The Architecture of Macular Neovascularizations Predicts Treatment Responses to Anti-VEGF Therapy in Neovascular AMD

Abstract

Introduction: Anti-VEGF therapy is an effective option for improving and stabilizing the vision in neovascular age-related macular degeneration (nAMD). However, the response to treatment is markedly heterogeneous. The aim of this study was therefore to analyze the vascular characteristics of type 1,2, and 3 macular neovascularizations (MNV) in order to identify biomarkers that predict treatment response, especially with regard to changes in intraretinal and subretinal fluid. Materials and Methods: Overall, 90 treatment-naive eyes with nAMD confirmed by optic coherence tomography (OCT), fluorescein angiography, and OCT angiography (OCTA) were included in this retrospective study. The MNV detected by OCTA were subjected to quantitative vascular analysis by binarization and skeletonization of the vessel using ImageJ. We determined their area, total vascular length (sumL), fractal dimension (FD), flow density, number of vascular nodes (numN), and average vascular diameter (avgW). The results were correlated with the treatment response to the initial three injections of anti-VEGF and the changes in intraretinal (IRF) and subretinal fluid (SRF) and the occurrence of pigment epithelial detachements (PED). Results: All patients found to have no subretinal or intraretinal fluid following the initial three injections of anti-VEGF showed a significantly smaller MNV area (p < 0.001), a lower sumL (p < 0.0005), and lesser FD (p < 0.005) before treatment than those who still exhibited signs of activity. These parameters also showed a significant influence in the separate analysis of persistent SRF (p < 0.005) and a persistent PED (p < 0.05), whereas we could not detect any influence on changes in IRF. The vascular parameters avgW, numN, and flow density showed no significant influence on SRF/IRF or PED changes. Conclusions: The size, the total vessel length, and the fractal dimension of MNV at baseline are predictors for the treatment response to anti-VEGF therapy. Therefore, particularly regarding the development of new classes of drugs, these parameters could yield new insights into treatment response.