Faricimab in Neovascular AMD Complicated by Pigment Epithelium Detachment: An AI-Assisted Evaluation of Early Morphological Changes

Introduction: This study investigates the early temporal changes in pigment epithelial detachment (PED) morphology following treatment with faricimab in patients with neovascular age-related macular degeneration (nAMD). Utilizing an artificial intelligence (AI)-assisted approach, we provide a detailed quantification and characterization of the dynamics of these morphological changes. Methods: A prospective observational study was conducted on 22 eyes from 22 treatment-naïve patients with nAMD-associated PED (presenting either type 1 or type 3 macular neovascularization). Participants were administered intravitreal faricimab (6 mg) at baseline and at days 30, 60, and 90. Comprehensive ophthalmic evaluations and spectral-domain optical coherence tomography (SD-OCT) imaging were conducted at baseline and at seven additional follow-up visits on days 1, 7, 14, 30, 60, 90, and 120. An AI-based automated segmentation algorithm was utilized to precisely quantify changes in PED volume, alongside intraretinal (IRF) and subretinal fluid (SRF) volumes, at each time point. Results: Treatment with faricimab resulted in a significant reduction in mean PED volume, with an average decrease of 12% at day 1, 29% at day 7, 51% at day 14, 68% at day 30, 72% at day 60, 79% at day 90, and 84% at day 120 (p < 0.0001 for all time points). Similarly rapid and marked reductions were noted in both mean IRF (23.5% at day 1, 90.7% at day 14) and SRF (14.4% at day 1, 91.2% at day 14) volumes. The study also showed a statistically significant improvement in best-corrected visual acuity (BCVA) over the follow-up period, correlating with the reduction in PED volume. Conclusion: Faricimab demonstrates early and significant efficacy in improving PED architecture in patients with nAMD. The rapid morphological improvements observed in this study suggest faricimab may represent a valid therapeutic option for PEDs associated with nAMD.