Effectiveness of disease-modifying treatment (DMT) in people affected by primary progressive multiple sclerosis (PPMS) is limited. Whether specific subgroups may benefit more from DMT in a real-world setting remains unclear. Our aim was to investigate the potential effect of DMT on disability worsening among patients with PPMS stratified by different disability trajectories. Within the framework of the Big MS Data network, we merged data from the Observatoire Français de la Sclérose en Plaques, the Swedish and Italian MS registries, and MSBase. We identified patients with PPMS that started DMT or were never treated during the observed period. Subpopulations with comparable baseline characteristics were selected by propensity score matching. Disability outcomes were analysed in time-to-recurrent event analyses, which were repeated in subclasses with different disability trajectories determined by latent class mixed models. Of the 3243 included patients, we matched 739 treated and 1330 untreated patients with a median follow-up of 3 years after pairwise censoring. No difference in the risk of confirmed disability worsening (CDW) was observed between the groups in the fully matched dataset (HR 1.11, 95% CI 0.97 to 1.23, p=0.127). However, we found a lower risk for CDW among the class of treated patients with an aggressive disability trajectory (n=360, HR 0.68, 95% CI 0.50 to 0.92, p=0.014). In line with previous studies, our data suggest that DMT does not ameliorate disability worsening in PPMS, in general. However, we observed a beneficial effect of DMT on disability worsening in patients with aggressive predicted disability trajectories.< Réduire