Title: Personalized response to treatment in patients with MS: Do different patients show benefits on different outcomes?
Authors: Francesca Bovis - University of Genoa (Italy) [presenting]
Abstract: Composite outcomes were proposed to study disability progression in MS, defined as the progression in at least one of a set of clinical variables as EDSS, T25FW, 9HPT and SDMT. Composite outcomes suffer from well-recognized limitations. Moreover, patients with different baseline demographic and clinical profiles can have a different propensity to respond to one specific domain. We combine the concept of defining responders to therapy according to their baseline profile with the concept of evaluating the treatment effect on multiple endpoints, with the idea that patients can have benefits from different outcomes. The treatment effect on 4 clinical endpoints (EDSS, T25FW, 9HPT, SDMT) was evaluated in an RCT assessing the Siponimod effect versus placebo. For each endpoint, a response score (RS) based on baseline characteristics was generated to characterize responders on the time to progression on that outcome, according to a previous method and using a training-validation procedure replicated on 500 bootstrap samples. Four different RS were obtained and validated, all showing a significant interaction with treatment, defining responders to each outcome. The scores were split into two groups according to a pre-specified cut-off. We showed that the treatment effect estimated in an RCT can be different on different outcomes and on different patients. This methodology allows personalizing the treatment effect according to the baseline patients' profile.