A multicondition polygenic risk report validated in the U.S. health care system could help clinicians detect inherited cardiovascular risks earlier, refine prevention strategies, and guide more individualized care.
Research: Development and validation of a clinical polygenic risk report for eight cardiovascular diseases in a US-based health system. Image credit: ArtemisDiana / Shutterstock
In a recent study published in Journal of the American College of Cardiology (JACC)researchers described the development and validation of a unified polygenic risk score (PRS) for eight cardiovascular diseases using data from 245,394 All of Us (AOU) Research Program participants and 53,306 Mass General Brigham Biobank (MGBB) participants.
The integrated PRS platform demonstrates robust risk stratification and clinically structured reporting that generally matches or exceeds individual input models, providing a transparent framework for identifying individuals at high genetic risk that may be missed by traditional clinical markers.
Background on cardiovascular polygenic risk scores
Cardiovascular disease (CVD) remains a major cause of mortality worldwide, but its incidence is characterized by a complex genetic structure, including substantial heritability and pleiotropy. Although some heart diseases are caused by rare, high-impact mutations in a single gene (monogenic), decades of research have demonstrated that the vast majority of cases are caused by thousands of common genetic variations across the genome, each with tiny individual effects.
Traditional clinical risk models, represented by pooled cohort equations (PCE), use demographic and phenotypic markers such as blood pressure and cholesterol to estimate risk, whereas PRS quantifies genetic risk from common genetic variants. However, comprehensive methodologies for risk stratification are still lacking. Systematic reviews and meta-analyses of available PRS approaches have shown that they often fail to capture the full range of genetic risk, especially in younger or “intermediate-risk” populations.
Therefore, there is an urgent need for a standardized “consensus” approach that can aggregate these scores across multiple conditions into a single, reliable report.
Integrated PRS study design and validation
This study aimed to address these knowledge gaps by creating a transparent pipeline to introduce genetic risk stratification into routine preventive care. The entire project consisted of multi-stage development and validation studies across three large-scale biobanks.
The training dataset was derived from genomic and electronic health record (EHR) data of 245,394 All of Us (AOU) participants (mean age = 51.7 ± 17.0 years). Seven trait models were trained using this dataset, but as standardized Lp(a) measurements were not available at AOU, the lipoprotein(a) elevation model was trained at UK Biobank. The training method focused on eight clinical conditions: atrial fibrillation (AF), coronary artery disease (CAD), type 2 diabetes mellitus (T2DM), thoracic aortic aneurysm (TAA), extreme hypertension, venous thromboembolism (VTE), severe hypercholesterolemia, and elevated lipoproteins (a).
The PRSmix software package was used to integrate published PRSs from the PGS catalog. For internal model testing prior to external validation, an 80/20 stratified split was applied to the AOU cohort and an equivalent UK Biobank-based approach was used for Lp(a).
External model performance validation was subsequently performed in an independent cohort of 53,306 participants from the Massachusetts General Brigham Biobank (MGBB). To account for genetic diversity, this study used computed principal components (PCs; derived from a 1000 genome-based shared PC space) to adjust for age, sex, and genetic ancestry.
Specifically, discrimination was assessed using the C statistic, and model calibration was assessed across age, gender, and ancestry subgroups.
Stratification of PRS risk across cardiovascular characteristics
The new integrated PRS platform demonstrated consistent risk stratification and nearly matched or exceeded the performance of individual input scores across eight characteristics. However, predictive performance varied by condition, with more modest discrimination for some outcomes such as VTE, TAA, and extreme hypertension.
The most striking finding of this study was elevated lipoprotein(a) levels, with those in the high genetic risk category (top 10%) having a significantly 41.0-fold increased odds of having elevated lipoprotein(a) levels (95% CI: 27.0-62.2) compared to those with average genetic risk (P < 0.0001).
Less dramatically, high-risk individuals (top 10%) for severe hypercholesterolemia (odds ratio (OR) = 4.1), CAD (OR = 3.73), T2DM (OR = 3.1), AF (OR = 3.0), and extreme hypertension (OR = 2.1) exhibited several times the risk of their average-risk counterparts. This study also showed that elevated genetic risk was common in this biobank-based analysis, with 71.2% of the MGBB population having at least one PRS-defined threshold corresponding to at least a 3-fold increased relative genetic risk for one or more of the eight traits.
Importantly, this study found that adding PRS to existing clinical tools, such as pooled cohort equations (PCE), significantly improved “net reclassification.” In CAD, incorporating genetic scores improved risk classification by 17% (P < 0.0001) for patients previously considered to be at "borderline" or "intermediate" risk. Prospective follow-up (over a median of 7.6 years) confirmed that higher PRS was associated with the development of CAD, AF, T2DM, VTE, and TAA, even in participants younger than 50 years.
Clinical significance of multi-condition PRS testing
This study represents an important step toward a clinically orderable multicondition cardiovascular PRS test. By validating an integrated PRS panel across eight conditions, this study’s novel approach provided a scalable framework to identify individuals who have normal conventional biomarkers but may harbor previously unrecognized genetic genetic risks.
However, the authors emphasize that limitations remain at this time. Although the score was performed across ancestry groups, predictive power remained strongest in European populations, highlighting the need for more diverse research data. The authors also noted that broader prospective validation and further evidence of clinical utility are needed before PRS-based treatment pathways are fully established.
In the future, this report will be available as a clinically orderable test, allowing physicians to use genetic “risk enhancers” to inform prevention discussions, targeted screening, lifestyle counseling, and clinically appropriate medication decisions for patients.
Polygenic risk scores for both eight cardiovascular traits @MassGenBrigham and @a_o_a_o_o_o_o_ – Stackable – Strongly indicates risk. Not yet introduced in clinical practice https://t.co/xEPwou7G5u pic.twitter.com/PkFkQ4dLvf
– Eric Topol (@EricTopol) April 29, 2026
Reference magazines:
- Misra, A. et al. (2026). Development and validation of a clinical polygenic risk report for eight cardiovascular diseases in a US-based health system. Journal of the American College of Cardiology. Advance online release. Doi: 10.1016/j.jacc.2026.03.035. https://www.jacc.org/doi/full/10.1016/j.jacc.2026.03.035
