Dr. Axe Wire · Study decoded · health

Metformin Shows No Gene Changes in Small Eye Disease Trial

A small trial of older adults with an eye disease found no measurable gene changes from metformin, though inflammation pathways showed early hints.

By Dr. Axe Editorial Team · October 29, 2026 · 1 source · Watch

A randomized trial in nondiabetic older adults with age-related macular degeneration, published October 20, 2026 in GeroScience, found that a course of metformin produced no statistically significant change in any individual gene in blood.

15total patients

1000mg twice daily

18months

0genes changed

Researchers studied 15 nondiabetic adults age 55 and older with geographic atrophy, a form of age-related macular degeneration, as part of a larger randomized trial (NCT02684578). Eight participants took 1000 milligrams of metformin twice daily for 18 months, while seven served as controls. The team drew whole blood samples and analyzed gene activity across the genome. After accounting for the number of genes tested, not one single gene showed a statistically meaningful difference between the metformin group and the control group.

In a secondary, exploratory analysis, three gene pathways ran lower in the metformin group: interferon alpha/beta signaling, which drives antiviral and inflammatory immune responses; cell junction organization, which affects how cells stick together and form barriers; and epithelial mesenchymal transition, a process where stable cells become more mobile. The authors describe these shifts as consistent with metformin possibly calming proinflammatory signals and activating anti-inflammatory ones.

Metformin is already prescribed for blood sugar control, and researchers have wondered whether it also quiets the low-grade inflammation that builds with age, sometimes called inflammaging. This trial's pathway results point that way, but the authors call them hypothesis-generating, and the main gene-by-gene test found nothing.

Metformin Dose in AMD Gene Study

GeroScience, 2026, Chen et al.

With only 15 people total and just 7 or 8 in each group, the trial was not built to detect gene-level changes with confidence, and the pathway signals need to be checked in larger groups before anyone can trust them. Whether any of this translates into a measurable benefit for eye health was not tested here.

Dr. Axe’s Take

Metformin lowers blood sugar, and in people without diabetes that does nothing for whatever drives geographic atrophy. The known costs are real: GI upset and, over time, lower vitamin B12, which matters for nerve and brain health in older adults. I wouldn't expect this thin pathway data to turn into an eye benefit, and it doesn't justify long-term use. If you're on metformin, don't stop it on your own, and have your B12 checked.

— Dr. Axe

Take the Cellular Health Quiz →

Study summary

Who: 15 nondiabetic adults, 55 and older, with geographic atrophy from age-related macular degeneration; 8 on metformin, 7 controls.
Design: Randomized trial (NCT02684578), sub-study with whole blood metagenomic sequencing and differential gene expression analysis.
Dose: 1000 mg metformin twice daily for 18 months, versus control.
Primary result: No individual gene showed a statistically significant difference between groups after correction for multiple comparisons.
Secondary: Interferon alpha/beta signaling, cell junction organization and epithelial mesenchymal transition pathways were lower in the metformin group (exploratory pathway analysis).
Funding / conflicts: Funders listed in the journal record: Research to Prevent Blindness; All May See Foundation; National Eye Institute. Competing interests: several authors consult for drug and device companies, including Boehringer Ingelheim and Roche/Genentech; one co-founded and holds equity in aging-therapy companies.
Date: Oct 20, 2026