Association between semaglutide use and thyroid eye disease-related outcomes in autoimmune thyroiditis with type II diabetes

Semaglutide use was associated with fewer thyroid eye disease-related diagnoses and treatments over five years in adults with autoimmune thyroiditis and type 2 diabetes, compared with other diabetes medications. The study used data from a large U.S. health records network but, because it was observational, cannot prove that semaglutide is protective.

Journal article — Retrospective propensity-matched cohort study. Population: Adults with thyrotoxicosis and type II diabetes mellitus in the TriNetX US Collaborative Network. Sample size: 23,279 patients per group after matching. Follow-up: 5 years. Interventions: Semaglutide.

After propensity matching, 23,279 patients were included per group (mean age 60.1 years; 75% female). At 5 years, semaglutide use was associated with lower rates of TED-related diagnoses (2.16% vs 2.82%; RR 0.76; 95% CI 0.68-0.86), systemic corticosteroid use (22.71% vs 26.39%; RR 0.86; 95% CI 0.82-0.90), TED-related surgical interventions (0.30% vs 0.42%; RR 0.70; 95% CI 0.51-0.96), and radiation treatment (1.17% vs 2.05%; RR 0.57; 95% CI 0.49-0.67). Teprotumumab use did not differ between groups (0.08% vs 0.08%; RR 1.00; 95% CI 0.52-1.92).

This paper adds large, propensity-matched real-world data on a potential protective association between semaglutide and thyroid eye disease-related outcomes in people with autoimmune thyroiditis and type 2 diabetes. It does not establish causation, dose effects, or mechanisms, so semaglutide's potential role in thyroid eye disease remains to be confirmed prospectively.

Key findings

• Semaglutide users had lower 5-year rates of TED-related diagnoses: 2.16% vs 2.82% (RR 0.76, 95% CI 0.68-0.86).
• Systemic corticosteroid use was lower in semaglutide users: 22.71% vs 26.39% (RR 0.86, 95% CI 0.82-0.90).
• TED-related surgical interventions were lower in semaglutide users: 0.30% vs 0.42% (RR 0.70, 95% CI 0.51-0.96).
• Radiation treatment was lower in semaglutide users: 1.17% vs 2.05% (RR 0.57, 95% CI 0.49-0.67).
• Teprotumumab use was rare and similar between groups: 0.08% vs 0.08% (RR 1.00, 95% CI 0.52-1.92).

Limitations

• Observational design; residual confounding cannot be excluded despite propensity matching.
• Reliance on ICD-10 coding and prescription records may lead to misclassification of exposure or outcomes.
• Teprotumumab use was rare, limiting interpretation of that outcome.
• Authors note that further studies are warranted to confirm these associations and clarify underlying mechanisms.

The record

• Journal: Canadian journal of ophthalmology. Journal canadien d'ophtalmologie 61(3): 665-669
• Published: 2026-06-01
• PMID: 41620207 — DOI: 10.1016/j.jcjo.2026.01.010
• MeSH terms: Humans, Diabetes Mellitus, Type 2, Thyroiditis, Autoimmune, Hypoglycemic Agents, Retrospective Studies, Adult, Aged, Middle Aged, Female, Male, Graves Ophthalmopathy, Glucagon-Like Peptides, Semaglutide
• Author countries: Israel

Peptide profiles: Semaglutide.

All indexed evidence: Semaglutide trials & papers.

Related studies: Pharmacological Interventions for Weight Reduction in Patients With Schizophrenia…, Effects of subcutaneous semaglutide on weight loss and inflammatory markers in adults…, Burden and Risk of Depression in Patients Receiving Semaglutide: A Systematic Review and…, Effect of tirzepatide and semaglutide on blood pressure: A systematic review and….