International Journal of Impotence Research (2005) 17, 450–454 & 2005 Nature Publishing Group All rights reserved 0955-9930/05 $30.00

www.nature.com/ijir

Efficacy of tadalafil in the treatment of erectile dysfunction in hypertensive men on concomitant thiazide diuretic therapy RA Kloner1*, R Sadovsky2, EG Johnson3, D Mo3 and S Ahuja3 1 Division of Cardiovascular Medicine, Heart Institute, Good Samaritan Hospital, Keck School of Medicine at University of Southern California, Los Angeles, California, USA; 2SUNY-Health Science Center at Brooklyn, Brooklyn, New York, USA; and 3Lilly Research Laboratories, Eli Lilly and Company, Indianapolis, Indiana, USA

Many men with erectile dysfunction (ED) have hypertension as a comorbid condition. Recent guidelines recommend thiazide diuretics as first-line therapy for hypertension. We analyzed data from 14 randomized, double-blind, placebo-controlled trials (N ¼ 2501) to evaluate the efficacy of tadalafil 20 mg for the treatment of ED in men on thiazides. Of the 2501 patients, 163 were on concomitant thiazides (116 tadalafil/47 placebo) and 159 (98%) were reported to have hypertension. The primary efficacy measures were mean change from baseline in the international index of erectile function (IIEF) erectile function (EF) domain and the proportion of ‘yes’ responses to sexual encounter profile (SEP) Questions 2 and 3. The tadalafil group showed a significantly (Po0.001) greater mean baseline to endpoint improvement on all efficacy outcome measures compared to placebo-treated patients regardless of concomitant thiazide use. More importantly, the responses to tadalafil were similar regardless of concomitant thiazide use. Additionally, responses to tadalafil were comparable between thiazide and nonthiazide users regardless of baseline ED severity (P40.05). International Journal of Impotence Research (2005) 17, 450–454. doi:10.1038/sj.ijir.3901360; published online 14 July 2005 Keywords: erectile dysfunction; hypertension; thiazide diuretics; tadalafil

Introduction The recently released ‘Seventh Report of the Joint National Committee on the Prevention, Detection, Evaluation and Treatment of High Blood Pressure (JNC 7)’ recommends thiazide-type diuretics to treat most patients with uncomplicated hypertension, either alone or in combination with drugs from other classes.1 One reason for noncompliance with antihypertensives is the potential occurrence of side effects, which include sexual dysfunction.2,3 Hypertension itself is associated with erectile dysfunction (ED) and many antihypertensive medications cause ED, including beta-blockers and centrally acting agents; however, thiazide diuretics seem to be the most commonly implicated. In the Treatment of Mild Hypertension Study (TOMHS) and the Trial of Antihypertensive Interventions and Management

(TAIM), the drug most commonly associated with ED was a thiazide diuretic, chlorthalidone.4,5 The mechanism for thiazide-induced ED has been hypothesized to be due to alterations in electrolytes, including the possibility of serum zinc deficiency, or volume depletion;6 however, the exact mechanism remains to be determined. Owing to the new recommendations for first-line treatment of hypertension with thiazide diuretics based on the JNC 7 report, recent results from the Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT),7 and a corresponding increase in the use of these agents, we examined whether tadalafil would demonstrate efficacy for the treatment of ED in men with hypertension receiving thiazide diuretics.

Methods *Correspondence: RA Kloner, Heart Institute, Good Samaritan Hospital (USC), 1225 Wilshire Blvd., 9th Floor Research, Los Angeles, CA 90017, USA. E-mail: [email protected] Received 7 April 2005; revised 24 May 2005; accepted 25 May 2005

Study design Data were analyzed from 14 randomized, doubleblind, placebo-controlled trials with a total of 2501

Efficacy of tadalafil in men taking thiazides RA Kloner et al

patients with ED. Of those patients, 163 were on concomitant thiazide-type diuretics (116 received tadalafil 20 mg and 47 placebo) and 159 of these 163 patients (97.5%) were reported to have hypertension. All studies involved 12 weeks of treatment, except for one, which lasted 26 weeks. For the purpose of the analyses presented here, only data gathered after 12 weeks of treatment are presented for the Erectile Function (EF) domain of the International Index of Erectile Function (IIEF) questionnaire and the Sexual Encounter Profile (SEP) diary. Since the Global Assessment Question (GAQ) was not gathered at 12 weeks in the study lasting 26 weeks, the GAQ results reported are from 13 studies. The 14 studies included treatment arms with 2.5, 5, 10 and/or 20 mg tadalafil; however, only the 20 mg data are reported because the number of patients on the other doses and concomitant thiazides was too small for meaningful statistical analyses. The study design of all 14 studies was similar: patients underwent a 4-week treatment-free run-in period followed by a 12-week treatment period in which tadalafil or placebo was taken as needed (a maximum of one dose per day) prior to sexual activity. Patients were asked to complete an SEP diary at the time of each sexual encounter and to complete the IIEF questionnaire at baseline after the treatment-free run-in period, and after 4, 8 and 12 weeks of treatment. All studies used the same three co-primary efficacy measures: mean change from baseline in the EF domain of the IIEF and the proportion of ‘yes’ responses to SEP Question 2 (SEP2: ‘Were you able to insert your penis into your partner’s vagina?’ (yes/no)) and SEP Question 3 (SEP3: ‘Did your erection last long enough for you to have successful intercourse?’ (yes/no)). Additionally, the GAQ was asked at the end of the study ‘Has the treatment you have been taking during this study improved your erections?’ (yes/no). Inclusion criteria and exclusion criteria were similar for all studies in this analysis. Inclusion criteria included men who were at least 18 y of age, who reported a minimum of 3 months history of ED and were to report on their sexual attempts with the same adult female partner. Exclusion criteria included unstable cardiovascular status (eg unstable angina, myocardial infarction or myocardial revascularization within the prior 90 days); a recent history of stroke or spinal cord trauma; use of nitrates, cancer chemotherapy or antiandrogens; failure to achieve any erection following radical prostatectomy or pelvic surgery; or clinically significant penile deformities or penile implants. All 14 of the studies excluded patients who, in the opinion of the investigator, were sildenafil nonresponders. Patients were included in the thiazide diuretic group if they took any one of the following medications: bendroflumethiazide, benzylhydrochlorthiazide, chlorothiazide, chlorthalidone, hydrochloro-

thiazide, indapamide (a thiazide-like drug), metolazone or trichlormethiazide. Patients with hypertension were identified by looking at the following diagnoses as defined by the Medical Dictionary for Regulatory Activities (MedDRA: version 6.1, MedDRA Maintenance and Support Services Organization, Reston, Virginia, USA) on the case report form during the screening for the study using the data set of pre-existing conditions:

451

1. Vascular hypertensive disorders (high-level group term). 2. Blood pressure disorders NEC (high-level term). 3. Cardiac hypertensive complication (high-level term). Of the four patients who were on thiazide diuretics but did not have a hypertension diagnosis, the diagnoses were: nephrolithiasis, renal colic, cardiac failure (congestive) and hyperlipidemia.

Statistical analysis The analysis of efficacy was conducted on an intentto-treat basis. Analyses of covariance (ANCOVA) models were used to evaluate treatment differences for all continuous outcomes. Changes from baseline in IIEF EF domain, SEP2 and SEP3 were treated as continuous outcomes. The analysis included all patients who had a baseline measurement and at least one postbaseline measurement. The percentage of patients responding positively to GAQ was based on the number of patients who responded to that question. To test whether there is a difference in outcome measurement between treatment groups, an ANCOVA model was used by including terms such as treatment group, study, baseline and interaction between treatment group and baseline. To test whether there is a difference in outcome measurement between subgroups as defined by thiazide use, an ANCOVA model was used by including the term subgroup in addition to the terms stated previously. In any model, if the interaction was not significant (if PZ0.1), then it was removed from the model and the betweentreatment-group P-value was obtained from the reduced model. Furthermore, a subgroup-by-treatment-group-interaction term was added to test whether treatment effect was similar between subgroups. Logistic regression models were used to evaluate treatment differences for categorical variables (eg response to GAQ). The logistic regression model included covariate terms described previously in ANCOVA models. A subgroup analysis was conducted to test if treatment effect as measured by changes from baseline in IIEF EF, SEP2 and SEP3 was similar between thiazide users and nonusers in each level of baseline ED severity as defined by the International Journal of Impotence Research

Efficacy of tadalafil in men taking thiazides RA Kloner et al 452

aggregated version of the Cappelleri categorization of IIEF EF domain of mild (17–25), moderate (11–16) and severe (1–10).8 If patients reported a normal IIEF EF domain score (26–30) at baseline, they were included in the mild group for purposes of data analyses, since ED is a self-reported condition. The analyses of safety in the 14 studies included all enrolled patients. Treatment-emergent adverse events (TEAEs), defined as events that first occurred or worsened after randomization, were recorded throughout the studies and were summarized by MedDRA 6.1 preferred term. Analyses of TEAEs used the Fischer exact test. Baseline and demographic characteristics of patients taking thiazides and not taking thiazides were compared by analysis of variance (ANOVA) for continuous variables and by w2 for categorical variables. Analyses were conducted using the SAS statistical package (SAS Institute, Cary, North Carolina, USA).

A higher percentage of patients on thiazide therapy tended to have severe ED at baseline, as measured by the EF domain of the IIEF, as well as a longer ED history than those not on thiazides, even though none of these trends were statistically significant (P40.05). It is notable that a greater proportion of patients in the tadalafil-plus-thiazide group were older, were of African descent, and had comorbid conditions than those patients taking tadalafil and no thiazide. Overall efficacy Tadalafil improved EF for all coprimary outcomes when compared to placebo (IIEF EF, SEP2 and SEP 3), whether patients were taking or not taking thiazide diuretics (Table 2). Results for GAQ are reported in Table 3. Overall, the effect of tadalafil relative to placebo was similar for thiazide users and nonusers (P40.10).

Results Efficacy analysis stratified by baseline severity Demographics and baseline characteristics Outcome measures were also examined based on ED severity scores at baseline. The efficacy analysis stratified by baseline ED severity consistently demonstrated that tadalafil effect relative to placebo,

The baseline demographics of the patients not taking thiazides and taking thiazides are included in Table 1. Table 1 Patient demographics and baseline characteristics

Not taking thiazides

Taking thiazides

Placebo (N ¼ 724)

Tadalafil 20 mg (N ¼ 1614)

Placebo (N ¼ 47)

Tadalafil 20 mg (N ¼ 116)

P-valuew

Mean age (range) Age 465 y, n (%) Duration of ED Z12 months, n (%)

54 (22–81) 132 (18%) 646 (89%)

54 (22–88) 276 (17%) 1386 (86%)

58 (37–74) 9 (19%) 42 (89%)

61 (32–78) 41 (35%) 106 (91%)

o0.001 o0.001

Ethnicity, n (%) African Caucasian Asian Hispanic Other

11 320 340 33 20

(2%) (44%) (47%) (5%) (3%)

49 909 500 106 50

(3%) (56%) (31%) (7%) (3%)

1 27 12 5 2

(2%) (57%) (26%) (11%) (4%)

12 70 17 8 9

(10%) (60%) (15%) (7%) (8%)

o0.001

Medical history, n (%) Coronary artery disease Depression Diabetes mellitus Hyperlipidemia Hypertension

26 18 147 79 154

(4%) (2%) (20%) (11%) (21%)

56 56 351 178 362

(3%) (3%) (22%) (11%) (22%)

2 1 18 15 44

(4%) (2%) (38%) (32%) (94%)

13 7 22 24 113

(11%) (6%) (19%) (21%) (97%)

o0.001

ED severity (EF domain score), n (%) Severe Moderate Mild

230 (32%) 214 (30%) 280 (39%)

501 (31%) 465 (29%) 648 (40%)

17 (36%) 13 (28%) 17 (36%)

46 (40%) 29 (25%) 41 (35%)

IIEF EF: International Index of Erectile Function Erectile Function domain. w P-value comparing patients on tadalafil/no thiazide vs tadalafil/thiazide; all P-values listed were statistically significant.

International Journal of Impotence Research

o0.01 o0.001

Efficacy of tadalafil in men taking thiazides RA Kloner et al 453

Table 2 Summary of mean efficacy scores in patients on tadalafil vs placebo in nonthiazide and thiazide groups Not taking thiazides

Taking thiazides

Placebo

Tadalafil 20 mg

Placebo

Tadalafil 20 mg

IIEF (mean scores) Baseline Change from baselinew End pointw

(n ¼ 704) 14.6 1.5 16.1

(n ¼ 1566) 14.7 8.8* 23.4*

(n ¼ 47) 14.2 1.3 15.5

(n ¼ 115) 14.0 9.4* 23.4*

SEP2 (mean-per-patient %) Baseline Change from baselinew End pointw

(n ¼ 708) 49.2% 3.6% 52.7%

(n ¼ 1579) 50.5% 30.7%* 81.2%*

(n ¼ 47) 43.5% 5.6% 49.1%

(n ¼ 115) 44.5% 34.3%* 78.8%*

SEP3 (mean-per-patient %) Baseline Change from baselinew End pointw

(n ¼ 708) 21.8% 11.8% 33.5%

(n ¼ 1579) 21.8% 47.6%* 69.4%*

(n ¼ 47) 18.2% 13.0% 31.2%

(n ¼ 115) 21.0% 45.1%* 66.1%*

w P40.05 efficacy comparison between nonthiazide and thiazide groups in ANOVA model including the terms treatment group, baseline, study, treatment by baseline interaction and subgroup. *Po0.001 efficacy comparison between tadalafil and placebo in ANOVA model including the terms treatment group, baseline, study and treatment by baseline interaction. The EF improvement among ED patients attributable to the therapy of tadalafil is similar for the nonthiazide and thiazide groups (P40.10).

Table 3 GAQ responsesa for patients on tadalafil vs placebo in nonthiazide and thiazide groups Not taking thiazides

% Yes responses

Taking thiazides

Placebo (N ¼ 641)

Tadalafil 20 mg (N ¼ 1444)

Placebo (N ¼ 43)

Tadalafil 20 mg (N ¼ 103)

38.2

85.2w

32.6

87.4w

a

Patients included in this analysis were those who responded to the GAQ at the end of 12-week therapy. w Po0.001 from comparing GAQ between tadalafil and placebo. The response among ED patients attributable to the therapy of tadalafil is similar for the nonthiazide and thiazide groups (P40.10).

as measured by the IIEF EF domain, SEP2 and SEP3 mean per-patient changes from baseline, were comparable between thiazide users and nonthiazide users in each ED severity stratum (P40.10). Change from baseline for the IIEF EF domain in the severe ED group was 12.1 in thiazide users vs 12.5 in nonusers, 10.0 vs 9.8 in the moderate ED group and 6.1 vs 5.1 in the mild ED group. Mean perpatient change from baseline for SEP2 in severe ED patients was 50.8% in thiazide users vs 53.6% in nonthiazide users, 43.4 vs 32.5% in moderate ED severity and 9.7 vs 11.9% in mild ED severity. Mean per-patient change from baseline for SEP3 in severe ED patients was 45.2% in thiazide users vs 50.3% in nonthiazide users, 61.6 vs 57.3% in moderate ED severity and 33.5 vs 38.5% in mild ED severity.

Adverse events The tolerability of tadalafil was evaluated based on reports of TEAEs (Table 4). No statistically significant difference was identified between thiazide users and nonusers in any of these events (P40.05), with the exception of upper respiratory tract infection (Po0.05 tadalafil/nonthiazide vs tadalafil/thiazide). Since the combination of tadalafil and an antihypertensive agent could potentially cause a lowering of blood pressure, terms indicating possible hypotension were compared: dizziness postural, syncope vasovagal, syncope, hypotension, orthostatic hypotension and loss of consciousness. There was no difference between the thiazide and nonthiazide groups in the incidence of these adverse events (P40.05).

Discussion Placebo-controlled clinical trials have demonstrated the efficacy and safety of tadalafil for the treatment of ED.9,10 In this analysis, we examined whether tadalafil was as efficacious in patients taking concomitant thiazide diuretics for hypertension as in those not taking thiazides. Robust efficacy was demonstrated by tadalafil when compared to placebo as measured by change from baseline in IIEF EF score, SEP2 and SEP3 responses, and tadalafil was as effective in those patients taking thiazides as in those not taking thiazides. While thiazides are widely implicated in drug-related ED, there is little data available in analyzing efficacy of phosphodiesterase type 5 (PDE5) inhibitors in patients on International Journal of Impotence Research

Efficacy of tadalafil in men taking thiazides RA Kloner et al 454

Table 4 Treatment-emergent adverse events (TEAE)a Not taking thiazides

Taking thiazides

Placebo Tadalafil Placebo Tadalafil (N ¼ 724) (N ¼ 1614) (N ¼ 47) (N ¼ 116) Headache (%) Dyspepsia (%) Back pain (%) Myalgia (%) Pain in extremity (%) Flushing (%) Nasal congestion (%) Upper respiratory tract infection (%) Cough (%)

4.0 0.7 1.9 0.7 0.3 1.0 0.7 0.8

15.1 7.1 5.5 3.1 2.9 2.9 2.3 0.9

6.4 0.0 6.4 0.0 0.0 0.0 0.0 0.0

14.7 8.6 5.2 4.3 4.3 3.4 1.7 3.4*

1.7

1.3

6.4

2.6

N ¼ all randomized patients. a TEAE occurring in Z2% of patients in any tadalafil treatment group. Data are presented as the percentage of patients who had a TEAE. Patients may be counted in more than one category. *Po0.05 tadalafil/no thiazide vs tadalafil/thiazide; all others were not statistically significant between tadalafil/no thiazide vs tadalafil/thiazide.

thiazides. Our data suggest that tadalafil was highly effective for the treatment of ED in men with hypertension on concomitant thiazide therapy. The importance of this finding is that thiazides are now considered to be the cornerstone of therapy for hypertension by many investigators. It is possible that tadalafil will improve compliance with thiazide diuretics in those men who are prone to develop ED with this kind of antihypertensive therapy. Our analysis also demonstrated that regardless of severity of ED, tadalafil was as effective in those patients taking thiazides as those not taking thiazides. Notably, there were trends for a higher percentage of those patients taking thiazides to have severe ED at baseline. An important concern has been whether addition of PDE5 inhibitors to antihypertensive medicines will be associated with a worsening of adverse events of either compound, especially adverse events related to hypotension. Our analyses show that there was no difference in adverse events, except for upper respiratory tract infection, whether patients on tadalafil were taking thiazides or not. In particular, there was no increase in the incidence of hypotensive adverse events. A limitation of this manuscript is that it is not known whether the ED in any of these patients was attributable to thiazides.

Conclusion Thiazide diuretics have historically been implicated in causing or worsening ED. Since many men with International Journal of Impotence Research

ED have concomitant hypertension and require therapy such as thiazide diuretics, it is important to determine the efficacy of PDE5 inhibitors in the setting of thiazide diuretics. In examining the data from 14 studies, it is evident that tadalafil provides robust efficacy regardless of whether patients are taking thiazide diuretics. Tadalafil is demonstrated to be an effective option for those men with ED who are on concomitant thiazide therapy.

Acknowledgements This study was supported by Lilly ICOS LLC (Bothell, Washington, Indianapolis, Indiana). We would like to acknowledge Dr Fanni Natanegara, PhD, for critically reviewing this manuscript. Disclosures: Dr Kloner is a consultant, researcher, and speaker for Lilly ICOS; Dr Kloner is also a consultant, speaker for Pfizer, and consultant to Bayer-GSK, Schering Plough, and Palatine/King Pharmaceuticals. Dr Sadovsky is a consultant and speaker for Lilly ICOS, Bayer-GSK, and Pfizer; Drs Johnson, Mo, and Ahuja are employees of Eli Lilly and Company.

References 1 Chobanian AV et al. The Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure: the JNC 7 report. JAMA 2003; 289: 2560–2572. 2 Weiss RJ. Effects of antihypertensive agents on sexual function. Am Fam Physician 1991; 44: 2075–2082. 3 Chang SW et al. The impact of diuretic therapy on reported sexual function. Arch Intern Med 1991; 151: 2402–2408. 4 Grimm Jr RH et al. Long-term effects on sexual function of five antihypertensive drugs and nutritional hygienic treatment in hypertensive men and women. Treatment of Mild Hypertension Study (TOMHS). Hypertension 1997; 29: 8–14. 5 Wassertheil-Smoller S et al. Effect of antihypertensives on sexual function and quality of life: the TAIM Study. Ann Intern Med 1991; 114: 613–620. 6 Keene LC, Davies PH. Drug-related erectile dysfunction. Adverse Drug React Toxicol Rev 1999; 18: 5–24. 7 The ALLHAT Officers and Coordinators for the ALLHAT Collaborative Research Group. Major outcomes in high-risk hypertensive patients randomized to angiotensin-converting enzyme inhibitor or calcium channel blocker vs diuretic: The Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT). JAMA 2002; 288: 2981–2997. 8 Cappelleri JC et al. Diagnostic evaluation of the erectile function domain of the International Index of Erectile Function. Urology 1999; 5: 346. 9 Brock GB et al. Efficacy and safety of tadalafil in the treatment of erectile dysfunction: results of integrated analyses. J Urol 2002; 168: 1332–1336. 10 Carson CC et al. The efficacy and safety of tadalafil: an update. Br J Urol Int 2004; 93: 1276–1281.