The idea of developing a genetic test to assess an individual’s risk of prostate cancer is attractive and feasible, but not practical or cost effective.
An interesting study has shown the feasibility of using a polygenic risk score for prostate cancer screening, but some practical issues remain.
The incidence and burden of prostate cancer is increasing, but there are no internationally accepted population-based screening approaches for its early detection. In an attempt to develop a polygenic risk score for prostate cancer, a group of UK researchers undertook the BARCODE1 study, a prospective, ongoing study involving 69 primary care centres.
Their initial findings, published in the New England Journal of Medicine, suggest their particular polygenic risk score can identify a greater proportion of patients with clinically significant prostate cancer compared to conventional screening methods, although two Australian experts have highlighted that there is still a long way to go before such a test can be used on a broader scale.
A polygenic risk score (created from a panel of 130 single-nucleotide polymorphisms previously linked to prostate cancer) was calculated for over 6000 European men with no personal history of prostate cancer.
“We think that about 60% of prostate cancers have a hereditary component,” said Professor Christopher Hovens, joint group leader of the prostate cancer research group at the Collaborative Centre for Genomic Cancer Medicine.
“The problem is that there’s very few single genes involved, like BRCA1 and BRCA2 for breast cancer. There are lots of variations in your genetics that we know have a small effect on their own but exert more of an influence when considered together as a polygenic risk score.”
The average participant was 61 years old, with one in five participants reporting a family history of prostate cancer. Twelve percent of participants (745 of 6393) had a polygenic risk score in the 90th percentile and were referred to a cancer centre for genetic counselling, where they were offered a PSA test, MRI and biopsy.
Roughly two thirds of patients referred for counselling (468/745) underwent an MRI and biopsy; 187 men (40%) were found to have prostate cancer. One hundred and three men with prostate cancer were deemed to be at intermediate risk or higher (i.e., a Gleason score of at least seven) – but almost three quarters of these cancers would not have been detected through the standard diagnostic and screening pathways used in the UK (which are based on PSA levels and/or Prostate Imaging Reporting and Data System [PI-RADS] scores).
Professor Hovens highlighted that the results of the current study demonstrate the feasibility of using a polygenic risk score in prostate cancer screening, but practical challenges still remained – such as determining whether the patient requires treatment.
“You may have a cancer that we call high-risk, and it may never do anything. Conversely, you can have a lower risk cancer that kills you in a few years,” the principal research fellow in the Department of Surgery at the University of Melbourne said.
“So, you end up trying to balance what’s good for public health with the cost trade off. The majority of men with prostate cancer fall into the low- or intermediate-risk category, and will be told that the best approach to treatment will be to monitor it and see if things get worse. There’s no other common cancer where the recommended treatment option is to do nothing.
“In at least 90% of prostate cancer cases that are diagnosed, the men die with the disease but not of it. However, there’s still about 10% who will die of it, and actually do need treatment… trying to create an accurate, non-invasive test for predicting whether someone has a lethal cancer or not is the Holy Grail.”
The researchers estimated that one in five patients with a polygenic risk score above the 90th percentile and screen-detected cancer would be overdiagnosed, meaning that their cancer would take longer than their remaining life to progress to a clinically significant disease stage.
Professor Niall Corcoran, a urological surgeon and joint leader of the prostate cancer research group at the Collaborative Centre for Genomic Cancer Medicine alongside Professor Hovens, questioned whether the researcher’s choice to measure the number of grade of cancers diagnosed from the polygenic risk score was the correct choice.
“If you were to design a study to show some sort of clinical impact, you’d have to include a control population that didn’t get screened, and you would have to show that there was some kind of survival difference between them. But that would be a huge, expensive study, so I can understand why [the researchers] did what they did,” said Professor Corcoran, who has previously collaborated with the senior authors of the current NEJM paper.
Another limitation of the study was that participation was limited to men of European ancestry, meaning that the results will not generalise to the more diverse general population of the United Kingdom. However, the researchers are not deterred.
“Genome wide association studies have provided data on risk SNPs across diverse ancestral groups, and research focused on the use of genetic ancestry-specific polygenic risk score for risk-based screening is in progress,” they wrote.
“Our study provides a framework on which to build further research on the role of genetic risk in screening for cancer in persons of non-European ancestries. These persons include those at higher risk for prostate cancer such as persons of Black African and Caribbean ancestry, whose lifetime risk in the United Kingdom is quoted as one in four as compared with one in eight among persons of European ancestry.”
Related
The fact that polygenic risk scores require continual updates over time as new links are established also presents challenges.
“The current study used 130 SNPs in their polygenic risk score, but we now know of at least 230 SNPs that are associated with prostate cancer – and this figure is increasing all the time,” Professor Hovens explained.
“If you are going to roll out a test, you have to fixate on what is known today, but in a few years’ time that test will become obsolete.”
While Professor Corcoran broadly supported the idea of personalised risk assessment for prostate cancer, but had concerns about their implementation.
“[One concern] I have about any kind of risk assessment is what age do you do it – when do you decide to pull the trigger to do the test? [The genetic risk] is there from conception, but different cancers have different peaks in incidence,” he said.
“So, you may want to do testing for colon cancer earlier than when you would want to do it for prostate. But if you’re going to do one, should [you] do them all at the same time to get a complete profile?”
