Cardiovascular risk prediction using metabolomic biomarkers and polygenic risk scores: A cohort study and modelling analyses

Metabolomic platforms using nuclear magnetic resonance (NMR) spectroscopy can now rapidly quantify many circulating metabolites which are potential biomarkers of cardiovascular disease (CVD). Here, we analyse ∼170,000 UK Biobank participants (5,096 incident CVD cases) without a history of CVD and not on lipid-lowering treatments to evaluate the potential for improving 10-year CVD risk prediction using NMR biomarkers in addition to conventional risk factors and polygenic risk scores (PRSs). Using machine learning, we developed sex-specific NMR scores for coronary heart disease (CHD) and ischaemic stroke, then estimated their incremental improvement of 10-year CVD risk prediction when added to guideline-recommended risk prediction models (i.e., SCORE2) with and without PRSs. The risk discrimination provided by SCORE2 (Harrell’s C-index = 0.718) was similarly improved by addition of NMR scores (ΔC-index 0.011; 0.009, 0.014) and PRSs (ΔC-index 0.009; 95% CI: 0.007, 0.012), which offered largely orthogonal information. Addition of both NMR scores and PRSs yielded the largest improvement in C-index over SCORE2, from 0.718 to 0.737 (ΔC-index 0.019; 95% CI: 0.016, 0.022). Concomitant improvements in risk stratification were observed in categorical net reclassification index when using guidelines-recommended risk categorisation, with net case reclassification of 13.04% (95% CI: 11.67%, 14.41%) when adding both NMR scores and PRSs to SCORE2. Using population modelling, we estimated that targeted risk-reclassification with NMR scores and PRSs together could increase the number of CVD events prevented per 100,000 screened from 201 to 370 (ΔCVDprevented: 170; 95% CI: 158, 182) while essentially maintaining the number of statins prescribed per CVD event prevented. Overall, we show combining NMR scores and PRSs with SCORE2 moderately enhances prediction of first-onset CVD, and could have substantial population health benefit if applied at scale. ### Competing Interest Statement During the course of this project P.S. became a full-time employee of GSK Plc. All significant contributions to this study were made prior to this role and GSK Plc had no input to the study. J.D. serves on scientific advisory boards for AstraZeneca, Novartis, and UK Biobank, and has received multiple grants from academic, charitable and industry sources outside of the submitted work. A.S.B. reports institutional grants from AstraZeneca, Bayer, Biogen, BioMarin, Bioverativ, Novartis, Regeneron and Sanofi. The remaining authors declare no competing interests. ### Funding Statement This work was performed using resources provided by the Cambridge Service for Data Driven Discovery (CSD3) operated by the University of Cambridge Research Computing Service ([www.csd3.cam.ac.uk][1]), provided by Dell EMC and Intel using Tier-2 funding from the Engineering and Physical Sciences Research Council (capital grant EP/P020259/1), and DiRAC funding from the Science and Technology Facilities Council ([www.dirac.ac.uk][2]). This work was supported by core funding from the: Cambridge BHF Centre of Research Excellence (RE/18/1/34212) and BHF Chair Award (CH/12/2/29428). S.C.R. and S.K. were funded by a British Heart Foundation (BHF) Programme Grant (RG/18/13/33946). S.C.R. was also funded by the National Institute for Health and Care Research (NIHR) Cambridge BRC (BRC-1215-20014; NIHR203312) [*]. X.J. was funded by British Heart Foundation (CH/12/2/29428) and Wellcome Trust (227566/Z/23/Z). L.P. and P.S. were supported by a Rutherford Fund Fellowship from the Medical Research Council grant MR/S003746/1. Y.X. and M.I. were supported by the UK Economic and Social Research Council (ES/T013192/1). S.A.L. was supported by a Canadian Institutes of Health Research postdoctoral fellowship (MFE-171279). E.D.A. holds a NIHR Senior Investigator Award. J.D. holds a BHF Professorship and a NIHR Senior Investigator Award. M.I. is supported by the Munz Chair of Cardiovascular Prediction and Prevention and the NIHR Cambridge Biomedical Research Centre (NIHR203312). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. *The views expressed are those of the authors and not necessarily those of the NIHR or the Department of Health and Social Care. ### Author Declarations I confirm all relevant ethical guidelines have been followed, and any necessary IRB and/or ethics committee approvals have been obtained. Yes The details of the IRB/oversight body that provided approval or exemption for the research described are given below: This study was approved under UK Biobank Projects 30418 and ethics approval was obtained from the North West Multi-Center Research Ethics Committee. I confirm that all necessary patient/participant consent has been obtained and the appropriate institutional forms have been archived, and that any patient/participant/sample identifiers included were not known to anyone (e.g., hospital staff, patients or participants themselves) outside the research group so cannot be used to identify individuals. Yes I understand that all clinical trials and any other prospective interventional studies must be registered with an ICMJE-approved registry, such as ClinicalTrials.gov. I confirm that any such study reported in the manuscript has been registered and the trial registration ID is provided (note: if posting a prospective study registered retrospectively, please provide a statement in the trial ID field explaining why the study was not registered in advance). Yes I have followed all appropriate research reporting guidelines, such as any relevant EQUATOR Network research reporting checklist(s) and other pertinent material, if applicable. Yes All data described are available through UK Biobank subject to approval from the UK Biobank access committee. See for further details. [1]: http://www.csd3.cam.ac.uk [2]: http://www.dirac.ac.uk