Kearns JT, Matulay JT, Anderson WE et al: Association of 5-alpha reductase inhibitor use with prostate specific antigen level at the time of urology referral in a retrospective cohort at a large, integrated health care system. Urol Pract 2021; 8: 619.

Five-alpha reductase inhibitor (5-ARI) use reduces serum prostate specific antigen (PSA) by 50% without a concomitant increase in prostate cancer (PCa) risk. For this reason, providers must double serum PSA in 5-ARI users prior to assessing PCa risk.¹ Men taking 5-ARIs have more severe prostate cancer at diagnosis and worse PCa outcomes, which may be the consequence of referring providers not accounting for 5-ARI use when assessing PSA and PCa risk.² In this article, we assessed whether 5-ARI users would have a higher overall corrected serum PSA level at time of referral to urology for further assessment of PCa risk.³

This retrospective cohort study was conducted using data from Atrium Health, an integrated health system with over 900 care locations in North Carolina, South Carolina and Georgia. Practice settings include urban, suburban and rural geographies. We included patients who were male, ≥40 years old and seen in an encounter in the ambulatory or outpatient clinical setting during the years 2018–2019. We excluded men who had received a prostate biopsy within the prior 24 months or a PCa diagnosis within the prior 18 months before their first Atrium outpatient or ambulatory encounter. 5-ARI users were defined as men with active orders for 5-ARIs ≥6 months from the time the PSA level was resulted, and nonusers were defined as men who had no order for a 5-ARI during the 18 months prior to PSA testing. We excluded men whose first order for a 5-ARI occurred 0–6 months prior to the PSA result, as it can take several months for 5-ARI use to fully impact PSA levels. Referral to urology for additional PCa diagnostic evaluation was defined as having a urology referral with an associated diagnosis suggesting PCa risk (eg elevated PSA, abnormal digital rectal examination).

Figure. Median serum PSA at index encounter and urology referral. PSA levels were corrected for 5-ARI use.

Characteristics of men referred to urology for additional PCa diagnostic evaluation are presented in the table 1. 5-ARI users had approximately double the serum PSA levels overall (2.2 vs 1.0 ng/ml, p <0.01) and at time of urology referral (11.6 vs 5.6 ng/ml, p <0.01), and were significantly older at time of urology referral (median 69.0 vs 64.0 years, p <0.01). After correction for age, race and body mass index (BMI), 5-ARI use was associated with increased PSA at urology referral, with 1.5–2 times corrected PSA values at urology referral compared to nonusers (see figure).

Previous work showed that men who used 5-ARIs prior to PCa diagnosis were diagnosed with more aggressive PCa and had worse PCa-specific outcomes.² A study of 80,875 men cared for in the Veterans Affairs system from 2011–2018 found that 5-ARI users had longer time from elevated PSA to diagnosis, higher PSA at prostate biopsy, higher risk PCa at diagnosis and higher PCa-specific mortality.² In contrast, the Prostate Cancer Prevention Trial and REDUCE Trial suggest no increased PCa-specific mortality in 5-ARI users with long-term followup.^4–6 However, the patients in these trials underwent standardized, per-protocol biopsies, which makes them a very different population from a screening cohort.

There were several limitations, including lack of randomization and use of the electronic health record (EHR) for data collection. Also, it has not been established that a 50% reduction in PSA levels with 5-ARI use is applicable to every patient. Furthermore, we did not have access to pharmacy dispensing data to measure patient compliance, so prescription orders were used as a surrogate.

In conclusion, men taking 5-ARIs have significantly higher serum PSA at time of referral to urology for additional PCa diagnostic evaluation than nonusers in this cohort. As the unadjusted PSA at referral to urology for PCa risk was the same between 5-ARI users and nonusers, the effect of 5-ARI use on serum PSA levels might not be routinely considered when assessing PCa risk. Future interventions to better inform patients and providers of the effect of 5-ARI use on serum PSA are warranted.