NEW YORK – Researchers want to prospectively validate a polygenic risk score that has shown promise in pinpointing which heart failure patients will benefit most from beta blockers.
While beta blockers are a first-line therapy for controlling blood pressure and heart rate in heart failure patients with reduced ejection fraction (HFrEF), many don't experience sustained heart function improvement. And despite years of research into the potential genomic underpinnings of that variation in response between individuals, there doesn't seem to be a single candidate gene that predicts beta-blocker response.
"There's an entire literature surrounding beta-blocker pharmacogenetics," said David Lanfear, first author of a retrospective study exploring the impact of a polygenic risk score (PRS) in predicting beta-blocker response in heart failure patients published in Circulation: Genomic and Precision Medicine last month. But while there have been interesting research findings, "there was never really a proven, clinically useful genetic tool to say, 'Should we give, or should we not give, the drug to [a particular patient] with heart failure?'" said Lanfear, an advanced heart failure and transplant cardiology specialist and codirector of the Center for Individualized and Genomic Medicine Research at Henry Ford Health in Detroit.
Lanfear had previously started a registry for collecting genetic data from heart failure patients at Henry Ford Health with plans to use the data to perform a genome-wide association study (GWAS) to identify potentially impactful pharmacogenomic variants. But when the research didn’t identify specific variants with an outsized impact on drug response, he and his colleagues were curious if an approach that captured variation across patients' genomes could prove more successful. The team thus created a PRS — which it referred to as a polygenic response predictor — to determine how beta blockers might impact HFrEF patients' survival.
The PRS is calculated based on 44 genetic markers identified from the GWAS conducted on data from the registry, dubbed the Henry Ford Heart Failure Pharmacogenomic Registry, but the function of many of the individual variants remains unknown. The model also accounts for clinical factors that contribute to patients' survival by incorporating Meta-Analysis Global Group in Chronic Heart Failure scores. Patients that fall in the 30th percentile of scores are determined to be beta-blocker responders who are likely to see a survival benefit from treatment.
In their published study, the investigators retrospectively validated the PRS using data from more than 7,100 heart failure patients of European ancestry who participated in the UK Biobank. Less than one-quarter of heart failure patients were determined to be responders to beta blockers based on the PRS, while the remaining 77 percent were deemed non-responders.
Patients predicted to be responders had a significantly reduced risk of death when on beta blockers, while non-responders showed little benefit, the study authors wrote in their paper.
Although more research is needed, the results suggest that the PRS could help clinicians "target the drug to people who get the most benefit," Lanfear said. Hopefully, "this is the future of precision medicine, where you can get a genomic profile and say, 'You have a really high likelihood of responding to drugs A and B, but [you don't need] drugs C and D.'"
Figuring out optimal alternative treatments for non-responders is another area for future research, according to Jasmine Luzum, a coauthor on the paper and an assistant professor at the University of Michigan College of Pharmacy. During a presentation at GenomeWeb's Updates in Precision Medicine and Pharmacogenomics conference in Orlando, Florida, last month, she suggested that since heart failure patients tend to be on multiple medications, this research, if confirmed in further studies, could allow doctors to de-escalate beta blockers in those the PRS deems non-responders. Polypharmacy prevalence ranges from 17.2 percent to 99 percent for heart failure patients, according to a 2021 systematic review.
This latest paper validates previous proof-of-concept findings in about 1,200 HFrEF patients published in Circulation: Heart Failure in 2020. "This genetic profile, once proven in trials, could be used to target beta blockers to the subset of heart failure [patients] with reduced ejection fraction who are likely to benefit and allow a large number of patients to forgo unnecessary treatment, fundamentally changing our treatment approach from beta blockers for all to genetically targeted therapy," the authors wrote in that paper, though they acknowledged that additional research in patients with non-European ancestries "is urgently needed." The earlier findings, like the more recent analysis, used data from patients of European ancestry.
Creating a model that's effective for patients of various ancestral backgrounds is a common challenge when constructing PRS, Lanfear said. Just because a PRS is validated in one ethnic group doesn't mean it will be equally accurate in other populations. The research team is seeking grant funding to develop a PRS for Black heart failure patients and has already started constructing a model using Henry Ford Heart Failure Pharmacogenomic Registry data. The investigators hope to also use large-scale datasets from the National Institutes of Health's All of Us Research Program and the Department of Veterans Affairs' Million Veteran Program.
"It's tricky constructing [PRS] across ancestral groups, because the prevalence of variants and linkage patterns are very different," Lanfear said. "We're finding it hard to have polygenic scores that are universally useful."
The team is also seeking funding for a prospective validation trial of the PRS they've developed for patients of European ancestry and wants to apply the PRS approach to predicting patients' responses to other heart failure drugs.
Luzum, at the March conference, noted that PRS are more complicated to study and implement than a single-gene pharmacogenetic approach. She predicted that PRS are the future of pharmacogenomics, though it'll be at least a decade, or longer, before the necessary validation studies are completed and guidelines and physician education modules are created facilitating their integration into clinical care.
"I feel like we've done a very good job in pharmacogenetics of identifying the low-hanging fruit of individual genes or variants that have a large enough effect size on their own to be clinically useful," she said in an interview after the conference. "We need to move the field forward and look at more of these polygenic effects of multiple variants that each have small effect sizes."