Comparison of two bevacizumab containing regimens in the first-line therapy of HER-negative metastatic breast cancer

TURANDOT, set up by the Central European Cooperative Oncology Group, is the first prospective trial to compare bevacizumab combined with either paclitaxel or capecitabine in the management of HER2-negative metastatic breast cancer. The first efficacy results of this study, presented at the 2012 annual meeting of the European Society for Medical Oncology (ESMO), show that the study failed to demonstrate non-inferiority in overall survival for bevacizumab-capecitabine compared with bevacizumab-paclitaxel.

Many agents are currently available in the first-line treatment of HER2-negative locally advanced or metastatic breast cancer. However, for the moment there is no gold standard. Previously reported studies demonstrated significant improvements in progression-free survival and response rate when bevacizumab was added to weekly paclitaxel (E2100), or capecitabine (RIBBON-1). However, to date no head-to-head comparison of bevacizumab in combination with paclitaxel or capecitabine data are available. In the TURNADOT study, a total of 564 women with HER2-negative metastatic breast cancer, who received no prior chemotherapy for their metastatic disease, were randomised to receive either bevacizumab plus paclitaxel (bevacizumab 10mg/kg on day 1 and 15; paclitaxel 90mg/m on day 1, 8 and 15 q4w) (N=285) or bevacizumab in combination with capecitabine (bevacizumab 15mg/kg d1; capecitabine 1000mg/m bid d1–14 q3w) (N=279). Patients in the study were treated until disease progression or unacceptable toxicity. The primary objective of the trial was to demonstrate non-inferior overall survival with bevacizumab-capecitabine versus bevacizumab-paclitaxel. Secondary endpoints of the study included response rate, progression-free survival, safety and quality of life.

After a median follow-up of 19 months, the 1-year overall survival rate was 81% in the bevacizumab-paclitaxel arm compared to 79% for bevacizumab-capecitabine treated patients (HR[95%CI]: 1.04[−∞ to 1.69]; p=0.0593). Given the pre-set boundary for non-inferiority of p=0.00105, the primary endpoint of non-inferiority for overall survival has thus not been met. After 18 months and two years, the overall survival rates for bevacizumab-paclitaxel and bevacizumab-capecitabine were 68% and 70% and 60% and 55% respectively. The overall response rate was shown to be better with bevacizumab-paclitaxel compared to bevacizumab-capecitabine (44% vs. 27% p<0.0001). Also the progression-free survival was shown to be longer when bevacizumab-paclitaxel was used compared with bevacizumab-capecitabine (11.0 vs. 8.1 months, HR[95%CI]: 1.36[1.09-1.68]; p=0.0052).

The observed undesirable effects were consistent with the known safety profiles of bevacizumab, paclitaxel and capecitabine. The most common grade ≥3 undesirable effects were neutropenia (18%), peripheral neuropathy (14%) and leucopenia (7%) with bevacizumab–paclitaxel and hand-foot syndrome (16%), hypertension (6%) and diarrhoea (5%) with bevacizumab–capecitabine.

In summary, this preplanned interim analysis showed that the non-inferiority criterion has not been met but overall survival results do not indicate relevant differences. The final overall survival data of this trial are expected in 2014. Progression-free survival and response rate were shown to be better for bevacizumab-paclitaxel. The observed toxicity was different for both study arms, but was in line with what has been observed in previous trials with paclitaxel and capecitabine. As such, the treatment selection must be based on the patients’ individual priorities.


Zielinski C, Láng I, Inbar MJ et al. First efficacy results from the TURANDOT phase III trial comparing two bevacizumab (Bev) containing regimens as first0line therapy for HER2-negative metastatic breast cancer (MBC). Presented at ESMO 2012, Abstract 317O.


Speaker Christoph Zielinski


Prof. Christoph C. Zielinski, MD,
Medical University Vienna-General Hospital, Vienna, Austria


See: Keyslides

Back to Top