Skip to main content
PMC home page
PLOS Medicine logoLink to PLOS Medicine
. 2021 Jan 14;18(1):e1003498. doi: 10.1371/journal.pmed.1003498

Polygenic risk scores in cardiovascular risk prediction: A cohort study and modelling analyses

Luanluan Sun 1,#, Lisa Pennells 1,#, Stephen Kaptoge 1,#, Christopher P Nelson 2, Scott C Ritchie 1,3, Gad Abraham 1,3, Matthew Arnold 1, Steven Bell 1, Thomas Bolton 1, Stephen Burgess 1, Frank Dudbridge 2,4, Qi Guo 1, Eleni Sofianopoulou 1, David Stevens 1, John R Thompson 2, Adam S Butterworth 1, Angela Wood 1, John Danesh 1,5,‡, Nilesh J Samani 2,4,‡, Michael Inouye 1,3,6,‡,*, Emanuele Di Angelantonio 1,‡,*
Editor: George Hindy7
PMCID: PMC7808664  PMID: 33444330

Abstract

Background

Polygenic risk scores (PRSs) can stratify populations into cardiovascular disease (CVD) risk groups. We aimed to quantify the potential advantage of adding information on PRSs to conventional risk factors in the primary prevention of CVD.

Methods and findings

Using data from UK Biobank on 306,654 individuals without a history of CVD and not on lipid-lowering treatments (mean age [SD]: 56.0 [8.0] years; females: 57%; median follow-up: 8.1 years), we calculated measures of risk discrimination and reclassification upon addition of PRSs to risk factors in a conventional risk prediction model (i.e., age, sex, systolic blood pressure, smoking status, history of diabetes, and total and high-density lipoprotein cholesterol). We then modelled the implications of initiating guideline-recommended statin therapy in a primary care setting using incidence rates from 2.1 million individuals from the Clinical Practice Research Datalink. The C-index, a measure of risk discrimination, was 0.710 (95% CI 0.703–0.717) for a CVD prediction model containing conventional risk predictors alone. Addition of information on PRSs increased the C-index by 0.012 (95% CI 0.009–0.015), and resulted in continuous net reclassification improvements of about 10% and 12% in cases and non-cases, respectively. If a PRS were assessed in the entire UK primary care population aged 40–75 years, assuming that statin therapy would be initiated in accordance with the UK National Institute for Health and Care Excellence guidelines (i.e., for persons with a predicted risk of ≥10% and for those with certain other risk factors, such as diabetes, irrespective of their 10-year predicted risk), then it could help prevent 1 additional CVD event for approximately every 5,750 individuals screened. By contrast, targeted assessment only among people at intermediate (i.e., 5% to <10%) 10-year CVD risk could help prevent 1 additional CVD event for approximately every 340 individuals screened. Such a targeted strategy could help prevent 7% more CVD events than conventional risk prediction alone. Potential gains afforded by assessment of PRSs on top of conventional risk factors would be about 1.5-fold greater than those provided by assessment of C-reactive protein, a plasma biomarker included in some risk prediction guidelines. Potential limitations of this study include its restriction to European ancestry participants and a lack of health economic evaluation.

Conclusions

Our results suggest that addition of PRSs to conventional risk factors can modestly enhance prediction of first-onset CVD and could translate into population health benefits if used at scale.

Luanluan Sun and colleagues investigate whether adding polygenic risk scores to conventional risk factors of cardiovascular disease helps predict disease risk.

Author summary

Why was this study done?

  • Application of polygenic risk scores (PRSs) has opened opportunities to enhance risk stratification and prevention for common diseases. The clinical utility of PRSs in cardiovascular disease (CVD) risk prediction is, however, uncertain.

  • Previous analyses have generally focused only on coronary heart disease (CHD) rather than the composite outcome of CHD and stroke, and have often lacked modelling of clinical implications of initiating guideline-recommended interventions (e.g., statin therapy).

What did the researchers do and find?

  • We quantified the incremental predictive gain with PRSs on top of conventional risk factors using data on 306,654 individuals from UK Biobank.

  • We modelled the population health implications of initiating statin therapy as recommended by current guidelines using data from 2.1 million individuals from the Clinical Practice Research Datalink.

  • Addition of information on PRSs to a conventional risk prediction model increased the C-index (a measure of risk discrimination) and improved risk classification of cases and non-cases.

  • We estimated that targeted assessment of PRSs among people at intermediate (i.e., 5% to <10%) 10-year CVD risk could help prevent 1 additional CVD event for approximately every 340 individuals screened, which would be almost 15 times more efficient than blanket assessment of PRS.

What do these findings mean?

  • Addition of PRSs to conventional risk factors provided modest improvement in prediction of first-onset CVD.

  • Nevertheless, these moderate improvements could translate into meaningful clinical benefit if applied at scale, and lead to the prevention of 7% more CVD events than conventional risk factors alone.

  • Our results also suggest that targeted use of PRSs would be more efficient than blanket population-wide use.

  • Future studies should seek to evaluate PRSs in non-European ancestry populations, and perform formal health economic evaluations.

Introduction

Advances in the application of polygenic risk scores (PRSs) have opened opportunities to enhance disease risk prediction by stratifying populations into risk groups using information on millions of variants across the genome [1–4]. The UK government’s Department of Health and Social Care green paper on disease prevention has stated: ‘As the evidence develops, complementing existing risk scores…with this kind of genetic information [i.e., PRSs] will be a priority for the UK healthcare system’ [5]. The US Centers for Disease Control and Prevention and the US National Institutes of Health are also considering the value of integrating PRSs into clinical practice [6].

A key strategy in the primary prevention of cardiovascular disease (CVD) is the use of risk prediction algorithms to target preventive interventions to people who may benefit from them most [7–12]. These algorithms typically include information on conventional risk factors, including age, sex, smoking history, history of diabetes, blood pressure, total cholesterol, and high-density lipoprotein (HDL) cholesterol [8–10]. The population health utility of PRSs in CVD risk prediction is, however, uncertain. Previous analyses have generally focused only on coronary heart disease (CHD) rather than the composite outcome of CHD and stroke, even though the composite outcome is the focus of most primary prevention guidelines. Furthermore, most previous PRS studies have lacked modelling of the clinical implications of initiating guideline-recommended interventions (e.g., statin therapy) [13,14], meaning that it has been difficult to judge the potential clinical gains of assessing PRSs.

Our study, therefore, aimed to address 2 questions. First, what is the improvement in CVD risk prediction when PRSs are added to risk factors used in conventional risk algorithms? We analysed 306,654 participants from UK Biobank (UKB) to assess the value of adding PRSs to several conventional risk factors. Second, what is the estimated population health impact of using information on PRSs for CVD prediction? We modelled data from 2.1 million individuals in the Clinical Practice Research Datalink (CPRD) to estimate the benefit of initiating statin therapy as recommended by guidelines. To contextualise our findings, we compared the incremental predictive gains afforded by PRSs with that provided by C-reactive protein (CRP), a plasma biomarker recommended for risk prediction in some CVD primary prevention guidelines [12,15].

Methods

Study design and overview

Our study involved several interrelated components (Fig 1). First, we constructed separate PRSs for CHD and stroke, using methods previously described [16,17]. Second, we calculated measures of risk discrimination and reclassification to quantify the incremental predictive gain with these PRSs on top of conventional risk factors. Third, to estimate the potential for disease prevention in a general population setting, we adapted (i.e., recalibrated) our findings to the context of a primary prevention population eligible for CVD screening, using incidence rates from contemporary computerised records from general practices in the UK. Fourth, we modelled the clinical implications of initiating statin therapy as recommended by current guidelines, comparing a ‘blanket’ approach (i.e., assessment of PRSs in all individuals eligible for CVD primary prevention) with a ‘targeted’ approach (i.e., focusing PRSs assessment only in people judged to be at intermediate 10-year risk of CVD after initial screening with conventional risk predictors alone). Fifth, to help contextualise the potential population health gains afforded by assessing PRSs, we compared them in the same dataset with gains afforded by assessment of CRP.

Fig 1. Study design and overview.

Fig 1

Open in a new tab

CHD, coronary heart disease; PRS, polygenic risk score.

Ethics statement

This research has been conducted using the UKB resource under application number 26865. The UKB study was approved by the North West Multi-centre Research Ethics Committee, and all participants provided written informed consent to participate in the UKB. This study is based in part on data from the CPRD obtained under licence from the UK Medicines and Healthcare products Regulatory Agency (protocol number 162RMn2). The data are provided by patients and collected by the National Health Service as part of their care and support.

Data sources

UK Biobank prospective study

Details of the design, methods, and participants of UKB have been described previously [18,19]. Briefly, participants aged 40 to 75 years identified through primary care lists were recruited across 22 assessment centres throughout the UK between 2006 and 2010. At recruitment, information was collected via a standardised questionnaire and selected physical measurements. Details of the data used from UKB are provided in S1 Text. Data were subsequently linked to Hospital Episode Statistics (HES), as well as national death and cancer registries. HES uses the International Classification of Diseases (ICD) 9th and 10th revisions to record diagnosis information, and the Office of Population, Censuses and Surveys: Classification of Interventions and Procedures, version 4 (OPCS-4), to code operative procedures. Death registries include deaths in the UK, with both primary and contributory causes of death coded according to ICD-10.

Genotyping was undertaken using a custom-built genome-wide array of approximately 826,000 markers [18,20]. Imputation to approximately 96 million markers was subsequently carried out using the Haplotype Reference Consortium and UK10K/1000 Genomes reference panels [20]. Clinical biochemistry markers, including total cholesterol, HDL cholesterol, and CRP, were measured at baseline in serum samples. Full details of the biochemistry sampling, handling and quality control protocol, and assay method have been provided previously [21].

UK Clinical Practice Research Datalink

To estimate the potential for disease prevention in a general population setting, we used data from the CPRD, a primary care database of anonymised medical records covering over 11.3 million individuals opting into data linkage from 674 general practices in the UK [22]. Individual-level data from consenting practices in the CPRD have been linked to HES and the national death registry. Details of the CPRD data used and endpoint definition are provided in S2 Text. The present analysis involved records of 2.1 million patients, a random sample of all CPRD data, working under the assumption that individuals in this database should be broadly representative of the UK general population.

Statistical analysis

To approximate populations relevant to CVD primary prevention, we focused on first-onset CVD outcomes among those with no prior history of CVD and not taking lipid-lowering treatments at recruitment. Analyses were performed according to a pre-specified analysis plan (S1 Analysis Plan) and restricted to participants of self-reported European ancestry, excluding those who (1) had missing genotype array or conventional risk factor information; (2) had a history of CVD at baseline (i.e., CHD, other heart disease, stroke, transient ischaemic attack, peripheral vascular disease, angina, or cardiovascular revascularization); (3) used lipid-lowering treatment at baseline; or (4) were included in the dataset to estimate component score mixing weights during PRS construction (see S1 Fig). The primary outcome was a first-onset CVD event, defined as the composite of CHD (i.e., myocardial infarction or fatal CHD) or any stroke. Secondary outcomes included each of CHD and stroke separately, and a combination of CHD, stroke, and cardiac revascularisation procedures (i.e., percutaneous transluminal coronary angioplasty [PTCA] and coronary artery bypass grafting [CABG]) (S1 Table).

We used separate PRSs for CHD and ischaemic stroke as 2 independent variables to predict the composite CVD outcome. PRSs were previously constructed using a meta-score approach based on external summary statistics from the previous largest genome-wide association studies (GWASs) of CHD and stroke [23,24]. Detailed information on PRS derivation has been previously provided [16,17], and the PRSs are publicly available and annotated at the PGS Catalog (http://www.pgscatalog.org) under accessions PGS000018 and PGS000039, respectively. The PRS for CHD comprised 1,743,979 variants where the mixing weights of component scores were estimated using 3,000 participants in UKB. The PRS for ischaemic stroke included 2,759,740 variants where the mixing weights of component scores were estimated using 12,000 participants in UKB (including the 3,000 participants mentioned above). Participants used in the training dataset were excluded from subsequent analysis. Previous analyses have not found evidence of overfitting [16,17], and independent replications have demonstrated consistent effect sizes [25–27]. The partial Pearson correlation coefficient between the PRS for CHD and the PRS for stroke was 0.32. In sensitivity analyses we (1) replaced the PRS for ischaemic stroke with a PRS for all stroke and (2) used a single PRS for the composite CVD outcome.

HRs were calculated using Cox proportional hazards models, stratified by UKB recruitment centre and sex, and using time since study entry as the timescale. Outcomes were censored if a participant was lost to follow-up or died from non-CVD causes, or if the end of available follow-up was reached (for England: 31 March 2017; Scotland: 30 October 2016; Wales: 30 May 2016). Predictors were entered as linear terms, after visual checking for log-linearity. No violation of the proportional hazards assumption was identified. Sensitivity analyses included calculation of cumulative incidence of CVD outcomes based on the cause-specific hazards estimated from Cox regression, in the presence of competing risk from non-CVD deaths [28,29].

The incremental predictive ability of PRSs for CHD and stroke was assessed upon addition (as 2 separate linear terms) to a model containing age, sex, systolic blood pressure, smoking status, history of diabetes, and total and HDL cholesterol (i.e., conventional risk factors). Risk discrimination was assessed using Harrell’s C-index, stratified by UKB recruitment centre and sex [30]. To avoid overestimation of the model’s ability to predict risk, we applied an internal/external validation approach by validating within a subset (i.e., 1 study centre or a 10% randomly selected population in UKB) the prediction model derived from the remaining datasets. Results were then meta-analysed across all validation subsets, weighted by the number of events in that specific subset. Improvements in risk prediction were also quantified by the net reclassification index (NRI), which summarises appropriate directional change in risk predictions for those who do and do not experience an event during follow-up (with increases in predicted risk being appropriate for cases and decreases being appropriate for non-cases) [31,32]. Calibration was assessed by comparing the observed and predicted risks across deciles of predicted risk, and by calculating calibration slope, root mean square error, and the Greenwood–Nam–D’Agostino p-value [13,14,33] using a 10-fold cross-validation approach to avoid optimism.

To assess the population health relevance of adding PRSs to conventional risk factors, we generalised our reclassification analyses to the context of a UK population eligible for primary prevention screening (S3 Text). Using CPRD data we recalibrated risk prediction models derived in UKB to give 10-year risks that would be expected in such a UK primary care setting, employing methods previously described [34]. (Since 10 years of follow-up was not available for all UKB recruitment centres, we used 9-year risk estimates in reclassification analyses.) Details are provided in S3 Text.

We modelled a population of 100,000 adults aged 40–75 years in the CPRD, with an age and sex profile matching that of the contemporary UK population (2017 mid-year population) [35], and CVD incidence rates as observed in individuals without previous CVD and not taking statins. We assumed an initial policy of statin allocation for people at ≥10% predicted 10-year risk as recommended by National Institute for Health and Care Excellence (NICE) guidelines [7]. We then modelled additional targeted assessment of PRSs, or CRP, among people at intermediate risk (5% to <10% predicted 10-year risk) to estimate the potential for additional treatment allocation and case prevention, assuming statin allocation would reduce CVD risk by 20% [36]. Details are in S3 Text. Analyses were performed with PLINK 2.0 [37] and Stata version 14, with 2-sided p-values and 95% confidence intervals. This study follows TRIPOD reporting guidelines (S1 TRIPOD Checklist).

Results

Characteristics of the study participants and association with CVD outcomes

Of the 502,219 participants initially enrolled in UKB, 306,654 participants met the inclusion criteria for this analysis: self-reported European ancestry, without a history of CVD, not on lipid-lowering treatment, and with complete information on genotype array data and conventional risk predictors (Table 1). During 2.6 million person-years at risk (median [5th, 95th percentile] follow-up of 8.1 [6.8–9.4] years), 5,680 CVD cases were recorded, including 3,333 CHD and 2,347 stroke events. Fig 2 shows the baseline characteristics of participants, as well as HRs for CVD adjusted for conventional risk factors. HRs for CHD and stroke outcomes separately and for the composite secondary outcome (including CHD, stroke, PTCA, and CABG) are presented in S2 Fig. Both PRSs showed log-linear associations with CVD outcomes, with HRs of 1.57 (95% CI 1.51–1.62) for CHD and 1.19 (95% CI 1.14–1.24) for stroke, after adjustment for age only (S3 Fig). HRs per 1-SD higher PRS did not materially change after adjustment for conventional risk factors; HRs were similar across people with different levels of risk factors, including family history of CVD (S4 and S5 Figs).

Table 1. Baseline characteristics of UK Biobank participants who had no prior history of vascular disease and were not on lipid-lowering treatment, by sex (n = 306,654).

Baseline characteristic Female Male Total
Number of participants 174,773 131,881 306,654
Age at recruitment, years 56.0 (7.9) 55.9 (8.2) 56.0 (8.0)
Cardiovascular risk factors
Current-smoker, percent 9.3 11.7 10.3
History of diabetes, percent 0.8 1.7 1.2
Treatment of hypertension, percent 10.9 11.7 11.2
Systolic blood pressure, mm Hg 134.2 (18.6) 140.4 (17.3) 136.9 (19.1)
Total cholesterol, mmol/l 6.0 (1.1) 5.8 (1.0) 5.9 (1.1)
HDL cholesterol, mmol/l 1.6 (0.4) 1.3 (0.3) 1.5 (0.4)
LDL cholesterol, mmol/l 3.7 (0.8) 3.7 (0.8) 3.7 (0.8)
C-reactive protein, Ln, mg/l 0.3 (1.1) 0.3 (1.0) 0.3 (1.1)
Incident cardiovascular outcomes
Follow-up, years, median (5th–95th percentile) 8.2 (6.8–9.4) 8.1 (6.5–9.3) 8.1 (6.8–9.4)
Number of coronary heart disease cases 2,453 880 3,333
Number of stroke cases 1,311 1,036 2,347
Open in a new tab

Data are shown as mean (SD), unless otherwise stated, adjusted for UK Biobank study centre.

HDL, high-density lipoprotein; LDL, low-density lipoprotein.

Fig 2. Adjusted hazard ratios of conventional cardiovascular risk factors and polygenic risk scores for first-onset cardiovascular outcomes.

Fig 2

Open in a new tab

Hazard ratios (HRs) were estimated using Cox regression, stratified by study centre and sex, and adjusted for age at baseline, smoking status, history of diabetes, systolic blood pressure, total cholesterol, and high-density lipoprotein (HDL) cholesterol levels, where appropriate. For continuous variables, HRs are shown per SD higher of each predictor to facilitate comparison. For categorical variables, HRs are shown for men versus women, for patients with diabetes versus without, and for current smokers versus others.

Incremental value in risk prediction

We assessed the incremental predictive ability of PRSs using measures of risk discrimination and reclassification, adding PRSs for CHD and stroke as 2 independent linear terms to a model containing conventional CVD risk factors. For the CVD outcome, the C-index was 0.710 (95% CI 0.703–0.717) for a prediction model containing conventional risk factors alone. The addition of information on PRSs increased the C-index by 0.012 (95% CI 0.009–0.015; Fig 3), yielding a continuous NRI of 10.2% (95% CI 7.2%–13.2%) among CVD cases and 12.6% (95% CI 12.2%–13.0%) among non-cases (Table 2). By comparison, the C-index increased by 0.004 (95% CI 0.003–0.006; Fig 3) after adding information on CRP to the conventional model. The improvement in NRI was also less with addition of CRP than with addition of PRSs, with incident cases more often correctly increased in risk by addition of PRSs (Table 2). Models including PRSs showed good calibration, with good agreement between the observed and predicted CVD risks (S6 Fig).

Fig 3. Incremental predictive ability of polygenic risk scores and C-reactive protein for cardiovascular disease, above conventional risk factors.

Fig 3

Open in a new tab

Conventional risk factors included age at baseline, sex, smoking status, history of diabetes, systolic blood pressure, total cholesterol, and high-density lipoprotein cholesterol. C-index and related changes were estimated using Cox regression, stratified by study centre and sex, adjusted for age at baseline, smoking status, history of diabetes, systolic blood pressure, total cholesterol, and high-density lipoprotein cholesterol. 95% confidence intervals (CIs) were estimated using the efficient jackknife approach.

Table 2. Net reclassification index (NRI) for cardiovascular disease (generalised to a primary prevention population) with addition of information on polygenic risk scores or C-reactive protein, above conventional risk factors.

Factors included Continuous NRI (95% CI) versus conventional risk factors alone
Cardiovascular disease
(n = 5,680)		 Coronary heart disease
(n = 3,333)		 Stroke
(n = 2,347)
Conventional risk factors plus polygenic risk scores
 Non-cases 12.6 (12.2, 13.0) 17.5 (17.1, 17.9) 6.6 (6.2, 7.0)
 Cases 10.2 (7.2, 13.2) 14.6 (10.8, 18.4) 3.5 (−1.2, 8.2)
Conventional risk factors plus C-reactive protein
 Non-cases 12.0 (11.6, 12.4) 12.6 (12.2, 13.0) 9.9 (9.5, 10.2)
 Cases 2.1 (−1.1, 4.9) 3.8 (0.1, 7.6) 0.8 (−4.0, 5.5)
Open in a new tab

Conventional risk factors included age at baseline, sex, smoking status, history of diabetes, systolic blood pressure, total cholesterol, and high-density lipoprotein cholesterol, with stratification by study centre and sex, where appropriate.

In hypothesis-generating analyses, the C-index changes with PRSs were possibly somewhat higher in men than women, and in participants with higher total cholesterol, lower HDL cholesterol, and higher predicted 10-year CVD risk (Fig 4; S2 Table). Among CVD cases and controls, continuous NRIs with assessment of PRSs were 11.5% (95% CI 7.8%–15.1%) and 14.1% (95% CI 13.5%–14.6%) in men, and 8.3% (95% CI 3.1%–13.5%) and 8.8% (95% CI 8.3%–9.3%) in women, respectively (S3 Table). The predictive value of PRSs was greater for CHD than for stroke outcomes (Table 2; Fig 3 and S7 Fig).

Fig 4. Incremental predictive ability of polygenic risk scores (PRSs) for cardiovascular disease (CVD) outcomes, beyond conventional risk predictors, across different baseline population characteristics.

Fig 4

Open in a new tab

The base model included information on the conventional risk factors, i.e., age at baseline, sex, smoking status, history of diabetes, systolic blood pressure, total cholesterol, and high-density lipoprotein (HDL) cholesterol, with stratification by study centre and sex, where appropriate. The prediction model within each subgroup was constructed using coefficients estimated among the entire population.

The results of adding information on PRSs were broadly similar to those observed overall in analyses that included (1) information on body-mass index, family history of CVD, use of blood-pressure-lowering treatment, or CRP in the prediction model (S4 Table; S8 Fig); (2) participants receiving lipid-lowering treatment at baseline (S5 Table; S9 Fig); (3) use of PRSs derived for the composite CVD outcome or for all stroke (S6 Table); and (4) a broader definition of the CVD outcome (i.e., CHD, stroke, PTCA, or CABG; S9 Fig). Furthermore, similar results were observed in analyses using the internal/external cross-validation approach (S10 and S11 Figs), the Pooled Cohort Equations (S7 Table), or competing risk models for non-CVD deaths (S8 Table).

Estimate of the potential for disease prevention

In population health modelling, we used age- and sex-specific incidence data from 2.1 million individuals in the CPRD without previous CVD and not taking statins to recalibrate risk models and achieve a predicted risk distribution as would be expected in this primary care population (S3 Text). We translated age- and sex-specific targeted assessment of PRSs to a population of 100,000 adults aged 40–75 years, assuming the age and sex structure of the current UK population, and CVD incidence rates observed in UK primary care. Under this scenario, we estimated that, using conventional risk factors alone, there would be 23,973 individuals classified as having intermediate 10-year (i.e., 5% to <10%) risk who were not already taking or eligible for statin treatment (i.e., people without a history of diabetes or CVD, and with low-density lipoprotein (LDL) cholesterol < 5.0 mmol/l; Fig 5). Additional assessment of PRSs in these individuals (i.e., a ‘targeted’ approach focusing only in people judged to be at intermediate 10-year risk of CVD after initial screening with conventional risk factors alone) would reclassify 3,115 intermediate-risk individuals as high-risk (i.e., ≥10%), of whom approximately 357 would be expected to have a CVD event within 10 years. This would correspond to an increase of about 7.1% (357/5,054) of the CVD events already classified at high risk using conventional risk predictors alone.

Fig 5. Estimated public health impact with targeted assessment of polygenic risk scores among 100,000 UK adults in a primary care setting.

Fig 5

Open in a new tab

CVD, cardiovascular disease; LDL, low-density lipoprotein; PRS, polygenic risk score.

Assuming statin allocation per current guidelines (i.e., those with 10-year CVD risk ≥ 10%) and statin treatment conferring a 20% relative risk reduction, such targeted assessment of PRSs among the intermediate-risk group would help prevent 72 (i.e., 357 × 0.2) events over the next 10-year period. In other words, targeted assessment of PRSs in individuals at intermediate risk for a CVD event could help prevent 1 additional event over 10 years for every 336 people so screened. For comparison, the number needed to screen with targeted assessment of CRP would be 491 (S9 Table). Similar results were observed when analysis involved cutoffs for clinical risk categories defined by other guidelines (S10 Table; S12 Fig).

In contrast with the targeted approach, we also modelled a blanket population-wide strategy of additional assessment of PRSs in all adults aged 40–75 years eligible for CVD primary prevention. In this scenario, compared to using conventional risk factors alone, 3,128 individuals would be reclassified from low or intermediate risk (i.e., <10%) to high risk (i.e., ≥10%), and 3,405 individuals would be reclassified from high risk to low or intermediate risk, of whom approximately 358 and 271 would be expected to have a CVD event within 10 years, respectively (S11 Table; S13 Fig), suggesting the need to screen 5,747 people with additional assessment of PRSs to help prevent 1 additional event over 10 years.

Discussion

We conducted complementary analyses in UKB, a purpose-designed prospective study of about 500,000 individuals, and the CPRD, a cohort of 2.1 million people derived from an extract of contemporary computerised records from general practices in the UK. Overall, our results suggest that the addition of PRSs to conventional risk factors can provide modest improvement in prediction of first-onset CVD, which, if applied at scale, could help prevent 7% more CVD events than use of conventional risk factors alone. Our results have potential implications for CVD risk prediction and for the evaluation of the potential population health utility of PRSs for disease.

First, our modelling suggests that, if applied to the contemporary UK population aged 40–75 years [38], additional use of PRSs could help prevent at least several thousand CVD events over the next 10 years beyond assessment of conventional risk factors alone.

Second, our results suggest that targeted use of PRSs would be almost 15 times more efficient than blanket population-wide use. In a modelled scenario in which PRSs were assessed in a primary care setting only among individuals considered at intermediate CVD risk after initial screening with conventional risk predictors alone, we estimated that such targeted assessment of PRSs could reclassify approximately 12% of screened individuals to the high-risk category, of whom 11% would be expected to have a CVD event within 10 years. If such a targeted approach were to be coupled with initiation of statin therapy in accordance with guidelines, our data suggest 1 extra CVD outcome could be prevented over a period of 10 years for approximately every 340 people in whom PRSs are assessed (compared with the need to screen approximately 5,700 people to achieve the same gain when using a blanket screening approach).

Third, as a benchmark, we compared the incremental predictive gains afforded by assessment of PRSs with those provided by CRP measurement (a plasma biomarker recommended for screening in some primary prevention guidelines) [12,15], with our results demonstrating a >1.5-fold greater gain in predictive accuracy with PRSs than CRP.

Fourth, we found that assessment of PRSs could improve prediction of CHD much more than prediction of stroke. Further work is needed to understand fully the reasons for such differential gains, which may relate both to the greater phenotypic heterogeneity of stroke outcomes [39–41] and the relatively lower statistical power of previous GWASs of stroke [24,41] compared with CHD [23]. It is likely that the composite outcome of CVD involves greater phenotypic and genetic heterogeneity than either CHD or stroke alone. Nevertheless, our study used the primary outcome of any first CVD event (defined as fatal or nonfatal CHD or stroke), in keeping with current CVD primary prevention guidelines that promote joint prediction and prevention of CHD and stroke.

Our study had major strengths. In the analysis of UKB, we approximated the targeted population for CVD primary prevention efforts by focusing on >300,000 participants without a history of CVD at baseline who were not taking lipid-lowering treatment. For these participants, we had access to concomitant and nearly complete information on several conventional CVD risk factors (e.g., lipid measurements) as well as on PRSs. We used multiple complementary metrics of risk discrimination and reclassification, as well as different absolute risk thresholds used in different clinical guidelines. The broadly concordant results we observed across these metrics supported the validity of our main conclusions. To extend the relevance of our findings to a UK primary care population, we also conducted modelling using the UK CPRD, adapting (recalibrating) our findings from UKB to be more representative of the general population. This adaptation was important because the general UK population has a higher baseline risk for CVD than the volunteers who enrolled in UKB, underscoring the need for recalibration when using established risk thresholds, and before making judgements about the population health utility of PRSs.

Our study also had limitations. We studied only middle-aged European ancestry participants in the UK, which limits the generalisability of our results. Hence, we (and others) are now addressing this gap by conducting studies of PRSs for CVD in different ethnic groups, as well as in other countries. Our study also lacked a health economic evaluation, which was beyond the scope of present analysis. We acknowledge the importance of health economic evaluations as part of future considerations to assess the clinical utility of PRSs for CVD prevention, noting that genome-wide array genotyping has a one-time cost (approximately £25 at current prices in the UK) and can be used to calculate PRSs for CVD as well as for many other chronic diseases. In particular, future studies (including health economic evaluations) are needed to evaluate a range of different CVD screening strategies, including a ‘genome first’ approach that inverts the current ‘conventional risk factors first’ approach to CVDs.

Our study did not assess potential psychological harms of using genetic information in CVD risk prediction. However, a previous randomised trial has excluded material effects of this type [42]. We used a conventional 10-year timeframe and standard clinical risk categories, acknowledging that reclassification analyses are intrinsically sensitive to choices of follow-up interval and clinical risk categories. Although we used 9-year risk estimates in reclassification analyses because 10 years of follow-up was not available for all UKB recruitment centres, it had minimal influence on our results. Somewhat greater population health impact than suggested by our analysis would be estimated if we had used less conservative modelling assumptions (e.g., more effective statin regimens, longer time horizons), conventional risk factor weights that were not fitted to UKB, or alternative disease outcomes (e.g., an exclusive focus on CHD). Conversely, our models could have overestimated the potential benefits of assessing PRSs because not all people eligible for statins will receive them or be willing and able to take them and adherent.

In conclusion, our results suggest that the addition of PRSs to conventional risk factors can modestly enhance the prediction of first-onset CVD and could translate into population health benefits if used at scale.

Supporting information

S1 Analysis Plan

(DOCX)

Click here for additional data file. (18.8KB, docx)
S1 Fig. Exclusion criteria applied in derivation of the primary analytic dataset.

*Prior history of cardiovascular disease at baseline included coronary heart disease, angina, other heart disease, stroke, transient ischaemic attack, peripheral vascular disease, and cardiac revascularisations.

(TIF)

Click here for additional data file. (83.4KB, tif)
S2 Fig. Hazard ratios for coronary heart disease, stroke, and the composite cardiovascular disease outcome (including coronary heart disease, stroke, and cardiac revascularisations), adjusted for conventional risk factors.

Hazard ratios (HRs) were estimated using Cox regression, stratified by study centre and sex, and adjusted for age at baseline, smoking status, history of diabetes, systolic blood pressure, total cholesterol, and HDL cholesterol, where appropriate. For continuous variables, HRs are shown for each SD higher of each predictor to facilitate comparison. For categorical variables, HRs are shown for men versus women, for patients with diabetes versus without, and for current smokers versus others.

(TIFF)

Click here for additional data file. (2.1MB, tiff)
S3 Fig. Shape and strength of associations of polygenic risk scores with risk of coronary heart disease and stroke.

The shape of association was estimated by dividing all participants into fifths. Hazard ratios were estimated using Cox regression, stratified by study centre and sex, and adjusted for age at baseline. Each square has an area inversely proportional to the effective variance of the log risk in that specific group, with vertical lines representing the 95% confidence intervals.

(TIFF)

Click here for additional data file. (1MB, tiff)
S4 Fig. Hazard ratios of polygenic risk scores (PRSs) for coronary heart disease and stroke, after progressive adjustment for conventional cardiovascular risk factors.

Hazard ratios were estimated using Cox regression, stratified by study centre and sex, and adjusted for conventional cardiovascular risk factors, where appropriate.

(TIFF)

Click here for additional data file. (1.3MB, tiff)
S5 Fig. Adjusted hazard ratios of polygenic risk scores for incident coronary heart disease and stroke by population characteristics at baseline.

Hazard ratios were estimated using Cox regression, stratified by study centre and sex, and adjusted for conventional cardiovascular risk factors, where appropriate.

(TIFF)

Click here for additional data file. (2.4MB, tiff)
S6 Fig. Observed and predicted cardiovascular risk when adding information on polygenic risk scores and/or C-reactive protein to conventional risk factors, in UK Biobank.

PRS, polygenic risk score; CRP, C-reactive protein; RMSE, root mean square error; GND Chi-sq, Greenwood–Nam–D’Agostino chi-squared index. Conventional risk factors included age at baseline, sex, smoking status, history of diabetes, systolic blood pressure, total cholesterol, and HDL cholesterol levels. Polygenic risk scores included the polygenic risk score for coronary heart disease and the one for ischaemic stroke (see Fig 2) as 2 linear predictors in the model throughout. To avoid optimism in assessing calibration of the models, we applied a 10-fold cross-validation approach. We divided the datasets into 10 random subsets with the same number of participants, with prediction models developed within 90% of the dataset, and validated using the remaining 10% of the dataset. Each blue square in the plots represents the mean value of the predicted/observed risk within each decile; these values were pooled across the 10 validation subsets and weighted by the number of events in that group. The ratios were calculated as the ratio of observed risks to predicted risks, with 1 representing perfect calibration. The RMSE was used to assess the differences between the predicted risks and the observed risks. The Greenwood–Nam–D’Agostino test is an extension of the Hosmer–Lemeshow test to situations with censored survival data, and tests the null hypothesis that the observed and expected probabilities are identical in each group.

(TIFF)

Click here for additional data file. (1.1MB, tiff)
S7 Fig. Incremental predictive ability of polygenic risk scores (PRSs) for cardiovascular disease, above conventional risk factors.

Conventional risk factors included age at baseline, sex, smoking status, history of diabetes, and systolic blood pressure, with or without baseline measurements of total cholesterol and HDL cholesterol.

(TIFF)

Click here for additional data file. (2MB, tiff)
S8 Fig. Incremental predictive values of polygenic risk scores (PRSs), above conventional risk factors, including body-mass index or family history of cardiovascular disease in the reference model.

CVD, cardiovascular disease. Conventional risk factors included age at baseline, sex, smoking status, history of diabetes, systolic blood pressure, total cholesterol, and HDL cholesterol. Polygenic risk scores included the polygenic risk score for CHD and the one for ischaemic stroke (see Fig 2) as 2 linear predictors in the model throughout.

(TIFF)

Click here for additional data file. (1.5MB, tiff)
S9 Fig. Incremental predictive values of polygenic risk scores (PRSs) above conventional risk factors, among individuals with or without lipid-lowering treatment, and for different cardiovascular outcomes.

Conventional risk factors included age at baseline, sex, smoking status, history of diabetes, systolic blood pressure, total cholesterol, and HDL cholesterol. Polygenic risk scores included the polygenic risk score for coronary heart disease and the one for ischaemic stroke (see Fig 2) as 2 linear predictors in the model throughout. Cardiac procedures included cardiovascular outcomes identified via OPCS-4: K40–K46, K49, K50.1, K50.2, K50.4, or K75.

(TIFF)

Click here for additional data file. (1.2MB, tiff)
S10 Fig. Incremental predictive value of polygenic risk scores, above conventional risk factors, by 10-fold cross-validation in UK Biobank.

Polygenic risk scores included the polygenic risk score for CHD and the one for ischaemic stroke (see Fig 2) as 2 linear predictors in the model throughout. Each subset represented a 10% randomly selected subset of UK Biobank participants from the entire study population. The prediction model was derived using 90% of the dataset, and validated among the remaining 10%. The overall C-index and relevant changes were estimated by meta-analysing the subset-specific results, weighted by the number of events in that subset.

(TIFF)

Click here for additional data file. (1.7MB, tiff)
S11 Fig. Incremental predictive value of polygenic risk scores, above conventional risk factors, when leaving 1 recruitment centre out per iteration in UK Biobank.

(TIFF)

Click here for additional data file. (2.4MB, tiff)
S12 Fig. Estimates of public health impact with targeted assessment of polygenic risk scores, among 100,000 UK adults using American Heart Association/American College of Cardiology guideline.

(DOCX)

Click here for additional data file. (37.8KB, docx)
S13 Fig. Estimates of public health impact of additional assessment of polygenic risk scores or C-reactive protein, above conventional risk factors, among 100,000 individuals.

Numbers in red are shown for individuals who were initially classified as being at high risk and were reclassified down to intermediate risk; numbers in blue are shown for individuals moving from intermediate risk to high risk. Among 100,000 individuals, 1,197 cases and 7,354 non-cases were treated, irrespective of their 10-year CVD risk, since they had history of diabetes or LDL cholesterol ≥ 5.0 mmol/l. The number of cases screened as high risk, or classified as high risk due to diabetes or LDL cholesterol level, using conventional risk factors alone was 5,054, and thus, the number of events prevented was 1,011 (5,054 × 0.2).

(TIF)

Click here for additional data file. (158.9KB, tif)
S1 Table. Definition of study outcomes.

(DOCX)

Click here for additional data file. (12.5KB, docx)
S2 Table. Incremental predictive ability of polygenic risk scores, and C-reactive protein, above conventional risk factors, by population characteristics at baseline.

Conventional risk factors included age at baseline, sex, smoking status, history of diabetes, systolic blood pressure, total cholesterol, and HDL cholesterol. Prediction model was developed using Cox regression for all participants, stratified by study centre and sex, adjusted for conventional risk predictors, where appropriate. Polygenic risk scores included the polygenic risk score for CHD and the one for ischaemic stroke (see Fig 2) as 2 linear predictors in the model throughout.

(DOCX)

Click here for additional data file. (16.6KB, docx)
S3 Table. Net reclassification index (NRI) for incident cardiovascular disease by addition of information on polygenic risk scores, and C-reactive protein, above conventional risk factors, for non-cases and cases.

ACC, American College of Cardiology; AHA, American Heart Association; NICE, National Institute for Health and Care Excellence. *Conventional risk factors included age, sex, smoking, systolic blood pressure, history of diabetes, total cholesterol, and HDL cholesterol. Polygenic risk scores included the polygenic risk score for CHD and the one for ischaemic stroke (see Fig 2) as 2 linear predictors in the model throughout. Calculations of the above categorical NRIs were <5%, 5% to <7.5%, and ≥7.5% according to the 2019 ACC/AHA guideline, and <5%, 5% to <10%, and ≥10% according to the 2014 NICE guideline.

(DOCX)

Click here for additional data file. (15.9KB, docx)
S4 Table. Partial likelihood ratio test for models with polygenic risk scores beyond conventional risk factors, C-reactive protein, and treatment of hypertension.

Conventional risk factors included age at baseline, sex, smoking status, history of diabetes, systolic blood pressure, total cholesterol, and HDL cholesterol, with stratification by study centre and sex, where appropriate.

(DOCX)

Click here for additional data file. (13.4KB, docx)
S5 Table. Net reclassification index (NRI) for incident cardiovascular disease by addition of information on polygenic risk scores, and C-reactive protein, above conventional risk factors, for non-cases and cases, including participants on statin treatment at baseline.

ACC, American College of Cardiology; AHA, American Heart Association; NICE, National Institute for Health and Care Excellence. *Conventional risk factors included age, sex, smoking, systolic blood pressure, history of diabetes, total cholesterol, and HDL cholesterol. Polygenic risk scores included the polygenic risk score for CHD and the one for ischaemic stroke (see Fig 2) as 2 linear predictors in the model throughout. Calculations of the above categorical NRIs were <5%, 5% to <7.5%, and ≥7.5% according to the 2019 ACC/AHA guideline, and <5%, 5% to <10%, and ≥10% according to the 2014 NICE guideline.

(DOCX)

Click here for additional data file. (16.2KB, docx)
S6 Table. Comparison of different polygenic risk scores (PRSs) on strength of associations, discriminative ability, and reclassification index for different cardiovascular outcomes, in UK Biobank.

CVD, cardiovascular disease; CHD, coronary heart disease; IS, ischaemic stroke; CI, confidence interval. Conventional risk factors included age at baseline, sex, smoking status, history of diabetes, systolic blood pressure, total cholesterol, and HDL cholesterol, with stratification by study centre and sex, where appropriate. The PRS for CHD and the PRS for IS were constructed using methods as in our previous work [1]. The PRS for stroke was constructed using the genome-wide significant variants in the MEGASTROKE consortium for total stroke, and linkage-disequilibrium-thinned in UK Biobank, with corresponding weights taken from the MEGASTROKE consortium [2]. Construction procedures of the 2 above PRSs did not include estimates from previous GWASs on other vascular risk factors. The PRS for IS was constructed using methods described in our previous work [3], by taking account of 19 phenotypes, and is publicly available (https://www.pgscatalog.org/score/PGS000039/). The PRS for CVD was constructed using the same approach as the PRS for IS but with CVD as the outcome.

(DOCX)

Click here for additional data file. (24.5KB, docx)
S7 Table. Incremental predictive ability of polygenic risk scores, and C-reactive protein, above the updated Pooled Cohort Equations (PCE) score.

*The PCE score for study participants in UK Biobank was calculated using the updated Pooled Cohort Equations score, i.e., the weights for each constituent predictor variable, as previously published [1]. **PCE variables included age at baseline, sex, smoking status, history of diabetes, systolic blood pressure, total cholesterol, HDL cholesterol, ethnicity, and treatment of high blood pressure, weighted by the Cox regression coefficients estimated in UK Biobank. Polygenic risk scores included the polygenic risk score for CHD and the one for ischaemic stroke (see Fig 2) as 2 linear predictors in the model throughout.

(DOCX)

Click here for additional data file. (15.9KB, docx)
S8 Table. Incremental predictive ability of polygenic risk scores, and C-reactive protein, with or without adjusting for competing risk from non-cardiovascular death.

Conventional risk factors included age at baseline, sex, smoking status, history of diabetes, systolic blood pressure, total cholesterol, and HDL cholesterol. Polygenic risk scores included the polygenic risk score for CHD and the one for ischaemic stroke (see Fig 2) as 2 linear predictors in the model throughout. Cumulative incidence of the composite CVD outcomes was estimated using the cause-specific hazards ratios from Cox regression, in the presence of competing risk from non-CVD deaths.

(DOCX)

Click here for additional data file. (14.8KB, docx)
S9 Table. Estimates of public health impact with targeted assessment (intermediate risk: 5% to <10%) of polygenic risk scores, and C-reactive protein, among 100,000 UK adults.

PRS, polygenic risk score; CRP, C-reactive protein. Conventional risk factors included age at baseline, sex, smoking, systolic blood pressure, history of diabetes, total cholesterol, and HDL cholesterol. Polygenic risk scores included the polygenic risk score for CHD and the one for ischaemic stroke (see Fig 2) as 2 linear predictors in the model throughout. The predicted 10-year cardiovascular risk categories used the 2014 NICE guideline. Estimates of public health impact for a hypothetical population of 100,000 individuals (40–75 years) were based on (1) the sex- and age-specific (5-year) profile of a standard UK population (2017 mid-year population) [35] and (2) sex-specific 5-year age-at-risk incidence rates of cardiovascular disease in the CPRD, among individuals without prior history of cardiovascular disease and not on statin treatment at baseline. Estimates for public health impact are shown before and after recalibration.

(DOCX)

Click here for additional data file. (15.2KB, docx)
S10 Table. Estimates of public health impact with targeted assessment (intermediate risk: 5% to <7.5%) of polygenic risk scores (PRSs), and C-reactive protein, among 100,000 UK adults.

CRP, C-reactive protein. Conventional risk factors included age at baseline, sex, smoking, systolic blood pressure, history of diabetes, total cholesterol, and HDL cholesterol. Polygenic risk scores included the polygenic risk score for CHD and the one for ischaemic stroke (see Fig 2) as 2 linear predictors in the model throughout. The predicted 10-year cardiovascular risk categories used the 2019 American Heart Association/American College of Cardiology guideline. Estimates of public health impact for a hypothetical population of 100,000 individuals (40–75 years) were based on (1) the sex- and age-specific (5-year) profile of a standard UK population (2017 mid-year population) [35] and (2) sex-specific 5-year age-at-risk incidence rates of cardiovascular disease in the CPRD, among individuals without prior history of cardiovascular disease and not on statin treatment at baseline. Estimates for public health impact are shown before and after recalibration.

(DOCX)

Click here for additional data file. (15.2KB, docx)
S11 Table. Numerical results for estimates of public health impact by additional assessment of polygenic risk scores (PRSs) or C-reactive protein, above conventional risk predictors, among 100,000 individuals.

*Among cases and non-cases, respectively, 1,197 and 7,354 participants had diabetes or LDL cholesterol measurement of 5.0 mmol/l or greater. Numbers in red are shown for individuals who were reclassified downwards with additional assessment, and numbers in blue are shown for individuals who were reclassified upwards with additional assessment. Polygenic risk scores included the polygenic risk score for CHD and the one for ischaemic stroke (see Fig 2) as 2 linear predictors in the model throughout.

(DOCX)

Click here for additional data file. (15.2KB, docx)
S1 Text. Description of analytic dataset from UK Biobank.

(DOCX)

Click here for additional data file. (20.8KB, docx)
S2 Text. Description of analytic dataset from Clinical Practice Research Datalink.

(DOCX)

Click here for additional data file. (119.3KB, docx)
S3 Text. Statistical methods used for estimating public health impact.

(DOCX)

Click here for additional data file. (4.3MB, docx)
S1 TRIPOD Checklist

(DOCX)

Click here for additional data file. (89.6KB, docx)

Acknowledgments

The views expressed are those of the authors and not necessarily those of the NHS, the NIHR, or the Department of Health and Social Care.

Abbreviations

CABG

coronary artery bypass grafting

CHD

coronary heart disease

CPRD

Clinical Practice Research Datalink

CRP

C-reactive protein

CVD

cardiovascular disease

GWAS

genome-wide association study

HDL

high-density lipoprotein

HES

Hospital Episode Statistics

LDL

low-density lipoprotein

NRI

net reclassification index

PRS

polygenic risk score

PTCA

percutaneous transluminal coronary angioplasty

UKB

UK Biobank

Data Availability

All data files are available from the UK Biobank and CPRD databases.

Funding Statement

This work was supported by core funding from the UK Medical Research Council (MR/L003120/1), the British Heart Foundation (RG/13/13/30194; RG/18/13/33946), and the National Institute for Health Research (NIHR) (Cambridge Biomedical Research Centre at the Cambridge University Hospitals NHS Foundation Trust and NIHR Leicester Biomedical Research Centre). This work was supported by Health Data Research UK, which is funded by the the UK Medical Research Council, the Engineering and Physical Sciences Research Council, the Economic and Social Research Council, the Department of Health and Social Care (England), the Chief Scientist Office of the Scottish Government Health and Social Care Directorates, the Health and Social Care Research and Development Division (Welsh Government), the Public Health Agency (Northern Ireland), the British Heart Foundation, and Wellcome. Luanluan Sun, Lisa Pennells, Stephen Kaptoge, and Matthew Arnold are funded by a British Heart Foundation Programme Grant (RG/18/13/33946). Christopher P. Nelson is funded by a British Heart Foundation Grant (SP/16/4/32697). Scott Ritchie, Mike Inouye, and Stephen Burgess are funded by the National Institute for Health Research (Cambridge Biomedical Research Centre at the Cambridge University Hospitals NHS Foundation Trust). David Stevens was funded by the National Institute for Health Research (Cambridge Biomedical Research Centre at the Cambridge University Hospitals NHS Foundation Trust). Thomas Bolton is funded by the NIHR Blood and Transplant Research Unit in Donor Health and Genomics (NIHR BTRU-2014-10024). Steven Bell was funded by the NIHR Blood and Transplant Research Unit in Donor Health and Genomics (NIHR BTRU-2014-10024). Angela Wood is supported by a BHF-Turing Cardiovascular Data Science Award and by the EC-Innovative Medicines Initiative (BigData@Heart). Professor John Danesh holds a British Heart Foundation Professorship and a National Institute for Health Research Senior Investigator Award.

References

  • 1.Knowles JW, Ashley EA. Cardiovascular disease: the rise of the genetic risk score. PLoS Med. 2018;15(3):e1002546 10.1371/journal.pmed.1002546 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Torkamani A, Wineinger NE, Topol EJ. The personal and clinical utility of polygenic risk scores. Nat Rev Genet. 2018;19(9):581–90. 10.1038/s41576-018-0018-x [DOI] [PubMed] [Google Scholar]
  • 3.Wise AL, Manolio TA, Mensah GA, Peterson JF, Roden DM, Tamburro C, et al. Genomic medicine for undiagnosed diseases. Lancet. 2019;394(10197):533–40. 10.1016/S0140-6736(19)31274-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Claussnitzer M, Cho JH, Collins R, Cox NJ, Dermitzakis ET, Hurles ME, et al. A brief history of human disease genetics. Nature. 2020;577(7789):179–89. 10.1038/s41586-019-1879-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.UK Department of Health and Social Care. Advancing our health: prevention in the 2020s. London: UK Department of Health and Social Care; 2019 [cited 2020 Dec 18]. https://www.gov.uk/government/consultations/advancing-our-health-prevention-in-the-2020s.
  • 6.Khoury MJ, Mensah GA. Is it time to integrate polygenic risk scores into clinical practice? Let’s do the science first and follow the evidence wherever it takes us! Atlanta: US Centers for Disease Control and Prevention; 2019 [cited 2020 Dec 18]. https://blogs.cdc.gov/genomics/2019/06/03/is-it-time/.
  • 7.National Institute for Health and Care Excellence. Cardiovascular disease: risk assessment and reduction, including lipid modification. London: National Institute for Health and Care Excellence; 2014 [cited 2020 Dec 18]. https://www.nice.org.uk/guidance/cg181.
  • 8.Conroy RM, Pyörälä K, Fitzgerald AP, Sans S, Menotti A, De Backer G, et al. Estimation of ten-year risk of fatal cardiovascular disease in Europe: the SCORE project. Eur Heart J. 2003;24(11):987–1003. 10.1016/s0195-668x(03)00114-3 [DOI] [PubMed] [Google Scholar]
  • 9.D’Agostino RB, Vasan RS, Pencina MJ, Wolf PA, Cobain M, Massaro JM, et al. General cardiovascular risk profile for use in primary care: the Framingham Heart Study. Circulation. 2008;117(6):743–53. 10.1161/CIRCULATIONAHA.107.699579 [DOI] [PubMed] [Google Scholar]
  • 10.Goff DC Jr, Lloyd-Jones DM, Bennett G, Coady S, D’Agostino RB, Gibbons R, et al. 2013 ACC/AHA Guideline on the Assessment of Cardiovascular Risk. Circulation. 2014;129(25 Suppl 2):S49–73. 10.1161/01.cir.0000437741.48606.98 [DOI] [PubMed] [Google Scholar]
  • 11.Bibbins-Domingo K, Grossman DC, Curry SJ, Davidson KW, Epling JW Jr, Garcia FA, et al. Statin use for the primary prevention of cardiovascular disease in adults: US Preventive Services Task Force recommendation statement. JAMA. 2016;316(19):1997–2007. 10.1001/jama.2016.15450 [DOI] [PubMed] [Google Scholar]
  • 12.Arnett DK, Blumenthal RS, Albert MA, Michos ED, Buroker AB, Miedema MD, et al. 2019 ACC/AHA guideline on the primary prevention of cardiovascular disease: executive summary: a report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. J Am Coll Cardiol. 2019;74(10):1376–414. 10.1016/j.jacc.2019.03.009 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Elliott J, Bodinier B, Bond TA, Chadeau-Hyam M, Evangelou E, Moons KGM, et al. Predictive accuracy of a polygenic risk score-enhanced prediction model vs a clinical risk score for coronary artery disease. JAMA. 2020;323(7):636–45. 10.1001/jama.2019.22241 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Mosley JD, Gupta DK, Tan J, Yao J, Wells QS, Shaffer CM, et al. Predictive accuracy of a polygenic risk score compared with a clinical risk score for incident coronary heart disease. JAMA. 2020;323(7):627–35. 10.1001/jama.2019.21782 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Anderson TJ, Gregoire J, Pearson GJ, Barry AR, Couture P, Dawes M, et al. 2016 Canadian Cardiovascular Society guidelines for the management of dyslipidemia for the prevention of cardiovascular disease in the adult. Can J Cardiol. 2016;32(11):1263–82. 10.1016/j.cjca.2016.07.510 [DOI] [PubMed] [Google Scholar]
  • 16.Inouye M, Abraham G, Nelson CP, Wood AM, Sweeting MJ, Dudbridge F, et al. Genomic risk prediction of coronary artery disease in 480,000 adults: implications for primary prevention. J Am Coll Cardiol. 2018;72(16):1883–93. 10.1016/j.jacc.2018.07.079 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Abraham G, Malik R, Yonova-Doing E, Salim A, Wang T, Danesh J, et al. Genomic risk score offers predictive performance comparable to clinical risk factors for ischaemic stroke. Nat Commun. 2019;10(1):5819 10.1038/s41467-019-13848-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Sudlow C, Gallacher J, Allen N, Beral V, Burton P, Danesh J, et al. UK biobank: an open access resource for identifying the causes of a wide range of complex diseases of middle and old age. PLoS Med. 2015;12(3):e1001779 10.1371/journal.pmed.1001779 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Fry A, Littlejohns TJ, Sudlow C, Doherty N, Adamska L, Sprosen T, et al. Comparison of sociodemographic and health-related characteristics of UK Biobank participants with those of the general population. Am J Epidemiol. 2017;186(9):1026–34. 10.1093/aje/kwx246 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Bycroft C, Freeman C, Petkova D, Band G, Elliott LT, Sharp K, et al. The UK Biobank resource with deep phenotyping and genomic data. Nature. 2018;562(7726):203–9. 10.1038/s41586-018-0579-z [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Allen NE, Arnold M, Parish S, Hill M, Sheard S, Callen H, et al. Approaches to minimising the epidemiological impact of sources of systematic and random variation that may affect biochemistry assay data in UK Biobank [version 1; peer review: 2 approved]. Wellcome Open Res. 2020;5:222 10.12688/wellcomeopenres.16171.1 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Herrett E, Gallagher AM, Bhaskaran K, Forbes H, Mathur R, van Staa T, et al. Data resource profile: Clinical Practice Research Datalink (CPRD). Int J Epidemiol. 2015;44(3):827–36. 10.1093/ije/dyv098 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Nikpay M, Goel A, Won HH, Hall LM, Willenborg C, Kanoni S, et al. A comprehensive 1,000 Genomes-based genome-wide association meta-analysis of coronary artery disease. Nat Genet. 2015;47(10):1121–30. 10.1038/ng.3396 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Malik R, Chauhan G, Traylor M, Sargurupremraj M, Okada Y, Mishra A, et al. Multiancestry genome-wide association study of 520,000 subjects identifies 32 loci associated with stroke and stroke subtypes. Nat Genet. 2018;50(4):524–37. 10.1038/s41588-018-0058-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Wünnemann F, Lo KS, Langford-Avelar A, Busseuil D, Dubé M-P, Tardif J-C, et al. Validation of genome-wide polygenic risk scores for coronary artery disease in French Canadians. Circ Genom Precis Med. 2019;12(6):e002481 10.1161/CIRCGEN.119.002481 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Dikilitas O, Schaid DJ, Kosel ML, Carroll RJ, Chute CG, Denny JA, et al. Predictive utility of polygenic risk scores for coronary heart disease in three major racial and ethnic groups. Am J Hum Genet. 2020;106(5):707–16. 10.1016/j.ajhg.2020.04.002 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Duncan L, Shen H, Gelaye B, Meijsen J, Ressler K, Feldman M, et al. Analysis of polygenic risk score usage and performance in diverse human populations. Nat Commun. 2019;10(1):3328 10.1038/s41467-019-11112-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Putter H, Fiocco M, Geskus RB. Tutorial in biostatistics: competing risks and multi-state models. Stat Med. 2007;26(11):2389–430. 10.1002/sim.2712 [DOI] [PubMed] [Google Scholar]
  • 29.Fine JP, Gray RJ. A proportional hazards model for the subdistribution of a competing risk. J Am Stat Assoc. 1999;94(446):496–509. 10.1080/01621459.1999.10474144 [DOI] [Google Scholar]
  • 30.Harrell FE, Lee KL, Mark DB. Multivariable prognostic models: issues in developing models, evaluating assumptions and adequacy, and measuring and reducing errors. Stat Med. 1996;15(4):361–87. [DOI] [PubMed] [Google Scholar]
  • 31.Leening MJ, Vedder MM, Witteman JC, Pencina MJ, Steyerberg EW. Net reclassification improvement: computation, interpretation, and controversies: a literature review and clinician’s guide. Ann Intern Med. 2014;160(2):122–31. 10.7326/M13-1522 [DOI] [PubMed] [Google Scholar]
  • 32.Kerr KF, Wang Z, Janes H, McClelland RL, Psaty BM, Pepe MS. Net reclassification indices for evaluating risk prediction instruments: a critical review. Epidemiology. 2014;25(1):114–21. 10.1097/EDE.0000000000000018 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Demler OV, Paynter NP, Cook NR. Tests of calibration and goodness-of-fit in the survival setting. Stat Med. 2015;34(10):1659–80. 10.1002/sim.6428 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Pennells L, Kaptoge S, Wood A, Sweeting M, Zhao X, White I, et al. Equalization of four cardiovascular risk algorithms after systematic recalibration: individual-participant meta-analysis of 86 prospective studies. Eur Heart J. 2019;40:621–31. 10.1093/eurheartj/ehy653 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Office for National Statistics. Estimates of the population for the UK, England and Wales, Scotland and Northern Ireland. Mid-2017 edition of this dataset. London: Office for National Statistics; 2020 [cited 2020 Dec 21]. https://www.ons.gov.uk/peoplepopulationandcommunity/populationandmigration/populationestimates/datasets/populationestimatesforukenglandandwalesscotlandandnorthernireland.
  • 36.Collins R, Reith C, Emberson J, Armitage J, Baigent C, Blackwell L, et al. Interpretation of the evidence for the efficacy and safety of statin therapy. Lancet. 2016;388(10059):2532–61. 10.1016/S0140-6736(16)31357-5 [DOI] [PubMed] [Google Scholar]
  • 37.Purcell S, Neale B, Todd-Brown K, Thomas L, Ferreira MAR, Bender D, et al. PLINK: a tool set for whole-genome association and population-based linkage analyses. Am J Hum Genet. 2007;81(3):559–75. 10.1086/519795 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.UK Office for National Statistics. Principal projection: UK population in age groups. London: UK Office for National Statistics; 2019 [cited 2020 Sep 1]. https://www.ons.gov.uk/peoplepopulationandcommunity/populationandmigration/populationprojections/datasets/tablea21principalprojectionukpopulationinagegroups.
  • 39.Emerging Risk Factors Collaboration, Di Angelantonio E, Sarwar N, Perry P, Kaptoge S, Ray KK, et al. Major lipids, apolipoproteins, and risk of vascular disease. JAMA. 2009;302(18):1993–2000. 10.1001/jama.2009.1619 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Sun L, Clarke R, Bennett D, Guo Y, Walters RG, Hill M, et al. Causal associations of blood lipids with risk of ischemic stroke and intracerebral hemorrhage in Chinese adults. Nat Med. 2019;25(4):569–74. 10.1038/s41591-019-0366-x [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Dichgans M, Pulit SL, Rosand J. Stroke genetics: discovery, biology, and clinical applications. Lancet Neurol. 2019;18(6):587–99. 10.1016/S1474-4422(19)30043-2 [DOI] [PubMed] [Google Scholar]
  • 42.Silarova B, Sharp S, Usher-Smith JA, Lucas J, Payne RA, Shefer G, et al. Effect of communicating phenotypic and genetic risk of coronary heart disease alongside web-based lifestyle advice: the INFORM randomised controlled trial. Heart. 2019;105(13):982–9. 10.1136/heartjnl-2018-314211 [DOI] [PMC free article] [PubMed] [Google Scholar]

Decision Letter 0

Helen Howard

3 Feb 2020

Dear Dr Di Angelantonio,

Thank you for submitting your manuscript entitled "Adding polygenic risk scores to conventional risk factors

in cardiovascular disease prediction" for consideration by PLOS Medicine.

Your manuscript has now been evaluated by the PLOS Medicine editorial staff and I am writing to let you know that we would like to send your submission out for external peer review.

However, before we can send your manuscript to reviewers, we need you to complete your submission by providing the metadata that is required for full assessment. To this end, please login to Editorial Manager where you will find the paper in the 'Submissions Needing Revisions' folder on your homepage. Please click 'Revise Submission' from the Action Links and complete all additional questions in the submission questionnaire.

Please re-submit your manuscript within two working days, i.e. by .

Login to Editorial Manager here: https://www.editorialmanager.com/pmedicine

Once your full submission is complete, your paper will undergo a series of checks in preparation for peer review. Once your manuscript has passed all checks it will be sent out for review.

Feel free to email us at plosmedicine@plos.org if you have any queries relating to your submission.

Kind regards,

Helen Howard, for Clare Stone PhD

Acting Editor-in-Chief

PLOS Medicine

plosmedicine.org

Decision Letter 1

Adya Misra

6 Mar 2020

Dear Dr. Di Angelantonio,

Thank you very much for submitting your manuscript "Adding polygenic risk scores to conventional risk factors

in cardiovascular disease prediction" (PMEDICINE-D-20-00284R1) for consideration at PLOS Medicine.

Your paper was evaluated by a senior editor and discussed among all the editors here. It was also discussed with an academic editor with relevant expertise, and sent to independent reviewers, including a statistical reviewer. The reviews are appended at the bottom of this email and any accompanying reviewer attachments can be seen via the link below:

[LINK]

In light of these reviews, I am afraid that we will not be able to accept the manuscript for publication in the journal in its current form, but we would like to consider a revised version that addresses the reviewers' and editors' comments. Obviously we cannot make any decision about publication until we have seen the revised manuscript and your response, and we plan to seek re-review by one or more of the reviewers.

In revising the manuscript for further consideration, your revisions should address the specific points made by each reviewer and the editors. Please also check the guidelines for revised papers at http://journals.plos.org/plosmedicine/s/revising-your-manuscript for any that apply to your paper. In your rebuttal letter you should indicate your response to the reviewers' and editors' comments, the changes you have made in the manuscript, and include either an excerpt of the revised text or the location (eg: page and line number) where each change can be found. Please submit a clean version of the paper as the main article file; a version with changes marked should be uploaded as a marked up manuscript.

In addition, we request that you upload any figures associated with your paper as individual TIF or EPS files with 300dpi resolution at resubmission; please read our figure guidelines for more information on our requirements: http://journals.plos.org/plosmedicine/s/figures. While revising your submission, please upload your figure files to the PACE digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at PLOSMedicine@plos.org.

We expect to receive your revised manuscript by Mar 27 2020 11:59PM. Please email us (plosmedicine@plos.org) if you have any questions or concerns.

***Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.***

We ask every co-author listed on the manuscript to fill in a contributing author statement, making sure to declare all competing interests. If any of the co-authors have not filled in the statement, we will remind them to do so when the paper is revised. If all statements are not completed in a timely fashion this could hold up the re-review process. If new competing interests are declared later in the revision process, this may also hold up the submission. Should there be a problem getting one of your co-authors to fill in a statement we will be in contact. YOU MUST NOT ADD OR REMOVE AUTHORS UNLESS YOU HAVE ALERTED THE EDITOR HANDLING THE MANUSCRIPT TO THE CHANGE AND THEY SPECIFICALLY HAVE AGREED TO IT. You can see our competing interests policy here: http://journals.plos.org/plosmedicine/s/competing-interests.

Please use the following link to submit the revised manuscript:

https://www.editorialmanager.com/pmedicine/

Your article can be found in the "Submissions Needing Revision" folder.

To enhance the reproducibility of your results, we recommend that you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see http://journals.plos.org/plosmedicine/s/submission-guidelines#loc-methods.

Please ensure that the paper adheres to the PLOS Data Availability Policy (see http://journals.plos.org/plosmedicine/s/data-availability), which requires that all data underlying the study's findings be provided in a repository or as Supporting Information. For data residing with a third party, authors are required to provide instructions with contact information for obtaining the data. PLOS journals do not allow statements supported by "data not shown" or "unpublished results." For such statements, authors must provide supporting data or cite public sources that include it.

We look forward to receiving your revised manuscript.

Sincerely,

Adya Misra, PhD

Senior Editor

PLOS Medicine

plosmedicine.org

-----------------------------------------------------------

Requests from the editors:

Title- Please revise your title according to PLOS Medicine's style. Your title must be nondeclarative and not a question. It should begin with main concept if possible. "Effect of" should be used only if causality can be inferred, i.e., for an RCT. Please place the study design ("A randomized controlled trial," "A retrospective study," "A modelling study," etc.) in the subtitle (ie, after a colon).

Author summary-At this stage, we ask that you include a short, non-technical Author Summary of your research to make findings accessible to a wide audience that includes both scientists and non-scientists. The Author Summary should immediately follow the Abstract in your revised manuscript. This text is subject to editorial change and should be distinct from the scientific abstract. Please see our author guidelines for more information: https://journals.plos.org/plosmedicine/s/revising-your-manuscript#loc-author-summary

Abstract methods and findings- the last sentence should be a limitation of your study design

Abstract background- please clearly state the aim of your study

Abstract conclusions- please tone down in view of the “modest gains” in prediction accuracy

Data availability- please provide details of how the data may be accessed, providing contact details, URLs or accession numbers as needed

Page 3- please include a header for “Introduction”

Reference square brackets must be placed before a full stop for example [2,5].

Methods- please introduce CHD on first view

A table reference is missing on Page 9line 6

Please complete the TRIPOD checklist, and include the completed checklist as Supporting Information. When completing the checklist, please use section and paragraph numbers, rather than page numbers. Please add the following statement, or similar, to the Methods: "This study is reported as per xxx guideline (S1 Checklist)."

Overall the language needs to be toned down as this is an observational study

Comments from the reviewers:

Reviewer #1: This study investigated the added values of polygenic risk scores in addition to conventional risk factors in prediction of cardiovascular disease. However, there are quite a few major issues needing attention.

1) Cox models were used throughout the study to predict CVD events. As the outcomes here were CVD events other than all cause mortality, there is a major issue of competing risk, which unfortunately was not addressed in the paper.

2) The authors developed the PRSs through 10-fold cross validation using the UK biobank data. However, there was no external validation (using independent external datasets) for the models, which is crucial for the validation of the developed models. Also, there is no calibration of the developed models. Could authors please read and follow the TRIPOD statement?

3) The prediction model containing conventional risk factors alone gave a C-index of 71% while the PRSs increased the C-index by 1% when adding to the model. Although this 1% is statistically significant, it is mainly because the huge sample size which will detect any tiny differences. Firstly this 1% increase hasn't gone through rigorous external validation, secondly the predictive model is not really convincing as hasn't adjusted for competing risk, and finally whole exercise of adding PRSs to the prediction model to increase the precision from 71% to 72% doesn't offer real practical benefits in clinical settings. It's more like a exploratory or association study to show the PRSs are of predictive value in predicting CVD outcomes, however this is already known.

4) The whole excise of using CPRD data to show the benefits of the added value of PRSs is very difficult to follow and not convincing at all, because everything was based on assumptions especially for different projected outcomes. The argument should be based on the solid outcomes not some arbitrary and hypothetical scenarios.

5) Figure 1 and 2 are a bit strange to put baseline characteristics and odd ratios together. It would be good to be straightforward and clear to have a baseline table to put UK biobank and CPRD data side by side, then have other tables for prediction models, validation, NPI, and etc.

6) A bit confused and unclear on which models the authors worked on, predict stroke? predict CHD? but the CVD events were defined as stroke or CHD? Also, need to describe exactly which are the derivation and validation datasets, and any independent external validation datasets?

Reviewer #2: Sun et al. study the performance of two previously developed and validated polygenic risk scores for coronary heart disease and stroke. The authors assess the improvement in risk prediction on top of conventional risk factors using around 300,000 UK Biobank participants and estimate the clinical benefits by modeling the impact on data from a primary care setting.

The paper is well written and in line with several recent studies. The added value of this work comes from the recalibration of risk and modeling to a primary care setting in the UK.

As the authors are projecting the improvements into general benefits in the primary care setting, it is important that the performance of the scores is additionally assessed in other ancestries and not restricted to individuals of European ancestry.

The stroke PRS was derived from GWAS summary statistics of 19 phenotypes including stroke and coronary artery disease. What is the correlation between the stroke PRS and coronary heart disease PRS? In previous work the authors have combined polygenic risk scores in a metaGRS, why haven't they followed similar strategy here and created such score for CVD?

Both scores were derived and validated in the UK Biobank which raises some questions regarding external validity. This should be discussed. How would the score perform in a non-UK Biobank dataset?

Use of antihypertensive medication is usually included with conventional risk factors and risk equations. Why is it not included here?

The authors use CRP for comparison with the PRS. How does the PRS perform on top of conventional risk factors that additionally include CRP?

Reviewer #3: In this manuscript, Sun and colleagues investigate (a) whether the addition of polygenic risk scores for cardiovascular disease to a clinical model yields a change in risk classification at a population level; and (b) whether using this modified score to influence statin prescriptions is likely to yield a reduction in incident cardiovascular disease in a primary prevention population. These questions are timely and important in the field, coming at a moment when the PRS literature is in transition from demonstrating disease association to showing potential public health impact from implementation.

A substantial body of prior work, with important contributions from these authors, has demonstrated that polygenic risk scores can function as independent risk factors for incident cardiovascular disease (Inouye, M. et al. Genomic Risk Prediction of Coronary Artery Disease in 480,000 Adults: Implications for Primary Prevention. Journal of the American College of Cardiology 2018) and can improve 10-year risk classification when compared to clinical risk factors alone in various cohorts, including similar work in the UK Biobank that was published after this manuscript's submission (Abraham, G. et al. Genomic prediction of coronary heart disease. Eur Heart J 2016; Elliott, J. et al. Predictive Accuracy of a Polygenic Risk Score-Enhanced Prediction Model vs a Clinical Risk Score for Coronary Artery Disease. JAMA 2020). Although the primary dataset and some of the content is similar to the now-published Elliott et al manuscript, this manuscript extends prior work by asking how many incident CVD events might be prevented in the general UK population if statin prescribing patterns were changed by using a model that adds a polygenic score to classical risk factors. Because the costs (such as those due to genotyping and the increased number of statin prescriptions) were not addressed, some important questions raised by this manuscript are left unanswered. Some amount of the novelty was eclipsed by Elliott et al. However, the authors make key novel contributions (by assessing the potential impact of genomic screening to influence preventive statin prescriptions in a general UK population). Nevertheless, a comprehensive assessment would look not only at the potential benefits as described in this manuscript, but also at the cost and potential harms.

Additional comments:

1) The description of the polygenic scores for CAD and stroke was insufficient to convey a complete understanding of how the scores were derived. The main text identified the source studies for the SNP weights (CARDIoGRAMplusC4D and MEGASTROKE, respectively) and indicated that plink 2.0 was used for genetic analysis. Supplementary Figure 4 indicated that "LD thinning" was performed in UK Biobank. "LD thinning" is not a named function in plink, and colloquially it can represent either (a) LD pruning on MAF or (b) LD clumping on GWAS P-value, both of which use r2 cutoffs and could therefore be the indicated method. Several other aspects of polygenic score creation in this manuscript were not clearly answered in the manuscript: most importantly, why were new scores created when polygenic scores using the same GWAS summary statistics have already been developed, tuned in a UK population, and publicly released (both by authors of this manuscript and others)? How was the r2 cutoff chosen and were other r2 cutoffs considered? The Data Availability section addresses UK Biobank and CPRD availability, but not the availability of the subsetted GWAS summary statistics necessary to compute the same scores in other populations; will those be made available? A more comprehensive treatment of the polygenic score will be important for permitting a complete evaluation, and for reproducibility. Alternatively, using polygenic scores that have already been derived and validated, such as scores previously published by coauthors of this manuscript, would sidestep this concern.

2) The authors compare a CVD risk model whose inputs are clinical risk factors (the same as those that are used in the Pooled Cohorts Equation [PCE]) with a model that includes those clinical risk factors plus polygenic scores. It would be useful to also compare the "PCE+polygenic scores" model to the original PCE that uses the original risk factor weights (i.e., the PCE that is actually in clinical use). Doing so would permit a comparison with typical clinical practice, while the current comparison assesses a hypothetical reweighting of the PCE (which may be more optimally calibrated for the studied population, but which is not actually in clinical use) compared to a PCE+polygenic scores model.

3) The authors exclude individuals who are already taking statins from their main analysis, but importantly they provide a sensitivity analysis that includes those individuals. Those without pre-existing disease but who are taking statins for primary prevention form an important subgroup of individuals whose treatment decisions might be impacted by a PRS-informed cardiovascular disease risk score. Their exclusion can influence the apparent effectiveness of risk scores. For example, in the evaluation of the 2013 ACC Pooled Cohorts Equation, a sensitivity analysis that excluded the ~15% of MESA participants on statins substantially improved the C-statistic (by 0.0135 on average), showing that excluding statin users from risk calculators can have an important influence on model performance (see Goff et al, "2013 ACC/AHA Guideline on the Assessment of Cardiovascular Risk," 2014, Circulation, supplement pages 41-42). Here, the concern would be that excluding current statin-takers might differentially affect a risk score with a polygenic score compared to a risk score without a polygenic score. The authors' sensitivity analysis addresses that concern. It would be helpful to provide more comprehensive results from this sensitivity analysis (e.g., sensitivity, specificity, reclassification counts) rather than just showing the AUC change in Supplementary Figure 9.

4) The likelihood ratio test (LRT) is the uniformly most powerful test comparing models with and without additional predictors. As such, while I believe that the authors have already chosen reasonable statistics to demonstrate clinical impact, for comprehensiveness it could be helpful to include a likelihood ratio test result to more precisely define the statistical improvement of adding the PRS (i.e., LRT of baseline vs baseline+PRS). In addition, a likelihood ratio test would permit a more concise answer to the question of whether the PRS adds information beyond C-reactive protein (i.e., LRT of baseline+CRP vs baseline+CRP+PRS). I raise this only as a suggestion for the authors' consideration.

Any attachments provided with reviews can be seen via the following link:

[LINK]

Decision Letter 2

Clare Stone

18 Aug 2020

Dear Dr. Di Angelantonio,

Thank you very much for submitting your manuscript "Polygenic risk scores in cardiovascular risk prediction: 

prospective cohort study and modelling analyses" (PMEDICINE-D-20-00284R2) for consideration at PLOS Medicine.

Your paper was evaluated by a senior editor and discussed among all the editors here. It was also discussed with an academic editor with relevant expertise, and sent to independent reviewers, including a statistical reviewer. The reviews are appended at the bottom of this email and any accompanying reviewer attachments can be seen via the link below:

[LINK]

In light of these reviews, I am afraid that we will not be able to accept the manuscript for publication in the journal in its current form, but we would like to consider a revised version that addresses the reviewers' and editors' comments. Obviously we cannot make any decision about publication until we have seen the revised manuscript and your response, and we plan to seek re-review by one or more of the reviewers.

In revising the manuscript for further consideration, your revisions should address the specific points made by each reviewer and the editors. Please also check the guidelines for revised papers at http://journals.plos.org/plosmedicine/s/revising-your-manuscript for any that apply to your paper. In your rebuttal letter you should indicate your response to the reviewers' and editors' comments, the changes you have made in the manuscript, and include either an excerpt of the revised text or the location (eg: page and line number) where each change can be found. Please submit a clean version of the paper as the main article file; a version with changes marked should be uploaded as a marked up manuscript.

In addition, we request that you upload any figures associated with your paper as individual TIF or EPS files with 300dpi resolution at resubmission; please read our figure guidelines for more information on our requirements: http://journals.plos.org/plosmedicine/s/figures. While revising your submission, please upload your figure files to the PACE digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at PLOSMedicine@plos.org.

We expect to receive your revised manuscript by Sep 08 2020 11:59PM. Please email us (plosmedicine@plos.org) if you have any questions or concerns.

***Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.***

We ask every co-author listed on the manuscript to fill in a contributing author statement, making sure to declare all competing interests. If any of the co-authors have not filled in the statement, we will remind them to do so when the paper is revised. If all statements are not completed in a timely fashion this could hold up the re-review process. If new competing interests are declared later in the revision process, this may also hold up the submission. Should there be a problem getting one of your co-authors to fill in a statement we will be in contact. YOU MUST NOT ADD OR REMOVE AUTHORS UNLESS YOU HAVE ALERTED THE EDITOR HANDLING THE MANUSCRIPT TO THE CHANGE AND THEY SPECIFICALLY HAVE AGREED TO IT. You can see our competing interests policy here: http://journals.plos.org/plosmedicine/s/competing-interests.

Please use the following link to submit the revised manuscript:

https://www.editorialmanager.com/pmedicine/

Your article can be found in the "Submissions Needing Revision" folder.

To enhance the reproducibility of your results, we recommend that you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see http://journals.plos.org/plosmedicine/s/submission-guidelines#loc-methods.

Please ensure that the paper adheres to the PLOS Data Availability Policy (see http://journals.plos.org/plosmedicine/s/data-availability), which requires that all data underlying the study's findings be provided in a repository or as Supporting Information. For data residing with a third party, authors are required to provide instructions with contact information for obtaining the data. PLOS journals do not allow statements supported by "data not shown" or "unpublished results." For such statements, authors must provide supporting data or cite public sources that include it.

We look forward to receiving your revised manuscript.

Sincerely,

Clare Stone, PhD

Acting Chief Editor

PLOS Medicine

plosmedicine.org

-----------------------------------------------------------

Requests from the editors:

Please address all points from Referee 1 - note we will only consult with this referee one more time.

Comments from the reviewers:

Reviewer #1: Thanks authors for their effort to improve the manuscript and I am satisfied with some of the responses. However, there are still remaining issues needing attention.

1) Regarding response #2, the authors didn't respond to my question on the calibration of the developed models. As described in the TRIPOD statement, the performance of clinical prediction models consists of both discrimination and calibration. So far in the paper the C-index and NPI are all about discrimination. However, there is no calibration of the models in the paper such as calibration plot/slope and other measures such as RMSE (root mean square error), therefore the performance evaluation is only half done.

2) Regrading response #3, I am still not convinced about the clinical usefulness of adding the PRSs to conventional risk factors as only marginal improvement of around 1% in C-index was shown in the study which is also subject to complete independent external validation (eg, in another European country). Of course, if apply to large population, the 1% will always convert to some numbers but not sure whether it is cost effective and practical. Authors haven't touched the point of implication and practical usefulness of this marginal improvement by PRSs in the discussion, which is not adequate. Basically, one would wonder whether it's worth it to use all these genetic tests to gain only marginal improvement in prediction which doesn't seem very useful as compared to conventional risk factors.

Reviewer #2: The authors addressed all my concerns

Any attachments provided with reviews can be seen via the following link:

[LINK]

Decision Letter 3

Adya Misra

17 Nov 2020

Dear Dr. Di Angelantonio,

Thank you very much for re-submitting your manuscript "Polygenic risk scores in cardiovascular risk prediction: 

prospective cohort study and modelling analyses" (PMEDICINE-D-20-00284R3) for review by PLOS Medicine.

I have discussed the paper with my colleagues and the academic editor and it was also seen again by reviewers. I am pleased to say that provided the remaining editorial and production issues are dealt with we are planning to accept the paper for publication in the journal.

The remaining issues that need to be addressed are listed at the end of this email. Any accompanying reviewer attachments can be seen via the link below. Please take these into account before resubmitting your manuscript:

[LINK]

Our publications team (plosmedicine@plos.org) will be in touch shortly about the production requirements for your paper, and the link and deadline for resubmission. DO NOT RESUBMIT BEFORE YOU'VE RECEIVED THE PRODUCTION REQUIREMENTS.

***Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.***

In revising the manuscript for further consideration here, please ensure you address the specific points made by each reviewer and the editors. In your rebuttal letter you should indicate your response to the reviewers' and editors' comments and the changes you have made in the manuscript. Please submit a clean version of the paper as the main article file. A version with changes marked must also be uploaded as a marked up manuscript file.

Please also check the guidelines for revised papers at http://journals.plos.org/plosmedicine/s/revising-your-manuscript for any that apply to your paper. If you haven't already, we ask that you provide a short, non-technical Author Summary of your research to make findings accessible to a wide audience that includes both scientists and non-scientists. The Author Summary should immediately follow the Abstract in your revised manuscript. This text is subject to editorial change and should be distinct from the scientific abstract.

We expect to receive your revised manuscript within 1 week. Please email us (plosmedicine@plos.org) if you have any questions or concerns.

We ask every co-author listed on the manuscript to fill in a contributing author statement. If any of the co-authors have not filled in the statement, we will remind them to do so when the paper is revised. If all statements are not completed in a timely fashion this could hold up the re-review process. Should there be a problem getting one of your co-authors to fill in a statement we will be in contact. YOU MUST NOT ADD OR REMOVE AUTHORS UNLESS YOU HAVE ALERTED THE EDITOR HANDLING THE MANUSCRIPT TO THE CHANGE AND THEY SPECIFICALLY HAVE AGREED TO IT.

Please ensure that the paper adheres to the PLOS Data Availability Policy (see http://journals.plos.org/plosmedicine/s/data-availability), which requires that all data underlying the study's findings be provided in a repository or as Supporting Information. For data residing with a third party, authors are required to provide instructions with contact information for obtaining the data. PLOS journals do not allow statements supported by "data not shown" or "unpublished results." For such statements, authors must provide supporting data or cite public sources that include it.

Please note, when your manuscript is accepted, an uncorrected proof of your manuscript will be published online ahead of the final version, unless you've already opted out via the online submission form. If, for any reason, you do not want an earlier version of your manuscript published online or are unsure if you have already indicated as such, please let the journal staff know immediately at plosmedicine@plos.org.

If you have any questions in the meantime, please contact me or the journal staff on plosmedicine@plos.org.

We look forward to receiving the revised manuscript by Nov 24 2020 11:59PM.

Sincerely,

Adya Misra

Senior Editor

PLOS Medicine

plosmedicine.org

------------------------------------------------------------

Requests from Editors:

Title- please reword to "Polygenic risk scores in cardiovascular risk prediction: a modelling study"

Abstract- please add brief participant demographics

Abstract -Might be helpful to describe the C-index?

The overall tone and language needs to be toned down to avoid overstating conclusions. Please add “our results indicate” or “modelling suggests” etc and adopt more cautious language

Conclusion- please add “our results suggest” or similar, to avoid overstating results. I’m not sure how this approach directly translates to application at scale, I suggest removing this part

Author summary- please add bullet points

Model assumptions mentioned on page 8 should be briefly mentioned in the abstract as well

Line 222, "to avoid optimism" should perhaps be "to critically evaluate this possibility ..." or similar?

Line 340-342 I suggest replacing this with a summary of the findings

Line 348- 20,000 CVD events prevented-- needs to be toned down as this is not directly tested in your study

Please remove financial information and data availability statements from the main text and add these to the article meta-data sections instead

Did your study have a prespecified protocol or analysis plan? Please state this (either way) early in the Methods section.

a) If a prespecified analysis plan (from your funding proposal, IRB or other ethics committee submission, study protocol, or other planning document written before analyzing the data) was used in designing the study, please include the relevant prospectively written document with your revised manuscript as a Supporting Information file to be published alongside your study, and cite it in the Methods section. A legend for this file should be included at the end of your manuscript.

b) If no such document exists, please make sure that the Methods section transparently describes when analyses were planned, and when/why any data-driven changes to analyses took place.

c) In either case, changes in the analysis-- including those made in response to peer review comments-- should be identified as such in the Methods section of the paper, with rationale.

TRIPOD checklist- please use paragraphs and sections instead of page numbers as these are likely to change

Please remove spaces from square brackets

Please ensure p-values are reported to up to three decimal spaces. For eg Fig4 contains - p<0.0001

Comments from Reviewers:

Reviewer #1: Many thanks authors for their great effort to improve the manuscript. I am satisfied with the response and revision. No further issues needing attention.

Any attachments provided with reviews can be seen via the following link:

[LINK]

Decision Letter 4

Adya Misra

14 Dec 2020

Dear Dr. Di Angelantonio,

I am writing concerning your manuscript submitted to PLOS Medicine, entitled “Polygenic risk scores in cardiovascular risk prediction:  a cohort study and modelling analyses.”

We have now completed our final technical checks and have approved your submission for publication. You will shortly receive a letter of formal acceptance from the editor.

Kind regards,

PLOS Medicine

Associated Data

    This section collects any data citations, data availability statements, or supplementary materials included in this article.

    Supplementary Materials

    S1 Analysis Plan

    (DOCX)

    Click here for additional data file. (18.8KB, docx)
    S1 Fig. Exclusion criteria applied in derivation of the primary analytic dataset.

    *Prior history of cardiovascular disease at baseline included coronary heart disease, angina, other heart disease, stroke, transient ischaemic attack, peripheral vascular disease, and cardiac revascularisations.

    (TIF)

    Click here for additional data file. (83.4KB, tif)
    S2 Fig. Hazard ratios for coronary heart disease, stroke, and the composite cardiovascular disease outcome (including coronary heart disease, stroke, and cardiac revascularisations), adjusted for conventional risk factors.

    Hazard ratios (HRs) were estimated using Cox regression, stratified by study centre and sex, and adjusted for age at baseline, smoking status, history of diabetes, systolic blood pressure, total cholesterol, and HDL cholesterol, where appropriate. For continuous variables, HRs are shown for each SD higher of each predictor to facilitate comparison. For categorical variables, HRs are shown for men versus women, for patients with diabetes versus without, and for current smokers versus others.

    (TIFF)

    Click here for additional data file. (2.1MB, tiff)
    S3 Fig. Shape and strength of associations of polygenic risk scores with risk of coronary heart disease and stroke.

    The shape of association was estimated by dividing all participants into fifths. Hazard ratios were estimated using Cox regression, stratified by study centre and sex, and adjusted for age at baseline. Each square has an area inversely proportional to the effective variance of the log risk in that specific group, with vertical lines representing the 95% confidence intervals.

    (TIFF)

    Click here for additional data file. (1MB, tiff)
    S4 Fig. Hazard ratios of polygenic risk scores (PRSs) for coronary heart disease and stroke, after progressive adjustment for conventional cardiovascular risk factors.

    Hazard ratios were estimated using Cox regression, stratified by study centre and sex, and adjusted for conventional cardiovascular risk factors, where appropriate.

    (TIFF)

    Click here for additional data file. (1.3MB, tiff)
    S5 Fig. Adjusted hazard ratios of polygenic risk scores for incident coronary heart disease and stroke by population characteristics at baseline.

    Hazard ratios were estimated using Cox regression, stratified by study centre and sex, and adjusted for conventional cardiovascular risk factors, where appropriate.

    (TIFF)

    Click here for additional data file. (2.4MB, tiff)
    S6 Fig. Observed and predicted cardiovascular risk when adding information on polygenic risk scores and/or C-reactive protein to conventional risk factors, in UK Biobank.

    PRS, polygenic risk score; CRP, C-reactive protein; RMSE, root mean square error; GND Chi-sq, Greenwood–Nam–D’Agostino chi-squared index. Conventional risk factors included age at baseline, sex, smoking status, history of diabetes, systolic blood pressure, total cholesterol, and HDL cholesterol levels. Polygenic risk scores included the polygenic risk score for coronary heart disease and the one for ischaemic stroke (see Fig 2) as 2 linear predictors in the model throughout. To avoid optimism in assessing calibration of the models, we applied a 10-fold cross-validation approach. We divided the datasets into 10 random subsets with the same number of participants, with prediction models developed within 90% of the dataset, and validated using the remaining 10% of the dataset. Each blue square in the plots represents the mean value of the predicted/observed risk within each decile; these values were pooled across the 10 validation subsets and weighted by the number of events in that group. The ratios were calculated as the ratio of observed risks to predicted risks, with 1 representing perfect calibration. The RMSE was used to assess the differences between the predicted risks and the observed risks. The Greenwood–Nam–D’Agostino test is an extension of the Hosmer–Lemeshow test to situations with censored survival data, and tests the null hypothesis that the observed and expected probabilities are identical in each group.

    (TIFF)

    Click here for additional data file. (1.1MB, tiff)
    S7 Fig. Incremental predictive ability of polygenic risk scores (PRSs) for cardiovascular disease, above conventional risk factors.

    Conventional risk factors included age at baseline, sex, smoking status, history of diabetes, and systolic blood pressure, with or without baseline measurements of total cholesterol and HDL cholesterol.

    (TIFF)

    Click here for additional data file. (2MB, tiff)
    S8 Fig. Incremental predictive values of polygenic risk scores (PRSs), above conventional risk factors, including body-mass index or family history of cardiovascular disease in the reference model.

    CVD, cardiovascular disease. Conventional risk factors included age at baseline, sex, smoking status, history of diabetes, systolic blood pressure, total cholesterol, and HDL cholesterol. Polygenic risk scores included the polygenic risk score for CHD and the one for ischaemic stroke (see Fig 2) as 2 linear predictors in the model throughout.

    (TIFF)

    Click here for additional data file. (1.5MB, tiff)
    S9 Fig. Incremental predictive values of polygenic risk scores (PRSs) above conventional risk factors, among individuals with or without lipid-lowering treatment, and for different cardiovascular outcomes.

    Conventional risk factors included age at baseline, sex, smoking status, history of diabetes, systolic blood pressure, total cholesterol, and HDL cholesterol. Polygenic risk scores included the polygenic risk score for coronary heart disease and the one for ischaemic stroke (see Fig 2) as 2 linear predictors in the model throughout. Cardiac procedures included cardiovascular outcomes identified via OPCS-4: K40–K46, K49, K50.1, K50.2, K50.4, or K75.

    (TIFF)

    Click here for additional data file. (1.2MB, tiff)
    S10 Fig. Incremental predictive value of polygenic risk scores, above conventional risk factors, by 10-fold cross-validation in UK Biobank.

    Polygenic risk scores included the polygenic risk score for CHD and the one for ischaemic stroke (see Fig 2) as 2 linear predictors in the model throughout. Each subset represented a 10% randomly selected subset of UK Biobank participants from the entire study population. The prediction model was derived using 90% of the dataset, and validated among the remaining 10%. The overall C-index and relevant changes were estimated by meta-analysing the subset-specific results, weighted by the number of events in that subset.

    (TIFF)

    Click here for additional data file. (1.7MB, tiff)
    S11 Fig. Incremental predictive value of polygenic risk scores, above conventional risk factors, when leaving 1 recruitment centre out per iteration in UK Biobank.

    (TIFF)

    Click here for additional data file. (2.4MB, tiff)
    S12 Fig. Estimates of public health impact with targeted assessment of polygenic risk scores, among 100,000 UK adults using American Heart Association/American College of Cardiology guideline.

    (DOCX)

    Click here for additional data file. (37.8KB, docx)
    S13 Fig. Estimates of public health impact of additional assessment of polygenic risk scores or C-reactive protein, above conventional risk factors, among 100,000 individuals.

    Numbers in red are shown for individuals who were initially classified as being at high risk and were reclassified down to intermediate risk; numbers in blue are shown for individuals moving from intermediate risk to high risk. Among 100,000 individuals, 1,197 cases and 7,354 non-cases were treated, irrespective of their 10-year CVD risk, since they had history of diabetes or LDL cholesterol ≥ 5.0 mmol/l. The number of cases screened as high risk, or classified as high risk due to diabetes or LDL cholesterol level, using conventional risk factors alone was 5,054, and thus, the number of events prevented was 1,011 (5,054 × 0.2).

    (TIF)

    Click here for additional data file. (158.9KB, tif)
    S1 Table. Definition of study outcomes.

    (DOCX)

    Click here for additional data file. (12.5KB, docx)
    S2 Table. Incremental predictive ability of polygenic risk scores, and C-reactive protein, above conventional risk factors, by population characteristics at baseline.

    Conventional risk factors included age at baseline, sex, smoking status, history of diabetes, systolic blood pressure, total cholesterol, and HDL cholesterol. Prediction model was developed using Cox regression for all participants, stratified by study centre and sex, adjusted for conventional risk predictors, where appropriate. Polygenic risk scores included the polygenic risk score for CHD and the one for ischaemic stroke (see Fig 2) as 2 linear predictors in the model throughout.

    (DOCX)

    Click here for additional data file. (16.6KB, docx)
    S3 Table. Net reclassification index (NRI) for incident cardiovascular disease by addition of information on polygenic risk scores, and C-reactive protein, above conventional risk factors, for non-cases and cases.

    ACC, American College of Cardiology; AHA, American Heart Association; NICE, National Institute for Health and Care Excellence. *Conventional risk factors included age, sex, smoking, systolic blood pressure, history of diabetes, total cholesterol, and HDL cholesterol. Polygenic risk scores included the polygenic risk score for CHD and the one for ischaemic stroke (see Fig 2) as 2 linear predictors in the model throughout. Calculations of the above categorical NRIs were <5%, 5% to <7.5%, and ≥7.5% according to the 2019 ACC/AHA guideline, and <5%, 5% to <10%, and ≥10% according to the 2014 NICE guideline.

    (DOCX)

    Click here for additional data file. (15.9KB, docx)
    S4 Table. Partial likelihood ratio test for models with polygenic risk scores beyond conventional risk factors, C-reactive protein, and treatment of hypertension.

    Conventional risk factors included age at baseline, sex, smoking status, history of diabetes, systolic blood pressure, total cholesterol, and HDL cholesterol, with stratification by study centre and sex, where appropriate.

    (DOCX)

    Click here for additional data file. (13.4KB, docx)
    S5 Table. Net reclassification index (NRI) for incident cardiovascular disease by addition of information on polygenic risk scores, and C-reactive protein, above conventional risk factors, for non-cases and cases, including participants on statin treatment at baseline.

    ACC, American College of Cardiology; AHA, American Heart Association; NICE, National Institute for Health and Care Excellence. *Conventional risk factors included age, sex, smoking, systolic blood pressure, history of diabetes, total cholesterol, and HDL cholesterol. Polygenic risk scores included the polygenic risk score for CHD and the one for ischaemic stroke (see Fig 2) as 2 linear predictors in the model throughout. Calculations of the above categorical NRIs were <5%, 5% to <7.5%, and ≥7.5% according to the 2019 ACC/AHA guideline, and <5%, 5% to <10%, and ≥10% according to the 2014 NICE guideline.

    (DOCX)

    Click here for additional data file. (16.2KB, docx)
    S6 Table. Comparison of different polygenic risk scores (PRSs) on strength of associations, discriminative ability, and reclassification index for different cardiovascular outcomes, in UK Biobank.

    CVD, cardiovascular disease; CHD, coronary heart disease; IS, ischaemic stroke; CI, confidence interval. Conventional risk factors included age at baseline, sex, smoking status, history of diabetes, systolic blood pressure, total cholesterol, and HDL cholesterol, with stratification by study centre and sex, where appropriate. The PRS for CHD and the PRS for IS were constructed using methods as in our previous work [1]. The PRS for stroke was constructed using the genome-wide significant variants in the MEGASTROKE consortium for total stroke, and linkage-disequilibrium-thinned in UK Biobank, with corresponding weights taken from the MEGASTROKE consortium [2]. Construction procedures of the 2 above PRSs did not include estimates from previous GWASs on other vascular risk factors. The PRS for IS was constructed using methods described in our previous work [3], by taking account of 19 phenotypes, and is publicly available (https://www.pgscatalog.org/score/PGS000039/). The PRS for CVD was constructed using the same approach as the PRS for IS but with CVD as the outcome.

    (DOCX)

    Click here for additional data file. (24.5KB, docx)
    S7 Table. Incremental predictive ability of polygenic risk scores, and C-reactive protein, above the updated Pooled Cohort Equations (PCE) score.

    *The PCE score for study participants in UK Biobank was calculated using the updated Pooled Cohort Equations score, i.e., the weights for each constituent predictor variable, as previously published [1]. **PCE variables included age at baseline, sex, smoking status, history of diabetes, systolic blood pressure, total cholesterol, HDL cholesterol, ethnicity, and treatment of high blood pressure, weighted by the Cox regression coefficients estimated in UK Biobank. Polygenic risk scores included the polygenic risk score for CHD and the one for ischaemic stroke (see Fig 2) as 2 linear predictors in the model throughout.

    (DOCX)

    Click here for additional data file. (15.9KB, docx)
    S8 Table. Incremental predictive ability of polygenic risk scores, and C-reactive protein, with or without adjusting for competing risk from non-cardiovascular death.

    Conventional risk factors included age at baseline, sex, smoking status, history of diabetes, systolic blood pressure, total cholesterol, and HDL cholesterol. Polygenic risk scores included the polygenic risk score for CHD and the one for ischaemic stroke (see Fig 2) as 2 linear predictors in the model throughout. Cumulative incidence of the composite CVD outcomes was estimated using the cause-specific hazards ratios from Cox regression, in the presence of competing risk from non-CVD deaths.

    (DOCX)

    Click here for additional data file. (14.8KB, docx)
    S9 Table. Estimates of public health impact with targeted assessment (intermediate risk: 5% to <10%) of polygenic risk scores, and C-reactive protein, among 100,000 UK adults.

    PRS, polygenic risk score; CRP, C-reactive protein. Conventional risk factors included age at baseline, sex, smoking, systolic blood pressure, history of diabetes, total cholesterol, and HDL cholesterol. Polygenic risk scores included the polygenic risk score for CHD and the one for ischaemic stroke (see Fig 2) as 2 linear predictors in the model throughout. The predicted 10-year cardiovascular risk categories used the 2014 NICE guideline. Estimates of public health impact for a hypothetical population of 100,000 individuals (40–75 years) were based on (1) the sex- and age-specific (5-year) profile of a standard UK population (2017 mid-year population) [35] and (2) sex-specific 5-year age-at-risk incidence rates of cardiovascular disease in the CPRD, among individuals without prior history of cardiovascular disease and not on statin treatment at baseline. Estimates for public health impact are shown before and after recalibration.

    (DOCX)

    Click here for additional data file. (15.2KB, docx)
    S10 Table. Estimates of public health impact with targeted assessment (intermediate risk: 5% to <7.5%) of polygenic risk scores (PRSs), and C-reactive protein, among 100,000 UK adults.

    CRP, C-reactive protein. Conventional risk factors included age at baseline, sex, smoking, systolic blood pressure, history of diabetes, total cholesterol, and HDL cholesterol. Polygenic risk scores included the polygenic risk score for CHD and the one for ischaemic stroke (see Fig 2) as 2 linear predictors in the model throughout. The predicted 10-year cardiovascular risk categories used the 2019 American Heart Association/American College of Cardiology guideline. Estimates of public health impact for a hypothetical population of 100,000 individuals (40–75 years) were based on (1) the sex- and age-specific (5-year) profile of a standard UK population (2017 mid-year population) [35] and (2) sex-specific 5-year age-at-risk incidence rates of cardiovascular disease in the CPRD, among individuals without prior history of cardiovascular disease and not on statin treatment at baseline. Estimates for public health impact are shown before and after recalibration.

    (DOCX)

    Click here for additional data file. (15.2KB, docx)
    S11 Table. Numerical results for estimates of public health impact by additional assessment of polygenic risk scores (PRSs) or C-reactive protein, above conventional risk predictors, among 100,000 individuals.

    *Among cases and non-cases, respectively, 1,197 and 7,354 participants had diabetes or LDL cholesterol measurement of 5.0 mmol/l or greater. Numbers in red are shown for individuals who were reclassified downwards with additional assessment, and numbers in blue are shown for individuals who were reclassified upwards with additional assessment. Polygenic risk scores included the polygenic risk score for CHD and the one for ischaemic stroke (see Fig 2) as 2 linear predictors in the model throughout.

    (DOCX)

    Click here for additional data file. (15.2KB, docx)
    S1 Text. Description of analytic dataset from UK Biobank.

    (DOCX)

    Click here for additional data file. (20.8KB, docx)
    S2 Text. Description of analytic dataset from Clinical Practice Research Datalink.

    (DOCX)

    Click here for additional data file. (119.3KB, docx)
    S3 Text. Statistical methods used for estimating public health impact.

    (DOCX)

    Click here for additional data file. (4.3MB, docx)
    S1 TRIPOD Checklist

    (DOCX)

    Click here for additional data file. (89.6KB, docx)
    Attachment

    Submitted filename: Responses_to_reviewers_20_May_2020.docx

    Click here for additional data file. (32.4KB, docx)
    Attachment

    Submitted filename: Responses_to_reviewers_2_Sept_2020v2.docx

    Click here for additional data file. (23.2KB, docx)
    Attachment

    Submitted filename: Responses_to_editors_30_Nov_2020.docx

    Click here for additional data file. (28.8KB, docx)

    Data Availability Statement

    All data files are available from the UK Biobank and CPRD databases.

    Articles from PLoS Medicine are provided here courtesy of PLOS

    RESOURCES