Association of HLA-C disparity with graft failure after marrow transplantation from unrelated donors.

Disparity for HLA-A or HLA-B antigens increases the risk of marrow graft rejection, but the relevance of HLA-C is unknown because typing methods have not been sufficiently accurate for clinical use. We designed a matched case-control study and employed DNA sequencing methods to evaluate the role of HLA-C disparity in 21 patients who experienced graft failure (cases) following transplantation with unmanipulated marrow from either HLA-A, B serologically matched, DRB1 matched (n = 14) or single locus mismatched (n = 7) unrelated donors. For each case, two patients who successfully engrafted were selected as controls based on similarity for factors known or suspected to influence engraftment. The estimated odds ratio (OR) of graft failure for an HLA-C mismatch relative to match (univariable model) was 5.2 (95% CI, 1.4, 19; P = .01). Serologically undetectable HLA-A or HLA-B allele disparity was also associated with graft failure. The association between HLA-C disparity and graft failure remained significant even after accounting for the contribution of HLA-A and/or HLA-B allele disparity (OR 4.0; 95% CI, 1.1, 15; likelihood ratio test P = .03). These results show that HLA-C functions as a transplantation antigen and that HLA-A and HLA-B allele mismatches are biologically important. Molecular-based methods for pretransplant assessment of class I compatibility should be implemented for the selection of unrelated marrow donors.