Review of emerging imaging findings to reveal a broader spectrum of lesions in AMD.

Age-related macular degeneration (AMD) is the leading cause of irreversible vision loss in developed countries, with prevalence expected to reach nearly 288 million by 2040. Accurate classification of AMD is critical for patient care and clinical research, guiding prognosis, therapeutic strategies, and the design of clinical trials. A widely adopted framework, the Beckman classification, stratifies AMD based primarily on color fundus photography (CFP) findings, defining stages from early to late disease. While simple and clinically applicable, such systems do not account for several key phenotypes revealed by advances in multimodal imaging. Such novel phenotypes include reticular pseudodrusen (RPD), acquired vitelliform lesions (AVL), non-exudative macular neovascularization (MNV), incomplete retinal pigment epithelium and outer retina atrophy (iRORA), and non-neovascular exudative fluid. Recent imaging modalities-including optical coherence tomography (OCT), fundus autofluorescence (FAF), and OCT angiography (OCTA)-have uncovered features with important prognostic implications that are currently misclassified with respect to AMD. For example, eyes with RPD or AVL in the absence of drusen are currently misclassified as "early AMD" or excluded altogether, despite their high risk of progression. Similarly, the ambiguous status of non-exudative MNV, which carries both protective and harmful potential, highlights the need for greater granularity. The Classification of Atrophy Meetings (CAM) group has also introduced refined OCT-based definitions such as iRORA and cRORA, underscoring early degenerative changes that precede geographic atrophy. Moreover, novel entities like non-neovascular intraretinal or subretinal fluid challenge the assumption that exudation is synonymous with neovascular AMD. This review synthesizes recent evidence highlighting the limitations of current classification systems in light of these advances. Furthermore, we emphasize that intermediate AMD, currently treated as a uniform category, actually encompasses highly heterogeneous phenotypes with distinct risks and trajectories. A more nuanced, imaging-integrated classification system is urgently needed to improve disease staging, identify high-risk eyes, and ensure appropriate patient selection for emerging therapies. Such a framework would not only better reflect the complex natural history of AMD but also facilitate the regulatory shift toward continuous, quantitative endpoints in clinical trials.