Share on Facebook
Share on Twitter
Share on LinkedIn
Share by Email
Print
Researchers are changing primary endpoints for trials of immune checkpoint inhibitors (ICIs) without making note of these changes, a systematic review suggests.
The review, published in JAMA Oncology, also showed that some ICI studies did not have their original protocol available.
Researchers conducted this review of 38 phase 3 randomized controlled trials of ICIs. The trials included a total of 31,647 patients with urothelial carcinoma, renal cell carcinoma, or non-small cell lung cancer.
Changes to at least 1 part of the primary endpoint were made in 24 trials (63%). Changes to the patient population were made in 19 trials (50%), and changes to the outcome measure were made in 13 trials (34%).
Of the 10 trials that included multiple groups in the endpoint comparison, 7 trials (70%) changed specifications of the groups.
There were 8 publications in which authors reported changing the primary endpoint. The reason for the change was reported in 5 of these cases.
The most common changes were switching to or adding overall survival (OS) as an outcome measure or adding different PD-L1 expression subgroups to the comparison.
“In some cases, these changes led to substantially altered conclusions,” the researchers wrote. “In CheckMate 9ER, the primary endpoint population was changed to also include favorable-risk patients (capped at 25%) in addition to poor/intermediate-risk patients. By including this group in the primary endpoint population, nivolumab and cabozantinib now can be used in all risk groups instead of only poor/intermediate risk, although subgroup analyses showed no significant benefit for progression-free survival or OS for patients with favorable risk.”
The researchers also found that the original protocol was not available for 19 trials (50%). For some protocols, text related to primary endpoints was blacked out.
“These problems require greater transparency; editors should insist that changes to study design and outcomes are reported as stipulated by CONSORT,” the researchers concluded. “[C]hanges to primary outcomes remain prevalent and may also extent to ICI randomized controlled trials in other cancer types.”
Another recent study showed that primary endpoints are being changed in other phase 3 cancer trials. The study, which included 755 trials, showed that about 19% had a primary endpoint changed, and about 70% of those changes were not disclosed in the manuscript.
Disclosures: Some study authors declared affiliations with biotech, pharmaceutical, and/or device companies. Please see the original reference for a full list of disclosures.
Reference
Richters A, Yildirim H, Booth CM, et al. Changes to primary end points in randomized clinical trials on immune checkpoint inhibitors in urothelial, renal cell, and lung cancers: A systematic review. JAMA Oncol. Published online June 8, 2023. doi:10.1001/jamaoncol.2023.1374
