FDA cancer drug approvals for solid tumors are increasingly relying on response rates rather than overall survival as the primary efficacy endpoint, a 20-year trends analysis found.

The proportion of FDA approvals based on overall survival fell by more than half between 2006-2010 and 2021-2025, while reliance on objective response rates almost quadrupled in that time. By 2021-2025, objective response rate had surpassed progression-free survival (PFS) as the most common primary endpoint supporting FDA approvals.

“What’s concerning is that the most common surrogate endpoint is not even PFS,” Bishal Gyawali, MD, PhD, of Queen’s University in Kingston, Ontario, Canada, told Medscape Medical News. “We have been criticizing PFS for its flaws and why it’s not a good surrogate for overall survival. But it seems like nowadays we don’t even get PFS. We get even a much weaker surrogate.”

Because cancer medicines represent the most common therapeutic class of new drugs and biologics approved by the FDA, these drugs — and the evidence for their approvals — are important for physicians, patients, industry, and policymakers. Gyawali and colleagues analyzed how the evidence supporting these treatments has changed over the past two decades, looking at original and supplemental indication approvals for adult solid tumors between 2006 and 2025.

The cross-sectional study included 385 indication approvals — 154 (40%) for novel drugs and 231 (60%) for supplemental indications for previously approved treatments. Overall, 280 approvals (72.7%) used the regular pathway and 105 (27.3%) used the accelerated approval pathway.

Over the 20-year period, the number of approvals increased more than fivefold from 30 in 2006-2010 to 155 in 2021-2025. In that time, just over a quarter (26.5%) were based on overall survival while the rest relied on surrogate endpoints as the primary endpoint.

The proportion of approvals based on overall survival decreased over time, from 40% in 2006-2010 to 18.7% in 2021-2025, while reliance on surrogate endpoints increased from 60% to 81.3%. PFS was the most common surrogate endpoint in 2006-2015 (60%), but objective response rate became the most common in 2016-2025 (46.7%). In the latest 5-year period, 42.6% of FDA approvals were supported by objective response rate as the primary endpoint compared with 32.9% supported by PFS.

In other words, “many of our drugs are approved on the basis of tumor shrinkage alone,” Gyawali said.

Approvals were increasingly supported by single-arm studies — 12.9% in 2006-2010 vs 37.2% in 2021-2025 — whereas approvals based on randomized trials declined from 87.1% to 62.8% over the same period.

Over the same period, approvals of chemotherapy agents fell from 40% to 3.9%, whereas approvals of targeted therapies rose from 23.3% to 32.9% and immunotherapy approvals rose from 20.0% to 37.4%.

Lung cancer accounted for the largest share of drug approvals (23.4%), followed by genitourinary (16.6%), gastrointestinal (15.3%), and breast cancers (11.9%). There were 11 (2.9%) tumor-agnostic approvals over the study period.

Gyawali expressed concern about another finding: the increasing use of regular approvals based on surrogate endpoints. While previous studies have found that most accelerated approvals were based on response rates from single-arm trials, the current analysis revealed an increasing number of surrogate-based approvals were granted through the regular approval pathway. The proportion of regular approvals based on surrogate endpoints rose from 53.8% in 2006-2010 to 74.8% in 2021-2025.

Gyawali said the distinction matters because accelerated approval requires confirmatory trials to verify clinical benefit, whereas regular approval generally carries no comparable requirement.

“At least if you give accelerated approval, you need to confirm the clinical benefit in a confirmatory trial, so you have that safety net of needing to confirm patient outcomes are improved. But we see that we are giving upfront regular approval for many of these drugs,” Gyawali told Medscape Medical News.

Some cancer indications granted accelerated approval have been withdrawn after confirmatory studies failed to verify benefit. Had those drugs received regular approval initially, their potential harm or lack of benefit might never have been discovered in subsequent confirmatory trials, Gyawali noted.

Gyawali said oncologists should talk to patients about whether an approved drug has been shown to improve survival or quality of life, delay progression, or merely shrink tumors.

Patients may “inherently assume that a drug is going to make them live longer,” Gyawali said. So “patients should know that some drugs may have been given regular approval based on surrogate endpoints alone, and the benefits may not be what it is claimed to be.”

That uncertainty should “100%” be part of shared decision-making, particularly when treatment carries substantial toxicity, inconvenience, or cost, he said.

The study had no specific funding. Disclosures for study authors are listed with the original study publication.