Changes in the Eye’s Blood Vessels Could Signal Early Risk for Alzheimer’s and Related Dementias

Researchers have found that changes in the eye’s blood vessels—more particularly vessel twisting and abnormal crossings—could serve as early warning signs of Alzheimer’s disease and related dementias (ADRD), according to a study published in Alzheimer’s and Dementia, The Journal of the Alzheimer’s Association.

Early diagnosis of ADRD remains one of the greatest challenges in neurology, the authors said. Current diagnostic tools, including MRI, PET scans and cerebrospinal fluid tests, provide valuable information about brain health, but they are costly, invasive and typically performed only after symptoms appear, according to the authors.

This has fueled interest in identifying noninvasive biomarkers that could signal risk earlier. The retina, part of the central nervous system and easily examined during routine eye exams, has emerged as a potential candidate.

Previous research has suggested that changes in the eye’s blood vessels could reflect vascular dysfunction in the brain, one of the earliest biological changes linked to ADRD. This study builds on evidence that abnormalities such as vessel twisting, or tortuosity, reduced vessel density and abnormal crossings (AVCs) in the retina could mirror similar vascular problems in the brain.

Because retinal imaging is already common in clinical practice, these features may offer a practical way to detect disease risk before memory problems or other symptoms come about. In this case, retinal vascular changes could serve as biomarkers for identifying individuals at risk of Alzheimer’s and related dementias.

In this study, all experiments were approved by The Jackson Laboratory Institutional Animal Care and Use Committee and followed standard guidelines for caring for animals. Mice carrying the Mthfr677C>T gene variant were bred on a C57BL/6J background to create litter-matched groups with CC, CT, and TT genotypes. The mice were also studied at 6 and 12 months of age.

Retinal imaging included optical coherence tomography, fundus photography and fluorescein angiography to measure vessel density, tortuosity and branching. Eyes were dilated, and mice were anesthetized for imaging. Retinas and brains were collected for immunohistochemistry and proteomic analysis, including staining for vascular markers (CD31, Collagen IV) and cell nuclei (DAPI).

Additionally, pattern electroretinography and intraocular pressure were measured to examine retinal function, and RNAscope was used to examine Mthfr expression across cell types. Proteins were extracted from retina and brain, digested and analyzed using liquid chromatography-tandem mass spectrometry to determine differential expression. Image analysis and statistical comparisons were also conducted using FIJI and Prism software.

It was found that mice carrying the Mthfr677C>T genetic variant displayed age- and sex-dependent changes in retinal blood vessels. At 6 months, the mice’s retinas appeared normal, but by 12 months, female TT mice showed fewer blood vessels and less branching, creating a simpler vascular network. Twists in the vessels and abnormal crossings between arteries and veins, both signs of vascular problems, were more common in TT mice, especially in 12-month females. Some veins were also pinched where arteries crossed them, a feature called “nicking.” Near the optic nerve, small arteries were narrower and veins were wider in 6-month TT mice.

Even with these changes in blood vessels, the overall thickness of the retina and the function of retinal ganglion cells were the same at baseline, suggesting the variant does not cause immediate nerve damage. Intraocular pressure was slightly higher in TT males, but this did not lead to glaucoma-related problems.

In addition, the Mthfr gene was active throughout both the retina and the brain, appearing in blood vessel cells, pericytes and smooth muscle cells. This supports the idea that similar processes affect blood vessels in both tissues. Proteomic analysis found 206 proteins that were altered in both the retina and brain. These proteins are linked to mitochondrial function, metabolism and the system that breaks down damaged proteins, according to the authors.

Together, these results suggest that changes in the retina’s blood vessels reflect underlying problems in the brain’s blood vessels caused by Mthfr677C>T, and they may serve as early indicators of Alzheimer’s disease and related dementias.

This study overall presents several strengths, including the use of a well-characterized Mthfr677C>T mouse model that mirrors human vascular and cerebrovascular changes linked with ADRD. By examining both retinal and brain vasculature, researchers found overlapping vascular phenotypes, including reduced vessel density, arteriovenous crossings and vessel tortuosity, highlighting the retina’s potential as a clinically accessible biomarker. Proteomic analysis strengthened the study even more by revealing shared molecular pathways, including mitochondrial dysfunction and protein homeostasis, supporting the relevance of retinal findings to brain health.

However, limitations include the delayed appearance of some retinal vascular changes compared to the brain, suggesting that vessel density alone could be a less reliable early biomarker. Fluorescein angiography also detected trends that didn’t reach statistical significance, and widespread Mthfr expression prevented identification of specific cell types driving the vascular changes.

The authors of this study recommend further research to measure blood flow with laser speckle contrast imaging, to explore cell-specific mechanisms using conditional models and to look at the interaction of vascular changes with amyloid and tau pathology to better understand early ADRD progression.