Third of cancer drugs without proven clinical benefit continue to be recommended for patients
According to a study published today in The BMJ, one-third of cancer drugs that received accelerated approval from the US Food and Drug Administration (FDA) continue to be recommended in clinical guidelines despite the fact that their confirmatory clinical trials failed to demonstrate improvement in their primary endpoints.
A primary endpoint is the principal outcome measure used at the conclusion of a study to determine if a given treatment was effective (eg. the number of deaths or the difference in survival between the treatment group and the control group). The primary endpoint is determined before to the study's commencement.
Clinical recommendations, the researchers argue, should be more aligned with the results of post-approval trials for cancer medications granted fast approval.
The FDA's fast approval procedure allows medications to be commercialised before their efficacy has been established, thereby expediting patient access to potential new drugs. However, as part of this approval process, the manufacturer is required to conduct post-approval research to verify clinical benefit (improved survival or quality of life in the case of cancer drugs). If these trials demonstrate a lack of benefit, the drug's approval can be revoked.
However, post-approval trials can take many years, and the FDA has historically been sluggish to withdraw a medicine or indication if post-approval trials fail to demonstrate therapeutic benefit.
As a result, a team of researchers from Canada and the United States set out to explore how the FDA handles cancer medications that were granted fast approval but later found to have unfavourable post-approval trials, as well as if these negative trials affect treatment guidelines.
They reviewed the FDA database from 1992 to December 2020 for all cancer treatments granted rapid approval and discovered 18 indications for ten cancer drugs that failed to demonstrate therapeutic benefit in post-approval trials.
Over an average of four years, 11 (61%) of these approvals were voluntarily withdrawn, one was revoked, and six (33%) remained on the drug's label.
The researchers next examined the most recent FDA and National Comprehensive Cancer Network (NCCN) guidelines and discovered that the majority of these medications retained high-level endorsement, sometimes even after permission for the particular indication was withdrawn or revoked.
These are observational studies, and the researchers acknowledge significant limitations, including reliance on publicly available data and awaiting FDA approval decisions, which may have impacted the accuracy of their findings.
Nonetheless, they assert that this is the most thorough study of its sort to date and that the findings "show the failure of the accelerated approval pathway's tradeoff between time and evidence."
They acknowledge that a recent wave of regulatory action "indicates that the FDA has paid increased attention to these situations over the last two years," but call for additional guidance and reforms to the accelerated approval pathway "to ensure that all FDA-approved drugs are demonstrated to be safe and effective for patients."