'Don’t abandon PSA testing, other prognostic indicators'

ORLANDO-Whom to treat vs whom not to treat remains a major dilemma in prostate cancer care, but distinguishing men who will benefit from treatment from those who will not is not a clear-cut prospect, according to a speaker at the 2009 ASCO Genitourinary Cancers Symposium.

“There are men who die despite curative treatment, and we need to be able to identify them earlier on so we can add additional therapy. There is a small proportion of men who are cured by our current therapy, and there is a huge proportion of men whose natural history is such that they could have avoided therapy completely,” said Eric A. Klein, MD, chair of the Glickman Urological and Kidney Institute in Cleveland.

“The key question we face in prognostication and risk stratification for prostate cancer is to be able to distinguish these three cohorts,” added Dr. Klein, who is also a professor of surgery at The Cleveland Clinic.

Data from the CapSure database show that 90% to 95% of newly diagnosed men with early-stage low-grade prostate cancer in the U.S. choose to be treated rather than observed. “Yet, with the rare exception of the very-high-grade tumors, we are still unable to distinguish a biologically significant cancer from a biologically insignificant cancer,” he said.

Despite CapSure data, more men are opting for active surveillance, making it crucial to be able to follow patients carefully and then institute treatment if it becomes indicated. Many groups suggest PSA parameters can be useful, but Dr. Klein disagreed.

“You cannot use PSA to predict progression and help you decide when to pull the trigger when the tumor is curable. The idea that you can follow the patient carefully with digital rectal exam, PSA, and repeat biopsy is a hypothesis that has not yet been proven,” he said (J Urol 178:2359-2365, 2007).

Dr. Klein said that one tool physicians will find very useful is the nomogram to predict curable disease preprostatectomy. One example of a nomogram was developed at Memorial Sloan-Kettering Cancer Center in New York (see Table, page 38). These prediction tools are free for downloading.

The nomograms also have links to relevant research papers, which is a great feature for clinicians and “sophisticated” patients, Dr. Klein added.

Right now, the best prognostic indicators available to clinicians are tumor grade, stage, PSA, and the pathological findings at the time of prostatectomy for those men who undergo surgery. “These are great predictors of cure, and we should not abandon them. PSA has its limitations as a screening test, and most men with elevated PSA levels do not have prostate cancer, but this does not mean that we should abandon this test,” he said.

Dr. Klein cited a study showing that the likelihood of lifetime risk of prostate cancer can be predicted when a man is in his mid-40s. The study raises the possibility that it might be more useful to get a baseline PSA at age 40 or 45, he said (Nature Cancer Reviews 8:266, 2008).

“There’s a tremendous amount of information that is contained just in that single PSA determination, because at that age, most men are not affected by benign prostatic hyperplasia,” Dr. Klein said. “This is a real window into the biology of what’s going on in the prostate. Perhaps we ought to change our screening paradigm and not wait until men are 50 or 55 to start screening them.”