Strengthening the reporting of genetic risk prediction studies (GRIPS): explanation and elaboration

Detection of gene-environment interactions in the presence of linkage disequilibrium and noise by using genetic risk scores with internal weights from elastic net regression

BACKGROUND

For the analysis of gene-environment (GxE) interactions commonly single nucleotide polymorphisms (SNPs) are used to characterize genetic susceptibility, an approach that mostly lacks power and has poor reproducibility. One promising approach to overcome this problem might be the use of weighted genetic risk scores (GRS), which are defined as weighted sums of risk alleles of gene variants. The gold-standard is to use external weights from published meta-analyses.

METHODS

In this study, we used internal weights from the marginal genetic effects of the SNPs estimated by a multivariate elastic net regression and thereby provided a method that can be used if there are no external weights available. We conducted a simulation study for the detection of GxE interactions and compared power and type I error of single SNPs analyses with Bonferroni correction and corresponding analysis with unweighted and our weighted GRS approach in scenarios with six risk SNPs and an increasing number of highly correlated (up to 210) and noise SNPs (up to 840).

RESULTS

Applying weighted GRS increased the power enormously in comparison to the common single SNPs approach (e.g. 94.2% vs. 35.4%, respectively, to detect a weak interaction with an OR ≈ 1.04 for six uncorrelated risk SNPs and n = 700 with a well-controlled type I error). Furthermore, weighted GRS outperformed the unweighted GRS, in particular in the presence of SNPs without any effect on the phenotype (e.g. 90.1% vs. 43.9%, respectively, when 20 noise SNPs were added to the six risk SNPs). This outperforming of the weighted GRS was confirmed in a real data application on lung inflammation in the SALIA cohort (n = 402). However, in scenarios with a high number of noise SNPs (>200 vs. 6 risk SNPs), larger sample sizes are needed to avoid an increased type I error, whereas a high number of correlated SNPs can be handled even in small samples (e.g. n = 400).

CONCLUSION

In conclusion, weighted GRS with weights from the marginal genetic effects of the SNPs estimated by a multivariate elastic net regression were shown to be a powerful tool to detect gene-environment interactions in scenarios of high Linkage disequilibrium and noise.

Critical appraisal and data extraction for systematic reviews of prediction modelling studies: the CHARMS checklist

Carl Moons and colleagues provide a checklist and background explanation for critically appraising and extracting data from systematic reviews of prognostic and diagnostic prediction modelling studies. Please see later in the article for the Editors' Summary.

Translating genetic risk factors for prostate cancer to the clinic: 2013 and beyond

Prostate cancer (PrCa) is the most commonly diagnosed cancer in the male UK population, with over 40,000 new cases per year. PrCa has a complex, polygenic predisposition, due to rare variants such as BRCA and common variants such as single nucleotide polymorphisms (SNPs). With the introduction of genome-wide association studies, 78 susceptibility loci (SNPs) associated with PrCa risk have been identified. Genetic profiling could risk-stratify a population, leading to the discovery of a higher proportion of clinically significant disease and a reduction in the morbidity related to age-based prostate-specific antigen screening. Based on the combined risk of the 78 SNPs identified so far, the top 1% of the risk distribution has a 4.7-times higher risk of developing PrCa compared with the average of the general population.

Prediction of age-related macular degeneration in the general population: the Three Continent AMD Consortium

PURPOSE

Prediction models for age-related macular degeneration (AMD) based on case-control studies have a tendency to overestimate risks. The aim of this study is to develop a prediction model for late AMD based on data from population-based studies.

DESIGN

Three population-based studies: the Rotterdam Study (RS), the Beaver Dam Eye Study (BDES), and the Blue Mountains Eye Study (BMES) from the Three Continent AMD Consortium (3CC).

PARTICIPANTS

People (n = 10,106) with gradable fundus photographs, genotype data, and follow-up data without late AMD at baseline.

METHODS

Features of AMD were graded on fundus photographs using the 3CC AMD severity scale. Associations with known genetic and environmental AMD risk factors were tested using Cox proportional hazard analysis. In the RS, the prediction of AMD was estimated for multivariate models by area under receiver operating characteristic curves (AUCs). The best model was validated in the BDES and BMES, and associations of variables were re-estimated in the pooled data set. Beta coefficients were used to construct a risk score, and risk of incident late AMD was calculated using Cox proportional hazard analysis. Cumulative incident risks were estimated using Kaplan-Meier product-limit analysis.

MAIN OUTCOME MEASURES

Incident late AMD determined per visit during a median follow-up period of 11.1 years with a total of 4 to 5 visits.

RESULTS

Overall, 363 participants developed incident late AMD, 3378 participants developed early AMD, and 6365 participants remained free of any AMD. The highest AUC was achieved with a model including age, sex, 26 single nucleotide polymorphisms in AMD risk genes, smoking, body mass index, and baseline AMD phenotype. The AUC of this model was 0.88 in the RS, 0.85 in the BDES and BMES at validation, and 0.87 in the pooled analysis. Individuals with low-risk scores had a hazard ratio (HR) of 0.02 (95% confidence interval [CI], 0.01-0.04) to develop late AMD, and individuals with high-risk scores had an HR of 22.0 (95% CI, 15.2-31.8). Cumulative risk of incident late AMD ranged from virtually 0 to more than 65% for those with the highest risk scores.

CONCLUSIONS

Our prediction model is robust and distinguishes well between those who will develop late AMD and those who will not. Estimated risks were lower in these population-based studies than in previous case-control studies.

The Rotterdam Study: 2014 objectives and design update

The Rotterdam Study is a prospective cohort study ongoing since 1990 in the city of Rotterdam in The Netherlands. The study targets cardiovascular, endocrine, hepatic, neurological, ophthalmic, psychiatric, dermatological, oncological, and respiratory diseases. As of 2008, 14,926 subjects aged 45 years or over comprise the Rotterdam Study cohort. The findings of the Rotterdam Study have been presented in over a 1,000 research articles and reports (see www.erasmus-epidemiology.nl/rotterdamstudy ). This article gives the rationale of the study and its design. It also presents a summary of the major findings and an update of the objectives and methods.

Genomic medicine in the prevention and treatment of atherosclerotic cardiovascular disease

Cardiovascular disease (CVD) is the leading cause of death in the world. Over the past decade considerable progress has been made in understanding the genomic basis of polygenic disorders including CVD. The future application of genomic medicine to the prevention and treatment of CVDs will ultimately lessen the burden of CVD. Given the complex nature of CVD, information derived from newer evolving fields, such as transcriptomics, proteomics and metabolomics, will allow us to fully interrogate features of the human genome to better understand disease pathogenesis and to identify new drug targets. In this article, we will review how genomics will allow enhanced risk prediction of cardiovascular events, provide personalized treatment options and hasten the drug development process, with a particular focus on atherosclerotic CVD.

Genetic variants associated with predisposition to prostate cancer and potential clinical implications

Prostate cancer is the commonest cancer in the developed world. There is an inherited component to this disease as shown in familial and twin studies. However, the discovery of these variants has been difficult. The emergence of genome-wide association studies has led to the identification of over 46 susceptibility loci. Their clinical utility to predict risk, response to treatment, or treatment toxicity, remains undefined. Large consortia are needed to achieve adequate statistical power to answer these genetic-clinical and genetic-epidemiological questions. International collaborations are currently underway to link genetic with clinical/epidemiological data to develop risk prediction models, which could direct screening and treatment programs.

PredictABEL: an R package for the assessment of risk prediction models

The rapid identification of genetic markers for multifactorial diseases from genome-wide association studies is fuelling interest in investigating the predictive ability and health care utility of genetic risk models. Various measures are available for the assessment of risk prediction models, each addressing a different aspect of performance and utility. We developed PredictABEL, a package in R that covers descriptive tables, measures and figures that are used in the analysis of risk prediction studies such as measures of model fit, predictive ability and clinical utility, and risk distributions, calibration plot and the receiver operating characteristic plot. Tables and figures are saved as separate files in a user-specified format, which include publication-quality EPS and TIFF formats. All figures are available in a ready-made layout, but they can be customized to the preferences of the user. The package has been developed for the analysis of genetic risk prediction studies, but can also be used for studies that only include non-genetic risk factors. PredictABEL is freely available at the websites of GenABEL ( http://www.genabel.org ) and CRAN ( http://cran.r-project.org/).