Hormone therapy for premature ovarian insufficiency: randomised trial and long-term evaluation

RESEARCH QUESTION What is the most effective hormone treatment for women with premature ovarian insufficiency [POI] in the short and long term? BACKGROUND POI is a clinical syndrome defined by the loss of ovarian activity before the age of 40 years. Long-term hormone treatment, at least to the age of natural menopause, is recommended to reduce menopausal symptoms and reduce the risk of adverse long-term outcomes. Treatment can be either the combined oral contraceptive pill [COCP] or hormone replacement therapy [HRT]. METHODS DESIGN Randomised, open, parallel, superiority trial with internal pilot SETTING UK gynaecology clinics, both specialist and general STUDY POPULATION Women with a diagnosis of POI (as per NICE guidance) regardless of cause HEALTH TECHNOLOGIES BEING ASSESSED: COCP: 30µg ethinylestradiol, with any progestogen, as an extended regimen. HRT: continuous estradiol, either oral or transdermal, dose no less than 2mg orally or 50mcg patch or 2 pumps gel, transdermally. For women with a uterus, progesterone or progestogen, either oral, intrauterine or transdermal, continuous or cyclical. PRIMARY OUTCOME Bone mineral density [BMD] at 2 years from a standard lumbar spine DEXA scan SECONDARY OUTCOMES Short to medium term: 1. Individual domains and total score of MENQOL-Intervention questionnaire, at 3, 6 and 12 months then annually 2. BMD in hip at 2 years, and hip and lumbar spine at 1 and 5 years 3. Bone metabolism markers: a subset from 3 clinics, at 3 and 12 months 4. Cardiovascular markers: Weight at 3 and 12 months then annually, blood pressure at 3, 6, 12 months and then annually. Fasting lipids, glucose and insulin will be collected in a subset, at 3, 12 months and then annually 5. Sexual function at 3, 6, 12 months then annually 6. Work productivity at 3, 6, 12 months then annually 7. Pregnancy and outcome 8. Satisfaction with treatment 9. Change or cessation of treatment 10. Adverse events Long-term outcomes: 1. Individual domains and total score of MENQOL-I at 10 years 2. BMD in hip and lumbar spine at 10 years 3. BP and weight at 10 years 4. Pregnancy and outcome 5. Change or cessation of treatment 6. Bone fractures (any location) 7. Cardiovascular disease, cancer, cognitive/ neurological conditions, osteoporosis 8. Mortality SAMPLE SIZE Assuming a SD of 0.15 for lumbar spine BMD, 382 patients will be required to detect a between–group difference of 0.05g/cm2 (a small effect size of 0.333) with a 90% power, 5% 2-sided significance level and 1:1 allocation. Allowing for up to 20% loss to follow-up, will recruit 480 patients. TIMELINES The research will have 4 stages: Years 1-6 Randomised trial (6 months set-up, 36 months recruitment, 2 year follow-up, report on primary outcome and biomarkers) Years 6-9 Randomised trial (medium-term data) Years 9-14 Cohort study (long-term data) Years 14-24 Routine data (very long-term data) ANTICIPATED IMPACT AND DISSEMINATION The trial will ensure that young women receive the optimum initial therapy for relief of symptoms, quality of life and sexual function, and protection against long-term adverse consequences of oestrogen deficiency. Results will be disseminated via the NIHR library, scientific papers, and conference presentations, and communicated to guideline development groups and commissioners. The Daisy Network and Turner’s Syndrome Support Society UK will lead on patient involvement.