The potential of polygenic risk scores for prostate cancer

We look at new research that introduces a genetic risk scoring system for the triage of prostate cancer patients in primary care

In a paper recently published in the British Journal of Cancer, researchers from the University of Exeter have used polygenic risk scores (also known as genetic risk scores) to triage patients with possible prostate cancer symptoms. The study explains that this method, deployable in local GP surgeries, is shown to be more reliable than current first-line tests for prostate cancer, and that it could remove the need for invasive tests for men presenting with prostate cancer symptoms and increase survivability for those testing positive.

Prostate cancer in the UK

Prostate cancer is on the rise, now accounting for about a quarter of new cancer diagnoses among UK men.

Symptoms, including an increased need to urinate, difficulty urinating and the presence of blood in the urine, commonly overlap with other conditions, making it difficult to diagnose. UK GPs refer around 800,000 men with symptoms for tests every year, of whom only about 52,000 will actually be diagnosed with the disease.

The condition is also highly variable: in some cases it is slow growing and can be lived with for years without causing symptoms, but at the same time it is the second most common cause of cancer deaths in male patients – of which 14% could be prevented by earlier diagnosis.

Like many types of cancer, a person’s risk of developing prostate cancer is affected by their genome. Men whose father or brother have had prostate cancer are 2.5 times more likely to develop the condition, and men with pathogenic BRCA gene variants are twice as likely to develop it.

Current testing methods

The current first-line test for prostate cancer is a blood test that detects prostate-specific antigen (PSA) levels. Everyone who has a prostate makes PSA, but how much of it gets into the blood varies according to age and the health of the prostate.

This test is only used for men who display symptoms, such as urination problems or erectile dysfunction, because the test can generate false positives (only one-third of patients with positive results actually have cancer) and false negatives (it fails to detect around 15% of cancers).

While those with prostate cancer have elevated levels of PSA, several non-cancerous conditions can cause elevated levels too. This unreliability means that a biopsy is generally needed to make a definitive diagnosis. Because biopsies are invasive and can be painful, the NHS aims to reduce the need for them by offering most men an MRI scan first. Not all negative cases can be eliminated this way, however.

What does the research tell us?

A new study from the University of Exeter demonstrates that clinicians can avoid reliance on invasive biopsies by using genomic-based polygenic risk scores to triage prostate cancer patients.

These scores were developed by researchers using genome-wide association study data. They incorporated the effects of more than 250 gene variants known to influence the risk of developing prostate cancer together with age risk predictors to come up with a triage system. Researchers then tested the system against a UK Biobank dataset of over 6,000 men who had seen a GP with prostate cancer symptoms, and found that it routinely flagged up men who went on to be diagnosed with prostate cancer within two years of their GP consultation.

Results indicated that, thanks to this genomic triage system, about one-fifth of symptomatic GP referrals could be fast-tracked for further investigation, while another 40% could be spared invasive biopsies.

Most prostate cancer diagnoses occur after a man reports his symptoms to his GP. So, while this new genomic triage system has not been tested on men without symptoms and cannot currently be used to predict future prostate cancer in healthy men, it builds a compelling case to explore how polygenic risk scores could be applied in those with symptoms. The potential benefits are considerable: reduced time to diagnosis, reduced need for invasive testing procedures, reduced time to effective treatment, and improved survivability rates of this increasingly common disease.