Eeles 2015.
| Methods | Randomized controlled trial | |
| Participants |
Country: United Kingdom centres, Spain, Hungary Setting: the Clinical Trials and Statistics Unit at The Institute of Cancer Research had overall responsibility for trial co‐ordination, data collation, central statistical monitoring, and all final analyses. The patients were recruited from 17 United Kingdom centres, with the remaining 26 recruited from single centres in Spain (n = 14. 9%) and Hungary (n = 12.8%). Computer‐generated random permuted blocks were used and stratification by treating centres, menopausal status and FIGO stage to assign in a 1:1 ratio to receive HRT and no HRT, from Febuary 1990 to November 1995. Baseline characteristics: the 150 participants were randomized into HRT and no HRT (75 participants in each group). The median age was 58.7 years (range: 29.3 to 89.6 years). Most participants were 50 to 59 years old (33.3%), postmenopausal (77.3%), of serous type (39.3%), grade 3 (38.0%), FIGO stage III (54.7%), with no residual lesion (41.3%) and receiving single‐agent platinum (47.3%). Inclusion criteria: eligible participants were women who had been diagnosed with EOC (any FIGO stage) fewer than 9 months previously. Both premenopausal and postmenopausal women were eligible. Exclusion criteria: the women who needed to preserve ovarian function and had a history of hormone‐dependent malignancy or with any contraindications to HRT. |
|
| Interventions |
Intervention characteristics: HRT
Control: no HRT (N=75) The median time receiving hormone therapy was 1.14 years (IQR: 0.46 to 5.08 years). |
|
| Outcomes | The median follow‐up time was 19.1 years. There were participants who were assigned to HRT but did not receive or discontinued HRT during follow‐up (49 participants (65.3%)) and participants who were assigned to no HRT but received HRT during follow‐up (8 participants (10.7%)). However, all analyses were performed on an intention‐to‐treat basis. Death rate: 121 participants (81%) died: 53 (71%) in the HRT group and 68 (91%) participants in the no HRT group. Recurrent rate: 122 participants (81%) had recurrence: 54 (72%) in the HRT group and 68 (91%) in the no HRT group. Adverse event: the adverse event rate was low, with no statistically significant difference between two groups. The presence of TIA, CVA, MI, fracture and second primary cancer in the HRT group was 2.7%, 2.7%, 0%, 2.7%, and 5.3%, respectively; whereas in the no HRT group it was 0%, 4.0%, 2.7%, 5.3% and 4.0 %, respectively. Breast cancer was found two participants receiving HRT and one participant receiving no HRT. |
|
| Notes | Sponsorship source: supported by the Institute of Cancer Research (to RE and JMB), by core funding (C1491/A8895) to the Clinical Trials and Statistics Unit at the Institute of Cancer Research from Cancer Research UK, and by the National Institute for Health Research support (to RE) to the Biomedical Research Centre at The Institute of Cancer Research and The Royal Marsden NHS Foundation Trust. The trial was sponsored by The Royal Marsden NHS Foundation Trust and was conducted in accordance with the principles of good clinical practice. | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Quote: "Computer‐generated random permuted blocks were used; stratification was by treating centre, menopausal status (pre versus post), and FIGO stage (I and II versus III and IV)." |
| Allocation concealment (selection bias) | Low risk | Quote: "Participants were centrally randomly assigned in a 1:1 ratio to receive either AHT or no AHT (control). Independent random assignment was performed via telephone (or fax for international sites) at the Clinical Trials and Statistics Unit at The Institute of Cancer Research." |
| Blinding of participants and personnel (performance bias) All outcomes | High risk | Participants were due to start treatment within 2 weeks of random assignment and to continue their treatment for a minimum of 5 years, if tolerated. Treatment was non‐blinded, and no placebo was given to control‐group participants. |
| Blinding of outcome assessment (detection bias) All outcomes | Unclear risk | There was no information on who were outcome assessors and whether they knew the treatment group. |
| Incomplete outcome data (attrition bias) All outcomes | High risk | There was no participant loss to follow‐up but after random assignment for HRT, 3 (4%) participants denied receiving HRT and 46 (61.3%) participants discontinued HRT up to death or last follow‐up. |
| Selective reporting (reporting bias) | Low risk | Quote: "All analyses were performed on an intention‐to‐treat basis, and a two‐sided significance level of 5% and corresponding 95% CIs were used throughout. All analyses were performed with Stata13 (StataCorp, College Station, TX)." |
| Other bias | Low risk | We do not suspect any other source of bias. |