Foundational analysis
Why better tools did not automatically mean better R&D
Scannell and colleagues described a historical decline in new drug approvals per inflation-adjusted R&D dollar, despite improvements in many discovery tools.
Model performance study
Putting a human Liver-Chip to a predictive test
Communications Medicine · Dec 6, 2022
What the source reports
Ewart and colleagues evaluated a human Liver-Chip for predictive toxicology using a defined set of compounds and an accompanying economic analysis.
Why it belongs here
A concrete example of assessing a model against a specific translational question rather than relying on biological resemblance alone.
WHAT THIS DOES NOT ESTABLISH
Performance is specific to the study design and compounds. Review the paper’s commercial disclosures and its linked author corrections.
Before you use it
Read the original source, check its population or model and intended use, and review uncertainty, disclosures and any subsequent updates. This note is a guide to reading, not a substitute for the paper or guidance.
Open the original sourceKeep exploring.
Draft guidance
A validation framework for human-relevant methods
FDA’s March 2026 draft guidance sets out general scientific considerations for validating New Approach Methodologies in drug development and regulatory submissions.
Final guidance
How model-informed evidence should support a decision
ICH M15 addresses planning, evaluation and documentation of model-informed drug-development evidence. FDA’s final guidance is dated June 2026.