Importance: The association of major lipid genes with and their potential as drug targets for age-related macular degeneration (AMD) is unknown. These associations are important to study because AMD is the leading cause of irreversible late-onset blindness in high-income countries. Objective: To determine whether the full range of structural genetic variation in apolipoprotein E (APOE), a master gene in peripheral and cerebral lipid metabolism, is associated with risk of AMD. Design, Setting, and Participants: This cohort study used data from the Copenhagen City Heart Study (CCHS) and the Copenhagen General Population Study (CGPS) cohorts. Participants were followed from study inclusion at the time of blood sampling to occurrence of event, death, emigration, or December 7, 2018, whichever came first. For participants in CCHS, the APOE gene was sequenced, and 9 variants with a heterozygote frequency of at least 0.0002 were genotyped in the CGPS. Observers were masked to patient groupings. Data were analyzed from March to September 2021. Exposures: The exposure was APOE status, and the direct gene product in plasma, apoE levels, was measured in all participants. Main Outcomes and Measures: Cox regression was applied to estimate risk of AMD associated with APOE genotype. Results: A total of 105 546 participants (mean [SD] age, 57.7 [13.4] years; 58 140 [55%] female participants) were included. Compared with participants with the common ɛ33 genotype, risk of AMD was lower in participants with ε44 (multifactorially adjusted hazard ratio [aHR], 0.66; 95% CI, 0.45-0.96) and ε43 (aHR, 0.80; 95% CI, 0.71-0.90) genotypes and higher in the ε32 (aHR, 1.15; 95% CI, 1.00-1.31) genotype. Compared with noncarriers, risk of AMD was higher for participants with Gly145Asp (aHR, 3.53; 95% CI, 1.14-10.96) and Arg154Cys (aHR, 4.52; 95% CI, 1-13-18.13) heterozygotes. Results were similar after further adjustment for lipid traits and after adjustment for the APOE ε2/ε3/ε4 variant. Combining all common and rare structural variants in a weighted allele score, risk of AMD per 1-mg/dL genetically higher plasma apoE was increased in the adjusted model (aHR, 1.12; 95% CI, 1.05-1.19), the adjusted model plus APOE ɛ2/ɛ3/ɛ4 status (aHR, 1.82; 95% CI, 1.20-2.76), and the adjusted model in individuals with the ε33 genotype only (aHR, 1.77; 95% CI, 1.14-2.75). Conclusions and Relevance: These findings highlight that structural variation in APOE beyond the ε2/ε3/ε4 variants may be important for risk of AMD in a population of European ancestry. Rare functional ɛ2-like variants in APOE have previously been reported to have protective associations for Alzheimer disease but the present findings suggest a simultaneous high risk of AMD. This would limit the drug target potential of mechanisms resembling these variants.