The FDA has announced a major change in its new drug application (NDA) policy supporting the regulatory approval of new drugs: an end to the “two-trial dogma” standard established to ensure the safety and efficacy of medical products.
The shift was announced in a New England Journal of Medicine (NEJM) article published earlier this month.
Let’s start with current policy (up until now).
Under Section 505(d) of the Federal Food, Drug, and Cosmetic Act, the FDA’s clinical data standard for NDAs has been based on the requirement for "substantial evidence” supporting an investigational candidate’s safety and effectiveness.
- Generally speaking, this has translated to at least two adequate and well-controlled clinical investigations that independently show statistically significant benefits.
However, in some cases, the FDA has also demonstrated disease-by-disease flexibility and granted approvals based on:
- One large, multicenter trial with confirmatory evidence
- Permitted under U.S. law since 1997
- Smaller datasets with compelling evidence (for rare diseases)
This rundown offers a look at how sponsors’ interpretation of the FDA guidance has shifted over the decades—particularly as the federal agency has amended its own interpretation of the law.
So what you’re saying is two trials haven't necessarily been mandated?
Correct. In fact, a 2022 analysis of new NDA-approved medications in 2020 found over 50% did so with only one pivotal trial.
- And we can’t forget the flexible, special circumstances for investigational candidates (such as ophthalmic drugs for rare diseases) that have received the FDA’s Breakthrough program designations or are being accelerated via Priority Review pathways.
Yet despite this: That aforementioned “two-trial dogma” has still been the unofficial, long-standing expectations … until now.
Do tell.
In the NEJM paper published last week, Vinay Prasad, MD, MPH, director of the Center for Biologics Evaluation and Research (CBER), and FDA Chief Martin Makary, MD, MPH, confirmed the major change in federal regulatory standards.
The new guidance: Allows sponsors to file for drug approval using clinical data from a single adequate and well-controlled study combined with confirmatory evidence.
The agency will also examine “all aspects of study design,” with a particular focus on:
- Controls
- Endpoints
- Effect size
- Statistical protocols
You mentioned confirmatory evidence earlier as well … what does this involve?
This could include data demonstrating a drug’s work on a mechanistic, biological level or in animal models, data supporting related indications or other drugs within the same class, as well as real-world data—any of which could qualify as supportive and confirmatory evidence.
See here for the FDA’s latest draft guidance on this.
Why implement this new standard now?
The authors reasoned that the updated one-trial standard is intended to resolve the ongoing “confusion from manufacturers regarding settings in which a single trial will be accepted” that has ensued over the last several decades.
In their argument against the two-trial requirement: They noted that, in 2026, “there are … alternative ways to feel assured that our products help people live longer or better than requiring manufacturers to test them yet again.”
- In other words: Their reasoning is in favor of quality (of a single study) over quantity (multiple studies).
Did the authors call out any specific benefits of the revision?
A major advantage (according to Drs. Prasad and Makary, at least): Requiring only one trial will "substantially reduce costs for sponsors” as well as speed drugs to market.
By their estimation: A single pivotal study could cost anywhere from $30 to 150 million—and require years to complete, further increasing the average time it takes to commercialize a drug.
- As such: The authors reasoned that, with just one (instead of two) trials to conduct, “lowering capital costs for drug developers may remove a persistent argument in justification of lofty and rising drug prices for everyday Americans — the onerous cost of research and development.”
This is pretty major … but also potentially controversial, right?
Indeed. A fact that the paper addresses, with the authors writing a two-part response for such criticism:
- First: “The FDA has never been perfect, and even with a default requirement of two trials, the FDA has approved numerous products that were later found to have serious safety concerns or lack efficacy.”
- Second: “As we note, the number of clinical studies is no safeguard against valid inference if all other aspects of trial design are deficient.”
- They added: “If the control arm is substandard, the endpoints dubious, the statistical plan generated post hoc, the power inadequate, or all of the above, erroneous conclusions may be reached even with two, three, or four studies.”
That being said: The authors added that a greater focus placed on just one study could potentially improve the FDA’s standards and “reduce the overall risk of approval of products that may later need to be withdrawn.”
So will the new one-trial default apply to all investigational drugs?
Not quite … the paper authors noted that there will still be cases in which clinical findings from two studies will be required.
Examples of this include if a candidate:
- Is perceived as having a “nebulous, pluripotent, or nonspecific mechanism of action,”
- Affects a “labile, short-term, or surrogate outcome,” or
- Has some form of “underlying limitation or deficiency.”
Noted. And looking at ophthalmology specifically …
We checked in with the ophthalmic community to gauge the general reaction to this new standard.
- Our source: David Almeida, MD, MBA, PhD, founder of Erie Retina Research, investigator-in-chief at The Centers for Advanced Surgical Exploration (CASEx), and interim CEO of ClinOmics AI.
His feedback: “This is the most important regulatory shift in ophthalmology drug development in 60 years.”
- “The FDA has not lowered the bar—they have moved it,” he said. “The question is no longer how many trials you ran; it is whether you can tell the complete biologic story: the molecular mechanism, the target engagement, the confirmatory proteomic evidence.”