NEW YORK – Polygenic risk scores (PRS) inched closer to clinical adoption over the past year, with several studies demonstrating their clinical utility and feasibility of implementing them into clinical workflows.
In particular, 2024 saw considerable growth of evidence supporting the clinical utility of PRS for cardiovascular disorders such as coronary artery disease (CAD), especially when combined with other health-related data such as social determinants of health and lifestyle factors.
At the American College of Cardiology's annual meeting in April, for example, numerous researchers showcased evidence for improving risk calculations using PRS in a variety of cardiac disease-related contexts, including high systolic blood pressure and subclinical CAD, as well as certain cerebrovascular events, such as ischemic stroke and transient ischemic attacks.
Scientists affiliated with the Mayo Clinic, which is actively pursuing strategies for integrating PRS into clinical practice, presented evidence that combining PRS with social determinants of health and lifestyle (SDLS) improved risk estimations for coronary heart disease (CHD, a term often used interchangeably with CAD) relative to each metric alone and compared to the pooled cohort equations (PCE), an established clinical risk tool based on known cardiovascular risk factors like age, sex, blood pressure, and cholesterol levels. Both PRS and SDLS correlated with clinical risk of CHD in that study and, when combined, showed an additive effect.
Two similar studies, one conducted by researchers at the Icahn School of Medicine at Mount Sinai in New York, the other carried out in China, found similar additive benefits to risk estimations for cardiovascular disease by combining PRS with other clinical factors. The Mount Sinai researchers reported improved prediction of certain cardiovascular disease subtypes, such as CAD characterized by high LDL and high lipoprotein(a), while the Chinese group reported enhanced risk estimations for both early- and late-onset CAD and for ischemic stroke.
More research from Mount Sinai, although separate from what was presented at ACC, identified 17 new CAD-related gene variants related to CAD risk by using a CAD-predictive machine-learning model called in silico scores for CAD (ISCAD) to analyze a combination of known risk factors, pooled cohort equations, and polygenic risk scores.
ISCAD incorporates laboratory measurements, vital signs, medications, symptoms, and genetic data from the UK Biobank, the All of Us biobank and the BioMe Biobank to collectively mitigate the influence of misclassification biases that stem from relying too heavily on diagnostic codes.
Experts praised the method for its integrative approach to refining CAD risk and expressed hope that ISCAD, which the Mount Sinai researchers stated they do not intend to patent, will be applied to more diseases in the future.
Researchers affiliated with large biobanks like the US National Institutes of Health-funded All of Us research program also released a bevy of studies this year, many of which advanced efforts to establish accurate PRS for CAD and other cardiovascular illnesses, as well as diabetes and individual drug responses.
Beyond heart disease
Researchers associated with All of Us also published a series of studies this year that collectively sought to refine PRS for numerous conditions and to address the underrepresentation of individuals of non-European ancestry in the program’s biobank.
One group of researchers from Vanderbilt University, Baylor College of Medicine, the Broad Institute, and elsewhere combined genome sequence with electronic health record data for over a quarter of a million All of Us individuals, approximately three-quarters of whom came from groups that have been traditionally underrepresented in biomedical research. The study unearthed more than 3,700 genetic variants with ties to 117 conditions and pointed to high replication between some of the associations in participants with European and African ancestry.
Another group combined genomic data from the All of Us biobank with that from the Million Veteran Program, Biobank Japan, and other large research efforts to calculate PRS for type 2 diabetes (T2D) clusters linked to CAD and other cardiometabolic traits.
Further support for PRS in cardiovascular disease and T2D over the last year came from the multinational Polygenic Risk Methods in Diverse Populations (PRIMED) consortium, which is developing methods to improve the use of PRS for predicting disease and health outcomes in populations of diverse ancestry.
Over the course of 2024, PRIMED investigators published studies that reinforced the notion that social determinants of health should be included with PRS in estimating risk, that PRS could be used to augment gestational T2D risk calculations, and that individuals with T2D share many genetic risk factors for cardiovascular disease with the general population.
Meanwhile, a group of researchers associated with the National Human Genome Research Institute (NHGRI)-funded Electronic Medical Records and Genomics (eMERGE) study published PRS for asthma, breast cancer, atrial fibrillation, chronic kidney disease, coronary heart disease, hypercholesterolemia, obesity, prostate cancer, and both type 1 and type 2 diabetes this year.
The eMERGE study mainly draws its data from the All of Us biobank and makes portability of PRS one of its key components. Nearly half of the eMERGE data consist of racial and ethnic minority groups, and the eMERGE investigators claim to have also made sure to include individuals from rural or low-income settings, another underserved group of people.
The eMERGE study investigators' method for recalibrating the tails of PRS distributions, which determine actionability, allowed them to optimize their calculations for diverse ancestries. They then integrated their PRS into a report to be returned to both patients and their physicians, which the researchers hope will help smooth the path toward clinical implementation of PRS.
Further paving the way toward clinical implementation, UK-based firm Genomics Ltd published several studies and entered into a number of partnerships this year, all aimed at demonstrating the feasibility and utility of several PRS applications.
Early in the year, the company published the results of its Healthcare Evaluation of Absolute Risk Testing (HEART) study on the feasibility of integrating PRS reports into routine clinical care. That study showed that both physicians and patients readily accepted the addition of a cardiovascular PRS and that in some cases, results triggered changes in treatment, which the firm said suggests clinical utility.
PRS in life insurance
Genomics said the HEART study was the first trial to put genetic risk tools into the hands of primary care physicians in the UK National Health System. In parallel, the company has brought its PRS tools across the pond to the US through a partnership with life insurance underwriter MassMutual.
Under that agreement, announced in April, MassMutual policyholders will have access to Genomics' at-home saliva-based PRS test for atrial fibrillation, breast or prostate cancer, cardiovascular disease, high blood pressure, high low-density lipoprotein cholesterol, low bone density, and type 2 diabetes.
Life and disability insurance underwriters, researchers, and policymakers have been debating the role of any genetic testing for these kinds of insurance for several years and have been evaluating the possible utility of PRS as a means to risk-stratify policyholders.
No life or disability insurance underwriter appears to have yet formally adopted PRS into their risk calculations nor do any appear to require genetic testing. Consumers and researchers have voiced concern about genetic discrimination over that possibility, and a patchwork of legislative acts and voluntary moratoria have arisen over the past few years.
In the case of the partnership between Genomics and MassMutual, the insurer sponsors Genomics' tests but does not have access to individuals' results. Both companies hope that by receiving individually tailored, actionable health advice, policyholders will be more likely to take steps to reduce their chances of developing each condition.
"The work we're doing with MassMutual is very explicitly not about using genetics for underwriting," said Peter Donnelly, CEO and cofounder of Genomics. "There are ways of using genetics in the insurance context that are good for the policyholders and good for the insurers because their interests are aligned in a way where the insurance company never sees the genetic data, they never see the results, but they benefit because of increased longevity on the part of the policyholders who had the test."
Challenges for clinical use
One criticism often levied against the potential clinical utility of PRS is that the risk they represent tends to be very marginal.
PRS related to cardiovascular disease and certain cancers –– breast and prostate cancer, in particular –– are the most commercially advanced largely because of the measures available to mitigate risk well in advance of disease occurrence through lifestyle changes and medical interventions.
In yet another study done by Genomics, the company found that PRS for coronary artery disease and breast cancer may reflect levels of risk similar to that of rare monogenic mutation carriers, which could argue for the inclusion of PRS in wider screening programs.
In fact, PRS for those and other indications are already beginning to enter large public screening programs. This past year, the state of Alabama, in partnership with nonprofit contract research organization Southern Research, MyOme, and Broad Clinical Labs, launched the Catalyst program, a public health initiative that provides genetic testing for personalized health risk assessments free of charge to Alabama residents. Risk assessments by PRS currently cover coronary artery disease, breast cancer, and type 2 diabetes, and the collaborators plan to expand these to include certain rare diseases, other cancers, and other cardiac and metabolic disorders, among other common diseases.
As part of the program, BCL and MyOme are jointly fielding a new method of blended genome-exome sequencing, a combined strategy of low-pass whole-genome sequencing and deep-coverage whole-exome sequencing that can be used for single-gene and pharmacogenomic testing, as well as issuing PRS.
Despite the forward momentum in PRS development, their clinical utility remains unclear in numerous contexts, including for cardiovascular diseases, especially when examining the differences in accuracy between populations and individuals.
A study led by investigators at the University of Pennsylvania, for example, revealed the need for better statistical methods to quantify the uncertainty inherent in PRS and strategies to communicate this uncertainty to physicians and their patients. Across 46 coronary heart disease PRS with "practically equivalent" population-level performance, the researchers found that the same individuals classified as having the highest risk by one PRS also had the lowest risk according to other PRS. Overall, 80 percent of participants had at least one risk estimate placing them in the top quintile for CHD risk, as well as one score placing them in the bottom quintile.
Although differences in the underlying genetic datasets used to develop the scores appeared likely to be at least one driver of score variability, the researchers noted that this was unlikely to fully explain the variability in individual-level risk predictions, underscoring the need for continued research before widespread clinical adoption of PRS.
Earlier in the year, the American College of Medical Genetics and Genomics (ACMG) expressed doubts about the clinical readiness of PRS for preimplantation genetic embryo testing, an application that has attracted a measure of commercial interest despite widespread scientific skepticism about its readiness for prime time.
Echoing a statement made last year, the ACMG said in its report that the clinical utility of PRS "remains unproven and PRS should not be offered as a clinical service for preimplantation genetic testing at this time … PRS may carry little practical information, and the risks of performing IVF solely for the purpose of PGT-P may outweigh the potential benefits."