Ian M. Thompson Jr., M.D., professor and chair of urology at the U T Health Science Center, is looking for a better test - or set of tests - to diagnose prostate cancer and to distinguish between cancers that will kill men if left untreated and those that will not threaten health and do not need treatment.
The widely used PSA test (prostate-specific antigen) is not getting the job done by itself, and more than ever the onus is on men to discuss risks and options with their physicians.
That is strongly confirmed in a new study reported in the July 6 issue of The Journal of the American Medical Association. Dr. Thompson is lead author among collaborators from the Fred Hutchinson Cancer Research Center, the University of Colorado, the National Cancer Institute, Cancer Research and Biostatistics (Seattle) and the Southwest Oncology Group.
Dr. Thompson spoke at a press briefing July 5 at San Antonio’s Cancer Therapy & Research Center (CTRC), where he is a member of the medical staff. Others in attendance included study co-author Charles A. Coltman Jr., M.D., of the Southwest Oncology Group, the San Antonio-based national clinical trials group, and Karen Fields, M.D., president and chief executive officer of the CTRC. Co-author Howard L. Parnes, M.D., of the National Cancer Institute, spoke by teleconference.
The study, which draws conclusions from the Prostate Cancer Prevention Trial (PCPT) conducted under the sponsorship of the National Cancer Institute, examines the performance characteristics of the PSA test. Dr. Thompson and his colleagues analyzed PCPT data taken from regular PSA screenings in participants and the outcomes of end-of-study prostate biopsies of more than 8,500 men who received a placebo rather than active medication.
“We once thought a PSA level of 4.0 (ng/ml) or lower was a magic figure for men, ensuring they were not at risk for prostate cancer,” Dr. Thompson said. “Now we are thinking ‘outside the 4.0 box.’ There is no level of PSA that gives us high enough sensitivity, meaning the ability to identify the man with cancer, while at the same time providing high enough specificity, meaning the ability to tell a man without cancer that he is fine. The bottom line is this: Each man needs to talk with his physician about his own risk factors - age, family history of prostate cancer, ethnicity and even weight - and decide where to go from there. PSA readings provide significant information but cannot be interpreted alone.”
The authors wrote in their paper that lowering the PSA cutoff value - the value that if exceeded would lead to further testing - to 1.1 ng/ml would result in detection of 83.4 percent of prostate cancers “but would subject 61.1 percent of men without cancer to prostate biopsy.” They also wrote that lowering the threshold, while it would result in finding more aggressive cancers earlier, would lead to the “unavoidable tradeoff [of] increased detection of biologically inconsequential cancers.”