International Urology and Nephrology

, Volume 45, Issue 4, pp 979–987

The effect of 5α-reductase inhibitors on prostate growth in men receiving testosterone replacement therapy: a systematic review and meta-analysis

Authors

  • Yuanshan Cui
    • Department of Urology, Beijing Tian-Tan HospitalCapital Medical University
  • Huantao Zong
    • Department of Urology, Beijing Tian-Tan HospitalCapital Medical University
  • Chenchen Yang
    • Department of Urology, Beijing Tian-Tan HospitalCapital Medical University
  • Huilei Yan
    • Department of Urology, Beijing Tian-Tan HospitalCapital Medical University
    • Department of Urology, Beijing Tian-Tan HospitalCapital Medical University
Urology - Original Paper

DOI: 10.1007/s11255-013-0477-0

Cite this article as:
Cui, Y., Zong, H., Yang, C. et al. Int Urol Nephrol (2013) 45: 979. doi:10.1007/s11255-013-0477-0

Abstract

Purpose

Androgen replacement therapy is a widely accepted form of treatment worldwide for aging men with late-onset hypogonadism (LOH) syndrome. Urologists have been concerned with the use of androgen supplements due to the possibility of enhancing prostate growth. We performed a systematic review and meta-analysis to assess the effect of 5α-reductase inhibitors on prostate growth in men receiving testosterone replacement therapy.

Methods

A literature review was performed to identify all published randomized placebo-controlled trials (RCT) that used exogenous testosterone combined with 5α-reductase inhibitor therapy for the treatment of hypogonadism. The search included the following databases: MEDLINE, EMBASE, and the Cochrane Controlled Trials Register. The reference lists of the retrieved studies were also investigated, and a systematic review and meta-analysis were conducted.

Results

Five publications involving a total of 250 patients were used in the analysis, including 4 RCTs that were short-term (≤6 mo) comparisons of testosterone plus a 5α-reductase inhibitor with testosterone plus placebo and 3 RCTs that were long-term (18–36 mo) comparisons of testosterone plus a 5α-reductase inhibitor with testosterone plus placebo. In our meta-analysis, we found that testosterone plus a 5α-reductase inhibitor may slow the progression of prostate growth. For the comparison of short-term testosterone plus 5α-reductase inhibitor treatment with testosterone plus placebo therapy, the prostate-specific antigen (PSA) level (the standardized mean difference (SMD) = −0.24, 95 % confidence interval (CI) = −0.45 to 0.04, p = 0.02)) and the prostate volume (SMD = −1.66, 95 % CI = −4.54 to 1.22, p = 0.26) indicated that, compared with testosterone plus placebo therapy, the testosterone plus 5α-reductase inhibitor may decrease the PSA level. For the comparison of long-term testosterone plus 5α-reductase inhibitor with testosterone plus placebo, the PSA level (SMD = −0.53, 95 % CI = −0.84 to 0.21, p = 0.001) and the prostate volume (SMD = −8.53, 95 % CI = −15.51 to 1.54, p = 0.02) showed that, compared with testosterone plus placebo therapy, the testosterone plus 5α-reductase inhibitor treatment may slow the progression of prostate growth.

Conclusions

Our meta-analysis indicates that the treatment of LOH patients with short-term testosterone plus 5α-reductase inhibitor therapy does not lead to prostate growth; however, this treatment could effectively decrease the PSA level. Additionally, long-term testosterone plus 5α-reductase inhibitor therapy could slow the progression of prostate growth.

Keywords

Testosterone5α-reductase inhibitorProstateMeta-analysisRandomized controlled trial

Introduction

Androgen deficiency in the aging male has become a topic of increasing interest and debate worldwide. Epidemiological data indicate that testosterone levels decrease progressively with age and that a significant percentage of men over the age of 60 years have serum testosterone levels that are below the lower limits of young adult (age 20–30 years) men [13]. Late-onset hypogonadism (LOH), also referred to as age-associated testosterone deficiency syndrome, is a clinical and biochemical syndrome associated with advancing age and is characterized by symptoms and a deficiency in serum testosterone levels [4]. This condition may result in a significant detriment to the quality of life and adversely affects the function of multiple organ systems, including the prostate [5]. Androgen replacement therapy (ART) is a widely accepted form of treatment to prevent or ameliorate many of the symptoms and conditions associated with LOH in aging men. The past decade has revealed the beneficial evidence of androgen treatment for hypogonadal men on multiple target organs, and recent studies demonstrate that the short-term beneficial effects of testosterone in older men are similar to those observed in younger men. However, ART has been suggested to potentially promote prostate growth [69]. In a systematic review of 19 randomized trials to determine the risks of adverse events associated with testosterone therapy in older men [10], the combined rate of all prostate events was revealed to be significantly greater in testosterone-treated men than in placebo-treated men (odds ratio, 1.78; 95 % CI, 1.07, 2.95).

The 5α-reductase enzyme is responsible for the conversion of testosterone to dihydrotestosterone (DHT) [11]. High activity of the 5α-reductase enzyme in humans results in excessive DHT levels in peripheral tissues, and hence, the suppression of androgen action by 5α-reductase inhibitors is a logical treatment for benign prostatic hyperplasia (BPH) because they inhibit the conversion of testosterone to DHT [12].

Occasionally, the gradual increase in prostate volume concomitant with the progressive decline in testosterone levels that occurs in middle age reflects the progression of BPH. A common clinical scenario most likely involves a proportion of men for whom 5α-reductase inhibitor therapy is prescribed together with testosterone replacement due to the coexistence of prostate hyperplasia and symptomatic hypogonadism [13].

The aim of the present study was to perform a meta-analysis to clarify the relationship between exogenous testosterone combined with 5α-reductase inhibitors and prostate growth, which may resolve some of the current problems regarding the effect of 5α-reductase inhibitors on prostate growth in men receiving testosterone replacement therapy.

Materials and methods

Search strategy

MEDLINE (from 1966 to December, 2013), EMBASE (from 1974 to December, 2013), the Cochrane Controlled Trials Register, and the reference lists of the retrieved studies were searched to identify RCTs that referred to the effect of exogenous testosterone combined with 5α-reductase inhibitors on the prostate. The following search terms were used: testosterone, 5α-reductase inhibitor, prostate, and randomized controlled trials.

Inclusion criteria

Randomized controlled trials (RCTs) that met the following criteria were included: (1) the study design included treatments with exogenous testosterone combined with a 5α-reductase inhibitor; (2) the study provided accurate data that could be analyzed, including the total number of subjects and the values of each index; and (3) the full text of the study could be accessed.

Trial selection

When the same RCT study was published in various journals or in different years, the most recent publication was used for the meta-analysis. Each study was included if the same group of researchers studied a group of subjects with multiple experiments. The authors of the present paper discussed each of the RCTs that were included and excluded studies that either failed to meet the inclusion criteria or could not be agreed upon. A flow diagram of the study selection process is presented in Fig. 1.
https://static-content.springer.com/image/art%3A10.1007%2Fs11255-013-0477-0/MediaObjects/11255_2013_477_Fig1_HTML.gif
Fig. 1

A flow diagram of the study selection process

Quality assessment

The quality of all the retrieved RCTs was assessed using the Jadad score [14]. All of the identified RCTs were included in the meta-analysis regardless of the quality score. The methodological quality of each study was assessed according to how patients were allocated to the arms of the study, the concealment of allocation procedures, blinding, and data loss due to attrition. The studies were then classified qualitatively according to the guidelines published in the Cochrane Handbook for Systematic Reviews of Interventions 5.1.0. Based on the quality assessment criteria, each study was rated and assigned to one of the three following quality categories: A, if all quality criteria were adequately met, the study was deemed to have a low risk of bias; B, if one or more of the quality criteria were only partially met or were unclear, the study was deemed to have a moderate risk of bias; or C, if one or more of the criteria were not met or were not included, the study was deemed to have a high risk of bias. Sensitivity analyses were then performed on the basis of whether these quality factors were adequate, inadequate, or unclear. Differences were resolved by discussion among the authors.

Data extraction

The following information from the databases was collected: (1) the name of the first author and the publication year, (2) the study design and sample size, (3) the therapy that the patients received, (4) the country of the patients, and (5) the data, including the changes in prostate volume and the prostate-specific antigen (PSA) levels.

Statistical analysis and meta-analysis

The meta-analysis of comparable data was performed using Review Manager 5.1.0 (The Cochrane Collaboration, Oxford, UK). Due to the large number of plots, we combined the 4 forest plots into 2 plots using Adobe Photoshop CS (Figs. 2, 3). We estimated the relative risk (RR) for dichotomous outcomes and the standardized mean difference (SMD) for continuous outcomes pooled across studies using the DerSimonian and Laird random-effects model [15]. We quantified inconsistencies using the I² statistic, which describes the proportion of heterogeneity across studies that is not due to chance and thus describes the extent of true inconsistencies in results across the trials [16]. An I² less than 25 % and an I² more than 50 % reflect small and significant inconsistencies, respectively. For the confidence interval (CI), we used a 95 % CI. If the result of an analysis showed a p > 0.05, we considered that the homogeneity of studies was satisfactory and chose the fixed-effects model. Otherwise, we chose the random-effects model. The presence of publication bias was evaluated using a funnel plot (Fig. 4).
https://static-content.springer.com/image/art%3A10.1007%2Fs11255-013-0477-0/MediaObjects/11255_2013_477_Fig2_HTML.gif
Fig. 2

Forest plots showing changes in a prostate-specific antigen levels and b prostate volume in the short-term treatment studies. T testosterone, 5α: 5α-reductase inhibitor, P placebo, SD standard deviation, IV inverse variance, CI confidence interval

https://static-content.springer.com/image/art%3A10.1007%2Fs11255-013-0477-0/MediaObjects/11255_2013_477_Fig3_HTML.gif
Fig. 3

Forest plots showing changes in a prostate-specific antigen levels and b prostate volume in the long-term treatment studies. T testosterone, 5α: 5α-reductase inhibitor, P placebo, SD standard deviation, IV inverse variance, CI confidence interval

https://static-content.springer.com/image/art%3A10.1007%2Fs11255-013-0477-0/MediaObjects/11255_2013_477_Fig4_HTML.gif
Fig. 4

Funnel plot of the studies represented in the meta-analysis. MD mean difference, SE standard error

Results

Characteristics of the individual studies

The database search revealed 32 articles that could have potentially been included in our meta-analysis. Based on the inclusion and exclusion criteria, 24 articles were excluded after reading the titles and abstracts of the articles. Two articles lacked useful data (the values of each determinant of prostate growth, such as PSA levels and prostate volume). Two articles described the same data; thus, one of them was excluded. In total, 5 articles [1721] with 7 RCTs were included in the analysis: 4 RCTs compared testosterone plus 5α-reductase inhibitor with testosterone plus placebo over the short term (≤6 mo), and 3 RCTs compared testosterone plus 5α-reductase inhibitor with testosterone plus placebo over the long term (18–36 mo). The included studies involved combination therapy that was started at the same time. The baseline characteristics of the studies included in our meta-analysis are listed in Tables 1, 2. The studies included in our analysis had demonstrated effective testosterone replacement (Table 3).
Table 1

Study and patient characteristics

Study

Therapy in experimental group

Therapy in control group

Country

Sample size

T cutoff level for study entry

Age (years)

Exclusion of PSA levels (ng/ml)

Short-term or long-term therapy

Forms of T replacement and dosing

Forms of 5αRI replacement and dosing

Experimental

Control

Amory JK 2004

T + 5αRI

T + P

USA

22

24

TT

12.1 nmol/L

≥65

>4

Long-term

I

200 mg/2wk

O

5 mg/d F

Vaughan C 2007

T + 5αRI

T + P

USA

22

24

TT

12.1 nmol/L

≥65

>4

Long -term

I 200 mg/2wk

O

5 mg/d F

Page ST 2011

T + 5αRI

T + P

USA

24

22

TT

9.7 nmol/L

≥50

>10

Short-term

TD

7.5 g/d

O

0.5 mg/d D

Bhasin S 2012

T + 5αRI

T + P

USA

16

18

TT

590 ng/dL

18–50

>4

Short-term

I

125 mg/wk

O

2.5 mg/d D

Mostaghel EA 2012

T + 5αRI

T + P

USA

9

7

TT

3.9 ng/mL

25–55

>2

Short-term

TD

10 g/d

O

0.5 mg/d D

T testosterone, 5αRI: 5α-reductase inhibitor, P placebo, PSA prostate-specific antigen, TT total testosterone, I injection, TD transdermal, O oral, F finasteride, D dutasteride

Table 2

Quality assessment of individual study

Study

Allocation sequence generation

Allocation concealment

Blinding

Loss to follow-up

Calculation of sample size

Statistical analysis

Intention-to-treat analysis

Level of quality

Amory JK 2004

A

A

A

14

Yes

Student’s t tests

Yes

A

Vaughan C 2007

A

A

A

15

Yes

Analysis of variance

No

A

Page ST 2011

A

A

A

7

Yes

Paired t tests

Yes

A

Bhasin S 2012

A

A

A

13

Yes

Analysis of variance

Yes

A

Mostaghel EA 2012

A

A

A

2

Yes

Analysis of variance

No

A

ITT intention-to-treat analysis

A—all quality criteria met (adequate): low risk of bias

B—one or more of the quality criteria only partly met (unclear): moderate risk of bias

C—one or more criteria not met (inadequate or not used): high risk of bias

Table 3

TT at baseline and at the study end point and ranges for normal testosterone measurements

Study

T + P

T + 5αRI

Ranges for normal testosterone measurements

Baseline

Study end point

Baseline

Study end point

Amory JK 2004

9.9 nmol/L

21 nmol/L

10.1 nmol/L

23 nmol/L

12–33 nmol/L

Vaughan C 2007

9.9 nmol/L

20.4 nmol/L

10 nmol/L

22.6 nmol/L

12–33 nmol/L

Page ST 2011

206 ng/dL

481 ng/dL

213 ng/dL

534 ng/dL

300–1,000 ng/dL

Bhasin S 2012

701 ng/dL

822 ng/dL

842 ng/dL

895 ng/dL

300–1,200 ng/dL

Mostaghel EA 2012

3.9 ng/mL

4.4 ng/mL

4.8 ng/mL

7.0 ng/mL

3–11 ng/mL

T testosterone, 5αRI: 5α-reductase inhibitor, P placebo, TT total testosterone

Quality of the individual studies

Among the studies included in the analysis, all of them described the randomization processes that they had employed. All of the studies used double-blinded RCTs, all studies performed a power calculation to determine the optimal sample size, and three studies used intention-to-treat analysis. The quality level of each identified study ranged from A to C.

Short-term testosterone plus 5α-reductase inhibitor versus testosterone plus placebo treatments

PSA levels

Four RCTs included PSA data representing a cohort of 105 participants (52 in the testosterone plus 5α-reductase inhibitor group and 53 in the testosterone plus placebo group) (Fig. 2). No heterogeneity was found between the trials; the SMD was −0.24, and the 95 % CI was −0.45 to 0.04 (p = 0.02) (Fig. 3). This result suggests that testosterone plus 5α-reductase inhibitor treatment is more likely to decrease PSA levels than treatment using testosterone plus placebo.

Prostate volume changes

Three RCTs included data representing the prostate volume changes in a cohort of 74 participants (38 in the testosterone plus 5α-reductase inhibitor group and 36 in the testosterone plus placebo group) (Fig. 2). According to our analysis, no heterogeneity was observed between the trials, a fixed-effects estimate of the SMD was −1.66, and the 95 % confidence interval (CI) was −4.54 to 1.22 (p = 0.26) (Fig. 2). Thus, comparison of the testosterone plus a 5α-reductase inhibitor treatment with a testosterone plus placebo treatment revealed no apparent differences in prostate volume changes.

Long-term testosterone plus a 5α-reductase inhibitor versus testosterone plus placebo treatments

PSA levels

Three RCTs included the PSA data, representing a cohort of 95 participants (44 in the testosterone plus 5α-reductase inhibitor group and 51 in the testosterone plus placebo group) (Fig. 3). No heterogeneity was found between the trials; the SMD was −0.53, and the 95 % CI was −0.84 to 0.21 (p = 0.001) (Fig. 3). This result suggests that testosterone plus 5α-reductase inhibitor treatment is more likely to decrease PSA levels than treatment with a testosterone plus placebo.

Prostate volume changes

Two RCTs included data representing the prostate volume changes in a cohort of 63 participants (29 in the testosterone plus 5α-reductase inhibitor group and 34 in the testosterone plus placebo group) (Fig. 3). According to our analysis, no heterogeneity was found between the trials, a fixed-effects estimate of the SMD was −8.53, and the 95 % CI was −15.51 to 1.54 (p = 0.02) (Fig. 3). Thus, comparing testosterone plus 5α-reductase inhibitor treatment with a testosterone plus placebo treatment revealed apparent differences in prostate volume changes. The former therapy clearly slowed prostate volume enlargement.

Discussion

Testosterone has a variety of well-known effects, including promoting bone accrual, building and maintaining muscle mass, and promoting erectile function and libido [22]. Of the 5 mg of testosterone manufactured daily by the testes, approximately 6–8 % is metabolized by 5α-reductase to make 0.3 mg of DHT [23]. A literature review by Gormley et al. [24] indicated that significant 5α-reductase expression occurs in the prostate but not in the muscle or bone, and it might lead to locally high concentrations of the potent androgen, DHT, which stimulates prostate growth. The 5α-reductase inhibitor is the principle treatment for BPH and is the only class of therapy to act on the pathophysiologic substrate of the disease to arrest the disease process, reduce prostate volume, improve symptoms, and reduce the risk of acute urinary retention and of BPH-related surgery.

We found that testosterone plus 5α-reductase inhibitor treatment decreases the PSA level compared with testosterone plus placebo during short-term therapy. Comparing long-term testosterone plus 5α-reductase inhibitor with testosterone plus placebo, we found that testosterone plus 5α-reductase inhibitor may slow the progression of prostate growth. As BPH progression is strongly related to prostate volume and PSA [2527], we can therefore state that testosterone plus 5α-reductase inhibitor therapy may slow the progression of prostate growth, and it may be clinically relevant, especially over years of treatment. The 5α-reductase inhibitor treatment may decrease the PSA level by half over a year [28]; although testosterone plus 5α-reductase inhibitor may have reduced the PSA levels in our study, it did not achieve the effect of the therapy with 5α-reductase inhibitor alone (10–33.3 %). In conclusion, the treatment of LOH patients with short-term testosterone plus 5α-reductase inhibitor therapy does not lead to prostate growth; however, it could effectively decrease the PSA level. Additionally, long-term testosterone plus 5α-reductase inhibitor therapy could slow the progression of prostate growth. Amory et al. [29] conducted a RCT comparing testosterone plus 5α-reductase inhibitor with testosterone plus placebo in hypogonadal men for 28 days and reported that testosterone plus 5α-reductase inhibitor could be appealing to patients and clinicians because it provides the benefits of testosterone therapy without stimulating prostate growth.

Urologists have been concerned about the use of androgen supplementation due to the possibility of fueling prostate growth, not only in cancer but also in benign prostatic disease. This concern has led to the specific exclusion of men with symptomatic BPH from most testosterone replacement trials reported to date. Thus, our article is among the first to study the effect of testosterone replacement and the co-administration of potent 5α-reductase inhibitor in hypogonadal men with documented prostatic enlargement. Only one of the included RCT reported patients with moderate BPH (prostate volume ≥30 mL, IPSS 8–20), and the rest reported mild BPH (prostate volume <30 mL, IPSS ≤8). Thus, our conclusion should be derived from patients with mild to moderate BPH.

The mechanism whereby the use of testosterone plus 5α-reductase inhibitors slows the progression of prostate growth is most likely attributable to changes in the intraprostatic hormonal milieu. Data from the included RCTs indicating that the changes in prostate volume were correlated with changes in serum DHT but not testosterone highlight the importance of DHT in prostate growth. 5α-reductase inhibitors decrease serum and prostate DHT concentrations [30], but because DHT is the predominant androgen within the prostate, 5α-reductase inhibitors appear to have related prostate selective physiological effects. Thus, combining physiological doses of testosterone with a 5α-reductase inhibitor might decrease prostate androgen concentrations and action with relative preservation of the beneficial organ-specific effects of testosterone. This attenuation of prostate growth by 5α-reductase inhibition might also be important in preventing symptomatic BPH.

This meta-analysis includes five studies that were all generated from double-blinded RCTs. According to the quality assessment scale that we developed, the individual studies in the meta-analysis are of conforming quality. The results of these analyses are potentially of great importance not only from a scientific standpoint but also for everyday clinical practice. However, the sample sizes of the included studies were small. Only three of the included studies used the various determinants of “prostate growth” measured in the present analysis (prostate volume and PSA levels) as their primary endpoints. Additionally, unpublished studies were not included in this analysis, which may have resulted in study bias. More high-quality trials with larger samples are proposed to learn more about the efficacy and safety of these combination therapies on the prostate.

Conclusions

This meta-analysis indicates that the treatment of LOH patients with short-term testosterone plus 5α-reductase inhibitor therapy does not lead to prostate growth but that it could effectively decrease the PSA level. Furthermore, long-term testosterone plus 5α-reductase inhibitor therapy could successfully slow the progression of prostate growth.

Conflict of interest

The authors had no conflicts of interest to declare in relation to this article.

Copyright information

© Springer Science+Business Media Dordrecht 2013