HLA-DPB1 TCE mismatch vector differentially affects HCT outcomes in 10/10 and ≤9/10 HLA matched patients in the UK.

BACKGROUND

Understanding permissive and beneficial HLA mismatches is a primary area of hematopoietic cell transplant (HCT) research. HLA-DPB1 is well studied in this context, due to its separation from HLA-A∼DQB1 by a recombination hot-spot.

OBJECTIVES

The primary study aim was to retrospectively assess the clinical impact of using the latest iteration of the HLA-DPB1 T cell epitope model (TCEv3) for selecting HLA-DPB1 mismatched donors in a UK cohort. The results were also compared to the previous, TCEv2 model.

STUDY

DESIGN

We used the TCEv3 model to assign core and non-core permissive mismatches, to a cohort of 1743 UK volunteer unrelated donor HCTs for malignant diseases between 1996 and 2020. We considered HLA-DPB1 mismatch vector (graft versus host (GvH), host versus graft (HvG), and bi-directional (Bi-D)) across TCE groups, and existing HLA-A∼DQB1 mismatching.

RESULTS

In 10/10 HLA matched patients, Bi-D HLA-DPB1 non-core permissive/non-permissive mismatches were associated with significantly worse 5-year overall survival, transplant related mortality, progression-free survival, chronic graft versus host disease (GvHD) and GvHD-free, relapse-free survival (GRFS) compared to HLA-DPB1 matched/permissively mismatched patients. In ≤9/10 HLA matched patients, both Bi-D and HLA-DPB1 matched/permissive mismatches were associated with similarly poor survival, while HvG HLA-DPB1 mismatches reduced mortality and improved GRFS. Overall risk of grades 2-4 acute GvHD (aGvHD) increased with the presence of GvH HLA-DPB1 mismatches and HLA-A∼DQB1 mismatching.

CONCLUSIONS

Assessing mismatch vector revealed that some canonical non-permissive HLA-DPB1 mismatches are equivalent to an HLA-DPB1 match/permissive mismatch and can be beneficial. Similarly, some canonical permissive mismatches and, surprisingly, HLA-DPB1 matching in an already mismatched setting were detrimental. Bi-D HLA-DPB1 mismatched patients consistently had the poorest outcomes. This UK first, multicenter study shifts our understanding of HLA-DPB1 permissibility in UK HCT, with potential to improve patient survival and quality of life post-HCT through alternative donor selection.