There are multiple lines of evidence suggesting that human recombinant erythropoietin (rEPO) could influence immune responses by direct effects of rEPO on T or B cells. The present study tested this hypothesis by measuring antibody responses after immunization to tetanus toxoid (TT, a T cell dependent antigen) or pneumococcal capsular polysaccharide antigen (PA, a T cell independent antigen). The patients chosen for this prospective study were chronic hemodialysis patients receiving chronic rEPO therapy, and a comparable group of chronic hemodialysis patients not receiving rEPO therapy. We found that the patients immunized with PA and receiving rEPO therapy (N = 15) had IgG anti-PA responses comparable to that of those not receiving rEPO therapy (N = 15). In contrast, in the patients immunized with TT, those receiving rEPO (N = 15) developed significantly higher IgG anti-TT levels than those not receiving rEPO (N = 14) (time-group interaction P = 0.005). The peak difference between these groups was at two weeks, where the rEPO-treated patients developed a 4.1-fold mean increase in IgG anti-TT level and those not receiving rEPO developed only a 1.4-fold mean increase in IgG anti-TT level (P < 0.01). The difference in immune response to TT in the rEPO compared to the non-rEPO-treated patients could not be explained by differences between the groups in any of the parameters measured at baseline or during the post-immunization period. In conclusion, rEPO therapy increased immune response to TT but not PA, which suggests that rEPO enhances immune response to T cell dependent antigens.