Chronic Glucocorticoid Use and Risk for Advanced Prostate Cancer at Presentation: A SEER-Medicare Cohort Study

Background: Examine the relationship between exposure to systemic glucocorticoids (steroids) and advanced prostate cancer (PCa) at presentation. Prior work suggested that steroid use may be associated with increased PCa risk. Materials and Methods: We queried the linked SEER-Medicare database (2004-2015) to identify PSA screened patients diagnosed with PCa. Criteria for screening included a PSA lab test or DRE exam in both the 12 month and 13 to 36 month periods prior to diagnosis of PCa. Steroid exposure was determined using Medicare Part D and groups were divided based on duration of use in the 3 years prior to diagnosis: controls with no exposure, <30 days, 30 days - 1 year, 1 to 2 years, and >2+ years. Advanced PCa was defined as systemic metastases or regional lymph node metastasis at presentation. Risk estimates for advanced PCa at presentation for steroid exposure groups vs. controls were assessed with univariable and multivariable logistic regression models. Results: We identified 22,920 PSA screened patients diagnosed with PCa of which 29% used glucocorticoids in the exposure period. The mean (SD) duration for glucocorticoid use (in days) among all steroid users was 76.7 days (192.1). On univariable and multivariable analyses, > 2 years of steroid exposure was associated with significantly increased risk for advanced PCa (OR 2.06, 95% CI 1.35-3.14 and OR 1.74, 95% CI 1.12-2.69, respectively). Conclusion: In this population-based PSA-screened cohort, prolonged steroid use was associated with increased risk of advanced PCa at diagnosis. With the widespread use of glucocorticoids, it is important to consider the role steroids may play in PCa pathogenesis.