The influence of graft composition on the outcome of reduced-intensity (RIC) allogeneic PBSC transplantation (allo-PBSC) remains controversial. In this study, we analyzed the impact of CD34 þ cell dose on the incidence of GVHD, and on the outcome after allo-PBSC, in 103 patients with hematological malignancies, using a uniform RIC regimen. The following variables were included in statistical analysis: (1) number of C34 þ cells, (2) highrisk vs low-risk disease status, (3) matched related vs matched unrelated donor, (4) female donor to male recipient vs any other combination, (5) age of recipient (above vs below the median). Univariate and multivariate analysis did not reveal any association between CD34 þ cell dose and acute grade-2 to grade-4, cGVHD, nonrelapse mortality (NRM), relapse rate (RR) and OS. High-risk disease status was the only variable independently associated with increased NRM (P ¼ 0.001), increased RR (P ¼ 0.012) and decreased OS (Po0.001). The same results were obtained when analysis was restricted to a subgroup of 55 patients with myeloid neoplasms. The influence of graft composition on the outcome of RIC allo-PBSC should be further investigated via well-controlled randomized prospective studies.