ORLANDO – A single prostate specific antigen level measurement before age 60 years in the setting of opportunistic screening predicted lethal prostate cancer in later life, according to findings from a nested case-control study involving Physicians’ Health Study participants.
The findings, which confirm prior observations that midlife PSA levels predict subsequent development of lethal prostate cancer in an unscreened population – and which extend the findings to a cohort where opportunistic screening occurs, could have implications for reducing unnecessary screening, biopsy, and treatment, Dr. Mark A. Preston reported at the annual meeting of the American Urological Association.
Of 14,916 men from the randomized, controlled Physicians’ Health Study who submitted a blood specimen prior to enrollment, 234 with total PSA levels available prior to age 60 years had prostate cancer diagnosed between 1983 and 1993, and 711 served as age-matched controls.
As of follow-up through 2012, metastatic or fatal disease had developed in 60 case patients.
"We went on to ask three main clinical questions about how [midlife PSA] might be clinically relevant," Dr. Preston, who is a urologist at Massachusetts General Hospital, Boston, said during a press briefing at the meeting.
The first question was whether baseline PSA predicts lethal cancer.
The median PSA levels at three age groups evaluated in the study (40-50 years, 50-55 years, and 55-60 years) were 0.68 ng/mL, 0.88 ng/mL, and 0.96 ng/mL, respectively, he said.
Using men with a PSA below the median as a reference group, the risk of lethal prostate cancer for those with PSA above the 75th percentile was "quite significant," with those aged 40-50 years having a sixfold increased risk, those aged 50-55 years having a fourfold increased risk, and those aged 55-60 having a 10-fold increased risk, he noted.
"This was even more significant for those with PSA above the 90th percentile," he said.
The next question was whether those with a very low PSA before age 60 years could forego further PSA testing.
Compared with men with a PSA above the median, those with PSA below the 25th percentile had a very low risk of lethal prostate cancer (odds ratio, 1.6), he said.
The third question was whether men with one very low PSA level measurement between ages 40-50 years could forego any further PSA testing.
"The numbers were small, but what we did find was that there were still men who had a PSA level at that very low level who still went on to die from prostate cancer at some point," he said, noting that this likely means that one PSA measurement is inadequate.
That one PSA, however, may be useful for risk stratification, allowing for a longer interval between testing for those with such low risk.
Early PSA was found in this study to have very good predictive value for lethal prostate cancer, with an area under the curve of about 0.80 for all of the age groups, he said.
"We concluded that a single baseline PSA among men at midlife strongly predicts the subsequent development of lethal prostate cancer in a U.S. population subject to opportunistic PSA screening, and also that there was no lower limit of PSA baseline at which no men developed lethal prostate cancer," he said.
The findings are notable, given the current controversy over the usefulness of PSA screening.
"The reason these data are so important – and this dovetails with a lot of other data that are out there – is that we’re trying to make a more sensible screening strategy for men," according to Dr. Scott E. Eggener, director of translational and outcomes research, section of urology at the University of Chicago, who moderated the press briefing.
This is very similar to how colonoscopy is used, he said, explaining that patients get a baseline colonoscopy and the findings determine the interval needed for follow-up.
"We’ve never done that in PSA screening for prostate cancer. It has always started at a certain age, everyone is treated the same way, it’s done once a year (at least in the United States)," he said.
The current data, when considered in light of other similar studies, are very compelling in that they demonstrate the need for a more individualized, tailored approach to screening based on factors such as race, family history, age, and baseline PSA.
"These data suggest that baseline PSA is an incredibly powerful predictor of events that are maybe destined to happen 10 or 20 years down the road, and we need to incorporate that in our thinking about screening," Dr. Eggener said.