ASCO observational analysis links GLP‑1 use with less stage IV progression in four cancers

What happened

At ASCO's 2026 annual meeting, researchers reported less recorded progression to stage IV disease among GLP-1 medicine users in four of seven cancer cohorts. The significant findings involved non-small cell lung, breast, colorectal, and hepatocellular cancers. Results for prostate, pancreatic, and kidney cancers did not meet the study's statistical threshold.

The investigators used the TriNetX health-record network to study people with stage I, II, or III cancer who began a GLP-1 receptor agonist after diagnosis. They compared them with propensity-matched patients who began a DPP-4 inhibitor, another class of diabetes medicine. The assembled matched cohorts covered more than 12,000 patient records across the seven cancer types.

Over five years, stage IV progression was recorded in 10% of GLP-1 users and 22% of DPP-4 inhibitor users in the lung-cancer comparison. The corresponding proportions were 10% and 20% for breast cancer, 13% and 22% for colorectal cancer, and 19% and 28% for liver cancer. The reported relative associations ranged from a 31% reduction for colorectal cancer to a 50% reduction for lung cancer.

A separate analysis used The Cancer Genome Atlas to examine tumor GLP-1 receptor expression and survival. That exploratory work concerns receptor expression in tumor samples, not whether prescribing a GLP-1 medicine improves survival. It should not be combined with the medication comparison as if both answered the same question.

What it means

The result is a signal worth testing prospectively. By using an active diabetes-drug comparator and propensity matching, the investigators tried to make the exposed groups more similar than a simple user-versus-nonuser comparison would.

The absolute progression percentages also give more context than relative reductions alone. Across the four statistically significant cohorts, the recorded differences ranged from 9 to 12 percentage points over the study window.

What it does not mean

The analysis does not show that GLP-1 medicines treat cancer or prevent metastasis. Medication assignment was not randomized. Differences in diabetes control, body weight, cancer stage within the broad I-to-III range, oncology treatment, surveillance, adherence, access, and other health factors may remain after matching.

DPP-4 inhibitors are an active comparator used for diabetes, not a placebo. The study therefore addresses one observational comparison among patients receiving two different medication classes. It does not show how GLP-1 users compare with every person who has one of these cancers.

The findings came from a conference abstract and real-world records, not a completed randomized oncology trial. They should not prompt anyone to start a GLP-1 medicine, replace cancer therapy, or change diabetes treatment without the clinicians managing both conditions.