In 1992, Bill Roher read an article about PSA, a relatively new screening test for prostate cancer. Roher was 65 at the time, which meant that he was high risk for developing the disease. So even though he hadn’t had any symptoms, he asked his doctor if he should take the prostate-specific antigen (PSA) test—a simple blood test that only costs about $30.
His doctor said no. “I said, well, I’m paying for the blood tests, let’s run it,” Roher says. By coincidence, his PSA level was high, and after a biopsy, Roher was diagnosed with prostate cancer. He has since founded Prostate Action to create awareness about the disease and the PSA test.
PSA is an enzyme secreted by the prostate. When the prostate is enlarged, infected, or hosting a tumor, PSA levels generally increase. Benign tumors also secrete PSA; in fact, less than 30 percent of patients with high PSAs actually have prostate cancer. And 30 percent of patients with cancer have normal PSAs. That's one reason why PSA is still controversial.
Before the 1990s, doctors mostly used PSA to measure the progress of patients already diagnosed with cancer. But in 1991, a paper written by Dr. William Catalona, a professor of urology at Washington University School of Medicine, reported that PSA could also be used to screen for cancer in at-risk men—which means, in effect, all older men. PSA cannot detect cancer. But, when used in conjunction with a digital rectal exam, it can help determine a man’s risk of developing cancer. High PSA levels may direct a doctor to perform a biopsy, which ultimately detects the tumor.
The American Cancer Society now recommends that men who are African American or have a family history of prostate cancer begin yearly PSA screening after age 40; all others should begin testing after 50.
Nevertheless, there is still a raging debate over its effectiveness. Dr. Catalona notes that prostate cancer mortality rates rose steadily for 30 years beginning in the 1960s, and only leveled off after PSA came into widespread use—but there is no evidence that increased testing is the reason. Many men have small tumors that will never become life-threatening, and removing them carries risk of impotence, incontinence and bowel problems. Unfortunately, there is no way to know whether a tumor will grow and spread, becoming dangerous, or stay small.
Some doctors do not recommend a biopsy until PSA is more than 9ng/mL, while others biopsy with levels as low as 2.5. Some recommend testing every four years for men who are not high risk, but others say that such conservatism could cost lives. And all sides of the debate have the evidence to back up their opinions.
Dr. Catalona is one of the staunchest defenders of frequent testing. In a 12-year long PSA study he conducted, he began by performing biopsies on men with levels over 4ng/mL. He found, however, that half of the men with PSAs above 2.5 eventually converted to PSAs over 4. And while PSA was rising, the tumor was growing, and in many cases it spread beyond the prostate. Now he biopsies anyone with a PSA above 2.5 ng/mL.
"Although only a small percentage of men are going to have their PSAs go up, those are the men whose cancers you want to detect, because they have the aggressive cancers," Dr. Catalona says. Failing to biopsy men with lower PSAs, he argues, "would be delaying the diagnosis of prostate cancer in a significant number of men. It would be a relatively small percentage of the population, but for those men it would be 100 percent."
Dr. Thomas Stamey of Stanford was one of the first to link PSA with prostate cancer, but now he believes that the test is overused. He recently published a study that said that PSA levels between 2 and 9 are “clinically useless” for determining tumor size. Now, he says, he regrets his previous advocacy of the test: “I’ve removed a few hundred prostates that I wish I hadn’t.”
But Dr. Catalona says that if you wait to biopsy until PSA is above 7, 30 percent of men will have cancer, and that’s unnecessary.
Dr. Stephen Strum, Medical Director of the Prostate Cancer Research Institute, agrees.
"PSA is an outstanding tool test to diagnose a disease early," he said. "It’s like picking up breast cancer early—no one ever complained about the mammogram. Nowhere in medicine do we do something that results in later diagnosis of the cancer and think that we’ve done good for the patient. It makes no sense."
Dr. Strum said that mortality rates would drop further if PSA were used correctly. PSA screens need to be taken often and tracked over time, he said. He added that the test should be taken not in isolation, but in conjunction with other tests, like the digital rectal exam.