A phase 2 liver trial where the add-on drug didn't help — and the base drug quietly did
Read the Sunday Brief →

Study wrapper · #430

Association of Tirzepatide versus Glucagon-Like Peptide-1 Receptor Agonists with Incident Glaucoma in Patients with Type 2 Diabetes: A Retrospective Cohort Study.

Pan SY, Weng CH, Sheen YJ, et al. Ophthalmology science. 2026.
MixedCohortMentions: Tirzepatide

Editor's note

In this retrospective cohort of 51,540 adults aged 60+ with type 2 diabetes drawn from a multinational database, tirzepatide was associated with a lower two-year risk of primary open-angle glaucoma than GLP-1 receptor agonists (hazard ratio 0.65, 95% CI 0.44–0.96). Groups were 1:1 propensity-matched, and the signal held across several sensitivity analyses. This is observational data: propensity matching cannot rule out residual confounding (for example, differences in who gets prescribed tirzepatide), and the wide confidence interval nudges close to 1.0. The absolute event numbers are modest, and the comparison is tirzepatide versus another active drug class, not versus placebo. The authors themselves call for further study to confirm clinical significance. Read this as a hypothesis-generating, safety-adjacent signal, not evidence that tirzepatide changes glaucoma risk.

Plain-language abstract

This study asked whether people with type 2 diabetes who take tirzepatide develop glaucoma at a different rate than those taking older GLP-1 medications. Using a large international database, researchers studied adults aged 60 and over who newly started one of these drugs between 2022 and 2024. People with prior glaucoma were excluded. Using statistical matching to make the two groups comparable on age, sex, blood sugar, weight and other factors, they followed just over 51,000 people for two years. Those on tirzepatide had about a 35% lower rate of open-angle glaucoma than those on GLP-1 drugs, and the pattern stayed consistent when the analysis was repeated in different ways. Because this is a look-back study rather than a randomized trial, it can show an association but not that tirzepatide caused the difference. The authors say more research is needed before drawing clinical conclusions.