Haie 1988.

Methods	Multicentre, 2‐armed parallel, RCT conducted by the Radiotherapy Cooperative Group of EORTC. Study duration: November 1977 to July 1981
Participants	441 previously untreated women with histologically confirmed carcinoma of the cervix with high risk of subclinical para‐aortic node metastases. 3 high‐risk groups of para‐aortic node metastasis applied in this study including: group 1: stages I and IIB proximal, i.e. with proximal vaginal or parametrial (or both) involvement, with histologically positive pelvic lymph nodes after surgery; group 2: stages I and IIB proximal without surgery and with positive pelvic lymph nodes on lymphangiogram; group 3: stages IIB distal, i.e. with distal vaginal or parametrial (or both) involvement and IIIb regardless of pelvic nodal status on lymphangiogram. Exclusion criteria: para‐aortic node involvement confirmed either by surgery or diagnosed on lymphangiogram; no lymphangiogram results or with a non‐interpretable lymphangiogram; urinary tract obstruction; higher risk of postoperative RT complications, i.e. previous abdominopelvic surgery or inflammatory bowel disease; previous history of other primary cancer, excluding basal cell carcinoma of the skin; pregnancy; and aged > 75 years.
Interventions	Control arm: pelvic RT. The upper limit of the field was the lower border of L4. Intervention arm: pelvic and para‐aortic RT. The pelvic field was extended to cover the para‐aortic nodes to the upper border of L1. RT technique: either 2 fields or 4 fields (box technique) for field configuration during EBRT. Source of beam energy was a megavoltage machine. The dose delivered to the pelvis was 40–50 Gy over 4–6 weeks using megavoltage machines. Further pelvic RT was given either by intracavitary brachytherapy or localised EBRT depending on each centre's practice. The external beam dose to the para‐aortic area was fixed at 45 Gy in 5 weeks. No chemotherapy administered.
Outcomes	Primary endpoint: 4‐year disease‐free survival Other outcomes: pelvic failure (no complete disappearance of all detectable pelvic disease or local recurrence); para‐aortic metastasis; distant metastasis other than para‐aortic metastasis; and 4‐year severe complication (grade 3 or 4) rate. However, this study did not report overall survival which is the most important outcome in research on cancer treatment.
Notes	Analyses based on an intention‐to‐treat basis. The distributions of important baseline characteristics of participants including participant age, histological subtypes, and FIGO stage were similar between groups.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Authors stated that order of treatment allocation was determined by computer‐generated random numbers.
Allocation concealment (selection bias)	Low risk	Authors stated that participants were randomised by drawing a sealed envelope.
Blinding of participants and personnel (performance bias) All outcomes	Low risk	No statement regarding blinding of participants and personnel; however, outcomes of interest were unlikely to be affected by lack of blinding of participants.
Blinding of outcome assessment (detection bias) All outcomes	Low risk	No statement regarding blinding of outcome assessor; however, outcomes of interest were unlikely to be affected by lack of blinding of outcome assessment. There was a well‐defined protocol for post‐treatment surveillance.
Incomplete outcome data (attrition bias) All outcomes	Low risk	214 (94%) participants in the pelvic RT group and 201 (94%) in the extended‐field RT group were followed for a minimum of 4 years and available for analysis of primary outcome.
Selective reporting (reporting bias)	High risk	No overall survival, quality of life, and cost‐effectiveness were reported which are important oncological outcomes.
Other bias	Low risk	Analyses based on an intention‐to‐treat basis.