Comparison of Outcomes of Allogeneic Hematopoietic Cell Transplantation for Primary Myelofibrosis Among Stem Cell Source Groups

Allogeneic hematopoietic cell transplantation (HCT) is the only treatment modality that can either cure or prolong life of patients with primary myelofibrosis (PMF). However, the issues of the choice of stem cell source, the choice of conditioning regimen, and the timing of HCT are currently under debate. To determine whether a difference in stem cell source affects the outcome of HCT for PMF patients, a retrospective study was conducted using the national registry data on 224 patients who received first allogeneic HCT in Japan with bone marrow (BM), peripheral blood stem cells (PBSC) or umbilical cord blood (UCB). This study was approved by the Data Management Committee of the Japan Society for Hematopoietic Cell Transplantation and by the ethics committee of the Nagoya University School of Medicine. One hundred fifty-two male and 72 female, with a median age of 55 years (range, 21-79 years), were treated with a myeloablative (48%) or non-myeloablative (52%) preconditioning and GVHD prophylaxis such as calcineurin inhibitor with methotrexate (78%) between 1993 and 2016. Stem cell sources were HLA-A, -B and -DR 6/6 matched related BM (Rtd-BM) (n = 22), HLA-A, -B and -DR 6/6 matched related PBSC (Rtd-PBSC) (n = 48), HLA-A, -B, -C and -DRB1 alleles 8/8 or 7/8 matched unrelated BM (UR-BM) (n = 91), unrelated UCB (UR-UCB) (n = 29) and others (n = 34) including HLA 5/6 or 4/6 matched related BM and PBSC, HLA-haploidentical related BM and PBSC, HLA alleles 6/8 or 5/8 matched unrelated BM, and HLA alleles 8/8 or 7/8 matched unrelated PBSC. All UR-UCB transplantation was performed with a single unit. The median follow-up term for living patients was 48 (0.3-202) months. Cumulative incidences of neutrophil recovery (≥0.5 × 10⁹/L) on day 60 were 100% in Rtd-BM (median days to recovery, 21 days), 94% in Rtd-PBSC (16 days), 86% in UR-BM (21 days), 79% in UR-UCB (25 days) and 91% in other transplantation (23 days). Cumulative incidences of grade II-IV and III-IV acute GVHD on day 100 were 23% and 5% in Rtd-BM, 27% and 19% in Rtd-PBSC, 27% and 10% in UR-BM, 31% and 10% in UR-UCB, and 26% and 15% in other transplantation, respectively. Cumulative incidences of chronic GVHD at 2 years after HCT were 35% in Rtd-BM, 41% in Rtd-PBSC, 34% in UR-BM, 19% in UR-UCB and 13% in other transplantation. Non-relapse mortality (NRM) at 2 years after HCT were 22% in Rtd-BM, 41% in Rtd-PBSC, 39% in UR-BM, 47% in UR-UCB and 48% in other transplantation. Multivariate analysis demonstrated that RBC transfusion ≥20 times before HCT (HR, 2.05; 95% CI, 1.06-3.98), PLT transfusion 10-19 times before HCT (3.56, 1.57-8.05), UR-UCB transplantation (4.70, 1.13-19.6) and other transplantation (4.38, 1.05-18.3) were predictive factors for higher NRM. Although performance status ≥2 at HCT was significant for higher NRM in univariate analysis, it was not significant in multivariate analysis. Relapse rates at 2 years after HCT were 14% in Rtd-BM, 17% in Rtd-PBSC, 13% in UR-BM, 24% in UR-UCB and 15% in other transplantation. Multivariate analysis demonstrated that no factor was associated with the incidence of relapse, although high-risk group of chromosome karyotype was significant for higher relapse rate in univariate analysis. Neither stem cell source groups nor DIPSS was not significant in univariate analysis. Overall survival (OS) rates at 2 and 5 years after HCT were 71% and 71% in Rtd-BM, 52% and 52% in Rtd-PBSC, 54% and 46% in UR-BM, 43% and 27% in UR-UCB, and 48% and 33% in other transplantation, respectively. Multivariate analysis demonstrated that age of 46-55 years (HR, 2.14; 95% CI, 1.00-4.56) and ≥56 years (2.69, 1.32-5.48), RBC transfusion ≥20 times before HCT (1.91, 1.13-3.25) and PLT transfusion 10-19 times before HCT (3.51, 1.64-7.52) predicted lower OS rate. Although performance status ≥2 at HCT, DIPSS intermediate-2 or high at HCT, high-risk group of chromosome karyotype, UR-UCB transplantation and other transplantation were significant for lower OS in univariate analysis, they were not significant in multivariate analysis. The present study could not find an advantage of use of JAK1/2 inhibitor before HCT, use of anti-thymoglobulin as a preconditioning, and non-myeloablative preconditioning regimen in terms of decreasing NRM or increasing OS rate. Our results suggest that allogeneic HCT provide a curative treatment for PMF patients, however careful management is required in HCT with other than Rtd-BM, Rtd-PB and UR-BM. Further analysis in a large cohort is required.