GLP-1 drugs alongside progestins tracked with lower endometrial cancer risk

Endometrial cancer incidence is rising, particularly among people with obesity and metabolic disorders, and the authors of a new cohort study open by arguing that strategies targeting both the hormonal and the metabolic side of that risk are needed [s1]. For women with endometrial hyperplasia who want to keep their uterus, the standard non-surgical answer is progestin therapy [s1].

A cohort study published in JAMA Network Open on February 10 asks whether adding a GLP-1 receptor agonist to that progestin changes what happens next [s1].

What was done

The authors used TriNetX, a federated network of de-identified electronic health records, to identify adult women with endometrial hyperplasia or benign uterine pathology who received progestins between May 1, 2005 — the date of first GLP-1 receptor agonist approval — and December 31, 2022 [s1]. Cases were identified by ICD-10 codes, and the data were extracted on February 23, 2025 [s1].

Four comparisons were specified [s1]:

• GLP-1 receptor agonist plus progestins versus progestins alone
• GLP-1 receptor agonist plus progestins versus metformin plus progestins
• Triple therapy (GLP-1 receptor agonist, metformin, progestins) versus metformin plus progestins
• Triple therapy versus progestins alone

The primary outcome was incident endometrial cancer; the secondary outcome was subsequent hysterectomy [s1].

The findings

A total of 18,414 women received a GLP-1 receptor agonist combined with progestin, against 426,406 who received progestin alone [s1].

Compared with progestins alone, the combination was associated with a significantly lower risk of endometrial cancer: hazard ratio 0.34 (95% CI, 0.27–0.44) [s1]. The authors report the association held across subgroups stratified by progestin route, baseline risk level, body mass index, and age [s1].

Against an active comparator it also held. GLP-1 receptor agonist plus progestins carried lower endometrial cancer risk than metformin plus progestins (HR, 0.30; 95% CI, 0.15–0.59) [s1]. Triple therapy outperformed both metformin plus progestins (HR, 0.37; 95% CI, 0.25–0.53) and progestin monotherapy (HR, 0.44; 95% CI, 0.29–0.66) [s1].

Hysterectomy rates were also lower in the GLP-1 group, at 2 years (HR, 0.47; 95% CI, 0.42–0.53) and at 5 years (HR, 0.59; 95% CI, 0.54–0.64) [s1].

The number that governs how this should be read

Mean age in the GLP-1-plus-progestin group was 43.1 years (SD 10.2). In the progestin-only group it was 35.2 years (SD 10.9) [s1].

That is an eight-year gap between the exposed and unexposed groups, and it runs in the direction that would ordinarily make the GLP-1 group look worse, not better — the exposed group was older. Finding a two-thirds reduction in the older group is, on its face, a point in favour of the association being something other than an artefact of age. The authors also report that the association held in age-stratified subgroups [s1].

But an eight-year difference in mean age between arms is also a marker of how different these two populations are in ways the records may not capture at all. Women prescribed a GLP-1 receptor agonist in this period were, by definition, women whose clinicians identified obesity or diabetes and acted on it — a group defined partly by engagement with care. Women who fill prescriptions for two drugs and return for follow-up differ systematically from women who fill one. In a records-based design, that difference does not disappear because it is inconvenient.

Other limits

This is a retrospective cohort study on administrative and EHR data. Endometrial cancer and hyperplasia were ascertained by diagnostic codes, not by pathology review. Dose, duration, and adherence for any of the drugs are not reported in the abstract. The comparison window opens in 2005, when GLP-1 receptor agonist use was rare and quite different in character from use in 2022.

Detection is the other hazard. Hysterectomy rates were lower in the GLP-1 group, which cuts against the simplest surveillance-bias story — more contact with gynaecology should mean more procedures, not fewer — but it also means the two groups had different amounts of tissue available for a cancer diagnosis over follow-up.

The authors' own conclusion is appropriately narrow: the combination was associated with reduced endometrial cancer risk, and further investigation is warranted to assess applicability and underlying mechanisms [s1].

What would settle it

A randomised trial in women with endometrial hyperplasia undergoing progestin therapy, with GLP-1 receptor agonist added or not, and pathologist-adjudicated outcomes. Until something like that exists, this is a hypothesis with a large sample behind it, not a treatment strategy.

Nothing here is a recommendation to start, stop, or change any medication. GLP-1 receptor agonists are not approved for cancer prevention.

Sources

• [s1] Yen TT, Hsieh TYJ, Lee GY, Toy EP, Wei JC, Tanner EJ. GLP-1 Receptor Agonists Plus Progestins and Endometrial Cancer Risk in Nonmalignant Uterine Diseases. JAMA Network Open, published online 2026-02-10.