Data support PSA screening, particularly in Black men, for whom there is a shortage of clinical trial evidence
Prostate-cancer screening based on prostate-specific antigen (PSA) levels in the blood was introduced-and readily adopted-in the United States around the late 1980’s.
But this screening method, in which higher PSA levels in the blood could indicate the presence of prostate cancer, became controversial, largely because of what are now recognized as faulty short-term data from randomized trials. PSA-screening trials also included very few Black men, who are at highest risk of dying of prostate cancer.
As a result, many experts in this field maintain that national prostate-cancer screening guidelines don’t accurately reflect the benefits of PSA screening.
Based on that assumption, University Hospitals’ Jonathan Shoag, MD, and a team of researchers from Case Western Reserve University and the Case Comprehensive Cancer Center, Weill Cornell Medical Center, the Fred Hutchinson Cancer Research Center, and others, set out to assess the tradeoffs of PSA screening using long-term epidemiologic data.
Shoag and his co-authors presented their findings in a late-breaking abstract at the American Urological Association’s annual meeting this month. Their study was published May 15 in The New England Journal of Medicine Evidence, accompanied by a Patient Platform.
Other researchers from CWRU/UH are Daniel Spratt, MD; Alexander Hill, School of Medicine student: Jennifer Cullen, PhD; Sarah Markt, ScD; and Erika Trapl, PhD.
Using death from prostate cancer as its endpoint, the team set to answer the question of how many more men needed to be diagnosed or treated to prevent one individual from dying of the disease. This calculation of the “number needed to diagnose” or “number needed to treat” was done using two approaches:
- The first is what the team called the “excess incidence” approach-taking the difference in diagnoses in years before and after PSA screening was introduced.
- The second involved building a model with inputs like age of diagnosis and stage of diagnosis to gain insight into whether or not the cancer would have been found regardless of PSA screening. Both methods gave the team roughly the same numbers of additional men diagnosed because of the PSA test.
With a near 50% reduction in death from prostate cancer since the introduction of PSA screening as the denominator, Shoag and his team said they found impressive numbers to support the benefits of that screening method.
“No matter the assumptions,” Shoag said, “the data showed lower numbers than prior estimates, many in the low single digits, for the number needed to treat to prevent a prostate-cancer death. This result was observed in all men, and especially for Black men.”
The team emphasized that including long-term data is important because the goal of screening is to find and treat a cancer early. Though men are often diagnosed with prostate cancer in their 50’s and 60’s, most die of the disease closer to 80 years old.
As such, studies using a 10-year interval from the start of screening don’t allow enough time to see a mortality benefit, but do see more cancers diagnosed early-the goal of screening.
Other than the ability to analyze longer-term data, using epidemiologic data has the distinctive benefit of being able to provide insight into the disease in Black men-despite their historic exclusion from clinical trials, the team concluded. “We show convincingly here that the tradeoffs of screening are remarkably favorable and make clear that young otherwise healthy men-in their 40s and 50s-should definitely have at least baseline PSA screening,” Shoag said. “The metric that people use to evaluate screening’s tradeoffs is often the number needed to diagnose or treat to prevent a death from cancer, and we find those numbers to be much more favorable towards PSA screening than prior estimates, remarkably so in Black men. The numbers quoted in many current guidelines are simply misleading. These should change.”