Background: Circulating epithelial tumor cells (CTCs) in peripheral blood are an ideal source for the detection of disseminated tumor cells of an easy sampling procedure. Their prognosis significance has been demonstrated in metastasic breast carcinoma and have potencial to influence the clinical management for pts. with breast cancer (Cristofanilli, NEJM 2004). The antiangiogenic agent bevacizumab (Bev.), in combination with CT, improves progression free survival (PFS) of first line treatments, may modify tumor cell intravasation and CTC count, and may change CEC levels. Aims of this study are the evaluation of the prevalence and kinetics of CTCs and CECs before and after antiangiogenic treatment with Bev in pts with metastatic breast cancer.

Methods: Eligible pts. received Bev (10mg/kg q2w) combined with paclitaxel 150 mg/m2 and gemcitabine 2000mg/m2 d 1 y 15 q28d as first line therapy, until disease progression, unacceptable toxicity or withdrawal. For pts. participating in the sub-study, CTC and CEC were measured in 7.5ml of blood at baseline and after the first cycle of treatment. Samples were subjected to imnumomagnetic enrichment with an anti-EpCAM antibody and fluorescence labelled. CTCs were defined as nucleated cells (DAPI+) expressing cytokeratin 8, 18 and 19 but CD45 negative phenotype. CECs were defined as nuclear cells (DAPI+) expressing CD105 PE and CD45 negative phenotype. A sample was considered positive when 1 or more cells were detected.

Results: Data are available for 37 pts. We found ≥1 CTCs before first cycle of treatment with bev in 73% of the patients (N=27). After first treatment, reduction of CTCs was found in 57% of the patients (N=16). The median number of CTCs was 34 cells/7.5 ml (min 0-max 845) of blood in the first determination and 4.79 cells/7.5 ml (min 0-max 99) in the second determination, p=0.0075. In 38% of the pts (N=14) we found ≥5 baseline CTCs and after treatment <5 CTCs were found in 89% of the pts (N=25). In 70% of pts with baseline ≥5 CTCs count, a reduction to < 5CTCs was observed in the second determination (N=10), p=0.20 (IC 34.15-93.33). In 10 pts we found CTCs=0 baseline value (35%) and in the second determination after treatment CTCs=0 cells/7.5 ml was observed in 53%, p=0.058. Baseline CECs ≥1 was observed in 100% of the pts (N=31). After first cycle of treatment with bev plus CT CECs=0 was found in 1 patient (3.4%). In 70% of pts (N=14) there was a reduction of baseline CECs count, p=0.3. The median value of baseline CECs was 123 cells/7.5 ml (min 4- max 1407) and the median value in second determination was 54 cells/7.5 ml (min 0-max 349). The median of reduction of CECs after treatment was 70 CECs, p=0.02.

Conclusions: The addiction of bev to 1st line CT was related with high reduction of the value of baseline CTCs and CECs count, statistically significant correlation. Reduction to < 5 CTCs of patients with baseline ≥5 CTCs (unfavourable prognosis) was observed in 70% of patients. The results of this explorative study are preliminary and a large number of pts and follow-up is required.

Citation Information: Cancer Res 2010;70(24 Suppl):Abstract nr P3-02-11.