Conference Coverage

Retinal changes may reflect brain changes in preclinical Alzheimer’s



– Changes in the retina seem to mirror changes that begin to reshape the brain in preclinical Alzheimer’s disease.

Manifested as a reduction in volume in the retinal nerve fiber layer, these changes appear to track the aggregation of beta amyloid brain plaques well before cognitive problems arise – and can be easily measured with a piece of equipment already in many optometry offices, Peter J. Snyder, PhD, said at the Clinical Trials in Alzheimer’s Disease conference.

Dr. Peter J. Snyder

The findings suggest that retinal scans might eventually be an easy, noninvasive, and inexpensive way to tag people who may be at elevated risk for Alzheimer’s disease, Dr. Snyder said in an interview.

“If we are lucky enough to live past age 45, then it’s a given that we’re all going to develop some presbyopia. So we all have to go to the optometrist sometime, and that may become a point of entry for broad screening and to track changes over time, to keep an eye on at-risk patients, and to refer those with retinal changes that fit the preclinical AD profile to specialty care for more comprehensive diagnostic evaluations.”

The retina begins to form in the third week of embryologic life, arising from the neural tube cells that also form the brain and spinal cord. It makes sense then that very early neuronal changes in Alzheimer’s disease could be occurring in the retina as well, said Dr. Snyder, professor of neurology and surgery (ophthalmology) at Rhode Island Hospital and Brown University, Providence.

“The retina is really a protrusion of the brain, and it is part and parcel of the central nervous system. In terms of the neuronal structure, the retina develops in layers with very specific cell types that are neurochemically and physiologically the same as the nervous tissue in the brain. That’s why it is, potentially, literally a window that could let us see what’s happening in the brain in early Alzheimer’s disease.”

OCT angiographic scan of the retina. The bottom two panels show structural imaging of the retinal layers. Retinal nerve fiber layer is the light-colored top layer. Red in the bottom left image indicates active blood flow over 3 seconds. Courtesy Dr. Peter Synder

OCT angiographic scan of the retina. The bottom two panels show structural imaging of the retinal layers. Retinal nerve fiber layer is the light-colored top layer. Red in the bottom left image indicates active blood flow over 3 seconds.

Dr. Snyder is not the first to focus on the retina as a predictive marker for Alzheimer’s. Last summer, at the Alzheimer’s Association International Conference in London, Fang Sara Ko, MD, an ophthalmologist in Tallahassee, Fla., presented 3-year data associating retinal nerve fiber layer thinning to cognitive decline in the U.K.’s ongoing prospective health outcomes study, U.K. Biobank.

Other researchers have explored amyloid in the lens and retina as a possible early Alzheimer’s identification tool. But Dr. Snyder’s study is the first to demonstrate a longitudinal association between neuronal changes in the eye and amyloid burden in the brain among clinically normal subjects.

For 27 months, he followed 56 people who had normal cognition but were beginning to experience subjective memory complaints. All subjects had at least one parent with Alzheimer’s disease. Everyone underwent an amyloid PET scan at baseline. Of the cohort, 15 had PET imaging evidence of abnormal beta-amyloid protein aggregation in the neocortex. This group was deemed to have preclinical Alzheimer’s disease, while the remainder served as a control group.

Dr. Snyder imaged each subject’s retinas twice – once at baseline and once at 27 months, when everyone underwent a second amyloid PET scan as well. He examined the retina with spectral domain optical coherence tomography, a relatively new method of imaging the retina.

These scanners are becoming increasingly more common in optometry practices, Dr. Snyder said. “Graduate optometrists tell me they would not want to be in a practice without one.” The scanners are typically used to detect retinal and ocular changes associated with diabetes, macular degeneration, glaucoma and multiple sclerosis.

Dr. Snyder used the scanner to examine the optic nerve head and macula at both baseline and 27 months in his cohort. He was looking for volumetric changes in several of the retinal layers: the peripapillary retinal nerve fiber layer (pRNFL), macular RNFL (mRNFL), ganglion cell layer (GCL), inner plexiform layer (IPL), outer nuclear layer (ONL), outer plexiform layer (OPL), and inner nuclear layer (INL). He also computed changes in total retinal volume.

Even at baseline, he found a significant difference between the groups. Among the amyloid-positive subjects, the inner plexiform layer was slightly larger in volume. “This seems a bit counterintuitive, but I think it suggests that there may be some inflammatory processes going on in this early stage and that we are catching that inflammation.”

Dr. Snyder noted that this finding has recently been replicated by an independent research group in Perth, Australia – with a much larger sample of participants – and will be reported at international conferences this coming year.

At 27 months, both the total retinal volume and the macular retinal nerve fiber layer volume were significantly lower in the preclinical AD group than in the control group. There was also a volume reduction in the peripapillary retinal nerve fiber layer, although the between-group difference was not statistically significant.

In a multivariate linear regression model that controlled for age and total amyloid burden, the mean volume change in the macular retinal nerve fiber layer accounted for about 10% of the variation in PET binding to brain amyloid by 27 months. Volume reductions in all the other layers appeared to be associated only with age, representing normal age-related changes in the eye.

Dr. Snyder said this volume loss in the retinal nerve fiber layer probably represents early demyelination and/or degeneration of the axons coursing from the cell bodies in the ganglion cell layer, which project to the optic nerve head.

“This finding in the retina appears analogous, and possibly directly related to, a similar loss of white matter that is readily observable in the early stages of Alzheimer’s disease. At the same time, patients are beginning to experience both cholinergic changes in the basal forebrain and the abnormal aggregation of fibrillar beta-amyloid plaques. I don’t know to what extent these changes are mechanistically dependent on each other, but they appear to also be happening, in the earliest stages of the disease course, in the retina.”

There is a lot of work left to be done before retinal scanning could be employed as a risk-assessment tool, however. With every new biomarker – and especially with imaging – the ability to measure change occurs far in advance of an understanding of what those changes mean, and how to judge them accurately.

“Every time we have a major advance in imaging, the technical engineering breakthroughs precede our detailed understanding of what we’re looking at and what to measure. This is where we are right now with retinal imaging. Biologically, it makes sense to be looking at this as a marker of risk in those who are clinically healthy, and maybe later as a marker of disease progression. But there is a lot of work to be done here yet.”

Dr. Snyder’s project was funded in part by a research award from Pfizer, with PET imaging supported in part by a grant from Avid Radiopharmaceuticals. He has no financial ties to the company, or other financial interest related to the study.

On Twitter @Alz_Gal

Next Article: