Outcomes of peripheral blood stem cells versus bone marrow in adult sickle cell disease patients undergoing haploidentical allogeneic stem cell transplantation
Haploidentical allogeneic hematopoietic cell transplantation (HCT) is a curative therapy for adults with sickle cell disease (SCD). In this setting, bone marrow (BM) is preferred due to a perceived lower risk of graft-versus-host disease (GvHD). Challenges include an invasive collection, insufficient CD34+ cell yield and higher risk of graft failure (GF). Peripheral blood stem cells (PBSC) offer a higher cell yield and faster engraftment. While PBSCs are associated with a higher risk of GvHD, it is unknown whether this risk persists with current conditioning regimens using robust lymphodepletion with both antithymocyte globulin (ATG) and post-transplant cyclophosphamide (PTCy). Given these developments, the preference for BM in haploidentical HCT for SCD should be reevaluated.
Adult SCD patients undergoing haploidentical HCT (2019-2024) at the Amsterdam University Medical Center and King Abdulaziz Medical City were retrospectively analyzed. All patients received the same reduced intensity conditioning consisting of ATG, thiotepa, fludarabine, cyclophosphamide, total body irradiation (2Gy) and PTCy. Patients receiving PBSC after insufficient BM harvest were assigned to the PBSC group. Log rank test was used to evaluate event-free survival (EFS), overall survival (OS) and engraftment period. GvHD incidence between groups was compared using Fisher’s exact test.
Out of 41 patients (median age 23 [IQR 19-31], 54% male), 26 (63%) received BM, and 15 (37%) received PBSC (including 4 after too low BM yield). Median follow-up was 800 days (IQR 611-1246) and 258 days (IQR 134-352) in BM and PBSC groups, respectively. Estimated 1-year EFS was 100% in the PBSC group vs 85% in the BM group (Figure 1A). All patients initially engrafted. Neutrophil and platelet recovery was faster in the PBSC group (20 vs 24 days, and 32 vs 36 days, respectively; Figure 1C/D). One BM recipient developed secondary GF with autologous reconstitution. Three BM recipients died from complications in the setting of poor graft function (n=2), or prolonged neutropenia (n=1). No grade III-IV aGvHD occurred. Incidence of grade II aGvHD at 1-year was 23% and 20% in BM and PBSC groups, respectively (p>0.99). One-year moderate/severe cGvHD occurred in 1 patient from each group (3.8% vs 6.7%; p>0.99).
In adult SCD patients undergoing reduced intensity haploidentical HCT, the use of PBSC may result in improved estimated 1-year EFS compared to BM. Comparable aGvHD and cGvHD rates between groups suggest that immunoablative conditioning with ATG and PTCy might abrogate the higher risk of GvHD associated with PBSC. Given the faster engraftment and less invasive collection procedure, PBSC appears a viable stem cell source in this setting. These results warrant validation in a prospective randomized trial with longer follow-up.
