Comparison of Outcomes after Haploidentical Relative and HLA Matched Unrelated Donor Transplantation with Post-Transplant Cyclophosphamide Containing Gvhd Prophylaxis Regimens

Introduction: Post-transplant cyclophosphamide (PT-Cy) in HLA-haploidentical hematopoietic cell transplantation (Haplo-HCT) results in acceptable engraftment and graft-versus-host disease (GVHD). Results of BMT CTN 1203 that compared matched unrelated donor (MUD) HCT with PT-Cy containing to calcineurin inhibitor (CNI) containing GVHD prophylaxis following reduced intensity conditioning, showed better 1-year GVHD free relapse free survival with PT-Cy. The relative value of a MUD versus a Haplo donor in the context of PT-Cy is not known. In this study we compared outcomes after Haplo- and MUD HCT with PT-Cy containing GVHD prophylaxis.

Methods: Eligible patients were aged >18 years and received haplo-HCT or MUD HCT for AML, ALL and MDS in the US from 2011-2018. Patients received myeloablative (n=1001) or reduced intensity (n=1398) regimens and were analyzed separately. All recipients of haplo-HCT and reduced intensity conditioning MUD HCT received PT-Cy + CNI + mycophenolate mofetil (MMF) for GVHD prophylaxis. Among recipients of myeloablative conditioning MUD HCT, 55% received PT-Cy + CNI + MMF and 45%, PT-Cy + CNI. Cox regression models were built to compare outcomes between donor types. As PT-Cy containing GVHD prophylaxis for myeloablative conditioning MUD HCT is recent; therefore, outcomes were censored at 1-year. A p-value ≤0.01 was considered significant.

Results: Myeloablative conditioning regimen: Age at transplant, sex, performance score, co-morbidity and CMV serostatus did not differ between recipients of haplo- and MUD HCT. Haplo-HCT recipients were less likely to be Caucasian (69% vs. 91%). They were also more likely to have AML (55% vs. 47) and less likely to have MDS (13% vs. 31%). Disease risk index did not differ between treatment groups. The predominant conditioning regimen for haplo-HCT was total body irradiation (TBI) + fludarabine (44%) for Haplo-HCT and for MUD HCT, fludarabine + busulfan ± thiotepa (59%). Engraftment rates did not differ between groups. Results of multivariate analyses are shown in Table 1. Compared to Haplo-HCT, grade II-IV acute GVHD risk was higher after MUD-HCT with PT-Cy + CNI GVHD prophylaxis. There were no differences in grade III-IV acute or chronic GVHD risk between treatment groups. Non-relapse mortality, relapse, disease-free and overall survival did not differ between treatment groups.

Reduced intensity conditioning regimen: Age at transplant, sex, performance score, co-morbidity and CMV serostatus did not differ between recipients of haplo- and MUD HCT. Haplo-HCT recipients were less likely to be Caucasian (72% vs. 95%). Haplo-HCT recipients were less likely to have MDS (19% vs. 24%). Disease risk index did not differ between treatment groups. The predominant conditioning regimen was TBI 200 cGy + fludarabine + cyclophosphamide (88%) for Haplo-HCT and for MUD HCT, TBI 200 cGy + fludarabine + cyclophosphamide (37%) and fludarabine + melphalan (33%). Engraftment rates did not differ between groups. Results of multivariate analyses are shown in Table 2. Compared to Haplo-HCT acute and chronic GVHD risks did not differ between treatment groups. However, non-relapse mortality was lower after MUD-HCT which led to higher disease-free and overall survival.

Conclusion: In patients who received myeloablative conditioning and PT-Cy + CNI + MMF GVHD prophylaxis, haploidentical related and matched unrelated donors are comparable. However, in patients who received less intense conditioning regimens, disease free and overall survival was higher after MUD HCT compared to Haplo-HCT.