#4104
Background: Dose dense (dd) q2wk AC-paclitaxel (P) is superior to q3wk AC-P. Both regimens are associated with a <1% incidence of CHF. In MBC, q3wk nanoparticle albumin-bound paclitaxel (nab-P) is superior to q3wk P and bevacizumab (B) improves progression-free survival when added to P. Based on the MBC data, B may be more effective when targeting minimal residual disease in the adjuvant setting and nab-P may offer superior efficacy with reduced toxicity. However, it is uncertain whether AC plus B increases the risk of CHF. We evaluated the cardiac safety of ddAC-nab-P with concurrent B as adjuvant therapy for BC.
 Methods: Based upon the accepted cardiac event (CE) rate of approximately 4% in trials with adjuvant trastuzumab, this study was designed with similar monitoring and tolerability thresholds. The primary endpoint is cardiac safety: symptomatic CHF or death from LV dysfunction. Secondary endpoints are toxicity, DFS, OS, and biomarkers for efficacy and toxicity. Eligible pts have resected HER2(-) BC and normal LVEF. B is administered concurrently (10 mg/kg IV q2wk x 8) with chemotherapy (AC at 60/600 mg/m2 q2wk x 4 then nab-P at 260 mg/m2 q2wk x 4) and continues at 15 mg/kg q3wk thereafter for a total one year of B therapy. Pegfilgrastim is administered Day 2 of each chemotherapy cycle. Radiation and endocrine therapy are administered per standard of care. MUGAs are obtained at baseline and mos 2, 6, 9 and 18. Although asymptomatic LVEF declines are not considered CEs, B may be held per protocol and long-term cardiac monitoring initiated. If 3 CEs or > 1 cardiac death from LV dysfunction occur, B + ddAC-nab-P will not be considered safe.
 Results: The target accrual of 80 pts has been met. Median age is 47y (27-75). As of May 31, 2008, median follow-up is 14 mo (1-21) and 51 pts have completed 1y of planned therapy. No patients have developed symptomatic LV dysfunction at any point during study therapy. Sixteen pts have discontinued treatment due to toxicity: asymptomatic LVEF decline (N=3), hypertension (N=3), wound healing (N=3), headache (N=2), sensory neuropathy (N=2), pain (N=1), hypersensitivity reaction (N=1), and pneumonitis (N=1). Three pts had disease progression, and 8 pts withdrew consent.
 
 Conclusions: At the time of this analysis, no symptomatic LV dysfunction has been observed with B + ddAC-nab-P. Follow-up is ongoing. Correlative studies, including analysis of troponin, renin, and circulating endothelial and tumor cells, are underway.
Citation Information: Cancer Res 2009;69(2 Suppl):Abstract nr 4104.