A pooled analysis of individual patient data from registrational trials of silodosin in the treatment of non-neurogenic male lower urinary tract symptoms (LUTS) suggestive of benign prostatic hyperplasia (BPH)

Authors


Abstract

Objective

To evaluate the efficacy and safety of silodosin in a pooled analysis based on individual patients data from three randomised controlled trials (RCTs) comparing silodosin and placebo.

Patients and methods

A pooled analysis of 1494 patients from three 12-week, similarly designed, parallel-group, multicentre, randomised, double-blind, placebo-controlled phase III RCTs (SI04009, SI04010, KMD3213-IT-CL 0215) was performed. Differences from placebo for the mean change from baseline to the end of treatment for the International Prostate Symptom Score (IPSS) and uroflowmetry data were tested using an analysis of covariance model.

Results

At study end, in the intention-to-treat population, silodosin was significantly more effective than placebo in improving IPSS total score (adjusted means differences [AMD] 2.7; P < 0.001). Silodosin was significantly more effective than placebo in improving storage, voiding, and quality-of-life-item subscores (all P < 0.001). Similarly, silodosin was more effective than placebo in improving maximum urinary flow rate (Qmax; AMD 0.8; P = 0.002). The most frequently reported adverse event (AE) was ejaculatory dysfunction, reported in 186 (22%) patients in the silodosin group and six (0.9%) in the placebo group (odds ratio 28.14; P < 0.001). Dizziness and orthostatic hypotension rates were similar in silodosin and placebo groups.

Conclusions

Silodosin is an effective treatment for male lower urinary tract symptoms suggestive of benign prostatic hyperplasia. The drug is able to improve total IPSS, all IPSS-related parameters, and Qmax at uroflowmetry. Ejaculatory dysfunction is the main treatment-related AE, whereas prevalence of cardiovascular AEs was similar to placebo.

Introduction

BPH is a common cause of LUTS in adult men [1]. α-Blockers, 5α-reductase inhibitors (5-ARIs), or their combination are the standard medical treatments for patients with uncomplicated bothersome LUTS [2, 3], with α-blockers being the most commonly prescribed category of drug in the European market [4].

All the available α-blockers (alfuzosin, doxazosin, tamsulosin, and terazosin) are currently regarded as equally clinically effective drugs in improving patient symptoms and quality of life due to LUTS, although differences in prevalence of adverse events (AEs) have been well reported among the different drugs [2, 3].

Silodosin is a novel α-blocker, recently approved by both the USA Food and Drug Administration and the European Medicines Agency [5, 6]. The main peculiarity of this novel drug is the higher selectivity for α-1A adrenoceptors and a very limited affinity for the α-1B and α-1D subtypes [7]. Consequently, silodosin is expected to be associated with a very limited number of cardiovascular AEs (mainly, orthostatic hypotension, dizziness, and vertigo), which may be more prevalent with other less selective α-blockers. To date, a few randomised controlled trials (RCTs) have reported a good profile of efficacy and safety for silodosin in comparison with placebo [8-10] or tamsulosin [8, 10, 11]. Moreover, two meta-analyses of RCTs were recently published [12, 13], showing that silodosin is an effective treatment for storage and voiding LUTS. However, both meta-analyses were based only on the overall data reported in the publications. More reliable information can indeed be obtained by evaluating individual patient data [14]. The purpose of the present study was to report a pooled analysis of three RCTs comparing silodosin and placebo based on individual patient's data.

Patients and Methods

Three 12-week, similarly designed, parallel group, multicentre, randomised, double-blind, placebo-controlled phase III studies (SI04009, SI04010, KMD3213-IT-CL 0215; Clinical Trials Registration Numbers NCT00224107, NCT00224120, NCT00359905) were conducted in the USA and Europe, to evaluate the efficacy and safety of silodosin in men with signs and symptoms of BPH [9, 10]. The main inclusion criteria were age ≥50 years, moderate-to-severe LUTS (defined by an IPSS total score of ≥13 points), and BOO (defined by a maximum urinary flow rate [Qmax] of 4–15 mL/s). After a screening period of up to 4 weeks, patients received single-blind treatment with placebo for 4 weeks, which was followed immediately by the 12-week double-blind treatment period. At 2 weeks after the start and at the end of the placebo run-in period, the IPSS and Qmax were determined to assess individual responses to placebo. Patients with a ≥30% decrease in IPSS or an increase in Qmax of ≥3 mL/s during the run-in period in the USA study or with improvement in the IPSS total score of ≥25% in the European study were excluded. In the USA RCTs, eligible patients were randomly assigned (1:1) to double-blind treatment with placebo or 8 mg silodosin once daily with breakfast; in the European study, patients who met the selection criteria were randomly assigned in a ratio of 2:2:1 to silodosin 8 mg, tamsulosin 0.4 mg, or placebo. Due to the absence of an active control arm in the two USA RCTs, only comparisons of silodosin and placebo were performed.

The intent-to-treat (ITT) population included all patients who had a baseline IPSS assessment and at least one valid IPSS assessment after baseline. The per-protocol (PP) population included all patients who completed the study without any major protocol violation. The safety population included all patients who were randomised and who received at least one dose of either silodosin or placebo.

Differences from placebo for the mean change from baseline to the end of treatment for the IPSS and uroflowmetry data were tested using an analysis of covariance model with terms for treatment and protocol as main effects and baseline value as a covariate. Stepwise forward and backward multivariable regression models including age, body mass index, baseline IPSS total score, baseline Qmax, baseline PSA level and treatment were generated to test the independent predictors of treatment response. The numerical variables were tested either as continuous variables or categorised according to the median value. Results of all these analyses were calculated as adjusted means with 95% CI. A two-sided significance level of 5% was applied for all statistical tests. All the statistical analyses were performed with SAS v.9.3 (SAS Institute, Cary, NC, USA) software.

Results

Table 1 summarises the demographic and baseline characteristics of the 837 patients randomised to silodosin 8 mg and 647 randomised to placebo included in the ITT analysis. Demographic and baseline characteristics were similar in the two study groups.

Table 1. Demographic and baseline characteristics
CharacteristicSilodosin 8 mgPlacebo
(n = 847)(n = 647)
Age, years  
Mean (sd)64.9 (8)64.7 (7.9)
Median (range)65.0 (50–87)64.0 (44–86)
Body mass index, kg/m2  
Mean (sd)27.8 (4.1)27.9 (4.0)
Median (range)27.2 (18–54)27.4 (20–53)
Race, n (%)  
Caucasian806 (95.2)589 (91)
African American14 (1.7)22 (3.4)
Hispanic20 (2.4)25 (3.9)
Asian5 (0.6)6 (0.9)
Other2 (0.2)5 (0.8)
Country, n (%)  
USA466 (55.0)457 (70.6)
Europe197 (23.3)100 (15.5)
Other non-European184 (21.7)90 (13.9)
Concomitant antihypertensive medications, n (%)283 (33.4)203 (31.4)
Past or current history of erectile dysfunction, n (%)303 (35.8)254 (39.3)
Past or current history of ejaculatory disorders, n (%)171 (20.2)137 (21.2)
IPSS total score  
Mean (sd)20.3 (4.9)20.7 (4.8)
Median (range)20 (13–35)20 (11–35)
IPSS storage subscore  
Mean (sd)8.7 (2.7)8.9 (2.6)
Median (range)9 (1–15)9 (1–15)
IPSS voiding subscore  
Mean (sd)11.7 (3.4)11.8 (3.5)
Median (range)11 (1–20)12 (3–20)
IPSS quality-of-life score  
Mean (sd)3.9 (1.0)4.0 (1.1)
Median (range)4.0 (1–6)4.0 (0–6)
PSA level, ng/mL  
Mean (sd)2.1 (1.8)2.2 (2.0)
Median (range)1.5 (0–9)1.6 (0–20)
Qmax, mL/s  
Mean (sd)9.0 (2.6)8.9 (2.8)
Median (range)9.0 (3–15)8.7 (3–17)

Efficacy Results

Table 2 summarises the change from baseline in total IPSS and other IPSS-related parameters in the two study groups.

Table 2. Change in IPSS total score and other IPSS-related parameters from baseline to study end
 Silodosin 8 mgPlaceboP
Change in total IPSS from baseline to study end
ITT populationn = 837n = 642 
Adjusted meansn = 836n = 639 
–6.5–3.8
AMD (95% CI)–2.7 (–3.3 to –2.1)<0.001
PP populationn = 712n = 547 
Adjusted meansn = 712n = 547 
–6.6–3.9
AMD (95% CI)–2.7 (–3.3 to –2.1)<0.001
Treatment responders to IPSS (decrease of ≥25% vs baseline at study end)
ITT population, n (%)n = 836n = 639<0.001
496 (59.3)240 (37.4)
PP population, n (%)n = 712n = 547<0.001
439 (61.7)218 (39.9)
Change in IPSS storage subscore from baseline to study end
ITT populationn = 837n = 642 
Adjusted meansn = 836n = 641 
–2.3–1.4
AMD (95% CI)–0.9 (–1.1 to –0.6)<0.001
PP populationn = 712n = 547 
Adjusted meansn = 711n = 547 
–2.3–1.5
AMD (95% CI)–0.8 (–1.1 to –0.5)<0.001
Change in IPSS voiding subscore from baseline to study end
ITT populationn = 837n = 642 
Adjusted meansn = 836n = 641 
–4.1–2.3
AMD (95% CI)–1.8 (–2.2 to –1.4)<0.001
PP populationn = 712n = 547 
Adjusted meansn = 711n = 547 
–4.3–2.4
AMD (95% CI)–1.9 (–2.3 to –1.5)<0.001
Change in IPSS quality of life item score from baseline to study end
ITT populationn = 837n = 642 
Adjusted meansn = 836n = 636 
–0.9–0.5
AMD (95% CI)–0.4 (–0.5 to –0.3)<0.001
PP populationn = 712n = 547 
Adjusted meansn = 711n = 544 
–0.9–0.5
AMD (95% CI)–0.4 (–0.5 to –0.3)<0.001

At study end, in the ITT population, the adjusted mean decrease in IPSS total score was –6.5 in patients receiving silodosin and –3.8 in those randomised to placebo (adjusted means differences [AMD] –2.7; 95% CI –3.3 to –2.1; P < 0.001). Similarly, in the PP population, there were adjusted mean IPSS total score decreases of 6.6 and 3.9 in patients receiving silodosin and placebo, respectively (AMD –2.7; 95% CI –3.3 to –2.1; P < 0.001) (Fig. 1A).

Figure 1.

Change in IPSS total score and other IPSS-related parameters.

Considering IPSS total score responders as those achieving a decrease of ≥25% vs baseline, at study end, 496 (59.3%) of the patients receiving silodosin were responders, compared with 240 (37.4%) in the placebo group (P < 0.001) in the ITT population. Similarly, in the PP population, 439 (61.7%) and 218 (39.9%) of patients receiving silodosin and placebo were responders, respectively (P < 0.001; Fig. 1B). Treatment received was the only predictive variable (odds ratio [OR] 0.382; 95% CI 0.303–0.481; P < 0.001), once adjusted for all the other covariates, either reported as continuous or categorical variables.

Looking at the IPSS subscores in the ITT population, silodosin was significantly more effective than placebo in improving storage (AMD –0.9; 95% CI –1.1 to –0.6; P < 0.001), voiding (AMD –1.8; 95% CI –2.2 to –1.4; P < 0.001), and quality-of-life-item (AMD –0.4; 95% CI –0.5 to –0.3; P < 0.001) subscores (Fig. 1C). Virtually identical results were achieved in the PP population (Table 2).

For Qmax, the adjusted mean increase was 2.7 mL/s in patients receiving silodosin and 1.9 mL/s in those randomised to placebo (AMD 0.8; 95% CI 0.3–1.3; P = 0.002) in the ITT population. Similarly, in the PP population, there were adjusted mean Qmax improvements of 2.8 and 2.1 mL/s in patients receiving silodosin and placebo, respectively (AMD 0.7; 95% CI 0.2–1.3; P = 0.001; Fig. 1D).

Safety Results

Table 3 summarises the adverse drug reactions (ADRs) and discontinuation rates in the silodosin and placebo groups.

Table 3. Summary of patients with ADRs and discontinuation rates
 Silodosin 8 mg (n = 847)Placebo (n = 647)P
No. eventsNo. patients (%)No. eventsNo. patients (%)
Any ADR363246 (29)7247 (7.3)<0.001
Retrograde ejaculation214186 (22)66 (0.9)<0.001
Dizziness1616 (1.9)54 (0.6)0.029
Orthostatic hypotension1111 (1.3)97 (1.1)0.449
Headache109 (1.1)55 (0.8)0.661
Nasal congestion98 (0.9)11 (0.2)0.027
Erectile dysfunction76 (0.7)42 (0.3)0.465
Diarrhoea86 (0.7)11 (0.2)0.079
Rhinitis64 (0.5)22 (0.3)0.685
Dry mouth44 (0.5)32 (0.3)0.684
Loss of libido84 (0.5)21 (0.2)0.527
Vertigo33 (0.4)0.221
Insomnia33 (0.4)0.111
Discontinuation rate78 (9.2)56 (8.7)0.5358
Discontinuation due to treatment-emergent AEs (related and not related)36 (4.3)11 (1.7)<0.001

Overall, 363 ADRs were reported in 246 (29%) patients in the silodosin group and 72 in 47 patients (7.3%) in the placebo group (OR 5.36; 95% CI 3.84–7.48; P < 0.001).

The most frequently reported ADR was ejaculatory dysfunction, reported in 186 (22%) patients in the silodosin group and six (0.9%) in the placebo group (OR 28.14; 95% CI 12.34–64.17; P < 0.001). Dizziness (1.9% in silodosin group vs 0.6% in the placebo group) and orthostatic hypothension (1.3% in silodosin group vs 1.1% in the placebo group) were the second and third most prevalent ADRs but both had similar prevalence rates in the silodosin and placebo groups. Notably, among patients with and without ejaculatory dysfunction, the prevalence of those who achieved at least a 30% improvement in IPSS and Qmax was similar in patients with and without ejaculatory dysfunction (27% vs 25% in those without; P = 0.557).

The number of patients who discontinued the study was similar in the silodosin (78, 9.2%) and placebo (56, 8.7%) groups. Discontinuation rates due to ADRs were low in both study groups (4.3% in silodosin group vs 1.7% in the placebo group), with the difference being statistically significant (OR 0.39; 95% CI 0.2–0.77; P < 0.001).

Discussion

In the present pooled analysis of individual patient data from USA and European registrational trials of silodosin, we found that silodosin was significantly more effective than placebo in improving the total IPSS and all IPSS-related parameters (i.e., IPSS responders, storage and voiding IPSS subscores, quality-of-life-item score), as well as Qmax at uroflowmetry. On the whole, the drug was well tolerated and the most prevalent ADR was ejaculatory dysfunction. However, cardiovascular AEs were similarly prevalent with silodosin and placebo, and the discontinuation rate in the silodosin group was only 2% higher than with placebo.

α-Blockers, 5-ARIs, or their combination are the standard medical treatments for patients with uncomplicated bothersome, moderate-to-severe LUTS suggestive of BPH who are unresponsive to behavioural management [2, 3]. α-Blockers, used alone or in combined therapy, are the most commonly prescribed category of drug, accounting for ≈70% of all the prescriptions in 2008 [4].

RCTs comparing different α-blockers are limited. However, alfuzosin, doxazosin, tamsulosin, and terazosin are commonly considered equally effective but with different profiles of AEs, with non-selective drugs, such as doxazosin and terazosin, requiring blood pressure monitoring and being associated with higher cardiovascular AEs [3]. Conversely, ejaculatory dysfunction is more common with selective drugs, such as tamsulosin [4].

Silodosin is a novel α-blocker, with the highest selectivity for α1A-adrenoceptors. Specifically, Shibata et al. [15] reported that silodosin's affinity for the α1A-adrenoceptor was 583-fold that for the α1B subtype and 56-fold that for the α1D subtype, whereas tamsulosin, the other selective α-blocker available in the market, was 15- and 3-fold more selective for the α1A-adrenoceptors than for the α1B and α1D subtypes. Three registrational RCTs are currently available on silodosin. Specifically, Marks et al. [9] reported on the two USA trials, showing significant improvements from baseline to week 12 in total IPSS (AMD –2.8, P < 0.001), IPSS storage (AMD –1, P < 0.001), and voiding (AMD –1.9, P < 0.001) subscores, as well as Qmax improvements (AMD 1 mL/s, P < 0.001) in the patients receiving silodosin as compared with placebo. Similarly, in the European registrational study which, notably, included an active-control arm with tamsulosin, Chapple et al. [10] reported that silodosin was significantly more effective than placebo in all the efficacy endpoints and as effective as tamsulosin. However, post hoc analyses of the same study found a significant difference in favour of silodosin compared with tamsulosin for thee IPSS symptoms (incomplete emptying, frequency, nocturia) considered simultaneously in patients with either moderate or severe LUTS [16]. Notably, all the registrational RCTs showed a favourable profile of AEs, with retrograde ejaculation being the most common ADR (28% in the USA studies and 14% in the European one) [9, 10]. Recently, two meta-analyses of RCTs including also other Japanese phase III RCTs corroborated such findings [12, 13].

In the present paper, we analysed the clustered datasets of USA and European registrational trials to provide more insights on the efficacy and safety of silodosin. Briefly, we were able to show an ≈3-point improvement in total IPSS, with ≈60% of the patients reporting improvements of >25% as compared with baseline scores. Such IPSS reduction was mainly due to improvements in voiding LUTS (2-point improvement in IPSS voiding subscore) but also the storage subscore and quality-of-life-item score improved significantly over placebo. Notably, such improvements were at the cost of relatively low ADR rates. Specifically, only ejaculatory dysfunction was significantly more prevalent with silodosin than with placebo. Conversely, all the other ADRs were rare and no more prevalent than with placebo.

The data of the present paper are important for several reasons. Although comparisons with other α-blockers are sparse, the data reconfirmed that silodosin was at least as effective as the other drugs in the category. Clinically speaking, the high silodosin affinity for the α1A-adrenoceptor translated into a minimal prevalence of orthostatic hypotension, dizziness, and cardiovascular AEs, with rates similar to those of the placebo arm. Such data are more relevant if we consider that about a third of the patients enrolled in the USA and European trials were taking antihypertensive medications [9, 10]. Moreover, modification of systolic blood pressure, diastolic blood pressure, and heart rate, were similar with silodosin and placebo [10]. Conversely, in a meta-analysis on the vascular-related safety of α-blockers, the odds of developing a cardiovascular-related AE were reported to be 1.66 for alfuzosin, 3.71 for terazosin, 3.32 for doxazosin, and 1.42 for tamsulosin as compared with placebo [17]. Consequently, silodosin emerges as a valuable drug for all patients needing an α-blocker for LUTS when preventing cardiovascular-related ADRs may have a major relevance (i.e., patients taking antihypertensive drugs, patients taking drugs with other cardiovascular-related AEs, patients with a history of other cardiovascular events, and elderly or frail patients). Such advantages in cardiovascular ADRs come at the cost of a higher prevalence of ejaculatory dysfunction. Ejaculatory dysfunction includes a broad spectrum of conditions ranging from absence of seminal emission, reduced ejaculate volume, and reduced ejaculation force. However, it has been suggested that patients with ejaculatory dysfunction are those with larger improvements in LUTS and Qmax, as compared with those without ejaculatory dysfunction [18, 19]. That data may explain the very low discontinuation rates seen in our present analysis. However, in the present study we demonstrated only a statistically non-significant difference, probably due to the lower prevalence of ejaculatory dysfunction within the European registrational trial, where the enrolled patients were older than those included in the USA studies. Moreover, it is also possible that ejaculatory dysfunction may have an inconsistent impact on patients’ sexual functions or that many patients with LUTS may have a limited interest in sexual activities, due to comorbidities or concomitant erectile dysfunction [20]. It is clear that all patients, and especially all those who are younger or healthier or more interested in sexual activities or without concomitant erectile dysfunction, should be counselled on the risk of ejaculatory dysfunction present with α-blocker and with other drug therapies for male LUTS.

The present study is not devoid of limitations. All the patients were treated for 12 weeks, which prevent us from drawing any conclusions on the long-term efficacy and safety of silodosin. However, the published open-label study demonstrated sustained efficacy over a 9-month treatment and no ADRs other than those reported in the RCTs evaluated in the present meta-analysis [21]. Further, long-term data are still lacking, although it will surely be produced in phase IV studies. Moreover, IPSS storage and voiding subscores were not validated, although they are extensively used in several RCTs on LUTS treatment.

In conclusion, silodosin is an effective treatment for male LUTS suggestive of BPH. The drug is able to improve total IPSS, all IPSS-related parameters, and Qmax at uroflowmetry. Ejaculatory dysfunction is the main treatment-related AE. The drug is well tolerated, with prevalence of cardiovascular ADRs as low as placebo.

Conflicts of interest

G.N. has been an advisory board member or speaker for Astellas, GlaxoSmithKleine, Lilly, Menarini, Nycomed, Pfizer Inc., Pierre Fabre, and Recordati.

C.R.C. is a consultant to AMS and Lilly, Consultant and Researcher to ONO; Consultant, Researcher and Speaker to Allergan, Astellas, Pfizer and Recordati.

F.M. is a consultant to Recordati, Eli Lilly, GSK, Astellas, Pierre Fabre, Takeda.

Abbreviations
ADR

adverse drug reaction

AE

adverse event

AMD

adjusted means differences

5-ARI

5α-reductase inhibitor

ITT

intent-to-treat

PP

per protocol

Qmax

maximum urinary flow rate

RCT

randomised controlled trial

Ancillary