Introduction While the tyrosine kinase inhibitors (TKIs) have dramatically improved outcomes for Philadelphia chromosome-positive B-cell acute lymphoblastic leukemia (Ph+ B-ALL), relapse remains a significant challenge. More recently, novel immunotherapies such as blinatumomab and CAR-T cell therapy have emerged as highly effective options. Nevertheless, hematopoietic stem cell transplantation (HCT) remains a vital curative strategy to consolidate remission, particularly for patients who are intolerant to these therapies, or in settings where these novel agents have limited accessibility. For patients lacking a matched sibling donor, both auto-HCT and haplo-HCT are feasible options. However, the optimal transplantation strategy between autologous and haploidentical HCT for this population remains to be explored.

Methods We retrospectively analyzed data from the European Society for Blood and Marrow Transplantation (EBMT) registry on adult patients with Ph+ B-ALL in first complete remission (CR1) who underwent their first auto-HCT or haplo-HCT between 2010 and 2022. All recipients had received a TKI for induction and/or consolidation therapy before HCT. To balance the cohorts, a 2:1 pair-matching algorithm was applied. Matching was based on an exact match for minimal residual disease (MRD) status prior to HCT and transplant region (China vs. other), as well as propensity score matching for age at HCT, year of HCT, and the time interval between diagnosis and HCT.

Results The study comprised 434 patients: 117 in the auto-HCT group and 317 in the haplo-HCT group. Auto-HCT recipients were older than those undergoing haplo-HCT (median age: 46 vs. 38 years, P < 0.001). Patients in the auto-HCT group underwent transplantation in earlier years (median: 2018 vs. 2020, P < 0.001) and had a longer interval from diagnosis to HCT (median: 6.8 vs. 5.8 months, P < 0.001). Regarding MRD evaluation, a higher proportion of auto-HCT recipients achieved MRD negativity prior to HCT compared to haplo-HCT recipients (81.4% vs. 62.3%, P < 0.001). The auto-HCT group had a lower rate of Cytomegalovirus (CMV) seropositivity (51.2% vs. 67.5%, P = 0.01) and used peripheral blood more frequently as the graft source (95.7% vs. 65.9%, P < 0.001). Myeloablative conditioning (MAC) was used less often in the auto-HCT group compared to the haplo-HCT group (72.8% vs. 87.1%, P < 0.001). Baseline characteristics such as sex and Karnofsky Performance Status (KPS) were similar between the two groups.

After matching, 97 auto and 159 haplo recipients were included in the analysis. Main causes of death were relapse (88.2% vs 38.7%) and infection (5.9% vs 38.7%) in the auto-HCT group and the haplo-HCT group, respectively. The cumulative incidence of neutrophil recovery at 30 days was 100% in the auto-HCT group and 97.6% (95% CI, 93%-99.1%) in the haplo-HCT group. Platelet recovery at 60 days was 97.8% (95% CI, 89.3%-99.6%) in the auto-HCT group and 94.1% (95% CI, 88.3%-97.1%) in the haplo-HCT group.

Regarding transplant outcomes, no significant difference was observed in 3-year Leukemia-Free Survival (LFS) between the auto-HCT and haplo-HCT groups (69.9% vs. 75%; HR, 0.78; P = 0.3). Similarly, 3-year Overall Survival (OS) was at 80.5% for auto-HCT and 78.6% for haplo-HCT; a formal hazard ratio was not calculated as the proportional hazards assumption was violated. Notably, compared to the auto-HCT group, the haplo-HCT group had a significantly lower 3-year relapse incidence (RI) (10.3% vs. 28%; HR, 0.35; P = 0.001) but a significantly higher 3-year non-relapse mortality (NRM) (14.8% vs. 2.1%; HR, 7.46; P = 0.008).

Conclusion For Ph+ B-ALL patients in CR1 with TKI therapy before transplantation, auto-HCT and haplo-HCT provide non different LFS, as the significantly lower relapse rate of haplo-HCT is counterbalanced by its higher NRM.

This content is only available as a PDF.
2025
Sign in via your Institution