Structural Biomarkers Influencing Drusenoid Pigment Epithelial Detachment Lifecycle and the Development of Late...

Drusenoid pigment epithelial detachment (dPED) is a notable phenotype in age-related macular degeneration (AMD), often evolving into macular complications such as macular neovascularization (MNV) and geographic atrophy (GA). The aim of this study was to identify potential prognostic biomarkers associated with the development of both MNV and GA.

A retrospective cohort study.

Patients with dPED in the setting of AMD.

This observational study analyzed OCT biomarkers to assess the dPED lifecycle and identify features predictive of macular complications. Seventy-one eyes with dPED from 51 patients were reviewed over an average follow-up of 37.5 ± 17.6 months (range: 24-104), examining structural alterations via multimodal imaging, which included color fundus photography, fundus autofluorescence, and OCT, while fluorescein angiography, indocyanine green angiography, or both were performed as needed. Associations between baseline biomarkers and macular complications were assessed using Cox proportional hazard models with a frailty term to account for intereye correlation. The Fine-Gray model was used to account for competing risk analysis.

Incidence and time to development of macular complications (MNV and GA) and their associations with baseline OCT biomarkers (cuticular drusen, hyperreflective foci, external limiting membrane/ellipsoid zone integrity, and retinal pigment epithelium [RPE] hypertransmission), modeled with frailty-adjusted Cox proportional hazards and Fine-Gray competing risks; secondary measures included dPED lifecycle features (collapse, timing) and morphometrics (height, width, volume).

Key findings included a 39.4% incidence of dPED collapse, with 60.7% of cases progressing to complications postcollapse. In the multivariable Cox proportional model, cuticular drusen (hazard ratio [HR]: 3.8, 95% confidence interval [CI]: 1.62-9.2, P = 0.002) and the presence of hyperreflective foci (HRF) at baseline (HR: 6.6, 95% CI: 1.97-22, P = 0.02) represented the main prognostic indicators of macular complications. In the Fine-Gray competing risks analysis, cuticular drusen remained a significant independent predictor (subdistribution hazard ratio [sHR] = 18.1, 95% CI: 1.89-174, P = 0.01) of MNV development, while HRF (sHR = 6.69, 95% CI: 1.98-22.61, P = 0.002) and external limiting membrane disruption at baseline (sHR = 3.69, 95% CI: 1.03-13.14, P = 0.044) were factors significantly associated with increased GA risk.

These results underscore the prognostic relevance of specific imaging biomarkers in dPED. Recognizing these features early may support timely treatment and help prevent irreversible photoreceptor and RPE damage.

Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.