- Email: [email protected]
Impact of 5α-Reductase Inhibitors on Men Followed by Active Surveillance for Prostate Cancer
EUROPEAN UROLOGY 59 (2011) 509–514
available at www.sciencedirect.com journal homepage: www.europeanurology.com
Platinum Priority – Prostate Cancer Editorial by Eugene K. Cha and Shahrokh F. Shariat on pp. 515–517 of this issue
Impact of 5a-Reductase Inhibitors on Men Followed by Active Surveillance for Prostate Cancer Antonio Finelli a,*, Greg Trottier a, Nathan Lawrentschuk a, Robert Sowerby a, Alexandre R. Zlotta b, Lenny Radomski a, Narhari Timilshina a, Andrew Evans a, Theodorus H. van der Kwast a, Ants Toi a, Micheal A.S. Jewett a, John Trachtenberg a, Neil E. Fleshner a a
Princess Margaret Hospital, University Health Network, University of Toronto, Ontario, Canada
b
Mount Sinai Hospital, University of Toronto, Toronto, Ontario, Canada
Article info
Abstract
Article history: Accepted December 15, 2010 Published online ahead of print on December 28, 2010
Background: In two large randomized controlled trials, 5a-reductase inhibitors (5-ARIs) were shown to prevent prostate cancer. No prior work had shown the effect of 5-ARIs on those already diagnosed with low-risk prostate cancer. Objective: Our aim was to determine the effect of 5-ARIs on pathologic progression in men on active surveillance. Design, setting, and participants: We conducted a single-institution retrospective cohort study comparing men taking a 5-ARI versus no 5-ARI while on active surveillance for prostate cancer. Measurements: Pathologic progression was evaluated and defined as Gleason score >6, maximum core involvement >50%, or more than three cores positive on a follow-up prostate biopsy. Kaplan-Meier analyses were conducted along with multivariable Cox proportional hazard regression modeling for predictors of pathologic progression. Results and limitations: A total of 288 men on active surveillance met the inclusion criteria. The median follow-up was 38.5 mo (interquartile range: 23.6–59.4) with 93 men (32%) experiencing pathologic progression and 96 men (33%) abandoning active surveillance. Men taking a 5-ARI experienced a lower rate of pathologic progression (18.6% vs 36.7%; p = 0.004) and were less likely to abandon active surveillance (20% vs 37.6%; p = 0.006). On multivariable Cox proportional hazards analysis, lack of 5-ARI use was most strongly associated with pathologic progression (hazard ratio: 2.91; 95% confidence interval, 1.5–5.6). The main study limitation was the retrospective design and variable duration of 5-ARI therapy. Conclusions: The 5-ARIs were associated with a significantly lower rate of pathologic progression and abandonment of active surveillance.
Keywords: Active surveillance Chemoprevention Prostate neoplasm
* Corresponding author. Division of Urology, Department of Surgical Oncology, Princess Margaret Hospital, 610 University Ave 3–130, Toronto, Ontario, Canada, M5G 2M9. Tel. +1 416 946 4501 ext. 2851; Fax: +1 416 946 6590. E-mail address: antonio.fi[email protected] (A. Finelli). 0302-2838/$ – see back matter. Crown Copyright # 2011 Published by Elsevier B.V. on behalf of European Association of Urology. All rights reserved.
doi:10.1016/j.eururo.2010.12.018
510
1.
EUROPEAN UROLOGY 59 (2011) 509–514
Introduction
first-time biopsies consisted of 6 cores before the year 2001 and 11 cores after the year 2001. Repeat biopsies consisted of 10 cores before 2001 and
Prostate cancer is the most common noncutaneous malignancy with >200 000 new cases diagnosed in the United States in 2010 [1]. Although screening for prostate cancer is controversial, there has been widespread use of prostate-specific antigen (PSA) testing that has resulted in earlier diagnoses. It is estimated that 1 in 6 men will develop prostate cancer, but it is believed only 1 in 34 will die from the disease [2]. There has been debate about the best management strategy for localized prostate cancer, and given that surgery and radiation have an adverse impact on quality of life [3], active surveillance has emerged as a popular strategy [4,5]. Low-risk prostate cancer, defined as a PSA level < 10 ng/ml, stage T1c–T2a, and Gleason score 6, accounts for approximately 50% of newly diagnosed prostate cancers [4]. Active surveillance involves close monitoring of patients with the initiation of curative treatment should there be signs of progression [6]. The National Comprehensive Cancer Network and the National Institute for Health and Clinical Excellence both suggest that active surveillance be considered a first-line treatment strategy in low-risk disease [7,8]. There has been a growing interest in 5a-reductase inhibitors (5-ARIs) for prostate cancer prevention and management. These drugs inhibit the conversion of testosterone to dihydrotestosterone, which drives both benign and malignant growth of prostatic tissue. Although these drugs are indicated for the treatment of benign prostatic hyperplasia, they have been studied in prostate cancer prevention. Two large randomized controlled trials demonstrated a significant reduction in prostate cancer detection in men on 5-ARIs [9,10]. Despite slight differences in study design, both 5-ARIs (finasteride and dutasteride) were associated with an approximately 25% relative reduction in prostate cancer diagnoses compared with placebo [9,10]. Given the benefits of the 5-ARIs, it has been hypothesized that they may delay or prevent progression of low-risk prostate cancer [11]. Although we await the results from an ongoing randomized controlled study of dutasteride versus placebo in men on active surveillance, the Reduction by Dutasteride of Clinical Progression Events in Expectant Management trial [11], to date there has been no report of 5-ARIs in this setting, and thus we sought to examine the relationship between 5-ARI use and progression while on active surveillance. 2.
Patients and methods
A prospectively maintained database approved by the ethics committee of all prostate biopsies performed at the Princess Margaret Hospital along with clinical records were used to construct a database of all men undergoing active surveillance. We identified men at our center with an initial diagnostic biopsy from 1995 onward who were then followed with active surveillance. Inclusion criteria included PSA <10 ng/ml, clinical stage T1c–T2a, Gleason score <6, and three or fewer cores positive with no more than 50% of a core involved at initial diagnostic biopsy. All patients had at least one repeat biopsy after diagnosis. Men were excluded if they were on a 5-ARI before being diagnosed with prostate cancer, did not have a follow-up biopsy, or had received active treatment at another center. All biopsies were performed by one of three dedicated uroradiologists using a standardized approach that did not depend on prostate volume [12]. All
15–16 cores after 2001. Because all biopsies were read by one of four genitourinary pathologists, maintaining a high level of grading consistency and synoptic reporting was used to further standardize the process. Five physicians in a nonstandardized fashion followed patients, although a relatively similar pattern of care was provided. For the most part, PSA was measured every 3 mo for 2 yr and then every 6 mo in stable patients. Digital rectal examination (DRE) was performed every 6 mo. A confirmatory biopsy was typically performed 12 mo after the initial biopsy and then every 2–3 yr until the patient reached 80 yr of age or refused treatment. Biopsy was performed earlier if PSA velocity was rapid, DRE was abnormal, or at the discretion of the caring physician. Pathologic progression was defined as increased grade, increased number of cores to more than three, or any core involvement >50%. Regarding 5-ARI use, all patients completed a questionnaire at each visit, which was used in conjunction with the clinical records to document 5-ARI initiation and use. The primary end point of this study was pathologic progression. The secondary end point was progression to active treatment. Continuous variables were analyzed with parametric statistics (t test, analysis of variance), and when data were not normally distributed they were analyzed with nonparametric tests (Wilcoxon signed rank test). Proportions were analyzed with a chi-square contingency test. A time-to-event analysis (Kaplan-Meier method) was performed in all patients with time to pathologic progression and active treatment as the events of interest. Univariable and multivariable Cox proportional hazards regression analyses were performed to determine predictors of pathologic progression. The 95% confidence interval was calculated; p values <0.05 were considered significant. Data were analyzed using SAS v.9.1 (SAS Institute Inc, Cary, NC, USA).
3.
Results
From 1995 to 2010, 389 patients diagnosed with low-risk prostate cancer were enrolled in active surveillance. Of the 389 men, 288 fulfilled the inclusion criteria for this study. These 288 patients had a mean age of 64 yr and a median follow-up of 38.5 mo (interquartile range [IQR]: 23.6–59.4 mo). In total, 93 men (32%) demonstrated pathologic progression after a median follow-up of 36.1 mo (IQR: 21.3–56.5 mo). For various reasons, 96 men (33.3%) came off surveillance after a median follow-up of 32.3 mo (IQR 17.3–52.9 mo). Pathologic progression was the most common reason with Gleason score >6, three cores or more, and >50% of core involvement the most common reasons. In addition, approximately 20% of men abandoned active surveillance because of anxiety. The median follow-up for men remaining on active surveillance was 43.2 mo (IQR: 25.2–60.9 mo). Of the 288 patients, 70 were started on a 5-ARI at variable time points after their diagnostic biopsy. Table 1 summarizes the clinical and pathologic variables of interest for men taking a 5-ARI and those who did not. Notably the men who took a 5-ARI had a larger median prostate size, 61 ml versus 41 ml ( p < 0.0001), and significantly higher PSA, 5.4 ng/ml versus 4.8 ng/ml ( p = 0.04). Ninety-three men experienced pathologic progression, 13 (18.6%) and 80 (36.7%) in the 5-ARI and non–5-ARI groups, respectively ( p = 0.004). The median time to progression was longer in the 5-ARI group, 41.3 versus 35.1 mo ( p = 0.013). Many of the patients had at least two of the factors considered as
511
EUROPEAN UROLOGY 59 (2011) 509–514
Table 1 – Baseline characteristics of active surveillance patients Variables No. of patients Age, yr, mean (SD) Year of diagnosis, n (%)
Family history of prostate cancer, n (%)
PSA at diagnosis, median (IQR) Prostate volume, ml, median (IQR) Positive core at diagnosis, n (%)
1995–1999 2000–2004 2005 or later Yes No Unknown
1 2 3
Maximum % core involvement*, mean (range) Median No. of biopsies (IQR) Overall median follow-up, mo (IQR)
No 5-ARI
5-ARI
218 63.8 (7.8) 22 (10.1) 74 (33.9) 122 (55.9) 48 (22.0) 103 (47.3) 67 (30.7) 4.8 (3.1–6.6) 41 (31.0–58.0) 155 (71.1) 47 (21.7) 16 (7.3) 6.0 (1.0–35.0) 3 (2–3) 36.7 (19.7–59.2)
70 65.6 (6.4) 6 (8.6) 19 (27.1) 45 (64.3) 10 (14.3) 31 (44.3) 29 (41.4) 5.4 (4.0–7.2) 61.0 (45.0–78.0) 50 (71.4) 13 (18.6) 7 (10.0) 6.8 (1.0–30.0) 3 (2–3) 45.8 (29.7–61.6)
p value
0.067 0.46
0.17 0.037 <0.001 0.71 0.37 0.76 0.014
5-ARI = 5a-reductase inhibitor; IQR = interquartile range; PSA = prostate-specific antigen; SD = standard deviation. This value is the core with the highest percentage involvement of a given biopsy from a given patient.
*
pathologic progression on follow-up biopsy (n = 45) with 48 having had only one of the pathologic features of progression on follow-up biopsy. Regarding progression to active treatment, 14 men (20%) on a 5-ARI versus 82 men (37.6%) who were not on a 5-ARI abandoned active surveillance ( p = 0.006). Median time to active treatment was significantly longer for those on a 5-ARI, 42.5 versus 31.5 mo ( p = 0.026). Fig. 1 depicts Kaplan-Meier curves for the two groups defined by 5-ARI status and time to pathologic progression and time to active treatment. Predictors of pathologic progression were analyzed with Cox proportional hazards regression analysis of baseline covariables (Table 2). On univariate analysis, baseline number of positive cores and family history were not associated with pathologic progression (data not shown). Multivariate analysis showed that lack of 5-ARI use was the strongest covariate associated with pathologic progression (hazard ratio [HR]: 2.91;
95% confidence interval [CI], 1.5–5.6), with increased age and PSA also statistically significant. Although the maximum percentage of any positive cores and lower prostate volume were associated with pathologic progression on univariate analysis, they were not predictive of progression in the multivariable model. Table 3 summarizes the clinical and pathologic features of men immediately before undergoing active treatment. Patients who went on to active treatment were similar regardless of 5-ARI status other than mean PSA was higher in the non–5-ARI group, 6.8 ng/ml versus 2.6 ng/ml ( p = 0.003), respectively. Reasons for initiating active treatment were similar between the two groups ( p = 0.22) with most men exiting active surveillance for pathologic progression. Because median prostate volume was significantly greater in men taking a 5-ARI, a repeat analysis was performed with prostate volume restricted to >40 ml in the non–5-ARI group. Even with prostate volume balanced between the patient
[()TD$FIG]
Fig. 1 – Kaplan-Meier analysis of (A) time to pathologic progression and (B) time to active treatment for patients on active surveillance for prostate cancer who are taking a 5a-reductase inhibitor (5-ARI) or not.
512
EUROPEAN UROLOGY 59 (2011) 509–514
Table 2 – Multivariable Cox proportional hazards regression analysis of baseline biopsy predictors of pathologic progression Variables
HR (95% CI)
Total group* No 5-ARI Age per year increase Baseline PSA Maximum % core at baseliney Prostate volume at baseline
2.91 1.04 1.14 1.01 0.99
p value
(5.64–1.50) (1.07–1.00) (1.26–1.03) (1.04–0.98) (1.00–0.98)
When prostate volume restricted to 40 ml in non–5-ARI groupz No 5-ARI 2.85 (5.42–1.50) Age at diagnosis 1.04 (1.08–0.99)
0.002 0.03 0.02 0.38 0.12
0.001 0.07
5-ARI = 5a-reductase inhibitor; CI = confidence interval; HR = hazard ratio; PSA = prostate-specific antigen. * Family history and number of positive cores were not significant in univariable Cox proportional hazards regression analysis and were not included in the multivariable model. y This value is the core with the highest percentage involvement of a given biopsy from a given patient. z Family history, maximum percentage of any core, and number of positive cores were not significant in univariable Cox proportional hazards regression analysis and were not included in the multivariable model.
groups, there continued to be a lesser rate of pathologic progression and active treatment in patients taking 5-ARIs (Table 4). For this subset of patients, multivariable Cox proportional hazards regression analysis (Table 2) showed that only lack of 5-ARI use was predictive of pathologic progression in patients with prostates >40 ml (HR: 2.85; 95% CI, 5.4–1.5; p = 0.001). 4.
Discussion
With the increase in PSA screening and lower thresholds for prostate biopsy, there has been a rise in the incidence of
low-risk prostate cancer [1]. Although many of these diagnosed men undergo treatment with curative intent, there is growing acceptance of a strategy of active surveillance. Based on two large randomized controlled trials, 5-ARIs were shown to diminish the risk of prostate cancer diagnoses [9,10]. We sought to evaluate the influence of 5-ARI use in our cohort of men on active surveillance. In general, the study population is similar to other published series of active surveillance [13–16]. The median follow-up for the entire cohort was 38.5 mo, and 33.3% of patients went on to active treatment after a median of 32.3 mo. The patient demographics, disease characteristics, and progression rates are similar to other studies, which indirectly validates the cohort and the generalizability of our findings. Our study suggests a benefit of 5-ARI use in men on active surveillance. There was an approximate 50% reduction in the rate of pathologic progression as well as progression to active treatment in those taking a 5-ARI. Multivariate analysis also demonstrated that a lack of 5-ARI use was the strongest predictor of pathologic progression (HR: 2.91), followed by higher baseline PSA and age. As expected, the baseline characteristics between groups were balanced aside from prostate volume and PSA (Table 1). This difference would bias against the 5-ARI group given the known prostate volume-grade bias that has been described [17]. Larger prostates are more likely undergraded at initial diagnostic biopsy [17]. Thus men in the 5-ARI group should theoretically have had a higher rate of pathologic progression with subsequent biopsies more likely to detect higher grade and/or higher volumes of disease that was not noted on initial biopsy [17]. Nonetheless, to balance the effect of prostate volume, prostate size was restricted to >40 ml in
Table 3 – Clinical and pathologic features of patients going on to active treatment Variables No. of patients Age, yr, mean (SD) Year of diagnosis, n (%)
1995–1999 2000–2004 2005 or later
PSA level, median (IQR) Prostate volume, median (IQR) Gleason score, n (%)
Positive cores before active treatment, n (%)
6 7 8 9 3 >3
*
Maximum % core involvement , mean (range) Primary reason for active treatment, n (%)
Increased Gleason LUTS Increased PSA Increased No. of cores Increased tissue Anxiety Unknown
No 5-ARI
5-ARI
82 68.2 (7.7) 16 (19.5) 29 (35.4) 37 (45.1) 6.8 (5.0–9.2) 40.5 (32.0–55.0) 46 (56.1) 29 (35.4) 5 (6.1) 2 (2.4) 52 (63.4) 30 (36.6) 23.0 (1.0–100.0) 45 (54.8) 1 (1.2) 7 (8.5) 9 (10.9) 6 (7.3) 14 (17.1) 0 (0.0)
14 66.9 (6.9) 3 (21.4) 4 (28.6) 7 (50.0) 2.6 (1.7–5.5) 48.5 (36.0–62.0) 8 (57.1) 5 (35.7) 0 (0.0) 1 (7.1) 10 (71.4) 4 (28.6) 26.7 (1.0–80.0) 7 (50.0) 0 (0.0) 2 (14.3) 0 (0.0) 1 (7.1) 3 (21.4) 1 (7.1)
p value
0.57 0.88 0.003 0.24 0.63
0.56 0.62 0.22
5-ARI = 5a-reductase inhibitor; IQR = interquartile range; LUTS = lower urinary tract symptoms; PSA = prostate-specific antigen; SD = standard deviation. This value is the core with the highest percentage involvement of a given biopsy from a given patient.
*
513
EUROPEAN UROLOGY 59 (2011) 509–514
Table 4 – Baseline characteristics and outcomes of patients when restricting no 5a-reductase group to prostate volume I40 ml (n = 199) Variables No. of patients Age, yr, mean (SD) Year of diagnosis, n (%)
Family history of prostate cancer, n (%)
PSA at diagnosis, median (IQR) Prostate volume, median (IQR) No. of positive cores at diagnosis, n (%)
Median No. of biopsies (IQR) Maximum % core involvement*, mean (range) Overall median follow-up, mo (IQR) Progression to active treatment, n (%) Median time to active treatment, mo (IQR) Primary reason for active treatment, n (%)
1995–1999 2000–2004 2005 or later Yes No Unknown
1 2 3
Yes No Increased Gleason LUTS Increased PSA Increased No. of cores Increased tissue Anxiety Unknown
Overall mean follow-up, mo (range) Mean follow-up patients to active treatment, mo (range) Median follow-up patients without active treatment, mo (IQR) Mean follow-up patients without active treatment, mo (range) Pathologic progression, n (%) Yes No Median time to pathologic progression, mo (IQR)
No 5-ARI
5-ARI
129 65.4 (7.1) 16 (12.4) 49 (37.9) 64 (49.6) 32 (24.8) 57 (44.2) 40 (31.0) 5.0 (4.2–6.9) 56.0 (45.0–66.0) 89 (68.9) 30 (23.3) 10 (7.8) 3 (2–3) 5.9 (1.0–35.0) 38.4 (23.0–60.2) 47 (36.4) 82 (63.6) 32.9 (17.5–49.7) 28 (59.6) 1 (2.1) 4 (8.5) 5 (10.6) 3 (6.4) 6 (12.8) 0 (0.0) 40.9 (1.4–119–6) 35.7 (1.4–112.2) 41.0 (25.2–61.3) 45.5 (4.4–119.6) 48 (37.2) 81 (62.8) 36.0 (19.3–56.3)
70 65.6 (6.5) 6 (8.6) 19 (27.1) 45 (64.3) 10 (14.3) 31 (44.3) 29 (41.3) 5.4 (4.0–7.2) 61.0 (40.0–78.0) 50 (71.4) 13 (18.6) 7 (10.0) 3 (2–3) 6.8 (1.0–30.0) 45.8 (29.7–61.6) 14 (20.0) 56 (80.0) 42.4 (25.4–60.7) 7 (50.0) 0 (0.0) 2 (1.43) 0 (0.0) 1 (7.1) 3 (21.4) 1 (7.1) 50.9 (7.4–176.1) 55.6 (13.9–176.1) 45.8 (29.8–61.8) 49.7 (7.4–154.3) 13 (18.6) 57 (81.4) 41.3 (28.4–58.4)
p value
0.85 0.14
0.15 0.72 0.13 0.68 0.89 0.31 0.032 0.016 0.033
0.34
0.032 0.033 0.42 0.4 0.006 0.034
IQR = interquartile range; LUTS = lower urinary tract symptoms; PSA = prostate-specific antigen; SD = standard deviation. This value is the core with the highest percentage involvement of a given initial biopsy from a given patient.
*
the non–5-ARI group and when compared with the 5-ARI group did not diminish the protective effect of 5-ARIs (Tables 2 and 4). Furthermore, the groups were balanced with regard to median follow-up and median number of biopsies per patient even with more patients having progressed in the non–5-ARI group. Thus a longer duration of follow-up or more intense and frequent prostate biopsy was not the cause for the greater progression rate in the non–5-ARI group. In summary, 5-ARIs appear to diminish progression and delay the time to progression as well. Men taking 5-ARIs were less likely to abandon active surveillance, and if they did, it was after a longer median follow-up, 55.5 versus 36.9 mo ( p = 0.026). Of the men who went on to active treatment, there was a similar distribution of reasons for abandoning active surveillance with pathologic progression the most common reason (Table 3). Of the men pursuing active treatment, Gleason scores of 7 on a follow-up biopsy was the reason for 50% and 55% of men in the 5-ARI and non–5-ARI groups, respectively. Seven patients in the non–5-ARI group developed Gleason scores >8 versus one in the 5-ARI group ( p = 0.63). In addition to upgrading, 15 patients in the non–5-ARI group and 1 in the 5-ARI group progressed to active treatment because of increased tissue involvement. Thus it appears that men in both groups exited active surveillance for similar reasons,
but 5-ARIs in general diminished the rate of progression to active treatment by approximately 50%. This study had limitations. It was nonrandomized, and certain data elements were obtained retrospectively. We attempted to overcome this bias by comparing all relevant variables (Table 1). The groups were similar other than prostate volume and PSA, but an analysis restricted to larger prostates in the non–5-ARI group demonstrated persistence of the benefit of 5-ARIs (Tables 2 and 4). The duration of 5-ARI therapy varied among patients as did the specific 5-ARI used. Also, this cohort was not prospectively enrolled in a protocol, and thus their follow-up was not identical, although practice patterns were quite similar among the five physicians. 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 any differences. 5.
Conclusions
In conclusion, 5-ARIs may be beneficial to men on active surveillance. Although we await the results of a large randomized controlled trial [11], we found that men taking 5-ARIs were nearly half as likely to experience pathologic progression or undergo active treatment than those not
514
EUROPEAN UROLOGY 59 (2011) 509–514
taking the drug. This is particularly relevant in the current era in which we are cognizant of overdetection and overtreatment of low-risk prostate cancer.
[4] Klotz L. Active surveillance with selective delayed intervention using PSA doubling time for good risk prostate cancer. Eur Urol 2005; 47:16–21. [5] Hardie C, Parker C, Norman A, et al. Early outcomes of active
Author contributions: Antonio Finelli had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
surveillance for localized prostate cancer. BJU Int 2005;95:956– 60. [6] Dall’Era MA, Cooperberg MR, Chan JM, et al. Active surveillance for early-stage prostate cancer: review of the current literature. Cancer
Study concept and design: Finelli, Trottier, Radomski, Timilshina, Fleshner, Lawrentschuk.
2008;112:1650–9. [7] NCCN clinical practice guidelines in oncology 2010: prostate
Acquisition of data: Trottier, Sowerby, Radomski, Lawrentschuk.
cancer. National Comprehensive Cancer Network Web site.
Analysis and interpretation of data: Trottier, Finelli, Timilshina, Radomski,
http://www.nccn.org/professionals/physician_gls/f_guidelines.asp.
Fleshner, Zlotta. Drafting of the manuscript: Finelli, Trottier. Critical revision of the manuscript for important intellectual content: Fleshner, Zlotta, Trachtenberg, Toi, Jewett, Evans, van der Kwast. Statistical analysis: Timilshina, Finelli, Trottier, Radomski. Obtaining funding: Finelli. Administrative, technical, or material support: Finelli, Trottier. Supervision: Finelli, Fleshner. Other (specify): None. Financial disclosures: I certify that all conflicts of interest, including specific financial interests and relationships and affiliations relevant to the subject matter or materials discussed in the manuscript (eg, employment/ affiliation, grants or funding, consultancies, honoraria, stock ownership or options, expert testimony, royalties, or patents filed, received, or pending), are the following: Neil E. Fleshner is a consultant/ advisor for Glaxo Smith Kline and Merck. Antonio Finelli is a consultant/ advisor for Glaxo Smith Kline. Funding/Support and role of the sponsor: None.
Accessed October 15, 2010. [8] Prostate cancer: diagnosis and treatment. National Institute for Health and Clinical Excellence Web site. http://guidance.nice.org. uk/CG58. Accessed October 15, 2010. [9] Andriole GL, Bostwick DG, Brawley OW, et al. Effect of dutasteride on the risk of prostate cancer. N Engl J Med 2010;362:1192–202. [10] Thompson IM, Goodman PJ, Tangen CM, et al. The influence of finasteride on the development of prostate cancer. N Engl J Med 2003;349:215–24. [11] Fleshner N, Gomella LG, Cookson MS, et al. Delay in the progression of low-risk prostate cancer: rationale and design of the Reduction by Dutasteride of Clinical Progression Events in Expectant Management (REDEEM) trial. Contemp Clin Trials 2007;28:763–9. [12] Babaian RJ, Toi A, Kamoi K, et al. A comparative analysis of sextant and an extended 11-core multisite directed biopsy strategy. J Urol 2000;163:152–7. [13] Eggener SE, Mueller A, Berglund RK, et al. A multi-institutional evaluation of active surveillance for low risk prostate cancer. J Urol 2009;181:1635–41. [14] Klotz L, Zhang L, Lam A, Nam R, Mamedov A, Loblaw A. Clinical
References [1] Jemal A, Siegel R, Xu J, Ward E. Cancer statistics, 2010. CA Cancer J Clin 2010;60:277–300. [2] Cancer facts and figures. American Cancer Society Web site. http://
results of long-term follow-up of a large, active surveillance cohort with localized prostate cancer. J Clin Oncol 2010;28:126–31. [15] Carter HB, Kettermann A, Warlick C, et al. Expectant management of prostate cancer with curative intent: an update of the Johns Hopkins experience. J Urol 2007;178:2359–64.
www.cancer.org/acs/groups/content/@epidemiologysurveilance/
[16] Dall’Era MA, Konety BR, Cowan JE, et al. Active surveillance for the
documents/document/acspc-026238.pdf. Accessed September 14,
management of prostate cancer in a contemporary cohort. Cancer
2010.
2008;112:2664–70.
[3] Sanda MG, Dunn RL, Michalski J, et al. Quality of life and satisfaction
[17] Kulkarni GS, Al-Azab R, Lockwood G, et al. Evidence for a biopsy
with outcome among prostate-cancer survivors. N Engl J Med
derived grade artifact among larger prostate glands. J Urol 2006;175:
2008;358:1250–61.
505–9.
available at www.sciencedirect.com journal homepage: www.europeanurology.com
Platinum Priority – Prostate Cancer Editorial by Eugene K. Cha and Shahrokh F. Shariat on pp. 515–517 of this issue
Impact of 5a-Reductase Inhibitors on Men Followed by Active Surveillance for Prostate Cancer Antonio Finelli a,*, Greg Trottier a, Nathan Lawrentschuk a, Robert Sowerby a, Alexandre R. Zlotta b, Lenny Radomski a, Narhari Timilshina a, Andrew Evans a, Theodorus H. van der Kwast a, Ants Toi a, Micheal A.S. Jewett a, John Trachtenberg a, Neil E. Fleshner a a
Princess Margaret Hospital, University Health Network, University of Toronto, Ontario, Canada
b
Mount Sinai Hospital, University of Toronto, Toronto, Ontario, Canada
Article info
Abstract
Article history: Accepted December 15, 2010 Published online ahead of print on December 28, 2010
Background: In two large randomized controlled trials, 5a-reductase inhibitors (5-ARIs) were shown to prevent prostate cancer. No prior work had shown the effect of 5-ARIs on those already diagnosed with low-risk prostate cancer. Objective: Our aim was to determine the effect of 5-ARIs on pathologic progression in men on active surveillance. Design, setting, and participants: We conducted a single-institution retrospective cohort study comparing men taking a 5-ARI versus no 5-ARI while on active surveillance for prostate cancer. Measurements: Pathologic progression was evaluated and defined as Gleason score >6, maximum core involvement >50%, or more than three cores positive on a follow-up prostate biopsy. Kaplan-Meier analyses were conducted along with multivariable Cox proportional hazard regression modeling for predictors of pathologic progression. Results and limitations: A total of 288 men on active surveillance met the inclusion criteria. The median follow-up was 38.5 mo (interquartile range: 23.6–59.4) with 93 men (32%) experiencing pathologic progression and 96 men (33%) abandoning active surveillance. Men taking a 5-ARI experienced a lower rate of pathologic progression (18.6% vs 36.7%; p = 0.004) and were less likely to abandon active surveillance (20% vs 37.6%; p = 0.006). On multivariable Cox proportional hazards analysis, lack of 5-ARI use was most strongly associated with pathologic progression (hazard ratio: 2.91; 95% confidence interval, 1.5–5.6). The main study limitation was the retrospective design and variable duration of 5-ARI therapy. Conclusions: The 5-ARIs were associated with a significantly lower rate of pathologic progression and abandonment of active surveillance.
Keywords: Active surveillance Chemoprevention Prostate neoplasm
* Corresponding author. Division of Urology, Department of Surgical Oncology, Princess Margaret Hospital, 610 University Ave 3–130, Toronto, Ontario, Canada, M5G 2M9. Tel. +1 416 946 4501 ext. 2851; Fax: +1 416 946 6590. E-mail address: antonio.fi[email protected] (A. Finelli). 0302-2838/$ – see back matter. Crown Copyright # 2011 Published by Elsevier B.V. on behalf of European Association of Urology. All rights reserved.
doi:10.1016/j.eururo.2010.12.018
510
1.
EUROPEAN UROLOGY 59 (2011) 509–514
Introduction
first-time biopsies consisted of 6 cores before the year 2001 and 11 cores after the year 2001. Repeat biopsies consisted of 10 cores before 2001 and
Prostate cancer is the most common noncutaneous malignancy with >200 000 new cases diagnosed in the United States in 2010 [1]. Although screening for prostate cancer is controversial, there has been widespread use of prostate-specific antigen (PSA) testing that has resulted in earlier diagnoses. It is estimated that 1 in 6 men will develop prostate cancer, but it is believed only 1 in 34 will die from the disease [2]. There has been debate about the best management strategy for localized prostate cancer, and given that surgery and radiation have an adverse impact on quality of life [3], active surveillance has emerged as a popular strategy [4,5]. Low-risk prostate cancer, defined as a PSA level < 10 ng/ml, stage T1c–T2a, and Gleason score 6, accounts for approximately 50% of newly diagnosed prostate cancers [4]. Active surveillance involves close monitoring of patients with the initiation of curative treatment should there be signs of progression [6]. The National Comprehensive Cancer Network and the National Institute for Health and Clinical Excellence both suggest that active surveillance be considered a first-line treatment strategy in low-risk disease [7,8]. There has been a growing interest in 5a-reductase inhibitors (5-ARIs) for prostate cancer prevention and management. These drugs inhibit the conversion of testosterone to dihydrotestosterone, which drives both benign and malignant growth of prostatic tissue. Although these drugs are indicated for the treatment of benign prostatic hyperplasia, they have been studied in prostate cancer prevention. Two large randomized controlled trials demonstrated a significant reduction in prostate cancer detection in men on 5-ARIs [9,10]. Despite slight differences in study design, both 5-ARIs (finasteride and dutasteride) were associated with an approximately 25% relative reduction in prostate cancer diagnoses compared with placebo [9,10]. Given the benefits of the 5-ARIs, it has been hypothesized that they may delay or prevent progression of low-risk prostate cancer [11]. Although we await the results from an ongoing randomized controlled study of dutasteride versus placebo in men on active surveillance, the Reduction by Dutasteride of Clinical Progression Events in Expectant Management trial [11], to date there has been no report of 5-ARIs in this setting, and thus we sought to examine the relationship between 5-ARI use and progression while on active surveillance. 2.
Patients and methods
A prospectively maintained database approved by the ethics committee of all prostate biopsies performed at the Princess Margaret Hospital along with clinical records were used to construct a database of all men undergoing active surveillance. We identified men at our center with an initial diagnostic biopsy from 1995 onward who were then followed with active surveillance. Inclusion criteria included PSA <10 ng/ml, clinical stage T1c–T2a, Gleason score <6, and three or fewer cores positive with no more than 50% of a core involved at initial diagnostic biopsy. All patients had at least one repeat biopsy after diagnosis. Men were excluded if they were on a 5-ARI before being diagnosed with prostate cancer, did not have a follow-up biopsy, or had received active treatment at another center. All biopsies were performed by one of three dedicated uroradiologists using a standardized approach that did not depend on prostate volume [12]. All
15–16 cores after 2001. Because all biopsies were read by one of four genitourinary pathologists, maintaining a high level of grading consistency and synoptic reporting was used to further standardize the process. Five physicians in a nonstandardized fashion followed patients, although a relatively similar pattern of care was provided. For the most part, PSA was measured every 3 mo for 2 yr and then every 6 mo in stable patients. Digital rectal examination (DRE) was performed every 6 mo. A confirmatory biopsy was typically performed 12 mo after the initial biopsy and then every 2–3 yr until the patient reached 80 yr of age or refused treatment. Biopsy was performed earlier if PSA velocity was rapid, DRE was abnormal, or at the discretion of the caring physician. Pathologic progression was defined as increased grade, increased number of cores to more than three, or any core involvement >50%. Regarding 5-ARI use, all patients completed a questionnaire at each visit, which was used in conjunction with the clinical records to document 5-ARI initiation and use. The primary end point of this study was pathologic progression. The secondary end point was progression to active treatment. Continuous variables were analyzed with parametric statistics (t test, analysis of variance), and when data were not normally distributed they were analyzed with nonparametric tests (Wilcoxon signed rank test). Proportions were analyzed with a chi-square contingency test. A time-to-event analysis (Kaplan-Meier method) was performed in all patients with time to pathologic progression and active treatment as the events of interest. Univariable and multivariable Cox proportional hazards regression analyses were performed to determine predictors of pathologic progression. The 95% confidence interval was calculated; p values <0.05 were considered significant. Data were analyzed using SAS v.9.1 (SAS Institute Inc, Cary, NC, USA).
3.
Results
From 1995 to 2010, 389 patients diagnosed with low-risk prostate cancer were enrolled in active surveillance. Of the 389 men, 288 fulfilled the inclusion criteria for this study. These 288 patients had a mean age of 64 yr and a median follow-up of 38.5 mo (interquartile range [IQR]: 23.6–59.4 mo). In total, 93 men (32%) demonstrated pathologic progression after a median follow-up of 36.1 mo (IQR: 21.3–56.5 mo). For various reasons, 96 men (33.3%) came off surveillance after a median follow-up of 32.3 mo (IQR 17.3–52.9 mo). Pathologic progression was the most common reason with Gleason score >6, three cores or more, and >50% of core involvement the most common reasons. In addition, approximately 20% of men abandoned active surveillance because of anxiety. The median follow-up for men remaining on active surveillance was 43.2 mo (IQR: 25.2–60.9 mo). Of the 288 patients, 70 were started on a 5-ARI at variable time points after their diagnostic biopsy. Table 1 summarizes the clinical and pathologic variables of interest for men taking a 5-ARI and those who did not. Notably the men who took a 5-ARI had a larger median prostate size, 61 ml versus 41 ml ( p < 0.0001), and significantly higher PSA, 5.4 ng/ml versus 4.8 ng/ml ( p = 0.04). Ninety-three men experienced pathologic progression, 13 (18.6%) and 80 (36.7%) in the 5-ARI and non–5-ARI groups, respectively ( p = 0.004). The median time to progression was longer in the 5-ARI group, 41.3 versus 35.1 mo ( p = 0.013). Many of the patients had at least two of the factors considered as
511
EUROPEAN UROLOGY 59 (2011) 509–514
Table 1 – Baseline characteristics of active surveillance patients Variables No. of patients Age, yr, mean (SD) Year of diagnosis, n (%)
Family history of prostate cancer, n (%)
PSA at diagnosis, median (IQR) Prostate volume, ml, median (IQR) Positive core at diagnosis, n (%)
1995–1999 2000–2004 2005 or later Yes No Unknown
1 2 3
Maximum % core involvement*, mean (range) Median No. of biopsies (IQR) Overall median follow-up, mo (IQR)
No 5-ARI
5-ARI
218 63.8 (7.8) 22 (10.1) 74 (33.9) 122 (55.9) 48 (22.0) 103 (47.3) 67 (30.7) 4.8 (3.1–6.6) 41 (31.0–58.0) 155 (71.1) 47 (21.7) 16 (7.3) 6.0 (1.0–35.0) 3 (2–3) 36.7 (19.7–59.2)
70 65.6 (6.4) 6 (8.6) 19 (27.1) 45 (64.3) 10 (14.3) 31 (44.3) 29 (41.4) 5.4 (4.0–7.2) 61.0 (45.0–78.0) 50 (71.4) 13 (18.6) 7 (10.0) 6.8 (1.0–30.0) 3 (2–3) 45.8 (29.7–61.6)
p value
0.067 0.46
0.17 0.037 <0.001 0.71 0.37 0.76 0.014
5-ARI = 5a-reductase inhibitor; IQR = interquartile range; PSA = prostate-specific antigen; SD = standard deviation. This value is the core with the highest percentage involvement of a given biopsy from a given patient.
*
pathologic progression on follow-up biopsy (n = 45) with 48 having had only one of the pathologic features of progression on follow-up biopsy. Regarding progression to active treatment, 14 men (20%) on a 5-ARI versus 82 men (37.6%) who were not on a 5-ARI abandoned active surveillance ( p = 0.006). Median time to active treatment was significantly longer for those on a 5-ARI, 42.5 versus 31.5 mo ( p = 0.026). Fig. 1 depicts Kaplan-Meier curves for the two groups defined by 5-ARI status and time to pathologic progression and time to active treatment. Predictors of pathologic progression were analyzed with Cox proportional hazards regression analysis of baseline covariables (Table 2). On univariate analysis, baseline number of positive cores and family history were not associated with pathologic progression (data not shown). Multivariate analysis showed that lack of 5-ARI use was the strongest covariate associated with pathologic progression (hazard ratio [HR]: 2.91;
95% confidence interval [CI], 1.5–5.6), with increased age and PSA also statistically significant. Although the maximum percentage of any positive cores and lower prostate volume were associated with pathologic progression on univariate analysis, they were not predictive of progression in the multivariable model. Table 3 summarizes the clinical and pathologic features of men immediately before undergoing active treatment. Patients who went on to active treatment were similar regardless of 5-ARI status other than mean PSA was higher in the non–5-ARI group, 6.8 ng/ml versus 2.6 ng/ml ( p = 0.003), respectively. Reasons for initiating active treatment were similar between the two groups ( p = 0.22) with most men exiting active surveillance for pathologic progression. Because median prostate volume was significantly greater in men taking a 5-ARI, a repeat analysis was performed with prostate volume restricted to >40 ml in the non–5-ARI group. Even with prostate volume balanced between the patient
[()TD$FIG]
Fig. 1 – Kaplan-Meier analysis of (A) time to pathologic progression and (B) time to active treatment for patients on active surveillance for prostate cancer who are taking a 5a-reductase inhibitor (5-ARI) or not.
512
EUROPEAN UROLOGY 59 (2011) 509–514
Table 2 – Multivariable Cox proportional hazards regression analysis of baseline biopsy predictors of pathologic progression Variables
HR (95% CI)
Total group* No 5-ARI Age per year increase Baseline PSA Maximum % core at baseliney Prostate volume at baseline
2.91 1.04 1.14 1.01 0.99
p value
(5.64–1.50) (1.07–1.00) (1.26–1.03) (1.04–0.98) (1.00–0.98)
When prostate volume restricted to 40 ml in non–5-ARI groupz No 5-ARI 2.85 (5.42–1.50) Age at diagnosis 1.04 (1.08–0.99)
0.002 0.03 0.02 0.38 0.12
0.001 0.07
5-ARI = 5a-reductase inhibitor; CI = confidence interval; HR = hazard ratio; PSA = prostate-specific antigen. * Family history and number of positive cores were not significant in univariable Cox proportional hazards regression analysis and were not included in the multivariable model. y This value is the core with the highest percentage involvement of a given biopsy from a given patient. z Family history, maximum percentage of any core, and number of positive cores were not significant in univariable Cox proportional hazards regression analysis and were not included in the multivariable model.
groups, there continued to be a lesser rate of pathologic progression and active treatment in patients taking 5-ARIs (Table 4). For this subset of patients, multivariable Cox proportional hazards regression analysis (Table 2) showed that only lack of 5-ARI use was predictive of pathologic progression in patients with prostates >40 ml (HR: 2.85; 95% CI, 5.4–1.5; p = 0.001). 4.
Discussion
With the increase in PSA screening and lower thresholds for prostate biopsy, there has been a rise in the incidence of
low-risk prostate cancer [1]. Although many of these diagnosed men undergo treatment with curative intent, there is growing acceptance of a strategy of active surveillance. Based on two large randomized controlled trials, 5-ARIs were shown to diminish the risk of prostate cancer diagnoses [9,10]. We sought to evaluate the influence of 5-ARI use in our cohort of men on active surveillance. In general, the study population is similar to other published series of active surveillance [13–16]. The median follow-up for the entire cohort was 38.5 mo, and 33.3% of patients went on to active treatment after a median of 32.3 mo. The patient demographics, disease characteristics, and progression rates are similar to other studies, which indirectly validates the cohort and the generalizability of our findings. Our study suggests a benefit of 5-ARI use in men on active surveillance. There was an approximate 50% reduction in the rate of pathologic progression as well as progression to active treatment in those taking a 5-ARI. Multivariate analysis also demonstrated that a lack of 5-ARI use was the strongest predictor of pathologic progression (HR: 2.91), followed by higher baseline PSA and age. As expected, the baseline characteristics between groups were balanced aside from prostate volume and PSA (Table 1). This difference would bias against the 5-ARI group given the known prostate volume-grade bias that has been described [17]. Larger prostates are more likely undergraded at initial diagnostic biopsy [17]. Thus men in the 5-ARI group should theoretically have had a higher rate of pathologic progression with subsequent biopsies more likely to detect higher grade and/or higher volumes of disease that was not noted on initial biopsy [17]. Nonetheless, to balance the effect of prostate volume, prostate size was restricted to >40 ml in
Table 3 – Clinical and pathologic features of patients going on to active treatment Variables No. of patients Age, yr, mean (SD) Year of diagnosis, n (%)
1995–1999 2000–2004 2005 or later
PSA level, median (IQR) Prostate volume, median (IQR) Gleason score, n (%)
Positive cores before active treatment, n (%)
6 7 8 9 3 >3
*
Maximum % core involvement , mean (range) Primary reason for active treatment, n (%)
Increased Gleason LUTS Increased PSA Increased No. of cores Increased tissue Anxiety Unknown
No 5-ARI
5-ARI
82 68.2 (7.7) 16 (19.5) 29 (35.4) 37 (45.1) 6.8 (5.0–9.2) 40.5 (32.0–55.0) 46 (56.1) 29 (35.4) 5 (6.1) 2 (2.4) 52 (63.4) 30 (36.6) 23.0 (1.0–100.0) 45 (54.8) 1 (1.2) 7 (8.5) 9 (10.9) 6 (7.3) 14 (17.1) 0 (0.0)
14 66.9 (6.9) 3 (21.4) 4 (28.6) 7 (50.0) 2.6 (1.7–5.5) 48.5 (36.0–62.0) 8 (57.1) 5 (35.7) 0 (0.0) 1 (7.1) 10 (71.4) 4 (28.6) 26.7 (1.0–80.0) 7 (50.0) 0 (0.0) 2 (14.3) 0 (0.0) 1 (7.1) 3 (21.4) 1 (7.1)
p value
0.57 0.88 0.003 0.24 0.63
0.56 0.62 0.22
5-ARI = 5a-reductase inhibitor; IQR = interquartile range; LUTS = lower urinary tract symptoms; PSA = prostate-specific antigen; SD = standard deviation. This value is the core with the highest percentage involvement of a given biopsy from a given patient.
*
513
EUROPEAN UROLOGY 59 (2011) 509–514
Table 4 – Baseline characteristics and outcomes of patients when restricting no 5a-reductase group to prostate volume I40 ml (n = 199) Variables No. of patients Age, yr, mean (SD) Year of diagnosis, n (%)
Family history of prostate cancer, n (%)
PSA at diagnosis, median (IQR) Prostate volume, median (IQR) No. of positive cores at diagnosis, n (%)
Median No. of biopsies (IQR) Maximum % core involvement*, mean (range) Overall median follow-up, mo (IQR) Progression to active treatment, n (%) Median time to active treatment, mo (IQR) Primary reason for active treatment, n (%)
1995–1999 2000–2004 2005 or later Yes No Unknown
1 2 3
Yes No Increased Gleason LUTS Increased PSA Increased No. of cores Increased tissue Anxiety Unknown
Overall mean follow-up, mo (range) Mean follow-up patients to active treatment, mo (range) Median follow-up patients without active treatment, mo (IQR) Mean follow-up patients without active treatment, mo (range) Pathologic progression, n (%) Yes No Median time to pathologic progression, mo (IQR)
No 5-ARI
5-ARI
129 65.4 (7.1) 16 (12.4) 49 (37.9) 64 (49.6) 32 (24.8) 57 (44.2) 40 (31.0) 5.0 (4.2–6.9) 56.0 (45.0–66.0) 89 (68.9) 30 (23.3) 10 (7.8) 3 (2–3) 5.9 (1.0–35.0) 38.4 (23.0–60.2) 47 (36.4) 82 (63.6) 32.9 (17.5–49.7) 28 (59.6) 1 (2.1) 4 (8.5) 5 (10.6) 3 (6.4) 6 (12.8) 0 (0.0) 40.9 (1.4–119–6) 35.7 (1.4–112.2) 41.0 (25.2–61.3) 45.5 (4.4–119.6) 48 (37.2) 81 (62.8) 36.0 (19.3–56.3)
70 65.6 (6.5) 6 (8.6) 19 (27.1) 45 (64.3) 10 (14.3) 31 (44.3) 29 (41.3) 5.4 (4.0–7.2) 61.0 (40.0–78.0) 50 (71.4) 13 (18.6) 7 (10.0) 3 (2–3) 6.8 (1.0–30.0) 45.8 (29.7–61.6) 14 (20.0) 56 (80.0) 42.4 (25.4–60.7) 7 (50.0) 0 (0.0) 2 (1.43) 0 (0.0) 1 (7.1) 3 (21.4) 1 (7.1) 50.9 (7.4–176.1) 55.6 (13.9–176.1) 45.8 (29.8–61.8) 49.7 (7.4–154.3) 13 (18.6) 57 (81.4) 41.3 (28.4–58.4)
p value
0.85 0.14
0.15 0.72 0.13 0.68 0.89 0.31 0.032 0.016 0.033
0.34
0.032 0.033 0.42 0.4 0.006 0.034
IQR = interquartile range; LUTS = lower urinary tract symptoms; PSA = prostate-specific antigen; SD = standard deviation. This value is the core with the highest percentage involvement of a given initial biopsy from a given patient.
*
the non–5-ARI group and when compared with the 5-ARI group did not diminish the protective effect of 5-ARIs (Tables 2 and 4). Furthermore, the groups were balanced with regard to median follow-up and median number of biopsies per patient even with more patients having progressed in the non–5-ARI group. Thus a longer duration of follow-up or more intense and frequent prostate biopsy was not the cause for the greater progression rate in the non–5-ARI group. In summary, 5-ARIs appear to diminish progression and delay the time to progression as well. Men taking 5-ARIs were less likely to abandon active surveillance, and if they did, it was after a longer median follow-up, 55.5 versus 36.9 mo ( p = 0.026). Of the men who went on to active treatment, there was a similar distribution of reasons for abandoning active surveillance with pathologic progression the most common reason (Table 3). Of the men pursuing active treatment, Gleason scores of 7 on a follow-up biopsy was the reason for 50% and 55% of men in the 5-ARI and non–5-ARI groups, respectively. Seven patients in the non–5-ARI group developed Gleason scores >8 versus one in the 5-ARI group ( p = 0.63). In addition to upgrading, 15 patients in the non–5-ARI group and 1 in the 5-ARI group progressed to active treatment because of increased tissue involvement. Thus it appears that men in both groups exited active surveillance for similar reasons,
but 5-ARIs in general diminished the rate of progression to active treatment by approximately 50%. This study had limitations. It was nonrandomized, and certain data elements were obtained retrospectively. We attempted to overcome this bias by comparing all relevant variables (Table 1). The groups were similar other than prostate volume and PSA, but an analysis restricted to larger prostates in the non–5-ARI group demonstrated persistence of the benefit of 5-ARIs (Tables 2 and 4). The duration of 5-ARI therapy varied among patients as did the specific 5-ARI used. Also, this cohort was not prospectively enrolled in a protocol, and thus their follow-up was not identical, although practice patterns were quite similar among the five physicians. 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 any differences. 5.
Conclusions
In conclusion, 5-ARIs may be beneficial to men on active surveillance. Although we await the results of a large randomized controlled trial [11], we found that men taking 5-ARIs were nearly half as likely to experience pathologic progression or undergo active treatment than those not
514
EUROPEAN UROLOGY 59 (2011) 509–514
taking the drug. This is particularly relevant in the current era in which we are cognizant of overdetection and overtreatment of low-risk prostate cancer.
[4] Klotz L. Active surveillance with selective delayed intervention using PSA doubling time for good risk prostate cancer. Eur Urol 2005; 47:16–21. [5] Hardie C, Parker C, Norman A, et al. Early outcomes of active
Author contributions: Antonio Finelli had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
surveillance for localized prostate cancer. BJU Int 2005;95:956– 60. [6] Dall’Era MA, Cooperberg MR, Chan JM, et al. Active surveillance for early-stage prostate cancer: review of the current literature. Cancer
Study concept and design: Finelli, Trottier, Radomski, Timilshina, Fleshner, Lawrentschuk.
2008;112:1650–9. [7] NCCN clinical practice guidelines in oncology 2010: prostate
Acquisition of data: Trottier, Sowerby, Radomski, Lawrentschuk.
cancer. National Comprehensive Cancer Network Web site.
Analysis and interpretation of data: Trottier, Finelli, Timilshina, Radomski,
http://www.nccn.org/professionals/physician_gls/f_guidelines.asp.
Fleshner, Zlotta. Drafting of the manuscript: Finelli, Trottier. Critical revision of the manuscript for important intellectual content: Fleshner, Zlotta, Trachtenberg, Toi, Jewett, Evans, van der Kwast. Statistical analysis: Timilshina, Finelli, Trottier, Radomski. Obtaining funding: Finelli. Administrative, technical, or material support: Finelli, Trottier. Supervision: Finelli, Fleshner. Other (specify): None. Financial disclosures: I certify that all conflicts of interest, including specific financial interests and relationships and affiliations relevant to the subject matter or materials discussed in the manuscript (eg, employment/ affiliation, grants or funding, consultancies, honoraria, stock ownership or options, expert testimony, royalties, or patents filed, received, or pending), are the following: Neil E. Fleshner is a consultant/ advisor for Glaxo Smith Kline and Merck. Antonio Finelli is a consultant/ advisor for Glaxo Smith Kline. Funding/Support and role of the sponsor: None.
Accessed October 15, 2010. [8] Prostate cancer: diagnosis and treatment. National Institute for Health and Clinical Excellence Web site. http://guidance.nice.org. uk/CG58. Accessed October 15, 2010. [9] Andriole GL, Bostwick DG, Brawley OW, et al. Effect of dutasteride on the risk of prostate cancer. N Engl J Med 2010;362:1192–202. [10] Thompson IM, Goodman PJ, Tangen CM, et al. The influence of finasteride on the development of prostate cancer. N Engl J Med 2003;349:215–24. [11] Fleshner N, Gomella LG, Cookson MS, et al. Delay in the progression of low-risk prostate cancer: rationale and design of the Reduction by Dutasteride of Clinical Progression Events in Expectant Management (REDEEM) trial. Contemp Clin Trials 2007;28:763–9. [12] Babaian RJ, Toi A, Kamoi K, et al. A comparative analysis of sextant and an extended 11-core multisite directed biopsy strategy. J Urol 2000;163:152–7. [13] Eggener SE, Mueller A, Berglund RK, et al. A multi-institutional evaluation of active surveillance for low risk prostate cancer. J Urol 2009;181:1635–41. [14] Klotz L, Zhang L, Lam A, Nam R, Mamedov A, Loblaw A. Clinical
References [1] Jemal A, Siegel R, Xu J, Ward E. Cancer statistics, 2010. CA Cancer J Clin 2010;60:277–300. [2] Cancer facts and figures. American Cancer Society Web site. http://
results of long-term follow-up of a large, active surveillance cohort with localized prostate cancer. J Clin Oncol 2010;28:126–31. [15] Carter HB, Kettermann A, Warlick C, et al. Expectant management of prostate cancer with curative intent: an update of the Johns Hopkins experience. J Urol 2007;178:2359–64.
www.cancer.org/acs/groups/content/@epidemiologysurveilance/
[16] Dall’Era MA, Konety BR, Cowan JE, et al. Active surveillance for the
documents/document/acspc-026238.pdf. Accessed September 14,
management of prostate cancer in a contemporary cohort. Cancer
2010.
2008;112:2664–70.
[3] Sanda MG, Dunn RL, Michalski J, et al. Quality of life and satisfaction
[17] Kulkarni GS, Al-Azab R, Lockwood G, et al. Evidence for a biopsy
with outcome among prostate-cancer survivors. N Engl J Med
derived grade artifact among larger prostate glands. J Urol 2006;175:
2008;358:1250–61.
505–9.