The role of HLA-E polymorphism in the outcome of unrelated hematopoietic stem cell transplantation : retrospective analysis of two large independent acute leukemia cohorts

Abstract

More than 50 years after its first implementation for the treatment of acute leukemia, allogeneic hematopoietic stem cell transplantation (HSCT) is still considered the only therapeutic option with life-long curative potential. Despite the undeniable progress made, HSCT remains a high risk treatment even in the optimal setting of 10/10 Human Leukocyte Antigen (HLA) compatibility between recipient and donor. In the search of immunogenetic factors beyond classical HLA antigens possibly impacting HSCT outcome, a limited number of studies have investigated the role of HLA-E in HSCT outcome. HLA-E is a ubiquitously transcribed, non-classical Major Histocompatibility Complex (MHC) Ib molecule with a multifaceted yet not fully understood role in both innate and adaptive immunity. Aim of this study was to investigate for the first time on such a large scale the effect of HLA-E polymorphism on HSCT outcome in a 10/10 HLA matched unrelated acute leukemia setting analyzing data from two large independent cohorts. The two cohorts (German cohort n=509, CIBMTR (Center for International Blood and Marrow Transplant Research) cohort n=1840) analyzed within this study consisted of adult acute leukemia patients receiving their first 10/10 HLA matched unrelated graft between 2000 and 2015. Both recipients and donors were HLA-E genotyped by Sanger and Next Generation Sequencing (NGS) using in-house kits in our laboratory in Ulm. Overall survival (OS), disease free survival (DFS), relapse (RI), transplant-related mortality (TRM), acute GvHD (aGvHD) and chronic GvHD (cGvHD) were evaluated; p<0.05 and p< 0.01 were considered significant in the German and CIBMTR cohort analysis, respectively. The effect of patient and donor HLA-E genotype as well as of patient/donor HLA-E match grade were assessed using univariate Kaplan-Meier (KM), multivariate Cox regression, and competing risks analyses in both cohorts. In the CIBMTR cohort the joint effect of donor/recipient (D/R) HLA-E genotype was additionally explored through a 4-level analysis of D/R HLA-E*01:03 homozygous vs other. The HLA-E allele frequencies as well as the relatively high rate of HLA-E discrepant transplant pairs (>32%) observed in both cohorts were in accordance with previously reported data. Regarding the effect of HLA-E polymorphism on HSCT outcome, the German cohort analysis revealed a significant correlation between HLA-E mismatch and improved HSCT outcome mainly in advanced disease patients. Specifically, patients transplanted by HLA-E mismatched donors exhibited significantly better OS and DFS rates due to lower TRM. The CIBMTR cohort analysis did not confirm this favorable effect, as HLA-E match status between recipient and donor did not have any impact on any outcome endpoint. With respect to the effect of recipient and/or donor HLA-E genotype on HSCT outcome, both cohort analyses identified the HLA-E*01:03/01:03 as an unfavorable prognostic factor. The results of the CIBMTR analysis as to this finding were significantly more enlightening indicating that the HLA-E*01:03/01:03 negative effect is mainly donor driven, with the patient HLA-E genotype effect being only weakly synergistic. Specifically, donor HLA-E*01:03/01:03 genotype was found to associate with significantly lower DFS, apparently due to higher relapse incidence rate. This study is the largest to date investigating the effect of HLA-E polymorphism on HSCT outcome. The main conclusions drawn are that: HLA-E matching between patients and donors in the setting of unrelated HSCT for acute leukemia confers no advantage to outcome; HLA-E incompatibility between patients and donors may improve unrelated HSCT outcome in a subset of advanced disease patients receiving anti-thymocyte globulin and; avoidance of HLA-E*01:03/01:03 donors may improve DFS prognostics in patients in complete remission undergoing first matched unrelated HSCT for acute leukemia. Although no definitive conclusions can be drawn before molecular-based studies shed light on the mechanisms implicated, these observations certainly set the basis for future consideration of HLA-E as relevant clinical prognostic factor in a matched unrelated HSCT setting for acute leukemia patients.