Evidence review

This evidence review is based on 2 studies (Roehrborn et al. 2008 and Donatucci et al. 2011) of tadalafil to treat men with lower urinary tract symptoms (LUTS) secondary to benign prostatic hyperplasia.

Roehrborn et al. (2008) published results from a 12-week, double-blind, placebo-controlled, dose-finding study (2.5 mg, 5 mg, 10 mg and 20 mg daily). After the 12-week, placebo-controlled treatment period, men were given the choice to participate in an open-label extension study (published by Donatucci et al. 2011) for a period of 1 year, in which all men received tadalafil 5 mg daily.

In addition, 1 systematic review (Gacci et al. 2012) was identified that looked at the use of phosphodiesterase type 5 (PDE5) inhibitors either alone (7 randomised controlled trials [RCTs], 3214 men) or in combination with alpha blockers (5 RCTs, 216 men) for the treatment of LUTS due to benign prostatic hyperplasia. This combined the results of 4 different PDE5 inhibitors, including but not limited to tadalafil, but did not report efficacy outcomes for tadalafil separately.

Other relevant studies of tadalafil have been published in men with LUTS suggestive of benign prostatic hyperplasia (Porst et al. 2011 and Oelke et al. 2012), and in men with both erectile dysfunction and LUTS of benign prostatic hyperplasia (Egerdie et al. 2012). These provide relevant evidence but contained a smaller number of participants randomised to receive tadalafil 5 mg. In addition, the study by Egerdie et al. (2012) focused on treating men with both erectile dysfunction and LUTS of benign prostatic hyperplasia, whereas this evidence summary focuses solely on the treatment of LUTS. The largest study identified using tadalafil 5 mg (Roehrborn et al. 2008) and the linked longer-term, open-label extension study (Donatucci et al. 2011) provide the evidence for this summary.

Roehrborn et al. (2008)

  • Design: multicentre, double-blind, randomised, placebo-controlled, dose-finding trial with a 12-week treatment period.

  • Population: 1058 men, 45 years or older (mean age 62.1 years) with a history of LUTS secondary to benign prostatic hyperplasia of 6 months or longer. Men had to have an International Prostate Symptoms Score (I-PSS) of 13 or more to be included in the study. The I-PSS has 7 symptom questions addressing incomplete bladder emptying, frequency, intermittency, urgency, weak stream, straining and nocturia during the past month. Each question is scored from 0 to 5 with a total score of up to 35 points. Scores range from 0 (asymptomatic) to 35 (very symptomatic) with the following categories of symptom severity: 1–7 mild, 8–19 moderate and 20–35 severe. The I-PSS contains 1 additional question on quality of life due to urinary symptoms, and is assigned a score of 0 to 6, with lower scores indicating a better quality of life. I-PSS total score is a validated tool but it is not clear whether subscores based on clusters of individual questions are also valid.

  • Previous treatments: men who received finasteride (within 3 months) or dutasteride (within 12 months) before the placebo run-in period were excluded from the trial. Alpha blockers had been used by 28.8% of men in the trial, and 1.1% had previously used 5-alpha-reductase inhibitors.

  • Intervention and comparison: 4-week, treatment-free, wash-out period, followed by a 4-week placebo run-in period, and then a 12-week treatment period with: tadalafil 2.5 mg, 5 mg, 10 mg or 20 mg daily or placebo daily.

  • Primary outcome: change in I-PSS from baseline to 12 weeks with tadalafil 5 mg compared with placebo using an intention-to-treat analysis.

  • Secondary outcomes included efficacy and dose-response relationship of different doses of tadalafil (assessed using I-PPS total score) and change from baseline to 12-week end point in I-PSS irritative subscore, voiding (obstructive) subscore, I-PSS quality of life, benign prostatic hyperplasia impact index (BPH−II), global assessment question (GAQ), uroflowmetry parameters, and international index of erectile function – erectile function domain (IIEF-EF).

Table 1 Summary of the trial: Roehrborn et al. (2008)

Tadalafil 5 mg daily

Placebo daily

Analysis

Randomised

n=212

n=211

ITT

n=212

n=210

Baseline and at least 1 post-baseline measurement available; the excluded patient in the placebo group had been randomised at 2 sites

Completed 12-week treatment

85.9% (182/212)

87.3% (185/212)

Baseline

I-PSS (±SD)

17.30 (±5.97)

17.08 (±6.36)

No significant difference

Efficacy

n=212

n=210

Primary outcome:

Change in I-PSSa, least squares mean (±SE)

−4.87 (±0.49)

−2.27 (±0.49)

p<0.001, absolute difference 2.6 points in favour of tadalafil

Selected secondary outcomes

p values below are for tadalafil (any dose) vs. placebo

Change in I-PSSa in patients with mild to moderate LUTS at baselineb, least squares mean (±SE)

−4.3 (±4.94)

−1.4 (±6.44)

p<0.001, absolute difference 2.9 points for tadalafil 5 mg vs. placebo

Change in I-PSSa in patients with severeLUTS at baselinec, least squares mean (±SE)

−6.2 (±6.71)

−3.9 (±5.3)

p<0.05, absolute difference 2.3 points for tadalafil 5 mg vs. placebo

Change in I-PSS irritative subscorea, least squares mean (±SE)

−1.89 (±0.23)

−0.99 (±0.23)

p<0.01, absolute difference 0.9 points for tadalafil 5 mg vs. placebo

Change in I-PSS obstructive subscorea, least squares mean (±SE)

−2.94 (±0.33)

−1.26 (±0.33)

p<0.001, absolute difference 1.68 points for tadalafil 5 mg vs. placebo

Change in I-PSS quality of lifea, least squares mean (±SE)

−0.86 (±0.11)

−0.49 (±0.11)

p<0.01, absolute difference 0.37 points for tadalafil 5 mg vs. placebo

Change in maximum urinary flow ratea, Qmax, cm/sec, least squares mean (±SE)

1.64 (±0.39)

1.24 (±0.40)

No significant change from baseline compared with placebo (p value not reported)

Safety

n=212

n=211

Patients reporting 1 or more SAE

0.5% (1/212)

2.8% (6/211)

No statistical test reported

Patients reporting 1 or more AE leading to discontinuation

5.7% (12/212)

2.4% (5/211)

No statistical test reported

Patients reporting 1 or more TEAE

30.7% (65/212)

21.2% (45/211)

No statistical test reported

Notes: The I-PSS irritative subscore sums the scores for: frequency, urgency and nocturia questions. The obstructive subscore sums the scores for: straining, weak stream, incomplete bladder emptying and intermittency questions.

Abbreviations: AE, adverse event; I-PSS, International Prostate Symptoms Score; ITT, intention to treat; LUTS, lower urinary tract symptoms; SAE, serious adverse event; SD, standard deviation; SE, standard error; TEAE, treatment-emergent adverse events.

a Change from baseline to 12 weeks.

b n=701, patients had mild to moderate I-PSS scores at baseline across all treatment groups (placebo, 2.5 mg, 5 mg, 10 mg and 20 mg). Patient numbers by treatment group not given.

c n=352, patients had severe I-PSS scores across all treatment groups (placebo, 2.5 mg, 5 mg, 10 mg and 20 mg). Patient numbers by treatment group not given.


Donatucci et al. (2011)

  • Design: 1-year, multicentre, open-label extension study. Safety and efficacy parameters assessed after 1 month and every 3 months thereafter.

  • Population: 427 men (from a total of 886 eligible men) who had completed the 12-week, placebo-controlled, dose-finding study (Roehrborn et al. 2008) and elected to take part in the open-label extension phase. Mean age 62.66 years (range 45.57 to 82.20 years). Around a third (32.79%) had previously used alpha blockers, whereas 1.41% had previously used 5-alpha-reductase inhibitors.

  • Intervention and comparison: all men used tadalafil 5 mg daily for the duration of the extension study regardless of treatment allocation in the 12-week trial (Roehrborn et al. 2008).

  • Concurrent treatments: men agreed not to use any other medical or herbal treatments to treat benign prostate hyperplasia or erectile dysfunction while enrolled in the study.

  • Primary outcome was not specified. Selected safety outcomes included: adverse events, changes in vital signs (blood pressure and pulse), clinical laboratory tests (serum chemistry and haematology, urinalysis), prostate specific antigen levels, post-void residual volume and electrocardiograms. Selected efficacy outcomes included: change in I-PSS score, I-PSS subscores, I-PSS quality of life and BPH-II.

Table 2 Summary of the trial: Donatucci et al. (2011)

Tadalafil 5 mg daily a

Analysis

Completed 12-week placebo study

n=886

Recruited to open-label extension

n=427

48.2% (427/886) of those competing the 12-week trial

Completed year follow-up

70.0% (299/427)

Secondary efficacy outcomes

No statistical tests reported

Mean change in I-PSS (±SD) week 0 to end pointb

−5.0 (±6.7)

Scores for 416 participants analysed

Mean change in I-PSS (±SD) week 12 to end pointb

−0.9 (±5.7)

Scores for 416 participants analysed

Mean change in I-PSS irritative subscore (±SD) week 0 to end pointb

−1.9 (±3.0)

Scores for 416 participants analysed

Mean change in I-PSS irritative subscore (±SD) week 12 to end pointb

−0.3 (±2.6)

Scores for 416 participants analysed

Mean change in I-PSS obstructive subscore (±SD) week 0 to end pointb

−3.1 (±4.4)

Scores for 416 participants analysed

Mean change in I-PSS obstructive subscore (±SD) week 12 to end pointb

−0.6 (±3.6)

Scores for 416 participants analysed

Mean change in I-PSS health-related quality of life (±SD) week 0 to end pointb

−1.1 (±1.4)

Scores for 414 participants analysed

Mean change in I-PSS health-related quality of life (±SD) week 12 to end pointb

−0.3 (±1.2)

Scores for 413 participants analysed

Change in maximum urinary flow rate, Qmax, cm/sec (±SE)

Not reported

Not reported

Safety

Patients reporting 1 or more SAE

4.7% (20/427)

No statistical tests reported

Patients reporting 1 or more AE leading to discontinuation

5.2% (22/427)

No statistical tests reported

Patients reporting 1 or more TEAE

57.6% (246/427)

No statistical tests reported

Notes: The I-PSS irritative subscore sums the scores for: frequency, urgency and nocturia. The obstructive subscore sums the scores for: straining, weak stream, incomplete bladder emptying and intermittency.

Abbreviations: AE, adverse event; I-PSS, International Prostate Symptoms Score; SAE, serious adverse events; SD, standard deviation; SE, standard error; TEAE, treatment-emergent adverse events.

a Tadalafil 5 mg daily from week 12 to week 64. Treatment before this (0–12 weeks) was either placebo or tadalafil at 2.5 mg, 5 mg, 10 mg or 20 mg daily (see Roehrborn et al. 2008).

b End point, the last non-missing post-baseline value after visit 7 (week 12).

Clinical effectiveness

In Roehrborn et al. (2008), 5 mg daily dosing of tadalafil statistically significantly improved LUTS compared with placebo. The least squares mean improvement in overall I-PSS was 2.27 points for placebo compared with 4.87 points for tadalafil after 12 weeks of treatment (p=0.001). A clinically meaningful improvement in score is defined in the study as a 3-point improvement in I−PSS from baseline. The improvement seen in the study is therefore clinically meaningful with tadalafil but not with placebo. However, the absolute difference between the groups is only 2.6 points, which is not a clinically meaningful difference. Statistically significant improvements in the primary outcome (tadalafil 5 mg compared with placebo) were detected at 4-, 8- and 12-week end points. The statistically significant benefit of tadalafil 5 mg over placebo in improving overall I-PSS at 12 weeks was maintained when the analysis was restricted to men with mild to moderate LUTS or severe LUTS at baseline.

Other statistically significant improvements seen for secondary outcomes at 12 weeks were: I-PSS obstructive subscore, I-PSS irritative subscore, I-PSS quality of life, BPH-II, GAQ and IIEF-EF for doses of 5 mg of tadalafil or greater (except for a non-significant finding for the 10 mg dose for BPH-II compared with placebo, p=0.056). No statistically significant improvement was seen for Qmax (peak urinary flow rate) using tadalafil 5 mg compared with placebo (p value not reported).

At baseline assessment, an average of 67.92% of men randomised to receive tadalafil 5 mg had a history of erectile dysfunction. This was numerically similar to the proportions reported in the other treatment arms of the trial. Therefore, the evidence is applicable largely to men with LUTS and erectile dysfunction, but also a proportion of men with LUTS without erectile dysfunction.

The results from Donatucci et al. (2011) indicated that most of the benefit in LUTS associated with tadalafil came within the first 12 weeks of treatment and was maintained or improved to a much smaller degree for up to 1 year. For example, the overall change in total I-PSS in men using tadalafil 5 mg from week 0 to the endpoint (the last non missing post baseline value after visit 7 [week 12]) was a mean of −5.0 points (standard deviation [SD] ± 7.2 points) with a reduction in score indicating an improvement. Between week 12 and the endpoint, scores changed by a mean of +0.2 points (SD ± 5.4 points). Considering all participants (those receiving placebo, or tadalafil 2.5 mg, 5 mg, 10 mg or 20 mg, during weeks 0 to 12), I-PSS scores changed by a mean of −5.0 points (SD ± 6.7 points) between week 0 and the endpoint, with a mean change of only −0.9 points (SD ± 5.7 points) seen between week 12 and the endpoint. A similar pattern was observed for the secondary outcomes I-PSS irritative and obstructive subscores.

The meta-analysis by Gacci et al. (2012) found that using PDE5 inhibitors alone (including but not limited to tadalafil, the only PDE5 inhibitor licensed for treating benign prostatic hyperplasia with LUTS) improved LUTS (mean change in I-PSS −2.8 points from baseline to end of study, p<0.0001) and erectile function (mean change in International Index of Erectile Function [IIEF] score from baseline to end of study +5.5, p<0.0001) compared with placebo but not maximum urinary flow rate. This meta-analysis compared the results of 2250 men (83.5%, 1879 completed the studies) using PDE5 inhibitors alone, with 964 men (90.2%, 870 completed the studies) using placebo. When used in combination with alpha blockers (109 men, 103 [94.5%] completed the studies), PDE5 inhibitors improved LUTS (I-PSS −1.8, p=0.05), erectile dysfunction (IIEF +3.6, p<0.0001) and maximum urinary flow rate (Qmax +1.5 ml/s, p<0.0001) compared with alpha blockers alone (107 men, 99 [92.5%] completed the studies).

Safety

Statistical analysis of safety data from Roehrborn et al. (2008) and the year-long extension study by Donatucci et al. (2011) was absent, limiting the conclusion that can be drawn.

In Roehrborn et al. (2008), relatively few men using tadalafil 5 mg daily reported 1 or more severe adverse events (1 out of 212 men, 0.5%) and adverse events leading to discontinuation (12 out of 212 men, 5.7%), whereas 65 out of 212 men (30.7%) reported 1 or more treatment-emergent adverse events of any severity.

During the 1-year open-label extension study (assessing 427 men), 20 men (4.7%) reported severe adverse events, 22 men (5.2%) reported adverse events leading to discontinuation, and 246 men (57.6%) reported 1 or more treatment-emergent adverse events of any severity.

The most commonly reported treatment-emergent adverse events were: dyspepsia (4%), gastro-oesophageal reflux disease (4.0%), back pain (3.7%), headache (3.0%), sinusitis (2.8%), hypertension (2.6%), and cough (2.1%).

Of the 246 men (57.6%) in the extension study reporting at least 1 treatment-emergent adverse event, most were classed as mild (44%) or moderate (45%) in severity.

Of the 20 men (4.7%) who reported at least 1 serious adverse event, 2 were considered to have serious adverse events possibly related to tadalafil: 1 man had worsening of coronary heart disease and 1 experienced global amnesia.

A meta-analysis of adverse events included in Gacci et al. (2012) reported a higher risk of flushing, gastro-oesophageal reflux, headache and dyspepsia after using PDE5 inhibitors alone (including but not limited to tadalafil) compared with placebo.

The full list of cautions, contraindications and side effects derived from all the clinical trials available when tadalafil was licensed are described in the summary of product characteristics. Of note, tadalafil is contraindicated in patients receiving nitrates; in patients with cardiac disease for whom sexual activity is inadvisable; and in patients with non-arteritic ischaemic optic neuropathy leading to loss of vision, uncontrolled arrhythmias, hypotension, uncontrolled hypertension, recent stroke, unstable angina, recent myocardial infarction or recent New York Heart Association class 2 or greater heart failure.

Evidence strengths and limitations

Roehrborn et al. (2008) had relatively high completion rates (85.9% in those using tadalafil 5 mg compared with 87.3% using placebo) after 12 weeks of treatment and was well powered to detect differences between tadalafil 5 mg and placebo treatments, suggesting that the results are reliable. A further strength is that longer-term safety and efficacy results (up to a year) were also published for those using tadalafil 5 mg daily (Donatucci et al. 2011); though the completion rates for this study were lower (69.9%).

I-PSS total score is a validated tool for assessing LUTS but it is not clear whether the subscores used in the studies (based on clusters of individual questions) are equally valid. Therefore, interpretation of these subscores should be more cautious.

Alpha blockers had been used previously in only 27.83% of men assigned tadalafil 5 mg in the 12-week, double-blind study by Roehrborn et al. (2008); therefore, the evidence is mainly applicable to tadalafil used as a first-line treatment for LUTS associated with benign prostatic hyperplasia. It is not clear whether tadalafil would lead to equivalent significant improvements in LUTS symptoms compared with placebo if the results were restricted only to those who had already used alpha blockers (that is, if tadalafil was used as a second-line treatment). Only 48.2% of eligible men self-selected to participate in the extension study after the initial 12-week treatment. This may have led to selection bias in the extension trial by including only men most responsive and/or tolerant to the drug during the initial 12-week trial. If this was the case, it could lead to an overestimate in the year-long efficacy of tadalafil and an underestimate in the prevalence of adverse events.

The lack of statistical analysis limits the interpretation of the safety and tolerability data from both studies.

Owing to the scope of this evidence summary, the summary focused on the results of only 2 study publications (Roehrborn et al. 2008 and Donatucci et al. 2011). Roehrborn et al. (2008) and the linked longer-term open-label extension study, Donatucci et al. (2011), were selected because together, they have reported outcomes from 1058 men initially randomised in 92 centres across 10 countries. As such, Roehrborn et al. (2008) was identified as the RCT with the largest group of men receiving tadalafil 5 mg (the dose of interest), and the open-label extension by Donatucci et al. (2011) provided the longest follow-up (1 year). Additional published studies also contribute to the evidence base on efficacy or safety, and were reviewed by the European Medicines Agency before issuing a marketing authorisation. These were not reviewed in this evidence summary.