Cancer - Prostate(구연) (E-138) Correlation between serum testosterone and prostate-specific antigen (PSA) in ordinary males 을지병원 , 국립경찰병원¹ 이준호, 박연원¹, 조인창¹, 김병훈¹, 최재덕, 조정만, 강정윤, 유탁근 <strong>Introduction</strong>: It is well known that the administration of exogenous testosterone could elevate serum PSA level and castration induce a reduction in serum PSA. However, it`s still unclear whether endogenous testosterone levels could affect serum PSA levels. This would be an important issue, because it could be related to the false positive rate and lead to unnecessary prostate biopsied. <strong>Methods</strong>: We retrospectively reviewed 9,318 male patients who visited our urology clinic for benign prostatic hyperplasia or regular health check-up. 582 patients were excluded because of PSA level > 4.0ng/mL, current usage of 5a-reductase inhibitors, pyuria, and lack of data. A total of 8736 patients were enrolled for analysis. We evaluated the relationship between serum testosterone and PSA with adjustments of age, prostate size, plasma volume, and metabolic syndrome. <strong>Results</strong>: Mean age, mean testosterone, and mean plasma volume were 50.2±5.6 years, 5.28±1.6 ng/mL, and 3137.7±177.6 mL, respectively. PSA showed a positive relationship with serum testosterone (Pearson correlation coefficient=0.450, P<0.001). After fully adjusting for age, total prostate volume, plasma volume, and metabolic syndrome, the geometric mean PSA linearly and significantly increased as a quartile testosterone increased (adjusted mean (interquartile range) ng/dL: 1st quartile: 0.833(0.810-0.857); 2nd quartile: 0.846(0.823-0.870); 3rd quartile: 0.853(0.829-0.876); 4th quartile: 0.880(0.856-0.904); P=0.047). Additionally, there was a significant difference in adjusted mean of PSA between 1st quartile and 4th quartile group of testosterone (P=0.006) and between 2nd quartile and 4th quartile group of testosterone (P=0.006). There was significant difference in adjusted mean of PSA between a group with hypogonadism (serum testosterone < 3.5 ng/mL) and a group without hypogonadism increased (adjusted mean (interquartile range) ng/dL: testosterone <3.5ng/mL: 0.801(0.766-0.835); testosterone ≥3.5ng/mL: 0.859(0.847-0.872); P=0.002). <strong>Conclusions</strong>: The present study suggests that PSA may have a positive correlation with serum testosterone. Careful considerations about serum testosterone are recommended for the interpretation of PSA results. keywords : Testosterone, Prostate-specific antigen, Hypogonadism