Many men who are told they do not have prostate cancer based on the results of a widely used blood test probably do have tumors, researchers reported yesterday.

The conclusion is based on a new study of nearly 3,000 men who were told by their doctors that they had nothing to worry about because their scores on the popular prostate-specific antigen (PSA) test fell within the "normal" range. The study found that 15 percent of them actually had tumors.

Although most of the cancers appeared to be so slow-growing that they were unlikely to ever require treatment, a small but significant portion of the malignancies were likely to become life-threatening, the study found.

"It's clear from our findings that there are many men with what is considered a normal PSA who are harboring cancer, and in some cases it may be aggressive prostate cancer," said Ian M. Thompson, chief of urology at the University of Texas Health Science Center at San Antonio, who led the study, published in today's issue of the New England Journal of Medicine.

The findings, the first to validate the suspicion that PSA testing is missing many cancers, rekindled a debate among physicians about how aggressively the tests should be used. Many experts argue that the PSA test is overused, frequently prompting men to undergo unnecessary biopsies and treatment for cancer that is growing so slowly it would never bother them. Others say that the opposite is true and that the guidelines for interpreting the results should be changed to catch more cancers.

"We're missing some dangerous cancers," said William J. Catalona, a professor of urology at Northwestern University who advocates lowering the cutoff point for ordering biopsies based on PSA results.

But others argued that there is still insufficient evidence that finding more cancers and treating them would save more lives.

"We are missing these prostate cancers, but we still don't have an answer to the fundamental question, which is: If we found them and treated them, would we make men live longer?" said Bruce "Ned" Calonge, chairman of the U.S. Preventive Services Task Force, which makes medical recommendations for the federal government. "That's that missing piece of evidence."

Prostate cancer strikes an estimated 220,900 U.S. men each year and kills about 28,900, making it the most common malignancy in males.

The prostate is a chestnut-size gland near the bladder that produces fluid for semen and surrounds the passageway that carries urine. The PSA test detects a protein produced by prostate cells called prostate-specific antigen, which can signal the presence of cancer.

But the test has long been controversial. Because prostate cancer often causes no problems, some experts think the test has resulted in many men undergoing unnecessary treatment. Aside from unneeded biopsies, surgery and radiation to remove or kill the cancer often leave men impotent and incontinent.

Recent research has hinted that many men may have tumors even though their PSA tests are below 4 nanograms per milliliter of blood -- the point at which most doctors order a biopsy to look for cancer. As a result, some experts have been arguing that the current threshold for doing a biopsy is too high, missing many cancers. They advocate that the level be reduced to as low as 2.5.

In the new study, researchers performed biopsies on the prostates of 2,950 men who had PSAs below 4. Doctors found prostate cancer in 449 men, or 15.2 percent. While most of the cancers appeared to be the type that would not be likely to go on to pose a health risk, 67 of those cancers -- 14.9 percent -- were found to be aggressive.

Some experts, however, said the findings should not prompt a lowering of the threshold for performing biopsies because that would result in many men undergoing the procedures and probably being treated for cancers that would never pose a risk.

"We really don't know if any of these cancers, including the high-grade cancers, would really go on to become significant cancer. The question is: Where are the trade-offs?" said Howard L. Parnes of the National Cancer Institute, who helped conduct the new study. "You will have to do a lot of extra biopsies to find a little more cancer. That's the problem with lowering the threshold."

But Jamie Bearse of the National Prostate Cancer Coalition, a patient advocacy group, said his organization advocates a much more aggressive approach.

"We strongly believe that it saves lives. I know if it were me, personally I would want to know and treat it rather than not know and play Russian roulette," he said.

Catalona agreed.

"Just to say that you don't want to open a Pandora's box and see what's there for fear you might be overtreating people -- I feel that does not make as much sense as recognizing that there are a lot of people who have PSA levels below 4 that do have cancer that can threaten to kill them. And the earlier you recognize that, the better it is for those patients," Catalona said.

Thompson, the lead author of the new study, said that until more research is done, men must decide for themselves, based on their individual circumstances, how to interpret their PSA results. For example, those at a higher risk of dying from prostate cancer, such as a man with a family history of the disease or African American men, may want to be more aggressive, he said.

Several experts said that evidence has been increasing that the best gauge of whether a PSA test is worrisome is if the level is rising rapidly over time.