MONDAY, April 15 (HealthDay News) -- Only one-third of men over age 65 who receive an abnormal result from their PSA test actually undergo prostate biopsy to look for disease, a new study finds.
Prostate-specific antigen (PSA) screening is a common test that measures the level of a key marker for prostate cancer in the blood. In general, the higher the level of this protein, the more likely it is that a man has prostate cancer, according to the U.S. National Cancer Institute.
The value of the PSA test has recently come into question, however, with several studies suggesting it causes men more harm than good -- spotting too many slow-growing tumors that, especially in older patients, may never lead to serious illness or death.
The new study focused on this issue once again, tracking outcomes for nearly 300,000 men, aged 65 and older, who underwent PSA screening in the U.S. Veterans Affairs health care system in 2003. The men's health was followed for up to five years.
There were more than 25,000 men with clinically abnormal PSA levels. According to the study authors, during the five-year follow-up period, only 33 percent of those men underwent at least one prostate biopsy to check for evidence of cancer. About 63 percent of those who did have a biopsy were diagnosed with prostate cancer, of whom 82 percent were treated for their cancer.
The older the man, the less likely he was to have a prostate biopsy after having an abnormal PSA screening test result. Men with other health problems were also less likely to undergo a prostate biopsy, the investigators reported.
The study was published online April 15 in the journal JAMA Internal Medicine.
Among men with biopsy-detected prostate cancer, the risk of death from causes other than prostate cancer increased with age and with the presence of other health problems, Dr. Louise Walter, of San Francisco Veterans Affairs Medical Center, and colleagues pointed out in a journal news release.
Two experts not connected to the study said the findings weren't surprising, given the patients' ages.
"PSA screening has been controversial as it has a relatively low yield for finding clinically significant cancer as well as potentialcomplications and expense related to diagnosis and treatment," said Dr. Louis Kavoussi, chairman of urology at North Shore-LIJ's Arthur Institute for Urology in New Hyde Park, N.Y.
In the new study, "as age and other chronic illnesses of aging increased, the less likely biopsy was performed," he said. "This makes sense as the authors report that older individuals and those with [other illnesses] are more likely to die of a non-prostate cancer-related cause."
Therefore, the decision to test for PSA levels in older men must take into account their relatively low risk of dying of prostate cancer, Kavoussi said. "Overall, it is known that about 10 percent of individuals diagnosed with prostate cancer succumb to the disease," he said. "In this older patient population study it was 2.2 percent -- much lower, but not zero."
Another expert agreed, saying that younger men may benefit most from regular PSA screening.
"For screening to be effective, we need to focus on men with a long life expectancy," said Dr. Stacy Loeb, assistant professor in the department of population health at NYU Langone Medical Center, New York City. "Screening allows us to diagnose the life-threatening cancers in time for cure [but] diagnosis does not mandate treatment," she explained.
"Once a diagnosis is made, many patients with low risk disease can be safely monitored conservatively," Loeb said. "Men should be actively involved in all of these choices, with a discussion about risks and benefits."
What's really needed, according to Kavoussi, is a screen that can tell a patient whether his prostate cancer is aggressive or not.
There's a "need for better ways of detecting clinically significant disease in this older population, both to avoid overtreatment and to minimize the risk of missing significant disease," Kavoussi said.
The U.S. National Cancer Institute has more about prostate cancer screening.