Answers are grounded in this paper and cite their sources.
Fast-track procedures involve an intermediate registration step that enables earlier patient access to innovative medical products. Applicants can first apply for conditional registration with only limited evidence for therapeutic effects, leading to a limited funding or approval period and time to prepare the application for permanent registration with more convincing evidence. This article investigates the underexplored option of achieving conditional registration with a kind of weak evidence in the same primary endpoint used later for permanent registration. This is possible with the EMA's conditional marketing authorization and is the preferred two-stage strategy in the German DiGA fast-track program for the reimbursement of digital therapeutics. We carefully discuss regulatory requirements, valid and efficient study designs as well as power considerations for such fast-track registration strategies. In particular, we propose two-stage adaptive designs that use the first-stage data for conditional registration and for planning the permanent registration study, where the decision on permanent registration can then be based on the data from both stages. The designs include an interim sample size recalculation for achieving a target conditional power and aim to achieve a target overall probability for conditional and permanent registration. This requires specific minimum and maximum sample sizes. By analyzing these sample sizes as well as expected ones, we show that adaptive designs usually perform much better than designs with separate studies for conditional and permanent registration.
No takes yet. Share an insight, caveat, or question.
Kluge et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: