FDA proposes guidance on using overall survival endpoints in cancer drug trials

The US Food and Drug Administration (FDA) on Monday issued draft guidance to assist sponsors in assessing overall survival in randomized clinical trials for oncology drugs and biologics. The agency cautions that while cancer treatments can improve overall survival, sponsors should look out for potential toxicity from the treatments that may have the opposite effect.

On 18 August, the FDA published draft guidance on assessing overall survival in cancer trials, which is considered both an efficacy and safety endpoint. While cancer treatments can positively affect overall survival, the agency notes that treatment toxicity can also negatively affect overall survival.

"Overall survival should be prioritized as the primary endpoint when feasible," said FDA. "It is most appropriate for consideration in randomized trials of oncologic diseases with a short natural history (e.g., metastatic pancreatic cancer), late-line disease settings, or in disease settings where there are other therapeutics known to prolong overall survival and it is important to demonstrate retention or improvement of this survival advantage."

FDA, however, also noted that it may not always be possible to use overall survival as the primary endpoint. The agency noted that in some cases where the drug can be highly efficacious with very long survival times, it may not be practical to use it. In other cases, using overall survival as the primary endpoint in single-arm trials may not be appropriate because time-to-event endpoints are generally hard to interpret without a randomized control arm.

FDA said that while the guidance discusses situations where overall survival can be used as the primary endpoint, its focus is to make recommendations for situations when overall survival is not the primary endpoint.

"When overall survival is not the primary endpoint and other endpoints such as response rate, progression-free survival, or event-free survival are the primary endpoint in an oncology randomized trial, FDA recommends collection and submission of overall survival data," said the guidance. "In some situations, overall survival may be a secondary endpoint included in a pre-specified hypothesis testing plan to permit a formal statistical test for an efficacy evaluation, which includes control of the overall study-wise Type I error rate.

"When it is not a secondary endpoint with an appropriate formal testing plan, FDA has historically evaluated the overall survival results in a descriptive manner, assessing overall survival as a safety measurement," the guidance added.

The guidance includes recommendations for designing clinical trials that assess overall survival. The FDA said that all randomized oncology clinical trials should be designed to assess overall survival to assess potential harm from cancer treatment. The agency also said the proposed statistical analysis plan assessing overall survival should factor in the futility or harm of the treatment when appropriate. It also proposes recommendations for crossover participants in the trials, and when researchers may need to use an unequal randomization scheme.

"Sponsors should include in their clinical trial protocols adequate plans for collecting data to minimize the amount of missing data for the assessment of overall survival and other long-term outcomes, including details on how data will be collected with regards to capturing long-term safety and efficacy information," said FDA in recommending clinical trial designs. "If non-proportional hazards are anticipated, then additional considerations for adequate overall survival assessment should be pre-specified, i.e., the study should be designed for long-term data collection, and sample size and power calculations should be based on justifiable assumptions."

Additionally, FDA provides significant guidance on statistical analysis considerations when designing trials and what to do with trial data. The agency said sponsors should include statistical analysis in their clinical trial protocol or other accompanying documents, such as the statistical analysis plan, before examining the study results or unblinded data. It noted that if the protocol and SAP are inadequate, sponsors can evaluate overall survival after obtaining the data.

"FDA also considers additional analyses conducted on overall survival outside the scope of those pre-specified in the SAP to be post hoc and for exploratory purposes (e.g., for hypothesis generation)," the agency added.

Stakeholders can comment on the draft guidance on www.regulations.gov under docket no. FDA-2024-D-5850 through 10 October.

Draft guidance

FDA proposes guidance on using overall survival endpoints in cancer drug trials

Related topics

Recommended content

Welcome to the new RAPS Digital Experience