Biomarqueurs anatomiques dynamiques de la progression de la DMLA


Santiago Costantino

CIUSSS de l'Est-de-l'Île-de-Montréal


Domaine : vieillissement

Programme Recherches sur la dégénérescence maculaire liée à l'âge (DMLA)

Concours 2019-2020

Age-related macular degeneration (AMD) is the lead cause of vision loss in elderly population in developed countries. Given that age is its primary risk factor, the prevalence and severity of this disease are likely to rise as life expectancy increases. Two types of late-stage AMD are possible: a proliferative neovascular form and a degenerative form. The only available treatment today is efficient for neovascular AMD alone, and all alternative cures for the degenerative have failed. 

The identification of AMD is mostly based on pictures of the back of the eye, and retina specialists emphasize the major important of earlier diagnosis methods. Optical coherence tomography (OCT) allows the visualisation of deep layers of the retina in a precise non-invasive way, and our labs have developed computer algorithms to automatically characterize the anatomy of such layers automatically. In particular we have created a new computational methods to trace the choroid – the exclusive source of oxygen and nutrients of cells targeted by AMD, and key anatomical features that characterize their eyes.

Previous attempts to create computer models that predict whether an AMD patient will soon evolve towards the most aggressive forms of the disease have not successful. However, these studies mostly focused on static imaging rather than the frequent examination of the eye to predict the evolution of the disease. We propose a two-year pilot prospective study of the healthy eye of patients who are receiving AMD treatment in the contralateral eye. This cohort of patients is at high risk of developing advanced AMD in the healthy eye, and will permit detecting subtle changes using the most sophisticated imaging tools available to date and artificial intelligence algorithms.