Biologically Randomized Comparison of Haploidentical Versus Human Leukocyte Antigen-Matched Related Donor Reduced-Intensity Conditioning Hematopoietic Cell Transplantation

Abstract

Using haploidentical donors for allogeneic hematopoietic cell transplantation (HCT) broadens transplant accessibility to a growing number of patients with hematologic disorders. Moreover, haploidentical HCT with post-transplant cyclophosphamide (PTCy) has become widespread practice due to accumulating evidence demonstrating favorable rates of survival and graft-versus-host disease (GvHD). Most studies comparing outcomes by donor sources have been confounded by variability in conditioning regimens, graft type (peripheral blood [PB] or bone marrow), and post-transplant GvHD prophylaxis (PTCy or non-PTCy), making it difficult to define the effect of donor source on outcomes. Levine Cancer Institute started a transplant and cellular therapy program in 2014, with both haploidentical and matched related donor (MRD) transplants initially performed using a uniform reduced-intensity conditioning (RIC) regimen, PB grafts, and PTCy-based GvHD prophylaxis. This retrospective observational study was conducted to compare the clinical outcomes associated with RIC haploidentical HCT and MRD HCT in patients receiving identical conditioning regimens, graft types, and supportive care. Our transplant database was queried to evaluate demographic characteristics, clinical features, and outcomes of RIC HCT for consecutive patients with hematologic malignancies who received haploidentical or MRD grafts between March 2014 and December 2017. An MRD was the preferred donor source; when unavailable, a haploidentical donor was used. Sixty-seven patients underwent haploidentical HCT and 25 MRD HCT. Overall, characteristics of transplant recipients were similar for the haploidentical and MRD groups; however, haploidentical donors were younger than MRDs (median 36 yr versus 57 yr, P < .0001). Results of univariable analysis showed similar overall survival (OS) for haploidentical and MRD HCT (hazard ratio [HR], 1.15; 95% CI, 0.61 to 2.15; P = .669). One-year, 1-yr, and 5-yr OS were 80.2%, 54.7%, and 41.2% for haploidentical HCT and 76.0%, 55.7%, and 51.1% for MRD HCT, respectively. With a median follow-up of 81.90 months, results of multivariable analysis revealed that donor source (haploidentical versus MRD) was not significantly associated with OS (HR, 0.97; 95% CI, 0.51 to 1.87; P = .933), relapse-free survival (HR, 0.75; 95% CI, 0.42 to 1.35; P = .337), cumulative incidence of relapse (HR, 0.81; 95% CI, 0.39 to 1.70; P = .579), or non-relapse mortality (HR, 1.12; 95% CI, 0.40 to 3.14; P = .827). Cumulative incidences of acute GvHD (aGvHD) and chronic GvHD (cGvHD) were not significantly different for haploidentical and MRD HCT (grades II to IV aGvHD: HR, 1.78; 95% CI, 0.72 to 4.37; P = .210; grades III to IV aGvHD: HR, 2.84; 95% CI, 0.34 to 23.63; P = .335; cGvHD: HR, 1.00; 95% CI 0.36 to 2.76; P = .995). With care that was homogenous in terms of conditioning regimens, graft type, GvHD prophylaxis, and supportive care, 92 patients who were biologically randomized to either haploidentical HCT or MRD HCT after RIC with PTCy had comparable outcomes.