GLP-1 drugs linked to fewer obesity-related cancers than insulin or sulfonylureas — but not versus metformin or newer drugs

A research letter in JAMA Network Open published on 3 November 2026 looked at whether starting a GLP-1 receptor agonist is associated with fewer obesity-related cancers in people with type 2 diabetes — and found the answer depends on what the drugs are compared with.

The study

• Type: observational cohort study using electronic health records from 359,595 adults seen at UT Southwestern Medical Center in Dallas between 2010 and 2025, analysed as five “target trial emulations”.
• Each analysis compared new GLP-1 users with new users of metformin, insulin, sulfonylureas, SGLT2 inhibitors or DPP-4 inhibitors; patients had no prior obesity-related cancer. GLP-1 groups ranged from 5,835 to 18,007 people.
• Diabetes severity was classed as low, moderate or high based on complications.

Results

• At 3 years, GLP-1 initiation was associated with lower risk versus insulin (hazard ratio 0.62; risk difference −1.25%) and sulfonylureas (HR 0.67; −1.14%); 5-year results were similar.
• No significant difference versus metformin (HR 0.91), SGLT2 inhibitors (HR 0.81) or DPP-4 inhibitors (HR 0.84).
• The lower risk was seen mainly in low-severity diabetes; in high-severity diabetes, risk differences favoured SGLT2 inhibitors over GLP-1 drugs.

The authors called the findings “broadly reassuring regarding oncologic safety” but said they “did not support selecting glucose-lowering therapy for cancer prevention”.

Limitations

This is a retrospective, single-health-system study, so residual confounding is possible — insulin users, for example, tended to have more advanced disease. Groups did not overlap perfectly after weighting, exposures or outcomes may be misclassified, severity categories were broad, some subgroups had few events, and 5-year estimates relied mostly on early GLP-1 users. It shows associations, not cause and effect. Discuss treatment choices with your doctor.