Abstract Study question How does the benefit/risk profile compare in women with endometriosis-associated pain who initiated relugolix combination therapy (Relugolix-CT) vs relugolix monotherapy with transition to Relugolix-CT? Summary answer Initiation with Relugolix-CT showed a more favourable benefit/risk profile than relugolix monotherapy: similar efficacy, lower rate of vasomotor symptoms and clinically significant bone loss. What is known already Treatment of endometriosis with a gonadotropin-releasing hormone antagonist monotherapy may be associated with vasomotor symptoms (e.g. hot flushes) and bone mineral density (BMD) loss. Relugolix-CT (relugolix 40 mg, estradiol 1 mg, norethisterone acetate 0.5 mg) was developed to minimise these hypoestrogenic adverse effects. In the pivotal SPIRIT 1 and 2 studies and long-term extension (LTE), once-daily Relugolix-CT demonstrated improved dysmenorrhoea early in treatment that was sustained for the duration of therapy over 104 weeks, with BMD loss <1%. This analysis evaluates the benefit of initiation with Relugolix-CT vs relugolix monotherapy and transition to Relugolix-CT on the 104-week benefit/risk profile. Study design, size, duration In the 24-week SPIRIT 1 and 2 studies, premenopausal women with endometriosis and moderate-to-severe pain at baseline were randomised 1:1:1 to placebo, Relugolix-CT, or delayed Relugolix-CT (relugolix 40 mg monotherapy then Relugolix-CT; 12 weeks each). Study completers were eligible to enrol in the open-label, 80-week LTE study where all women received once-daily Relugolix-CT. Efficacy and safety data over 104 weeks from the SPIRIT studies were assessed in this ad hoc analysis. Participants/materials, setting, methods Efficacy and safety were analysed as cumulative proportions, evaluating time to benefit and time to harm. Time to benefit was defined as median time to minimal-to-no dysmenorrhoea (Numerical Rating Scale score ≤1), time to harm as time to first vasomotor symptoms (hyperhidrosis, feeling hot, hot flush, night sweats and flushing), and time to lumbar spine BMD loss from baseline to > 3% among those meeting protocol-specified BMD criteria. Median time was not estimable for safety. Main results and the role of chance In total, 1251 women were treated in the SPIRIT 1 and 2 studies. Of 418 and 417 women in the Relugolix-CT and delayed Relugolix-CT groups, time to minimal-to-no dysmenorrhoea was 8 weeks in 56.3% and 75.7% of women, respectively, increasing to 82.5% and 86.4% at Week 24, 94.5% and 95.9% at Week 52, and 95.3% and 95.9% at Week 104. In both groups, median time to minimal/no dysmenorrhea-associated pain occurred within two menstrual cycles. Cumulative proportion of patients with first vasomotor symptoms in the Relugolix-CT and delayed Relugolix-CT groups was 4.1% and 16.8%, respectively, at Week 4, 13.5% and 35.3% at Week 24, 16.4% and 38.3% at Week 52, and 16.8% and 38.8% at Week 104. Cumulative proportion of patients with first BMD loss >3% in the Relugolix-CT and delayed Relugolix-CT groups was 1.1% and 5.9%, respectively, at Week 12, 10.2% and 23.5% at Week 24, 15.5% and 29.4% at Week 52, and 21.5% and 32.1% at Week 104. Given similar time to pain improvement with higher risk of vasomotor symptoms or clinically significant BMD loss with delayed Relugolix-CT treatment, these findings support the benefit of initiation of Relugolix-CT to mitigate hypoestrogenic adverse events. Limitations, reasons for caution The 80-week, open-label SPIRIT LTE did not have a control group. This evaluation was an ad hoc investigation of SPIRIT data and was not part of the prespecified statistical analysis plan. Comparisons between groups were qualitative, and no statistical inferences were generated. Wider implications of the findings Although Relugolix-CT and relugolix monotherapy showed similar efficacy benefit, monotherapy was associated with a higher risk of vasomotor symptoms and clinically significant BMD loss. The more favourable Relugolix-CT benefit/risk profile vs monotherapy helps fulfil an unmet need for long-term, effective, and well-tolerated hormonal treatment of symptoms of endometriosis. Trial registration number SPIRIT 1 (ClinicalTrials.gov: NCT03204318; EudraCT: 2017–001588–19); SPIRIT 2 (NCT03204331; 2017–001632–19); SPIRIT LTE (NCT03654274; 2017-004066-10)