Using genetic association data to guide drug discovery and development: Review of methods and applications

Evidence on the validity of drug targets from randomized trials is reliable but typically expensive and slow to obtain. In contrast, evidence from conventional observational epidemiological studies is less reliable because of the potential for bias from confounding and reverse causation. Mendelian randomization is a quasi-experimental approach analogous to a randomized trial that exploits naturally occurring randomization in the transmission of genetic variants. In Mendelian randomization, genetic variants that can be regarded as proxies for an intervention on the proposed drug target are leveraged as instrumental variables to investigate potential effects on biomarkers and disease outcomes in large-scale observational datasets. This approach can be implemented rapidly for a range of drug targets to provide evidence on their effects and thus inform on their priority for further investigation. In this review, we present statistical methods and their applications to showcase the diverse opportunities for applying Mendelian randomization in guiding clinical development efforts, thus enabling interventions to target the right mechanism in the right population group at the right time. These methods can inform investigators on the mechanisms underlying drug effects, their related biomarkers, implications for the timing of interventions, and the population subgroups that stand to gain the most benefit. Most methods can be implemented with publicly available data on summarized genetic associations with traits and diseases, meaning that the only major limitations to their usage are the availability of appropriately powered studies for the exposure and outcome and the existence of a suitable genetic proxy for the proposed intervention.

Author Correction: New genetic signals for lung function highlight pathways and chronic obstructive pulmonary disease associations across multiple ancestries

An amendment to this paper has been published and can be accessed via a link at the top of the paper.

New genetic signals for lung function highlight pathways and chronic obstructive pulmonary disease associations across multiple ancestries

Reduced lung function predicts mortality and is key to the diagnosis of chronic obstructive pulmonary disease (COPD). In a genome-wide association study in 400,102 individuals of European ancestry, we define 279 lung function signals, 139 of which are new. In combination, these variants strongly predict COPD in independent populations. Furthermore, the combined effect of these variants showed generalizability across smokers and never smokers, and across ancestral groups. We highlight biological pathways, known and potential drug targets for COPD and, in phenome-wide association studies, autoimmune-related and other pleiotropic effects of lung function-associated variants. This new genetic evidence has potential to improve future preventive and therapeutic strategies for COPD.

Polymorphisms in Natural Killer Cell Receptor Protein 2D (NKG2D) as a Risk Factor for Cholangiocarcinoma

BACKGROUND AND AIMS

Understanding of the significant genetic risk factors for Cholangiocarcinoma (CC) remains limited. Polymorphisms in the natural killer cell receptor G2D (NKG2D) gene have been shown to increase risk of CC transformation in patients with Primary Sclerosing Cholangitis (PSC). We present a validation study of NKG2D polymorphisms in CC patients without PSC.

METHODS

Seven common Single Nucleotide Polymorphisms (SNPs) of the NKG2D gene were genotyped in 164 non-PSC related CC subjects and 257 controls with HaploView. The two SNPs that were positively identified in the previous Scandinavian study, rs11053781 and rs2617167, were included.

RESULTS

The seven genotyped SNPs were not associated with risk of CC. Furthermore, haplotype analysis revealed that there was no evidence to suggest that any haplotype differs in frequency between cases and controls (P > 0.1).

CONCLUSION

The common genetic variation in NKG2D does not correlate significantly with sporadic CC risk. This is in contrast to the previous positive findings in the Scandinavian study with PSC-patients. The failure to reproduce the association may reflect an important difference between the pathogenesis of sporadic CC and that of PSC-related CC. Given that genetic susceptibility is likely to be multifaceted and complex, further validation studies that include both sporadic and PSC-related CC are required.

Adult height, coronary heart disease and stroke: a multi-locus Mendelian randomization meta-analysis

Background: We investigated causal effect of completed growth, measured by adult height, on coronary heart disease (CHD), stroke and cardiovascular traits, using instrumental variable (IV) Mendelian randomization meta-analysis.

Methods: We developed an allele score based on 69 single nucleotide polymorphisms (SNPs) associated with adult height, identified by the IBCCardioChip, and used it for IV analysis against cardiovascular risk factors and events in 21 studies and 60 028 participants. IV analysis on CHD was supplemented by summary data from 180 height-SNPs from the GIANT consortium and their corresponding CHD estimates derived from CARDIoGRAMplusC4D.

Results: IV estimates from IBCCardioChip and GIANT-CARDIoGRAMplusC4D showed that a 6.5-cm increase in height reduced the odds of CHD by 10% [odds ratios 0.90; 95% confidence intervals (CIs): 0.78 to 1.03 and 0.85 to 0.95, respectively],which agrees with the estimate from the Emerging Risk Factors Collaboration (hazard ratio 0.93; 95% CI: 0.91 to 0.94). IV analysis revealed no association with stroke (odds ratio 0.97; 95% CI: 0.79 to 1.19). IV analysis showed that a 6.5-cm increase in height resulted in lower levels of body mass index (P < 0.001), triglycerides (P < 0.001), non high-density (non-HDL) cholesterol (P < 0.001), C-reactive protein (P = 0.042), and systolic blood pressure (P = 0.064) and higher levels of forced expiratory volume in 1 s and forced vital capacity (P < 0.001 for both).

Conclusions: Taller individuals have a lower risk of CHD with potential explanations being that taller people have a better lung function and lower levels of body mass index, cholesterol and blood pressure.

Generalized estimating equations: A hybrid approach for mean parameters in multivariate regression models

We propose an extension of the generalized estimating equation approach to multivariate regression models (Liang and Zeger, 1986) which allows the estimation of dispersion and association parameters in the covariance matrix partly using estimating equations as in Prentice and Zhao (1991), and partly by the direct use of consistent estimators. The advantages of this hybrid approach over that of Prentice and Zhao (1991) are a reduction in the number of fourth moment assumptions that must be made, and the consequent reduction in numerical complexity. We show that the type of estimation used for covariance parameters does not affect the asymptotic efficiency of the mean parameter estimates. The advantages of the hybrid model are illustrated by a simulation study. This work was motivated by problems in statistical genetics, and we illustrate our approach using a twin study examining association between the osteocalcin receptor and various osteoporisis-related traits.

Selecting instruments for Mendelian randomization in the wake of genome-wide association studies

Mendelian randomization (MR) studies typically assess the pathogenic relevance of environmental exposures or disease biomarkers, using genetic variants that instrument these exposures. The approach is gaining popularity-our systematic review reveals a greater than 10-fold increase in MR studies published between 2004 and 2015. When the MR paradigm was first proposed, few biomarker- or exposure-related genetic variants were known, most having been identified by candidate gene studies. However, genome-wide association studies (GWAS) are now providing a rich source of potential instruments for MR analysis. Many early reviews covering the concept, applications and analytical aspects of the MR technique preceded the surge in GWAS, and thus the question of how best to select instruments for MR studies from the now extensive pool of available variants has received insufficient attention. Here we focus on the most common category of MR studies-those concerning disease biomarkers. We consider how the selection of instruments for MR analysis from GWAS requires consideration of: the assumptions underlying the MR approach; the biology of the biomarker; the genome-wide distribution, frequency and effect size of biomarker-associated variants (the genetic architecture); and the specificity of the genetic associations. Based on this, we develop guidance that may help investigators to plan and readers interpret MR studies.

Plasma urate concentration and risk of coronary heart disease: a Mendelian randomisation analysis

BACKGROUND

Increased circulating plasma urate concentration is associated with an increased risk of coronary heart disease, but the extent of any causative effect of urate on risk of coronary heart disease is still unclear. In this study, we aimed to clarify any causal role of urate on coronary heart disease risk using Mendelian randomisation analysis.

METHODS

We first did a fixed-effects meta-analysis of the observational association of plasma urate and risk of coronary heart disease. We then used a conventional Mendelian randomisation approach to investigate the causal relevance using a genetic instrument based on 31 urate-associated single nucleotide polymorphisms (SNPs). To account for potential pleiotropic associations of certain SNPs with risk factors other than urate, we additionally did both a multivariable Mendelian randomisation analysis, in which the genetic associations of SNPs with systolic and diastolic blood pressure, HDL cholesterol, and triglycerides were included as covariates, and an Egger Mendelian randomisation (MR-Egger) analysis to estimate a causal effect accounting for unmeasured pleiotropy.

FINDINGS

In the meta-analysis of 17 prospective observational studies (166 486 individuals; 9784 coronary heart disease events) a 1 SD higher urate concentration was associated with an odds ratio (OR) for coronary heart disease of 1·07 (95% CI 1·04-1·10). The corresponding OR estimates from the conventional, multivariable adjusted, and Egger Mendelian randomisation analysis (58 studies; 198 598 individuals; 65 877 events) were 1·18 (95% CI 1·08-1·29), 1·10 (1·00-1·22), and 1·05 (0·92-1·20), respectively, per 1 SD increment in plasma urate.

INTERPRETATION

Conventional and multivariate Mendelian randomisation analysis implicates a causal role for urate in the development of coronary heart disease, but these estimates might be inflated by hidden pleiotropy. Egger Mendelian randomisation analysis, which accounts for pleiotropy but has less statistical power, suggests there might be no causal effect. These results might help investigators to determine the priority of trials of urate lowering for the prevention of coronary heart disease compared with other potential interventions.

FUNDING

UK National Institute for Health Research, British Heart Foundation, and UK Medical Research Council.

The role and interaction of imprinted genes in human fetal growth

Identifying the genetic input for fetal growth will help to understand common, serious complications of pregnancy such as fetal growth restriction. Genomic imprinting is an epigenetic process that silences one parental allele, resulting in monoallelic expression. Imprinted genes are important in mammalian fetal growth and development. Evidence has emerged showing that genes that are paternally expressed promote fetal growth, whereas maternally expressed genes suppress growth. We have assessed whether the expression levels of key imprinted genes correlate with fetal growth parameters during pregnancy, either early in gestation, using chorionic villus samples (CVS), or in term placenta. We have found that the expression of paternally expressing insulin-like growth factor 2 (IGF2), its receptor IGF2R, and the IGF2/IGF1R ratio in CVS tissues significantly correlate with crown–rump length and birthweight, whereas term placenta expression shows no correlation. For the maternally expressing pleckstrin homology-like domain family A, member 2 (PHLDA2), there is no correlation early in pregnancy in CVS but a highly significant negative relationship in term placenta. Analysis of the control of imprinted expression of PHLDA2 gave rise to a maternally and compounded grand-maternally controlled genetic effect with a birthweight increase of 93/155 g, respectively, when one copy of the PHLDA2 promoter variant is inherited. Expression of the growth factor receptor-bound protein 10 (GRB10) in term placenta is significantly negatively correlated with head circumference. Analysis of the paternally expressing delta-like 1 homologue (DLK1) shows that the paternal transmission of type 1 diabetes protective G allele of rs941576 single nucleotide polymorphism (SNP) results in significantly reduced birth weight (−132 g). In conclusion, we have found that the expression of key imprinted genes show a strong correlation with fetal growth and that for both genetic and genomics data analyses, it is important not to overlook parent-of-origin effects.

Sixty-five common genetic variants and prediction of type 2 diabetes

We developed a 65 type 2 diabetes (T2D) variant-weighted gene score to examine the impact on T2D risk assessment in a U.K.-based consortium of prospective studies, with subjects initially free from T2D (N = 13,294; 37.3% women; mean age 58.5 [38-99] years). We compared the performance of the gene score with the phenotypically derived Framingham Offspring Study T2D risk model and then the two in combination. Over the median 10 years of follow-up, 804 participants developed T2D. The odds ratio for T2D (top vs. bottom quintiles of gene score) was 2.70 (95% CI 2.12-3.43). With a 10% false-positive rate, the genetic score alone detected 19.9% incident cases, the Framingham risk model 30.7%, and together 37.3%. The respective area under the receiver operator characteristic curves were 0.60 (95% CI 0.58-0.62), 0.75 (95% CI 0.73 to 0.77), and 0.76 (95% CI 0.75 to 0.78). The combined risk score net reclassification improvement (NRI) was 8.1% (5.0 to 11.2; P = 3.31 × 10(-7)). While BMI stratification into tertiles influenced the NRI (BMI ≤24.5 kg/m(2), 27.6% [95% CI 17.7-37.5], P = 4.82 × 10(-8); 24.5-27.5 kg/m(2), 11.6% [95% CI 5.8-17.4], P = 9.88 × 10(-5); >27.5 kg/m(2), 2.6% [95% CI -1.4 to 6.6], P = 0.20), age categories did not. The addition of the gene score to a phenotypic risk model leads to a potentially clinically important improvement in discrimination of incident T2D.

IL8 polymorphisms and overall survival in pazopanib- or sunitinib-treated patients with renal cell carcinoma

BACKGROUND

We evaluated germline single nucleotide polymorphisms (SNPs) for association with overall survival (OS) in pazopanib- or sunitinib-treated patients with advanced renal cell carcinoma (aRCC).

METHODS

The discovery analysis tested 27 SNPs within 13 genes from a phase III pazopanib trial (N=241, study 1). Suggestive associations were then pursued in two independent datasets: a phase III trial (COMPARZ) comparing pazopanib vs sunitinib (N=729, study 2) and an observational study of sunitinib-treated patients (N=89, study 3).

RESULTS

In study 1, four SNPs showed nominally significant association (P≤0.05) with OS; two of these SNPs (rs1126647, rs4073) in IL8 were associated (P≤0.05) with OS in study 2. Because rs1126647 and rs4073 were highly correlated, only rs1126647 was evaluated in study 3, which also showed association (P≤0.05). In the combined data, rs1126647 was associated with OS after conservative multiple-test adjustment (P=8.8 × 10(-5); variant vs reference allele hazard ratio 1.32, 95% confidence interval: 1.15-1.52), without evidence for heterogeneity of effects between studies or between pazopanib- and sunitinib-treated patients.

CONCLUSIONS

Variant alleles of IL8 polymorphisms are associated with poorer survival outcomes in pazopanib- or sunitinib-treated patients with aRCC. These findings provide insight in aRCC prognosis and may advance our thinking in development of new therapies.

Comparison of methods and sampling designs to test for association between rare variants and quantitative traits

Genome-wide association studies succeeded in finding genetic variants associated with various phenotypes, but a large portion of the predicted genetic contribution to many traits remains unknown. One plausible explanation is that some missing variation is due to rare variants. Latest sequencing technology facilitates the investigation of such rare variants, but their statistical analysis remains challenging. For quantitative traits, a commonly used approach is to contrast the frequency of putatively functional rare variants between subjects in the two tails of the trait distribution. The contrast is usually performed by Fisher's exact or similar test. These tests are conservative as they discard trait rank information and are most useful under the unrealistic homogeneity assumption (i.e., variants have similar effects). We propose, and investigate via simulations, various designs for resequencing studies and statistical methods that incorporate information about rank, predicted function and allow for heterogeneity of effects. We propose designs which accommodate heterogeneity by sequencing both tails and the middle of the trait and novel statistical tests for trend, for heterogeneity and for a combination of the two. The conclusions of the simulations are four fold: (1) sequencing both tails and the middle of the trait distributions is desirable when heterogeneity is suspected, (2) trend and heterogeneity statistics should be used alongside other methods, (3) using rank information improves power over Fisher's exact test when the number of rare variants is not very large and (4) due to high misclassification rates, incorporating current predictions of a variant's function does not improve power.

HMG-coenzyme A reductase inhibition, type 2 diabetes, and bodyweight: evidence from genetic analysis and randomised trials

BACKGROUND

Statins increase the risk of new-onset type 2 diabetes mellitus. We aimed to assess whether this increase in risk is a consequence of inhibition of 3-hydroxy-3-methylglutaryl-CoA reductase (HMGCR), the intended drug target.

METHODS

We used single nucleotide polymorphisms in the HMGCR gene, rs17238484 (for the main analysis) and rs12916 (for a subsidiary analysis) as proxies for HMGCR inhibition by statins. We examined associations of these variants with plasma lipid, glucose, and insulin concentrations; bodyweight; waist circumference; and prevalent and incident type 2 diabetes. Study-specific effect estimates per copy of each LDL-lowering allele were pooled by meta-analysis. These findings were compared with a meta-analysis of new-onset type 2 diabetes and bodyweight change data from randomised trials of statin drugs. The effects of statins in each randomised trial were assessed using meta-analysis.

FINDINGS

Data were available for up to 223 463 individuals from 43 genetic studies. Each additional rs17238484-G allele was associated with a mean 0·06 mmol/L (95% CI 0·05-0·07) lower LDL cholesterol and higher body weight (0·30 kg, 0·18-0·43), waist circumference (0·32 cm, 0·16-0·47), plasma insulin concentration (1·62%, 0·53-2·72), and plasma glucose concentration (0·23%, 0·02-0·44). The rs12916 SNP had similar effects on LDL cholesterol, bodyweight, and waist circumference. The rs17238484-G allele seemed to be associated with higher risk of type 2 diabetes (odds ratio [OR] per allele 1·02, 95% CI 1·00-1·05); the rs12916-T allele association was consistent (1·06, 1·03-1·09). In 129 170 individuals in randomised trials, statins lowered LDL cholesterol by 0·92 mmol/L (95% CI 0·18-1·67) at 1-year of follow-up, increased bodyweight by 0·24 kg (95% CI 0·10-0·38 in all trials; 0·33 kg, 95% CI 0·24-0·42 in placebo or standard care controlled trials and -0·15 kg, 95% CI -0·39 to 0·08 in intensive-dose vs moderate-dose trials) at a mean of 4·2 years (range 1·9-6·7) of follow-up, and increased the odds of new-onset type 2 diabetes (OR 1·12, 95% CI 1·06-1·18 in all trials; 1·11, 95% CI 1·03-1·20 in placebo or standard care controlled trials and 1·12, 95% CI 1·04-1·22 in intensive-dose vs moderate dose trials).

INTERPRETATION

The increased risk of type 2 diabetes noted with statins is at least partially explained by HMGCR inhibition.

FUNDING

The funding sources are cited at the end of the paper.

Refinement of variant selection for the LDL cholesterol genetic risk score in the diagnosis of the polygenic form of clinical familial hypercholesterolemia and replication in samples from 6 countries

BACKGROUND

Familial hypercholesterolemia (FH) is an autosomal-dominant disorder caused by mutations in 1 of 3 genes. In the 60% of patients who are mutation negative, we have recently shown that the clinical phenotype can be associated with an accumulation of common small-effect LDL cholesterol (LDL-C)-raising alleles by use of a 12-single nucleotide polymorphism (12-SNP) score. The aims of the study were to improve the selection of SNPs and replicate the results in additional samples.

METHODS

We used ROC curves to determine the optimum number of LDL-C SNPs. For replication analysis, we genotyped patients with a clinical diagnosis of FH from 6 countries for 6 LDL-C-associated alleles. We compared the weighted SNP score among patients with no confirmed mutation (FH/M-), those with a mutation (FH/M+), and controls from a UK population sample (WHII).

RESULTS

Increasing the number of SNPs to 33 did not improve the ability of the score to discriminate between FH/M- and controls, whereas sequential removal of SNPs with smaller effects/lower frequency showed that a weighted score of 6 SNPs performed as well as the 12-SNP score. Metaanalysis of the weighted 6-SNP score, on the basis of polymorphisms in CELSR2 (cadherin, EGF LAG 7-pass G-type receptor 2), APOB (apolipoprotein B), ABCG5/8 [ATP-binding cassette, sub-family G (WHITE), member 5/8], LDLR (low density lipoprotein receptor), and APOE (apolipoprotein E) loci, in the independent FH/M- cohorts showed a consistently higher score in comparison to the WHII population (P < 2.2 × 10(-16)). Modeling in individuals with a 6-SNP score in the top three-fourths of the score distribution indicated a >95% likelihood of a polygenic explanation of their increased LDL-C.

CONCLUSIONS

A 6-SNP LDL-C score consistently distinguishes FH/M- patients from healthy individuals. The hypercholesterolemia in 88% of mutation-negative patients is likely to have a polygenic basis.

Testing for non-linear causal effects using a binary genotype in a Mendelian randomization study: application to alcohol and cardiovascular traits

Background: Mendelian randomization studies have so far restricted attention to linear associations relating the genetic instrument to the exposure, and the exposure to the outcome. In some cases, however, observational data suggest a non-linear association between exposure and outcome. For example, alcohol consumption is consistently reported as having a U-shaped association with cardiovascular events. In principle, Mendelian randomization could address concerns that the apparent protective effect of light-to-moderate drinking might reflect ‘sick-quitters’ and confounding.

Methods: The Alcohol-ADH1B Consortium was established to study the causal effects of alcohol consumption on cardiovascular events and biomarkers, using the single nucleotide polymorphism rs1229984 in ADH1B as a genetic instrument. To assess non-linear causal effects in this study, we propose a novel method based on estimating local average treatment effects for discrete levels of the exposure range, then testing for a linear trend in those effects. Our method requires an assumption that the instrument has the same effect on exposure in all individuals. We conduct simulations examining the robustness of the method to violations of this assumption, and apply the method to the Alcohol-ADH1B Consortium data.

Results: Our method gave a conservative test for non-linearity under realistic violations of the key assumption. We found evidence for a non-linear causal effect of alcohol intake on several cardiovascular traits.

Conclusions: We believe our method is useful for inferring departure from linearity when only a binary instrument is available. We estimated non-linear causal effects of alcohol intake which could not have been estimated through standard instrumental variable approaches.

Association of vitamin D status with arterial blood pressure and hypertension risk: a mendelian randomisation study

BACKGROUND

Low plasma 25-hydroxyvitamin D (25[OH]D) concentration is associated with high arterial blood pressure and hypertension risk, but whether this association is causal is unknown. We used a mendelian randomisation approach to test whether 25(OH)D concentration is causally associated with blood pressure and hypertension risk.

METHODS

In this mendelian randomisation study, we generated an allele score (25[OH]D synthesis score) based on variants of genes that affect 25(OH)D synthesis or substrate availability (CYP2R1 and DHCR7), which we used as a proxy for 25(OH)D concentration. We meta-analysed data for up to 108 173 individuals from 35 studies in the D-CarDia collaboration to investigate associations between the allele score and blood pressure measurements. We complemented these analyses with previously published summary statistics from the International Consortium on Blood Pressure (ICBP), the Cohorts for Heart and Aging Research in Genomic Epidemiology (CHARGE) consortium, and the Global Blood Pressure Genetics (Global BPGen) consortium.

FINDINGS

In phenotypic analyses (up to n=49 363), increased 25(OH)D concentration was associated with decreased systolic blood pressure (β per 10% increase, -0·12 mm Hg, 95% CI -0·20 to -0·04; p=0·003) and reduced odds of hypertension (odds ratio [OR] 0·98, 95% CI 0·97-0·99; p=0·0003), but not with decreased diastolic blood pressure (β per 10% increase, -0·02 mm Hg, -0·08 to 0·03; p=0·37). In meta-analyses in which we combined data from D-CarDia and the ICBP (n=146 581, after exclusion of overlapping studies), each 25(OH)D-increasing allele of the synthesis score was associated with a change of -0·10 mm Hg in systolic blood pressure (-0·21 to -0·0001; p=0·0498) and a change of -0·08 mm Hg in diastolic blood pressure (-0·15 to -0·02; p=0·01). When D-CarDia and consortia data for hypertension were meta-analysed together (n=142 255), the synthesis score was associated with a reduced odds of hypertension (OR per allele, 0·98, 0·96-0·99; p=0·001). In instrumental variable analysis, each 10% increase in genetically instrumented 25(OH)D concentration was associated with a change of -0·29 mm Hg in diastolic blood pressure (-0·52 to -0·07; p=0·01), a change of -0·37 mm Hg in systolic blood pressure (-0·73 to 0·003; p=0·052), and an 8·1% decreased odds of hypertension (OR 0·92, 0·87-0·97; p=0·002).

INTERPRETATION

Increased plasma concentrations of 25(OH)D might reduce the risk of hypertension. This finding warrants further investigation in an independent, similarly powered study.

FUNDING

British Heart Foundation, UK Medical Research Council, and Academy of Finland.

Population genomics of cardiometabolic traits: design of the University College London-London School of Hygiene and Tropical Medicine-Edinburgh-Bristol (UCLEB) Consortium

Substantial advances have been made in identifying common genetic variants influencing cardiometabolic traits and disease outcomes through genome wide association studies. Nevertheless, gaps in knowledge remain and new questions have arisen regarding the population relevance, mechanisms, and applications for healthcare. Using a new high-resolution custom single nucleotide polymorphism (SNP) array (Metabochip) incorporating dense coverage of genomic regions linked to cardiometabolic disease, the University College-London School-Edinburgh-Bristol (UCLEB) consortium of highly-phenotyped population-based prospective studies, aims to: (1) fine map functionally relevant SNPs; (2) precisely estimate individual absolute and population attributable risks based on individual SNPs and their combination; (3) investigate mechanisms leading to altered risk factor profiles and CVD events; and (4) use Mendelian randomisation to undertake studies of the causal role in CVD of a range of cardiovascular biomarkers to inform public health policy and help develop new preventative therapies.

The impact of targeting all elderly persons in England and Wales for yearly influenza vaccination: excess mortality due to pneumonia or influenza and time trend study

OBJECTIVE

To investigate the impact on mortality due to pneumonia or influenza of the change from risk-based to age group-based targeting of the elderly for yearly influenza vaccination in England and Wales.

DESIGN

Excess mortality estimated using time series of deaths registered to pneumonia or influenza, accounting for seasonality, trend and artefacts. Non-excess mortality plotted as proxy for long-term trend in mortality.

SETTING

England and Wales.

PARTICIPANTS

Persons aged 65-74 and 75+ years whose deaths were registered to underlying pneumonia or influenza between 1975/1976 and 2004/2005.

OUTCOME MEASURES

Multiplicative effect on average excess pneumonia and influenza deaths each winter in the 4-6 winters since age group-based targeting of vaccination was introduced (in persons aged 75+ years from 1998/1999; in persons aged 65+ years from 2000/2001), estimated using multivariable regression adjusted for temperature, antigenic drift and vaccine mismatch, and stratified by dominant circulating influenza subtype. Trend in baseline weekly pneumonia and influenza death rates.

RESULTS

There is a suggestion of lower average excess mortality in the six winters after age group-based targeting began compared to before, but the CI for the 65-74 years age group includes no difference. Trend in baseline pneumonia and influenza mortality shows an apparent downward turning point around 2000 for the 65-74 years age group and from the mid-1990s in the 75+ years age group.

CONCLUSIONS

There is weakly supportive evidence that the marked increases in vaccine coverage accompanying the switch from risk-based to age group-based targeting of the elderly for yearly influenza vaccination in England and Wales were associated with lower levels of pneumonia and influenza mortality in older people in the first 6 years after age group-based targeting began. The possible impact of these policy changes is observed as weak evidence for lower average excess mortality as well as a turning point in baseline mortality coincident with the changes.

A comparison of Bayesian and frequentist approaches to incorporating external information for the prediction of prostate cancer risk

We present the most comprehensive comparison to date of the predictive benefit of genetics in addition to currently used clinical variables, using genotype data for 33 single-nucleotide polymorphisms (SNPs) in 1,547 Caucasian men from the placebo arm of the REduction by DUtasteride of prostate Cancer Events (REDUCE®) trial. Moreover, we conducted a detailed comparison of three techniques for incorporating genetics into clinical risk prediction. The first method was a standard logistic regression model, which included separate terms for the clinical covariates and for each of the genetic markers. This approach ignores a substantial amount of external information concerning effect sizes for these Genome Wide Association Study (GWAS)-replicated SNPs. The second and third methods investigated two possible approaches to incorporating meta-analysed external SNP effect estimates - one via a weighted PCa 'risk' score based solely on the meta analysis estimates, and the other incorporating both the current and prior data via informative priors in a Bayesian logistic regression model. All methods demonstrated a slight improvement in predictive performance upon incorporation of genetics. The two methods that incorporated external information showed the greatest receiver-operating-characteristic AUCs increase from 0.61 to 0.64. The value of our methods comparison is likely to lie in observations of performance similarities, rather than difference, between three approaches of very different resource requirements. The two methods that included external information performed best, but only marginally despite substantial differences in complexity.

The benefits of using genetic information to design prevention trials

Clinical trials for preventative therapies are complex and costly endeavors focused on individuals likely to develop disease in a short time frame, randomizing them to treatment groups, and following them over time. In such trials, statistical power is governed by the rate of disease events in each group and cost is determined by randomization, treatment, and follow-up. Strategies that increase the rate of disease events by enrolling individuals with high risk of disease can significantly reduce study size, duration, and cost. Comprehensive study of common, complex diseases has resulted in a growing list of robustly associated genetic markers. Here, we evaluate the utility--in terms of trial size, duration, and cost--of enriching prevention trial samples by combining clinical information with genetic risk scores to identify individuals at greater risk of disease. We also describe a framework for utilizing genetic risk scores in these trials and evaluating the associated cost and time savings. With type 1 diabetes (T1D), type 2 diabetes (T2D), myocardial infarction (MI), and advanced age-related macular degeneration (AMD) as examples, we illustrate the potential and limitations of using genetic data for prevention trial design. We illustrate settings where incorporating genetic information could reduce trial cost or duration considerably, as well as settings where potential savings are negligible. Results are strongly dependent on the genetic architecture of the disease, but we also show that these benefits should increase as the list of robustly associated markers for each disease grows and as large samples of genotyped individuals become available.

Association between the chromosome 9p21 locus and angiographic coronary artery disease burden: a collaborative meta-analysis

OBJECTIVES

This study sought to ascertain the relationship of 9p21 locus with: 1) angiographic coronary artery disease (CAD) burden; and 2) myocardial infarction (MI) in individuals with underlying CAD.

BACKGROUND

Chromosome 9p21 variants have been robustly associated with coronary heart disease, but questions remain on the mechanism of risk, specifically whether the locus contributes to coronary atheroma burden or plaque instability.

METHODS

We established a collaboration of 21 studies consisting of 33,673 subjects with information on both CAD (clinical or angiographic) and MI status along with 9p21 genotype. Tabular data are provided for each cohort on the presence and burden of angiographic CAD, MI cases with underlying CAD, and the diabetic status of all subjects.

RESULTS

We first confirmed an association between 9p21 and CAD with angiographically defined cases and control subjects (pooled odds ratio [OR]: 1.31, 95% confidence interval [CI]: 1.20 to 1.43). Among subjects with angiographic CAD (n = 20,987), random-effects model identified an association with multivessel CAD, compared with those with single-vessel disease (OR: 1.10, 95% CI: 1.04 to 1.17)/copy of risk allele). Genotypic models showed an OR of 1.15, 95% CI: 1.04 to 1.26 for heterozygous carrier and OR: 1.23, 95% CI: 1.08 to 1.39 for homozygous carrier. Finally, there was no significant association between 9p21 and prevalent MI when both cases (n = 17,791) and control subjects (n = 15,882) had underlying CAD (OR: 0.99, 95% CI: 0.95 to 1.03)/risk allele.

CONCLUSIONS

The 9p21 locus shows convincing association with greater burden of CAD but not with MI in the presence of underlying CAD. This adds further weight to the hypothesis that 9p21 locus primarily mediates an atherosclerotic phenotype.

Causal relationship between obesity and vitamin D status: bi-directional Mendelian randomization analysis of multiple cohorts

BACKGROUND

Obesity is associated with vitamin D deficiency, and both are areas of active public health concern. We explored the causality and direction of the relationship between body mass index (BMI) and 25-hydroxyvitamin D [25(OH)D] using genetic markers as instrumental variables (IVs) in bi-directional Mendelian randomization (MR) analysis.

METHODS AND FINDINGS

We used information from 21 adult cohorts (up to 42,024 participants) with 12 BMI-related SNPs (combined in an allelic score) to produce an instrument for BMI and four SNPs associated with 25(OH)D (combined in two allelic scores, separately for genes encoding its synthesis or metabolism) as an instrument for vitamin D. Regression estimates for the IVs (allele scores) were generated within-study and pooled by meta-analysis to generate summary effects. Associations between vitamin D scores and BMI were confirmed in the Genetic Investigation of Anthropometric Traits (GIANT) consortium (n = 123,864). Each 1 kg/m(2) higher BMI was associated with 1.15% lower 25(OH)D (p = 6.52×10⁻²⁷). The BMI allele score was associated both with BMI (p = 6.30×10⁻⁶²) and 25(OH)D (-0.06% [95% CI -0.10 to -0.02], p = 0.004) in the cohorts that underwent meta-analysis. The two vitamin D allele scores were strongly associated with 25(OH)D (p≤8.07×10⁻⁵⁷ for both scores) but not with BMI (synthesis score, p = 0.88; metabolism score, p = 0.08) in the meta-analysis. A 10% higher genetically instrumented BMI was associated with 4.2% lower 25(OH)D concentrations (IV ratio: -4.2 [95% CI -7.1 to -1.3], p = 0.005). No association was seen for genetically instrumented 25(OH)D with BMI, a finding that was confirmed using data from the GIANT consortium (p≥0.57 for both vitamin D scores).

CONCLUSIONS

On the basis of a bi-directional genetic approach that limits confounding, our study suggests that a higher BMI leads to lower 25(OH)D, while any effects of lower 25(OH)D increasing BMI are likely to be small. Population level interventions to reduce BMI are expected to decrease the prevalence of vitamin D deficiency.

Predictive models of choroidal neovascularization and geographic atrophy incidence applied to clinical trial design

PURPOSE

To develop comprehensive predictive models for choroidal neovascularization (CNV) and geographic atrophy (GA) incidence within 3 years that can be applied realistically to clinical practice.

DESIGN

Retrospective evaluation of data from a longitudinal study to develop and validate predictive models of CNV and GA.

METHODS

The predictive performance of clinical, environmental, demographic, and genetic risk factors was explored in regression models, using data from both eyes of 2011 subjects from the Age-Related Eye Disease Study (AREDS). The performance of predictive models was compared using 10-fold cross-validated receiver operating characteristic curves in the training data, followed by comparisons in an independent validation dataset (1410 AREDS subjects). Bayesian trial simulations were used to compare the usefulness of predictive models to screen patients for inclusion in prevention clinical trials.

RESULTS

Logistic regression models that included clinical, demographic, and environmental factors had better predictive performance for 3-year CNV and GA incidence (area under the receiver operating characteristic curve of 0.87 and 0.89, respectively), compared with simple clinical criteria (AREDS simplified severity scale). Although genetic markers were associated significantly with 3-year CNV (CFH: Y402H; ARMS2: A69S) and GA incidence (CFH: Y402H), the inclusion of genetic factors in the models provided only marginal improvements in predictive performance.

CONCLUSIONS

The logistic regression models combine good predictive performance with greater flexibility to optimize clinical trial design compared with simple clinical models (AREDS simplified severity scale). The benefit of including genetic factors to screen patients for recruitment to CNV prevention studies is marginal and is dependent on individual clinical trial economics.

Comparison of statistical tests for association between rare variants and binary traits

Genome-wide association studies have found thousands of common genetic variants associated with a wide variety of diseases and other complex traits. However, a large portion of the predicted genetic contribution to many traits remains unknown. One plausible explanation is that some of the missing variation is due to the effects of rare variants. Nonetheless, the statistical analysis of rare variants is challenging. A commonly used method is to contrast, within the same region (gene), the frequency of minor alleles at rare variants between cases and controls. However, this strategy is most useful under the assumption that the tested variants have similar effects. We previously proposed a method that can accommodate heterogeneous effects in the analysis of quantitative traits. Here we extend this method to include binary traits that can accommodate covariates. We use simulations for a variety of causal and covariate impact scenarios to compare the performance of the proposed method to standard logistic regression, C-alpha, SKAT, and EREC. We found that i) logistic regression methods perform well when the heterogeneity of the effects is not extreme and ii) SKAT and EREC have good performance under all tested scenarios but they can be computationally intensive. Consequently, it would be more computationally desirable to use a two-step strategy by (i) selecting promising genes by faster methods and ii) analyzing selected genes using SKAT/EREC. To select promising genes one can use (1) regression methods when effect heterogeneity is assumed to be low and the covariates explain a non-negligible part of trait variability, (2) C-alpha when heterogeneity is assumed to be large and covariates explain a small fraction of trait’s variability and (3) the proposed trend and heterogeneity test when the heterogeneity is assumed to be non-trivial and the covariates explain a large fraction of trait variability.

Evaluation of genetic markers as instruments for Mendelian randomization studies on vitamin D

BACKGROUND

Mendelian randomization (MR) studies use genetic variants mimicking the influence of a modifiable exposure to assess and quantify a causal association with an outcome, with an aim to avoid problems with confounding and reverse causality affecting other types of observational studies.

AIM

We evaluated genetic markers that index differences in 25-hydroxyvitamin D (25(OH)D) as instruments for MR studies on vitamin D.

METHODS AND FINDINGS

We used data from up-to 6,877 participants in the 1958 British birth cohort with information on genetic markers and 25(OH)D. As potential instruments, we selected 20 single nucleotide polymorphisms (SNP) which are located in the vitamin D metabolism pathway or affect skin pigmentation/tanning, including 4 SNPs from genome-wide association (GWA) meta-analyses on 25(OH)D. We analyzed SNP associations with 25(OH)D and evaluated the use of allele scores dividing genes to those affecting 25(OH)D synthesis (DHCR7, CYP2R1) and metabolism (GC, CYP24A1, CYP27B1). In addition to the GWA SNPs, only two SNPs (CYP27B1, OCA2) showed evidence for association with 25(OH)D, with the OCA2 association abolished after lifestyle adjustment. Per allele differences varied between -0.02 and -0.08 nmol/L (P≤0.02 for all), with a 6.1 nmol/L and a 10.2 nmol/L difference in 25(OH)D between individuals with highest compared lowest number of risk alleles in synthesis and metabolism allele scores, respectively. Individual SNPs but not allele scores showed associations with lifestyle factors. An exception was geographical region which was associated with synthesis score. Illustrative power calculations (80% power, 5% alpha) suggest that approximately 80,000 participants are required to establish a causal effect of vitamin D on blood pressure using the synthesis allele score.

CONCLUSIONS

Combining SNPs into allele scores provides a more powerful instrument for MR analysis than a single SNP in isolation. Population stratification and the potential for pleiotropic effects need to be considered in MR studies on vitamin D.

An abundance of rare functional variants in 202 drug target genes sequenced in 14,002 people

Rare genetic variants contribute to complex disease risk; however, the abundance of rare variants in human populations remains unknown. We explored this spectrum of variation by sequencing 202 genes encoding drug targets in 14,002 individuals. We find rare variants are abundant (1 every 17 bases) and geographically localized, so that even with large sample sizes, rare variant catalogs will be largely incomplete. We used the observed patterns of variation to estimate population growth parameters, the proportion of variants in a given frequency class that are putatively deleterious, and mutation rates for each gene. We conclude that because of rapid population growth and weak purifying selection, human populations harbor an abundance of rare variants, many of which are deleterious and have relevance to understanding disease risk.

Potential impact of adding genetic markers to clinical parameters in predicting prostate biopsy outcomes in men following an initial negative biopsy: findings from the REDUCE trial

BACKGROUND

Several germline single nucleotide polymorphisms (SNPs) have been consistently associated with prostate cancer (PCa) risk.

OBJECTIVE

To determine whether there is an improvement in PCa risk prediction by adding these SNPs to existing predictors of PCa.

DESIGN, SETTING, AND PARTICIPANTS

Subjects included men in the placebo arm of the randomized Reduction by Dutasteride of Prostate Cancer Events (REDUCE) trial in whom germline DNA was available. All men had an initial negative prostate biopsy and underwent study-mandated biopsies at 2 yr and 4 yr. Predictive performance of baseline clinical parameters and/or a genetic score based on 33 established PCa risk-associated SNPs was evaluated.

OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS

Area under the receiver operating characteristic curves (AUC) were used to compare different models with different predictors. Net reclassification improvement (NRI) and decision curve analysis (DCA) were used to assess changes in risk prediction by adding genetic markers.

RESULTS AND LIMITATIONS

Among 1654 men, genetic score was a significant predictor of positive biopsy, even after adjusting for known clinical variables and family history (p = 3.41 × 10(-8)). The AUC for the genetic score exceeded that of any other PCa predictor at 0.59. Adding the genetic score to the best clinical model improved the AUC from 0.62 to 0.66 (p<0.001), reclassified PCa risk in 33% of men (NRI: 0.10; p=0.002), resulted in higher net benefit from DCA, and decreased the number of biopsies needed to detect the same number of PCa instances. The benefit of adding the genetic score was greatest among men at intermediate risk (25th percentile to 75th percentile). Similar results were found for high-grade (Gleason score ≥ 7) PCa. A major limitation of this study was its focus on white patients only.

CONCLUSIONS

Adding genetic markers to current clinical parameters may improve PCa risk prediction. The improvement is modest but may be helpful for better determining the need for repeat prostate biopsy. The clinical impact of these results requires further study.

Deep resequencing unveils genetic architecture of ADIPOQ and identifies a novel low-frequency variant strongly associated with adiponectin variation

Increased adiponectin levels have been shown to be associated with a lower risk of type 2 diabetes. To understand the relations between genetic variation at the adiponectin-encoding gene, ADIPOQ, and adiponectin levels, and subsequently its role in disease, we conducted a deep resequencing experiment of ADIPOQ in 14,002 subjects, including 12,514 Europeans, 594 African Americans, and 567 Indian Asians. We identified 296 single nucleotide polymorphisms (SNPs), including 30 amino acid changes, and carried out association analyses in a subset of 3,665 subjects from two independent studies. We confirmed multiple genome-wide association study findings and identified a novel association between a low-frequency SNP (rs17366653) and adiponectin levels (P = 2.2E-17). We show that seven SNPs exert independent effects on adiponectin levels. Together, they explained 6% of adiponectin variation in our samples. We subsequently assessed association between these SNPs and type 2 diabetes in the Genetics of Diabetes Audit and Research in Tayside Scotland (GO-DARTS) study, comprised of 5,145 case and 6,374 control subjects. No evidence of association with type 2 diabetes was found, but we were also unable to exclude the possibility of substantial effects (e.g., odds ratio 95% CI for rs7366653 [0.91-1.58]). Further investigation by large-scale and well-powered Mendelian randomization studies is warranted.

Maternal inheritance of a promoter variant in the imprinted PHLDA2 gene significantly increases birth weight

Birth weight is an important indicator of both perinatal and adult health, but little is known about the genetic factors contributing to its variability. Intrauterine growth restriction is a leading cause of perinatal morbidity and mortality and is also associated with adult disease. A significant correlation has been reported between lower birth weight and increased expression of the maternal PHLDA2 allele in term placenta (the normal imprinting pattern was maintained). However, a mechanism that explains the transcriptional regulation of PHLDA2 on in utero growth has yet to be described. In this study, we sequenced the PHLDA2 promoter region in 263 fetal DNA samples to identify polymorphic variants. We used a luciferase reporter assay to identify in the PHLDA2 promoter a 15 bp repeat sequence (RS1) variant that significantly reduces PHLDA2-promoter efficiency. RS1 genotyping was then performed in three independent white European normal birth cohorts. Meta-analysis of all three (total n = 9,433) showed that maternal inheritance of RS1 resulted in a significant 93 g increase in birth weight (p = 0.01; 95% confidence interval [CI] = 22-163). Moreover, when the mother was homozygous for RS1, the influence on birth weight was 155 g (p = 0.04; 95% CI = 9-300), which is a similar magnitude to the reduction in birth weight caused by maternal smoking.

Deep sequencing of the LRRK2 gene in 14,002 individuals reveals evidence of purifying selection and independent origin of the p.Arg1628Pro mutation in Europe

Genetic variation in LRRK2 predisposes to Parkinson disease (PD), which underpins its development as a therapeutic target. Here, we aimed to identify novel genotype-phenotype associations that might support developing LRRK2 therapies for other conditions. We sequenced the 51 exons of LRRK2 in cases comprising 12 common diseases (n = 9,582), and in 4,420 population controls. We identified 739 single-nucleotide variants, 62% of which were observed in only one person, including 316 novel exonic variants. We found evidence of purifying selection for the LRRK2 gene and a trend suggesting that this is more pronounced in the central (ROC-COR-kinase) core protein domains of LRRK2 than the flanking domains. Population genetic analyses revealed that LRRK2 is not especially polymorphic or differentiated in comparison to 201 other drug target genes. Among Europeans, we identified 17 carriers (0.13%) of pathogenic LRRK2 mutations that were not significantly enriched within any disease or in those reporting a family history of PD. Analysis of pathogenic mutations within Europe reveals that the p.Arg1628Pro (c4883G>C) mutation arose independently in Europe and Asia. Taken together, these findings demonstrate how targeted deep sequencing can help to reveal fundamental characteristics of clinically important loci.

The interleukin-6 receptor as a target for prevention of coronary heart disease: a mendelian randomisation analysis

BACKGROUND

A high circulating concentration of interleukin 6 is associated with increased risk of coronary heart disease. Blockade of the interleukin-6 receptor (IL6R) with a monoclonal antibody (tocilizumab) licensed for treatment of rheumatoid arthritis reduces systemic and articular inflammation. However, whether IL6R blockade also reduces risk of coronary heart disease is unknown.

METHODS

Applying the mendelian randomisation principle, we used single nucleotide polymorphisms (SNPs) in the gene IL6R to evaluate the likely efficacy and safety of IL6R inhibition for primary prevention of coronary heart disease. We compared genetic findings with the effects of tocilizumab reported in randomised trials in patients with rheumatoid arthritis.

FINDINGS

In 40 studies including up to 133,449 individuals, an IL6R SNP (rs7529229) marking a non-synonymous IL6R variant (rs8192284; p.Asp358Ala) was associated with increased circulating log interleukin-6 concentration (increase per allele 9·45%, 95% CI 8·34-10·57) as well as reduced C-reactive protein (decrease per allele 8·35%, 95% CI 7·31-9·38) and fibrinogen concentrations (decrease per allele 0·85%, 95% CI 0·60-1·10). This pattern of effects was consistent with IL6R blockade from infusions of tocilizumab (4-8 mg/kg every 4 weeks) in patients with rheumatoid arthritis studied in randomised trials. In 25,458 coronary heart disease cases and 100,740 controls, the IL6R rs7529229 SNP was associated with a decreased odds of coronary heart disease events (per allele odds ratio 0·95, 95% CI 0·93-0·97, p=1·53×10(-5)).

INTERPRETATION

On the basis of genetic evidence in human beings, IL6R signalling seems to have a causal role in development of coronary heart disease. IL6R blockade could provide a novel therapeutic approach to prevention of coronary heart disease that warrants testing in suitably powered randomised trials. Genetic studies in populations could be used more widely to help to validate and prioritise novel drug targets or to repurpose existing agents and targets for new therapeutic uses.

FUNDING

UK Medical Research Council; British Heart Foundation; Rosetrees Trust; US National Heart, Lung, and Blood Institute; Du Pont Pharma; Chest, Heart and Stroke Scotland; Wellcome Trust; Coronary Thrombosis Trust; Northwick Park Institute for Medical Research; UCLH/UCL Comprehensive Medical Research Centre; US National Institute on Aging; Academy of Finland; Netherlands Organisation for Health Research and Development; SANCO; Dutch Ministry of Public Health, Welfare and Sports; World Cancer Research Fund; Agentschap NL; European Commission; Swedish Heart-Lung Foundation; Swedish Research Council; Strategic Cardiovascular Programme of the Karolinska Institutet; Stockholm County Council; US National Institute of Neurological Disorders and Stroke; MedStar Health Research Institute; GlaxoSmithKline; Dutch Kidney Foundation; US National Institutes of Health; Netherlands Interuniversity Cardiology Institute of the Netherlands; Diabetes UK; European Union Seventh Framework Programme; National Institute for Healthy Ageing; Cancer Research UK; MacArthur Foundation.

How informative is a negative finding in a small pharmacogenetic study?

Many pharmacogenetic studies fail to yield any statistically significant associations. Such negative findings may be due to the absence of, or inadequate statistical power to test for, an effect at the genetic variants tested. In many instances, sample sizes are small, making it unclear how to interpret the absence of statistically significant findings. We demonstrate that the amount of information that can be drawn from a negative study is improved by incorporating statistical power and the added context of well-validated pharmacogenetic effects into the interpretation process. This approach permits clearer inferences to be made about the possible range of genetic effects that may be present in, or are likely absent from, small drug studies.

Performance of genotype imputation for rare variants identified in exons and flanking regions of genes

Genotype imputation has the potential to assess human genetic variation at a lower cost than assaying the variants using laboratory techniques. The performance of imputation for rare variants has not been comprehensively studied. We utilized 8865 human samples with high depth resequencing data for the exons and flanking regions of 202 genes and Genome-Wide Association Study (GWAS) data to characterize the performance of genotype imputation for rare variants. We evaluated reference sets ranging from 100 to 3713 subjects for imputing into samples typed for the Affymetrix (500K and 6.0) and Illumina 550K GWAS panels. The proportion of variants that could be well imputed (true r(2)>0.7) with a reference panel of 3713 individuals was: 31% (Illumina 550K) or 25% (Affymetrix 500K) with MAF (Minor Allele Frequency) less than or equal 0.001, 48% or 35% with 0.0010.05. The performance for common SNPs (MAF>0.05) within exons and flanking regions is comparable to imputation of more uniformly distributed SNPs. The performance for rare SNPs (0.01 Collapse

Key Words Collapse

MESH Headings

• Exons/genetics
• Gene Frequency
• Genes/genetics
• Genetic Variation
• Genome-Wide Association Study
• Genotype
• Humans
• Polymorphism, Single Nucleotide/genetics

Collapse

Grants

• R01 HG004960 NHGRI NIH HHS
• R01 HG005701 NHGRI NIH HHS
• R01HG005701 NHGRI NIH HHS
• R01HG004960 NHGRI NIH HHS

Collapse

Effect modification by population dietary folate on the association between MTHFR genotype, homocysteine, and stroke risk: a meta-analysis of genetic studies and randomised trials

BACKGROUND

The MTHFR 677C→T polymorphism has been associated with raised homocysteine concentration and increased risk of stroke. A previous overview showed that the effects were greatest in regions with low dietary folate consumption, but differentiation between the effect of folate and small-study bias was difficult. A meta-analysis of randomised trials of homocysteine-lowering interventions showed no reduction in coronary heart disease events or stroke, but the trials were generally set in populations with high folate consumption. We aimed to reduce the effect of small-study bias and investigate whether folate status modifies the association between MTHFR 677C→T and stroke in a genetic analysis and meta-analysis of randomised controlled trials.

METHODS

We established a collaboration of genetic studies consisting of 237 datasets including 59,995 individuals with data for homocysteine and 20,885 stroke events. We compared the genetic findings with a meta-analysis of 13 randomised trials of homocysteine-lowering treatments and stroke risk (45,549 individuals, 2314 stroke events, 269 transient ischaemic attacks).

FINDINGS

The effect of the MTHFR 677C→T variant on homocysteine concentration was larger in low folate regions (Asia; difference between individuals with TT versus CC genotype, 3·12 μmol/L, 95% CI 2·23 to 4·01) than in areas with folate fortification (America, Australia, and New Zealand, high; 0·13 μmol/L, -0·85 to 1·11). The odds ratio (OR) for stroke was also higher in Asia (1·68, 95% CI 1·44 to 1·97) than in America, Australia, and New Zealand, high (1·03, 0·84 to 1·25). Most randomised trials took place in regions with high or increasing population folate concentrations. The summary relative risk (RR) of stroke in trials of homocysteine-lowering interventions (0·94, 95% CI 0·85 to 1·04) was similar to that predicted for the same extent of homocysteine reduction in large genetic studies in populations with similar folate status (predicted RR 1·00, 95% CI 0·90 to 1·11). Although the predicted effect of homocysteine reduction from large genetic studies in low folate regions (Asia) was larger (RR 0·78, 95% CI 0·68 to 0·90), no trial has evaluated the effect of lowering of homocysteine on stroke risk exclusively in a low folate region.

INTERPRETATION

In regions with increasing levels or established policies of population folate supplementation, evidence from genetic studies and randomised trials is concordant in suggesting an absence of benefit from lowering of homocysteine for prevention of stroke. Further large-scale genetic studies of the association between MTHFR 677C→T and stroke in low folate settings are needed to distinguish effect modification by folate from small-study bias. If future randomised trials of homocysteine-lowering interventions for stroke prevention are undertaken, they should take place in regions with low folate consumption.

FUNDING

Full funding sources listed at end of paper (see Acknowledgments).

Comparative analysis of genome-wide association studies signals for lipids, diabetes, and coronary heart disease: Cardiovascular Biomarker Genetics Collaboration

Aims

To evaluate the associations of emergent genome-wide-association study-derived coronary heart disease (CHD)-associated single nucleotide polymorphisms (SNPs) with established and emerging risk factors, and the association of genome-wide-association study-derived lipid-associated SNPs with other risk factors and CHD events.

Methods and results

Using two case–control studies, three cross-sectional, and seven prospective studies with up to 25 000 individuals and 5794 CHD events we evaluated associations of 34 genome-wide-association study-identified SNPs with CHD risk and 16 CHD-associated risk factors or biomarkers. The Ch9p21 SNPs rs1333049 (OR 1.17; 95% confidence limits 1.11–1.24) and rs10757274 (OR 1.17; 1.09–1.26), MIA3 rs17465637 (OR 1.10; 1.04–1.15), Ch2q36 rs2943634 (OR 1.08; 1.03–1.14), APC rs383830 (OR 1.10; 1.02, 1.18), MTHFD1L rs6922269 (OR 1.10; 1.03, 1.16), CXCL12 rs501120 (OR 1.12; 1.04, 1.20), and SMAD3 rs17228212 (OR 1.11; 1.05, 1.17) were all associated with CHD risk, but not with the CHD biomarkers and risk factors measured. Among the 20 blood lipid-related SNPs, LPL rs17411031 was associated with a lower risk of CHD (OR 0.91; 0.84–0.97), an increase in Apolipoprotein AI and HDL-cholesterol, and reduced triglycerides. SORT1 rs599839 was associated with CHD risk (OR 1.20; 1.15–1.26) as well as total- and LDL-cholesterol, and apolipoprotein B. ANGPTL3 rs12042319 was associated with CHD risk (OR 1.11; 1.03, 1.19), total- and LDL-cholesterol, triglycerides, and interleukin-6.

Conclusion

Several SNPs predicting CHD events appear to involve pathways not currently indexed by the established or emerging risk factors; others involved changes in blood lipids including triglycerides or HDL-cholesterol as well as LDL-cholesterol. The overlapping association of SNPs with multiple risk factors and biomarkers supports the existence of shared points of regulation for these phenotypes.

Sequencing of Lp-PLA2-encoding PLA2G7 gene in 2000 Europeans reveals several rare loss-of-function mutations

Elevated plasma levels of lipoprotein-associated phospholipase A₂ (Lp-PLA2) activity have been shown to be associated with increased risk of coronary heart disease and an inhibitor of this enzyme is under development for the treatment of that condition. A Val279Phe null allele in this gene, that may influence patient eligibility for treatment, is relatively common in East Asians but has not been observed in Europeans. We investigated the existence and functional effects of low frequency alleles in a Western European population by re-sequencing the exons of PLA2G7 in 2000 samples. In all, 19 non-synonymous single-nucleotide polymorphisms (nsSNPs) were found, 14 in fewer than four subjects (minor allele frequency <0.1%). Lp-PLA2 activity was significantly lower in rare nsSNP carriers compared with non-carriers (167.8±63.2 vs 204.6±41.8, P=0.01) and seven variants had enzyme activities consistent with a null allele. The cumulative frequency of these null alleles was 0.25%, so <1 in 10 000 Europeans would be expected to be homozygous, and thus not potentially benefit from treatment with an Lp-PLA2 inhibitor.

Pazopanib efficacy in renal cell carcinoma: evidence for predictive genetic markers in angiogenesis-related and exposure-related genes

PURPOSE

Pazopanib, an oral angiogenesis inhibitor, is approved for the treatment of advanced renal cell carcinoma (RCC). Response to pazopanib monotherapy varies between patients, and no validated biomarkers predictive of treatment outcome have been identified. We tested the hypothesis that this variability is partially dependent on germline genetic variants that may affect pazopanib exposure or angiogenesis pathways.

PATIENTS AND METHODS

Twenty-seven functional polymorphisms within 13 genes were evaluated in 397 patients with RCC. Genetic association with progression-free survival (PFS) and objective response rate (RR) was analyzed using the Cox proportional hazards model and proportional odds model, respectively.

RESULTS

Three polymorphisms in IL8 and HIF1A and five polymorphisms in HIF1A, NR1I2, and VEGFA showed nominally significant association (P ≤ .05) with PFS and RR, respectively. Compared with the wild-type AA genotype (median PFS, 48 weeks), the IL8 2767TT variant genotype showed inferior PFS (27 weeks, P = .009). The HIF1A 1790AG genotype was associated with inferior PFS and reduced RR, compared with the wild-type GG genotype (median PFS, 20 v 44 weeks; P = .03; RR, 30% v 43%, P = .02). Reductions in RR were detected for the NR1I2 -25385TT genotype, compared with the wild-type CC genotype (37% v 50%, P = .03), and for the VEGFA -1498CC genotype compared with the TT genotypes (33% v 51%).

CONCLUSION

Germline variants in angiogenesis- and exposure-related genes may predict treatment response to pazopanib monotherapy in patients with RCC. If validated, these markers may explain why certain patients fail antiangiogenesis therapy and they may support the use of alternative strategies to circumvent this issue.

High-dose vitamin D supplements are not associated with linear growth in a large Finnish cohort

High vitamin D intake in childhood has been suggested to have an adverse influence on linear growth. In Finland, in the mid-1960s the official recommendation for infant vitamin D supplementation was 2000 IU/d (50 μg/d). We investigated whether high-dose vitamin D supplementation in infancy was associated with subsequent growth in height. We used data from a prospective population-based birth cohort study including all children due to be born in the 2 northernmost provinces in Finland in 1966 (12,058 live-births, coverage 96%). Information on each participant's height was collected at birth and ages 1, 14, and 31 y, as were possible confounding factors (data for analyses available from 10,060 singletons). Information on the frequency and dose of vitamin D supplementation was collected in 1967 when participants were 1 y of age. A weak association was found between frequency of vitamin D supplementation with greater height at age 1 y (P = 0.005), which was explained by birth characteristics and maternal and social factors (adjusted P = 0.34). Neither frequency nor dose of vitamin D supplementation was associated with height at 14 or 31 y (P > 0.13). To conclude, contrary to proposed evidence suggesting that vitamin D has a negative influence on growth rate at a dosage of ~2000 IU/d, supplementation at this level in the Northern Finland Birth Cohort was not associated with reduced height at any age studied.

HLA-DQA1*02:01 is a major risk factor for lapatinib-induced hepatotoxicity in women with advanced breast cancer

PURPOSE

Hepatobiliary adverse events (AEs) have been observed in a small proportion of patients with metastatic breast cancer (MBC) treated with lapatinib. This study sought to identify gene variants associated with lapatinib-induced ALT elevation and hepatobiliary AEs.

PATIENTS AND METHODS

A two-stage pharmacogenetic investigation of ALT elevation was conducted in lapatinib-treated patients with MBC. Exploratory marker identification evaluated classical HLA alleles, candidate genes, and genome-wide screening in 37 cases with ALT greater than 3 times the upper limit of normal (ULN) and 286 controls with ALT ≤ 1× ULN, selected from 901 lapatinib-treated patients in 12 trials. Markers achieving prespecified association thresholds were progressed to an independent confirmatory data set of 24 ALT cases and 155 controls selected from a subsequent trial of 374 lapatinib-treated patients.

RESULTS

Of 58 variants associated with ALT elevation in the exploratory data set, four exceeded the prespecified significance threshold in the confirmatory analysis. These variants reside in the same MHC genomic locus and include HLA-DQA1*02:01. In the confirmatory study, DQA1*02:01 allele carriage was present in 71% of ALT cases and in 21% of controls (P < .001; odds ratio, 9.0; 95% CI, 3.2 to 27.4). As a predictor of liver safety risk in ALT cases versus noncases, DQA1*02:01 had negative and positive predictive values of 0.97 (95% CI, 0.95 to 0.99) and 0.17 (95% CI 0.10 to 0.26), respectively.

CONCLUSION

These results support a role for immune mechanisms in lapatinib-induced hepatotoxicity. Further work is required to determine whether testing for DQA1*02:01 allele carriage is clinically useful in managing liver safety risk during lapatinib treatment.

A variance components factor model for genetic association studies: a Bayesian analysis

Studies of gene-trait associations for complex diseases often involve multiple traits that may vary by genotype groups or patterns. Such traits are usually manifestations of lower-dimensional latent factors or disease syndromes. We illustrate the use of a variance components factor (VCF) model to model the association between multiple traits and genotype groups as well as any other existing patient-level covariates. This model characterizes the correlations between traits as underlying latent factors that can be used in clinical decision-making. We apply it within the Bayesian framework and provide a straightforward implementation using the WinBUGS software. The VCF model is illustrated with simulated data and an example that comprises changes in plasma lipid measurements of patients who were treated with statins to lower low-density lipoprotein cholesterol, and polymorphisms from the apolipoprotein-E gene. The simulation shows that this model clearly characterizes existing multiple trait manifestations across genotype groups where individuals' group assignments are fully observed or can be deduced from the observed data. It also allows one to investigate covariate by genotype group interactions that may explain the variability in the traits. The flexibility to characterize such multiple trait manifestations makes the VCF model more desirable than the univariate variance components model, which is applied to each trait separately. The Bayesian framework offers a flexible approach that allows one to incorporate prior information.

Genome-wide association study of major recurrent depression in the U.K. population

OBJECTIVE

Studies of major depression in twins and families have shown moderate to high heritability, but extensive molecular studies have failed to identify susceptibility genes convincingly. To detect genetic variants contributing to major depression, the authors performed a genome-wide association study using 1,636 cases of depression ascertained in the U.K. and 1,594 comparison subjects screened negative for psychiatric disorders.

METHOD

Cases were collected from 1) a case-control study of recurrent depression (the Depression Case Control [DeCC] study; N=1346), 2) an affected sibling pair linkage study of recurrent depression (probands from the Depression Network [DeNT] study; N=332), and 3) a pharmacogenetic study (the Genome-Based Therapeutic Drugs for Depression [GENDEP] study; N=88). Depression cases and comparison subjects were genotyped at Centre National de Génotypage on the Illumina Human610-Quad BeadChip. After applying stringent quality control criteria for missing genotypes, departure from Hardy-Weinberg equilibrium, and low minor allele frequency, the authors tested for association to depression using logistic regression, correcting for population ancestry.

RESULTS

Single nucleotide polymorphisms (SNPs) in BICC1 achieved suggestive evidence for association, which strengthened after imputation of ungenotyped markers, and in analysis of female depression cases. A meta-analysis of U.K. data with previously published results from studies in Munich and Lausanne showed some evidence for association near neuroligin 1 (NLGN1) on chromosome 3, but did not support findings at BICC1.

CONCLUSIONS

This study identifies several signals for association worthy of further investigation but, as in previous genome-wide studies, suggests that individual gene contributions to depression are likely to have only minor effects, and very large pooled analyses will be required to identify them.

Genetic variation in complement factor H and risk of coronary heart disease: eight new studies and a meta-analysis of around 48,000 individuals

OBJECTIVES

To investigate the association of polymorphisms in complement factor H (CFH) and coronary heart disease (CHD) using meta-analysis.

BACKGROUND

Age-related macular degeneration (AMD) and CHD may share partially overlapping pathogenesis. A non-synonymous SNP (rs1061170/Y402H) in CFH encoding complement factor H (fH) is robustly associated with increased AMD risk but associations with CHD risk have been inconsistent.

METHODS

We conducted de novo genotyping and genetic association analyses of incident and prevalent CHD in four studies, and in silico analysis of the same association in a further four cohorts. We pooled these data with information from all published studies using random effects meta-analysis, including a total of 48,646 participants of which 9097 were CHD cases. We also evaluated the association of Y402H with known risk factors for CHD by pooling results from new and in silico studies providing relevant data.

RESULTS

CFH genotype was not associated with CHD. Compared to the reference TT homozygote group the pooled odds ratio (OR) for individuals homozygous for the C allele was 1.02, 95% CI (0.91, 1.13) and that for heterozygote TC individuals was 1.04 (0.98, 1.10). There was no association of CFH with systolic and diastolic blood pressure, total-, LDL- and HDL-cholesterol, or body mass index. Individuals who were CC compared to TT had higher triglyceride levels: pooled mean difference 0.06 (0.02, 0.10) mmol/L, p=0.005.

CONCLUSIONS

The AMD-associated CFH genotype is not associated with CHD. With the possible exception of triglycerides, this CFH SNP was not associated with a wide range of other CHD risk factors.

Evidence for both copy number and allelic (NA1/NA2) risk at the FCGR3B locus in systemic lupus erythematosus

The Fcgamma-receptor locus on chromosome 1q23 shows copy-number variation (CNV), and it has previously been shown that individuals with reduced numbers of copies of the Fcgamma-receptor-IIIB gene (FCGR3B) have an increased risk of developing systemic lupus erythematosus (SLE). It is not understood whether the association arises from FCGR3B (CD16b) itself, is observed because of linkage disequilibrium with actual causal alleles and/or is an effect of CNV on flanking FCGR genes. Thus, we extended this previous work by genotyping the FCGR3B alleles NA1/NA2 and re-assaying CNV using a paralogue ratio test assay in a family study (365 families). We have developed a novel case/pseudo-control approach to analyse family data, as the phase of copy number (CN) is not known in parents and cannot always be inferred in offspring. The results, obtained by fitting logistic regression models, confirm the association of low CN of FCGR3B with SLE (P=0.04). The risk conferred by low copies (<2) was contingent on FCGR3B allotype, being greater for deletion of NA1 than the for lower-affinity NA2. The simpler model with just CN was rejected in favour of the biallelic-CN model (P=0.03). We observed a correlation (R(2)=0.75, P<0.0001) between FCGR3B CNV and neutrophil expression in both healthy controls and patients with SLE. Our results suggest that one mechanism by which CNV at this locus confers disease risk is directly as a result of reduced FcgammaRIIIb function, either because of reduced expression (related to CNV) or because of reduced affinity for its ligand (NA1/NA2 allotype).

An ecological correlation study of late age-related macular degeneration and the complement factor H Y402H polymorphism

PURPOSE

To investigate whether variation in the distribution of the risk allele frequency of the Y402H single-nucleotide polymorphism (SNP) across various ethnicities and geographic regions reflects differences in the prevalence of late age-related macular degeneration (AMD) in those ethnicities.

METHODS

Published data were obtained via a systematic search. Study samples were grouped into clusters by ethnicity and geographic location and the Spearman correlation coefficient of the prevalence of late AMD and risk allele frequencies was calculated across clusters.

RESULTS

Across all ethnicities, AMD prevalence was seen to increase with age. Populations of European descent had both higher risk allele frequencies and prevalence of late AMD than did Japanese, Chinese, and Hispanic descendants. Results for African descendants were anomalous: although allele frequency was similar to that in European populations, the age-specific prevalence of late AMD was considerably lower. The correlation coefficient for the association between allele frequency and AMD prevalence was 0.40 (95% confidence interval [CI] = -0.36 to 0.84, P = 0.28) in all populations combined and 0.71 (95% CI = 0.02-0.94, P = 0.04) when people of African descent were excluded.

CONCLUSIONS

Evidence was found at the population level to support a positive association between the Y204H risk allele and the prevalence of AMD after exclusion of studies undertaken on persons of African ancestry. Data in African, Middle Eastern, and South American populations are needed to provide a better understanding of the association of late AMD genetic risk across ethnicities.

Evidence for an interaction between familial liability and prenatal exposure to infection in the causation of schizophrenia

OBJECTIVE

The authors sought to determine whether prenatal exposure to infection and a positive family history of psychotic disorders interact synergistically to increase the risk of later developing schizophrenia.

METHOD

The authors linked two national registers, the Medical Birth Register and the Finnish Population Register, to identify all women in Helsinki who received hospital treatment during pregnancy for an upper urinary tract infection (N=9,596) between 1947 and 1990. The Finnish Hospital Discharge Register was used to ascertain psychiatric outcomes in adulthood of offspring exposed to infection prenatally. Family history of psychotic disorders was determined by linking the Hospital Discharge Register and the Population Register. The authors used an additive statistical interaction model to calculate the amount of biological synergism between positive family history and prenatal exposure to infection.

RESULTS

Prenatal exposure to infection did not significantly increase the risk of schizophrenia. However, the effect of prenatal exposure to pyelonephritis was five times greater in those who had a family history of psychosis compared to those who did not. The synergy analysis suggested that an estimated 38%-46% of the offspring who developed schizophrenia and had both prenatal exposure to infection and a positive family history of psychotic disorders did so as a result of the synergistic action of both risk factors.

CONCLUSIONS

These findings support a mechanism of gene-environment interaction in the causation of schizophrenia.

The effect of exclusion of cases with unrecorded best estimate of gestational age on the estimates of preterm birth rate

OBJECTIVES

To investigate the effect of excluding cases with unrecorded best estimate of gestational age at birth on pregnancy outcome reporting and to determine the reasons for unrecorded gestational age data.

DESIGN

Prospective study.

SETTING

Fifteen maternity units in North West London.

POPULATION

497,105 women who booked for antenatal care from 1988 to 1998.

METHOD

Multiple logistic regression analysis.

MAIN OUTCOME MEASURES

Preterm birth rate of, and the factors associated with, cases with unrecorded best estimate of gestational age at birth.

RESULTS

Of the 53,981 cases with an unrecorded best estimate of gestational age at birth, by using additional data, it was possible to compute a new best estimate of gestational age in 80%. In this latter group, the preterm birth rate was 42% (95% CI 41.5-42.6). The corrected, overall preterm birth rate in North West London (9.8%, 9.7-9.9) was higher than the original estimate (7.6%, 7.5-7.7), which included only cases with recorded data on gestational age at birth. The most significant factors associated with an unrecorded gestational age were no ultrasound scan (OR 49, P < 0.001), and preterm birth <31 weeks (OR 30, P < 0.001).

CONCLUSIONS

The incidence of preterm birth are likely to be under-reported in studies where only cases with readily available gestational age data are included. In routinely collected maternity data, human omission is an important contributing factor for an unrecorded best estimate of gestational age at birth. This is associated with the urgent transfer of babies to the neonatal intensive care unit.

Multilocus Bayesian meta-analysis of gene-disease associations

Meta-analysis is a vital tool in genetic epidemiology. However, meta-analyses to identify gene-disease associations are compromised when contributing studies have typed partially overlapping sets of markers. Currently, only marginal analyses are possible, and these are restricted to the subset of studies typing that marker. This does not allow full use of available data and leads to the confounding of marker effects by closely associated markers. We present a Bayesian approach that exploits prior information on underlying haplotypes to allow multi-marker analysis incorporating data from all relevant studies of a gene or region, irrespective of the markers typed. We present results from application of our approach to data on a possible association between PDE4D and ischemic stroke.

The clonal evolution of metastases from primary serous epithelial ovarian cancers

Several models of evolution from primary cancers to metastases have been proposed; but the most widely accepted is the clonal evolution model proposed for colorectal cancer in which tumors develop by a process of linear clonal evolution driven by the accumulation of somatic genetic alterations. Various other models of cancer progression and metastasis have been proposed, including parallel evolution and the same gene model. The aim of this study was to investigate the evolution of metastases from primary cancer in 22 patients diagnosed with high-grade serous epithelial ovarian cancer. We established somatic genetic profiles based on the pattern of loss of heterozygosity, in several different regions of tumor tissue within the primary tumor and metastatic deposits from each case. Maximum parsimony tree analysis was used to examine the evolutionary relationship between the primary and metastatic samples for each patient. In addition, we investigated the extent of genetic heterogeneity within and between metastatic tumors compared with primary ovarian tumors. Our data suggest that most, if not all, metastases are clonally related to the primary tumors. However, the data oppose a single model of linear-clonal evolution whereby a late stage clone within the primary tumor acquires additional genetic changes that enable metastatic progression. Instead, the data support a model in which primary ovarian cancers have a common clonal origin, but become polyclonal, with different clones at both early and late stages of genetic divergence acquiring the ability to progress to metastasis.