And if a molecule created to regulate sleep and wakefulness were to succeed, thanks also toArtificial intelligenceto open a new path against one of the most aggressive forms of multiple sclerosis? It is from this question that the scientists of the BRAVEinMS project started, coordinated by the Vita-Salute San Raffaele University and the Irccs San Raffaele Hospital in Milan. Through an approach of pharmacological repositioning supported by artificial intelligence, researchers have identified in bavisanthistamine H3 receptor antagonist, a promising therapeutic candidate for progressive multiple sclerosis.
The results in the experimental models
In experimental models of the disease, including man-mouse chimerasthe drug has shown the ability to act on two key fronts: the protection of neurons and the myelin repair. “We bet on an idea: combining AI, modeling based on stem cells and collaborative science to accelerate the discovery of new therapies for this disease,” explained Gianvito Martino, vice-rector of research at the Vita-Salute San Raffaele University and scientific director of the hospital during the presentation of the results. «Today that bet has produced not only a real candidate who is now only missing the last mile to get to the patient’s bed, but also further 30 possible new candidates potentially usable in progressive multiple sclerosis”.
From in silico analysis to final selection
The work looked further 1,500 drugs already known: the analysis in silico selected 273, and after a long series of tests aimed at establishing the toxicity of the molecules on nerve cells, this number was reduced to thirty-two candidates and finally to six compounds. Among these, bavisant stood out for the efficacy and safety profile.
The role of computational tools
«The platform Spokes it was fundamental for the prioritization of candidates”, underlined Sergio Baranzini, of the University of California San Francisco. «Integrating the graph theory and the machine learningwe were able to reduce thousands of compounds to a few hundred, simplifying subsequent testing steps in vitro And alive. This work highlights the power of computational tools in accelerating drug discovery.”
A response to a still open therapeutic need
There progressive multiple sclerosis it affects over one million people worldwide and approximately 15–20 thousand in Italy. It is a form of the disease for which therapeutic options remain limited: for this reason, the approach developed by the BRAVEinMS project represents an important step in the search for new treatment strategies. The time has come for theArtificial intelligence show us what you really can do.
