Selecting likely causal risk factors from high-throughput experiments using multivariable Mendelian randomization.

Modern high-throughput experiments provide a rich resource to investigate causal determinants of disease risk. Mendelian randomization (MR) is the use of genetic variants as instrumental variables to infer the causal effect of a specific risk factor on an outcome. Multivariable MR is an extension of the standard MR framework to consider multiple potential risk factors in a single model. However, current implementations of multivariable MR use standard linear regression and hence perform poorly with many risk factors. Here, we propose a two-sample multivariable MR approach based on Bayesian model averaging (MR-BMA) that scales to high-throughput experiments. In a realistic simulation study, we show that MR-BMA can detect true causal risk factors even when the candidate risk factors are highly correlated. We illustrate MR-BMA by analysing publicly-available summarized data on metabolites to prioritise likely causal biomarkers for age-related macular degeneration.

Systemic and Ocular Determinants of Peripapillary Retinal Nerve Fiber Layer Thickness Measurements in the European Eye Epidemiology (E3) Population.

PURPOSE: To investigate systemic and ocular determinants of peripapillary retinal nerve fiber layer thickness (pRNFLT) in the European population. DESIGN: Cross-sectional meta-analysis. PARTICIPANTS: A total of 16 084 European adults from 8 cohort studies (mean age range, 56.9±12.3-82.1±4.2 years) of the European Eye Epidemiology (E3) consortium. METHODS: We examined associations with pRNFLT measured by spectral-domain OCT in each study using multivariable linear regression and pooled results using random effects meta-analysis. MAIN OUTCOME MEASURES: Determinants of pRNFLT. RESULTS: Mean pRNFLT ranged from 86.8±21.4 µm in the Rotterdam Study I to 104.7±12.5 µm in the Rotterdam Study III. We found the following factors to be associated with reduced pRNFLT: Older age (ß = -0.38 µm/year; 95% confidence interval [CI], -0.57 to -0.18), higher intraocular pressure (IOP) (ß = -0.36 µm/mmHg; 95% CI, -0.56 to -0.15), visual impairment (ß = -5.50 µm; 95% CI, -9.37 to -1.64), and history of systemic hypertension (ß = -0.54 µm; 95% CI, -1.01 to -0.07) and stroke (ß = -1.94 µm; 95% CI, -3.17 to -0.72). A suggestive, albeit nonsignificant, association was observed for dementia (ß = -3.11 µm; 95% CI, -6.22 to 0.01). Higher pRNFLT was associated with more hyperopic spherical equivalent (ß = 1.39 µm/diopter; 95% CI, 1.19-1.59) and smoking (ß = 1.53 µm; 95% CI, 1.00-2.06 for current smokers compared with never-smokers). CONCLUSIONS: In addition to previously described determinants such as age and refraction, we found that systemic vascular and neurovascular diseases were associated with reduced pRNFLT. These may be of clinical relevance, especially in glaucoma monitoring of patients with newly occurring vascular comorbidities.

Joint Contribution of Genetic Susceptibility and Modifiable Factors to the Progression of Age-Related Macular Degeneration over 10 Years: The Three Continent AMD Consortium Report.

PURPOSE: To assess joint effects of genetic and modifiable factors on the 10-year progression of age-related macular degeneration (AMD). DESIGN: Individual and pooled data analyses of 2 population-based cohorts. PARTICIPANTS: Blue Mountains Eye Study (BMES) and Rotterdam Study (RS) participants (n = 835). METHODS: Participants of the BMES and RS were followed up over 10 years or more. At baseline and follow-up visits, interviews using questionnaires and eye examinations with retinal photography were performed. Age-related macular degeneration was assessed by trained photographic graders and verified by retinal specialists. Genetic susceptibility to AMD meant carrying 2 or more risk alleles of the CFH or ARMS2 SNPs, or both (rs1061170 and rs10490924), relative to 0 or 1 risk allele. Discrete logistic regression models were used to investigate the joint associations of genetic susceptibility and either smoking, fish consumption, dietary intake of lutein-zeaxanthin, or combined environmental risk scores from the 3 modifiable factors with the risk of AMD progression. Odds ratios (ORs) with 95% confidence intervals (CIs) and synergy indexes are reported. MAIN OUTCOME MEASURE: Ten-year progression of AMD, categorized as any (≥1 step) or 2-step (≥2 steps) progression on the Three Continent AMD Consortium 5-step severity scale. RESULTS: Older age, the presence of AMD genetic susceptibility, and baseline AMD status were associated strongly with AMD progression (P < 0.0001). In analyses of pooled data, each additional score from the combined environmental risk scores was associated with an increased risk of 2-step progression over 10 years (OR, 1.26; 95% CI, 1.02-1.56). The copresence of AMD genetic susceptibility and combined risk score of 3 or more was associated with a substantially higher risk of 2-step progression compared with the presence of either factor alone. There was a significant synergistic effect (OR, 4.14; 95% CI, 1.07-15.95) and interaction (P = 0.025) between genetic susceptibility and environmental risk score of 3 or more. CONCLUSIONS: Among persons with AMD genetic susceptibility and pre-existing early AMD lesions, presenting with high environmental risk scores from 3 modifiable factors (smoking, infrequent consumption of fish, low lutein-zeaxanthin intake) were associated with an increased risk of 2-step progression over 10 years.

Five-year progression of unilateral age-related macular degeneration to bilateral involvement: the Three Continent AMD Consortium report.

PURPOSE: To assess the 5-year progression from unilateral to bilateral age-related macular degeneration (AMD) and associated risk factors. DESIGN: Pooled data analyses of three prospective population-based cohorts, the Blue Mountains Eye Study, Beaver Dam Eye Study and Rotterdam Study. METHODS: Retinal photography and interview with comprehensive questionnaires were conducted at each visit of three studies. AMD was assessed following the modified Wisconsin AMD grading protocol. Progression to bilateral any (early and late) or late AMD was assessed among participants with unilateral involvement only. Factors associated with the progression were assessed using logistic regression models while simultaneously adjusting for other significant risk factors. RESULTS: In any 5-year duration, 19-28% of unilateral any AMD cases became bilateral and 27-68% of unilateral late AMD became bilateral. Factors associated with the progression to bilateral involvement of any AMD were age (per year increase, adjusted OR 1.07), carrying risk alleles of the complement factor H and age-related maculopathy susceptibility 2 genes (compared with none, OR 1.76 for 1 risk allele and OR 3.34 for 2+ risk alleles), smoking (compared with non-smokers, OR 1.64 for past and OR 1.67 for current smokers), and the presence of large drusen area or retinal pigmentary abnormalities in the first eye. CONCLUSION: One in four to one in five unilateral any AMD cases, and up to one in two unilateral late AMD cases, progressed to bilateral in 5 years. Known AMD risk factors, including smoking, are significantly associated with the progression to bilateral involvement.