Comparison of Two Conditioning Regimens for Haploidentical Stem Cell Transplantation in Patients with Severe Aplastic Anemia

Allogeneic hematopoietic cell transplantation (HCT) is the best curative approach for patients with severe aplastic anemia (SAA). The outcomes of haploidentical donor HCT (haplo-HCT) have improved, making it a favorable option for patients lacking an HLA-identical donor. However, the optimal conditioning regimen remains largely undetermined. Studies have identified several risk factors, including poor performance status, higher HCT Comorbidity Index (HCT-CI), longer disease duration, older recipient, etc., that are associated with poorer transplant outcomes in SAA patients. This study retrospectively analyzes the outcomes of 461 patients who underwent haplo-HCT for SAA, comparing two conditioning regimens [conventional Busulfan and Cyclophosphamide (cBuCy) regimen, and reduced toxicity Busulfan, low-dose Cyclophosphamide and Fludarabine (BFC) regimen].

Overall, we found that recipients using either cBuCy regimen or BFC regimen achieved comparable overall survival [OS, 89.3% (85.2-92.5%) vs 87.1% (81.4-93.1%), P = 0.406] and failure-free survival (FFS, 87.1% (81.5-89.5%) vs 86.6% (79.2-91.1%), P = 0.670). Multivariable analysis identified that the interval between SAA diagnosis to HCT [≥ 12 months, hazard ratio (HR) 2.44 for OS (P = 0.046) and for 2.02 for FFS (P = 0.042)], recipients physical performance status [ECOG score ≥ 2, HR 2.97 for OS (P < 0.001) and 2.45 for FFS (P = 0.005)], and HCT-CI [HCT-CI ≥ 1, HR 3.45 for OS (P = 0.013) and 3.92 for FFS (P = 0.026)] were risk factors for both OS and FFS. Additionally, older recipient (≥ 40 years) was associated with inferior FFS (HR 1.62, P = 0.044). The probabilities of OS and FFS decreased significantly with increasing number of risk factors for survival (P < 0.05 comparing 0-1 vs ≥ 2 risk factors). Therefore, we defined patients with 0-1 risk factors for survival as lower-risk group and ≥ 2 risk factors as high-risk group.

Based on these findings, we performed a subgroup analysis. In the subgroup with a longer disease duration or an older recipient, no significant differences were found when comparing the probabilities of OS and FFS between the cBuCy regimen and the BFC regimen. For patients with worse performance status, there was a trend toward better survival in those receiving the BFC regimen compared to the cBuCy cohort [77.8% (62.6-87.2%) vs 63.3% (48.2-74.9%), P = 0.130 for both OS and FFS]. Of note, in patients with HCT-CI of ≥ 1, the BFC regimen was associated with higher probabilities of OS [84.4% (67.5-92.2%) vs 65.4% (54.3-74.7%), P = 0.034] and FFS [82.8% (66.7-90.1%) vs 65.4% (54.3-74.7%), P = 0.040]. In subgroup with disease duration of < 12 months, ECOG score of 0-1, or HCT-CI of 0, the probabilities of OS and FFS were comparable between patients receiving either of the two conditioning regimens. Furthermore, we found that high-risk SAA recipients (with ≥ 2 risk factors for survival) were more likely to benefit from the BFC regimen [79.4% (69.6-84.7%) vs 71.1% (65.9-77.8%), P = 0.043 for OS; and 78.6% (68.2-84.0%) vs 69.9% (62.7-74.8%), P = 0.037 for FFS].

In conclusion, our study suggests that reduced toxicity BFC conditioning regimen is a reasonable option for high-risk SAA recipients (with ≥ 2 risk factors for survival), in whom optimization of conditioning might further improve survival outcomes.