As many as two of every five men whose prostate cancer was caught through a PSA screening test have tumors too slow-growing to ever be a threat, says a new study that raises more questions about the controversial tests.

The work "reinforces the message that we are overdiagnosing prostate cancer," said Dr. Len Lichtenfeld of the American Cancer Society, who was not involved in the new study.

More than 186,000 U.S. men will be diagnosed with prostate cancer this year, and nearly 29,000 will die, according to cancer society estimates. Most men over 50 have had a blood test that measures prostate specific antigen, or PSA, mostly for routine screening.

There begins the list of problems: Most men who undergo a biopsy for an abnormal PSA test don't turn out to have prostate cancer; high PSAs often signal a benign enlarged prostate. Of those who do have cancer, there's no proof yet that early detection saves lives - as most prostate tumors grow so slowly that had they not been screened, those men would have died of something else without the anxiety.

How many? Estimates vary widely. Enter the new study, which tracked prostate cancer diagnosed in U.S. men ages 54 to 80 between 1985 and 2000, and used three different models developed by cancer centers to more accurately estimate overdiagnosis.

Depending on how it's calculated, anywhere from 23 percent to 42 percent of PSA-detected cancers would otherwise never have been detected in the man's lifetime, concluded the team led by researchers at Erasmus University Medical Center in the Netherlands.

The study was published online Tuesday by the Journal of the National Cancer Institute.

Why is overdiagnosis such a concern? Because finding an early tumor forces men to choose among contested treatments - "watchful waiting," surgery, hormone therapy, radiation. And because some treatments can cause incontinence and impotence, men whose tumors wouldn't have been a threat can suffer serious side effects for no gain.

In fact, national health guidelines issued last year said men over age 75 shouldn't undergo PSA screening, while younger men should make an individual choice after hearing the pros and cons and weighing their own cancer risk.

The new study's estimate of U.S. overdiagnosis probably is too low - because since 2000, doctors have begun performing biopsies for lower PSA levels than once were the trigger, wrote Dr. Michael Barry of Massachusetts General Hospital in an accompanying editorial.

It's a confusing issue, acknowledged the cancer society's Lichtenfeld.

It boils down to: "If we diagnose this disease, are we making your life better? We know that for other cancers," such as breast, cervical and colorectal, which have strong evidence showing early detection hugely improves survival, he said.

Major studies are under way that in a few years should offer better guidance for prostate cancer screening, and scientists are furiously hunting new tests that might help pinpoint who has a worrisome tumor and who can relax.

"We're waiting for that evidence. Hopefully we'll have it in the not too distant future, but we really don't have the best answer right now," added Lichtenfeld, who stressed the importance of discussing potential benefits and risks with a doctor.