Hi, my name is Drew Adamek. I'm a medical student at the University of Chicago and we're presenting our work on retinopathy of prematurity and the effect of metformin. Retinopathy of prematurity is the leading cause of childhood blindness in the US and our current treatments work, but they're not perfect. So we started asking, is there something we could do earlier, even before birth, to reduce this risk?

We started considering metformin. It's one of the most widely used diabetes drugs in the world and it has well documented anti-inflammatory and anti-angiogenic properties. We know it's shown a protective effect in age-related macular degeneration, which shares key pathogenic pathways with ROP. In addition, a computational drug gene analysis has already flagged it as a candidate for ROP prevention and has been shown to be safe in early pregnancy.

However, what hadn't been studied yet was the actual clinical outcomes of metformin exposure on ROP and that's what we wanted to look at. To do this, we conducted a retrospective cohort study at the University of Chicago. We looked at premature infants screened for ROP and compared those whose mothers had taken any metformin to those just on insulin without metformin during pregnancy. We identified 41 applicable pregnancies with ROP data and manually verified exposure of flagged mothers. We identified 13 metformin-exposed and 28 insulin-only babies. We looked at any ROP, severity staging of ROP, and severe ROP defined as stage three or higher.

In the insulin-only group, 46% of infants developed ROP. While in the metformin-exposed group, it was just 15%. That's a statistically significant 79% reduction in odds. Furthermore, metformin exposure was significantly associated with a lower overall severity stage. Notably, none of the metformin-exposed infants developed stage two or higher. When we looked at severe ROP, there were none in the metformin-exposed group compared to nearly 11% in the insulin-only group, though it did not reach significance. When we adjusted for gestational age and birth weight, the effect did not quite reach statistical significance, but we'd expect that modeling as a small sample. But the magnitude of the odds ratio stayed large at an odds ratio of 0.24.

Metformin is the most common type two diabetes drug in the world, making it accessible and inexpensive. It crosses the placenta and its effects on angiogenesis are well established. While our current study has limitations, namely the small sample size, the retrospective design, and gestational age as a confounder, we now have clinical outcomes that back up a biologically plausible mechanism. We hope these findings prompt larger investigations into repurposing this highly accessible drug, potentially preventing ROP for countless premature infants in the future.