BACKGROUND Recombinant human erythropoietin is indicated for the treatment of anemia resulting from chronic renal failure or chemotherapy. It is also used for patients at high risk for transfusions because of significant blood loss during surgery. A new recombinant human erythropoietin (epoetin alfa) that excludes fetal bovine serum and human serum albumin from among its ingredients was developed in Korea. This study was planned as part of a product development project at the request of the Korean regulatory agency. OBJECTIVE The aim of this study was to compare the pharmacokinetic and pharmacodynamic characteristics of a new recombinant human erythropoietin (test) formulation with an existing branded (reference) formulation after a single subcutaneous administration. METHODS An open-label, sequence-randomized, 2-period, 2-sequence, 2-treatment crossover study was conducted. Healthy male subjects were randomly assigned with a random number table into 1 of 2 sequence groups, and each subject was given recombinant human erythropoietin 4000 IU SC in the upper arm as the test formulation in one period and the reference formulation in the other period, according to the sequence group. Each period was separated by a 4-week washout period. Serial blood samples were taken up to 120 hours after drug administration for the pharmacokinetic assessments and up to 240 hours for reticulocyte counts as the pharmacodynamic end point. Pharmacokinetic analysis was performed without baseline correction. Adverse events (AEs) were collected by spontaneous reporting of the subjects or solicited by asking general health-related questions. RESULTS Twenty healthy men (mean [range] age, 25.6 [21-36] years; height, 175 [167-187] cm; weight, 70 [57.6-85.5] kg) were enrolled in and completed the study. The mean (SD) baseline erythropoietin plasma concentrations were 10.4 (2.4) mIU/mL for the test formulation and 10.8 (3.5) mIU/mL for the reference formulation. After the injection of 4000 IU SC per subject, the erythropoietin plasma concentrations reached a maximum at a median T(max) of 10 hours for both formulations (range: test formulation, 7.00-95.95 hours; reference formulation, 6.98-24.13 hours). The mean (SD) C(max) values for the test and reference formulations were 74.34 (30.63) and 80.46 (30.56) mIU/mL, respectively; the mean AUC(0-last) values were 3664 (731.5) and 3553 (723.2) mIU·h/mL. The ratios of the geometric mean (test/reference) for C(max) and AUC(0-last) were 0.92 (90% CI, 0.81-1.05) and 1.03 (90% CI, 0.98-1.09). The mean baseline hemoglobin, hematocrit, and reticulocyte counts were 15.4 g/dL, 45.5%, and 49.6 · 10(3)/μL, respectively, for the test formulation and 15.5 g/dL, 45.3%, and 47.5 · 10(3)/μL for the reference formulation. The mean reticulocyte counts slowly reached T(max) for both formulations at a median of 120 hours after administration (test formulation, 120.0 hours [range, 95.5-240.8 hours]; reference formulation, 120.1 hours [range, 72.0-240.5 hours]). The mean (SD) maximum reticulocyte counts for the test and reference formulations were 77.7 (12.2) · 10(3)/μL and 80.7 (15.2) · 10(3)/μL, respectively; values for area under the effect curve to the last observation (AUEC(0-last)) were 14,781.5 (2439.2) · 10(3)/μL · h and 14,783.8 (2415.4) · 10(3)/μL · h. The 2 agents did not exhibit any significant differences in maximum reticulocyte counts or AUEC(0-last). During the study, a total of 6 AEs were reported, which were mild in severity. After the administration of test formulation, 1 case each of rhinorrhea, epigastric discomfort, and joint sprain (left ankle) were reported. After the administration of reference formulation, 2 cases of rhinorrhea and 1 case of cough were reported. CONCLUSIONS In this small, selected group of healthy male volunteers, there were no significant differences in pharmacokinetic parameters or effects on reticulocytes between a test formulation and a reference formulation of recombinant human erythropoietin.