Major Finding: NSAIDs, statins, and thiazide diuretics were associated with lower PSA levels of 1%, 3%, and 6%, respectively.
Data Source: 1,846 men aged 40 years or older who completed the National Health and Nutrition Examination Survey for 2003-2006.
Disclosures: Some of the investigators are consultants to Veridex LLC, a manufacturer of diagnostic tests.
SAN FRANCISCO — Commonly used medications were associated with clinically important reductions in prostate-specific antigen levels among roughly 2,000 middle-aged and older men in a cross-sectional study.
After 1 year of regular use, PSA levels were 1% lower in users of nonsteroidal anti-inflammatory drugs (NSAIDs), 3% lower in statin users, and—an apparently novel observation—6% lower in thiazide diuretic users, according to data reported at a symposium on genitourinary cancers. The difference in PSA levels among users and nonusers of the common medications increased over time, with reductions of 6%, 13%, and 26% seen with 5 years of regular use of NSAIDs, statins, and thiazide diuretics, respectively.
“If taking these medications alters serum PSA, it could affect the quality of prostate cancer screening,” said lead investigator Dr. Steven L. Chang of Stanford (Calif.) University. On the other hand, “perhaps these medications may influence prostate growth.”
Using data from the National Health and Nutrition Examination Survey (NHANES) for 2003-2006, the researchers assessed associations between medication use and log-transformed PSA levels in 1,846 men aged 40 years or older who had a serum PSA measurement; did not have a history of prostate cancer, prostatitis, or recent prostate manipulation; and were not taking 5-alpha reductase inhibitors or hormone therapy.
Statins topped the list of the 10 medications most commonly used in the study cohort (taken by 20% of the men), according to study results, which were reported in a poster session.
They were followed by beta-blockers (13%), angiotensin-converting enzyme (ACE) inhibitors (11%), NSAIDs (9%), proton pump inhibitors (9%), calcium channel blockers (6%), selective serotonin reuptake inhibitors (6%), thiazide diuretics (5%), alpha-blockers (4%), and sulfonylureas (4%).
In multivariate analyses, PSA levels after 1 year of regular use were 1% lower in NSAID users (P = .03), 3% lower in statin users (P = .01), and 6% lower in thiazide diuretic users (P = .03), relative to those in the respective nonusers. The remaining medications were not independently associated with PSA levels.
The effects of statins and NSAIDs on PSA have been previously reported, but the finding for thiazide diuretics appears to be new and was somewhat surprising in magnitude, Dr. Chang commented in an interview at the symposium, which was sponsored by the American Society of Clinical Oncology, American Society for Radiation Oncology, and Society of Urologic Oncology.
Certain medication combinations also were associated with reduced PSA levels, including fixed-dose combinations of beta-blockers plus thiazide diuretics (6% reduction, P = .03) and ACE inhibitors plus thiazide diuretics (7%, P = .02), and concurrently used statins and beta-blockers (3%, P = .03), statins and ACE inhibitors (3%, P = .04), and statins and thiazide diuretics (8%, P = .002).
Among the combinations, the reduction was greatest for concurrently used statins and thiazide diuretics, with a 36% difference after 5 years of regular use of both medications.
However, the link between statin use and lower PSA levels was minimized or negated in men who were concurrently taking calcium channel blockers, Dr. Chang noted.
“It's unclear as to the true mechanism behind what we are observing here, but it certainly raises a number of questions that should be addressed,” Dr. Chang said.
One possibility is that these medications simply reduce PSA levels without influencing the development or growth of prostate cancer.
“In that case, patients who develop prostate cancer would be identified later because their PSA levels would be lower than in others (all other things being equal) who are not on these medications,” Dr. Chang observed.
Alternatively, the medications might have some effect on the prostate gland, for example, reducing cancer development or prostate size and thereby lowering PSA levels.
In sum, Dr. Chang concluded, if the observed associations are proved to be causal, “future work is necessary to determine how medication use should be factored into prostate cancer screening. If any of these medications actually affect prostate cells, perhaps they may have a role in prevention or therapy of prostatic diseases.”
“If taking these medications alters serum PSA, it could affect the quality of prostate cancer screening,” said lead investigator Dr. Steven L. Chang.
Source Courtesy Dr. Steven L. Changhttp://circ.ahajournals.org/cgi/reprint/circulationaha.105.555482v1