PSA doubling time: When is it useful as a treatment guide?

PSA doubling time is an accepted tool to determine the probability that prostate cancer will or will not be clinically significant following primary treatment. However, the definition and proper use of PSA doubling time in clinical practice has become controversial. In this interview, W. Scott McDougal, MD, discusses where the controversy lies and explains when and how this measurement can (and cannot) be used to guide treatment decisions. Dr. McDougal is chief of the department of urology at Massachusetts General Hospital and professor of urology at Harvard Medical School, Boston. He was interviewed by former Urology Times Editorial Consultant Robert C. Flanigan, MD, who is professor and chairman of the department of urology, Loyola University, Maywood, IL.

Q. How do you define PSA doubling time?

A. First, it's important to understand that the kinetics of prostate cancer cells are not accurately described by a simple monoexponential equation. Cell kinetics are generally described by a complex set of equations. One cannot use two or three points to define PSA doubling time, as some clinicians have suggested.

A second problem in understanding what the rate of rise in PSA means is that many studies have used surrogates of death, rather than death itself to define outcomes for stratifying doubling time. Surrogates that have been used, for example, include a rising PSA while the patient is receiving ablative hormonal therapy or documentation of metastatic disease, neither of which is particularly accurate in predicting death from disease, as we all know.

Understanding these problems, I believe, allows one to make some sense of the PSA following definitive therapy in that it allows one to utilize some of this data when it is meaningful, and use other parameters when it is not. As mentioned above, for patients at the extremes, one can make a prognosis. For a large portion of patients in the middle group, the physician needs to use other factors brought to bear to advise the patient with respect to prognosis. "PSA doubling time" in this group of patients does not play a primary role.

Q. Are you suggesting that the standard definition most people use, based on two or three data points, is probably not reliable?