Posted By: Ann Klinck, Communications & Marketing Assistant, ASHG
With growing interest in polygenic risk scores (PRS) and questions arising about its clinical relevance, ASHG and The American Journal of Human Genetics (AJHG) hosted the webinar The Development and Application of Polygenic Risk Scores. The archived video is now available for viewing.
Bruce Korf, AJHG editor and webinar moderator, was joined by Eimear Kenny, PhD, Director of the Center for Genomic Health at the Icahn School of Medicine at Mount Sinai, and Sekar Kathiresan, MD, Director of the Center for Genomic Medicine at Massachusetts General Hospital, and Director of the Cardiovascular Disease Initiative at the Broad Institute.
Origins of Polygenic Risk Scores
With much research leading to the growth of PRS, Dr. Kenny believes the research paper The Correlation between Relatives on the Supposition of Mendelian Inheritance (1918) from R.A. Fisher was a huge contributor to the modern-day field complex trait genetics. Fisher defined the term variance, introduced the analysis of variance to partition observed variation into underlying casual factors, and theorized that several genes could contribute to variation using mendelian segmentation. This work led to understanding the genetic architecture and the prediction of complex traits.
Dr. Kenny went into further detail on the background of PRS, which you can see in the archived webinar (skip to 4:53).
What is a PRS?
A PRS is defined as “a genetic prediction of an individuals’ phenotype. It is calculated by summing across the products of a genome-wide association study (GWAS) effect sizes and number of trait-increasing alleles.” Dr. Kenny said that the ultimate goal of a PRS is to predict the phenotype in an out of sample individual who does not have a recorded phenotype.
So, Where Does GWAS Fit In?
Dr. Kenny emphasized that the explosion of GWAS was a tipping point for risk prediction and has expanded in both number of studies and size. GWASs provide evidence supporting Fisher’s theory that most heritable variation is due to thousands of genetic variants each with a tiny marginal effect.
In other words, the genetic architecture of complex traits is almost always highly polygenic. Essentially, the expansion of PRS and GWAS go hand-in-hand.
Where is PRS Being Seen in the Field?
PRS are influencing a variety of subject areas including epidemiology, statistics, public health, social science, and medicine. Dr. Kathiresan shared the real-world case of a 42-year-old male patient, who suffered a heart attack after being told six months prior that he had only a 1.7% chance of having a heart attack within the next ten years. In this case, other methods of predicting heart attack risk like cholesterol, triglycerides, or blood pressure were not helpful preventatives. As the current heart attack risk prediction model is largely driven by age, heart attack in people under 55 is difficult to predict.
“For early-onset disease, stratifying individuals based on inborn DNA variation is an important option, being as most diseases have an inherited component,” Dr. Kathiresan noted. What this could lead to is earlier interventions outside of lifestyle changes, like the use of a statin medication.
Essentially, over the last 15 years, it’s been found that using a polygenic risk model can identify other at-risk individuals. Dr. Kathiresan believes that in the next five years, polygenic risk will start to be incorporated when calculating a patient’s heart attack risk.
To hear more about the work on heart attack risk and other common complex diseases, check out the full webinar (skip to 35:02).
Representing the Global Population
Though there are millions of GWAS participants, most databases driving the GWAS and PRS research are of European ancestry, while only one-seventh of the world’s population has European ancestry. PRS can be adjusted for ancestry, but if based on current data, scores may not be as accurate for non-European populations.
Dr. Kenny described increasing diversity amongst genetic study participants as a multi-pronged issue (skip to 1:00:12). There are many efforts to recruit diverse populations, not only in ancestry, but in community type and socioeconomic status. Still, there is a lot of work to be done. Dr. Kenny believes that increasing diversity in the scientific community as a whole will allow labs to think about participant diversity more clearly.
Dr. Kathiresan agrees and pointed out that participant diversity is important for not only common disease genetics, but also for rare disease genetics in terms of who has been sequenced and the ability to then interpret rare variation.
PRS is a fast-moving area of interest but is just one developing approach to genomics and health, and there are many others such as transcriptional risk scores, epigenetics, and copy-number variation.
“I am optimistic about our ability as a field to really tackle these questions, problems, and challenges,” said Dr. Kenny in conclusion. “There are reasonable solutions, and we can pursue them. I’m hopeful we as a scientific community engage a lot more with the public, clinicians, and other stakeholders that think about these similar questions.”