Merck and Mayo Clinic Partner on AI for MS Drug Development
Merck and the Mayo Clinic have formed a partnership to use machine learning and clinical datasets for AI-driven drug development in multiple sclerosis. The collaboration will leverage large, harmonized patient data with advanced analytics to accelerate discovery and de-risk the development process. This alliance underscores the strategic value of combining clinical data with AI to improve therapeutic outcomes.
- The collaboration will utilize Mayo Clinic's Platform_Orchestrate, a program providing access to de-identified clinical and multimodal datasets, including registries and biorepositories, alongside advanced AI and analytics tools. This marks the first time Mayo Clinic has engaged in a strategic collaboration of this magnitude with a global biopharmaceutical company. - Merck will integrate data from the Mayo Clinic Platform with its own internal "virtual cell" platforms, which are computer-based models designed to simulate cellular behavior in both healthy and diseased states to improve drug target identification. - In addition to multiple sclerosis, the partnership will initially focus on developing treatments for atopic dermatitis and inflammatory bowel disease, conditions also identified as having high unmet medical needs. - This collaboration follows a recent setback in Merck's multiple sclerosis pipeline; in late 2023, their investigational oral Bruton's tyrosine kinase (BTK) inhibitor, evobrutinib, failed to meet its primary endpoints in two Phase III trials for relapsing MS. - The multifactorial and heterogeneous nature of multiple sclerosis presents significant challenges to traditional drug discovery, making AI-driven approaches that can analyze high-dimensional biomedical data particularly valuable for identifying new therapeutic targets and personalizing treatments. - For the biomanufacturing sector, this partnership exemplifies a broader industry trend of leveraging digital twins and AI to de-risk development. These technologies can simulate and optimize complex biological processes like cell culture and purification, predicting critical quality attributes and reducing out-of-spec events in a cGMP environment. - A significant hurdle in applying AI to cell and gene therapy manufacturing is the lack of data standardization across fragmented systems, which complicates the development of robust, generalizable predictive models. This collaboration's focus on large, harmonized datasets directly addresses this critical infrastructure challenge. - The application of AI in a GMP-regulated environment is a key focus for the industry, with efforts to use existing frameworks like Computerized System Validation to manage and mitigate potential new risks associated with machine learning and other advanced algorithms.
