How effective is a PSA test for detecting prostate cancer?

Two separate and large research studies were conducted that investigated the effectiveness of prostate specific antigen (PSA) testing for detecting prostate cancer. The Prostate, Lung, Colorectal, and Ovarian (PLCO) Cancer Screening Trial in the United States enrolled 77,000 men ages 55 to 74, half of whom were randomized to receive yearly screening with both digital rectal examination (DRE) and prostate specific antigen testing. Unfortunately, a very large proportion of men in the PLCO study who were randomized to receive no screening did actually have PSA testing performed, thereby clouding the results. This study revealed no decrease in the death rate due to prostate cancer in the men who were randomized to screening after an average follow-up of seven years.

Quite to the contrary, the European Randomized Study of Screening for Prostate Cancer (ERSPC) revealed a 20% decrease in the rate of deaths due to prostate cancer over an average follow-up of nine years. This trial enrolled 162,000 men ages 55 to 69 who were screened mainly with PSA only once every four years (not annually with DRE as in the U.S. study). Patients in the ERSPC study had biopsies recommended if their PSA was 3 ng/ml or higher, rather than at a threshold of 4 ng/ml as used in the PLCO study. Also, more than 85% of men in the ERSPC study who had an abnormal PSA underwent recommended biopsies, as opposed to less than 50% of men in the PLCO trial. These discrepancies within the trials may at least partially account for the differing results.

A truly specific test is one that only generates an abnormal result when the disease in question is present. Such a test is very useful at ruling disease in, because it will almost never be positive when disease is absent -- and thus if positive, the test reliably rules in disease. Alas, the PSA is not such a test. The result of the test’s low specificity is that many men without prostate cancer wind up having unnecessary biopsies, and sometimes, complications of those biopsies.

The PSA might still be useful if it were highly sensitive. A highly sensitive test is reliably abnormal when disease is present. When a highly sensitive test result is normal, it is a very reliable indicator that disease is truly absent. But again, the PSA falls short. Levels in many cases of prostate cancer are not appreciably different from levels seen in normal men -- and thus the test cannot be counted on to rule disease out either.

A prostate-specific antigen (PSA) test is a blood test that measures the level of a protein called prostate-specific antigen (PSA), which is made by cells in the prostate. Doctors use the test to detect prostate cancer, but it does not provide a definitive diagnosis.

The downside to testing is that an elevated PSA cannot differentiate aggressive tumors from slow-growing, harmless ones that might never cause symptoms during a man's life. Nor can it differentiate cancer from BPH or prostatitis. As a result, it can cause needless worry -- and may lead to costly and invasive procedures, such as biopsies, to determine if cancer is present. Conversely, PSA screening doesn't detect all cancers, so a normal PSA level may offer a false sense of security. Many men with cancer confined to the prostate have normal PSA values. Even advocates of PSA testing doubt its value in men with less than a 10-year life expectancy (ages 75 or older, for men in average health) because these men are more likely to die of something else first.

Studies question the benefit of PSA screening -- does it save lives by enabling doctors to treat aggressive prostate cancers early, or does it harm men who would never die from the disease by subjecting them to the side effects of treatment?

Even though PSA screening increased the diagnosis of prostate cancer, it did not improve the prostate cancer survival rate, and showed no real differences in the numbers of deaths from other causes.

In the United States, the lifetime risk of dying from prostate cancer is 3%. A 27% relative risk reduction would mean that, with screening, the risk of dying from prostate cancer would drop from 3% to 2.19%.

In Europe, that modest benefit came at a steep price. Researchers calculated that 48 men who are not at risk of dying from prostate cancer would have to be treated in order for screening to prevent one death from the disease over nine years. In other words, 48 men would risk the side effects of treatment to save one life.

One conclusion that can be drawn from the studies: men should think carefully about PSA screening and discuss the risks and potential benefits with their doctors before having the test.

Although the prostate-specific antigen (PSA) test is not perfect, it is the best currently available test for early detection of prostate cancer. Since doctors started using the test, the number of prostate cancers found at an early, curable stage has increased. Because most men have normal test results, they can be reassured that they are unlikely to have prostate cancer, especially if the digital rectal exam (DRE) is also negative.

While the PSA test is considered a major advance in diagnosing early-stage prostate cancer, it has some drawbacks. For 100 men over 50 at average risk for prostate cancer, the following would be found if they all had a PSA test:

• Ten of the 100 men would have a PSA level higher than normal (over 4.0). The 10 men would need further testing to clarify their abnormal levels.
• Three of the 10 men would be found to have prostate cancer.
• Seven of the 10 men would be found not to have prostate cancer. They would have an elevated PSA for other reasons -- most likely benign prostatic hyperplasia (BPH).
• Ninety of the 100 men would have PSA levels in the normal range (<4.0).
• One or two of these 90 men would be found to have significant prostate cancer that becomes life threatening.

This shows that the PSA test is moderately sensitive. Of 100 men with prostate cancer, it will detect only about 70 of them, a sensitivity of about 70%. But the positive predictive value of the PSA test is low. Only three out of 10 positive results were cancer (PPV of about 30%). And seven out of 10 positive PSA results (i.e., greater than 4.0) are false-positive results; this means that two out of three men who are told that they may have cancer after taking the PSA test actually do not have it. When the PSA is greater than 10.0, the test is more accurate. There is about a 50-50 chance of having cancer at this level of PSA.

Negative predictive value is high reflecting the low likelihood of having significant prostate cancer with a normal PSA level (NPV of about 98%).

Take the RealAge Test!