Erythropoiesis-stimulating Agents and Myeloma: Bad News or No News?

Abstract & Commentary

By William B. Ershler, MD

Synopsis: In a retrospective analysis of 323 patients with multiple myeloma, of which 169 patients received one or another form of ESA for treatment of disease or treatment-associated anemia, multivariate analysis, using a proportional hazard model, indicated worse progression-free and overall survival associated with ESA treatment. This report is quite different from another recently reported analysis, and highlights the problems associated with retrospective studies. Nonetheless, there is a prevailing sensibility that ESA treatment for anemic myeloma patients, as with all cancer patients, must be cautiously applied and carefully monitored.

Source: Katodritou E, et al. Erythropoiesis-stimulating agents are associated with reduced survival in patients with multiple myeloma. Am J Hematol. 2008;83:697-701.

There has been much concern in recent months regarding the possibility that the use of erythropoiesis-stimulating agents (ESA) may negatively influence overall survival for patients with cancer. The focus has been on solid tumors, for which such evidence has been mounting.1-3 Yet, for hematologic malignancies, the issue has been less well studied. In the current retrospective analysis, Katodritou et al from the Greek Myeloma Study Group reviewed their experience with 323 multiple myeloma patients followed between 1988 and 2007 to determine whether ESA treatment influenced progression-free or overall survival. One hundred and sixty-nine patients received ESA and 154 did not. The median duration of ESA administration was six weeks (range 4-10 weeks), and the median baseline hemoglobin (Hb) levels of patients who received ESA were 9.4 g/dL (range 7.3-11.9). Anemia response for the patients who received ESA was 68%.

Established myeloma prognostic factors (age, serum LDH, hemoglobin concentration, platelet number, serum creatinine, International Scoring Scale [ISS] score, and beta-2 microglobulin) and ESA use were each proven to be of prognostic significance in terms of overall survival (OS) by univariate analysis (p < 0.05, for all parameters). Using a proportional hazards model for multivariate analysis, including all of the parameters found significant by univariate analysis (as above), only beta-2 microglobulin, age, and ESA use predicted survival (p < 0.05 for each). The median OS of patients in the ESA group was 31 months (95% CI: 25- 37 months), whereas in the group without ESA administration, it was 67 months (95% CI: 55-79 months; p < 0.001). Deep vein thrombosis (DVT) was observed in 11 patients (6.5%) in the ESA group and in seven patients (4.5%) in the group without ESA (p = 0.59). The odds ratio for thrombosis after ESA administration was 1.46 (95% CI: 0.51-4.3).


This report runs counter to that of Baz et al,4 who described the Cleveland Clinic experience with regard to outcomes of myeloma patients in the context of ESA use. Their study population consisted of 257 patients treated from 1997-2003, and followed for at least one month. Of these, 127 patients received an ESA for at least one month, whereas the remainder had not. On average, patients who received ESA were older, had more advanced disease, higher serum creatinine, lower serum hemoglobin, higher beta 2-microglobulin, lower platelet counts, and a longer time from diagnosis to enrollment at the myeloma program (p < 0.001 for all). They found, however, that after adjusting for age, months from diagnosis to enrollment, serum creatinine, hemoglobin, platelet count, and beta 2-microglobulin, the use of an ESA was associated with improved overall survival (hazard ratio = 0.6; 95% CI = 0.38-0.94) in patients with intermediate or advanced disease but not in patients with early stage disease.

On the surface, it's hard to reconcile the findings from these two studies. Whereas the current report from Greece is somewhat larger and the study population a little more homogenous (care had been taken to exclude those who had been transfused as well as those who were treated beyond a hemoglobin of 13 g/dL), the Cleveland Clinic report reflected patients with more sustained ESA use (median ESA treatment duration was 36 weeks, compared with only six weeks in the Greek study). Thus, inherent differences in the population may explain the different findings.

It is disconcerting that we don't have an answer on this, but it is unlikely that a retrospective review, as large as it might be, will provide the solution. When myeloma patients develop anemia, it usually reflects advancing disease. Physicians rely on a whole host of factors, including the presence or absence of comorbidities, socioeconomic factors, and the ability to afford expensive therapy, that the confounders obviate comparable groupings, even in a well constructed multivariate analysis, such as attempted in the current report. Instead, a prospective, randomized trial would be the most direct way to answer this question, but don't look for this to happen anytime soon. Once again, clinicians will need to rely on their experience and clinical judgment rather than established evidence when confronted with the decision about how best to manage symptomatic anemia in patients with multiple myeloma, or any cancer for that matter.


1. Henke M, et al. Erythropoietin to treat head and neck cancer patients with anaemia undergoing radiotherapy: randomised, double-blind, placebo-controlled trial. Lancet. 2003;362:1255-1260.

2. Leyland-Jones B, et al. Maintaining normal hemoglobin levels with epoetin alfa in mainly non-anemic patients with metastatic breast cancer receiving first-line chemotherapy: a survival study. J Clin Oncol. 2005;23:5960-5972.

3. Wright JR, et al. Randomized, double-blind, placebo-controlled trial of erythropoietin in non-small-cell lung cancer with disease-related anemia. J Clin Oncol. 2007;25:1027-1032.

4. Baz R, et al. Recombinant human erythropoietin is associated with increased overall survival in patients with multiple myeloma. Acta Haematol. 2007;117: 162-167.