Fountzilas 2001.

Study characteristics
Methods	Accrual dates: October 1997 to May 1999 Sample size: 183 Number of centres: unknown (Greece) Randomisation method:done at central office and based on a random number list, stratification Baseline comparability: no significant imbalance reported
Participants	Female Age range: 26‐77 years (median 57.5 years) MBC First‐line chemotherapy
Interventions	Arm I: EP versus E→P Arm I (EP) epirubicin + paclitaxel Arm II (E→P) epirubicin followed by paclitaxel (dose dense)
Outcomes	Overall response rate Complete response rate (complete disappearance of all clinical symptoms and signs of disease for at least 4 weeks) Time to progression Overall survival (from start of chemotherapy to date of death or last follow‐up) Toxicity
Notes	Schema 2 Randomised phase 2 Follow‐up: minimum 6 months, maximum 84.5 months Median relative dose intensity in both groups was similar (96‐97) 80% completed planned combination therapy, 85% completed planned sequential therapy
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	1‐ adequate
Allocation concealment (selection bias)	Unclear risk	2‐ not reported
Blinding of participants and personnel (performance bias) All outcomes	Unclear risk	2‐ not reported
Blinding of outcome assessment (detection bias) All outcomes	Low risk	1‐ all imaging reviewed by 3 independent radiologists at the end of the study
Incomplete outcome data (attrition bias) All outcomes	Low risk	1‐ no missing data
Selective reporting (reporting bias)	Low risk	1‐ all pre‐specified outcomes are reported
Other bias	Low risk	1‐ treatment groups similar at baseline regarding important prognostic factors The trial was not stopped early The analysis was by intention to treat