Total Body Irradiation-Based Versus Chemotherapy-Alone Myleloblative Conditioning Regimen Before Hematopoietic Allogeneic Stem Cell Transplantation in Adults With Acute Lymphoblastic Leukemia: A Multi-Institutional Study From the Francophone Society of Bone Marrow Transplantation and Cellular Therapy (SFGM-TC).

[PURPOSE] Total body irradiation (TBI) has been considered the myeloablative standard for fit adults undergoing allogeneic hematopoietic stem cell transplantation (HSCT). However, the only randomized phase III recently reported no benefit of TBI in adult patients. This report aims to investigate the role of TBI in patients from a large European database.

[MATERIALS AND METHODS] This European multicenter retrospective study included 3575 adults with acute lymphoblastic leukemia (1979-2019) in the SFGM-TC registry who received myeloablative conditioning before HSCT: 3166 TBI-based and 409 chemotherapy (CT)-only. Primary endpoints included overall survival (OS), progression-free survival (PFS), relapse incidence (RI), and nonrelapse mortality (NRM). Competing-risks and Cox models were applied. Prespecified subgroup analyses included a favorable cohort (CR1, age ≤ 40 years, 2010-2019) with inverse probability of treatment weighting (IPTW).

[RESULTS] With a median follow-up of 71 months, all efficacy outcomes were improved with TBI-based conditioning. The 5-year OS, PFS, RI, and NRM were 67.9% versus 55.5%, (P = .0002), 62.5% versus 46.9% (P < .0001), 37.5% versus 53.1% (P < .0001), and 32.6% versus 33.4% (P = .32) in TBI and the CT-alone groups, respectively. In multivariable analyses, TBI independently improved OS (hazard ratio [HR], 0.61), PFS (HR, 0.61), and RI (HR, 0.61). In the favorable subgroup (CR1, ≤40 years, 2010-2019; n = 496), IPTW‑adjusted OS (HR, 0.76; 95% confidence interval [CI], 0.44-1.31; P =.317), and PFS (HR, 0.73; 95% CI, 0.45-1.17; P = .187) were similar between regimens; 5‑year relapse was lower with TBI (20.3% vs 46.6%; cause‑specific HR, 0.65; 95% CI, 0.36-1.15; P = .135), with comparable 5‑year NRM (10.3% vs 10.4%; HR, 1.05; 95% CI, 0.41-2.65; P = .925).

[CONCLUSION] In this multicenter registry, TBI-based myeloablative conditioning was associated with lower relapse and improved OS/PFS versus CT-only, with similar NRM. In a contemporary favorable subgroup (CR1, ≤40 y, 2010-2019), survival rates were comparable, with relapse numerically lower after TBI. Randomized trials are needed to define who benefits most.