Use of 5?-reductase inhibitors for benign prostatic hypertrophy and risk of high-grade prostate cancer: A French population-based study.
Association between 5-ARI use and high-grade prostate cancer is still debated since the unanticipated signal reported in PCPT and REDUCE trials. Observational studies with different designs all reported an increased risk for high grade prostate cancer, though not statistically significant. Even if somehow reassuring, those findings did not rule out an association. We then set up the CANARI study focusing on men with symptoms and complications related to benign prostatic hypertrophy.
This population-based nested matched case-control study used comprehensive French Health Insurance Data (SNIIRAM) linked to data from all path labs located in Brittany, France. Among 74,596 men living in Brittany with at least one reimbursement for a drug licensed for symptomatic or complicated benign prostatic hypertrophy (5-ARI monotherapy or in association, alpha-blockers or phytotherapy) and free of prostate cancer in 2010-2011, 859 cases of incident cases (in 2012-2013) were identified and linked to pathology results; five controls per case were randomly selected using an incidence density sampling design. Conditional adjusted odds ratio (95% confidence interval) was used as a measure of association; cumulative duration exposure to 5-ARI was categorised into three classes (<1 year, 1-2 years, >= 2 years) and homogeneity across high- (Gleason score >= 8) and low-grade (Gleason score < 8) prostate cancer was investigated when comparing 5-ARI users to 5-ARI non-users. ClinicalTrials.gov identifier: NCT02873117
767 cases were eligible for the analysis, matched to 3835 controls; 963 men (153 cases and 810 controls) had been exposed to 5-ARI; 3 639 (614 cases and 3025 controls) were 5-ARI non-users, mostly (67%) treated by alpha-adrenoreceptor antagonists. Mean age combined was 69.3 years; men with Gleason score <8 were younger (mean age 68.2 years than men with Gleason >=8 (76.0 years). Patients exposed to 5ARI had a similar number of PSA measurement (85%) than 5ARI non-users, and a quite similar number of prostate samples (respectively 28% and 26 %). Conditional adjusted odds ratio were 0.64 (95%CI, 0.44 to 0.93) and 1.76 (95%CI, 0.97 to 3.21) for low grade and high-grade prostate cancer, respectively, when comparing 5-ARI users for at least 2 years to non users (test for heterogeneity, p = 0.0048).
Our results supported a qualitative heterogeneity when considering an association between long-term exposure to 5-ARI and prostate cancer. These results are in line with previously published randomized clinical trials and, though not yet explained, should be considered when managing long-term treatment for symptomatic benign prostatic hypertrophy.
Funding: This work was supported by the Fondation pour la Recherche Medicale (FRM), grant number "IMD-2013-1228948".