White matter hyperintensities (WMH) are brain lesions that show up brightly on MRI scans, indicating areas of potential structural abnormalities in the brain’s white matter. They are highly prevalent in older age, present in up to 20-95% of people aged 60-90 years old¹. Past studies have hinted that these lesions may relate to the thinning of the cerebral cortex, but comprehensive genetic insights were missing. When we set out to study WMH and their potential link to cortical atrophy, a core feature in dementia, we were motivated by a critical question: could genetic factors underlie this association, and might they also explain part of the observed risk for dementia?

Our international team used data from over 50,000 participants across ten cohorts, forming a large-scale investigation into the genetics of WMH-associated cortical thinning. The findings confirmed that increased WMH volume was consistently (across all cohorts) associated with decreased cortical thickness, even when adjusting for typical vascular risk factors like hypertension and diabetes. Particularly notable was the insula, a key brain region for integrating various sensory, emotional, and cognitive functions, where we observed the strongest correlation.

A genome-wide association study (meta-GWAS) was performed to explore genetic loci associated with this WMH-cortical atrophy relationship. We identified 20 significant loci, of which 15 influence genes within the cortex, particularly those involved in axonal transport and cytoskeletal organization – essential for maintaining neuronal health. Moreover, these genetic signals were enriched in vascular-related cell types and cells supporting neuronal health, such as astrocytes and oligodendrocytes. This pattern aligns well with the hypothesis that disruptions in small blood vessels and axonal health may contribute to cortical thinning, connecting vascular risk to cortical atrophy.

One particularly enlightening aspect of our study was constructing a polygenic risk score (PRS) based on our meta-GWAS of WMH-associated cortical thinning. This PRS, when applied to an independent dataset of 500,000 individuals, showed that higher genetic vulnerability to WMH-related cortical atrophy predicts increased risk for both vascular and all-cause dementia. This finding underscores the need to consider vascular health in understanding dementia risk, as well as the genetic susceptibility that some individuals may carry.

While this study is among the first large-scale investigations into the genetics of WMH-associated cortical thinning, it represents only the beginning of what we hope will be a productive line of research.