Doubts Cast on Common Prostate Cancer Test
High PSA readings may not indicate a tumor, study says
FRIDAY, Sept. 10, 2004 (HealthDayNews) -- A review of 20 years' worth of tissue samples casts doubt on the value of prostate-specific antigen (PSA) testing for prostate cancer, a new study contends.
In fact, the resulting treatments that often follow a worrisome PSA test result may do more harm than good, according to the study by Stanford University researchers.
"I have here in my laboratory a 20-year collection of thin sections that allow us to construct the extent of cancer in the prostate," said lead study author Dr. Thomas A. Stamey, a professor of urology at Stanford. "We can look at the relationship between PSA levels and the size of the cancer.
"In the first five-year period, there was a 60 percent relationship between the size of the cancer and the level of PSA. In the last five years, that has fallen to 2 percent, which is why we say the PSA era is over. We are no longer finding significant cancers by PSA screening," Stamey said.
A key word in that sentence is "significant." Over the past decades, research has shown that "all men will get prostate cancer if they live long enough," Stamey said. The incidence is 8 percent for men in their 20s, compared to 70 percent to 80 percent for men in their 70s, he said.
Yet prostate cancer is rarely fatal, Stamey said, with a death toll of only 226 per 100,000 men over the age of 60. In many cases, it progresses so slowly that it poses no major danger.
Yet doing a biopsy to see whether cancer is present has some risks, while treatment of prostate cancer by radiation or surgery can cause major problems, such as impotence, he said.
As a result, "lots of the cancers we diagnose don't need to be diagnosed. Far more men die with prostate cancer than of prostate cancer," said Dr. Howard L. Parnes, chief of the prostate group in the division of cancer prevention at the National Cancer Institute.
The Stanford study appears in the October issue of the Journal of Urology.
Stamey said there was a strong correlation between a high PSA level and prostate cancer 20 years ago because the test was usually done only when a physician suspected cancer. Today, the test is usually done as a matter of routine, he said. Since PSA production is related to the size of the prostate, a high level usually is related to benign prostatic hyperplasia, the harmless increase in prostate size that occurs with aging, he said.
Another important thing to note about PSA screening, Parnes said, is that "there is no magic threshold of PSA below which you can be assured you do not have prostate cancer and none above which a biopsy should automatically be performed."
Right now, he said, "the National Cancer Institute does not have a position on prostate-cancer screening. We are conducting studies to determine if prostate-cancer screening actually saves lives."
One such study has enrolled 75,000 men who are undergoing regular PSA testing, to determine whether testing can reduce deaths from prostate cancer. Results will be available "probably at least five years from now," Parnes said.
Another study, started 10 years ago, is intended to tell whether surgery provides a benefit over watchful waiting for men with high PSA readings, he said. Again, results are not expected for several years.
Meanwhile, Parnes suggests a cautious approach to PSA testing.
Often, the test is given because "the perception of physicians is that patients think they are not good doctors if they don't give the test," Parnes said. "What doctors should do is to find the time to discuss the pros and cons of screening. A PSA test should be done with the informed consent of the patient."
The U.S. Centers for Disease Control and Prevention has more on prostate cancer screening.