Bone Marrow Versus Peripheral Blood in HLA-Matched Sibling Transplantation: Does the Stem Cell Source Matter in Severe Aplastic Anemia?

Background

Selecting a graft source is critical in HLA-matched sibling donor hematopoietic stem cell transplantation (MSD-HCT) for severe aplastic anemia (SAA). Since graft-versus-host-disease (GVHD) is the major cause of mortality after MSD-HCT, and SAA patients do not need the graft-versus-leukemia effect, the selection of a suitable source of stem cells becomes critical. Although bone marrow (BM) is the preferable stem cell source based on the registry-based studies, mobilized peripheral blood stem cells (PBSC) may offer faster engraftment with rapid immune reconstitution. Moreover, the recent advances in conditioning regimens have significantly reduced transplantation-related complications. Here, we compared the long-term outcomes of BM vs PBSC as a graft source in adult patients with SAA who received MSD-HCT with the reduced intensity conditioning (RIC) regimen.

Method

We conducted a retrospective analysis of 171 consecutive SAA adult patients who received MSD-HCT with Flu6/Cy2/ATG4 RIC regimen (Flu, 30 mg/m ² IV for 6 days: Cy, 50 mg/kg IV for 2 days; rabbit ATG, 2.5 mg/kg IV for 4 days) or total nodal irradiation (TNI) of 750 cGy combined with ATG as GVHD prophylaxis (TNI-750/ATG) from March 2002 to March 2021 at Seoul St. Mary's Hospital, Seoul, Korea. BM was the primary graft source, but PBSC was used in donors who refused BM harvest, were unsuitable for general anesthesia, or received TNI-750/ATG conditioning. The primary endpoint of this study was to compare the overall survival (OS), and the secondary endpoints included speed of hematologic recovery, primary or secondary graft failure (GF), and failure-free survival (FFS) between the BM and PBSC group.

Results

The median patient age in the BM group was 38 years (range, 15 - 59) and 44 years (range, 19 - 64) in the PBSC group (p=0.005). There were no differences of baseline characteristics between the two groups except for a donor-to-recipient sex discrepancy (female to male, BM vs. PBSC; 16.4% vs. 30.9%, p=0.029). The majority of BM group (114/116, 98.2%) received the Flu6/Cy2/ATG4 RIC regimen, while 38.2% (21/55) of PBSC group received TNI-750/ATG for the conditioning (p<0.001). The PBSC contained a median 4.71 (range, 1.22 - 17.03) x 10 ⁶/kg CD34 ⁺ cells and 368.82 (range, 25.12 - 1234.70) x 10 ⁶/kg CD3 ⁺ cells which were about 1.7-fold (BM 2.80 x 10 ⁶/kg CD34 ⁺ cells; range, 0.25 - 10.05) and 10-fold (BM 39.61 x 10 ⁶/kg CD3 ⁺ cells; range, 1.34 - 463.68) higher than the BM graft (p<0.001). Consequently, the PBSC group exhibited faster engraftment of neutrophils (median 13 vs. 11 days, p<0.001) and platelets (median 18 vs. 12 days, p=0.012).

However, there were no differences in the incidence of secondary GF (19.0% vs. 12.7%, p=0.386) and donor-type aplasia (4.3% vs. 10.9%, p=0.178) between two groups. Furthermore, the incidence of acute grade I-II (8.6% vs. 10.9%, p=0.619) and moderate to severe chronic graft-versus-host disease (GVHD) (3.5% vs. 5.5%, p=0.542) were also not significantly different between two groups. However, the BM group has shown more CMV reactivation (44.0% vs. 24.9%, p=0.005), and higher incidence of respiratory infection (14.7% vs. 3.6%, p=0.037) than those of the PBSC group. The incidences of transplantation-related complications between the BM and PBSC group are summarized in Table. The 10-year OS (93.8% vs. 98.1%, p=0.246) and FFS (65.7% vs. 69.0%, p=0.455) were comparable (Figure). Moreover, there was no significant difference in FFS between BM and PBSC group by potential risk factors of patient and donor age at MSD-HCT, sex incompatibility, ABC mismatch, PNH clone presence, HCT-CI ≥3, infused CD34 ⁺ cell count, interval from diagnosis to transplant, prior IST history, heavy transfusion before allo-HCT, or severity of AA.

Conclusions

Our results suggest that survival outcomes of MSD-HCT showed no difference between BM and PBSC as graft sources, and using PBSC may be an efficient alternative offering faster engraftment without significant difference in graft failure or transplantation-related adverse events such as GVHD, and survival. Current analysis and results were drawn from the Asian population with a fludarabine-based conditioning regimen and therefore, need to validate in the Western population. In addition, well-designed prospective studies in a larger cohort are required to confirm our findings.