Comparison of Haploidentical Donor Hematopoietic Cell Transplantation Using Post-Transplant Cyclophosphamide to Matched-Sibling, Matched-Unrelated, Mismatched-Unrelated, and Umbilical Cord Blood Donor Transplantation in Adults with Acute Lymphoblastic Leukemia: A CIBMTR Study
Matthew J. Wieduwilt,Leland Metheny,Mei-Jie Zhang,Hai-Lin Wang,Noel Estrada-Merly,Marcos de Lima,Mark Litzow,Partow Kebriaei,Chris S. Hourigan,Daniel J. Weisdorf,Wael Saber
DOI: https://doi.org/10.1182/blood-2020-137186
IF: 20.3
2020-11-05
Blood
Abstract:Background The relative benefits of haploidentical HCT using PTCy compared to traditional matched-sibling donor, traditional unrelated donor and umbilical cord blood (UCB) allogeneic HCT for adult acute lymphoblastic leukemia (ALL) is still being defined. Methods We performed a retrospective analysis using the Center for International Blood and Marrow Transplant Research (CIBMTR) database comparing post-HCT outcomes of haploidentical donors with PTCy to HLA-identical sibling donors, 8/8 HLA-MUDs, 7/8 HLA-MUDs, and UCB. Outcomes were overall survival (OS) leukemia-free survival (LFS), relapse, non-relapse mortality (NRM), grade 2-4 and grade 3-4 acute graft-versus-host disease (aGVHD), and chronic GVHD (cGVHD). Eligible patients were adults with ALL in complete remission (CR1, CR2, or CR3) undergoing first allogeneic HCT from 2013 to 2017. Key exclusion included not receiving PTCy for haploidentical HCT and ex vivo T-cell depleted or CD34-selected grafts. Cox proportional hazards multivariable regression analysis was used to compare the treatment groups. Results 4201 patients were studied: 393 haploidentical donor (92 centers), 1627 HLA-identical sibling donor (206 centers), 1646 8/8 HLA-MUD (181 centers), 230 7/8 HLA-MUD (90 centers), and 305 UCB (79 centers). Groups were well matched, however, notable differences between included race, time from diagnosis to HCT (CR1 only), conditioning regimen intensity, donor age, graft source for non-cord (peripheral blood or bone marrow), GVHD prophylaxis modality, and the use of in vivo T-cell depletion. Compared to other groups, haploidentical HCT had the lowest percentage of non-Hispanic white patients (43% vs. 49-74%), was more likely to use reduced-intensity conditioning (42% vs 17-25%), and was more likely to use bone marrow (41% vs. 14-29%). In multivariate analysis comparing haploidentical to matched-sibling donor HCT, no differences were observed in OS, LFS, NRM, relapse, or aGVHD. Compared to haploidentical HCT, the risk of cGVHD was significantly higher with matched-sibling donor HCT (D/R sex match F/M, HR 2.59, P<0.001, D/R sex match, other, HR 1.37, P=0.003). Comparing haploidentical to 8/8 HLA-MUD HCT, no differences were seen in OS, LFS, relapse, or Grade 2-4 aGVHD but 8/8 HLA-MUD HCT was associated with increased incidences of NRM (HR 1.42, P=0.02), grade 3-4 aGVHD (HR 1.59, P=0.005), and cGVHD (HR D/R sex match F/M, HR 2.91, P<0.001, D/R sex match, other, HR 1.38, P=0.001). Comparing haploidentical to 7/8 HLA-MUD HCT, no differences were seen in LFS or relapse but 7/8 HLA-MUD HCT was associated with worse OS (HR 1.38, P=0.01) and increased incidences of NRM (HR 2.13, P=<0.001), grade 2-4 aGVHD (HR 1.33, P=0.04) , grade 3-4 aGVHD (HR 1.86, P=0.003), and cGVHD (HR 1.72, P=<0.001). Comparing haploidentical to UCB HCT, no differences were seen in late OS (>18 months), late LFS, relapse, or cGVHD. Compared to haploidentical HCT, UCB HCT was associated with worse early OS (≤18 months, HR 1.93, P<0.001), worse early LFS (HR 1.40, P=0.007) and increased incidences of NRM (HR 2.08, P<0.001), grade 2-4 aGVHD (HR 1.83, P<0.001), and grade 3-4 aGVHD (HR 1.97, P<0.001). Similar outcomes were observed in sensitivity analyses restricting cohorts to (1) MA conditioning with peripheral blood stem cells and (2) calcineurin inhibitor-based GVHD prophylaxis for matched-sibling, unrelated donor, and UCB HCT. Conclusions For OS and LFS, haploidentical HCT using PTCy for adult ALL was superior to traditional HCT using 7/8 HLA-MUDs and UCB but equivalent to traditional HCT using matched-sibling donors and 8/8 HLA-MUDs. Results were consistent across all subgroups. Haploidentical HCT had a lower incidence of aGVHD compared to 8/8 HLA-MUD, 7/8 HLA-MUD, and UCB HCT and a lower incidence of cGVHD compared to matched-sibling donor, 8/8 HLA-MUD, and 7/8 HLA-MUD HCT. Based on these data, haploidentical HCT with PTCy results in similar survival with less GVHD compared to traditional matched-sibling and 8/8 HLA-matched donor HCT and is the superior alternative donor option for adult patients with ALL needing allogeneic HCT in CR. Figure Disclosures Wieduwilt: Leadiant: Research Funding; Shire: Research Funding; Merck: Research Funding; Macrogeneics: Research Funding; Amgen: Research Funding; Daiichi Sankyo: Membership on an entity's Board of Directors or advisory committees. de Lima:Incyte: Other: Personal Fees, advisory board; BMS: Other: Personal Fees, advisory board; Pfizer: Other: Personal fees, advisory board, Research Funding; Celgene: Research Funding; Kadmon: Other: Personal Fees, Advisory board. Kebriaei:Novartis: Other: Served on advisory board; Pfizer: Other: Served on advisory board; Kite: Other: Served on advisory board; Ziopharm: Other: Research Support; Amgen: Other: Research Support; Jazz: Consultancy. Hourigan:Merck: Other: Research Support (CRADA); Sellas: Other: Research Support (CRADA). Weisdorf:FATE Therapeutics: Consultancy; Incyte: Research Funding.
hematology