Long-term Use of 5-alpha-reductase Inhibitors is Safe and Effective in Men on Active Surveillance for Prostate Cancer - Full-Text Article

Background - Although 5-alpha-reductase inhibitors (5ARIs) have been shown to benefit men with prostate cancer (PCa) on active surveillance (AS), their long-term safety remains controversial. Our objective is to describe the long-term association of 5ARI use with PCa progression in men on AS.

Materials/subjects and methods - The cohort of men with low-risk PCa was derived from a prospectively maintained AS database at the Princess Margaret (1995–2016). Pathologic, grade, and volume progression were the primary end points. Kaplan–Meier time-to-event analysis was performed and Cox proportional hazards regression was used to determine predictors of progression where 5ARI exposure was analyzed as a time-dependent variable. Patients who came off AS prior to any progression events were censored at that time.

Results - The cohort included 288 men with median follow-up of 82 months (interquartile range: 37–120 months). Among non-5ARI users (n = 203); 114 men (56.2%) experienced pathologic progression compared with 24 men (28.2%) in the 5ARI group (n = 85), (p < 0.001). Grade and volume progression were higher in the non-5ARI group compared with the 5ARI group (n = 82; 40.4% vs. n = 19; 22.4% respectively, p = 0.003 for grade progression; n = 87; 43.1% and n = 15; 17.7%, respectively for volume progression p < 0.001). Lack of 5ARI use was independently positively associated with pathologic progression (HR: 2.65; CI: 1.65–4.24), grade progression (HR: 2.75; CI: 1.49–5.06), and volume progression (HR: 3.15; CI: 1.78–5.56). The frequency of progression to high-grade (Grade Group 4–5) tumors was not significantly different between the groups.

Conclusions - Use of 5ARIs diminished both grade and volume progression without an increased risk of developing Grade Groups 4–5 disease.


The management of low-risk prostate cancer (PCa) has changed dramatically over the past 10 years. More and more men are being offered active surveillance (AS) as an initial treatment strategy for localized PCa.1,2 This paradigm shift is largely due to the long-term safety demonstrated with this approach by two large North American studies.3-5 Although there is some regional variation in the adoption of AS, acceptance for this strategy is now almost universal.2,6-8 There continues to be debate as to the criteria that should be used to select patients for AS with advantages and disadvantages based on the stringency of inclusion,9 but in general much of Grade Group (GG) 1 (Gleason 6) PCa is now initially managed with AS.9

Although AS serves many men well, some do eventually go on to definitive treatment. On average 20% of men reclassify (upgrade to a higher GG) at the confirmatory biopsy and an additional 2% per year progress as they are found to have Gleason 7 (GG 2 or 3) or higher-grade disease on follow-up biopsies.10-12 There have been a number of efforts to diminish the rate of progression with chemoprevention13,14 using statins, 5-alpha-reductase inhibitors (5ARIs), metformin, and other agents.14-16 Of all the agents, 5ARIs have had the most dramatic benefit in this setting.17-20

5ARIs have been shown to diminish the rate of progression and all-cause reclassification to definitive treatment in men on AS.17-20 For example, we have reported the rate of progression for men taking 5ARI vs. those who did not, 18.6% vs. 36.7%, respectively.17 A follow-up analysis of our original report17 demonstrated that the benefit persisted when measuring 5ARI exposure as a time dependent covariate.21 Although encouraging, there is a lack of evidence on long-term outcomes and a general concern about the safety of 5ARI use remains. Given that the United States Food and Drug Administration continues to have a “black box” warning (the strictest warning put in the labeling of prescription drugs or drug products by the Food and Drug Administration when there is reasonable evidence of an association of a serious hazard with the drug) on 5ARI labeling due to the possible association with highgrade disease, there is merit in continuing to evaluate the safety of 5ARIs in the AS setting. Although many other men have joined our AS cohort and took a 5ARI, for the purpose of this study, we present long-term follow-up data on our initial cohort.17

Materials/subjects and methods

Patient cohort

This cohort has been previously described.17 The study and waiver of consent was approved by the University Health Network institutional review board. All patients were identified using a prospectively maintained AS database at Princess Margaret Cancer Centre (PM) between 1995 and 2016 (an additional 6 years of follow-up on the original cohort). Inclusion criteria for AS at PM are as follows: PSA < 10 ng/ml, clinical stage T1c–T2a, Group 1 (Gleason score < 6), and three or fewer positive cores with no more than 50% of a core involved at initial diagnostic biopsy. For the purposes of this study, the cohort was restricted to all patients who had at least one repeat biopsy after diagnosis. Men were excluded if they were on a 5ARI before being diagnosed with PCa, or had received definitive treatment at another center. All biopsies were performed by one of three dedicated uro-radiologists using a standardized approach that did not depend on prostate volume.22

Initial biopsies consisted of 6 cores before 2001 and 11 cores after 2001, while repeat biopsies were 10 cores before 2001 and 15–16 cores after 2001.The follow-up protocol for AS at PM is a confirmatory biopsy 12–18 months after the initial biopsy followed by rebiopsy every 2–3 years until the patient reaches 80 years of age, is no longer medically fit or refuses treatment. PSA is measured every 3 months for 2 years and then every 6 months in stable patients. Digital rectal examination (DRE) is performed every 6 months. A rapid PSA velocity, abnormal DRE, or discretion of the treating physician may also trigger an earlier rebiopsy. MRI is used at the discretion of the treating urologist.

Study end points

Overall pathologic progression was the primary end point of this study and was defined as increased grade, increased number of cores to more than three, or any core involvement of more than 50% at confirmatory biopsy (reclassification) or later biopsies. We also assessed grade progression and volume progression (increase in number of positive cores (>3) or percent core involvement (>50%) separately. The secondary end point was progression to definitive treatment, which could have been triggered by pathologic progression or patient/physician preference.

5ARI exposure

All patients completed a questionnaire at each clinic visit, which was used in conjunction with the clinical records to document 5ARI initiation and discontinuation. A patient was considered a 5ARI user if they commenced 5ARI use after their diagnostic biopsy. For descriptive purposes, nonusers were patients who never used 5ARI after diagnostic biopsy nor prior to definitive treatment or censoring. The main indications for 5ARI use in this cohort were benign prostate hyperplasia and lower urinary tract symptoms.

Statistical analysis

Variables were compared between 5ARI users and nonusers with analysis of variance for continuous variables and chisquare contingency test for categorical variables. The association between 5ARI use and the primary end point, pathologic progression, was explored using the Kaplan–Meier method. Univariable and multivariable Cox proportional hazards regression analyses were performed to determine predictors of pathologic, grade, and volume progression. In these models, 5ARI use was modeled as a time-dependent covariate allowing patients who may have started a 5ARI after a period of nonuse, or discontinuing a 5ARI after a period of use, to contribute person-time to both the user and nonuser groups. Patients who stopped AS prior to any progression events were censored at the time of stopping AS. To account for differences in prostate volume at baseline between 5ARI and non-5ARI groups, sensitivity analyses were performed restricting men in the non-5ARI group to those with larger glands (prostate volume > 40 ml). PH assumption was checked graphically by plotting log (-log(survival)) vs. log(time). The 95% confidence intervals were calculated; p values < 0.05 were considered significant. Data were analyzed using SAS v.9.4 (SAS Institute Inc, Cary, NC, USA).


From 1995 to 2010, a total of 288 men were identified that fulfilled the study inclusion criteria and represent the cohort included in our previous report [17, 21]. Mean age at diagnosis was 64.2 years (n = 288) and median follow-up time was 82 months (interquartile range [IQR]: 37–120 months). With further follow-up until 2016, pathologic progression was evident in 138 men (54.5%). Thirty-five (12.2%) men came off surveillance after a median follow-up of 52 months (IQR: 31–78 months) prior to any progression events. Reasons for stopping surveillance included: lost to follow-up (n = 4), transition to watchful waiting (n = 7), transfer of care to another facility (n = 11), and death from non-PCa-related causes (n = 13). The median follow-up for men remaining on AS was 113 months (IQR: 78–135 months).

In the update of the 288 patients, 85 were identified as 5ARI users, their baseline clinical and pathologic characteristics are summarized in Table 1. Notably at baseline 5ARI users had a larger median prostate size, 58 ml vs. 41 ml (p < 0.001), yet PSA at diagnosis, was not significantly higher (p = 0.06). 

Table 1. Baseline characteristics (n = 288).

table 1 baseline characteristics

Table 2 summarizes the follow-up of these patients. Of the 5ARI users, 23 (27.1%) discontinued AS vs. 104 (51.2%) who were non-5ARI users (p = 0.001). Median time to definitive treatment was longer for those on a 5ARI compared with those who were not (60 vs. 43 months; p = 0.039). From the original cohort [17], 15 (6.8%) men who were initially nonusers of 5ARI started taking the drug, and 17 (24.3%) who initially used 5ARI stopped using them prior to progression. The median duration of ARI use was 75 months (IQR: 24–112 months). Prostate MRI has only been recently available, and in the subset of patients (n = 222) with information on prostate MRI, the frequency of MRI during follow-up was 23.8% for 5ARI users compared with 21.8% for nonusers of 5ARI (p = 0.74).

Table 2. Follow-up variables (n = 288).

table 2 follow up variables

The frequency of pathologic progression was lower in 5ARI users as compared with nonusers (28.2% vs. 56.2%, p < 0.001, Table 3).

Table 3. Overall progression, reclassification, and progression after confirmatory biopsya.

table 3 Overall progression reclassification and progression

The median follow-up time was nearly twice as long in the 5ARI group compared with the non5ARI group, 114 vs. 61 months (p < 0.001). Figure 1 shows the Kaplan–Meier survival curves for pathologic progression for men who ever took 5ARIs and those that did not (p < 0.001).

fig 1 product limit survival estimates

Fig. 1 Pathologic progression in the 5ARI vs. non-5ARI group. Kaplan Meier survival estimate curve to demonstrate. Top curve = 5ARI users, bottom curve = non-5ARI users.

Grade progression was higher in the non-5ARI group compared with the 5ARI group (n = 82; 40.4% vs. n = 19; 22.4% respectively, p = 0.003, Table 3). There were eleven men who progressed to high-grade disease (GG 4–5/Gleason score 8 or higher) after a mean follow-up of 65 months. Two of the eleven were taking 5ARI and nine were not; however, these numbers represented a similar proportion of men in each group (11%) (Supplemental Table 1).

Supplemental Table 1. Distribution of grade progression.

distribution of grade progression

Volume progression was also higher in the non5ARI group compared with the 5ARI group (n = 87; 43.1% and n = 15; 17.7%, respectively, p < 0.001) (Table 3). Reclassification at confirmatory biopsy and progression after confirmatory biopsy were also more frequent in the non-5ARI users compared with 5ARI users, although the difference was not statistically significant for grade progression after confirmatory biopsy (Table 3).

Table 3. Overall progression, reclassification, and progression after confirmatory biopsya.

table 3 Overall progression reclassification and progression

Cox proportional hazards regression modeling was used to examine predictors for pathologic, grade, and volume progression (Table 4).

Table 4. Univariable and multivariable analysis of predictors of progression (Cox proportional hazards regression, time-dependent analysis).

table 4 univariable and multivariable analysis of predictors

5ARI use was analyzed as a time dependent covariate where the time when men were not on a 5ARI during AS was analyzed as part of the non-5ARI group. Lack of 5ARI use was positively associated with pathologic progression (univariable HR: 2.63; CI: 1.65–4.19; p ≤ 0.001 and multivariable HR: 2.65; CI: 1.65–4.24; p < 0.001). Lack of 5ARI use was also significantly positively associated with grade and volume progression separately. Results were similar when 5ARI use was treated as a fixed variable (Supplemental Table 2).

Supplemental Table 2. Univariable and multivariable analysis of predictors of progression (Cox Proportional Hazards regression, 5ARI treated as a fixed variable)

a) Pathologic progression (288 patients, 138 events)

pathologic progression

b) Grade progression (288 patients, 101 events)

grade progression

c) Volume progression (288 patients, 102 events)


In multivariable analysis, age at diagnosis was significantly associated with all types of progression, and number of positive cores and maximum percent core were positively associated with volume progression (Table 4). Family history of PCa was not associated with any of the progression endpoints (data not shown). PSA density and prostate volume were not significantly associated with progression and were not included in the multivariable models. However, because prostate volume differed substantially between 5ARI users and nonusers, we performed a sensitivity analysis where we restricted the sample to patients with prostate volume larger than 40 ml in the non-5ARI group and observed a similar association of 5ARI with pathologic progression (Supplemental Table 3).

Supplemental Table 3: Predictors of pathologic progression in sample restricted to patients with prostate volume (> 40 ml in non-5ARI patients) (Cox Proportional Hazards regression).



Several groups have investigated the use of 5ARIs to mitigate risk of disease progression during AS.17,19,23,24 While evidence supporting 5ARI use is encouraging, some argue that data regarding safety of long-term use of 5ARIs are limited. Furthermore, the clinical benefit of 5ARIs with respect to pathological progression or grade reclassification with time remains an area of controversy. This is especially true when considering the high misclassification rate inherent to early surveillance biopsies,10,25–27 the uncertainty regarding the time required for 5ARIs to exert a potentially meaningful effect and the tendency of patients to pursue treatment in the absence of pathological triggers due to anxiety. In this study, we show a protective effect of 5ARI use on disease progression in men on AS with a median follow-up of ~7 years.

Primary prevention trials (The Prostate Cancer Prevention Trial (PCPT) and REDUCE) showed that 5ARIs reduced the incidence of PCa compared with placebo.28,29 The protective effect of finasteride persisted through 16 years of follow-up.30 An observed increase of Gleason Scores 7–10 in the finasteride study arm compared with the placebo arm noted in a secondary analysis at 7 years of follow-up of the PCPT triggered concern about harm. In addition, the REDUCE trial showed an increase of GG 4–5 (Gleason Scores 8–10) associated with dutasteride.29 However, large cohort studies have reported that that the protective effect of 5ARI on PCa incidence is not associated with higher frequency of high-grade or lethal prostate cancer.31,32

In the setting of secondary prevention, some observational studies failed to demonstrate a benefit of 5ARI in AS. For example, Ross et al.19 looked at 587 men with very low-risk PCa enrolled in an AS program and found that treatment with 5ARIs was not significantly associated with the grade reclassification in men with very low-risk PCa. However, the REDEEM study, a randomized trial of 5ARI compared with placebo in men with low-risk disease on AS, showed a reduced frequency of progression (pathologic or therapeutic) in the 5ARI (38%) compared with the control group (48%) by 3 years.23 Time to pathologic progression specifically was also lower in the 5ARI compared with the placebo group, however the result did not attain traditional statistical significance (p = 0.079). In agreement with the REDEEM results, the original report from our cohort17,21,23 showed that after a median follow-up of 36.1 months, men taking 5ARIs were less likely to have experienced pathologic progression than men not on a 5ARI. Furthermore, multivariable analysis showed that 5ARI use was the strongest covariate associated with pathologic progression and that the rate of high-grade disease (Groups 4 and 5; Gleason 8–10) was not different in patients taking 5ARI compared with those who were not.

This study reports long-term follow-up of our original cohort, accounting for variable and sustained exposure to 5ARI, and confirms a reduced risk of progression with no added risk of developing high-grade (GG 4–5) disease in men who use 5ARIs. There was an ~50% reduction in the rate of volume progression (after confirmatory biopsy) as well as a reduction in progression to definitive treatment in those taking a 5ARI. Multivariable analysis also demonstrated that non-5ARI users were more likely to experience pathologic progression (HR: 2.65). As expected, the baseline characteristics between groups were balanced aside from PSA and prostate volume at diagnosis (Table 1). The difference in prostate volume would actually bias against the 5ARI group given the known prostate volume-grade bias that has been described.33 According to previous work, larger prostates are more likely to be under-graded at initial diagnostic biopsy.33 Thus men in the 5ARI group should have had a higher rate of pathologic progression with subsequent biopsies, as higher grade and/or higher volumes of disease that was not detected on initial biopsy would be found on subsequent ones. We performed a sensitivity analysis restricting the sample to patients with prostate volume to ≥40 ml and found that the protective effect of 5ARIs persisted. Furthermore, the groups were balanced with regard to median number of biopsies per patient and the time between biopsies on surveillance thus minimizing the risk of ascertainment bias explaining the reduced progression rate in the 5ARI group.

The current study has limitations. It was nonrandomized; however, we attempted to overcome potential bias by accounting for all relevant variables. The time on 5ARIs varied among patients as did the specific 5ARI used. To account for men who took the drug inconsistently, a timedependent analysis was carried out to account for time while on AS but not on a 5ARI and to diminish the likelihood of overestimating the benefit of 5ARI. Also, though no fixed follow-up protocol was followed on surveillance, AS practice patterns are similar among the five physicians following these patients. Moreover, the mean number of PSA readings, biopsies, and follow-up was similar between the groups, and thus a specific pattern of practice was unlikely to account for differences between 5ARI users and nonusers.


In our cohort, which chronicles the longest follow-up of men taking 5ARI while on AS, we report an association with favorable outcomes. Specifically, 5ARI use was associated with reduced risk of both grade and volume progression. Most importantly, 5ARI use was not associated with the development of high-grade (GG 4–5; GS 8–10) disease. Thus, 5ARIs should be considered for secondary prevention in men on AS for low-grade disease.

Compliance with ethical standards - Conflict of interest: The authors declare that they have no conflict of interest.

Authors: A. Finelli1, M. Komisarenko1, L. J. Martin1, N. Timilshina2, K. Jain1, J. Morris1, A. Zlotta2, G. Kulkarni1, N. Perlis1, T. van der Kwast1, A. Evans1, S. Ghai1, N. Fleshner1, S. M. H. Alibhai3, R. J. Hamilton1
1. Division of Urology, Department of Surgical Oncology, Princess Margaret Cancer Centre, University of Toronto, Toronto, ON, Canada
2. Division of Urology, Department of Surgical Oncology, Mount Sinai Hospital and University Health Network, University of Toronto, Toronto, ON, Canada
3. Department of Medicine, Institute of Health Policy, Management, and Evaluation, University Health Network, University of Toronto, Toronto, ON, Canada

Finelli, A., Komisarenko, M., Martin, L.J. et al. Long-term use of 5-alpha-reductase inhibitors is safe and effective in men on active surveillance for prostate cancer. Prostate Cancer Prostatic Dis (2020). https://doi.org/10.1038/s41391-020-0218-2
Received: 25 November 2019 / Revised: 11 February 2020 / Accepted: 19 February 2020 © The Author(s), under exclusive licence to Springer Nature Limited 2020


1. Tosoian JJ, Carter HB, Lepor A, Loeb S. Active surveillance for prostate cancer: current evidence and contemporary state of practice. Nat Rev Urol. 2016;13:205–15.
2. Cooperberg MR. Active Surveillance for low-risk prostate cancer —an evolving international standard of care. JAMA Oncol. 2017;3:1398–9.
3. Klotz L, Vesprini D, Sethukavalan P, Jethava V, Zhang L, Jain S, et al. Long-term follow-up of a large active surveillance cohort of patients with prostate cancer. J Clin Oncol. 2015;33:272–7.
4. Tosoian JJ, Trock BJ, Landis P, Feng Z, Epstein JI, Partin AW, et al. Active surveillance program for prostate cancer: an update of the Johns Hopkins experience. J Clin Oncol. 2011;29:2185–90.
5. Tosoian JJ, Mamawala M, Epstein JI, Landis P, Wolf S, Trock BJ, et al. Intermediate and longer-term outcomes from a prospective active-surveillance program for favorable-risk prostate cancer. J Clin Oncol. 2015;33:3379–85.
6. Bruinsma SM, Bangma CH, Carroll PR, Leapman MS, Rannikko A, Petrides N, et al. Active surveillance for prostate cancer: a narrative review of clinical guidelines. Nat Rev Urol. 2016;13:151–67.
7. Loeb S. Active surveillance offers functional advantages without impacting survival for low-risk prostate cancer. Eur Urol. 2018;73:868–9.
8. Loeb S, Folkvaljon Y, Curnyn C, Robinson D, Bratt O, Stattin P. Uptake of active surveillance for very-low-risk prostate cancer in Sweden. JAMA Oncol. 2017;3:1393–8.
9. Komisarenko M, Timilshina N, Richard PO, Alibhai SM, Hamilton R, Kulkarni G, et al. Stricter active surveillance criteria for prostate cancer do not result in significantly better outcomes: a comparison of contemporary protocols. J Urol. 2016;196: 1645–50.
10. Wong LM, Ferrara S, Alibhai SM, Evans A, Van der Kwast T, Trottier G, et al. Diagnostic prostate biopsy performed in a nonacademic center increases the risk of re-classification at confirmatory biopsy for men considering active surveillance for prostate cancer. Prostate Cancer Prostatic Dis. 2015;18:69–74.
11. Wong LM, Trottier G, Toi A, Lawrentschuk N, Van der Kwast TH, Zlotta A, et al. Should follow-up biopsies for men on active surveillance for prostate cancer be restricted to limited templates? Urology. 2013;82:405–9.
12. Weiss B, Loeb S. Standard and targeted biopsy during follow-up for active surveillance. Rev Urol. 2015;17:112–3.
13. Violette PD, Saad F. Chemoprevention of prostate cancer: myths and realities. J Am Board Fam Med. 2012;25:111–9.
14. Vemana G, Hamilton RJ, Andriole GL, Freedland SJ. Chemoprevention of prostate cancer. Annu Rev Med. 2014;65:111–23.
15. Richard PO, Finelli A. 5-Alpha reductase inhibitors in active surveillance. Curr Opin Urol. 2014;24:324–8.
16. Jayalath VH, Nayan M, Finelli A, Komisarenko M, Timilshina N, Kulkarni GS, et al. Statin use and time to progression in men on active surveillance for prostate cancer. Prostate Cancer Prostatic Dis. 2018;21:509–15.
17. Finelli A, Trottier G, Lawrentschuk N, Sowerby R, Zlotta AR, Radomski L, et al. Impact of 5alpha-reductase inhibitors on men followed by active surveillance for prostate cancer. Eur Urol 2011;59:509–14.
18. Shelton PQ, Ivanowicz AN, Wakeman CM, Rydberg MG, Norton J, Riggs SB, et al. Active surveillance of very-low-risk prostate cancer in the setting of active treatment of benign prostatic hyperplasia with 5alpha-reductase inhibitors. Urology. 2013;81:979–84.
19. Ross AE, Feng Z, Pierorazio PM, Landis P, Walsh PC, Carter HB, et al. Effect of treatment with 5-alpha reductase inhibitors on progression in monitored men with favourable-risk prostate cancer. BJU Int. 2012;110:651–7.
20. Vidal AC, Howard LE, Moreira DM, Castro-Santamaria R, Andriole GL, Freedland SJ. Aspirin, NSAIDs, and risk of prostate cancer: results from the REDUCE study. Clin Cancer Res. 2015;21:756–62.
21. Wong LM, Fleshner N, Finelli A. Impact of 5-alpha reductase inhibitors on men followed by active surveillance for prostate cancer: a time-dependent covariate reanalysis. Eur Urol. 2013;64:343.
22. Babaian RJ, Toi A, Kamoi K, Troncoso P, Sweet J, Evans R, et al. A comparative analysis of sextant and an extended 11-core multisite directed biopsy strategy. J Urol. 2000;163:152–7.
23. Fleshner NE, Lucia MS, Egerdie B, Aaron L, Eure G, Nandy I, et al. Dutasteride in localised prostate cancer management: the REDEEM randomised, double-blind, placebo-controlled trial. Lancet. 2012;379:1103–11.
24. Dai C, Ganesan V, Zabell J, Nyame YA, Almassi N, Greene DJ, et al. Impact of 5alpha-reductase inhibitors on disease reclassification among men on active surveillance for localized prostate cancer with favorable features. J Urol. 2018;199:445–52.
25. Komisarenko M, Wong LM, Richard PO, Timilshina N, Toi A, Evans A, et al. An increase in gleason 6 tumor volume while on active surveillance portends a greater risk of grade reclassification with further followup. J Urol. 2016;195:307–12.
26. Iremashvili V, Pelaez L, Manoharan M, Jorda M, Rosenberg DL, Soloway MS. Pathologic prostate cancer characteristics in patients eligible for active surveillance: a head-to-head comparison of contemporary protocols. Eur Urol. 2012;62:462–8.
27. Andriole GL, Bostwick D, Brawley OW, Gomella L, Marberger M, Montorsi F, et al. The effect of dutasteride on the usefulness of prostate specific antigen for the diagnosis of high grade and clinically relevant prostate cancer in men with a previous negative biopsy: results from the REDUCE study. J Urol. 2011;185:126–31.
28. Thompson IM, Goodman PJ, Tangen CM, Lucia MS, Miller GJ, Ford LG, et al. The influence of finasteride on the development of prostate cancer. N Engl J Med. 2003;349:215–24.
29. Andriole GL, Bostwick DG, Brawley OW, Gomella LG, Marberger M, Montorsi F, et al. Effect of dutasteride on the risk of prostate cancer. N Engl J Med. 2010;362:1192–202.
30. Unger JM, Hershman DL, Till C, Tangen CM, Barlow WE, Ramsey SD, et al. Using Medicare claims to examine long-term prostate cancer risk of finasteride in the prostate cancer prevention trial. J Natl Cancer Inst. 2018.
31. Preston MA, Wilson KM, Markt SC, Ge R, Morash C, Stampfer MJ, et al. 5Alpha-reductase inhibitors and risk of high-grade or lethal prostate cancer. JAMA Intern Med. 2014;174:1301–7.
32. Wallerstedt A, Strom P, Gronberg H, Nordstrom T, Eklund M. Risk of prostate cancer in men treated with 5alpha-reductase inhibitors—a Large Population-Based Prospective Study. J Natl Cancer Inst. 2018;110:1216–21.
33. Kulkarni GS, Lockwood G, Evans A, Toi A, Trachtenberg J, Jewett MA, et al. Clinical predictors of Gleason score upgrading: implications for patients considering watchful waiting, active surveillance, or brachytherapy. Cancer. 2007;109:2432–8.

Read an Editorial by J. Kellogg Parsons, MD, MHS: The Efficacy and Safety of Long-term Use of 5-alpha-reductase Inhibitors for Preventing Progression for Prostate Cancer Patients on Active Surveillance – Editorial