Prognostic analysis of salvage allogeneic HCT in acute leukemia patients relapsing after autologous HCT:  a study from the acute leukemia working party of the EBMT

Abstract Background While autologous hematopoietic cell transplantation (auto-HCT) is a feasible post-remission strategy for patients with acute myeloid leukemia (AML), its primary limitation is the high risk of disease relapse. This study aims to provide an overview of outcomes following salvage allogeneic HCT for patients who experience relapse after auto-HCT, and to identify relevant prognostic factors. Method This retrospective study analyzed consecutive adult patients (≥18 years) with AML who underwent auto-HCT in complete remission (CR) and subsequently received salvage allo-HCT due to relapse between 2014 and 2022, as recorded in the European Society for Blood and Marrow Transplantation (EBMT) registry. Key exclusion criteria included an interval between allo-HCT and auto-HCT of less than 30 days or ex vivo T-cell-depleted allo-HCT. Survival analysis and competing risk analysis were employed to assess clinical outcomes, while Cox regression models were used to identify prognostic factors. Results A total of 233 patients were included in the analysis with a median follow-up of 3.6 years (IQR, 2.9-4.1) post-allo-HCT. The cohort comprised 107 patients (45.5%) who received reduced-intensity conditioning (RIC) and 127 patients (54.5%) who underwent myeloablative conditioning (MAC). Among these groups, 69.8% and 75.6% of patients, respectively, were in second complete remission (CR) at the time of allo-HCT. 28.3% of RIC recipients and 13.4% of MAC recipients transitioned from chemotherapy-based conditioning during auto-HCT to TBI-based conditioning for allo-HCT. Overall, the 3-year post-allo-HCT outcomes showed overall survival (OS) of 54.4% (95% CI, 46.8-61.4%), leukemia-free survival (LFS) of 44.2% (95% CI, 36.8-51.2%), relapse incidence of 29.6% (95% CI, 23.2-36.2%), and non-relapse mortality (NRM) of 26.2% (95% CI, 20.2-32.7%). Multivariate analysis revealed no significant differences between RIC and MAC in terms of OS (HR 1.22, 95% CI 0.73-2.05; P=0.454), LFS (HR 1.32, 95% CI 0.85-2.05; P=0.215), relapse (HR 1.32, 95% CI 0.74-2.37; P=0.345), NRM (HR 1.33, 95% CI 0.67-2.65; P=0.411), or chronic GVHD (HR 1.18, 95% CI 0.63-2.24; P=0.605). Notably, MAC was associated with a significantly higher risk of grade 2-4 acute GVHD compared to RIC (HR 2.4, 95% CI 1.05-5.46; P=0.037). While relapse timing (in years) after auto-HCT was not observed associated with outcomes, patients with active disease had significantly worse OS (HR 3.42, 95% CI 1.94–6.03; P < 0.001) and LFS (HR 2.44, 95% CI 1.46–4.07; P = 0.001) compared to those in second CR. Conclusion Salvage allo-HCT is an effective treatment option for patients who relapse after auto-HCT. Achieving a second CR prior to allo-HCT is associated with improved survival. MAC results in a higher incidence of grade 2–4 aGVHD compared to RIC, while no significant differences in NRM, OS, or LFS are observed between the two conditioning regimens.