In This Section

Fewer Than Half of Accelerated Approval Drugs Showed Clinical Benefit in Confirmatory Trials After Five Years

SAN DIEGO – Of the 46 cancer drugs that the U.S. Food and Drug Administration (FDA) granted accelerated approval between 2013-2017, 63% were converted to regular approval even though only 43% demonstrated a clinical benefit in confirmatory trials after more than five years of follow-up, according to a study presented at the American Association for Cancer Research (AACR) Annual Meeting 2024, held April 5-10.

The study was simultaneously published in JAMA.

“Accelerated approvals make up a large percentage of new cancer drug approvals, but we knew that oncology accelerated approvals are often not subsequently confirmed as having an effect on ‘hard’ clinical endpoints such as overall survival (OS),” said Ian T. T. Liu, MD, JD, MPH, MS, who worked on the study and is a postdoctoral research fellow with the Program On Regulation, Therapeutics And Law (PORTAL) within the Division of Pharmacoepidemiology and Pharmacoeconomics at Brigham and Women’s Hospital and Harvard Medical School.

The FDA’s accelerated approval pathway was established more than 30 years ago to allow drugs that treat serious conditions and fill an unmet medical need to be approved early based on a surrogate endpoint. The FDA defines a surrogate endpoint as a marker or measure that is “thought to predict clinical benefit but is not itself a measure of clinical benefit.” Any drug approved under this pathway is still required to undergo follow-up trials to confirm the drug can achieve its anticipated clinical benefit. If the drug does not provide this evidence, then the FDA could remove it from the market.

Liu and his colleagues sought to evaluate how a more recent set of cancer drugs approved through the accelerated approval pathway performed after approval while also considering whether these drugs were found to improve patient quality of life (QOL), which previous studies did not examine.

They conducted two analyses. The first focused on all drugs granted accelerated approval from 2013-2017, meaning they had more than five years to complete a confirmatory trial. The second examined the evidence used to convert accelerated approvals to regular approvals for all drugs between 2013-2023, even if they had less than five years to evaluate their efficacy in clinical trials. They used publicly available FDA data to identify cancer drugs granted accelerated approval during this period, including information on the FDA website, industry press releases,, and peer-reviewed journal articles. The researchers pulled out relevant information from the trials including National Clinical Trial number, trial design, primary endpoint, and primary and secondary outcomes.

In total, Liu and colleagues found that 129 drugs were granted accelerated approval for a cancer-related indication from 2013-2023. Forty-six of those had more than five years for follow-up trials (the cohort examined between 2013-2017). Among those 46, 63% were converted to regular approval, 22% were withdrawn, and 15% remained ongoing after a median of 6.3 years. Only 43% demonstrated a clinical benefit in confirmatory trials.

During this period, the time to regular approval conversion increased from 1.6 to 3.6 years, while the time to a drug being withdrawn from the market decreased from 9.9 to 3.6 years.

“Faster, appropriate withdrawal decisions are a good thing for patients, as they ensure that ineffective drugs are on the market for a shorter period of time,” Liu explained. “While our study showed an increase in the time between accelerated approval and conversion to regular approval, we believe that conversion decisions should be both timely but—more importantly—supported by high-quality clinical outcomes, and that this is critical to the proper functioning of the accelerated approval pathway.”

In the second analysis, the researchers found that 48 of the 129 drugs granted accelerated approval from 2013-2023 were converted to regular approval. Of these, conversion was based on OS in 40% of the cases, progression-free survival in 44%, response rate plus duration of response in 10%, and response rate in 4%. In one case, a drug was converted despite a negative confirmatory trial. They also looked at whether the regular approval indications matched the accelerated approval indications and found that 63% were converted to regular approval for a different indication, usually a broader indication and often an earlier line of therapy in the same cancer type.

“We hope these findings will encourage greater communication between patients and physicians about the uncertainty surrounding cancer drugs approved on preliminary surrogate measures and the potential risks and benefits of a given treatment,” said Liu. “Our findings may also encourage regulators to scrutinize the commonplace practice of converting accelerated to regular approvals based on limited evidence, to invest resources in robustly validating more oncology surrogate measures, and to ensure that confirmatory trials will all be powered to show improvements in endpoints that matter to patients and their families such as overall survival and quality of life measures.”

The study’s senior author, Edward Cliff, MBBS, MPH, a hematology fellow and postdoctoral fellow at PORTAL, who presented the study, added, “The FDA relinquishes significant leverage once it converts accelerated approvals to regular approval—it is harder to ensure timely completion of further trials, and harder to withdraw drugs—so the evidence used to justify these decisions is important. We found seven drugs that were converted to regular approval based on response rate, i.e., tumor shrinkage, but this leaves a great deal of uncertainty regarding whether the drugs ultimately benefit patients.”

Limitations of this study include the fact that the researchers only looked at confirmatory trial data to evaluate evidence of clinical benefit and it is possible that larger studies, or studies with greater follow-up, may show different results. Additionally, seven of the drugs in the first cohort still had ongoing confirmatory trials, so efficacy data were not yet available. The researchers also used published OS and QOL data and it is possible that clinical benefit may have been overestimated since statistical improvements in endpoints may not be clinically meaningful to patients.

The study was funded by Arnold Ventures, but the funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; or decision to submit the manuscript for publication. Liu and Cliff report no conflicts of interest.

Download a photo of Liu Download a photo of Cliff