To assess the safety and efficacy of tadalafil once daily on lower urinary tract symptoms suggestive of clinical benign prostatic hyperplasia (BPH-LUTS) in men without erectile dysfunction (ED).
To compare these with effects in men with ED.
Patients and Methods
After a 4-week washout period and 4-week placebo run-in period, 1089 men without ED (n = 338) and with ED (n = 751) were randomly assigned to placebo or tadalafil 5 mg once daily for 12 weeks in three global clinical studies with similar designs.
In the pooled dataset, post hoc analyses of covariance assessed the impact and severity of BPH-LUTS using the International Prostate Symptom Score (IPSS) and the BPH Impact Index (BII) and IPSS quality-of-life (IPSS-QoL) subscores.
Safety was assessed using treatment-emergent adverse events.
The treatment-by-ED-status interaction was used to assess efficacy differences between the with/without ED subgroups.
Men without ED were similar in BPH-LUTS severity/previous therapy to men with ED.
Tadalafil significantly reduced BPH-LUTS from baseline when compared with placebo in men without ED (IPSS −5.4 vs −3.3, P < 0.01; IPSS voiding subscore −3.5 vs −2.0, P < 0.01; IPSS storage subscore −1.9 vs −1.3, P < 0.05).
Tadalafil also significantly improved quality of life from baseline when compared with placebo in men without ED (IPSS-QoL −1.0 vs −0.7, BII −1.4 vs −1.0; both P < 0.05).
Between-ED-subgroup interactions were not significant (all P > 0.68).
Tadalafil was safe and well tolerated.
Tadalafil 5 mg once daily improved BPH-LUTS in men without ED by a magnitude similar to that observed in men with ED.
The adverse event profile in men without ED was consistent with that observed in men with ED.
Lower urinary tract symptoms suggestive of BPH (BPH-LUTS), are increasingly common in aging men [1-3]. While the majority will experience coexisting erectile dysfunction (ED) , many will have BPH-LUTS without ED . Irrespective of coexisting ED, men with BPH-LUTS often report a diminished overall quality of life .
Phosphodiesterase type 5 (PDE5) inhibitors, especially tadalafil, have been investigated for the treatment of BPH-LUTS and ED coexisting with BPH-LUTS. Tadalafil 5 mg once daily was approved in 2011 in the USA and in 2012 in the European Union for the treatment of BPH-LUTS and ED with BPH-LUTS and has achieved regulatory approval in other countries for the treatment of men with one or both of these conditions. Treatment with tadalafil may provide an alternative to current recommendations such as α1-adrenergic blockers (α-blockers) as monotherapy for the management of bothersome BPH-LUTS or in combination with 5α-reductase inhibitors (5-ARIs) . Commonly reported adverse events (AEs) associated with α-blockers include orthostatic hypotension, dizziness, asthenia, nasopharyngitis and abnormal ejaculation [7, 8]. Ejaculatory dysfunction, characterized by Hellstrom and Sikka  as decreased ejaculate volume and anejaculation, occurs in 10 and 30% of men taking tamsulosin and silodosin, respectively [10, 11]. Common AEs associated with 5-ARIs include decreased libido, decreased ejaculate volume and ED [7, 8, 12, 13]. Reports of persistent sexual dysfunction (including libido, ejaculation and orgasm disorders) after drug discontinuation or male infertility and/or poor semen quality that normalized or improved after drug discontinuation resulted recently in a revision of the finasteride US package insert [14, 15].
While PDE5-inhibitors successfully improve ED, the approval of tadalafil once daily for the treatment of BPH-LUTS supports the additional effects of tadalafil on other key structures of the lower urinary tract. Several studies have confirmed the efficacy and tolerability of tadalafil in the treatment of men with BPH-LUTS, irrespective of ED [16-19]; however, sample sizes in individual studies in men with BPH-LUTS but without ED have been too small to fully understand the impact of tadalafil in this subpopulation. Thus, the focus of this integrated analysis of international studies in men with BPH-LUTS [17-19] was to further evaluate the efficacy and safety of tadalafil 5 mg once daily vs placebo on BPH-LUTS in a large sample of men without ED and to compare these effects with those found in men with ED.
Materials and Methods
Study Design and Participants
The present integrated analysis comprised three randomized, double-blind, placebo-controlled, 12-week studies with similar inclusion and exclusion criteria performed at centres in Argentina, Australia, Austria, Belgium, France, Canada, Germany, Greece, Italy, Mexico, the Netherlands, Poland, Spain, Sweden and the USA. Men were eligible to participate if they were ≥45 years of age with a history of LUTS secondary to BPH for >6 months, an IPSS ≥13, and maximum urinary flow rate (Qmax) 4–15 mL/s. ED reflected a clinical diagnosis by the investigator after a physical examination and a functional inquiry confirming a consistent change in the quality of erection adversely affecting satisfaction with sexual intercourse, but this was not a required inclusion criterion. Exclusion criteria included PSA >10.0 ng/mL (or PSA ≥4.0 to ≤10.0 ng/mL if malignancy had not been excluded), postvoid residual urine volume ≥300 mL at screening, and the use of finasteride within the previous 3 months or dutasteride within the previous 6 months or 12 months, respectively. Additional exclusion criteria were as described previously [17-19].
A 4-week washout period preceded a 4-week placebo lead-in period before the baseline visit. At baseline, patients were randomly assigned either to tadalafil 5 mg once daily or placebo for a 12-week treatment period. Additional study design details were as reported previously .
The studies were conducted in accordance with the Declaration of Helsinki and were approved by the institutional review boards at each site. All participants provided written informed consent before undergoing any study procedure or receiving study therapy.
The primary efficacy measure of BPH-LUTS severity was assessed by the mean change in total IPSS from baseline to endpoint (12 weeks or last measurement). The IPSS is a validated, seven-part, self-administered questionnaire. Higher scores are indicative of greater symptom severity.
Secondary efficacy measures included IPSS obstructive (voiding) and irritative (storage) subscores, the BPH Impact Index (BII), and the IPSS quality-of-life (IPSS-QoL) subscore. Safety measures included patient-reported treatment-emergent AEs (TEAEs). Other efficacy and safety parameters have been reported previously in the individual studies [18-20].
The efficacy analysis population included all randomized patients who started double-blind study medication with either tadalafil 5 mg or placebo once daily. The demographic and safety summaries were based on randomized patients.
We used ancova to assess the least squares (LS) mean change in total IPSS from baseline to endpoint (12 weeks or last non-missing measurement). Model terms were included for baseline, treatment, region, protocol, ED history indicator and treatment-by-ED-status interaction. Baseline-by-treatment and treatment-by-region interactions were included if P < 0.10.
Similar to the analysis for total IPSS, secondary analyses were conducted using ancova to evaluate LS mean change from baseline to endpoint (12 weeks or last non-missing measurement) for tadalafil 5 mg compared with placebo for IPSS voiding and storage, BII and IPSS-QoL subscores.
To determine whether treatment response patterns across therapies (tadalafil or placebo) were different for men with or without ED, a treatment-by-ED-status interaction was assessed in the ancova model described above. P values for treatment-by-ED-status interaction were reported regardless of their statistical significance. Insignificant interaction P values were indicative of similar treatment effects across the two ED status subgroups.
A total of 1092 men entered the studies and were randomly assigned to treatment groups. Of these, 1089 reported their ED status: 338/1089 (31%) reported no history of ED, and 751/1089 (69%) reported a history of ED. Of the 1089 patients, 971 (89%) completed the study (placebo, n = 485; tadalafil, n = 486).
The sample of men without ED was similar to the sample of men with ED in age, body mass index, PSA level, sexual activity and previous α-blocker therapy (Table 1). Previous PDE5 therapy was reported by 1% of men without ED (3/338) and 30% of men with ED (224/751). Men without ED reported a lower incidence of comorbid conditions than men with ED (Table 1): diabetes mellitus (7% [23/338 patients] vs 15% [115/751 patients]), cardiovascular disease (41% [139/338 patients] vs 50% [373/751 patients]), and hypertension (35% [119/338 patients] vs 43% [326/751 patients]).
Table 1. Demographics and baseline characteristics for all randomly assigned patients without and with ED
Patients without ED
Patients with ED
Placebo, N = 171
Tadalafil 5 mg, N = 167
Total, N = 338
Placebo, N = 374
Tadalafil 5 mg, N = 377
Total, N = 751
*Sexually active patients are those who were sexually active with an adult female partner and who expected to remain sexually active with the same female partner for the duration of the study. n = number of patients with nonmissing data at baseline and at least one postbaseline visit; percentages are based on the number of patients with non-missing data in each subgroup (with or without ED).
Mean (sd) age, years
Mean (sd) BMI, kg/m2
Mean (sd) PSA, ng/mL
Mean (sd) total IPSS, before placebo run-in = screening
In men without ED, treatment with tadalafil 5 mg resulted in a significant improvement in total IPSS mean change from baseline (–5.4) compared with placebo (–3.3; treatment difference, –2.1, P < 0.01 [Table 2; Fig. 1]). Significant improvements in changes from baseline were observed compared with placebo in IPSS subscores (voiding, –3.5 vs –2.0, P < 0.01; storage, –1.9 vs –1.3, P < 0.05, respectively) and quality of life (BII, –1.4 vs –1.0; IPSS-QoL, –1.0 vs –0.7; both P < 0.05 [Table 2; Figs 1, 2]).
Table 2. IPSS change, BPH and quality of life from baseline to endpoint by ED status
IPSS score type
Patients without ED
Patients with ED
Tadalafil 5 mg
Tadalafil 5 mg
LS Mean (se)
LS Mean (se)
Base, baseline; End, endpoint; Δ, change; n, number of subjects with non-missing data at baseline and at least one post-baseline visit.
a–eTreatment-by-ED-status interaction: aP = 0.731; bP = 0.687; cP = 0.781; dP = 0.809; eP = 0.886.
Individual treatment group n values do not add up to subgroup N values because of missing IPSS data for several patients.
In men with ED, treatment with tadalafil 5 mg resulted in a significant improvement in total IPSS mean change from baseline to endpoint (–5.7) compared with placebo (–3.3; treatment difference, –2.3, P < 0.001 [Table 2; Fig. 1]). Significant improvements in changes from baseline were observed compared with placebo in IPSS subscores (voiding, –3.5 vs –1.9; storage, –2.2 vs –1.3, both P < 0.001) and quality of life (BII, –1.6 vs –0.9; IPSS-QoL, –1.1 vs –0.7; both P < 0.001 [Table 2; Figs 1, 2]).
Improvement in BPH-LUTS was similar between men without ED and with ED, as P values for treatment-by-ED-status interactions were insignificant for total IPSS (P = 0.73), IPSS voiding subscore (P = 0.69), IPSS storage subscore (P = 0.78), BII (P = 0.81), and IPSS-QoL (P = 0.89).
Common TEAEs reported by men without ED and with ED were similar (Table 3). Of 338 men without ED, 91 (26.9%) reported at least one TEAE; 53 (15.7%) were treated with tadalafil and 38 (11.2%) with placebo. Headache (n = 7), dyspepsia (n = 6), and nasopharyngitis (n = 6) were the most commonly reported TEAEs in men without ED treated with tadalafil; and nasopharyngitis (n = 4) and back pain (n = 3) were the most commonly reported with placebo. Of the 751 men with ED, 174 (23.2%), reported at least one TEAE; 96 (12.8%) were treated with tadalafil and 78 (10.4%) with placebo. Headache (n = 10), dyspepsia (n = 9), and hypertension (n = 9) were the most commonly reported TEAEs in men with ED treated with tadalafil; and nasopharyngitis (n = 9) and headache (n = 7) were the most commonly reported with placebo.
Table 3. Safety and tolerability of tadalafil 5 mg once daily, by ED status
Patients without ED
Patients with ED
Placebo, N = 171
Tadalafil 5 mg, N = 167
Total, N = 338
Placebo, N = 374
Tadalafil 5 mg, N = 377
Total, N = 751
n, number of patients with non-missing data at baseline and at least one post-baseline visit. Note: TEAEs ≥1% in either 5-mg tadalafil group. MedDRA Version 13.1.
Patients with ≥1 TEAE
Pain in extremity
Gastroesophageal reflux disease
In men without ED, tadalafil 5 mg once daily significantly reduced BPH-LUTS and improved quality of life; these changes were similar to those observed in men with ED. The AE profile in men without ED was consistent with that of men with ED treated with tadalafil. The results from this integrated analysis of data from three global clinical trials provide further evidence of tadalafil as an efficacious and safe treatment option, not only in men with coexisting ED and BPH-LUTS, but also in men without ED.
Men with and without ED were well balanced in baseline BPH severity; ∼ 60% had moderate BPH severity in each group. Not surprisingly, men with ED more often had pre-existing diabetes mellitus and cardiovascular diseases, including hypertension, than men without ED. In the present analysis, there was no difference in reported TEAEs between the two patient samples. As currently available treatments for BPH-LUTS may lead to AEs associated with hypotension and sexual dysfunction [7, 8, 12, 13] or potentially could negatively impact underlying ED, tadalafil provides an alternative risk-to-benefit profile in men with BPH-LUTS, with and without ED, and would present the only single medication approved to treat men with both conditions.
The impact of increasing doses was assessed post hoc in a previous dose-finding study in men without ED and men with ED. Tadalafil (2.5, 5, 10 and 20 mg) improved IPSS in both groups of men and to similar magnitudes per insignificant interaction analyses . After 12 weeks of treatment with tadalafil 5 mg once daily, mean improvements in IPSS in men without and with ED were −5.3 and −4.8, respectively. BII and IPSS-QoL scores also improved in both groups of men and by similar magnitudes; however, a dose-dependent tendency was not observed in the smaller samples of men without ED . As the study assessed dose-ranging data, sample sizes were small, ranging from 64 to 74 men without ED per treatment arm, thus providing insufficient statistical power to demonstrate significant improvement from baseline. To achieve appropriate statistical power for baseline-to-endpoint assessment, the present integrated analysis combined data from three trials. Improvement in total IPSS in this analysis was similar in men without ED compared with men with ED (−5.4 vs −5.7, respectively).
Multiple studies have consistently shown improvement in BPH-LUTS and quality of life with tadalafil treatment, despite modest Qmax improvement [16-19, 21]. Although a change in Qmax with tadalafil treatment did not reach significance in several studies, the consistency of tadalafil treatment in improving BPH-LUTS severity in conjunction with small improvements in Qmax was consistent with the updated BPH guideline that notes a poor correlation between BPH-LUTS and Qmax [7, 22].
While it is reasonable to consider that effective treatment of either condition (BPH-LUTS or ED) improves the overall quality of life, the degree of contribution and the relationship of responses of the two conditions were beyond the scope of this analysis; however, multiple tadalafil studies have reported significant improvement in quality of life based on specific questionnaires assessing changes in urinary symptoms [16-19].
As tadalafil is an erectogenic drug, concerns about unintentional unblinding have been raised. Such concerns are not unusual among efficacious agents with easily identifiable efficacy outcomes or those with significant or easily identifiable adverse reactions. In the present analysis, a similar reduction in BPH-LUTS and associated improvement in quality of life was observed both in men without ED and in men with ED, suggesting that unintentional unblinding did not play a significant role.
Regular intercourse may have a role in preserving erectile function among elderly men ; however, whether an improvement in erection leads to increased sexual activity is not clear and could not be confirmed in a post hoc analysis based on a large dose-finding tadalafil study in men with BPH-LUTS, which showed that increasing tadalafil doses from 2.5 to 20 mg increasingly improved ED . In that study, frequency of sexual intercourse attempts was similar across doses, including placebo, suggesting no association between erection improvement and number of intercourse attempts.
It has been proposed that increased frequency of ejaculation may indirectly improve LUTS, but in an assessment of LUTS severity in a cross-sectional analysis of 2115 white male subjects between the ages of 40 and 79 from the Olmsted study population, Jacobsen et al.  could not confirm an association between frequency of ejaculation and LUTS, peak urinary flow rates or prostate volume after accounting for the confounding effects of age.
Several key mechanisms through which tadalafil or other PDE5 inhibitors may affect BPH-LUTS or ED independently have been identified [26, 27]. PDE5-isoenzymes are known to suppress nitric oxide/cGMP) signalling, thereby impairing the normal micturition cycle and leading to LUTS [26, 27]. The effect of PDE5 inhibition leading to increased NO/cGMP concentration in the corpus cavernosum and pulmonary arteries has been observed in the smooth muscle of the prostate and bladder and their vascular supply. Vascular relaxation results in increased blood perfusion and may reduce BPH symptoms [28, 29]. Relaxation of stromal smooth muscle of the prostate and bladder may complement these vascular effects [26, 27, 30]. Recently, a novel L-cysteine/hydrogen sulphide pathway in the human bladder, modulated through a mostly NO-independent mechanism and activated by PDE5 inhibition, has been identified . Animal studies have suggested that PDE5-inhibition may also modulate the afferent-nerve activity from the lower urinary tract, decreasing the perception of bladder fullness and feeling of urgency [27, 32]. Whether or not similar mechanisms as reported in the bladder are also represented in the penile cavernous tissue is unclear, but these findings may support independence of the ED and BPH-LUTS response.
The present analysis includes an integrated population of subjects both with and without ED in an ancova model to allow interaction assessment; however, the efficacy evaluation within subjects without ED was the primary objective. A strength of the analysis is its sufficient statistical power that is based on a larger sample size of integrated data to assess tadalafil's effect on BPH-LUTS in men without ED, which was not achievable in any of the single studies. A limitation of this analysis may be the methodology in determining the ED population. For the purposes of this analysis, ED reflected the clinical diagnosis of ED based on a physical examination and functional inquiry. Additional assessment of ED was made by collecting data from the erectile function (EF) domain of the International Index of Erectile Function (IIEF). By definition, an IIEF-EF domain score <26  would substantiate the clinical diagnosis of ED. In this dataset, the IIEF questionnaire was only administered to those patients who were sexually active with a female partner, with the intent of evaluating response to therapy. While representing ∼80% of the population deemed to meet the clinical diagnosis of ED, other populations were unable to benefit from this assessment. Furthermore, when sexually active men with BPH-LUTS but no history of ED were treated with tadalafil in blinded placebo-controlled studies, the IIEF-EF change was only 0.8 point, suggesting that the methodology for determining ED history is reasonably reliable . Nevertheless, neither clinical determination of ED nor the IIEF tool alone can fully rule out ED in this population recruited for BPH-LUTS.
In this integrated analysis of 1089 men with BPH-LUTS, of whom 338 did not have ED, tadalafil was well tolerated and significantly improved BPH-LUTS and quality of life. These results suggest that PDE5 inhibitors may provide an effective therapeutic alternative for the treatment of BPH-LUTS in men both with and without coexisting ED.
This study was supported by Eli Lilly and Co. Nicole Johnston and Joseph Giaconia of INC Research (Raleigh, NC) provided writing assistance.
Conflict of Interest
Lei Xu, David Wong and Lars Viktrup are employees and stockholders of Eli Lilly and Company (Lilly). Gerald Brock and Claus Roehrborn are paid consultants to Lilly.