Higher Versus Standard Doses of Recombinant Erythropoietin for the Treatment of Anemia in Patients with Myelodysplastic Syndromes: Results of a Retrospective Survey from the Italian Registry of Myelodisplastic Syndromes (FISM)

Introduction: Erythropoietic stimulating agents (ESAs) have not yet received FDA nor EMA approval to treat anemia of myelodysplastic syndromes (MDS), but ESAs are in use in this clinical setting since more than 25 years. Early studies, using various but usually "standard" doses of recombinant EPO (rEPO) (30-40.000 IU weekly), showed a disappointing overall response-rate of no more than 15-25%. In recent years most studies reported a response rate of more than 50%, provided MDS patients were selected according to favorable clinical variables determined during the years ( ie low transfusion requirement, absence of blasts, endogenous EPO levels < 500). Higher response rates with respect to early studies are probably due to higher, not weight-adjusted doses of rEPO (60-80.000 UI weekly). Nevertheless, a direct comparison between the two different schedules of rEPO treatment is still lacking, and the superiority of the so called higher doses of rEPO to standards regimens is inferred only by results of meta-analysis comparing studies performed at different times with various schedules and heterogenous cohorts of MDS patients.

Objectives: We aimed at clarifying whether the erythroid response rate differed in patients exposed to higher doses versus standard doses of rEPO.

Methods: Within the framework of the Italian Network of regional MDS registries established by Fondazione Italiana Sindromi Mielodisplastiche (FISM) a cohort of 103 MDS patients (pts) with anemia treated with higher doses of rEPO (40.000 IU twice a week, H) for at least 3 months within 6 months from diagnosis were identified; a second cohort of 206 pts, similar for clinical parameters known to influence response to rEPO (i.e. Hb concentrations at the time of starting treatment, IPSS score, transfusion-dependency, endogenous Epo levels at diagnosis and time of treatment from diagnosis) and treated with standard doses (40.000 IU weekly, S) were compared to the first cohort. Univariate and multivariate analysis were performed as appropriate in order to identify factors influencing clinical response to treatment.

Results: Characteristics of subjects were: median Hb pre-treatment 8.9 g/dL in H cohort and 9.1 g/dL in S cohort, IPSS score Intermediate-2/high in 5% of H cohort and 8% of S cohort, transfusion-dependency in 25% of H cohort and 26% of S cohort, median EPO at diagnosis 79 IU in H cohort and 69 IU in S cohort. According to IWG 2006 criteria, after 3 months of rEPO treatment the overall erythroid response-rate among all the pts in the two cohorts was 53% (163 out of 309 pts). No difference in erythroid response-rate was found between MDS pts treated with higher doses (49 (48%) responders out of 103 cases) when compared to cases treated with standard doses (114 (55%) responders out of 206 pts, p= 0.23). As expected, IPSS score, transfusion-dependency and EPO serum levels at diagnosis were statistically associated with response. In particular, at multivariate analysis, significantly lower response-rates to rEPO were associated with transfusion-dependency (yes vs no, OR= 0.59 (95%CI: 0.44-0.79, p<0.001), higher endogenous serum EPO levels at diagnosis (>500 vs <=500, OR= 0.36 (95%CI: 0.19-0.68, p=0.002) and higher IPSS score (intermediate 2/ high vs, intermediate 1 / low, OR=0.42 (95%CI: 0.24-0.74, p=0.003).

Conclusions: Our data, although derived by a retrospective analysis, seem to indicate that standard doses of rEPO are at least as effective as higher-doses for correcting anemia in MDS patients; in this clinical scenario, a standard-doses rEPO treatment allows for a consistent reduction of costs without precluding to achieve a durable erythroid response to rEPO. Prospective, randomized studies addressing this point are necessary to definitely address this topic.