Impact of HLA-Mismatch in Unrelated Donor Hematopoietic Stem Cell Transplantation: A Comprehensive Meta-Analysis

Abstract

Mismatched unrelated donor (MMURD) transplantation is widely used in patients who lack a matched unrelated donor (MUD). Despite several published observational studies, the magnitude of risk associated with 9 out of 10 MMURD transplant and that of HLA-specific mismatches is still unclear. We performed a meta-analysis to assess the difference in outcomes of MUD to those of MMURD transplant. A comprehensive search of Medline and EMBASE was performed through April 2014. We included published manuscripts that reported outcomes of adult patients with hematologic malignancies who underwent MUD (matched at HLA-A, B, C and DRB1 at a minimum) or MMURD transplant (single antigen or allele mismatch only). The pooled relative risks (RR) from hazard ratios and risk ratios and 95% confidence intervals (CI) were estimated using a DerSimonian-Laird random-effects model. Heterogeneity was tested using the I2 statistic and Cochran Q test, and sources of heterogeneity were evaluated using univariate meta-regression. Statistical analyses were performed using Stata v13.1. Twenty-two studies were included, representing 38,877 transplants. Nine out of 10 MMURD transplant was associated with a higher risk of overall mortality compared with 10 out of 10 MUD transplant (n=10; RR 1.32, 95% CI 1.17–1.49). No significant sources of heterogeneity were identified. MMURD transplant was associated with worse disease-free survival compared with MUD transplant (n=6; RR 1.20, 95% CI, 1.08–1.33). The pooled effect estimates for grade II-IV acute and chronic GVHD were 1.23 (n=4; 95% CI 0.99–1.53) and 1.08 (n=3; 95% CI 0.76–1.54) respectively. Single antigen/allele mismatches at HLA-A, B, C or DRB1 were associated with a higher risk of grade III-IV acute GVHD and overall mortality, but not chronic GVHD as compared with MUD transplant (Table). Disease-free survival was only worse for HLA-B mismatch. MMURD transplant is associated with significantly higher risk of overall mortality and acute GVHD, but not chronic GVHD compared with MUD transplant. This risk is associated with HLA-A, -B, -C, and –DRB1, but not -DQB1 and -DPB1 mismatches.TableClinical outcomes of 9/10 MMURD versus 10/10 MUD by HLA-specific allele.Pooled RR (95% CI)Grade II-IVAcute GVHDGrade III-IVAcute GVHDChronicGVHDDisease-FreeSurvivalOverallSurvivalHLA-A1.12 (0.75-1.66)1.57 (1.29-1.90)0.99 (0.68-1.45)1.20 (1.00-1.45)1.47 (1.31-1.65)HLA-B1.41 (0.63-3.12)1.67 (1.37-2.03)0.87 (0.56-1.36)1.20 (1.04-1.38)1.33 (1.18-1.51)HLA-C1.16 (1.04-1.28)1.44 (1.25-1.65)1.04 (0.94-1.16)1.01 (0.73-1.38)1.28 (1.13-1.46)HLA-DRB11.60 (1.23-2.09)1.39 (1.13-1.72)1.11 (0.68-1.83)1.16 (0.94-1.44)1.36 (1.22-1.51)HLA-DQB10.89 (0.37-2.14)insufficient datainsufficient data0.99 (0.71-1.37)0.95 (0.69-1.31)HLA-DPB12.68 (1.16-6.19)1.20 (0.94-1.52)1.06 (0.76-1.46)0.85 (0.75-0.97)0.98 (0.89-1.07) Open table in a new tab