The phase II AVASTEM trial explored the impact of chemotherapy-bevacizumab combination on breast cancer stem cells in the neoadjuvant setting. We aimed to identify biological features associated with preoperative chemotherapy efficacy and prognosis by analyses of circulating tumor cells (CTCs) and bevacizumab pharmacokinetics (PK). The main objective was to assess the prognostic (relapse-free survival and overall survival) and predictive (pathological complete response, pCR) values of CTCs (CellSearch technology) and bevacizumab PK (ELISA). Seventy-five patients were included. Out of them 50 received bevacizumab-chemotherapy and 25 received chemotherapy alone. CTC results were available for 60 patients and PK data for 29 patients in the experimental arm. The absence of CTC at inclusion was correlated to better outcome. Five-years overall survival (OS) was 91% for CTC-negative patients vs. 54% for CTC-positive cases (HR = 6.21; 95%CI (1.75-22.06), = 0.001, log-rank test). Similar results were observed for RFS with 5 y-RFS of 78% vs. 44% (HR = 3.51; 95%CI (1.17-10.52), = 0.017, log-rank test). However, CTC status at baseline was not predictive of pCR ( = 0.74). CTC status after one cycle was not a significant prognostic factor (HR = 1.56; 95%CI (0.19-12.67); = 0.68 for OS and HR = 2.76; 95%CI (0.60-12.61); = 0.17 for RFS, log-rank test). Bevacizumab serum levels could not predict pCR and survival. PK values were not associated with treatment-related toxicities. In conclusion, CTCs detection at baseline is a prognostic marker for breast cancer receiving a neoadjuvant chemotherapy-bevacizumab combination independently of tumor response.