Type 1 Choroidal Neovascularization Is Associated with Reduced Localized Progression of Atrophy in Age-Related Macular Degeneration

Purpose: To investigate the association between the presence of type 1 choroidal neovascularization (CNV) and the localized progression of atrophy in age-related macular degeneration (AMD).

Design: Analysis of patients' data collected in the context of 2 noninterventional, prospective studies conducted at the Department of Ophthalmology, University of Bonn, Germany.

Participants: A total of 98 eyes diagnosed with AMD of 59 patients (40 female, 19 male) with a mean (±standard deviation) age at baseline of 76.60±6.65 years and median (interquartile range) review period of 1.17 years (1.01-1.55) were included. Eyes were subdivided into 3 categories based on multimodal imaging and ocular history: retinal pigment epithelium (RPE) atrophy with treatment-naïve quiescent CNV (n=7), RPE atrophy with a history of exudative CNV (n=10), and RPE atrophy without evidence of coexisting CNV (n=81).

Methods: Retinal pigment epithelium atrophy was delineated on the basis of serial fundus-autofluorescence and infrared-reflectance images. If CNV was detected by OCT angiography (OCTA), its location and dimension were spatially mapped to RPE atrophy. The localized progression of RPE atrophy in topographic relation to the CNV lesion was then analyzed using mixed-effects logistic regression. The spatial overlap (Dice coefficient) between predicted and observed RPE atrophy progression was evaluated to estimate the model accuracy.

Main outcome measures: Odds ratio (OR) for localized RPE atrophy progression in areas overlying type 1 CNV.

Results: The prediction model achieved a high overlap between predicted and observed RPE atrophy progression with a cross-validated Dice coefficient of 0.87 (95% confidence interval [CI], 0.85-0.89) reflecting a high accuracy. The odds for future RPE atrophy involvement were reduced by a factor of 0.21 (95% CI, 0.19-0.24) in the presence of treatment-naïve quiescent type 1 CNV and by a factor of 0.46 (95% CI, 0.41-0.51) in the presence of exudative type 1 CNV.

Conclusions: The results indicate that there is markedly reduced RPE atrophy progression in areas co-localizing with quiescent and exudative type 1 CNV. This observation is compatible with a potential protective effect of type 1 CNV on the RPE and overlying neurosensory retina. These results may have relevant clinical implications for the management of CNV and lead to new therapeutic strategies to prevent atrophy progression.