FDA Schedules Bayesian Guidance Webinar For Regulatory Teams

Regulatory affairs and clinical trial teams working on pediatric and rare disease programs face a tightening documentation standard: FDA's January 2026 draft guidance on Bayesian methodology sets explicit expectations for prior distributions, estimands, missing data handling, and software disclosure, and a CDER-hosted panel on October 14 offers direct access to the biostatisticians who wrote it.

The event, hosted by CDER's Small Business and Industry Assistance (SBIA) program, runs from 12:00 p.m. to 1:00 p.m. ET and is structured as a pre-recorded overview followed by a live panel discussion. Attendees are required to complete the on-demand video before joining the live session. Presenters include John Scott, PhD (CBER Division of Biostatistics), James Travis, PhD, and Van Tran, PhD (both from CDER's Office of Biostatistics), with panelists Mark Rothmann, PhD and John Alexander, MD, MPH from the Division of Pediatric and Maternal Health.

The draft guidance itself addresses a PDUFA VII commitment and covers five substantive areas: success criteria and operating characteristics, use of external data in prior distributions, estimands and missing data in a Bayesian framework, software and computation requirements, and documentation and reporting standards. For regulatory submissions involving adaptive designs or small-population trials, each of these areas carries direct implications for protocol design and statistical analysis plan construction.

Bayesian approaches are positioned in the guidance as most applicable where leveraging historical or external study data is scientifically justified, a condition that arises frequently in pediatric extrapolation and rare disease development, where conventional frequentist power calculations are often infeasible given enrollment constraints. Regulatory affairs leads preparing IND or NDA submissions that incorporate Bayesian elements should treat the guidance's documentation expectations as a pre-submission checklist, particularly around prior specification justification and sensitivity analyses.

Continuing education credit is available through RAPS, SOCRA, SQA, and ACRP for real-time attendees; certificates are issued within two weeks post-event and require live participation, not on-demand viewing alone.

The October 14 session represents a direct line to the FDA reviewers who will evaluate Bayesian statistical packages in submissions, making pre-registration and advance completion of the on-demand module the practical first step for any team with an active or planned Bayesian trial design.

Source: FDA CDER SBIA via FDA.gov What's New: Drugs RSS Feed, published August 24, 2026. Event scheduled October 14, 2026, 12:00–1:00 p.m. ET.