New blood test for prostate cancer ‘ninety-nine percent accurate’

Researchers from University Hospitals of Leicester NHS Trust have shown that a new blood test can diagnose prostate cancer and identify what stage the disease is at with 99 per cent accuracy.    

The new blood test builds on previous methods used to detect prostate cancer, that look for an elevated level of a ‘prostate specific antigen’ (PSA) and if the test is positive then a patient is sent for a biopsy.  

The method developed by scientists at Nottingham Trent University, aims to reduce the use of a biopsy, while still accurately confirming prostate cancer.  

In the new test after the blood is extracted, computers analyse the sample for signs of the disease and categorise it as either low, intermediate or high-risk. 

Seventy two men with no symptoms of prostate cancer participated in the study. Using the new blood test, 31 of the men were diagnosed with low risk cancer, and 41 with high risk cancer.

Professor Masood Khan, consultant urologist at University Hospitals Leicester NHS Trust and an honorary professor at Nottingham Trent University, said: “Improving our ability to detect men harbouring clinically significant prostate cancer is vital to save lives. MRI scans can help spot a tumour but are not accurate enough to make a conclusive diagnosis on their own. PSA blood tests can give false positives, which can cause unnecessary psychological impact through being misdiagnosed.”

“If this test can be proven to work at scale, then it will not only reduce the burden on the NHS but also spare men from having unnecessary invasive procedures and help clinicians to decide whether to ‘watch’ or ‘actively manage’ patients, even when they are asymptomatic but have mildly higher PSA levels.”

The next phase is to conduct a larger study.

The research, ‘Identifying prostate cancer and its clinical risk in asymptomatic men using machine learning of high dimensional peripheral blood flow cytometric natural killer cell subset phenotyping data’, is published in the journal, eLife https://elifesciences.org/articles/50936.