Polygenic risk scores (PRSs) are an emerging genetic tool designed to estimate an individual’s risk of developing certain diseases, such as coronary artery disease (CAD). By analyzing numerous genetic markers across the genome, PRSs aim to provide a comprehensive assessment of a person’s genetic predisposition to conditions like heart disease. These scores hold promise for enabling personalized healthcare strategies. However, new research from the Perelman School of Medicine at the University of Pennsylvania, published in JAMA and presented at the American Heart Association’s Scientific Sessions in Chicago, highlights critical variability in PRSs, potentially undermining their reliability for individual risk predictions.

The study analyzed genetic data from over 260,000 individuals of diverse backgrounds to evaluate how different PRSs perform in predicting CAD risk. While most of the scores delivered consistent predictions at the population level, they showed significant variability at the individual level. Some participants were categorized as both high and low risk by different PRSs, raising concerns about inconsistent clinical advice depending on the score used.

Researchers from Penn Medicine, along with collaborators from the NIH’s All of Us Research Program, the Penn Medicine Biobank, and UCLA ATLAS Precision Health Biobank, compared 48 different PRSs for CAD. While 46 scores showed similar predictions for population-wide risk, 20% of participants had at least one PRS categorizing them simultaneously in the highest and lowest 5% of risk. These discrepancies highlight the challenges of using PRSs for individual-level predictions.

“The goal of PRSs is to help identify people at higher genetic risk for diseases like heart disease,” said Dr. Scott M. Damrauer, Vice Chair for Clinical Research in Penn’s Department of Surgery and a vascular surgeon at Penn and the CMCVAMC. “But for clinical use, it’s important that the results are consistent and reliable, especially when decisions about someone’s health are on the line.”

Challenges in Individual-Level Predictions

Although PRSs are effective at identifying population-level trends, their reliability for individual predictions remains limited. The variability in predictions poses challenges for personalized medicine, where accurate risk assessments are crucial for tailoring prevention and treatment strategies.

Sarah Abramowitz, the study’s lead author and a medical student at the Zucker School of Medicine at Hofstra/Northwell, emphasized the need to refine these tools. “Our research underscores a critical gap in our understanding of PRSs, which has implications for their use in personalized medicine,” she explained. “While these scores show promise for population-level CAD risk assessment, we need more robust methods to quantify and communicate the uncertainty of individual-level predictions.”

The inconsistency of PRSs highlights a gap in the methods used to translate genetic data into actionable clinical insights. To avoid misleading patients, healthcare providers must address these uncertainties before PRSs can be widely implemented in CAD risk assessment.

Recommendations for Future Use

The findings suggest that while PRSs have potential, they should be integrated cautiously into clinical care. Researchers recommend that clinicians use PRSs as part of a comprehensive risk assessment strategy, incorporating other factors such as clinical history and lifestyle. This holistic approach would mitigate the impact of inconsistencies in PRSs and provide a more accurate overall assessment of an individual’s risk.

Further research is needed to refine PRSs and develop methods for quantifying the uncertainties in individual predictions. This work will ensure that PRSs can reliably guide clinical decisions and support their broader adoption in personalized healthcare.

Reference:

1. Abramowitz SA, Boulier K, Keat K, et al. Evaluating Performance and Agreement of Coronary Heart Disease Polygenic Risk Scores. JAMA. Published online November 16, 2024. doi:10.1001/jama.2024.23784

(Input from various sources)

(Rehash/Yash Kamble/MSM)