Relapse Risks and the Impact of Second Transplantation on Outcomes in Pediatric Patients Undergoing Allogeneic HCT for Acute Leukemia

Abstract

Background: Allogeneic hematopoietic cell transplantation (HCT) is a potentially curative therapy for high-risk hematologic malignancies. Despite being curative, the post-transplant phase is often associated with complications. For example relapse. Up to 40% of pediatric patients undergoing transplant for acute leukemia suffer from relapse. To date, static prognostic models and biomarker monitoring have improved early relapse detection, but they do not dynamically integrate evolving patient data to predict relapse risk over time. Objective: In this retrospective time-to-event study, we aimed to find the characteristics and longitudinal biomarker patterns that distinguish patients who relapse from those who recover after their first HCT. Methods: We analyzed 255 consecutive pediatric patients receiving their first HCT. Relapse occurred in 51 patients (20%). Elastic Net regression was used to identify the most important biomarkers for the distinction between the relapse and the non-relapse group. The longitudinal patterns were investigated with generalized estimation equations (GEE) to characterize the trajectories of the immunomonitoring markers (e.g. lymphocytes, neutrophils). Cox proportional hazards model with time-varying covariates were used to estimate the effect of both baseline and dynamic factors on time to relapse. Results: Elastic Net identified absolute blast count, platelet count, percentage of reticulated platelet count, bilirubin levels and CD8 absolute counts as the most differentiating markers between relapse and non-relapse patients. GEE univariately indicated platelet count as a prognostic factor for relapse (p = 0.01). A linear predictor score stratified patients into high- and low-risk groups. At one year post-transplant, the high-risk group had a significantly lower relapse-free survival than the low-risk group (HR = 1.65, CI = 0.91-2.99). Conclusion: Integrating longitudinal biomarkers into a unified time-to-event framework improves relapse risk stratification after pediatric HCT. Continuous monitoring of the identified markers may enable earlier detection of high-risk patients and support personalized post-transplant interventions. External validation in multicenter cohorts is warranted before clinical implementation.