Prostate Cancer Screening May Not Reduce Deaths
Studies Cast Doubt on Usefulness of Common Test for Disease

By Rob Stein
Washington Post Staff Writer
Thursday, March 19, 2009

Men are being urged to carefully consider risks before undergoing prostate cancer screening in the wake of two large, long-awaited studies that did not produce convincing evidence that routine testing significantly reduces the chance of dying from the disease.

The PSA blood test, which millions of men undergo each year, did not lower the death toll from the disease in the first decade of a U.S. government-funded study involving more than 76,000 men, researchers reported yesterday. The second study, released simultaneously, was a European trial involving more than 162,000 men that did find fewer deaths among those tested. But the reduction was relatively modest and the study showed that the tests resulted in a large number of men undergoing needless, often harmful treatment.

Together, the studies -- released early by the New England Journal of Medicine to coincide with presentations at a scientific meeting in Stockholm -- cast new doubt on the utility of one of the most widely used tests for one of the most common cancers.

"Americans have been getting screened for prostate cancer because there is this religious faith that finding it early and cutting it out saves lives," said Otis W. Brawley of the American Cancer Society. "We've been doing faith-based screening instead of evidence-based screening. These findings should make people realize that it's a legitimate question about whether we should be screening for prostate cancer."

Other experts were more circumspect, arguing that the European study did indicate at least some benefit for some men, and that the U.S. trial could eventually confirm those findings as it follows the men for longer periods. But they agreed that the new findings should prompt patients and their doctors to discuss the risks and benefits of the testing.

"It shouldn't be a knee-jerk response to get tested," said Christine D. Berg of the National Cancer Institute, which sponsored the U.S. study. "We should be telling these guys to go talk to their physician and say, 'In light of the current evidence and what you know about me and my health, what should I do?' "

Some researchers, however, remain supportive of routine testing, saying the U.S. study has flaws that could have limited its ability to detect a reduction in deaths.

"I don't think that screening should be summarily dismissed based on these trials," said E. David Crawford, a urology professor at the University of Colorado at Denver who helped conduct the U.S. study and heads the Prostate Conditions Education Council, which promotes testing. "I think they say we should be more smart when we screen."

The findings address perhaps the most important and contentious issue in men's health: how best to detect and treat prostate cancer. The disease is diagnosed in more than 218,000 U.S. men each year and about 28,000 die of it, making it the most common cancer after skin cancer and the second-leading cancer killer among men.

The PSA test, which measures a protein produced by prostate tissue called the prostate-specific antigen, has significantly increased the number of prostate cancer cases being caught at early stages. But it has been far from clear whether that translates into a reduction in deaths from the disease. Prostate cancer often grows so slowly that many men die from something else without ever knowing they had it.

Because it is not clear precisely what PSA level signals the presence of cancer, many men experience stressful false alarms that lead to surgical biopsies, which can be painful and in rare cases can cause serious complications.

Even when the test detects a real cancer, doctors are uncertain what, if anything, men should do about it. Many are simply monitored. Many others, however, undergo surgery, radiation and hormone treatment, which often leave them incontinent, impotent and experiencing other sometimes debilitating or even possibly life-threatening complications.

"I know guys who are morbidly depressed because of the complications of their prostate cancer treatment," Brawley said. "I know three people who attempted suicide. I know widows of guys who died from their treatment. There are significant harms associated with over-treatment of prostate cancer."

Because of the uncertainty, many major medical groups have stopped recommending routine PSA testing. Nevertheless, its use remains widespread, and many experts were hoping the two large trials would help settle the issue.

In the U.S. study, researchers randomly assigned 76,693 men ages 55 to 74 at 10 centers, including Georgetown University, to receive either six annual screenings consisting of PSA testing and physical examinations or whatever their doctors recommended on their own, which could include screening.

After seven years, 17 percent more prostate cancers were diagnosed after 10 years. But there was no significant difference in deaths from the disease between the two groups.

Although the men will continue to be followed for at least 13 years, and a benefit might emerge with more time, an independent panel monitoring the study decided the researchers had a duty to inform the participants of the interim findings and make the results public.

The researchers noted that there were actually more deaths overall in the screened group -- 312 vs. 225 -- and they could not rule out that the excess may have been the result of over-treatment.

In the European study, 162,243 men ages 55 to 69 in seven countries were randomly assigned to undergo PSA screening every four years, or no screening. After a median follow-up time of nine years, 20 percent fewer prostate cancer deaths were found among those screened. Because of the study's design, however, several experts said that reduction was hard to interpret. At best, it means about 10,000 men would have to be screened for about 10 years to prevent seven deaths. Put another way, 1,410 men would need to be screened and 48 would have to be treated to prevent one death.

"It's very disturbing," said Fritz H. Schroder of the Erasmus Medical Center in the Netherlands, who led the study. "That means in order to save one life, you treat a very large number of men."

Experts cautioned that the decision to undergo screening remains individual. For men whose family members have died from prostate cancer, are relatively young and know they are at risk, the downside of potentially undergoing unnecessary treatment may be worthwhile. For others, especially older men with shorter life spans, it may not.

"Some men would say, 'If I can reduce my chance of dying from cancer, I'll take that risk and face the music.' Other men would say, 'Gee if you have to diagnose 50 to save one life, my chances are high I'll be part of the 49. I'll take my chances without it.' I think that's reasonable. This isn't a one-size-fits-all result," said Michael J. Barry of Harvard Medical School, who wrote an editorial on the study.

View all comments that have been posted about this article.

© 2009 The Washington Post Company