In the Journals Plus

Midlife PSA levels predict risk for aggressive prostate cancer among black men

Show Citation

November 27, 2018

Baseline PSA levels obtained during midlife appeared to predict the long-term risk for aggressive prostate cancer among a cohort of black men, according to a study published in European Urology.

PSA levels between 1 ng/ml and 3 ng/ml were associated with highest risk.

“The totality of evidence from this study and previous work provides strong support for use of midlife PSA level to determine a personalized screening strategy. Targeted screening based on a midlife PSA might identify men at high risk while minimizing screening in those men at low risk,” Mark A. Preston, MD, MPH, assistant professor of surgery in the division of urology at Brigham and Women’s Hospital, and colleagues wrote.

Although data have shown baseline PSA levels obtained between the ages of 40 and 60 years strongly predict prostate cancer incidence and mortality in cohorts of primarily white men, few studies have examined the risk specifically among black men. Because of known racial disparities in prostate cancer, research on screening among black men has become a priority for the U.S. Preventive Services Task Force, according to study background.

For this reason, Preston and colleagues evaluated whether baseline PSA levels obtained during 2002 and 2009 predicted long-term aggressive prostate cancer risk among black men in the Southern Community Cohort Study — a prospective cohort of 86,000 men and women recruited through community health centers across 12 southern states, with the highest representation of black populations among existing cohorts.

Median follow-up was 9 years.

Overall, 197 men were diagnosed with incident prostate cancer by 2015, of whom 91 had aggressive disease. Researchers matched these men according to age, blood draw date and enrollment site to 569 controls from the study who did not have prostate cancer at the time of the matched case’s diagnosis.

Researchers then compared PSA levels in blood samples within blinded case-control data sets.

Median PSA levels among controls were 0.72 ng/ml for those aged 40 to 49 years, 0.8 ng/ml for those aged 50 to 54 years, 0.94 ng/ml for those aged 55 to 59 years, and 1.03 ng/ml for those aged 60 to 64 years. PSA levels in the 90th percentile for these age groups were 1.68 ng/ml, 1.85 ng/ml, 2.73 ng/ml and 3.33ng/ml.

Ninety-five percent of all men with prostate cancer and 97% of men with aggressive prostate cancer had baseline PSA levels above their age-specific median.

The OR for total prostate cancer comparing PSA levels higher than the 90th percentile vs. less than or equal to median levels was 83.6 (95% CI, 21.2-539) for those aged 40 to 54 years and 71.7 (95% CI, 23.3-288) for those aged 55 to 64 years.


For aggressive prostate cancer, ORs were 174 (95% CI, 32.3-infinity) for the 40- to 54-year age group and 51.8 (95% CI, 11-519) for the 55- to 64-year age group.

The investigators noted the use of a composite endpoint of aggressive prostate cancer based on stage, grade and mortality as a study limitation.

“While these findings do not imply that these younger men should immediately undergo prostate biopsy or definitive treatment, they suggest that this group would benefit from more intensive PSA screening for earlier identification of cancer while it is potentially curable,” Travis Gerke, ScD, epidemiologist at Moffitt Cancer Center, said in a press release. “This ‘smarter screening’ approach may allow us to identify and follow men at higher risk while reducing the population-wide harms of screening low-risk men.”

Several questions remain regarding the use of midlife PSA levels in developing personalized screening strategies, Rebecca E. Graff, PhD, Linda Kachuri, PhD, and John S. Witte, PhD, all of University of California, San Francisco, wrote in an accompanying editorial.

“One is the particular parameters of the screening program,” they wrote. “When should the midlife PSA measurement be taken? What PSA cutoffs would dictate varying screening frequencies? What should the frequency of PSA screening be for the different risk groups?

A multipronged randomized control trial would provide evidence to address these questions, but it would also be prohibitively expensive and take decades to complete,” they added. “It will thus be important to leverage existing large-scale observational data from diverse, well-characterized populations to attempt to identify optimal screening parameters.” – by Jennifer Southall

Disclosures: Preston and Gerke report no relevant financial disclosures. Please see the study for all authors’ relevant financial disclosures. The editorial authors report no relevant financial disclosures.

Search for articles by this author

itj+ Infographic

itj+ Perspective