FDA panels: real‑world data and evidence can substitute where trials are infeasible — examples include alpelisib and Zolgensma

FDA experts explained when and how real‑world data (RWD) and real‑world evidence (RWE) can support regulatory decisions: the data source must be fit for purpose (relevant, reliable, traceable), study designs must guard against bias (clear eligibility, covariate capture, prespecified analysis), and sponsors should submit patient‑level data so reviewers can reanalyze results.

Panelists highlighted use cases. CDER described alpelisib’s accelerated approval for severe PIK3CA‑related overgrowth syndrome supported in part by an expanded‑access registry whose protocol, blinded central review and prespecified statistical plan enabled reviewers to conclude a meaningful benefit. CBER described Zolgensma’s approval, where a well‑characterized natural history registry provided the external comparator for open‑label trials in spinal muscular atrophy. CDRH described device examples (newborn screening assays) where hospital records and registries were used to confirm control‑status in device evaluation.

Challenges and mitigation: presenters emphasized the importance of prespecifying protocols and statistical analysis plans, demonstrating data provenance and traceability, handling missing data and carefully defining index dates. They recommended early pre‑submission engagement with FDA to agree on data sources and analytic approaches.

What to do next: sponsors seeking to use RWD/RWE should consult FDA guidances, engage reviewers early, post prespecified protocols and SAPs when appropriate, and be prepared to submit patient‑level datasets for review.