The FDA proposes ditching comparative efficacy studies for biosimilars

Under Commissioner Marty Makary, M.D., M.P.H., the FDA has focused on accelerating drug development, including piloting artificial intelligence (AI) efforts, updating processes and reevaluating what evidence is needed to approve new drugs and therapies.

This effort to accelerate development can be clearly seen in how the agency is approaching biosimilars. In October 2025, the FDA issued a guidance update indicating that “switching” studies — a trial that requires patients to be switched between the reference product and the biosimilar to demonstrate interchangeability — are no longer required.

At the same time, the FDA proposed eliminating comparative efficacy studies for a biosimilar application, according to a draft guidance issued in October 2025. Instead, manufacturers could rely on comparative analytical assessments to support biosimilarity.

If the agency makes this change to elimate comparative efficacy studies, laboratory analytics could be used to demonstrate that a potential biosimilar is structurally similar in terms of quality, purity and potency, which would indicate that it has the same safety and efficacy profile.

This would represent a significant change in how biosimilars are developed and approved by the regulatory agency. Currently, biosimilar developers have to conduct clinical trials comparing their biosimilar products against the reference biologics to demonstrate there is no meaningful difference. These studies can take anywhere from one to three years and cost about $24 million on average.

Eliminating comparative efficacy studies would speed development and reduce development costs by 80%, which would lower prices and increase access, Sarfaraz K. Niazi, Ph.D., adjunct professor at the University of Illinois College of Pharmacy, said in an interview. “Cost is the reason why we don’t have more biosimilars,” he said.

Biologic medications make up only 5% of prescriptions in the United States but account for 51% of total drug spending as of 2024, according to an FDA factsheet. Medicare and Medicaid are facing cost pressures to provide access to these medications and in the commercial market, high deductibles mean patients face high costs.

Biosimilars can generate savings for the healthcare system and can expand access to biologics. Savings attributable to U.S. biosimilars have been estimated at $36 billion through 2023, with $12.4 billion occurring in 2023 alone, according to the Association for Accessible Medicine.

The uptake and market share of biosimilars have grown slowly and have varied across therapeutic categories. But as a whole, they account for only 24% of competitive molecule volume, with ranges from 8% of market share for insulin lispro (which references Humalog and treats diabetes) to 82% of market share for bevacizumab (which references Avastin to treat certain cancers) three years after biosimilar entry, according to the most recent Use of Medicines report from IQVIA Institute for Human Data Science.

The biosimilar market is expected to grow 13.8% and be valued at $122 billion by 2034, according to one estimate. Niazi says if some hurdles for biosimilar development were eliminated, competition would increase and this could lower prices.

Scientific understanding changed

Niazi, who has published 10 research papers on biosimilar statistics, argues efficacy studies for biosimilars are scientifically meaningless. He explained that since biosimilars must already demonstrate they are analytically nearly identical to reference products — 99.99% similar — the clinical studies are using a measurement that is less sensitive than the analytical tests already required.

“Any study that is for the purpose of comparison of any property must have a pre-agreed, acceptable difference. This is part of the statistics,” he said. “Many studies had a 20% to 40% difference that was accepted. Theoretically, they can never fail, no matter what you do. Two products that are analytically similar, which means 99.99% similar, will have the same response [in a comparative efficacy trial].”

Since the FDA’s 2015 publication of its guidance for developing biosimilars and the approval of Zarxio (filgrastim-sndz) — the first biosimilar, which references Neupogen and is used to treat neutropenia and boost infection-fighting white blood cells — scientific understanding has evolved. The agency has gained significant experience in evaluating biosimilarity.

"Science continues to evolve, and the FDA remains committed to advancing commonsense policies that further promote efficient and effective biosimilar and interchangeable biosimilar development, without compromising safety and effectiveness,” George Tidmarsh, M.D., Ph.D., director, FDA Center for Drug Evaluation and Research, said in an FDA news release.

The FDA points out in its recent draft guidance that a comparative analytical assessment is more sensitive than a comparative efficacy study. “Currently available analytical technologies can structurally characterize highly purified therapeutic proteins and model in vivo functional effects with a high degree of specificity and sensitivity using in vitro biological and biochemical assays,” regulators said in the draft guidance.

Removing the requirement for comparative efficacy studies would align with how other regulators are reviewing biosimilars. The European Medicines Agency (EMA) and the UK’s Medicines and Healthcare Products Regulatory Agency (MHRA) already have adopted similar positions.

There is also precedent within the FDA. In 2019, the agency issued similar draft guidance to rely on comparative analytical assessment for biosimilar insulins.

Niazi argues that further changes can be made to the biosimilar approval process, including removing the requirement for a pharmacokinetic (PK) analysis, which studies how the body interacts with a drug. These studies look at absorption, distribution, metabolism, and excretion of a drug.

“A PK study is not bioequivalence,” Niazi said. “If two products have the same half-life and the same clearance, then to the body, it is a molecule. The process for how the molecule came to be doesn’t matter. It can be different, but as long as what you’re delivering at the end of the day is the same active molecule, then it should have the same side effects and the same efficacy.”

He says PK studies should be required if a biologic affects the immune system, because that alters the kinetics of a drug. But in many cases, it would not be needed. Niazi has filed a Citizen’s Petition with the FDA that asks the agency to establish criteria for when a PK study can be waived, especially for intravenous, ocular, or tissue-specific products.

Intraocular biologics, he argues, have negligible systemic exposure. PK studies are done by testing blood samples. He says in his petition that this means PK studies are scientifically irrelevant and put study participants at unwarranted risk. Additionally, he says waiving PK studies of intravenous biologics has a precedent in the current requirement for intravenous delivery of small molecules.

The Citizen Petition also suggests that the US Pharmacopeia develop nonproprietary product specifications for licensed biologics. By using these specifications, biosimilar manufacturers would not have to independently acquire and test multiple lots of the reference product. This process, he argues, results in inconsistent analytical comparisons.

The draft guideline for eliminating the comparative effectiveness was issued for comment purposes only and is not for implementation. Agency officials will review comments, but it may take a few months, Niazi said, to revise if necessary.