Antiangiogenic therapy in recurrent breast cancer with lymphangitic spread to the chest wall: a randomized phase II trial of bevacizumab with sequential or concurrent oral vinorelbine and capecitabine

Objectives: To assess efficacy of bevacizumab in combination with oral chemotherapy in patients with breast cancer with lymphangitic spread to the chest wall (LBC). To identify surrogate biomarkers of response to bevacizumab. Patients and methods: We randomly assigned patients to receive bevacizumab plus either sequential or concurrent oral vinorelbine and capecitabine every 3 weeks. The primary endpoint was time to ultimate progression (TTP); the response rate and overall survival (OS) were secondary endpoints. We performed gene expression profiling on baseline tissue samples collected from triple negative LBC. We assessed circulating endothelial cells (CEC), circulating endothelial progenitors (CEP) and circulating pericyte progenitors (CPP). Results: A total of 66 patients were enrolled. There was no difference in TTP (median TTP 5.3 vs. 4.8 months, p=0.21) and in OS (median OS 15.8 vs 11.9 months; p=0.25) when comparing concurrent vs sequential treatment, respectively. Response rate was 25% vs 28% in the concurrent vs sequential arm (p=1.00), respectively. A set of 16 genes predictive of response to bevacizumab was identified. The counts of CEPs and viable CECs below the median value were associated with an improved overall survival: 26.6 vs 9.5 months for CEPs and 22.6 vs 11.0 months for viable CECs, respectively (p=0.02). Conclusions: Oral chemotherapy and bevacizumab (BEVIX) is an active regimen in patients with LBC. We support the importance of using LBC as a biological model for investigating angiogenesis inhibitors. CECs and CEPs biomarkers have been identified as predictive markers of outcome and warrant further investigation.