Recently, the administration of high-dose cyclophosphamide (Cy) after T cell–replete haploidentical stem cell infusion has been reported to be feasible and effective. In the original study, bone marrow (BM) was used as the source of stem cells. Here, we retrospectively analyzed the use of BM versus peripheral blood stem cells (PBSCs) in a cohort of patients receiving haploidentical T cell–replete transplantation after a nonmyeloablative conditioning regimen with postinfusion Cy. In the PBSC versus BM groups, the incidence of acute graft-versus-host disease (GVHD) was 33% versus 25%, respectively, and the incidence of chronic GVHD was 13% versus 13%, respectively. The median time to achieve a safe and unsupported absolute neutrophil and platelet count was 20 versus 21 days and 27 versus 29 days, respectively. The incidence of engraftment was also similar in the 2 cohorts. The 1-year nonrelapse mortality rate was 12% versus 22%, respectively (P = .96). Finally, nonsignificant differences in survival were observed. In conclusion, the use of PBSCs instead of BM after T cell–replete haploidentical transplantation did not appear to be detrimental in terms of either GVHD or engraftment rate. PBSCs could be a valid alternative to BM after transplantation from a haploidentical donor using postinfusion Cy.