A Deep Dive Into HLA Typing For Unrelated Donor Hematopoietic Cell Transplantation

This podcast comments on a large registry study evaluating the effect of ultra-high-resolution HLA typing on outcomes of unrelated donor transplantation.   TRANSCRIPT This JCO Podcast provides observations and commentary on the JCO article 'Impact of Previously Unrecognized HLA Mismatches Using Ultra-High-Resolution Typing in Unrelated Donor Hematopoietic Cell Transplantation" by Mayor et al. My name is Navneet Majhail, and I am the Director of the Blood and Marrow Transplant Program and the Vice-Chair for the Department of Hematology and Medical Oncology at the Taussig Cancer Institute, Cleveland Clinic. My oncologic specialty is transplantation and cellular therapy.    For patients who are potential candidates for allogeneic hematopoietic cell transplantation, the first critical step is finding an HLA-matched donor. The HLA genes are located within one of the most gene rich regions of the human genome, are highly polymorphic, and encode proteins that critically modulate the body's immune responses against a variety of stimuli. When selecting an unrelated bone marrow or peripheral blood stem cell donor, we typically try to identify donors who are matched at HLA genes where polymorphism is predominantly present, namely, class I genes HLA-A, -B, -C, and class II genes HLA-DRB1, -DQB1 and –DPB1. Guidelines recommend using a donor who is matched at least at HLA-A, -B, -C, and –DRB1, and preferentially at -DPB1 and -DQB1 as well.    Graft-versus-host disease or GVHD is an immune-mediated complication that continues to be a major threat to successful patient outcomes after hematopoietic cell transplantation. Better matching between the donor and the recipient lowers risk of GVHD, and guidelines recommend use of an HLA 8/8 matched unrelated donor, though in clinical practice we prefer an HLA 10/10 and even a 12/12 matched donor if one is available. In the past, HLA typing methods used 'antigen-level' serological testing. However, with advances in technology, the field has moved on to 'allele-level' or high-resolution typing which can discriminate among HLA genes that encode cell-surface proteins that ultimately constitute the antigen recognition domain, which is the functionally active portion of the HLA molecule that ultimately interacts with T-cell and NK-cell receptors. Research has shown that matching at allele-level is associated with lower risks of GVHD and better survival compared to historical serotyping-based methods, and DNA-based HLA-typing is the current standard of care.    Further advances in technology to next-generation sequencing or ultra-high-resolution typing can now allow characterization of the full HLA gene sequence. This has raised the question of the clinical significance of HLA polymorphisms in regions outside the antigen recognition domain. Prior studies in smaller cohorts of patients have raised the possibility that transplants using ultra-high-resolution matched donors may be associated with better survival and lower risks of acute GVHD. To definitively validate these findings, Mayor et al conducted a study in a cohort of >5,000 allogeneic transplant recipients from the Center for International Blood and Marrow Transplant Research. Patients had received a matched unrelated donor transplant for acute leukemia or MDS between 2008 and 2017. The manuscript that accompanies this podcast provides details of their study population, but overall their cohort was fairly representative of unrelated bone marrow and peripheral blood stem cell transplant recipients.     To summarize some key findings of their study, first, among donor-recipient pairs deemed HLA 10/10 match using high-resolution typing, 18% were found to not be a 10/10 match on ultra-high-resolution typing. Overall, only 12% of patients had a 12/12 ultra-high-resolution matched donor. Second, overall survival was comparable between patients receiving 12/12 ultra-high-resolution matched and mismatched transplants. Furthermore, there was no association of survival with the degree of ultra-high-resolution mis-match, that is, the number of loci where there was a mismatch. Similarly, when considering a subgroup of patients who were ultra-high-resolution matched at 10/10 loci, there was no difference in survival between patients who were 12/12 matched, that is, matched at DPB1, and those permissively or non-permissively mis-matched at DPB1. The authors did report an association of ultra-high-resolution matching with acute GVHD for their whole cohort, and associations with GVHD and transplant-related mortality in some subgroups, and I refer you to their JCO manuscript for details.      There are some caveats to consider in applying their findings to clinical practice, and a good study always leads to more questions. The probability of finding an adult HLA 8/8 high-resolution matched unrelated donor varies from 16% to 75% depending on recipients race and ethnicity – the chances of finding a 10/10 or 12/12 donor who is ultra-high-resolution matched is going to be significantly lower. How do we factor in the role of other known non-HLA unrelated donor selection factors such as donor age, donor sex, CMV status, ABO type, and graft source vis-à-vis ultra-high-resolution matching? Even in this highly selected cohort of patients who were actually able to get a transplant, nearly 90% did not have a 12/12 ultra-high-resolution matched donor – in this setting, how do mismatches at different loci compare with respect to outcomes? Do we change our transplant conditioning and GVHD prophylaxis regimens in ultra-high-resolution mis-matched unrelated donor transplants to reduce the risk of GVHD? Several studies have indicated comparable survival between matched unrelated donor and haploidentical related donor transplants – does the use of an ultra-high-resolution 10/10 or 12/12 HLA matched donor offer any outcome advantage compared to the haploidentical transplantation? Taken together and at this time, their findings are primarily applicable to patients who have the luxury of choosing from several young 10/10 HLA matched unrelated donors. Cost of HLA typing using next-generation sequencing is also a factor that needs to be considered.    Notwithstanding these 'yet to be answered' questions, there are advantages to ultra-high-resolution typing, and current technology does allow for rapid and unambiguous characterization of HLA genes with a rapid turnaround time. Many HLA labs are already implementing third-generation typing methods, and with increasing use and demand, it is expected to become cheaper and will no longer be cost-prohibitive.    Overall, with HLA-identical sibling, matched unrelated, haploidentical related, mis-matched unrelated, and umbilical cord blood, nearly all patients who need a transplant have a donor. It is heartening to see that our research has pivoted from "Is there a donor available?" to "What is the best donor?" for a given patient.     This concludes this JCO podcast. Thank you for listening.