Experts recommend increased visibility for regulatory analysis of primary clinical trial data

The authors of a recent JAMA Oncology article said the US Food and Drug Administration (FDA) and other regulatory authorities should publish their reviews of primary clinical trial data on academic platforms alongside published trial data from investigators and sponsors.

Researchers from the Clinical Trials Center at the National Cancer Center and Cancer Hospital at the Chinese Academy of Medical Sciences in Beijing, China, published a Viewpoint article stating that confirmatory clinical trial findings significantly affect medical practices and treatment decisions.

“Although peer review processes are stringent, reviewers often base their evaluations on secondary data disclosed by authors and sponsors,” the article states. “This method creates a vulnerability, particularly with industry-sponsored trials, in which the published narrative may not present a fully balanced account.”

The authors compared results from published research articles with FDA’s review of three drugs discussed in recent Oncologic Drugs Advisory Committee (ODAC) meetings: the chimeric antigen receptor T-cell therapy ciltacabtagene autoleucel, the KRAS^G12C inhibitor sotorasib, and the poly(adenosine diphosphate–ribose) polymerase inhibitor olaparib.

In the case of olaparib, FDA raised concerns during the ODAC meeting about the trial population’s heterogeneity, and identified potential worse outcomes for patients with metastatic castration-resistant prostate cancer without BRCA variation status.

“This discovery, indiscernible in the published report, led the FDA to restrict approval to patients with BRCA variation, a decision contingent on scrutiny beyond standard peer review,” said the authors.

They also noted that during the ODAC meeting for sotorasib, FDA identified issues with systemic bias in the open-label trial, which included asymmetrical dropout as well as “investigator imaging assessments favoring the sotorasib group, early crossover according to investigator assessment that affected blinded independent central review assessments, applicant-triggered radiologic rereadings altering interim progression-free survival results, and potential misuse of the confirmation of progression procedure.”

The authors said that while the early dropout was reported in published studies prior to the ODAC meeting, the other concerns were not.

Although sotorasib’s accelerated approval status was not revoked, the authors said FDA's concerns will continue to influence physicians' assessments of its risk-benefit profile during clinical practice.

Ciltacabtagene autoleucel’s ODAC meeting highlighted several issues not identified in published journal studies, which included a higher rate of progression-free survival events in the treatment group leading to death, a higher rate of treatment-emergent adverse events leading to death in the 90 days after starting treatment, and a trend of declining overall survival in the first 10 months following treatment.

“These findings highlighted apprehensions regarding patient mortality linked to treatment delays and adverse reactions, aspects not clearly elucidated in the published trial results,” the authors said, noting the discussion from the ODAC meeting “remains highly informative for clinicians considering use of this therapy.”

They also emphasized that the issues highlighted were not isolated.

“They show a pattern in which pivotal trials achieve their primary end points, yet subsequent FDA regulatory analysis of the primary data reveals critical risks or methodological flaws,” said the authors. “This gap suggests that conventional peer review, although essential, is insufficiently stringent to vet complex raw clinical trial data.”

The authors also raised concerns that “some manuscripts may be crafted to support a specific narrative rather than to provide a balanced assessment.”

To address these concerns, the researchers suggested that investigators publish primary data from confirmatory clinical trials in published journal studies to improve the peer review process.

“[W]e advocate for regulatory bodies to disseminate the outcomes of their evaluations of primary clinical trial data via academic platforms, aiming to empower clinicians with enhanced information for making well-informed medical decisions,” the authors said. “We encourage clinicians to pay close attention to regulatory opinions, particularly those based on reanalysis of primary data and assessments of study conduct.”

Should FDA do more to share its independent review data?

Commenting on the analysis, Aaron Kesselheim, professor of medicine at Harvard Medical School and director of the Program On Regulation, Therapeutics, And Law (PORTAL) group at Brigham and Women’s Hospital and Harvard Medical School, said FDA publishing the results of analyses presented at advisory committee meetings “is a natural next step.”

“The minutes of [advisory committee] meetings are available for public viewing, but few people know how to access them and published articles are more accessible,” he told Focus.

However, Kesselheim said he is not sure if FDA would be amenable to the idea.

“Most significantly, such a step would take more time and resources, and the FDA recently cut its workforce substantially and its budget is being contracted,” he explained.

FDA is also significantly cutting down on advisory committee meetings, even though advisory committee meetings were held for between 5% and 10% of new drug approvals, Kesselheim noted.

“If we want to better communicate the basis for FDA decisions with clinicians, we need to think more strategically than either more [advisory committees] or publishing FDA analyses using direct outreach, drug fact boxes, and tools specifically designed for communication,” he said.

Joseph Ross, professor of medicine and public health at Yale School of Medicine, also commented on the study and said he strongly agrees with the importance of disseminating FDA’s independent reviews.

“There’s lots of information that sometimes diverts from what the companies have summarized or published in the independent reviews from the FDA,” Ross told Focus.

However, he noted that publishing them in journal articles may not be the right approach.

Rossu pointed out that one issue is that companies have not shared their data with FDA with the expectation that the agency would publish it.

“They’ve shared the data in order to get marketing authorization of their product,” he noted.

However, Ross said that it is important that FDA publicly posts its independent reviews of every drug that is approved, and there may be better ways of sharing that data in addition to posting them on the agency’s website and having the results data available on ClinicalTrials.gov.

“Perhaps the solution instead is to have the National Library of Medicine index the FDA's review packages in some way, so that when somebody searches for evidence on a particular drug, they're brought to the FDA's page where they publicly posted all of their independent assessments,” he said.

JAMA Oncology article

Experts recommend increased visibility for regulatory analysis of primary clinical trial data

Related topics

Recommended content

Welcome to the new RAPS Digital Experience