The group examined data from over 1,600 patients who received hematopoietic cell transplants from related haploidentical donors and sought to identify clinically significant allele variations on different HLA haplotypes. “We were really interested in testing the clinical importance of a particular sequence feature of HLA-DR molecules, and a novel paradigm for HLA-DQ, as a prelude to constructing extended haplotypes,” Dr. Petersdorf said. The research team investigated a specific sequence feature of HLA-DR molecules — residue 86 of the HLA-DRβ chain. This residue is said to be dimorphic as it usually encodes either a glycine (Gly) or valine (Val), which can influence the peptide-binding domain of HLA-DR molecules. “Although there has been information on the structural impact of the Gly86Val dimorphism, data regarding the clinical relevance of this feature is lacking,” Dr. Petersdorf exclaims. She continued, “we first tested the hypothesis that among haploidentical patients and transplant donors, the risks associated with HLA-DRB1 mismatching depended on whether the molecules were 86Gly or 86Val.” Consistent with their hypothesis they “learned that risks depend on the patient’s HLA-DR molecule and whether the donor has the same or different residue 86. This permitted us to simplify the definition of HLA-DR, reducing over 3588 unique DRB1 alleles to either 86Gly or 86Val.”
In addition to HLA-DR molecules, the group recently discovered that the αβ heterodimer of HLA-DQ is “a clinically relevant paradigm in unrelated donor hematopoietic cell transplantation,” as “some haplotypes permit trans-heterodimerization between one parental DQα with the opposite parental DQβ, and vice versa,” Dr. Petersdorf explains. “We tested the relevance of (mis)matching for HLA-DQ in haploidentical transplantation and discovered that the patient and donor heterodimers predicted risk of mortality for certain αβ heterodimer combinations,” which again enabled the group to reduce thousands of unique HLA-DQ alleles to a much simpler functional model. The research team extended their definition of class II haplotypes to include HLA-DM, a molecule that participates in the presentation of peptides by HLA-DR and -DQ. They discovered that haplotypes defined by all three molecules, HLA-DR, HLA-DQ and HLA-DM, provide new information on the risk of relapse.
This work enhances our understanding of how certain HLA class II genes and molecules contribute to leukemia relapse after transplantation. Here, the Petersdorf group “discovered that the risk of the recurrence of the patient’s leukemia was higher with the presence of certain haplotypes compared to others,” notes Dr. Petersdorf. She continues and explains that the “data suggest that if HLA-DR and HLA-DQ, which are currently characterized in patients and candidate donors prospectively, are visualized not simply as ‘matched’ or ‘mismatched’ allele sequences, but by what they contribute functionally at Gly86Val or as an αβ heterodimer, then we might be able to select haploidentical family members to lower the risk of leukemia recurrence.” While Petersdorf acknowledges that there is still much to understand about the role of HLA molecules in transplantation, she states, “we hope that refined approaches to the selection of donors and use of HLA information for risk prediction may contribute to improved outcomes for our patients. At the same time, it is very exciting to add to the fundamental biology of HLA genes and proteins.”