Updated review finds PSA screening linked to lower prostate cancer mortality over the long term
A new review led by University of Toronto researchers has shown that prostate-specific antigen (PSA) screening reduces long-term mortality associated with prostate cancer.
These findings, which represent the highest quality evidence to date, were drawn from up to 23 years of follow-up data from over 721,000 clinical trial participants across North America and Europe. Published recently in European Urology, the study’s conclusions aim to settle the controversy around PSA screening and will inform prostate cancer screening guidelines around the world.
“Screening for prostate cancer using PSA has been a matter of debate for many, many years” says Borna Tadayon Najafabadi, a urology resident at U of T’s Temerty Faculty of Medicine.
He co-led the new review paper along with David-Dan Nguyen, a fellow urology resident at Temerty Medicine, and Girish Kulkarni, a surgeon-scientist at the University Health Network and professor of surgery at Temerty Medicine.
Prostate-specific antigen is a protein produced by the prostate that can be measured in a blood test. High levels of PSA can be a sign of prostate cancer, and the PSA test can, in some cases, be an appropriate tool to detect cancer earlier. However, because prostate cancer is often very slow growing, the benefits of screening may not appear until decades later, when some tumours become metastatic.
“Previous evidence syntheses only followed participants for about 10 to 15 years and the effect was not really observable or very small,” says Nguyen, who is completing his PhD as part of Temerty Medicine’s surgeon-scientist training program.
In the new review, the researchers analysed the most recent data — representing the longest follow up thus far — from five large-scale, randomized clinical trials. They found that PSA screening was associated with a clear reduction in the risk of prostate cancer-specific death. According to the study findings, one death is avoided for every 1,000 persons screened.
A Cochrane review — conducted independently and published two weeks later — arrived at similar conclusions with the authors rating the evidence as less certain which, Nguyen says, is likely due to differences in methodology between the two groups.
Both Nguyen and Tadayon Najafabadi acknowledge an important caveat to their work: all five clinical trials included in their review had different approaches to screening. Some trials only screened participants once with a PSA test whereas other trials offered repeat tests. Taking these and other factors into account, the researchers say their findings likely underestimate the true benefit of PSA screening.
“We now know, with high certainty, there is a benefit so we can shift our focus to how do we deliver screening in an effective, safe and equitable way,” says Nguyen.
Currently the Canadian Task Force on Preventive Health Care does not recommend PSA screening at any age because of the previously weak evidence about the test’s benefits and the known harms related to the overdiagnosis and overtreatment of prostate cancer.
Nguyen points out that new approaches to screening — including innovations in microultrasound-guided prostate biopsy and MRI-based risk stratification developed by U of T researchers — have reduced the risks associated with the PSA test and, combined with the newest data, should shift the conversation towards more nuanced discussions around who should undergo PSA screening.
Earl Miller, a patient partner and co-author on the new study, says that he often hears stories from Black men whose requests for a PSA test are turned down by their family doctors, despite Black men having almost double the risk of developing prostate cancer.
“There needs to be a smarter, risk-stratified approach to screening that is centralized and offers the test to people who will benefit the most,” says Nguyen, who previously co-authored an article in The Lancet Regional Health calling for Canada to revisit its approach to prostate cancer screening.
“The Canadian guidelines in place right now are case-by-case, inefficient and inequitable.”
For Nguyen, this research was also an opportunity to test out ottoSR, an AI-driven large language model co-developed by U of T and University of Calgary researchers to speed up the process for systematic reviews.
The researchers teamed up with Christian Cao, a medical student at Temerty Medicine who co-led the creation and testing of the AI tool, to use ottoSR to independently perform a database search, filter the search results and extract the appropriate data.
“When we compared ottoSR’s results with the human results, it was essentially a perfect match,” says Nguyen.
He notes that within a couple of hours, ottoSR completed tasks that took their team of human researchers one to two weeks to finish.
“Systematic reviews are the foundation for clinical guidelines, but they are often slow, expensive and difficult to keep up to date,” says Cao. “This work shows how AI can help accelerate that process while preserving the rigour needed for decisions that affect patient care.”
As a result of this work, Nguyen and OttoSR are part of a consortium that was recently awarded €2 million to develop evidence-based European guidelines, quality indicators and requirements for prostate cancer screening, diagnosis and care.
Closer to home, Nguyen and Tadayon Najafabadi are hoping to take their findings to policymakers at the national and provincial level to advocate for updated recommendations that increase PSA screening access to those who need it.
“This is a very important milestone for prostate cancer screening,” says Tadayon Najafabadi. “Having higher quality evidence will help unify the recommendations and create consistency in what physicians and health care providers do.”
