|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Originally published as JCO Early Release 10.1200/JCO.2005.05.036 on August 8 2005 © 2005 American Society of Clinical Oncology.
Erythropoietin Use in Cancer Patients: A Matter of Life and Death?Mayo Clinic, Rochester, MN It has now been 15 years since the first formal demonstration that recombinant human erythropoietin can ameliorate anemia associated with cancer.1 Since then, dozens of clinical trials in diverse groups of cancer patients have convincingly demonstrated the power of erythropoietin to increase hemoglobin (Hb) levels, and to reduce the need for potentially risky RBC transfusions.2 Three nonrandomized, community-based, open-label studies, enrolling more than 7,000 patients with nonmyeloid malignancy,3-6 and at least one randomized study,7 also suggested that erythropoietin can improve patients' quality of life (QOL). A subsequent placebo-controlled study, however, indicated that this QOL benefit may be more modest than was initially hoped.8 Because recombinant erythropoietin is viewed by most clinicians as relatively nontoxic, and because erythropoietic agents are widely believed to have both objective and subjective benefits, the chief barriers to the universal use of this class of drugs have been relatively high cost, the lack of a uniform Hb response, clinical inertia, and patient inconvenience.9 In addition, there are several efficacy hurdles that, if cleared, could lead to a more robust recommendation for the use of erythropoietin in all patients with cancer: a clear demonstration that erythropoietin improves tumor response to therapy; an unequivocal demonstration that erythropoietin improves overall survival; and convincing evidence that erythropoietin therapy can benefit patients with normal or near-normal Hb levels, just as it helps those with moderate or severe anemia. Because improvement in overall survival is the gold standard for any drug treatment in people suffering from malignancy, trials were developed to try to demonstrate that recombinant erythropoietin could positively influence life expectancy in this group of patients. Improvement in cancer survival with erythropoietin therapy seemed a lofty goal but did not appear to be too much to ask. There was good reason to believe that successful treatment of anemia might augment the effectiveness of cancer therapy and prolong life, given that there is little question that the presence of moderate to severe anemia is associated with reduced survival in persons with a malignancy.10,11 It is less clear, however, whether anemia itself is the proximate cause of reduced survival or is instead a surrogate marker for other adverse prognostic features.12 Correction of Hb to near-normal levels with RBC transfusions might help clarify this question, but is not practical and has multiple potential toxicities. Thus, the possibility of improving survival by ameliorating anemia is best addressed by the use of recombinant erythropoietin. However, an important caveat with respect to erythropoietins is that the effects of this class of drugs on cancer outcomes from anemia correction alone may never be understood clearly because they appear to have biologic effects other than just increasing hemoglobin.13 Interest in a possible positive association between erythropoietin treatment and survival was fueled by a prospective, randomized, double-blind, allocation-concealed, multinational study of 375 anemic patients receiving chemotherapy for nonmyeloid cancer, which suggested a trend toward improved survival with erythropoietin use (median survival, 17 months with epoetin alfa, 11 months with placebo; Kaplan-Meier estimates, P = .13, log-rank test).7 However, that study's primary end points were transfusions, Hb response, and QOL; it was not powered to detect survival differences. A similar but also not statistically significant trend toward improved survival with erythropoietin was observed in the recently published, placebo-controlled epoetin alfa study of the North Central Cancer Treatment Group, which also enrolled anemic patients with nonmyeloid cancer receiving chemotherapy.8 Despite these two large trials and other smaller studies suggesting a potential survival benefit with the use of recombinant erythropoietin, a meta-analysis that included 19 studies in cancer patients (randomized trials published through 2001) could not reach a definitive conclusion about whether there might be a treatment effect on overall survival.14,15 The analysis did report a trend for improved survival, with an adjusted hazard ratio for death of 0.81 with erythropoietin (95% CI, 0.67 to 0.99). Most trials included in this meta-analysis had a relatively short observation period (median, 84 days), and therefore had limited power to uncover a survival advantage with erythropoietin. The need to resolve this issue definitively has become more pressing with the recent publication of two studies suggesting that erythropoietin therapy might actually be harmful in certain groups of cancer patients. One of these studies is the Breast Cancer Erythropoietin Survival Trial (BEST) by Leyland-Jones et al, first published as a preliminary report in 200316 and now presented in greater detail in this issue of the Journal of Clinical Oncology.17 In contrast to most other studies of erythropoietin in cancer patients, this large clinical trial had overall survival as the primary end point. It enrolled 939 metastatic breast cancer patients in 20 countries. Those in whom the Hb decreased below 13 g/dL were treated with either epoetin alfa or placebo for 12 months, with the goal Hb range of between 12 and 14 g/dL. Although the investigators' preliminary report had stated "Patients were included in the study if their hemoglobin concentration was 13 g/dL,"16 which was somewhat ambiguous, the full report published here makes it clear that patients were enrolled regardless of baseline Hb. This point may be important because although the mean Hb in each study group was identical (12.5 g/dL), there was a trend for more severe baseline anemia in the treatment arm (Hb < 10.5 g/dL in 14% of those patients v 11% in the placebo arm). The BEST study was stopped early by a data safety monitoring committee because of poorer overall survival in the epoetin alfa arm. This rather surprising result received considerable coverage in high-profile medical journals16,18 and in the lay press at the time the data were first released, and caused concern among many patients and physicians who pay attention to such things. Because of the limited data set collected prospectively in the BEST study, an independent retrospective chart review was performed to try to shed additional light on this result, which confirmed that the baseline performance status was worse in the erythropoietin-treated group, for unclear reasons. Most of the excess deaths in the BEST trial were believed to be the result of early disease progression, though there was also an imbalance in deaths as a result of chemotherapy toxicity and thromboembolic events (14 suspected clots in the erythropoietin arm v four in the placebo arm). Almost all of the excess deaths in the erythropoietin arm occurred in the first 4 months of the trial (ie, an additional 25 events in the epoetin alfa group: 41 v 16 during that period), with only two additional excess deaths over the subsequent study period (138 v 111 deaths overall). Thus, most of the excess deaths occurred at a time point when it is difficult to imagine how erythropoietin might have been a factor, short of contributing to a catastrophic event; yet the independent chart review disclosed no consistent evidence for such events. A few other features of the trial remain puzzling: despite the early deaths and survival distinction, the overall time to disease progression and tumor response rates were not different between groups. In addition, even though one group received placebo, only minimal Hb differences developed overall between arms. The detrimental survival effect was also not long lasting. By 19 months into the study, when patients were no longer receiving study drug, the proportion of the surviving population in each arm was virtually the same. It is tempting to attribute these results to bad luck—a statistical fluke that perhaps happened because the study did not include performance status or the degree of baseline anemia as a stratification factor in the randomization. Imbalances in treatment populations can occur randomly even in large trials such as this one, although real-world experience with the law of large numbers suggests that most of the time, if a clinical trial is large enough, the important differences will balance out. In addition, if enough studies are performed with a single therapy, there will always be some results that are outliers. Still, the BEST results do give one pause. BEST was a relatively large study, designed specifically to look at a survival end point; not only was survival not better with erythropoietin, it was significantly worse. The 2003 preliminary report of the BEST trial16 came on the heels of another clinical study that challenged the sanguine conventional wisdom that erythropoietin was almost uniformly beneficial for patients with cancer. Henke et al19 studied the effects of epoetin beta in 351 anemic head and neck cancer patients receiving radiotherapy; the therapeutic goal was normalization of Hb (> 14 g/dL in women, > 15 g/dL in men). Locoregional progression-free survival was unexpectedly poorer with epoetin beta treatment than with placebo (adjusted relative risk, 1.62; 95% CI, 1.22 to 2.14; P = .0008). Although the methodology of this study was heavily criticized in a series of letters published in The Lancet in 2004, the results are still noteworthy. It may not be a coincidence that the BEST study and the study by Henke et al that observed poorer treatment outcomes with recombinant erythropoietin had a higher Hb target than most other studies in cancer patients. The toxicity profile of erythropoietin at higher Hb levels may be distinct: at least three other studies of erythropoietic agents in cancer patients that had a normal or near-normal goal Hb, and were not looking specifically at survival as an end point, have been terminated because of higher thromboembolic risk in the active therapy arm. Collectively, these findings have called into question the prevailing view of erythropoietin as a totally benign drug. From a mechanistic standpoint, how might erythropoietin affect survival in cancer patients, for better or for worse? Clinical outcomes might represent a complex balance between the downstream effects of erythropoietin on anemia, specific adverse events attributable to the drug (such as thromboembolism), and direct effects of the erythropoietin itself on tumor cells. In vitro data support several competing possibilities. Tumor oxygenation can be improved with alleviation of anemia; this might increase the effectiveness of radiotherapy or chemotherapeutic agents that are dependent on oxygen tension in the tumor microenvironment, or might shut down the hypoxia-inducible factor 1 alpha–vascular endothelial growth factor angiogenesis pathway that promotes tumor growth and metastasis. In contrast, diverse tumor cell lines have erythropoietin receptors,20 and erythropoietin has effects on intracellular signaling pathways such as the Janus kinase–signal transducer activator of transcription (JAK-STAT) signaling cascade and phosphoinositol-3 kinase activity, which might be important in altering the biologic behavior of tumor cells, such as inhibiting apoptosis.21 An immunomodulatory role has also been proposed, on the basis of research in the mouse.22 Because some of these effects would be expected to promote tumor growth and others to inhibit it, the overall in vivo effects of erythropoietin on human cancer can only be resolved by well-designed clinical studies. In evaluating the risks and benefits of erythropoietic agents, it is important to recognize that epoetin alfa, epoetin beta, and darbepoetin alfa—the three most commonly used erythropoietic agents at present—are not identical molecules. They have unique pharmacokinetics and response and toxicity profiles, and might be expected to have distinct effects on survival. The dose and schedule of recombinant erythropoietin therapy could also matter, as might the goal Hb and the rate of Hb increase during treatment; the majority of treated patients in both the BEST and the European head and neck cancer trial achieved an Hb level higher than that recommended by evidence-based guidelines, such as those published by the American Society of Clinical Oncology and the American Society of Hematology (12 g/dL).23 One can even imagine that there is an important effect related to the specific type of tumor: some pathologies might be more sensitive to oxygen tension than others, or might be more often treated with chemotherapy drugs with an oxygenation-dependent cytotoxic mechanism, or might have intracellular signaling pathways with distinct responsiveness to erythropoietin. It is no secret that the use of erythropoietic agents in clinical practice consumes a colossal share of health care resources. In the United States, erythropoietin represents the largest single drug expense for the Center for Medicare & Medicaid Services—on the order of one billion dollars per year, although not all of this is for cancer patients (hemodialysis is the other major indication for erythropoietic agent therapy).24 Amgen's (Thousand Oaks, CA) 2004 sales of epoetin alfa (Epogen) were estimated at $2.6 billion, and sales of darbepoetin alfa (Aranesp) were estimated at $2.4 billion.25 Johnson & Johnson (New Brunswick, NJ), the parent company of Ortho Biotech (Bridgewater, NJ), is believed by analysts to have sold $3.6 billion worth of Procrit (epoetin alfa) and Eprex (epoetin alfa) last year.26 To put these numbers into perspective, the cumulative worldwide box office gross for all five of George Lucas' Star Wars films amounted to a mere $3.4 billion, and the total spending on the 2004 American Presidential and Congressional elections has been estimated to be between $3 and $4 billion.27 If recombinant erythropoietic agents improve cancer survival in addition to their proven benefits on Hb and transfusion requirements, then they might be worth every penny. But with such high stakes, we owe it to our patients and payers to get it right. Ideally, an adequately powered study in a carefully balanced group of cancer patients with overall survival as the primary end point would go a long way toward answering the important question of whether erythropoietin therapy improves survival, is harmful, or has no substantive effect on life expectancy. It is probably not necessary for such a trial to enroll patients with an identical tumor type and stage, as long as well-recognized general prognostic features such as performance status are balanced, and the study is large enough that tumor types and other characteristics are evenly distributed. In addition, because of the possible distinct biologic behavior of erythropoietins at various Hb ranges, more than one trial with different goal Hb levels and different agents might be required. In May 2004, an advisory panel of the US Food and Drug Administration urged that more trials of erythropoietic agents be performed in cancer patients in the United States, given that most large randomized trials of these drugs have been conducted in Europe. Unfortunately, given the profits that are reaped from erythropoietin therapy, years of high-penetration direct-to-consumer advertising, and the great comfort that clinicians have developed with use of this class of drugs, it may be extraordinarily difficult to accrue patients to such trials. We may have to rely on well-conducted future meta-analyses, which should be able to include at least twice as many patients as the most recently conducted meta-analysis. In August 2004, the US Food and Drug Administration's MedWatch safety information program announced a letter addressed to oncologists, hematologists, and other health care professionals regarding the safety of epoetin alfa (Procrit). The associated new prescribing information for epoetin alfa recommends that the target Hb level in patients with cancer should be "individually determined for each patient" (although it is not clear how such determinations should be made), that this target should not exceed 12 g/dL, and that epoetin alfa doses should be withheld if the Hb level is 13 g/dL or higher. In addition, a new recommendation about the rate of Hb increase was made, with dosing interruption and modification also recommended if the rate of Hb increase exceeds 1 g/dL during a 2-week period. Similar warnings were added to the darbepoetin alfa (Aranesp) label in December 2004, as a precaution, although the adverse events prompting the label change were observed with other drugs in the class. Given the available evidence, these interim recommendations seem sensible and probably should be applied to all erythropoietic agents. Authors' Disclosures of Potential Conflicts of Interest The authors indicated no potential conflicts of interest. REFERENCES 1. Ludwig H, Fritz E, Kotzmann H, et al: Erythropoietin treatment of anemia associated with multiple myeloma. N Engl J Med 322:1693-1699, 1990[Abstract]
2. Seidenfeld J, Piper M, Flamm C, et al: Epoetin treatment of anemia associated with cancer therapy: A systematic review and meta-analysis of controlled clinical trials. J Natl Cancer Inst 93:1204-1214, 2001
3. Glaspy J, Bukowski R, Steinberg D, et al: 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, 1997 4. Demetri GD, Kris M, Wade J, et al: 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, 1998[Abstract]
5. Gabrilove JL, Cleeland CS, Livingston RB, et al: 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, 2001 6. Desai J, Demetri GD: Recombinant human erythropoietin in cancer-related anemia: An evidence-based review. Best Pract Res Clin Haematol 18:389-406, 2005[Medline]
7. Littlewood TJ, Bajetta E, Nortier JW, et al: 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, 2001 8. Witzig TE, Silberstein PT, Loprinzi CL, et al: Phase III, randomized, double-blind study of epoetin alfa versus placebo in anemic patients with cancer undergoing chemotherapy. J Clin Oncol 23:2606-2617, 2004 9. Glaspy JA: Overcoming barriers to erythropoietic therapy: Characteristics of darbepoetin alfa and potential advances in scheduling, dosing, and dose-efficiency. Oncology (Williston Park) 16:71-77, 2002 10. Caro JJ, Salas M, Ward A, et al: Anemia as an independent prognostic factor for survival in patients with cancer: A systemic, quantitative review. Cancer 91:2214-2221, 2001[CrossRef][Medline] 11. Ludwig H, Van Belle S, Barrett-Lee P, et al: 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, 2004 12. Glaspy JA: The potential for anemia treatment to improve survival in cancer patients. Oncology (Williston Park) 16:35-40, 2002
13. Maiese K, Li F, Chong ZZ: New avenues of exploration for erythropoietin. JAMA 293:90-95, 2005 14. Bohlius J, Langensiepen S, Schwarzer G, et al: Erythropoietin for patients with malignant disease. [Cochrane Database System Review]. Oxford, England, Cochrane Library, CD003407, 2004
15. Bohlius J, Langensiepen S, Schwarzer G, et al: Recombinant human erythropoietin and overall survival in cancer patients: Results of a comprehensive meta-analysis. J Natl Cancer Inst 97:489-498, 2005 16. Leyland-Jones B: Breast cancer trial with erythropoietin terminated unexpectedly. Lancet Oncol 4:459-560, 2003[CrossRef][Medline]
17. Leyland-Jones B, Semiglazov V, Pawlicki M, et al: 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, 2005 18. Brower V: Erythropoietin may impair, not improve, cancer survival. Nat Med 9:1439, 2003 19. Henke M, Laszig R, Rube C, et al: Erythropoietin to treat head and neck cancer patients with anaemia undergoing radiotherapy: Randomised, double-blind, placebo-controlled trial. Lancet 362:1255-1260, 2003[CrossRef][Medline]
20. Yasuda Y, Fujita Y, Matsuo T, et al: Erythropoietin regulates tumour growth of human malignancies. Carcinogenesis 24:1021-1029, 2003 21. Leyland-Jones B: Evidence for erythropoietin as a molecular targeting agent. Semin Oncol 29:145-154, 2002
22. Mittelman M, Neumann D, Peled A, et al: Erythropoietin induces tumor regression and antitumor immune responses in murine myeloma models. Proc Natl Acad Sci U S A 98:5181-5186, 2001
23. Rizzo JD, Lichtin AE, Woolf SH, et al: 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, 2002 24. Centers for Medicare & Medicaid Services: Testimony of Thomas A. Scully Administrator Centers For Medicare & Medicaid Services on reimbursement & access to prescription drugs under Medicare Part B House Ways and Means Committee Subcommittee on Health, October 3, 2002. http://www.cms.hhs.gov/media/press/testimony.asp?Counter=741. 25. Standard & Poor's Stock Report: Amgen (AMGN). April 2, 2005, New York, NY, McGraw-Hill 26. Standard & Poor's Stock Report: Johnson & Johnson (JNJ), April 3, 2005, New York, NY, McGraw-Hill 27. Report of the Nonpartisan Center for Responsive Politics: '04 Election expected to cost nearly $4 billion, October 21, 2004. http://www.opensecrets.org/pressreleases/2004/04spending.asp
Related Article
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2005 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
