512 Background: W P is active, well tolerated and feasible with high efficacy alone or with T (Seidman et al. JCO 1998 and 2001). The primary aims of this trial were to determine whether: 1) W P improves response rate (RR) vs. S P, regardless of HER2 status and assignment to T, and 2) T added to P improves RR vs. P alone for HER2 normal tumors. Secondary objectives were to evaluate time to progression (TTP) and overall survival (OS) with respect to the above comparisons. Methods: Patients (pts) with measurable MBC, 0–1 prior chemotherapy regimens for locally advanced or MBC, ≥ 1 year from adjuvant taxane (if any) were stratified by line of therapy and randomized unequally (40:60) to S P 175 mg/m2 or W P 80 mg/m2. After accrual of the first 171 pts, when T became accepted as standard therapy for pts with HER2 positive MBC (Slamon et al NEJM 2001), HER2 status was assessed locally, pts with HER2 positive MBC received T and pts with HER2 normal MBC were randomized for T (standard weekly dosing). 158 pts treated with S P in the prior CALGB 9342 trial (Winer et al. JCO 2004) contributed data to the S P group in C9840 by pre-designed analyses. Results: 585 pts entered (488 first-line/97 second-line), and 577 were treated (n=735 with C9342 pts). Efficacy was analyzed as prescribed in adjusted multivariate models, controlling for relevant covariates including line of therapy and trastuzumab use. W P was superior to S P with respect to RR: 40% vs. 28%, (OR=1.61, p=0.017), and TTP: 9 vs. 5 mos, (HR=1.45, p = 0.0008). In this model, OS was 24 mos for W P vs. 16 mos for S P, (HR=1.19, p = 0.17). For pts with HER2 normal MBC, T did not improve RR: 35% vs. 29%, p=0.34, TTP: 7 vs. 6 mos, p=0.09, or OS: 22 vs. 20 mos, p=0.67. W P caused more grade 3 sensory/motor neuropathy (23/8% vs. 12/4%, p=0.001/0.04), and less grade ≥ 3 granulocytopenia (8 vs. 15%, p=0.013). 5 pts (0.8%) experienced significant cardiac dysfunction, 4 on T arms. There were 2 treatment related deaths, both due to pneumonia, on W P. Conclusions: W P is superior to S P in the management of MBC. T does not improve efficacy of P in HER2 normal MBC. Characterization of tumor tissue for HER2 (immunohistochemistry and fluorescent in situ hybridization) and other potential correlates of T sensitivity is ongoing centrally, as are analyses of serial changes in serum HER2 extracellular domain and quality of life. Author Disclosure Employment or Leadership Consultant or Advisory Stock Ownership Honoraria Research Funding Expert Testimony Other Remuneration Bristol-Myers Squibb Bristol-Myers Squibb