Predicting stroke through genetic risk functions: the CHARGE Risk Score Project.
ABSTRACT: BACKGROUND AND PURPOSE:Beyond the Framingham Stroke Risk Score, prediction of future stroke may improve with a genetic risk score (GRS) based on single-nucleotide polymorphisms associated with stroke and its risk factors. METHODS:The study includes 4 population-based cohorts with 2047 first incident strokes from 22,720 initially stroke-free European origin participants aged ?55 years, who were followed for up to 20 years. GRSs were constructed with 324 single-nucleotide polymorphisms implicated in stroke and 9 risk factors. The association of the GRS to first incident stroke was tested using Cox regression; the GRS predictive properties were assessed with area under the curve statistics comparing the GRS with age and sex, Framingham Stroke Risk Score models, and reclassification statistics. These analyses were performed per cohort and in a meta-analysis of pooled data. Replication was sought in a case-control study of ischemic stroke. RESULTS:In the meta-analysis, adding the GRS to the Framingham Stroke Risk Score, age and sex model resulted in a significant improvement in discrimination (all stroke: ?joint area under the curve=0.016, P=2.3×10(-6); ischemic stroke: ?joint area under the curve=0.021, P=3.7×10(-7)), although the overall area under the curve remained low. In all the studies, there was a highly significantly improved net reclassification index (P<10(-4)). CONCLUSIONS:The single-nucleotide polymorphisms associated with stroke and its risk factors result only in a small improvement in prediction of future stroke compared with the classical epidemiological risk factors for stroke.
Project description:Multiple studies have identified single-nucleotide polymorphisms (SNPs) that are associated with coronary heart disease (CHD). We examined whether SNPs selected based on predefined criteria will improve CHD risk prediction when added to traditional risk factors (TRFs).SNPs were selected from the literature based on association with CHD, lack of association with a known CHD risk factor, and successful replication. A genetic risk score (GRS) was constructed based on these SNPs. Cox proportional hazards model was used to calculate CHD risk based on the Atherosclerosis Risk in Communities (ARIC) and Framingham CHD risk scores with and without the GRS.The GRS was associated with risk for CHD (hazard ratio [HR] = 1.10; 95% confidence interval [CI]: 1.07-1.13). Addition of the GRS to the ARIC risk score significantly improved discrimination, reclassification, and calibration beyond that afforded by TRFs alone in non-Hispanic whites in the ARIC study. The area under the receiver operating characteristic curve (AUC) increased from 0.742 to 0.749 (? = 0.007; 95% CI, 0.004-0.013), and the net reclassification index (NRI) was 6.3%. Although the risk estimates for CHD in the Framingham Offspring (HR = 1.12; 95% CI: 1.10-1.14) and Rotterdam (HR = 1.08; 95% CI: 1.02-1.14) Studies were significantly improved by adding the GRS to TRFs, improvements in AUC and NRI were modest.Addition of a GRS based on direct associations with CHD to TRFs significantly improved discrimination and reclassification in white participants of the ARIC Study, with no significant improvement in the Rotterdam and Framingham Offspring Studies.
Project description:The utility of genetic risk scores (GRS) as independent risk predictors remains inconclusive. Here, we evaluate the additive value of a multi-locus GRS to the Framingham risk score (FRS) in coronary artery disease (CAD) risk prediction. A total of 2888 individuals (1566 coronary patients and 1322 controls) were divided into three subgroups according to FRS. Multiplicative GRS was determined for 32 genetic variants associated to CAD. Logistic Regression and Area Under the Curve (AUC) were determined first, using the TRF for each FRS subgroup, and secondly, adding GRS. Different models (TRF, TRF+GRS) were used to classify the subjects into risk categories for the FRS 10-year predicted risk. The improvement offered by GRS was expressed as Net Reclassification Index and Integrated Discrimination Improvement. Multivariate analysis showed that GRS was an independent predictor for CAD (OR = 1.87; p<0.0001). Diabetes, arterial hypertension, dyslipidemia and smoking status were also independent CAD predictors (p<0.05). GRS added predictive value to TRF across all risk subgroups. NRI showed a significant improvement in all categories. In conclusion, GRS provided a better incremental value in intermediate subgroup. In this subgroup, inclusion of genotyping may be considered to better stratify cardiovascular risk.
Project description:BACKGROUND AND PURPOSE:Limited data exist on the performance of the revised Framingham Stroke Risk Score (R-FSRS) and the R-FSRS in conjunction with nontraditional risk markers. We compared the R-FSRS, original FSRS, and the Pooled Cohort Equation for stroke prediction and assessed the improvement in discrimination by nontraditional risk markers. METHODS:Six thousand seven hundred twelve of 6814 participants of the MESA (Multi-Ethnic Study of Atherosclerosis) were included. Cox proportional hazard, area under the curve, net reclassification improvement, and integrated discrimination increment analysis were used to assess and compare each stroke prediction risk score. Stroke was defined as fatal/nonfatal strokes (hemorrhagic or ischemic). RESULTS:After mean follow-up of 10.7 years, 231 of 6712 (3.4%) strokes were adjudicated (2.7% ischemic strokes). Mean stroke risks using the R-FSRS, original FSRS, and Pooled Cohort Equation were 4.7%, 5.9%, and 13.5%. The R-FSRS had the best calibration (Hosmer-Lemeshow goodness-of-fit, ?2=6.55; P=0.59). All risk scores were predictive of incident stroke. C statistics of R-FSRS (0.716) was similar to Pooled Cohort Equation (0.716), but significantly higher than the original FSRS (0.653; P=0.01 for comparison with R-FSRS). Adding nontraditional risk markers individually to the R-FSRS did not improve discrimination of the R-FSRS in the area under the curve analysis, but did improve category-less net reclassification improvement and integrated discrimination increment for incident stroke. The addition of coronary artery calcium to R-FSRS produced the highest category-less net reclassification improvement (0.36) and integrated discrimination increment (0.0027). Similar results were obtained when ischemic strokes were used as the outcome. CONCLUSIONS:The R-FSRS downgraded stroke risk but had better calibration and discriminative ability for incident stroke compared with the original FSRS. Nontraditional risk markers modestly improved the discriminative ability of the R-FSRS, with coronary artery calcium performing the best.
Project description:To assess the performance of a panel of common single nucleotide polymorphisms (genotypes) associated with type 2 diabetes in distinguishing incident cases of future type 2 diabetes (discrimination), and to examine the effect of adding genetic information to previously validated non-genetic (phenotype based) models developed to estimate the absolute risk of type 2 diabetes.Workplace based prospective cohort study with three 5 yearly medical screenings.5535 initially healthy people (mean age 49 years; 33% women), of whom 302 developed new onset type 2 diabetes over 10 years.Non-genetic variables included in two established risk models-the Cambridge type 2 diabetes risk score (age, sex, drug treatment, family history of type 2 diabetes, body mass index, smoking status) and the Framingham offspring study type 2 diabetes risk score (age, sex, parental history of type 2 diabetes, body mass index, high density lipoprotein cholesterol, triglycerides, fasting glucose)-and 20 single nucleotide polymorphisms associated with susceptibility to type 2 diabetes. Cases of incident type 2 diabetes were defined on the basis of a standard oral glucose tolerance test, self report of a doctor's diagnosis, or the use of anti-diabetic drugs.A genetic score based on the number of risk alleles carried (range 0-40; area under receiver operating characteristics curve 0.54, 95% confidence interval 0.50 to 0.58) and a genetic risk function in which carriage of risk alleles was weighted according to the summary odds ratios of their effect from meta-analyses of genetic studies (area under receiver operating characteristics curve 0.55, 0.51 to 0.59) did not effectively discriminate cases of diabetes. The Cambridge risk score (area under curve 0.72, 0.69 to 0.76) and the Framingham offspring risk score (area under curve 0.78, 0.75 to 0.82) led to better discrimination of cases than did genotype based tests. Adding genetic information to phenotype based risk models did not improve discrimination and provided only a small improvement in model calibration and a modest net reclassification improvement of about 5% when added to the Cambridge risk score but not when added to the Framingham offspring risk score.The phenotype based risk models provided greater discrimination for type 2 diabetes than did models based on 20 common independently inherited diabetes risk alleles. The addition of genotypes to phenotype based risk models produced only minimal improvement in accuracy of risk estimation assessed by recalibration and, at best, a minor net reclassification improvement. The major translational application of the currently known common, small effect genetic variants influencing susceptibility to type 2 diabetes is likely to come from the insight they provide on causes of disease and potential therapeutic targets.
Project description:Whether using both traditional risk factors and genetic variants for stroke as opposed to using either of the 2 alone improves the prediction of stroke risk remains unclear. The purpose of this study was to compare the predictability of stroke risk between models using traditional risk score (TRS) and genetic risk score (GRS).We used a case-cohort study from the Korean Cancer Prevention Study-II (KCPS-II) Biobank (n=156,701). We genotyped 72 single nucleotide polymorphisms (SNPs) identified in genome-wide association study (GWAS) on the KCPS-II sub-cohort members and stroke cases. We calculated GRS by summing the number of risk alleles. Prediction models with or without GRS were evaluated in terms of the area under the receiver operating characteristic curve (AUROC).Sixteen out of 72 SNPs identified in GWAS showed significant associations with stroke, with an odds ratio greater than 2.0. For participants aged <40 years, AUROCs for incident stroke were 0.58, 0.65, and 0.67 in models using modifiable TRS only, GRS only, and TRS plus GRS, respectively, showing that GRS only model had better prediction than TRS only. For participants aged ≥40 years, however, TRS only model had better prediction than GRS only model. Favorable levels of traditional risk were associated with significantly lower stroke risks within each genetic risk category.TRS and GRS were both independently associated with stroke risk. Using genetic variants in addition to traditional risk factors may be the most accurate way of predicting stroke risk, particularly in relatively younger individuals.
Project description:The American Heart Association has established criteria for the evaluation of novel markers of cardiovascular risk. In accordance with these criteria, we assessed the association between a multi-locus genetic risk score (GRS) and incident coronary heart disease (CHD), and evaluated whether this GRS improves the predictive capacity of the Framingham risk function.Using eight genetic variants associated with CHD but not with classical cardiovascular risk factors (CVRFs), we generated a multi-locus GRS, and found it to be linearly associated with CHD in two population based cohorts: The REGICOR Study (n=2351) and The Framingham Heart Study (n=3537) (meta-analyzed HR [95%CI]: ~1.13 [1.01-1.27], per unit). Inclusion of the GRS in the Framingham risk function improved its discriminative capacity in the Framingham sample (c-statistic: 72.81 vs.72.37, p=0.042) but not in the REGICOR sample. According to both the net reclassification improvement (NRI) index and the integrated discrimination index (IDI), the GRS improved re-classification among individuals with intermediate coronary risk (meta-analysis NRI [95%CI]: 17.44 [8.04; 26.83]), but not overall.A multi-locus GRS based on genetic variants unrelated to CVRFs was associated with a linear increase in risk of CHD events in two distinct populations. This GRS improves risk reclassification particularly in the population at intermediate coronary risk. These results indicate the potential value of the inclusion of genetic information in classical functions for risk assessment in the intermediate risk population group.
Project description:Genome-wide association studies have identified numerous single nucleotide polymorphisms (SNPs) as associated with colorectal cancer (CRC) risk in populations of European descent. However, their utility for predicting risk to CRC in Asians remains unknown. A case-cohort study (random sub-cohort N=1,685) from the Korean Cancer Prevention Study-II (KCPS-II) (N=145,842) was used. Twenty-three SNPs identified in previous 47 studies were genotyped on the KCPS-II sub-cohort members. A genetic risk score (GRS) was calculated by summing the number of risk alleles over all SNPs. Prediction models with or without GRS were evaluated in terms of the area under the receiver operating characteristic curve (AUROC) and the continuous net reclassification index (NRI).Seven of 23 SNPs showed significant association with CRC and rectal cancer in Koreans, but not with colon cancer alone. AUROCs (95% CI) for traditional risk score (TRS) alone and TRS plus GRS were 0.73 (0.69-0.78) and 0.74 (0.70-0.78) for CRC, and 0.71 (0.65-0.77) and 0.74 (0.68-0.79) for rectal cancer, respectively. The NRI (95% CI) for a prediction model with GRS compared to the model with TRS alone was 0.17 (-0.05-0.37) for CRC and 0.41 (0.10-0.68) for rectal cancer alone.Our results indicate genetic variants may be useful for predicting risk to CRC in the Koreans, especially risk for rectal cancer alone. Moreover, this study suggests effective prediction models for colon and rectal cancer should be developed separately.
Project description:Atrial fibrillation (AF) is prevalent and there is a clinical need for biomarkers to identify individuals at higher risk for AF. Fixed throughout a life course and assayable early in life, genetic biomarkers may meet this need. Here, we investigate whether multiple single nucleotide polymorphisms together as an AF genetic risk score (AF-GRS) can improve prediction of one's risk for AF.In 27 471 participants of the Malmö Diet and Cancer Study, a prospective, community-based cohort, we used Cox models that adjusted for established AF risk factors to assess the association of AF-GRS with incident AF and ischemic stroke. Median follow-up was 14.4 years for incident AF and 14.5 years for ischemic stroke. The AF-GRS comprised 12 single nucleotide polymorphisms that had been previously shown to be associated with AF at genome-wide significance.During follow-up, 2160 participants experienced a first AF event and 1495 had a first ischemic stroke event. Participants in the top AF-GRS quintile were at increased risk for incident AF (hazard ratio, 2.00; 95% confidence interval, 1.73-2.31; P=2.7×10(-21)) and ischemic stroke (hazard ratio, 1.23; 95% confidence interval, 1.04-1.46; P=0.02) when compared with the bottom quintile. Addition of the AF-GRS to established AF risk factors modestly improved both discrimination and reclassification (P<0.0001 for both).An AF-GRS can identify 20% of individuals who are at ?2-fold increased risk for incident AF and at 23% increased risk for ischemic stroke. Targeting diagnostic or therapeutic interventions to this subset may prove clinically useful.
Project description:<h4>Objective</h4>A genetic risk score (GRS) comprised of single nucleotide polymorphisms (SNPs) and metabolite biomarkers have each been shown, separately, to predict incident type 2 diabetes. We tested whether genetic and metabolite markers provide complementary information for type 2 diabetes prediction and, together, improve the accuracy of prediction models containing clinical traits.<h4>Research design and methods</h4>Diabetes risk was modeled with a 62-SNP GRS, nine metabolites, and clinical traits. We fit age- and sex-adjusted logistic regression models to test the association of these sources of information, separately and jointly, with incident type 2 diabetes among 1,622 initially nondiabetic participants from the Framingham Offspring Study. The predictive capacity of each model was assessed by area under the curve (AUC).<h4>Results</h4>Two hundred and six new diabetes cases were observed during 13.5 years of follow-up. The AUC was greater for the model containing the GRS and metabolite measurements together versus GRS or metabolites alone (0.820 vs. 0.641, P < 0.0001, or 0.820 vs. 0.803, P = 0.01, respectively). Odds ratios for association of GRS or metabolites with type 2 diabetes were not attenuated in the combined model. The AUC was greater for the model containing the GRS, metabolites, and clinical traits versus clinical traits only (0.880 vs. 0.856, P = 0.002).<h4>Conclusions</h4>Metabolite and genetic traits provide complementary information to each other for the prediction of future type 2 diabetes. These novel markers of diabetes risk modestly improve the predictive accuracy of incident type 2 diabetes based only on traditional clinical risk factors.
Project description:<h4>Objective</h4>The aim of this study is to discover common variants in 6 lipid metabolic genes and construct and validate a genetic risk score (GRS) based on the joint effects of genetic variants in multiple genes from lipid and other pathobiologic pathways.<h4>Background</h4>Explaining the genetic basis of coronary artery disease (CAD) is incomplete. Discovery and aggregation of genetic variants from multiple pathways may advance this objective.<h4>Methods</h4>Premature CAD cases (n = 1,947) and CAD-free controls (n = 1,036) were selected from our angiographic registry. In a discovery phase, single nucleotide polymorphisms (SNPs) at 56 loci from internal discovery and external reports were tested for associations with biomarkers and CAD: 28 promising SNPs were then tested jointly for CAD associations, and a GRS consisting of SNPs contributing independently was constructed and validated in a replication set of familial cases and population-based controls (n = 1,320).<h4>Results</h4>Five variants contributed jointly to CAD prediction in a multigenic GRS model: odds ratio 1.24 (95% CI 1.16-1.33) per risk allele, P = 8.2 x 10(-11), adjusted OR 2.03 (1.53-2.70), fourth versus first quartile. 5-SNP genetic risk score had minor impact on area under the receiver operating characteristic curve (P > .05) but resulted in substantial net reclassification improvement: 0.16 overall, 0.28 in intermediate-risk patients (both P < .0001). GRS(5) predicted familial CAD with similar magnitude in the validation set.<h4>Conclusions</h4>The Intermountain Healthcare's Coronary Genetics study demonstrates the ability of a multigenic, multipathway GRS to improve discrimination of angiographic CAD. Genetic risk scores promise to increase understanding of the genetic basis of CAD and improve identification of individuals at increased CAD risk.