Erectile dysfunction and lower urinary tract symptoms associated with benign prostatic hyperplasia (LUTS/BPH) combined responders to tadalafil after 12 weeks of treatment Claus G. Roehrborn, Kathryn B. Egan*, Martin M. Miner, Xiao Ni, David G. Wong and Raymond C. Rosen* Department of Urology, University of Texas Southwestern Medical Center, Dallas, TX, *New England Research Institutes, Inc., Watertown, MA, Men s Health Center, The Miriam Hospital, Providence, RI, Global Statistical Sciences and Advanced Analytics, and Eli Lilly and Company, Indianapolis, IN, USA Objective To analyse the proportion of men taking tadalafil 5 mg once daily who experience a combined improvement in symptoms of both erectile dysfunction (ED) and lower urinary tract symptoms associated with benign prostatic hyperplasia (LUTS/BPH). Materials and Methods The data from men aged 45 years randomized to tadalafil 5 mg once daily or placebo enrolled in one of four randomized, placebo-controlled LUTS/BPH clinical trials were analysed (N = 927). A novel classification of combined responders to ED and LUTS/BPH treatment was defined, based on published criteria for men who showed improvement in both International Index of Erectile Function Erectile Function domain (IIEF-EF) score and total International Prostate Symptom Score (IPSS). Descriptive analyses assessed the covariate distribution by responder status. Unadjusted and adjusted logistic regressions provided odds ratios with 95% confidence intervals comparing combined responders with all others (partial and nonresponders). Results Among men randomized to tadalafil 5 mg, 40.5% were combined responders (n = 189). Among placebo randomized men, 18.3% were combined responders (n = 84). Combined responders, in the total population, had the highest baseline IPSS and lowest baseline IIEF-EF scores, corresponding to the highest level of dysfunction. The majority of men were aged 65 years, white, non-obese, non-smokers, and regular alcohol consumers. Only treatment, baseline IPSS, baseline IIEF-EF, obesity and psychoactive medication use were significantly associated with responder status (P 0.05). Tadalafil-treated men had 2.8 times significantly increased adjusted odds of being combined responders vs nonresponders (P < 0.001). For each unit decrease in baseline IIEF-EF or alcoholic drink consumption per week there was a 4% significant increase in the adjusted odds of being a combined responder to tadalafil therapy. Conclusions This novel measure of combined response is useful in differentiating patients with clinically relevant symptom improvement for both ED and LUTS/BPH after treatment with tadalafil 5 mg once daily vs placebo. This combined responder measure may be useful in future assessment of treatment benefits across patient groups after various types of treatment intervention (e.g. surgical vs pharmacotherapy vs non-pharmacological intervention). Keywords LUTS, BPH, erectile dysfunction, tadalafil Introduction Erectile dysfunction (ED) and LUTS, associated with BPH (LUTS/BPH) are prevalent urological conditions in older men. The association between ED and LUTS/BPH and the increasing prevalence of each has been previously shown in population-based epidemiological studies [1 4] and summarized in a recent systematic review [5]. An estimated 70% of men with LUTS/BPH have coexisting ED [6]. Placebo-controlled studies have confirmed that tadalafil, a BJU International 2016 BJU International doi:10.1111/bju.13406 BJU Int 2016; 118: 153 160 Published by John Wiley & Sons Ltd. www.bjui.org wileyonlinelibrary.com
Roehrborn et al. phosphodiesterase-5 (PDE5) inhibitor, is efficacious and safe in treating men with either ED or LUTS/BPH, or in men presenting with both [6 10]. Previously, responder definitions for tadalafil treatment in men with ED or LUTS/BPH, using a change in Erectile Function domain of the International Index of Erectile Function (IIEF-EF) score or IPSS, respectively, have been reported [11,12]; however, the proportion of tadalafil-treated patients who reach an improvement in symptoms for both ED and LUTS/BPH has not been investigated. Patients in the tadalafil clinical trials pooled dataset were classified as combined responders for ED and LUTS/BPH, ED-only responders, or LUTS/BPH-only responders. All responder categories were compared with non-responders to tadalafil 5 mg treatment after 12 weeks of follow-up to determine if their symptom improvement was different, and whether this association differed for particular population subgroups. Men who were combined responders for ED and LUTS/BPH were hypothesized to show distinctive characteristics (i.e. unique phenotype) in comparison with partial/non-responders. We were interested in identifying phenotypic predictors to assist clinicians in the management of their patients by targeting patients to treatments and/or setting patient expectations. The aims of the study were (i) to define classification for a combined ED/BPH treatment responder to PDE5 inhibitor therapy with tadalafil 5 mg once daily in men with both ED and LUTS/BPH symptoms before treatment, henceforth known as combined responders; (ii) to identify descriptive characteristics of combined responders, ED-only responders, LUTS/BPH-only responders and complete non-responders, and (iii) to assess the efficacy of treatment with tadalafil 5 mg for ED and LUTS/BPH by comparing combined responders with all partial and non-responders, controlling for age, comorbid conditions and other relevant covariates on the outcomes of interest [7,11]. Methods Study Design A subset of the pooled tadalafil LUTS/BPH randomized, double-blind, placebo-controlled, parallel-design, 12-week clinical trials dataset was used. Details of the study design and trial results have been published previously [7]. All participants were men aged 45 years who had 6 month LUTS/BPH history, an IPSS 13 and peak urinary flow 4 or <15 ml/s at the start of the placebo run-in period. As ED status was not included in the trial criteria, participants who were sexually active with an adult female partner received the IIEF questionnaire. Participants with baseline and 1 post-baseline measurement for both IIEF-EF and IPSS were eligible for inclusion (N = 2 500). Studies were approved by institutional review boards and each subject gave written informed consent. Exclusion criteria included randomization to a dose of tadalafil other than 5mg(n = 832) or the tamsulosin arm from LVID (n = 168). Additionally, men who were not sexually active with an adult female partner (n = 198) or were missing this data (n = 1) were excluded, as were men with no history of ED (n = 269). After exclusion of those with incomplete data (n = 105), the final analysis sample was 927 men. During trial participation, men were randomized to one of two treatment groups: tadalafil 5 mg (n = 460) or placebo (n = 467). Outcome A novel classification of combined ED and LUTS/BPH responders (combined responder) to tadalafil 5 mg once-daily treatment was defined based on existing ED and LUTS/BPH efficacy criteria. ED responders were defined using a change in the IIEF-EF score, a domain of the validated selfadministered IIEF questionnaire based on six questions that examine erectile function. At baseline, men with mild ED (IIEF-EF score 17 25), moderate ED (IIEF-EF score 11 16), and severe ED (IIEF-EF score 0 10) required 2, 5 and 7- point increases in IIEF-EF scores, respectively, to be responders [12]. LUTS/BPH was assessed by the total IPSS, a validated self-report LUTS severity questionnaire based on seven urinary symptom questions and one question on quality of life [13]. Total scores ranged from 0 to 35 (asymptomatic to highly symptomatic). A reduction of 25% in the total IPSS between baseline and 12-week follow-up was the measure used to define a LUTS/BPH responder [11]. Combined responders were men who were responders in both ED and LUTS/BPH. ED-only responders were ED responders who were not LUTS/BPH responders. LUTS/BPHonly responders were LUTS/BPH responders who were not ED responders. If men failed to meet both the ED and LUTS/ BPH responder criteria, they were classified as nonresponders. Covariates Baseline IPSS, baseline IIEF-EF, age (years), body mass index (BMI; kg/m 2 ), systolic and diastolic blood pressure (mmhg), total cholesterol (mmol/l), non-fasting glucose (mmol/l), haemoglobin A1c, and PSA were all assessed as continuous variables. Categorical BMI (<30 kg/m 2 and 30 kg/m 2 ) was considered, as was race/ethnicity and treatment. Smoking, alcohol consumption and frequency, cardiovascular disease, diabetes, hypertension, hyperlipidaemia and medication use were assessed as binary yes/no variables. Medications considered were antihypertensive drugs (a- and b-blockers, calcium channel blockers, angiotensin-converting enzyme inhibitor, diuretics and/or angiotensin-receptor blocker), psychoactive drugs (selective serotonin reuptake inhibitor, 154 BJU International 2016 BJU International
ED and LUTS/BPH combined responders to tadalafil serotonin-norepinephrine reuptake inhibitors, antidepressants, antipsychotics, tricyclic antidepressants and/or monoamine oxidase inhibitors), antidiabetic drugs (insulin, sulphonylureas, a-glucosidase inhibitors, amylin analogues, incretin mimetics, dipeptidyl peptidase-4 inhibitors, biguanides, meglitinides and/or thiazolidinediones), and statins or other lipid-lowering medications. Prostate volume or median lobe observations were not available for inclusion, nor were data on pelvic injuries as result of previous surgeries. Analysis Responder status by ED or LUTS/BPH response was crosstabulated (Table 1). Descriptive analyses of covariates overall and by responder status were assessed (Table 2). No outliers were identified. For continuous variables, means, standard deviations and t-test P values are presented. Frequencies (%) and chi-squared P values were used to describe categorical variables. Odds ratios (ORs) and 95% CIs were assessed using unadjusted and adjusted logistic regression models where combined responders were compared with all partial and/or non-responders, while accounting for cohort and treatment. Backwards stepwise selection was used to create multivariable logistic regression models predicting combined responder status based on covariates of interest (P 0.1; Fig. 1). No adjustment was made for multiple comparisons and no interaction terms were included in final models. P values 0.05 were taken to indicate statistical significance. Data were analysed using SAS software, version 9.3, SAS Institute Inc., Cary, NC, USA. Table 1 Cross-tabulation of erectile dysfunction and LUTS/BPH response.* LUTS/BPH response ED response Non-responder Responder Total Total sample, n (%) Non-responder 281 (30.3) 163 (17.6) 444 (47.9) Responder 210 (22.7) 273 (29.5) 483 (52.1) Total 491 (53.0) 436 (47.0) 927 (100.0) Placebo treatment arm only, n (%) Non-responder 185 (40.2) 71 (15.4) 256 (55.7) Responder 120 (26.1) 84 (18.3) 204 (44.3) Total 305 (66.3) 155 (33.7) 460 (100.0) Tadalafil 5 mg treatment arm only, n (%) Non-responder 96 (20.6) 92 (19.7) 188 (40.3) Responder 90 (19.3) 189 (40.5) 279 (59.7) Total 186 (39.8) 281 (60.2) 467 (100.0) ED, erectile dysfunction; IIEF-EF, International Index of Erectile Function Erectile Function domain. *As a result of rounding not all percentages sum. Those with mild ED at baseline (17 IIEF-EF 25) required at least a 2-point increase in IIEF-EF score to be a responder, those with moderate ED (IIEF-EF score 11 16) required at least a 5-point increase, and those with severe ED (IIEF-EF score 0 10) required at least a 7- point increase [12]. A reduction of 25% in the IPSS total score was required between baseline and 12-week follow-up to determine a positive response to tadalafil 5 mg treatment. Results The percentage of patients who were classified as combined responders to tadalafil therapy was 40.5% among treatmentrandomized men, 18.3% among placebo-randomized men and 29.5% in the overall study population (Table 1). Of the 460 placebo-randomized men, 40% were treatment nonresponders (chi-squared P < 0.001) and, among 467 men randomized to active drug, 21% were non-responders (chisquared P < 0.001; Table 1). The distribution of responder status by each covariate of interest is shown in Table 2. Importantly, the ED-only and LUTS/BPH-only responders (partial responders) were not analysed as independent categories because these classifications were affected in a positive or negative direction, contingent on our criteria for defining combined responders. The more liberal these criteria, the greater number of men would accordingly be classified as combined responders and a lower number of men would necessarily be BPH-only or ED-only responders. In light of these necessary limitations to independence, we do not report logistic results for these latter categories in isolation. Combined responders had a significantly greater likelihood of being on active treatment, being non-white (i.e. black/other race), and having a higher BMI than other responder categories. They had lower baseline IIEF-EF scores and higher baseline IPSSs (P 0.05). Unadjusted Odds Ratios When comparing combined responders with all others, treatment with tadalafil was associated with a 3.04 times greater unadjusted odds of being classified as a combined response (Table 3). A 1-unit increase in baseline IPSS significantly increased the odds of achieving a combined response by 3% (OR 1.03). Baseline IIEF-EF score was borderline significantly associated with a 3% lower odds of combined response (OR 0.97). Obese men had a 30% significantly lower odds of achieving a combined response and the number of alcoholic drinks consumed weekly was significantly associated with achieving a combined response because each additional drink decreased the odds of being a combined responder by 3%. Only alcohol consumption frequency (drinks/week) was significantly associated with responder status after stratification by treatment at the P < 0.1 level. For every 1-unit increase in alcoholic drinks consumed per week, the unadjusted odds of being a combined responder for men randomized to placebo was 0.86 times the odds of being a partial or non-combined responder. Adjusted Odds Ratios The final adjusted multivariable model included treatment, age, baseline IIEF-EF, alcohol consumption frequency, diabetes and psychoactive medication use which were all BJU International 2016 BJU International 155
Roehrborn et al. Table 2 Descriptive characteristics of analysis sample overall and by responder status (N = 927). Characteristics Overall Combined responder* (n = 273) ED only responder* (n = 163) LUTS/BPH only responder* (n = 210) Non-responder* (n = 281) P Treatment 467 (50.4) 189 (69.2) 92 (56.4) 90 (42.9) 96 (34.2) <0.001 Baseline IPSS score 17.3 5.5 18.0 5.4 17.6 5.3 16.5 5.4 17.0 5.8 0.020 Baseline IIEF-EF domain 16.0 7.8 14.9 6.4 15.6 6.7 17.1 8.9 16.6 8.7 0.012 Demographic characteristics Age, years 63.0 8.3 62.8 8.3 61.2 7.9 63.2 8.7 63.4 8.2 0.248 Black or other race 70 (7.7) 25 (9.2) 10 (6.3) 18 (8.6) 17 (6.3) 0.519 Obese: BMI 30 kg/m 2 688 (74.2) 215 (78.8) 127 (77.9) 160 (76.2) 186 (66.2) 0.003 Smoker 127 (13.7) 38 (13.9) 27 (16.6) 34 (16.2) 28 (10) 0.135 Alcohol consumption frequency, drinks/week 6.0 5.6 5.4 5.4 6.6 5.1 5.4 4.8 6.8 6.8 0.042 Laboratory measurements Systolic blood pressure, mmhg 133.4 13.1 132.3 13.6 132.9 13.4 133.6 12.8 134.7 12.7 0.172 Diastolic blood pressure, mmhg 80.8 8.1 80.4 8.7 80.3 8.1 81.3 8.0 81.2 7.7 0.381 Total cholesterol, mmol/l 5.2 1.0 5.3 1.1 5.3 1.0 5.2 1.1 5.1 1.0 0.389 Non-fasting glucose, mmol/l 5.9 1.6 5.9 1.4 6.0 2.0 5.9 1.7 5.8 1.5 0.864 Haemoglobin A1c, mmol/mol 0.057 0.007 0.06 0.005 0.058 0.008 0.058 0.007 0.057 0.007 0.722 PSA, ng/ml 2.0 1.9 2.0 2.2 2.0 1.9 2.0 1.7 2.0 1.7 0.996 Diseases Cardiovascular disease 424 (45.7) 115 (42.1) 68 (41.7) 106 (50.5) 135 (48) 0.172 Diabetes 127 (13.7) 32 (11.7) 26 (16) 30 (14.3) 39 (13.9) 0.643 Hypertension 361 (38.9) 101 (37) 57 (35) 88 (41.9) 115 (40.9) 0.431 Hyperlipidaemia 210 (22.7) 61 (22.3) 36 (22.1) 47 (22.4) 66 (23.5) 0.983 Medications Antihypertensive, 486 (52.4) 139 (50.9) 79 (48.5) 109 (51.9) 159 (56.6) 0.355 Psychoactive, 30 (3.2) 2 (0.7) 6 (3.7) 5 (2.4) 17 (6) 0.004 Diabetes, ** 102 (11) 27 (9.9) 22 (13.5) 23 (11) 30 (10.7) 0.703 Statins and other lipid-lowering 202 (21.8) 59 (21.6) 32 (19.6) 48 (22.9) 63 (22.4) 0.883 BMI, body mass index; ED, erectile dysfunction; IIEF-EF, International Index of Erectile Function Erectile Function domain. *Columns are frequency (column %) for categorical variables and mean (SD) for continuous variables. P values are from t-test for continuous variables and chi-squared test for categorical variables. Reference category was never, none, no symptoms, no disease or no use for each condition, disease or medication, as appropriate. Antihypertensive medication includes a-blockers, b-blockers, calcium channel blockers, angiotensin-converting enzyme inhibitors, diuretics, angiotensin-receptor blocker and other antihypertensive drugs. Psychoactive medication includes selective serotonin reuptake inhibitors, serotonin-norepinephrine reuptake inhibitors, antidepressants, antipsychotics, tricyclic antidepressants and monoamine oxidase inhibitors. **Diabetes medication includes insulin use, insulin TR phase, insulin XT phase, sulphonylureas, a-glucosidase inhibitors, amylin analogues, incretin mimetics, dipeptidyl peptidase-4 inhibitors, biguanides, meglitinides and thiazolidinediones. significant at the P 0.1 level (Fig. 1). Men randomized to treatment had 2.80 times significantly increased odds of being combined responders in comparison with non-combined responders (95% CI 1.89, 4.16). Each unit increase in baseline IIEF-EF significantly decreased the odds of being a combined responder by 5% vs a non-combined responder (OR 0.95, 95% CI 0.93, 0.98). Each additional reported alcoholic drink per week significantly decreased the odds of being a combined responder by 4% vs a non-combined responder (OR 0.96, 95% CI 0.92, 1.00). Psychoactive medication use and diabetes remained in the final model as they were borderline significantly associated with responder status (OR 0.27, P = 0.091; and, OR = 0.47, P = 0.019, respectively). Age was forced into the final adjusted model (OR 0.99, 95% CI 0.97, 1.02). All other covariates of interest in Table 2 were not significantly associated (P 0.1) with the combined responder outcome. Discussion A novel classification has been developed and used for classifying patients with both ED and LUTS/BPH as combined responders, defined as a clinically relevant improvement in both available measures of ED and LUTS/ BPH. In summary, 40.5% of patients in the pooled study population who received tadalafil 5 mg once daily, a PDE5 inhibitor approved for use on a daily basis, could be classified as combined responders. Although less than the number who usually achieve positive outcomes for either condition alone, it is substantially greater than 18.3% of those randomized to placebo (P < 0.001). Taking tadalafil 5 mg therapy vs placebo more than doubled a man s chances of achieving a combined response. An additional 39.0% of treatment-randomized men had a partial response classified as an improvement in ED or LUTS/BPH, but not both, and 20.6% did not achieve benefit on either LUTS/BPH or ED response based on published definitions. ED responders were defined using a previously published standard for clinically meaningful improvement in IIEF-EF domain scores, and LUTS/BPH responders were similarly defined using a published standard for clinical improvement in IPSS scores [11,12]. Combined responders were required to achieve the threshold for both of these outcomes independently in order to be classified as a combined responder. 156 BJU International 2016 BJU International
ED and LUTS/BPH combined responders to tadalafil Fig. 1 Forest plot of adjusted odds of combined responder status vs non-combined responder status (odds ratio, 95% CI). IIEF-EF, International Index of Erectile Function Erectile Function domain. Treatment arm Age (years) Covariates Baseline IIEF - EF domain Alcohol consumption frequency (drinks/week) Diabetes Psychoactive medications 0 1 2 3 Odds Ratio 4 5 Table 3 Unadjusted associations between combined responders vs non-combined responders and covariates of interest, overall and stratified by treatment assignment (n = 927). Characteristics Combined responders Stratified sample Total sample* Treatment* Placebo* OR (95% CI) OR (95% CI) OR (95% CI) Treatment 3.04 (2.45, 4.10) Baseline IPSS 1.03 (1.01, 1.06) 1.03 (1.00, 1.07) 1.03 (0.99, 1.08) Baseline IIEF-EF score 0.97 (0.96, 0.99) 0.97 (0.95, 1.00) 0.98 (0.95, 1.01) Age, years 1.00 (0.98, 1.01) 0.99 (0.97, 1.02) 1.01 (0.98, 1.04) Black or other race 1.37 (0.82, 2.30) 1.32 (0.67, 2.58) 1.38 (0.57, 3.36) Obese: BMI 30 kg/m 2 0.70 (0.50, 0.98) 0.81 (0.51, 1.28) 0.76 (0.44, 1.29) Smoker 1.02 (0.67, 1.53) 1.07 (0.62, 1.83) 0.91 (0.45, 1.85) Alcohol consumption frequency (drinks/week) 0.97 (0.93, 1.00) 1.00 (0.96, 1.04) 0.86 (0.79, 0.95) Laboratory measurements Systolic blood pressure (mmhg) 0.99 (0.98, 1.00) 0.99 (0.98, 1.01) 0.99 (0.97, 1.01) Diastolic blood pressure (mmhg) 0.99 (0.97, 1.01) 0.99 (0.97, 1.01) 0.99 (0.96, 1.02) Total cholesterol (mmol/l) 1.07 (0.94, 1.23) 1.12 (0.94, 1.33) 1.03 (0.81, 1.30) Non-fasting glucose (mmol/l) 1.00 (0.92, 1.09) 1.02 (0.91, 1.14) 0.99 (0.85, 1.15) Haemoglobin A1c (mmol/mol) PSA (ng/ml) 0.99 (0.92, 1.07) 1.01 (0.92, 1.10) 0.98 (0.86, 1.13) Diseases Cardiovascular 0.82 (0.62, 1.09) 0.83 (0.57, 1.21) 0.83 (0.51, 1.34) Diabetes 0.78 (0.51, 1.20) 0.63 (0.36, 1.10) 1.00 (0.49, 2.02) Hypertension 0.90 (0.67, 1.20) 0.84 (0.57, 1.23) 1.01 (0.62, 1.66) Hyperlipidaemia 1.00 (0.71, 1.40) 0.92 (0.59, 1.43) 1.01 (0.56, 1.81) Medication Antihypertensive 0.92 (0.80, 1.06) 0.90 (0.74, 1.09) 0.96 (0.77, 1.19) Psychoactive 0.17 (0.04, 0.71) 0.56 (0.13, 2.50) Diabetes 0.85 (0.53, 1.35) 0.68 (0.37, 1.23) 1.09 (0.50, 2.35) Statins and other lipid-lowering 1.00 (0.71, 1.41) 1.05 (0.67, 1.63) 0.80 (0.43, 1.48) BMI, body mass index; ED, erectile dysfunction; IIEF-EF, International Index of Erectile Function Erectile Function domain; OR, odds ratio. *N = 927 overall; n = 467 treatment; n = 460 placebo. Compared with men with either partial (ED-only, LUTS/BPH-only) response or non-response to treatment (n = 654). Compared with men with either no ED or no LUTS/BPH response or who were complete non-responders (n = 278 for treatment). Compared with men with either no ED or no LUTS/BPH response or who were complete non-responders (n = 376 for placebo). Reference category was no symptoms, no disease or no use for each condition, disease or medication, as appropriate. The rate of positive outcomes for ED in our analysis was consistent with rates reported in other studies, taking into account differences in study designs or treatment responder definitions. Baseline IIEF-EF score was significantly associated with combined responder status. A 1-unit increase in IIEF-EF score (improvement) was significantly associated with a 5% decreased adjusted odds of achieving a combined response compared with being a treatment non-responder. This finding suggests, that contrary to being a treatment contraindication, men with more severe IIEF-EF symptoms are more likely to be combined responders than men with more mild ED symptoms. A similar study reported tadalafil 5 mg to be efficacious in ED treatment because 74.8% of patients with mild baseline IIEF-EF scores achieved a clinically meaningful BJU International 2016 BJU International 157
Roehrborn et al. improvement in ED symptoms. In this same study, among men with baseline moderate or severe IIEF-EF scores, 63.1 and 44.5%, respectively, reported improvements in their ED symptoms [14]. When treated with tadalafil 5 mg, about two-thirds of men were likely to achieve clinically significant benefit in their LUTS symptoms, defined by a 25% decrease in IPSS from baseline to endpoint, as shown previously in one recent publication [11]. The overall LUTS/BPH responder 52.1% prevalence (combined responders added to the LUTS/BPHonly responders) was fairly consistent with rates published in previous LUTS/BPH research [11]. The criterion of 25% improvement over baseline IPSS has been adopted in a recent study by Nickel et al. [11] in men taking tadalafil or placebo. These authors showed that patients receiving tadalafil 5mg have a 2.0 times significantly higher odds of being LUTS/BPH responders in comparison with patients randomized to placebo. Using the 25% reduction in IPSS definition of LUTS/BPH responders is an intermediate assumption for defining a threshold for LUTS/BPH improvement because it is relative to baseline IPSS [15 17]. These baseline IPSSs may be important because symptom severity at baseline has been shown to play a significant role in determining outcomes of treatment [18]. Tadalafil 5 mg is efficacious for use in men with a combination of ED and LUTS/BPH, or either condition alone. Treatment significantly increased the odds of achieving a combined response compared with a non-combined response (OR 2.80), confirming this treatment efficacy for simultaneous improvement in both ED and LUTS/BPH. Previously, it had yet to be determined what proportion of tadalafil-treated patients with both conditions would show a positive treatment response across both conditions and whether there were significant correlates or predictors for a combined outcome. The present findings are broadly consistent with those of other large placebo-controlled studies that have indicated tadalafil tobeefficacious and safe in treating men with either ED or LUTS/BPH, or both [6 10]. Among men randomized to placebo, fewer than one in five (18.3%) achieved a combined response. Men on placebo had a < 20% (i.e. 15.4%) likelihood of being ED-only responders and a 26.1% likelihood of being LUTS/BPH-only responders. Placebo responses are commonly observed in clinical trials of both ED and LUTS/BPH in the range of 15 30%, depending on the response criterion, severity of disease and duration of the trial. In that respect, our rate of observed placebo response is consistent with rates in previous studies in LUTS/ BPH or ED [19,20]. This relatively typical placebo response was observed in a previous tadalafil study to occur for 4 6 months before attenuating [21], and could be attributed to behaviour changes during the clinical trial or regression to the mean. Sech et al. [22] have identified a predictable placebo effect across BPH trials that the authors attributed to regression to the mean and other factors. Men enrolled in trials are often more aware of their symptoms and have more interaction with healthcare professionals than they would without clinical trial participation [19]. Given that many men treated for ED or LUTS/BPH are older, with higher BMI and multiple comorbid conditions, the effects of these covariates on responder status were assessed (Table 2) [7,11]. In the overall study population, the majority of men were aged 65 years, white, non-obese, non-smokers, and consumed alcohol regularly. Most men did not have diabetes, hypertension, hyperlipidaemia or use psychoactive medications, diabetic medications or statins. Only treatment, baseline IPSS, baseline IIEF-EF, obesity and psychoactive medication use were significantly associated with responder status (P 0.10). Combined responders had the highest baseline IPSS and lowest baseline IIEF-EF values, which is not surprising as LUTS severity can influence the amount of improvement that a patient considers meaningful and, therefore, is willing to report [18]. As in a previous study of LUTS/BPH responders, we did not find baseline characteristics or demographics to be different between patients responding or not to treatment [11]. Interestingly, only three covariates of interest that we investigated were associated in adjusted analyses in the total population with being a combined responder (P 0.1). First, for each additional reported alcoholic drink per week, there was a 4% significantly decreased odds of being a combined responder vs a non-combined responder. Second, current psychoactive medication use was borderline significantly associated with a decrease in the odds of being a combined responder (OR 0.27, P=0.091); however, this finding should be cautiously interpreted as there were relatively few men overall on psychoactive medications in this population (n = 30; 3.2% of total). Third, having diabetes was borderline significantly associated with a 53% decreased odds of being a combined responder vs a non-combined responder (P = 0.019). Other covariates of interest that we investigated were not significantly associated with responder status and, therefore, did not predict who would be a combined responder or not (Fig. 1). As most men who seek treatment for either LUTS/BPH or ED are likely to have both conditions [23], it is important for clinicians and insurers to recognize the coexistence and opportunity for combined benefits of treatment. With the availability of a single, approved treatment (tadalafil5mg once daily) that is potentially effective for either condition alone or for both, the present analysis shows a novel measure for clinicians to assess combined responders. With this, healthcare practitioners could assess patients with respect to treatment success in their own practice compared with published data, and to manage individual patients with 158 BJU International 2016 BJU International
ED and LUTS/BPH combined responders to tadalafil precision and maximum attention to detail. Because many clinicians use IPSS and IIEF scales routinely in their practice, this may be feasible and practical to implement. Patients could be informed of potential treatment benefits in both areas and should discuss expectations and treatment outcomes with their physicians. The IPSS and IIEF tools may assist in this discussion in order to maximize the patient s overall quality of life. The major strengths of the present study are the use of an integrated pooled population dataset of 927 men from four randomized, placebo-controlled clinical trials. As with previous studies [24], this large and diverse patient population in the combined dataset provides a unique opportunity to assess ED and/or LUTS/BPH responder status and covariates of interest in association with this binary responder outcome. This is the first study to combine the ED IIEF-EF responder definition with the LUTS/BPH IPSS responder definition for tadalafil 5 mg once-daily treatment to create a novel combined responder definition for evaluating treatment outcomes [11,12]. Previously, a study reported no appreciable difference in response based on selfreport of ED on LUTS/BPH response and reported no appreciable difference in erectile function on LUTS [25]. Multivariate modelling techniques were also used to assess the influence of a broad range of potential covariates. We also recognize several limitations of the present study. ED and LUTS/BPH symptom data were self-reported and patient satisfaction with improvement is subjective; however, the use of the validated IIEF-EF for ED and IPSS for LUTS/BPH for this post hoc analysis was scientifically appropriate to assess ED and LUTS/BPH symptom severity and treatment responder status [11,12]. IPSS was the primary endpoint in the respective clinical trials. The relatively short 12-week duration of treatment may have obscured the potential longterm effects of some baseline covariates of interest on longterm treatment outcomes. Another limitation is the lack of information about prostate volume, median lobe characteristics, or previous surgeries in our sample. Finally, our findings were developed from retrospective analysis of a large, combined clinical trial database of men randomized to tadalafil 5 mg or placebo, and warrant replication and further validation in future prospective studies with tadalafil 5 mg or other treatment agents. For descriptive analyses, a definition of combined responders, ED-only responders, LUTS/BPHonly responders, and complete non-responders was used where each category was mutually exclusive. Men could have had more or less of one component or the other. This is a potential limitation of the characteristics analyses for the EDonly and LUTS/BPH-only categories. These categories should be interpreted with caution because they only include men who were responders to one category while also being nonresponders to the other category. Not all men who were ED responders were included in the ED-only group and not all men who were LUTS/BPH responders were included in the LUTS/BPH-only group as men in the combined responders group were by definition ED responders and LUTS/BPH responders as well. In conclusion, 40% of men aged 45 years who were enrolled in one of four LUTS/BPH clinical trials were combined responders for ED and LUTS/BPH to treatment with tadalafil 5 mg once daily. Tadalafil 5 mg is an efficacious treatment that improves both ED and LUTS/BPH symptoms after 12 weeks. Clinicians could use this combined response standard to judge if a patient has an improvement after treatment. Only alcohol use, baseline IIEF-EF scores, and diabetes significantly decreased the odds of being a combined responder in comparison with a non-combined responder in this population. Acknowledgements NERI Lilly BPH Publication Initiative was supported by Lilly USA, LLC. Conflict of Interest Dr Roehrborn received partial compensation for time and effort dedicated to the work on this manuscript consisting of data analysis and interpretation, writing of the manuscript and editing, and proofing of the final version. References 1 Boyle P, Robertson C, Mazzetta C et al. The association between lower urinary tract symptoms and erectile dysfunction in four centres: the UrEpik study. BJU Int 2003; 92: 719 25 2 Rosen R, Altwein J, Boyle P et al. Lower urinary tract symptoms and male sexual dysfunction: the multinational survey of the aging male (MSAM-7). Eur Urol 2003; 44: 637 49 3 Brookes ST, Link CL, Donovan JL, McKinlay JB. Relationship between lower urinary tract symptoms and erectile dysfunction: results from the Boston Area Community Health Survey. J Urol 2008; 179: 250 5 4 Braun MH, Sommer F, Haupt G, Mathers MJ, Reifenrath B, Engelmann UH. Lower urinary tract symptoms and erectile dysfunction: co-morbidity or typical Aging Male symptoms? Results of the Cologne Male Survey. Eur Urol 2003; 44: 588 94 5 Park HJ, Joanne Won JE, Sorsaburu S, Rivera PD, Lee SW. Urinary tract symptoms (LUTS) secondary to benign prostatic hyperplasia (BPH) and LUTS/BPH with erectile dysfunction in Asian men: a systematic review focusing on tadalafil. World J Mens Health 2013; 31: 193 207 6 Oelke M, Bachmann A, Descazeaud A et al. EAU guidelines on the treatment and follow-up of non-neurogenic male lower urinary tract symptoms including benign prostatic obstruction. Eur Urol 2013; 64: 118 40 7 Porst H, Kim ED, Casabe AR et al. Efficacy and safety of tadalafil once daily in the treatment of men with lower urinary tract symptoms suggestive of benign prostatic hyperplasia: results of an international randomized, double-blind, placebo-controlled trial. Eur Urol 2011; 60: 1105 13 8 Egerdie RB, Auerbach S, Roehrborn CG et al. Tadalafil 2.5 or 5 mg administered once daily for 12 weeks in men with both erectile dysfunction and signs and symptoms of benign prostatic hyperplasia: BJU International 2016 BJU International 159
Roehrborn et al. results of a randomized, placebo-controlled, double-blind study. J Sex Med 2012; 9: 271 81 9 Oelke M, Giuliano F, Mirone V, Xu L, Cox D, Viktrup L. Monotherapy with tadalafil or tamsulosin similarly improved lower urinary tract symptoms suggestive of benign prostatic hyperplasia in an international, randomised, parallel, placebo-controlled clinical trial. Eur Urol 2012; 61: 917 25 10 Roehrborn CG, McVary KT, Elion-Mboussa A, Viktrup L. Tadalafil administered once daily for lower urinary tract symptoms secondary to benign prostatic hyperplasia: a dose finding study. J Urol 2008; 180: 1228 34 11 Nickel JC, Brock GB, Herschorn S, Dickson R, Henneges C, Viktrup L. Proportion of tadalafil-treated patients with clinically meaningful improvement in lower urinary tract symptoms associated with benign prostatic hyperplasia - integrated data from 1499 study participants. BJU Int 2015; 115: 815 21 12 Rosen RC, Allen KR, Ni X, Araujo AB. Minimal clinically important differences in the erectile function domain of the International Index of Erectile Function scale. Eur Urol 2011; 60: 1010 6 13 Berry SJ, Coffey DS, Walsh PC, Ewing LL. The development of human benign prostatic hyperplasia with age. J Urol 1984; 132: 474 9 14 Porst H, Gacci M, Buttner H, Henneges C, Boess F. Tadalafil once daily in men with erectile dysfunction: an integrated analysis of data obtained from 1913 patients from six randomized, double-blind, placebo-controlled, clinical studies. Eur Urol 2014; 65: 455 64 15 Lepor H. Phase III multicenter placebo-controlled study of tamsulosin in benign prostatic hyperplasia. Tamsulosin Investigator Group. Urology 1998; 51: 892 900 16 Narayan P, Tewari A. A second phase III multicenter placebo controlled study of 2 dosages of modified release tamsulosin in patients with symptoms of benign prostatic hyperplasia. United States 93-01 Study Group. J Urol 1998; 160: 1701 6 17 Chapple CR, Montorsi F, Tammela TL, Wirth M, Koldewijn E, Fernandez Fernandez E. Silodosin therapy for lower urinary tract symptoms in men with suspected benign prostatic hyperplasia: results of an international, randomized, double-blind, placebo- and active-controlled clinical trial performed in Europe. Eur Urol 2011; 59: 342 52 18 Barry MJ, Williford WO, Chang Y et al. Benign prostatic hyperplasia specific health status measures in clinical research: how much change in the American Urological Association symptom index and the benign prostatic hyperplasia impact index is perceptible to patients? J Urol 1995; 154: 1770 4 19 van Leeuwen JH, Castro R, Busse M, Bemelmans BL. The placebo effect in the pharmacologic treatment of patients with lower urinary tract symptoms. Eur Urol 2006; 50: 440 52 20 Nickel JC. Placebo therapy of benign prostatic hyperplasia: a 25-month study. Canadian PROSPECT Study Group. Br J Urol 1998; 81: 383 7 21 Hansen BJ, Meyhoff HH, Nordling J, Mensink HJ, Mogensen P, Larsen EH. Placebo effects in the pharmacological treatment of uncomplicated benign prostatic hyperplasia. The ALFECH Study Group. Scand J Urol Nephrol 1996; 30: 373 7 22 Sech SM, Montoya JD, Bernier PA et al. The so-called placebo effect in benign prostatic hyperplasia treatment trials represents partially a conditional regression to the mean induced by censoring. Urology 1998; 51: 242 50 23 Seftel AD, de la Rosette JJ, Birt J, Porter V, Zarotsky V, Viktrup L. Coexisting lower urinary tract symptoms and erectile dysfunction: a systematic review of epidemiological data. Int J Clin Pract 2013; 67: 32 45 24 Brock G, Broderick G, Roehrborn CG, Xu L, Wong D, Viktrup L. Tadalafil once daily in the treatment of lower urinary tract symptoms (LUTS) suggestive of benign prostatic hyperplasia (BPH) in men without erectile dysfunction. BJU Int 2013; 112: 990 7 25 Brock GB, McVary KT, Roehrborn CG et al. Direct effects of tadalafil on lower urinary tract symptoms versus indirect effects mediated through erectile dysfunction symptom improvement: integrated data analyses from 4 placebo controlled clinical studies. J Urol 2014; 191: 405 11 Correspondence: Claus G. Roehrborn, Department of Urology, UT Southwestern Medical Center, 5323 Harry Hines Blvd, J8 142, Dallas, TX 75390-9110, USA. e-mail: Claus.roehrborn@utsouthwestern.edu Abbreviations: ED, erectile dysfunction; IIEF-EF, International Index of Erectile Function Erectile Function domain; PDE5, phosphodiesterase-5; BMI, body mass index. 160 BJU International 2016 BJU International