Summary
A cross-sectional UK Biobank study found that white matter lesion volume had a different association with brain parenchymal fraction in people carrying certain NOTCH3 variants. The findings come from a medRxiv preprint and need to be interpreted in light of the small number of carriers studied.
A cross-sectional UK Biobank study reported a different relationship between white matter lesions and brain tissue volume in people carrying certain NOTCH3 variants. In the 44,943 participants with complete brain MRI data, white matter hyperintensity volume was associated with a lower brain parenchymal fraction among non-carriers, but with a higher fraction among the 96 variant carriers. The results are reported in a medRxiv preprint.
White matter hyperintensities (WMHs) are bright areas on MRI commonly used as markers of cerebral small vessel disease. They usually occur alongside lower brain volume in the general population. The authors note that in CADASIL, a small-vessel disease caused by cysteine-altering NOTCH3 mutations, lesions may also reflect tissue water accumulation and dilation of spaces around blood vessels. The study examined whether a different lesion–volume pattern could also be seen among carriers identified in a general-population cohort.
What the MRI analysis found
The researchers analysed MRI data acquired between 2014 and 2023. Participants had a mean age of 64 years, with a standard deviation of 8 years. The 96 carriers had heterozygous cysteine-altering variants in NOTCH3 epidermal growth factor-like repeat domains 7 to 34; 44,847 participants were non-carriers.
WMH volume was automatically measured and normalized to total intracranial volume. Brain parenchymal fraction—the combined volume of grey and white matter divided by total intracranial volume—was used as an imaging measure of brain tissue volume. The models adjusted for age, sex, lacune count and vascular risk factors; lacunes, small areas of tissue damage, were manually assessed in carriers.
Carriers had a higher mean normalized WMH volume than non-carriers: 0.83% compared with 0.36% (P<10⁻⁸). Their mean brain parenchymal fraction was also slightly higher, at 81.1% versus 80.5% (P=0.03).
The association between the two measures differed by genotype. Among non-carriers, each 1% increase in normalized WMH volume was associated with a 0.116% lower brain parenchymal fraction (β = −0.116%; P=3.10⁻⁶). Among carriers, the association was positive (β = +0.51%; P=0.019), and was stronger after accounting for lacune count (β = +0.61%; P=0.006). A genotype-specific interaction was also reported (P=0.036). The difference persisted across age groups and after adjustment for vascular risk factors.
Why the pattern matters—and what it can show
The results suggest that WMH volume may not have the same relationship to brain tissue volume in every genetic context. The authors interpret the carrier pattern as consistent with lesions reflecting tissue water accumulation rather than only irreversible tissue loss, particularly at an early, predominantly preclinical stage. That interpretation matters because MRI lesions are often used as indicators of small-vessel disease burden.
This was a cross-sectional imaging analysis, not a study tracking changes in lesions or brain volume over time. It therefore identifies an association, rather than showing how the measures change as disease progresses. The carrier group was also small compared with the non-carrier group, and the report is a preprint. Whether this pattern predicts later progression or clinical outcomes remains to be tested.