ATLANTA – Men in their 40s with a low prostate specific antigen can probably safely delay additional testing for 10-15 years.
Young men with a higher baseline level, however, are twice as likely to develop prostate cancer over the same time period and should probably have their PSA tested at regular intervals, Dr. Christopher Weight said at the annual meeting of the American Urologic Association.
Dr. Christopher Weight
His prospective study of 268 men in their 40s showed that none of the men with a baseline PSA of 1.0 ng/mL or less developed high-risk disease by 10 years and only 3% developed it by 15 years. The findings could provide an effective way to risk-stratify young populations, reducing unnecessary testing and the consequences that sometimes follow it, he said.
"We have to admit that we overdiagnose and overtreat men," Dr. Weight said at a press briefing. "But there is danger in completely throwing out the PSA test. Testing men early can help us identify those who can safely delay additional testing and those who will benefit from more frequent tests."
Dr. Weight, a urology oncology fellow at the Mayo Clinic, Rochester, Minn., turned to the Olmsted County cohort for the study data. Since 1990, most of the residents in the county have received their medical care through the Mayo Clinic and its affiliate centers. A linked health records database provides information for long-term population-based studies.
His analysis included 268 men, all of whom had a baseline PSA drawn sometime during their 40s (median age 45 years). The men also had a transrectal ultrasound and digital rectal exam. They have been followed now for up to 20 years, with a median time of 16 years.
Among the cohort, 192 had a baseline PSA of 1.0 ng/mL or lower and 76 had a level of more than 1.0 ng/mL. There were no significant between-group differences in either family history or the results of the rectal exam.
Over the full follow-up period, men with the lower PSA level had a significantly lower risk of exceeding the age-specific cut points for PSA than did men with the higher levels (10% vs. 50%).
By the end of the follow-up period, there were six incident cases of prostate cancer in the low-PSA group, all of which were low-risk disease. This translated to an incidence rate of 1.6 per 1,000 patient/ years, with a mean of 15 years until diagnosis.
Twice as many men in the high-PSA group developed prostate cancer (12). Of these cases, 10 were low-risk disease and 2 high-risk. This translated into a rate of 8/1,000 patient-years, with a mean of 10 years to diagnosis.
The baseline measurement was fairly predictive of 15-year outcomes, Dr. Weight said. A cutoff of 1.0 ng/mL at the initial test had a sensitivity of 67% and a specificity of 74% for predicting the occurrence of prostate cancer. Changing the cutoff to 0.7 ng/mL on the initial test resulted in a sensitivity of 83% and a specificity of 46%
The results show that this single, early PSA level may be helpful in counseling patients about follow-up, said Dr. Scott Eggener, who moderated the briefing.
"The goals of any test are to identify the cohort of people most likely to benefit, and those people in whom screening can be limited to minimize the potential harms of the test," said Dr. Eggener, director of urology outcomes and translational research at the University of Chicago Medical Center. "If a young man has a very low PSA, we can feel comfortable in recommending that he have another test in several years, somewhat like what’s done with a screening colonoscopy. If the level is higher, this patient probably needs to be followed more frequently."
Neither Dr. Weight nor Dr. Eggener had any financial disclosures.