Comparative data on reduced-dose (8 Gy) versus standard-dose (12 Gy) total body irradiation (TBI) before allogeneic haematopoietic stem cell transplantation (allo-HSCT) for adults with acute lymphoblastic leukaemia (ALL) remain limited. We retrospectively analysed 101 adults undergoing first allo-HSCT between 1999 and 2024 after 8 Gy or 12 Gy fractionated TBI-based conditioning. Relapse and non-relapse mortality (NRM) were analysed using multivariable Fine–Gray models. Median follow-up was 8.1 years. Although the 8 Gy cohort was older, had greater comorbidity burden and a higher-risk disease profile, unadjusted NRM, relapse incidence and overall survival did not differ between groups. In multivariable analyses, 12 Gy TBI was associated with increased NRM (sub-distribution hazard ratio 5.97, 95% confidence interval [CI] 1.77–20.15; p = 0.004) without evidence of reduced relapse incidence. The association between 12 Gy TBI and higher NRM was consistent across sensitivity analyses, including adjustment for transplant year and conditioning regimen, cause-specific Cox regression and propensity score overlap weighting. The association of 12 Gy TBI with inferior overall survival in the multivariable model was not confirmed after overlap weighting. Consistent pre-transplant measurable residual disease data were unavailable, limiting adjustment for this prognostic factor. These findings support prospective evaluation of reduced-dose TBI in selected adults with ALL.