Transplant Outcomes in Myelofibrosis: Impact of Donor Type (Cord Blood Grafts Supported by CD34+ selected Cells [Haplo-Cord] Versus Matched Donors)

Abstract

Despite the established potentially curative role of allogeneic hematopoietic cell transplantation (allo-HCT) in managing myelofibrosis (MF), the choice of alternative donors for patients lacking matched donors remains a challenge, and the optimal graft source in this disease entity continues to be an ongoing debate. We aimed to evaluate the impact of donor type: umbilical cord blood transplant supported with CD34+ selected haploidentical donor (haplo-cord) versus adult matched related donor (MRD) and matched unrelated donor (MUD) in 40 adult patients with primary or secondary MF, including those progressing to accelerated phase (AP) or blast phase (BP), who underwent their first allo-HCT. The primary objective of this study was to analyze the impact of stem cell source on primary endpoints of overall survival (OS), graft-versus-host disease, and non-relapse mortality (NRM). Median follow-up for all alive patients was 53 months (range 0.3-63 months). Nine patients (22.5%) underwent an MRD allo-HCT, 15 patients (37.5%) underwent a MUD allo-HCT, and 16 patients (40%) underwent a haplo-cord allo-HCT. Four patients died without neutrophil engraftment: 3 (19%) in haplo-cord group and one (4%) in MRD/MUD group. The cumulative incidence of absolute neutrophil engraftment by day 60 was 80% (95% CI 45-94) in the haplo-cord group and 92% (95% CI 65-98) in the MRD/MUD group (P = .09). The cumulative incidence of platelet engraftment by day 60 was 59% (95% CI 27-81) in haplo-cord group and 75% (95% CI 51-88) in MRD/MUD group (P = .4). OS was 62% at 1 year (95% CI 49-79) and 34% at 3 years (95% CI 21-55). The 3-year OS was similar between the haplo-cord group and the MRD/MUD (37% versus 32%, P = .9). The 1-year OS for AP/BP patients was 50% (95% CI 27-93) in the haplo-cord group, compared to 40% (95% CI 19-86) in the MRD/MUD. The 1-year OS for chronic phase CP patients was 83% (95% CI 58-100) in the haplo-cord group, compared to 79% (95% CI 60-100) in the MRD/MUD group. The cumulative incidence of relapse at 3 years in the haplo-cord group was 13% (95% CI 1.8-34), and in the MRD/MUD group was 28% (95% CI 10-49) (P = .36). One-year NRM was 38% (95% CI 15-61) in the haplo-cord group and 33% (95% CI 15-52) in the MRD/MUD group. Three-year NRM was 48% (95% CI 19-72) in the haplo-cord group and 54% (95% CI 29-73) in MRD/MUD group (P = .95). We showed no significant difference in OS, relapse, and NRM outcomes after haplo-cord transplant compared to adult matched donors’ grafts (MRD or MUD) in MF patients. However, there were more graft failures in patients transplanted with a haplo-cord transplants and delayed engraftments with inadequate haplo myeloid bridges. Despite the small sample size in our study, considering our findings and the availability of other alternative donors, using haplo-cord platforms may no longer be justified for MF unless the UCB engraftment dynamics can be optimized.