Tadalafil Once Daily for Erectile Dysfunction
Tadalafil Once Daily for Erectile Dysfunction
This prospective, multicenter, longitudinal, naturalistic observational study (H6D-EW-LVIU) was conducted in Germany, France, Italy and Greece. Centres were invited sequentially from a randomised list of 205 candidate centres that had expressed interest in participation; 97 centres signed the contract, 59 actually enrolled patients. Patients were consecutively enrolled between November 2011 and June 2012. Adult male patients who met the criteria for ED according to the investigator and presented within the normal course of care were eligible to participate in the study, if they had decided together with their physician to either initiate PDE5-inhibitor treatment for the first time (treatment-naĂ¯ve) or to switch from any previous PDE5-inhibitor. Patients who had previously used tadalafil OaD were excluded. The observational nature of the study was confirmed by the appropriate ethical review boards in each participating country and patients provided written informed consent to data collection, storage and release of anonymised data.
Assessment and treatment of patients was based solely on the investigator's routine practice in the provision of care to ED patients.
At baseline (T1), data were collected for all patients (overall cohort). At two postbaseline observational time points (T2, T3), data were collected only for those patients who initiated or switched to tadalafil OaD at baseline (tadalafil OaD cohort). T2 and T3 data were collected at routine visits within 1–3 months and 4–6 months only for patients switching to or starting tadalafil OaD treatment. Patients who initiated or switched to tadalafil OaD at baseline but switched or discontinued this treatment during the observation were followed up until the end of the 6-month observation period. A telephone follow-up call was performed if a patient had no visit within 4–6 months after baseline.
The primary objective was to assess the time to treatment discontinuation of tadalafil OaD in men who had initiated or switched to tadalafil OaD at baseline (i.e. in the tadalafil OaD cohort). Time to discontinuation was measured as the number of days from start of tadalafil OaD treatment until switch to any other PDE5-inhibitor PRN or discontinuation of any PDE5-inhibitor treatment. Treatment breaks ≤ 6 weeks without switch to a different ED medication were not considered as discontinuation of tadalafil OaD. Patients lost to follow-up and without response to telephone follow-up were censored at the date of the last contact.
Secondary outcomes based on the data collected for the overall cohort include patient characteristics and reasons for the choice/switch of PDE5 inhibitor treatment at baseline. Secondary outcomes based on the longitudinal observation of the tadalafil OaD cohort include treatment switching patterns and reasons for switches or discontinuations, and the following EF assessments: (i) International Index of Erectile Dysfunction – Erectile Function (IIEF-EF) domain score, (ii) percent 'yes' responses to Global Assessment Questions (GAQ-1: 'Has the treatment you have been taking during this study improved your erections?'; GAQ-2: 'Has the treatment improved your ability to engage in sexual activity?').
Safety was evaluated by assessing treatment-emergent adverse events (TEAEs) reported during tadalafil OaD treatment (Medical Dictionary for Regulatory Activities; MedDRA version 15.0).
Sample Size. Assuming that 50–80% of patients who initiated tadalafil OaD at baseline would continue on tadalafil OaD until Month 6, a sample size of 200 patients in the tadalafil OaD cohort would allow an estimation of the continuation rate and the associated 95% confidence interval (CI) with an appropriate precision ranging between ± 5.5% and ± 6.9%. Accounting for 20% loss to follow-up, it was planned to enrol 250 patients initiating tadalafil OaD at baseline. Assuming that 10% of all patients enrolled would initiate tadalafil OaD at baseline, then for 250 tadalafil OaD patients to be enrolled into the 6-month observation period, a total of 2500 baseline data collection forms for patients initiating/switching to any PDE5-inhibitor would be needed. Assuming a questionnaire response rate of 85%, placement of 3000 questionnaires was required for a return of 2500 completed baseline data collection forms. Therefore, consecutive enrolment was planned to be stopped as soon as either 3000 patients were enrolled overall, or 600 patients were enrolled into the tadalafil OaD cohort.
Primary Analysis. All patients prescribed tadalafil OaD treatment at baseline were included in the analysis. The distribution of time to discontinuation of tadalafil OaD was estimated using the Kaplan–Meier product-limit method. The Kaplan–Meier proportions and associated 95% CIs of patients on tadalafil OaD at 2, 4 and 6 months were reported. An additional exploratory analysis investigated the association between time to discontinuation of tadalafil OaD and selected baseline factors using a Cox proportional hazards model; hazard ratios (HR) and the corresponding 95% CIs are reported. The final model was developed as follows: factors associated with treatment discontinuation were identified using backward selection (removed if p > 0.1). These factors were then reviewed to define a final model. The final model included baseline factors for presence or absence of relevant comorbidities, type of physician who initially diagnosed the ED, country, duration of living arrangement, age, ED aetiology, ED severity and work status. Additional sensitivity analyses were conducted to check if the goodness of fit based on the Akaike Information Criterion (AIC) could be improved by exchanging or deleting variables from the model. Also, these analyses were complemented by corresponding logistic regression analyses to investigate the association between continuation of tadalafil OaD treatment at T3, following the same procedure as for the Cox proportional hazards model except using forward selection (included if p ≤ 0.1).
Secondary Analyses. All patients who provided consent to release information and fulfilled the study entry criteria were included in the cross-sectional baseline data analysis. All patients prescribed tadalafil OaD at baseline were included in the longitudinal analyses. Reasons for initiating or switching to different PDE5-inhibitors at T1 were assessed descriptively. Logistic regression models were developed to investigate factors potentially associated with the initial choice of or switch to tadalafil OaD at baseline; odds ratios (ORs) and 95% CIs are reported. Factors associated with the choice of or switch to tadalafil OaD were initially identified using forward selection (included if p ≤ 0.1), and then reviewed to define the final model. The final model for the initial choice of tadalafil OaD included the factors body mass index (BMI), decreased duration of erection compared with pre-ED status, prior pelvic surgery, ED severity, duration of living arrangement, country, presence vs. absence of concomitant antihypertensive treatment, presence vs. absence of concomitant insulin treatment, type of physician who diagnosed the ED and education status. The final model investigating switches to tadalafil OaD included the factors age, country, ED aetiology, decreased libido compared with pre-ED status, ED severity, duration of living arrangement and work status.
Changes in IIEF-EF domain scores were assessed using a mixed model for repeated measures, including the following prespecified covariates as fixed effects: age (18–65 vs. > 65 years), benign prostate hyperplasia (BPH) (yes vs. no), cardiovascular disease (yes vs. no), country, diabetes (yes vs. no), dyslipidaemia (yes vs. no), aetiology of ED, pretreatment with PDE5-inhibitors (yes vs. no), ED severity and visit. A corresponding model was calculated including only those patients who remained on tadalafil OaD throughout the observation. GAQ- and safety data were evaluated descriptively.
In case of missing data, the number of patients with data available is given in the figures/tables. For Cox proportional hazards and logistic regression models, patients with missing data had to be excluded from the specific model(s) that needed the missing variables. Data were analysed using the SAS 9.2 software (SAS Institute Inc., Cary, NC).
Methods
Patients and Study Design
This prospective, multicenter, longitudinal, naturalistic observational study (H6D-EW-LVIU) was conducted in Germany, France, Italy and Greece. Centres were invited sequentially from a randomised list of 205 candidate centres that had expressed interest in participation; 97 centres signed the contract, 59 actually enrolled patients. Patients were consecutively enrolled between November 2011 and June 2012. Adult male patients who met the criteria for ED according to the investigator and presented within the normal course of care were eligible to participate in the study, if they had decided together with their physician to either initiate PDE5-inhibitor treatment for the first time (treatment-naĂ¯ve) or to switch from any previous PDE5-inhibitor. Patients who had previously used tadalafil OaD were excluded. The observational nature of the study was confirmed by the appropriate ethical review boards in each participating country and patients provided written informed consent to data collection, storage and release of anonymised data.
Assessment and treatment of patients was based solely on the investigator's routine practice in the provision of care to ED patients.
At baseline (T1), data were collected for all patients (overall cohort). At two postbaseline observational time points (T2, T3), data were collected only for those patients who initiated or switched to tadalafil OaD at baseline (tadalafil OaD cohort). T2 and T3 data were collected at routine visits within 1–3 months and 4–6 months only for patients switching to or starting tadalafil OaD treatment. Patients who initiated or switched to tadalafil OaD at baseline but switched or discontinued this treatment during the observation were followed up until the end of the 6-month observation period. A telephone follow-up call was performed if a patient had no visit within 4–6 months after baseline.
Outcome Measures
The primary objective was to assess the time to treatment discontinuation of tadalafil OaD in men who had initiated or switched to tadalafil OaD at baseline (i.e. in the tadalafil OaD cohort). Time to discontinuation was measured as the number of days from start of tadalafil OaD treatment until switch to any other PDE5-inhibitor PRN or discontinuation of any PDE5-inhibitor treatment. Treatment breaks ≤ 6 weeks without switch to a different ED medication were not considered as discontinuation of tadalafil OaD. Patients lost to follow-up and without response to telephone follow-up were censored at the date of the last contact.
Secondary outcomes based on the data collected for the overall cohort include patient characteristics and reasons for the choice/switch of PDE5 inhibitor treatment at baseline. Secondary outcomes based on the longitudinal observation of the tadalafil OaD cohort include treatment switching patterns and reasons for switches or discontinuations, and the following EF assessments: (i) International Index of Erectile Dysfunction – Erectile Function (IIEF-EF) domain score, (ii) percent 'yes' responses to Global Assessment Questions (GAQ-1: 'Has the treatment you have been taking during this study improved your erections?'; GAQ-2: 'Has the treatment improved your ability to engage in sexual activity?').
Safety was evaluated by assessing treatment-emergent adverse events (TEAEs) reported during tadalafil OaD treatment (Medical Dictionary for Regulatory Activities; MedDRA version 15.0).
Statistical Analysis
Sample Size. Assuming that 50–80% of patients who initiated tadalafil OaD at baseline would continue on tadalafil OaD until Month 6, a sample size of 200 patients in the tadalafil OaD cohort would allow an estimation of the continuation rate and the associated 95% confidence interval (CI) with an appropriate precision ranging between ± 5.5% and ± 6.9%. Accounting for 20% loss to follow-up, it was planned to enrol 250 patients initiating tadalafil OaD at baseline. Assuming that 10% of all patients enrolled would initiate tadalafil OaD at baseline, then for 250 tadalafil OaD patients to be enrolled into the 6-month observation period, a total of 2500 baseline data collection forms for patients initiating/switching to any PDE5-inhibitor would be needed. Assuming a questionnaire response rate of 85%, placement of 3000 questionnaires was required for a return of 2500 completed baseline data collection forms. Therefore, consecutive enrolment was planned to be stopped as soon as either 3000 patients were enrolled overall, or 600 patients were enrolled into the tadalafil OaD cohort.
Primary Analysis. All patients prescribed tadalafil OaD treatment at baseline were included in the analysis. The distribution of time to discontinuation of tadalafil OaD was estimated using the Kaplan–Meier product-limit method. The Kaplan–Meier proportions and associated 95% CIs of patients on tadalafil OaD at 2, 4 and 6 months were reported. An additional exploratory analysis investigated the association between time to discontinuation of tadalafil OaD and selected baseline factors using a Cox proportional hazards model; hazard ratios (HR) and the corresponding 95% CIs are reported. The final model was developed as follows: factors associated with treatment discontinuation were identified using backward selection (removed if p > 0.1). These factors were then reviewed to define a final model. The final model included baseline factors for presence or absence of relevant comorbidities, type of physician who initially diagnosed the ED, country, duration of living arrangement, age, ED aetiology, ED severity and work status. Additional sensitivity analyses were conducted to check if the goodness of fit based on the Akaike Information Criterion (AIC) could be improved by exchanging or deleting variables from the model. Also, these analyses were complemented by corresponding logistic regression analyses to investigate the association between continuation of tadalafil OaD treatment at T3, following the same procedure as for the Cox proportional hazards model except using forward selection (included if p ≤ 0.1).
Secondary Analyses. All patients who provided consent to release information and fulfilled the study entry criteria were included in the cross-sectional baseline data analysis. All patients prescribed tadalafil OaD at baseline were included in the longitudinal analyses. Reasons for initiating or switching to different PDE5-inhibitors at T1 were assessed descriptively. Logistic regression models were developed to investigate factors potentially associated with the initial choice of or switch to tadalafil OaD at baseline; odds ratios (ORs) and 95% CIs are reported. Factors associated with the choice of or switch to tadalafil OaD were initially identified using forward selection (included if p ≤ 0.1), and then reviewed to define the final model. The final model for the initial choice of tadalafil OaD included the factors body mass index (BMI), decreased duration of erection compared with pre-ED status, prior pelvic surgery, ED severity, duration of living arrangement, country, presence vs. absence of concomitant antihypertensive treatment, presence vs. absence of concomitant insulin treatment, type of physician who diagnosed the ED and education status. The final model investigating switches to tadalafil OaD included the factors age, country, ED aetiology, decreased libido compared with pre-ED status, ED severity, duration of living arrangement and work status.
Changes in IIEF-EF domain scores were assessed using a mixed model for repeated measures, including the following prespecified covariates as fixed effects: age (18–65 vs. > 65 years), benign prostate hyperplasia (BPH) (yes vs. no), cardiovascular disease (yes vs. no), country, diabetes (yes vs. no), dyslipidaemia (yes vs. no), aetiology of ED, pretreatment with PDE5-inhibitors (yes vs. no), ED severity and visit. A corresponding model was calculated including only those patients who remained on tadalafil OaD throughout the observation. GAQ- and safety data were evaluated descriptively.
In case of missing data, the number of patients with data available is given in the figures/tables. For Cox proportional hazards and logistic regression models, patients with missing data had to be excluded from the specific model(s) that needed the missing variables. Data were analysed using the SAS 9.2 software (SAS Institute Inc., Cary, NC).
Source...