Impact of Human Leukocyte Antigen Epitope Matching on Outcomes after Unrelated Bone Marrow Transplantation

Background

The immunological effect of human leukocyte antigen (HLA) disparity has not been fully elucidated by the number and locus of HLA antigens or alleles. Several methods to predict epitopes recognized by the immune system have been developed to understand the immunogenicity of amino acid sequences in mismatched HLA pairs. We investigated the association between mismatching of HLA antibody-identified epitopes and hematopoietic stem cell transplantation (HSCT) outcomes.

Patients and Methods

This was a retrospective cohort study with 9,991 patients who underwent their first HSCT for hematologic malignancies from unrelated bone marrow donors in the Transplant Registry Unified Management Program (TRUMP). HLA epitope mismatches (EMM) were quantified using HLA-matchmaker (HLAMM). We conducted a multivariate analysis using Cox proportional hazard regression for the overall survival and the Fine-Gray regression model for competing risks. Statistical analyses were performed with Stata version 15.1 and R version 4.0.2.

Results

The median follow-up period for survivors was 6.3 years (interquartile range [IQR], 2.5-9.2 years). The most common indication for HSCT was acute myeloid leukemia (n=3,917, 39.2%), followed by acute lymphoblastic leukemia (n=1,913, 19.2%), and mature lymphoid malignancies (n=1,906, 19.1%). HSCT from HLA-matched, HLA 1 allele mismatched, and HLA 2 or more allele mismatched donors was received by 6,200, 2660, and 1131 patients, respectively. The number of EMM in recipient-donor pairs in our study population ranged from 0 to 37 in HLA class I (median, 0) and 0 to 60 in HLA class II (median, 1). Patients were categorized into two groups, low and high EMM, using the median value for each epitope matching as the threshold.

Higher HLA class I EMM in the graft-versus-host (GVH) direction was associated with a significantly higher risk of grade III-IV acute GVHD (aGVHD) (hazard ratio [HR] 1.69, 95% confidence interval [CI] 1.21-2.36). Higher EMM for both class I and class II in the host-versus-graft (HVG) direction was associated with a significantly longer time to neutrophil engraftment (HR 0.78, 95% CI 0.65-0.95). In subgroup analysis limited to HLA 1 allele mismatch, class I high EMM group in the GVH direction had a significant association with higher risk for grade II-IV and grade III-IV aGVHD compared with both class I and class II low-EMM group (HR 1.69, 95% CI 1.16-2.47).

Patients with HLA-C EMM accounted for 94.5% (n=1,603) of patients with HLA class I EMM. We further investigated the impact of HLA-C on severe aGVHD in relation to other known high-risk mismatch patterns. All killer immunoglobulin-like receptor (KIR)-ligand mismatched recipient-donor pairs (n=376) had HLA-C EMM. In multivariate analysis, patients with KIR-ligand mismatches and EMM did not show a higher incidence of grade III-IV aGVHD compared with KIR-ligand-matched patients with EMM (HR 0.96, 95% CI 0.74-1.25). In addition to the known high-risk mismatch patterns in the Japanese cohort (Kawase et al. Blood. 2007, Morishima et al. Haematologica. 2016), EMM was associated with a higher risk for grade III-IV aGVHD (Figure 1A, compared with HLA-C allele-matched patients (Match), HLA-C allele-mismatched patients without antigen mismatches, and EMM (group A): HR 0.78, 95% CI 0.41-1.48; HLA-C antigen-mismatched patients without EMM (group B): HR 0.56, 95% CI 0.28-1.15; HLA-C epitope-mismatched patients without high-risk mismatches (group C): HR 1.67, 95% CI 1.44-1.95; Patients with high-risk mismatches other than patient mismatched HLA-C*14:02 (group D): HR 2.01, 95% CI 1.50-2.69; Patients with patient mismatched HLA-C*14:02 (group E): HR 3.38, 95% CI 2.39-4.78). HLA-C epitope-mismatched patients without high-risk mismatches also showed a higher incidence of non-relapse mortality (NRM) and lower overall survival (OS) than HLA-C allele-matched patients (NRM (Figure 1B): HR 1.39, 95% CI 1.25-1.54; OS (Figure 1C): HR 1.20, 95% CI 1.10-1.30).

Conclusion

HLAMM-based epitope matching might be useful for identifying high-risk groups who can develop serious complications after HSCT from HLA-mismatched unrelated donors. In the HLA-C locus, epitope-mismatched recipient-donor pairs are non-permissive mismatched patterns along with known high-risk amino acid substitutions. Our findings might be helpful for clinicians in selecting permissive donors from alternative donor options.

Figure 1

View largeDownload slide

Figure 1

View largeDownload slide

Close modal