We evaluated the efficacy and safety of sequential therapy with trastuzumab monotherapy (H-mono) followed by H plus docetaxel (D) after disease progression (H --> H + D) versus combination therapy with H + D as first-line therapy. Patients with human epidermal growth factor receptor type 2 (HER2)-positive metastatic breast cancer (MBC) and left ventricular ejection fraction >50% were randomly assigned to either (a) H --> H + D [H, once weekly 2 mg/kg (loading dose, 4 mg/kg); D, once every 3 weeks 60 mg/m(2)] or (b) H + D. Primary endpoints were progression-free survival (PFS) for the H-mono stage of the H --> H + D group and H + D group and overall survival (OS) for both groups. Secondary endpoints were overall response rate, time to treatment failure, second PFS and safety. The planned number of patients was 160 patients in total. Of 112 patients enrolled, 107 were eligible. After 112 patients were enrolled, the Independent Data Monitoring Committee recommended stopping enrollment because PFS and OS were greater in the H + D group than the H --> H + D group. Median PFS was 445 days in the H + D group versus 114 days for H-mono in the H --> H + D group [hazard ratio (HR), 4.24; P < 0.01]. OS was significantly longer in the H + D group (HR, 2.72; P = 0.04). H + D therapy is significantly superior to H --> H + D therapy as first-line therapy in patients with HER2-positive MBC, especially in terms of OS.