Finasteride is commonly prescribed for hair loss and benign prostate hyperplasia. Prior studies suggest that finasteride can also lower risk of low grade prostate cancer. However, finasteride patients who do develop prostate cancer may have more aggressive disease. We hypothesized that finasteride is associated with increased prostate cancer-specific mortality (PCSM) in a large national cohort.
We identified 111,933 prostate cancer patients with stage I-IV disease diagnosed between 2000 and 2015 from a national Veterans Affairs (VA) database. Of this cohort 11,085 patients had taken finasteride prior to diagnosis and 100,848 did not. Patient-level covariates included Charlson comorbidity score, marital status, employment status, age, zip-code income and education, body mass index, race, alcohol history, and tobacco history. Tumor-related covariates included clinical T, N, and M stage, Gleason score, and prostate specific antigen concentration. Treatment-related covariates included use of hormone therapy and treatment with radiation therapy or surgery. We compared PCSM, overall survival (OS), and non-cancer mortality (NCM) between groups using multivariable Cox regression, controlling for potential confounders. All tests of statistical significance were two sided.
Median follow-up was 6.3 years. Patients who had taken finasteride prior to diagnosis were more likely have clinically T3-4 (5.44% vs. 3.45%), clinically-node positive (3.53% vs. 1.86%), clinically-metastatic (7.35% vs. 3.14%), and Gleason 8-10 (24.79% vs. 16.13%) disease. In a multivariable Cox regression, finasteride patients had worse PCSM (hazard ratio [HR] 1.37; 95% confidence interval [CI] 1.28-1.46; p<0.0001) and OS (HR 1.1; 95% CI 1.07-1.15; p<0.0001), but equivalent NCM (HR 1.03; 95% CI 0.98-1.08; p=0.21). When limiting analysis to individual T, nodal, and metastatic stage, and Gleason score categories the pattern of worse PCSM, worse OS, and equivalent NCM held across all subgroups.
In this cohort of VA patients, prostate cancer patients who took finasteride prior to diagnosis presented with higher stage and grade disease. After controlling for potential confounders, prostate cancer finasteride patients had worse prostate cancer-specific mortality and overall survival. Further study is needed to externally validate these findings.
Author Disclosure: R.R. Sarkar: None. A.K. Bryant: None. K. Kader: None. R. Mckay: Research Grant; Bayer, Pfizer, Genentech. Advisory Board; Norvartis, Janssen. J.P. Einck: Independent Contractor; American College of Radiation Oncology. Board Member, Advisor; Cure Cervical Cancer. oRG. A.J. Mundt: Honoraria; Up to Date, US Oncology, Varian Medical Systems. C.J. Kane: Advisory Board; Janssen, SNP Bio. J.K. Parsons: None. J.D. Murphy: None. B.S. Rose: None.
© 2018 Published by Elsevier Inc.