A major new genetic study overturns key assumptions about how often gene variants thought to cause inherited retinal degenerations (IRDs) actually lead to disease in the general population.¹ The study was published in the American Journal of Human Genetics.

Inherited retinal degenerations have traditionally been considered “Mendelian,” meaning that carrying a disease-causing variant in a single critical gene was assumed to inevitably lead to disease. But researchers now report that many of these variants have much lower penetrance—the likelihood that a genetic change produces symptoms—than previously believed.

Using genomic and health record data from two massive volunteer biobanks (the U.S. All of Us Research Program and the U.K. Biobank) scientists looked at hundreds of thousands of participants to see how often so-called pathogenic variants in known IRD genes actually result in clinical disease. They curated a list of 167 variants in 33 genes previously reported to cause IRDs and scanned more than 317,000 participants to see who carried them.

The results showed only a small fraction of individuals who carried one of these IRD-associated genotypes had evidence of retinal disease. When using strict diagnostic criteria based on electronic health records, just about 9% of carriers had a documented IRD. When using broader diagnostic codes that might capture additional forms of vision loss, penetrance rose to about 28%—still well below the assumption that these variants cause disease nearly 100% of the time.

Researchers also validated their findings using retinal imaging data from the U.K. Biobank, which showed a similar penetrance range of roughly 16% to 28% among carriers with detectable retinal abnormalities. Across both datasets, individual characteristics like age, smoking status, or other health conditions did not reliably predict whether a carrier would develop disease, suggesting that other genetic or environmental factors, yet to be identified, modify risk.

“This study provides empirical evidence that the classical Mendelian model does not always hold true at the population level,” said senior author Dr. Eric Pierce of Mass Eye and Ear and Harvard Medical School. According to the team, the discrepancy likely stems in part from ascertainment bias. Most prior estimates of penetrance came from clinical studies of affected families, which tend to overestimate risk because they focus on people already showing symptoms.

The study authros stated that as large-scale genomic screening becomes more common, clinicians may need to rethink how results are interpreted for conditions long viewed as monogenic and fully penetrant. Patients found to carry an IRD-associated variant might face very different risks than previously communicated, and genetic counseling could shift toward incorporating polygenic and environmental risk modifiers.

Reference

1. Zaslavsky K, Chen L, Park C, et al. Low population penetrance of variants associated with inherited retinal degenerations. Am J Hum Genet. 2025. Published online December 22, 2025. doi:10.1016/j.ajhg.2025.11.015