrease in ferritin to levels 50–75% below baseline [20]. Ferritin also decreased rapidly even after intravenous iron was administrated in normal subjects treated with EPO [80]. Under these conditions, ferritin most likely reﬂects the iron content of a smaller, more labile pool in equilibrium with both the erythropoietic compartment and storage iron. Many patients have underlying disorders with “inappropriately high” serum ferritin levels. Ferritin levels are elevated in conditions such as hyperthyroidism, inﬂammation/infection, hepatocellular disease, malignancies, alcohol consumption, and oral contraceptives [76]. Two- thirds of renal dialysis patients respond to intravenous iron therapy, with mean ferritins of 94 μg/L and mean transferrin saturations of 22% that are no different than the patients who are not responsive to intravenous iron [110]. This has led to suggested guidelines [91] for renal dialysis patients, in which a ferritin of less than 200 μg/L alone, or less than 400 μg/L with a transferrin saturation <20%, are used to determine the need for intravenous iron therapy; only at transferrin saturations greater than 50% or ferritin levels in excess of 800 μg/L, are these patients considered unlikely to beneﬁt from iron therapy.

688

L. T. Goodnough

Table 5. Serum levels that differentiate anemia of chronic disease from irondeﬁciency anemia* Variable

Anemia of Chronic Disease

Iron-Deﬁciency Anemia

Both Conditions†

Iron Transferrin

Reduced Reduced to normal Reduced Normal to increased

Reduced Increased

Reduced Reduced

Reduced Reduced

Reduced Reduced to normal

Normal

Increased

Low (<1)

High (>2)

Normal to increased High (>2)

Increased

Normal

Increased

Transferrin saturation Ferritin Soluble transferrin receptor Ratio of soluble transferrin receptor to log ferritin Cytokine levels

* Relative changes are given in relation to the respective normal values. † Patients with both conditions include those with anemia of chronic disease and true iron deﬁciency. Reproduced, with permission [115].

In patients with the anemia of cancer, up to 50% are unresponsive to EPO therapy [78]. After two weeks of EPO therapy, ferritin levels greater than 400 μg/L correctly predicted lack of response in 88% of the cases, while levels <400 μg/L correctly predicted response in 75% of cases [78]. However, several studies have failed to show a role for ferritin in either predicting response to EPO therapy or in identifying functional iron deﬁciency in patients with cancer-related anemia [25,70,72]. Parenteral iron has been demonstrated to enhance rates of response (compared to a placebo or oral iron supplementation) to therapy with erythropoietic agents in cancer patients who are undergoing chemotherapy [4]. Further clinical studies to determine the level of ferritin that would predict response to intravenous iron in oncology patients undergoing EPO therapy are needed.

Erythrocyte and reticulocyte parameters Since reticulocytes are normally released from the marrow 18–36 hours before their ﬁnal maturation into erythrocytes, they provide a real-time assessment of the functional state of erythropoiesis. However, in the early phases of stimulated erythropoiesis, changes in absolute reticulocyte counts reﬂect the release from the marrow of immature reticulocytes rather than

Erythropoiesis, iron metabolism and iron supplementation

689

true expansion of erythropoiesis [9,65,26]. It has been suggested that response to EPO therapy could be assessed by measuring hemoglobin and reticulocyte counts after 4 weeks of therapy: a change in hemoglobin level greater than 1.0 g/dl and/or a change in absolute reticulocyte count more than 40 × 109/L could indicate that the patient is a “responder” to EPO therapy [24,70]. Flow cytometric analysis of reticulocytes allows precise measurements of reticulocyte cell volume (MCVr), hemoglobin concentration (CHCMr) and hemoglobin content (CHr) [73]. In normal subjects, EPO therapy induces an increase in MCVr and a decrease in CHCMr [10]. Normal subjects treated with erythropoietin with baseline serum ferritin >100 μg/L have almost no production of hypochromic reticulocytes. Detection of iron-restricted erythropoiesis takes place at an earlier stage if reticulocyte parameters are used rather than red cell indices [18,20]. Among the latter, the proportion of hypochromic red blood cells, deﬁned as red blood cells with a hemoglobin concentration of less than 28 g/dl, can be used to identify iron-restricted erythropoiesis, but this parameter, used with a cutoff point of 10% appears to be less reliable than CHr [44]. CHr has been studied in dialysis patients. CHr demonstrated 100% sensitivity and 80% speciﬁcity and was a more accurate predictor of response to iron therapy than serum ferritin, transferrin saturation, or percentage of hypochromic erythrocytes [44]. Another study showed that a baseline CHr < 28 pg had 78% sensitivity and 71% speciﬁcity to detect iron-restricted erythropoiesis, compared with 50% and 39% for traditional biochemical measures [90]. In dialysis patients treated with EPO, CHr increases during intravenous iron therapy, indicating value as an early indicator of ironrestricted erythropoiesis [13], even with normal serum ferritin or transferrin saturation [14]. Measurements of total reticulocyte hemoglobin, an integrated index which is derived from the absolute reticulocyte count and the CHr [34], showed that reticulocyte-hemoglobin was much higher in subjects treated with intravenous iron [80]. Moreover, administration of intravenous iron along with EPO therapy in cardiac surgery patients abolished the production of hypochromic reticulocytes, and CHr remained within the normal range [105]. A recent study concluded that CHr was the strongest predictor of iron deﬁciency in children [21], and should be considered an alternative to standard iron studies for diagnosis of iron deﬁciency.

Transferrin receptor The soluble transferrin receptor (TfR) is derived primarily from red cell precursor normoblasts [46] and provides an estimate of the erythroid compartment mass. Both enhanced erythropoiesis and iron deﬁciency elevate

690

L. T. Goodnough

TfR. Serum ferritin is the most sensitive and speciﬁc index of iron status when there are residual iron stores, whereas TfR is most sensitive in the presence of iron-restricted erythropoiesis [81]. In a study of 43 healthy, nonanemic adult women, 17 (40%) had signiﬁcant changes in TfR in response to oral iron therapy, indicating the presence of subclinical iron deﬁciency [108]. Moreover, in another study, twenty-ﬁve percent of patients undergoing routine ferritin tests who were also studied for TfR measurements were categorized as iron-deﬁcient by TfR (>2.8 mg/L) but not by ferritin (>12 μg/L) [81]. These values could represent iron-replete individuals with increased erythropoiesis, or iron-deﬁcient patients with an acute-phase increase of the ferritin value. The clinical utility of the TfR may, therefore, be limited to the subset of ill patients in whom iron deﬁciency is suspected but whose ferritin values are normal or raised [81], seen commonly in the anemia of chronic disease; a number of studies [40,81,98] have shown TfR to be of value in differentiating iron deﬁciency anemia (where TfR is usually increased) from the anemia of chronic disease (where TfR is usually normal). The value of TfR in predicting the response to EPO therapy and the adequacy of iron availability is modest. Lower baseline or low-normal TfR levels may predict the initial response to EPO therapy in patients on dialysis [11]. Other studies, however, have shown little predictive value for this assay in patients receiving EPO, since serum TfR values above normal are observed both in iron deﬁciency and during erythropoietin-induced expansion of erythropoietic activity [30]. Further studies are required to delineate the clinical usefulness of TfR measurements in patients undergoing EPO therapy.

Erythropoietin assay A classiﬁcation of anemias has been proposed around the concept of an adequate or an inadequate erythropoietin response to the degree of anemia, using patients with iron deﬁciency or chronic hemolytic anemia as reference populations [8]. The correlation between the percentage of patients showing an “inadequate” erythropoietin response to anemia and the percentage of patients responding to EPO therapy (according to the author’s criteria) can be illustrated (Fig. 3) for several diseases, with a range in response in myelodysplastic syndromes [111], multiple myeloma [79], and rheumatoid arthritis [96]. There are several problems with the use of erythropoietin levels in the management of individual patients. The interpretation of an erythropoietin level must take into account the degree of anemia at the time of measurement; commercial assay results do not take this into consideration, so that clinicians need familiarity with mathematical corrections such as observed/ predicted ratios [8]. A retrospective analysis of EPO therapy in anemic

Erythropoiesis, iron metabolism and iron supplementation Responsiveness to r-Hu-Epo (% of cases)

100

691

ARF RA

80 MM HIV 60 CANCER 40

MDS/MMM 20

0 0

20

40

60

80

100

Inadequate Epo response to anemia (% of cases)

Fig. 3. Correlation between the percentage of cases showing “inadequate EPO response to anemia”: and the percentage of cases responding to EPO therapy (according to the author’s criteria). The numbers are derived directly or calculated from reported data (ARF) anemia of renal failure, (RA) anemia of rheumatoid arthritis, (HIV) anemia in HIV-infected patients, (MM) anemia in multiple myeloma, (Cancer), anemia of cancer, (MDS/MMM) anemia in myelodysplastic syndromes and myeloﬁbrosis with myeloid metaplasia. Reprinted with permission [11]

cancer patients not receiving chemotherapy [1] found that pretreatment erythropoietin levels of <200 mu/mL were correlated with red cell response to EPO therapy; however, subsequent analyses have found that erythropoietin levels are not predictive for response in cancer patients receiving chemotherapy [48,49]. Since almost all anemic cancer patients and renal dialysis patients have erythropoietin levels that are inadequate for the degree of anemia [8], measuring erythropoietin levels is not useful in these settings. Furthermore, guidelines recommend that EPO therapy be instituted before hemoglobin levels fall below 10 g/L, a level at which interpretation of erythropoietin level

692

L. T. Goodnough

is not valid [8]. The erythropoietin assay may be most useful as a determinant of response to therapy in certain settings such as patients with multiple myeloma, malignant lymphoma or myelodysplasia [25,93,111].

Iron therapy strategies Intravenous iron therapy has been closely scrutinized for risks and adverse events. Imferon (Iron Dextran BP) was previously approved for parenteral use [6]. This product was associated with a 0.6% risk of anaphylactoid reactions and a 1.7% risk of severe, serum sickness-like reactions characterized by fever, arthralgias and myalgias [66]. An incidence of delayed reactions of up to 30% and severe reactions of 5.3% were subsequently described, attributed to changes in manufacturing processes [119]; this product was withdrawn from use. InFed (Iron dextran USP, Schein Pharm Corp, Florham Park, NJ) is currently approved for parenteral (intramuscular or intravenous) use in the U.S. InFed administered intravenously during dialysis is associated with signiﬁcant adverse reactions in 4.7% of patients, of which 0.7% are serious or lifethreatening, and another 1.7% are characterized as anaphylactoid [45]. The prevalence of these reactions does not differ among patients receiving lowdose (100 mg) or higher-dose (250–500 mg) infusions [5]. A recent review reported 196 allergic/anaphylaxis cases with use of iron dextran in the U.S. between 1976 and 1996, of which 31 (15.8%) were fatal [39]. Safety aspects of parenteral iron dextran, ferric gluconate, and iron saccharate have been scrutinized [102,106,107]. Iron saccharate is available in Europe (Venofer®, Fresenius Medical Care, Germany) but not the US. The drug can be applied intravenously over one hour up to 200 mg at each administration. Ferric gluconate has been available in Europe for more than 20 years and was approved for use in the US in 1999 for intravenous administration (Ferrlecit, Schein Pharm Corp, Florham Park, NJ) in renal dialysis patients. Dosage is limited to 125 mg infused over one hour at each administration. The rate of allergic reactions (3.3 episodes per million doses) appears lower than iron dextran (8.7 episodes per million doses) and the safety proﬁle is substantially better; among seventy-four severe adverse events reported from 1976 to 1996, there were no deaths [92]. Adverse events associated with ferric gluconate include hypotension, rash, chest or abdominal pain, with an incidence of 1.3% for serious reactions [22,95]. Intravenous iron therapy can cause a clinical syndrome (nausea, facial reddening, and hypotension) which may be attributed to acute iron toxicity due to oversaturation (>100%) of transferrin [120] or due to nonspeciﬁc drug toxicity [101]. The increased erythropoietic effect (4.5 to 5.5 times basal) of intravenous iron dextran (with an estimated half-life of 60 hrs) is transient and lasts 7–10 days, after which the remaining iron is sequestered

Erythropoiesis, iron metabolism and iron supplementation

693

in the RES and erythropoiesis returns to basal rates [71]. Iron measurements and intravenous iron therapy are, therefore, optimal at two-week intervals. A dose-response relationship between EPO therapy and erythropoiesis that is affected favorably by intravenous iron has important implications for EPO dosage and cost [69]. Current total recommended EPO dosage for patients scheduled for elective surgery [55] range from 1800 u/kg [69] to 4200 u/kg [50], which for a 70 kg patient would cost $1300 to $3000 [35]. As in renal dialysis patients [43], intravenous iron may reduce EPO dosage in oncology patients by improving iron-restricted erythropoiesis, even in patients with demonstrable iron stores; clinical trials to study this are underway.

Conclusion The development of new laboratory methods to assess iron-restricted erythropoiesis, along with clinical trials of patients undergoing EPO therapy, have furthered our understanding of the relationship between erythropoiesis and iron metabolism. Reticulocyte parameters hold promise in the evaluation of iron-restricted erythropoiesis, but more studies are needed in order to deﬁne their role in patients undergoing iron and/or EPO therapy. The erythropoietin assay and transferrin receptor assay are valuable tools for clinical research, but their roles in routine clinical practice remain undeﬁned. The availability of a safer intravenous iron preparation allows an opportunity to study its value in oncology patients undergoing EPO therapy. Given the low prevalence but potential side effects, the use of intravenous iron in oncology patients needs to be deﬁned by controlled clinical trials.

References 1. Abels R, Larholt K, Krantz K, Bryant E (1991) Recombinant human erythropoietin for the treatment of the anemia of cancer. In: Murphy MJ Jr (ed) Blood cell growth factors: their present and future use in hematology and oncology. Alpha Med, Dayton, OH, pp 121–141 2. Adamson JW (1968) The erythropoietin/hematocrit relationship in normal and polycythemic man: implications of marrow regulation. Blood 32: 597–609 3. Adamson JW, Eschbach JW (1998) Erythropoietin for end-stage renal disease. N Engl J Med 339: 625–627 4. Auerbach M, Ballard H, Trout JR, McIlwain M, Ackerman A, Bahrain H, Balan S, Barker L, Rana J (2004) Intravenous Iron optimizes the response to recombinant human erythropoietin in cancer patients with chemotherapy-related anemia: A multicenter, open-label randomized trial. J Clin Oncol 22: 1301–1307 5. Auerbach M, Winchester J, Wahab A, Richards K, McGinley M, Hall F, Anderson J, Briefel G (1998) A randomized trial of three iron dextran infusion methods for anemia in erythropoietin-treated dialysis patients. Am J Kidney Dis 31: 81–86

694

L. T. Goodnough

6. Auerbach M, Witt D, Toler W, Fierstein M, Lerner RG, Ballard H (1988) Clinical use of the total dose intravenous of iron dextran. J Lab Clin Med 111: 566–569 7. Baron JF (1999) Primum non nocere. In: Baron JF (ed) Transfusion alternatives in transfusion medicine. LM Sante, Chattilon Cedex, Fr.; 1: 3–5 8. Barosi G (1994) Inadequate erythropoietin response to anemia: deﬁnition and clinical relevance. Ann Hematol 68: 215–223 9. Beamish MR, Davies AG, Eakins JD, Jacobs A, Trevett D (1971) The measurement of reticuloendothelial iron release using iron-dextran. Br J Haematol 21: 617–622 10. Beguin Y, Loo M, R’Zik S, Sautois B, Lejeune F, Rorive G (1995) Quantitative assessment of erythropoiesis in haemodialysis patients demonstrates gradual expansion of erythroblasts during constant treatment with recombinant human EPO. Br J Haematol 89: 17–23 11. Beguin Y, Loo M, R’Zik S, Sautois B, Lejeune F, Rorive G, Fillet G (1993) Early prediction of response to recombinant human erythropoietin in patients with the anemia of renal failure by serum transferrin receptor and ﬁbrinogen. Blood 82: 2010–2016 12. Beutler E (1958) The utilization of saccharated Fe59 oxide in red cell formation. J Lab Clin Med 51: 415–419 13. Bhandari S, Norfolk D, Brownjohn A, Turney J (1997) Evaluation of RBC ferritin and reticulocyte measurements in monitoring response to intravenous iron therapy. Am J Kidney Dis 30: 814–821 14. Bhandari S, Turney JH, Brownjohn AM, Norfolk D (1998) Reticulocyte indices in patients with end stage renal disease on hemodialysis. J Nephrol 11: 78–82 15. Biesma DH, Kraaijenhagen RJ, Poortman J, Marx JJM, Van De Wiel A (1992) The effect of oral iron supplementation on erythropoiesis in autologous blood donors. Transfusion 32: 162–165 16. Biesma DH, Van De Wiel A, Beguin Y, Kraaijenhagen RJ, Marx JJM (1994) Erythropoietic activity and iron metabolism in autologous blood donors during recombinant human erythropoietin therapy. Eur J Clin Invest 24: 426–432 17. Brecher MA, Monk TG, Goodnough LT (1997) A standardized method for calculating blood loss. Transfusion 37: 1070–1074 18. Brugnara C (1998) Use of reticulocyte cellular indices in the diagnosis and treatment of hematological disorders. Int J Clin Lab Res 28: 1–11 19. Brugnara C, Chambers LA, Malynn E, Goldberg MA, Kruskall MS (1993) Red blood cell regeneration induced by subcutaneous recombinant erythropoietin: iron-deﬁcient erythropoiesis in iron-replete subjects. Blood 81: 956–964 20. Brugnara C, Colella GM, Cremins JC, Langley RC, Schneider TJ, Rutherford CJ, Goldberg MA (1994) Effects of subcutaneous recombinant human erythropoietin in normal subjects: development of decreased reticulocyte hemoglobin content and iron-deﬁcient erythropoiesis. J Lab Clin Med 123: 660–664 21. Brugnara C, Zurakowski D, DiCanzio J, Boyd T, Platt O (1999) Reticulocyte hemoglobin content to diagnose iron deﬁciency in children. JAMA 281: 2225– 2247 22. Calvar C, Mata D, Alonso C, Ramos B, Lopez de Novales E (1997) Intravenous administration of iron gluconate during haemodialysis. Nephrol Dial Transplant 12: 574–575

Erythropoiesis, iron metabolism and iron supplementation

695

23. Cavill I (1982) Disorders of iron metabolism. Diagnostic methods. Clin Haematol 11: 259–273 24. Cazzola M, Mercuriali F, Brugnara C (1997) Use of recombinant human erythropoietin outside the setting of uremia. Blood 89: 4248–4267 25. Cazzola M, Messinger D, Battistel V, Bron D, Cimino R, Enller-Ziegler L, Essers U, Greil R, Grossi A, Jäger G, LeMevel A, Najman A, Silingardi V, Spriano M, van Hoof A, Ehmer B (1995) Recombinant human erythropoietin in the anemia associated with multiple myeloma or non-Hodgkin’s lymphoma: dose-ﬁnding and identiﬁcation of predictors of response. Blood 86: 4446–4453 26. Cazzola M, Ponchio L, Pedrotti C, Farina C, Cerani P, Lucottic C, Novella A, Rovati A, Bergamaschi G, Beguin Y (1996) Prediction of response to recombinant human erythropoietin (rHuEpo) in anemia of malignancy. Haematologica 81: 434–441 27. Cazzola M, Pootrakul P, Bergamaschi G, Huebers HA, Eng M, Finch CA (1987) Adequacy of supply for erythropoiesis: in vivo observations in humans. J Lab Clin Med 110: 734–739 28. Coleman PH, Stevens AR, Dodge HT, Finch CA (1953) Rate of regeneration after blood loss. Arch Intern Med 92: 341–348 29. Cook JD, Lipschitz DA, Miles LEM, Finch CA (1974) Serum ferritin as a measure of iron stores in normal subjects. Am J Clin Nutr 27: 681–687 30. Cook JD, Skikne BS, Baynes RD (1993) Serum transferrin receptor. Annu Rev Med 44: 63–74 31. Crosby WH (1955) The metabolism of hemoglobin and bile pigment in hemolytic disease. Am J Med 18: 112–122 32. Crosby WH (1958) Treatment of haemochromatosis by energetic phlebotomy. One patient’s response to the letting of 55 liters of blood in 11 months. Br J Haematol 4: 82–88 33. d’Onofrio G, Chirillo R, Zini G, Caenaro G, Tommasi M, Micciulli G (1995) Simultaneous measurement of reticulocyte and red blood cell indices in healthy subjects and patients with microcytic and macrocytic anemia. Blood 85: 818–823 34. Davis BH, Bigelow N, Ball ED, Mills L, Cornwell GG 3rd (1989) Utility of ﬂow cytometric reticulocyte quantiﬁcation as a predictor of engraftment in autologous bone marrow transplantation. Am J Hematol 32: 81–87 35. Doolittle RF (1991) Biotechnology – the enormous cost of success. N Engl J Med 324: 360–362 36. Eckfeldt JH, Witte DL (1994) Serum iron: would analytical improvement enhance patient outcome? Clin Chem 40: 505–507 37. Eschbach J, Egrie J, Downing M, Browne JK, Adamson JW (1987) Correction of anemia of end-stage renal disease with recombinant human erythropoietin. N Engl J Med 316: 73–78 38. Eschbach JW, Adamson JW (1999) Iron overload in renal failure patients: Changes since the introduction of erythropoietin therapy. Kidney Int 55 [Suppl]: S35–S43 39. Faich G, Strobos J (1999) Sodium ferric gluconate complex in sucrose: safer intravenous iron therapy than iron dextran. Am J Kidney Dis 33: 464–470 40. Ferguson BJ, Skikne BS, Simpson KM, Baynes RD, Cook JD (1992) Serum transferrin receptor distinguishes the anemia of chronic disease from iron deﬁciency anemia. J Lab Clin Med 19: 385–390

696

L. T. Goodnough

41. Finch CA (1982) Erythropoiesis, erythropoietin, and iron. Blood 60: 1241– 1246 42. Finch CA (1982) Perspectives in iron metabolism. N Engl J Med 25: 1520– 1528 43. Fishbane S, Frei GL, Maesaka J (1995) Reduction in recombinant human erythropoietin doses by the use of chronic intravenous iron supplementation. Am J Kidney Dis 26: 41 44. Fishbane S, Galgano C, Langley RC Jr, Canﬁeld W, Maesaka JK (1997) Reticulocyte hemoglobin content in the evaluation of iron status of hemodialysis patients. Kidney International. 52: 217–222. 45. Fishbane S, Ungureanu VD, Maeska JK, Kaupke CJ, Lim V, Wish S (1996) The safety of intravenous iron dextran in hemodialysis patients. Am J Kidney Dis 28: 529–534 46. Flowers CH, Skikne BS, Covell AM, Cook JD (1989) The clinical measurement of serum transferrin receptor. J Lab Clin Med 114: 368–377 47. Gasche C, Dejaco C, Waldhoer T, Tilinger W, Reinisch W, Fueger GF, Gangl A, Lochs T (1997) Intravenous iron and erythropoietin for anemia associated with Crohn’s disease. Ann Int Med 126: 782–787 48. Glaspy J, Bukowski R, Steinberg D, Taylor C, Tchekmedyians S, Vadhan–Raj S (1997) Impact of therapy with epoietin alfa on clinical outcomes in patients with nonmyeloid malignancies during cancer chemotherapy in community oncology practice. J Clinic Oncol 15: 1218–1234 49. Glaspy J, Demetri G, Cremieux P (1999) Does baseline endogenous erythropoietin level impact hemoglobin (Hb) response to rHuEPO in cancer patients? Blood 94 [Suppl 1]: 15b (Abstr 3210) 50. Goldberg MA, McCutchen JW, Jove M, DiCesare P, Friedman RJ, Poss R, Guilfoyle M, Freid D, Young DC (1996) A safety and efﬁcacy comparison study of two dosing regimens of epoietin alfa in patients undergoing major orthopedic surgery. Am J Orth 25: 544–597 51. Goodnough LT, Bravo J, Hsueh Y, Keating L, Brittenham GM (1989) Red blood cell volume in autologous and homologous blood units. Implications for risk beneﬁt assessment for autologous blood “crossover” and directed blood transfusion. Transfusion 29: 821–822 52. Goodnough LT, Brittenham G (1990) Limitations of the erythropoietic response to serial phlebotomy: Implications for autologous blood predeposit programs. J Lab Clin Med 115: 28–35 53. Goodnough LT, Marcus RE (1997) The erythropoietic response to erythropoietin therapy in patients with rheumatoid arthritis. J Lab Clin Med 130: 381–386 54. Goodnough LT, Marcus RE (1998) Erythropoiesis in iron-depleted and ironreplete patients stimulated with erythropoietin: the relevance of storage iron. Vox Sang 75: 128–133 55. Goodnough LT, Monk TG, Andriole GL (1997) Erythropoietin therapy. N Engl J Med 336: 933–938 56. Goodnough LT, Price TH, Friedman KD, Johnston M, Ciavarella D, Khan N, Sacher R, Vogler WR, Wissel M, Abels RI (1994) A phase III trial of recombinant human erythropoietin therapy in non-anemic orthopedic patients subjected to aggressive autologous blood phlebotomy: dose, response, toxicity, efﬁcacy. Transfusion 34: 66–71

Erythropoiesis, iron metabolism and iron supplementation

697

57. Goodnough LT, Price TH, Parvin CA (ö1995) The endogenous erythropoietin response and the erythropoietic response to blood loss anemia: The effects of age and gender. J Lab Clin Med 126: 57–64 58. Goodnough LT, Price TH, Parvin CA, Friedman KD, Vogler WR, Khan N, Sacher R, Johnston M, Wissel M, Ciavarella D (1994) Erythropoietin response to anaemia is not altered by surgery or recombinant human EPO therapy. Br J Haematol 87: 695–699 59. Goodnough LT, Price TH, Rudnick S, Soegiarso RW (1992) Preoperative red blood cell production in patients undergoing aggressive autologous blood phlebotomy with and without erythropoietin therapy. Transfusion 32: 441–445 60. Goodnough LT, Price TH, Rudnik S (1991) Iron-restricted erythropoiesis as a limitation to autologous blood donation in the erythropoietin-stimulated bone marrow. J Lab Clin Med 118: 289–296 61. Goodnough LT, Rudnick S, Price TH, Ballas SK, Collins M, Crowley JP, Kosmin M, Kruskall MS, Lenes BA, Menitove JE, Silbertein CE, Smith KJ, Wallas CH, Abels R, Von Tress M (1989) Increased preoperative collection of autologous blood with recombinant human erythropoietin therapy. N Engl J Med 321: 1163–1168 62. Goodnough LT, Skikne B, Brugnara C (2000) Erythropoietin, iron, and erythropoiesis. Blood 96: 823–833 63. Goodnough LT, Verbrugge D, Marcus RE, Goldberg V (1994) The effect of patient size and dose of recombinant human erythropoietin therapy on red blood cell expansion. J Am Coll Phys 179: 171–176 64. Hallberg L, Bengtsson C, Lapidus L, Lindstedt G, Lundberg PA, Hulten L (1993) Screening for iron deﬁciency: an analysis based on bone-marrow examinations and serum ferritin determinations in a population sample of women. Br J Haematol 85: 787–798 65. Hamstra RD, Block MH (1969) Erythropoiesis in response to blood loss in man. J Appl Physiol 27: 503–507 66. Hamstra RD, Block MH, Schocket AL (1980) Intravenous iron dextran in clinical medicine. JAMA 243: 1726–1729 67. Hastka J, Lasserre JJ, Schwarzbeck A, Reiter A, Hehlmann R (1996) Laboratory tests of iron status: correlation or common sense? Clin Chem 42: 718–724 68. Heiss MM, Tarabichi A, Delanoff C, Allgayer H, Jauch KW, Hernandez-Richter T, Mempel W, Beck KG, Schildberg FW, Messmer K (1996) Perisurgical erythropoietin application in anemic patients with colorectal cancer: A doubleblind randomized study. Surgery 119: 523–527 69. Henderson PA, Hillman RS (1969) Characteristics of iron dextran utilization in man. Blood 34: 357–375 70. Henry D, Abels R, Larholt K (1995) Prediction of response to recombinant human erythropoietin (Epoietin-a) therapy in cancer patients. Blood 85: 1676–1678 71. Hillman RS, Henderson PA (1969) Control of marrow production by the level of iron supply. J Clin Invest 48: 454–460 72. Kasper C, Terhaar A, Fossa A, Welt A, Seeber S, Nowrousian MR (1997) Recombinant human erythropoietin in the treatment of cancer-related anaemia. Eur J Haematol 58: 251–256

698

L. T. Goodnough

73. Kasper SM, Gerlich W, Buzello W (1997) Preoperative red cell production in patients undergoing weekly autologous blood donation. Transfusion 37: 1058–1062 74. Kasper SM, Lazansky H, Stark C, Klimek M, Laubinger R, Börner Y (1998) Efﬁcacy of oral iron supplementation is not enhanced by additional intravenous iron during autologous blood donation. Transfusion 38: 764–770 75. Kickler TS, Spivak JL (1988) Effect of repeated whole blood donations on serum immunoreactive erythropoietin levels in autologous donors. JAMA 260: 65– 67 76. Leggett BA, Brown NN, Bryant SJ, Duplock L, Powell LW, Halliday JW (1990) Factors affecting the concentrations of ferritin in serum in a healthy Australian population. Clin Chem 36: 1350–1355 77. Lipshitz DA, Cook JD, Finch CA (1994) A clinical evaluation of serum ferritin as an index of iron stores. N Engl J Med 290: 1213–1216 78. Ludwig H, Fritz E, Leitgeb C, Pecherstorfer M, Samonigg H, Schuster J (1994) Prediction of response to erythropoietin treatment in chronic anemia of cancer. Blood 84: 1056–1063 79. Ludwig H, Fritze, Kotzmann H, Hocker P, Gisslinger H, Barnas U (1990) Erythropoieitin treatment of anemia assocated with multiple myeloma. N Engl J Med 322: 1693–1699 80. Major A, Mathez-Loic F, Rohling R, Gautschi K, Brugnara C (1997) The effect of intravenous iron on the reticulocyte response to recombinant human EPO. Br J Haematol 98: 292–294 81. Mast AE, Blinder MA, Gronowski AM, Chumley C, Scott MG (1998) Clinical utility of the soluble transferrin receptor and comparison with serum ferritin in several populations. Clin Chem 44: 45–51 82. Mays T, Mays T (1976) Intravenous iron dextran therapy in the treatment of anemia occurring in surgical, gynecologic, and obstetric patients. Surg Gynecol Obstet 143: 381–384 83. Means RT (1999) Commentary: An anemia of chronic disease, after all? J Invest Med 47: 203–207 84. Megens JGN, Olijhoek G, Ziekenhuis V, Musto P, Nogarin L (1999) Comparison of efﬁcacy of oral versus intravenous iron supplementation with epoietin alfa treatment in arthroplasty patients. Br J Anaesth 82: 82–95 85. Menitove J (ed) (1999) Standards for blood banks and transfusion services, 19th edn. American Association of Blood Banks, Bethesda, MD, p76 86. Mercuriali F, Inghilleri G, Bifﬁ E, Colotti MT, Vinci A, Sinigaglia L, Gualtieri G (1997) Comparison between intravenous and subcutaneous recombinant human erythropoietin (epoietin alfa) administration in presurgical autologous blood donation in anemic rheumatoid arthritis patients undergoing major orthopedic surgery. Vox Sang 72: 93–100 87. Mercuriali F, Inghilleri, Bifﬁ E, Colotti MT, Vinci A, Oriani G (1998) Epoietin alfa in low hematocrit patients to facilitate autologous blood donation in total hip replacement: A randomized, double-blind, placebo-controlled, dose-ranging study. Acta Haematol 100: 69–76 88. Mercuriali F, Zanella A, Barosi G, Inghilleri G, Bifﬁ E, Vinci A, Colotti MT (1993) Use of erythropoietin to increase the volume of autologous blood donated by orthopedic patients. Transfusion 33: 55–60

Erythropoiesis, iron metabolism and iron supplementation

699

89. Messmer K (1996) Consensus statement: Using epoietin alfa to decrease the risk of allogeneic blood transfusion in the surgical setting. Semin Hematol 33 [Suppl 2]: 78–80 90. Mittman N, Sreedhara R, Mushnick R, Chattopadlyay S, Zelmonovic P, Vasagli M, Avram MM (1997) Reticulocyte hemoglobin content predicts functional iron deﬁciency in hemodialysis patients receiving rHuEPO. Am J Kidney Dis 30: 912–913 91. NFK – DOQI (1997) Clinical practice guidelines for the treatment of anemia of chronic renal failure. Am J Kidney Dis 30 [Suppl]: S192–S237 92. Nissenson AR, Lindsay RM, Swan S, Seligman P, Strobos J (1999) Sodium ferric gluconate complex in sucrose is safe and effective in hemodialysis patients: North American Trial. Am J Kidney Dis 33: 471–482 93. Österborg A, Boogaerts MA, Cimino R, Essers U, Holowiecki J, Juliusson G, Jäger G, Majman A, Peest D (1996) Recombinant human erythropoietin in transfusion-dependent anemic patients with multiple myeloma and nonHodgkin’s lymphoma – a randomized multicenter study. The European Study Group of Erythropoietin (Epoetin Beta) Treatment in Multiple Myeloma and Non-Hodgkin’s Lymphoma. Blood 87: 2675–2682 94. Olivieri NF, Brittenham GM, Matsui D, Berkautch M, Blendis LM, Cameron RG, McCelland RA, Liu PD (1995) Iron-chelation therapy with oral deferiprone in patients with thalassemia major. N Engl J Med 332: 918–922 95. Pascual J, Teruel JL, Liano F, Sureda A, Ortuna J (1992) Serious adverse reactions after intravenous ferric gluconate. Nephrol Dial Transplant 7: 271–272 96. Pincus T, Olson NJ, Russel IJ, Wolfe F, Harris ER, Schnitzer TJ, Boccagno JA, Krantz SB (1990) Multicenter study of recombinant human erythropoietin in correction of anemia in patients with rheumatoid arthritis. Am J Med 89: 161– 168 97. Price TH, Goodnough LT, Vogler WR, Sacher RA, Hellman RM, Johnston MFM, Bolgiano DC, Abels RI (1996) The effect of recombinant human erythropoietin on the efﬁcacy of autologous blood donation in patients with low hematocrits: A multicenter, randomized, double-blind, controlled trial. Transfusion 36: 29–36 98. Punnonen K, Irjala K, Rajamaki A (1997) Serum transferrin receptor and its ratio to serum ferritin in the diagnosis of iron deﬁciency. Blood 89: 1052–1057 99. Sans T, Boﬁll C, Joven J, Cliville X, Simo JM, Llobet X, Pero A, Galbany J (1996) Effectiveness of very low doses of subcutaneous recombinant human erythropoietin in facilitating autologous blood donation before orthopedic surgery. Transfusion 36: 822–826 100. Schreiber S, Howaldt S, Schnoor M, Nikolaus S, Bauditz J, Gasche G, Lochs H, Raedler A (1996) Recombinant erythropoietin for the treatment of anemia in inﬂammatory bowel disease. N Eng J Med 334: 619–623 101. Seligman PA, Schleicher RB (1999) Comparison of methods used to measure serum iron in the presence of iron gluconate or iron dextran. Clin Chem 45: 898–900 102. Silverberg DS, Blum M, Peer G, Kaplan E, Iaina A (1996) Intravenous ferric saccharate as an iron supplementation in dialysis patients. Nephron 72: 413–417 103. Silverberg DS, Iaina A, Peer G, Kaplan E, Levi BA, Frank N, Steinbruch S, Blum M (1996) Intravenous iron supplementation for the treatment of the anemia of

700

104. 105.

106. 107. 108.

109.

110.

111.

112. 113.

114. 115. 116.

117.

118.

119.

L. T. Goodnough moderate to severe chronic renal failure patients not receiving dialysis. Am J Kidney Dis 27: 234–238 Skikne BS, Cook JD (1992) Effect of enhanced erythropoiesis on iron absorption. J Lab Clin Med 120: 746–751 Sowade O, Sowade B, Brilla K, Franke W, Stephan P, Gross J, Scigalla P, Warnke H (1997) Kinetics of reticulocyte maturity fractions and indices and iron status during therapy with epoetin beta in cardiac surgery patients. Am J Hematol 55: 89–96 Sunder-Plassmann G, Horl WH (1995) Importance of iron supply for erythropoietin therapy. Nephrol Dial Transplant 10: 2070–2076 Sunder-Plassmann G, Horl WH (1997) Safety aspects of parenteral iron in patients with end stage renal disease. Drug Safety 17: 241–250 Suominen P, Punnonen K, Rajamaki A, Irjala K (1998) Serum transferrin receptor and transferrin receptor-ferritin index identify healthy subjects with subclinical iron deﬁcits. Blood 92: 2934–2938 Tarng DC, Hyang TP, Chen TW, Yang WC (1999) Erythropoietin hyporesponsiveness: from iron deﬁciency to iron overload. Kidney Int 55 [Suppl]: S107– S118 Taylor JE, Peat N, Porter C, Morgan AG (1996) Regular, low-dose intravenous iron therapy improves response to erythropoietin in haemodialysis patients. Nephrol Dial Transplant 11: 1079–1083 Thompson JA, Gilliland DG, Prchal J, Bennett JM, Larholt K, Nelson RA, Rose EH, Dugan MH (2000) Effects of recombinant human erythropoietin combined with granulocyte/macrophage colony-stimulating factor in the treatment of patients with myelodysplastic syndrome. Blood 95: 1175–1179 Tietz NW, Rinker AD, Morrison SR (1994) When is a serum iron really a serum iron? The status of serum iron measurements. Clin Chem 40: 546–551 Weisbach V, Skoda P, Rippel R, Layer G, Glaser A, Zingsem J, Zimmerman R, Eckstein R (1999) Oral or intravenous iron as an adjunct to autologous blood donation in elective surgery: A randomized, controlled study. Transfusion 39: 465–472 Weiss G (1999) Iron and anemia of chronic disease. Kidney Int 55 [Suppl]: S12–S17 Weiss G, Goodnough LT (2005) Anemia of chronic disease. N Engl J Med 352: 43–55 Weiss G, Houston T, Kastner S, Johrer K, Grunewald K, Brock JH (1997) Regulation of cellular iron metabolism by erythropoietin: activation of iron- regulatory protein and upregulation of transferrin receptor expression in erythroid cells. Blood 89: 680–687 Wingard RL, Parker RA, Ismail N, Hakim RM (1995) Efﬁciency of oral iron therapy in patients receiving recombinant human erythropoietin. Am J Kidney Dis 25: 433–439 Wood JK, Milner FA, Pathak UN (1968) The metabolism of iron dextran given as a total dose infusion to iron deﬁcient Jamaican subjects. Br J Haematol 14: 119–129 Woodman J, Shaw RJ, Shipman AJ, Edwards AM (1987) A surveillance program on a long-established product: Imferon (Iron Dextran BP). Pharmaceut Med 1: 289–296

Erythropoiesis, iron metabolism and iron supplementation

701

120. Zanen AL, Adriaansen HJ, Van Bommel EFH, Posthuma R, Th de Jong GM (1996) “Over saturation” of transferrin after intravenous ferric gluconate (Ferrlecit) in haemodialysis patients. Nephrol Dial Transplant 11: 820–824 Correspondence: Dr. Lawrence T. Goodnough, M.D., Professor, Pathology and Medicine, Stanford University, 300 Pasteur Drive, H-1402, Stanford, CA 94305-5626, USA, E-mail: [email protected]

Chapter 27

Are there risks for use of iron in cancer patients? P. Gascón Division of Medical Oncology, Hematology-Oncology Department (ICMHO), IDIBAPS, Hospital Clínic-Barcelona University, Barcelona, Spain

Many cancer patients with anemia require recombinant human erythropoietins (EPO) to correct their anemia. Cancer is considered an inﬂammatory process and it is associated with the anemia of chronic disease, secondary to inﬂammation. Over the last few years, the concept of functional iron deﬁciency has emerged and as a consequence the need to administer iron, in one or another form, to provide soluble, mobilized iron for the formation of new erythrocytes. Auerbach et al. (2004) reported the treatment of chemotherapy-related anemia in cancer patients (n = 157) with epoetin alfa and has shown that those patients who received iron i.v. (either bolus or continuous infusion) had a much better response than those patients who received p.o. iron (p < 0.02). In a more recent study Henry et al. (2007) reported that i.v. ferric gluconate signiﬁcantly improves response to epoetin alfa versus oral iron in anemic patients (n = 187) with cancer receiving chemotherapy (73% vs 46%, p = 0.009). With these data in mind, it is expected that the use of i.v. iron will increase dramatically among oncologists in the near future. But, is iron administration safe? The reality is that iron as a metal, it is a rather toxic substance (Table 1). Life in planet Earth is highly dependent on the use of iron and oxygen (O2) to drive multiple energy storage processes. It is well known that the combination of iron and oxygen, in nature, is a notorious corrosive process that, under physiologic conditions, yields harmful compounds such as rust and oxygen free radicals, Theil (2003). Ferritin is nature’s solution to the chemical problem. It will keep both compounds separated. Ferritin will store iron in a safe, soluble manner and will allow for regulated release of iron as needed for normal metabolism. Studies of oxidative DNA damage and iron by Theil (2003) have shown interesting ﬁndings. Using 8OHdG (a marker of oxidative damage), its levels have been found to be inversely associated with life expectancy. Furthermore, Nakano (2003) demonstrated in men and in women that the urinary concentration of 8OHdG increases with the serum ferritin level.

704

P. Gascón

Table 1. Natural mechanisms to reduce the toxic effects of iron Primary prevention: intracellular Fe: extracellular Fe: extracellular heme: cell-to cell:

Storage of iron ferritin (normal) and hemosiderin (overload) transferrin haptoglobin, hemopexin and albumin carrier proteins (transferritin, apoferritin); chelators (pyrophosphate)

Secondary prevention: Prevention of the formation of free radicals enzyme systems intracellular: SOD*, GPX**, catalase at lipid membranes: PL-hydroperoxide-GPX*** antioxidants: vitamins (A, C, E) cellular chelators of iron: citrate, ADP, pyrophosphates * superoxide dysmutase. ** glutathione peroxidase. *** phospholipid hydroperoxide glutathione peroxidase. Table 2. Iron risks • Potential: – ischemic heart disease21,24–26* – cancer11,22,23,28 – diabetes8,15,29 – infectious disease16,27,30 – neurodegenerative disorders19,20 • Known: – serious allergic reactions including anaphylaxis (<1%)1,2,33 anaphylaxis: iron dextrane (0,61%) iron gluconate (0,04%) – delayed hypersensitivity reaction (1,2%)1 – delayed arthalgia/myalgia (8%)1 – nausea, vomiting, constipation, asthenia (∼20%–42%)7 * reference number.

Iron potential risks A role for iron has been proposed in the pathogenesis of ischemic heart disease Sullivan (1981, 2001, 2003), Salonen (1992); cancer, Stevens (1994), Shaheen (2003), Kallianpur (2004), Toyokuni (2002); diabetes, Hua (2001), Lee (2004), Tuomainen (1997); infectious diseases, Marx (2002), Weinberg (1978), Sunder-Plassmann (1999) and, neurodegenerative disorders, Richardson (2004), Robson (2004) (Table 2).

Are there risks for use of iron in cancer patients?

705

Ischemic heart disease Chronic iron intake may have multiple side effects in association with inﬂammation, Sullivan (2004). These include the formation of highly toxic hydroxylradicals with subsequent lipid peroxidation. This effect has been postulated to be involved in the pathogenesis of atherosclerosis via induction of tissue damage and endothelial dysfunction and, thus a risk factor for stroke or acute myocardial infarction. Studies by Sullivan (1981) have suggested that iron depletion protects endothelial function, lowering stored iron results in substantial elevations of nitric oxide concentration in humans, potentially resulting in enhanced endothelium-dependent vasodilatation. Iron infusion acutely causes endothelial dysfunction in vivo in healthy humans as demonstrated by Sullivan (1981) and, endothelial cell injury by doxorubicin is augmented by iron-dependent mechanisms as reported by Xu (2005). These are the bases for recommending the avoidance of giving iron and doxorubicin together. However, a recent publication by Kaiserova (2006) challenges the putative role of iron and hydroxyl radicals in the oxidative stress-mediated cytotoxicity of doxorubicin and bleomycin using cell culture experiments in A549 cells.

Cancer A role for iron has been also proposed in the pathogenesis of cancer, Weinberg (1996, 1999) (Fig. 1). Several prospective studies in animals with excesive iron either injected or orally have demonstrated that there was an increased incidence of adenocarcinomas of colon-rectum, breast, hepatomas, mesotheliomas, renal carcinomas and sarcomas. In humans with hemochromatosis there is a 200 fold increase in hepatocarcinoma and an increased risk of non-hepatic carcinoma. Some tribes in Africa who cook with iron tools or drink beer that is brewed in nongalvanized steel drums, develop siderosis that has been associated with a 15-fold increase in developing cancer (Gangaidzo 1995; Weinberg 1996). In a 15 year-follow-up study on 8.500 men and women, Stevens et al. (1994) found that a moderate elevation of body iron level was associated with an increased risk of cancer occurrence and death. Among 379 men who developed cancer over the study period, the mean transferrin saturation at enrollment was 32.1% whereas among 2,908 who remained cancer-free was 30.7%; the difference for mortality was 32.3% among 233 deaths vs. 30.8% among 3,054 men not dying of cancer. Is iron carcinogenic? There are several pre-clinical reports that favour that postulate as Chen et al. (2006) and Knöbel et al. (2006) have demonstrated. Iron is essential for the proliferation of normal and neoplastic cells. H1299 lung cancer cells overexpressing IRP1 (iron regulatory protein 1) show a dramatic impairment of the capacity of the cells to form tumor

706

P. Gascón

Fig. 1. Suggested mechanisms of iron carcinogenesis. Two mechanisms have been proposed. A direct damaging effect on the DNA and, an indirect mechanism by the formation of reactive oxygen species (ROS), free radicals

xenografts in nude mice suggesting a link between IER/IRP and cancer, Chen et al. (2006). A recent report by Knöbel et al. (2006) determined a genotoxic effect of iron on non-transformed and in pre-neoplastic human colon cells by the same group that has previously reported a genotoxic effect of ferric iron on the colorectal cell line HT29.

Infectious disease Microorganisms have developed multiple strategies to acquire iron as has been postulated by Genco et al. (2001). Therapeutically administered iron that is not immediately cleared from the circulation by the RES will be captured by microorganisms for their growth and proliferation. It has been shown, Gordon (2003) that iron negatively affects the activity of cellmediated immune effector function. In vitro and in vivo observations indicate that iron loading of macrophages reduces their responsiveness to interferon-gamma, an important cytokine to control acute infections. Oexle et al. (2003) demonstrated that following treatment with iron salts or transferrin-bound iron, the induction of interferon-gamma-mediated immune effector pathways is impaired. In addition, macrophages loaded with iron lose their ability to kill intracellular organisms such as Legionella, Listeria and

Are there risks for use of iron in cancer patients?

707

Ehrlichia also by interferon-gamma mediated pathways, Weiss (2002). Thus, it is now well documented that, by various mechanisms, iron therapy may increase the risk of infectious complications, even septicaemia, in anemia secondary to inﬂammation or chronic anemia. This association is the main reason to discourage people to use iron while a patient has acute infection.

Known iron risks It is well known the side effects of oral iron: constipation, gastritis, nauseas, vomiting, asthenia (Table 2). Nowadays, many patients are receiving parenteral iron for optimal correction of their anemia with or without epoetins. Available parenteral iron therapy includes: iron dextran, iron gluconate, and iron sucrose. By far, the most commonly used has been the dextran form, although, the other two preparations are associated with much less side effects. It is well know that the use of intravenous iron has been associated to serious adverse effects (SAE) such as anaphylactic reactions. Have been the occurrence of these reactions over-exagerated? It is then very important, due to the increased use of iron among oncologists and nephrologists, to analyze what is in the literature with regards its intravenous administration and to determine the evidence of its side-effects. The risk of anaphylactic reactions occurs only in <1%. The risks for allergic reactions, including full-blown anaphylaxis, is much lower with iron gluconate (0,04%) than with iron dextran (0,61). Lacan et al. (2005) have demonstrated that parenteral iron is relatively safe. This was a one single institution 5 year study. A total of 121 patients received 444 infusions of parenteral iron over this period. Iron dextran was the most commonly used product (85 patients), iron sucrose (2 patients) and iron gluconate (34 patients). Overall adverse event rates per patient with iron dextran and iron gluconate were 16.5% and 5.8%, respectively (P = .024). Test doses of iron dextran were used 88% of the time for initial infusions of iron dextran. All adverse events for all parenteral iron products were mild or moderate. There were no serious adverse events and no anaphylaxis was observed. In the study by Auerbach (2004) only one out of 78 patients (1,2%) presented a delayed hypersensitivity reaction manifested as chest and back pain, nausea, vomiting, ﬂushing and hypotension (the event occurred immediately after the test dose of iron dextrane). With regards other side effects in the same study, an 8% events of delayed arthralgia/myalgia was reported. In a recent study by Henry et al. (2007) giving i.v. ferric gluconate vs. oral iron vs. no iron with epoetin alfa in 187 cancer patients with anemia receiving chemotherapy, there were no signiﬁcant differences in SAE among the three groups of patients.

708

P. Gascón

Iron safety and tolerability in cancer patients have been found similar to the observed in iron deﬁciency anemia and in chronic renal failure patients.

Discusion Increased iron availability either in blood or in tissues has been associated with an increased risk for the development of several types of tumors and may promote carcinogenesis as it has been discussed previously. Although animal studies using high doses of iron have suggested that it may promote tumour growth, these ﬁndings have not been reﬂected in clinical practice and probably do not apply to our cancer population. Furthermore, the doses of iron given in oncology are relatively low and are given for a short period of time (Table 3). It appears that except for the allergic/anaphylactoid reactions which is an acute phenomenon, stimulation of invasive organisms leading to bacteriemia and sepsis or enhancing of tumour growth may be related to Table 3 Study author

Iron formulation

Dose

Hematopoietic Response (%)

QoL

Auerback et al. (2004) Vandebroek et al. (2006)

Iron dextrane Unknown

68% iv vs. 36% po

↑

Higher in the iv groupthan the po group

ND

Hedenus et al. (2007)

Iron sucrose

93% iv vs. 53% no iron

ND

Henry et al. (2007)

Ferric Gluconate

100 mg i.v. bolus each visit* 200 mcg i.v. Q3W or 200 mcg i.v. in two doses within a 3W period 100 mg iv QW for W 0–6 and then 100 mg iv for W 8–14 125 mg i.v. weekly for 8 weeks**

73% iv vs: 46% po

ND

• Auerbach et al. (J Clin Oncol 2004; 22:1301). Eligibility: Hb ≤ 10.5 g/dL. Iron was given i.v. bolus in each visit to the calculated dose for iron replacement. Iron (ferrous sulphate) was given 325 mg p.o. BID daily. EPO: EPO alpha 40,000 U/QW. • Vandebroek et al. (J Clin Oncol 2006; 24:18S, abstract #8612). Eligibility: Hb < 11 g/dL. EPO: Darbepoetin alfa 500 mcg Q3W. • Hedenus et al. (Leukemia 2007; 21:627). Elegibility: Hb < 11 g/dL. EPO: EPO beta 30,000 U/QW. • Henry et al. (The Oncologist 2007; 12:231). Eligibility: Hb < 11 g/dL. Iron was given i.v. bolus weekly for 8 weeks. Iron (ferrous sulphate) was given 325 mg p.o. TID daily. EPO: EPO alpha 40,000 U/QW. ND: not done; QoL: Quality of Life.

Are there risks for use of iron in cancer patients?

709

chronic exposure and accumulation of tissue iron. There has been a safety concern on the issue of potential “iron overload” in patients receiving i.v. iron and, the possible risk of developing cancer or infections as a consequence. In Henry et al. (2007) the highest serum ferritin levels were 3,586 ng/ml in the ferric gluconate arm vs. 6,186 in the p.o. iron arm vs. 3,830 in the non-iron arm. Based on these data, it does not appear that i.v. iron in cancer patients with anemia drives the body into an iron-overload situation. Most of the literature addressing these two important side effects (cancer and infections) in iron-overloaded patients comes from hemochromatosis or hemodialysis patients. Hepatocarcinoma has been associated in patients with hemochromatosis and, typically only in patients who ﬁrst developed cirrhosis, Kowdley (2004). Similarly, very few data support any increase in the rate of common or uncommon infections, Besarab (1999). Furthermore, Kalantar-Zadeh et al. (2005) have found no increased death rate for serum ferritin levels as high as 1,200 ng/ml in a study conducted with more than 58,000 hemodialysis patients to analyze any association between iron and mortality.

Conclusion In cancer associated anemia, we may give iron i.v. 5–8 times (100–200 mg) during the full course of treatment. The amount of total iron accumulated is minimal and the iron that reaches the tumour should be extremely low. With this in mind, the risks for cancer growth or infections due to iron administration, in particular i.v. administration, have to be considered, at this stage, probably more an academic point than a real issue. Until, prospective studies are done to clarify the potential harmful effect of using iron, we should adopt an approach that takes into account the beneﬁts/hazard ratio of iron therapy on cancer and on the well-being of our patients.

References 1. Auerbach M, Ballard H, Truout JR, McIlwain M, Ackerman A, Bahrain H, Balan S, Barker L, Rana J (2004) Intravenous iron optimizes the response to recombinant human erythropoietin in cancer patients with chemotherapy-related anemia: a multicenter, open-label, randomized trial. J Clin Oncol 22(7): 1301–1307 2. Besarab A, Frinak S, Yee J (1999) An indistinct balance: The safety and efﬁcacy of parenteral iron therapy. J Am Soc Nephrol 10: 2029–2043 3. Chen G, Fillebeen C, Wang J, Pantopoulos K (2007) Overexpression of iron regulatory protein 1 suppresses growth of tumor xenografts. Carcinogenesis 28(4): 785–91 4. Gangaidzo IT, Gordeuk VR (1995) Hepatocellular carcinoma and African iron overload. Gut 37: 727–730

710

P. Gascón

5. Genco CA, White-Dixon D (2001) Emerging strategies in microbial haem capture. Mol Microbiol 61: 1830–1839 6. Gordon S (2003) Alternative activation of macrophages. Nat Immunol Rev 3: 133–146 7. Henry DH, Dahl NV, Auerbach M, Tchekmedyian S, Laufman LR (2007) Intravenous ferric gluconate signiﬁcantly improves response to epoetin alfa versus oral iron or no iron in anemic patients with cancer receiving chemotherapy. Oncologist 12(2): 231–242 8. Hua NW, Stoohs RA, Facchini FS (2001) Low iron status and enhanced insulin sensitivity in bacto-ovo vegetarians. Br J Nutr 86: 515–519 9. Kalantar-Zadeh K, Regidor DL, McAllister CJ, Michael B, Warnock DG (2005) Time-dependent associations between iron and mortality in hemodialysis patients. J Am Soc Nephrol 16(10): 3070–80 10. Kaiserova H, den Hartog GJ, Simunek T, Schroterova L, Kvasnickova E, Bast A (2006) Iron is not involved in oxidative stress-mediated cytotoxicity of doxorubicin and bleomycin. Br J Pharmacol 149(7): 920–930 11. Kallianpur AR, Hall LD, Yadav M, Christman BW,Dittus RS Haines JL, Parl FF, Summar ML (2004) Increased prevalence of the HFE C282Y hemochromatosis allele in women with breast cancer. Cancer Epodemiol Biomarkers Prev 13: 205–212 12. Knobel Y, Weise A, Glei M, Sedt W, Claussen U, Pool-Zobel BL (2007) Ferric iron is genotoxic in non-transformed and preneoplastic human colon cells. Food Chem Toxicol 45(5): 804–811 13. Kowdley KV (2004) Iron, hemochromatosis, and hepatocellular carcinoma. Gastroenterol 127: S79–S86 14. Laman CA, Silverstein SB, Rodgers GM (2005) Parenteral iron therapy: a single institution’s experience over a 5-year period. J Natl Compr Canc Netw 3(6): 791–795 15. Lee DH, Folsom AR, Jacobs DR Jr (2004) Dietary iron intake and type 2 diabetes incidence in postmenopausal women:the Iowa Women’s Health Study. Diabetologia 47: 185–194 16. Marx JJ (2002) Iron and infection: competition between host and microbes for a precious element. Best Pract Res Clin Haematol 15: 411–426 17. Nakano M, Kawanishi Y, Kamohara S, Uchida Y, Shiota M, Inatomi Y, Komori T, Miyazawa K, Gondo K, Yamasawa I (2003) Oxidative DNA damage(8hydroxydeoxyguanosine) and body iron status: a study on 2507 healthy people. Free Radic Biol Med 35: 826–832 18. Oexle HA, Kaser A, Most J, Bellmann-Weiler R, Werner ER, Werner-Felmayer G (2003) Pathways for the regulation of interferon-gamma-inducible genes by iron in human monocytic cells. J Leukoc Biol 74: 287–294 19. Richardson DR (2004) Novel chelators for central nervous system disorders that involve alterations in the metabolism of iron and other metal ions. Ann N Y Acad Sci 1012: 326–341 20. Robson KJH, Lehmann DJ, Wimhurst VLC, Livesey KJ, Combrinck M, Merryweather-Clarke AT, Warden DR, Smith AD (2004) Synergy between the C2 allele of transferring and the C282Y allele of the haemochromatosis gene (HFE) as risk factors for developing Alzheimer’s disease. J Med Genet 41: 261– 265

Are there risks for use of iron in cancer patients?

711

21. Salonen JT, Nyyssonen K, Korpela H, Tuomiletho J, Seppanen R, Salonen R (1992) High stored iron levels are associated with excess risk of myocardial infarction in eastern Finnish men. Circulation 86: 803–811 22. Shaheen NJ, Silvermann LM, Keku T, Lawrence LB, Rohlfs EM, Martin CF, Galanko J, Sandler RS (2003) Association between hemochromatosis (HFE) gene mutation carrier status and the risk of colon cancer. J Natl JCancer Inst 95: 154–159 23. Stevens RG, Graubard BI, Micozzi MS, Keriishi K, Blumberg BS (1994) Moderate elevation of body iron level and increased risk of cancer occurrence and death. Int J Cancer 56: 364–369 24 Sullivan JL (1981) Iron and the sex differences in heart disease risk. Lancet 1: 1293–1294 25. Sullivan JL (2003) Are menstruating women protected from heart disease because of, or in spite of, estrogen? Relevance to the iron hypothesis. Am Heart J 145: 190–194 26. Sullivan JL (2004) Is stored iron safe? J Lab Clin Med 144(6): 280–284 27. Sunder-Plassmann G, Patruta SI, Hörl WH (1999) Pathobiology of the role of iron in infection. Am J Kidney Dis 34: 25–29 28. Toyokuni S (2002) Iron and carcinogenesis: From Fenton reaction to target genes. Redox Rep 7: 189–197 29. Tuomainen TP, Nyyssonen K, Salonen R, Tervahauta A, Korpela H, Lakka T, Kaplan GA, Salonen JT (1997) Body iron stores are associated with serum insulin and blood glucose concentrations. Population study in 1,013 eastern Finnish men. Diabetes Care 20: 426–428 30. Weinberg ED (1978) Iron and infection. Microbiol Rev 42: 45–46 31. Weinberg ED (1996) The role of iron in cancer. Eur J Cancer Prev 5: 19– 36 32. Weinberg ED (1999) Iron therapy and cancer. Kidney Intl 55 (S69): 131– 134 33. Weiss G, Gordeuk VR (2005) Beneﬁts and risks of iron therapy for chronic anaemias. Eur J Clin Invest 35 [Suppl 3]: 36–45 34. Theil EC (2003) Ferritin: at the crossroads of iron and oxygen metabolism. J Nutr 133 [5 Suppl 1]: 1549S–1553S 35. Xu X, Persson HL, Richardson DL (2005) Molecular pharmacology of the interaction of anthracyclines with iron. Molecular Pharmacol 68: 261–271 Correspondence: Prof. Dr. Pere Gascón, Division of Medical Oncology, HematologyOncology Department (ICMHO), Hospital Clínic, Villarroel 170, escalera 2, planta 5, Barcelona, Spain, E-mail: [email protected]

Chapter 28

Metabolic and physiologic effects of rhEPO in anemic cancer patients K. Lundholm and P. Daneryd Surgical Metabolic Research Laboratory at Lundberg Laboratory for Cancer Research, Department of Surgery, Sahlgrenska University Hospital, Göteborg University, Göteborg, Sweden

Anemia in progressive cancer Anemia in cancer disease is usually regarded a consequence of disease progression explained by several factors, but may as well be a contributing factor and promoter behind the progression of wasting in cancer patients (Mittelman 1996; Moliterno and Spivak 1996; Bertero and Caligaris-Cappio 1997). This unusual concept goes back to our previous studies, which have revealed both increased adrenergic activity and sensitivity in cancer patients (Drott et al. 1989), a phenomenon probably related to the cardiovascular system as indicated by either elevated resting heart rate and increased excretion of stress hormones in cancer patients with progressive weight loss, including altered cardiac adrenoceptor expression in heart tissue from tumor bearing animals (Drott et al. 1987; Drott et al. 1988; Hyltander et al. 1991). Results from our laboratory have, thus, suggested anemia as a promoter behind increased energy expenditure in cancer patients, although anemia is also a consequence of the host systemic inﬂammation in cancer (Fig. 1). However, it has never been evaluated whether anti-inﬂammatory treatment in itself, by reducing cytokine production, would be sufﬁcient to prevent or attenuate anemia to any signiﬁcant extent in patients with progressive cancer (Nieken et al. 1995). Our recent data suggest that anti-inﬂammatory treatment can attenuate resting energy expenditure in weight-losing cancer patients (Lundholm et al. 2004b), in agreement with ﬁndings that provision of ﬁsh oils attenuate systemic inﬂammation in patients with pancreatic carcinoma leading to improved status (Barber et al. 1999). Thus, inﬂammation or the consequences of inﬂammation such as anemia can promote energy drain of the host in association with progressive wasting. This phenomenon should undoubtedly lead to more rapid deterioration of body composition, which in turn would hamper functions such as standing or walking. Such a compromise may theoretically decrease quality of life in cancer

714

K. Lundholm and P. Daneryd 3 7,5 35

REE (Kcal/kg/day)

32 ,5 30 27 ,5 25 22 ,5 20 17 ,5 15 12 ,5 60

80

100

120

140

160

180

200

220

Hemoglobin conc. (g/L) Fig. 1. Relationship between blood hemoglobin concentration and resting energy expenditure in unselected cancer patients (r = 0.42, p < 0.001)

patients with sustained inﬂammation due to both systemic and local tumor factors with anemia as the consequence.This concept has support in the ﬁndings that cancer patients have reduced self-estimated quality of life. Thus, treatment leading to restitution of blood hemoglobin may improve the situation and reduce the subjective feeling of fatigue in a large number of cancer patients, particularly when anemia is the combined result of disease and treatment. Our recent research has, therefore, focused on the question whether rhEPO treatment attenuates anemia in unselected cancer patients on systemic anti-inﬂammatory drugs as NSAID; and if so, whether alleviation of anemia in cancer patients with progressive disease will protect or improve physical functioning evaluated by exercise testing and concomitant measurements of whole body metabolism and energy efﬁciency during a near maximum work load (Daneryd et al. 1998). Self-reported quality of life in cancer patients is a widely used approach to evaluate this clinically important endpoint (Glaspy 1997). However, perception of quality of life is highly subjective, and adaptation to disease state or to normal physiologic changes (as in normal aging) are common phenomena. We have observed that young individuals may regard themselves normally ﬁt even after major intestinal resections due to Crohn’s disease, despite signiﬁcantly reduced functional ability (Brevinge et al. 1995). Such results may therefore imply that it is important to evaluate effectiveness of medical treatment in both objective and subjective dimensions. We have

Metabolic and physiologic effects of rhEPO in anemic cancer patients

715

accordingly combined the use of both exercise testing (treadmill walking) as an objective measure of integrated performance for all our patients, as well as instruments for subjective measures of HRQL: one general instrument for assessment of health and performance (SF-36) and one more cancer-speciﬁc (EORTC QLC-C30) (Lindholm et al. 2004), on the role anemia has to attenuate physical performance and HRQL in cancer.

rhEPO treatment More than eleven hundred weight-losing cancer patients, mainly with gastrointestinal cancer, were evaluated in our laboratory by measurement of body composition and energy metabolism during both resting and exercise conditions. Such unselected weight-losing cancer patients have, as a group, elevated resting energy expenditure, although it is still an unresolved matter whether energy expenditure is truly elevated or merely represents a phenomenon of insufﬁcient adaptation to fully account for decreased food intake and weight loss in such patients. Nevertheless, energy balance eventually becomes increasingly negative, and any physical strain on such patients will impose an extra metabolic burden to the resting situation, particularly when exercise or physical activities are performed in the presence of anemia. In this context we found it interesting to consider to what extent resting wholebody metabolism and particularly exercise-related metabolism is abnormal in weight-losing cancer patients with anemia compared to weight-losing individuals without anemia. The results in our previous studies demonstrate that both anemia and inﬂammation cause signiﬁcant alterations in resting metabolism. We have also speculated to what extent restricted physical functioning in cancer patients can be explained by nutritional disorders, which secondarily lead to alterations in body composition. One of the largest components of body constituents, lean body mass (LBM) of which the skeletal muscles represent a major constituent, may well be a limiting factor for exercise capacity (Fig. 2). LBM is thus a measure of the protein content in the body. This mass, including skeletal muscles is susceptible to decreased food intake. Thus, undernutrition and not primarily anemia may be a major restricting factor for walking ability or physical functioning in cancer patients. However, there is a correlation between exercise and hemoglobin concentrations in blood, actually suggesting that exercise capacity of patients starts to decline whenever subnormal blood hemoglobin concentrations appear (Fig. 3). From this perspective it should, therefore, not be appropriate to allow patients to become anemic at all, since this should imply deterioration of patients’ physical functioning (Fig. 4). It is obvious that physical capacity in cancer patients is directly related to VO2 as it is in normal individuals according to a large body of information emerging from the area of sport medicine (Fig. 5).

716

K. Lundholm and P. Daneryd

Maximum exercise capacity (watt)

250 225 200 175 150 125 100 75 50 25 0 -25 25000

35000

45000

55000

65000

75000

LBM (g) Fig. 2. Relationship between maximum exercise capacity and lean body mass (LBM) in unselected cancer patients (r = 0.60, p < 0.001)

225

Max exercise power (watt)

200 175 150 125 100 75 50 25 0 -25 60

80

100

120

140

160

180

200

220

Blood hemoglobin conc. (g/L) Fig. 3. Relationship between blood hemoglobin concentration (g/L) and maximum exercise capacity (watt) in unselected cancer patients (r = 0.36, p < 0.0001)

Metabolic and physiologic effects of rhEPO in anemic cancer patients

717

120 100

Watt

80 60 40 20

Hb(<89)

Hb(90-99)

Hb(100-109)

Hb(110-119)

Hb(130-160)

Hb(120-129)

0

Fig. 4. Maximum exercise capacity in unselected cancer patients when grouped according to blood hemoglobin concentration (Hb) between 160 g/L to below 89 g/L (ANOVA: p < 0.001)

3500

mL oxygen uptake/min

3000

2500

2000

< 9 5 g/l > 1 2 5 g /l

1500

1000

500

0 -25

0

25

50

75

100

125

150

175

200

225

250

Watt Fig. 5. Relationships between maximum exercise capacity (watt) and near maximum oxygen uptake (mL/min) in cancer patients with hemoglobin concentration above 125 g/L and below 95 g/L respectively (r = 0.92, p < 0.001)

718

K. Lundholm and P. Daneryd

Thus, it appears that anemic patients had a strict relationship between exercise and VO2 as in patients with normal blood hemoglobin covering a large span of power output during exercise (Fig. 4). Therefore, anemia will restrict further expansion of power output beyond the limit deﬁned by a certain hemoglobin level (Fig. 3). It is then tempting to conclude that blood hemoglobin determines to what extent increasing exercise is possible. Consequently, an important clinical question would be whether normalization of anemia, or rather the prevention of the development of anemia in cancer patients is associated with preserved exercise capacity and metabolic efﬁcacy in cancer patients with progressive disease. And if so, whether this improvement is translated into improved HRQL.

Study design With the above purpose in mind, one hundred and eight cancer patients, mainly suffering from gastrointestinal malignancy, were randomized for treatment with either oral indomethacin alone (50 mg × 2 per day) or the combination of indomethacin (50 mg × 2 per day) and subcutaneous injections of rhEPO (Eprex®, Janssen-Cilag, Stockholm, Sweden, range 12,000–30,000 IU per week) at three occasions per week. RhEPO injections were continued until blood hemoglobin concentration became normalized or remained within reference values for healthy individuals (Daneryd et al. 1998). RhEPO treatment was thus instituted when blood hemoglobin concentration fell below the lower limits of normal hemoglobin concentrations (12.8 g/dL for men, 12.0 g/dL for women) and was maintained until hemoglobin was normalized during iterated courses of therapy. Hence, some patients randomized to the treatment group never received rhEPO, or received a low frequency of drug provision. The treatment was continued until the physician regarded the hemoglobin value as stable within deﬁned values. RhEPO doses were escalated in patients who had inadequate hemoglobin response (Kasper et al. 1997). Some patients needed maintenance doses of rhEPO to keep hemoglobin within normal values, while in others treatment could be withdrawn for a signiﬁcant period of time before recurrence of anemia. Oral anti-inﬂammatory treatment continued until death or until patients were unable for other reasons to take the tablets (Lundholm et al. 1994). None of the study and control patients received blood transfusions or any other speciﬁc tumor treatment besides general palliative support, including the cyclo-oxygenase inhibition by indomethacin according to good clinical practice at our institution. Patients were stratiﬁed to treatment groups by a computer based algorithm for tumor type, previous treatment, expected survival, age, sex, nutritional state, liver and kidney function tests, systemic inﬂammation and blood hemoglobin concentration before the start of treatment (Pocock and Simon 1975).

Metabolic and physiologic effects of rhEPO in anemic cancer patients

719

All patients were treated according to our protocol and followed until death. Measurements were performed at inclusion and repeatedly every second month. Treatment response variables included heart rate, blood pressure, body temperature, respiratory rate, energy expenditure, food intake, urinary excretions, exercise test, and HRQL.

Exercise testing Exercise testing started with the patient standing on the treadmill for one minute and then walking 1.5 km/h for two minutes. The test continued with walking at 1.5 km/h at a 12% elevation for one minute, and the treadmill speed was increased 0.1 km/h every 10th second until the person ﬁnished the test. The speed at which patients ﬁnished the test was deﬁned as maximal exercise power. Oxygen uptake and carbon dioxide production were measured in each patient during exercise (Daneryd et al. 1998).

Hemoglobin Blood hemoglobin concentration improved signiﬁcantly in patients on rhEPO treatment, while it declined in the control group (Fig. 6). In general, results demonstrated that rhEPO has the power and efﬁciency to protect unselected cancer patients with solid gastrointestinal tumors and progressive disease from becoming anemic despite severe paraneoplastic symptoms of disease.

Exercise capacity Body weight and resting energy expenditure became signiﬁcantly lower among control patients compared to rhEPO treated patients, while food intake, body fat and LBM were not statistically different between the two groups during follow-up. As we expected, prevention of anemia in our study patients was associated with a signiﬁcantly higher exercise capacity, suggesting that patients’ functional ability and oxygen transporting system were preserved despite disease progression (Fig. 7). Maximum exercise capacity, ventilation, whole-body oxygen uptake and carbon dioxide production during exercise were all signiﬁcantly higher in erythropoietin treated patient, compared to controls. Derived values of exercise per LBM (watt/kg) were signiﬁcantly lower in controls and oxygen uptake per watt produced (ml 02/ watt) was signiﬁcantly higher in the controls compared to study patients during maximum exercise (Fig. 8). This suggests that metabolic efﬁciency was improved in patients subjected to prevention of anemia. It is important to emphasize that rhEPO treated patients had a

720

K. Lundholm and P. Daneryd 150 145 140

Hemoglobin (g/L)

135 130 125 120 115 110 105 100 0

2-4

Follow-up (months)

6-8

10-30 In do Epo

Fig. 6. Blood hemoglobin during follow-up in cancer patients randomized to receive either oral intake of indomethacin (50 mg × 2/day, indo) alone or indomethacin in combination with rhEPO (Epo) (12,000–20,000 U/week) (ANOVA: p < 0.001)

similar degree of inﬂammation as the controls. Thus, metabolic abnormalities in our control patients were most likely a matter of adaptation to anemia, hypoxia and inefﬁcient metabolism compared to nonanemic rhEPO-treated patients.

Health-related quality of life (HRQL) Two self-administrated questionnaires were used to measure HRQL: Medical Outcomes Study Short-Form Health Survey (SF-36) and EORTC QLQ-C30 (+3) Questionnaire. The SF-36 contains 36 questions that assess eight aspects of HRQL (physical functioning, role-physical functioning, bodily pain, general health, vitality, social functioning, role emotional functioning, and mental health). The EORTC QLQ-C30 is a 30-item core questionnaire for cancer patients that contains ﬁve functional scales (physical functioning, role functioning, emotional functioning, cognitive functioning, and social functioning), as well as three symptom scales (fatigue, nausea/vomiting, and pain), six single items (dyspnoea, insomnia, loss of appetite, constipation, diarrhea,

Metabolic and physiologic effects of rhEPO in anemic cancer patients

721

140

Maximum exercise capacity (w)

120

100

80

60

40

20

0

0

2-4

Follow-up (months)

6-8

10-30 Indo Epo

Fig. 7. Maximum exercise capacity (watt) in the same patients as illustrated in Fig. 6 (ANOVA: p < 0.0001)

and ﬁnancial difﬁculties) and two global questions concerning health status and overall HRQL. The evaluation of treatment of disease-related anemia showed some discrepancies between objective and subjective measurements (Fig. 4 vs 9) and different relationships between each instrument for evaluation of HRQL and exercise power, respectively. Unexpectedly few of the items in either instrument for HRQL appeared statistically different in rhEPO treated patients, despite the fact that changes in hemoglobin concentrations, even within the narrow ranges of normality, represent a limiting factor for exercise capacity relevant for physical functioning in daily activities. The objective measure of the exercise capacity in our study could detect an early decrease in exercise performance even within normal ranges of hemoglobin concentration (Fig. 4), while the subjective assessments only sensed reduction in physical functioning (PF) below the normal range of hemoglobin concentration (Fig. 9). Self-assessment of patients’ General Health was proportional to performance expressed as exercise power (Fig. 10), whereas patients could obviously lose up to one third of their exercise power without reporting a change in selfassessment of physical functioning (Fig. 11).

722

K. Lundholm and P. Daneryd 3

Watt/kg

2

1

0

0

2-4

6-8

Follow-up (months)

10-30 In do Epo

Fig. 8. Maximum exercise capacity (watt) per lean body mass (LBM, kg) in the patients illustrated in Fig. 7 (ANOVA: p < 0.003) 80 70 60

PF

50 40 30 20 10

gr5(<89)

gr4(90-99)

gr3(100-109)

gr2(110-119)

gr1(120-129)

gr0(130-160)

0

Fig. 9. Self-reported physical functioning (PF) in unselected cancer patients grouped according to blood hemoglobin concentration. Observe that the patients did not appreciate any difference within normal blood concentrations (>120 g/L) as compared to differences in objective measurements of exercise capacity shown in Fig. 4 (ANOVA: p < 0.01)

Metabolic and physiologic effects of rhEPO in anemic cancer patients

723

120

100

GH-SF36

80

60 40

20

0

-20 -25

0

25

50

75

100

125

150

175

200

225

Max exercise power (watt) Fig. 10. Relationship between maximum exercise power and self-reported general health (GH) in unselected cancer patients (r = 0.37, p < 0.001)

120

100

PF-SF36

80

60 40

20

0

-20 -25

0

25

50

75

100

125

150

175

200

225

Max exercise power (watt) Fig. 11. The relationship between maximum exercise power (watt) and self-reported physical functioning (PF) in unselected cancer patients (r = 0.49, Y = 0.58x − 0.002x2 + 27.5, p < 0.001)

724

K. Lundholm and P. Daneryd

Subjectively scored physical functioning and general health were signiﬁcantly predicted by C-reactive protein (CRP), maximum exercise power, whole body fat, and caloric intake. The only principal difference between physical functioning and general health as predictors, was that alterations in whole body fat accounted for alterations in physical functioning, but not in the scoring of general health. Maximum exercise capacity was signiﬁcantly predicted by LBM, whole body fat, and plasma insulin-like growth factor.

Concluding remarks Studies on cancer patients have demonstrated that rhEPO can overcome both cytotoxic and disease-related depression of erythropoiesis (Eguchi 1995; Mittelman 1996). Cytotoxicity following radiation and chemotherapy is, thus, a well-deﬁned indication for rhEPO treatment. Pure disease-related anemia is more complex, with fundamentally different mechanisms, such as occult bleeding, nutritional impairment, as well as tumor-related depression of blood formation (Moliterno and Spivak 1996). Conventional anti-inﬂammatory drugs are effective to decrease and attenuate biochemical and tissue alterations secondary to both local and systemic inﬂammation. Therefore, it may be conceptually possible to protect patients from being anemic by provision of anti-inﬂammatory treatment in those situations where systemic inﬂammation dominates. However, a number of observations indicate that systemic inﬂammation in cancer patients is multifactorial and may, therefore, be drug resistant. Erythropoietin treatment should, thus, be a superior and more direct approach to treat anemia compared to anti-inﬂammatory intervention. Conventional anti-inﬂammatory treatment by steroids and cyclo-oxygenase inhibitors may be biochemically, although not clinically, efﬁcient since treatment effectiveness may be blurred by induction of occult gastrointestinal bleeding and other complications in a substantial number of patients. Interestingly, exercise capacity per LBM was more preserved in rhEPO treated patients, and oxygen uptake per watt produced during exercise was signiﬁcantly higher in control patients, suggesting that muscle contractions and mobility were less efﬁcient in anemic cancer patients, although little is known about metabolic efﬁciency in walking and exercise among humans and particularly in sick patients. Our observations are compatible with a less efﬁcient oxygen transporting system in anemic cancer patients, imposing a larger energy strain on the cardiovascular system during anemia when normalized to a given work load. Thus, an efﬁcient organism should consume less energy during both resting and moving conditions. Being more energy efﬁcient should inevitably mean a less pronounced wasting of the host as long as appetite does not compensate for energy deﬁcits due to subnormal

Metabolic and physiologic effects of rhEPO in anemic cancer patients

725

energy states. This concept is also supported by our previous observations, which imply that anorectic cancer patients have less pronounced adaptation in resting energy expenditure along with decreased food intake (Hyltander et al. 1991). Weight loss and hypermetabolism is thus not compensated for by increased appetite. A loss of up-regulation in dietary intake may explain in part development of cancer cachexia (Bosaeus et al. 2001). The dietary macronutrient composition does not change in most cancer patients, but long-term COX-treatment attenuated increased resting energy metabolism and improved appetite in weight-losing cancer patients, due to decreased systemic inﬂammation (Lundholm et al. 2004a). Individualized nutritional support can protect the integrated metabolism and metabolic function as well as increase survival in these weight-losing cancer patients (Lundholm et al. 2004b). The role of anemia in skeletal muscle metabolism and function in patients with end-stage renal disease has been studied by several authors (Barany et al. 1991; Moore et al. 1993; Park et al. 1993; Marrades et al. 1996a; Thompson et al. 1996). Patients were re-evaluated after treatment with rhEPO in order to correct anemia. Uremic patients demonstrated a lower aerobic capacity as reduction in power output during dynamic exercise and at onset of metabolic acidosis compared with untreated controls (Thompson et al. 1996). The availability of oxygen was not an exclusive limiting factor, since increased hemoglobin concentration did not entirely explain improvements in metabolism in uremia (Marrades et al. 1996b). Thus, oxygen transporting systems seemed to remain below normal in uremia indicating undeﬁned alterations in skeletal muscle cell function when adapted to chronic anemia and uremia. By contrast, cancer related anemia in our patients seemed less treatment resistant. Our previous studies have indicated that anemia is a promoter behind weight loss, and that systemic inﬂammation may be another factor behind elevated energy expenditure, although we have not succeeded to discriminate between the two as independent factors. An interesting question for the future would be whether a metabolically less efﬁcient and compromised organism compensates for energy inefﬁciencies by increased adrenergic activity and sensitivity as observed earlier in cancer patients (Drott et al. 1988, 1989) and directly in animals suffering from either cancer disease or pure malnutrition (Drott et al. 1987). If so, treatment of anemia should reduce endogenous stress and thereby simply attenuate progression of host wasting. In conclusion, anemia is effectively overcome by rhEPO in unselected cancer patients, even when patients are on anti-inﬂammatory treatment (Daneryd et al. 1998). Prevention of anemia preserves whole-body metabolic efﬁciency and physical function, deﬁned as exercise power (Wasserman et al. 1999). It is likely that these functional improvements are due to both maintained circulatory effectiveness and skeletal muscle

726

K. Lundholm and P. Daneryd

function, which to some extent contribute to improved over-all quality of life.

Acknowledgment Supported in parts by grants from Swedish Cancer Society (2014), Swedish Research Council (08712), Assar Gabrielsson Foundation (AB Volvo), Jubileumskliniken foundation, IngaBritt & Arne Lundberg Research Foundation, Swedish and Göteborg Medical Societies and the Medical Faculty, Göteborg University.

References 1. Barany P, Wibom R, Hultman E, Bergstrom J (1991) ATP production in isolated muscle mitochondria from haemodialysis patients: effects of correction of anaemia with erythropoietin. Clin Sci (Colch) 81: 645–653 2. Barber MD, Ross JA, Voss AC, Tisdale MJ, Fearon KC (1999) The effect of an oral nutritional supplement enriched with ﬁsh oil on weight-loss in patients with pancreatic cancer. Br J Cancer 81: 80–86 3. Bertero MT, Caligaris-Cappio F (1997) Anemia of chronic disorders in systemic autoimmune diseases. Haematologica 82: 375–381 4. Bosaeus I, Daneryd P, Svanberg E, Lundholm K (2001) Dietary intake and resting energy expenditure in relation to weight loss in unselected cancer patients. Int J Cancer 93: 380–383 5. Brevinge H, Berglund B, Bosaeus I, Tolli J, Nordgren S, Lundholm K (1995) Exercise capacity in patients undergoing proctocolectomy and small bowel resection for Crohn’s disease. Br J Surg 82: 1040–1045 6. Daneryd P, Svanberg E, Korner U, Lindholm E, Sandstrom R, Brevinge H, Pettersson C, Bosaeus I, Lundholm K (1998) Protection of metabolic and exercise capacity in unselected weight-losing cancer patients following treatment with recombinant erythropoietin: a randomized prospective study. Cancer Res 58: 5374–5379 7. Drott C, Persson H, Lundholm K (1989) Cardiovascular and metabolic response to adrenaline infusion in weight-losing patients with and without cancer. Clin Physiol 9: 427–439 8. Drott C, Svaninger G, Lundholm K (1988) Increased urinary excretion of cortisol and catecholamines in malnourished cancer patients. Ann Surg 208: 645–650 9. Drott C, Waldenstrom A, Lundholm K (1987) Cardiac sensitivity and responsiveness to beta-adrenergic stimulation in experimental cancer and undernutrition. J Mol Cell Cardiol 19: 675–683 10. Eguchi K (1995) Management of chemotherapy-induced anemia. Curr Opin Oncol 7: 316–319 11. Glaspy J (1997) The impact of epoetin alfa on quality of life during cancer chemotherapy: a fresh look at an old problem. Semin Hematol 34 [Suppl 2]: 20–26

Metabolic and physiologic effects of rhEPO in anemic cancer patients

727

12. Hyltander A, Drott C, Korner U, Sandstrom R, Lundholm K (1991) Elevated energy expenditure in cancer patients with solid tumours. Eur J Cancer 27: 9–15 13. Kasper C, Terhaar A, Fossa A, Welt A, Seeber S, Nowrousian MR (1997) Recombinant human erythropoietin in the treatment of cancer-related anaemia. Eur J Haematol 58: 251–256 14. Lindholm E, Daneryd P, Korner U, Hyltander A, Fouladiun M, Lundholm K (2004) Effects of recombinant erythropoietin in palliative treatment of unselected cancer patients. Clin Cancer Res 10: 6855–6864 15. Lundholm K, Daneryd P, Bosaeus I, Körner U, Lindholm E (2004a) Palliative nutritional intervention in addition to cyclooxygenase and erythropoietin treatment for patients with malignant disease: Effects on survival, metabolism and function. A randomized prospective study. Cancer 100: 1967–1977 16. Lundholm K, Daneryd P, Korner U, Hyltander A, Bosaeus I (2004b) Evidence that long-term COX-treatment improves energy homeostasis and body composition in cancer patients with progressive cachexia. Int J Oncol 24: 505–512 17. Lundholm K, Gelin J, Hyltander A, Lonnroth C, Sandstrom R, Svaninger G, Korner U, Gulich M, Karrefors I, Norli B, Hafström L, Kewenter J, Olbe L, Lundell L (1994) Anti-inﬂammatory treatment may prolong survival in undernourished patients with metastatic solid tumors. Cancer Res 54: 5602–5606 18. Marrades RM, Alonso J, Roca J, Gonzalez de Suso JM, Campistol JM, Barbera JA, Diaz O, Torregrosa JV, Masclans JR, Rodriguez-Roisin R, Wagner PD (1996a) Cellular bioenergetics after erythropoietin therapy in chronic renal failure. J Clin Invest 97: 2101–2110 19. Marrades RM, Roca J, Campistol JM, Diaz O, Barbera JA, Torregrosa JV, Masclans JR, Cobos A, Rodriguez-Roisin R, Wagner PD (1996b) Effects of erythropoietin on muscle O2 transport during exercise in patients with chronic renal failure. J Clin Invest 97: 2092–2100 20. Mittelman M (1996) Anemia of cancer: pathogenesis and treatment with recombinant erythropoietin. Isr J Med Sci 32: 1201–1206 21. Moliterno AR, Spivak JL (1996) Anemia of cancer. Hematol Oncol Clin North Am 10: 345–363 22. Moore GE, Bertocci LA, Painter PL (1993) 31P-magnetic resonance spectroscopy assessment of subnormal oxidative metabolism in skeletal muscle of renal failure patients. J Clin Invest 91: 420–424 23. Nieken J, Mulder NH, Buter J, Vellenga E, Limburg PC, Piers DA, de Vries EG (1995) Recombinant human interleukin-6 induces a rapid and reversible anemia in cancer patients. Blood 86: 900–905 24. Park JS, Kim SB, Park SK, Lim TH, Lee DK, Hong CD (1993) Effect of recombinant human erythropoietin on muscle energy metabolism in patients with endstage renal disease: a 31P-nuclear magnetic resonance spectroscopic study. Am J Kidney Dis 21: 612–618 25. Pocock SJ, Simon R (1975) Sequential treatment assignment with balancing for prognostic factors in the controlled clinical trial. Biometrics 31: 103–115 26. Thompson RT, Muirhead N, Marsh GD, Gravelle D, Potwarka JJ, Driedger AA (1996) Effect of anaemia correction on skeletal muscle metabolism in patients with end-stage renal disease: 31P magnetic resonance spectroscopy assessment. Nephron 73: 436–441

728

K. Lundholm and P. Daneryd: Metabolic and physiologic effects

27. Wasserman K, Hansen JE, Sue DY, Casaburi R, Whipp BJ (1999) Principles of exercise testing and interpretation. Lippincott Williams & Wilkins, Baltimore, Maryland Correspondence: Prof. Kent Lundholm, Department of Surgery, Sahlgrenska University Hospital, 413 45 Göteborg, Sweden, E-mail: [email protected]

Chapter 29

Effects of rhEPO on quality of life in anemic cancer patients S. Chowdhury, J. F. Spicer, and P. G. Harper Medical Oncology, Guy’s Hospital, London

Introduction Anemia is one of the commonest conditions associated with cancer, with 50 to 60% of cancer patients experiencing anemia at some stage during the course of their illness (Bokemeyer and Foubert 2004). The incidence and severity of anemia depends on the tumor type, stage of disease, patient age, and type and intensity of treatment. Cancer-related anemia may occur due to the malignancy itself (tumor bleeding, marrow inﬁltration) or due to myelosuppressive chemotherapy and/or radiotherapy. Other contributing factors include intercurrent illness, poor nutritional status, gastrointestinal blood loss and decreased red blood cell (RBC) survival. In addition, cancer patients have lower than expected serum erythropoietin concentrations for their corresponding levels of hemoglobin (Hb) compared with patients with other forms of anemia (Miller et al. 1990; Beguin et al. 1992). Increasing erythropoietin levels is therefore an attractive approach to the treatment of anemia in this patient group. In cancer-related anemia, symptom severity is determined by the rapidity of onset of anemia, the underlying malignancy, the intensity of cancer treatment, and the patient’s cardiovascular and respiratory reserve. The European Cancer Anemia Survey (ECAS) reported on the high prevalence of anemia among patients with cancer (see chapter 7) (Ludwig et al. 2004). This study investigated 15,367 cancer patients at 748 cancer centers throughout Europe, who were followed for up to 6 months to establish the incidence and prevalence of anemia. 72% of patients with hematological malignancies and 66% of patients with solid tumors were anemic (Hb <12 g/dL) at some point during the 6-month survey. Prospective clinical trials have shown that mild-to-moderate anemia (Hb 8–12 g/dL) occurs in up to 75% of cancer patients receiving chemotherapy and/or radiotherapy (Groopman and Itri 1999). Patients with lymphoma, lung, ovarian and genitourinary cancer have the highest incidence of severe anemia, with as many as 50–60% of these patients requiring blood transfusions. This risk is increased in those patients with a low Hb (10–12 g/dL) at the start of chemotherapy and those who are

730

S. Chowdhury et al.

scheduled to receive platinum-containing regimens (Groopman and Itri 1999).

Burden of disease Anemia is a complex condition that affects many body systems producing a wide range of symptoms. In cancer-related anemia the main symptoms include fatigue, dyspnoea, palpitations, dizziness, depression, impaired cognitive function, sleeping disturbance and loss of libido (Bokemeyer and Foubert 2004). Fatigue is the most common symptom reported by cancer patients with over 75% of patients experiencing typical symptoms (Vogelzang et al. 1997). Fatigue may present as various subjective symptoms including despondency, sluggishness and indifference. It may also result in a number of psychosocial symptoms including anxiety, depression and sleep disturbance (Vogelzang et al. 1997). Cognitive impairment, memory loss and inattention can be disturbing for the patient and increase anxiety. Fatigue has also been associated with an increased incidence of concomitant symptoms such as pain (Blesch et al. 1991). Fatigue is neither relieved by sleep nor exacerbated by exercise and may be extremely debilitating with a marked impact on the patient’s level of functioning. A study by the Fatigue Coalition, a multidisciplinary group of clinicians, researchers and patient advocates was aimed at understanding the relative implications and impact of cancer-related fatigue (Vogelzang et al. 1997). In this survey of perceptions of cancer-related fatigue 32% of patients reported experiencing daily fatigue. The areas of daily life most severely affected included the ability to work, physical and emotional well-being, ability to enjoy life in the moment and intimacy with their partner (Fig. 1) (Vogelzang et al. 1997). While fatigue can have a serious impact upon quality of life, cancer patients may experience difﬁculty in conveying this notion to clinicians and carers (Vogelzang et al. 1997). The study from the Fatigue Coalition showed that while oncologists perceived that 76% of their patients experienced fatigue, they believed that pain adversely affected their patients to a greater degree than fatigue (Fig. 2). In contrast, the patients themselves felt that fatigue affected their daily activities more than pain. Although the subject of fatigue was discussed between half of the patients in the study with their oncologist, only one quarter of oncologists believed there was effective treatment for fatigue. 80% of oncologists in this study stated that fatigue was often overlooked or under-treated (Vogelzang et al. 1997). This divergence in views of clinicians and patients was reﬂected in later studies. Two surveys report that fatigue is the most common and important problem experienced by cancer patients ahead of pain and nausea and vomiting (Curt et al. 1999;

Effects of rhEPO on quality of life in anemic cancer patients

731

61%

Ability to work

60%

Physical well-being Ability to enjoy life in the moment

57% 51%

Emotional well-being 44%

Intimacy with partner

42%

Ability to take care of family Relationships with family and friends

38%

Concerns about mortality and survival

33%

0

20

40

60

% of Patients Fig. 1. The negative impact of fatigue on daily routine of cancer patients. Percentage of cancer patients (n = 419) reporting aspects of daily routine that were affected ‘very much or ‘somewhat’ by fatigue. Adapted from Vogelzang (1997).

61%

Fatigueeg

37% 19%

Pain

61% 5% 2%%

Both

0

20

40

60

80

% Response

Patients (n = 419) Oncologists (n = 197)

Fig. 2. Percentage of patients and oncologists reporting that fatigue or pain affect patients’ daily life more. Adapted from Vogelzang (1997).

Stone et al. 2000). Stone and colleagues showed that one in two patients had never reported fatigue to the hospital clinician and of the 14% who did receive advice about management of fatigue; the most common advice was “to take a rest”. One third of patients in this study reported that their fatigue

732

S. Chowdhury et al.

was not well-managed, whereas management of pain and nausea and vomiting were only described as “not well managed” by 9% and 7% of patients respectively (Curt et al. 1999; Stone et al. 2000). A consequence of improved anti-emetics and analgesics has been the recognition of other symptoms, especially those related to daily function and well-being. Fatigue alone does not always mean that anemia is present but the level of Hb has been shown to have a strong correlation with the incidence of fatigue and quality of life (QOL). In a study of patients with solid or hematological malignancies (n = 50), patients with Hb <12 g/dL reported signiﬁcantly more fatigue, worse physical and functional well-being and generally reduced QOL than those who had Hb >12 g/dL (Cella 1998).

Assessment of cancer-related anemia and quality of life Accurate assessment of anemia and fatigue is important to ensure that patients are optimally managed. Assessment should be comprehensive incorporating laboratory, physical and quality of life indicators. Current assessment of anemia is often based on a single evaluation usually a qualitative measure such as Hb, with considerably less emphasis placed on patient symptoms, changes in function or QOL. As a result the symptoms that most affect a patient’s life, such as fatigue, are not fully evaluated and the full impact of anemia on the patient can often be underestimated. Evaluation of anemia should be ongoing, involving assessment of changes in patient well-being as well as long-term monitoring of anemia indicators (e.g. Hb concentration). While fatigue is a subjective sensation, it can be evaluated using various tests based on the patients’ perception of their quality of life. The measurement of quality of life (QOL) can be difﬁcult for clinicians to assess accurately. A well-established measure of patient QOL is the linear analogue self-assessment scale (LASA), considered to be a valid and reliable test for measuring subjective experience (Huskisson 1974; Huskisson 1982) often used in clinical trial settings (McCormack et al. 1988). It consists of a 100 mm horizontal bar on which the patient draws a line at a point which indicates his/her energy level, ranging from “as low as could be” at zero up to “as high as could be” at one hundred (Fig. 3). The same scale is used for “ability to undertake daily activities” and for “overall QOL”. The “numerical” effect “sizes” the magnitude of changes in perceived quality of life, it provides a standardized measure either of perceived health status change within a group, or a “difference” in changes between groups (Kazis et al. 1989). Some clinicians, however, feel that the results of these scales are difﬁcult to interpret and apply to the everyday clinical setting. There are other well-validated and accurate measurement systems that may be used in conjunction with the simpler LASA systems. Brieﬂy, these include the general Functional Assessment of Cancer Therapy (FACT-G) questionnaire consisting of twenty-seven questions (Cella 1998). The ques-

Effects of rhEPO on quality of life in anemic cancer patients

733

TO BE COMPLETED BY THE PATIENT

HOW WOULD YOU RATE YOUR ENERGY LEVEL DURING PAST WEEK? AS LOW AS COULD BE

10

20

30

40

50

60

70

80

90

100

AS HIGH AS COULD BE

SAME SCALE FOR: • ABILITY TO DO DAILY ACTIVITIES • OVERALL QUALITY OF LIFE

Fig. 3. Quality of Life (QOL) Linear Analogue Scale Assessment (LASA)

tions cover the four principle areas of QOL generally accepted by researchers in this area, that is, physical, functional, social and emotional ﬁelds. Subscales of this questionnaire have been developed addressing the issues of fatigue and anemia in cancer therapy: the FACT-F questionnaire (FACT-G plus thirteen questions relating to fatigue) and the FACT-An questionnaire (FACTF with an additional seven questions pertaining to anemia in cancer patients) (see chapter 14) (Yellen et al. 1997). These questionnaires are invaluable tools in assessing the quality of life of cancer patients with respect to the symptoms and distress resulting from anemia.

Treatment options Successful management of anemia in cancer patients requires accurate recognition of the condition as well as an understanding of the burden of the disease for the patient and an appreciation of the beneﬁt and limitations of the different forms of treatment. Treatment should be based on the severity of associated symptoms and the likelihood that anemia will worsen with continued treatment. If anemia is mild and asymptomatic the appropriate management may simply be observation and regular monitoring. However, anemia is commonly symptomatic and despite being commonly recognized is treated in less than one in 10 (9%) cancer patients (Vogelzang et al. 1997).

RBC transfusions RBC transfusions have historically been the main form of therapy when severe anemia develops in patients with cancer (Glaspy 2000). RBC transfusions rapidly raise the RBC count and Hb concentration and are effective

734

S. Chowdhury et al.

in the majority of patients. However, the beneﬁts are usually short-lived leading to a “roller-coaster” of unstable Hb levels that is reﬂected in the patient’s symptoms. In addition, treatment is not standardized with no trigger value (mean Hb concentration) established between clinical centers at which transfusion should be implemented, leading to inconsistencies in treatment. This was reﬂected in a large-scale UK audit of 2,719 cancer patients with solid tumours from 28 specialist centers which showed that even when Hb levels are <10 g/dL prior to chemotherapy, only a minority of patients receive a blood transfusion before the second cycle of chemotherapy (Barrett-Lee et al. 2000). This practice is largely because of a lack of understanding that fatigue is a treatable symptom but also takes into account the risks associated with RBC transfusion. These include potential transmission of infectious diseases, alloimmunisation, allergic reactions, hemolytic reactions, iron and circulatory overload and possible immunosuppression (Groopman and Itri 1999). Other disadvantages of RBC transfusions are that the treatment requires a hospital visit, cannulation of the patient and the resulting increase in Hb is shortlived (Itri 2000). Interestingly, the UK audit detected a trend for larger proportions of testicular cancer patients to complain of tiredness than those of other tumor types. The testicular cancer group, which generally has younger patients than those of other tumor types, had the lowest average Hb concentrations prior to RBC transfusion and perhaps reﬂected a reluctance by clinicians to transfuse patients who have a good prognosis unless their symptoms were severe (Barrett-Lee et al. 2000). Those critically ill patients who do receive RBC transfusions are generally transfused to maintain an Hb level of at least 8 g/dL (Glaspy 2000). However, several recent studies suggest that maintenance of higher Hb levels can improve the quality of life of patients (Abels 1993; Glaspy 1997; Demetri et al. 1998; Littlewood et al. 1999) and outcomes of cancer treatment (Littlewood et al. 2000).

Erythropoietic proteins With the introduction of recombinant human erythropoietin (rhEPO) in the late 1980’s, a valuable alternative treatment for anemia became available. Erythropoietic proteins such as epoetin alpha, epoetin beta and darbepoetin alpha alleviate anemia by mimicking the role of endogenous erythropoietin and stimulating the development of RBC precursor cells. The subcutaneous administration of epoetin alfa or beta mimics the physiological erythropoietin production (Halstenson et al. 1991; Dunn and Markham 1996; Cazzola et al. 1997; Storring et al. 1998). Erythropoietic proteins are well-tolerated and produce stable Hb levels thus avoiding the ﬂuctuations associated with RBC transfusions. Clinical trials have demonstrated that epoetin alfa or beta signiﬁcantly increases Hb levels in patients receiving a range of chemother-

Effects of rhEPO on quality of life in anemic cancer patients

735

apy regimens (both platinum and non-platinum-based) (Abels 1993; Cazzola et al. 1995; Glaspy et al. 1997; Demetri et al. 1998; Glimelius et al. 1998; Oberhoff et al. 1998; Gabrilove et al. 1999; Johansson et al. 2001; Littlewood et al. 2001; Osterborg et al. 2002). This effect is observed irrespective of tumor origin and is independent of tumor response (Demetri et al. 1998). Several trials have also been performed using darbepoetin alpha with similarly encouraging results (Vansteenkiste et al. 2002; Hedenus et al. 2003; Kotasek et al. 2003).

Effect of erythropoietic proteins on quality of life Several studies have shown that erythropoietic proteins increase Hb levels and reduce RBC transfusions in anemic cancer patients receiving chemotherapy and also in those not receiving chemotherapy (Abels 1993; Glaspy et al. 1997; Demetri et al. 1998; Gabrilove et al. 1999). In the study by Abels and colleagues (Abels 1993), the hematocrit and Hb of patients receiving platinum-based chemotherapy plus epoetin alfa (150 IU/kg three times weekly) increased signiﬁcantly more than in the placebo group receiving only chemotherapy. The increase in Hb was maintained for the duration of erythropoietin treatment (12 weeks) at levels of about 12 g/dL. Those patients responding to erythropoietin treatment also experienced a signiﬁcant increase in energy levels, activity levels and overall quality of life. The improvement experienced is notable as the patients studied were in advanced stages of cancer and two of the three subgroups were undergoing intensive courses of cyclic chemotherapy (platinum or non-platinum-based regimens). An additional patient group included those not receiving chemotherapy at all. The pooled data from all three patient groups showed that there was no difference in response rate to erythropoietin according to tumor type or tumor inﬁltration of the bone marrow (Abels 1993). Three large-scale, community-based studies on the treatment of cancer patients with epoetin alpha have also shown an association between increase in Hb levels and improved quality of life measurements (Glaspy et al. 1997; Demetri et al. 1998; Gabrilove et al. 1999). In all three studies, the baseline Hb was between 9.2 to 9.5 g/dL; the baseline transfusion requirements, activity, energy levels and overall quality of life were also similar. The patient groups in each of the three studies included patients with hematological and non-hematological malignancies with lung cancer being the most frequent tumor type (21.6–25.7%). Other tumor types included breast, gynecological, head and neck, gastrointestinal and prostate cancers. The erythropoietin dosing schedule differed in the studies. Demetri et al. and Glaspy et al., used a three times weekly dosing of epoetin alpha (Glaspy et al. 1997; Demetri et al. 1998) while Gabrilove et al., used a once weekly dosage (Gabrilove et al. 1999). In the community-based setting of these

736

S. Chowdhury et al.

studies, the majority of patients self-administered or had their erythropoietin administered to them by a carer. Not only did erythropoietin treatment signiﬁcantly increase Hb levels by 1.8 (Glaspy et al. 1997) or 2.0 g/dL (Demetri et al. 1998; Gabrilove et al. 1999) from baseline (p < 0.001), but its administration to patients reduced their overall requirement for RBC transfusion and the number of units transfused (Glaspy et al. 1997; Demetri et al. 1998). Quality of life as measured by the LASA scale showed a signiﬁcant improvement of 24% (Glaspy et al. 1997), 21% (Demetri et al. 1998) or 20% (Gabrilove et al. 1999) with erythropoietin treatment (p < 0.001). Energy and activity levels also improved with erythropoietin treatment (Glaspy et al. 1997; Demetri et al. 1998; Gabrilove et al. 1999). Furthermore, these studies showed a signiﬁcantly positive correlation (p ≤ 0.001) between Hb levels and QOL measurements, that is the greater the magnitude of change in Hb level from baseline, the better the improvement in QOL scores, using the LASA scales (Glaspy et al. 1997; Demetri et al. 1998; Gabrilove et al. 1999) and FACT-An scales (Demetri et al. 1998; Gabrilove et al. 1999). It is notable that Demetri and colleagues showed that although patients treated with erythropoietin perceived their quality of life to have improved, as indicated by higher FACT questionnaire and LASA scores compared to those of the placebo group, there was no positive correlation with improved Karnofsky scores given to them by the clinician (Demetri et al. 1998). These results highlight the difference in perception by patients of their quality of life and their performance status compared with observations by their clinician. It appears that anemic cancer patients could obtain a therapeutic beneﬁt from treatment of anemia with rhEPO and treatment of the anemia may signiﬁcantly improve patient quality of life. Several subsequent studies have shown that erythropoietic proteins increase hemoglobin levels and reduce RBC transfusions in anemic cancer patients receiving chemotherapy (Hedenus et al. 2001; Littlewood et al. 2001; Osterborg et al. 2002) and also in those not receiving chemotherapy (Smith 2003). These studies have also shown that treatment with erythropoietic proteins improves quality of life in anemic cancer patients. In a randomized, double-blind trial of 375 cancer patients receiving non-platinum-containing chemotherapy treatment with epoetin alfa (150 IU/kg three times per week) signiﬁcantly improved quality of life (Littlewood et al. 2001). In this study, erythropoietin treated patients scored better on all the measurements of quality of life, that is on the LASA scales (energy level, activity levels and overall quality of life) and also on the FACT-G, FACT-F (fatigue speciﬁc) and FACT-An (anemia speciﬁc) questionnaires (Littlewood et al. 2001). Erythropoietin was effective in raising Hb levels compared with placebo (p <0.01) with a mean increase of 2.2 g/dL Hb and reducing transfusion requirements of these cancer patients. A planned multiple linear regression

Effects of rhEPO on quality of life in anemic cancer patients

737

analysis, which accounted for the effects of disease progression and several other potential confounding variables on QOL, showed a signiﬁcant advantage for erythropoietin over placebo for all primary cancer and anemia speciﬁc QOL domains (all, p < 0.05) conﬁrming the results of the univariate analysis (Fallowﬁeld et al. 2002). In contrast in the placebo arm, changes in QOL measurements (for all categories of QOL measured) were universally negative. This group of patients received the current standard for management of anemia in cancer patients, that is, no management of anemia until rescue intervention with a RBC transfusion at an Hb level determined by the physician. This suggests that under the current standard of care for management of anemia the quality of life and functional capacity of cancer patients deteriorates over time (Littlewood et al. 2001). In contrast, QOL improved in anemic patients receiving erythropoietin treatment, due to increased and sustained Hb levels. Randomized, controlled studies using epoetin beta in the treatment of cancer-related anemia have also shown it to be effective in increasing Hb signiﬁcantly and reducing the requirement for blood transfusion (Cazzola et al. 1995; Glimelius et al. 1998; Oberhoff et al. 1998; Osterborg et al. 2001). In a randomized study in 349 anemic patients receiving chemotherapy for hematological malignancies, statistically signiﬁcant differences were observed in QOL (based on total FACT-An and FACT-G scores) after 12 and 16 weeks of treatment with epoetin beta compared with placebo (p < 0.05) (Osterborg et al. 2002). This improvement in QOL correlated with a signiﬁcantly higher Hb response rate in patients who received erythropoietin compared with placebo (67% vs. 27%, p < 0.0001) (Osterborg et al. 2002). Despite the clinical beneﬁts of erythropoietic proteins, response rates are variable, with approximately 50% of patients failing to respond to treatment (Fallowﬁeld et al. 2002). It is difﬁcult to predict which patients will respond to treatment and frequent dosing, once to three times a week, is required. Longer acting versions of rhEPO, such as darbepoetin alpha, would allow less frequent injections. Darbepoetin has been shown to improve QOL and increase Hb rates relative to placebo following once-weekly administration in two large placebocontrolled, phase III studies. The ﬁrst study in 320 patients with lung cancer showed an improvement in FACT-Fatigue scores in patients who received darbepoetin compared to patients receiving placebo (Vansteenkiste et al. 2002) (56% versus 44% overall improvement; 32% versus 19% with ≥25% improvement; mean difference = 13%; 95% CI = 2% to 23%, p = .019). The second study in 344 patients with lymphoproliferative malignancies experienced a statistically signiﬁcant increase in FACT-F score compared with placebo (p = 0.032) (Hedenus et al. 2003). Darbepoetin has also been found to be effective relative to placebo in a three-week schedule in anemic cancer patients receiving chemotherapy (Kotasek et al. 2003).

738

S. Chowdhury et al.

Timing of erythropoietin therapy A number of recent studies have shown that rhEPO can prevent the negative effects of anemia on quality of life in patients at risk of developing mildto-moderate anemia during chemotherapy (Beguin 2002; Littlewood et al. 2002; Straus 2002). Two studies compared the effect of early vs. delayed rhEPO therapy (Littlewood et al. 2002; Straus 2002). Littlewood and colleagues showed that patients with higher Hb levels (>10.5 g/dL) at initiation of rhEPO (150 IU/kg three times/week) had a lower transfusion rate (7.1% vs. 28.1%), reached a higher peak Hb level (13.8 vs. 12.7 g/dL) and reached a target Hb level of >12 g/dL more quickly (4 vs. 12 weeks) than patients who started treatment when their Hb fell to <10.5 g/dL (Littlewood et al. 2002). Similarly, Straus and colleagues reported that immediate rhEPO therapy (40,000 IU/kg weekly) resulted in signiﬁcant increases in Hb levels (mean change +0.8, p = 0.007) (Straus 2002). Patients with delayed rhEPO treatment (observation until Hb < 9.0 g/dL) had signiﬁcant decreases in Hb levels (mean change −0.8 g/dL, p = 0.007). In addition, patients receiving early rhEPO treatment reported a signiﬁcant (p < 0.05) improvement in quality of life that was signiﬁcantly correlated with change in Hb level (p < 0.05). Patients receiving early intervention had signiﬁcantly fewer subsequent clinic visits and days requiring general assistance with usual daily activities (p < 0.001) (Straus 2002). A recent randomized study supports early intervention with rhEPO to prevent anemia and maintain Hb levels (Chang et al. 2005). In this study patients receiving chemotherapy for breast cancer (n = 354, patients could be treated in the adjuvant or metastatic setting) were randomly assigned to epoetin alpha treatment (40,000 IU/kg weekly) or standard of care. At week 12 the change in FACT-An scores were signiﬁcantly higher in the rhEPO group (mean change +2.16 epoetin alfa vs. −4.43 for standard of care; p < 0.0001). Hemoglobin responses were signiﬁcantly higher in the rhEPO group resulting in a signiﬁcant reduction in transfusion use (Chang et al. 2005). These ﬁndings suggest that rhEPO can maintain or improve Hb levels and attenuate decreases in quality of life when administered early during the course of chemotherapy and/or to patients with mild-to-moderate anemia. In particular early intervention allows patients to maintain a relatively high QOL, including a positive sense of well-being and ability to function, to experience less fatigue during chemotherapy administration while decreasing health resource utilization.

Optimal hemoglobin for intervention The American Society of Clinical Oncology (ASCO) and American Society of Hematology (ASH) guidelines recommend the use of rhEPO for treat-

Effects of rhEPO on quality of life in anemic cancer patients

739

ment of anemia in patients with Hb < 10 g/dL and suggest that treatment should be determined by the clinical situation in patients with Hb levels between 10 and 12 g/dL. The EORTC has also recently published guidelines on the use of erythropoietic proteins in anemic patients with cancer (Bokemeyer et al. 2004). They recommend the initiation of erythropoietic proteins at Hb 9–11 g/dL in both patients receiving or not receiving chemotherapy based on anemia-related symptoms (Bokemeyer et al. 2004). They also suggest that erythropoietic proteins may be considered in asymptomatic patients with Hb 9–11 g/dL to prevent a further decline in hemoglobin according to individual factors. A target Hb of 12–13 g/dL is recommended and the major goals of treatment are to improve quality of life and reduce transfusions. In transfusion-dependent patients erythropoietic proteins should be initiated in addition to transfusions, although the prophylactic use of epoetins is not recommended (Bokemeyer et al. 2004). Both sets of guidelines, whilst establishing a minimum trigger Hb value for intervention, also make it clear that treatment with recombinant erythropoietins at higher Hb levels may be necessary or preferred. The optimal Hb level for quality of life as well as for treatment outcomes is still being actively investigated. The human homeostatic mechanism results in an increase in erythropoietin production when Hb levels fall below 12 g/dL, suggesting the importance of maintaining Hb above this level (Finch 1982). Retrospective analyses of study data have shown that the greatest improvement in QOL occurs when the hemoglobin level increases from 11 to 12 g/dL (range 11–13 g/dL) (Crawford et al. 2002). Results from the studies from Demetri and Glaspy were combined to examine the incremental improvements in QOL for each incremental increase in Hb. The aim was to identify the Hb level where optimal increases in QOL measurements occurred. The greatest incremental improvement in overall QOL occurred between Hb levels of 11 and 12 g/dL (Finch 1982), although some further beneﬁt was seen by increasing the Hb to “normal” values. These results suggest that the low patient Hb levels that are currently tolerated by oncologists before intervention occurs should no longer be accepted. Treatment with erythropoietic proteins to sustain increases in Hb to levels of 12 g/dL or above would be ideal for optimal improvement in patient QOL and to prevent deterioration in the physical and functional capacity of the patient, allowing them to remain well enough to receive appropriate cytotoxic treatment.

Summary Several randomized, placebo-controlled blinded studies and open-label community-based studies have shown that the development of anemia in cancer patients undergoing chemotherapy has a major impact on quality of life and functional capacity of patients. Studies have shown repeatedly that fatigue is

740

S. Chowdhury et al.

the symptom that patients rank most important and which most adversely affects their lives. Despite this, patient needs remain underappreciated, underrecognized and undertreated. Erythropoietic proteins have a welldocumented positive impact on quality of life and patient-reported outcomes when used appropriately and should be considered in every cancer anemia management plan. Their role in the treatment of anemia in cancer patients is likely to gain prominence as greater emphasis is placed on quality of life in the management of cancer patients.

References 1. Abels R (1993) Erythropoietin for Anemia in Cancer Patients. Eur J Cancer 29A [Suppl 2]: S2–S8 2. Barrett-Lee PJ, Bailey NP, O’Brien ME, Wager E (2000) Large-scale UK audit of blood transfusion requirements and anemia in patients receiving cytotoxic chemotherapy. Br J Cancer 82: 93–97 3. Beguin Y (2002) Prediction of response and other improvements on the limitations of recombinant human erythropoietin therapy in anemic cancer patients. Haematologica 87: 1209–1221 4. Beguin Y, Yerna M, Loo M, Weber M, Fillet G (1992) Erythropoiesis in multiple myeloma: defective red cell production due to inappropriate erythropoietin production. Br J Haematol 82: 648–653 5. Blesch KS, Paice JA, Wickham R, et al (1991) Correlates of fatigue in people with breast or lung cancer. Oncol Nurs Forum 18: 81–87 6. Bokemeyer C, Aapro MS, Courdi A, et al (2004) EORTC guidelines for the use of erythropoietic proteins in anemic patients with cancer. Eur J Cancer 40: 2201–2216 7. Bokemeyer C, Foubert J (2004) Anemia impact and management: focus on patient needs and the use of erythropoietic agents. Semin Oncol [Suppl 8] 31: 4–11 8. Cazzola M, Mercuriali F, Brugnara C (1997) Use of recombinant human erythropoietin outside the setting of uremia. Blood 89: 4248–4267 9. Cazzola M, Messinger D, Battistel V, et al (1995) Recombinant human erythropoietin in the anemia associated with multiple myeloma or non-Hodgkin’s lymphoma: dose ﬁnding and identiﬁcation of predictors of response. Blood 86: 4446–4453 10. Cella D (1998) Factors inﬂuencing quality of life in cancer patients: anemia and fatigue. Semin Oncol [Suppl 7] 25: 43–46 11. Chang J, Couture F, Young S, McWatters KL, Lau CY (2005) Weekly epoetin alfa maintains hemoglobin, improves quality of life, and reduces transfusion in breast cancer patients receiving chemotherapy. J Clin Oncol 23: 2597–2605 12. Crawford J, Cella D, Cleeland CS, et al (2002) Relationship between changes in hemoglobin level and quality of life during chemotherapy in anemic cancer patients receiving epoetin alfa therapy. Cancer 95: 888–895 13. Curt GA, Breitbart W, Cella DF, et al (1999) Impact of cancer-related fatigue on the lives of patients. Proc Am Soc Clin Oncol 18: 573a (Abstr)

Effects of rhEPO on quality of life in anemic cancer patients

741

14. Demetri GD, Kris M, Wade J, Degos L, Cella D (1998) Quality-of-life beneﬁt in chemotherapy patients treated with epoetin alfa is independent of disease response or tumor type: results from a prospective community oncology study. Procrit Study Group. J Clin Oncol 16: 3412–3425 15. Dunn CJ, Markham A (1996) Epoetin beta. A review of its pharmacological properties and clinical use in the management of anemia associated with chronic renal failure. Drugs 51: 299–318 16. Fallowﬁeld L, Gagnon D, Zagari M, et al (2002) Multivariate regression analyses of data from a randomised, double-blind, placebo-controlled study conﬁrm quality of life beneﬁt of epoetin alfa in patients receiving non-platinum chemotherapy. Br J Cancer 87: 1341–1353 17. Finch CA (1982) Erythropoiesis, erythropoietin, and iron. Blood 60: 1241–1246 18. Gabrilove JL, Einhorn LH, Livingston RB, Winder E, Cleeland CS (1999) Onceweekly dosing of epoietin alfa is similar to three-times-weekly dosing in increasing hemoglobin and quality of life. Proc Am Soc Clin Oncol 18: 574a (Abstr) 19. Glaspy J, Bukowski R, Steinberg D, Taylor C, Tchekmedyian S, Vadhan-Raj S (1997) Impact of therapy with epoetin alfa on clinical outcomes in patients with nonmyeloid malignancies during cancer chemotherapy in community oncology practice. Procrit Study Group. J Clin Oncol 15: 1218–1234 20. Glaspy JA (1997) Epoietin alfa improves hematological parameters and quality of life (QOL) in breast cancer patients. Proceedings of the European Society of Oncology, October 31st 2000 21. Glaspy JA (2000) Hematologic supportive care of the critically ill cancer patient. Semin Oncol 27: 375–383 22. Glimelius B, Linne T, Hoffman K, et al (1998) Epoetin beta in the treatment of anemia in patients with advanced gastrointestinal cancer. J Clin Oncol 16: 434–440 23. Groopman JE, Itri LM (1999) Chemotherapy-induced anemia in adults: incidence and treatment. J National Cancer Institute 91: 1616–1634 24. Halstenson CE, Macres M, Katz SA, et al (1991) Comparative pharmacokinetics and pharmacodynamics of epoetin alfa and epoetin beta. Clin Pharmacol Ther 50: 702–712 25. Hedenus M, Adriansson M, San Miguel J, et al (2003) Efﬁcacy and safety of darbepoetin alfa in anemic patients with lymphoproliferative malignancies: a randomized, double-blind, placebo-controlled study. Br J Haematol 122: 394–403 26. Hedenus M, Brandberg Y, Molostova V, et al (2001) Efﬁcacy of epoetin beta in treating the anemia of cancer in patients with haematological malignancies. Proceedings of the European Haematology Association, Frankfurt, Germany 27. Huskisson EC (1974) Measurement of pain. Lancet 2: 1127–1131 28. Huskisson EC (1982) Measurement of pain. J Rheumatol 9: 768–769 29. Itri LM (2000) Optimal hemoglobin levels for cancer patients. Semin Oncol [Suppl 4] 27: 12–15 30. Johansson JE, Wersall P, Brandberg Y, Andersson SO, Nordstrom L (2001) Efﬁcacy of epoetin beta on hemoglobin, quality of life, and transfusion needs in patients with anemia due to hormone-refractory prostate cancer – a randomized study. Scand J Urol Nephrol 35: 288–294 31. Kazis LE, Anderson JJ, Meenan RF (1989) Effect sizes for interpreting changes in health status. Medical Care 27: S178–S189

742

S. Chowdhury et al.

32. Kotasek D, Steger G, Faught W, et al (2003) Darbepoetin alfa administered every 3 wekks alleviates anaemia in patients with solid tumours receiving chemotherapy; results of a double-blind, placebo-controlled, randomised study. Eur J Cancer 39: 2026–2034 33. Littlewood TJ, Bajetta E, Cella D (1999) Efﬁcacy and quality of life outcomes of epoetin alfa in a double-blind, placebo-controlled multicenter study of cancer patients receiving non-platinum containing chemotherapy. Proc Am Soc Clin Oncol 18: 574a (Abstr) 34. Littlewood TJ, Bajetta E, Nortier JW, Vercammen E, Rapoport B (2001) Effects of epoetin alfa on hematologic parameters and quality of life in cancer patients receiving nonplatinum chemotherapy: results of a randomized, double-blind, placebo-controlled trial. J Clin Oncol 19: 2865–2874 35. Littlewood TJ, Bajetta E, Rapoport V, Norties J, et al (2002) Early administration of epoetin alfa optimizes anemia management with respect to hematologic and quality of life (QOL) outcomes in anemic cancer patients (pts) undergoing chemotherapy. Blood 100: 18b–19b (Abstr) 36. Littlewood TJ, Rapoport B, Bajetta J, Nortier J (2000) Possible relationship of hemoglobin levels with survival in anemic cancer patients receiving chemotherapy. Proc Am Soc Clin Oncol 19: 605a (Abstr) 37. Ludwig H, Van Belle S, Barrett-Lee P, et al (2004) The European Cancer Anemia Survey (ECAS): a large, multinational, prospective survey deﬁning the prevalence, incidence, and treatment of anemia in cancer patients. Eur J Cancer 40: 2293–2306 38. McCormack HM, Horne DJ, Sheather S (1988) Clinical applications of visual analogue scales: a critical review. Psychological Medicine 18: 1007–1019 39. Miller CB, Jones RJ, Piantadosi S, Abeloff MD, Spivak JL (1990) Decreased erythropoietin response in patients with the anemia of cancer. N Engl J Med 322: 1689–1692 40. Oberhoff C, Neri B, Amadori D, et al (1998) Recombinant human erythropoietin in the treatment of chemotherapy-induced anemia and prevention of transfusion requirement associated with solid tumors: a randomized, controlled study. Ann Oncol 9: 255–260 41. Osterborg A, Brandberg Y, Molostova V, et al (2001) Efﬁcacy of epoetin beta in treating the anemia of cancer in patients with haematological malignancies. Proceedings of the European Haematology Association (submitted) 42. Osterborg A, Brandberg Y, Molostova V, et al (2002) Randomized, double-blind, placebo-controlled trial of recombinant human erythropoietin, epoetin Beta, in hematologic malignancies. J Clin Oncol 20: 2486–2494 43. Smith RE, Jr (2003) Erythropoietic agents in the management of cancer patients. Part 1: Anemia, quality of life, and possible effects on survival. J Support Oncol 1: 249–256, 258–259 44. Stone P, Richardson A, Ream E, Smith AG, Kerr DJ, Kearney N (2000) Cancerrelated fatigue: Inevitable, unimportant and untreatable? Results of a multicentre patient survey. Ann Oncol 11: 971–975 45. Storring PL, Tiplady RJ, Gaines Das RE, et al (1998) Epoetin alfa and beta differ in their erythropoietin isoform compositions and biological properties. Br J Haematol 100: 79–89

Effects of rhEPO on quality of life in anemic cancer patients

743

46. Straus DJ (2002) Epoetin alfa as a supportive measure in hematologic malignancies. Semin Hematol 39 [Suppl 3]: 25–31 47. Vansteenkiste J, Pirker R, Massuti B, et al (2002) Double-blind, placebocontrolled, randomized phase III trial of darbepoetin alfa in lung cancer patients receiving chemotherapy. J Natl Cancer Inst 94: 1211–1220 48. Vogelzang NJ, Breitbart W, Cella D, et al (1997) Patient, caregiver, and oncologist perceptions of cancer-related fatigue: results of a tripart assessment survey. The Fatigue Coalition. Semin Hematol [Suppl 2] 34: 4–12 49. Yellen SB, Cella DF, Webster K, Blendowski C, Kaplan E (1997) Measuring fatigue and other anemia-related symptoms with the Functional Assessment of Cancer Therapy (FACT) measurement system. J Pain Symptom Manage 13: 63–74 Correspondence: Dr. Peter G. Harper, Medical Oncology, 3rd Floor Thomas Guy House, Guy’s Hospital, St Thomas Street, London SE1 9RT, UK, E-mail: [email protected]

Chapter 30

Thrombosis during therapy with erythropoiesis stimulating agents in cancer J. Glaspy Division of Hematology-Oncology, UCLA School of Medicine, Los Angeles, California, USA

The beneﬁts of therapy with erythropoiesis stimulating agents (ESAs) for the treatment of anaemia during cancer chemotherapy, in terms of reduction in the symptoms of anemia and in transfusions, are well documented. Recently, there has been an increased focus on the safety of ESAs in this setting, with one of the issues being an apparent increase in the incidence of thrombosis. When ESAs were developed for the treatment of the anaemia occurring in patients with chronic renal failure, an increase in the incidence of hypertension and thromboses, particularly involving vascular access devices, was observed. There was some evidence that these complications were more likely to occur when haemoglobin levels increased rapidly and it was assumed that the physiology involved a rapid expansion in red cell mass in patients lacking the ability to respond with a compensatory reduction in plasma volume. If this were the only mechanism of thrombosis associated with ESA therapy, it would not be expected to occur in patients, such as the typical patient with anaemia due to cancer chemotherapy, with reasonably intact renal function. The biology involved is clearly more complicated than initially expected and remains incompletely understood. This chapter will review the relevant aspects.

The hypercoaguable state in cancer Since Trousseau’s classic report, it has been clear that there is an association between cancer and activation of the coagulation system and well recognized that cancer patients have an increased risk of thromboembolism. There are several mechanisms involved in this increase in observed thrombosis rates in cancer patients (Adess et al. 2006; Zwicker et al. 2007). Increased production of tissue factor (tissue thromboplastin, which among other things binds to factor VII to generate thrombin) (Rak et al. 2006; Tesselaar et al. 2007)

746

J. Glaspy

and cancer procoagulant (a vitamin K dependent protease that activates factor X) (Gale and Gordon 2001) by tumor cells is common, may be important in cancer progression and metastasis (Franchini et al. 2007) and activates coagulation. Direct compression of major veins by tumor or ﬂuid collections and the implantation of indwelling venous catheters represent physical factors increasing thrombotic potential. Therapy with hormonal agents such as tamoxifen, anti-angiogenic agents such as bevacizumab and several chemotherapeutic agents (Lee 2003) can further increase thrombosis risk. Finally, cancer patients frequently have co-morbidities that increase clotting risk and are frequently hospitalized or otherwise immobile (Khorana et al. 2006; Stein et al. 2006). One of the signiﬁcant challenges facing accurate estimation of the incremental thrombotic risk posed by ESA therapy for a speciﬁc cancer patient is the relative lack of published data regarding baseline risk in speciﬁc clinical settings (Lee 2003). In general, thrombosis is more common in patients with primary CNS neoplasms, lung cancer (Tagalakis et al. 2007) and gynaecological malignancies, less common in patients with breast, prostate or head and neck cancers, and more common in patients with metastatic disease, regardless of primary site (Fig. 1). Chemotherapy is associated with a ﬁfty

0

5

10

15

20

25

30

Central nervous system Gastrointestinal Lung Low-grade lymphoma Ovary Sarcoma Colon and rectum Leukemia Other High-grade lymphoma Urinary tract Prostate Head and neck Breast Metastatic tumors Two or more cancers

Fig. 1. Estimated rates (in percents) of thrombosis in patients with cancers of various primary sites based upon a medical record database. (Source: Medstat and General Practice Research Database, Amgen Brieﬁng Document to the U.S. Oncology Drug Advisory Committee, May 4, 2004.)

Thrombosis during therapy with erythropoiesis stimulating agents

747

percent increase in the relative risk of thrombosis compared to cancer patients not receiving chemotherapy (Lee 2003). There is also some evidence that anaemia is a risk factor for thrombosis in patients with cancer receiving chemotherapy. Clearly, patients with cancer who are anaemic and receiving chemotherapy and who are therefore candidates to receive ESAs are at substantial baseline risk for thrombotic events, with rates at least 6 fold higher than the general population (Lee 2003).

The effects of ESAs on thrombosis risk in cancer In the initial randomized, placebo-controlled trials of ESAs for the treatment of chemotherapy-induced anaemia, no statistically signiﬁcant increase in the rate of thrombosis was observed in treated patients. This failure to document an effect on thrombosis rates was probably due to the multiplicity of covariates determining risk among cancer patients, coupled with sample sizes that did not confer sufﬁcient power to detect a modest effect in the presence of this tremendous variability in baseline risk. Subsequently, an observational study in patients with gynaecological malignancies undergoing chemo-radiotherapy suggested thrombosis risk were higher in patients receiving ESAs compared to historical controls at the same institution (Wun et al. 2003). Similar observations were made in comparing thrombosis rates in clinical trials of chemo-radiotherapy for cervical cancer before and after the integration of ESAs into patient management (Lavey et al. 2004), but not in studies of patients with ovarian cancer (Westin et al. 2007). Several meta-analyses have now been carried out on randomized trials of ESAs in patients with cancer (Bohlius et al. 2006; Ross et al. 2006). Overall, in these pooled analyses including patients with a wide variety of cancers and baseline thrombosis risk, there is an observed increase in the incidence of thrombosis associated with ESA therapy, with an estimated relative risk of approximately 1.6 (Fig. 2). This reﬂects an absolute risk increase of two to three percent, from between three and four percent to between ﬁve and six percent. This appears to be a class effect, with similar risks observed with recombinant erythropoietins and with darbepoetin alfa (Ross et al. 2006). The relative increase in thrombosis risk associated with ESA therapy appears to be similar in patients with cancer, whether or not they are receiving chemotherapy (Ross et al. 2006). The majority of these thrombotic events are venous thromboses, catheter-associated clots and pulmonary emboli; ESA therapy has not been linked to an increase in arterial events, such as cerebrovascular accidents and myocardial infarctions, in cancer patients.

748

J. Glaspy

Study

Treatment n/N

Thatcher 1999a (37) P-174 J&J (16) Cascinu 1994 (12) Bamias 2003 (47) Smith 2003 (45) EPO-CAN-20 J&J (62) Case J&J (13) Henry J&J (14) N93 004 FDA (66) Chang 2005 (48) Littlewood J&J (6) Thompson 2000 (27) Vansteenkiste 2002 (53) INT-1 J&J (64) Witzig J&J (46) Leyland-jones J&J (8) Osterborg 1996b (24) Henke 2003 Roche (7) Ten Bokkel 1998a (31) Rose J&J (17) Thatcher 1999b (37) Osterborg 1996a (24) Adess J&J (15) Throuvalas 2000 (32) GOG-0191 FDA (63) Italian 1998 (21) Razzouk 2004 (44) Welch 1995 (38) Osterborg 2002 (25) Ten Bokkel 1998b (31) Vadhan-Raj FDA (56) INT-3 J&J (65) Savonje 2004 (51) Machtay 2004 (54) EPO-GBR-7 FDA (57) Dammacco J&J (20) EPO-CAN-15 FDA (61) Rosenzweig 2004 (60)

0/42 0/33 0/50 0/72 1/64 1/31 2/81 6/67 24/109 19/175 14/251 1/45 7/155 3/164 9/168 36/448 1/48 10/180 2/45 9/142 2/44 2/47 1/65 1/28 9/58 1/44 6/112 1/15 1/170 4/42 7/29 8/135 9/211 2/71 5/151 5/69 16/53 4/14

3728 Total (95% Cl) Total events: 229 (Treatment), 118 (Control)

Control n/N

RR (fixed) 95% Cl

Weight %

0/22 0/12 0/50 1/72 1/22 2/31 3/76 8/65 26/115 14/175 5/124 0/21 5/159 1/80 6/165 25/456 0/24 6/171 0/17 2/79 0/22 0/25 0/59 0/26 3/55 0/43 2/110 0/15 0/173 0/16 2/31 1/65 1/104 0/70 1/149 1/76 2/53 0/13

1.16 1.15 1.55 2.39 6.28 19.56 10.82 5.17 0.52 3.82 1.04 4.68 19.15 0.51 4.76 0.56 1.99 0.51 0.50 0.40 0.40 2.38 0.39 1.56 0.39 0.38 0.55 1.49 1.04 1.04 0.39 0.78 0.74 1.55 0.40

3041

100.00

0.01

0.1 Control

1

10

RR (fixed) 95% Cl Not estimable Not estimable Not estimable 0.33 [0.01, 8.05] 0.34 [0.02, 5.27] 0.50 [0.05, 5.23] 0.63 [0.11, 3.64] 0.73 [0.27, 1.98] 0.97 [0.60, 1.59] 1.36 [0.70, 2.62] 1.38 [0.51, 3.75] 1.43 [0.06, 33.82] 1.44 [0.47, 4.43] 1.46 [0.15, 13.85] 1.47 [0.54, 4.05] 1.47 [0.89, 2.40] 1.53 [0.06, 36.23] 1.58 [0.59, 4.26] 1.96 [0.10, 38.79] 2.50 [0.55, 11.30] 2.56 [0.13, 51.05] 2.71 [0.14, 54.32] 2.73 [0.11, 65.68] 2.79 [0.12, 65.66] 2.84 [0.81, 9.96] 2.93 [0.12, 70.08] 2.95 [0.61, 14.28] 3.00 [0.13, 68.26] 3.05 [0.13, 74.41] 3.56 [0.20, 62.58] 3.74 [0.85, 16.56] 3.85 [0.49, 30.15] 4.44 [0.57, 34.55] 4.93 [0.24, 100.89] 4.93 [0.58, 41.73] 5.51 [0.66, 45.98] 8.00 [1.93, 33.09] 8.40 [0.50, 142.27] 1.67 [1.35, 2.06]

100

Treatment

Fig. 2. A forest plot for the meta-analysis of rates of thromboembolism in randomized studies of ESAs in cancer patients. [by permission of Oxford University Press from Bohlius et al. (2006) Recombinant Human Erythropoietins and Cancer Patients; Updated meta-Analysis of 57 Studies Including 9353. Journal of National Cancer Institute 98 (10): 708–714]

Mechanisms of ESA-induced thrombosis in cancer patients Understanding the precise mechanisms through which ESAs promote venous thrombosis in cancer patients would aid us greatly in predicting these events and developing rationale approaches to their prevention. Several possibilities have been proposed, with varying degrees of supportive evidence, but the deﬁnitive answer has remained unclear (Table 1). It is often assumed that thromboses during ESA therapy in cancer patients must be due simply to alterations in blood rheology induced by the increasing red cell mass. While there was some early evidence that this might explain some of the hypertension and thromboses occurring in patients with chronic renal failure who are unable to appropriately adjust their total blood

Thrombosis during therapy with erythropoiesis stimulating agents

749

Table 1. Proposed mechanisms of ESA-induced thrombosis and the resulting rational prediction and prevention strategies. (CRP = c-reactive protein) Proposed Mechanism of ESA-Induced Thrombosis

Prediction Strategy Suggested by that Mechanism

Prevention Strategy Suggested by that Mechanism

Alterations in blood rheology

Expect in patients with a rapid rise in haemoglobin or“overshoot” polycythemia None

Develop ESA treatment algorithms that minimize a given rise and/or overshoot Possibly anti-platelet agents

Expect in patients with a rapid rise in haemoglobin or“overshoot” polycythemia None

Anti-platelet agents

Expect in patients with elevated CRP levels

Possibly anti-inﬂammatory therapy

Activation of vascular endothelium Red cell – platelet interactions

Thrombopoietinerythropoietin interactions Increases in inﬂammation

Anti-platelet agents

volume as red cell mass rises, there is little evidence that this is an important factor in thromboses observed during ESA therapy in the oncology setting. Obviously, if rapid expansion of red cell mass in cancer patients were the major factor driving thrombotic risk in this setting, a signiﬁcant incidence of thrombosis would be observed immediately follow red blood cell transfusion. Although the risks of red cell transfusion have been extensively studied and well documented, a signiﬁcant increase in venous thrombosis has not been demonstrated. Moreover, there is not a clear relationship observed between haemoglobin response following initiation of ESA therapy and thrombosis; thrombosis occurs in non-responders as well as responders. In one detailed analysis of data from more than 2,000 patients, no signiﬁcant relationship between initial ESA dose or rate of hemoglobin rise and observed thrombosis could be identiﬁed (Hedenus et al. 2005). These authors concluded that, if a predictor of thrombosis based upon haemoglobin rise must be chosen, the best predictor would be a rise of 2 g/dL in 28 days; a rise of 1 g/dL in 14 days had no predictive value. Because very few patients treated with ESAs in the oncology setting develop “overshoot” polycythemia, avoiding supranormal haemoglobin levels, while a reasonable goal, would not be expected to substantially reduce the observed thrombosis risk. There is clear evidence that vascular endothelial cells express functional erythropoietin receptors. Clinically, an increase in blood pressure is

750

J. Glaspy

120

Δ% cE-selectin

100 80 60 40 20 0 –20 0

5

10

15

20

Study days

Fig. 3. E-selectin levels in human volunteers treated with recombinant erythropoietin, three times weekly (arrows indicate erythropoietin treatment). [this research was originally published in Blood. Stohlawetz et al., Effects of erythropoietin on platelet reactivity and thrombopoiesis in humans. Blood. 2000; 95 (9): 2983–2989. © the American Society of Hematology]

sometimes noted when subjects are treated with ESAs; this is frequently observed before any increase in haemoglobin level has occurred, strongly suggesting a direct effect on vasculature. Human endothelial cells express erythropoietin receptor messenger RNA (Anagnostou et al. 1994), and activation of these receptors by erythropoietin results in a unique molecular response in these cells (Fodinger et al. 2000). In vitro, erythropoietin induces nitric oxide production by endothelial cells (Banerjee et al. 2000; BeleslinCokic et al. 2004); in vivo, erythropoietin promotes vasculogenesis (Ashley et al. 2002) and is involved in endothelial cell recruitment and mobilization (Satoh et al. 2006; Urao et al. 2006). In one study in volunteers treated with recombinant erythropoietin three times weekly, increases in serum e-selectin levels (Fig. 3) was consistent with endothelial activation and with evidence of increased platelet reactivity (Stohlawetz et al. 2000). A direct effect of ESAs in vascular endothelial cells, which could increase the likelihood of interactions with platelets initiating primary hemostasis, is one of the best demonstrated potential mechanisms for ESA-induced thrombosis. Unfortunately, this is not helpful in identifying patients at risk, although it suggests that anti-platelet agents may be effective in reducing this effect of ESAs. There is evidence that red blood cells may play a role in enhancing the recruitment of platelets by activated platelets in thrombus formation (Valles et al. 1991). In one study, the effects of releasates from collagen-stimulated platelets in platelet rich plasma on the activation of unstimulated platelets, as measured by CD-62 expression, was studied (Valles et al. 2002). When the stimulated platelet preparation included red blood cells, the activating effects of the releasates were enhanced. The enhancing effects of red cells on platelet

Thrombosis during therapy with erythropoiesis stimulating agents

751

% CD62-Positive Platelets

A. 45 * * 30

15

0

PRP

10%

20%

30%

40%

Hematocrit

% CD62-Positive Platelets

B.

45 Without ASA

With ASA

30

15 * 0

PRP

PRP+ RBC

PRP

* PRP+ RBC

Fig. 4. A The effects of varying concentrations of red blood cells on the activating effects of releasates from collagen stimulated platelets. B Aspirin blocks this effect. [This research was originally published in Blood. Vallés et al., Platelet-erythrocyte interactions enhance α1Ibβ3 integrin receptor activation and P-selectin expression during platelet recruitment: down-regulation by aspirin ex vivo. Blood. 2002; 99(11): 3978–3984. © the American Society of Hematology]

recruitment increased as the haematocrit of the platelet rich plasma increased and was blocked by aspirin (Fig. 4). Whether this mechanism is operative at all in the increase in thrombotic events associated with ESA therapy in cancer patients is not known. If it is involved, the effect would be limited to that subset of patients in whom ESA therapy is associated with a signiﬁcant increase in haemoglobin levels. This would not explain the failure to date to demonstrate a signiﬁcant correlation between haemoglobin rise and thrombotic events. Due to the structural similarities of erythropoietin and thrombopoietin, there has been speculation that erythropoietin may interact with the thrombopoietin receptor and induce platelet production and/or activation. Initial reports of increases in platelet counts in ESA treated patients actually reﬂected developing functional iron deﬁciency, and there has not been a convincing demonstration of human erythropoietin alone inducing a cellular

752

J. Glaspy

Actual Expected

(T PO

10

0

TP O

10 0 EP ng ng O /m /m 2 L L U + (T /m EP PO L O 10 2 0 U /m ng SC L) /m F L 10 + 0 SC ng (T /m F PO 10 L 0 10 ng 0 /m IL ng -3 L) /m 10 L + 0 n IL g/ -3 m (T L 10 PO 0 10 ng 0 / m G ng L) -C /m SF L + 10 G 0 -C ng SF /m L 10 0 ng /m L)

% CD62+ platelets

100 90 80 70 60 50 40 30 20 10 0

Fig. 5. The effects of human thrombopoietin (TPO) alone, or in combination with human erythropoietin (EPO) or other cytokines on platelet activation as measured by CD62 expression. (SCF = stem cell factor, IL-3 = interleukin-3, G-CSF = granulocyte colony-stimulating factor) [From Wun et al. (1997) Thrombopoietin is synergistic with other hematopoietic growth factors and physiologic platelet agonists for platelet activation in vitro. Am J Hematol 54: 225–232. © 1997 Wiley-Liss, Inc. Reprinted with permission of Wiley-Liss, Inc., a subsidiary of John Wiley & Sons, Inc.]

response acting through the human thrombopoietin receptor. However, there is some evidence that erythropoietin may synergize with thrombopoietin in the activating effects of that ligand on mature platelets (Wun et al. 1997). In vitro, exposure of human platelets to thrombopoietin results in an expected activation as reﬂected by increased CD62 expression. When platelets are incubated with both thrombopoietin and erythropoietin, the magnitude of increased CD62 expression is greater than expected; similar effects are observed with stem cell factor (c-kit ligand) but not with interleukin 3 or granulocyte colony-stimulating factor (Fig. 5). It is not clear whether this effect of erythropoietin is mediated through the thrombopoietin receptor or erythropoietin receptors on platelets. It is important to point out that there is no clinical evidence that platelet activation by thrombopoietin is associated with an increased incidence of thrombosis in any clinical setting. Treatment of cancer patients with recombinant thrombopoietin has not been reported to increase thrombosis rates. Although this is an interesting observation deserving of further study, there is no supporting evidence that this mechanism is a signiﬁcant contributor to thromboses observed in ESA treated cancer patients.

Thrombosis during therapy with erythropoiesis stimulating agents

753

The biologies of cancer and of coagulation have interesting overlaps, which have led some authors to speculate that coagulation is meaningfully involved in cancer progression (Rak et al. 2006; Franchini et al. 2007; Rak et al. 2006). One area of overlap is the central role of inﬂammation in both processes. As an example, tissue factor, which as noted earlier is clearly involved in the hyper-coaguable state in cancer, is a transmembrane receptor that activates pro-inﬂammatory intracellular pathways and malignancyassociated thrombosis is associated with increases in markers of inﬂammation (Fareed et al. 2004). In dialysis patients, levels of c-reactive protein (CRP), a biomarker for active inﬂammation, have been shown to correlate with increased levels of thrombin-activatable ﬁbrinolytic inhibitor, resulting in a relative deﬁciency in ﬁbrinolysis with ongoing inﬂammation (Tobu et al. 2004). These investigators have hypothesized that chronic therapy with ESAs results in increased inﬂammation and therefore thrombosis risk, although the evidence for a causal link between ESA therapy and inﬂammation is lacking. Nevertheless, given the interrelationships between coagulation, malignancy and inﬂammation, if ESAs were found to be inﬂammatory, they would be expected to be particularly thrombogenic in patients with cancer.

Thrombosis prophylaxis The issue of thrombosis prophylaxis in cancer patients is a difﬁcult one, even when ESA therapy is not contemplated. Many cancer patients have an increased risk of bleeding, due either to their disease or its treatment and the risks posed by bleeding and thrombosis must be assessed and balanced on an individualized basis in making these difﬁcult decisions (Leonardi et al. 2007). As noted previously, recent data have suggested that coagulation may play a role in cancer progression, and there is some evidence that therapeutic anticoagulation may improve cancer outcomes in patients with lung cancer (Akl et al. 2007) or in the general oncology population (Akl et al. 2007; de Lorenzo et al. 2006; Piccioli et al. 2006). If this effect is further validated and becomes more widely recognized, the proportion of cancer patients receiving anticoagulation is likely to increase signiﬁcantly, and the thrombotic effects of ESAs may no longer be an issue for these patients. In the meantime, it is not uniformly agreed upon which patients should receive primary prophylaxis based solely upon and increased risk. If primary or secondary prophylaxis is decided upon, there is controversy as to which agent, warfarin, anti-platelet drugs (Cloonan et al. 2007), heparin, low molecular weight heparin (LMWH) (Burris 2006; Cunningham 2006; Stine et al. 2007), or the newer factor X antagonists offers the best balance of efﬁcacy, toxicity, and cost in a given clinical setting. In the speciﬁc case of patients with central venous catheters (one of the settings in which ESAs may increase thrombosis rates) there is controversy regarding the value of

754

J. Glaspy

prophylaxis and treatment of catheter-associated thrombosis (Akl et al. 2007; Chan et al. 2007; Cunningham et al. 2006; Fagnani et al. 2007; Freytes 2007; Karthaus et al. 2006; Kovacs et al. 2007; Lee et al. 2006). There are currently no speciﬁc strategies for the prevention of ESA associated thrombosis in cancer patients. It is important, although not surprising, that low dose warfarin is not effective in reducing the incidence of thromboses observed in ESA treated cancer patients (Lin et al. 2006). Given the possible mechanisms noted above, clinical trials of anti-platelet agents should be a high priority in this ﬁeld.

Conclusion Therapy with ESAs is associated with a modest increased risk of venous thrombosis. The mechanism is incompletely understood, but may involve activation of vascular endothelium. Additional studies are needed to improve our ability to predict and prevent this complication.

References 1. Adess M, Eisner R, Nand S, Godwin J, Messmore HL, Wehrmacher WH (2006) Thromboembolism in cancer patients: pathogenesis and treatment. Clin Appl Thromb Hemost 12: 254–266 2. Akl E, Karmath G, Yosuico V, et al (2007) Anticoagulation for thrombosis prophylaxis in cancer patients with central venous catheters. Cochrane Database Syst Rev CD006468 3. Akl E, van Doormaal F, Barba M, et al (2007) Parenteral anticoagulation for prolonging survival in patients with cancer who have no other indication for anticoagulation. Cochrane Database Syst Rev CD006652 4. Akl EA, Kamath G, Kim SY, et al (2007) Oral anticoagulation for prolonging survival in patients with cancer. Cochrane Database Syst Rev CD006466 5. Anagnostou A, Liu Z, Steiner M, et al (1994) Erythropoietin receptor mRNA expression in human endothelial cells. Proc Natl Acad Sci USA 91: 3974–3978 6. Ashley RA, Dubuque SH, Dvorak B, Woodward SS, Williams SK, Kling PJ (2002) Erythropoietin stimulates vasculogenesis in neonatal rat mesenteric microvascular endothelial cells. Pediatr Res 51: 472–478 7. Banerjee D, Rodriguez M, Nag M, Adamson JW (2000) Exposure of endothelial cells to recombinant human erythropoietin induces nitric oxide synthase activity. Kidney Int 57: 1895–1904 8. Beleslin-Cokic BB, Cokic VP, Yu X, Weksler BB, Schechter AN, Noguchi CT (2004) Erythropoietin and hypoxia stimulate erythropoietin receptor and nitric oxide production by endothelial cells. Blood 104: 2073–2080 9. Bohlius J, Wilson J, Seidenfeld J, et al (2006) Recombinant human erythropoietins and cancer patients: updated meta-analysis of 57 studies including 9353 patients J Natl Cancer Inst 98: 708–714

Thrombosis during therapy with erythropoiesis stimulating agents

755

10. Burris HA (2006) Low-molecular-weight heparins in the treatment of cancerassociated thrombosis: a new standard of care? Semin Oncol 33: S3–S16; quiz S41–S42 11. Chan A, Iannucci A, Dager WE (2007) Systemic anticoagulant prophylaxis for central catheter-associated venous thrombosis in cancer patients. Ann Pharmacother 41: 635–641 12. Cloonan ME, DiNapoli M, Mousa SA (2007) Efﬁcacy of anticoagulants and platelet inhibitors in cancer-induced thrombosis. Blood Coagul Fibrinolysis 18: 341–345 13. Cunningham MS, White B, Hollywood D, O’Donnell J (2006) Primary thromboprophylaxis for cancer patients with central venous catheters – a reappraisal of the evidence. Br J Cancer 94: 189–194 14. Cunningham RS (2006) The role of low-molecular-weight heparins as supportive care therapy in cancer-associated thrombosis. Semin Oncol 33: S17–S25; quiz S41–S42 15. De Lorenzo F, Dotsenko O, Scully MF, Tymoshchuk M (2006) The role of anticoagulation in cancer patients: facts and ﬁgures. Anticancer Agents Med Chem 6: 579–587 16. Fagnani D, Franchi R, Porta C, et al (2007) Thrombosis-related complications and mortality in cancer patients with central venous devices: an observational study on the effect of antithrombotic prophylaxis. Ann Oncol 18: 551–555 17. Fareed D, Iqbal O, Tobu M, Hoppensteadt DA, Fareed J (2004) Blood levels of nitric oxide, C-reactive protein, and tumor necrosis factor-alpha are upregulated in patients with malignancy-associated hypercoagulable state: pathophysiologic implications. Clin Appl Thromb Hemost 10: 357–364 18. Fodinger M, Fritsche-Polanz R, Buchmayer H, et al (2000) Erythropoietininducible immediate-early genes in human vascular endothelial cells. J Investig Med 48: 137–149 19. Franchini M, Montagnana M, Targher G, Manzato F, Lippi G (2007) Pathogenesis, clinical and laboratory aspects of thrombosis in cancer. J Thromb Thrombolysis 24: 29–38 20. Freytes CO (2007) Thromboembolic complications related to central venous access catheters in cancer patients. Semin Thromb Hemost 33: 389–396 21. Gale AJ, Gordon SG (2001) Update on tumor cell procoagulant factors. Acta Haematol 106: 25–32 22. Hedenus M, Canon J, Kotasek D, et al (2005) Effects of dose adjustment rules on safety during erythropoietic therapy: a retrospective analysis of darbepoetin alfa administered either every 3 weeks or weekly. Proceedings of the American Society of Hematology Annual Meeting; Abstract 3376 23. Karthaus M, Kretzschmar A, Kroning H, et al (2006) Dalteparin for prevention of catheter-related complications in cancer patients with central venous catheters: ﬁnal results of a double- blind, placebo-controlled phase III trial. Ann Oncol 17: 289–296 24. Khorana AA, Francis CW, Culakova E, Fisher RI, Kuderer NM, Lyman GH (2006) Thromboembolism in hospitalized neutropenic cancer patients. J Clin Oncol 24: 484–490 25. Kovacs MJ, Kahn SR, Rodger M, et al (2007) A pilot study of central venous catheter survival in cancer patients using low-molecular-weight heparin

756

26.

27. 28.

29.

30.

31. 32.

33.

34.

35.

36.

37.

38. 39.

40.

J. Glaspy (dalteparin) and warfarin without catheter removal for the treatment of upper extremity deep vein thrombosis (The Catheter Study). J Thromb Haemost 5: 1650–1653 Lavey RS, Liu PY, Greer BE, et al (2004) Recombinant human erythropoietin as an adjunct to radiation therapy and cisplatin for stage IIB-IVA carcinoma of the cervix: a Southwest Oncology Group study. Gynecol Oncol 95: 145– 151 Lee AY (2003) Epidemiology and management of venous thromboembolism in patients with cancer. Thromb Res 110: 167–172 Lee AY, Levine MN, Butler G, et al (2006) Incidence, risk factors, and outcomes of catheter-related thrombosis in adult patients with cancer. J Clin Oncol 24: 1404–1408 Leonardi MJ, McGory ML, Ko CY (2007) A systematic review of deep venous thrombosis prophylaxis in cancer patients: implications for improving quality. Ann Surg Oncol 14: 929–936 Lin A, Ryu J, Harvey D, Sieracki B, Scudder S, Wun T (2006) Low-dose warfarin does not decrease the rate of thrombosis in patients with cervix and vulvo-vaginal cancer treated with chemotherapy, radiation, and erythropoeitin. Gynecol Oncol 102: 98–102 Piccioli A, Falanga A, Prandoni P (2006) Anticoagulants and cancer survival. Semin Thromb Hemost 32: 810–813 Rak J, Milsom C, May L, Klement P, Yu J (2006) Tissue factor in cancer and angiogenesis: the molecular link between genetic tumor progression, tumor neovascularization, and cancer coagulopathy. Semin Thromb Hemost 32: 54–70 Rak J, Yu JL, Luyendyk J, Mackman N (2006) Oncogenes, trousseau syndrome, and cancer-related changes in the coagulome of mice and humans. Cancer Res 66: 10643–10646 Ross SD, Allen IE, Henry DH, Seaman C, Sercus B, Goodnough LT (2006) Clinical beneﬁts and risks associated with epoetin and darbepoetin in patients with chemotherapy-induced anemia: a systematic review of the literature. Clin Ther 28: 801–831 Satoh K, Kagaya Y, Nakano M, et al (2006) Important role of endogenous erythropoietin system in recruitment of endothelial progenitor cells in hypoxiainduced pulmonary hypertension in mice. Circulation 113: 1442–1450 Stein PD, Beemath A, Meyers FA, Skaf E, Sanchez J, Olson RE (2006) Incidence of venous thromboembolism in patients hospitalized with cancer. Am J Med 119: 60–68 Stine KC, Saylors RL, Saccente CS, Becton DL (2007) Treatment of deep vein thrombosis with enoxaparin in pediatric cancer patients receiving chemotherapy. Clin Appl Thromb Hemost 13: 161–165 Stohlawetz PJ, Dzirlo L, Hergovich N, et al (2000) Effects of erythropoietin on platelet reactivity and thrombopoiesis in humans. Blood 95: 2983–1989 Tagalakis V, Levi D, Agulnik JS, Cohen V, Kasymjanova G, Small D (2007) High risk of deep vein thrombosis in patients with non-small cell lung cancer: a cohort study of 493 patients. J Thorac Oncol 2: 729–734 Tesselaar ME, Romijn FP, Van Der Linden IK, Prins FA, Bertina RM, Osanto S (2007) Microparticle-associated tissue factor activity: a link between cancer and thrombosis? J Thromb Haemost 5: 520–527

Thrombosis during therapy with erythropoiesis stimulating agents

757

41. Tobu M, Iqbal O, Fareed D, et al (2004) Erythropoietin-induced thrombosis as a result of increased inﬂammation and thrombin activatable ﬁbrinolytic inhibitor. Clin Appl Thromb Hemost 10: 225–232 42. Urao N, Okigaki M, Yamada H, et al (2006) Erythropoietin-mobilized endothelial progenitors enhance reendothelialization via Akt-endothelial nitric oxide synthase activation and prevent neointimal hyperplasia. Circ Res 98: 1405– 1413 43. Valles J, Santos MT, Aznar J, et al (1991) Erythrocytes metabolically enhance collagen-induced platelet responsiveness via increased thromboxane production, adenosine diphosphate release, and recruitment. Blood 78: 154–162 44. Valles J, Santos MT, Aznar J, et al (2002) Platelet-erythrocyte interactions enhance alpha(IIb)beta(3) integrin receptor activation and P-selectin expression during platelet recruitment: down-regulation by aspirin ex vivo. Blood 99: 3978–3984 45. Westin SN, Skinner EN, Jonsson Funk M, Gehrig PM, Van Le L (2007) Incidence of symptomatic deep venous thrombosis with epoetin alfa or darbepoetin alfa treatment of anemia in patients with ovarian or primary peritoneal cancer. Gynecol Oncol 105: 414–417 46. Wun T, Law L, Harvey D, Sieracki B, Scudder SA, Ryu JK (2003) Increased incidence of symptomatic venous thrombosis in patients with cervical carcinoma treated with concurrent chemotherapy, radiation, and erythropoietin. Cancer 98: 1514–1520 47. Wun T, Paglieroni T, Hammond WP, Kaushansky K, Foster DC (1997) Thrombopoietin is synergistic with other hematopoietic growth factors and physiologic platelet agonists for platelet activation in vitro. Am J Hematol 54: 225–232 48. Zwicker JI, Furie BC, Furie B (2007) Cancer-associated thrombosis. Crit Rev Oncol Hematol 62: 126–136 Correspondence: John Glaspy, MD, Professor of Medicine, Division of HematologyOncology, UCLA School of Medicine, Los Angeles, California 90095, USA, E-mail: [email protected]

Chapter 31

The effect of rhEPO on survival in anemic cancer patients T. J. Littlewood Department of Haematology, John Radcliffe Hospital, Oxford, UK

Introduction Anaemia is common in patients with malignant disease and most commonly will be attributed to the anaemia of chronic disease (Spivak 1994) and to the myelosuppressive effects of treatment with chemo- and/or radiotherapy. Erythropoietin deﬁciency is found in the majority of patients with cancer related anaemia (Miller et al. 1990). The frequency of anaemia in patients with cancer varies according to tumour type and treatment (Table 1). Anaemia results in important symptoms for the patient (Thomas 1998). Common symptoms include fatigue, breathlessness, swollen feet, chest pain and loss of mental acuity. Anaemia not only results in an impaired quality of life but can have an adverse effect on the patient’s life expectancy (Manegold 1998). The presence of anaemia at diagnosis is an adverse prognostic ﬁnding for patients with many tumour types. For example, a haemoglobin of <10.0 g/dl places patients with myeloma (Durie and Salmon 1975) and chronic lymphatic leukaemia (Binet et al. 1981) into the most advanced stages of disease (Durie-Salmon and Binet staging systems, respectively) irrespective of other disease features. Data from Albain et al. (1991) indicate that in patients with non-small cell lung cancer a haemoglobin level of <11.0 g/dl at the time of diagnosis was one of the most important adverse prognostic factors for eventual survival, and Ohlhauser et al. (1997), in the same patient group, showed an adverse prognosis for patients with a haemoglobin of <12.7 g/dl at diagnosis. A low haemoglobin concentration may impair survival by: i) Reducing the effectiveness of chemo- and radiotherapy, ii) Indirect mechanisms resulting from a decrease in the patients’ quality of life, iii) Decreasing the amount of treatment delivered to the patient possibly because of ii above, iv) Impairing tumour oxygenation.

760

T. J. Littlewood

Table 1. Frequency of anaemia in various types of malignancies Malignancy

% of patients

Hb level

Myeloma (San Miguel et al. 2001)

50% on presentation. Majority during initial chemotherapy 40% at diagnosis. >50% after 3 cycles of chemotherapy 38% during treatment. (33% required at least one blood transfusion)

<10.5 g/dl

Non-Hodgkin’s lymphoma (Coifﬁer et al. 2001) Solid tumours (n = 2,719 varied types) (Barrett-Lee et al. 2000)

<12.0 g/dl <11.0 g/dl

Tumour hypoxia occurs when tumour growth exceeds the ability of the local microvasculature to supply oxygen to tumour cells and several studies have shown that tumours are generally more hypoxic than surrounding normal tissues (Molls et al. 1998). Tumour oxygenation is mainly affected by the rate of blood ﬂow, by the microcirculation and by the haemoglobin concentration. Correcting the haemoglobin level may improve tumour oxygenation (Kelleher et al. 1996). Many studies show that, irrespective of treatment, patients with hypoxic tumours are likely to have a smaller chance of local disease control and are less likely to be cured of their disease compared to patients with better oxygenated tumours of the same size and stage (Höckel et al. 1996; Brizel et al. 1997). Explanations for the adverse impact of tumour hypoxia on survival (Höckel et al. 1999) include: 1) Hypoxia may induce changes in expression of oxygen dependent proteins within the tumour cells such as vascular endothelial growth factor (VEGF), which stimulates angiogenesis and increases the potential for tumour growth and metastases. 2) Ionising radiation results in the formation of free radicals within cells. In the presence of oxygen the free radicals are ﬁxed and interact with DNA and cell membranes to cause cell death. When cells are hypoxic, free radicals are not ﬁxed and cell death may not occur (Hall 1994). 3) Hypoxia may produce a growth advantage for tumour cells which are resistant to apoptosis with a decrease in the potential for cure or control, and hypoxic tumours may overexpress the tumour suppressor gene p53, a cell phenotype with a higher malignant potential. The response of cells exposed to hypoxia is to decrease overall protein synthesis, which in turn will lead to slow cell growth and eventual apoptosis. This

The effect of rhEPO on survival in anemic cancer patients

761

beneﬁcial effect (in terms of tumour control) is, however, counterbalanced by the promotion of tumour progression induced by proteome and genomic changes caused by the hypoxic environment. Because there is a clear relationship between anaemia and poor outcome for many tumour types, many investigators have looked at the impact of anaemia correction on tumour outcomes. Most studies have investigated the role of recombinant erythropoietin (rHuEpo) but despite a decade or more of research many unanswered questions remain. The most important of these is whether anaemia correction has a positive impact on survival and, if so, whether this applies to all tumours treated with either chemotherapy and radiotherapy or just some. Also, if there is a beneﬁcial effect of rHuEpo, is this just via correction of anaemia or are there other possible biological effects? In the next part of this chapter I will review the impact on tumour control of anaemia correction in laboratory animals and humans being treated with either chemo- or radiotherapy.

Preclinical studies in radiation treated animals Thews et al. (1998) set up an experiment to look at the radiosensitivity of sarcomas implanted into rats. Three models were used. Rats implanted with tumour, which were not anaemic, rats implanted with tumour, which had previously been made anaemic by treatment with carboplatin and, lastly, rats implanted with tumour, pretreated with carboplatin and then treated with rHuEpo to correct anaemia. After noncurative radiotherapy, the delay before tumour regrowth occurred was 4.5 days in the anaemic rats, 9.5 days in the rHuEpo treated rats and 12.0 days in the nonanaemic rats. These results suggest that correcting anaemia increases the antitumour effect of the radiotherapy. A study in mice by Stüben et al. (2002) produced identical results. In this study, nude mice were implanted with a human glioblastoma. One group of mice was made anaemic by pretreatment with radiotherapy, one group treated with radiotherapy had anaemia corrected by rHuEpo and the third group were not pretreated and were never anaemic. The control (never anaemic mice) showed the greatest radiosensitivity, followed by the rHuEpo treated mice. The anaemic mice had the most radioresistant tumours. Both this study and that by Thews et al. (1998) also noted that treatment with rHuEpo did not stimulate tumour growth.

Preclinical studies of chemotherapy Just as the data above suggest that the tumours of anaemic animals are more radio-resistant than those from nonanaemic ones, similar data exist in chemotherapy-treated animals.

762

T. J. Littlewood

Severe combined immunodeﬁcient (SCID) mice, with human ovarian cancer xenografts, were divided into four groups. Group 1 was treated with cisplatin, group 2 with rHuEpo, group 3 with both cisplatin and rHuEpo and group 4 with saline solution. Mice treated with combined cisplatin and rHuEpo had a greater antitumour response than cisplatin-alone mice. Tumour growth was similar in the control and rHuEpo groups, showing that erythropoietin did not stimulate tumour growth (Silver and Piver 1999). In another study, in a murine ﬁbrosarcoma model, hypoxic and normoxic tumour cells were exposed to a variety of cytotoxic drugs. Hypoxic cells were found to be 2-6-fold more chemoresistant to cytotoxic drugs such as cyclophosphamide, BCNU, carboplatin and melphalan than normoxic tumour cells (Teicher et al. 1990). The impact of anaemia on the cytotoxicity of cyclophosphamide has also been investigated in a rat model (Thews et al. 2001). Anaemic rats had evidence of tumour regrowth at 8.6 days after cyclophosphamide treatment compared to re-growth at 13.3 days for the nonanaemic control group and the group with anaemia corrected by rHuEpo. This study again suggests that this tumour was more sensitive to treatment with cyclophosphamide if either the animal had never been anaemic or if anaemia had been corrected by rHuEpo than in the anaemic control animals. If treatment with rHuEpo has a beneﬁcial effect on survival, the most likely explanation is thought to be by anaemia correction resulting in a reduction in tumour hypoxia. A study in a mouse myeloma model has hypothesized that treatment with erythropoietin prolongs survival compared to a placebo-treated group and that the therapeutic effect was attributed to a T-cell triggered mechanism (Mittelman et al. 2001). This possible immunomodulatory role of rHuEpo requires further investigation.

Clinical studies of radiotherapy A large study conducted in Canada (Grogan et al. 1999) in patients with cervical cancer treated with radiotherapy showed that the 5-year survival was 74% in patients whose mean haemoglobin during the course of treatment was >12.0 g/dl, 52% for those whose mean haemoglobin was >11.0 and <12.0 g/dl and 45% for those patients whose mean haemoglobin was <11.0 g/dl. An interesting aspect of this study was that patients who were transfused and who attained a speciﬁc haemoglobin level had a survival rate similar to those patients who achieved that haemoglobin level spontaneously. This study was a retrospective analysis but provides a pointer that anaemia correction (by blood transfusion in this case) improves outcomes in this population of patients with cervical cancer. In another study, 889 patients with squamous cell tumours of the head and neck were treated with radiotherapy (Frommhold et al. 1998). Anaemic patients had an estimated 5-year chance of survival of 28.4% versus 58.2%

The effect of rhEPO on survival in anemic cancer patients

763

Table 2. Outcome of chemotherapy (using 5-FU and mitomycin C) and radiotherapy in anaemic patients with head and neck cancer receiving or not receiving (historical control group) rhEPO (Glaser et al. 2001)

Number of patients Pathological response (%) 2-Year locoregional control (%) 2 Year overall survival (%)

RhEPO

No rhEPO

p value

57 61 95 88

87 17 72 60

<0.001 <0.001 <0.001

Pathological complete response to neoadjuvant chemotherapy, 2 year actuarial locoregional tumor control; and 2 year actuarial overall survival rates

for nonanaemic patients. In a multivariate analysis the effect of haemoglobin on survival was independent of other risk factors such as tumour site, stage, or type of treatment. A study from Vienna (Glaser et al. 2001) reported on the effectiveness of rhEPO in treating the anaemia of patients with head and neck cancer being treated with chemotherapy (using 5-FU and mitomycin C) and radiotherapy. Resection of the oral cavity tumour and bilateral neck dissection was carried out 5 weeks following completion of the neoadjuvant treatment. The results in the group treated with rhEPO were compared with a historical control group not treated with rhEPO. The results (Table 2) indicate that rhEPO improves the haemoglobin level, reduces transfusion need but also signiﬁcantly improves the response rate and the chance of local control and overall survival. The conﬁdent mood based on these studies was jolted by a paper published by Henke et al. (2003) reporting on a study of patients with head and neck tumours treated with radiotherapy and randomized to either rHuEpo or placebo. The survival in the rHuEpo-treated patients was poorer than in the placebo group. Detailed analysis of this report suggested that there may have been imbalances between the treatment and placebo arms in the type and grade of tumours and for whether the patients were cigarette smokers or not. It was also noted that the rHuEpo group had a signiﬁcant percentage of patients with a haemoglobin concentration above normal. All of these factors may have had an effect on the quality and importance of the data collected. Nevertheless, this study made all investigators in this ﬁeld reﬂect.

Clinical studies of chemotherapy A large, placebo-controlled, randomized study of rhEPO was conducted in patients receiving non-platinum-containing chemotherapy (Littlewood et al. 2001). The original objectives of this trial were to assess the effects of rhEPO

764

T. J. Littlewood

on transfusion requirements, haemoglobin level, quality of life and safety. Before the study was unblinded an additional objective of exploring a possible relationship between increased haemoglobin and survival was included. A total of 375 patients with solid or nonmyeloid haematological malignancies was entered. All patients were anaemic at study entry either with a baseline haemoglobin of <10.5 g/dl or with a haemoglobin of >10.5 but less than 12.0 g/dl following a Hb decrease of >1.5 g/dl in the previous month. Patients were randomized to receive rhEPO or placebo in a 2 : 1 ratio in favour of rhEPO. Study medication was given by subcutaneous injection three times per week at a dose of 150–300 u/kg body weight for 12–24 weeks. Compared with placebo, patients treated with rhEPO signiﬁcantly decreased transfusion requirements, increased the haemoglobin concentration and improved quality of life. The survival analysis was conducted 12 months after the last patient had completed the trial and before the study was unblinded. At the 12-month assessment, 135 patients were alive and 237 had died. For the whole group the median survival times were 17 months for the rhEPO group compared to 11 months for the placebo group (Fig. 1). The Kaplan-Meier 12-month estimates showed a trend towards better overall survival favouring rhEPO (p = 0.13; log rank test). An analysis of survival according to whether the patient had a haematological malignancy or solid tumour still showed an apparent survival advantage for the rhEPO treated group (Fig. 2).

Probability of being alive N Alive/Lost Died Epoetin alfa 251 96 155 Placebo 124 42 82

1.0

0.8

0.6

0.4

0.2

0.0 0 (p =0.126, log-rank test)

5

10

15

20

25

30

35

Months since start of study

Fig. 1. Kaplan-Meier estimates of survival

40

The effect of rhEPO on survival in anemic cancer patients N

Alive/Lost

Died

Hematologic Epoetin alfa

115

61

54

Hematologic Placebo

58

28

30

136

35

101

66

14

52

Probability of being alive 1.0

0.8

765

Solid Epoetin alfa Solid Placebo

0.6

0.4

0.2

0.0 0

5

10

15

20

25

30

35

40

Months since start of study

Fig. 2. Kaplan-Meier estimates of survival by tumor type

The survival data reported in this study provide additional support for the results from patients treated with radiotherapy with or without chemotherapy. However, there needs to be caution in interpretation. The study was neither designed, nor powered to determine survival, and a number of variables which could have inﬂuenced survival such as tumour stage, intensity of chemotherapy, extent of bone marrow involvement and disease progression were neither controlled nor stratiﬁed in the study and these data were not collected during the follow-up period. Several other studies have looked at the impact of treatment with rHuEpo on survival in patients receiving chemotherapy. One, in patients with lung cancer, reported by Vansteenkiste et al. (2002) showed that rHuEpo patients had a median life expectancy of 46 weeks compared to 34 weeks in the placebo group. A very recent report in patients with lymphoproliferative diseases has shown no difference in life expectancy between the treatment and placebo groups (Österborg et al. 2005) and a study in patients with advanced, incurable cancer also showed no survival difference between the erythropoietin and placebo groups (Witzig et al. 2005). Finally, a study in metastatic breast cancer was terminated early because of a worse survival in the rHuEpo than the placebo group (Leyland Jones 2003). All of these studies, both with a positive, negative or neutral outcome, can be criticized for ﬂaws in their designs, and the question of whether

766

T. J. Littlewood

anaemia correction improves outcome in anaemic patients with cancer remains uncertain.

Conclusions The majority of doctors were trained to believe that mild to moderate anaemia in patients with cancer was to be expected and that the anaemia did not cause the patient any symptoms until the haemoglobin fell to less than 10.0 g/dl or even less than 8.0 g/dl. Until the last decade, blood transfusion was the only reliable treatment for the anaemia but transfusions are still rather sparingly used. The incidence of anaemia in patients with cancer, and its impact on quality of life, has been excellently conﬁrmed by a large European study in which over 15,000 patients with cancer were studied (Ludwig et al. 2004). This study also noted that most patients with anaemia do not receive corrective treatment. Recent data from large community-based studies (Glaspy et al. 1997; Demetri et al. 1998) and from randomized trials (Hedenus et al. 2003) strongly support the notion that anaemia has a negative impact on the patients’ quality of life and that correcting the anaemia using erythropoietin will provide an objective improvement in the patients’ well-being. There remains considerable uncertainty as to whether anaemia correction (by transfusion or recombinant erythropoietin) has a beneﬁcial, negative or no impact on tumour response to treatment and overall survival. A Cochrane meta-analysis of cancer outcomes in patients randomized to rHuEpo or placebo reported beneﬁt for the erythropoietin treated patients (Bohlius et al. 2004). This analysis, however, did not include the two negative studies recently reported and cited above (Henke et al. 2003; LeylandJones 2003). In these studies, on the other hand, mainly nonanaemic cancer patients were treated, and it is therefore questionable, if they should be included in a meta-analysis evaluating the effects of rhEPO on survival in anaemic cancer patients. However, in a recent update of the abovementioned meta-analysis including these two studies, no conclusive results were obtained with regard to the impact of rhEPO on survival (Bohlius et al. 2006). Erythropoietin receptors have been found on the surface of some tumour cells, raising concerns that treatment with erythropoietin might stimulate these and enhance tumour growth. In vitro data have also suggested that treatment with erythropoietin might simulate vascular endothelial cells and angiogenesis and, hence, contribute to tumour growth. No such evidence has been found in clinical studies (Farrell and Lee 2004). In addition, the validity of antibodies used to identify EPO receptors and the functionality of receptors detected have been questioned in some recent studies (Elliott et al. 2006; Osterborg et al. 2007).

The effect of rhEPO on survival in anemic cancer patients

767

The main problem in this ﬁeld is the lack of excellently designed trials. My personal opinion is that, at worst, erythropoietin will not be disadvantageous as long as it is used to correct the haemoglobin concentration to normal and not supra-normal levels. I also believe that there may be some tumour types (such as cervical cancer) where anaemia correction has an important positive impact on survival. The reality is that for many tumours we will probably never be sure whether anaemia correction is helpful in increasing the antitumour response to treatment and in prolonging life expectancy.

References 1. Albain KS, Crowley JJ, LeBlanc M, Livingston RB (1991) Survival determinants in extensive-stage non-small-cell lung cancer: The Southwest Oncology Group experience. J Clin Oncol 9: 1618–1626 2. Barrett-Lee PJ, Bailey NP, O’Brien MER, Wager E (2000) Large scale UK audit of blood transfusion requirements and anaemia in patients receiving cytotoxic chemotherapy. Br J Cancer 82: 93–97 3. Binet JL, Auquier A, Dighiero G, Chastang C, Piguet H, Goasguen J, Vaugier G, Potron G, Colona P, Oberling F, Thomas M, Tchernia G, Jacquillat C, Boivin P, Lesty C, Duault MT, Monconduit M, Belabbes S, Gremy F (1981) A new prognostic classiﬁcation of chronic lymphocytic leukemia derived from a multivariate survival analysis. Cancer 48: 198–206 4. Bohlius J, Langensiepen S, Schwarzer G, Seidenfeld J, Piper M, Bennett C, Engert A (2005) Recombinant human erythropoietin and overall survival in cancer patients: results of a comprehensive meta-analysis. J Natl Cancer Inst 97: 489–498 5. Bohlius J, Wilson J, Seidenfeld J, Piper M, Schwarzer G, Sandercock J, Trelle S, Weingart O, Bayliss S, Djulbegovic B, Bennett C, Langensiepen S, Hyde C, Engert A (2006) Recombinant human erythropoietins and cancer patients: updated meta-analysis of 57 studies including 9353 patients. J Natl Cancer Inst 98: 708–714 6. Brizel DM, Sibley GS, Prosnitz LR, Scher RL, Dewhurst MW (1997) Tumor hypoxia adversely affects the prognosis of carcinoma of the head and neck. Int J Radiat Oncol Biol Phys 38: 285–289 7. Coifﬁer B, Guastalla JP, Pujade-Lauraine E, Bastit P (2001) Predicting cancerassociated anaemia in patients receiving non-platinum chemotherapy: results of a retrospective survey. Eur J Cancer 37: 1617–1623 8. Demetri GD, Kris M, Wade J, Degos L, Cella D (1998) Quality-of-life beneﬁt in chemotherapy patients treated with epoetin alfa is independent of disease response or tumor type: results from a prospective community oncology study. J Clin Oncol 16: 3412–3425 9. Durie BGM, Salmon SE (1975) A clinical staging system for multiple myeloma. Correlation of measured myeloma cell mass with presenting clinical features, response to treatment, and survival. Cancer 36: 842–854 10. Farrell F, Lee A (2004) The erythropoietin receptor and its expression in tumor cells and other tissues. Oncologist 9 [Suppl 5]: 18–30

768

T. J. Littlewood

11. Frommhold H, Guttenberger R, Henke M (1998) The impact of blood haemoglobin content on the outcome of radiotherapy. The Freiburg experience. Strahlenther Onkol 174 [Suppl 4]: 31–34 12. Elliott S, Busse L, Bass M, Lu H, Sarosi I, Sinclair A, Spahr C, Um M, Van G, Begley C (2006) Anti-Epo receptor antibodies do not predict Epo receptor expression. Blood 107: 1892–1895 13. Glaspy J, Bukowski R, Steinberg D, Taylor C, Tchekmedyian S, Vadhan-Raj S for the Procrit Study Group (1997) Impact of therapy with epoetin alfa on clinical outcomes in patients with nonmyeloid malignancies during cancer chemotherapy in community oncology practice. J Clin Oncol 15: 1218–1234 14. Glaser C, Millesi W, Kornek G, Lang S, Schull B, Watzinger F, Selzer E, Lavey RS (2001) Impact of hemoglobin level and use of recombinant erythropoietin on efﬁcacy of preoperative chemoradiation therapy for squamous cell carcinoma of the oral cavity and oropharynx. Int J Radiat Oncol Biol Phys 50: 705–715 15. Grogan M, Thomas GM, Melamed I, Wong FL, Pearcey RG, Joseph PK, Portelance L, Crook J, Jones Kd (1999) The importance of hemoglobin levels during radiotherapy for carcinoma of the cervix. Cancer 86: 1528–1536 16. Hall EJ (1994) The oxygen effect and reoxygenation. In: Hall EJ (ed) Radiobiology for the radiologist, 4th edn. Lippincott, Philadelphia, pp 133–152 17. Hedenus M, Adriannson M, San Miguel J, Kramer MH, Schipperus MR, Juvonen E, Taylor K, Belch A, Altes A, Martinelli G, Watson D, Matcham J, Rossi G, Littlewood TJ (2003) Efﬁcacy and safety of darbepoietin alfa in anaemic patients with lymphoproliferative malignancies: a randomized, double-blind, placebocontrolled study. Br J Haematol 122: 394–403 18. Henke M, Laszig R, Rube C, Schafer U, Haase ID, Schilcher B, Mose S, Beer KT, Burger U, Dougherty C, Frommholz H (2003) Erythropoietin to treat head and neck cancer patients with anaemia undergoing radiotherapy: randomised, double-blind, placebo-controlled trial. Lancet 362: 1255–1260 19. Höckel M, Schlenger K, Höckel S, Vaupel P (1999) Association between tumor hypoxia and malignant progression: the clinical evidence in cancer of the uterine cervix. In: Vaupel P, Kelleher DK (eds) Tumor hypoxia. Wissenschaftliche Verlagsgesellschaft, Stuttgart, pp 65–74 20. Hockel M, Schlenger K, Arral B, Mitze M, Scaffer U, Vaupel P (1996) Association between tumor hypoxia and malignant progression in advanced carcinoma of the uterine cervix. Cancer Res 56: 4509–4515 21. Kelleher DK, Matthiensen U, Thews O, Vaupel P (1996) Blood ﬂow, oxygenation and bioenergetic status of tumors after erythropoietin treatment in normal and anemic rats. Cancer Res 56: 4728–4734 22. Leyland-Jones B, BEST Investigators and Study Group (2003) Breast cancer trial with erythropoietin terminated unexpectedly. Lancet Oncol 4: 459–460 23. Littlewood TJ, Bajetta E, Nortier JWR, Vercammen E, Rapoport B (2001) Effects of epoetin alfa on hematologic parameters and quality of life in cancer patients receiving nonplatinum chemotherapy: results of a randomized, doubleblind, placebo-controlled trial. J Clin Oncol 19: 2865–2874 24. Ludwig H, Van Belle S, Barrett-Lee P, Birgegard G, bokemeyer C, Gascon P, Kosmidis P, Nortier J, Olmi P, Schneider M, Schrijvers D (2004) The European cancer anaemia survey (ECAS) : a large, multinational, prospective survey deﬁn-

The effect of rhEPO on survival in anemic cancer patients

25. 26.

27.

28. 29. 30.

31.

32.

33.

34. 35.

36.

37.

38.

39. 40.

769

ing the prevalence, incidence and treatment of anaemia in cancer patients. Eur J Cancer 40: 2293–2306 Manegold C (1998) The causes and prognostic signiﬁcance of low haemoglobin levels in tumor patients. Strahlenther Oncol 174 [Suppl 4]: 17–19 Miller CB, Jones RJ, Piantadosi S, Abeloff MD, Spivak JL (1990) Decreased erythropoietin response in patients with the anemia of cancer. N Engl J Med 322: 1689–1692 Mittelman M, Neumann D, Peled A, Kanter P, Haran-Ghera N (2001) Erythropoietin induces tumor regression and antitumor responses in murine myeloma model. Proc Natl Acad Sci USA 98: 5181–5186 Molls M, Stadler P, Becker A, Feldmann HJ, Dunst J (1998) Relevance of oxygen in radiation oncology. Strahlenther Onkol 174 [Suppl 4]: 13–16 Ohilhauser C, Bulzebruck H, Ebert W, et al (1997) Prognostic factors for survival in inoperable non small cell lung cancer. Oncology 20: 126–131 Osterborg A, Brandberg Y, Hedenus M (2005) Impact of epoetin-beta on survival of patients with lymphoproliferative malignancies: long term follow-up of a large randomized study. Br J Haematol 129: 206–209 Osterborg A, Aapro M, Cornes P, Haselbeck A, Hayward CR, Jelkmann W (2007) Preclinical studies of erythropoietin receptor expression in tumour cells: impact on clinical use of erythropoietic proteins to correct cancer-related anaemia. Eur J Cancer 43: 510–519 San Miguel JF, Garcia-Sanz R, González M, Moro MJ, Hernández JM, Ortega F, Borrego D, Carnero M, Casanova F, Jiménez R, Portero JA, Orfão (1995) A new staging system for multiple myeloma based on the number of S-phase plasma cells. Blood 85: 448–455 Silver DF, Piver MS (1999) Effects of recombinant erythropoietin on the antitumor effect of cisplatin in SCID mice bearing human ovarian cancer: a possible oxygen effect. Gynecol Oncol 73: 280–284 Spivak JL (1994) Recombinant human erythropoietin and the anemia of cancer. Blood 84: 997–1004 Stuben G, Pottgen C, Knuhmann K, Schmidt K, Stuschke M, Thews O, Vaupel P (2003) Erythropoietin restores the anaemia-induced reduction in radiosensitivity of experimental human tumours in nude mice. Int J Radiat Oncol Biol Phys 55: 1358–1362 Teicher BA, Holden SA, Al-Achi A, Herman TS (1990) Classiﬁcation of antineoplastic treatments by their differential toxicity toward putative oxygenated and hypoxic tumor subpopulations in vivo in the FSaIIC murine ﬁbrosarcoma. Cancer Res 50: 3339–3344 Thews O, Koenig R, Kelleher DK, Kutzner J, Vaupel P (1998) Enhanced radiosensitivity in experimental tumours following erythropoietin treatment of chemotherapy-induced anaemia. Br J Cancer 78: 752–756 Thews O, Kelleher DK, Vaupel P (2001) Erythropoietin restores the anemiainduced reduction in cyclophosphamide cytotoxicity in rat tumors. Cancer Res 61: 1358–1361 Thomas ML (1998) Impact of anemia and fatigue on quality of life in cancer patients: a brief review. Med Oncol 15 [Suppl 1]: S3–S7 Vansteenkiste J, Pirker R, Massuti B, Barata F, Fout A, Fiegl M, Siena S, Gateley J, Tomita D, Colowick AB, Musil J (2002) Double-blind, placebo-controlled,

770

T. J. Littlewood: The effect of rhEPO on survival

randomized phase III trial of darbepoietin alfa in lung cancer patients receiving chemotherapy. J Natl Cancer Inst 94: 1211–1220 41. Witzig TE, Silberstein PT, Loprinzi CL, Sloan JA, Novotny PJ, Mailliard JA, Kendrith MR, Alberts SR, Krook JE, Levitt R, Morton RF (2005) Phase III, randomized, double-blind study of epoetin alfa compared with placebo in anemic patients receiving chemotherapy. J Clin Oncol 23: 2606–2617 Correspondence: Tim J. Littlewood, M.D., Department of Haematology, John Radcliffe Hospital, Oxford OX3 9DU, UK, E-mail: [email protected].

Chapter 32

From bench to bedside: Neuroprotective effects of erythropoietin H. Ehrenreich and C. Bartels Division of Clinical Neuroscience, Max-Planck-Institute of Experimental Medicine, Göttingen, Germany

Introduction Western industrialized countries have to deal with ever increasing numbers of patients suffering from diseases of the nervous system. This growing problem poses a growing challenge for novel medical strategies and is based on an increasingly aging population due to considerable improvements in general health care/treatment outcome as well as alterations in lifestyle. The prototype of an acute brain disease, stroke, constitutes together with cardiovascular diseases, the second leading cause of death after cancer in industrialized nations. Among the prototype chronic disorders of the brain is Alzheimer’s disease with its dramatically increasing prevalence due to the increased life expectancy of the population (Ferri et al. 2005). Diseases of the nervous system are mostly heterogeneous in origin, i.e. they are the result of an array of multiple possible causes of disease, ranging from a large number of different genetically predisposing factors to a variety of environmental cofactors. Environmental cofactors include social factors, stress, neurotrauma, infections, drug abuse, or comorbidity, to just name a few. In addition, there may be dozens of genes variably interacting to create what we call “genetic predisposition”. This array of causes with some diseaseoverlapping factors leads to a disease-speciﬁc common clinical phenotype. This common clinical phenotype is the basis of our classiﬁcation of neuropsychiatric disorders (Fig. 1). The complex etiology of neuropsychiatric diseases makes it very clear that there is virtually no etiological treatment available. Also, there is not too much expectation that there will be a speciﬁc causal therapy or a cure within the next decades for any of the most frequent diseases of the nervous system. There might, at best, be some retrospective identiﬁcation of potential risk factors, useful to develop preventive strategies. The most relevant diseases of the nervous system share not only the features of being multifactorial, but also of being frequent, expensive, devastating, non-curable, and allow no hope for a cure within the lifespan of presently

772

H. Ehrenreich and C. Bartels Multiple possible causes of disease Development into a common* clinical phenotype

* clinically classifiable Fig. 1. Most diseases of the brain are heterogeneous in origin

Multiple neuropsychiatric diseases "Final common pathway"

TARGET of NEUROPROTECTION Fig. 2. Most diseases of the brain share pathophysiological features

living generations. Thus, rather than aiming at a cure, we have to (1) intensify all kinds of prophylactic approaches to prevent/delay disease onset, and (2) search for improvement of these conditions and/or cessation of their progression. Therefore, therapeutic targets based on the concept of the “ﬁnal common pathway” appear promising. As illustrated in Fig. 2, diseases of the brain share pathophysiological features that lead to disease progression, independent of the actual disease etiology. These common features, building the ﬁnal common pathway, range from increased neuronal apoptosis, decreased neurogenesis, formation of radical oxygen or nitrogen species, incapacitated axonal sprouting, altered synaptogenesis and synapse function, disturbed calcium metabolism, to accumulation of cellular aggregates or neurovascular dysfunction, to just name the most important mechanisms. Novel therapeutic approaches target one or several components of this ﬁnal common pathway, thereby providing what we call “neuroprotection”. Neuroprotection can be deﬁned as the “attempt to maintain the highest possible integrity of cellular interactions in the brain, resulting in a protection of neural function” (Ehrenreich and Sirén 2001). Candidate compounds for neuroprotection have to fulﬁll certain properties that enable them to efﬁciently combat components of the ﬁnal common pathway. Having shown their neuroprotective potential both in vitro and in vivo, they are expected

From bench to bedside: Neuroprotective effects of erythropoietin

773

to have favorable pharmacokinetics in man, supporting their availability in areas at distress. Moreover, they have to be well tolerated and safe. Based on these conditions, the best strategy may always be to test these compounds in ﬁrst proof-of-principle studies in man. Provided that these are successful, it is reasonable and easy to go back to more extensive experiments in animals and cell culture in order to speciﬁcally explore the respective mechanisms of action. Among a plethora of potential neuroprotective factors, erythropoietin (EPO) has turned out in recent years to be one of the most promising agents (Jelkmann 1992; Campana et al. 1998; Dame et al. 2001; Ehrenreich and Sirén 2001; Sirén and Ehrenreich 2001; Gassmann et al. 2003; Ehrenreich 2004; Ehrenreich et al. 2004a; Juul 2004; Brines and Cerami 2005). In fact, the renal hormone EPO, stimulator of red blood cell production, has started a new carrier.

Evidence for a role of EPO in the nervous system: Experience from in vitro and in vivo studies Expression of EPO and its receptor in the nervous system and role of HIF-1 EPO and EPO receptor (EPO-R) were found to be abundantly expressed not only in neurons and glial cells in culture (Masuda et al. 1993; Masuda et al. 1994; Masuda et al. 1997; Lewczuk et al. 2000; Nagai et al. 2001), but also in the embryonic rodent brain and spinal cord (Knabe et al. 2004; Knabe et al. 2005) as well as in the developing human central nervous system (Juul et al. 1998) and to decrease considerably after birth (Ehrenreich et al. 2005). Their expression in postnatal and adult brain tissue is low under normal conditions (Marti et al. 1996), but a stimulation of EPO and EPO-R expression is observed upon hypoxia, ischemia, inﬂammation or neurodegeneration (Sirén et al. 2001b; Ehrenreich et al. 2004b; Eid et al. 2004). It appears that the powerful multifaceted role of the EPO system in the developing nervous system ranges from (1) termination of waves of physiological apoptosis in the brain to (2) inﬂuence on production and differentiation of neuronal progenitors. This role seems to gradually vanish during brain maturation, to be potentially re-activated in conditions of distress within the nervous system. Independent of particular etiological factors like blood loss, anemia, alterations in circulatory functions or inﬂammatory conditions, low oxygen tension of tissues is the most prominent stimulator of the expression of the EPO system (Bauer and Kurtz 1989; Chikuma et al. 2000). Hypoxia-induced EPO expression is mediated by the transcription factor hypoxia-inducible factor-1 (HIF-1), which consists of two subunits: HIF-1α and HIF-1β (Semenza 2001; Sharp and Bernaudin 2004; Eckardt and Kurtz 2005). Whereas HIF-1β is constitutively expressed in all cells and tissues and is

774

H. Ehrenreich and C. Bartels

non-responsive to hypoxia, it is essential for HIF-1 transcriptional activity. In contrast, HIF-1α is rapidly hydroxylated under normoxia, leading to its degradation. Under hypoxia, HIF-1α remains stable and, as part of the HIF1 complex, provokes the transcription/activation of a large number of hypoxia-inducible genes, among them EPO. Interestingly, similar to hypoxia, iron chelators are able to lead to HIF-1 stabilization since iron is needed for enzymes degrading HIF-1α. This iron dependency of HIF-1α degrading enzymes may be of therapeutic interest when applying EPO for neuroprotective strategies (Semenza 2001; Sharp and Bernaudin 2004). Mild to moderate iron deﬁciency caused by long-term EPO treatment without iron substitution does not only limit the hematopoietic “side effects” of EPO in neurodegenerative indications, but may also add to the neuroprotective EPO actions, e.g. in multiple sclerosis (unpublished observations). Indeed, an increased iron turnover in non-anemic chronic progressive multiple sclerosis patients has been associated with an ongoing inﬂammatory process (Sfagos et al. 2005). Mechanisms underlying the activation of EPO-R are less clear as compared to activation of EPO. Although it has been shown that hypoxia induces EPO-R expression in endothelial (Beleslin-Cokic et al. 2004) and neuronal cells (Lewczuk et al. 2000; Ehrenreich et al. 2005), the conditions of its expression in brain tissue are still obscure. In fact, HIF-1 may not be the only inducer of EPO-R expression in these situations (Sharp and Bernaudin 2004). We have systematically studied the cellular expression pattern of EPO and EPO-R in the brains of patients who suffered an ischemic stroke and died at various time-points thereafter. It turned out that in the early phase of stroke there is a predominantly neuronal and endothelial expression pattern of both EPO and EPO-R, and even brain-invading mononuclear cells show positive immunoreactivity (Sirén et al. 2001b). During that early phase after ischemic stroke in man, apoptosis in the periinfarct area plays a major role (Sairanen et al. 2006), and EPO/EPO-R are very likely to act here in an antiapoptotic fashion, thus representing a powerful endogenous neuroprotective system. In fact, removing endogenous EPO after experimental stroke by intracerebroventricular infusion of soluble EPO-R leads to a dramatic increase in lesion size (Sakanaka et al. 1998). In addition, regarding neuronal EPO-R, the strong expression by nerve ﬁbers may point to a prominent function of the EPO system in axonal protection in the brain (Sirén et al. 2001b). Weeks after a stroke, the cellular expression pattern shifts to mainly glial expression of both EPO and EPO-R, which appears to persist even many months after an ischemic event in the brain (Sirén et al. 2001b). An upregulation of the glial EPO system has been interpreted as one feature of an acquired ischemic tolerance (Bernaudin et al. 2002; Dawson 2002; Ruscher et al. 2002; Dirnagl et al. 2003). Age-matched control brains from patients who died of non-brain-related conditions showed only faint and scattered neuronal expression of EPO/EPO-R (Sirén et al. 2001b).

From bench to bedside: Neuroprotective effects of erythropoietin

775

Not only ischemia leads to a distinct upregulation of the EPO system in the brain, but also degenerative and inﬂammatory conditions go hand in hand with a re-activation of the EPO system (Ehrenreich et al. 2004b; Eid et al. 2004). As compared to healthy adults, schizophrenic patients show a strong EPO-R expression in frontal cortex and hippocampus (Ehrenreich et al. 2004b). An upregulation of EPO-R can also be observed in areas of inﬂammation in the brain, as for instance in multiple sclerosis (unpublished observations).

EPO in the nervous system: Cellular pathways The hematopoietic effects of EPO have been reported in great detail in other chapters of this book. Interestingly, several mechanisms of action that had been identiﬁed for cells of the hematopoietic system, have in recent years been found to similarly underlie EPO actions on cells of the nervous system (Digicaylioglu and Lipton 2001; Sirén et al. 2001a). In any tissue, EPO exerts its effects by binding to speciﬁc cell surface receptors, EPO-R, belonging to the cytokine type 1 receptor family (Fisher 2003; Jelkmann 1992). Dimerization of EPO-R before/upon ligand binding (Livnah et al. 1999) enables autophosphorylation of the receptor-associated Janus tyrosine kinase 2 and leads to activation of several downstream signaling cascades. In contrast to the hematopoietic EPO-R, consisting in its activated form of a homodimer, the neuroprotective EPO-R has recently been hypothesized to be different (Yamaji et al. 1996; Chin et al. 2000) and to consist of an EPO-R monomer (Tsai et al. 2006) that dimerizes with another receptor subtype, e.g. the beta common receptor (betaCR), to form a heterodimer (Jubinsky et al. 1997; Brines et al. 2004). Whereas betaCR is not a component of the hematopoietic EPO-R, it seems to be required for the tissue protective effects of EPO (Brines et al. 2004; Brines and Cerami 2005). EPO signaling in neuronal cells appears to predominantly involve activation of the phosphatidylinositol3-kinase(PI3-K)-Akt/protein kinase B pathway, the Ras-mitogen-activated protein kinases, signal transducers and activators of transcription-5 (STAT5), and NF-kappa B-dependent transcription (Digicaylioglu and Lipton 2001; Sirén et al. 2001a; Mori et al. 2003; Kilic et al. 2005a; Kretz et al. 2005; Lee et al. 2004; Park et al. 2006; Um and Lodish 2006).

Neuroprotective properties of EPO in cell culture and animal models of neuropsychiatric disease Its potent antiapoptotic effects in cells of the nervous system as well as in endothelial cells (Anagnostou et al. 1990; Morishita et al. 1997; Lewczuk et al. 2000; Sirén et al. 2001a; Ghezzi and Brines 2004; Kumral et al. 2005a),

776

H. Ehrenreich and C. Bartels

together with its anti-oxidative power (Chattopadhyay et al. 2000; Kawakami et al. 2000; Kawakami et al. 2001; Kumral et al. 2005b; Liu et al. 2006) make EPO an attractive candidate for neuroprotection, targeting some of the major determinants of the ﬁnal common detrimental pathway (Marrero et al. 1998; Digicaylioglu et al. 2004; Ghezzi and Brines 2004; Yamamoto et al. 2004). In addition, EPO acts in a neurotrophic fashion, protects axons, promotes their outgrowth, and enhances synaptic sprouting and synaptogenesis (Konishi et al. 1993; Tabira et al. 1995; Koshimura et al. 1999; Bocker-Meffert et al. 2002; Juul 2004). Furthermore, EPO stimulates angiogenesis (Ribatti et al. 1999; Kertesz et al. 2004) and neurogenesis (Shingo et al. 2001; Yu et al. 2002; Wang et al. 2004a). This variety of properties, obviously derived from its prominent role during brain development, explains the tremendous neuroprotective potential that EPO has shown in a number of different animal studies: It has not only been demonstrated to be beneﬁcial in rodent models of brain or spinal cord hypoxia/ischemia (Sadamoto et al. 1998; Sakanaka et al. 1998; Bernaudin et al. 1999; Brines et al. 2000; Sirén et al. 2001a; Bernaudin et al. 2002; Celik et al. 2002; Kumral et al. 2003; Villa et al. 2003; Kumral et al. 2005a; Tsai et al. 2006), but also of subarachnoid hemorrhage (Grasso et al. 2002; Springborg et al. 2002), of brain or spinal cord trauma (Brines et al. 2000; Gorio et al. 2002), of multiple sclerosis (EAE, experimental autoimmune encephalitis) (Brines et al. 2000; Agnello et al. 2002; Li et al. 2004; Sattler et al. 2004; Diem et al. 2005), of Parkinson’s disease (Genc et al. 2001; Genc et al. 2002; Csete et al. 2004), of epilepsy (Brines et al. 2000), of retinal degeneration (BockerMeffert et al. 2002; Grimm et al. 2002; Junk et al. 2002; Rex et al. 2004; Kilic et al. 2005b), of various axotomy models (Campana and Myers 2001; Campana and Myers 2003; Weishaupt et al. 2004), of neuropathies including diabetes (Bianchi et al. 2004; Keswani et al. 2004; Orhan et al. 2004), to just name the most important ﬁndings. In most of these conditions, the phenomenon of symptomatic improvement is convincingly described, without identifying the responsible molecular/cellular mechanisms that would explain the beneﬁcial effect of EPO. A lot more work has to go into understanding of these particular mechanisms, since this understanding will allow an even more comprehensive therapeutic exploitation of EPO. For example, the bell-shaped dose response curve observed for some neuroprotective EPO actions in vitro (Weishaupt et al. 2004) and in vivo (Sakanaka et al. 1998) is of highest practical relevance but still unexplained. In fact, in neuroprotective indications, there are no “predictors of the right dose” in humans yet.

EPO and the blood-brain-barrier: Transfer and inﬂuence on function One of the most prominent issues that arose both in animal studies and in human trials was the question of whether EPO can cross the blood-brain-

From bench to bedside: Neuroprotective effects of erythropoietin

777

barrier to directly exert its neuroprotective effects in the central nervous system. It seemed unlikely that a molecule over 30,000 Dalton in size with a high degree of glyosylation would be able to readily overcome the bloodbrain-barrier (Ballabh et al. 2004; Juul et al. 2004). Although the mechanisms of how EPO crosses the blood-brain-barrier are still not entirely clear, there is no doubt anymore that EPO does cross this barrier in amounts sufﬁcient to exert its neuroprotective effects (Brines et al. 2000; Banks et al. 2004; Ehrenreich et al. 2004b). This has been shown (1) in vitro, using explant cultures (Martinez-Estrada et al. 2003), (2) in rat studies, where peripherally applied EPO reaches a peak in the cerebrospinal ﬂuid (CSF) after 3.5 hours (Ehrenreich et al. 2004b), and (3) also in humans, upon application of indium111-labeled EPO (Ehrenreich et al. 2004b). In in vitro studies, EPO itself inﬂuenced blood-brain-barrier function and prevented its breakdown (Martinez-Estrada et al. 2003). In fact, penetration of EPO across the bloodbrain-barrier is saturable, pointing to a receptor-mediated transfer mechanism via the cerebral endothelium. Once EPO has reached the brain, the amount that will stay there appears to depend on the amount of EPO-R expressed. In situations of dramatic EPO-R upregulation as in hypoxia or ischemia (Sirén et al. 2001b), virtually all molecules that reach the cerebral circulation and pass through the endothelium get “trapped” in the brain. Therefore, EPO in the CSF represents mainly “spillover” from brain tissue. Nevertheless, measuring an EPO level of 60–100 times over baseline in the CSF of patients suffering from a stroke after intravenous infusion of highdose EPO, has helped getting the ﬁrst proof-of-principle study on EPO in stroke approved (Ehrenreich et al. 2002). Even more importantly, in preparation of our proof-of-concept study in schizophrenia, we determined brain and CSF concentrations of EPO both in schizophrenic patients and healthy controls after intravenous bolus injection of indium-111-labeled EPO (Ehrenreich et al. 2004b). We did not only conﬁrm that there was a high uptake of labeled EPO in the brain of schizophrenic patients, in line with their increased EPO-R expression pattern, but also found uptake of labeled EPO in the brain of healthy controls, albeit to a lower degree. These ﬁndings are in line with others who reported transfer of EPO over the bloodbrain-barrier (Brines et al. 2000; Martinez-Estrada et al. 2003; Banks et al. 2004), although the amount measured was in the range of albumin uptake. Since EPO degradation in the brain seems to be slower as compared to the circulation, the concentrations maintained in the brain after peripheral application are obviously sufﬁcient to provide neuroprotection. Despite the as yet unclear role of the classical EPO-R in brain, its increased expression, as observed in schizophrenia, coincides with increased indium-111 EPO uptake in the brain. This would point to the classical EPOR being present at least as an immunoreactive monomer in a heterodimeric or even multimeric brain EPO-R (Ehrenreich et al. 2004b; Tsai et al. 2006). As mentioned above, there is increasing evidence that EPO does not only

778

H. Ehrenreich and C. Bartels

cross the blood-brain-barrier but also inﬂuences its function. It appears that its binding to brain endothelial EPO-R is involved in the transfer process of EPO over the blood-brain-barrier, and inﬂuences endothelial function itself. Speciﬁcally, it has been observed that EPO is capable of reducing migration of inﬂammatory cells over the blood-brain-barrier (Villa et al. 2003) and of counteracting the breakdown of the blood-brain-barrier in conditions like hypoxia in vitro (Martinez-Estrada et al. 2003) or seizures in vivo (Uzum et al. 2005). In line with these experimental results, we found Technetium-99m uptake in the lesion area reduced in stroke patients after EPO treatment (unpublished observations). A temporary tightening of the blood-brainbarrier may be an additional protective effect of EPO. To conclude, the concept of large molecules being unable to cross the intact blood-brainbarrier may have to be revised, at least for certain growth factors like EPO.

Evidence for neuroprotective effects of EPO in man: Experience from proof-of-principle studies Prototype of an acute brain disease: Stroke Based on our own promising preclinical data, we decided as early as 1997 to initiate a pilot study on EPO in human stroke. For proof-of-principle in a small number of subjects, we determined unique inclusion criteria to reach a homogeneous population of stroke patients. Patients were eligible if they suffered from an acute stroke in the middle cerebral artery (MCA) territory. In addition, we requested magnetic resonance imaging (MRI) for inclusion to conﬁrm diagnosis and obtain objective control of the time of symptom onset. These two very strict inclusion criteria, stroke in the MCA territory and MRI for conﬁrmation of diagnosis, most likely explain the success of this small trial. The trial consisted of a safety and a double-blind proof-of-concept part, with identical design (Ehrenreich et al. 2002). The design of the “Göttingen EPO Stroke Trial” is presented in Fig. 3. In the safety part of the trial, we aimed at estimating potential risks associated with application of EPO in this new indication. In addition to general safety measures, we speciﬁcally focused (1) on blood pressure, to detect potential hypertensive effects of EPO (Carlini et al. 1993a; Carlini et al. 1993b; Vaziri et al. 1995; Quaschning et al. 2003; Miyashita et al. 2004), (2) on hematocrit, to uncover additional rheological risk factors as caused by increased blood viscosity, and (3) on the potential of EPO to cross the blood-brain-barrier, to guarantee a sufﬁciently high amount of EPO in the brain under conditions of human stroke. The answers of the safety trial were all reassuring, with no serious adverse events, no signs of hypertension following treatment, a stable hematocrit, as well as a 60–100 times over baseline concentration of EPO in the CSF after intravenous infusion (Ehrenreich et al. 2002).

Intervention

D

ay

30

779

18 D

ay

7 D

ay

3

2

ay D

ay

ay D

• History • Physical examination • Routine laboratory • cCT scan • cMRI

D

Inclusion

1

From bench to bedside: Neuroprotective effects of erythropoietin

Erythropoietin (33 333 IU iv, once daily, 30 min)

Outcomes

Functional Outcome Barthel Index, Rankin Scale

Neurological Outcome NIH Stroke Scale Scandinavian Stroke Scale

Infarct Size Magnetic resonance imaging

Laboratory Routine parameters, S100B ICU or stroke unit

Hospital ward or rehabilitation facility

Fig. 3. Study design The double-blind, placebo-controlled, randomized “Göttingen EPO Stroke Trial”

The design of the double-blind, placebo-controlled, randomized proofof-concept trial was unchanged as compared to the safety part of the study. Again, EPO treatment proved to be safe and well tolerated in this stroke population. EPO treated patients displayed better remission of neurological symptoms during post-stroke recovery (NIH-Stroke Scale), a less prominent evolution of lesion size as determined by MRI volumetrical analysis, and a faster return to baseline of the circulating glial damage marker S100B (Ehrenreich et al. 2002; Herrmann and Ehrenreich 2003). Most importantly, the Barthel Index, estimating the degree of independence in daily living retained after stroke, showed better scores in EPO as compared to placebo patients. Interestingly, rather than inducing an increase in hematocrit, EPO prevented the decrease in hematocrit observed in the placebo group (Ehrenreich et al. 2002). Following these encouraging and stimulating results, the “German Multicenter EPO Stroke Trial”, involving the centers Hannover, Bremen, Göttingen, and recently also Celle, Braunschweig, and Erlangen, has been initiated in early 2003. First results of this trial will most likely be available in spring of 2007. Provided that these demonstrate again beneﬁcial effects of EPO application in stroke, this treatment will go into clinical practise. A safety interim analysis at the state of 60 patients conﬁrmed that EPO treatment was again well tolerated and safe. There were no EPO-related serious adverse events reported or observed. This ﬁnding is particularly important with respect to the fact that in the multicenter trial, rtPA (recombinant tissue plasminogen activator) treatment of stroke patients (Group TNIoNDaSr-PSS

780

H. Ehrenreich and C. Bartels

1995) is allowed wherever applicable. The trial can therefore be expected to yield information on the safety not only of EPO but also of a combination of EPO and rtPA in ischemic stroke. The design of the “German Multicenter EPO Stroke Trial” differs from the pilot trial in a few points: (1) There is a cut-off of NIHSS score of 5 for patients to be included in the trial, i.e. only moderate to severe strokes are investigated; (2) Barthel Index on day 90 serves as primary outcome measure; (3) MRI analysis has been reduced to two time-points (days 1 and 7); (4) There is no longer an upper age limit of eligibility of patients; (5) The EPO preparation has been switched from EPO-β to EPO-α and a dose of 40,000 IU (former trial 33,333 IU) on each of the three consecutive days after stroke. All other inclusion criteria remained identical. Although, at this time, there are no efﬁcacy data to be reported, an overall mortality rate of 12% (at the state of 200 patients) may at least indicate a high quality study (Note that a mortality of around 20% could be expected in a stroke population comparable to the one included in this trial (Heinsius et al. 1998; Higashida 2005)).

Prototype of a chronic brain disease: Schizophrenia Building on the beneﬁcial outcome of our pilot trial in human stroke as well as on our conceptual understanding of the EPO system as an endogenous neuroprotective system in the brain, we decided to provide the prerequisites for long-term application of EPO in a chronic degenerative brain disease, schizophrenia (Andreasen 2000; Benes 2000). Regarding safety, we could base our study on the well-known safety of chronic EPO application in renal anemia. Here, long-term treatment of millions of people with EPO has proven safe and well tolerated (e.g. Eschbach et al. 1987). We next had to demonstrate that EPO penetrates into the brain of patients suffering from schizophrenia. Using indium-111-labeled EPO, we showed that EPO enters the brain in appreciable amounts both in schizophrenic patients and healthy controls. The stronger signal detectable in the brain of schizophrenics is most likely due to the more pronounced expression of EPO-R as compared to healthy control brains (Ehrenreich et al. 2004b). In preparation of the EPO schizophrenia study, we further found that EPO was protective in haloperidol treated primary neuronal cultures, reducing the amount of cell death following exposure to this antipsychotic agent (Ehrenreich et al. 2004b). Finally, a number of cognitive tests both in mice and rats convinced us that EPO is potentially capable of improving functions that are known to be affected in schizophrenia (Ehrenreich et al. 2004b; Sirén et al. 2006). Most importantly, in our parietal lobe cryo-lesion model of juvenile mice, early EPO application prevented neurodegeneration/brain atrophy from progressing (Sirén et al. 2006). Since neurodegeneration starting in the

From bench to bedside: Neuroprotective effects of erythropoietin

781

parietal lobe and spreading all over the cortex is known to take place in schizophrenia (Lieberman et al. 2001; Thompson et al. 2001), these ﬁndings also supported our planned human trial. Finally, in April 2003, a double-blind, placebo-controlled, randomized multicenter trial was initiated in Germany, including chronic schizophrenic men with stable disease state and clear-cut cognitive deﬁcit. Cooperating centers were Göttingen, Kiel, Homburg, Köln, and Marburg. A total of 39 patients were treated over 12 weeks with weekly 40,000 IU of EPO versus placebo. The trial was concluded after a total duration of two years as planned, and study results will be published end of 2006. In brief, EPO treatment led to an improvement in schizophrenia-related cognitive dysfunction without inﬂuencing psychopathology. This selective improvement together with the fact that no other compound as yet has been clearly demonstrated to improve cognition in this disease may encourage further clinical research along these lines. The observation of surprisingly stable hematocrit values despite high weekly dosage of EPO in schizophrenic patients (as well as in multiple sclerosis patients – unpublished data) also awaits further investigation.

Evidence of cognitive effects of EPO: Facts and hypotheses Effects of EPO on cognitive performance have been observed as early as during the days when EPO was ﬁrst introduced into the clinic for treatment of renal anemia (Eschbach et al. 1987; Sundal et al. 1991; Di Paolo et al. 1992; Jelkmann 1992; Nissenson 1992; Kramer et al. 1996; Pickett et al. 1999). Over more than a decade, however, this effect has been attributed to the effect of EPO on hematopoiesis and, therefore, on tissue oxygenation (Nissenson 1992; Pickett et al. 1999; Elwood et al. 2001). Only end of the nineties, the ﬁrst papers came out that provided evidence for EPO inﬂuencing cognition via its action on the brain (Sadamoto et al. 1998; Sakanaka et al. 1998). These studies, however, have all been studies in rodents, and observations in humans were not made in prospective trials but, at best, derived from trial data in retrospect. Cognitive tests in rodents improving during or after EPO treatment in healthy animals or various animal models of neuropsychiatric disease are multifaceted, ranging from e.g. eight-arm maze, taste aversion testing, hole-board, elevated plus maze, to Morris water maze (Ehrenreich et al. 2004b; Sirén et al. 2006). In our own hands, apart from positive results of animal studies (Ehrenreich et al. 2004b; Sirén et al. 2006), EPO demonstrated to be effective in augmenting cognitive function in schizophrenia (paper to be published end of 2006) as well as in multiple sclerosis (unpublished observations). Figure 4 gives an overview of cognitive domains presumably responsive to EPO treatment in these chronic neurodegenerative human diseases. The molecular/cellular basis of EPO effects on cognitive performance remains to be elucidated.

782

H. Ehrenreich and C. Bartels

SCHIZOPHRENIC PATIENTS ng ki ry or mo W e m

rm te y rt or o m h S me

m er y gt or n m Lo e m

e iv ut ons c i xe t E unc f

l ia at s p n s o uo ti is c V fun

Ps y fu ch nc om ti o on to s r

COGNITIVE DOMAINS

n io nt e tt A

MULTIPLE SCLEROSIS PATIENTS cognitive improvement in schizophrenic patients,

cognitive improvement in MS patients,

cognitive improvement in both groups.

Fig. 4. Pattern of improvement in various cognitive domains in chronic schizophrenic (N = 20) and chronic progressive multiple sclerosis (N = 4) patients after 12 weeks of weekly EPO treatment (unpublished observations)

Future neuroprotective studies in man Encouraged by (1) the positive results of our proof-of-principle studies and (2) the safety of EPO also in the multicenter stroke trial, even allowing for co-application of rtPA, together with (3) our increasing experience on long-term high-dose treatment in neurodegenerative diseases like schizophrenia and multiple sclerosis, we are in the process of performing ﬁrst exploratory studies including dose-ﬁnding approaches on Parkinson’s disease, Alzheimer’s disease and major depression (Ehrenreich et al. 2004a). All over the world, studies on EPO in central nervous system indications are planned or even ongoing, with stroke being the “favourite indication”. In Denmark, a study on patients with subarachnoid hemorrhage has been started several years ago. In Switzerland, a trial exploring the neuroprotective effect of EPO in hypoxic neonates has started in summer of 2005. Studies on chemotherapy-associated neuropathy and on acute myelitis are planned at Johns Hopkins University, Baltimore. Various centers are interested in exploring the effect of EPO in oncology, focusing on the so-called “chemobrain”, i.e. damage to the brain caused by chemo- and/or radiotherapy. In Italy, a multicenter spinal cord trauma study has been announced. Despite the world-wide interest in neuroprotective trials using EPO, the studies planned or running are few and slowly advancing, since ﬁnancing is difﬁcult. Public funding is virtually impossible since investigators are referred to industry. Established manufacturers of erythropoiesis-stimulating proteins hesitate to support the trials due to the expired/expiring patents, the upcoming “biosimilars” production, and the tremendous black market being fed from various countries world-wide. In this regard, the introduction of EPO analogues/derivatives with slightly modiﬁed properties (Dong et al. 1992, Erbayraktar et al. 2003, Leist et al. 2004, Wang et al. 2004b) may help to ﬁnally ﬁnance and perform the necessary trials. A promising compound

From bench to bedside: Neuroprotective effects of erythropoietin

783

may be carbamylated EPO (CEPO) which preferentially binds to the neuroprotective/tissue protective EPO-R, therefore being devoid of EPO’s (in neuroprotective applications potentially disturbing) hematopoietic effects (Ehrenreich 2004, Leist et al. 2004).

Acknowledgements The clinical studies reported here have been funded by the Max-PlanckSociety (MPG), and by research grants from Orthobiotech and Lundbeck. The basic research has received grant support from MPG, DFG, Orthobiotech, Lundbeck, and Roche.

References 1. Agnello D, Bigini P, Villa P, Mennini T, Cerami A, Brines ML, Ghezzi P (2002) Erythropoietin exerts an anti-inﬂammatory effect on the CNS in a model of experimental autoimmune encephalomyelitis. Brain Res 952(1): 128–134 2. Anagnostou A, Lee ES, Kessimian N, Levinson R, Steiner M (1990) Erythropoietin has a mitogenic and positive chemotactic effect on endothelial cells. Proc Natl Acad Sci U S A 87(15): 5978–5982 3. Andreasen NC (2000) Schizophrenia: the fundamental questions. Brain Res Brain Res Rev 31(2–3): 106–112 4. Ballabh P, Braun A, Nedergaard M (2004) The blood-brain barrier: an overview: structure, regulation, and clinical implications. Neurobiol Dis 16(1): 1–13 5. Banks WA, Jumbe NL, Farrell CL, Niehoff ML, Heatherington AC (2004) Passage of erythropoietic agents across the blood-brain barrier: a comparison of human and murine erythropoietin and the analog darbepoetin alfa. Eur J Pharmacol 505(1–3): 93–101 6. Bauer C, Kurtz A (1989) Oxygen sensing in the kidney and its relation to erythropoietin production. Annu Rev Physiol 51: 845–856 7. Beleslin-Cokic BB, Cokic VP, Yu X, Weksler BB, Schechter AN, Noguchi CT (2004) Erythropoietin and hypoxia stimulate erythropoietin receptor and nitric oxide production by endothelial cells. Blood 104(7): 2073–2080 8. Benes FM (2000) Emerging principles of altered neural circuitry in schizophrenia. Brain Res Brain Res Rev 31(2–3): 251–269 9. Bernaudin M, Marti HH, Roussel S, Divoux D, Nouvelot A, MacKenzie ET, Petit E (1999) A potential role for erythropoietin in focal permanent cerebral ischemia in mice. J Cereb Blood Flow Metab 19(6): 643–651 10. Bernaudin M, Nedelec AS, Divoux D, MacKenzie ET, Petit E, Schumann-Bard P (2002) Normobaric hypoxia induces tolerance to focal permanent cerebral ischemia in association with an increased expression of hypoxia-inducible factor1 and its target genes, erythropoietin and VEGF, in the adult mouse brain. J Cereb Blood Flow Metab 22(4): 393–403 11. Bianchi R, Buyukakilli B, Brines M, Savino C, Cavaletti G, Oggioni N, Lauria G, Borgna M, Lombardi R, Cimen B, Comelekoglu U, Kanik A, Tataroglu C, Cerami A, Ghezzi P (2004) Erythropoietin both protects from and reverses experimental diabetic neuropathy. Proc Natl Acad Sci U S A 101(3): 823–828

784

H. Ehrenreich and C. Bartels

12. Bocker-Meffert S, Rosenstiel P, Rohl C, Warneke N, Held-Feindt J, Sievers J, Lucius R (2002) Erythropoietin and VEGF promote neural outgrowth from retinal explants in postnatal rats. Invest Ophthalmol Vis Sci 43(6): 2021– 2026 13. Brines M, Grasso G, Fiordaliso F, Sfacteria A, Ghezzi P, Fratelli M, Latini R, Xie QW, Smart J, Su-Rick CJ, Pobre E, Diaz D, Gomez D, Hand C, Coleman T, Cerami A (2004) Erythropoietin mediates tissue protection through an erythropoietin and common beta-subunit heteroreceptor. Proc Natl Acad Sci U S A 101(41): 14907–14912 14. Brines M, Cerami A (2005) Emerging biological roles for erythropoietin in the nervous system. Nat Rev Neurosci 6(6): 484–494 15. Brines ML, Ghezzi P, Keenan S, Agnello D, de Lanerolle NC, Cerami C, Itri LM, Cerami A (2000) Erythropoietin crosses the blood-brain barrier to protect against experimental brain injury. Proc Natl Acad Sci U S A 97(19): 10526–10531 16. Campana WM, Misasi R, O’Brien JS (1998) Identiﬁcation of a neurotrophic sequence in erythropoietin. Int J Mol Med 1(1): 235–241 17. Campana WM, Myers RR (2001) Erythropoietin and erythropoietin receptors in the peripheral nervous system: changes after nerve injury. Faseb J 15(10): 1804–1806 18. Campana WM, Myers RR (2003) Exogenous erythropoietin protects against dorsal root ganglion apoptosis and pain following peripheral nerve injury. Eur J Neurosci 18(6): 1497–1506 19. Carlini R, Obialo CI, Rothstein M (1993a) Intravenous erythropoietin (rHuEPO) administration increases plasma endothelin and blood pressure in hemodialysis patients. Am J Hypertens 6(2): 103–107 20. Carlini RG, Dusso AS, Obialo CI, Alvarez UM, Rothstein M (1993b) Recombinant human erythropoietin (rHuEPO) increases endothelin-1 release by endothelial cells. Kidney Int 43(5): 1010–1014 21. Celik M, Gokmen N, Erbayraktar S, Akhisaroglu M, Konakc S, Ulukus C, Genc S, Genc K, Sagiroglu E, Cerami A, Brines M (2002) Erythropoietin prevents motor neuron apoptosis and neurologic disability in experimental spinal cord ischemic injury. Proc Natl Acad Sci U S A 99(4): 2258–2263 22. Chattopadhyay A, Choudhury TD, Bandyopadhyay D, Datta AG (2000) Protective effect of erythropoietin on the oxidative damage of erythrocyte membrane by hydroxyl radical. Biochem Pharmacol 59(4): 419–425 23. Chikuma M, Masuda S, Kobayashi T, Nagao M, Sasaki R (2000) Tissue-speciﬁc regulation of erythropoietin production in the murine kidney, brain, and uterus. Am J Physiol Endocrinol Metab 279(6): E1242–1248 24. Chin K, Yu X, Beleslin-Cokic B, Liu C, Shen K, Mohrenweiser HW, Noguchi CT (2000) Production and processing of erythropoietin receptor transcripts in brain. Brain Res Mol Brain Res 81(1–2): 29–42 25. Csete M, Rodriguez L, Wilcox M, Chadalavada S (2004) Erythropoietin receptor is expressed on adult rat dopaminergic neurons and erythropoietin is neurotrophic in cultured dopaminergic neuroblasts. Neurosci Lett 359(1–2): 124–126 26. Dame C, Juul SE, Christensen RD (2001) The biology of erythropoietin in the central nervous system and its neurotrophic and neuroprotective potential. Biol Neonate 79(3–4): 228–235

From bench to bedside: Neuroprotective effects of erythropoietin

785

27. Dawson TM (2002) Preconditioning-mediated neuroprotection through erythropoietin? Lancet 359(9301): 96–97 28. Di Paolo B, Di Liberato L, Fiederling B, Catucci G, Bucciarelli S, Paolantonio L, Albertazzi A (1992) Effects of uremia and dialysis on brain electrophysiology after recombinant erythropoietin treatment. Asaio J 38(3): M477–480 29. Diem R, Sattler MB, Merkler D, Demmer I, Maier K, Stadelmann C, Ehrenreich H, Bahr M (2005) Combined therapy with methylprednisolone and erythropoietin in a model of multiple sclerosis. Brain 128(Pt 2): 375–385 30. Digicaylioglu M, Lipton SA (2001) Erythropoietin-mediated neuroprotection involves cross-talk between Jak2 and NF-kappaB signalling cascades. Nature 412(6847): 641–647 31. Digicaylioglu M, Garden G, Timberlake S, Fletcher L, Lipton SA (2004) Acute neuroprotective synergy of erythropoietin and insulin-like growth factor I. Proc Natl Acad Sci U S A 101(26): 9855–9860 32. Dirnagl U, Simon RP, Hallenbeck JM (2003) Ischemic tolerance and endogenous neuroprotection. Trends Neurosci 26(5): 248–254 33. Dong YJ, Kung C, Goldwasser E (1992) Receptor binding of asialoerythropoietin. J Cell Biochem 48(3): 269–276 34. Eckardt KU, Kurtz A (2005) Regulation of erythropoietin production. Eur J Clin Invest 35 [Suppl 3]: 13–19 35. Ehrenreich H, Sirén AL (2001) Neuroprotection – what does it mean? – What means do we have? Eur Arch Psychiatry Clin Neurosci 251(4): 149– 151 36. Ehrenreich H, Hasselblatt M, Dembowski C, Cepek L, Lewczuk P, Stiefel M, Rustenbeck HH, Breiter N, Jacob S, Knerlich F, Bohn M, Poser W, Ruther E, Kochen M, Gefeller O, Gleiter C, Wessel TC, De Ryck M, Itri L, Prange H, Cerami A, Brines M, Sirén AL (2002) Erythropoietin therapy for acute stroke is both safe and beneﬁcial. Mol Med 8(8): 495–505 37. Ehrenreich H (2004) Medicine. A boost for translational neuroscience. Science 305(5681): 184–185 38. Ehrenreich H, Aust C, Krampe H, Jahn H, Jacob S, Herrmann M, Sirén AL (2004a) Erythropoietin: novel approaches to neuroprotection in human brain disease. Metab Brain Dis 19(3–4): 195–206 39. Ehrenreich H, Degner D, Meller J, Brines M, Behe M, Hasselblatt M, Woldt H, Falkai P, Knerlich F, Jacob S, von Ahsen N, Maier W, Bruck W, Ruther E, Cerami A, Becker W, Sirén AL (2004b) Erythropoietin: a candidate compound for neuroprotection in schizophrenia. Mol Psychiatry 9(1): 42–54 40. Ehrenreich H, Hasselblatt M, Knerlich F, von Ahsen N, Jacob S, Sperling S, Woldt H, Vehmeyer K, Nave KA, Sirén AL (2005) A hematopoietic growth factor, thrombopoietin, has a proapoptotic role in the brain. Proc Natl Acad Sci U S A 102(3): 862–867 41. Eid T, Brines ML, Cerami A, Spencer DD, Kim JH, Schweitzer JS, Ottersen OP, de Lanerolle NC (2004) Increased expression of erythropoietin receptor on blood vessels in the human epileptogenic hippocampus with sclerosis. J Neuropathol Exp Neurol 63(1): 73–83 42. Elwood PC, Pickering J, Gallacher JE (2001) Cognitive function and blood rheology: results from the Caerphilly cohort of older men. Age Ageing 30(2): 135–139

786

H. Ehrenreich and C. Bartels

43. Erbayraktar S, Grasso G, Sfacteria A, Xie QW, Coleman T, Kreilgaard M, Torup L, Sager T, Erbayraktar Z, Gokmen N, Yilmaz O, Ghezzi P, Villa P, Fratelli M, Casagrande S, Leist M, Helboe L, Gerwein J, Christensen S, Geist MA, Pedersen LO, Cerami-Hand C, Wuerth JP, Cerami A, Brines M (2003) Asialoerythropoietin is a nonerythropoietic cytokine with broad neuroprotective activity in vivo. Proc Natl Acad Sci U S A 100(11): 6741–6746 44. Eschbach JW, Egrie JC, Downing MR, Browne JK, Adamson JW (1987) Correction of the anemia of end-stage renal disease with recombinant human erythropoietin. Results of a combined phase I and II clinical trial. N Engl J Med 316(2): 73–78 45. Ferri CP, Prince M, Brayne C, Brodaty H, Fratiglioni L, Ganguli M, Hall K, Hasegawa K, Hendrie H, Huang Y, Jorm A, Mathers C, Menezes PR, Rimmer E, Scazufca M (2005) Global prevalence of dementia: a Delphi consensus study. Lancet 366(9503): 2112–2117 46. Fisher JW (2003) Erythropoietin: physiology and pharmacology update. Exp Biol Med (Maywood) 228(1): 1–14 47. Gassmann M, Heinicke K, Soliz J, Ogunshola OO, Marti HH, Hofer T, Grimm C, Heinicke I, Egli B (2003) Non-erythroid functions of erythropoietin. Adv Exp Med Biol 543: 323–330 48. Genc S, Kuralay F, Genc K, Akhisaroglu M, Fadiloglu S, Yorukoglu K, Fadiloglu M, Gure A (2001) Erythropoietin exerts neuroprotection in 1-methyl-4-phenyl1,2,3,6-tetrahydropyridine-treated C57/BL mice via increasing nitric oxide production. Neurosci Lett 298(2): 139–141 49. Genc S, Akhisaroglu M, Kuralay F, Genc K (2002) Erythropoietin restores glutathione peroxidase activity in 1-methyl-4-phenyl-1,2,5,6-tetrahydropyridineinduced neurotoxicity in C57BL mice and stimulates murine astroglial glutathione peroxidase production in vitro. Neurosci Lett 321(1–2): 73–76 50. Ghezzi P, Brines M (2004) Erythropoietin as an antiapoptotic, tissue-protective cytokine. Cell Death Differ 11 [Suppl 1]: S37–44 51. Gorio A, Gokmen N, Erbayraktar S, Yilmaz O, Madaschi L, Cichetti C, Di Giulio AM, Vardar E, Cerami A, Brines M (2002) Recombinant human erythropoietin counteracts secondary injury and markedly enhances neurological recovery from experimental spinal cord trauma. Proc Natl Acad Sci U S A 99(14): 9450–9455 52. Grasso G, Buemi M, Alafaci C, Sfacteria A, Passalacqua M, Sturiale A, Calapai G, De Vico G, Piedimonte G, Salpietro FM, Tomasello F (2002) Beneﬁcial effects of systemic administration of recombinant human erythropoietin in rabbits subjected to subarachnoid hemorrhage. Proc Natl Acad Sci U S A 99(8): 5627–5631 53. Grimm C, Wenzel A, Groszer M, Mayser H, Seeliger M, Samardzija M, Bauer C, Gassmann M, Reme CE (2002) HIF-1-induced erythropoietin in the hypoxic retina protects against light-induced retinal degeneration. Nat Med 8(7):718– 724 54. Group TNIoNDaSr-PSS (1995) Tissue plasminogen activator for acute ischemic stroke. N Engl J Med 333(24): 1581–1587 55. Heinsius T, Bogousslavsky J, Van Melle G (1998) Large infarcts in the middle cerebral artery territory. Etiology and outcome patterns. Neurology 50(2): 341–350

From bench to bedside: Neuroprotective effects of erythropoietin

787

56. Herrmann M, Ehrenreich H (2003) Brain derived proteins as markers of acute stroke: their relation to pathophysiology, outcome prediction and neuroprotective drug monitoring. Restor Neurol Neurosci 21(3–4): 177–190 57. Higashida RT (2005) Recent advances in the interventional treatment of acute ischemic stroke. Cerebrovasc Dis 20 [Suppl 2]: 140–147 58. Jelkmann W (1992) Erythropoietin: structure, control of production, and function. Physiol Rev 72(2): 449–489 59. Jubinsky PT, Krijanovski OI, Nathan DG, Tavernier J, Sieff CA (1997) The beta chain of the interleukin-3 receptor functionally associates with the erythropoietin receptor. Blood 90(5): 1867–1873 60. Junk AK, Mammis A, Savitz SI, Singh M, Roth S, Malhotra S, Rosenbaum PS, Cerami A, Brines M, Rosenbaum DM (2002) Erythropoietin administration protects retinal neurons from acute ischemia-reperfusion injury. Proc Natl Acad Sci U S A 99(16): 10659–10664 61. Juul S (2004) Recombinant erythropoietin as a neuroprotective treatment: in vitro and in vivo models. Clin Perinatol 31(1): 129–142 62. Juul SE, Anderson DK, Li Y, Christensen RD (1998) Erythropoietin and erythropoietin receptor in the developing human central nervous system. Pediatr Res 43(1): 40–49 63. Juul SE, McPherson RJ, Farrell FX, Jolliffe L, Ness DJ, Gleason CA (2004) Erytropoietin concentrations in cerebrospinal ﬂuid of nonhuman primates and fetal sheep following high-dose recombinant erythropoietin. Biol Neonate 85(2): 138–144 64. Kawakami M, Iwasaki S, Sato K, Takahashi M (2000) Erythropoietin inhibits calcium-induced neurotransmitter release from clonal neuronal cells. Biochem Biophys Res Commun 279(1): 293–297 65. Kawakami M, Sekiguchi M, Sato K, Kozaki S, Takahashi M (2001) Erythropoietin receptor-mediated inhibition of exocytotic glutamate release confers neuroprotection during chemical ischemia. J Biol Chem 276(42): 39469–39475 66. Kertesz N, Wu J, Chen TH, Sucov HM, Wu H (2004) The role of erythropoietin in regulating angiogenesis. Dev Biol 276(1): 101–110 67. Keswani SC, Buldanlioglu U, Fischer A, Reed N, Polley M, Liang H, Zhou C, Jack C, Leitz GJ, Hoke A (2004) A novel endogenous erythropoietin mediated pathway prevents axonal degeneration. Ann Neurol 56(6): 815–826 68. Kilic E, Kilic U, Soliz J, Bassetti CL, Gassmann M, Hermann DM (2005a) Brainderived erythropoietin protects from focal cerebral ischemia by dual activation of ERK-1/-2 and Akt pathways. Faseb J 19(14): 2026–2028 69. Kilic U, Kilic E, Soliz J, Bassetti CI, Gassmann M, Hermann DM (2005b) Erythropoietin protects from axotomy-induced degeneration of retinal ganglion cells by activating ERK-1/-2. Faseb J 19(2): 249–251 70. Knabe W, Knerlich F, Washausen S, Kietzmann T, Sirén AL, Brunnett G, Kuhn HJ, Ehrenreich H (2004) Expression patterns of erythropoietin and its receptor in the developing midbrain. Anat Embryol (Berl) 207(6): 503–512 71. Knabe W, Sirén AL, Ehrenreich H, Kuhn HJ (2005) Expression patterns of erythropoietin and its receptor in the developing spinal cord and dorsal root ganglia. Anat Embryol (Berl) 210(3): 209–219 72. Konishi Y, Chui DH, Hirose H, Kunishita T, Tabira T (1993) Trophic effect of erythropoietin and other hematopoietic factors on central cholinergic neurons in vitro and in vivo. Brain Res 609(1–2): 29–35

788

H. Ehrenreich and C. Bartels

73. Koshimura K, Murakami Y, Sohmiya M, Tanaka J, Kato Y (1999) Effects of erythropoietin on neuronal activity. J Neurochem 72(6): 2565–2572 74. Kramer L, Madl C, Stockenhuber F, Yeganehfar W, Eisenhuber E, Derﬂer K, Lenz K, Schneider B, Grimm G (1996) Beneﬁcial effect of renal transplantation on cognitive brain function. Kidney Int 49(3): 833–838 75. Kretz A, Happold CJ, Marticke JK, Isenmann S (2005) Erythropoietin promotes regeneration of adult CNS neurons via Jak2/Stat3 and PI3K/AKT pathway activation. Mol Cell Neurosci 29(4): 569–579 76. Kumral A, Ozer E, Yilmaz O, Akhisaroglu M, Gokmen N, Duman N, Ulukus C, Genc S, Ozkan H (2003) Neuroprotective effect of erythropoietin on hypoxicischemic brain injury in neonatal rats. Biol Neonate 83(3): 224–228 77. Kumral A, Genc S, Ozer E, Yilmaz O, Gokmen N, Koroglu TF, Duman N, Genc K, Ozkan H (2005a) Erythropoietin Downregulates Bax and DP5 ProApoptotic Gene Expression in Neonatal Hypoxic-Ischemic Brain Injury. Biol Neonate 89(3): 205–210 78. Kumral A, Tugyan K, Gonenc S, Genc K, Genc S, Sonmez U, Yilmaz O, Duman N, Uysal N, Ozkan H (2005b) Protective effects of erythropoietin against ethanol-induced apoptotic neurodegenaration and oxidative stress in the developing C57BL/6 mouse brain. Brain Res Dev Brain Res 160(2): 146– 156 79. Lee SM, Nguyen TH, Park MH, Kim KS, Cho KJ, Moon DC, Kim HY, Yoon do Y, Hong JT (2004) EPO receptor-mediated ERK kinase and NF-kappaB activation in erythropoietin-promoted differentiation of astrocytes. Biochem Biophys Res Commun 320(4): 1087–1095 80. Leist M, Ghezzi P, Grasso G, Bianchi R, Villa P, Fratelli M, Savino C, Bianchi M, Nielsen J, Gerwien J, Kallunki P, Larsen AK, Helboe L, Christensen S, Pedersen LO, Nielsen M, Torup L, Sager T, Sfacteria A, Erbayraktar S, Erbayraktar Z, Gokmen N, Yilmaz O, Cerami-Hand C, Xie QW, Coleman T, Cerami A, Brines M (2004) Derivatives of erythropoietin that are tissue protective but not erythropoietic. Science 305(5681): 239–242 81. Lewczuk P, Hasselblatt M, Kamrowski-Kruck H, Heyer A, Unzicker C, Sirén AL, Ehrenreich H (2000) Survival of hippocampal neurons in culture upon hypoxia: effect of erythropoietin. Neuroreport 11(16): 3485–3488 82. Li W, Maeda Y, Yuan RR, Elkabes S, Cook S, Dowling P (2004) Beneﬁcial effect of erythropoietin on experimental allergic encephalomyelitis. Ann Neurol 56(6): 767–777 83. Lieberman J, Chakos M, Wu H, Alvir J, Hoffman E, Robinson D, Bilder R (2001) Longitudinal study of brain morphology in ﬁrst episode schizophrenia. Biol Psychiatry 49(6): 487–499 84. Liu J, Narasimhan P, Song YS, Nishi T, Yu F, Lee YS, Chan PH (2006) Epo protects SOD2-deﬁcient mouse astrocytes from damage by oxidative stress. Glia 53(4): 360–365 85. Livnah O, Stura EA, Middleton SA, Johnson DL, Jolliffe LK, Wilson IA (1999) Crystallographic evidence for preformed dimers of erythropoietin receptor before ligand activation. Science 283(5404): 987–990 86. Marrero MB, Venema RC, Ma H, Ling BN, Eaton DC (1998) Erythropoietin receptor-operated Ca2+ channels: activation by phospholipase C-gamma 1. Kidney Int 53(5): 1259–1268

From bench to bedside: Neuroprotective effects of erythropoietin

789

87. Marti HH, Wenger RH, Rivas LA, Straumann U, Digicaylioglu M, Henn V, Yonekawa Y, Bauer C, Gassmann M (1996) Erythropoietin gene expression in human, monkey and murine brain. Eur J Neurosci 8(4): 666–676 88. Martinez-Estrada OM, Rodriguez-Millan E, Gonzalez-De Vicente E, Reina M, Vilaro S, Fabre M (2003) Erythropoietin protects the in vitro blood-brain barrier against VEGF-induced permeability. Eur J Neurosci 18(9): 2538– 2544 89. Masuda S, Nagao M, Takahata K, Konishi Y, Gallyas F Jr, Tabira T, Sasaki R (1993) Functional erythropoietin receptor of the cells with neural characteristics. Comparison with receptor properties of erythroid cells. J Biol Chem 268(15): 11208–11216 90. Masuda S, Okano M, Yamagishi K, Nagao M, Ueda M, Sasaki R (1994) A novel site of erythropoietin production. Oxygen-dependent production in cultured rat astrocytes. J Biol Chem 269(30): 19488–19493 91. Masuda S, Chikuma M, Sasaki R (1997) Insulin-like growth factors and insulin stimulate erythropoietin production in primary cultured astrocytes. Brain Res 746(1–2): 63–70 92. Miyashita K, Tojo A, Kimura K, Goto A, Omata M, Nishiyama K, Fujita T (2004) Blood pressure response to erythropoietin injection in hemodialysis and predialysis patients. Hypertens Res 27(2): 79–84 93. Mori M, Uchida M, Watanabe T, Kirito K, Hatake K, Ozawa K, Komatsu N (2003) Activation of extracellular signal-regulated kinases ERK1 and ERK2 induces Bcl-xL up-regulation via inhibition of caspase activities in erythropoietin signaling. J Cell Physiol 195(2): 290–297 94. Morishita E, Masuda S, Nagao M, Yasuda Y, Sasaki R (1997) Erythropoietin receptor is expressed in rat hippocampal and cerebral cortical neurons, and erythropoietin prevents in vitro glutamate-induced neuronal death. Neuroscience 76(1): 105–116 95. Nagai A, Nakagawa E, Choi HB, Hatori K, Kobayashi S, Kim SU (2001) Erythropoietin and erythropoietin receptors in human CNS neurons, astrocytes, microglia, and oligodendrocytes grown in culture. J Neuropathol Exp Neurol 60(4): 386–392 96. Nissenson AR (1992) Epoetin and cognitive function. Am J Kidney Dis 20(1) [Suppl 1]: 21–24 97. Orhan B, Yalcin S, Nurlu G, Zeybek D, Muftuoglu S (2004) Erythropoietin against cisplatin-induced peripheral neurotoxicity in rats. Med Oncol 21(2): 197–203 98. Park MH, Lee SM, Lee JW, Son DJ, Moon DC, Yoon do Y, Hong JT (2006) ERK-mediated production of neurotrophic factors by astrocytes promotes neuronal stem cell differentiation by erythropoietin. Biochem Biophys Res Commun 339(4): 1021–1028 99. Pickett JL, Theberge DC, Brown WS, Schweitzer SU, Nissenson AR (1999) Normalizing hematocrit in dialysis patients improves brain function. Am J Kidney Dis 33(6): 1122–1130 100. Quaschning T, Ruschitzka F, Stallmach T, Shaw S, Morawietz H, Goettsch W, Hermann M, Slowinski T, Theuring F, Hocher B, Luscher TF, Gassmann M (2003) Erythropoietin-induced excessive erythrocytosis activates the tissue endothelin system in mice. Faseb J 17(2): 259–261

790

H. Ehrenreich and C. Bartels

101. Rex TS, Allocca M, Domenici L, Surace EM, Maguire AM, Lyubarsky A, Cellerino A, Bennett J, Auricchio A (2004) Systemic but not intraocular Epo gene transfer protects the retina from light-and genetic-induced degeneration. Mol Ther 10(5): 855–861 102. Ribatti D, Presta M, Vacca A, Ria R, Giuliani R, Dell’Era P, Nico B, Roncali L, Dammacco F (1999) Human erythropoietin induces a pro-angiogenic phenotype in cultured endothelial cells and stimulates neovascularization in vivo. Blood 93(8): 2627–2636 103. Ruscher K, Freyer D, Karsch M, Isaev N, Megow D, Sawitzki B, Priller J, Dirnagl U, Meisel A (2002) Erythropoietin is a paracrine mediator of ischemic tolerance in the brain: evidence from an in vitro model. J Neurosci 22(23): 10291–10301 104. Sadamoto Y, Igase K, Sakanaka M, Sato K, Otsuka H, Sakaki S, Masuda S, Sasaki R (1998) Erythropoietin prevents place navigation disability and cortical infarction in rats with permanent occlusion of the middle cerebral artery. Biochem Biophys Res Commun 253(1): 26–32 105. Sairanen T, Karjalainen-Lindsberg ML, Paetau A, Ijas P, Lindsberg PJ (2006) Apoptosis dominant in the periinfarct area of human ischaemic stroke – a possible target of antiapoptotic treatments. Brain 129(Pt 1): 189–199 106. Sakanaka M, Wen TC, Matsuda S, Masuda S, Morishita E, Nagao M, Sasaki R (1998) In vivo evidence that erythropoietin protects neurons from ischemic damage. Proc Natl Acad Sci U S A 95(8): 4635–4640 107. Sattler MB, Merkler D, Maier K, Stadelmann C, Ehrenreich H, Bahr M, Diem R (2004) Neuroprotective effects and intracellular signaling pathways of erythropoietin in a rat model of multiple sclerosis. Cell Death Differ 11 [Suppl 2]: S181–192 108. Semenza GL (2001) HIF-1 and mechanisms of hypoxia sensing. Curr Opin Cell Biol 13(2): 167–171 109. Sfagos C, Makis AC, Chaidos A, Hatzimichael EC, Dalamaga A, Kosma K, Bourantas KL (2005) Serum ferritin, transferrin and soluble transferrin receptor levels in multiple sclerosis patients. Mult Scler 11(3): 272–275 110. Sharp FR, Bernaudin M (2004) HIF1 and oxygen sensing in the brain. Nat Rev Neurosci 5(6): 437–448 111. Shingo T, Sorokan ST, Shimazaki T, Weiss S (2001) Erythropoietin regulates the in vitro and in vivo production of neuronal progenitors by mammalian forebrain neural stem cells. J Neurosci 21(24): 9733–9743 112. Sirén AL, Ehrenreich H (2001) Erythropoietin – a novel concept for neuroprotection. Eur Arch Psychiatry Clin Neurosci 251(4): 179–184 113. Sirén AL, Fratelli M, Brines M, Goemans C, Casagrande S, Lewczuk P, Keenan S, Gleiter C, Pasquali C, Capobianco A, Mennini T, Heumann R, Cerami A, Ehrenreich H, Ghezzi P (2001a) Erythropoietin prevents neuronal apoptosis after cerebral ischemia and metabolic stress. Proc Natl Acad Sci U S A 98(7): 4044–4049 114. Sirén AL, Knerlich F, Poser W, Gleiter CH, Bruck W, Ehrenreich H (2001b) Erythropoietin and erythropoietin receptor in human ischemic/hypoxic brain. Acta Neuropathol (Berl) 101(3): 271–276 115. Sirén AL, Radyushkin K, Boretius S, Kammer D, Riechers CC, Natt O, Sargin D, Watanabe T, Sperling S, Michaelis T, Price J, Meyer B, Frahm J, Ehrenreich

From bench to bedside: Neuroprotective effects of erythropoietin

116.

117.

118.

119.

120.

121.

122.

123.

124.

125.

126.

127.

128.

129.

791

H (2006) Global brain atrophy after unilateral parietal lesion and its prevention by erythropoietin. Brain 129(Pt 2): 480–489 Springborg JB, Ma X, Rochat P, Knudsen GM, Amtorp O, Paulson OB, Juhler M, Olsen NV (2002) A single subcutaneous bolus of erythropoietin normalizes cerebral blood ﬂow autoregulation after subarachnoid haemorrhage in rats. Br J Pharmacol 135(3): 823–829 Sundal E, Businger J, Kappeler A (1991) Treatment of transfusion-dependent anaemia of chronic renal failure with recombinant human erythropoietin. A European multicentre study in 142 patients to deﬁne dose regimen and safety proﬁle. Nephrol Dial Transplant 6(12): 955–965 Tabira T, Konishi Y, Gallyas F Jr (1995) Neurotrophic effect of hematopoietic cytokines on cholinergic and other neurons in vitro. Int J Dev Neurosci 13(3–4): 241–252 Thompson PM, Vidal C, Giedd JN, Gochman P, Blumenthal J, Nicolson R, Toga AW, Rapoport JL (2001) Mapping adolescent brain change reveals dynamic wave of accelerated gray matter loss in very early-onset schizophrenia. Proc Natl Acad Sci U S A 98(20): 11650–11655 Tsai PT, Ohab JJ, Kertesz N, Groszer M, Matter C, Gao J, Liu X, Wu H, Carmichael ST (2006) A critical role of erythropoietin receptor in neurogenesis and post-stroke recovery. J Neurosci 26(4): 1269–1274 Um M, Lodish HF (2006) Antiapoptotic effects of erythropoietin in differentiated neuroblastoma SH-SY5Y cells require activation of both the STAT5 and AKT signaling pathways. J Biol Chem 281(9): 5648–5656 Uzum G, Sarper Diler A, Bahcekapili N, Ziya Ziylan Y (2005) Erythropoietin prevents the increase in blood-brain barrier permeability during pentylentetrazol induced seizures. Life Sci (in press) Vaziri ND, Zhou XJ, Smith J, Oveisi F, Baldwin K, Purdy RE (1995) In vivo and in vitro pressor effects of erythropoietin in rats. Am J Physiol 269(6 Pt 2): F838–845 Villa P, Bigini P, Mennini T, Agnello D, Laragione T, Cagnotto A, Viviani B, Marinovich M, Cerami A, Coleman TR, Brines M, Ghezzi P (2003) Erythropoietin selectively attenuates cytokine production and inﬂammation in cerebral ischemia by targeting neuronal apoptosis. J Exp Med 198(6): 971–975 Wang L, Zhang Z, Wang Y, Zhang R, Chopp M (2004a) Treatment of stroke with erythropoietin enhances neurogenesis and angiogenesis and improves neurological function in rats. Stroke 35(7): 1732–1737 Wang X, Zhu C, Wang X, Gerwien JG, Schrattenholz A, Sandberg M, Leist M, Blomgren K (2004b) The nonerythropoietic asialoerythropoietin protects against neonatal hypoxia-ischemia as potently as erythropoietin. J Neurochem 91(4): 900–910 Weishaupt JH, Rohde G, Polking E, Sirén AL, Ehrenreich H, Bahr M (2004) Effect of erythropoietin axotomy-induced apoptosis in rat retinal ganglion cells. Invest Ophthalmol Vis Sci 45(5): 1514–1522 Yamaji R, Okada T, Moriya M, Naito M, Tsuruo T, Miyatake K, Nakano Y (1996) Brain capillary endothelial cells express two forms of erythropoietin receptor mRNA. Eur J Biochem 239(2): 494–500 Yamamoto M, Koshimura K, Sohmiya M, Murakami Y, Kato Y (2004) Effect of erythropoietin on nitric oxide production in the rat hippocampus using in vivo brain microdialysis. Neuroscience 128(1): 163–168

792

H. Ehrenreich and C. Bartels: Neuroprotective effects of erythropoietin

130. Yu X, Shacka JJ, Eells JB, Suarez-Quian C, Przygodzki RM, Beleslin-Cokic B, Lin CS, Nikodem VM, Hempstead B, Flanders KC, Costantini F, Noguchi CT (2002) Erythropoietin receptor signalling is required for normal brain development. Development 129(2): 505–516 Correspondence: Prof. Hannelore Ehrenreich, MD, DVM, Head, Division of Clinical Neuroscience, Max-Planck-Institute of Experimental Medicine, Hermann-Rein Straße 3, 37075 Göttingen, Germany, E-mail: [email protected]

Chapter 33

rhEPO in patients with anemia and congestive heart failure D. S. Silverberg1, D. Wexler2, A. Iaina1, S. Steinbruch1, Y. Wollman1, and D. Schwartz1 1

Department of Nephrology and 2Department of Cardiology and Heart Failure Clinic, From the Department of Nephrology and Cardiology, Tel Aviv Medical Center, Tel Aviv, Israel

Summary Many patients with congestive heart failure (CHF) fail to respond to maximal CHF therapy and progress to end stage CHF with many hospitalizations, very poor quality of life, end stage renal failure, or die of cardiovascular complications within a short time. One factor that has generally been ignored in many of these people is the fact that they are often anemic. The anemia is due mainly to renal failure but also to the inhibitory effects of cytokines on the bone marrow. Anemia itself may further worsen the cardiac function and make the patients resistant to standard CHF therapies. Indeed anemia has been associated with increased severity of CHF, increased hospitalization, worse cardiac function and functional class, higher doses of diuretics, worsening of renal function and reduced quality of life. In both controlled and uncontrolled studies the correction of the anemia with recombinant human erythropoietin (rhEPO) and oral or intravenous (IV) iron is associated with improvement in all these parameters. EPO itself may also play a direct role in improving the heart unrelated to the improvement of the anemia. Anemia may also play a role in the worsening of coronary heart disease even without CHF.

Introduction CHF is a major health burden in the western world [6,13,49,56,59]. Some recent data suggest that almost 1 in 3 people who have reached the age of 55 will develop CHF during their remaining life span [6]. The burden of CHF is particularly high in the elderly – about 80% of new diagnoses are in patients age 65 and older – the mean age of CHF patients being about 74 [6,13,49,56,59]. Despite many advances in the treatment of CHF, mortality is still very high, reaching 30–40% at one year in some studies [6,13,49,56,59].

794

D. S. Silverberg et al.

Although most deaths are due to cardiac or cardiovascular causes, a substantial number of patients progress to ESRD, which may be partially related to the progressive CHF [54]. In addition, many remain severely symptomatic despite optimal CHF therapy, including treatment with angiotensin converting enzyme inhibitors (ACEIs), beta blockers, angiotensin receptor blockers (ARBs), and aldospirone [6,13,49,56,59]. Anemia is frequently seen in CHF patients [75–83,96]. Is it possible that this anemia is an important contributor to the failure to respond to standard CHF treatment? Anemia (as deﬁned by us as a hemoglobin (Hb) of less than 12 g/dl) has been found in our own studies in about half the cases of CHF seen in our CHF outpatient clinic [75] and in about half the CHF cases requiring hospitalization [96]. Anemia in CHF has been associated with increased mortality, increased hospitalization, and a greater severity of CHF compared to nonanemic CHF patients [75,78,80–82,96]. That anemia may actually be contributing to the severity of the CHF is suggested by several controlled and uncontrolled studies which we and others have carried out in which correction of the anemia in severely resistant CHF patients by subcutaneous (sc) rhEPO used together with IV iron [12,63,75–77,79,83] or oral iron [18,53] has been associated with an improvement in functional capacity and cardiac function [12,53,75–77,79,83], a reduction in hospitalizations [12,18,63,75–77,79,83], a stabilization or improvement in renal function [18,53,63,75–77,79,83], a reduction in the dose of diuretics needed [12,63,75–77,79,83] and an improvement in the quality of life [12].

How common are anemia and chronic kidney insufﬁciency in CHF? Many studies in addition to our own have appeared recently which evaluated the relationship between anemia and CHF [75,78,80–82,96]. Analysis of these studies suggests that about 40% of patients with CHF have Hb levels below 12 g/dl. In the many studies that have examined the relationship between CHF and the level of mean Hb or hematocrit (Hct) and/or the prevalence of anemia in CHF patients, there is a very wide variation in prevalence of anemia ranging from 2.72% to 61% [75,78,80–82,96]! Why the enormous variation? Examination of these studies shows that the anemia was generally more common in the elderly, in diabetics and in those with more severe renal damage or more severe CHF. It was also more common in those who were hospitalized than in those treated in the community, and more common in those in whom the anemia was deﬁned as a Hb level of <12–13.5 g/dl as compared to <11 g/dl. In many of the larger controlled intervention studies of ACEIs, ARBs, and beta blockers in CHF, patients with chronic kidney insufﬁciency (CKI) or severe anemia were speciﬁcally excluded, which could partially explain the low prevalence of anemia in these studies. In some studies anemia was not present when patients were ﬁrst seen but developed over the

rhEPO in patients with anemia and congestive heart failure

795

period of follow-up. Some of the CHF studies deﬁned anemia as merely a physician’s recorded diagnosis of anemia in the medical discharge chart without actual values of Hb being given. The problem with this is that some doctors may recognize the presence of anemia only if it is very severe. Although CKI is often seen in CHF there is also a wide variation in the mean serum creatinine or creatinine clearance in the studies, but it would appear that about half the patients with CHF when ﬁrst seen have a serum creatinine of ≥1.5 mg/dl or a creatinine clearance of ≤60 ml/min/m2 both of these values ﬁtting the criteria for moderate to severe CKI [75,78,80–82,96].

How does anemia cause CHF? It has been known for years that anemia, if severe enough, can cause heart failure even in normal individuals [2]. Indeed, one recent study of over 1 million elderly US Medicare patients showed that anemia was an independent predictor of the development of CHF over a one year period [71]. The tissue hypoxia and peripheral vasodilation present in anemia causes a lowering of blood pressure, leading to an increased sympathetic response, which leads to tachycardia, increased stroke volume, renal vasoconstriction, reduced renal blood ﬂow, and salt and water retention (Fig. 1). This will lead to an increase in extra cellular ﬂuid (ECF) including an increase in plasma volume [2]. The reduced renal blood ﬂow will also cause an increased

Anemia CHF

Tissue hypoxia

LVH and eventual cell death

Peripheral vasodilation

≠ Ventricular

Ø Blood pressure

diameter and ≠BNP

≠ Sympathetic activity

≠ Plasma volume

≠

Heart rate and stroke volume

Ø Renal blood flow

Fluid retention ≠ Renin angiotensin aldosterone ADH

Fig. 1. The mechanism for ﬂuid retention and heart failure in anemia [75,78,80–82,96]

796

D. S. Silverberg et al.

secretion of renin, angiotensin, aldosterone and antidiuretic hormone (a), further augmenting the renal vasoconstriction and salt and water retention, and further increasing the ECF volume [2]. In addition, norepinephrine, renin, angiotensin, and aldosterone are all toxic to renal, cardiac, endothelial and other cells [8,38–39]. The tachycardia and increased stroke volume can eventually lead to ventricular dilation and hypertrophy [1,5], myocardial cell death, cardiac ﬁbrosis, and CHF [8,38–39]. All these ﬁndings are consistent with animal studies on the effect of anemia. Anemia in rats has been demonstrated to result in eccentric cardiac hypertrophy associated with increased capillary proliferation, abnormal diastolic wall stress, interstitial ﬁbrosis, increased left ventricular and diastolic pressure, increased left ventricular mass and decreased cardiac functional reserve [65–66,68,70]. Another mechanism of edema in anemia may be increased loss of ﬂuid from capillaries. Edema in fetal sheep is associated with an increase in capillary hydrostatic pressure. This is brought about by the combination of reduced arteriolar resistance without any associated reduction in postcapillary venular resistance. Thus anemia can cause extravascular ﬂuid accumulation because of “leaky capillaries” even without obvious heart failure [15–16].

Abnormalities associated with anemia in CHF [75,78,80–82,96] Compared to CHF patients without anemia, the presence of anemia in CHF patients has been associated with many abnormalities [75,78,80–82,96], including higher mortality, more hospitalizations, longer hospitalizations, higher hospitalization costs, a worse New York Heart Association (NYHA) Functional Class, lower left ventricular systolic function (as judged by a lower left ventricular ejection fraction (LVEF), and worse ventricular diastolic function, lower exercise capacity, reduced oxygen consumption (MVO2) during peak exercise, lower quality of life, higher serum beta natriuretic peptide (BNP) levels, (which suggest more severe CHF), elevated plasma volume and total body water, lower red cell volume and more severe peripheral edema, lower blood pressure, higher heart rate, poorer peripheral perfusion, higher ventricular ﬁlling pressures, higher pulmonary capillary wedge pressures, and a greater resistance to therapy as judged by the need for higher doses of diuretics and for digoxin. Anemia in CHF is also often associated with lower renal function and more rapid deterioration of renal function, and with signs of malnutrition such as a low body mass index, low serum albumin, low serum total protein, and low serum cholesterol [75,78,80–82,96]. The anemia and malnutrition seen in CHF may be partly caused by increased inﬂammatory cytokines such as tumor necrosis factor alpha (TNFα) and interleukin 6 (IL6), all signs of inﬂammation [75,78,80–82,96]. It is ironic that something as simple and as simply measured as Hb may be as powerful a predictor of cardiovascular events in CHF as LVEF, cardiac

rhEPO in patients with anemia and congestive heart failure

797

catheterization variables such as left and right ventricular pressures, BNP, exercise capacity and MVO2 [75,78,80–82,96]!

Is anemia actually causing the worsening of the CHF? Results of correcting the anemia Does the anemia actually contribute to the worsening of the CHF or is it just an innocent bystander, merely a marker of more severe CHF? One way of ﬁnding out is to actually treat the anemia and see if this improves the CHF. In both uncontrolled [75–77,79] and controlled [83] studies we showed that when the anemia is corrected to a Hb of 12.0 to 13.5 g/dl by sc rhEPO and IV iron (Ferric sucrose – Venofer – Vifor Int), the CHF improved, as evidenced by improvement of the NYHA functional class, increased LVEF, reduced number and days of hospitalization, reduced doses of oral and IV furosemide required, and improved self assessed shortness of breath and fatigue. In the uncontrolled studies [75–77,79] we also found that the creatinine clearance (CCr), that had been falling at a rate of about 1 ml/min/month before the anemia was corrected, stabilized after correction of the anemia. All these patients had been under a cardiologists care before we intervened to treat the anemia and had been on maximally tolerated doses of all the recommended CHF medications but were still resistant to therapy and were highly symptomatic. In the controlled study, the group in which the anemia was treated had no change in mean serum creatinine, whereas in the untreated group the mean serum creatinine levels increased signiﬁcantly [83]. In addition, one quarter of the patients in the untreated group died – all due to severe progressive CHF, whereas none died in the group in which the anemia had been corrected. In a randomized placebo controlled study of 22 patients with anemia and very severe CHF [53], Mancini et al. evaluated the use of sc rhEPO and oral iron over a 3-month period. Exercise duration, the distance walked in 6 minutes, peak MVO2, and the quality of life all improved in the treated group (whose mean Hb increased from 11.0 to 14.3 g/dl) and either stayed the same or worsened in the placebo group. The degree of improvement in MVO2 was proportional to the degree of change in the Hb. This is important since MVO2 is an important prognostic indicator for CHF survival [53]. In another study by the same group the anemia was found to be associated with a reduced red cell mass in the majority of cases and with an increased plasma volume (PV) in the rest [3]. Correction of the anemia reduced the PV to normal and increased the red cell mass [53]. In a preliminary US study [63], 84 CHF patients with anemia (Hb < 12 g/dl) and CKI (serum creatinine ≥1.5 mg/dl) were treated with IV iron (ferric gluconate-Ferrlecit) and rhEPO over a period of up to 15 months. By the end of the treatment period, compared to a similar period of time

798

D. S. Silverberg et al.

before the treatment, 37% had a decrease in serum creatinine and 30% had a decreased oral diuretic dose. The number of admissions to hospital had decreased by 43% and the number of hospital days had decreased by 33%. In another preliminary US study [18], 81 patients with predominantly NYHA III/IV CHF and anemia (Hb < 11 g/dl) were treated with rhEPO and oral iron. Mean follow-up was 438 ± 336 days. The mean initial blood urea nitrogen (BUN) was 51 ± 31 mg/dl and the mean initial serum creatinine was 2.1 ± 1.8 mg/dl. The Hb increased from a mean of 9.9 ± 1.1 before to 11.7 ± 1.7 g/dl after treatment. The mean BUN fell to 38 ± 23 mg/dl. The number of hospital days compared to an equal period before treatment fell by 50%. In yet another preliminary study [12], 10 patients with severe CHF and a Hb of less than 12 g/dl were treated for a mean of 5.0 ± 2.7 months with rhEPO and IV iron. All were receiving maximal medication for CHF. The results were compared to 13 similar patients in whom the anemia was not treated. The Hb in the treated group increased from 10.2 ± to 13.7 ± 1.2 g/dl and remained unchanged at 10.6 ± 0.9 in the untreated group. Compared to the untreated group, correction of the anemia was associated with a marked improvement in NYHA (1.7 in the treated vs 3.2 in the untreated), 90.3% less episodes of severe CHF, 88.7% less hospitalizations, and a 61% reduction in the need for IV diuretics. The serum creatinine was 1.4 ± 0.2 mg/dl in the treated and 1.7 ± 0.7 mg/dl in the untreated group (not signiﬁcant). Another argument in favor of anemia being partly responsible for the worsening of CHF is the effect of anemia on BNP. BNP is now used for both the diagnosis of CHF and for assessing prognosis [72]. It reﬂects the volume, stretch and pressure in the ventricles. Studies in CHF have found that the BNP levels in cardiac patients both with [9,89–90,92] and without CHF [43,89] was related inversely to Hb levels – the lower the Hb the higher the BNP. In patients with anemia but without cardiac disease [2,42] atrial natriuretic peptide (ANP) was also elevated, and when the anemia was corrected the ANP returned to normal. In one study anemia was found to be a better predictor of long-tem survival in CHF than the BNP level [92]. All these data clearly suggest that anemia indeed does play a role in the worsening of CHF, but large randomized placebo-controlled studies are needed to conﬁrm this, and these are indeed currently in progress.

Can treatment of anemia prevent CHF in CKI patients? Anemia in CKI is associated with an increase in hospitalizations [32,41]. Collins and his associates have shown that the consistent use of rhEPO in CKI patients in the 2 years before dialysis is associated with reduced rates of hospitalization for CHF and other heart diseases and with less CHF and hospitalizations after onset of dialysis [11,85]. In a similar study, the PRESAM study carried out in Europe and elsewhere, the use of rhEPO in the predial-

rhEPO in patients with anemia and congestive heart failure

799

ysis period was associated with a lower incidence of CHF, angina pectoris and myocardial infarction (MI) [91]. The mortality rate has also been shown to be lower in dialysis period in patients received rhEPO in the predialysis phase [22,98], as is the hospitalization rate [11,85]. The clinical improvement in our own studies of severe CHF patients, most of whom also had CKI [75–83,96], suggests that the multifaceted approach we used, with close cooperation between cardiologists and nephrologists, and maximal therapy of the CHF and the anemia, seems to prevent the progression of both the cardiac and the renal complications of CHF.

What is the effect of anemia on severity of CHF and on mortality? Many studies have examined the relationship between anemia, CHF severity, and mortality [75,78,80–82,96]. Of 25 studies that looked at the relationship between severity of CHF and anemia, 20 (80%) showed that the presence of anemia was associated with a more severe degree of CHF as judged by NYHA functional class [75,78,80–82,96]. Of 46 studies that examined the relationship between mortality and anemia in CHF, 44 (95.7%) showed a positive relationship [75,78,80–82,96]. In many of these studies the relationships between anemia and CHF were still statistically signiﬁcant by multivariate analysis where renal function and age were also taken into consideration – that is anemia was found to be an independent risk factor for cardiac mortality. Indeed renal failure and anemia were found in some studies to have an additive effect on mortality [75,78,80–82,96].

The additive effects of CHF, CKI and anemia In a study of over 1 million US Medicare elderly patients [31] it was found that CHF, CKI and anemia are additive in increasing mortality and the risk of developing end stage renal disease (ESRD). It is perhaps not surprising that anemia and CHF are such a lethal combination, considering the fact that both cause hypotension which activates the sympathetic and the renin, angiotensin and aldosterone system (RAAS) causing tachycardia and increased cardiac work, reduced renal function, increased salt and water absorption, increased ECF and increased plasma volume [2–3]. In addition, all three conditions, CHF and CKI [8] as well as anemia [2,7,27,50], are associated with an increase in four toxic mechanisms: sympathetic activity, activity of the RAAS system, oxidative stress and inﬂammation. Even more worrisome is the fact that all four of these mechanisms not only attack and destroy body cells but that they also activate each other [8]! Thus, the sympathetic system activates the other three, as do all the other three mechanisms [8]. This suggests that the only way to improve CHF and CKI is by a

800

D. S. Silverberg et al.

broad – based attack on all four of the mechanisms – something that we do when we use ACEI/ARBs, beta blockers and aldospirone, as well as control the anemia.

The effect of anemia on hospitalization, hospital mortality and hospital expenses in CHF Nine studies that examined the relationship between anemia and the number and/or duration of hospitalizations [75,78,80–82,96], all showed a signiﬁcant relationship. Kosiborod et al. [45] found that a 1% drop in Hct was associated with a 2% greater risk of rehospitalization. Golden et al. [24] found that 53% of those CHF patients in a heart clinic required hospitalization if their Hct was <35% compared to 21.4% where the initial Hct was >42% [24]. Uber et al. [90] found that 50% of anemic CHF patients required either hospitalization or an emergency room visit compared to only 15% of nonanemics [90]. These ﬁndings are consistent with the striking reduction in hospitalization that we found with correction of the anemia in our intervention studies [75–77,79,83]. The degree of anemia in CHF may also play a role in in-hospital mortality and hospital costs. A recent analysis of 9.107 patients hospitalized with the primary diagnosis of CHF was performed in 21 US hospitals [62]. Hb had an independent effect on in-hospital mortality. Multivariate remodeling showed that a 1 g/dl increase in Hb was associated with a 10.2% reduction in mortality risk, a 5.1% reduction in length of stay and a 5.3% decrease in hospital charges. In another CHF study, Felker et al. [21] found that the inhospital mortality was 6.1% in those with a Hb of ≤11.3 g/dl, 2.4% in those with a Hb of 11.4–13.9 g/dl and 1.4% in those with a Hb of >13.9 g/dl [21]. In a trial of ACEI in CHF, the SOLVD trial, Hct levels were signiﬁcantly related to hospital expenses – the hospital expenses were 19.9% less for an Hct of ≥36% compared to an Hct of <33% [26].

Effects of anemia on renal function in CHF Does anemia cause progression of the renal failure in CHF? In a study of 1004 consecutive patients admitted with a primary diagnosis of CHF, anemia was more prevalent in those who experienced a worsening of renal function during admission [23]. In a case control study of some of these same CHF patients, anemia was an independent predictor of worsening of renal function [10]. As in our previous studies [75–77,83], in a recent study of 78 anemic CHF patients [79], we found that 91.0% of anemic CHF patients when ﬁrst seen by us in an outpatient setting already had moderate to severe renal failure as deﬁned by a creatinine clearance of <60 ml/min/1.73 m2. The mean

rhEPO in patients with anemia and congestive heart failure

801

serum creatinine and creatinine clearance initially were 2.2 ± 0.9 mg/dl and 32.5 ± 26.5 ml/min/1.73 m2 respectively, and these did not change signiﬁcantly over a mean of 20.7 ± 12.1 months during which the anemia was corrected. It is a striking observation that the renal function in our anemic CHF – CKI patients in all our studies [75–77,79,83], which had been deteriorating at a rate of about 1 ml/month despite maximal CHF medications when they were anemic, appeared to stabilize in most patients when the anemia was controlled and when the patient received optimal CHF medications in optimal doses. These ﬁndings suggest that aggressive and early medical therapy of CHF and correction of the associated anemia may prevent the worsening not only of the CHF but of the CKI as well. Our ﬁndings on renal function are consistent with the effect of anemia correction in CKI without CHF seen in other studies. Anemia in CKI is associated with a more rapid rate of deterioration of renal function than in non anemics [40,48] and in a recent controlled study of anemia in CKI it was found that correction of the anemia with rhEPO could greatly slow down the progression of the renal failure [25]. A lower Hct before or shortly after percutaneous coronary intervention (PCI) has also been found to be an independent and important risk factor for the development of contrast-induced nephropathy, increasing the risk by about 50–100% [61]. Thus, it is possible that correction of the anemia with rhEPO may prevent contrast-induced nephropathy. This clearly deserves a controlled study since contrast-induced nephropathy is a common cause of acute renal failure.

The effect of anemia on other cardiac and noncardiac conditions The sensitivity of the damaged heart to anemia has been shown in many animal studies – CHF develops at much milder degrees of anemia in those with damaged hearts than in those with normal hearts [29,95]. This is consistent with our ﬁndings [75,79,83] and those of others [75,78,80–82,96]. For example, as mentioned above [75,78,80–82,96], many recent studies of patients with CHF show that survival is much worse in the anemic CHF patient than in those with a normal Hb level. Even patients with heart disease but without CHF may be very sensitive to the damaging effects of anemia. In a recent study, patients who had asymptomatic left ventricular systolic dysfunction without CHF were 2 to 3 times more likely to develop symptomatic CHF, be hospitalized or die over a period of several years if they were also anemic, independent of other factors such as age or renal function [14]. Anemia has also been found to be associated with markedly reduced survival in patients one month [47,60,97] and one year [46,60] after a MI. In addition, two years after elective percutaneous coronary interventions (PCI) in male patients with proven coronary heart disease (CHD) [69], patients in the

802

D. S. Silverberg et al.

lowest Hb quintile showed, in Cox regression analysis, a markedly higher risk for death (adjusted hazard rate ratio of 4.09). Indeed, in those CHD patients with an initial Hb of ≤10.9 g/dl the mortality at 2 years was 55% compared to only 3% in those with an Hb of 14–14.9 g/dl. Remarkably, 48% of all deaths occurred in the lowest Hb quintile which included only 21% of all PCI patients. In 3 other studies of patients with suspected CHD who underwent coronary angiography, anemia was also found to be an independent risk factor for adverse outcomes including death and cardiovascular complications [4,28,55]. All this suggests that anemia may be a common and important contributor to death and morbidity in patients with CHD unrelated to CHF. One study suggests that this anemia may actually be causative of the mortality in CHD and not merely a casual association [97]. In that study, correction of anemia by blood transfusion during hospitalization for an acute MI was associated with a marked reduction in the mortality rate over a 1month period compared to those anemic patients not transfused. Clearly this ﬁeld needs more investigation.

The etiology of the anemia in CHF The main cause of the anemia is most likely renal damage produced by the poor cardiac function. The reduced cardiac output and renal vasoconstriction leads to prolonged renal ischemia. This causes renal damage and reduced production of EPO in the kidneys. However, studies in animals have shown that CHF itself may cause anemia [37]. The damaged heart may secrete cytokines such as TNFα [30,37] which can cause anemia in four ways [17,37,57–58]: by reducing EPO production in the kidneys, by interfering with EPO activity at the level of the bone marrow, by inhibiting the release of Fe from the reticuloendothelial system so that it cannot get to the bone marrow to be utilized in Hb production, and by reducing iron absorption from the gut. Indeed, it has recently been shown that the higher the TNFα in CHF the lower the Hb level [7]. The reduced iron absorption from the gut is probably due to the release of hepcidin from the liver. This peptide is released by IL6 and goes to the gut where it prevents the absorption of iron [17,58]. We recently found an inhibitor to erythropoiesis that was secreted by cultured rat cardiomyocytes after they were exposed to anoxia for 2 hours (Wollman Y and Fibach E, unpublished data). The supernatant ﬂuid was added to cultures of normal human erythroid progenitors and this ﬂuid inhibited these red cell progenitors from multiplying and maturing to red cells. This suggests that the damaged heart cells can indeed produce substances that inhibit erythropoiesis. There are many other possible causes of anemia in CHF. Many CHF patients take aspirin which may cause blood loss. CHF patients often have proteinuria, and EPO, iron and transferrin can all be lost in signiﬁcant amounts in the urine [94], also contributing to the anemia. EPO production

rhEPO in patients with anemia and congestive heart failure

803

can be inhibited by ACEIs and ARBs and thus cause anemia [36,51,64]. Indeed, in a recent study, the SOLVD randomized controlled study of ACEI in CHF, the use of ACEIs increased the risk of developing anemia by 56%, and this was associated with a signiﬁcant increase in mortality [36]. Over one year the Hct fell by at least 4% in 14% of those taking enalapril but in only 9.5% of those on placebo. The anemia of ACEI is probably due mainly to 1) inhibition of angiotensin II production and to lowering of insulin growth factor levels, both of which can cause inhibition of erythroid precursors, and also to 2) an ACEI-induced increase in N acetyl-seryl-aspartyl-lysyl proline, a natural peptide that also decreases red cell precursors [36]. As previously mentioned, many patients with CHF have CKI, which itself is known to cause reduced iron absorption from the gut [44]. Diabetics are about twice as likely to develop anemia as nondiabetics [87–88]. This is probably due mainly to the fact that the elevated blood sugar damages the EPO-producing cells in the kidney, lowering the secretion of EPO. However many anemic diabetics also show signs of iron deﬁciency [87–88]. Finally, part of the anemia in CHF may be due to hemodilution, but recent studies showed that the majority of anemic CHF patients actually have a reduced red cell volume [3].

Iron deﬁciency in CKI In the Third National Health and Nutrition Examination Survey (NHANES III) (1988–1994 data), among those with CrCl of 20 to 30 ml/min, 46% of women and 19% of men had iron deﬁciency as deﬁned by transferring saturation (TSat) <20% and 47% and 44% as deﬁned by serum ferritin <100 ug/l, respectively [33]. The iron deﬁciency was independently associated with a lower Hb. Those men with the most severe iron deﬁciency anemia (as judged by having high erythrocyte protoporphyrin levels) had a mean Hb 1.8 g/dl lower than those with normal erythrocyte protoporphyrin levels, i.e. no iron deﬁciency. In a subsequent analysis of their data [34], the same investigators found that in the range of CrCl 30–50 ml/min less than one third of men with a Hb of <12 g/dl and women with an Hb of <11 g/dl had a serum ferritin ≥100 ug/l and a %Tsat ≥20%. In addition, the %TSat above 20% was independently associated with higher Hb levels. All this again suggests that iron deﬁciency may affect over half the patients with moderate or severe CKI. Several studies have shown that treatment of iron – deﬁcient CKI patients with IV iron even without EPO can cause a substantial increase in the Hb level [73–74].

Attitude of cardiologists and internists to anemia in CHF In a preliminary report from the Cleveland Clinic [86], 2011 consecutive ambulatory patients with chronic CHF patients seen in tertiary care

804

D. S. Silverberg et al.

cardiology or internal medicine clinics were studied. Anemia was deﬁned as a Hb ≤ 12 g/dl in men and ≤11 g/dl in women. 29% of these CHF patients had or developed anemia. Yet anemia was only recognized as a diagnosis in 11.1% of these cases by the internists and in only 4.4% of the cases by cardiologists. Diagnostic evaluation was only performed in 6% of these patients and only 10% received medical therapy for the anemia. The conclusion of the investigators was that anemia in ambulatory patients with CHF was underrecognized, under-diagnosed and under-treated by cardiologists and internists. In another study, anemia was found as a physician-recorded diagnosis in 17% of the records of CHF patients [19] but when actual Hb values were examined in the charts of such patients by the same group in another study the prevalence of anemia was actually 38% [20]. Clearly, anemia is often not recognized by physicians as a problem in CHF. These ﬁndings are consistent with recent American guidelines in detection and treatment of CHF, which do not mention even a single word about anemia [35,67]. Clearly cooperation must exist between cardiologists and nephrologists in treating these patients’ anemia and CHF early and vigorously. Our own program is based on such mutual cooperation.

Nonhematopoietic actions of erythropoietin EPO modulates a broad array of cellular processes outside of the hematopoietic system including Endothelian Progenitor Cell development, cellular integrity and angiogenesis. EPO also inhibits the apoptotic mechanisms of injury and inﬂammatory damage. It may offer protection against several kinds of injury in the heart, the central and peripheral nervous system, the eyes, the kidney and the blood vessels [52,84,93]. It protects and repairs endothelial cells throughout the body [52,84,93].

The antioxidant effects of erythropoietin The improvement in the cardiac function, renal and patient function that we see when the anemia is corrected with EPO may not be related only to the increased oxygen carrying capacity of the blood and the delivery of more oxygen to the tissues. The red blood cells (RBC) have many antioxidants which can neutralize the oxidative stress produced by radical oxygen species (ROS) that are produced in excessive amounts in CHF and CKI and can damage all the cells of the body by causing increased collagen synthesis and ﬁbrosis, causing lipid peroxidation, releasing inﬂammatory cytokines and causing apoptosis of endothelial cells and smooth muscle cells. The neutralization of ROS is produced mainly by the glutathione system in the RBC along with enzymes such as superoxide dismutase or catalase which react

rhEPO in patients with anemia and congestive heart failure

805

with the ROS and limit their effect on surrounding tissues [27,50]. Thus, simply increasing the number of erythrocytes can cause a major improvement in oxidative stress.

Method of correction of anemia in CHF and CKI When we see patients with anemia (Hb < 12 g/dl) and CHF and/or CKI we attempt to rule out secondary causes such as gastrointestinal bleeding and, whenever possible, taking their age and general condition into consideration, we usually send them for a gastrointestinal workup including gastroscopy and colonoscopy. We assess renal function, serum folic acid and B12 levels and thyroid and parathyroid function routinely and treat these abnormalities as well. If the serum ferritin is <500 mg/dl and the TSat <35%, we will start IV iron as Venofer – iron sucrose – two 5cc ampoules each containing 100 mg elemental iron, which we place into 150cc of normal saline. We then give a test dose of 25 mg over 45 minutes and, if this is tolerated (as in our experience it is in well over 99% of cases), we then give the rest over another 45 minutes. On subsequent visits we give the whole 150cc of normal saline with the 200 mg iron over 45 minutes. We give this treatment at weekly intervals until either the serum ferritin reaches 800 mg/dl, the TSat reaches 45% or the Hb reaches 13.5 g/dl. We start erythropoietin in a dose of 10,000 IU or equivalent doses of darbepoetin and maintain this dose at weekly intervals subcutaneously until the target of 13.5 g/dl has been reached and reduce the frequency to once every 2 weeks and once every 3 or 4 weeks depending on the Hb level. At times we can stop the EPO entirely. The Hb usually rises at a rate of about 1 g/dl per month. When the target Hb is reached we stop the IV iron and only give it again when the Hb falls below target and the iron parameters are in the safe range.

The vicious circle of CHF, CKI and anemia – the cardiorenal anemia (CRA) syndrome A vicious circle therefore appears to be present in CHF, where CHF itself causes both anemia and CKI. The CKI causes more anemia and the anemia and CKI act back to further worsen the CHF, which then further worsens the anemia and CKI and so on. In other words, each of the three can cause or be caused by the other. We have suggested calling this relationship the CRA syndrome [75–83,96]. This vicious circle is caused by overactivity of four systems, the sympathetic, RAAS, oxidative stress and inﬂammatory systems. These not only work together but further stimulate each other [71]. The importance of this concept is that if the anemia is not treated in CHF patients, there will

806

D. S. Silverberg et al.

Cardio Renal Anemia syndrome: a vicious circle of destruction CHF

CKI

Anemia

Fig. 2. The interaction between congestive heart failure (CHF), chronic kidney insufﬁciency (CKI) and anemia – The CRA syndrome

likely be resistance to any other form of CHF therapy and there will be progression of both the CHF and the CKI. Thus, correction of anemia may be crucial in the prevention of the progression of both CHF and CKI. It also follows that the failing heart needs maximal protection with all the CHF medications in the recommended doses.

The challenge of detecting and treating anemia and CHF in the community All the above suggests that if CHF is diagnosed early and treated with adequate doses of diuretics, ACE inhibitors and/or with angiotensin receptor blockers, with those beta blockers that have a proven effect in CHF (carvedilol, metoprolol, or bisoprolol), and, wherever possible, Aldospirone, the resultant control of CHF in the community will be far better than it is today. The evidence is mounting, however, that control of the associated anemia may also be another crucial element in the treatment of CHF. Indeed in our experience the correction of anemia is often associated with a profound improvement in cardiac function, patient function and quality of life in patients who were resistant to all of the usually recommended treatments. In addition, improvement in the Hb and cardiac function may also prevent deterioration of the renal function. This holds out the possibility that control of CHF is an important means of preventing CKI, as well.

Conclusion The use of aggressive therapy of CHF with maximally tolerated doses of CHF medications in combination with anemia correction by subcutaneous

rhEPO in patients with anemia and congestive heart failure

807

erythropoietin and IV iron in patients with the combination of anemia, CHF and CKI was associated with an improvement in CHF, a stabilization of renal function, a low mortality, a reduction in hospitalization and an improvement in quality of life. Close cooperation between nephrologists, cardiologists, diabetologists, internists and family physicians will help maximize the care of these complicated CHF-CKI patients. The beneﬁts to the patients in terms of quality of life, improved physical function, less hospitalizations and avoidance of dialysis and early death make it worth the effort.

References 1. Aessopos A, Deftereos S, Farmakis D, et al (2004) Cardiovascular adaptation to chronic anemia in the elderly. Echocardiographic study. Clin Invest Med 27: 265–273 2. Anand IS, Chandrashekhar Y, Ferrari R, Poole-Wilson PA, Harris PC (1993) Pathogenesis of edema in chronic anemia: Studies of body water and sodium, renal function, haemodynamics and plasma hormones. Br Heart J 70: 357–362 3. Androne AS, Katz SD, Lund L, et al (2003) Hemodilution is common in patients with advanced heart failure. Circulation 107: 226–229 4. Arant CB, Wessel TR, Olson MB, et al (2004) Hemoglobin level is an independent predictor for adverse cardiovascular outcomes in women undergoing evaluation for chest pain. J Am Coll Cardiol 43: 2009–2014 5. Astor BC, Arnett DK, Brown A, Coresh J (2004) Association of kidney function with left ventricular morphology among African Americans: The Atherosclerotic Risk in Communities (ARIC) Study. Am J Kidney Dis 43: 836–845 6. Bleumink GS, Knetsch AM, Stirkenboom MCJM, et al (2004) Quantifying the heart failure epidemic: prevalence, incidence rate, lifetime risk and prognosis of heart failure. Eur Heart J 25: 1614–1619 7. Bolger AP, Doehner W, Sharma R, Coate JS, Anker S (2002) Anemia in chronic heart failure: The relationship to inﬂammatory cytokine production and prognostic importance. Circulation 106 [Suppl]: 2819A (Abstr) 8. Bongartz LG, Cramer MJ, Doevendans PA, Joles JA, Braam B (2005) The severe cardiorenal syndrome: “Guyton revisited”. Eur Heart J 26: 11–17 9. Brucks S, Little WC, Chao T, et al (2004) Relation of anemia to diastolic heart failure and the effect on outcome. Am J Cardiol 93: 1055–1057 10. Butler J, Forman DE, Abraham WT, et al (2004) Relationship between heart failure treatment and development of worsening renal function among hospitalized patients. Am Heart J 147: 331–338 11. Collins AJ (2003) Anaemia management prior to dialysis: cardiovascular beneﬁts and cost-beneﬁt observations. Nephrol Dial Transplant 18 [Suppl 2]: ii2–ii6 12. Comin-Colet J, Bruguera J, Garcia C, et al (2005) Effect on hospitalization and functional status of correction of mild anemia with subcutaneous erythropoietin and intravenous iron in patients with severe advanced chronic heart failure and mild renal dysfunction. Heart Failure Meeting of the European Society of Cardiology. Lisbon 2005. (Abstr) 13. Cowie MR, Mosterd A, Wood DA, et al (1997) The Epidemiology of heart failure. Eur Heart J 18: 208–225

808

D. S. Silverberg et al.

14. Das SR, Dries DL, Drazner MH, Yancy CW, Chae CU (2004) Low hematocrit is associated with risk of progression from ALVD to symptomatic heart failure. Am J Cardiol 2005, 96: 827–831 15. Davis LE, Hohimer AR (1991) Hemodynamics and organ ﬂow in fetal sheep subjected to chronic anemia. Am J Physiol 261: R152–R158 16. Davis LE, Hohimer R, Brace RA (1996) Changes in left thoracic lymph ﬂow during progressive anemia in the ovine fetus. Am J Obstet Gynecol 174: 1469–1476 17. Deicher R, Horl WH (2004) Hepcidin: a molecular link between inﬂammation and anaemia. Nephrol Dial Transplant 19: 521–524 18. Delgado RM, Radovancevic R, van Rossem CC, Bracey A, Radovancevic R (2004) Effect of administration of recombinant human erythropoietin in heart failure patients. Eur Heart J 25 [Suppl 1]: 386A 19. Ezekowitz JA, McAlister FA, Armstrong PW (2003) Anemia is common in heart failure and is associated with poor outcomes: insights from a cohort of 12065 patients with new onset heart failure. Circulation 107: 223–225 20. Ezekowitz JA, McAlister FA, Armstrong PW (2005) The interaction among sex, hemoglobin and outcomes in a specialty heart failure clinic. Can J Cardiol 21: 165–171 21. Felker GM, Gattis WA, Leimberger JD, et al (2003) Usefulness of anemia as a predictor of death and rehospitalization in patients with decompensated heart failure. Am J Cardiol 92: 625–628 22. Fink JC, Blahut SA, Reddy M, Light PD (2001) Use of erythropoietin before the initiation of dialysis and its impact on mortality. Am J Kidney Dis 37: 348– 355 23. Forman DE, Butler J, Wang Y, et al (2004) Incidence, predictors of admission, and impact of worsening renal function among patients hospitalized with heart failure. J Am Coll Cardiol 43: 61–67 24. Golden JS, Fallick CC, Josephson SR, et al (2002) Baseline hematocrit predicts clinical outcomes in heart failure. Circulation 106 [Suppl]: 3358A (Abstr) 25. Gouva C, Nikolopoulos P, Ioannidis JPA, Siamopoulos KC (2004) Treating anemia early in renal failure slows the decline of renal function: A randomized controlled study. Kidney Int 66: 753–760 26. Gregory DD, Sarnak MJ, Konstam MA, Pereira B, Salem D (2003) Impact of chronic kidney disease and anemia on hospitalization expense in patients with left ventricular dysfunction. Am J Cardiol 92: 1300–1305 27. Grune T, Sommerburg O, Siems WG (2000) Oxidative stress in anemia. Clin Nephrol 53 [Suppl 1]: S18–S22 28. Gurm HS, Lincoff AM, Kleiman NS, et al (2004) Double jeopardy of renal insufﬁciency and anemia in patients undergoing percutaneous coronary interventions. Am J Cardiol 94: 30–34 29. Hebert PC (2004) Physiological aspects of anemia. Crit Care Clin 20: 187–212 30. Herrera-Garza EH, Stetson SJ, Cubillos-Garzon A, Vooltich MT, Farmer JA, Torre-Amione G (1999) Tumor necrosis factor: A mediator of disease progression in the failing human heart. Chest 115: 1170–1174 31. Herzog CA, Muster HA, Li S, Collins AJ (2004) Impact of congestive heart failure, chronic kidney disease and anemia on survival in the Medicare population. J Card Fail 10: 467–472

rhEPO in patients with anemia and congestive heart failure

809

32. Holland DC, Lam M (2000) Predictors of hospitalization and death among predialysis patients: a retrospective study. Nephrol Dial Transplant 15: 650–658 33. Hsu C-Y, McCullough CE, Curhan GC (2002) Epidemiology of anemia associated with chronic renal insufﬁciency among adults in the United States: Results from the Third National Health and Nutrition Examination Survey. J Am Soc Nephrol 13: 504–410 34. Hsu C-Y, McCullough CE, Curhan GC (2002) Iron status and hemoglobin level in chronic renal failure. J Am Soc Nephrol 13: 2783–2786 35. Hunt SA, Baker DW, Chin MC, et al (2001) ACC/AHA guidelines for the evaluation and management of chronic heart failure: Executive summary. J Am Coll Cardiol 38: 2101–2113 36. Ishani A, Weinhandl E, Zhao Z, et al (2005) Angiotensin-converting enzyme inhibitor as a risk factor for the development of anemia and the impact of incident anemia on mortality in patients with left ventricular dysfunction. J Am Coll Cardiol 45: 391–399 37. Iverson PO, Woldbaek PR, Tonnessen T, Christensen G (2002) Decreased hematopoiesis in bone marrow of mice with congestive heart failure. Am J Physiol Regul Integr Comp Physiol 282: R166–R172. 38. Johnson DB, Dell’Italia LJ (1996) Cardiac hypertrophy and failure in hypertension. Curr Opin Nephrol Hypertens 5: 186–191 39. Katz AM (1994) The cardiomyopathy of overload: An unnatural growth response in the hypertrophied heart. Ann Intern Med 121: 363–371 40. Keane WF, Brenner BM, de Zeeuw D, et al (2003) The risk of developing end-stage renal disease in patients with type 2 diabetes and nephropathy: The RENAAL study. Kidney Int 63: 149–150 41. Khan SS, Kazmi WH, Abichandani R, Tighiouart H, Pereira BJ, Kausz AT (2002) Health care utilization among patients with chronic kidney disease. Kidney Int 62: 229–236 42. Kikuchi M, Inagaki T (1999) Atrial natriuretic peptide in aged patients with iron deﬁciency anemia. Arch Gerontol Geriat 28: 105–115 43. Knudsen CW, Vik-Mo H, Omland T (2005) Blood hemoglobin is an independent predictor of B-type natriuretic peptide levels. Clin Sci 109: 69–74 44. Kooistra MP, Niemantverdriet EC, van Es A, et al (1998) Iron absorption in erythropoietin-treated haemodialysis patients: effects of iron availability, inﬂammation and aluminium. Nephrol Dial Transplant 13: 82–88 45. Kosiborod M, Smith GL, Radford MJ, Foody JM, Krumholz HM (2003) The prognostic importance of anemia in patients with heart failure. Am J Med 114: 112–119 46. Langston RD, Presley R, Flanders WD, McClellan WM (2003) Renal insufﬁciency and anemia are independent risk factors for death among patients with acute myocardial infarction. Kidney Int 64: 1398–1405 47. Lee PC, Kini AS, Ahsan C, Fisher E, Sharma SK (2004) Anemia is an independent predictor of mortality after percutaneous coronary intervention. J Am Coll Cardiol 44: 541–546 48. Levin A (2002) The relationship of hemoglobin level and survival: direct or indirect effects. Nephrol Dial Transplant 17 [Suppl 5]: 8–13 49. Levy D, Kenchaiah S, Larson MG, et al (2002) Long term trends in the incidence of and survival with heart failure. N Engl J Med 347: 1397–1402

810

D. S. Silverberg et al.

50. Ludat K, Sommerburg O, Grune T, Siems WG, Riedel E, Hampl H (2000) Oxidation parameters in complete correction of renal anemia. Clin Nephrology 53 [Suppl 1]: S30–S35 51. Macdougall IC (1999) The role of ACE inhibitors and angiotensin II receptor blockers in the response to erythropoietin. Nephrol Dial Transplant 14: 1836–1841 52. Maiese K, Li F, Chong ZZ (2005) New avenues of exploration for erythropoietin. JAMA 293: 90–95 53. Mancini DM, Katz SD, Lang C, et al (2003) Effect of erythropoietin on exercise capacity in patients with moderate to severe chronic heart failure. Circulation 107: 294–299 54. McClellan WM, Langston RD, Presley R (2004) Medicare patients with cardiovascular disease have a high prevalence of chronic kidney disease and a high rate of progression to end-stage renal disease. J Am Soc Nephrol 15: 1912–1919 55. McKechnie RS, Smith D, Montoye C, et al (2004) Prognostic implications of anemia on in-hospital outcomes after percutaneous coronary interventions. Circulation 110: 271–277 56. McMurray JJV, Stewart S (2002) The burden of heart failure. Eur Heart J 4 [Suppl D]: D50–D58 57. Means RT (1999) Advances in the anemia of chronic disease. Int J Hematol 70: 7–12 58. Means RT (2004) Hepcidin and anaemia. Blood Rev 18: 219–225 59. Miller LW, Missov ED (2001) Epidemiology of heart failure. Cardiol Clin 19: 547–555 60. Nikolsky E, Aymong ED, Halkin A, et al (2004) Impact of anemia in patients with acute myocardial infarction undergoing percutaneous coronary intervention. J Am Coll Cardiol 44: 547–553 61. Nikolsky E, Mehran R, Lasic Z, et al (2005) Low hematocrit predicts contrastinduced nephropathy after percutaneous coronary intervention. Kidney Internat 67: 706–713 62. Nordyke RJ, Kim JJ, Goldberg GA, et al (2004) Impact of anemia on hospitalization time, charges and mortality in patients with heart failure. Value in Health 7: 464–471 63. Novak B, Sumodi V, Bynane M, Fagnilli K, Hanna MA, Hixson E (2003) Anemia management in an outpatient heart failure center. J Card Fail 9 [Suppl 1]: Abstr 386 64. Odabas AR, Cetinkaya R, Selcuk Y, Keles S, Bilen H (2003) The effect of high dose Losartan on erythropoietin resistance in patients undergoing hemodialysis. Panminerva Med 45: 59–62 65. Olivetti G, Lagrasta C, Quaini F, et al (1989) Capillary growth in anaemia induced ventricular remodelling in the rat heart. Circ Res 65: 1182–1192 66. Olivetti G, Quaini F, Lagrasta C, et al (1992) Myocyte cellular hypertophy and hyperplasia contribute to ventricular wall modelling in anemia induced cardiac hypertrophy in rats. Am J Pathol 141: 227–239 67. Packer M, Cohn JN, et al (1999) Consensus recommendations for the management of chronic heart failure. Am J Cardiol 83: 1A–38A 68. Rakusan K, Cicutti N, Kolar F (2001) Effect of anemia on cardiac function, microvascular structure and capillary hematocrit in rat hearts. Am J Physiol Heart Circ Physiol 280: H1407–H1414

rhEPO in patients with anemia and congestive heart failure

811

69. Reinecke H, Trey T, Wellmann J, et al (2003) Hemoglobin-related mortality in patients undergoing percutaneous coronary interventions. Eur Heart J 24: 2142–2150 70. Rossi MA, Carillo SV (1983) Electron microscopic study on the cardiac hypertrophy induced by iron deﬁciency in the rat. Br J Exp Pathol 64: 373–387 71. Sandgren E, Murray AM, Herzog CA, et al (2005) Anemia and new-onset congestive heart failure in the general Medicare population. J Card Fail 11: 99–105 72. Shapiro BP, Chen HH, Burnett JC, Redﬁeld MM (2003) Use of plasma brain natriuretic peptide concentration to aid in the diagnosis of heart failure. Mayo Clin Proc 78: 481–486 73. Silverberg DS, Blum M, Agbaria Z, et al (1999) Intravenous iron for the treatment of predialysis anemia. Kidney Int 55 [Suppl 67]: S79–S85 74. Silverberg DS, Blum M, Agbaria Z, et al (2001) The effect of IV iron alone or in combination with low dose erythropoietin in the rapid correction of anemia of chronic renal failure in the predialysis period. Clin Nephrol 55: 212–219 75. Silverberg DS, Wexler D, Blum M, et al (2000) The use of subcutaneous erythropoietin and intravenous iron for the treatment of the anemia of severe, resistant congestive heart failure improves cardiac and renal function, functional cardiac class, and markedly reduces hospitalizations. J Am Coll Cardiol 35: 1737–1744 76. Silverberg DS, Wexler D, Blum M, et al (2001) Aggressive therapy of congestive heart failure and associated chronic renal failure with medications and correction of anemia stops or slows the progression of both diseases. Perit Dial Int 21 [Suppl 3]: S236–S240 77. Silverberg DS, Wexler D, Blum M, et al (2003) The effect of correction of anemia in diabetics and nondiabetics with severe resistant congestive heart failure and chronic renal failure by subcuteanous erythropoietin and intravenous iron. Nephrol Dial Transplant 18: 141–146 78. Silverberg DS, Wexler D, Blum M, et al (2004) The interaction between heart failure, renal failure and anemia – the cardio-renal anemia syndrome. Blood Purif 22: 277–284 79. Silverberg DS, Wexler D, Blum M, et al (2005) Effects of treatment with EPO beta on outcomes in patients with anaemia and chronic renal failure. Kidney Blood Press Res 28: 41–47 80. Silverberg DS, Wexler D, Blum M, Iaina A (2003) The cardio-renal anemia syndrome: correcting anemia inpatients with resistant congestive heart failure can improve both cardiac and renal function and reduces hospitalization. Clin Nephrol 60 [Suppl 1]: S93–S102 81. Silverberg DS, Wexler D, Blum M, Schwartz D, Wollman Y, Iaina A (2003) Erythropoietin should be part of congestive heart failure management. Kidney Int 64 [Suppl 87]: S40–S47 82. Silverberg DS, Wexler, Iaina A (2004) The role of anemia in the progression of congestive heart failure. Is there a place for erythropoietin and intravenous iron? J Nephrol 17: 749–761 83. Silverberg DS, Wexler, Sheps D, et al (2001) The effect of correction of mild anemia in severe, resistant congestive heart failure using subcutaneous erythropoietin and intravenous iron: A randomized controlled study. J Am Coll Cardiol 37: 1775–1780

812

D. S. Silverberg et al.: rhEPO in patients with anemia

84. Smith KJ, Bleyer AJ, Little WC, Sane DC (2003) The cardiovascular effects of erythropoietin. Cardiovasc Res 59: 538–548 85. St. Peter WL, Xue J, Ebben J, Collins A (2001) Pre-end stage renal disease erythropoietin use predicts hospitalization in the periods before and after end-stage renal disease diagnosis. J Am Soc Nephrol 12 [Suppl]: Abstr 1274 86. Tang WHW, Miller H, Partin M, et al (2003) Anemia in ambulatory patients with chronic heart failure: A single-center clinical experience derived from electronic medical records. J Am Coll Cardiol 41 [Suppl A]: 157A 87. Thomas MC, MacIsaac RJ, Tsalamandris C (2004) Anemia in patients with type 1 diabetes. J Clin Endocrinol Metab 89: 4359–4363 88. Thomas MC, MacIsaac RJ, Tslalmandris C, Power D, Jerums G (2003) Unrecognized anemia in patients with diabetes. Diabetes Care 26: 1164–1169 89. Tsuji H, Nishino N, Kimura Y, et al (2004) Haemoglobin level inﬂuences plasma brain natriuretic peptide concentration. Acta Cardiol 59: 527–531 90. Uber PA, Park MH, Scott RL, Mehra MR (2003) B-type natriuretic peptide levels are closely associated with anemia in chronic heart failure independent of underlying renal function. J Card Fail 9 [Suppl 1]: 133A, (Abstr) 91. Valderabano F, Horl W, Macdougall IC, Rossert J, Rutkowski B, Wauters J-P (2003) Pre-dialysis survey on anemia management. Nephrol Dial Transplant 18: 89–100 92. Van der Meer P, Voors AA, Lipsic E, Smilde TDJ, van Gilst WH, van Veldhuisen DJ (2004) Prognostive value of plasma erythropoietin on mortality in patients with chronic heart failure. J Am Coll Cardiol 44: 63–67 93. van der Meer P, Voors AA, Lipsic E, van Gilst WH, van Veldhuisen DJ (2004) Erythropoietin in cardiovascular diseases. Eur Heart J 13: 285–291 94. Vaziri ND (2001) Erythropoietin and transferrin metabolism in nephrotic syndrome. Am J Kidney Dis 38: 1–8 95. Wahr JA (1998) Myocardial ischemia in anaemic patients. Br J Anaes 81 [Suppl]: 10–15 96. Wexler D, Silverberg DS, Sheps D, et al (2004) Prevalence of anemia in patients admitted to hospital with a primary diagnosis of congestive heart failure. Int J Cardiol 96: 79–87 97. Wu W-C, Rathore SS, Wang Y, Radford MJ, Krumholtz H (2001) Blood transfusion in elderly patients with acute myocardial infarction. N Engl J Med 345: 1230–1236 98. Xue JL, St. Peter WL, Ebben JP, Everson SE, Collins AJ (2002) Anemia treatment in the pre-ESRD period and associated mortality in elderly patients. Am J Kid Dis 40: 1153–1161 Correspondence: Donald S. Silverberg, Department of Nephrology, The Department of Nephrology, Tel Aviv Medical Center, Weizmann 6, Tel Aviv 64239, Israel, E-mail: [email protected]

Chapter 34

Cost-effectiveness of treating cancer anaemia P. Cornes Institution – Bristol Haematology & Oncology Centre, Bristol, BS2 8ED, United Kingdom

Introduction It may seem strange that the efﬁcacy of erythropoietins in treating cancer related anaemia was decided beyond doubt in 1995 yet more than a decade later there is still debate about whether they are cost-effective. In 1995, cumulative meta-analysis reviewing randomised trials against date of trial completion showed an unequivocal positive effect (Clark 2002). Health economic study is an evolving area of medicine. In contrast to the familiar standard format of phase I, II and III clinical trials, there is no agreed methodology or unequivocal deﬁnition of a positive or negative result. This may seem frustrating to haematologists and oncologists, but the importance of reimbursement and economic appraisal in determining patients’ access to treatment makes this an important area of erythropoietin and anaemia research. A recent report showed large international variations in the uptake of cancer treatments (Wilking and Jönsson 2005). All 19 countries studied shared a central drugs regulatory authority, the European Medicines Agency (EMA). The EMA approves treatments on the basis of three levels of assessment; efﬁcacy, safety, and quality of manufacture. When the EMA approves a drug, it has to be made available in all member states within 180 days. The inequality of uptake of treatments suggests that another fourth level of regulation of access to treatment occurs, which is economic. Despite being approved for use by the EMA, and American FDA and other regulatory bodies, the reimbursement agreements and uptake of erthropoietins for treating cancer anaemia vary widely, from 5% of anaemic cancer cases treated in one European Union country to 66% of cases in another (Cavill 2002). This suggests that despite regulator approval, there is no agreement on the cost effectiveness of erthropoietins for treating cancer anaemia. This is in clear contrast to the use of erthropoietins to treat anaemia of renal disease, where there is near universal access in the developed world.

814

P. Cornes

In this article the different methods for estimating the cost-effectiveness of treatments are reviewed, and the results of published studies of treating cancer anaemia are discussed to discover why this disparity occurs.

Methods – literature search strategy Published studies for this article were identiﬁed through medical subject heading and textword searches at the US National Library of Medicine (www.nlm.nih.gov/mesh), and further searched through related item links. Review articles were retrieved and references cross checked. National and international guidelines were reviewed, along with formal health technology appraisals identiﬁed from Pubmed and internet searches and through guideline archive websites at the the UK National Library of Guidelines database (www.library.nhs.uk/guidelinesFinder) the Agency for Healthcare Research and Quality of the United States National Guideline Clearinghouse searchable database of clinical practice guidelines (www.guideline.gov/), the Danish Centre for Evaluation and Health Technology Assessment (DACEHTA) (www.dacehta.dk/) the UK National Institute for Health and Clinical Excellence (NICE) (www.nice.org.uk/) and the TRIP database of guidelines (www.tripdatabase.com). Unpublished studies were searched from online trials databases (www.clinicaltraials.com) (www.cancer.gov), from presentation abstracts of international cancer conferences from 2000–2007 and systematic reviews from the Cochrane Collaboration (www.cochrane.org/reviews/) and published metanalysis (Bohlius et al. 2006; Wilson et al. 2007). Internet sites were searched via Google and Google scholar (www.google.co.uk). My colleagues at the group for optimizing research in cancer anaemia (ORCA) provided expert advice.

Introduction – why cost effectiveness studies are required The demand for cancer treatment is predicted to rise signiﬁcantly over the next twenty years as the population proﬁle of developed countries age due to reduced birth rates and increased longevity. For a country such as the United Kingdom, the change in demographics predicts that cancer cases will treble in the next 20 years. Added to the demographic trends, costs will increase further as new technologies allow diagnosis of future disease in young adults, and cancer treatment relies on biotechnology agents and individually designed treatments (Latief 2004). These predicted changes will have a signiﬁcant effect on the overall national health budgets of developed nations, because cancer is already such a common disease. In 2002–2003, because of its age proﬁle of an elderly population, the United Kingdom spent the highest proportion of national total healthcare expenditure on cancer

Cost-effectiveness of treating cancer anaemia

815

Table 1. Direct costs for cancer in EU countries, 2002/2003 (from N. Wilking and B. Jonsson, 2005) Country

Direct costs for cancer (Million Euros)

Direct costs for cancer per capita (Euros)

Cancer cost as a % of total healthcare costs

Total Austria Belgium Czech Republic Denmark Finland France Germany Greece Hungary Ireland Italy The Netherlands Norway Poland Portugal Spain Sweden Switzerland United Kingdom

54,236 923 1,469 663 748 587 7,091 12,100 1,112 556 468 6,578 1,525 871 1,300 943 3,885 1,253 1,391 10,823

120 114 142 65 139 113 119 150 101 56 118 114 94 191 34 90 92 140 189 182

6.4 6.5 6.5 6.5 6.5 6.9 5.3 5.4 6.5 6.5 6.5 6.5 4.1 6.5 6.5 6.5 6.5 7.0 6.5 10.6

treatment in the European Union (10.6% of total healthcare expenditure). The mean proportion for the EU is 6.4% (Table 1) (Wilking and Jönsson 2005). Other countries are expected to see similar trends as the average population age rises, since age is the greatest risk factor for cancer. Cancer medicine is therefore a priority for overall cost control by governments or health insurance organisations.

The social costs of cancer care It is not only hospital costs that are predicted to increase, as social care will become more expensive in the developed world. The ageing population proﬁle, coupled with low birth rates and the trend to small family size or single households, will reduce the pool of community carers for the sick and inﬁrm. The higher social care costs will fall increasingly on the state, insurers or individuals, leaving a smaller pool of funding for direct medical care. In

816

P. Cornes

this changing world, developed countries will look to prioritise care as a way of controlling costs. Properly conducted health economic studies should help to direct care to where it has the potential to do the greatest good. Oncologists should become involved with planning, conducting, evaluating and reviewing critically these studies to ensure that future cancer medicine will receive its appropriate portion of the national health, social and research budgets. Already the effect of increased health spending has had negative effects. The cost of health insurance premiums in the USA has risen faster than the rate of general inﬂation and the national wealth (GDP) (Appleby 2001). In 1970 medical care represented less than one-tenth of U.S. personal consumption spending, the ﬁfth-largest component after food, housing, transportation, and household operation. In that year Americans spent roughly the same portion of their personal consumption on clothing as they did on medical care. Since that time, medical care has been a steadily increasing share of personal consumption. By 2001 medical care represented 18.2 percent of personal consumption spending and was the largest component (Reinhardt et al. 2004). The result has been that middle-income families in the USA have been increasingly unable to afford a comprehensive health insurance plan. In 2001, 28% of people with household incomes of $20,000 to $40,000 (£22,000 to 32,000) lacked insurance for at least part of the year. By 2005, the proportion was 41% (Tanne 2006). Inﬂation in the German State heath Insurance Drug Budget has shown a similar trend. In 2004 the German national drug budget was 21.7 Billion Euro ($27.2 Billion), but this rose by 16% in 2005 and 15% in 2006 and emergency drug budget controls had to be introduced (Tuffs 2006). Erythropoietins are recombinant biotechnology products, and so are more costly to produce than earlier generations of pharmaceuticals. Their clinical success has been demonstrated by sales. At the end of 2004 the world market for erythropoietin was valued at $11.1 billion with a growth of 8% over 2003. The market has been growing at an average annual growth rate of 18% over the previous 5 years (ASInsights 2007). This combination of high individual cost drug and high volume sales makes this an important area of interest for health economists.

Reasons for increased pharmaceutical expenditure by health services The rise in drug costs can be explained only in part by more patients being treated. New generation drugs have cost more than their predecessors. Pharmaceutical expenditure in the USA showed inﬂation of 18% in 1999, 16% in 2000 and 17% in 2001, while the actual volume of medicines prescribed increased by only 5–6% each year (Hogerzeil 2004). Canadian data

Cost-effectiveness of treating cancer anaemia

817

for 1987 to 1993 conﬁrms this trend. Over that period the cost per prescription rose 93%. Price rises of existing medicines accounted for 33%, increased quantity per prescription for 15%, while new medicines added 55% (Hogerzeil 2004). Health economists and insurers have noted these trends and have paid special attention to the costs and beneﬁts of the new generation of biotechnology medicines, such as Erythropoietins (Epo). Despite regulatory approval of erythropoietins and both national and international guidelines for its use, the different reimbursement status in different countries seems to explain the unequal access of patients to medically effective treatment (Bokemeyer et al. 2004). In Europe the ECAS survey (Table 2) has shown only 17.4% of anaemic patients undergoing cancer treatment received Epo (Ludwig et al. 2004). Within Europe the access to Epo varied between 5% and 66% depending upon the patient’s country of origin (Table 3) (Cavill 2002). Readers are reminded that economics is not primarily about saving money. It is about using scarce resources as efﬁciently as possible. There are a wide variety of different tools for economic analysis, and this diversity may explain why there has been such signiﬁcant divergence of opinion of the worth of erythropoietin.

Table 2. The treatments given to anaemic cancer patients undergoing treatment in the European Cancer Anaemia Survey Epo (Ludwig et al. 2004) Treatment

Patient distribution (%)

No treatment Erythropoietin Transfusion Iron

61.1 17.4 14.9 6.5

17% 15%

7%

No treatment Transfusion

61%

Erythropoietin Iron

818

P. Cornes

Table 3. The proportion of anaemic cancer patients treated with transfusion or erythropoietin and iv iron by European country, from Cavill 2002 Country

Proportion transfused (%)

Proportion given erythropoietin and iv iron (%)

Italy France Spain Germany United Kingdom

34 59 59 67 95

66 42 41 33 5

Willingness to pay for treatment One simple test of economic advantage is just to ask patients how much they would pay themselves for a treatment. An early 1998 cost-beneﬁt analysis from Canada used willingness to pay as a tool to see if erythropoietins were cost effective. Regarding the beneﬁts they would experience after 3 months of erythropoietin administration, 100 patients receiving cisplatin and noncisplatin therapy stated that they would be willing to pay an average of 587 U.S. dollars (U.S.$587) (95%CI: $300–$875) and U.S.$613 (95%CI: $324–$902), respectively. These beneﬁts were then subtracted from the total cost of the drug when administered to patients receiving cisplatin (U.S.$3530) and noncisplatin (U.S.$3653) therapy. This produced a net incremental treatment cost of U.S.$2943 (95%CI: $2655–$3230) and U.S.$3039 (95%CI: $2750–$3328) for the respective treatment groups, which the report authors felt to represent only a modest beneﬁt to patients (Ortega et al. 1998). Few patients understand the true costs of their care when it is provided by an employment or social insurer or national health service, and this test is therefore not in common use to decide reimbursement.

What is the cost of not treating anaemia in cancer? There is a high cost associated with cancer related anaemia. This suggests that clinically effective treatment of anaemia could be cost effective, with a potential for lower overall health costs through active management of the condition. The European cancer Anaemia survey showed that 61.1% of anaemic cancer patients received no speciﬁc treatment at all (Ludwig et al. 2004). The consequence of this was reduced quality of life, absence from work, loss of wages, increased morbidity and possibly mortality (Barnett et al. 2000, 2002).

Cost-effectiveness of treating cancer anaemia

819

Table 4. The annual insurance costs for patients with cancer increases with anaemic, compared with non-anaemic patients, 1998–2001 from Nissenson AR et al. 2005 Patient group

Number of patients studied

Costs in US dollars per patient

Anaemic Non-anaemic

22,030 17,542

34,009 9,034 (p < 0.001)

All this had a ﬁnancial cost, and anaemic patients cost signiﬁcantly more to care for than the general cancer population (Chaves et al. 2003; Penninx et al. 2003). Nissenson and colleagues studied this in an insured population in the USA (Nissenson et al. 2005). Using claims for care of 1998–2001, annual costs for 22,030 anaemic cancer patients were compared with a random sample of 17,542 cancer patients without anaemia. Anaemic cancer patients cost more by a factor of 3.8 (Table 4). These extra costs were spread across a wide range of extra work, with signiﬁcantly increased costs for outpatient visits, emergency room consultations, admissions, hospitalisation and laboratory tests (p < 0.001). Using a health insurance database, anaemic cancer patients having chemotherapy in 1999 showed signiﬁcantly increased healthcare costs compared with the non-anaemic population (Lyman et al. 2005; Berndt et al. 2005). Twenty-ﬁve percent of the 619 newly diagnosed cancer patients treated with chemotherapy had anaemia. Data for the 619 individuals were linked to their employers’ short-term disability records. Patients with anaemia were identiﬁed by a diagnosis codes recorded for anaemia or treatment of the patient with transfusion or erythropoietin. The presence of anaemia and longer length of transfusion therapy were associated with increased expenditures. In contrast longer length of erythropoietin treatment was associated with lower expenditures. The incremental costs due to anaemia among patients receiving chemotherapy were $5,538 per month at 2001 values in the ﬁrst 6 months following cancer diagnosis. Costs related to short-term disability leave made up 10.8%. A similar picture was seen with an American employer’s health care scheme. With more than 100,000 employees covered, total medical and disability claim costs for care between anaemic and non-anaemic new cancer patients were compared for 1996–1998 (Barnett et al. 2000, 2002). Anaemia increased annual costs by $3775 per patient. Only 36% of the extra costs were for direct medical care of anaemia such as transfusions, laboratory tests or erythropoietin. The majority of the cost, $2422 (64%), was spent on treating the symptoms and side effects of anaemia such as fatigue. Anaemic patients who were working suffered excess time off work, at an additional cost of $1117.

820

P. Cornes

Concept of the “total cost of illness” These studies suggest that the total cost of illness, for anaemic cancer patients, might be reduced with investment in an active treatment programme. One aim of health care should be to minimise the total costs of illness, not simply the direct medical expenditures. To minimise costs of an illness, studies should measure its impact on the total costs of illness to the patient, family, employer, and society. Extra investment in treatment from the hospital might be repaid by reduced social or work absence costs. This concept of “total cost of illness” is new in oncology. In a case study examining the total costs of back injury illness to employers it was found that medical expenditures accounted for less than half of the total costs of illness. The average total costs of illness varied by over 350 percent among employers, and a simple parameter (days off work) explained 62.5 percent of the variance in total costs of illness (Gustafson et al. 1995). A study by Straus and colleagues suggests that this “total cost of illness” approach might be useful in assessing cancer anaemia (Table 5). The trial randomised 269 patients having cancer chemotherapy to start erythropoietin for up to 16 weeks at the ASCO guideline trigger of haemoglobin <12 g/dl, while the other arm did not start until typical transfusion trigger levels of <9 g/dl were reached (Straus et al. 2003, 2006). Comparisons of baseline-adjusted quality of life scores between groups were signiﬁcant for the group starting erythropoietin at Hb <12 g/dl. The results, shown in Table 5, conﬁrm that erythropoietin given in accordance with the international guidelines would improve physical well being, functional well being, fatigue improved along with overall energy levels and activity (all p < 0.05). The patients starting at Table 5. The beneﬁt to starting erythropoietin during chemotherapy at Hb <12 g/dl compared with waiting to Hb <9 g/dl from Straus DJ et al. 2003 Parameter studied

Beneﬁt seen to Epo at Hb <12 g/dl compared with at <9 g/dl

Therapy (FACT)-General (G) physical well being FACT-Anemia (An) Fatigue FACT-An Total improved The Linear Analog Scale Assessment (LASA) Energy improved LASA Activity Days in bed Days with reduced activity Nights in hospital, clinic visits, and calls to physicians

Improved +1.1 (P = .024) Improved +2.7 (P = .005) Improved +3.2 (P = .008) Improved +6.0 (P = .007) Improved +6.0 (P = .008) Reduced 49.1% (P = .017) Reduced 29.4% (P = .042) Trend to reduction (P > 0.05)

Cost-effectiveness of treating cancer anaemia

821

Table 6. Cost categories included in cost-effectiveness studies Direct costs

Indirect costs

Hospital costs Hospital stay Inpatient consultations Nursing time Outpatient physician visits

Work days lost For patient For carer

Other consultations

Travel costs for treatment Personal time lost To cancer To anaemia To chemotherapy

Transfusions Tests Laboratory tests Imaging Diagnostic tests/procedures

Hb <12 g/dl had signiﬁcantly greater 49% reduction in the number of days spent in bed (52.2% vs 3.1%; P = .017) and 29% reduction in restricted activity days (41.6% vs 12.2%; P = .042). There was also a trend toward reductions in nights in hospital, clinic visits, and calls to physicians. With half the number of days spent in bed and a third less days with reduced activity it is likely that these patients required signiﬁcantly less social support in the form of carer time and missed employment of the patients and family carers. With social and community costs the major component of increased expenditure by anaemic cancer patient on treatment this reduced support might offer a signiﬁcant overall cost saving. It is difﬁcult to capture the full costs and beneﬁts of treatment in a cost effectiveness study, as so much varied data is needed from hospital, family physicians, home, carers and employers. The wide range of data needed is summarised in Table 6. For this reason most published studies compare the hospital costs of the treatment to the beneﬁt estimated by the added Quality Adjusted Life Year (QALY) method.

The QALY method for comparing treatments The data needed for a full cost-effectiveness study is complex and timeconsuming. The Quality Adjust Life Year method (QALY) simpliﬁes the study but is controversial. Figure 1 shows the outcomes of three different

822

P. Cornes 100 90 80 70 60

Treatment A

50

Treatment B

40 Treatment C

30 20 10 0 0

1

2

3

4

5

TIME IN YEARS

Fig. 1. Demonstration of the Quality Adjusted Life Year (QALY); the quality of life over time for patients treated with 3 different treatment protocols, A, B and C The standard treatment (A) results in a rapid fall in quality of life and death at 3 years from diagnosis. Economists score death as zero, and full quality of life as 1.0 (100%). Treatment B improves quality of life for a time, but does not increase survival. Treatment C improves quality of life and also prolongs survival by 18 months compare with standard care to 4.5 years from the time of diagnosis QALY scores for the treatments B and C, compared with standard care are: Treatment QALY gain compared with treatment A B C

1.5 years × (0.9 − 0.45) = +0.675 QALY gain (1.5 years × (0.9 − 0.45)) + (1.5 years × (0.9 − 0)) = +2.025 QALY gain

treatments, A, B and C. Treatment A is the hypothetical standard of care, treatment B can temporarily halt the decline in quality of life as the disease progresses but not prolong life, while treatment C can both palliate the symptoms of disease and prolong life. All three situations can be described with a single score, the QALY. The advantage of the QALY is that, in principle, it allows comparison between interventions that otherwise would have different outcomes because the diseases being treated and compared are different in their effects. In a rational world, the application of QALY analysis should lead to optimized health care as a ﬁnite resource would then be spent ﬁrst on treatments with the lowest cost per QALY, then the next cheapest until the health budget had been allocated. Given a ﬁxed budget this maximizes the total number of QALY’s gained (Ament 1993). This comparison of different diseases and treatments with the same QALY system is not possible using costeffectiveness Analysis (CEA), where incremental effects are assessed in natural units such as lives saved, years of life gained, blood pressure mea-

Cost-effectiveness of treating cancer anaemia

823

sured in mm of Hg, etc. Cost/QALY analysis is sometimes also called costutility analysis (CUA), where the quality of life beneﬁt of a treatment is give value by the degree to which the public would prefer it to other treatment outcomes (Drummond et al. 1997). Decision making at the level of a hospital department is complicated by the fact that knowledge about the value of a QALY is scare, and the value of the treatment may be seen from a hospital perspective, rather than considering overall costs. Prior to the development of erythropoietins, a hospital would have offered no active treatment to a patient with mild or moderate cancer anaemia. By spending hospital ﬁnance on erythropoietins the hospital would have invested in a treatment that might beneﬁt best the patient at home, or their employer, or the employer of their family member who has been acting as a carer instead of working. Since mild anaemia is not treated by blood transfusion, erythropoietins will not be seen to reduce the cost of future admissions. Since few physicians know the state of their hospital budget at any given time, nor the likely patient workload for the remaining ﬁnancial period, deciding treatment priorities on a case by case basis on cost/QALYs will be inherently unfair. For this reason governments and health insurance schemes are turning increasingly to agree beforehand the treatments that will be funded or reimbursed on the basis of a cost/QALY threshold and a central or national assessment of the treatment or technology. The QALY method of comparing treatment adds potential bias, as it is not disease speciﬁc. To estimate the value of a new treatment over the standard therapy, the preference value of the new treatment is calculated. This is the amount by which the new treatment is preferred over the standard therapy. This “preference value” is not typically assigned by the patients who gather data for the study. The value is often assigned by a panel of the lay public chosen to represent society and not by the patient (Greenhalgh 1997; Bosanquet and Tolley 2003). This value can be expressed in direct monetary terms. However the difference in wealth between individuals in society means that a common shared value is preferred. This can be assessed as the risk of immediate death from the treatment that the surrogate patient might accept to have the new treatment. In this latter case, the value is called the “standard gamble”. A panel of the public might, for example, chose to risk a chance of death of 1 in 20 to have the a treatment (treatment A), and for another treatment, might risk 1 in 10 (treatment B). This suggests that the public panel prefer treatment B, as they are prepared to gamble double the risk of death to obtain it. The ratio of risk of death is expressed between 1.0 (no risk accepted, and no “utility” to be gained by treating it) and O (would accept death to escape the current situation, with every utility to be gained by having a rare successful treatment result). Discovering the “preferred” treatment by a “standard gamble” of the risk of death may seem a very abstract term, but for many medical treatments it

824

P. Cornes

can be a realistic appraisal. Cancer patients might risk a higher chance of dying of increased complications from high dose chemotherapy a bone marrow transplant compared with standard intensity chemotherapy if the intensive but risky treatment could result in higher chances of cure or prolonged life. In the same way, oncologists will have seen patients refuse adjuvant chemotherapy with cancer, preferring a higher risk of death from the cancer in the future to avoid the risks of toxicity now. Economists call this latter choice a “time-trade-off” as patients may prefer to live better in the short term by avoiding chemotherapy toxicity but sacriﬁce length of life at a date in the future, when the cancer could return. Using these tools, economists can gain parametric measures of the effect of treatment of quality and length of life, and of treatment preference free of direct monetary value. From this the value, the “utility” of the treatment is determined. A treatment with medical beneﬁts that few patients are willing to accept any risks to attain, is clearly a less preferred treatment (with a lower utility) that one which patients will risk a signiﬁcant chance of immediate death or shortening of life. Within health and health care, the greater the preference for a particular health state, the greater the “utility” associated with it. “Utilities” of health states are generally expressed on a numerical scale ranging from 0 to 1, in which 0 represents the “utility” of the state “Dead” and 1 the utility of a state lived in “perfect health”. The utilities assigned to a speciﬁc state of health can be estimated using a series of techniques such as Standard Gamble or Time Trade-Off. Another method is to use a validated questionnaire. One example of this is the EQ-5D, where a 5 question panel about health is scored to provide a scale of results for different choices. This system has been validated, and then costed in monetary terms for different countries and cultures, and may help ensure a consistency between health economic studies where preferences are used to determine value. For the USA the scores have been veriﬁed in 1998 (Johnson et al. 1998; Johnson and Coons 1998; Agency for Healthcare Research and Quality 2005). For the UK, the scores were validated in 1996 (Kind 1996). For this reason, the ﬁve question EQ-5D scale is used frequently in economic analyses, and has been used for recent systematic evaluation of erythropoietin treatment. To help readers, the ﬁve question EQ-5D scale is described in Table 7, and examples of the utility or value scores of typical results are shown in Table 8. Health services can use these measures to decide if a treatment is sufﬁciently cost effective to be funded. Using measures of affordability, a national provider or insurance company could set a monetary threshold above which treatment would not be paid for. The threshold to funding would typically be expressed as a maximum cost per QALY, or maximum cost/utility. When there is a choice between treatments, it can be described as an incremental cost-effectiveness ratio (ICER), the ratio of change in costs from providing the new treatment to the change in QALY that could be gained from it. Not

Cost-effectiveness of treating cancer anaemia

825

Table 7. Using the EQ-5D ﬁve question panel to produce a series of “health states”: Scores for the EQ-5D are generated from the ability of the individual to function in ﬁve dimensions. These are: Mobility

Pain/discomfort

Self-care

Anxiety/depression

Usual activities (work, study, housework, leisure activities)

1. 2. 3. 1. 2. 3. 1. 2. 3. 1. 2. 3. 1. 2. 3.

No problems walking about. Some problems walking about. Conﬁned to bed. No pain or discomfort. Moderate pain or discomfort. Extreme pain or discomfort. No problems with self-care. Some problems washing or dressing. Unable to wash or dress self. Not anxious or depressed. Moderately anxious or depressed. Extremely anxious or depressed. No problems in performing usual activities. Some problems in performing usual activities. Unable to perform usual activities.

Each of the ﬁve dimensions used has three levels – no problem, some problems and major problems – making a total of 243 possible health states, to which ‘unconscious’ and ‘dead’ are added to make 245 in total. This creates a “health Utility Index”. Each of the 245 health states has been reviewed by a citizens’ panel in different European countries to determine the utility of each state and the monetary value of improving it. Reference: The EQ-5D is a public domain instrument from the EuroQol Group at www.euroqol.org

all health providers have formal economic evaluations of treatments, but with the escalating costs of modern medical care, their use is increasing over time. A cost/QALY study from one health system or country cannot be assumed to be correct in another. This is because of the different wealth and make up of societies. For example, in a country with large average family size and a culture of extended family social support, the value given to home independence by a treatment may be less than in a country where home and social care is difﬁcult to provide without a monetary cost. In the USA, the low level of acute hospital medical beds, nurses and physicians compared with the averages for developed countries in the OECD means that the value of a treatment that can reduce the rate of hospitalization is may be higher than in countries with hospital capacity to admit patients for episodes of care such as a blood transfusion (Anderson et al. 2006; Reinhardt et al. 2004). Each study will therefore express what economists call “societal values”. Despite these difﬁculties, it is possible to highlight differences in the threshold value

826

P. Cornes

Table 8. Using the EQ-5D ﬁve question panel to produce health state values. Six examples are shown with different scores. Value has been described both by the time trade off method and using a visual analogue scale to produce a single ﬁgure for each of the 245 possible health states in 8 different European countries and the USA Health state score

Description

Valuation

11111 11221

No problems No problems walking about; no problems with self-care; some problems with performing usual activities; some pain or discomfort; not anxious or depressed Some problems walking about; some problems washing or dressing self some problems with performing usual activities; moderate pain or discomfort; moderately anxious or depressed No problems walking about; some problems washing or dressing self; unable to perform usual activities; some pain or discomfort; not anxious or depressed Some problems walking about; no problems with self care; no problems with performing usual activities; moderate pain or discomfort; extremely anxious or depressed Some problems walking about, unable to wash or dress self, unable to perform usual activities, moderate pain or discomfort, moderately anxious or depressed

1.000 0.760

22222

12321

21123

23322

0.516

0.329

0.222

0.079

for treatments assigned by different countries to decide reimbursement (Table 9). The majority of countries, even if they use cost effectiveness studies to decide reimbursement have not published them as a ﬁxed threshold. They can however be implied by reviewing the recent approvals for treatment with the associated published health economic studies. For example, USA Medicare rules for the provision of renal dialysis show that interventions costing $50,000 USD per QALY are mandated by federal law (Ubel 2003; McGregor 2003). From these studies a pattern of cost-effectiveness thresholds are expected to emerge (Eichler et al. 2004). The threshold for adoption is thought to be somewhere between $20,000 (£11,300, €16,500)/QALY and $100,000/QALY, with thresholds of $50–60,000/QALY frequently proposed (Bell et al. 2006). Cost-beneﬁt analysis (CBA) provides a clear decision to undertake an intervention if the monetary value of its beneﬁts exceed its costs. Policymakers in many countries have avoided setting a public explicit cut-off price per QALY (or other measure of outcome) above which tax or insurance will not be used to purchase health care. Such a pronouncement would probably

Cost-effectiveness of treating cancer anaemia

827

Table 9. Comparison of the thresholds for maximum cost per QALY funded by country. Currency conversion is quoted from the source reference where available, and from conversion at www.xe.com on 7 August 2007 where equivalent cost was not given (indicated with an asterisk*) Country

Cost maximum in referenced document

Euro equivalent

GP pound equivalent

USA Dollar equivalent

Reference

Holland Sweden United Kingdom USA

Euro 80,000 SEK 500,000 £30,000 GB Pounds $50,000

€80,000 €53,384 €44,134*

£55,000 £36,780* £30,000

$105,000 $74,384 * $60,661 *

[42] [43] [44] [45]

€36,378*

£24,724.24*

$50,000

[37] [38] [46] [47]

be politically controversial. Instead, countries typically have sought to set that upper limit implicitly, through a mixture of price controls and limits on capacity (Hirth et al. 2000; Reinhardt et al. 2004). The Dutch government has however signalled an openness in the debate over rationing national health care resources by publicly proposing a €80,000 euro per QALY maximum for socially funded health insurance (Sheldon 2006). For Sweden the threshold is 500,000 Swedish crowns, approximately €53,384, £36,780, $74,384 * (Dewilde et al. 2006). Using data for all payments for healthcare, both hospital and community, the UK National Health service (NHS) has been able to discover the total cost paid for care of cancer patients in England. Joining this data to the outcome results has enabled it to discover the current expenditure to gain an extra year of life for an average cancer patient is £13,100 at 2004/5 costs (€19,333 EUR, $26,670 USD). Very importantly, this ﬁgure is not adjusted for quality of life. However, it is noteworthy that this current spending is signiﬁcantly less than the UK accepted threshold of £30,000 per QALY for reimbursement of new treatments (€44,288 EUR, $61,079 USD) (Martin et al. 2007). This suggests that the UK National Health service will approve up to approximately twice the average cancer patient spend to secure better treatments for its patients. The European network for Health Technology Assessment, EUnetHTA, coordinates the efforts of 27 European countries including 24 Member States of the European Union in evaluating health technology in Europe. It was created in 2006 and aims at developing an organisation for a European network for health technology assessment along with practical tools to produce, disseminate and transfer technology assessment results into policy advice to the Member States and European Union.

828

P. Cornes Table 10. Examples of the utility placed on different diseases (health states) by the Standard gamble method of eliciting QALY value for a health state (from Wang AJ. Key concepts in evaluating outcomes of ATP funding of medical technologies. J Technol Transfer 23 (2): 61–65) Health State

QALY Value

Full Health (Reference state) Mild angina Moderate angina Home dialysis Insulin-dependent diabetes Rheumatoid arthritis Severe angina Blind, deaf, or dumb Chronic obstructive pulmonary disease Death (Reference state)

1.00 0.90 0.70 0.64 0.58 0.52 0.50 0.39 0.38 0.00

Examples of the utility placed on different health states can give readers an idea of the utilities of common illness states. Table 10 lists some examples. A treatment that controlled moderate angina completely for two years would be expected to gain 0.6 of a QALY (2 × (1.0 − 0.7) = 0.6). A treatment that controlled the symptoms of rheumatoid arthritis for 4 years would have gained 1.92 QALY (4 × (1.0 − 0.52) = 1.92). If that treatment cost €18,000 euros more than the standard care, the cost per QALY would be €9375 (18,000/1.92). Economists regard full health as a state of score 1.0, suggesting that no person would gamble to risk life to improve it. They also regard death as 0.0 (zero), suggesting that a person about to die will take all risks to get a cure to escape it. This is not always seen in practice. Few people regard themselves as fully ﬁt and enjoying a full quality of life. Using utility scores in health members of the public in the “Healthy People 2000 survey” showed that the average utility of persons age 45 to 50 was 0.86, and age 65 to 70 was 0.77 (Erickson et al. 1995). Furthermore, all cancer physicians will at some time have met a patient who preferred to die than face a continued life of suffering. This suggests that there can be a “health state” with a utility of a minus score that may not be recognised by economists. Countries such as Holland and Switzerland have recognised this by developing a medical service to deliver euthanasia for patients who would prefer to die than continue in a poor state of incurable health. This state of preference for death has been recorded for patients with cancer anaemia, suggesting that for some patients, the QALY of their health state is very low, and that effective symptom relief would be a highly valued treatment.

Cost-effectiveness of treating cancer anaemia

829

In theory, a Quality Adjusted Life Year (QALY) is of the same value no matter to whom it accrues. This implies that the social value of a QALY, and hence of health, is invariant with age, sex, occupation or responsibility for the current health crisis. This egalitarian view is not supported in studies of the general public, who make up the panels of citizens that determine the utility of potential treatments. Treatment for the young is valued higher than for the old, with health at infancy given approximately twice the value of health at advanced ages (65 years) (Busschbach et al. 1993). Readers need to understand that cost effectiveness studies produced by the pharmaceutical industry may be restricted in publication. For example, the New England Medical Journal blocks the publication of cost effectiveness analyses funded by industry if at least one author has direct ﬁnancial ties with the sponsoring company (Kassirer and Angell 1994). To view these studies it may be necessary to review the small print of health economic appraisal documents, where such analyses are frequently recorded and referenced. Readers will need to be reassured that association with pharmaceutical companies does not imply poor practice. A recent review of decisions made by the Food and drugs Administration (FDA) of the USA showed that in 221 meetings involving 1957 advisory committee members, excluding advisory committee members and voting consultants with conﬂicts would not have altered the overall vote outcome at any meeting studied (Lurie et al. 2006). For presentations of pharmaceutical company studies at UK health economic appraisals, there is an agreed standard economic model. For the 2007 UK National Institute for Clinical Excellence (NICE) appraisal of cancer anaemia this was the “Birmingham Model” (www.nice.org/). Different data input to that model will however produce different results. For example, assuming a both a survival and quality of life advantage to a supportive care treatment such as erythropoietin will result in a much smaller cost per QALY than if the beneﬁt was restricted solely to quality of life (Miners et al. 2005). At reimbursement approval hearings, readers will ﬁnd frequently that a range of models will be submitted. These will explore different subgroup data analysis, looking to see if a favourable patient group can be identiﬁed for which economically effective beneﬁts can be conﬁrmed.

Transfusion and erythropoietin compared The European cancer anaemia study (Table 2) showed that two clinically effective treatments for anaemia were in common use, transfusion and erythropoietin. Systematic reviews conclude that the use of erythropoietin signiﬁcantly reduced the relative risk of receiving red blood cell transfusions by O.64 (RR 0.64; 95% CI 0.60 to 0.68, 42 trials, with 6,510 patients randomised). Since so few patients with cancer anaemia actually receive blood transfusion, for example 14.9% in the ECAS survey, the average patient saved only one

830

P. Cornes

Table 11. International guidelines for blood transfusion thresholds Guideline group

Haemoglobin level for transfusion in g/dl

Ofﬁce of Medical Applications NIH, USA, 1998 College of American Pathologists’ Canadian Expert Working Group American Society of Anesthesiologists Task Force

<7

UK Guidelines for Blood Transfusion practice, January 2001. CREST Secretariat

5 to 8 7 to 8 “rarely needed if the hemoglobin concentration is greater than 10 g/dL, and almost always needed when the level is less than 6 g/dL.” <7 Usually transfuse 7–10 Individual decision for each patient >10 Usually do not transfuse

unit of blood less than the control group (WMD −1.05; 95% CI −1.32 to −0.78, 14 trials with 2,353 patients randomised) (Bohlius et al. 2006). Low levels of transfusion are in line with international guidelines (Table 11). Blood is a scarce and precious resource, and although it can be critical in supporting circulation and saving lives with severe anaemia, it has only a transient effect when transfused and has not been demonstrated to deliver medium or longterm quality of life advantages to anaemic patients with cancer. A study published in 1997 by Shefﬁeld and colleagues proposed that transfusion was the cost effective option for correcting anaemia during cancer chemotherapy (Shefﬁeld et al. 1997). The study was not a prospective clinical design, but a modelling study based on published clinical data. Patients received either leukocyte-depleted packed red blood cells (PRBCs) or subcutaneous erythropoietin 150 units/kg three times per week for 6 months (24 wk). After 6 weeks, if erythropoietin recipients did not display a response, they received erythropoietin 300 units/kg three times weekly for the duration of therapy. If erythropoietin recipients still exhibited no response, they were given blood transfusions. For a treatment period of 24 weeks, approximately 64% of erythropoietin recipients responded at an average expected cost of $12,971 per patient. One hundred percent of transfusion recipients responded at a cost of $481; this was reported to predict a cost saving of $8,490 for a transfusion policy. This study had signiﬁcant ﬂaws in that it assumed that transfusion was 100% effective for correcting haemoglobin, and assumed that any requirement for transfusion of patients using erythropoietin indicated a failure of that treatment, even if overall the quality of life and transfusion requirements of the anaemic patients were lower overall. It did not compare the mean haemoglobin levels over the six months of the study, not the effect on quality of life. It is difﬁcult to compare transfusion and erythropoietin in cancer anaemia, as anaemic cancer patients may need access to both treatments for

Cost-effectiveness of treating cancer anaemia

831

1

rhEPO 1

Hb (g/ dl)

RBC Transfusions

1 8 6

= transfusions given

4 0

30

60

90

120

150

180

210

Days of treatment Fig. 2. The Hb proﬁles (in g/dl) of patients receiving red blood cell (RBC) transfusions or recombinant human erythropoietin (rhEPO) (redrawn from Littlewood T, et al. 2006)

optimal clinical care. Transfusion is usually reserved for the rapid correction of severe or life threatening anaemia with Hb <8 g/dl where the primary aim is circulatory support rather than improved quality of life (Pirker et al. 2003). Average transfusions are of 2 units, taking cancer patients to an Hb target of 10 g/dl in a matter of hours. Erythropoietin is advocated for mild or moderate anaemia, with starting Hb at between 9 and 11 g/dl with the aim of improving quality of life by returning patients to normal Hb levels of >11 g/dl. The effects of transfusion are of rapid onset but are short lived, while erythropoietin produces sustained responses over weeks, such that the mean Hb of an anaemic patient supported by a transfusion policy runs typically lower than a patient supported by erythropoietin (Fig. 2) (Littlewood et al. 2006). Where repeated transfusions have been advocated, to support radical chemoradiotherapy by maintaining an average Hb of 11–12 g/dl, the strategy fails a signiﬁcant proportion of patients. Even weekly transfusion is unable to sustain Hb correction during pelvic treatment for cervix cancer (Kapp et al. 2002). To transfuse blood is an expensive option, and many publications seem to underestimate this. The average treatment is a two-unit transfusion. At 1998 prices the mean cost per transfusion was estimated at $938, not the $481 of the previous study described (Cremieux et al. 2000). The blood itself was only 19% of the expense, while labour and overhead costs made up 35% and 46% respectively. This transfusion requires an average of 4.5 hours to deliver, and an overnight stay in hospital for 1 in 4 patients is required (Barrett-Lee et al. 2000). This underestimate of the cost of transfusion, by ignoring the overhead costs of hospital care and attending staff, is widespread. Forbes and colleagues suggested a cost of $155 per unit of blood transfused, made up of

832

P. Cornes

Table 12. Costs of transfusion of 2 units in UK hospitals, 2004 estimates, from Agrawal et al. 2006. Currency conversion as may 2007 using the online service at www.ex.com* Item

Cost £

Cost Euro*

Cost US $*

Blood Staff cost per transfusion Disposables Wastage Derived costs, such as hospital stay Total

287.56 37.24

421.547 54.5830

573.517 74.2626

13.25 11.86 201.11

19.4194 17.3822 294.755

26.4209 23.6492 400.893

546.12

800.474

Proportion of total cost as % 53% 7% 2% 2% 37%

1,088.64

Changes in the relative costs of blood transfusion and erythropoietins over time.

material costs 37%, Lab tests 43%, blood bank handling 13% and administration 7% (Forbes et al. 1991). A study from the UK in 2004 estimated that the average transfusion cost hospitals £ 546.12 (€800.22 or $1,088.24) (Agrawal et al. 2006). Transfusion sessions for patients receiving units of red blood cells, within either haematology or oncology departments, were followed using time and motion techniques to measure the direct costs. Other data were collected from the centres to calculate the cost of disposables, blood wastage and blood bank machines. The proportion of the costs are shown in Table 12. It should be noted that signiﬁcant indirect costs, such as those incurred by patients, their carers and societal costs, were not considered. The cost of blood has been increasing over time. There are signiﬁcant costs involved in the collection and testing of donated blood and these costs continue to increase with new safety and puriﬁcation standards (Allain 2003). Blood transfusions are also associated with the potential for transmission of bloodborne infections, haemolytic reactions, volume overload, supply deﬁcits for certain blood types, allergic-type reactions, and iron overload (Provan 1999; Kushner et al. 2001; Mortimer 2002; Allain 2003; Sandler et al. 2003; Ludlam and Turner 2006). Moreover, single transfusions are not effective for the correction of chronic anaemia (Österborg 1998). Multiple hospital attendances are required, with 25% of transfusions requiring hospitalization (Barrett-Lee et al. 2000) leading to multiple exposures to the health risks of transfusion. A retrospective chart review of 219 patients with solid tumours found that transfusion aberrations, necessitating unusual laboratory monitoring or resulting in a transfusion reaction, were quite frequent (Mohandas and Aledort 1995). Over a 12-month period, 812 units of red cells were

Cost-effectiveness of treating cancer anaemia

833

transfused and 100 aberrations were recorded. A positive antibody screen that required further work-up was recorded in 22 patients (10%) and transfusion reactions occurred in 19 patients (8.7%). In view of the risk of adverse events, the necessity for each transfusion should be carefully weighed up beforehand (Mortimer 2002). In contrast to the rising costs of transfusion over time, the real costs of erythropoietin has been falling (Remak et al. 2003). This comes from both the reduction in cost per unit of drug, and also because more recent protocols for treatment have increased effectiveness and reduced the mean doses delivered. True cost-effectiveness of a treatment can change over time, and early analysis can reach incorrect conclusions because of data deﬁciencies.

Review of published cost/QALY health economic analyses The ﬁst step to a economic utility analysis is to review the trials with the best quality of evidence. This translates typically into a systematic review of randomised clinical trials. From this it should be possible to estimate: • The response rate to treatment and whether this is homogeneous across patient subgroups (for example solid cancers vs haematological tumours). • The duration of treatment (12–16 weeks is typical for most published studies) • The costs of treatment per unit time • Outcomes in terms of gain in quality of life and length of life (QALY) and in patient preference (utility), converted to costs per QALY • Potential downstream costs saved by this treatment and costs increased from adverse events or failure to respond to the treatment. • From this the incremental cost effectiveness ratio (ICER) of erythropoietins compared with standard care can be estimated. The ICER is the extra costs required to gain one QALY. This might be estimated for the whole patient group represented in the clinical trials, and then again for different subgroups of patients to see if the beneﬁt is restricted to a minority. • Using agreed thresholds for reimbursement, to decide if reimbursement should be approved. Since 1998 there have been several systematic reviews by guideline development groups, but only 2 formal national economic evaluations for reimbursement have been identiﬁed, and only in two studies was a formal incremental cost effectiveness ratio declared. Italian 1998 review The Laboratory of Medical Informatics in Pavia, Italy, performed a formal cost/QALY analysis of erythropoietins in 1998. The clinical beneﬁts were

834

P. Cornes

assessed as reduction in the need for transfusions with chemotherapy. Erythropoietins were judged to reduce the probability of transfusions from 21.9% to 10.4%, and the number of RBC units per patient per month from 0.55 to 0.29. This gave an ICER of $189,652/QALY (Barosi et al. 1998). USA Agency for Healthcare Research and Quality 2001 review The Technology Evaluation Center, Blue Cross and Blue Shield Association, Chicago performed a systematic review in 2001 for the Agency for Healthcare Research and Quality of the USA (AHRQ) (Seidenfeld et al. 2001). They concluded that erythropoietins decreased the percentage of patients transfused by 9%–45% in adults with mean baseline hemoglobin concentrations of 10 g/dL or less (seven trials; n = 1,080), by 7%–47% in those with hemoglobin concentrations greater than 10 g/dL but less than 12 g/dL (seven trials; n = 431). The number of patients needed to treat to prevent one transfusion was 4.4 for all studies, 5.2 for higher quality studies. AHRA advised that quality-of-life data were insufﬁcient for meta-analysis, and so no formal cost/QALY estimate could be determined. Canadian Cancer Care Ontario 2003 review Systematic reviews were conducted in 2003 (and earlier in 1997) by the Canadian Cancer Care Ontario Practice Guidelines Initiative (The Role of Erythropoietin in the Management of Cancer Patients with NonHematologic Malignancies Receiving Chemotherapy (Practice Guideline #12-1 (Version 2.2003), but no formal economic evaluation was performed (Cancer Care Ontario Practice Guidelines Initiative 2003). Danish Centre for Evaluation and Health Technology Assessment 2004 review The Danish Centre for Evaluation and Health Technology Assessment (DACEHTA) reviewed Erythropoietin for Anaemic Cancer Patients in 2004, and a summary is available in English (Danish Centre for Evaluation and Health Technology 2004). Analysis involved a systematic review of the randomized trials, and a cost evaluation of guidelines developed by the Danish Society for Oncology. These guidelines recommend that EPO be offered to all those cancer patients with a haemoglobin concentration <6.0 mmol/l (approximately 9.7 g/dl), who suffer from chemotherapy-induced anaemia, and that EPO, depending on the clinical circumstances, may be given to patients suffering from less severe anaemia (haemoglobin between 6.0–7.5 mmol/l (9.7 to 12.4 g/dl)).

Cost-effectiveness of treating cancer anaemia

835

The systematic review was able to demonstrate that a policy of reimbursement for erythropoietin would save blood transfusions, but the effect of erythropoietin on quality of life was uncertain. This was blamed on the very different methodologies used in the different randomized trials. In one half of the studies examined, the applied methods of measurement showed no improvement in the quality of life among patients treated with EPO. The other half showed a variable positive effect. However, the applied methods of measurement were not felt to be comparable. For this reason, no formal cost/QALY assessment was performed. The average cost per patient treated with EPO was estimated at DKK 46,000–64,000/EURO 6,175–8,591 (DKK 35,000–41,000/EURO 4,698–5,503 per patient with a quantity discount rate of 30% on the medicine). The average cost of blood transfusions has been estimated at approximately DKK 4,000/537 EURO per patient. Employing EPO instead of blood transfusions involves an average cost of approximately DKK 109,000/EURO 14,631 per additional patient (DKK 77,755/EURO 10,437 with a quantity discount rate of 30% on the medicine), who responds to the treatment when receiving EPO instead of blood transfusions. The high estimated cost, along with uncertain utility beneﬁts because no cost/QALY analysis was performed, meant that DACEHTA advised against

Table 13. Summary of economic estimates from 2004 DACEHTA review (2004) currency conversions at August 2007 rates*

Average cost of transfusion per cancer patient Average cost of erythropoietin per patient (assuming 30% discount for medicines) Cost QALY estimate Employing EPO instead of blood transfusions (assuming 30% discount for medicines)

Cost in reference currency

Euro equivalent

GP Pound equivalent

USA dollar equivalent

DKK 4,000

€537 EURO

£364

$734,488

DKK 35,000–41,000

€4,698–5,503

£3,180–£3,725

$6,427–7,529

uncertain

uncertain

uncertain

uncertain

DKK 77,755/

€10,437

£7,063

$14,278

836

P. Cornes

the routine use of erythropoietins to treat cancer anaemia, and requested that further randomized trials assessing quality of life and economic beneﬁts be performed before DACEHTA decided whether EPO treatment should be recommended generally to a larger group of anaemic cancer patients undergoing chemotherapy.

United Kingdom National Institute for Health and Clinical Effectiveness review – 2007 Using an update of the the Cochrane collaboration meta-analysis of 2005 and 2006, the United Kingdom National Institute for Health and Clinical Effectiveness (NICE) discovered the beneﬁts expected from erythropoietin treatment (Danish Centre for Evaluation and Health Technology 2004). This is summarised in Table 14. Table 14. Key data produced by the systematic review of randomised trials that was performed for the 2007 UK NICE appraisal – Wilson J, et al. 2007 Data measure

Result

The best estimate of haematological response (improvement by 2 g/dL) with erythropoietin Haemoglobin (Hb) change with erythropoietin Response rate for erythropoietin Reduction in the number of patients who receive a red blood cell transfusion with erythropoietin The trial duration was most commonly Statistical heterogeneity in the estimate of haematological response Quality of life outcomes: data on quality of life collected by 20 trials

RR 3.4 (95% CI 3.0 to 3.8, 22 RCTs) weighted mean difference of +1.63 g/dl (95% CI 1.46 to 1.80) 50% 18%

Survival outcomes

Risk of a serious adverse event

16–20 weeks No important differences between any subgroups examined. A broadly positive effect was analysed using vote counting and qualitative assessment for erythropoietin. It was not clear how these results translated into utility gains. This is a crucial data problem, as it makes QALY and utility estimates inaccurate No difference, with HR 1.03 and 95% CI 0.88 to 1.21 (variance estimate inﬂated for substantive heterogeneity, χ 2 = 37.75; 27 df, p = 0.08). 0.05 (1 in 20 patients treated)

Cost-effectiveness of treating cancer anaemia

837

Calculations were done to estimate the cost of treating an average adult cancer patient, 70 kg in weight, undergoing erythropoietin treatment in the context of a four weekly chemotherapy regime, lasting for six courses, with erythropoietin starting in the second cycle. Administration costs were estimated at £8.01 per week. The costs of a serious adverse event were calculated at £101 with a chance occurrence in 1 in 20 patients treated (risk 0.05) per month. Using list prices the costs of a four week erythropoietin course was estimated at £1,106.7. Costs varied slightly between Erythropoetin types with a range of £1,087.9 to £1,102.2 per month (€1,605.93 to €1,627.03 EUR, $2,215.61–$2,244.73 USD). (Euro and US Dollar equivalent at August 2007 rates). Erythropoietin was estimated to save blood transfusions, with a blood unit cost of £120.22 (€178 EUR, $243 USD), and an administration charge of £413 per hospital transfusion (€611 EUR, $836 USD). The NICE Assessment Group’s economic evaluation used a 3-year time horizon. The model evaluated the use of erythropoietin analogues (with the possibility of blood transfusion) versus blood transfusion alone. Patients included in the model were characterised only by their baseline haemoglobin concentration at the start of chemotherapy. No other characteristics, such as cancer type or type of anti-cancer treatment, were assumed to inﬂuence outcome. In the treatment arm, the erythropoietin analogue was assumed to be given when haemoglobin concentration fell below 13 g/dl. A full dose was assumed when haemoglobin concentration was less than 12 g/100 ml and half doses when haemoglobin concentration was between 12 and 13 g/dl. The baseline distribution of haemoglobin was restricted to levels less than or equal to 11 g/dl in sensitivity analyses, following the change to the marketing authorisations during 2005 (starting haemoglobin restricted to 11 g/dl or lower for all products). Erythropoietin analogue treatment was assumed to stop if and when haemoglobin concentration reached 13 g/dl. Response to treatment was deﬁned as a 2 g/dl increase in a given haemoglobin concentration. Blood transfusion was considered when haemoglobin concentration was below 10 g/dl. Readers will be aware that this model is very different from the contemporaneous guidelines for treatment from the EORTC that advise initiating erythropoietins at an Hb level of 9–11 g/dl depending on symptoms (Bokemeyer et al. 2004). The decision to transfuse when the Hb fell below 10 g is a direct odds with the advice from the UK chief medical ofﬁcer who wrote that “there is little evidence that transfusion of someone with a haemoglobin level of 8 g/dL is necessary” and with international guidelines (Donaldson 2004) that suggests that the model is based on an overuse of both erythropoietin and transfused blood, compared with current guidelines for care. Data was submitted to the NICE review in 2007 using models from the manufacturers of the different erythropoietins. They can be reviewed in detail on the NICE website at www.nice.org.uk, but the Utility estimates in relation to haemaglobin levels are interesting and reproduced in Table 15. All show that a close relationship can be found between the Hb and quality of life and

<8 8–9 9–10 10–11 11–12 12–13 >13 Incremental gain in taking a patient from Hb 8–9 to 12–13

Health state by Hb (g/dL)

0.564 0.608 0.629 0.665 0.715 0.750 0.750 0.75 − 0.608 = 0.142

Company model B

Company model A 0.564 0.639 0.623 0.699 0.728 0.75 0.75 0.75 − 0.639 = 0.11

EQ5D

EQ5D

0.466 0.563 0.631 0.692 0.749 0.789 0.810 0.789 − 0.563 = 0.226

Company model C

Time-trade-off

Measure of utility used

0.481 0.615 0.615 0.781 0.781 0.781 0.856 0.781 − 0.615 = 0.166

Company model D

Time-trade-off

Table 15. The relationship between utility of treatment and Hb in 4 different models from 3 different pharmaceutical companies submitted to the UK NICE 2007 appraisal

838 P. Cornes

Cost-effectiveness of treating cancer anaemia

839

utility, with an incremental gain of 0.11 to 0.226 for correcting anaemia from haemoglobins of 8–9 g/dl back to the normal levels of 12–13 g/dl. The utility revealed for an anaemic cancer patient with Hb of 8–9 is similar to that of patients with Insulin-dependent diabetes, Rheumatoid arthritis and severe angina, all of which would be considered severe enough diseases to require an attempt at treatment. If a patient with an Hb of 8–9 g/dl could be corrected to normal, the duration of clinic beneﬁt over standard care would need to be between 4 and 9 years (4.43–9.09 years) to gain one whole incremental unit of utility. The NICE 2007 economic model assumes a treatment period of 6 months and a clinical beneﬁt after treatment of a further 3 months. It also assumes that only 50% of patients respond. This suggests a range of total utility that could be gained from erythropoietin treatment in this situation is from 0.11 × 0.5 × (9/12) = 0.04215 to 0.226 × 0.5 × (9/12) = 0.08475. With a threshold of willingness to pay ICER of £30,000, and a low total utility gain probable, treatment would need to be inexpensive in this model to gain approval for reimbursement. The actual NICE analysis produced an ICER of over £100,000 per additional QALY. The results of the sensitivity analysis demonstrated that erythropoietin analogues became more cost effective as the baseline distribution of haemoglobin was restricted to lower levels but still remained above

1 0.9 0.8 A B C D Linear (D) Linear (C) Linear (B) Linear (A)

0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 6

7

8

9

10

11

12

13

14

Hb

Fig. 3. There is a strong correlation between utility and haemoglobin using 4 different economic models (A, B, C, D) with lines of best ﬁt estimated using a linear function from data submitted to NICE appraisal, 2007

Cost of a 4 week course of erythropoietin (slight variation between erythropoietin types) ICER Cost QALY estimates All studies Starting Hb <8 g/dl Platinum chemotherapy With iron supplements Ovarian cancer treated with platinum chemotherapy

Average cost of transfusion per cancer patient

blood unit cost of $243 and an administration charge of $836 per hospital transfusion $2,215.61–$2,244.73

blood unit cost of €178 and an administration charge of €611 per hospital transfusion €1,605.93 to €1,627.03

€147,887/QALY €96,127 to 118,310/QALY €57,676/QALY €59,155/QALY €26,620/QALY

blood unit cost of £120.22 and an administration charge of £413 per hospital transfusion £1,087.9 to £1,102.2 at UK list price (no discount)

>£100,000/QALY £65,000 to £80,000/QALY £39,000/QALY £40,000/QALY £18,000/QALY

$202,321/QALY $131,509 to 161,857/QALY $78,905/QALY $80,928/QALY $36,418/QALY

USA dollar equivalent

Euro equivalent

Cost in GB pounds

Table 16. Summary of economic estimates from 2007 UK NICE appraisal (with Euro and US Dollar equivalents at August 2007 rates)

840 P. Cornes

Cost-effectiveness of treating cancer anaemia

841

conventional cost-effectiveness thresholds. The most favorable ICERs were obtained when a baseline haemoglobin concentration of 8 g/dl was assumed for all patients, and these were in the range of £65,000 to £80,000. Subgroup analyses were performed for patients treated with platinum chemotherapy, for the addition of iron supplements, and for patients with ovarian cancer. When incorporating the clinical evidence from the subgroup analyses and strategies explored, the ICER estimate for the subgroup of patients who received platinum-based chemotherapy was £39,000 per additional QALY, and the ICER for the treatment strategy of including supplemental iron was £40,000 per additional QALY. Patients receiving platinum-based chemotherapy for ovarian cancer showed and ICER which was £18,000 per additional QALY. The Assessment Group cautioned that this latter result as modeled was particularly sensitive to the survival beneﬁt seen to ovarian cancer patients in trials to date. The hazard ratio of death was 0.71 (95% CI, 0.44 to 1.14) derived from the systematic review and subsequent meta-analysis. The UK NICE interim appraisal as of August 2007 advised that the use of erythropoietins in cancer anaemia was not cost effective to its usual thresholds of £30,000/QALY.

Conclusions Formal health economic studies have been unable to demonstrate a signiﬁcant cost/QALY advantage for erythropoietins, yet other forms of economic analysis suggest that there may be signiﬁcant costs beneﬁts when the treatment is viewed from a family, employer or insurer standpoint (Lyman et al. 2005; Berndt et al. 2005). This is a recurring problem for supportive care treatments that are designed to improve only the quality of life, and not prolong it. Cost/QALY analysis was designed to prioritise resources when chosing life-extending treatments. A review has been reported of 110 interventions for cancer treatment or prevention in 39 articles for which both cost/life-year and cost/QALY were reported (Tengs 2004). Assuming a $50,000 USD decision threshold, adjustment for quality of life would affect the implied choice in 5% of cases. With a $400,000 USD threshold, adjustment for quality of life would affect choice for only 2% of interventions. In another study of 228 interventions, where both added years of life and QALY were estimated, quality-adjustment led to a ratio moving either above or below 50,000 US dollars/LY (or QALY) in 8% cases (Chapman et al. 2004). This suggests that the decisions made by cost/QALY analysis are weighted heavily towards prioritisation of treatments that extend life, over those that improve its quality. Although QALY analysis is designed to strike a balance between length of life and its enjoyment, in practice it seems to make little difference to reimbursement decisions.

842

P. Cornes

While the formal cost/QALY appraisal documents reviewed in this article cite frequently the need for higher quality research, and with better validated quality of life tools, there has been little recognition that the QALY model of evaluation may also need improvement. The impact on local decisionmakers has been that extra resources for cancer treatment are directed to treatments that offer marginal advantages to survival, with few resources left for short and longer-term supportive care targeted primarily on improving quality of life. Other economic tools, such as the concept of the “total cost of illness” suggest that supportive care may be signiﬁcantly undervalued (Bosanquet and Tolley 2003). All the formal economic appraisals of erythropoietin have suggested that there is a difﬁculty in interpreting quality of life data from the existing trials (Bottomley et al. 2003). The lack of accessible data prompted the Danish DACEHTA and USA Agency for Healthcare Research and Quality to state that accurate cost/QALY analysis was not possible. The Italian and UK studies have all suggested that cost/QALY estimates lie above the conventional thresholds of affordability of £30,000 GB Pounds, €80,000 Euro, $50,000 US Dollars. Subgroups such as patients treated with platinum chemotherapy and when supplemental iron is used have lower costs, but still above threshold. Ovarian cancer patients did appear to lie below the thresholds for reimbursement, but the increased cost effectiveness lay partly in a survival beneﬁt seen to patients randomised to chemotherapy with erythropoietin support. This data was based on meta-analysis of only 262 patients in 4 randomised controlled trials, and so the result had statistical uncertainty. It appears that emerging data from the most recent randomised trials may challenge this poor appraisal of the health economic beneﬁts. The following improvements in outcomes would all be expected to contribute to a reduced ICER, and greater cost effectivness then reported in the NICE appraisal • Response rates in trials with erythropoietin and iron supplementation now exceed 80%, wheras the NICE model assumed only 50% from the pooling of all the earlier studies in the meta-analysis when this aspect of treatment was not known (Demarteau et al. 2007). • Duration of treatment is typically 16–20 weeks of erythropoietin, wheras a longer 24 week course was expected in the model. • Duration of clinical beneﬁt after erythropoietin treatment ended has been recorded at 6 months, wheras a 12 week duration until anaemia and QOL beneﬁts normalised was expected in the study. Twelve weeks was an estimate as at the time of the appraisal as no published data was available. • Costs of erythropoietin were taken from list prices and national tarrifs, wheras price discounts are often available to bulk purchasers. In contrast to the rising costs of transfusion over time, the real costs of erythropoietin has been falling (Remak et al. 2003). True cost-effectiveness of a treatment

Cost-effectiveness of treating cancer anaemia

•

•

•

•

843

can change over time, and early or incomplete analysis can reach incorrect conclusions because of data deﬁciencies. Overall cost beneﬁts from treating anaemia have been demonstrated for the patient at home, and for employers and insurers, wheras the review assessed costs from a hospital perspective (Barnett et al. 2000, 2002). The utility of treating anaemia from a patient’s perspective may be higher than that of a panel of the public who may not appreciate the high symptom load of patients that is described as “fatigue”. A signiﬁcant difference of understating has been recorded between doctors and patients (Vogelzang et al. 1997). Using utility scores from cancer patients with anaemia would be likely to demonstrate a greater preference for treatment that the studies that relied on a surrogate panel of patients to chose. Quality of life scoring systems, with validated utility estimates are now available, such as the EQ-5D ﬁve question panel. Older trials have been criticised for failing to use such tools, but this is an evolving area of research and few consensus guidelines have been created. It was only in 2006 that the European network for Health Technology Assessment (EUnetHTA), was established. It coordinates the efforts of 27 European countries including 24 Member States of the European Union in evaluating health technology in Europe. It aims to develop practical tools to perform future studies. Subgroup analysis rather than overall meta-analysis has been advocated in several appraisals. Current clinical guidance, for example from the EORTC, advoctes starting treatment when the haemoglobin is in the range 9 to 11, and stopping treatment when normal levels have been achieved. Randomised trials using erythropoietins have explored very different aims, such as prevention of anaemia and attempting to create supra-normal Hb levels to augment radiotherapy response with increased doses of erythropoietin. These studies are outside both the current published guidelines for erythropoietins, and their product licences, and yet were analysed within the same meta-analysis. Whilst inclusive meta-analyses with the largest patient numbers are preferred, there has to be a balance taken when trials can be so heterogenous. No oncologist would suggest a meta-analysis where adjuvant tamoxifen was studied for both ER positive and negative breast cancer patients in the same study, without also performing the logical step of a subgroup analysis. The laboratory work showing a nonlinear relationship between Hb and tumour oxygenation where Hb levels above normal actually worsens tumour oxygenation suggests that excluding “off license” studies with high Hb targets has a valid scientiﬁc explanation (Vaupel et al. 2002). Using a pooled analysis of four separate randomised trials with darbepoietin, it has been possible to explore the relationship between tumour control and overall survival at different starting Hb levels, and with the magnitude of Hb response to treatment with active or placebo (Hedenus et al. 2005). Stratiﬁcation of patients

844

P. Cornes

randomized in future trials should make further analysis along these lines possible. • The UK NICE reviewers point out however that the clinical guidelines for the use of erythropoietins have evolved over time, and only a limited number of the completed and published studies followed strictly such current guidance as from the EORTC in 2006. These points, along with the reporting of newer trials mean that much work still needs to be done before the true cost-effectiveness of erythropoietins are understood. Trials that reﬂect the current clinical guidelines, with appropriate testing for case selection, iron correction, and validated quality of life and utility scoring tools will still be needed to satisfy formal cost/QALY reviewers. The “total cost of illness” model opens up another area for work, which may be increasingly relevant as we work to convert cancer from an acute life threatening illness into a manageable chronic disease. Each health system and country will need to make its own appraisal. Different countries will have different drug regulatory and pricing processes which will directly affect the cost/QALY estimates. Even with the same randomised trials analysed, the utility that a country might decide for a treatment may be based on different social and cultural issues that are not shared with another country, and hence attract different values. Where a country has few hospital facilities and staff to run them, then the costs of treatments that can avoid hospital admissions may be more acceptable. To outsiders, it is remarkable to observe that In 2003, the United States of America had fewer practicing physicians, practicing nurses, and acute care bed days per capita than the median country in the Organization for Economic Cooperation and Development (OECD) (Anderson et al. 2006). The United States is in the bottom quartile of hospital beds per capita (Reinhardt et al. 2004). In the USA, for example, spending on medicine to avoid hospital care might offer an extra health beneﬁt that altered the economic considerations about erythropoietins.

References 1. Agency for Healthcare Research and Quality, Rockville, MD: Calculating the U.S. Population-based EQ-5D Index Score. August 2005. http://www.ahrq.gov/ rice/EQ5Dscore.htm 2. Agrawal S, Davidson N, Walker M, Gibson S, Lim C, Morgan CL, Cowell W (2006) Assessing the total costs of blood delivery to hospital oncology and haematology patients. Curr Med Res Opin 22(10): 1903–1909 3. Allain J-P (2003) Transfusion risks of yesterday and of today. Transfus Clin Biol 10: 1–5 4. Ament AJ (1993) Allocation of scarce resources: the limitations of cost per Qaly. Abstr Int Soc Technol Assess Health Care Meet 9: 57

Cost-effectiveness of treating cancer anaemia

845

5. Anderson GF, Frogner BK, Johns RA, Reinhardt UE (2006) Health care spending and use of information technology in OECD countries. Health Affairs 25(89): 819–831 6. Appleby J (2001) Health insurance prices to soar. USA TODAY 08/27/2001 7. ASInsights. Erythropoietin: Current market dynamics and future outlook. Publisher: ASInsights Available for download as a PDF at http://www. pharmainstitute.com/coms2/summary_0283-1335_ITM (accessed 2007 may 8) 8. Barnett A, Birnbaum H, Cremieux PY, Fendrick AM, Slavin M (2000) The costs of cancer to a major employer in the United States: a case-control analysis. Am J Manag Care 6(11): 1243–1251 9. Barnett A, Cremieux PY, Fendrick AM, et al (2002) Anemia related costs for cancer patients. J Manag Care Med 6: 20–28 10. Barosi G, Marchetti M, Liberato NL (1998) Cost-effectiveness of recombinant human erythropoietin in the prevention of chemotherapy-induced anaemia. Br J Cancer 78(6): 781–787 11. Barrett-Lee PJ, Bailey NP, O’Brien ME, Wager E (2000) Large-scale UK audit of blood transfusion requirements and anaemia in patients receiving cytotoxic chemotherapy. Br J Cancer 82(1): 93–97 12. Bell CM, Urbach DR, Ray JG, Bayoumi A, Allison B Rosen AB, Greenberg D, Neumann PJ (2006) Bias in published cost effectiveness studies: systematic review. BMJ 332: 699–703 13. Berndt E, Crown W, Kallich J, Long S, Song X, Lyman GH (2005) The impact of anaemia and its treatment on employee disability and medical costs. Pharmacoeconomics 23(2): 183–192 14. Bohlius J, Wilson J, Seidenfeld J, Piper M, Schwarzer G, Sandercock J, Trelle S, Weingart O, Bayliss S, Djulbegovic B, Bennett CL, Langensiepen S, Hyde C, Engert A (2006) Recombinant Human Erythropoietins and Cancer Patients: Updated Meta-Analysis of 57 Studies Including 9353 Patients. JNCI Journal of the National Cancer Institute 98(10): 708–714 15. Bohlius J, Wilson J, Seidenfeld J, Piper M, Schwarzer G, Sandercock J, Trelle S, Weingart O, Bayliss S, Brunskill S, Djulbegovic B, Benett CL, Langensiepen S, Hyde C, Engert E (2006) Erythropoietin or darbepoetin for patients with cancer. Cochrane Database Syst Rev 3: CD003407 16. Bokemeyer C, Aapro MS, Courdi A, Foubert J, Link H, Osterborg A, Repetto L, Soubeyran P (2004) EORTC guidelines for the use of erythropoietic proteins in anaemic patients with cancer. Eur J Cancer 40(15): 2201–2216 17. Bokemeyer C, Aapro MS, Courdi A, Foubert J, Link H, Osterborg A, Repetto L, Soubeyran P (2004) EORTC guidelines for the use of erythropoietic proteins in anaemic patients with cancer. Eur J Cancer 40(15): 2201–2216 18. Bosanquet N, Tolley K (2003) Treatment of anaemia in cancer patients: implications for supportive care in the National Health Service Cancer Plan. Curr Med Res Opin 19(7): 643–650 19. Bottomley A, Thomas R, Van Steen K, Flechtner H, de Graeff A (2003) Guidelines for the use of Epoetin: Have Quality-of-life beneﬁts been proven? JCO Jun 1: 2223 20. Busschbach JJV, Hessing DJ, de Charro FT (1993) The utility of health at different stages in life: A quantitative approach, Social Science & Medicine 37(2): 153–158

846

P. Cornes

21. Cancer Care Ontario Practice Guidelines Initiative (The Role of Erythropoietin in the Management of Cancer Patients with Non-Hematologic Malignancies Receiving Chemotherapy (Practice Guideline #12-1) (Version 2.2003) 22. Danish Centre for Evaluation and Health Technology. Epoetin (EPO) for Anaemic Cancer Patients – A Rapid Assessment. Health Technology Assessment 2004; 6(1) (available online at www.sst.dk/publ/Publ2004/CEMTV_eng/epo_ 1805.pdf 23. Cavill I (2002) Reducing blood transfusion. Focus should be on improving patients’ ability to make own blood. BMJ 21;325(7365): 655 24. Chapman RH, Berger M, Weinstein MC, Weeks JC, Goldie S, Neumann PJ (2004) When does quality-adjusting life-years matter in cost-effectiveness analysis? Health Econ 13(5): 429–436 25. Chaves PHM, Mody SH, Najib MM, Siegartel LR, Blasi MV, Woodman RC (2003) Anemia in the elderly is associated with increased healthcare costs and resource utilization. Blood 102: abstract 2769 26. Clark O (2002) Erythropoietin, uncertainty principle and cancer related anaemia. BMC Cancer 24;2(1): 23 27. Cremieux PY, Barrett B, Anderson K, Slavin MB (2000) Cost of outpatient blood transfusion in cancer patients. J Clin Oncol 18(14): 2755–2761 28. Demarteau N, Annemans L, Mossman T, Bracco A (2007) Cost-effectiveness of darbepoetin alfa (DA) 500*T*mcg every three weeks (Q3W) with IV iron compared to DA Q3W alone in cancer patients (pts) with chemotherapy-induced anaemia (CIA). J Clin Oncol 2007 ASCO Annual Meeting Proceedings (PostMeeting Edition) 25 (18S) (June 20 Supplement): 19531 29. Devlin N, Parkin D (2004) Does NICE have a cost-effectiveness threshold and what other factors inﬂuence its decisions? A binary choice analysis. Health Economics 13(5): 437–452 30. Dewilde S, Turk F, Tambour M, Sandström T (2006) The economic value of antiIgE in severe persistent, IgE-mediated (allergic) asthma patients: adaptation of INNOVATE to Sweden. Curr Med Res Opin 22(9): 1765–1776. 31. Donaldson L (2004) On the state of the public health: Annual report of the Chief Medical Ofﬁcer 2003. UK Department of Health, 28 July 2004 32. Drummond MF, O’Brien B, Stoddart GL, Torrance GW (1997) Methods for the Economic Evaluation of Health Care Programmes. 2. Oxford Medical Publications, Oxford 33. Eichler H-G, Kong SX, Gerth WC, Mavros P, Jönsson P (2004) Use of costeffectiveness analysis in health-care resource allocation decision-making: how are cost-effectiveness thresholds expected to emerge? Value in Health 7(5): 518–528 34. Erickson P, Wilson R, Shannon I (1995) Years of healthy life. Healthy People 2000 Statistical Notes. Atlanta (GA): Centers for Disease Control and Prevention 35. Evans C, Tavakoli M, Crawford B (2004) Use of quality adjusted life years and life years gained as benchmarks in economic evaluations: a critical appraisal. Health Care Manage Sci 7: 43–49 36. Forbes JM, Anderson MD, Anderson GF, Bleecker GC, Rossi EC, Moss GS (1991) Blood transfusion costs: a multicenter study. Transfusion 31(4): 318–323 37. Greenhalgh T (1997) How to read a paper: papers that tell you what things cost (economic analyses). BMJ 315: 596–599

Cost-effectiveness of treating cancer anaemia

847

38. Gustafson DH, Helstad CP, Hung CF, Nelson G, Batalden P (1995) The total costs of illness: a metric for health care reform. Hosp Health Serv Adm 40(1): 154–171 39. Hedenus M, Vansteenkiste J, Kotasek D, Austin M, Amado RG (2005) Darbepoetin alfa for the treatment of chemotherapy-induced anemia: disease progression and survival analysis from four randomized, double-blind, placebocontrolled trials. J Clin Oncol 23(28): 6941–6948 40. Hirth RA, Chernew ME, Miller E, Fendrick AM, Weissert WG (2000) Willingness to pay for a quality-adjusted life year: in search of a standard. Med Decis Making 20(3): 332–342 41. Hirth RA, Chernew ME, Miller E, Fendrick AM, Weissert WG (2000) Willingness to pay for a quality-adjusted life year: in search of a standard. Med Decis Making 20: 332–342 42. Hogerzeil HV (2004) BMJ 329: 1169 43. Johnson JA, et al (1998) Valuation of EuroQOL (EQ-5D) health states in an adult US sample. Pharmacoeconomics 13: 421–433 44. Johnson JA, Coons SJ (1998) Comparison of the EQ-5D and SF-12 in an adult US sample Quality of Life Research 7(2): 155–166 45. Kapp KS, Poschauko J, Geyer E, Berghold A, Oechs AC, Petru E, Lahousen M, Kapp DS (2002) Evaluation of the effect of routine packed red blood cell transfusion in anemic cervix cancer patients treated with radical radiotherapy. Int J Radiat Oncol Biol Phys 54(1): 58–66 46. Kassirer JP, Angell M (1994) The journal’s policy on cost-effectiveness analyses. N Engl J Med 331: 669–670 47. Kind P (1996) The EuroQol instrument: An index of health-related quality of life. In: Spilker B (ed) Quality of life and pharmacoeconomics in clinical trials, 2nd edition. Lippincott-Raven Publishers, Philadelphia, pp 191–201 48. Kushner JP, Porter JP, Olivieri NF (2001) Secondary iron overload. American Society of Hematology Education Program Book 2001, 1: 47–61 49. Latief K (2004) Cancer 2025 report, in Cancer Services Insight. 2004 Autumn, p8 50. Littlewood T, Zambrowski J-J, Cornes P (2006) Curent Medical Research & Opinion 22(s4): s23–s34 51. Ludlam CA, Turner ML (2006) Managing the risk of transmission of variant Creutzfeldt Jakob disease by blood products. Br J Haematol 132: 13–24 52. Ludwig H, Van Belle S, Barrett-Lee P, Birgegard G, Bokemeyer C, Gascon P, Kosmidis P, Krzakowski M, Nortier J, Olmi P, Schneider M, Schrijvers D (2004) The European Cancer Anaemia Survey (ECAS): a large, multinational, prospective survey deﬁning the prevalence, incidence, and treatment of anaemia in cancer patients. Eur J Cancer 40: 2293–2306 53. Lurie P, Almeida CM, Stine N, Stine AR, Wolfe SM (2006) Financial conﬂict of interest disclosure and voting patterns at Food and Drug Administration Drug Advisory Committee meetings. JAMA 295(16): 1921–1928 54. Lyman GH, Berndt ER, Kalich JD, et al (2005) The economic burden of anemia in cancer patients receiving chemotherapy. Value Health 8: 149–156 55. Martin S, Rice N, Smith PC (2007) The link between healthcare spending and health outcomes Evidence from English programme budgeting data. Centre for Health Economics, University of York, Published by: The Health Foundation, 90 Long Acre, London WC2E 9RA, June 2007

848

P. Cornes

56. McGregor M (2003) Cost-utility analysis: use QALYs only with great caution. CMAJ 168(4): 433–434 57. Miners AH, Garau M, Fidan D, Fischer AJ (2005) Comparing estimates of cost effectiveness submitted to the National Institute for Clinical Excellence (NICE) by different organisations: retrospective study. BMJ 330: 65 58. Mohandas K, Aledort L (1995) Transfusion requirements, risks, and costs for patients with malignancy. Transfusion 35: 427–430 59. Mortimer PP (2002) Making blood safer. BMJ 325: 400–401 60. NICE. Anaemia (cancer-treatment induced) – erythropoietin (alpha and beta) and darbepoetin: Appraisal consultation document. 29 Jun 2007 [web ref http://guidance.nice.org.uk/page.aspx?o=439014] 61. Nissenson AR, Wade S, Goodnough T, Knight K, Dubois RW J (2005) Economic burden of anemia in an insured population. Manag Care Pharm 11(7): 565–574 62. Österborg A (1998) Recombinant human erythropoietin (rHuEPO) therapy in patients with cancer-related anaemia: what have we learned? Med Oncol 15[Suppl 1]: S47–S49 63. Ortega A, Dranitsaris G, Puodziunas AL (1998) What are cancer patients willing to pay for prophylactic epoetin alfa? A cost-beneﬁt analysis. Cancer 83(12): 2588–2596 64. Penninx BW, Pahor M, Woodman RC, Ferrucci L, Guralnik JM (2003) Late-life anemia identiﬁes persons at risk for mortality and hospitalization. Blood 102: abstract 881 65. Pirker R, Wiesenberger K, Pohl G, Minar W (2003) Anemia in lung cancer: clinical impact and management. Clin Lung Cancer 5: 90–97 66. Provan D (1999) Better blood transfusion. BMJ 318: 1435–1436 67. Reinhardt UE, Hussey PS, Anderson GF (2004) U.S. Health Care Spending In An International Context. Health Affairs 23(3): 10–25 68. Reinhardt UE, Hussey PS, Anderson GF, U.S. (2004) Health care spending in an international context. Health Affairs 2(3): 10–25 69. Remak E, Hutton J, Jones M, Zagari M (2003) Changes in cost-effectiveness over time. The case of Epoetin Alfa for renal replacement therapy patients in the UK. Eur J Health Econ 4(2): 115–121 70. Sandler SG, Yu H, Rassai N (2003) Risks of blood transfusion and their prevention. Clin Adv Hematol Oncol 1: 307–313 71. Seidenfeld J, Piper M, Flamm C, Hasselblad V, Armitage JO, Bennett CL, Gordon MS, Lichtin AE, Wade JL 3rd, Woolf S, Aronson N (2001) Epoetin treatment of anemia associated with cancer therapy: a systematic review and metaanalysis of controlled clinical trials. J Natl Cancer Inst 93(16): 1204–1214 72. Shefﬁeld R, Sullivan SD, Saltiel E, Nishimura L (1997) Cost comparison of recombinant human erythropoietin and blood transfusion in cancer chemotherapy-induced anemia. Ann Pharmacother 31(1): 15–22 73. Sheldon T (2006) Dutch consider excluding costly treatments from health insurance. BMJ 333: 113 (Zinnige en duurzaame zorg (Sensible and Lasting Care), is available in Dutch at www.rvz.net) 74. Straus DJ, Testa M, Riggs SA, Tulpule A, Sarokhan B (2003) Early Treatment with Epoetin Alfa Improves Anemia, Quality of Life (QOL), and Productivity in Patients (Pts) with Hematologic Malignancies and Mild Anemia during Chemotherapy (CT). Blood 102(11): A1811

Cost-effectiveness of treating cancer anaemia

849

75. Straus DJ, Testa MA, Sarokhan BJ, Czuczman MS, Tulpule A, Turner RR, Riggs SA (2006) Quality-of-life and health beneﬁts of early treatment of mild anemia: a randomized trial of epoetin alfa in patients receiving chemotherapy for hematologic malignancies. Oc 107(8): 1909–1917 76. Tanne JH (2006) Number of uninsured middle class US citizens grows. BMJ 332: 1047 77. Tengs TO (2004) Cost-effectiveness versus cost-utility analysis of interventions for cancer: does adjusting for health-related quality of life really matter? Value Health 7(1): 70–78 78. Tuffs A (2006) German law to cuts drug costs angers doctors. BMJ 332: 1051 79. Ubel PA (2003) What is the price of life and why doesn’t it increase at the rate of inﬂation? Arch Intern Med (163): 1640–1641 80. Vaupel P, Thews O, Mayer A, Höckel S, Höckel M (2002) Oxygenation status of gynecologic tumors: what is the optimal hemoglobin level? Strahlenther Onkol 178(12): 727–731 81. Vogelzang NJ, et al (1997) Semin Hematol 34 [Suppl 2]: 4–12 82. Wang AJ (1998) Key concepts in evaluating outcomes of ATP funding of medical technologies. J Technol Transfer 23(2): 61–65. 83. Wilking N, Jönsson B (2005) A pan-European comparison regarding patient access to cancer drugs. Stockholm, Sweden, Karolinska Institutet in collaboration with Stockholm School of Economics, 2005 84. Wilking N, Jonsson B (2005) A Pan European comparison regarding patient access to cancer drug, Karolinska Institute and Stockholm School of Economics, Stockholm, Sweden 2005, p12 85. Wilson J, Yao GL, Raftery J, Bohlius J, Brunskill S, Sandercock J, Bayliss S, Moss P, Stanworth S, Hyde C (2007) A systematic review and economic evaluation of epoetin alfa, epoetin beta and darbepoetin alfa in anaemia associated with cancer, especially that attributable to cancer treatment. Health Technol Assess 11(13): 1–220 86. Wilson J, Yao GL, Raftery J, Bohlius J, Brunskill S, Sandercock J, Bayliss S, Moss P, Stanworth S, Hyde C (2007) A systematic review and economic evaluation of epoetin alfa, epoetin beta and darbepoetin alfa in anaemia associated with cancer, especially that attributable to cancer treatment. Health Technol Assess 11(13): 1–220 Correspondence: Dr. Paul Cornes, Bristol Haematology and Oncology Centre, Horﬁeld Road, Bristol, BS2 8ED, United Kingdom, E-mail: [email protected]

Addendum 1

Serum EPO level in healthy persons and cancer patients Study

Study subjects

Nowrousian et al. 1996 Chiesa and Fink 2000 Cheung et al. 2001

Healthy adults, (mean ± SD), solid-phase enzyme immunoassay Healthy adults, (median, reference range), enzymelinked immunoassay Healthy adults Epoetin alfa 150 U/kg sc tiw, Cmax (mean ± SD) Epoetin alfa 40,000 U once a week, Cmax (mean ± SD) Anemic patients without treatment Chronic myeloproliferative diseases, (median, range) Chronic lymphocytic leukemia, (median, range) Multiple myeloma, (median, range) Malignant lymphomas, (median, range) Solid tumors, (median, range) Anemic patients receiving Ch/ChP Baseline, breast cancer, (median, range) Baseline, gynecologic tumors, (median, range) Baseline, lung cancer, (median, range) Anemic patients with solid tumors and ChP + rhEPO Epoetin alfa 150 U/kg sc tiw Baseline, responding to epoetin alfa, (mean, range) Baseline, nonresponding to epoetin alfa, (mean, range) Week 2 of treatment, responding to epoetin alfa, (mean, range) Week 2 of treatment, nonresponding to epoetin alfa, (mean, range) Week 4 of treatment, responding to epoetin alfa, (mean, range) Week 4 of treatment, nonresponding to epoetin alfa, (mean, range) Anemic lung cancer patients receiving ChP + rhEPO Epoetin beta 9000 U sc once a week, Cmax (mean + SD) Epoetin beta 18,000 U sc once a week, Cmax (mean + SD) Epoetin beta 36,000 U sc once a week, Cmax (mean + SD)

Nowrousian et al. 1996

Demetri et al. 1998

GonzalezBaron et al. 2002

Fujisaka et al. 2006

Number

Serum EPO level (mU/ml)

99

4.4 ± 2.9

40

18, 12–50

17 17

143 ± 54.2 861 ± 445.1

22 36 94 61 124

72, 1–10,000 87, 7–7,078 42, 1–3,244 39, 1–4,000 26, 2–5,110

382 297 545

44, 5–1,124 36, 2–701 57, 0–1,557

57 33

69.1, 10.6–425 84.0, 3.61–506

57

208.7, 33–567

33

204.5, 16.26–999.8

57

211.4, 14.81–620.9

33

261.8, 32–999.1

3

308 ± 117

6

678 ± 86.7

6

1316 ± 766

rhEPO = recombinant human erythropoietin; sc = subcutaneous; tiw = three times a week; Cmax = maximal concentration achieved in serum; Ch = nonplatinum-based chemotherapy; ChP = platinumbased chemotherapy

852

Addendum 1

References 1. Cheung W, Minton N, Gunawardena K (2001) Pharmacokinetics and pharmacodynamics of epoetin alfa once weekly and three times weekly. Eur J Clin Pharmacol 57: 411–418 2. Chiesa ME, Fink NE (2000) Reference intervals for erythropoietin measurement by the JCL EIA assay need to be determined locally. Clin Chem Lab Med 38: 65–66 3. Demetri GD, Kris M, Wade J, et al (1998) Quality-of-life beneﬁt in chemotherapy patients treated with epoetin alfa is independent of disease response or tumor type: results from a prospective community oncology study. Procrit Study Group. J Clin Oncol 16: 3412–3425 4. González-Barón M, Ordón˜ez A, Ranquesa R, et al (2002) Response predicting factors to recombinant human erythropoietin in cancer patients undergoing platinum-based chemotherapy. Cancer 95: 2408–2413 5. Fuijsaka Y, Tamura T, Ohe Y, et al (2006) Pharmacokinetics and pharmacodynamics of weekly epoetin beta in lung cancer patients. Jpn J Clin Oncol 36: 477–482 6. Nowrousian MR, Kasper C, Oberhoff C, et al (1996) Pathophysiology of cancer related anemia. In: Smyth JF, Boogaerts MA, Ehmer BRM (eds) rhErythropoietin in cancer supportive treatment. Marcel Dekker, New York Basel Hong Kong, pp 13–34

Index

2,3-diphosphoglycerate (2,3-DPG) 317, 336 5-ﬂuorouracil 451, 763 5q- syndrome 531 8-OHdG (a marker of oxidative damage) 703 ABVD 210 acidosis 289–290, 292 acquired treatment resistance 290 actinomycin D 451 activator protein-1 289 acute iron toxicity 692 acute lymphocytic leukemia (ALL) 642–643, 652, 655 acute myelogenous leukemia 642 acute myeloid leukemia (AML) 340, 531 acute normovolemic hemodilution 665 ADAMTS13 130 administration of rhEPO to donors 604 adrenergic activity 713 adverse effects of IV iron therapy 477 adverse prognostic ﬁnding 759 afﬁnity of Hb for oxygen 317 aggressive tumor subclones 292 AKT 47 allergic reactions 707 allogeneic blood exposure 665 alloimmunization 338 Alzheimer’s disease 782 amenorrhea 311

American Society of Clinical Oncology (ASCO) 509, 738 American Society of Hematology (ASH) 509, 738 aminolevulinic acid 295 anaemia and poor outcome 761 anaemia correction 766 analogues 410 anaphylactic reactions 707 anemia 265, 285, 288, 291, 293, 296 anemia correction 267 anemia in cancer 123, 189 anemia in elderly cancer patients 346, 380 anemia in patients receiving chemotherapy 212 anemia model 267 anemia of aging 383 anemia of chronic disease (ACD) 149, 382, 433, 551, 688 anemia-inducing substance (AIS) 161 anemia-related quality of life 370 anemia-related symptoms 369 anemic hypoxia 275 angina pectoris 308, 322 angiogenesis 297 anorexia 479 anoxia 284 antibody C-20 107 antibody techniques used for detecting EPO-R 104 anticancer agents 294 anticoagulation 753

854 anti-EPO antibodies 418 antioxidant effects of erythropoietin 804 antithymoglobulin (ATG) 533 AP-1 20 apoptosis 290 apoptotic potential 293–294 arterial remodeling 325 ascorbic acid (vitamin C) 478 association between anemia and QOL 384 association between iron overload and infections 340 association between RBC transfusions and adverse clinical outcomes 338 asthenia 479 ATP depletion 284 autoimmune diseases 338 autologous blood donation 680 autologous blood phlebotomy 680 autologous transfusions 393 baseline Hb level 200 basic helix-loop-helix (bHLH) 18 Bcl-2 23 Bcl-XL 23, 25 BCNU 762 BEACOPP 210, 214 BFU-E (burst-forming unit-erythroid) 2, 5, 171, 534, 602 BH3-only pro-apoptotic proteins 25 bioavailability 414 biochemical markers of storage iron 679, 685 bladder cancer 211, 257 bleomycin 451 blood ﬂow 295 blood loss 124 blood pressure 418 blood transfusion 272 blood transfusion thresholds 830 blood transition 267 blood viscosity 270–271 blood-brain-barrier 776 body composition 715 body weight 719 bone marrow erythropoiesis 2 bone marrow inﬁltration 124

Index bone pain 479 brain, cardiovascular tissues 412 brain endothelial EPO-R 778 brain or spinal cord hypoxia/ischemia 776 brain or spinal cord trauma 776 brain tumors 284 breast cancer 196, 211, 268, 284 breathlessness 190 C-20 (sc-695) rabbit polyclonal antipeptide anti-human EPO-R antibody 106 CA IX 293 cancer-associated anemia 81 cancer cachexia 725 cancer linear analog scale (CLAS) 371 cancer of head and neck 265 cancer of the uterine cervix 269, 284 cancer of the vulva 272 cancer recurrence 338, 666 cancer-related anemia (CRA) 131, 149 cancer-related fatigue 313, 369–370 carbon dioxide production 719 carbon monoxide 293 carboplatin 212, 215, 451, 762 carboxyhemoglobin 285 carcinoma of the cervix 394 cardiac failure 310, 318, 325 cardiac output 320 cardiac remodeling 325–326, 341 cardiac remodeling: left ventricular hypertrophy 318 cardiomegaly 324 cardiotoxicity of doxorubicin 459 cardiovascular consequences of chronic anemia 318 cardiovascular system 318 caspase activation 43 caspases in 42 catheter-associated clots 747 catheter-associated thrombosis 754 CD34+ cells 27, 602 CD-62 expression 750 CEBOPP 210 cell cycle effects 290 cell cycle position 289

Index cell proliferation 294 cellular response to hypoxia 69 CERA 408, 421 cerebral blood ﬂow 343 cerebrospinal ﬂuid (CSF) 777 cervical cancer 211, 257 cervical carcinoma 272 cervix cancer 265, 285 CFU-E (colony-forming unit-erythroid) 2, 5, 67, 167, 171, 411, 534 CFU-GM 603 change in hemoglobin level and erythropoietin response 680 changes in gene expression proteome changes 291 chemoresistance 294, 762 chemosensitivity 295 chemotherapy 190, 289, 291, 294, 296, 417 chemotherapy cycles 195 chemotherapy induced anemia 459, 645 CHOEP 210 CHOEP-14 214 CHOEP-21 214 CHOP 210, 213 CHOP-14 214 CHOP-21 214 CHr 689 chromosomal rearrangements 292 chronic brain disease 780 chronic heart failure 341 chronic lymphocytic leukemia (CLL) 198, 157, 173, 211, 433 chronic myelocytic leukemia 211, 223 chronic myelomonocytic leukemia (CMML) 532 chronic myeloproliferative diseases 157, 173 chronic obstructive pulmonary disease (COPD) 342 cisplatin 212, 215, 762 clinical ﬁndings in anemia 323 clinical manifestations of anemia 308 clonal dominance 292 clonal heterogeneity 290 clonal selection 290, 292, 297 CNS tumors 649

855

cognitive abnormalities 343 cognitive dysfunction 345, 637 cognitive effects of EPO 781 cognitive event-related potential (ERP) 327 cognitive function 327 cognitive performance 343 cold agglutinins 127 cold intolerance 311 colony-stimulating factors (CSFs) 534 colorectal cancer 211 combined modality regimens 253 commercially available anti-EPO-R antibodies 106 comparative effects of rhEPO in autologous and allogeneic transplantation 598 comparison of epoetin and RBT 403 compensatory mechanisms in anemia 317 conditions illustrating functional iron deﬁciency 551 congestive heart failure (CHF) 326, 345, 793 constipation 479 COPP 210 COPP/ABVD 214 coronary artery disease 342, 345, 801 coronary artery ﬂow 342 correction of functional iron deﬁciency by intravenous iron 553 correlation between fatigue and hemoglobin level 350 correlation between hemoglobin and quality of life (QOL) in cancer patients receiving chemotherapy 349 cost categories included in costeffectiveness studies 821 cost effectiveness of erthropoietins 813 cost of not treating anaemia in cancer 818 cost per QALY 823, 827 cost/QALY health economic analyses 833 cost-beneﬁt analysis (CBA) 826 cost-effectiveness analysis (CEA) 822

856

Index

cost-effectiveness of treating cancer anaemia 813 costs associated with anemia 511 costs for cancer in EU countries 815 costs of transfusion 832 cough 479 counterfeit rhEPO preparations 419 C-reactive protein (CRP) 554, 753 creatinine clearance (CCr) 797 criteria of response 542 cumulative effect of chemotherapy 203 cyclic changes in tumor oxygenation 348 cyclophosphamide 213, 762 cyclosporine A 533 cytochrome c 535 cytokines 295 cytotoxic chemotherapy 190 cytotoxic drugs 212, 294 damage of EPO producing proximal renal tubular cells 215 darbepoetin 267, 295, 408 darbepoetin alfa 410, 415, 439, 452, 655 decrease in Hb level during chemotherapy 219 decreased exercise capacity 329 decreased libido and fertility 331 decreased muscle strength 329 decreased peripheral vascular resistance 318 decreased work tolerance 308 defective endogenous EPO production 334 deﬁcient EPO production 215 delayed intervention 523 deoxyribonuclease (DNase) IIα 46 depression 308, 322, 730 detection of EPO-R 104 detection of iron-restricted erythropoiesis during EPO therapy 688 diabetes mellitus 346, 776 diarrhea 479 differentiation 2, 290 diffusion geometry 296 diffusion-limited hypoxia 285

diseases of the brain 772 disseminated intravascular coagulopathy (DIC) 129 divalent metal transporter 1 (DMT1) 165 dizziness 308, 310, 479, 637, 730 DNA damage 291 DNA repair enzymes 289, 294 dosage of rhEPO 642 dose of epoetin 440 dose-response relation, response rate 454 dose-response relationship between erythropoietin and red blood cell expansion 681 dose-response relationship of darbepoetin alfa 458 dosing 416 doxorubicin 451 drug delivery 295 duodenal cytochrome b (Dcytb) 165 dyspnea 309–310, 322, 479, 637, 730 early initiation of ESAs 519 early intervention 738 early intervention versus delayed treatment 514 early intervention versus no treatment 514 echocardiography 324 edema 310, 479, 796 effect of allogenic transfusion on cancer recurrence 338 effect of rhEPO on survival 759 effects of EPO on cognitive performance 781 effects of EPO on various tissues and organ systems 328 effects of erythropoietin on tumor growth and survival 107 effects of ESAs on thrombosis risk in cancer 747 effects of rhEPO on quality of life 729 effects of treatment with ESAs on survival 481 EKLF 19 elderly patients with cancer 345 elimination of EPO 416

Index embryonic erythropoiesis 1 endogenous erythropoietin production after allogenic peripheral blood hematopoietic stem cell transplantation (HSCT) 593 endometrial carcinoma 211 energy expenditure 713, 719 energy metabolism 715 EORTC 118, 739 EORTC guidelines 400–401, 480 EPO degradation 413 EPO gene transfer 420 EPO in the CSF 777 EPO in the nervous system 775 EPO levels 6 EPO mimetics 420 EPO production 133, 172, 68 EPO receptor (EPO-R) expression on tumor cells 69, 82, 88, 103, 411, 412, 481, 773, 774 EPO receptor mRNA 82 EPO receptor status 82 EPO response 172, 690 EPO signalling 412 EPO therapy 682–683, 691 EPO/EPO-R complex 74 EPO/EPO-receptor complex 412 EPO-induced intracellular signalling 70 EPO-R 12, 14, 67 EPO-R expression 481 EPO-R mRNA 104 epoetin 408 epoetin alfa 409, 438, 452 epoetin beta 409, 436, 452 epoetin omega 409 EQ-5D 824 erectile problems in men 331 ERK1, ERK2 27 erythrocyte and reticulocyte indices 679 erythroid differentiation 3 erythroid precursors 3, 5 erythropoiesis 1 erythropoiesis during blood loss and erythropoietin (EPO) therapy 682 erythropoiesis stimulating agents (ESAs) 745

857

erythropoietic parameters 559 erythropoietic proteins 734 erythropoietic responses 681 erythropoietin (EPO) and erythropoiesis in patients with anemia* of chronic disease 683 erythropoietin 81, 267, 272, 293, 385 erythropoietin assay 690 erythropoietin doses used in preclinical studies 107 erythropoietin during chemotherapy 820 erythropoietin levels 690 ESA therapy and thrombosis 749 ESA-induced thrombosis 750 esophageal cancer 211 etoposide 212, 451 European Cancer Anemia Survey (ECAS) 194, 212, 396 European Medicines Agency (EMA) 813 European Organisation for Research and Treatment of Cancer (EORTC) 465 Ewing’s sarcoma 211, 224, 648–649 exercise capacity 715, 719 exercise performance 721 exercise testing 715, 719 facial ﬂushing 479 FACIT-An 372 FACIT-Anemia 372 FACIT-Fatigue (or FACIT-F) 372 FACT-An 373, 374, 380, 441, 733 FACT-Anemia 313, 385, 736 FACT-F 373, 733 FACT-Fatigue 372–374, 380, 385, 441, 733, 736–737 FACT-G 380, 441 factors inﬂuencing response to EPO 542 factors potentially limiting response to rhEPO 543 faintness 308, 637 Fas 36 Fas; TRAIL 31 Fas-L 36, 169 Fas-L/TRAIL system 170

858

Index

fatigue 308–309, 312, 322, 327, 343, 345, 369, 479, 510, 636–637, 654, 730 fatigue questionnaires 371 ferric gluconate 692, 708 ferritin 161, 470, 688 Ferrlecit 692 ferroportin 161–162 fever 479 FIGO stage 287 ﬁxed weekly dosages of rhEPO 457 ﬂuctuations in Hb level 347 ﬂuid retention 795 ﬂulike syndrome 479 FOG-1 14 folate deﬁciencies 122 follicle-stimulating hormone (FSH) 331 Food and Drug Administration (FDA) 480 formation of free radicals 704 free radicals 290, 294, 760 frequency and severity of anemia associated with chemotherapy 209 frequency and severity of anemia associated with cytotoxic agents 208 frequency of anemia 252, 760 functional ability 719 functional ability of RBCs 336 Functional Assessment of Cancer Therapy (FACT) 372 Functional Assessment of Cancer Therapy-General (FACT-G) 313, 372 Functional Assessment of Chronic Illness Therapy – Anemia Scale (FACIT-An) 371, 372 functional dependence 345 functional iron deﬁciency 471, 549, 680 functional value of RBC units 338 G-CSF + EPO 595, 604 gastric cancer 211 gastrointestinal cancer 715 GATA target genes 16 GATA-1 14–15, 40, 169 GATA-2 11 GATA-4 12 gemcitabine 213, 451

gene ampliﬁcation 292 gene expression 289, 297 general Functional Assessment of Cancer Therapy (FACT-G) 732 genetic engineering 420 genome changes 290–292, 297 genomic instability 291–292 German Multicenter EPO Stroke Trial 779 Gﬁ-B 20 glucose depletion 292 GLUT-1 289–290, 293 glutathione 289, 294 glycolysis 289–290 glycosylation sites 408 GM-/G-CSF + EPO 604 grading systems for anemia 118 graft-versus-host disease (GVHD) 585 granulocyte colony-stimulating factor (G-CSF) 535 granulocyte/macrophage colonystimulating factor (GM-CSF) 535, 594 growth factors 289, 291 guidelines for the use of the erythropoietic stimulating agents (ESAs) 509 gynecologic cancer 194, 196 haemoglobin level 764 hairy cell leukemia 211, 223 half-life 409 Hb content of reticulocytes (CHr) 470 Hb levels 118, 597 Hb levels and survival 255 Hb levels during radiotherapy 258 Hb response 545 head and neck cancer 194, 200, 211, 272, 284, 394 headache 308, 310, 322, 637 health economic study 813 health-related quality of life (HRQL) 510, 720 heart failure 795 heat shock protein (HSP)70–2 107, 293 hematocrit (Hct) 117, 267 hematocrit, optimal 271

Index hematologic malignancies 651 hematological features of CRA 157 hematological malignancies 340 hematopoiesis 11 hematopoietic response 545 hematopoietic stem cell transplantation (HSCT) 583 hematoporphyrin 295 hemoconcentration 271 hemoglobin (Hb) 117 hemoglobin concentration 266, 286, 295 hemoglobin level 266, 293 hemoglobin, optimal level 271 hemoglobin optimum 274 hemoglobin prognostic value 266 hemoglobin response 517 hemoglobin-oxygen dissociation curve 317 hemolysis 125, 134 hemolytic uremic syndrome (HUS) 129 hemosiderosis 340 hemostasis 419 heparin 753 hepcidin 155, 162, 164–165, 167, 171 HIFα 10 HIF-α 10 HIF-1 69, 289 HIF-1α 68, 293, 774 HIF-1βb 68 high-molecular-weight iron dextran (DexFerrum®) 477 high-risk MDS 531 Hodgkin lymphoma 649 Hodgkin’s disease (HD) 197, 211, 221 homologous transfusions 393 hospitalization 800 HRQOL outcomes 517 human hematopoiesis 2 human stroke 780 hypercoaguable state in cancer 745 hyperlipidemia 329 hypertension 479 hyperviscosity 271 hypochromic RBCs 470, 689 hypo-responsiveness to EPO therapy 685

hypothalamic-pituitary-gonadal axis 331 hypoxia 8, 259, 275, 296, 417, 760 hypoxia factor 1α 8 hypoxia in tumors 249 hypoxia, acute 285 hypoxia, deﬁnition 283 hypoxia, diffusion limited 285 hypoxia, enemic 285 hypoxia, hypoxemic 285 hypoxia, ischemic 285 hypoxia, perfusion-limited 285 hypoxia-induced drug resistance 452 hypoxic cells 249 hypoxic conditions 451 hypoxic environment 266 hypoxic fraction 267, 269, 272–273 hypoxic stress proteins 294 hypoxic thresholds 284 hypoxic tissue fraction 275 hypoxic tumours 760 i.v. ferric gluconate 707 IFN-γ 36, 42, 152, 161, 167, 559, 586 ifosfamide 213 IL-1 153, 160, 163, 167, 175, 559, 586 IL-6 149, 153–154, 163–165, 586 immune hemolysis 125, 134 immune system and other defence mechanisms 332 immunogenicity 418 immunohistochemistry 104 immunosuppression 434 immunotherapy 290, 295 impact of anemia on organ function 322, 324, 330, 341 impact of anemia on outcome 257 impact of Hb on tumor oxygenation 255 impact of pretreatment Hb levels on radiotherapeutic results 256 impact of treatment with rhEPO on survival 461 impaired cognitive function 730 improvement in QOL 464

859

860

Index

inactivity 637 incidence of anemia 195, 203, 450 incidence of thrombosis associated with ESA therapy 747 increased cardiac output 318, 320 increased mortality related to anemia in various types of cancer 226 increased risk of infections 340 incremental beneﬁt in QOL 464 incremental cost-effectiveness ratio (ICER) 824 INFeD® 692 initial Hb 201 insulin resistance 329 interaction between anemia and other pathophysiologic or disabling processes 341 interferons (IFNs) 149 interleukin 6 (IL6) 796 interleukin-1 beta (IL-1βb) 149 the International Prognostic Scoring System (IPSS) 532 intratumor variability 287 intravenous iron 671 intravenous iron administration 475, 685 intravenous iron administration in patients receiving EPO 686 intravenous iron in cancer patients 473 intravenous iron supplementation 682 intravenous iron therapy 685, 692 intravenous iron therapy for anemia 686 ionizing radiation 296 irinotecan 213 iron availability 708 iron deﬁciency (ID) 120, 687, 803 iron dextran (INFeD® and DexFerrum®), iron gluconate (Ferrlecit®) 475, 692, 707, 708 iron formulations for IV iron therapy 475 iron gluconate 707 iron metabolism 11, 551 iron overload 339 iron potential risks 704 iron regulatory protein 1 705

iron risks 704 iron saccharate 692 iron sucrose (saccharate) (Venofer®) 475, 708 iron supplementation 440, 469, 642 iron supplementation during rhEPO administration 671 iron, transferrin 688 iron-restricted erythropoiesis 475, 682, 685 irradiation 268 irregular menstrual cycles 311 irritability 637 ischemic hypoxia 285 IV iron 473, 669 IV iron supplementation 475 Jehovah’s Witnesses 653 kB nuclear factor (NFkB) 12 kidney function 6, 8, 345 laboratory ﬁndings in functional ID 550 lactate accumulation 292 lean body mass (LBM) 715 left ventricular hypertrophy (LVH) 310, 325 lenalidomide 534 leptin 330 lethargy 190, 322 level of hemoglobin <12 g/dl at the start of chemotherapy 217 LH/FSH ratio 331 limitation to intravenous iron therapy 685 limitations in cognitive function 322 Linear Analog Scale Assessment (LASA) 313, 732 liver tumors 6, 285 local concentrations of EPO 93 locoregional spread 292 long term efﬁcacy of RBT 400 long-acting rhEPO analogues 421 long-term survival of surgical cancer patients 667 loss of appetite 637 loss of concentration 637

Index loss of differentiation 291 loss of libido 308, 730 low haemoglobin concentration 759 low molecular weight heparin (LMWH) 753 low-molecular-weight iron dextran (INFeD®) 477 low-risk MDS 531 LTC-IC 604 lung cancer (small-cell and non-smallcell) 195, 211, 257 luteinizing hormone (LH) 331 lymphoma 190, 194 MACOP-B 210 macrocytic anemia 122 major depression 782 malignant lymphomas 157, 173 malignant melanoma 284 malignant phenotype 266 malignant progression 290–292, 296–297 malnutrition 329, 345, 796 maturation 3 mature RBC 44 MCH 158 MCHC (mean corpuscular Hb concentration) 158–159 MCV (mean corpuscular volume) 158–159 MDR-1 290 mean corpuscular volume (MCV) 119 mean FACT scores by hemoglobin 377 mechanisms of ESA-induced thrombosis in cancer patients 748, 749 mechanisms of iron carcinogenesis 706 medulloblastoma 653 melphalan 762 membrane transporters 289, 291 memory problems 637 menorrhagia 311 menstrual abnormalities including amenorrhea 331 mental alertness 637

861

meta-analysis of rates of thromboembolism in randomized studies of ESAs 748 metabolic abnormalities observed in anemic patients 329 metabolic efﬁciency 719 metastasis 292 methotrexate 451 microangiopathic hemolysis (MAH) 128, 135 microchimerism 338 microcytic anemia 120 mild anaemia, 397 minimal clinically important difference (MCID) in QOL 375 mistransfusion 335 mitomycin C 763 moderate anaemia 397 MOPP 210 multidimensional QOL measures 371 multi-drug resistance proteins 294 multiple myeloma (MM) 157, 165, 173, 197, 211, 220, 340, 433 multiple sclerosis 776, 782 myelodysplasia 340 myelodysplastic syndrome (MDS) 157, 173, 531, 564 myeloma 194 myelosuppression 132, 207, 215 myelosuppressive effects of chemotherapy 212 myocardial diseases 341 National Comprehensive Cancer Network (NCCN) 465 National Institute for Health and Clinical Effectiveness (NICE) 836 nausea 479 NCI, ECOG CALGB, GOG 118 need for RBC transfusions 450 neopterin 152 NESP 410 neurodegenerative diseases 782 neuroprotection 772 neuroprotective effects of EPO in man 778 neuroprotective effects of erythropoietin 771

862

Index

neuroprotective factors 773 neuropsychiatric disease 775 neurotoxicity of vincristine 459 N-glycans 409 NF-κB 19 NF-E2 19 Non Small Cell Lung Cancer (NSCLC) 395 non-Hodgkin lymphoma (NHL) 197, 211, 221, 285, 433, 642 non-platinum chemotherapy 196 non-platinum-based chemotherapy 213, 218 non-platinum-containing chemotherapy 763 normal breast tissue 269 normocytic anemia 121 nosocomial and postoperative infections 338 Novel Erythropoiesis Stimulating Protein 410 nuclear factor 289 O/P = ratio of observed to predicted log serum EPO levels 173 O2 availability 271, 283 O2 consumption rate of 285 O2 depletion 296 O2 extraction 269 O2 microsensors 267 O2 supply 271 O2 transport capacity 265, 270, 296 O2 transport index 275 O2-dependent chemotherapy 268 octamer-binding protein-1 (Oct-1) 21 of the cytokine receptor I family 93 ontogenesis 1 optimal Hb level for quality of life 739 optimal level of hemoglobin in cancer patients 346 oral iron 671, 707 oral iron supplementation 682 osteosarcoma 649 ovarian cancer 211, 224 overall survival 766 oxidative DNA damage 703

oxidative hemolysis 136 oxidative stress 330 oxygen (O2) 5, 703 oxygen consumption 295–796 oxygen delivery 270 oxygen free radicals 703 oxygenation status 267, 286, 295 oxygenation status, hemoglobin 286 oxygenation status, tumors 283, 286 oxygen-enhancement effect 293 P300 327 paclitaxel 213, 451 palliative care 654 pallor 308, 311, 324, 636 palpitation 308, 310, 322, 730 pancreatic cancer 284 parenteral iron 688, 707 Parkinson’s disease 776, 782 pathological tumor stage 287 patient selection, prediction of response 461 PBPC transplantation 600, 601 perfusion 285 perioperative allogeneic blood transfusion 666 perioperative anemia 667 perioperative blood salvage 665 peri-operative blood transfusions 393 peripheral vascular resistance 319 peripheral vasodilatation 318–319 pharmacokinetic effects of anemia 345 pharmacokinetics 290, 407 pharmacological properties of ESAs 452 photodynamic therapy 268, 289, 291, 295–296 photosensitizer 295 physical disability 345 physical function 725 plasma elimination half-life 413 platinum compounds 198 platinum treatment 196 platinum-based chemotherapy 196, 213, 217 pO2 histography 284

Index point mutations 292 polarographic technique 284 polyethylene glycol 421 poor concentration 637 poor feeding 637 postoperative infections 666 post-stroke recovery 779 postural hypotension 322 potential risks of allogenic blood transfusions 335 pounding pulse 322 predicting response of anemia to ESAs 469 prediction of anemia 198 prediction of response to EPO in the anemia of renal failure 556 prediction of response to rhEPO 442, 467, 541 predictive algorithm of response to rhEPO 556 predictive algorithms 560 predictive factors for anemia 201, 202 predictive models 555 predictor of anemia in patients receiving cisplatin 218 predictors for cisplatin-induced anemia 200 pretreatment Hb levels 204, 251 prevalence and incidence of anemia 212 prevalence of anemia 190–191, 252–253 preventing chemotherapy-induced anemia 461 prevention of anemia, early intervention 461 probability of RBC transfusions 200 problems in identifying functional EPO-R in tumor tissues 103 prognosis 265, 296–297 prognostic factor 265, 294, 296 prognostic impact of anemia 254 prognostic parameter 266 progression-free survival 481 proliferation kinetics 289 proliferation potential 293 ProMACE 210 promiscuous activation 93

863

proportion of anaemic cancer patients treated with transfusion or erythropoietin 818 proportion of anaemic patients by each chemotherapy cycle 396 proportion of patients with Hb < 11 g/dl over the course of chemotherapy 395 prostate cancer 211, 284 proteinuria 310, 802 proteome 289 proteome changes 290, 297 pulmonary emboli 747 pure red cell aplasia (PRCA) 418, 479 QALY Value 828 QOL scores 375 Quality Adjusted Life Year (QALY) 821, 829 quality of life (QOL) 311, 369, 434, 441, 713–714, 732, 764 RA (refractory anemia) 532 RA with excess of blasts (RAEB) 565 RA with ring sideroblasts (RAS) 565 radiation 291, 653 radiation-induced anemia 252 radical oxygen species (ROS) 804 radiochemotherapy 254 radioresistance 293–294, 761 radiosensitivity 292, 294, 761 radiotherapy 250, 289, 292, 763 RAEB with excess blasts in transformation (RAEB-t) 532 rapid correction of anaemia 402 RBC transfusions 450 RCAS1 168 real-time assessment of the functional state of erythropoiesis 688 receptor-binding cancer antigen expressed on SiSo cells 168 recombinant DNA technology 407 recombinant DNA-derived EPO 407 rectal cancer 284 recurrence of colorectal cancer 393 recurrent cancers 287

864

Index

red blood cell “storage lesion” 336 red blood cell (RBC) transfusion 3, 117, 119, 393, 434, 642 red blood cell (RBC) transfusions or recombinant human erythropoietin (rhEPO) 831 red blood transfusion (RBT) 393 reduced exercise capacity 322 refractory anemia (RA) 565 refractory anemia with excess blasts (RAEB) 532 refractory anemia with ringed sideroblasts (RARS) 532 relationship between anemia, cognitive dysfunction and fatigue 344 relationship between baseline hemoglobin level and the need for RBC transfusions during chemotherapy 219 relationship between erythropoietin, iron, and erythropoiesis 679 relationship between Hb level and cognitive function 343 relationship between initial storage iron (mg) and red blood cell volume expansion 684 relationship between renal function, EPO production, hemoglobin level, and RBC transfusions 216 relationship between the hematocrit and the bleeding time (BT) 342 relationship between utility of treatment and Hb 838 relative iron deﬁciency 680 relative risk of subsequent transfusion 402 renal cell carcinoma 211, 285 renal failure 346, 556, 800 renal toxicities and anemia 216 renin, angiotensin and aldosterone system (RAAS) 799 repair enzymes 291, 293 resistance to ESAs 461 resistance-related proteins 290–291 resistant clonal variants 292 respiratory disorders 342 response criteria 545 resting metabolism 715

reticulocyte count 119, 157, 159, 597, 688 retinal degeneration 776 retinoblastoma (Rb) 20 reverse transcriptase-polymerase chain reaction (RT-PCR) 104 rhabdomyosarcoma 649 rheological properties 271 rhEPO 166, 267 RhEPO after allogeneic transplantation 586, 592 RhEPO before transplantation 600 rhEPO dosages 457 rhEPO in hematopoietic stem cell transplantation 583 rhEPO in MDS 565 RhEPO to mobilize PBSC 602 rheumatoid diseases 346 the risk/beneﬁt proﬁle of intravenous iron 685 risk factors for anemia 190, 199, 525 risk of infection 334 risk of venous thromboembolism 525, 747 risks and limitations of RBC transfusion 334 role of EPO in the nervous system 773 safety of epoetin 417, 442, 764 safety of ESAs 479, 480 safety of systematic IV iron supplementation 553 sarcomas 644 schizophrenia 780, 782 secondary iron overload 340 selection pressure 292 serum creatinine 175 serum EPO concentration 171 serum EPO 557 serum EPO, Hb and sTfR after a chemotherapy 557 serum EPO levels 173 serum EpO levels after HSCT 584 serum ferritin 120, 159, 682 serum folate level 122 serum homocysteine 122 serum iron 159

Index serum iron levels 680 serum levels that differentiate anemia of chronic disease from irondeﬁciency anemia 688 serum transferrin receptor (sTfR) 159, 556, 597 severe anaemia 397 severity of anemia 118 sexual dysfunctions 331 shortness of breath 308, 310 sialic acid residues 409 sickle trait 270 side effects of rhEPO administration in HSCT 606 signiﬁcance of anemia in cancer chemotherapy 201 singlet oxygen species 295 site-directed mutagenesis 410 sites of the degradation of EPO 415 skeletal muscle 725 skin rush 479 sleeping disorders 322 sleeping disturbance 730 social costs of cancer care 815 soft tissue sarcoma 265, 284 solid tumor 157, 173, 250, 642–643, 645, 647–648, 650–651, 654 soluble forms of the EPO receptor 90 speciﬁc in vivo biological activity 409 speciﬁcity of antibody C-20 107 speciﬁcity of EPO-R detection techniques 103 squamous cell carcinoma 266, 274 STAT (Signal Transducer and Activator of Transcription) proteins 72 Stat5 22 storage lesion 335 storage of iron 704 stroke 778 superiority of IV iron 473 surgical intervention in cancer patients 663 survival 481, 761, 764 survival according to Hb level 225 SWOG 118 symptoms of anemia 321 syncope 322 systolic and diastolic dysfunctions 325

865

tachycardia 308, 310 tachypnoe 310 Tal-1 18, 19 telomerase 290 telomerase inhibitors 290 terminally ill cancer patients 398 TGF-b 586 thalassemia 120 thalidomide 533 therapeutic efﬁcacy 268 therapeutic resistance 283, 289, 291, 296 thrombocytosis 120 thromboembolic events (TEE) 480 thromboses during ESA therapy 748 thrombosis in patients with cancers 746 thrombosis prophylaxis 753 thrombotic risk posed by ESA 746 thrombotic thrombocytopenic pupura (TTP) 129 tinnitus 322, 637 tiredness 309 tissue acidosis 294 tissue hypoxia 318 tissue oxygenation 336 TNF 586 TNF-α 34, 153, 160, 162, 167, 175, 559 total Anemia Risk Score 201 toxic effects of iron 704 toxicity of drugs in older persons 345 TRAIL 169 transferrin receptor (TfR) 161, 689 transferrin saturation (TFS) 159, 470, 680, 682, 688 transfusion and erythropoietin compared 829 transfusion “trigger” 395–396 transfusion outcome 514 transfusion reactions 338 transfusion requirements 663, 764 transfusion-associated graft-versus-host disease (TAGVHD) 338 transfusion-related acute lung injury (TRALI) 338, 666 transient cerebral ischemia 322

866

Index

treatment of cancer (EORTC) 465 treatment outcome 293 treatment resistance 289 treatment with rhEPO after HSCT 586 treatment-induced anemia 253 treatment-related anemia 132 treatments given to anaemic cancer patients 817 Trial Outcome Index-Fatigue (TOI-F) 372 tumor aggressiveness 265 tumor-associated macrophages (TAMs) 151 tumor growth 417 tumor hypoxia 259, 265, 283, 284, 291–293, 295–297 tumor microcirculation 285 tumor microvasculature 296 tumor microvessels 271 tumor necrosis factor (TNF) 30 tumor necrosis factor alfa (TNF-α) 149, 554, 796 tumor oxygenation 259, 265, 267, 274–275, 283, 288 tumor progression 296–297 tumor therapy, O2 dependent 296 tumor vascularization 294 tumor-associated macrophages (TAMs) 150 tumor-to-tumor variability 287 tumour hypoxia 760 TWEAK 42 ultraﬁlterable (UF) platinum concentration 200 urothelial cancer 211 use of iron in cancer patients 703 use of rhEPO (300 IU/kg/day) perioperatively 670

VAS score 398 vascular density 294 vascular endothelial cells 749 vascular endothelial growth factor (VEGF) 259, 760 vascular permeability 271 VEGF 293 VEGF in anemic tumor patient 260 Venofer® 692 venous thromboses 747 ventilation 719 vertigo 308, 310, 322, 637 viability of stored RBCs 337 VIML 210 vinorelbine 213 viscous resistance 271, 274 vitamin B12 122 vitamin B12 deﬁciency 122 VO2 715 volume of RBCs that a patient can tolerate 664 vomiting 479 von Hippel-Lindau tumor suppressor protein (pVHL) 10 vulvar cancer 272 Waldenström macroglobulinemia 211, 223 warm antibodies 126 wasting 329, 713 weakness 309, 322 weight loss 322, 713, 725 Western blotting 104 whole-body metabolic efﬁciency 725 whole-body oxygen uptake 719 World Health Organization (WHO) 118, 346 World Health Organization (WHO) performance status 194

Immunoendocrinology: Scientific and Clinical Aspects

Immunoendocrinology: Scientific and Clinical Aspects

Clinical Trials in Oncology

Clinical Trials in Oncology

Clinical Trials in Oncology

Clinical Trials in Oncology

Clinical Trials in Oncology

Clinical Trials in Oncology

Clinical Trials in Oncology

Clinical Trials in Oncology

Cancer Neurology in Clinical Practice (Current clinical Oncology)

Cancer Neurology in Clinical Practice (Current clinical Oncology)

Cancer Neurology in Clinical Practice 2nd Edition - Neurologic Complications of Cancer and Its Treatment (Current Clinical Oncology)

Cancer Neurology in Clinical Practice 2nd Edition - Neurologic Complications of Cancer and Its Treatment (Current Clinical Oncology)

Breast Cancer: Atlas of Clinical Oncology

Breast Cancer: Atlas of Clinical Oncology

Clinical Pain Management Cancer Pain, 2nd edition

Clinical Pain Management Cancer Pain, 2nd edition

Manual of Clinical Oncology

Manual of Clinical Oncology

Infections in Cancer Patients (Basic and Clinical Oncology)

Infections in Cancer Patients (Basic and Clinical Oncology)

Supportive Care in Cancer (Basic and Clinical Oncology)

Supportive Care in Cancer (Basic and Clinical Oncology)

Manual of Clinical Oncology

Manual of Clinical Oncology

Manual of Clinical Oncology

Manual of Clinical Oncology

Abeloff's Clinical Oncology, 4th Edition

Abeloff's Clinical Oncology, 4th Edition

Abeloff's Clinical Oncology, 4th Edition

Abeloff's Clinical Oncology, 4th Edition

Cancer: Basic Science and Clinical Aspects

Cancer: Basic Science and Clinical Aspects

Abeloff's Clinical Oncology, 4th Edition

Abeloff's Clinical Oncology, 4th Edition

Clinical Radiation Oncology, 3rd Edition

Clinical Radiation Oncology, 3rd Edition

Clinical Radiation Oncology, 3rd Edition

Clinical Radiation Oncology, 3rd Edition

Clinical Scenarios in Surgical Oncology (Clinical Scenarios in Surgery Series)

Clinical Scenarios in Surgical Oncology (Clinical Scenarios in Surgery Series)

Intraperitoneal Cancer Therapy (Current Clinical Oncology)

Intraperitoneal Cancer Therapy (Current Clinical Oncology)

Handbook of Statistics in Clinical Oncology

Handbook of Statistics in Clinical Oncology

Handbook of Statistics in Clinical Oncology

Handbook of Statistics in Clinical Oncology

Cancer neurology in clinical practice

Cancer neurology in clinical practice

Handbook of Statistics in Clinical Oncology

Handbook of Statistics in Clinical Oncology

Cancer Chemotherapy in Clinical Practice

Cancer Chemotherapy in Clinical Practice

Clinical Progress in Renal Cancer

Clinical Progress in Renal Cancer

Thyroid Cancer in Clinical Practice

Thyroid Cancer in Clinical Practice

Thyroid Cancer in Clinical Practice

Thyroid Cancer in Clinical Practice

Our partners will collect data and use cookies for ad personalization and measurement. Learn how we and our ad partner Google, collect and use data. Agree & close