Haploidentical stem cell transplantation (Haplo-SCT) with post-transplant cyclophosphamide (PT-Cy) represents a potential curative strategy for patients with Hodgkin lymphoma (HL) when a matched related or unrelated donor is not available1. While bone marrow (BM) was originally the preferred stem cell source, more recently peripheral blood stem cell (PBSC) is more often used. Some retrospective studies suggest that the risk to develop acute and chronic graft-versus-host-disease (GVHD) is higher with PBSC than BM, while PBSC may reduce the risk of relapse2.
Here we analyzed the effect of stem cell source in 91 HL patients receiving Haplo-SCT with PT-Cy, with the aim to evaluate if the final outcome is modified by the use of PBSC or BM
From April 2009 to January 2017, 91 consecutive patients with poor prognosis HL received a Haplo-SCT with PT-Cy either from a PBSC (n = 38) or BM (n = 53). The two cohorts were similar for most characteristics, but the PBSC group had more patients with an unfavorable hematopoietic stem cell transplant comorbidity index (HCT-CI) score ≥3 (p = 0.002) and had received a non myeloablative conditioning (NMAC; p = 0.001).
Cumulative incidence of neutrophil>500/ul at day +30 and of platelet >20000/ul at day +60 were 96% (95% CI: 89–98) and 96% (95% CI: 88–99), respectively, with no significant differences between the PBSC and BM cohorts. With a median follow-up of 22.8 months, there was no difference between PBSC and BM graft in terms of cumulative incidence of grade 2–4 acute GVHD (29% vs 21%, p = 0.3), grade 3–4 acute GVHD (3% vs 4%, p = 0.7) and moderate-severe chronic GVHD (9% vs 7%, p = 0.7). This was also confirmed by multivariate analysis (data not shown). In the whole population, the 2-year overall survival (OS), 2-year progression-free survival (PFS) and 1-year GVHD/relapse free survival (GRFS) rates were 67%, 58% and 58%, respectively. We observed a trend for improved OS (74% vs 62%, p = 0.07) and PFS (62% vs 56%, p = 0.1) for recipients of PBSC relative to BM cells, but pre-transplant disease status was the only significant variable by univariate analysis (Table I). By multivariate analysis, PBSC resulted a protective factor both for OS (HR: 0.29, p = 0.006), PFS (HR: 0.38, p = 0.001) and GRFS (HR: 0.44, p = 0.020). The other independent variables affecting the final outcome were pre-transplant disease status and hematopoietic cell transplant comorbidity index (HCT-CI) (Table I). Non-relapse mortality was not affected by graft source both by univariate and multivariate analysis, while pre-transplant disease status was the only variable affecting the chance of disease relapse.
Overall these data suggest that PBSC is associated with better outcome, in terms of OS, PFS and GRFS, relative to BM cells as graft source for patients undergoing Haplo-SCT with PT-Cy. In addition, the risk of acute and chronic GVHD is not increased after PBSC relative to BM graft.