AI drug discovery [Jackie Niam / 2181661830 / iStock / Getty Images Plus]
Developing an effective cure for Rett syndrome has been extremely challenging, in large part due to the complexity of cognitive and physical impairments that come with the rare, genetic, childhood disorder. People with Rett syndrome (1 out of 10,000 girls—and far fewer boys) have mutations in the MeCP2 gene on the X chromosome that lead them to suffer from repetitive hand motions, speech difficulties, and seizures, but also problems in non-neurological organs, including the digestive, musculoskeletal, and immune system.
Now, a team at the Wyss Institute has made a significant breakthrough by leveraging an AI-driven drug discovery process. Their findings identified that the drug known as vorinostat—an HDAC inhibitor FDA-approved for the treatment of cutaneous T cell lymphoma—is a promising treatment for Rett syndrome, demonstrating disease-modifying abilities across multiple neuronal and non-neuronal tissues in preclinical models.
The findings are published in Communications Medicine in the paper, “AI-enabled drug prediction and gene network analysis reveal therapeutic use of vorinostat for Rett Syndrome in preclinical models.”
The Wyss Institute’s computational nemoCAD pipeline was key to the discovery of vorinostat as a potential Rett syndrome therapy. It enabled the team to predict drug candidates not based on a specific target molecule of the disease (like most traditional drug discovery approaches) but on changes that occur in the entire gene network across multiple organ systems in Rett syndrome.
“The identification and further development of vorinostat as the potentially first curative treatment for Rett syndrome would not have been possible without our unique AI-enabled computational approach to drug discovery, and its combination with an innovative disease model that broadly mimics the features of Rett syndrome,” said Donald Ingber, MD, PhD, Wyss founding director. “This new target-agnostic approach for drug discovery proved to be extremely fast and effective and, together with our unique technology translation capabilities, creates a model for us to tackle other diseases with unmet need that pose similarly enormous challenges.”
The team first created a small animal model consisting of tadpoles from Xenopus laevis, in which they generated various mutations inactivating the MeCP2 gene to reflect the diverse patient population. The engineered tadpoles recapitulated a range of critical features of Rett syndrome, including developmental and behavioral delay, seizures, as well as intestinal, muscle, and brain anomalies.
The researchers then used nemoCAD to compare all gene expression changes that occurred in MeCP2-defective tadpoles vs. healthy tadpoles. They then used a public database curated by the NIH named LINCS, which contains gene expression signatures induced by more than 19,800 drug compounds in a large variety of human cell lines. In doing so, they could predict drug compounds that could reverse the pathological changes in the same gene expression networks.
Vorinostat scored the highest on the list and produced the strongest therapeutic effects in the genetically engineered tadpoles, which showed an impressive reversal of their disease features on a whole-organism level.
The Wyss-enabled startup Unravel Biosciences will be initiating a proof-of-concept clinical trial to assess the drug’s efficacy and safety in 15 female patients with Rett syndrome in Colombia later this year.
Unravel Biosciences’ lead pipeline asset, RVL-001, is a proprietary formulation of vorinostat. Because vorinostat has already been approved by the FDA to treat a blood disease, the company has been able to rapidly repurpose this drug as a therapy for Rett syndrome.
It will also test an “n-of-1 trial design” to evaluate different vorinostat treatments within individual patients, which is more appropriate to the complexity of the disease and to rare disease communities in general.
