The accelerated approval program of the U.S. Food and Drug Administration (FDA) allows certain medications to be marketed if they are indicated for serious disease and there has been preliminary evidence of the drug’s efficacy. Pharmaceutical companies must then conduct postapproval trials to confirm the medication’s clinical benefit. If the postapproval trial fails, the indication should be removed from the product’s label. Some of these medications can remain available if they have other approved indications and may continue to be prescribed off-label for the withdrawn indication.
Hwang et al performed a cohort study to investigate changes in cancer drug use following accelerated approval of three medications for four indications after publication of negative confirmatory trials. Results of the study were published as a research letter in JAMA Oncology.
Study Details
Using the Optum deidentified Clinformatics DataMart Database, a national commercial claims database, cancer medications were identified that had withdrawn accelerated approval indications from January 2020 to December 2022, but remained on the market for at least one other approved indication that could be distinguished from the withdrawn indication on the basis of diagnostic codes. The medications included atezolizumab for breast cancer, atezolizumab for two urothelial cancer indications, idelalisib for follicular lymphoma, and romidepsin for peripheral T-cell lymphoma.
The study included 762,752 patients, grouped into four cohorts: breast cancer, urothelial cancer, follicular lymphoma, and peripheral T-cell lymphoma. Changes were measured in the monthly proportion of incident (no use in the prior 365 days) and prevalent medication use following the first public reporting of negative trial results.
The investigators reported that findings showed that the use of medications in the four groups increased following accelerated approval and before any negative results were reported from postapproval studies. Once the confirmatory studies published negative findings, however, use of these medications decreased significantly despite ongoing availability of the medications. The use of atezolizumab, for example, increased by 16 patients per 1 million patients (95% confidence interval [CI] = 12 to 21 patients/million) with breast cancer before the negative confirmatory findings were published; following publication of negative findings, use of atezolizumab decreased by 28 patients per million (95% CI = –35 to –20).
The time between accelerated approval and negative confirmatory trial results were published varied, ranging from 12 to 113 months; further, the time between publication of the negative trial and withdrawal of FDA approval also varied, ranging from 1 to 47 months. The investigators stated: “The use of the oncology medications we studied responded quickly to cautionary signals and decreased after confirmatory trials that failed to demonstrate efficacy, despite their ongoing availability via off-label prescribing…. This may be because oncologists rapidly incorporated such data into clinical practice or because payers restricted coverage of ineffective medications.”
Conclusions
The authors concluded, “Ensuring timely completion of postapproval studies and enforcing appropriate regulatory actions based on negative trial results is necessary to minimize patient exposure to ineffective and potentially unsafe medications.”
Catherine S. Hwang, MD, MSPH, Division of Pharmacoepidemiology and Pharmacoeconomics, Department of Medicine, Brigham and Women’s Hospital, is the corresponding author for the JAMA Oncology article.
Disclosure: This study was supported by a grant from Arnold Ventures and the Commonwealth Fund to Brigham and Women’s Hospital. For full disclosure of all study authors, visit JAMA.org.
