Liz Szabo
USA Today
Screening men with the PSA test increases their chances of being diagnosed with prostate cancer, but doesn't reduce their overall risk of death, according to a large, long-running government study.
The results - the newest update to a study launched in 1993 - are the latest to question the value of using the PSA to routinely screen all older men for prostate cancer, a common practice since the 1980s.
"It doesn't seem as if the PSA has any impact on mortality or death," says Thomas Kirk, president of Us TOO, a patient advocacy group not involved in the new study, which included 76,685 men ages 55 to 74. "It would be nice if we could say that a PSA screening could save your life, but we can't say that."
Critics of the PSA, a blood test that measures a protein called prostate-specific antigen, note that the screenings lead many men to be diagnosed with relatively harmless cancers that never threaten their lives, but whose treatment leaves them incontinent or impotent. In the new study, researchers found that men randomly assigned to get PSA tests were 12% more likely to be diagnosed with prostate cancer, but no more likely to die. That suggests that 12% of men screened with the PSA were "overdiagnosed," or diagnosed with cancer that didn't need to be found, leading to unnecessary treatment, says co-author Philip Prorok, of the National Cancer Institute, which funded the $250 million trial.
Based on concerns like these, the U.S. Preventive Services Task Force in October recommended against routine PSA screenings for most men. A National Institutes of Health panel last month also suggested that many men with early prostate tumors could safely delay treatment. Because the PSA detects so many harmless tumors, about 100,000 of the 240,000 men diagnosed with prostate cancers a year could opt for "active surveillance" rather than immediate treatment, the panel said.
Researchers are looking for better early warning signs of prostate cancer, called biomarkers, that may someday provide a more accurate screening test, says Prorok.
Yet study co-author Gerald Andriole says the PSA, while imperfect, is still the best screening test available for prostate cancer, which kills more than 33,000 Americans a year. And they note it's possible that their study could have underestimated the benefits of screening. That's because only 85% of the men randomly assigned to get routine PSA screening actually did so. In the comparison arm of the trial, however, about half of the men not assigned to get a PSA decided to get one on their own, Prorok says.
The new study's findings also differ in some ways from other trials. A large European study of prostate cancer screening, whose early results were released in 2009, found a 20% decline in deaths from prostate cancer among screened men, compared to unscreened men, but no reduction in overall death rates, says Derek Raghavan, a cancer specialist and spokesman for the American Society of Clinical Oncology, who was not involved in either study.
In spite of the new study's limitations, the American Cancer Society's Otis Brawley described it as "well-designed and well-run."
While screening average men may not save lives, Andriole says men at higher risk of death from the disease - such as African-Americans or those with a father or brother who had advanced prostate cancer - may still benefit from screenings. Andriole also would like to see the PSA used in "smarter," more selective ways.
Today, men often undergo yearly screenings beginning at age 50.
Doctors such as Raghavan would like to see studies evaluating the merits of screening younger men, such as those in their 40s. Those with negligible PSA levels could skip screening for a time, while those with concerning PSA levels would be evaluated more closely. But Prorok, who notes the high cost of such trials, says it will be very difficult to launch another study to thoroughly answer those questions.
Brawley acknowleges that the new results may complicate men's decisions about screening. As long as patients are well-informed, Brawley says, they can make reasonable decisions either way.
"If a man wants to be screened, he should be screened," says Andriole, of the Washington University School of Medicine. "But he should be advised, before the blood is drawn, what the pros and cons are."
USA Today