Getting screened for Alzheimer’s disease could soon mean taking a trip to the eye doctor. Decreased retinal thickness, the presence of abnormal proteins, and changes in how the retinal blood vessels respond to light all appear to be signs of neurodegenerative disease, according to researchers who spoke at the recent Alzheimer’s Association International Conference (AAIC 2016) in Toronto.

All of these could be detected with non-invasive eye exams, which would represent a huge leap forward for patients and Alzheimer’s researchers alike.

Alzheimer’s is the most common cause of dementia, and it’s irreversible. It affects an estimated 5 million Americans, and the numbers are growing. But right now, there’s no perfect way to diagnose it: Doctors perform memory tests on their patients, or take a detailed family history, which means the disease sometimes isn’t caught until it’s progressed. A definitive diagnosis generally can’t be done until after the patient’s death, when clusters of abnormal proteins called amyloid plaques (a hallmark of the disease) can be found in brain tissue samples.

Earlier detection would mean that patients and their families could plan ahead, and that researchers could better study the disease. Improved screening methods would enable doctors to identify who’s at risk, maybe even before their symptoms start to show.

Read More: Can Learning to Code Delay Alzheimer's?

The eyes are attracting attention as a portal to what’s happening in the brain. At a session at the AAIC 2016, researchers focused on the retina, which sits in the back of the eye and is made up of nerve tissue. The eyes are like windows into the brain, said Melanie Campbell, professor of optometry and vision science at the University of Waterloo. She told Motherboard in an interview that amyloid plaques can appear in the back of the eyes on the retina.

It’s possible amyloids leak into the vitreous fluid of the eye from the cerebrospinal fluid, Campbell said. Researchers also hypothesize that amyloid proteins are synthesized by neural cells within the eye, a similar process to what happens in the brains of Alzheimer’s patients, appearing in both the retina and the vitreous fluid.

Right now, in the lab, amyloids can be detected on retinas using rather complicated and expensive eye-imaging techniques. But Campbell and colleagues developed a prototype device that does the job more easily and cheaply. This new technology, called polarimetry, uses polarized light.

“It turns out amyloids show up very clearly under polarized light,” she said.

She presented results of a series of proof-of-concept scans done on human and canine retinas. The scans were conducted on a series of cadaver retinas from the Eye Bank of Canada (20 from people who had Alzheimer’s, and 22 controls), as well on living and postmortem canine retinas.

The researchers found that amyloid deposits were not only easy to detect with this new technology, but it was relatively easy to count them, and to measure their size—something other imaging techniques can’t do. The next step will be testing the device clinically on patients with Alzheimer’s disease, Campbell said. However, the presence of amyloids isn’t a guaranteed way to diagnose it; they show risk so this would be for screening.


Another clue of the disease is thin retinal nerve fiber layers (RNFL). In fact, the thinner RNFLs are, the poorer the cognition levels of subjects, according to Fang Ko, clinical associate professor of ophthalmology, Florida State University and Moorfields Eye Hospital in the UK, who also spoke at the conference.

Here, researchers used data from the UK Biobank, which included medical and health details of 500,000 volunteers aged between 40 to 69 years from across England. Of these, 67,000 underwent eye exams, which included retinal imaging. Many were ultimately excluded (including those with diabetes or other conditions that affect the retina), leaving about 32,000 subjects. They completed four different cognitive tests. Of those, a total of 1,251 participants went on to repeat the cognitive tests after three years.

Researchers found that people with thinner RNFLs performed worse on each of the cognitive tests than those whose RNFLs were thicker. And those who started the study with thinner RNFL had greater cognitive decline at the three year follow-up than those who had thicker ones.

It may be possible to use thin RNFL as a predictor of cognitive decline, she said, but it isn’t a surefire method: diseases like glaucoma can also affect its thickness, so once again, this could be a useful tool for screening rather than diagnosis.

A third technique, using a flickering light exam of the retinal blood vessels, could also help screen for Alzheimer’s, according to Konstantin Kotliar, a biomedical engineer at the Aachen University of Applied Sciences in Germany.

In healthy eyes, a flickering light shone on the retina causes immediate dilation of both retinal arteries and veins. “In people with Alzheimer’s disease, retinal arteries and veins have a delayed reaction to a flickering light test,” he said. But, they undergo greater dilation than in people without the disease. (Diminished and sometimes delayed dilation is also seen in eye diseases like glaucoma, he said.)

At the conference, Kotliar presented a study (unpublished as of yet) measuring and comparing retinal vessel reactions to flickering light in patients aged 60 to 79. Fifteen had mild-to-moderate dementia due to Alzheimer’s; 24 had mild cognitive impairment, also from Alzheimer’s, and 15 were healthy controls with no cognitive impairment. Retinal artery and vein reactions to 20-second-long flicker stimulation were measured.

Both arteries and veins dilated more in people with mild to moderate Alzheimer’s than in controls. Also, the start of dilation in the retinal arteries took longer in people with Alzheimer’s than in controls—though the delay wasn’t as pronounced in the veins. How the retinal vessels behaved in Alzheimer’s patients was a surprise, and this might contribute to another screening test, he said.