Study Questions Benefit of PSA Tests

Article date: January 10, 2006

Better Answers Await Results of Randomized Clinical Trials

Summary: Screening for prostate cancer by measuring blood levels of prostate-specific antigen (PSA) doesn't cut down on deaths from the disease, according to a study in the Archives of Internal Medicine. Men should carefully weigh the pros and cons of screening before deciding whether to get tested, say researchers from Yale University and other institutions. That recommendation is in line with American Cancer Society guidelines for prostate cancer screening (the article misstates the ACS position). Still, says ACS screening director Robert Smith, PhD, the new study isn't strong enough to say definitively that prostate cancer screening isn't valuable.

Why it's important: The question of whether to screen for prostate cancer is controversial in the medical community. For some cancers (breast and colon especially), checking people for cancer even when they have no symptoms can reduce deaths by finding tumors at an early stage, when they are easier to treat. But with prostate cancer, the situation isn't so clear-cut. While some prostate cancers are aggressive, others grow so slowly that many men are more likely to die of something else before the cancer becomes a problem. And treating prostate cancer may cause side effects like impotence and incontinence that can lower a man's quality of life.

Screening for prostate cancer may find cancers that wouldn't cause any problems if left untreated. But there's no good way to tell which cancers need treatment and which don't. That means many men who get a prostate cancer diagnosis will be treated, and possibly suffer from side effects of treatment, unnecessarily. For this reason, many researchers are conducting studies to help figure out if the benefits of prostate cancer screening outweigh the harms. Until there's more evidence, ACS and many other medical groups recommend that men make an individual, informed decision about whether to get screened.

What's already known: Prostate cancer can be found by measuring the amount of prostate-specific antigen, or PSA, in a man's blood. If the level is above 4, there's a reasonable chance a man will have prostate cancer. But other conditions that aren't cancer can cause PSA levels to rise, such as benign enlargement of the prostate or infection. Also, men with a PSA level below 4 can still have prostate cancer. A digital rectal examination (DRE), which allows the doctor to feel if the prostate gland is enlarged, is another way to screen for prostate cancer, but it is not completely accurate either. A positive result from either of these screening procedures usually must be followed up by a biopsy.

Some studies done in other countries have suggested that screening for prostate cancer can lower the death rate from the disease. Other studies, like this one, have found no benefit. There are currently 2 large randomized, controlled trials (the most reliable type of study) examining the question. In these studies, some men are randomly assigned to get screened for prostate cancer with a PSA test, while others are randomly assigned to get no screening. The results should give a much better picture of the effect of prostate cancer screening on prostate cancer death rates, but those studies won't be completed for several more years. Until then, other types of studies may provide clues.

How this study was done: The researchers looked at medical records to see if screening made a difference in whether men died of prostate cancer or not. The records came from men age 50 and older treated at 10 different VA hospitals in New England. The researchers randomly chose 501 prostate cancer patients who died within 9 years of their diagnosis (136 of them died of prostate cancer, while the rest died of other causes). Each of these men was matched to a randomly selected "control," a man who was the same age and had been a patient at the same hospital and was still alive. These men may or may not have had prostate cancer. The researchers compared screening with PSA and DRE in both groups.

What was found: The researchers found no differences in screening rates among the prostate cancer patients who had died and the control group. Among the cases, 70 men (14%) had been screened for the disease with a PSA test, and among the controls 65 men (13%) had been screened with a PSA test. Even when the researchers looked only at the men who had died specifically of prostate cancer, screening did not seem to have an impact. Similar numbers of men in both groups reported being screened for prostate cancer.

But the way the study was done could have missed a more favorable result, Smith said. The time frame of the study, from 1991 to 1999, might have been too brief to see a benefit from screening; few men in either group actually got screened for prostate cancer; and the number of prostate cancer deaths was relatively small.

Another concern is the age of the men in the study. More than half were older than 70. Most doctors question whether men this old would benefit from PSA screening. Prostate cancer is typically a slow-growing cancer and men over 70 would likely die of other causes.

The bottom line: Because their study didn't find a benefit to prostate cancer screening, the authors say routine testing should not be recommended for men who do not have symptoms. Instead, doctors should explain to men what screening can and cannot accomplish, and help men make an informed decision about whether to get screened. Smith agrees with that approach.

"Before the end of this decade, it is likely that we will have results from 2 large, prospective, randomized trials of prostate cancer screening, one conducted in the US and the other in Europe," he said. "Until such time, men should be informed about what is known and not known about early prostate cancer detection so that they have an opportunity to make an informed decision about testing."

Citation: "The Effectiveness of Screening for Prostate Cancer: A Nested Case-Control Study." Published in the Jan. 9, 2006, Archives of Internal Medicine (Vol. 166, No. 1: 38-43). First author: John Concato, MD, MPH, Department of Veterans Affairs Connecticut Healthcare System and Yale University School of Medicine.

Reviewed by: Members of the ACS Medical Content Staff

ACS News Center stories are provided as a source of cancer-related news and are not intended to be used as press releases.