Annals of Oncology

letters to the editor

Circulating tumor cells, colon cancer and bevacizumab: the meaning of zero
It has been recently observed that in metastatic breast cancer patients treated with bevacizumab, circulating tumor cells (CTC) are often undetected [1]. Due to the large use of bevacizumab as rst-line treatment for metastatic colorectal cancer (mCRC), we investigated the predictive value of CTC count in patients with mCRC treated with a rst-line chemotherapy plus bevacizumab compared with those treated with chemotherapy plus cetuximab. The predictive value of CTC was dened as the effect of change in CTC number after 23 months of treatment on response to therapies evaluated by RECIST criteria. CTC count was carried out through CellSearch system (Veriplex Corporation, Warren, NJ). In a group of 27 patients treated with bevacizumab, the median number of baseline CTC was 2.7 (range 037). After 612 weeks of treatment, CTC count dropped to 0 in 24 patients (89%) and 1 in 3 patients (11%). Surprisingly, 56% of patients with 0 CTC at follow-up evaluation had a progressive disease (PD) (Table 1, A).

Downloaded from annonc.oxfordjournals.org at Universidade Federal de S?o Paulo on August 20, 2011

Table 1. Correlation between change of CTC number (baseline and after 612 weeks of treatment) and clinical outcome in mCRC patients treated with chemotherapybevacizumab (A) and chemotherapycetuximab (B) (A) CTC count (bevacizumab) First blood draw 0 12 3 0 12 3 5 9 13 24 3 0 Imaging response SD/PR PD

612 weeks blood draw

(B) CTC count (cetuximab) First blood draw 0 12 3 0 12 3 4 8 8 3 11 6

10 14 2 1 0 0 Imaging response SD/PR PD

612 weeks blood draw

3 10 0

0 1 6

CTC, circulating tumor cells; mCRC, metastatic colorectal cancer; SD, stable disease; PR, partial response; PD, progressive disease.

Volume 22 | No. 8 | August 2011

letters to the editor | 1929

letters to the editor

In a group of 20 patients treated with cetuximab, the median number of baseline CTC was 2.6 (range 211). After 612 weeks of treatment, CTC count dropped to 0 in 3/20 (15%) patients; to 1 or 2 in 11/20 patients (55%) and in 6/20 patients (30%) remained persistently high (3). After 12 weeks of treatment, 6/6 (100%) patients with persistently high CTC count were found in PD, whereas 13/14 (93%) patients with decreased number of CTC were found with stable disease or partial response (Table 1, B). Similar to what observed in breast cancer, we suppose that the lack of predictive value of CTC in mCRC may depend on the use of bevacizumab. We hypothesize that bevacizumab may alter CTC biological features, thus preventing CellSearch from capturing them. Indeed, it is well recognized that bevacizumab, through the inhibition of tumor angiogenesis, leads to intratumoral hypoxia [2], which is a crucial trigger of tumor progression, invasive cell behavior and epithelial mesenchymal transition (EMT) [3]. Some provocative studies have also reported that vascular endothelial growth factor inhibitors, although efcient to reduce primary tumor growth, may promote tumor metastasis [4]. Thus, bevacizumab-induced hypoxia may generate in the primary tumor a selected population of cells undergoing EMT, which loose cellcell contact, become motile and acquire mesenchymal features. During this process, epithelial markers as EpCam and cytokeratins are often down-regulated, making these cells undetectable by CellSearch system. Thus, the meaning of 0 CTC in our patients with progression of disease may be simply due to underestimation of CTC by CellSearch. Recently, Mego et al. [5] for the rst time used the term undetectable CTC instead of negative CTC in breast cancer patients, suggesting that this status may be due to underestimation by CellSearch of CTC, which undergone EMT. Our results are in agreement with those from Tol et al. [6] who have reported that 80% of patients with mCRC with PD on computed tomography scan imaging have low CTC count at 12 weeks and concluded that CTC counts alone cannot accurately predict the clinical outcome. We believe that the lack of predictive value of CTC count in mCRC patients treated with bevacizumab may represent the rst example of drug-related undetectable CTC status and suggest caution when using CellSearch to evaluate CTC count as predictive factor for patients with mCRC receiving bevacizumab. P. Gazzaniga1*, C. Raimondi1, A. Gradilone1, M. Di Seri1, F. Longo1, E. Cortesi2 & L. Frati1
Department of Molecular Medicine, 2Division of Oncology, Department of Radiological, Oncological and Pathological Sciences, Sapienza University of Rome, Rome, Italy (*E-mail: paola.gazzaniga@uniroma1.it)
1

Annals of Oncology

references
1. Bidard FC, Mathiot C, Degeorges A et al. Clinical value of circulating endothelial cells and circulating tumor cells in metastatic breast cancer patients treated rst line with bevacizumab and chemotherapy. Ann Oncol 2010; 21: 17651771. 2. Kerbel R, Folkman J. Clinical translation of angiogenesis inhibitors. Nat Rev 2002; 2: 727739. 3. Peinardo H, Cano A. A hypoxic twist in metastasis. Nat Cell Biol 2008; 10: 253254. 4. Paez Ribes M, Allen E, Hudock J et al. Antiangiogenic therapy elicits malignant progression of tumors to increased local invasion and distant metastasis. Cancer Cell 2009; 15(3): 220231. 5. Mego M, De Giorgi U, Dawood S et al. Characterization of metastatic breast cancer patients with nondetectable circulating tumor cells. Int J Cancer 2010; doi:10.1002/ijc.25690. 6. Tol J, Koopman M, Miller MC et al. Circulating tumour cells early predict progression-free and overall survival in advanced colorectal cancer patients treated with chemotherapy and targeted agents. Ann Oncol 2010; 21(5): 10061012.

doi:10.1093/annonc/mdr292
Downloaded from annonc.oxfordjournals.org at Universidade Federal de S?o Paulo on August 20, 2011

funding
Fondazione per la Ricerca Oncologica (FO.R.O.onlus)

disclosure
The authors declare no conict of interest.

1930 | letters to the editor

Volume 22 | No. 8 | August 2011