Erythropoietin Levels and Androgens Use: What Is Their Relationship in the Correction of Anemia?

While Daniell¹ writes that androgen therapy is a proven treatment of anemia in patients with end-stage renal disease (ESRD) through the stimulation of endogenous erythropoietin levels or some other erythropoietic cells, Spivak² notes in his reply that such an assertion contradicts the known physiology of erythropoiesis. Perhaps I can help resolve some of the conflict. Thirty years ago when we used androgens in the treatment of ESRD, we had just begun to realize that the kidney participated in erythrocyte regulation. At that time, we used a mouse bioassay for detecting the functional presence of erythropoietin, but it could not distinguish between the absence of erythropoietin and the lack of its efficacy due to inhibiters. We then developed an enzyme-linked immunosorbent assay to detect the actual quantities of erythropoietin in patients with ESRD and found that they had higher erythropoietin levels than our control patients and that those levels seemed to be independent of androgen use, transfusion dependency, and hematocrit level (Figure).³ Like Daniell,¹ we even postulated at that time that androgens might act by a separate mechanism from erythropoietin. Because the development of biorecombinant erythropoietin was still many years away, we had not yet learned that there were factors such as age, hyperparathyroidism, inflammation, inadequate dialysis, angiotensin antagonists, acidosis, and hyperphosphatemia⁴ that can produce erythropoietin resistance and distort such relationships as we had chosen to observe. Spivak² is correct that androgens are neither necessary nor sufficient in the correction of anemia. Since the introduction of biorecombinant erythropoietin, most of the anemia in ESRD is corrected, regardless of endogenous erythropoietin levels. Such a feat was almost never accomplished with androgen therapy before the introduction of biorecombinant erythropoietin, as noted in the Figure. Endogenous erythropoietin levels in ESRD do not seem to predict hematocrit levels or a response to biorecombinant erythropoietin. Recently, we found that increased age alone, unrelated to sex, is associated with resistance to biorecombinant erythropoietin,⁴ which could explain the association that Daniell¹ notes in his figure and has mistakenly attributed to the effect of diminishing endogenous androgens in elderly men. The National Kidney Foundation does not recommend androgens as an adjunctive therapy for erythropoietin resistance because the adverse effects of androgen therapy appear to outweigh its potential benefits. Therefore, investigation of reversal of inhibiters of erythropoietin, such as cytokines like tumor necrosis factor α and defensins like hepcidin, may be a more fruitful pathway in the treatment of erythropoietin resistance.⁵