A Novel Therapeutic Regimen Based on the Constrained-Disorder Principles May Improve the Response to Dimethyl Fumarate in Multiple Sclerosis: Findings from an Exploratory Feasibility Open-Label Clinical Trial.

Introduction Dimethyl fumarate (DMF) is an oral disease-modifying therapy recommended for patients with relapsing-remitting multiple sclerosis (MS). However, a lack of efficacy, safety concerns, tolerability issues, and poor compliance often lead to discontinuation of DMF. In an innovative approach, this exploratory proof-of-concept clinical trial aimed to assess the feasibility and safety of introducing variability into treatment regimens to overcome compensatory mechanisms that underlie the loss of drug effectiveness and to generate preliminary signals of potential improvements in response to DMF. Methods Seven MS patients treated with DMF were enrolled in an open-label, uncontrolled feasibility clinical trial in which an app provided a personalized therapeutic regimen, resulting in variability in dosages and administration times within predefined ranges. Prespecified feasibility endpoints included: (1) successful app installation and use; (2) patient adherence to the algorithmic dosing schedule; (3) completion of the 12-week follow-up; and (4) absence of serious adverse events. Exploratory outcome measures included the Expanded Disability Status Scale (EDSS), neurofilament light chain (NfL) levels, and MRI, which were assessed only as hypothesis-generating signals. Results Although this study is a small exploratory proof of concept, the data suggest that AI-assisted personalized treatment is feasible and could improve clinical responses to DMF. The EDSS score improved in 2 patients (29%, p = 0.371), and the MRI remained stable in 6 patients (100%). These findings did not reach statistical significance and should be interpreted as exploratory only. A high patient engagement rate with the app was recorded during the study. No serious adverse events were observed. Summary The results of this feasibility trial support the safety and feasibility of using a personalized algorithm that randomizes DMF regimens and provide preliminary hypothesis-generating signal data regarding potential improvements in therapy response. These findings are insufficient to support conclusions about efficacy. Controlled studies are needed to confirm these findings. (NCT06385197)