Revisiting the role of second allogeneic HCT for post-transplant relapse in acute lymphoblastic leukemia: Comparative outcomes in the era of targeted therapies

Abstract

Abstract Background: Relapse after allogeneic hematopoietic cell transplantation (HCT) remains the leading cause of treatment failure in adult patients with acute lymphoblastic leukemia (ALL). Historically, a second HCT has been considered the standard curative approach for relapsed patients. However, with the increasing availability of novel agents, including ponatinib, blinatumomab, inotuzumab ozogamicin, and chimeric antigen receptor (CAR) T-cell therapy, the relative benefit of second HCT versus non-HCT salvage in the modern era requires re-evaluation.Aims: We aimed to compare the outcomes of second HCT versus modern non-HCT salvage strategies in adults with relapsed ALL after first allo-HCT, with subgroup analyses by treatment era, molecular subtype (Ph-positive vs. Ph-negative), remission status, relapse pattern, and next-generation sequencing findings.Methods: We retrospectively analyzed 249 adult patients with ALL who relapsed after first allo-HCT at our institution between 2010 and 2024. Patients were grouped into two treatment eras –conventional (2010 -2015, n=76) and contemporary era (2016-2024, n=173)– based on the clinical adoption of novel salvage agents including ponatinib, blinatumomab, inotuzumab ozogamicin, and CAR T-cell therapy. Patients were further classified into two cohorts: those treated with chemo-immunotherapy alone (non-HCT group) and those who subsequently underwent second allo-HCT by an intention-to-treat manner after achieving remission. Outcomes were compared using Mantel-Byar and Simon-Makuch models to account for immortal time bias. In a limited subset of patients (n=30), next-generation sequencing data were reviewed to identify high-risk molecular features.Results: Among 249 relapsed patients, 172 had bone marrow relapse (BMR), 20 had BMR + extramedullary relapse (EMR), and 57 had isolated EMR (iEMR) including 17 isolated CNS relapses. Median time-to-progression was 8.1 months (92 patients relapsing ≤6 months, 72 between 6-12 months, and 85 >12 months). In conventional era, the second HCT showed superior overall survival (OS) compared to non-HCT (p=0.014). However, in the contemporary era, 2-year OS was superior with non-HCT among patients achieving CR (61.3% vs. 40.9%, p=0.010). Among the Ph-negative ALL patients, 2-year OS did not statistically differ in two groups (23.1% vs. 47.2%, p=0.202) and among those achieving CR (62.9% vs. 47.2%, p=0.508). Among Ph-positive ALL patients, 2-year OS was superior in non-HCT group (46.1% vs. 23.1%, p=0.005), particularly among those in CR (61.3% vs. 23.1%, p=0.002). Targeted sequencing revealed enrichment of TP53 and TET2 mutations among patients with early post-relapse mortality (TP53: p=0.001; TET2: p=0.049). In multivariate analysis, poor minimal residual disease (MRD) response (HR 7.41; [95% CI 3.98–13.79]; p<0.001) were associated with inferior OS, while late relapse with longer time-to-progression (>12 months) showed improved survival (HR 0.35; [95% CI 0.21–0.58]; p<0.001). Second HCT was not independently associated with improved survival, but was associated with inferior OS in Ph-positive ALL (HR 3.41; [95% CI 1.33–8.76]; p=0.011).Conclusions: Our data challenges the traditional paradigm that second allo-HCT is the optimal salvage strategy for relapsed ALL following transplantation. In the contemporary era, expanded availability of new targeted agents has led to improved survival, particularly in Ph-positive ALL, without the morbidity of repeated transplantation. These findings underscore the need for individualized, MRD-guided, and mutation-informed salvage strategies and warrant prospective validation.