Statin drugs and PSA concentration in the National Health and Nutrition Examination Survey (NHANES) 2001 - 2004

B127

Purpose: Observational and laboratory studies suggest that statin drugs (HMG-CoA reductase inhibitors) may protect against advanced prostate cancer. However, little is known about the influence of these common cholesterol-lowering drugs on circulating PSA levels in men free of a prostate cancer diagnosis. Because PSA is widely used as a screening test for prostate cancer, an effect of statins on circulating PSA levels independent of any true influence on prostate cancer could create a detection bias; that is, a different probability of detection of asymptomatic prostate cancer between users and nonusers.
 >Materials and Methods: Serum PSA concentration was measured in 2,574 men aged 40+ years who participated in NHANES 2001-2004 and who did not have a prostate cancer diagnosis or recent prostate biopsy or exam. Participants were asked whether they were taking any prescription medications to lower cholesterol. Although the question did not differentiate between statins and other types of cholesterol-lowering drugs, data on prescribing patterns indicate that by 1998, 90% of cholesterol-lowering drug prescriptions were for statins. All statistical analyses were weighted and conducted using SUDAAN (v. 9.0.1) to account for the NHANES complex survey design. We estimated the geometric mean PSA concentration among statins users and nonusers using linear regression to adjust for age, race/ethnicity, body mass index (BMI), and cigarette smoking. To limit the influence of possible health correlates of statins use on PSA, we conducted sub-analyses stratified by diagnosis with major co-morbidities (cardiovascular disease, congestive heart failure, diabetes, or cancer); statistical interaction was assessed using the Wald test.
 >Results: The weighted prevalence of current statin use in the population was 18.8% (SE = 0.97%). Statins users had a slightly lower PSA concentration compared with nonusers (geometric mean: 0.89 vs 0.95, p = 0.22), although the difference was not statistically significant. However, when the analysis was limited to the 1,680 men without major co-morbidities, the inverse association between statins use and PSA concentration was more pronounced and was statistically significant (0.86 vs 0.99 p = 0.02, p for interaction = 0.003). In contrast, among men with major co-morbidities, statins users seemed to have a higher PSA concentration than nonusers, although the difference was not statistically significant (0.91 vs. 0.83 p = 0.14).
 >Conclusions: These cross-sectional data suggest that men who use statin drugs may have a lower PSA concentration compared with men who do not take statin drugs, especially when the potential influence of co-morbid conditions is removed. These results are consistent with the two other preliminary epidemiologic studies on this topic. If statins lower a man’s PSA, then the probability of detecting asymptomatic prostate cancer would be lower among statin users, and if a tumor were present then, in the future, users would be more likely to be diagnosed at a higher stage than nonusers. Thus, detection bias due to this observed association between statin use and serum PSA concentration cannot explain the inverse association between statin use and advanced prostate cancer that has been observed in several cohort studies. However, this association may have important clinical implications for prostate cancer screening.