Retinal thickness changes over time by SD-OCT in a AD mouse model (APP^NL-F/NL-F)

Abstract

Purpose : Alzheimer’s disease (AD) presents retinal neurodegenerative changes even before they appear in the brain, suggesting the retina as an accessible biomarker of AD. The present work is a diachronic study using spectral domain optical coherence tomography (SD-OCT) to determine the total retinal thickness and retinal nerve fiber layer (RNFL) thickness at 6, 9, 12, 15, 17, and 20 months old, in an APP^NL-F/NL-F mouse model of AD compared to wild type (WT) animals.

Methods : Total retinal thickness and RNFL thickness were determined in male APP^NL-F/NL-F (n=55) and WT (n=41) animals. The mean total retinal thickness was analyzed following the Early Treatment Diabetic Retinopathy Study sectors. RNFL was measured in six sectors of axonal ring scans around the optic nerve. Microglial and astroglial activation was evaluated at 17-months-old, by analyzing vertical retinal sections immunostained with Iba-1 and GFAP.

Results : In the APP^NL-F/NL-F group compared to WT animals, for total retinal thickness was observed: i) At 6-months-old, a significant thinning in the outer temporal sector; ii) at 15-months-old a significant thinning in the inner temporal and in the inner and outer inferior retinal sectors; iii) at 17-months-old, a significant thickening in the inferior and nasal sectors in both inner and outer rings; and iv) at 20-months-old, a significant thinning in the inner ring of nasal, temporal and inferior retina and in the outer ring of superior and temporal retina. In RNFL thickness, a significant thinning in the global analysis and in nasal and inner-temporal sectors at 6 months old were observed. In addition, in the APP^NL-F/NL-F group we found activation of microglial cells (thicker somas and processes and amoeboid forms) and astroglial cells (higher GFAP+ immunostaining with astrocyte clusters in certain areas).

Conclusions : In the APP^NL-F/NL-F model, the thickness of the retina over time exhibited a thinning, probably produced by neurodegeneration alternating with the thickening possibly caused in addition to the deposits by neuroinflammation observed in some areas of the retina in this model. These changes over time are similar to those found in the human retina and could constitute a biomarker of AD. The APP^NL-F/NL-F AD model can provide us a better understanding of the different retinal alterations during the progression of AD.

This is a 2021 ARVO Annual Meeting abstract.