Mass General Brigham has started offering a blood test that estimates a person’s inherited risk for eight different cardiovascular conditions, moving polygenic risk scoring from research labs into clinical practice. The test, built with Broad Clinical Labs, was validated using data from two of the largest biobank cohorts in the United States, including the NIH’s All of Us Research Program. Patients can now ask a clinician to order the test on a self-pay basis, but open questions about how stable these genetic risk labels remain over time could shape whether the results lead to better care or unnecessary worry.
Why polygenic cardiac risk testing is entering clinics now
Heart disease remains the leading cause of death in the United States, and standard risk calculators rely on factors like cholesterol, blood pressure, and smoking status that can change over a lifetime. Polygenic risk scores take a different approach: they aggregate the small effects of thousands of genetic variants to estimate how likely someone is to develop a condition. Until recently, these scores lived almost entirely inside research databases. The new test from Mass General Brigham and Broad Clinical Labs is designed to change that by packaging a validated polygenic risk report into a format clinicians can order and act on.
According to a company announcement, the assay is currently available as a clinician-ordered blood test that patients pay for out of pocket, outside of typical insurance coverage pathways. The report estimates inherited risk for eight conditions, including coronary artery disease and atrial fibrillation, and is intended to complement-rather than replace-traditional clinical risk assessments. Clinicians receive an interpretive summary that places a patient’s score into population-based categories such as elevated, average, or reduced genetic risk.
The practical question for patients is whether learning about elevated genetic risk early enough can reduce the chance of a heart attack, stroke, or arrhythmia. If a person lands in a high-risk category for coronary artery disease or atrial fibrillation, a doctor might recommend earlier statin therapy, closer monitoring, or lifestyle changes years before symptoms appear. That is the promise. The tension is that polygenic scores are not static labels. A preprint on medRxiv examining the instability of high polygenic risk classification found that as scoring models improve, some patients initially flagged as high risk get reclassified downward. If those patients had already started preventive medications or undergone additional testing based on the earlier score, the interventions may have been unnecessary. Patients whose initial high-risk classification is later downgraded after integrative scoring could, in theory, show lower rates of unnecessary preventive interventions than those whose scores remain unchanged, but no published cohort data yet confirms that hypothesis.
Validation data from All of Us and Mass General Brigham Biobank
The scientific foundation for the test is laid out in a peer-reviewed study in the cardiology literature. Researchers developed and validated a clinical polygenic risk report for eight cardiovascular conditions using large biobank cohorts, including the All of Us Research Program and the Mass General Brigham Biobank. The study compared people in the highest genetic risk group against those with average risk and reported elevated odds for conditions such as coronary artery disease and atrial fibrillation, though the exact hazard ratios for each of the eight conditions are not reproduced in available source summaries.
The All of Us cohort gave the researchers a particularly large and diverse validation dataset. The NIH describes All of Us as the largest integrated genomics and health database in the world, linking genomic data directly to electronic health records. That linkage allowed the research team to check whether the polygenic scores actually predicted real diagnoses recorded in clinical charts, not just self-reported outcomes. The Mass General Brigham Biobank served a similar function, grounding the validation in a health system where patients’ genetic data could be matched against years of documented care.
A separate peer-reviewed paper published earlier in Nature Medicine described the methodological groundwork for translating polygenic risk scores into a clinical assay, covering laboratory workflow, quality control, and reporting design. That earlier work established the technical pipeline that the new eight-condition test builds on, including how results are formatted for clinicians who may not have genetics training. Together, these studies form the evidentiary backbone for moving from research-grade scoring to a regulated, billable test.
Unstable risk labels and the limits of current scoring
The biggest unresolved problem is reliability across time. Polygenic risk scores are calculated using statistical models trained on existing genetic datasets. As those datasets grow and models are retrained, the boundary between “high risk” and “average risk” can shift. The medRxiv preprint on classification instability directly addresses this issue, proposing integrative scoring as one way to reduce the chance that a patient’s risk category flips with the next model update. But the preprint has not yet undergone peer review, and no large-scale clinical trial has measured whether integrative scoring actually reduces unnecessary interventions in practice.
For now, Mass General Brigham’s offering represents a snapshot of risk based on the best-available models at a particular moment. Clinicians will have to decide how to handle the possibility that a patient’s category might change as algorithms are refined. One option is to treat the polygenic result as a modest risk modifier-similar to a family history note-rather than a decisive trigger for long-term medication. Another is to document clearly that the score reflects current knowledge and may be updated in the future, much like reinterpreting some genetic variants as new evidence emerges.
A second gap involves equity. Polygenic risk scores have historically performed better in populations of European ancestry because the genome-wide association studies used to build them were disproportionately drawn from those groups. The All of Us program was designed in part to address that imbalance by enrolling a more diverse participant pool, but whether the eight-condition test performs equally well across all ancestry groups has not been fully detailed in the available source material. Without transparent performance metrics by ancestry, there is a risk that the test could overestimate or underestimate risk for some patients, potentially widening existing disparities in cardiovascular care.
Cost and access add another layer of uncertainty. The test is positioned as clinician-ordered and self-pay, which may limit uptake to patients who can afford discretionary genetic testing. Insurance coverage policies for polygenic risk scoring are still emerging, and payers typically require evidence that a test changes management and improves outcomes before agreeing to reimburse it. Until such data exist, the clinical use of this assay is likely to be concentrated in specialty prevention clinics and among patients who are already highly engaged with their cardiovascular risk.
How clinicians might use the results
Despite these caveats, the test could influence care in several concrete ways. For a middle-aged patient with borderline cholesterol and no symptoms, a high polygenic risk score for coronary artery disease might nudge a clinician toward earlier statin therapy or more aggressive lifestyle counseling. For someone with a strong family history of atrial fibrillation, an elevated score could justify closer rhythm monitoring or a lower threshold for evaluating palpitations. Conversely, a low genetic risk result might provide reassurance in cases where traditional risk calculators give ambiguous answers, though experts caution against using a favorable score as a reason to ignore standard prevention guidelines.
Effective implementation will likely depend on decision-support tools that integrate polygenic results into existing risk calculators rather than presenting them in isolation. Clinicians will also need educational resources to explain what the scores do-and do not-mean. A “high” category does not guarantee disease, and an “average” category does not confer immunity. Communicating probabilistic risk in a way that avoids fatalism or false reassurance will be a central challenge as more patients encounter these reports.
What patients should know before ordering the test
For patients considering the new blood test, several practical questions are worth raising with a clinician. First, how might the result change current care? If a doctor cannot identify any decision that would be altered by knowing the polygenic score, the benefit of testing may be limited. Second, what follow-up would be recommended for a high-risk result, and are those interventions supported by evidence independent of the genetic score? Third, how will the result be documented and revisited if scoring models evolve?
Patients should also ask about ancestry-specific performance, especially if they come from groups that have historically been underrepresented in genetic research. Understanding that the score is one piece of a broader risk picture-not a definitive verdict-can help keep expectations realistic. As with many emerging genomic tools, the value of polygenic cardiovascular risk testing will depend less on the raw number than on how thoughtfully it is used in the context of each person’s overall health, preferences, and access to care.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
