Integration of Biomarker Polygenic Risk Score Improves Prediction of Coronary Heart Disease

Meta-prediction of coronary artery disease risk

Coronary artery disease (CAD) is a leading cause of morbidity and mortality worldwide, and accurately predicting individual risk is critical for prevention. Here we aimed to integrate unmodifiable risk factors, such as age and genetics, with modifiable risk factors, such as clinical and biometric measurements, into a meta-prediction framework that produces actionable and personalized risk estimates. In the initial development of the model, ~2,000 predictive features were considered, including demographic data, lifestyle factors, physical measurements, laboratory tests, medication usage, diagnoses and genetics. To power our meta-prediction approach, we stratified the UK Biobank into two primary cohorts: first, a prevalent CAD cohort used to train predictive models for cross-sectional prediction at baseline and prospective estimation of contributing risk factor levels and diagnoses (baseline models) and, second, an incident CAD cohort using, in part, these baseline models as meta-features to train a final CAD incident risk prediction model. The resultant 10-year incident CAD risk model, composed of 15 derived meta-features with multiple embedded polygenic risk scores, achieves an area under the curve of 0.84. In an independent test cohort from the All of Us research program, this model achieved an area under the curve of 0.81 for predicting 10-year incident CAD risk, outperforming standard clinical scores and previously developed integrative models. Moreover, this framework enables the generation of individualized risk reduction profiles by quantifying the potential impact of standard clinical interventions. Notably, genetic risk influences the extent to which these interventions reduce overall CAD risk, allowing for tailored prevention strategies.

A framework for detecting causal effects of risk factors at an individual level based on principles of Mendelian randomisation: applications to modelling individualised effects of lipids on coronary artery disease

BACKGROUND

Mendelian Randomisation (MR) has been widely used to study the causal effects of risk factors. However, almost all MR studies concentrate on the population's average causal effects. With the advent of precision medicine, the individualised treatment effect (ITE) is often of greater interest. For instance, certain risk factors may pose a higher risk to some individuals than others, and the benefits of treatments may vary across individuals. This study proposes a framework for estimating individualised causal effects in large-scale observational studies where unobserved confounding factors may be present.

METHODS

We propose a framework (MR-ITE) that expands the scope of MR from estimating average causal effects to individualised causal effects. We present several approaches for estimating ITEs within this MR framework, primarily grounded on the principles of the "R-learner". To evaluate the presence of causal effect heterogeneity, we also proposed two permutation testing methods. We employed polygenic risk score (PRS) as instruments and proposed methods to improve the accuracy of ITE estimates by removal of potentially pleiotropic single nucleotide polymorphisms (SNPs). The validity of our approach was substantiated through comprehensive simulations. The proposed framework also allows the identification of important effect modifiers contributing to individualised differences in treatment effects. We applied our framework to study the individualised causal effects of various lipid traits, including low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C), triglycerides (TG), and total cholesterol (TC), on the risk of coronary artery disease (CAD) based on the UK-Biobank (UKBB). We also studied the ITE of C-reactive protein (CRP) and insulin-like growth factor 1 (IGF-1) on CAD as secondary analyses.

FINDINGS

Simulation studies demonstrated that MR-ITE outperformed traditional causal forest approaches in identifying ITEs when unobserved confounders were present. The integration of the contamination mixture (ConMix) approach to remove invalid pleiotropic SNPs further enhanced MR-ITE's performance. In real-world applications, we identified positive causal associations between CAD and several factors (LDL-C, Total Cholesterol, and IGF-1 levels). Our permutation tests revealed significant heterogeneity in these causal associations across individuals. Using Shapley value analysis, we identified the top effect modifiers contributing to this heterogeneity.

INTERPRETATION

We introduced a new framework, MR-ITE, capable of inferring individualised causal effects in observational studies based on the MR approach, utilizing PRS as instruments. MR-ITE extends the application of MR from estimating the average treatment effect to individualised treatment effects. Our real-world application of MR-ITE underscores the importance of identifying ITEs in the context of precision medicine.

FUNDING

This work was supported partially by a National Natural Science Foundation of China grant (NSFC; grant number 81971706), the KIZ-CUHK Joint Laboratory of Bioresources and Molecular Research of Common Diseases, Kunming Institute of Zoology and The Chinese University of Hong Kong, China, and the Lo Kwee Seong Biomedical Research Fund from The Chinese University of Hong Kong.

Association of genetically predicted levels of circulating blood lipids with coronary artery disease incidence

BACKGROUND AND AIMS

Estimating the genetic risk of coronary artery disease (CAD) is now possible by aggregating data from genome-wide association studies (GWAS) into polygenic risk scores (PRS). Combining multiple PRS for specific circulating blood lipids could improve risk prediction. Here, we sought to evaluate the performance of PRS derived from CAD and blood lipids GWAS to predict the incidence of CAD.

METHODS

This study included individuals aged between 40 and 69 from UK Biobank. We conducted GWAS for blood lipids measured by nuclear magnetic resonance in individuals without lipid-lowering treatments (n = 73,915). Summary statistics were used to derive PRS in the remaining participants (n = 318,051). A PRSCAD was derived using the CARDIoGRAMplusC4D GWAS. Hazard ratios (HR) for CAD (n = 9017 out of 301,576; median follow-up: 12.6 years) were calculated per standard deviation increase in each PRS. Models' discrimination capacity and goodness-of-fit were evaluated.

RESULTS

Out of 30 PRS, 27 were significantly associated with the incidence of CAD (p < 0.0017). The optimal combination of PRS included PRS for CAD, VLDL-C, total cholesterol and triglycerides. Discriminative capacities were significantly increased in the model including PRSCAD and clinical risk factors (CRF) (C-statistic = 0.778 [0.773-0.782]) compared to the model with CRF only (C-statistic = 0.755 [0.751-0.760], difference = 0.022 [0.020-0.025]). Although the C-statistic remained similar when independent lipids PRS were added to the model with PRSCAD and CRF (C-statistic = 0.778 [0.773-0.783]), the goodness-of-fit was significantly increased (chi-square test statistic = 20.18, p = 1.56e-04).

CONCLUSIONS

Although independently associated with CAD incidence, blood lipids PRS provide modest improvement in the predictive performance when added to PRSCAD.

Exploring autism spectrum disorder and co-occurring trait associations to elucidate multivariate genetic mechanisms and insights

BACKGROUND

Autism spectrum disorder (ASD) is a partially heritable neurodevelopmental trait, and people with ASD may also have other co-occurring trait such as ADHD, anxiety disorders, depression, mental health issues, learning difficulty, physical health traits and communication challenges. The concomitant development of ASD and other neurological traits is assumed to result from a complex interplay between genetics and the environment. However, only a limited number of studies have performed multivariate genome-wide association studies (GWAS) for ASD.

METHODS

We conducted to-date the largest multivariate GWAS on ASD and 8 ASD co-occurring traits (ADHD, ADHD childhood, anxiety stress (ASDR), bipolar (BIP), disruptive behaviour (DBD), educational attainment (EA), major depression, and schizophrenia (SCZ)) using summary statistics from leading studies. Multivariate associations and central traits were further identified. Subsequently, colocalization and Mendelian randomization (MR) analysis were performed on the associations identified with the central traits containing ASD. To further validate our findings, pathway and quantified trait loci (QTL) resources as well as independent datasets consisting of 112 (45 probands) whole genome sequence data from the GEMMA project were utilized.

RESULTS

Multivariate GWAS resulted in 637 significant associations (p < 5e-8), among which 322 are reported for the first time for any trait. 37 SNPs were identified to contain ASD and one or more traits in their central trait set, including variants mapped to known SFARI ASD genes MAPT, CADPS and NEGR1 as well as novel ASD genes KANSL1, NSF and NTM, associated with immune response, synaptic transmission, and neurite growth respectively. Mendelian randomization analyses found that genetic liability for ADHD childhood, ASRD and DBT has causal effects on the risk of ASD while genetic liability for ASD has causal effects on the risk of ADHD, ADHD childhood, BIP, WA, MDD and SCZ. Frequency differences of SNPs found in NTM and CADPS genes, respectively associated with neurite growth and neural/endocrine calcium regulation, were found between GEMMA ASD probands and controls. Pathway, QTL and cell type enrichment implicated microbiome, enteric inflammation, and central nervous system enrichments.

CONCLUSIONS

Our study, combining multivariate GWAS with systematic decomposition, identified novel genetic associations related to ASD and ASD co-occurring driver traits. Statistical tests were applied to discern evidence for shared and interpretable liability between ASD and co-occurring traits. These findings expand upon the current understanding of the complex genetics regulating ASD and reveal insights of neuronal brain disruptions potentially driving development and manifestation.

Population Performance and Individual Agreement of Coronary Artery Disease Polygenic Risk Scores

Importance

Polygenic risk scores (PRSs) for coronary artery disease (CAD) are a growing clinical and commercial reality. Whether existing scores provide similar individual-level assessments of disease liability is a critical consideration for clinical implementation that remains uncharacterized.

Objective

Characterize the reliability of CAD PRSs that perform equivalently at the population level at predicting individual-level risk.

Design

Cross-sectional Study.

Setting

All of Us Research Program (AOU), Penn Medicine Biobank (PMBB), and UCLA ATLAS Precision Health Biobank.

Participants

Volunteers of diverse genetic backgrounds enrolled in AOU, PMBB, and UCLA with available electronic health record and genotyping data.

Exposures

Polygenic risk for CAD from previously published PRSs and new PRSs developed separately from the testing cohorts.

Main Outcomes and Measures

Sets of CAD PRSs that perform population prediction equivalently were identified by comparing calibration and discrimination (Brier score and AUROC) of generalized linear models of prevalent CAD using Bayesian analysis of variance. Among equivalently performing scores, individual-level agreement between risk estimates was tested with intraclass correlation (ICC) and Light's Kappa, measures of inter-rater reliability.

Results

50 PRSs were calculated for 171,095 AOU participants. When included in a model of prevalent CAD, 48 scores had practically equivalent Brier scores and AUROCs (region of practical equivalence = 0.02). Across these scores, 84% of participants had at least one score in both the top and bottom risk quintile. Continuous agreement of individual risk predictions from the 48 scores was poor, with an ICC of 0.351 (95% CI; 0.349, 0.352). Agreement between two statistically equivalent scores was moderate, with an ICC of 0.649 (95% CI; 0.646, 0.652). Light's Kappa, used to evaluate consistency of assignment to high-risk thresholds, did not exceed 0.56 (interpreted as 'fair') across statistically and practically equivalent scores. Repeating the analysis among 41,193 PMBB and 50,748 UCLA participants yielded different sets of statistically and practically equivalent scores which also lacked strong individual agreement.

Conclusions and Relevance

Across three diverse biobanks, CAD PRSs that performed equivalently at the population level produced unreliable individual risk estimates. Approaches to clinical implementation of CAD PRSs must consider the potential for discordant individual risk estimates from otherwise indistinguishable scores.

Multi-organ imaging-derived polygenic indexes for brain and body health

The UK Biobank (UKB) imaging project is a crucial resource for biomedical research, but is limited to 100,000 participants due to cost and accessibility barriers. Here we used genetic data to predict heritable imaging-derived phenotypes (IDPs) for a larger cohort. We developed and evaluated 4,375 IDP genetic scores (IGS) derived from UKB brain and body images. When applied to UKB participants who were not imaged, IGS revealed links to numerous phenotypes and stratified participants at increased risk for both brain and somatic diseases. For example, IGS identified individuals at higher risk for Alzheimer's disease and multiple sclerosis, offering additional insights beyond traditional polygenic risk scores of these diseases. When applied to independent external cohorts, IGS also stratified those at high disease risk in the All of Us Research Program and the Alzheimer's Disease Neuroimaging Initiative study. Our results demonstrate that, while the UKB imaging cohort is largely healthy and may not be the most enriched for disease risk management, it holds immense potential for stratifying the risk of various brain and body diseases in broader external genetic cohorts.

Integration of risk factor polygenic risk score with disease polygenic risk score for disease prediction

Polygenic risk score (PRS) is useful for capturing an individual's genetic susceptibility. However, previous studies have not fully exploited the potential of the risk factor PRS (RFPRS) for disease prediction. We explored the potential of integrating disease-related RFPRSs with disease PRS to enhance disease prediction performance. We constructed 112 RFPRSs and analyzed the association of RFPRSs with diseases to identify disease-related RFPRSs in 700 diseases, using the UK Biobank dataset. We uncovered 6157 statistically significant associations between 247 diseases and 109 RFPRSs. We estimated the disease PRSs of 70 diseases that exhibited statistically significant heritability, to generate RFDiseasemetaPRS-a combined PRS integrating RFPRSs and disease PRS-and compare the prediction performance metrics between RFDiseasemetaPRS and disease PRS. RFDiseasemetaPRS showed better performance for Nagelkerke's pseudo-R2, odds ratio (OR) per 1 SD, net reclassification improvement (NRI) values and difference of R2 considered by variance of R2 in 31 out of 70 diseases. Additionally, we assessed risk classification between two models by examining OR between the top 10% and remaining 90% individuals for the 31 diseases; RFDiseasemetaPRS exhibited better R2, NRI and OR than disease PRS. These findings highlight the importance of utilizing RFDiseasemetaPRS, which can provide personalized healthcare and tailored prevention strategies.