Time-dependent analysis of adoptive immunotherapy following sequential FLAMSA-reduced intensity conditioning and allogeneic hematopoietic stem cell transplantation in patients with high-risk myeloid neoplasia.

Prophylactic donor lymphocyte infusions (DLI) are part of the sequential FLAMSA-reduced intensity conditioning (RIC) regimen to cure high risk myeloid neoplasia with allogeneic hematopoietic stem cell transplantation (HSCT). Although DLI themselves carry significant risks, their prophylactic use has not been analyzed in a time-dependent manner. One hundred and fourteen patients underwent FLAMSA-RIC HSCT between 2013 and 2020. Next to Kaplan-Meier estimation of overall, disease-free, and graft-versus-host relapse-free survival (OS, DFS, GRFS), cumulative incidences of relapse and death in remission were calculated in a competing risk model. Additionally, the contribution of prophylactic and preemptive DLI as time-dependent covariates was assessed using a time-varying model toward DFS (Simon-Makuch method, Mantel-Byar test). At 2 years, OS was 45.2% [95% CI 36.7-55.7%], DFS 31.8% [95% CI 24-42.2%] and GRFS 11.3 [95% CI 6.5-19.8]. Neither prophylactic nor preemptive DLI showed a significant influence on DFS when considered time-dependent covariates (Mantel-Byar, p = .3). This was further corroborated in competing risk analysis with DLI as time-dependent covariates. Both prophylactic and preemptive DLI miss significance in their impact on survival within a high-risk cohort in a time-varying model. Controlled trials to address the impact of postgrafting immunotherapy approaches are needed.