Abstract Most candidates for reduced-intensity allogeneic stem cell transplantation (RIC) do not have a human leukocyte antigen (HLA)-matched sibling donor available. Improved HLA-typing techniques allow more stringent matching of unrelated donors (UD), but have reduced the likelihood of identifying a fully matched UD (MUD), defined as a 10/10 locus match by molecular typing. Whilst donors matching at 8–9/10 HLA loci (MMUD) are considered appropriate by some groups, it is unclear the degree to which HLA-mismatch increases procedural toxicity and how this is influenced by different regimens. Availability of alternative stem cell sources such as umbilical cord blood makes it important to clearly define such risks. We therefore analyzed outcomes of 99 consecutive patients who underwent RIC using an UD in our institution from October 1998 to July 2008. Donor selection was performed according to standard criteria, including molecular typing for HLA-A, -B, -C, DRB1 and DQB1. Donors were 10/10 HLA-matched (n=62) or mismatched at up to 2 HLA loci (n=37). Conditioning was 20mg/d alemtuzumab for 5 days, fludarabine 30mg/m2 for 5 days and melphalan 140mg/m2 for 1 day, with cyclosporine from day -1. Peripheral blood stem cells were used in 51 (30 MUD, 21 MMUD) and bone marrow in 48 (32 MUD, 16 MMUD). Median age was 47 years (13– 67), and underlying disease was AML (n=9), myeloma (n=13), Hodgkin (n=18) or non- Hodgkin lymphoma (n=40), and others (n=19). Patients had a median of 4 previous lines of treatment (0–9) and 49 had failed a prior autograft. There were no significant differences between the groups in terms of age, sex, prior autograft, number of treatment lines, or chemosensitivity at transplantation. There were significantly more CMV seropositive recipients in the MMUD group (P.002). Neutrophil recovery (>0.5 × 109/l) occurred at a median of 12 days (9–57) and did not differ between MUD and MMUD (median 11 vs 12 days). Two patients, both MUDs, had primary graft failure and engrafted following CD34 top-up. Three patients rejected the graft, all MMUDs (P.07 for MUD vs MMUD). Secondary graft failure occurred in 3 cases, all MMUDs (P.07 for MUD vs MMUD). All 3 received CD34 top-up with subsequent improvement in counts in one, another died shortly after infusion of cells and 1 remains cytopenic 4 years following the transplant. Of 40 at risk cases, CMV reactivation requiring treatment occurred in 35 (88%; 85% MUD and 90% MMUD). Donor lymphocyte infusions (DLI) were administered in 35 cases, 23 for disease relapse and 12 for mixed chimerism alone, with no significant difference between MUD (n=25, 40%) and MMUD (n=10, 27%, P.4). There was no difference in the incidence of grade 2–4 acute graft-versus-host disease (aGvHD) between MUD (27%) and MMUD (30%, P.7). Prior to DLI, only 3 cases of grade 3 aGvHD occurred (2 MUD and 1 MMUD) and no grade 4 aGvHD. Chronic GvHD (cGvHD) at 1 year occurred in 30%, with a non-significant trend for a lower incidence in MUD versus MMUD patients (24% vs 40%, P.1). Extensive cGvHD at 1 year was 15% (MUD 13% and MMUD 18%, P.6) excluding post-DLI GvHD, and was 26% at 1 year (MUD 26% and MMUD 27%, P.4) and 40% at 3 years (MUD 34% and MMUD 46%) including post-DLI GvHD. With a median follow-up of 3.0 years, transplant related mortality (TRM) for the whole cohort was 14% at 100 days and 28% at 1 year, with no significant difference between MUD (13% and 30%) and MMUD (16% and 27%). Overall survival (OS) at 3 years was 46% for the whole cohort (45% for MUD, 48% for MMUD). The outcome of 1 (n=20) versus 2 (n=17) locus mismatched transplants was also compared. There was a non-significant trend to increased grade 2–4 aGvHD (16% versus 39%, P.2) in the 2 locus MMUD group but extensive cGvHD (including post-DLI GvHD) at 1 year was not different (30% vs 24%, P.6). Survival was not inferior in the 2 locus MMUD group relative to the single mismatches, with 100 day TRM (20% 1 locus vs 13% 2 locus, P.5), 1-year TRM (43% 1 locus vs 19% 2 locus, P.5), and 3 year OS (32% 1 locus vs 63% 2 locus, P.3). We conclude that 8–9/10 MMUD-RIC is a viable option using T-cell depletion with 100mg alemtuzumab in vivo, without a significant adverse impact on TRM or OS compared with 10/10 MUD. The long-term OS of 48% following MMUD-RIC is encouraging given the inclusion mainly of patients with multiply relapsed/refractory hematological malignancy.