HLA-C matching is an important determinant of outcome after myeloablative unrelated donor (URD) hematopoietic stem cell transplantation. However, its importance in non-myeloablative stem cell transplantation (NST) is not known. We report a retrospective analysis of 111 patients who underwent URD NST, of whom 78 were 10/10 matched at HLA-A, B, C, DRB1, DQB1 and 33 were mismatched at one or more HLA-C antigen/allele (24 HLA-C only; nine HLA-C+other locus mismatch). Patients were conditioned with busulfan (0.8 mg/kg/day i.v. x 4 days) and fludarabine (30 mg/m(2)/day i.v. x 4 days). Graft-versus-host disease prophylaxis included cyclosporine/prednisone- or tacrolimus/mini-methotrexate-based regimens. HLA-C disparity did not impair engraftment. Median marrow donor chimerisms were >or=90% donor at day+30 and +100 in both groups. Overall survival at 2 years was 30% in HLA-C-mismatched and 51% in 10/10-matched patients (P=0.008). In Cox regression, HLA-C mismatch was an independent predictor of death (hazard ratio 1.85, P=0.04). Treatment-related mortality was higher in the HLA-C-mismatched group: 48 versus 16% (P=0.0001). Cumulative relapse incidence was 35% in the HLA-C-mismatched and 55% in the 10/10-matched cohort, P=0.09. HLA-C mismatch is associated with inferior survival after URD NST.