Recently, a well-publicized study, which appeared in the New England Journal of Medicine, sent shockwaves through much of the medical community, and raised many questions about screening for prostate cancer. The study showed that in men with very low PSA levels (less than 4.0 ng/ml) who underwent needle biopsies of the prostate,15 percent had prostate cancer — and of these men, 15 percent had worrisome Gleason scores of 7 or higher. “Basically, this study showed that for men with a PSA greater than 1.0 ng/ml, there is no threshold PSA,” explains Patrick C. Walsh, M.D., University Distinguished Service Professor of Urology.“ There is no magic cutoff number to guarantee that a man does not have cancer — and most specifically, a life-threatening cancer.”

So, what’s the best way to screen for prostate cancer? The key is to look beyond cutoff numbers, says H. Ballentine Carter, M.D., professor of urology and oncology, and one of the world’s top experts in the study and understanding of PSA. “Today, it is impossible to base any recommendations for screening on scientific outcomes, because they simply don’t exist.” The trouble, he adds, is that PSA is prostate-specific and not cancer-specific, and so any prostate disease— including infection and benign enlargement, as well as cancer — can cause PSA levels to rise.

PSA is most valuable as a marker for cancer when the slate is relatively clean —“when the other confounding, benign conditions do not exist.” With this in mind, the best time for a man to have his first PSA test is when he’s young and not even terribly

worried about prostate cancer, “about 35 to 40 years of age, when he is unlikely to have benign prostatic enlargement. This will provide a valuable baseline upon which other measurements of PSA can be compared for the rest of his life.”

After this first test, Carter recommendsthat a man have follow-up PSA measurements every two to five years. Just how farapart to space these subsequent tests depends on the baseline level: “If the initial PSA is greater than the median for his age, then PSA levels should probably be checked every two years.” For men in their forties, the median PSA is 0.6 ng/ml, and for men in their fifties, it is 0.7 ng/ml. “At present,”Carter says, “this seems to be the most reasonable approach for all men based on available data, whether they’re high-risk (if prostate cancer runs in their family, or if they are of African descent) or not. From this baseline, the rate of increase in PSA every year can be calculated.” This concept, called PSA velocity, was pioneered by Carter in 1992, in a landmark article in the Journal of the American Medical Association.

“For men with low PSA levels (between 1 and 3), any increase is alarming,” warns Carter. In a study presented this year at the American Urological Association, Carter found that increases as small as 0.2 ng/ml a year were a predictor of death from prostate cancer.

In his 1992 article, Carter reported that for men with PSA levels between 4 and 10, a PSA velocity of 0.75ng/ml per year suggested that cancer was present, and for men with these PSA levels, this remains a critical guideline. “Change in PSA over time is the most valuable tool we have for interpreting the PSA — for predicting both the presence of cancer, and whether or not it is life-threatening.”

What if it’s my first PSA test? I don’t have abaseline. How do I know I don’t have cancer?“ For men in this situation, given the prevalence of life-threatening cancers at very low PSA levels, the guidelines commonly being used are of questionable help,” says Carter. His recommendation: “If you are in your fifties or sixties, have never before had a PSA test, and have a PSA level greater than 3.0per ng/ml — and you are otherwise healthy, and could expect to live 15 or 20 years — a biopsy is indicated. If you are in your forties, a biopsy is a good idea if your PSA exceeds 2 to 2.5ng/ml.”

As long as PSA testing has been used as a screen for prostate cancer, “there have been naysayers who have said it is not valuable,” notes Walsh, “and many of them spoke up when the New England Journal of Medicine article came out. But nothing could be further from the truth. PSA testing is valuable. But you need to know how to test, how often to test, and how to interpret the results. There is no question, however, that we could also use a new marker.” (For an exciting breakthrough in this area, see the article on Robert Getzenberg’s work on EPCA.)