Comparable Survival and Transplant-Related Mortality Following Allogeneic Hematopoietic Cell Transplantation from HLA Allele-Matched Unrelated and HLA-Identical Sibling Donors.

Comprehensive analyses of unrelated donor (URD) and recipient HLA-matching for allogeneic hematopoietic cell transplantation (AHCT) have demonstrated better outcomes when allele typing is performed using high-resolution nucleotide sequence-based techniques. To evaluate survival following myeloablative AHCT using allele-level HLA-matched URD as compared with HLA-identical sibling donors, we analyzed outcomes for 430 patients treated at our center between March 1991 and April 2005. Sequence-based allele typing was retrospectively performed for HLA-A, B, C, DR and DQ when not done at time of AHCT for URD (n = 124; 29%) and non-sibling related donors (n = 19; 4%). Donors were HLA-identical siblings (n = 276; 64%), HLA allele-matched URD (n = 52; 12%), single HLA-locus mismatched donors (n = 52; 12%), or > 1 locus mismatched donors (n = 50; 12%). The median age at transplant was 23.4 years (range: 0.2 – 61). The most common diagnoses were AML (n = 107; 25%), CML (n = 90; 21%), ALL (n = 86; 20%) and MDS (n = 50; 12%). Total body irradiation-based preparative regimens were used for 283 patients (66%). Bone marrow (BM) was the graft for 388 patients (90%) and GCSF-mobilized peripheral blood stem cells (PBSC) for the remaining 42 (10%). Graft-versus-host disease (GVHD) prophylaxes were cyclosporine and methotrexate (n = 327; 76%), long methotrexate (n = 42; 10%), T-cell depletion (n = 19; 4%), or other regimens (n = 42; 10%). With a median follow-up of 4.8 years (range: 0.2 – 12.1), the 5-year estimate of overall survival (OS) for the entire group was 48.2% (95% CI: 45.7 – 50.7) and transplant-related mortality (TRM) was 31.4% (95% CI: 28.8 – 34). As shown in the Table, OS and TRM were indistinguishable between AHCT performed with HLA-identical siblings compared with HLA allele-matched URD. There was also no difference in grade III – IV acute GVHD (P = .46) between these two groups whereas there was a trend towards more extensive chronic GVHD (HR 1.8; 95% CI: 0.9 – 3.6; P = 0.12) for the URD recipients. Using a multivariate analysis to adjust for advanced disease, age (> vs ≤ 30 years), graft (BM vs PBSC) and female-to-male gender mismatch, there remained no difference in OS between HLA-identical siblings and HLA allele-matched URD (P = 0.67). These results demonstrate that key outcomes (OS, TRM, and severe acute GVHD) are equivalent in recipients of grafts from either allele-level 10/10 HLA-matched URD or HLA-identical siblings.