Abstract
Determining the potency of MSCs is a critical component of their application as cellular therapies. The function of MSCs does not rely on a single mechanism but rather on overlapping and cumulative effector pathways, which necessitates the assay matrix strategy in potency analysis. Artificial intelligence (AI) tools can significantly enhance the assay matrix strategy by generating novel potency scores that capture unified critical quality attributes that may not be readily discernible through human analysis. AI can provide precise potency metrics for investigational MSC products by comparing them to appropriate controls. The next generation of MSC potency analysis will increasingly rely on AI tools, as they can match patients with MSC products exhibiting the most appropriate potency profiles for personalized and targeted therapies. A significant challenge in deploying AI tools is the need for robust predictors of efficacy that relates to the potency of investigational MSC products. Nevertheless, AI has the potential to stratify patients who are most likely to respond to MSC therapy by leveraging clinical data in combination with detailed potency analyses. We discuss these opportunities and challenges in this perspective article.