Survival Outcomes after Relapse in Patients Who Underwent Haploidentical Stem Cell Transplantation (haplo-SCT)

Background Relapse remains the most common cause of relapse-related mortality following hematopoietic stem cell transplantation (HSCT) with limited treatment options and an overall poor prognosis. Therefore, it is important to understand the characteristics, survival outcome, predictive factors, and emerging treatment options for these high-risk patients. There is limited data in this regard for patients who underwent haplo-SCT. We aimed from this study to assess survival outcomes of patients with acute myeloid leukemia (AML), acute lymphoblastic leukemia (ALL) and myelodysplastic syndrome (MDS) who underwent haplo-SCT. Methods We included all consecutive adult AML, MDS and ALL patients who had relapse after their first haplo-SCT between February 2009 and March 2019. Primary objective: overall survival (OS) after relapse. Secondary objectives were to explore predictive factors for survival. Age, gender, KPS, DRI, HCT-CI, conditioning regimen, stem cell source, and presence of acute GVHD before relapse, targeted therapy and time-to-relapse were included in the predictive risk-factor analysis for overall survival Results A total of 338 patients were identified during the study period, of which 88 patients (26%) progressed after transplant. Median age for relapsed patients was 48 (range, 19-70) years, 58 had AML, 18 had ALL and 12 had MDS. Table 1 summarizes patient and disease characteristic. Majority of patients had high/very-high DRI. Median time to relapse after haplo-SCT was 5 (range, 1-61) months. Forty-seven patients (47%) received targeted-based therapy after relapse, and only 5 (6%) and 4 (5%) received second-SCT and DLI, respectively. With a median follow-up of 11 months after relapse, the 1-year OS was 27%. In UVA, high/very high DRI and HCT-CI >3 were significantly associated with inferior OS. Receiving targeted-based salvage therapy and time to relapse after 5 months were associated with improved OS. In MVA, high/very high DRI was predictive for inferior OS (HR 1.862, 95% CI: 1.060-3.269; p=0.031) and time to relapse >5 months after haplo-SCT was associated with improved OS (HR 0.403, 95% CI: 0.235-0.609; p=0.001). Conclusion Relapse after haplo-SCT is associated with poor prognosis. Time to relapse (<5 months) and high/very high DRI predict poor prognosis, but integrating new targeted therapies seem to improve survival. Our findings highlight the continued unmet need for new therapeutics, not only to treat relapsed disease but also to reduce relapse rates after transplant (e.g. new conditioning platforms, maintenance therapies).