Castration was defined as serum oestradiol (E2) ≤184 picomoles/litre (pmol/L) or 50 picograms/millilitre (pg/mL).

The primary endpoint was evaluated based on centralised blinded bioanalysis of serum samples for E2. The percentage of subjects castrated and the 95% asymptotic confidence intervals (CIs), calculated from binomial distribution, are presented.

The per protocol set included all randomised subjects who received at least one dose of study medication with at least one baseline and at least one post-baseline assessment of the primary efficacy parameter, and without major protocol violations/deviations affecting the primary efficacy endpoint.

The percentages of subjects who were castrated at Weeks 4 and 8 where castration was defined as serum E2 ≤184 pmol/L or 50 pg/mL are presented. The 95% asymptotic CIs were calculated from the binomial distribution.

The full analysis set included all randomised subjects who received at least one dose of study medication with at least one baseline and at least one post-baseline assessment of the primary efficacy parameter.

The percentages of subjects who were castrated at Weeks 4, 8 and 12 where castration was defined as serum E2 ≤110 pmol/L or 30 pg/mL are presented. The 95% asymptotic CIs were calculated from the binomial distribution.

The full analysis set included all randomised subjects who received at least one dose of study medication with at least one baseline and at least one post-baseline assessment of the primary efficacy parameter.

Endometriosis-associated pelvic pain was assessed using a 100 millimetres (mm) visual analogue scale (VAS) where subjects indicated the subjective level of their most severe endometriosis pain over the last 4 weeks by making a single vertical mark on the line ranging from 'absence of pain' (0 mm) to 'unbearable pain' (100 mm). Lower scores indicated a better outcome. Baseline was defined as the last available assessment prior to the first dose of study medication. The least squares (LS) mean change from baseline at each timepoint as measured by the VAS is presented.

The full analysis set included all randomised subjects who received at least one dose of study medication with at least one baseline and at least one post-baseline assessment of the primary efficacy parameter.

The mean serum E2 concentrations at baseline and Weeks 4, 8 and 12 are presented.

The full analysis set included all randomised subjects who received at least one dose of study medication with at least one baseline and at least one post-baseline assessment of the primary efficacy parameter.

The mean FSH concentrations at baseline and Weeks 4, 8 and 12 are presented.

The full analysis set included all randomised subjects who received at least one dose of study medication with at least one baseline and at least one post-baseline assessment of the primary efficacy parameter.

The mean LH concentrations at baseline and Weeks 4, 8 and 12 are presented.

The full analysis set included all randomised subjects who received at least one dose of study medication with at least one baseline and at least one post-baseline assessment of the primary efficacy parameter.

Time to menses recovery was defined as the time (in days) between the date of the last dose of study medication and the date of the first day the subject observed menstrual bleeding of the next menstrual period. Menses recovery status was assessed at all study visits from Day 1 to the end of study visit. The median time to menses recovery was analysed using the Kaplan-Meier method.

The full analysis set included all randomised subjects who received at least one dose of study medication with at least one baseline and at least one post-baseline assessment of the primary efficacy parameter.