PSA Screening Fails Says Creator

"My one wish with prostate cancer," says Dr. Thomas Stamey, a veteran researcher at Stanford University, "is that before a doctor does anything aggressive, he would tell his patient that all men will develop the disease eventually." He pauses to let the bad news sink in. "The good news," he wants doctors to add, "is that the rate of dying from prostate cancer is infinitesimal."

Stamey has been in a reflective mood of late because of the growing realization, by him and others, that the screening test he helped discover is far less useful for detecting prostate cancer than many had once believed. In 1987, a team led by Stamey found that high levels of prostate-specific antigen (PSA) circulating in the blood were a strong indication of prostate tumors.


Before then, the only way to detect prostate cancer was with a painful biopsy, and this is still used to confirm the disease. But with a simple PSA test , doctors thought they could weed out men who don't need the more invasive procedure, as well as spot tumors at an earlier, more treatable stage.


The test has proven a powerful draw: about half of all men over the age of 50 get annual PSA tests. Now, Stamey hopes that men will be open to the older means of screening for prostate cancer.


"I don't think PSA adds very much," he says.


The End of the PSA Era?

These second thoughts stem from a troubling variety of evidence that suggests widespread testing is possibly causing more harm than good. Although death rates from prostate cancer are lower than they were before PSA screening, these rates have also declined in countries where this type of testing is not commonly used.


If one were to randomly biopsy men, as Dr. Wael Sakr of Wayne State University did on a group who were accidentally killed on the streets of Detroit, about 8 percent of those in their 20s would have prostate cancer, with the rates steadily increasing as men age. Indeed, about 80 percent will develop the disease by the age of 70.


Some of these tumors are clearly dangerous. But most are slow moving, and many prostate cancer patients can go 20 years without any need for treatment, according to a recent study led by Dr. Peter Albertson. With widespread PSA screening picking up these relatively benign tumors, Stamey fears that the tests are leading to unnecessary treatment and worry.


"I've been as guilty as anyone else," he says.


Not everyone is willing to give up on PSA screening quite yet.


"We know it's not a perfect test," says Jamie Bearse, a spokesperson for National Prostate Cancer Coalition, which advocates annual PSA screening for men 40 years and older. Bearse is hopeful that newer screening tests will prove more discerning. In the meantime, he says, the PSA test is the best early detection option men have. "They would rather know, than not know."


Normal vs. Abnormal PSA

Yet it is increasingly hard to determine even a broad risk prediction from PSA testing. Traditionally, doctors used a PSA measurement of 4 as a key cut off point: lower than 4 nanograms per milliliter (ng/mL) in the blood meant that men were considered cancer free, whereas higher or equal to 4 suggested the need for a biopsy to confirm the disease.


But looking at nearly 5,500 men who had a PSA test and then at least one biopsy, researchers found this cutoff point often missed tumors or implied that men had cancers that weren't in fact really there, according to recent results published in the Journal of the American Medical Association.


Indeed, a PSA level of 4.1 ng/mL accurately predicts only 20 percent of prostate cancers and leads to false alarms about 6 percent of the time. Lowering the threshold will detect more cancers, but at the price of causing men to undergo biopsies for no reason or finding many benign tumors that only need to be monitored. A PSA cutoff of 2.1, for example, would yield false readings more than 85 percent of the time to catch only slightly more than half of all tumors.


The authors of the study, led by Dr. Ian Thompson of the University of Texas Health Science Center in San Antonio, call for a substantial "reeducation" effort on the increasingly murky role of PSA. "It will be a challenge to the medical community to change the long held notion that there is a 'normal' PSA level," the authors write.


Researchers still hold out hope for measuring PSA velocity, the rise of PSA over time that gives an indication if the tumor is growing. As an initial detection tool, Stamey says that doctors should return to looking at increasing age, along with a family history of the disease, as a way of determining who may be at greater risk for prostate cancer and in need of a biopsy to catch tumors early.


Other than that, he adds, "We have as much a way of predicting who will need a biopsy based on looking at someone's eyes." Ongoing studies are continuing to look at whether PSA testing saves lives, and the American Cancer Society and other test supporters urge men to discuss the pros and cons of screening with their doctors. But the creator of the PSA test is calling for the end.


"All men will develop a prostate cancer," says Stamey. "That's a given."