Seeking men without prostates whose PSA will double nonetheless

April 13, 2010

Most men who have had radical prostatectomy can look forward to at least 15 years free from prostate cancer. But which patients are likely to see their PSA levels double within nine months, and therefore have high risk of recurrence? Oncologists are calling for new ways to predic, and researchers are on the case.

After years of debate about what PSA means to prostate cancer, for post-surgical patients, at least, it has a decisive significance. Most men who have had radical prostatectomy can look forward to at least 15 years free from prostate cancer if their postoperative PSA values rise only moderately or take longer than nine months to double. Last month, oncologists from Johns Hopkins announced at the ASCO Genitourinary Cancers Symposium that they have confirmed the predictive value of PSA doubling time, using multiethnic data from a large armed-services database to supplement less ethnically diverse records from Hopkins' own population.

The overwhelming majority of men will survive many years after radical prostatectomy. Authors of a retrospective multicenter analysis of more than 13,000 prostate cancer patients reported in the Journal of Clinical Oncology last September that only 4% of them could be expected not to survive for 15 years after surgery. Very few such patients, therefore, need to worry about recurrence. But which patients are likely to see their PSA levels double within the nine months?

The ability of clinical factors to identify these unlucky few is "limited," wrote Peter Scardino, MD, FACS of Memorial Sloan-Kettering Cancer Center et al in JCO. They cited a "need for novel markers specifically associated with the biology of lethal prostate cancer."

Others are on the case already, of course. A powerful asset to identify the men still at risk is the SEARCH (Shared Equal Access Regional Cancer Hospital) database. Two teams of Duke University Medical Center oncologists have scoured the SEARCH database for clinical clues to post-surgical progression, with some success. One reported last month in The Journal of Urology that among patients who had small prostates (volume less than 35 cc.), tumor involvement greater than 20% was a strong predictive factor for post-operative PSA recurrence. The other team found that PSA recurrence correlated with estimated blood loss during surgery of between 1,500 mL and 3,500 mL-but not, oddly, with blood loss above 3,500 mL.

Is moderately elevated blood loss a marker of aggressive cancer, of poor surgical technique, or of something else entirely? "We have no new data to support one over the other," says lead investigator Stephen Freedland, MD, of the Duke Prostate Center. "My opinion is that high EBL may be a marker of poor technique, but the margin data did not support that."

Others are testing PSA assays with sensitivity levels beyond "ultra." Chemical engineers at Northwestern University's Feinberg School of Medicine have devised an assay that detects PSA at femtogram/mL levels and reported results of a pilot test on serum from 18 prostate cancer patients. The test detects PSA by trapping it between minute particles studded with anti-PSA antibodies and adorned with DNA molecules of a known sequence, which are measured in the elution. They call it the "nanoparticle-based bio-barcode assay," likening the custom DNA sequence to a barcode identifier.

Chad Mirkin, a professor of medicine at Northwestern, says the barcode test has "redefined zero" in PSA measurement, and that it can identify (and reassure) the 42% of men whose PSA levels are not going to rise after surgery. The Chicago researchers are currently testing the same assay on post-operative serum samples from a larger retrospective sample from over 400 prostate cancer patients after radical prostatectomy, and say that the preliminary results reinforce their previously published data. They are planning a prospective study.

None of these approaches, however promising, features the "novel markers obviously associated with the biology of lethal prostate cancer" that Scardino and his co-authors want to see. But molecular biologists have also been working the problem.

The well-known phenomenon of telomere loss (the "wearing away" of the ends of chromosomes) in cancer appears to correlate with PSA recurrence, according to tests of prostate biopsy samples at the University of New Mexico School of Medicine in Albuquerque. Eric Treat and coauthors from the Department of Urology revealed last month in the Journal of Oncologythat a finding of telomere DNA content less than 30% of standard levels reflected an increased risk of PSA recurrence (defined as two consecutive readings above 0.2 ng/mL, not as PSA doubling).

Oncologist Timothy Johnson, MD, and coworkers at Baylor College of Medicine and MD Anderson, collaborating with scientists in Japan, predict that the secretable protein cav-1 may be the sought-after biomarker. A major structural component of caveloae, cell-surface invaginations that are crucial to processes such as cell adhesion and molecular transport, cav-1 is secreted by human prostate cell lines and promotes the survival of cancer cells in in vitro tests. In a previous study involving a large population of men with prostate cancer, pretreatment levels correlated with faster times to recurrence.

The researchers found recently, to their surprise, that cav-1 also stimulates angiogenesis, leading them to believe that the protein is capable of significant and varied effects on the tumor microenvironment. Do these effects pertain even when the prostate has been removed? Do presurgical cav-1 levels foreshadow recurrence?

The Baylor team reported in 2006 that preoperative cav-1 levels correlate with risk of recurrence among men with PSA levels of at least 10 ng/mL. "We are in the process of completing a large prospective study that will hopefully have the numbers to make very definitive statements in this context," Thompson says.