A phase II study of neoadjuvant bevacizumab plus capecitabine and concomitant radiotherapy in patients with locally advanced rectal cancer

To assess safety and activity of neoadjuvant bevacizumab, capecitabine and standard radiotherapy in locally advanced rectal cancer as well as potential predictive biomarkers. The multicentric phase II study enrolled 43 patients who received bevacizumab infusion (5 mg/kg) every 2 weeks for 4 cycles; oral capecitabine at 825 mg/m(2) twice a day for 5.5 weeks with external-beam irradiation (50.4 Gy in 28 fractions over 5.5 weeks). We determined certain biomarkers before and after therapy for correlation with response. Post-operative histologic examination revealed no residual cancer cells in 6 of the 43 patients (14%; 95% confidence limits 3.60-24.31%). In another 22 patients (51.2%) a varying percentage of cancer cells in residual areas of fibrosis/ necrosis was found, corresponding to Mandard TRG 2 or 3 classification. Tumor resection with negative circumferential margin was achieved in 38/40 (95%) operated patients. Sphincter-sparing surgery was obtained in 31 (72.1%) patients. Primary tumor and lymph nodes downstaging was observed in 15 (34.9%) and 16 (37.2%) cases, respectively. Neoadjuvant therapy was safe and well tolerated. The most frequent side effects were G1-2 diarrhea, proctitis, rectal bleeding and hypertension. No biomarker tested was significantly predictive of both pathological complete response and disease-free survival. Pre-treatment CD-34 vessel density, post-treatment Ki-67 labeling index and VEGFR-2 cancer cells expression significantly correlated with residual tumor area. The schedule of neoadjuvant therapy tested was safe and active. Pre-treatment vessel density by the panendothelial marker anti CD-34 antibody, post-treatment Ki-67 labeling index and VEGFR-2 expression were significantly associated to residual tumor area. The biomarkers correlations warrant further evaluation in prospective clinical trials.