Head-to-Head Comparison of Epoetin Alfa (EPO) 40,000 U QW vs Darbepoetin Alfa (DARB) 200 μg Q2W in Anemic Cancer Patients Receiving Chemotherapy (CT): Final Results of a Planned Interim Analysis (IA).

Abstract

Although pharmacokinetic differences between EPO and DARB are known, clinical comparisons are best determined through randomized clinical trials. This randomized, open-label, multicenter study was designed to compare the efficacy of EPO and DARB for the treatment of CT-induced anemia using dosing consistent with the 2004 NCCN Cancer and Treatment-Related Anemia Guidelines. Hb response, transfusions (TFNs), and safety were assessed in patients (pts): ≥18 y with solid tumors; with baseline (BL) Hb ≤11 g/dL; scheduled to receive CT for ≥12 wks; and no prior erythropoietin therapy within 3 months. Pts were stratified by CT type (platinum/nonplatinum) then randomized 1:1 to EPO 40,000 U SC QW or DARB 200 μg SC Q2W for up to 16 wks. Dose increases were required for nonresponders (<1 g/dL Hb increase) after 4 wks for EPO (40,000 U to 60,000 U) or 6 wks for DARB (200 μg to 300 μg). If Hb increased to >13 g/dL, study drug was held and then reduced, and if Hb increased >1.3 g/dL (EPO) or >1.0 g/dL (DARB) in a 2-wk period, dose was reduced. Pts were withdrawn if CT was discontinued or CT type changed from platinum to nonplatinum or vice versa. Primary efficacy endpoint is Hb response rate (HRR; proportion of pts achieving an Hb increase of ≥1 g/dL from BL during the first 4 weeks) excluding Hb values within 28 days following a TFN. Primary analysis is a logistic regression model (treatment group and CT type as factors) comparing HRRs between the 2 treatment groups. A maximum of 400 pts could have been randomized; however, a prespecified IA of the primary endpoint using logistic regression was to be performed after the first 300 pts completed 4 wks of therapy. If the IA demonstrated that P≤.0125 (1-sided), then the treatment difference between EPO and DARB was considered statistically significant in favor of EPO and enrollment would be terminated. 339 pts (170 EPO, 169 DARB) comprised a modified intent-to-treat (MITT) population (pts with ≥1 dose of study drug and ≥1 postbaseline Hb or TFN value). 305 (151 EPO, 154 DARB) of these pts had completed 4 wks of therapy by the same date and were included in the IA. The IA demonstrated a statistically significantly higher HRR for EPO pts, 47%, compared with DARB pts, 33%, P=.0078 (1-sided). Enrollment was terminated; some patients remain on study. Demographics for all randomized patients were similar for EPO/DARB (mean age: 62/63 y; men: 38/31%; ECOG PS 0–1: 86/82%; and nonplatinum CT: 62/58%). Most common tumor types were breast, lung, and colorectal. In the MITT population, mean BL Hb was 10.1 g/dL in each group and Hb increases for EPO and DARB, respectively, were 0.7 and 0.3 g/dL after 4 wks, 1.0 and 0.5 g/dL after 8 wks, 1.3 and 0.7 g/dL after 12 wks, and 1.3 and 0.7 g/dL after 16 wks; last-value-carried-forward method used to impute missing Hb values and Hb values within 28 days following RBC TFN. After study day 28, 22/170 (13%) EPO pts and 28/169 (17%) DARB pts received RBC or whole blood TFN. Available data suggest a similar safety profile for the 2 agents. This study demonstrated that pts treated with EPO 40,000 U SC QW had a significantly greater HRR during the first 4 wks of therapy compared with DARB 200 μg SC Q2W. In addition, EPO-treated pts had greater Hb increases throughout the study compared with DARB.