Mikhail And Davis Back Rare-Disease Regulatory Adaptation, Reframing How FDA Reviews Small-Population Therapies

The FDA’s permanent appointments of Karim Mikhail to the Center for Biologics Evaluation and Research and Michael Davis to the Center for Drug Evaluation and Research point to a more defined regulatory posture for rare disease therapies: not a lower standard, but a framework adapted to the realities of tiny patient populations and difficult trial execution.

That distinction is more than word choice. At an FDA Town Hall at the BIO International Convention in June, BIO President John Crowley argued that calling the approach “flexibility” wrongly suggests weaker evidence requirements. Mikhail agreed with the framing, saying the goal is an adapted regulatory framework for rare diseases. His argument was that too much flexibility can reduce predictability, while an adapted framework could make decisions more predictable for sponsors.

The Regulatory Shift

Earlier this month, the FDA appointed Mikhail and Davis as permanent center directors after both had served in acting roles. In a note shared the same day, RBC Capital Markets said both appointments would have a positive impact on the rare disease space. The firm described Mikhail as supportive of “flexible trial design for cell and gene therapies, especially for rare indications,” and said Davis’ appointment could benefit CNS, neuropsychiatry, rare disease and novel modalities.

The practical importance of that view is already visible in uniQure’s Huntington’s disease gene therapy. Under former Commissioner Marty Makary and former CBER Director Vinay Prasad, the company’s program had drawn public criticism, and the FDA had insisted on a new trial with a sham surgery control before it would consider a regulatory filing for approval. Under Mikhail, the agency instead concluded that uniQure’s existing data were sufficient for review, and the company filed its biologics license application earlier this month.

That approach resembles how CBER operated under Peter Marks, who argued in 2024 that non-randomized, single-arm trials may be the best option for some rare disease gene therapies because randomized trials can be extremely difficult or impossible in populations numbering only dozens to a few hundred people in the United States.

The Tools In Play

At the BIO Town Hall, Crowley pointed to several mechanisms that could support an adapted framework, including accelerated approval, adaptive study designs, Bayesian statistics and a range of endpoints agreed upon in advance. UniQure used what the FDA has called adaptive study designs by relying on external controls rather than the sham surgery control that Makary and Prasad later said would be required.

The FDA under Makary and Prasad also introduced policies meant to support rare disease development despite their objections in the uniQure case. One example is the plausible mechanism pathway, introduced in November 2025, which is intended to provide alternate criteria to determine efficacy in exceedingly small patient populations where randomized, controlled trials are not possible.

Taken together, these examples suggest the agency is not abandoning evidentiary discipline. It is moving toward a model that tries to match the evidentiary tool to the feasibility of the disease setting.

The Endpoint Problem

Trial design is only part of the issue. Endpoint alignment remains a major source of regulatory risk, as shown by Capricor Therapeutics’ Duchenne muscular dystrophy cell therapy deramiocel. The therapy was rejected in July 2025 after Prasad reportedly canceled an advisory committee. After Capricor resubmitted and the FDA accepted the BLA application in March, the agency held an adcomm meeting in July.

There, Capricor and the FDA disagreed over which endpoints mattered most. Deramiocel met the Phase 3 trial’s primary endpoint with statistically significant benefits in upper-limb function. But much of the advisory discussion centered on a secondary endpoint involving decline in cardiac function measured by left ventricular ejection fraction. Advisers were asked to vote on that question and recommended against approval by a 9-3 tally.

Nicholas Richardson of Precision for Medicine said the functional outcomes from the trial were meaningful, but they were not the main focus of the discussion. He raised the possibility of a multi-component endpoint within an adaptive framework, meaning multiple clinically meaningful outcomes that may not match an existing FDA precedent but can be objectively measured.

The signal for drug developers is that rare disease strategy now depends less on generic calls for regulatory flexibility and more on securing early alignment around evidence structure, controls and endpoints. If Mikhail and Davis can turn that philosophy into consistent review practice, the payoff would be lower execution risk in categories where traditional trial models often break down.