Project timeline
Where we are today
We have been working on mathematical techniques to make new clinical trial designs more legible. Michael demonstrates in his PhD thesis (100-page PDF here - chapter 5 is the relevant part) that for a large class of designs, if run large-scale simulations of a trial can be run, one can turn those simulations into a statistically rigorous proof about its Type I Error function [which is one of the biggest sticking points for FDA reviewers].
There are a few other requirements. For further information on the methodology see this seminar recording from May 2022, or slides from Michael's talk given to the FDA in September 2021.
Event
Barriers
Time
Initial use cases enter the FDA CID program
Specialized and optimized software delivered
Initial peer-reviewed paper to be published
+1-2 years
Specialized FDA teams de facto approve this approach
A number of peer reviewed papers published
FDA leadership engagement and support
Referenceable track record
+1-3 years
Typical drug development organization can use software
Typical FDA reviewer expected to accept
Execution risk and infrastructure for complex trials
Demand for complex trials grows
Low regulatory risk
Knowledge and comfort with use
Ecosystem of statistical methods
Accessible papers
Previous successes to point to
Engagement with all FDA reviewers
+2-10 years