The retina is a part of the central nervous system, originating as an outgrowth of the developing brain and shares many similarities with the brain. The retina is easily accessible for direct and noninvasive imaging with high spatial resolution and sensitivity, providing an ideal diagnostic target.^15,16  It has been shown that Aβ plaques occur in the retinas in addition to the brains of Alzheimer’s sufferers, and furthermore exhibit autofluorescence that is easily visible with slightly modified standard ophthalmic equipment.  There is potential to visualize amyloid at an earlier point in disease progression based on histologic evidence showing retinal amyloid's existence in early-state cases of disease. ¹⁷

Aβ appears in the neural retina due to the same processing pathways of amyloid precursor protein (APP) into amyloid beta fragments as occurs in the brain¹⁸. In the retina, APP is synthesized by retinal ganglion cells.^19,20 In addition to the production of Aβ in place, the connection of the retina and brain via the optic nerve presents the possibility of retrograde (or anterograde) transport of intracellular amyloids and APP.

The presence of amyloid beta in the neural retina of Alzheimer’s patients has been demonstrated by multiple research groups.  In 2011, Koronyo et al. reported that Aβ plaques were present in the retina of subjects who had postmortem confirmation of AD diagnosis using histology. Their work used immunohistochemistry labeling of Aβ in the retina and found plaque concentrations and morphology similar to those in the brain, in stark contrast with the absence of plaques in the retinas and brains of a group of non-AD subjects.

In an independent study of 16 post mortem eyes and brains from 16 donors with neurological disorders, including 13 with AD, Campbell et al. report that amyloid deposits were present in all the AD eyes and neither of the two non-AD eyes.²¹ Subsequently Campbell et al. reported that retinal amyloid had a 100% sensitivity for AD in a group of matched eyes and brains from 26 donors analyzed post mortem.¹⁵  Campbell’s method was analogous to Koronyo et al. in that ex vivo staining of human retinas was conducted on retinal flat mounts and the retinal area surveyed for evidence of Aβ.

In 2017 Koronyo et al. published further work showing the strong correlation between retinal and cerebral plaque load in AD subjects and non-AD matched controls, and the significantly larger retinal amyloid burden in AD subjects versus controls, using post-mortem histology. In addition, retinal autofluorescent imaging was used for the first time qualitatively and quantitatively to demonstrate with high significance that both autofluorescent spot quantity and spot intensity was greater in AD patients than in younger, healthy controls.²³

Additional studies have further demonstrated the potential for retinal AF measurement as a diagnostic tool in AD. In a feasibility study, Ngolab et al. showed that retinal AF quantification correlated in a strong and positive way with amyloid PET standardized uptake value ratio (SUVr), and clearly separated amyloid positive from amyloid negative subjects.²⁴ Retinal AF quantification was shown by Dumitrascu et al. to correlate negatively with hippocampal volume and positively with increasing cognitive decline as measured by standardized cognitive tests.²⁵ Kile et al. showed the feasibility of using retinal AF to detect changes in cerebral amyloid for patients undertaking an AD disease modifying therapy.²⁶