Predicting stroke through genetic risk functions: the CHARGE Risk Score Project

Stroke risk prediction models: A systematic review and meta-analysis

BACKGROUND

Prediction algorithms/models are viable methods for identifying individuals at high risk of stroke across diverse populations for timely intervention. However, evidence summarizing the performance of these models is limited. This study examined the performance and weaknesses of existing stroke risk-score-prediction models (SRSMs) and whether performance varied by population and region.

METHODS

PubMed, EMBASE, and Web of Science were searched for articles on SRSMs from the earliest records until February 2022. The Prediction Model Risk of Bias Assessment Tool was used to assess the quality of eligible articles. The performance of the SRSMs was assessed by meta-analyzing C-statistics (0 and 1) estimates from identified studies to determine the overall pooled C-statistics by fitting a linear restricted maximum likelihood in a random effect model.

RESULTS

Overall, 17 articles (cohort study = 15, nested case-control study = 2) comprising 739,134 stroke cases from 6,396,594 participants from diverse populations/regions (Asia; n = 8, United States; n = 3, and Europe and the United Kingdom; n = 6) were eligible for inclusion. The overall pooled c-statistics of SRSMs was 0.78 (95%CI: 0.75, 0.80; I2 = 99.9%), with most SRSMs developed using cohort studies; 0.78 (95%CI: 0.75, 0.80; I2 = 99.9%). The subgroup analyses by geographical region: Asia [0.81 (95%CI: 0.79, 0.83; I2 = 99.8%)], Europe and the United Kingdom [0.76 (95%CI: 0.69, 0.83; I2 = 99.9%)] and the United States only [0.75 (95%CI: 0.72, 0.78; I2 = 73.5%)] revealed relatively indifferent performances of SRSMs.

CONCLUSION

SRSM performance varied widely, and the pooled c-statistics of SRSMs suggested a fair predictive performance, with very few SRSMs validated in independent population group(s) from diverse world regions.

The Inclusion of Underrepresented Populations in Cardiovascular Genetics and Epidemiology

Novel genetic risk markers have helped us to advance the field of cardiovascular epidemiology and refine our current understanding and risk stratification paradigms. The discovery and analysis of variants can help us to tailor prognostication and management. However, populations underrepresented in cardiovascular epidemiology and cardiogenetics research may experience inequities in care if prediction tools are not applicable to them clinically. Therefore, the purpose of this article is to outline the barriers that underrepresented populations can face in participating in genetics research, to describe the current efforts to diversify cardiogenetics research, and to outline strategies that researchers in cardiovascular epidemiology can implement to include underrepresented populations. Mistrust, a lack of diverse research teams, the improper use of sensitive biodata, and the constraints of genetic analyses are all barriers for including diverse populations in genetics studies. The current work is beginning to address the paucity of ethnically diverse genetics research and has already begun to shed light on the potential benefits of including underrepresented and diverse populations. Reducing barriers for individuals, utilizing community-driven research processes, adopting novel recruitment strategies, and pushing for organizational support for diverse genetics research are key steps that clinicians and researchers can take to develop equitable risk stratification tools and improve patient care.

Associations of polygenic risk scores with risks of stroke and its subtypes in Chinese

BACKGROUND AND PURPOSE

Previous studies, mostly focusing on the European population, have reported polygenic risk scores (PRSs) might achieve risk stratification of stroke. We aimed to examine the association strengths of PRSs with risks of stroke and its subtypes in the Chinese population.

METHODS

Participants with genome-wide genotypic data in China Kadoorie Biobank were split into a potential training set (n=22 191) and a population-based testing set (n=72 150). Four previously developed PRSs were included, and new PRSs for stroke and its subtypes were developed. The PRSs showing the strongest association with risks of stroke or its subtypes in the training set were further evaluated in the testing set. Cox proportional hazards regression models were used to estimate the association strengths of different PRSs with risks of stroke and its subtypes (ischaemic stroke (IS), intracerebral haemorrhage (ICH) and subarachnoid haemorrhage (SAH)).

RESULTS

In the testing set, during 872 919 person-years of follow-up, 8514 incident stroke events were documented. The PRSs of any stroke (AS) and IS were both positively associated with risks of AS, IS and ICH (p<0.05). The HR for per SD increment (HRSD) of PRSAS was 1.10 (95% CI 1.07 to 1.12), 1.10 (95% CI 1.07 to 1.12) and 1.13 (95% CI 1.07 to 1.20) for AS, IS and ICH, respectively. The corresponding HRSD of PRSIS was 1.08 (95% CI 1.06 to 1.11), 1.08 (95% CI 1.06 to 1.11) and 1.09 (95% CI 1.03 to 1.15). PRSICH was positively associated with the risk of ICH (HRSD=1.07, 95% CI 1.01 to 1.14). PRSSAH was not associated with risks of stroke and its subtypes. The addition of current PRSs offered little to no improvement in stroke risk prediction and risk stratification.

CONCLUSIONS

In this Chinese population, the association strengths of current PRSs with risks of stroke and its subtypes were moderate, suggesting a limited value for improving risk prediction over traditional risk factors in the context of current genome-wide association study under-representing the East Asian population.

The prediction of Alzheimer's disease through multi-trait genetic modeling

To better capture the polygenic architecture of Alzheimer's disease (AD), we developed a joint genetic score, MetaGRS. We incorporated genetic variants for AD and 24 other traits from two independent cohorts, NACC (n = 3,174, training set) and UPitt (n = 2,053, validation set). One standard deviation increase in the MetaGRS is associated with about 57% increase in the AD risk [hazard ratio (HR) = 1.577, p = 7.17 E-56], showing little difference from the HR for AD GRS alone (HR = 1.579, p = 1.20E-56), suggesting similar utility of both models. We also conducted APOE-stratified analyses to assess the role of the e4 allele on risk prediction. Similar to that of the combined model, our stratified results did not show a considerable improvement of the MetaGRS. Our study showed that the prediction power of the MetaGRS significantly outperformed that of the reference model without any genetic information, but was effectively equivalent to the prediction power of the AD GRS.

Integrating polygenic and clinical risks to improve stroke risk stratification in prospective Chinese cohorts

The utility of the polygenic risk score (PRS) to identify individuals at higher risk of stroke beyond clinical risk remains unclear, and we clarified this using Chinese population-based prospective cohorts. Cox proportional hazards models were used to estimate the 10-year risk, and Fine and Gray's models were used for hazard ratios (HRs), their 95% confidence intervals (CIs), and the lifetime risk according to PRS and clinical risk categories. A total of 41,006 individuals aged 30-75 years with a mean follow-up of 9.0 years were included. Comparing the top versus bottom 5% of the PRS, the HR was 3.01 (95%CI 2.03-4.45) in the total population, and similar findings were observed within clinical risk strata. Marked gradients in the 10-year and lifetime risk across PRS categories were also found within clinical risk categories. Notably, among individuals with intermediate clinical risk, the 10-year risk for those in the top 5% of the PRS (7.3%, 95%CI 7.1%-7.5%) reached the threshold of high clinical risk (⩾7.0%) for initiating preventive treatment, and this effect of the PRS on refining risk stratification was evident for ischemic stroke. Even among those in the top 10% and 20% of the PRS, the 10-year risk would also exceed this level when aged ⩾50 and ⩾60 years, respectively. Overall, the combination of the PRS with the clinical risk score improved the risk stratification within clinical risk strata and distinguished actual high-risk individuals with intermediate clinical risk.

Dissecting Polygenic Etiology of Ischemic Stroke in the Era of Precision Medicine

Ischemic stroke (IS), the leading cause of death and disability worldwide, is caused by many modifiable and non-modifiable risk factors. This complex disease is also known for its multiple etiologies with moderate heritability. Polygenic risk scores (PRSs), which have been used to establish a common genetic basis for IS, may contribute to IS risk stratification for disease/outcome prediction and personalized management. Statistical modeling and machine learning algorithms have contributed significantly to this field. For instance, multiple algorithms have been successfully applied to PRS construction and integration of genetic and non-genetic features for outcome prediction to aid in risk stratification for personalized management and prevention measures. PRS derived from variants with effect size estimated based on the summary statistics of a specific subtype shows a stronger association with the matched subtype. The disruption of the extracellular matrix and amyloidosis account for the pathogenesis of cerebral small vessel disease (CSVD). Pathway-specific PRS analyses confirm known and identify novel etiologies related to IS. Some of these specific PRSs (e.g., derived from endothelial cell apoptosis pathway) individually contribute to post-IS mortality and, together with clinical risk factors, better predict post-IS mortality. In this review, we summarize the genetic basis of IS, emphasizing the application of methodologies and algorithms used to construct PRSs and integrate genetics into risk models.

Polygenic Risk, Midlife Life's Simple 7, and Lifetime Risk of Stroke

Background

Recent genetic discoveries in stroke have unleashed the potential of using genetic information for risk prediction and health interventions aimed at disease prevention. We sought to estimate the lifetime risk of stroke (LTRS) by levels of genetic risk and to investigate whether optimal cardiovascular health can offset the negative impact of high genetic risk on lifetime risk of stroke.

Methods and Results

Study participants were 11 568 middle‐aged adults (56% women, 23% Black adults), who were free of stroke at baseline and were followed up for a median of 28 years. The remaining LTRS was estimated according to levels of genetic risk based on a validated stroke polygenic risk score, and to levels of cardiovascular health based on the American Heart Association Life's Simple 7 recommendations. At age 45, individuals with high, intermediate, and low polygenic risk score had a remaining LTRS of 23.2% (95% CI, 20.8%–25.5%), 13.8% (95% CI, 11.7%–15.8%), and 9.6% (95% CI, 7.3%–11.8%), respectively. Those with both a high genetic risk and an inadequate Life's Simple 7 experienced the highest LTRS: 24.8% (95% CI, 22.0%–27.6%). Across all polygenic risk score categories, those with an optimal Life's Simple 7 had a ≈30% to 43% lower LTRS than those with an inadequate Life's Simple 7. This corresponded to almost 6 additional years lived free of stroke.

Conclusions

The LTRS varies by levels of polygenic risk and cardiovascular health. Maintaining an optimal cardiovascular health can partially offset a high genetic risk, emphasizing the importance of modifiable risk factors and illustrating the potential of personalizing genetic risk information to motivate lifestyle changes for stroke prevention.

2022 Prevention of chronic non-communicable diseases in Of the Russian Federation. National guidelines

.

Genome-wide polygenic risk impact on intracranial aneurysms and acute ischemic stroke

Polygenic risk scores (PRSs) have an important relevance to approaches for clinical usage in intracranial aneurysm (IA) patients. Hence, we aimed to develop IA-predicting PRS models including the genetic basis shared with acute ischemic stroke (AIS) in Korean populations. We applied a weighted PRS (wPRS) model based on a previous genome-wide association study (GWAS) of 250 IA patients in a hospital-based multicenter cohort, 222 AIS patients in a validation study, and 296 shared controls. Risk predictability was analyzed by the area under the receiver operating characteristic curve (AUROC). The best-fitting risk models based on wPRSs were stratified into tertiles representing the lowest, middle, and highest risk groups. The weighted PRS, which included 29 GWASs (p < 5×10⁻⁸) and two reported genetic variants (p < 0.01), showed a high predictability in IA patients (AUROC = 0.949, 95% CI: 0.933–0.966). This wPRS was significantly validated in AIS patients (AUROC = 0.842, 95% CI: 0.808–0.876; p < 0.001). Two-stage risk models stratified into tertiles showed an increased risk for IA (OR = 691.25, 95% CI: 241.77–1976.35; p = 3.1×10⁻³⁴; sensitivity/specificity = 0.728/0.963), which was replicated in AIS development (OR = 39.76, 95% CI: 16.91–93.49; p = 3.1×10⁻¹⁷; sensitivity/specificity = 0.284/0.963). A higher wPRS for IA may be associated with an increased risk of AIS in the Korean population. These findings suggest that IA and AIS may have a shared genetic architecture and should be studied further to generate a precision medicine model for use in personalized diagnosis and treatment.

Associations of carotid intima media thickness with gene expression in whole blood and genetically predicted gene expression across 48 tissues

Carotid intima media thickness (cIMT) is a biomarker of subclinical atherosclerosis and a predictor of future cardiovascular events. Identifying associations between gene expression levels and cIMT may provide insight to atherosclerosis etiology. Here, we use two approaches to identify associations between mRNA levels and cIMT: differential gene expression analysis in whole blood and S-PrediXcan. We used microarrays to measure genome-wide whole blood mRNA levels of 5647 European individuals from four studies. We examined the association of mRNA levels with cIMT adjusted for various potential confounders. Significant associations were tested for replication in three studies totaling 3943 participants. Next, we applied S-PrediXcan to summary statistics from a cIMT genome-wide association study (GWAS) of 71 128 individuals to estimate the association between genetically determined mRNA levels and cIMT and replicated these analyses using S-PrediXcan on an independent GWAS on cIMT that included 22 179 individuals from the UK Biobank. mRNA levels of TNFAIP3, CEBPD and METRNL were inversely associated with cIMT, but these associations were not significant in the replication analysis. S-PrediXcan identified associations between cIMT and genetically determined mRNA levels for 36 genes, of which six were significant in the replication analysis, including TLN2, which had not been previously reported for cIMT. There was weak correlation between our results using differential gene expression analysis and S-PrediXcan. Differential expression analysis and S-PrediXcan represent complementary approaches for the discovery of associations between phenotypes and gene expression. Using these approaches, we prioritize TNFAIP3, CEBPD, METRNL and TLN2 as new candidate genes whose differential expression might modulate cIMT.

A Novel Afrocentric Stroke Risk Assessment Score: Models from the Siren Study

BACKGROUND

Stroke risk can be quantified using risk factors whose effect sizes vary by geography and race. No stroke risk assessment tool exists to estimate aggregate stroke risk for indigenous African.

OBJECTIVES

To develop Afrocentric risk-scoring models for stroke occurrence.

MATERIALS AND METHODS

We evaluated 3533 radiologically confirmed West African stroke cases paired 1:1 with age-, and sex-matched stroke-free controls in the SIREN study. The 7,066 subjects were randomly split into a training and testing set at the ratio of 85:15. Conditional logistic regression models were constructed by including 17 putative factors linked to stroke occurrence using the training set. Significant risk factors were assigned constant and standardized statistical weights based on regression coefficients (β) to develop an additive risk scoring system on a scale of 0-100%. Using the testing set, Receiver Operating Characteristics (ROC) curves were constructed to obtain a total score to serve as cut-off to discriminate between cases and controls. We calculated sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) at this cut-off.

RESULTS

For stroke occurrence, we identified 15 traditional vascular factors. Cohen's kappa for validity was maximal at a total risk score of 56% using both statistical weighting approaches to risk quantification and in both datasets. The risk score had a predictive accuracy of 76% (95%CI: 74-79%), sensitivity of 80.3%, specificity of 63.0%, PPV of 68.5% and NPV of 76.2% in the test dataset. For ischemic strokes, 12 risk factors had predictive accuracy of 78% (95%CI: 74-81%). For hemorrhagic strokes, 7 factors had a predictive accuracy of 79% (95%CI: 73-84%).

CONCLUSIONS

The SIREN models quantify aggregate stroke risk in indigenous West Africans with good accuracy. Prospective studies are needed to validate this instrument for stroke prevention.

Development and Validation of a Polygenic Risk Score for Stroke in the Chinese Population

OBJECTIVE

To construct a polygenic risk score (PRS) for stroke and evaluate its utility in risk stratification and primary prevention for stroke.

METHODS

Using a meta-analytic approach and large genome-wide association results for stroke and stroke-related traits in East Asians, we generated a combined PRS (metaPRS) by incorporating 534 genetic variants in a training set of 2,872 patients with stroke and 2,494 controls. We then validated its association with incident stroke using Cox regression models in large Chinese population-based prospective cohorts comprising 41,006 individuals.

RESULTS

During a total of 367,750 person-years (mean follow-up 9.0 years), 1,227 participants developed stroke before age 80 years. Individuals with high polygenic risk had an about 2-fold higher risk of incident stroke compared with those with low polygenic risk (hazard ratio [HR] 1.99, 95% confidence interval [CI] 1.66-2.38), with the lifetime risk of stroke being 25.2% (95% CI 22.5%-27.7%) and 13.6% (95% CI 11.6%-15.5%), respectively. Individuals with both high polygenic risk and family history displayed lifetime risk as high as 41.1% (95% CI 31.4%-49.5%). Individuals with high polygenic risk achieved greater benefits in terms of absolute risk reductions from adherence to ideal fasting blood glucose and total cholesterol than those with low polygenic risk. Maintaining favorable cardiovascular health (CVH) profile could substantially mitigate the increased risk conferred by high polygenic risk to the level of low polygenic risk (from 34.6% to 13.2%).

CONCLUSIONS

Our metaPRS has great potential for risk stratification of stroke and identification of individuals who may benefit more from maintaining ideal CVH.

CLASSIFICATION OF EVIDENCE

This study provides Class I evidence that metaPRS is predictive of stroke risk.

A systematic review of the status and methodological considerations for estimating risk of first ever stroke in the general population

AIMS

The methodological quality of development, validation, and modification of those models have not been evaluated via a thoroughly literature review. This study aims to describe the overall status and evaluate the methodological quality of risk prediction models for stroke incidence in the general population.

METHODS

We searched the database of EMBASE and MEDLINE by the combination of subject words and key words to collect the research on stroke risk prediction model in the general population. The retrieval time was from the establishment of the database to September 2019. It should be mentioned that risk of bias for each model was assessed, and data on population characteristics and model performance was also extracted.

RESULTS

The search screened 11,386 peer-reviewed publications and 57 citation searching, of which 48 were included in the review, describing the development of 51 prediction models, 47 external validation models, and 12 modification models. Among 51 development models, the predicted outcome concentrated on fatal or non-fatal stroke (n = 37, 73%). Thirty-nine development models (76%) were without internal validation. C-statistic or AUC was adopted for discrimination in 80% models, and Hosmer-Lemeshow test (n = 25, 49%) was also performed for calibration. Twenty-six development models (53%) were externally validated, among which only 2 (8%) were validated by independent researchers. Risk prediction performance was improved when models were modified by adding novel risk factors, such as the internal carotid artery plaque and intima-media thickness.

CONCLUSION

Models for predicting stroke occurrence need further external validation, recalibration, or modification in different populations, to help interpret those models in the practice of stroke prevention.

Polygenic Risk Scores Augment Stroke Subtyping

Objective

To determine whether the polygenic risk score (PRS) derived from MEGASTROKE is associated with ischemic stroke (IS) and its subtypes in an independent tertiary health care system and to identify the PRS derived from gene sets of known biological pathways associated with IS.

Methods

Controls (n = 19,806/7,484, age ≥69/79 years) and cases (n = 1,184/951 for discovery/replication) of acute IS with European ancestry and clinical risk factors were identified by leveraging the Geisinger Electronic Health Record and chart review confirmation. All Geisinger MyCode patients with age ≥69/79 years and without any stroke-related diagnostic codes were included as low risk control. Genetic heritability and genetic correlation between Geisinger and MEGASTROKE (EUR) were calculated using the summary statistics of the genome-wide association study by linkage disequilibrium score regression. All PRS for any stroke (AS), any ischemic stroke (AIS), large artery stroke (LAS), cardioembolic stroke (CES), and small vessel stroke (SVS) were constructed by PRSice-2.

Results

A moderate heritability (10%–20%) for Geisinger sample as well as the genetic correlation between MEGASTROKE and the Geisinger cohort was identified. Variation of all 5 PRS significantly explained some of the phenotypic variations of Geisinger IS, and the R² increased by raising the cutoff for the age of controls. PRSLAS, PRSCES, and PRSSVS derived from low-frequency common variants provided the best fit for modeling (R² = 0.015 for PRSLAS). Gene sets analyses highlighted the association of PRS with Gene Ontology terms (vascular endothelial growth factor, amyloid precursor protein, and atherosclerosis). The PRSLAS, PRSCES, and PRSSVS explained the most variance of the corresponding subtypes of Geisinger IS suggesting shared etiologies and corroborated Geisinger TOAST subtyping.

Conclusions

We provide the first evidence that PRSs derived from MEGASTROKE have value in identifying shared etiologies and determining stroke subtypes.

Combined lifestyle factors are associated with the risk of ischaemic stroke in a Chinese population

PURPOSE

Lifestyle plays an important role in the development of ischaemic stroke (IS). The objective of this study was to evaluate the association between combined lifestyle factors and the risk of IS in an eastern Chinese population.

METHODS

We collected lifestyle information from 191 patients with IS admitted to the second affiliated hospital of Soochow University and 575 control subjects from community clinics using a structured questionnaire. After division into training and test datasets, regularised stepwise logistic regression, least absolute shrinkage and selection operator regression and genetic algorithm models were used to identify lifestyle factors associated with IS. The improved discriminative ability by adding the lifestyle factors was determined by c-index and reclassification analysis. The final model was validated in the test dataset.

RESULTS

After controlling for conventional cardiovascular disease (CVD) risk factors, preferences for fruits (OR (95% CI): 0.29 (0.18-0.46), p<0.001) and soy products (0.47 (0.29-0.75), p=0.002) were negatively associated with IS, while lower life satisfaction (mildly satisfied: 2.15 (1.27-3.63), p=0.004; not satisfied: 6.39 (1.76-25.44), p=0.006) was positively associated with IS. Adding these factors to a basic CVD risk model improved the c-index (0.825 vs 0.753, p<0.001) and reclassification for IS (net reclassification index (95% CI): 18.49% (7.90%-29.08%), integrated discrimination index (95% CI): 0.11% (0.08%-0.14%), p<0.001). The model with lifestyle factors achieved a c-index of 0.813 and good calibration in the test dataset.

CONCLUSIONS

Our results showed that combined lifestyle factors including dietary pattern and life satisfaction are independently associated with the risk of IS.

Clinical Application of a Novel Genetic Risk Score for Ischemic Stroke in Patients With Cardiometabolic Disease

BACKGROUND

Genome-wide association studies have identified single-nucleotide polymorphisms that are associated with an increased risk of stroke. We sought to determine whether a genetic risk score (GRS) could identify subjects at higher risk for ischemic stroke after accounting for traditional clinical risk factors in 5 trials across the spectrum of cardiometabolic disease.

METHODS

Subjects who had consented for genetic testing and who were of European ancestry from the ENGAGE AF-TIMI 48 (Effective Anticoagulation with Factor Xa Next Generation in Atrial Fibrillation), SOLID-TIMI 52 (Stabilization of Plaques Using Darapladib), SAVOR-TIMI 53 (Saxagliptin Assessment of Vascular Outcomes Recorded in Patients with Diabetes Mellitus), PEGASUS-TIMI 54 (Prevention of Cardiovascular Events in Patients With Prior Heart Attack Using Ticagrelor Compared to Placebo on a Background of Aspirin), and FOURIER (Further Cardiovascular Outcomes Research With PCSK9 Inhibition in Patients With Elevated Risk) trials were included in this analysis. A set of 32 single-nucleotide polymorphisms associated with ischemic stroke was used to calculate a GRS in each patient and identify tertiles of genetic risk. A Cox model was used to calculate hazard ratios for ischemic stroke across genetic risk groups, adjusted for clinical risk factors.

RESULTS

In 51 288 subjects across the 5 trials, a total of 960 subjects had an ischemic stroke over a median follow-up period of 2.5 years. After adjusting for clinical risk factors, a higher GRS was strongly and independently associated with increased risk for ischemic stroke (P trend=0.009). In comparison with individuals in the lowest third of the GRS, individuals in the middle and top tertiles of the GRS had adjusted hazard ratios of 1.15 (95% CI, 0.98-1.36) and 1.24 (95% CI 1.05-1.45) for ischemic stroke, respectively. Stratification into subgroups revealed that the performance of the GRS appeared stronger in the primary prevention cohort with an adjusted hazard ratio for the top versus lowest tertile of 1.27 (95% CI, 1.04-1.53), in comparison with an adjusted hazard ratio of 1.06 (95% CI, 0.81-1.41) in subjects with previous stroke. In an exploratory analysis of patients with atrial fibrillation and CHA₂DS₂-VASc score of 2, high genetic risk conferred a 4-fold higher risk of stroke and an absolute risk equivalent to those with CHA₂DS₂-VASc score of 3.

CONCLUSIONS

Across a broad spectrum of subjects with cardiometabolic disease, a 32-single-nucleotide polymorphism GRS was a strong, independent predictor of ischemic stroke. In patients with atrial fibrillation but lower CHA₂DS₂-VASc scores, the GRS identified patients with risk comparable to those with higher CHA₂DS₂-VASc scores.

Genomic risk score offers predictive performance comparable to clinical risk factors for ischaemic stroke

Recent genome-wide association studies in stroke have enabled the generation of genomic risk scores (GRS) but their predictive power has been modest compared to established stroke risk factors. Here, using a meta-scoring approach, we develop a metaGRS for ischaemic stroke (IS) and analyse this score in the UK Biobank (n = 395,393; 3075 IS events by age 75). The metaGRS hazard ratio for IS (1.26, 95% CI 1.22-1.31 per metaGRS standard deviation) doubles that of a previous GRS, identifying a subset of individuals at monogenic levels of risk: the top 0.25% of metaGRS have three-fold risk of IS. The metaGRS is similarly or more predictive compared to several risk factors, such as family history, blood pressure, body mass index, and smoking. We estimate the reductions needed in modifiable risk factors for individuals with different levels of genomic risk and suggest that, for individuals with high metaGRS, achieving risk factor levels recommended by current guidelines may be insufficient to mitigate risk.

Clinical Variables and Genetic Risk Factors Associated with the Acute Outcome of Ischemic Stroke: A Systematic Review

Stroke is a complex disease and one of the main causes of morbidity and mortality among the adult population. A huge variety of factors is known to influence patient outcome, including demographic variables, comorbidities or genetics. In this review, we expound what is known about the influence of clinical variables and related genetic risk factors on ischemic stroke outcome, focusing on acute and subacute outcome (within 24 to 48 hours after stroke and until day 10, respectively), as they are the first indicators of stroke damage. We searched the PubMed data base for articles that investigated the interaction between clinical variables or genetic factors and acute or subacute stroke outcome. A total of 61 studies were finally included in this review. Regarding the data collected, the variables consistently associated with acute stroke outcome are: glucose levels, blood pressure, presence of atrial fibrillation, prior statin treatment, stroke severity, type of acute treatment performed, severe neurological complications, leukocyte levels, and genetic risk factors. Further research and international efforts are required in this field, which should include genome-wide association studies.

Risks associated with the stroke predisposition at young age: facts and hypotheses in light of individualized predictive and preventive approach

Stroke is one of the most devastating pathologies of the early twenty-first century demonstrating 1-month case-fatality rates ranging from 13 to 35% worldwide. Though the majority of cases do occur in individuals at an advanced age, a persistently increasing portion of the patient cohorts is affected early in life. Current studies provide alarming statistics for the incidence of "young" strokes including adolescents. Young stroke is a multifactorial disease involving genetic predisposition but also a number of modifiable factors, the synergic combination of which potentiates the risks. The article analyzes the prevalence and impacts of "traditional" risk factors such as sedentary lifestyle, smoking, abnormal alcohol consumption, drug abuse, overweight, hypertension, abnormal sleep patterns, and usage of hormonal contraceptives, among others. Further, less explored risks such as primary vascular dysregulation and associated symptoms characteristic for Flammer syndrome (FS) are considered, and the relevance of the FS phenotype for the stroke predisposition at young age is hypothesized. Considering the high prevalence of known genetic and modifiable risk factors in the overall predisposition to the young stroke, the risk mitigating measures are recommended including innovative screening programs by application of specialized questionnaires and biomarker panels as well as educational programs adapted to the target audiences such as children, adolescents, and young adults.

Monogenic, Polygenic, and MicroRNA Markers for Ischemic Stroke

Ischemic stroke (IS) is a leading disease with high mortality and disability, as well as with limited therapeutic window. Biomarkers for earlier diagnosis of IS have long been pursued. Family and twin studies confirm that genetic variations play an important role in IS pathogenesis. Besides DNA mutations found previously by genetic linkage analysis for monogenic IS (Mendelian inheritance), recent studies using genome-wide associated study (GWAS) and microRNA expression profiling have resulted in a large number of DNA and microRNA biomarkers in polygenic IS (sporadic IS), especially in different IS subtypes and imaging phenotypes. The present review summarizes genetic markers discovered by clinical studies and discusses their pathogenic molecular mechanisms involved in developmental or regenerative anomalies of blood vessel walls, neuronal apoptosis, excitotoxic death, inflammation, neurogenesis, and angiogenesis. The possible impact of environment on genetics is addressed as well. We also include a perspective on further studies and clinical application of these IS biomarkers.

Genetic risk, incident stroke, and the benefits of adhering to a healthy lifestyle: cohort study of 306 473 UK Biobank participants

OBJECTIVE

To evaluate the associations of a polygenic risk score and healthy lifestyle with incident stroke.

DESIGN

Prospective population based cohort study.

SETTING

UK Biobank Study, UK.

PARTICIPANTS

306 473 men and women, aged 40-73 years, recruited between 2006 and 2010.

MAIN OUTCOME MEASURE

Hazard ratios for a first stroke, estimated using Cox regression. A polygenic risk score of 90 single nucleotide polymorphisms previously associated with stroke was constructed at P<1×10^-5 to test for an association with incident stroke. Adherence to a healthy lifestyle was determined on the basis of four factors: non-smoker, healthy diet, body mass index <30 kg/m², and regular physical exercise.

RESULTS

During a median follow-up of 7.1 years (2 138 443 person years), 2077 incident strokes (1541 ischaemic stroke, 287 intracerebral haemorrhage, and 249 subarachnoid haemorrhage) were ascertained. The risk of incident stroke was 35% higher among those at high genetic risk (top third of polygenic score) compared with those at low genetic risk (bottom third): hazard ratio 1.35 (95% confidence interval 1.21 to 1.50), P=3.9×10^-8. Unfavourable lifestyle (0 or 1 healthy lifestyle factors) was associated with a 66% increased risk of stroke compared with a favourable lifestyle (3 or 4 healthy lifestyle factors): 1.66 (1.45 to 1.89), P=1.19×10^-13. The association with lifestyle was independent of genetic risk stratums.

CONCLUSION

In this cohort study, genetic and lifestyle factors were independently associated with incident stroke. These results emphasise the benefit of entire populations adhering to a healthy lifestyle, independent of genetic risk.

Traditional and Genetic Risk Score and Stroke Risk Prediction in Korea

BACKGROUND AND OBJECTIVES

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).

METHODS

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).

RESULTS

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.

CONCLUSIONS

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.

[Complex assessment of the contribution of genetic factors to the risk of ischemic stroke]

AIM

To develop a method of the complex assessment of genetic risk for ischemic stroke (IS) and evaluate its effectiveness.

MATERIAL AND METHODS

Genotyping of 182 patients with atherothrombotic and cardioembolic subtypes of IS and 360 healthy individuals of 48 single nucleotide polymorphic loci (SNP) associated with the risk of II and its subtypes was performed.

RESULTS AND CONCLUSION

In each group of SNPs, composite indicators of genetic risk of IS in groups of patients and healthy controls were identified. Differences between the calculated values of the genetic risk in these groups were significant (p <0,05). The quality of the binary classification validated by ROC-analysis confirmed the predictive potential of the proposed method of risk calculation for determining the genetic predisposition to the development of IS.

Body fat distribution and risk of incident ischemic stroke in men and women aged 50 to 74 years from the general population. The KORA Augsburg cohort study

Background

It remains controversial whether measures of general or abdominal adiposity are better risk predictors for ischemic stroke. Furthermore, so far it is unclear whether body fat mass index (BFMI) and fat free mass index (FFMI) are risk predictors for ischemic stroke. This study examined the sex-specific relevance of body mass index (BMI), BROCA Index, waist circumference (WC), waist-height ratio (WHtR), BFMI and FFMI for the development of ischemic stroke in a Caucasian population.

Material and methods

The prospective population-based cohort study was based on 1917 men and 1832 women (aged 50 to 74 years) who participated in the third (1994/95) or fourth (1999/2001) MONICA/KORA Augsburg survey. Subjects were free of stroke at baseline. Standardized anthropometric and bioelectric impedance measurements were obtained at baseline. Hazard ratios (HR) were estimated from Cox proportional hazard models.

Results

During a median follow-up of 9.3 years 128 ischemic strokes occurred in men and 81 in women, respectively. Coded as quartiles WC and WHtR were significantly associated with incident stroke in multivariable analyses in women (comparing the 4^th vs. the bottom quartile), but none of the adiposity measures was significantly associated with incident stroke in multivariable adjusted analyses in men. When anthropometric measures were used as continuous variables, these findings were confirmed. After multivariable adjustment the associations between obesity measures and incident ischemic stroke were statistically significant only for WC (HR 1.39, 95%CI 1.12-1.72) and WHtR in women (HR 1.39, 95%CI 1.12-1.73) per increase of 1 standard deviation. In both sexes the measures BFMI and FFMI were no independent predictors for incident ischemic stroke.

Conclusions

Abdominal obesity measures are independent predictors of incident ischemic stroke in women but not in men from the general adult population. Thus, it may be of particular importance for women to prevent central obesity in order to reduce their risk of ischemic stroke.

Coronary Artery Calcification and Rheumatoid Arthritis: Lack of Relationship to Risk Alleles for Coronary Artery Disease in the General Population

OBJECTIVE

Coronary artery disease (CAD) in the general population is characterized by an increased frequency of particular susceptibility single-nucleotide polymorphisms (SNPs). Because the frequency of CAD is increased among patients with rheumatoid arthritis (RA), we sought to determine whether the frequency of these SNPs is increased in RA patients with CAD, hypothesizing that RA could enhance CAD risk by acting through established genetic pathways predisposing to CAD.

METHODS

Coronary artery calcification (CAC) as detected by computed tomography was used as a measure of CAD in 561 patients with RA. One hundred SNPs associated with CAD in the general population were genotyped or imputed, and their relationship to CAC was established through multiple regression analysis for individual SNPs and a genetic risk score representing their cumulative effect.

RESULTS

Ninety-one CAD-related SNPs were genotyped successfully; of these, 81 exhibited no association with CAC (Agatston units) or different CAC categorizations, either individually or collectively, in the genetic risk score. Only rs579459 (ABO) and rs17676451 (HAL) had a consistent positive association between genotype and CAC, with a significant increase in the frequency of the effect allele in both homozygous and heterozygous genotype distributions. Five were variably negatively associated. Furthermore, a positive association between the Disease Activity Score in 28 joints and CAC was observed, and after adjustment for traditional cardiovascular risk factors, it was not modified by correcting for the CAD-related SNP genetic risk score.

CONCLUSION

The increased risk of CAC in patients with RA does not appear to operate primarily through established genetically regulated atherogenic mechanisms that are preponderant in the general population.

Utilizing nutritional genomics to tailor diets for the prevention of cardiovascular disease: a guide for upcoming studies and implementations

INTRODUCTION

Personalized diets based on an individual's genome to optimize the success of dietary intervention and reduce genetic cardiovascular disease (CVD) risk, is one of the challenges most frequently discussed in the scientific community. Areas covered: The authors gathered literature-based evidence on nutritional genomics and CVD phenotypes, our own results and research experience to provide a critical overview of the current situation of using nutritional genomics to tailor diets for CVD prevention and to propose guidelines for future studies and implementations. Expert commentary: Hundreds of studies on gene-diet interactions determining CVD intermediate (plasma lipids, hypertension, etc.) and final phenotypes (stroke, etc.) have furnished top-level scientific evidence for claiming that the genetic effect in cardiovascular risk is not deterministic, but can be modified by diet. However, despite the many results obtained, there are still gaps in practically applying a personalized diet design to specific genotypes. Hence, a better systemization and methodological improvement of new studies is required to obtain top-level evidence that will allow their application in the future precision nutrition/medicine. The authors propose several recommendations for tackling new approaches and applications.

Revised Framingham Stroke Risk Profile to Reflect Temporal Trends

BACKGROUND

Age-adjusted stroke incidence has decreased over the past 50 years, likely as a result of changes in the prevalence and impact of various stroke risk factors. An updated version of the Framingham Stroke Risk Profile (FSRP) might better predict current risks in the FHS (Framingham Heart Study) and other cohorts. We compared the accuracy of the standard (old) and of a revised (new) version of the FSRP in predicting the risk of all-stroke and ischemic stroke and validated this new FSRP in 2 external cohorts, the 3C (3 Cities) and REGARDS (Reasons for Geographic and Racial Differences in Stroke) studies.

METHODS

We computed the old FSRP as originally described and a new model that used the most recent epoch-specific risk factor prevalence and hazard ratios for individuals ≥55 years of age and for the subsample ≥65 years of age (to match the age range in REGARDS and 3C studies, respectively) and compared the efficacy of these models in predicting 5- and 10-year stroke risks.

RESULTS

The new FSRP was a better predictor of current stroke risks in all 3 samples than the old FSRP (calibration χ2 of new/old FSRP: in men: 64.0/12.1, 59.4/30.6, and 20.7/12.5; in women: 42.5/4.1, 115.4/90.3, and 9.8/6.5 in FHS, REGARDS, and 3C, respectively). In the REGARDS, the new FSRP was a better predictor among whites compared with blacks.

CONCLUSIONS

A more contemporaneous, new FSRP better predicts current risks in 3 large community samples and could serve as the basis for examining geographic and racial differences in stroke risk and the incremental diagnostic utility of novel stroke risk factors.

Genetics of stroke in a UK African ancestry case-control study: South London Ethnicity and Stroke Study

Objective:

Despite epidemiologic data showing an increased stroke incidence in African ancestry populations, genetic studies in this group have so far been limited, and there has been little characterization of the genetic contribution to stroke liability in this population, particularly for stroke subtypes.

Methods:

We evaluated the evidence that genetic factors contribute to stroke and stroke subtypes in a population of 917 African and African Caribbean stroke cases and 868 matched controls from London, United Kingdom. We (1) estimated the heritability of stroke in this population using genomic-relatedness matrix-restricted maximum likelihood approaches, (2) assessed loci associated with stroke in Europeans in our population, and (3) evaluated the influence of genetic factors underlying cardiovascular risk factors on stroke using polygenic risk scoring.

Results:

Our results indicate a substantial genetic contribution to stroke risk in African ancestry populations (h² = 0.35 [SE = 0.19], p = 0.043). Polygenic risk scores indicate that cardiovascular risk scores contribute to the genetic liability (odds ratio [OR] 1.09 [95% confidence interval (CI) 1.01–1.17], p = 0.029) and point to a strong influence of type 2 diabetes in large vessel stroke (OR 1.62 [95% CI 1.19–2.22], p = 0.0024). Single nucleotide polymorphisms associated with ischemic stroke in Europeans shared direction of effect in SLESS (p = 0.031), suggesting that disease mechanisms are shared across ancestries.

Conclusions:

Stroke in African ancestry populations is highly heritable and influenced by genetic determinants underlying cardiovascular risk factors. In addition, stroke loci identified in Europeans share direction of effect in African populations. Future genome-wide association studies must focus on incorporating African ancestry individuals.

Multiple Genetic Associations with Irish Wolfhound Dilated Cardiomyopathy

Cardiac disease is a leading cause of morbidity and mortality in dogs and humans, with dilated cardiomyopathy being a large contributor to this. The Irish Wolfhound (IWH) is one of the most commonly affected breeds and one of the few breeds with genetic loci associated with the disease. Mutations in more than 50 genes are associated with human dilated cardiomyopathy (DCM), yet very few are also associated with canine DCM. Furthermore, none of the identified canine loci explain many cases of the disease and previous work has indicated that genotypes at multiple loci may act together to influence disease development. In this study, loci previously associated with DCM in IWH were tested for associations in a new cohort both individually and in combination. We have identified loci significantly associated with the disease individually, but no genotypes individually or in pairs conferred a significantly greater risk of developing DCM than the population risk. However combining three loci together did result in the identification of a genotype which conferred a greater risk of disease than the overall population risk. This study suggests multiple rather than individual genetic factors, cooperating to influence DCM risk in IWH.

Beyond Endometriosis Genome-Wide Association Study: From Genomics to Phenomics to the Patient

Endometriosis is a heritable, complex chronic inflammatory disease, for which much of the causal pathogenic mechanism remains unknown. Genome-wide association studies (GWAS) to date have identified 12 single nucleotide polymorphisms at 10 independent genetic loci associated with endometriosis. Most of these were more strongly associated with revised American Fertility Society stage III/IV, rather than stage I/II. The loci are almost all located in intergenic regions that are known to play a role in the regulation of expression of target genes yet to be identified. To identify the target genes and pathways perturbed by the implicated variants, studies are required involving functional genomic annotation of the surrounding chromosomal regions, in terms of transcription factor binding, epigenetic modification (e.g., DNA methylation and histone modification) sites, as well as their correlation with RNA transcription. These studies need to be conducted in tissue types relevant to endometriosis-in particular, endometrium. In addition, to allow biologically and clinically relevant interpretation of molecular profiling data, they need to be combined and correlated with detailed, systematically collected phenotypic information (surgical and clinical). The WERF Endometriosis Phenome and Biobanking Harmonisation Project is a global standardization initiative that has produced consensus data and sample collection protocols for endometriosis research. These now pave the way for collaborative studies integrating phenomic with genomic data, to identify informative subtypes of endometriosis that will enhance understanding of the pathogenic mechanisms of the disease and discovery of novel, targeted treatments.

Third European Stroke Science Workshop

Lake Eibsee, Garmisch-Partenkirchen, November 19 to 21, 2015: The European Stroke Organization convened >120 stroke experts from 27 countries to discuss latest results and hot topics in clinical, translational, and basic stroke research. Since its inception in 2011, the European Stroke Science Workshop has become a cornerstone of European Stroke Organization's academic activities and major highlight for researchers in the field. Participants include stroke researchers at all career stages who convene for plenary lectures and discussions, thus facilitating crosstalk among researchers from different fields. As in previous years, the workshop was organized into 7 scientific sessions each focusing on a major research topic. All sessions started with a keynote lecture that provided an overview on current developments and set the scene for the following presentations. The latter were short focused talks on a timely topic and included the most recent findings, including unpublished data. A new element at this year's meeting was a hot topic session in which speakers had to present a provocative concept or update sharply within 5 minutes. In the following, we summarize the key contents of the meeting. The program is provided in the online-only Data Supplement.

Epidemiology of cardiovascular disease: recent novel outlooks on risk factors and clinical approaches

INTRODUCTION

Cardiovascular (CVD) risk assessment with traditional risk factors (age, sex, blood pressure, lipids, smoking and diabetes) has remained relatively invariant over the past decades despite some inaccuracies associated with this approach. However, the search for novel, robust and cost-effective risk markers of CVD risk is ongoing.

AREAS COVERED

A large share of the major developments in CVD risk prediction during the past five years has been made in large-scale biomarker discovery and the so called 'omics' - the rapidly growing fields of genomics, transcriptomics, epigenetics and metabolomics. This review focuses on how these new technologies are helping drive primary CVD risk estimation forward in recent years, and speculates on how they could be utilized more effectively for discovering novel risk factors in the future. Expert commentary: The search for new CVD risk factors is currently undergoing a significant revolution as the simple relationship between single risk factors and disease will have to be replaced by models that strive to integrate the whole field of omics into medicine.

Genetic Associations With White Matter Hyperintensities Confer Risk of Lacunar Stroke

Supplemental Digital Content is available in the text.

Background and Purpose—

White matter hyperintensities (WMH) are increased in patients with lacunar stroke. Whether this is because of shared pathogenesis remains unknown. Using genetic data, we evaluated whether WMH-associated genetic susceptibility factors confer risk of lacunar stroke, and therefore whether they share pathogenesis.

Methods—

We used a genetic risk score approach to test whether single nucleotide polymorphisms associated with WMH in community populations were associated with magnetic resonance imaging–confirmed lacunar stroke (n=1,373), as well as cardioembolic (n=1,331) and large vessel (n=1,472) Trial of Org 10172 in Acute Stroke Treatment subtypes, against 9,053 controls. Second, we separated lacunar strokes into those with WMH (n=568) and those without (n=787) and tested for association with the risk score in these 2 groups. In addition, we evaluated whether WMH-associated single nucleotide polymorphisms are associated with lacunar stroke, or in the 2 groups.

Results—

The WMH genetic risk score was associated with lacunar stroke (odds ratio [OR; 95% confidence interval [CI]]=1.14 [1.06–1.22]; P=0.0003), in patients both with and without WMH (WMH: OR [95% CI]=1.15 [1.05–1.26]; P=0.003 and no WMH: OR [95% CI]=1.11 [1.02–1.21]; P=0.019). Conversely, the risk score was not associated with cardioembolic stroke (OR [95% CI]=1.03 [0.97–1.09]; P=0.63) or large vessel stroke (OR [95% CI]=0.99 [0.93,1.04]; P=0.39). However, none of the WMH-associated single nucleotide polymorphisms passed Bonferroni-corrected significance for association with lacunar stroke.

Conclusions—

Genetic variants that influence WMH are associated with an increased risk of lacunar stroke but not cardioembolic or large vessel stroke. Some genetic susceptibility factors seem to be shared across different radiological manifestations of small vessel disease.

Improved prediction of complex diseases by common genetic markers: state of the art and further perspectives

Reliable risk assessment of frequent, but treatable diseases and disorders has considerable clinical and socio-economic relevance. However, as these conditions usually originate from a complex interplay between genetic and environmental factors, precise prediction remains a considerable challenge. The current progress in genotyping technology has resulted in a substantial increase of knowledge regarding the genetic basis of such diseases and disorders. Consequently, common genetic risk variants are increasingly being included in epidemiological models to improve risk prediction. This work reviews recent high-quality publications targeting the prediction of common complex diseases. To be included in this review, articles had to report both, numerical measures of prediction performance based on traditional (non-genetic) risk factors, as well as measures of prediction performance when adding common genetic variants to the model. Systematic PubMed-based search finally identified 55 eligible studies. These studies were compared with respect to the chosen approach and methodology as well as results and clinical impact. Phenotypes analysed included tumours, diabetes mellitus, and cardiovascular diseases. All studies applied one or more statistical measures reporting on calibration, discrimination, or reclassification to quantify the benefit of including SNPs, but differed substantially regarding the methodological details that were reported. Several examples for improved risk assessments by considering disease-related SNPs were identified. Although the add-on benefit of including SNP genotyping data was mostly moderate, the strategy can be of clinical relevance and may, when being paralleled by an even deeper understanding of disease-related genetics, further explain the development of enhanced predictive and diagnostic strategies for complex diseases.

Boosting for high-dimensional two-class prediction

Background

In clinical research prediction models are used to accurately predict the outcome of the patients based on some of their characteristics. For high-dimensional prediction models (the number of variables greatly exceeds the number of samples) the choice of an appropriate classifier is crucial as it was observed that no single classification algorithm performs optimally for all types of data. Boosting was proposed as a method that combines the classification results obtained using base classifiers, where the sample weights are sequentially adjusted based on the performance in previous iterations. Generally boosting outperforms any individual classifier, but studies with high-dimensional data showed that the most standard boosting algorithm, AdaBoost.M1, cannot significantly improve the performance of its base classier. Recently other boosting algorithms were proposed (Gradient boosting, Stochastic Gradient boosting, LogitBoost); they were shown to perform better than AdaBoost.M1 but their performance was not evaluated for high-dimensional data.

Results

In this paper we use simulation studies and real gene-expression data sets to evaluate the performance of boosting algorithms when data are high-dimensional. Our results confirm that AdaBoost.M1 can perform poorly in this setting, often failing to improve the performance of its base classifier. We provide the explanation for this and propose a modification, AdaBoost.M1.ICV, which uses cross-validated estimates of the prediction errors and outperforms the original algorithm when data are high-dimensional. The use of AdaBoost.M1.ICV is advisable when the base classifier overfits the training data: the number of variables is large, the number of samples is small, and/or the difference between the classes is large. To a lesser extent also Gradient boosting suffers from similar problems. Contrary to the findings for the low-dimensional data, shrinkage does not improve the performance of Gradient boosting when data are high-dimensional, however it is beneficial for Stochastic Gradient boosting, which outperformed the other boosting algorithms in our analyses. LogitBoost suffers from overfitting and generally performs poorly.

Conclusions

The results show that boosting can substantially improve the performance of its base classifier also when data are high-dimensional. However, not all boosting algorithms perform equally well. LogitBoost, AdaBoost.M1 and Gradient boosting seem less useful for this type of data. Overall, Stochastic Gradient boosting with shrinkage and AdaBoost.M1.ICV seem to be the preferable choices for high-dimensional class-prediction.

Electronic supplementary material

The online version of this article (doi:10.1186/s12859-015-0723-9) contains supplementary material, which is available to authorized users.

Current Applications of Genetic Risk Scores to Cardiovascular Outcomes and Subclinical Phenotypes

Genetic risk scores are a useful tool for examining the cumulative predictive ability of genetic variation on cardiovascular disease. Important considerations for creating genetic risk scores include the choice of genetic variants, weighting, and comparability across ethnicities. Genetic risk scores that use information from genome-wide meta-analyses can successfully predict cardiovascular outcomes and subclinical phenotypes, yet there is limited clinical utility of these scores beyond traditional cardiovascular risk factors in many populations. Novel uses of genetic risk scores include evaluating the genetic contribution of specific intermediate traits or risk factors to cardiovascular disease, risk prediction in high-risk populations, gene-by-environment interaction studies, and Mendelian randomization studies. Though questions remain about the ultimate clinical utility of the genetic risk score, further investigation in high-risk populations and new ways to combine genetic risk scores with traditional risk factors may prove to be fruitful.

Genomic risk prediction of complex human disease and its clinical application

Recent advances in genome-wide association studies have stimulated interest in the genomic prediction of disease risk, potentially enabling individual-level risk estimates for early intervention and improved diagnostic procedures. Here, we review recent findings and approaches to genomic prediction model construction and performance, then contrast the potential benefits of such models in two complex human diseases, aiding diagnosis in celiac disease and prospective risk prediction for cardiovascular disease. Early indications are that optimal application of genomic risk scores will differ substantially for each disease depending on underlying genetic architecture as well as current clinical and public health practice. As costs decline, genomic profiles become common, and popular understanding of risk and its communication improves, genomic risk will become increasingly useful for the individual and the clinician.

A colorectal cancer prediction model using traditional and genetic risk scores in Koreans

Background

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).

Results

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.

Conclusion

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.

Electronic supplementary material

The online version of this article (doi:10.1186/s12863-015-0207-y) contains supplementary material, which is available to authorized users.

Genetic diagnosis and prognosis of Alzheimer's disease: challenges and opportunities

Alzheimer's disease (AD), the most common form of dementia in western societies, is a pathologically and clinically heterogeneous disease with a strong genetic component. The recent advances in high-throughput genome technologies allowing for the rapid analysis of millions of polymorphisms in thousands of subjects has significantly advanced our understanding of the genomic underpinnings of AD susceptibility. During the last 5 years, genome-wide association and whole-exome- and whole-genome sequencing studies have mapped more than 20 disease-associated loci, providing insights into the molecular pathways involved in AD pathogenesis and hinting at potential novel therapeutic targets. This review article summarizes the challenges and opportunities of when using genomic information for the diagnosis and prognosis of AD.

Genetics of cerebral small vessel disease

Cerebral small vessel disease (SVD) is an important cause of stroke and cognitive impairment among the elderly and is a more frequent cause of stroke in Asia than in the US or Europe. Although traditional risk factors such as hypertension or diabetes mellitus are important in the development of cerebral SVD, the exact pathogenesis is still uncertain. Both, twin and family history studies suggest heritability of sporadic cerebral SVD, while the candidate gene study and the genome-wide association study (GWAS) are mainly used in genetic research. Robust associations between the candidate genes and occurrence of various features of sporadic cerebral SVD, such as lacunar infarction, intracerebral hemorrhage, or white matter hyperintensities, have not yet been elucidated. GWAS, a relatively new technique, overcomes several shortcomings of previous genetic techniques, enabling the detection of several important genetic loci associated with cerebral SVD. In addition to the more common, sporadic cerebral SVD, several single-gene disorders causing cerebral SVD have been identified. The number of reported cases is increasing as the clinical features become clear and diagnostic examinations are more readily available. These include cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy, cerebral autosomal recessive arteriopathy with subcortical infarcts and leukoencephalopathy, COL4A1-related cerebral SVD, autosomal dominant retinal vasculopathy with cerebral leukodystrophy, and Fabry disease. These rare single-gene disorders are expected to play a crucial role in our understanding of cerebral SVD pathogenesis by providing animal models for the identification of cellular, molecular, and biochemical changes underlying cerebral small vessel damage.

Genetics of ischaemic stroke in young adults

Background

Stroke may be a clinical expression of several inherited disorders in humans. Recognition of the underlined genetic disorders causing stroke is important for a correct diagnosis, for genetic counselling and, even if rarely, for a correct therapeutic management. Moreover, the genetics of complex diseases such the stroke, in which multiple genes interact with environmental risk factors to increase risk, has been revolutionized by the Genome-Wide Association Study (GWAS) approach.

Scope of review

Here we review the single-gene causes of ischemic stroke, bringing the reader from the candidate gene method toward the exciting new horizons of genetic technology.

Major conclusions

The aetiological diagnosis of ischemic stroke in young adults is more complex than in the elderly. The identification of a genetic cause is important to provide appropriate counseling and to start a correct therapy, when available. The advent of GWAS technology, such as for other complex pathological conditions, has contributed enormously to the understanding of many of these genetic bases. For success large, well phenotyped case cohorts are required, and international collaborations are essential.

General significance

This review focuses on the main causes of genetically-based ischemic stroke in young adults, often classified as indeterminate, investigating also the recent findings of the GWAS, in order to improve diagnostic and therapeutic management.

•

The aetiological diagnosis of stroke in young adults needs a different and more complex diagnostic work up than in older adults.

•

Stroke may be a clinical expression of several inherited disorders in humans.

•

The most common genetic causes of stroke are CADASIL, Fabry and mitochondrial diseases.

•

Recognition of the underlined genetic disorders causing stroke is important for the correct management of the patient.

A genetic risk score for hypertension associates with the risk of ischemic stroke in a Swedish case-control study

Genetic risk scores (GRS), summing up the total effect of several single-nucleotide polymorphisms (SNPs) in genes associated with either coronary risk or cardiovascular risk factors, have been tested for association with ischemic stroke with conflicting results. Recently an association was found between a GRS based on 29 SNPs discovered by genome-wide association studies and hypertension. The aim of our study was to investigate the possible association of the same GRS with ischemic stroke on top of other 'traditional risk factors', also testing its potential improvement in indices of discrimination and reclassification, in a Swedish case-control study. Twenty-nine SNPs were genotyped in 3677 stroke cases and 2415 controls included in the Lund Stroke Register (LSR), the Malmö Diet and Cancer (MDC) study and the Sahlgrenska Academy Study on Ischemic Stroke (SAHLSIS). The analysis was conducted in the combined sample, and separately for the three studies. After adjustment for hypertension, diabetes mellitus and smoking habits, the GRS was associated with ischemic stroke in the combined sample (OR (95% CI) 1.086 (1.029-1.147) per SD increase in the GRS P=0.003) with similar trends in all three samples: LSR (1.050 (0.967-1.140); P=0.25), MDC (1.168 (1.060-1.288); P=0.002) and SAHLSIS (1.124 (0.997-1.267); P=0.055). Measures of risk discrimination and reclassification improved marginally using the GRS. A blood pressure GRS is independently associated with ischemic stroke risk in three Swedish case-control studies, however, the effect size is low and adds marginally to prediction of stroke on top of traditional risk factors including hypertension.

[Prediction in cerebrovascular diseases]

Prediction of the outcome of cerebrovascular diseases or of the effects and complications of various forms of treatment are essential components of all stroke treatment regimens. This review focuses on the prediction of the stroke risk in primary prevention, the prediction of the risk of secondary stroke following a transient ischemic attack (TIA), the estimation of the outcome following manifest stroke and the treatment effects, the prediction of secondary cerebrovascular events and the prediction of vascular cognitive impairment following stroke. All predictive activities in cerebrovascular disease are hindered by the translation of predictive results from studies and patient populations to the individual patient. Future efforts in genetic analyses may be able to overcome this barrier and to enable individual prediction in the area of so-called personalized medicine. In all the various fields of prediction in cerebrovascular diseases, three major variables are always important: age of the patient, severity and subtype of the stroke. Increasing age, more severe stroke symptoms and the cardioembolic stroke subtype predict a poor outcome regarding both survival and permanent disability. This finding is somewhat banal and will therefore never replace the well experienced clinician judging the chances of a patient and taking into account the personal situation of this patient, e.g. for initiation of a rehabilitation program. Besides the individualized prediction, in times of restricted economic resources and increasing tendency to clarify questions of medical treatment in court, it seems unavoidable to use prediction in economic and medicolegal interaction with clinical medicine. This tendency will be accompanied by difficult ethical problems which neurologists must be aware of. Improved prediction should not be used to allocate or restrict resources or to restrict medically indicated treatment.