Abstract

To quantify the effect of statins’ use on Prostate Specific Antigen (PSA) levels in patients referred to prostate biopsy and to determinate if the exposure to statins must be considered to improve the prostate cancer diagnostic accuracy of PSA. We selected 551 subjects with PSA <10.0 ng/mL, referred to ultrasound guided trans-rectal prostate biopsy and classified as cancer or non-cancer patients after biopsy. Information regarding statins’ use was obtained from clinical records. We used path analysis to quantify the direct (reflects the influence on PSA biology and metabolism) indirect (reflects the influence on PSA through the effect on the risk of prostate cancer) and total effects (net result of direct and indirect effects) of statins’ use on PSA. We used Receiver Operating Characteristic curves to assess the global predictive accuracy of models including PSA, age, body mass index, 5-α-reductase inhibitors, aspirin and statins’ use for distinguishing between prostate cancer and benign conditions. We observed a negative total effect of statins on PSA levels (users vs. non-users: −0.633 ng/mL, 95% CI: −1.087, −0.179), which corresponds to approximately 8.9% lower levels among statins’ users, mostly due to the direct effect (−0.588 ng/mL, 95% CI: −1.034, −0.141) rather than that by the indirect effect (−0.045 ng/mL, 95% CI: −0.152, 0.061). There were no statistically significant differences between the area under the curve corresponding to the models with or without statins (P = 0.274). In patients referred to prostate biopsy, statins’ use contributed to lower Prostate Specific Antigen levels, but the clinical impact in these patients is low.

Abstract

The impact of using ASA (acetylsalicylic acid), metformin and statins on the prostate cancer may be significant, but is not clear and findings from previous studies are inconsistent. This study aims to evaluate the relationship between the use of ASA, metformin and statins and the pathological characteristics and risk of biochemical recurrence (BCR) of prostate cancer. A total of 311 patients submitted to radical prostatectomy (RP) at Hospital de Braga between January 2010 and June 2014 were analyzed. Data were obtained from clinical records and the crude and adjusted association between the use of a specific drug and prostate specific antigen (PSA), pathological stage, Gleason score, positive surgical margin and risk of biochemical recurrence were calculated. Overall, 26 (8.4%) of the patients used ASA, 35 (11.3%) metformin and 112 (36%) statins. The ASA users had a lower PSA compared with the non users (5.9 vs. 8.9 ng/m, p = 0.008). Its use was an independent predictor of positive surgical margin (OR = 3.77, IC 95%: 1.45‐9.78). The use of metformin was associated with advanced pathological stages, more precisely pT3b (20% vs. 7.7%, p = 0.048). No other differences were detected between ASA, metformin and statins users. The use of ASA, metformin and statins has no beneficial effect on prostatic cancer patients. On the opposite, an association between the use of metformin and advanced pathological stages was observed. These results should be validated in other larger samples and longer follow‐up. The association between the use of ASA and lower PSA was already detected in other studies, whose mechanism should be clarified in future