It’s a controversy that has been roiling among research physicians for two decades: Does the prostate-specific antigen, or PSA, blood test for prostate cancer cause more harm than good? Does the early detection of prostate cancer—that is, before symptoms appear—actually save lives? These questions still haven’t been answered, though many physicians automatically test their middle-aged and elderly male patients as if they have.

A new study conducted in Europe was initiated to see whether PSA screening symptom-free men every two years has any advantage over screening at four-year intervals. The short answer is no, but this study, published in the September 5 issue of the Journal of the National Cancer Institute, suggests that the PSA test is more likely to find non-life threatening cancers without reducing the number of advanced and potentially lethal cancers.

Monique J. Roobol, PhD, and colleagues at medical centers in Holland and Sweden based their findings on data from a large trial called the European Randomized Study of Screening for Prostate Cancer. Initiated in 1993, this trial will answer the question of whether early detection of prostate cancer reduces mortality from this disease.

The reason that this trial and its American equivalent, the Prostate, Lung, Colorectal and Ovarian Cancer trial ongoing since 1992, will take so long to provide answers is the fact that most prostate cancers found on the basis of PSA screening are either non-progressive or so slow-growing that they will never become life threatening or symptomatic. A minority of the prostate cancers diagnosed during the course of this trial will be the deadly aggressive version, and it is unclear whether early detection and prompt treatment of these fast-growing cancers will result in prolonged lives. About 17% of U.S. men contract prostate cancer and 3% of them will die of the disease.

Context Needed

Any discussion of the new study about screening intervals requires some background information on prostate cancer itself. Pathologists have long known that many of the cancers that show up in the prostate will never become lethal had they remained undetected for a lifetime. Autopsy studies of men who died in accidents show that the older they are, the more likely they will have cancer in the prostate.

That’s why the measure of a screening test’s value is not how many cancers it can find; instead it is a reduction in prostate cancer deaths. But that’s not all. Harm must also be taken into account. The detection of prostate cancers that would never progress or become lethal clearly results in a high rate of biopsies and unnecessary treatment. The latter—irradiation or surgical removal of the prostate—is associated with an increased risk of urinary incontinence, bowel dysfunction and impotence.

The diagnosis of new cases of prostate cancer increased substantially in all countries after PSA screening was introduced. In the U.S., for example, PSA screening began in the late 1980s. But contrary to expectations, the prostate cancer death rate has decreased only modestly. And it is not known whether the drop in prostate cancer mortality is due to early detection or improvements in treatment, or both. Such uncertainties can be resolved by a long-term clinical trial that randomly assigns men to be screened (with a PSA test and a digital rectal examination) or not. This describes the European and the American ongoing trials.

“Interval Cancers” Studied

The new study by Roobol and colleagues takes on yet-another unanswered question about prostate cancer screening: What is the appropriate interval for screening? In the U.S. trial, men are usually screened annually.

Not so in Europe. Some medical centers participating in the European trial screen participants every four years, and others screen every two years. This difference in intervals provided the opportunity for Roobol and colleagues to see whether one has any advantage over the other. Of particular interest was the number of prostate cancers detected in men between screening tests. These are what researchers call “interval cancers.” And they are worrisome because they are likely to be the potentially fatal, advanced cancers. A reduction in their detection would be a good sign that screening might eventually reduce the number of prostate cancer deaths.

Roobol and colleagues found that, after 10 years, more cancers were diagnosed in the men (12%) screened at the shorter (two-year) intervals than in the men (8%) screened at longer (four-year) intervals. Contrary to expectations, the number of potentially life-threatening prostate cancers was the same in both groups. The downside of too frequent screening is overdiagnosis, i.e., the increased detection of cancers that do not progress, resulting in more unnecessary biopsies and treatments.

Asked by e-mail for the bottom line message from her study, Dr. Roobol wrote, “Screening should not be done at fixed intervals but through a more personal approach. Men with very low PSA levels (<1.0 ng/ml) can, depending on their age, be screened at much longer intervals. Screening every year for me looks [like] too much for the majority of men in the age group 55 or higher.”

E. David Crawford, MD, University of Colorado Health Sciences Center, author of the editorial that accompanied this study, had this observation: “Although many of us believe that early detection is saving lives, definitive evidence is lacking.”

This uncertainty is reflected in the treatment options listed for all stages of prostate cancer on the National Cancer Institute Web site (www.cancer.gov). Whether prostate cancer is diagnosed at stage I (early) or stage IV (advanced)—no treatment is as reasonable a choice as having the prostate surgically removed or irradiated.

That raises the obvious question: Why test symptomless men when—once cancer is found—no treatment is a valid option? We still don’t know whether the man whose cancer was detected after symptoms appeared is any worse off than the symptom-less man whose cancer was detected with a screening PSA test.

More Needle Biopsies, More Prostate Cancers Detected

“Despite the considerable attention given to the prostate-specific antigen (PSA) as a screening test for prostate cancer, it is needle biopsy—and the PSA test result—that actually establishes the diagnosis of prostate cancer.”

With that as background, H. Gilbert Welch and colleagues at Dartmouth and other Medical Centers studied Medicare claims to determine the proportion of men, aged 65 years and older, who have prostate cancer after a needle biopsy (removal of prostatic tissue by inserting a thin needle through the rectum and into several sections of the prostate). Their study was published in the September 19 issue of the Journal of the National Cancer Institute

The researchers found: Over 10,000 needle biopsies were performed in 1993 through 2001 in 8,273 men, age 65 and older. The overall proportion of needle biopsies found to contain cancer was 32%, and it increased with age—35% for men in their late 70s and 41% for men over age 80.

The risk of finding prostate cancer increased with repeat biopsies in those initially found to be cancer-free, with 50% of the men diagnosed with prostate cancer after two biopsies, 62% after three biopsies and 68% after four biopsies.

The odds of finding cancer go up not only with the age of the man undergoing a needle biopsy but also the number of tissue samples taken. Over the years, the needle biopsy has become more and more aggressive. Initially, needle biopsies consisted of jabs for prostate tissue in six different sites. Now many urologists advocate 12 or more and some even advocate “saturation biopsy,” which involves 32 to 38 jabs, say Welch and colleagues, “the more biopsies taken, the more cancer is found.” The incidence of biopsy-induced infection was not recorded in this study.

Maryann Napoli, Center for Medical Consumers ©
November 2007