FDA Drafts Guidance on Bayesian Statistics in Drug and Biologic Trials

The US Food and Drug Administration has published draft guidance on how drug and biologic developers can use Bayesian statistical methods in clinical trials, including to support the main conclusions of pivotal studies. The agency announced the document in a press release dated January 12, the same day a notice of availability appeared in the Federal Register. Regulatory Focus reported that the draft was issued on Friday, January 9.

"Bayesian methodologies help address two of the biggest problems of drug development: high costs and long timelines," FDA Commissioner Marty Makary said in the release. "Providing clarity around modern statistical methods will help sponsors bring more cures and meaningful treatments to patients faster and more affordably."

What the draft covers

According to the Federal Register notice, the guidance is aimed at sponsors and applicants submitting investigational new drug applications, new drug applications, biologics license applications and supplemental applications. "The primary focus of this guidance is on the use of Bayesian methods to support primary inference in pivotal clinical trials designed to support the effectiveness and safety of drug and biological products," the notice says.

In a Bayesian analysis, the notice explains, data collected in a study are combined with a prior distribution expressing a pre-study position about a parameter of interest, producing a posterior distribution used to draw conclusions. The prior often summarises data or information available before the study begins.

The draft gives recommendations on choosing success criteria, evaluating a trial's operating characteristics and determining the prior distribution, and includes detailed considerations for using an informative prior to borrow external information in the primary analysis of a prospective trial, according to the notice. It also discusses estimands and missing data, software and computation, and documentation and reporting.

The FDA's release lists uses such as determining futility or success earlier in adaptive trials, informing dose selection in later trials, and incorporating information from previous studies, real-world evidence and external or nonconcurrent controls. The agency said the methods may be especially valuable for sponsors targeting rare or pediatric indications, where patient populations are smaller.

A user-fee commitment and an example from a past approval

The FDA said publishing the draft is intended to satisfy its commitment under PDUFA VII, the sixth reauthorization of the Prescription Drug User Fee Act, in which the agency and industry agreed to performance goals on reviewing complex innovative trial designs. Regulatory Focus noted that the agency's Complex Innovative Trial Design initiative, which began under PDUFA VI, promotes Bayesian analysis in trials.

As an example, the guidance describes how the phase 3 study of Rebyota, a fecal microbiota product for preventing recurrent Clostridioides difficile infection, used a Bayesian model to formally incorporate data from an earlier phase 2 placebo-controlled study, Regulatory Focus reported. Rebyota was approved in 2022.

Sponsors are expected to justify the study design and planned analyses in the protocol and to provide supporting information on the proposed prior, any external data being borrowed, success criteria and operating characteristics, and to submit it to the FDA as early as possible in the design phase, according to Regulatory Focus. Written or electronic comments on the draft are due by March 13, 2026, under docket number FDA-2025-D-3217, the Federal Register notice says. The guidance was issued by the FDA's drug and biologics centers.