Surrogate Endpoints Remain a Challenge in Interventional Oncology Trials

Interventional oncology (IO) investigators are increasingly relying on surrogate biomarkers to accelerate cancer trial evaluation; however, questions remain about whether these measures reliably predict long-term clinical benefits, conclude the authors of “Survival and Surrogate Biomarkers in Interventional Oncology Trials: Pitfalls, Challenges, and Future Directions,” a review article published in the May 2026 issue of CardioVascular and Interventional Radiology.¹

Authored by Ana P. Gonzalez, Adam Swersky, and Riad Salem, the article examined the place of surrogate endpoints in IO trials, with a focus on liver-directed cancer therapies.

According to the authors, overall survival (OS) remains the “gold standard” endpoint in oncology research.¹ However, obtaining OS data can require years of follow-up, prompting broader use of surrogate measures that provide earlier indications of treatment response. Commonly used surrogate endpoints include progression-free survival, time to progression, objective response rate, duration of response, and quality-of-life metrics.

The review noted that surrogate endpoints have shown variable predictive performance, reproducibility, and reliability across studies, specifically in those involving hepatocellular carcinoma and metastatic liver disease. These inconsistencies extend to endpoint definitions and reporting, which in turn make broad adoption difficult.

The investigators further described statistical and methodological challenges in validating surrogate markers. Earlier frameworks, such as Prentice’s 1989 paper² and the 2001 guidance from the NIH Biomarkers Working Group, established criteria for surrogate validation, but these are rarely consistently met in modern oncology trials.

The review highlights an ongoing tension in oncology research between accelerating therapeutic development and ensuring that surrogate markers accurately reflect patient outcomes. While surrogate biomarkers play an important role in IO studies, there is a pressing need for improved validation of these markers and greater standardization in how they are applied and reported. This is exacerbated by IO’s characteristic lack of large randomized controlled trials, which provide the most reliable source of validation.

The paper also emphasized the growing role of advanced analytics. Emerging technologies such as radiomics, artificial intelligence, and synthetic control arms could strengthen future endpoint validation and improve the design of oncology trials.

“Advancing IO will require the integration of modern trial methodologies, synthetic control arms, radiomics, and artificial intelligence to strengthen surrogate endpoint validation and facilitate broader clinical and regulatory acceptance,” the authors wrote. They add that the field is at “a pivotal juncture” as investigators seek to balance innovation with rigorous evaluation of clinical outcomes.¹

As IO continues to stake its place in cancer care, reliance on surrogate endpoints becomes increasingly important. However, stronger validation and standardized reporting—made possible in part by technological advancements—will be imperative for broader clinical adoption.

1. Gonzalez AP, Swersky A, Salem R. Survival and surrogate biomarkers in interventional oncology trials: pitfalls, challenges, and future directions. Cardiovasc Intervent Radiol. 2026;49(5):862-868. doi:10.1007/s00270-025-04281-7

2. Prentice RL. Surrogate endpoints in clinical trials: definition and operational criteria. Stat Med. 1989;8(4):431-440. doi:10.1002/sim.4780080407