Comment from Matt Mahoney

Thank you for the opportunity to provide comments on the FDA Draft Guidance, General Considerations for the Use of New Approach Methodologies in Drug Development. The CARDIOVERSE project is an Advanced Research Projects Agency for Health-funded initiative focused on combinatorial NAMs that more accurately predict cardiotoxicity and safety profiles for drug candidates. Our feedback reflects our experience on this project. •The guidance would benefit from additional clarification of FDA’s thinking on the use of NAMs before qualification within a weight-of-evidence framework especially with respect to situations where NAM findings differ from large-animal study results. For example, a human stem cell–based assay may be consistent with safety while a traditional animal study suggests a risk. Such a result could reflect an important species difference that human-relevant NAMs are supposed to overcome or, alternatively, a failure of NAM platform. •FDA should further emphasize the importance of incorporating and evaluating human population and genetic variability in NAM development. While the draft guidance references donor variability in cell-based assays, additional recommendations on the use of multiple stem cell donors, genetically diverse datasets, or virtual patient populations would strengthen confidence in the predictive value and generalizability of NAMs. •Additional guidance on integrating evidence across multiple NAM platforms would also be valuable. Specifically, FDA could outline best practices for weight-of-evidence frameworks, especially for combinatorial in silico and in vitro NAMs. •Greater clarity would be welcome regarding validation expectations for advanced computational approaches, including AI-driven predictive models and virtual population simulations, and when in a development program would be appropriate to discuss these approaches with FDA. Example validation strategies, performance benchmarks, and regulatory expectations would help support consistent implementation and evaluation of these technologies. On behalf of the CARDIOVERSE project, Matt Mahoney