To access the results of clinical trials, researchers use “endpoints,” specific events or outcomes that help evaluate the benefit of the intervention under investigation. Traditionally, overall survival (OS), the time from diagnosis (or treatment initiation) until death, and quality of life (QoL), a measure of how much an individual enjoys life, based on both a sense of well-being and the ability to carry out daily activities, serve as endpoints for phase 3 clinical trials.
Recently, alternative endpoints, including progression-free survival (PFS), have emerged as popular readouts for many phase 3 clinical trials. PFS indicates the time from diagnosis (or treatment initiation) that an individual patient still has the disease, but it does not get worse or “progress” to a more severe disease state. However, the suitability and relevance of PFS endpoints remain unclear.
To assess the validity of PFS endpoints for phase 3 clinical trials, a team of researchers set out to determine how many trials have OS or QoL superiority. A recent publication in the Journal of the American Medical Association (JAMA) Oncology shows the results of their analysis.
The researchers identified phase 3 oncology interventional clinical trials comparing two groups. All the studies included in this assessment employed a “superiority design,” which is a type of clinical trial design that aims to demonstrate that the intervention under investigation is more effective than the control. Results from all the trials used in the study were published between 2002 and 2024.
The researchers identified 791 phase 3 clinical trials encompassing 555,580 patients for inclusion in the current study. Of these published trials, 63% used alternative endpoints. Just over half (420) of the trials met the primary endpoint, a predetermined measure used to determine whether the intervention under investigation was effective.
Fifty-five percent of the studies demonstrated alternative endpoint superiority, while 28% showed OS superiority. Only 34% of the studies included global QoL analyses, and the majority of these did not adjust for baseline indications. Thus, only 11% of studies showed global QoL superiority.
Taken together, 32% of the trials demonstrated OS or QoL superiority, and 6% showed both. Importantly, among all trials with a positive alternative endpoint, fewer than half demonstrated OS or global QoL superiority (43% and 15%, respectively).
The authors conclude that improvements in alternative endpoints, such as PFS, may not be sufficient to interpret the results of oncology-related phase 3 clinical trials. Positive alternative endpoint findings did not correlate highly with OS or QoL endpoints. The researchers suggest that late-phase clinical trial design should focus on OS and QoL improvements to obtain the most valid readout of interventional benefit.
Sources: JAMA Oncol