Self-reported erectile function over the past 4 weeks. IIEF- EF is the sum of Questions 1-5 and 15 of the IIEF. Questions 1-5 are scored 0 (low/no erectile function) to 5 (high erectile function) and Question 15 is scored 1 (very low confidence) to 5 (very high confidence), for a total score ranging from 1 to 30. Higher scores represent better erectile function. Data presented are the percentage of participants who return to normal erectile function (IIEF-EF domain score ≥26) at end of double-blind treatment period (Period III).

All randomized participants who received at least 1 dose of study drug, had a baseline and at least 1 post-baseline IIEF-EF measurement, excluding those with IIEF-EF domain score ≥26 ("normal") at baseline. Last observation carried forward (LOCF) principle was used.

Self-reported erectile function over the past 4 weeks. IIEF- EF is the sum of Questions 1-5 and 15 of the IIEF. Questions 1-5 are scored 0 (low/no erectile function) to 5 (high erectile function) and Question 15 is scored 1 (very low confidence) to 5 (very high confidence), for a total score ranging from 1 to 30. Higher scores represent better erectile function. Least Squares (LS) mean of the change from baseline to endpoint is from an analysis of covariance (ANCOVA). The model included centered-baseline as a covariate and factors of study, treatment group, and pooled site within study, and centered-baseline-by-treatment-group interaction.

All randomized participants who received at least 1 dose of study drug, had a baseline and at least 1 post-baseline IIEF-EF measurement, excluding those with IIEF-EF domain score ≥26 ("normal") at baseline. Last observation carried forward (LOCF) principle was used.

Self-reported intercourse satisfaction over the past 4 weeks. IIEF-IS is the sum of Questions 6, 7 and 8 of the IIEF. Scores range from 0 (low/no satisfaction) to 5 (high satisfaction) for each question, with the total possible score for the 3 questions of 0 to 15. Least Squares (LS) mean of the change from baseline to endpoint is from an analysis of covariance (ANCOVA). The model included centered-baseline as a covariate and factors of study, treatment group, and pooled site within study, and centered-baseline-by-treatment-group interaction.

All randomized participants who received at least 1 dose of study drug, had a baseline and at least 1 post-baseline IIEF-IS measurement, excluding those with International Index of Erectile Function - Erectile Function (IIEF-EF) domain score ≥26 ("normal") at baseline. Last observation carried forward (LOCF) principle was used.

Self-reported overall satisfaction over the past 4 weeks. IIEF-OS is the sum of Questions 13 and 14; each question scored as 1 (low/no satisfaction) through 5 (high satisfaction) with total subscore for the 2 questions of 2 to 10. Least Squares (LS) mean of the change from baseline to endpoint is from an analysis of covariance (ANCOVA). The model included centered-baseline as a covariate and factors of study, treatment group, and pooled site within study, and centered-baseline-by-treatment-group interaction.

All randomized participants who received at least 1 dose of study drug, had a baseline and at least 1 post-baseline IIEF-OS measurement, excluding those with International Index of Erectile Function - Erectile Function (IIEF-EF) domain score ≥26 ("normal") at baseline. Last observation carried forward (LOCF) principle was used.

Assessed was the mean change from baseline in the percentage of Yes responses to the SEP diary Questions 1-5. Least Squares (LS) mean of the change from baseline to endpoint is from an analysis of covariance (ANCOVA). The model included centered-baseline as a covariate and factors of study, treatment group, and pooled site within study, and centered-baseline-by-treatment-group interaction.

All randomized participants who received at least 1 dose of study drug, had a baseline and at least 1 post-baseline SEP Questions assessment, excluding those with International Index of Erectile Function - Erectile Function (IIEF-EF) domain score ≥26 ("normal") at baseline. Last observation carried forward (LOCF) principle was used.

The TSS measured participant satisfaction with treatment based on a 13-item questionnaire. The overall score for each of five TSS domains (confidence to complete sexual activity, ease of erection, pleasure from sexual activity, erectile function satisfaction, and satisfaction with orgasm) was converted to a 0-100 scale with higher numbers on the scale indicating greater satisfaction. Least Squares (LS) mean of the change from baseline to endpoint is from an analysis of covariance (ANCOVA). The model included centered-baseline as a covariate and factors of study, treatment group, and pooled site within study, and centered-baseline-by-treatment-group interaction.

All randomized participants who received at least 1 dose of study drug, had a baseline and at least 1 post-baseline TSS measurement, excluding those with International Index of Erectile Function - Erectile Function (IIEF-EF) domain score ≥26 ("normal") at baseline. Last observation carried forward (LOCF) principle was used.

The TSS - patient satisfaction with medication measured participant satisfaction with treatment based on a 13-item questionnaire. The overall score was converted to a 0-100 scale with higher numbers on the scale indicating greater satisfaction. Satisfaction with medication was analyzed using analysis of covariance (ANOVA). The model included factors of study, treatment group, and pooled site within study.

All randomized participants who received at least 1 dose of study drug, had a baseline and at least 1 post-baseline participant's satisfaction with medication measurement, excluding those with IIEF-EF domain score ≥26 ("normal") at baseline. Last observation carried forward (LOCF) principle was used.

Self-reported erectile function over the past 4 weeks. Question 15, confidence in the ability to get an erection, is scored from 1 (very low confidence) to 5 (very high confidence). Least Squares (LS) mean of the change from baseline to endpoint is from an analysis of covariance (ANCOVA). The model included centered-baseline as a covariate and factors of study, treatment group, and pooled site within study, and centered-baseline-by-treatment-group interaction.

All randomized participants who received at least 1 dose of study drug, had a baseline and at least 1 post-baseline IIEF question 15 (Sexual Confidence) assessment, excluding those with IIEF-EF domain score ≥26 ("normal") at baseline. Last observation carried forward (LOCF) principle was used.

The TSS measured participant's partner satisfaction with treatment based on a 13-item questionnaire. The overall score for each of five TSS domains (confidence to complete sexual activity, ease of erection, pleasure from sexual activity, erectile function satisfaction, and satisfaction with orgasm) was converted to a 0-100 scale with higher numbers on the scale indicating greater satisfaction. Least Squares (LS) mean of the change from baseline to endpoint is from an analysis of covariance (ANCOVA). The model included centered-baseline as a covariate and factors of study, treatment group, and pooled site within study, and centered-baseline-by-treatment-group interaction.

All randomized participants who received at least 1 dose of study drug, had a baseline and at least 1 post-baseline partner's TSS measurement, excluding those with IIEF-EF domain score ≥26 ("normal") at baseline. Last observation carried forward (LOCF) principle was used.

The TSS - partner satisfaction with medication measured participant's partner satisfaction with treatment based on a 13-item questionnaire. The overall score was converted to a 0-100 scale with higher numbers on the scale indicating greater satisfaction. Satisfaction with medication was analyzed using analysis of covariance (ANOVA). The model included factors of study, treatment group, and pooled site within study.

All randomized participants who received at least 1 dose of study drug, had a baseline and at least 1 post-baseline partner's satisfaction with medication measurement, excluding those with IIEF-EF domain score ≥26 ("normal") at baseline. Last observation carried forward (LOCF) principle was used.

Self-reported erectile function over the past 4 weeks. IIEF- EF is the sum of Questions 1-5 and 15 of the IIEF. Questions 1-5 are scored 0 (low/no erectile function) to 5 (high erectile function) and Question 15 is scored 1 (very low confidence) to 5 (very high confidence), for a total score ranging from 1 to 30. Higher scores represent better erectile function. Data presented are the percentage of participants who return to normal erectile function (IIEF-EF domain score ≥26) at end of open-label extension treatment period (Period IV).

All participants who were entered the open-label treatment, received at least 1 dose of study drug during open-label treatment, had Week 12 and at least 1 IIEF-EF measurement during open-label treatment, excluding those with IIEF-EF domain score ≥26 ("normal") at baseline. Last observation carried forward (LOCF) principle was used.

Self-reported erectile function over the past 4 weeks. IIEF- EF is the sum of Questions 1-5 and 15 of the IIEF. Questions 1-5 are scored 0 (low/no erectile function) to 5 (high erectile function) and Question 15 is scored 1 (very low confidence) to 5 (very high confidence), for a total score ranging from 1 to 30. Higher scores represent better erectile function.

All participants who were entered the open-label treatment, received at least 1 dose of study drug during open-label treatment, had Week 12 and at least 1 IIEF-EF measurement during open-label treatment, excluding those with IIEF-EF domain score ≥26 ("normal") at baseline. Last observation carried forward (LOCF) principle was used.

Assessed the percentage of Yes responses to the SEP diary Question 3 "Did your erection last long enough for you to have successful intercourse?" from Week 12 (end of double-Blind treatment) to Week 16 (end of open-label treatment).

All participants who were entered the open-label treatment, received at least 1 dose of study drug during open-label treatment, had Week 12 and at least 1 SEP diary Question 3 measurement during open-label treatment, excluding those with IIEF-EF domain score ≥26 ("normal") at baseline. Last observation carried forward (LOCF) principle was used.