Outcome of Hematopoietic Cell Transplant from HLA-Matched Siblings Compared to a Volunteer Unrelated Donors Matched for HLA-A, B, C, DRB1, and DQB1 Alleles.

Hematopoietic cell transplantation from an HLA-matched unrelated donor (URD) is an alternative when an HLA-matched sibling donor (MSD) is not available. Matching of donor HLA alleles is associated with improved outcome after unrelated HCT compared to mismatching. Whether HLA matching can result in outcomes similar to those observed after HLA-identical sibling HCT remains an important question. We compared clinical outcomes after HCT in 704 recipients with HLA-A, B, C, DRB1, DQB1 allele matched URD and 1211 recipients of HLA-identical MSD. Patients received myeloablative conditioning for treatment of AML, ALL, CML, or MDS. The two groups were similar in phase of disease (advanced: acute leukemia>CR1, CML>CP,MDS>RA; early: all others), source of stem cells, age, and patient CMV serology. Male patients with a female donors occurred more frequently in the MSD group (27%) relative to the URD group (18%), and MSD were more often CMV-seropositive (52%) than URD (29%). Multivariable regression models were fit for the endpoints overall survival (OS), acute graft-vs-host disease (GVHD), and relapse, adjusting for each of the factors listed above. Overall, 350 (50%) of URD and 597 (49%) of MSD recipients survive for a median of 7.1 years follow-up (range 0.1–15.2), leading to an adjusted hazard ratio (HR) for mortality of 1.16 (confidence interval (CI), 1.00–1.34, p=.05). The relative risk was similar among pediatric (<18 years) and adult patients (interaction test, p>0.90), but there was a suggestion that the risk differed according to phase of disease (interaction test, p=0.07). In particular, the adjusted HR in early disease phase was 1.39 (CI, 1.08–1.78, p=.01) compared to 1.09 (CI, 0.91–1.30, p=0.37) in advanced phase. Grades II–IV GVHD occurred in 83% of URD and 67% of MSD patients, for an adjusted odds ratio (OR) of 2.37 (CI, 1.84–3.06, p<0.0001), while grades III–IV GVHD occurred in 25% and 18% of recipients, respectively (adjusted OR=1.98 (CI, 1.52–2.57, p<0.0001)). As with mortality, there was a suggestion that the difference in GVHD between URD and MSD differed according to disease phase (p=0.07 for grades II–IV, p=0.04 for grades III–IV). The OR for grades II–IV GVHD was 2.79 (CI, 1.94–4.01, p<0.0001) among patients with early phase disease and 1.99 (CI, 1.38–2.88, p=0.0003) among those with advanced phase, while OR for grades III–IV GVHD was 2.58 (CI, 1.75–3.80, p<0.0001) among early phase and 1.59 (CI, 1.10–2.30, p=0.01) among advanced phase patients. Relapse was observed in 25% of unrelated recipients compared to 29% of sibling recipients (HR=0.88; CI, 0.72–1.08; p=0.22). The association of relapse with donor type was not significantly affected by disease phase (interaction test, p=0.84). When an unrelated donor is matched for HLA-A, B, C, DRB1, and DQB1 with the recipient, survival after HCT is similar to that with an HLA-identical sibling donor for patients with advanced disease despite the increased incidence of severe GVHD. Mortality is higher in the URD group for patients with early disease phase. While the difference in mortality for early disease phase may reflect the importance of alloreactivity among patients with few other competing risks for mortality, selection bias for patients with unfavorable prognostic features also may contribute to higher mortality in the URD group.