Geographic atrophy in age-related macular degeneration: phenotypic characterisation for clinical trial consideration

Geographic atrophy (GA) is an advanced form of age-related macular degeneration (AMD) and a leading cause of central vision loss. Advances in multimodal imaging for GA have improved its phenotypic characterisation, enabling more precise assessment of disease. This is increasingly important for identifying features predictive of progression to inform prognosis and guide patient counselling, enable selection for clinical trials and for disease monitoring both in routine clinical practice and in a research setting. In addition, accurately determining foveal involvement is crucial for selection of patients suitable for emerging therapies. High-resolution imaging is also important to recognise and distinguish GA subtypes such as pachychoroid GA from conventional GA, given their genetic and phenotypic differences and possible variation in response to therapy.Imaging modalities include colour fundus photography, which is widely available and allows an initial assessment of GA lesions. Fundus autofluorescence imaging permits clear visualisation of GA borders and provides an accurate topographical map of GA pattern and extent, whereas near-infrared reflectance imaging may be superior for evaluation of foveal involvement. Optical coherence tomography (OCT) allows for measurement of the ellipsoid zone which may correlate to visual function and permits differentiation between biomarkers such as nascent GA, incomplete and complete retinal pigment epithelium and outer retinal atrophy (iRORA and cRORA respectively), and identification of pachychoroid GA. Each of these have important prognostic implications and enable accurate selection for clinical trials, monitoring progression and treatment response. Emerging approaches such as red excitation light and high-resolution OCT, may provide more accurate and reliable assessment of atrophic changes. Alongside these advances, artificial intelligence-based tools show great potential in automating GA detection, characterising of structural biomarkers, measuring progression rates and screening patients for clinical trials, increasingly reliability and reproducibility. A better understanding of the important role of multimodal imaging in the classification and assessment of GA, and detection of factors that affect progression will enable clinicians to advise, monitor and, where possible, appropriately treat this major cause of sight loss.