Background: In PHEREXA (NCT01026142), adding P to H + X did not significantly improve independent review facility-assessed progression-free survival (IRF-PFS; primary endpoint) in patients with HER2-positive MBC who received a prior taxane and progressed during or after H-based therapy (hazard ratio [HR] 0.82; 95% confidence interval [CI] 0.65–1.02; p= .0731). An 8-month increase in median OS to 36.1 months was observed with P, but due to hierarchical testing of IRF-PFS, and subsequently of OS, statistical significance could not be claimed. No new safety signals were identified. We now report the final prespecified analysis. Methods: Randomization arms were A: intravenous H 8 mg/kg→6 mg/kg every 3 weeks (q3w) + oral X 1250 mg/m² twice daily (2 weeks on, 1 week off q3w) and B: intravenous P 840 mg→420 mg q3w + intravenous H per Arm A + oral X 1000 mg/m² (same schedule as Arm A). Treatment was given until disease progression, unmanageable toxicity, or patient request for discontinuation. OS and investigator-assessed PFS (INV-PFS) were assessed in the intent-to-treat population (all randomly assigned patients); adverse events (AEs), in the safety population (patients who received ≥1 dose of study drug). Results are descriptive. Results: At clinical cutoff (20-Sep-17), median time on study, including follow-up, was 23 months in Arm A and 33 months in Arm B. Efficacy is shown in the table. There was a small increase in AE incidence with longer follow-up, and no new symptomatic left ventricular systemic dysfunction (0 patients in Arm A and 5 [2.2%] in Arm B). Conclusions: While final OS results of PHEREXA are descriptive, median OS of 37.3 months in Arm B, with a 9.1-month increase versus Arm A, shows that clinical efficacy of H + P is maintained with longer follow-up. There were no new safety signals and no evidence of late cardiac toxicity. Clinical trial information: NCT01026142