However, nearly twice as many men who had surgery reported incontinence and impotence after two years, researchers report in the New England Journal of Medicine.

"These are very compelling data," said Dr. Durado Brooks, director of prostate and colorectal cancers at the American Cancer Society, who was not involved in the research.

Both he and study author Dr. Timothy Wilt of the University of Minnesota School of Medicine said the results suggest that many men who have received surgery in the past probably didn't need it. "I don't think there's any question that many of those cancers were overtreated," said Brooks, who stressed that there is no way to know on an individual basis.

Prostate cancer is the second-leading killer in men, with more than 240,000 new diagnoses and 28,000 deaths every year. While many doctors still screen for the disease, more and more medical organizations are recommending against routinely doing so. And even with a cancer diagnosis, it's not clear that aggressive treatments such as surgery or radiation are always helpful.

That's because treatments have side effects, whereas the tumor might never have caused problems if left alone.

The new study is based on 731 men who had been diagnosed with prostate cancer, often as a result of screening with the prostate-specific antigen (PSA) test. Researchers then randomly assigned the men to prostate removal or observation only.

More than a decade after the tumor was discovered, 5.8 percent of the men who received surgery had died from prostate cancer or its treatment, compared to 8.4 percent of patients where doctors just kept an eye on the tumor. Overall, 47 percent of the men in the surgery group died during the study, compared to 50 percent of the others.

Both differences could have been due to chance, the researchers found. By contrast, more than one in five of the men who went under the knife experienced a complication of the surgery, including one death.

"Death from prostate cancer, with observation, was very uncommon," Wilt told Reuters Health.

"We think our results apply to the vast majority of men diagnosed with prostate cancer today," he said.

CANCERS OVERTREATED?

The study, which only dealt with localized tumors, leaves open the possibility that men with very high PSA levels - greater than 10 nanograms per milliliter - may still benefit from surgery.

About 17 percent of men will be diagnosed with prostate cancer, but the risk of dying from the disease is only about three percent. That's because prostate tumors tend to grow so slowly that other health problems often kill a patient first.

"There has been a greater consideration on the part of many urologists to take less aggressive approaches with some low-grade cancer, particularly in older men," Brooks said. "However it's been very difficult because cancer is such a scary word. When people hear 'cancer,' they want something done about it and they want something done yesterday."

The study is known as PIVOT -- the Prostate Cancer Intervention Versus Observation Trial, and was first presented more than a year ago at a cancer meeting in Washington, D.C.

The men were enrolled in the study between 1994 and 2002, when PSA testing was becoming popular. Men diagnosed today, said Wilt, are probably even less likely to benefit from surgery "because patients are picked up a lot earlier with smaller tumors. They're very unlikely to be bothered by their cancer in their lifetime."

At the two-year mark, 17 percent of those who had undergone surgery reported urinary incontinence. The rate was six percent in the observation group. In addition, 81 percent of surgery patients had impotence problems compared with 44 percent in the observation group.

Brooks told Reuters Health that, in the past, nine out of ten U.S. men with a prostate cancer diagnosis have chosen aggressive treatment, frequently regretting the long-term side effects.

"Probably 30 to 40 percent of men could reasonably take a less-aggressive approach to their prostate cancer," he said.

"My hope is that men will look at this information and try to make some decisions in a less emotional way," said Brooks. "But the reality is, it usually takes 10 years for significant shifts in medical practice as a result of new findings in the medical literature. So it's going to be a slow transition to get people to back away from this type of diagnosis."

SOURCE: bit.ly/MmFuWK New England Journal of Medicine, online July 18, 2012.

• A Permanent Economic Emergency