Purpose To assess the efficacy and safety of trastuzumab plus capecitabine with or without pertuzumab in patients with human epidermal growth factor receptor 2-positive metastatic breast cancer who experienced disease progression during or after trastuzumab-based therapy and received a prior taxane. Patients and Methods Patients were randomly assigned to arm A: trastuzumab 8 mg/kg → 6 mg/kg once every 3 weeks plus capecitabine 1,250 mg/m2 twice a day (2 weeks on, 1 week off, every 3 weeks); or arm B: pertuzumab 840 mg → 420 mg once every 3 weeks plus trastuzumab at the same dose and schedule as arm A plus capecitabine 1,000 mg/m2 on the same schedule as arm A. The primary end point was independent review facility-assessed progression-free survival (IRF PFS). Secondary end points included overall survival (OS) and safety. Hierarchical testing procedures were used to control type I error for statistical testing of IRF PFS, OS, and objective response rate. Results Randomly assigned (intent-to-treat) populations were 224 and 228 patients in arms A and B, respectively. Median IRF PFS at 28.6 and 25.3 months' median follow-up was 9.0 v 11.1 months (hazard ratio, 0.82; 95% CI, 0.65 to 1.02; P = .0731) and interim OS was 28.1 v 36.1 months (hazard ratio, 0.68; 95% CI, 0.51 to 0.90). The most common adverse events (all grades; incidence of ≥ 10% in either arm and ≥ 5% difference between arms) were hand-foot syndrome, nausea, and neutropenia in arm A, and diarrhea, rash, and nasopharyngitis in arm B. Conclusion The addition of pertuzumab to trastuzumab and capecitabine did not significantly improve IRF PFS. An 8-month increase in median OS to 36.1 months with pertuzumab was observed. Statistical significance for OS cannot be claimed because of the hierarchical testing of OS after the primary PFS end point; however, the magnitude of OS difference is in keeping with prior experience of pertuzumab in metastatic breast cancer. No new safety signals were identified.