A landmark analysis of phase 3 oncology trials reveals a troubling disconnect between what clinical research measures and what patients value most. Only 28% of trials published over the past two decades demonstrated overall survival improvement, while just 11% showed quality of life benefits, according to findings presented at the ASCO Annual Meeting.
The meta-epidemiological analysis, conducted by researchers at The University of Texas MD Anderson Cancer Center, examined 791 two-arm superiority phase 3 oncology trials registered on ClinicalTrials.gov between 2002 and 2024, encompassing 555,580 enrolled patients.
Alternative Endpoints Dominate Trial Design
The study found that 63% of trials used alternative primary endpoints rather than overall survival or quality of life measures. Progression-free survival has emerged as the dominant primary endpoint in modern phase 3 oncology trials, driven by practical advantages including higher event rates, smaller required patient populations, faster data readouts, and reduced trial costs.
"The primary endpoints that we select for phase 3 trials will be used to determine whether we move forward with a new drug or therapeutic strategy in clinical practice, so they have enormous importance," said Alexander D. Sherry, MD, a radiation oncology resident at MD Anderson and lead researcher. "What I really want to know from trial data is whether an intervention will help my patient live longer or live better."
Concerning Gap in Patient-Centered Outcomes
While approximately half (53%) of trials met their primary endpoints, with 55% showing alternative endpoint superiority, the results for patient-centered outcomes were far less encouraging. Only 48 trials—representing just 6% of those evaluated—demonstrated improvements in both survival and quality of life.
The analysis revealed significant methodological issues in quality of life assessments. The majority (82%) of quality-of-life analyses did not adjust for baseline scores, representing substantial potential for loss of efficiency and statistical power.
Even when researchers expanded their analysis to examine whether trials improved any aspect of quality of life—such as physical functioning or mental well-being—only 10% of trials showed superiority in the experimental arm across any quality-of-life subscale domain.
Clinical Implications and Regulatory Concerns
"There is increasing concern that alternative endpoints like PFS are not adequate surrogates for OS or quality of life—the two things different studies and surveys show patients, their families and patient advocates care most about," Sherry explained. "Changes in the radiologic dimension of a tumor on a scan may not correlate to how long a patient feels or how long they'll live."
The findings highlight fundamental questions about the regulatory approval process for cancer drugs. Many treatments receive approval based on statistically significant improvements in surrogate endpoints without demonstrating tangible benefits in life expectancy or patient well-being.
Balancing Speed and Meaningful Outcomes
Sherry acknowledged the complex challenges facing clinical trial design. "We certainly need to have trials read out faster, we want to bring new drugs to patients, and there is competition among pharmaceutical companies to get their drugs on the market quickly. Those are all valid points, and this has driven the rise in use of alternative endpoints."
However, he emphasized that the conversation about designing trials to answer clinically relevant questions has been ongoing for years. "Everyone at the table is trying to help our patients. We're just trying to find the best way to do that."
Study Limitations and Future Directions
The researchers noted several study limitations, including the possibility that trials missing their primary endpoints may be less likely to publish secondary endpoint analyses. Additionally, the findings may not be generalizable beyond trials registered on ClinicalTrials.gov.
"We are still improving OS in at least a quarter to a third of studies, which is notable, but quality of life is vastly underreported and underpublished, and those gains are very rare," Sherry said.
The results suggest that the heightened focus on alternative endpoints has overshadowed the importance of evaluating overall survival and quality of life in a robust manner, potentially undermining the clinical relevance of late-phase oncology research for both patients and clinicians.
