Improved Outcomes with Class I and II DNA-Based HLA Typing for Matched Unrelated Donor (MUD) Allogeneic Bone Marrow Transplant (ABMT) in Hematologic Diseases.

Class I and II DNA-based HLA typing has resulted in improved outcomes in MUD ABMT compared to that observed with HLA serologic typing methods. Since MUDs do not necessarily possess the same linkage of HLA genes on haplotypes as found in matched sibling donors (MSD) it is important to determine the specific alleles present through DNA-based methods in order to improve matching between recipients and donors. We analyzed 146 consecutive hematologic malignancy patients (pts) who underwent ABMT at our institution from 1/7/99–2/4/04, to compare outcomes of 41 MUD and 105 MSD pts. All pts were matched at HLA-A, B, C, and DRB1 by at least low to intermediate resolution using either PCR-sequence specific priming or PCR-sequence specific oligonucleotide probing using Class I and II DNA-based HLA typing. All pts received a busulfan/cyclophosphamide (Bu/Cy)-based preparative regimen. All MUD ABMTs were CD8+ T cell depleted while all MSD ABMTs were T cell replete. GVHD prophylaxis for MUD pts consisted of FK506 and methotrexate while MSD pts received cyclosporine and methotrexate or mycophenolate mofetil. There were no differences between MUD or MSD pts in terms of sex, age, race, prior radiation therapy, or donor-pt gender. Primary diagnoses were comparable except the MSD group had more CML pts (19 vs. 0) (p=0.01 for all diagnoses). There were no differences between MUD and MSD pts regarding infection, acute GVHD, acute grade 3–4 GVHD, chronic GVHD or extensive chronic GVHD. MUD pts as compared to MSD pts had higher graft failure rates (12% vs. 3%, p = 0.019); lower relapse-free survival (median 5.4 mos vs. 10.7 mos, p=0.037); higher 100-day mortality (37% vs. 21%, p=0.05), and lower overall survival (median 5.7 mos vs.13.6 mos, p=0.036). Deaths occurred in 30 (73%) of MUD pts and 61 (58%) of the MSD pts with disease relapse followed by GVHD as the most common causes of death. Since CML pts were only in the MSD group, we repeated the analysis excluding these pts and there were no longer any significant differences between the MUD and MSD pts in 100 day mortality (p=0.12), relapse-free survival (median 5.4 mos vs. 6.5 mos, p=0.22), or overall survival (median 5.7 mos vs. 10.0 mos, p=0.18). For further purposes of comparison, we also analyzed 112 consecutive ABMT pts (31 MUD, T cell depleted; 81 MSD, T cell replete; Bu/Cy-based preparative regimens for all pts) previously transplanted at our institution from 4/93–11/98 during which time only HLA serologic and Class II DNA-based typing was used. In this earlier pt cohort, the MUD pts had a significantly higher incidence of acute GVHD (p<0.001), lower 100-day mortality (52% vs. 12%, p<0.001), lower relapse-free survival (p=0.001) and lower overall survival (2.6 mos vs. 32.5 mos, p=0.001) than the MSD patients. We conclude that Class I and II DNA-based HLA typing is important and can improve outcome in MUD ABMT. Further investigation with high resolution HLA typing methods may continue to improve outcomes with MUD ABMT.