Blood biomarkers for age-related macular degeneration – BIOMAC

Age-related macular degeneration (AMD) is a major cause of blindness, with dramatic personal and societal consequences. The advanced forms of the disease, associated with major visual loss, are preceded by early forms, generally asymptomatic, which represent a window for prevention. In addition to aging and genetic susceptibility, nutrition and smoking are major drivers of this complex disease. Being modifiable, they represent important levers for the prevention of AMD, especially of its advanced forms. Indeed, smokers exhibit a 3-fold increased risk for advanced AMD, with a 5-fold increase in heavy smokers. Strong associations of the progression to advanced AMD with several nutrients have been identified, in particular with lutein and zeaxanthin (which accumulate actively in the macular area, constituting the macular pigment), antioxidants (vitamins C, E and zinc) and omega-3 polyunsaturated fatty acids (PUFAs). However, assessment of lifestyle exposures is difficult, as it mainly relies on questionnaires, which are subject to numerous biases and approximations. We hypothesize that the use of specific blood biomarkers would represent more precise and objective assessments of nutritional and smoking exposures, thus improving the estimation of the risk of progression to advanced AMD. This project will leverage the cohorts Alienor (1109 elderly participants from the general population of Bordeaux, France, with multimodal diagnosis of AMD during a follow-up of 14 years) and Macustar (767 participants with intermediate AMD from 20 clinical centers in 7 European countries, with multimodal diagnosis of AMD during a follow-up of up to 5 years). Using blood samples stored in the biobanks of the 2 cohorts, we propose to measure: 1) plasma lipophilic biomarkers (including the patented biomarker for retinal omega-3 PUFAs content from our ANR BLISAR project and other key AMD nutrients (lutein and zeaxanthin, vitamins A, D, and E)), using a novel methodology based on Supercritic Fluid Chromatography coupled to tandem mass spectrometry (SFC-QqQ/MS); 2) blood concentrations of 9 metals (including zinc, a key AMD nutrient, and cadmium, a biomarker of smoking exposure), using a sensitive and accredited method based on triple quadrupole inductively coupled plasma mass spectrometry. Several machine learning techniques (elastic net, PLS-Cox, survival random forest...) will be used to develop a combination of these biomarkers with high predictive value for progression to advanced AMD. This combination of biomarkers will be developed on the Alienor data and validated on the Macustar data (external validation). It will then be introduced in our multidimensional prediction model resulting from the H2020 Eye-Risk project, in order to determine if its prediction performances can be improved. This model is based on age, a polygenic AMD risk score, presence of early retinal alterations and lifestyle and clinical factors (smoking, Mediterranean diet score and blood pressure), and has been implemented in a digital platform (www.macutest.net). Finally, the specific associations of the identified blood biomarkers with cutting-edge multimodal retinal imaging biomarkers will also be explored. Hence, the BIOMAC project relies on innovative promising biological assays, which have a great potential for assessment of personal risk for AMD in large groups of population. It also builds on existing cohorts with available biobanks and long-term follow-up, allowing assessment of the predictive value of these assays. In the future, the identified biomarkers will represent a major asset for the development of personalized preventive strategies, helping to target, tailor and monitor lifestyle interventions.