Small Hyperreflective Retinal Foci as in vivo imaging feature of resident microglia activation in geographic atrophy

Geographic atrophy (GA), the atrophic late stage of age-related macular degeneration (AMD), is one of the leading causes of vision loss in developed countries. Based on genetic, histological and preclinical studies, the role of the innate immune system in the development and progression of GA is well established. Microglia, the principal resident immune cells, are recognized as key players in innate immunity and contributors to AMD development. Optical coherence tomography (OCT) allows to identify small hyperreflective retinal foci (HRF) with specific features known as aggregates of activated microglial cells as possible in vivo imaging feature of local neuroretinal inflammation. The purpose of this study was to evaluate the presence and amount of small HRF in the eyes of patients with different macular atrophic phenotypes. Patients with GA in both eyes (bilateral GA: B-GA group), patients with GA in one eye and macular new vessels (MNV) in the fellow-eye (unilateral GA: U-GA group) and patients with extensive macular atrophy with pseudodrusen (EMAP), a rare and aggressive variant of atrophic AMD, were retrospectively analyzed. HRF, defined as isolated punctiform elements of small dimensions (≤30 μm) with intermediate reflectivity (similar to that of the nerve fiber layer) and without a shadow cone, were manually identified and quantified. The amount of HRF was correlated to best corrected visual acuity (BCVA), GA lesion size, measured both at near infrared reflectance (NIR), and blue wavelength fundus autofluorescence (FAF) images, to some GA features (multifocal versus unifocal GA; presence versus absence of foveal sparing) and to central retinal thickness (CRT). Forty-six patients (26 in the B-GA group, 16 in the U-GA group and 4 in the EMAP group) were studied. Patients with EMAP were younger compared to patients with B-GA and to patients with U-GA (63.5 ± 6.8 years vs 80.4 ± 8.4 years B-GA, and vs 83.3 ± 6.1 years U-GA; p = 0.0004 and p= <0.0001, respectively). Mean BCVA, mean GA area at NIR and at FAF images, foveal sparing and multifocal versus unifocal GA distribution and mean CRT were not significantly different among groups. GA area was wider on NIR versus FAF in all groups, significantly in B-GA and U-GA groups (11.7 ± 7.6 mm² vs 10.6 ± 7.1 mm², p = 0.0087 in B-GA; 7.8 ± 9.2 mm² vs 7.7 ± 9.4 mm², p = 0.004 in U-GA). The number of HRF was significantly higher in U-GA compared to B-GA and to EMAP (47.4 ± 7.1 vs 31.6 ± 7.3 B-GA and 28.0 ± 4.9 EMAP, p < 0.0001 for both), while mean HRF number did not significantly differ between B-GA and EMAP (p = 0.1960). HRF count correlated only to CRT, positively in B-GA and negatively in U-GA group. The increase of small HRF, which mirrors retinal microglial activation, characterizes eyes with unilateral GA (and MNV in the fellow eye) but not eyes with bilateral GA or EMAP. The role of activated microglia in the retina of GA eyes needs to be better investigated, mainly considering the actual and new therapeutic strategies with which to reduce either the development or progression of the atrophic macular changes.