Association of Treatment With 5 α -Reductase Inhibitors and Prostate Cancer Mortality Among Older Adults

This cohort study examines the association between use of 5α-reductase inhibitors and prostate cancer mortality among US Medicare beneficiaries.


Introduction
5α-reductase inhibitors (5-ARIs) are used to treat benign prostatic enlargement, a common condition causing urinary outflow obstruction. 1 They also reduce prostate-specific antigen (PSA) by approximately 50%. 2 Our group has recently published that among US military veterans, 5-ARIs are associated with delays in prostate cancer (PC) diagnoses, higher grade and stage at presentation, and worse PC-specific mortality (PCSM), presumably because of misinterpreted PSA values. 3We hypothesized that these results are generalizable to the broader US population.Patients were required to have at least 1 year of prediagnosis Medicare enrollment for comorbidity assessment.All patients were followed up until death or December 31, 2015.

Methods
We extracted demographic and comorbidity variables, prescription drug information, and tumor-level variables.We defined 5-ARI use as any prescription of finasteride or dutasteride at least 6 months before a PC diagnosis.We doubled the PSA level for 5-ARI users per the Reduction by Dutasteride of Prostate Cancer Events (REDUCE) trial. 4We did not include in our models staging or treatment information as covariates because they are on the causal pathway between exposure and outcome. 4 tested for differences in covariates between exposure groups using χ 2 and Wilcoxon rank sum tests.We used multivariable Fine-Gray competing risk regression models to obtain estimates of PCSM and noncancer mortality accounting for competing risk of death.We used multivariable Cox proportional hazards regression models to obtain hazard ratio (HR) estimates of all-cause mortality (ACM).Statistical tests were 2-sided, with P < .05considered statistically significant, and were conducted with SAS statistical software version 9.4 (SAS Institute Inc) and R statistical software version 3.5.1 (R Project for Statistical Computing).

Results
The final cohort included 30 313 patients, with median (interquartile range) follow-up of 3.75 (2.33-5.25)years.A total of 2373 patients (7.83%) were prescribed 5-ARIs at least 6 months before PC diagnosis, with a median (interquartile range) treatment duration of 2.46 (1.45-3.74)years.The 4-year cumulative incidence of death from PC was 5.3% in 5-ARI users and 2.8% in nonusers.
Compared with participants who did not receive 5-ARI, those who did were more likely to present with disease that was high grade (Gleason score 8-10) (18% vs 29%, respectively; difference, Author affiliations and article information are listed at the end of this article. Open Access.This is an open access article distributed under the terms of the CC-BY License.1).

Discussion
This cohort study found that 5-ARI users presented with higher adjusted PSA levels and PC disease burden.They also had worse PCSM and ACM, but not worse noncancer mortality.These results are consistent with our recently published findings 3 that observed that among US veterans, 5-ARI use was associated with worse PCSM (subdistribution HR, 1.39; 95% CI, 1.27-1.52;P < .001)and ACM (HR, 1.10; 95% CI, 1.05-1.15;P < .001).Like other studies, [3][4][5][6] we found that comorbidities, unmarried status, old age, low income, and black race were risk factors for PCSM, which adds validity to our findings.One study limitation included the possibility of misclassification bias, in which 5-ARIs were not used as prescribed.
Our results suggest a need for increased awareness of 5-ARI-induced PSA suppression, clearer guidelines for early PC detection, and systems-based mechanisms to help improve care for men using 5-ARIs.
This cohort study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.The study was considered exempt from review by the University of California, San Diego institutional review board because it used deidentified data.Data analysis was performed from February 1, 2019, through April 30, 2019.We used the Surveillance, Epidemiology, and End Results Program-Medicare linked database and identified patients with stage I to IV PC and known PSA level at diagnosis between January 1, 2008, and December 31, 2013; no missing covariate information; and Medicare Part D coverage.

Table 2 .
Prostate Cancer-Specific, Noncancer, and All-Cause Mortality

Table 1 .
Baseline Patient Characteristics at Prostate Cancer Diagnosis (continued)