PSA Interpretation and Testing Can Be Made More Effective

NEW YORK--Although it has revolutionized the early detection ofprostate cancer, PSA testing can, and must, be improved to becomethe powerful tool needed in the fight against this disease, H.Ballentine Carter, MD, said at a conference on prostate cancerat Lenox Hill Hospital.

"PSA testing is virtually synonymous with diagnosis today.It is the most sensitive indicator we have, though not the mostspecific," said Dr. Carter, associate professor of urology,Johns Hopkins Medical School. The main issue is its inabilityto unerringly distinguish between prostate cancer and benign prostatichypertrophy (BPH), he said.

Cancer or BPH?

Currently, the most effective way to optimize interpretation ofPSA levels is to look at the rate of PSA change, Dr. Carter said.He pointed out that although PSA levels vary with both prostatecancer and BPH, the deviation between consecutive PSA measurementsrelative to elapsed time is distinctively different between thetwo conditions.

Less than 5% of men without prostate cancer have increases inPSA greater than 0.75 ng/mL per year; therefore, an increase greaterthan this should raise suspicion of prostate cancer, Dr. Cartersaid. Accurate interpretation requires at least three measurementsat an interval of no less than 6 months and, ideally, 9 monthsto a year.

"That means it would take almost 2 years to accumulate agood PSA history," he said, adding that "if more menstart having PSA at age 50, it will be reasonable and feasibleto do this."

He also promoted age-specific PSA standards (rather than levelalone) as a means of decreasing false positives in older men andincreasing detection in younger men.

Dr. Carter predicted that in the near future it will be possibleto distinguish prostate cancer from BPH by molecular form, sincePSA exists in several molecular forms, and they are distinctivelydifferent in the two conditions.

Reasons for False Positives

Dr. Carter warned that PSA level interpretation must take intoaccount factors other than malignancy that may affect levels.He cited prostate manipulation, trauma, and various therapies.

Dr. Carter asserted that although performance of a digital rectalexamination (DRE) rarely causes false-positive PSA elevation,it does elevate it somewhat, so it is prudent to draw blood fortesting before performing DRE.

Surgery, radiation therapy, and androgen deprivation (hormone)therapy can all affect PSA and must be taken into account whenpost-treatment levels are evaluated, he said.

Dr. Carter described PSA as a valuable tool for evaluation ofdisease extent and follow-up, since increasing PSA levels indicatedisease progression. He pointed out that, after radical prostatectomy,a detectable level of PSA indicates residual disease. By 6 monthsafter radiotherapy, PSA should be normal; if it is not, then residualdisease is likely.

PSA measurement after hormone therapy is a marker for the androgensensitivity of the tumor, he said. Recurrent disease is virtuallyalways detectable with PSA and often earlier than by clinicalmanifestations.

The timing of a detectable PSA after radical prostatectomy mayhelp distinguish between localized and metastatic recurrence,he noted. A detectable PSA more than 1 year after prostatectomysuggests local recurrence; if it occurs earlier, it suggests metastases.