Predicting Prostate Cancer Risk: Are Polygenic Risk Scores Ready for Prime Time?

Howard Wolinsky

September 05, 2023

DNA testing for prostate cancer — of the patients' inherited DNA and their tumors' somatic DNA — is increasingly used in the US to determine whether and how to treat low-grade, localized prostate cancers.

Another genetic approach, known as the polygenic risk score (PRS), is emerging as a third genetic approach for sorting out prostate cancer risks.

PRS aims to stratify a person's disease risk by going beyond rare variants in genes, such as BRCA2, and compiling a weighted score that integrates thousands of common variants whose role in cancer may be unknown but are found more frequently in men with prostate cancer. Traditional germline testing, by contrast, looks for about 30 specific genes directly linked to prostate cancer.

Essentially, "a polygenic risk score estimates your risk by adding together the number of bad cards you were dealt by the impact of each card, such as an ace vs a deuce," said William Catalona, MD, a urologist at Northwestern University Feinberg School of Medicine, Chicago, known as the father of prostate-specific antigen (PSA) screening.

In combination, these variants can have powerful predictive value.

Having a tool that can mine the depths of a person's genetic makeup and help doctors devise a nuanced risk assessment for prostate cancer seems like a winning proposition.

Despite its promise, PRS testing is not yet used routinely in practice. The central uncertainty regarding its use lies in whether the risk score can accurately predict who will develop aggressive prostate cancer that needs to be treated and who won't. The research to date has been mixed, and experts remain polarized.

"PRS absolutely, irrefutably can distinguish between the probability of somebody developing prostate cancer or not. Nobody could look at the data and argue with that," said Todd Morgan, MD, a genomics researcher from the University of Michigan in Ann Arbor. "What [the data] so far haven't really been able to do is distinguish whether somebody is likely to have clinically significant prostate cancer vs lower-risk prostate cancer."

The Promise of PRS in Prostate Cancer?

Prostate cancer — the most common type of solid tumor in men and the second most common cancer killer — is a major target for PRS because it is considered one of the most heritable cancers, according to Burcu Darst, PhD, a genetic epidemiologist at Fred Hutchinson Cancer Center in Seattle.

Research in the area has intensified in recent years as genome-wide association studies (GWAS) have become more affordable and the genetic information from these studies increasingly has been aggregated in biobanks.

"Because the sample sizes now are so much bigger than they used to be for GWAS studies, we're able to develop much better polygenic risk scores than we were before," said Darst.

Darst is lead author on the largest, most diverse prostate GWAS analysis, which led to the development of a PRS that is highly predictive of prostate cancer risk across diverse populations.

In the 2021 meta-analysis, which included 107,247 case patients and 127,006 control patients, Darst and colleagues identified 86 new genetic risk variants independently associated with prostate cancer risk, bringing the total to 269 known risk variants.

Compared with men at average genetic risk for prostate cancer — those in the 40% to 60% genetic risk score category — men in the top 10% of the risk score (90% to 100%) had between a 3.74-fold to fivefold higher risk for prostate cancer. However, the team did not find evidence that the genetic risk score could differentiate a person's risk for aggressive vs nonaggressive disease.

As Darst's team continues to improve the PRS, Darst says it will get better at predicting aggressive disease. One recent analysis from Darst and colleagues found that "although the PRS generally did not differentiate aggressive versus nonaggressive prostate cancer," about 40% of men who will develop aggressive disease have a PRS in the top 20%, whereas only about 7% of men who will develop aggressive tumors have a PRS in the bottom 20%. Another recent study from Darst and colleagues found that PRS can distinguish between aggressive and nonaggressive disease in men of African ancestry.

These findings highlight "the potential clinical utility of the polygenic risk score," Darst said.

Although the growing body of research makes Catalona, Darst, and others optimistic about PRS, the landscape is also littered with critics and studies showcasing its limitations.

An analysis published in JAMA Internal Medicine earlier this year found that, compared with a contemporary clinical risk predictor, PRS did not improve prediction of aggressive prostate cancers. Another recent study, which used a 6.6 million-variant PRS to predict the risk of incident prostate cancer among 5701 healthy men of European descent older than age 69, found that men in the top 20% of the PRS distribution "had an almost three times higher risk of prostate cancer" compared with men in the lowest quintile; however, a higher PRS was not associated with a higher Gleason grade group, indicative of more aggressive disease.

"While a PRS for prostate cancer is strongly associated with incident risk" in the cohort, "the clinical utility of the PRS as a biomarker is currently limited by its inability to select for clinically significant disease," the authors concluded.

Utility in Practice?

Although PRS has been billed as a predictive test, Catalona believes PRS could have a range of uses both before and after diagnosis.

PRS may, for instance, guide treatment choices for men diagnosed with prostate cancer, Catalona noted. For men with a PRS that signals a higher risk for aggressive disease, a positive prostate biopsy result could help them decide whether to seek active treatment with surgery or radiation or go on active surveillance.

PRS could also help inform cancer screening. If a PRS test found a patient's risk for prostate cancer was high, that person could decide to seek PSA screening before age 50 — the recommended age for average-risk men.

However, Aroon Hingorani, MD, a professor of genetic epidemiology at the University College London, expressed concern over using PRS to inform cancer screenings.

Part of the issue, Hingorani and colleagues explained in a recent article in The BMJ, is that "risk is notoriously difficult to communicate."

PRS estimates a person's relative risk for a disease but does not factor in the underlying population risk. Risk prediction should include both, Hingorani said.

People with high-risk scores may, for instance, discuss earlier screening with their clinician, even if their absolute risk for the disease — which accounts for both relative risk and underlying population disease risk — is still small, Hingorani and colleagues said. "Conversely, people who do not have 'high risk' polygenic scores might be less likely to seek medical attention for concerning symptoms, or their clinicians might be less inclined to investigate."

Given this, Hingorani and colleagues believe polygenic scores "will always be limited in their ability to predict disease" and "will always remain one of many risk factors," such as environmental influences.

Another caveat is that PRS generally is based on data collected from European populations, said Eric Klein, MD, chairman emeritus of urology at the Cleveland Clinic and now a scientist at the biotechnology company Grail, which developed the Galleri blood test, which screens for 50 types of cancer. While a valid concern, "that's easy to fix ultimately," he said, as the diversity of inputs from various ethnicities increases over time.

Although several companies offer PRS products, moving PRS into the clinic would require an infrastructure for testing, which does not yet exist in the US, said Catalona.

Giordano Botta, PhD, CEO of New York–based PRS software start-up Alleica, which bills itself as The Polygenic Risk Score Company, said "test demand is growing rapidly." His company offers PRS scores that integrate up to 700,00 markers for prostate cancer depending on ancestry and charges patients $250 out of pocket for testing.

Botta noted that thousands of American patients have undergone PRS testing through his company. Several health systems, including Penn Medicine, Brigham and Women's Hospital, and the University of Alabama at Birmingham, have been using the test to help "see beyond what traditional risk factors allow," Botta said.

However, this and other PRS tests are not yet widely used in the primary care setting.

A major barrier to wider adoption is that experts remain divided on its clinical utility. "People either say it's ready and it should be implemented, or they say it's never going to work," said Sowmiya Moorthie, PhD, a senior policy analyst with the PHG Foundation, a Cambridge University–associated think tank.

Klein sits in the optimistic camp. He envisions a day soon when patients will undergo whole-genome testing to collect data on risk scores and incorporate the full genome into the electronic record. At a certain age, primary care physicians would then query the data to determine the patient's germline risk for a variety of diseases.

"At age 45, if I were a primary care physician seeing a male, I would query the PRS for prostate cancer, and if the risks were low, I would say, 'You don't need your first PSA probably until you're 50,' " Klein said. "If your risk is high, I'd say, 'Let's do a baseline PSA now.' "

We would then have the data to watch these patients a little more closely, he said.

Moorthie, however, remains more reserved about the future of PRS. "I take the middle ground and say, I think there is some value because it's an additional data point," Moorthie said. "And I can see it having value in certain scenarios, but we still don't have a clear picture of what these are and how best to use and communicate this information."

For more news, follow Medscape on Facebook, X, Instagram, and YouTube.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.