PSA May Be Poor Predictor of Lethal Prostate Cancer
But a U.S. expert challenges new Swedish finding
WEDNESDAY, April 4, 2007 (HealthDay News) -- The standard prostate-specific antigen (PSA) blood test is not helpful in predicting lethal prostate cancers in men who are not treated but are placed under "watchful waiting" by their doctors, Swedish researchers say.
"Unfortunately, it turned out to be an inaccurate tool," said Dr. Katja Fall, a researcher at the Karolinska Institute in Stockholm, Sweden, and lead author of a report in the April 4 issue of the Journal of the National Cancer Institute.
But one U.S. expert, Dr. Anthony D'Amico of Harvard Medical School, Boston, said it was too soon to write off the usefulness of the PSA test for these patients, noting that the study has had far too short a follow-up to draw any solid conclusions.
Watchful waiting is a common strategy in prostate cancer, which often progresses slowly -- especially so in many older men. Because the cancer is often diagnosed in older men, one common saying in the field is that "more men die with their prostate cancer than of it." Still, watchful waiting requires periodic assessment of the state of the tumor by, among other things, measurements of PSA. A rapid rise in PSA level is regarded as a major warning sign.
The study included 267 men diagnosed with early prostate cancer. The men received no treatment but were closely watched for signs of progression for two years.
Initial PSA values and the rate of change were associated with later development of cancer, the researchers reported. However, PSA readings were found to "perform poorly in distinguishing between those who develop a lethal prostate cancer from those at low or no risk of disease progression," they wrote.
That finding drew a quick rebuttal from D'Amico, professor of radiation oncology at the Dana-Farber Cancer Institute and Harvard Medical School, in Boston. His team published a report two years ago that found PSA levels were an accurate predictor of serious trouble ahead.
The Swedish study has a basic flaw, D'Amico said: It was reported too early, when only a small fraction of the participants would have developed lethal prostate cancer or died of it.
Just 34 of the men in the study had died of prostate cancer when the report was written, D'Amico noted. "They took 13 percent of the participants to predict what would happen to the other 87 percent," D'Amico said. "Their conclusion has to be tempered by the fact that it is based on a simulation of what is likely to happen in the vast majority of events."
But Fall said the study results "have us hoping that we can find some sort of [non-PSA] decision tool to help us define treatment choice." Researchers are looking for such tools among molecular markers that are present in tumors, she said.
PSA markers are among the tools used at Dana-Farber in monitoring men under watchful waiting, D'Amico said. "We have recommended that if someone's PSA rises by more than two points a year, they go into active therapy," he said.
And the Swedish researchers may yet change their conclusion as data accumulates, D'Amico said. He said the results were similar to the early findings of his own study at Harvard, which ended up following more than 1,000 men for more than seven years.
"They have to wait until more follow-up occurs to conclude that PSA is not an accurate indicator," D'Amico said.
Full information on prostate cancer is available from the U.S. National Cancer Institute.