Aspirin Use May Lower PSA Levels

February 16, 2009
Kurt Ullman

Research from Vanderbilt University shows that NSAID use is significantly associated with lower PSA levels. This raises questions about whether this class of medications might delay detection of prostate cancers.

Nonsteroidal anti-inflammatory drugs (NSAIDs) are commonly used, especially aspirin, which has cardioprotective effects in older adults. Research from Vanderbilt University shows that NSAID use is significantly associated with lower levels of prostate-specific antigen (PSA). This raises questions about whether this class of medications might delay detection of prostate cancers.

“There have been a number of studies suggesting from their analyses that NSAID use can reduce prostate cancer,” said Jay H. Fowke, PhD, assistant professor of medicine at Vanderbilt University in Nashville. “We were concerned, though, that the use of these medications was not effecting cancer per se, but rather our ability to detect cancers using PSA levels. The latter could show up like a protective association in those prior studies.”

In conjunction with The Nashville Men’s Health Study, researchers enrolled 1277 men aged >40 years who were scheduled for diagnostic prostate biopsies. Surveys and clinical interviews were used to assess NSAID use. Investigator also reviewed medical charts to determine current PSA levels, prostate volume, and clinical diagnoses following biopsy.

Approximately 46% of participants reported using NSAIDs, with aspirin being the most common (37% of subjects reported using aspirin). After adjusting for age, race, family prostate cancer history, obesity, and treatment for benign prostatic hypertrophy (BPH), cardiovascular disease, hyperlipidemia, and diabetes, aspirin was significantly associated with lower PSA levels. The effect of aspirin on PSA levels was greatest among those with a prostate volume of at least 60 ml, those diagnosed with prostate cancer, and those with concomitant prostate cancer and prostate enlargement. NSAID use was not associated with increased or decreased prostate volume.

Dr. Fowke stressed that there were not enough people using NSAIDs other than aspirin to attribute findings to a class effect of all NSAIDs. Investigators also were unable to assess the routine dose of aspirin used. “Men who were taking aspirin on a regular basis had significantly lower PSA levels,” said Dr. Fowke. “This opened the door to the idea that those prior associations with protection in other studies may be just an artifact of this detection bias.”

“We found an association between aspirin use only in those with latent cancer,” Dr. Fowke added. “This suggests that aspirin is having some benefit on the cancer itself, but it does leave open the door for concerns about the detection issue. The clinician may want to consider retesting PSA levels after discontinuing aspirin in those with marginal or suspicious PSA results.”

“This is not the first time somebody has linked NSAIDs with prostate cancer,” said Al Barqawi, MD, assistant professor in surgery /urology and director of the Research Section of Urologic Oncology at the University of Colorado Denver. “Association does not equal causation, and it is very easy to show something that is not there. Even taking into account the results of this study, there is not enough evidence to show that taking aspirin will prevent or effect prostate cancer.”

Dr. Barqawi, who was not involved in the study, also said concerns that the lower PSA levels seen with aspirin might be delaying diagnosis might be overblown. The PSA itself is not a very accurate measure to begin with, he noted. “Physicians need to look closely at how much aspirin is needed to bring PSA [levels] down to the point where we don’t do a biopsy or miss a cancer,” he said. “This abstract doesn’t give us much guidance on how far the PSA will fall. Obviously, if it is reduced by .1 ng/mL, there is little clinical difference between 7 ng/mL or 6.9 ng/mL.”

Even if aspirin intake is significantly affecting the PSA level, Dr. Barqawi suggested this does not necessarily translate to a significant reduction in actual prostate cancer risk. Further, Dr. Barqawi said oncologists have to keep in mind that a drug’s ultimate preventative benefit should be measured by reduction in mortality.

“To see such an effect we need to conduct a prospective, randomized, double-blinded, placebo-controlled study first,” he cautioned. “Overall, this study is only showing a possible correlation with a marker (PSA) that is known to have significant lack of specificity and sensitivity in detecting prostate cancer. It is like you are trying to move a whole mountain, but you are focusing on a little stone.”

Fowke JH, et al. Association between NSAIDs and PSA, prostate volume, and prostate cancer. The American Association for Cancer Research, Seventh Annual International Conference on Frontiers in Cancer Prevention Research. Washington, DC. November 17, 2008. Abstract A113.