Omega-3 Fatty Acids as Protective Factors for Age-related Macular Degeneration: Prospective Cohort and Mendelian Randomization Analyses.

Purpose: Epidemiological studies and clinical trials have reported inconsistent findings regarding the protective role of omega-3 fatty acids in age-related macular degeneration (AMD), we investigated their association in a prospective cohort and examined causality using Mendelian randomization (MR) analyses.

Design: Prospective cohort study and two-sample MR analyses.

Participants: We included individuals of European descent from UK Biobank with plasma omega-3 and docosahexaenoic acid (DHA) measurement. Our MR analyses used genome-wide association study data on plasma level of omega-3 and DHA (UK Biobank, n=115,006) and AMD (dry, wet, and any) (FinnGen, n=208,690 to 209,122).

Methods: Cox regression models was used to assess the association between plasma omega-3/DHA levels and AMD incidence, adjusting for systemic covariates and AMD polygenetic risk score (PRS). Interaction effects of AMD genetic risk (PRS, CFH/ARMS2 genotypes) and plasma omega-3/DHA levels were tested. For MR analyses, we used random-effect inverse-variance weighted (RE-IVW) model as primary model and five other models for sensitivity. Causality between omega-3/DHA and AMD was considered significant if P-values were <0.05 in the primary model and at least 2 other models.

Main Outcome Measures: The risk of AMD.

Results: Among 258,350 AMD-free individuals, 5,068 (1.9%) developed AMD over 12.9-year follow-up. Higher plasma levels (in mmol/L) of omega-3 (HR, 0.80; 95%CI, 0.72, 0.95; P=0.006) and DHA (HR, 0.65; 95%CI, 0.44, 0.96; P=0.029) were associated with lower risk of being diagnosed with AMD. MR analyses showed causality between genetic predisposition to higher plasma omega-3 levels and lower risk of dry AMD (OR, 0.83; 95% CI, 0.73, 0.96, P=0.010), wet AMD (OR, 0.76; 95%CI, 0.65,0.88; P<0.001), and any AMD (OR, 0.82; 95%CI, 0.74, 0.92; P<0.001). Likewise, genetic predisposition to higher plasma DHA levels were causally associated with a lower risk of wet AMD (OR, 0.79; 95%CI, 0.65, 0.96; P= 0.017) and any AMD (OR, 0.84; 95%CI, 0.72, 0.98; P=0.030). No significant interaction was found between omega-3/DHA levels and AMD genetic risks (all P>0.05).

Conclusions: Our prospective cohort and MR analyses both suggested a protective effect of omega-3 and DHA on AMD, supporting the need for further clinical trials to test their effectiveness in AMD prevention and treatment.