Redefining Regulatory Proof: The End of the Two‐Trial Paradigm

Résumé fourni par la source

The recent decision to establish a single pivotal trial as the default evidentiary standard for Food and Drug Administration (FDA) approval marks a substantive shift in the architecture of regulatory proof. This marks a shift from replication-based validation to a coherence-based model of sufficiency, beyond a procedural adjustment. Because FDA standards likely influence global development strategies, this change will likely extend beyond the United States. Although accelerating access and reducing development costs are legitimate objectives, the reform raises methodological and epistemological concerns. The statistical rationale does not fully incorporate the contextual factors that influence false-positive risk, including prior plausibility and statistical power. Prior plausibility remains informal, and post-marketing evidence seldom restores the epistemic security provided by independent replication. Replication is not merely a statistical redundancy but rather a structural safeguard of robustness and external validity. Without explicit safeguards, increased regulatory flexibility may generate heterogeneity in evidentiary thresholds and progressively weaken the discriminative function of health technology assessment.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.