/04/ /0 Reprinted from Vol. 172, , August 2004 THE JOURNAL OF UROLOGY

Transcription

1 /04/ /0 Reprinted from Vol. 172, , August 2004 THE JOURNAL OF UROLOGY Printed in U.S.A. Copyright 2004 by AMERICAN UROLOGICAL ASSOCIATION DOI: /01.ju d7 RANDOMIZED STUDY OF TESTOSTERONE GEL AS ADJUNCTIVE THERAPY TO SILDENAFIL IN HYPOGONADAL MEN WITH ERECTILE DYSFUNCTION WHO DO NOT RESPOND TO SILDENAFIL ALONE R. SHABSIGH,* J. M. KAUFMAN, C. STEIDLE AND H. PADMA-NATHAN From the Department of Urology, Columbia University, New York, New York (RS), Urology Research Options, Aurora, Colorado (JMK), Northeast Indiana Research, Fort Wayne, Indiana (CS), and The Male Clinic, Keck School of Medicine at the University of Southern California School of Medicine, Beverly Hills, California (HP-N) ABSTRACT Purpose: We compare the efficacy of testosterone gel (T-gel) versus placebo as adjunctive therapy to sildenafil in hypogonadal men with erectile dysfunction who do not respond to sildenafil alone. Materials and Methods: A randomized, placebo controlled, double-blind, parallel group, multicenter study was performed. A total of 75 hypogonadal men (18 to 80 years old, morning serum total testosterone 400 ng/dl or less) with confirmed lack of response to sildenafil monotherapy were randomized (1:1) to receive a daily dose of 1% T-gel or 5 gm placebo gel as adjunctive therapy to 100 mg sildenafil during a 12-week period. Subjects were evaluated for sexual function, primarily based on the International Index of Erectile Function (IIEF), quality of life and serum testosterone levels at baseline and weeks 4, 8 and 12. Results: Testosterone treated subjects had greater improvement in erectile function compared to those who received placebo, reaching statistical significance at week 4 (4.4 vs 2.1, p 0.029, 95.1% CI 0.3, 4.7). Similar trends were observed for improvements in orgasmic function, overall satisfaction, total IIEF score and percentage of IIEF responders. T-gel significantly (p 0.004) increased total and free testosterone levels throughout the study, although no significant correlations were made between testosterone levels and the IIEF at end point. Conclusions: T-gel taken with sildenafil may be beneficial in improving erectile function in hypogonadal men with erectile dysfunction who are unresponsive to sildenafil alone. KEY WORDS: erectile dysfunction; hypogonadism; testosterone; drug therapy, combination Erectile dysfunction (ED) is the persistent inability to achieve and maintain an erection sufficient for satisfactory sexual activity. 1 Sildenafil citrate, an orally administered phosphodiesterase-5 (PDE-5) inhibitor commonly used to treat ED, selectively inhibits cyclic guanosine monophosphate specific PDE type 5, leading to increased cyclic guanosine monophosphate levels which result in muscle relaxation and penile tumescence. 2 Despite its effectiveness, however, approximately 30% to 50% of subjects receiving sildenafil do not adequately respond to therapy. 3 Based on the observed correlation between low serum free testosterone (T) and impaired cavernous vasodilatation in men with ED, 4 androgen replacement was hypothesized to be an effective concomitant therapy to sildenafil in hypogonadal men with ED who did not adequately respond to sildenafil alone. The objective of our study was to evaluate the efficacy and safety of transdermal 1% testosterone gel (T-gel) (AndroGel, Unimed/Solvay Pharmaceuticals, Inc., Marietta, Georgia) versus placebo as adjunctive therapy to sildenafil in this patient population. For the sake of brevity, only the efficacy findings are reported. MATERIALS AND METHODS Study design and population. This study followed the Declaration of Helsinki (1996) and the US 21 Code of Federal Accepted for publication March 26, Supported by Solvay Pharmaceuticals, Inc., the parent company of Unimed Pharmaceuticals, Inc. and manufacturer of the testosterone gel used in this study. Study received institutional review board approval. * Correspondence and requests for reprints: Department of Urology, Columbia Presbyterian Medical Center; 161 Fort Washington Ave., New York, New York (telephone: ; FAX: ; rs66@columbia.edu). Regulations. Institutional review board approval was obtained before initiation and written informed consent was obtained from each subject. This randomized, placebo controlled, double-blind, parallel group study involved 7 study visits. Visits 1 and 2 (screening) were completed within 14 days of visit 3, visit 3 was on day 28 at the start of the 4-week lead-in period with 100 mg sildenafil monotherapy, visit 4 was baseline, day 1 and the start of double-masked adjunctive therapy and visits 5 to 7 were on weeks 4, 8 and 12 of combined therapy. At visit 4 hypogonadal men with a confirmed inadequate response to 100 mg sildenafil monotherapy were randomized in a 1:1 ratio to receive 1% (5 gm) T-gel or placebo gel as adjunctive therapy to sildenafil tablets (Viagra, Pfizer Inc., New York, New York) during a 12-week period. The T and placebo gels were administered once daily per the manufacturer labeling. Subjects took 100 mg sildenafil once daily for each day with sexual activity per the manufacturer labeling. Randomization was based on a computer generated schedule stratified by investigator. Containers of study gel were identical in appearance and construction. Eligible subjects were males 18 to 80 years old who had ED for 3 months or greater, were involved in a stable sexual relationship with a female partner for 6 months or greater, had a morning serum total T level of 400 ng/dl or less at visits 1 and 2 and were nonresponders to sildenafil monotherapy (defined as a score of 2 or 3 on questions 3 and 4 of the International Index of Erectile Function [IIEF]). Main reasons for study exclusion were history of prostate or breast cancer, clinically significant or uncontrolled medical or psychiatric conditions (including diabetes mellitus), neurological disorders that cause ED, generalized S97 Please cite this article as J Urol 2008;179: S97 S102. DOI: /j.juro Copyright 2008 by AMERICAN UROLOGICAL ASSOCIATION.

2 S98 ADJUNCTIVE USE OF TESTOSTERONE GEL WITH SILDENAFIL Variable TABLE 1. IIEF scores at baseline (ITT population) Overall Range* Mean (SD) Placebo Gel (33 pts) Mean (SD) Testosterone Gel (37 pts) IIEF total score (10.3) 41.1 (8.4) Erectile function domain (4.0) 12.9 (3.0) Orgasmic function domain (2.8) 5.4 (1.8) Sexual desire domain (2.1) 6.3 (2.0) Intercourse satisfaction domain (2.4) 9.4 (1.9) Overall satisfaction domain (2.1) 4.9 (1.8) Question (0.8) 2.6 (0.7) Question (0.8) 2.2 (0.6) * Range of possible scores for each category; higher scores are indicative of better status; individual question scores are based on 5- (1 to 5) or 6- (0 to 5) point scales. Over the past 4 weeks, when you attempted sexual intercourse, how often were you able to penetrate (enter) your partner? Over the past 4 weeks, during sexual intercourse, how often were you able to maintain your erection after you had penetrated (entered) your partner? skin disease that may affect T-gel absorption, hyperprolactinemia (serum prolactin greater than 25 ng/ml) and drug or alcohol abuse. Subjects underwent a complete urological examination at screening, and those with symptoms of prostate disease and a concomitant urine flow rate of less than 12 ml per second were excluded from the study independently of prostate specific antigen levels. Efficacy criteria. ED and treatment outcomes were as- FIG. 1. Flow chart of subject disposition

3 ADJUNCTIVE USE OF TESTOSTERONE GEL WITH SILDENAFIL S99 sessed using the IIEF, a validated, multidimensional, selfadministered 15-item questionnaire. 5 Ranges of response for the IIEF are outlined in table 1. The primary efficacy parameter was response to the IIEF erectile function domain. Other IIEF domain scores were evaluated as secondary parameters, as were individual question scores, total score, and the proportion of responders and improvers. Responders were defined as subjects with a response of 4 (most times) or 5 (almost always or always) to questions 3 and 4 of the IIEF (table 1). Improvers were defined as subjects who increased functioning response by at least 1 category over baseline for either question 3 or 4. Other secondary efficacy parameters included frequency of successful sexual activity, subject global assessment of overall treatment efficacy (GAQ) (3 Yes/No questions), quality of life (QOL) 6 and serum total (and free) T levels. Statistical analysis. The planned sample size of 80 subjects was based on detecting a difference of 6 points or greater between groups in mean change in the erectile function domain score from baseline to week 12 with 90% power at a significance level of using an 8-point change as the estimated standard deviation. A statistical analysis plan was finalized before unblinding. All tests for the IIEF and QOL were 2-sided with an adjusted critical alpha level of Final analyses of all other efficacy measures were performed at a level of Computations were performed using SAS software (SAS Institute, Carey, North Carolina). Efficacy analyses were conducted on an intent-to-treat (ITT) basis with last observations carried forward. The efficacy population was defined as subjects who received at least 1 dose of study gel and had at least 1 post-baseline efficacy assessment. Analyses were performed at weeks 4, 8 and 12 and at end point. For continuous variables, betweentreatment comparisons were made for baseline and mean change from baseline values using an analysis of variance (ANOVA) model. If the data were not normally distributed (Shapiro-Wilk p 0.001), then analysis of variance on ranks was used (Wilcoxon rank sum test). 7 The Cochran-Mantel- Haenszel test was used to analyze treatment differences for nonordered categorical variables. 8 Subgroup analyses of the primary efficacy parameter were conducted for age, severity of baseline ED and hypogonadism, concurrent disease and smoking status. Within-group changes and between-treatment differences in serum T levels were evaluated using paired t tests and ANOVA, respectively. Analysis of variance on ranks was used for between-treatment comparisons if data were not normally distributed. Spearman rank correlation coefficients were used to explore the relationship between T levels and IIEF domain and total scores. RESULTS Disposition and background characteristics. The study was conducted from October 2, 2001 to November 15, A total of 75 subjects (15 centers) were randomized and received at least 1 dose of study gel in conjunction with sildenafil (fig. 1). Seventy subjects were included in the ITT efficacy analyses. Treatment groups were generally well matched with regard to demographic and background characteristics (table 2). Efficacy findings. Erectile Function Domain of the IIEF: Mean erectile function domain scores were comparable between groups at baseline (table 1). Throughout the study subjects who received T-gel with sildenafil had greater improvement in the erectile function domain score than those who received sildenafil with placebo (fig. 2). This difference reached statistical significance at week 4 when mean change from baseline values were 2.1 and 4.4 in the placebo and T groups, respectively (p 0.029, 95.1% CI 0.3, 4.7). Improvement in erectile function occurred within the first TABLE 2. Demographic and background characteristics (ITT population) Variable Placebo Gel T-Gel Totals No. pts No. ethnic origin (%): White 24 (73) 29 (78) 53 (76) Black 3 (9) 4 (11) 7 (10) Asian 1 (3) 1 (3) 2 (3) Hispanic 5 (15) 3 (8) 8 (11) Age: Mean SD Range Body mass index (kg/m 2 ): Mean SD Range No. ED etiology (%): Psychogenic 2 (6) 1 (3) 3 (4) Organic 14 (42) 20 (54) 34 (49) Mixed 17 (52) 16 (43) 33 (47) No. ED duration (%): 3 to 6 Mos 0 (0) 0 (0) 0 (0) 6 12 Mos 4 (12) 2 (5) 6 (9) Greater than 12 mos 29 (88) 35 (95) 64 (91) No. ED severity (%):*, Mild to moderate 7 (21) 3 (8) 10 (14) Moderate 16 (48) 27 (73) 43 (61) Severe 8 (24) 7 (19) 15 (21) Unknown 2 (6) 0 (0) 2 (3) No. hypogonadism severity (%):*, None 2 (6) 1 (3) 3 (4) Mild 7 (21) 15 (41) 22 (31) Moderate 13 (39) 19 (51) 32 (46) Severe 9 (27) 2 (5) 11 (16) Extreme 2 (6) 0 (0) 2 (3) No. concurrent condition (%):* Smoking 6 (18) 4 (11) 10 (14) Diabetes 4 (12) 7 (19) 11 (16) Hypertension 13 (39) 12 (32) 25 (36) Hyperlipidemia 4 (12) 10 (27) 14 (20) Prostate disease 1 (3) 5 (14) 6 (9) * At baseline. Severity of ED was based on the erectile function domain score of the IIEF as mild to moderate (17 to 21 points), moderate (11 to 16) and severe (6 to 10); severity of hypogonadism was based on serum total testosterone levels as mild (301 to 400 ng/dl), moderate (201 to 300), severe (101 to 200) and extreme (0 to 100). 4 weeks and was relatively stable in the T-gel group over time, whereas a smaller and gradual (yet progressive) improvement was observed in the placebo group. Subgroup analyses did not reveal any trends in either group to indicate that change in the erectile function domain score is affected by age, smoking history, severity of ED or hypogonadism at baseline, or concurrent diabetes, hypertension or hyperlipidemia. However, the small number of subjects in some of the subgroups precluded any definitive analysis (table 2). At each evaluation the percentage of IIEF responders was numerically higher in the T-gel group versus placebo (table FIG. 2. Mean change from baseline in IIEF erectile function domain score by ITT population. Asterisk denotes significance between groups at level (ANOVA). Positive changes from baseline were indicative of improvement.

4 S100 ADJUNCTIVE USE OF TESTOSTERONE GEL WITH SILDENAFIL Time Point TABLE 3. IIEF responders and improvers (ITT population) No. Placebo Gel (%) No. T-Gel (%) p Value (Cochran-Mantel- Haenszel test stratified by pooled center). IIEF responder:* Wk 4 9 (27) 19 (51) Wk 8 12 (39) 16 (47) Wk (42) 19 (58) End point 13 (39) 19 (51) IIEF improver (evaluated at wk 12 only) 22 (71) 26 (79) * Defined as subjects with a response of 4 or 5 to both questions 3 and 4 of the IIEF; response to both questions was based on a 6-point scale with higher scores indicative of better status. Defined as subjects who increased functioning response by at least 1 category over baseline for either question 3 or 4 of the IIEF. 3). As with mean change in the erectile function domain score, the difference between groups reached statistical significance at week 4 when the percentage of responders was nearly 2-fold higher in the testosterone treated subjects (p 0.040). At week 12 the majority of subjects in both arms were classified as IIEF improvers but without significant difference between groups. Other Domains of the IIEF and Response to Individual Questions: Mean scores of all other IIEF domains, as well as the total score, were comparable between groups at baseline (table 1). With each domain subjects who received T-gel with sildenafil showed greater improvement throughout the study compared to those who received placebo. For the orgasmic function domain, improvement was significantly superior in favor of T-gel at week 4 (p 0.009, 95.1% CI 0.3, 2.4) and approached statistical significance at all other time points (p 0.10). Improvements were also significantly superior in the T-gel group over placebo for the overall satisfaction domain score (p 0.020, 95.1% CI 0.2, 2.1) and total IIEF score (p 0.011, 95.1% CI 1.6, 12.3) at week 4 (figs. 3 and 4). Throughout the study, responses to individual IIEF questions were frequently more favorable in the T-gel group compared to placebo. Response to question 5 ( Over the past 4 weeks, during sexual intercourse, how difficult was it to maintain your erection to completion of intercourse? ) was significantly better in T-treated subjects at week 4 (p 0.012), at week 12 (p 0.008) and at end point (p 0.004), as was response to question 10 ( Over the past 4 weeks, when you had sexual stimulation or intercourse, how often did you have the feeling of orgasm [with or without ejaculation]? ) at weeks 4 (p 0.003), 8 (p 0.014) and 12 (p 0.018). Statistically significant differences were also observed between groups (in favor of T-gel) for response to question 13 ( Over the past 4 weeks, how satisfied have you been with your overall sex life? ) at week 4 (p 0.001) and question 15 ( Over the past 4 weeks, how do you rate your confidence that you can get and keep your erection? ) at week 12 (p 0.045). Serum Testosterone Levels and Correlation to the IIEF: Mean serum total T levels were comparable between groups at baseline as were mean serum free T levels. Subjects in the T-gel group showed statistically significant (p 0.004) increases from baseline in serum total and free T at weeks 4, 8, 12 and end point. These subjects also showed significantly superior (p 0.001) improvement in T levels over placebo at each evaluation (fig. 5). At end point no statistically significant correlations were observed in either group between serum total and free T levels and any of the IIEF total or domain scores. Other Efficacy Measures and Quality of Life: Throughout the adjunctive phase, mean changes from baseline in the average weekly number of successful sexual attempts and the proportion of successful attempts were comparable between groups. During this period the average weekly number of successful sexual attempts ranged from 1.5 to 2.4 for the placebo group and from 1.7 to 2.1 for the T-gel group. Likewise, the proportion of successful sexual attempts ranged from 43% to 50% and from 49% to 59%, respectively. Although response was generally in favor of the T-gel for GAQ question 1 (ie Did your gel improve erections?), no statistically significant differences were noted between groups at any time point. For question 2 ( Did your gel improve your response to sildenafil? ) a significantly higher percentage of subjects in the T-gel group answered yes compared to the placebo group at week 4 (59% vs 27%, p 0.005), at week 12 (73% vs 52%, p 0.049) and at end point (70% vs 48%, p 0.045). Likewise, the percentage of subjects who answered yes to question 3 ( Did your gel improve your erections at time when sildenafil was not taken within 4 hours before sexual activity? ) was numerically in favor of the T-gel group at each evaluation, reaching statistical significance at week 4 (41% vs 33%, p 0.009). Mean total QOL scores were similar between groups at baseline. Throughout the study changes in QOL were generally small but were consistently better in subjects who received T-gel versus placebo (fig. 6), reaching significance at week 12 (p 0.028) 95.1% CI 0.0, 0.5) and approaching significance at end point (p 0.059). FIG. 3. Mean change from baseline in IIEF overall satisfaction domain score by ITT population. Asterisk denotes significance between groups at level (ANOVA). Positive changes from baseline were indicative of improvement. FIG. 4. Mean change from baseline in IIEF total score by ITT population. Asterisk denotes significance between groups at level (ANOVA). Positive changes from baseline were indicative of improvement.

5 ADJUNCTIVE USE OF TESTOSTERONE GEL WITH SILDENAFIL S101 FIG. 5. Mean change from baseline in serum testosterone levels by ITT population. A, total testosterone. B, free testosterone. Asterisk denotes significance between groups at level (Wilcoxon rank sum test). DISCUSSION In this study adjunctive use of transdermal T-gel with sildenafil improved sexual activity in hypogonadal men with ED who were unresponsive to sildenafil alone. Improvements over placebo were observed throughout the 12-week period for each domain of the IIEF and the total score. Differences between study groups were most marked after 4 weeks of therapy, reaching statistical significance for improvements in the total IIEF score and 3 of the 5 domain scores (including erectile function). These findings were also reflected in the percentage of IIEF responders. Results of a small 1-month pilot study confirm that adjunctive administration of transdermal T in sildenafil nonresponders with low-normal androgen levels significantly improves erectile function. 9 FIG. 6. Mean change from baseline in quality of life (total score) by ITT population. Asterisk denotes significance between groups at level (ANOVA). Positive changes from baseline were indicative of improvement. The loss of statistical significance observed in this proofof-concept trial between groups over time is possibly due to small sample size, dropout rate or the steady improvement observed among placebo treated subjects during the evaluation period. Such a placebo effect is often observed in clinical trials and has been reported with ED therapy. 10 The lack of significant difference between groups could also be attributed to the fixed dose study design, limiting dosing to the manufacturer recommended starting dose. Future trials should address these imperfections, and include larger populations and a dose escalating design. The high prevalence of diabetics in this study may have minimized the overall effect of adjunctive T therapy. Although an inclusion criterion was diabetic control, 13 subjects (placebo 5; T-gel 8) had increased HgbA1c (HgbA1c 8% or greater) at study entry, suggestive of poorly controlled disease. Prolonged disease in the past may have also resulted in peripheral neuropathy that was nonreversible by current well-adjusted treatment. All of these subjects had a decreased potential for responding to either monotherapy or combination therapy. Another confounding factor was introduced by the inclusion of 8 subjects (placebo 5 T-gel 3) who could not be classified as nonresponders to sildenafil at baseline. While improved erectile function is the main goal of ED therapy, quality of life is also enhanced as subjects become more satisfied with sexual activity. In this study QOL scores were numerically better throughout the adjunctive therapy period in subjects who received T-gel versus placebo, and by week 12 such improvements were statistically superior in testosterone treated subjects. Subjects who received T-gel also reported better treatment efficacy based on response to

6 S102 ADJUNCTIVE USE OF TESTOSTERONE GEL WITH SILDENAFIL the GAQ, particularly when response was associated with use of sildenafil. CONCLUSIONS Despite the therapeutic advance of PDE-5 inhibitors, many men with ED do not adequately respond to therapy. More than 20% of subjects screened for our study had low T levels (less than 400 ng/dl). Such levels may contribute to continued sexual dysfunction when treated with sildenafil alone. T-gel (1%) taken with sildenafil may be beneficial in improving erectile function in hypogonadal men with ED who are unresponsive to sildenafil alone. Further studies using T-gel in conjunction with alternative PDE-5 inhibitors are warranted. Given our current knowledge, it seems prudent to screen men presenting with ED for hypogonadism before initiating therapy. REFERENCES 1. NIH Consensus Conference. Impotence. NIH Consensus Development Panel on Impotence. JAMA, 270: 83, Accessed September 17, Salonia, A., Rigatti, P. and Montorsi, F.: Sildenafil in erectile dysfunction: a critical review. Curr Med Res Opin, 19: 241, Aversa, A., Isidori, A. M., De Martino, M. U., Caprio, M., Fabbrini, E., Rocchietti-March, M. et al: Androgens and penile erection: evidence for a direct relationship between free testosterone and cavernous vasodilation in men with erectile dysfunction. Clin Endocrinol, 53: 517, Rosen, R. C., Riley, A., Wagner, G., Osterloh, I. H., Kirkpatrick, J. and Mishra, A.: The international index of erectile function (IIEF): a multidimensional scale for assessment of erectile dysfunction. Urology, 49: 822, Fugl-Meyer, A. R., Lodnert, G., Branholm, I. B. and Fugl-Meyer, K. S.: On life satisfaction in male erectile dysfunction. Int J Impot Res, 9: 141, Quade, D.: Rank analysis of covariance. J Am Stat Assoc, 62: 1187, Mantel, N.: Chi-square tests with one degree of freedom: extensions of the Mantel-Haenszel procedure. J Am Stat Assoc, 58: 690, Aversa, A., Isidori, A. M., Spera, G., Lenzi, A. and Fabbri, A.: Androgens improve cavernous vasodilation and response to sildenafil in patients with erectile dysfunction. Clin Endocrinol, 58: 632, Moyad, M. A.: The placebo effect and randomized trials: analysis of conventional medicine. Urol Clin North Am, 29: 125, 2002