January 10, 2006

By NICHOLAS BAKALAR
Men who have been screened for prostate cancer by the most commonly
used tests have no greater chance of surviving the disease than those
who have not been screened at all, new research has found.

A report on the research, published yesterday in The Archives of
Internal Medicine, found that neither a prostate specific antigen test,
known as a P.S.A., nor a rectal examination worked to reduce deaths
from prostate cancer.

The value of the screening has long been a matter of debate. A P.S.A.
level can be abnormal even when a man does not have prostate cancer.
But when an abnormal P.S.A. level is discovered, typically the next
step is a biopsy.

Even a biopsy is inconclusive, however. The tissue samples from a
negative screening, for example, may by chance have come from parts of
the prostate that are free of cancerous cells.

Even if a cancer is found, an operation, which often causes erectile
dysfunction and incontinence as side effects, may not be necessary
since many cancers are so slow growing that they will never cause a
problem.

Dr. John Concato, a researcher at the Veterans Affairs Connecticut
Healthcare System and the lead author on the paper, stressed that a
physician was obligated to clarify all the issues for patients.

"He should explain the benefits and risks, in the context of each
patient's values," Dr. Concato said. "For example, some patients place
such a high premium on avoiding incontinence and impotence that a
positive P.S.A. test can be problematic."

The research involved nearly 72,000 men over 50 who received outpatient
care at any of 10 Veterans Affairs hospitals in New England.

Of this group, the scientists identified 1,425 men with prostate cancer
that was diagnosed from 1991 to 1995 and then studied the records of
501 patients who had died of the disease as of 1999.

For each case, the researchers randomly selected a living patient to be
part of a control group.

Screening with P.S.A. had been performed for 70 of the men who died and
for 65 men in the control group used for comparison.

If screening had been effective, a lower proportion of screened
patients would have been found among the group of men who had died. But
this was not the case.

An editorial accompanying the report states that 78 percent of male
primary care physicians and 95 percent of urologists over 50 have
themselves had at least one P.S.A. screening, so they apparently have
decided that the test is useful.

"Many urologists and other physicians have received the P.S.A. test,
perhaps because they don't consider the issue of screening to be
uncertain," said Dr. Concato. "They believe the test works, but our
results don't support that position."

Dr. Michael J. Barry, the author of the editorial and an associate
professor of medicine at Harvard, said that a doctor's personal
decision to have the test or to decline it need not affect his ability
to inform a patient properly.

"Whatever the doctor's beliefs, he should still be capable of giving a
patient enough objective information to arrive at an informed
conclusion for himself," Dr. Barry said in an e-mail message.

"I think it's quite feasible to present the pros and cons of a medical
intervention like a P.S.A. test and have the patient reach a different
decision than the doctor would," he continued.

Dr. Barry said that at age 52, he had not had a P.S.A. test himself,
but he added that he routinely presented the option to his patients.
"After a discussion," he added, "many decide to go ahead and be
tested."

The authors said their study was carefully controlled to eliminate bias
and included a large population, increasing the validity of the
results.

At the same time, they acknowledged that there are good studies with
different findings, and that more research would be needed to settle
the question of whether prostate cancer screening does more harm than
good, or the reverse