Background: Safety and efficacy of gemcitabine plus docetaxel (GD) and capecitabine plus docetaxel (CD) were compared in patients with metastatic breast cancer, where the alternate crossover monotherapy (GD→C or CD→G) was predetermined. Patients and methods: Patients were randomly assigned to 3-week cycles of either gemcitabine 1000 mg/m2 on days 1 and 8 plus docetaxel 75 mg/m2 on day 1 or capecitabine 1000 mg/m2 twice daily on days 1-14 plus docetaxel 75 mg/m2 day 1. Upon progression, patients received crossover monotherapy. Primary end point was time to progression (TtP). Secondary end points evaluated overall response rate (ORR), overall survival (OS), and adverse events (AEs). Results: Despite over-accrual of 475 patients, the trial matured with only 324 of 385 planned TtP events due to patient discontinuations. Human epidermal growth factor receptor 2 status was not captured in this study. More CD patients (28%) discontinued due to AEs than GD patients (18.0%, P = 0.009). TtP [hazard ratio (HR) = 1.101, 95% confidence interval (CI) 0.885-1.370, P = 0.387] and OS (HR = 1.031, 95% CI 0.830-1.280, P = 0.785) were not significantly different comparing GD and CD. ORR was not statistically different (P = 0.239) comparing GD (72 of 207, 34.8%) and CD (78 of 191, 40.8%). TtP, OS, and ORR were not significantly different comparing crossover groups. GD caused greater fatigue, hepatotoxicity, neutropenia, and thrombocytopenia but not febrile neutropenia; CD caused more hand-foot syndrome, gastrointestinal toxicity, and mucositis. Conclusions: GD and CD produced similar efficacy and toxicity profiles consistent with prior clinical experience. © The Author 2010. Published by Oxford University Press on behalf of the European Society for Medical Oncology. All rights reserved.