ABSTRACT
AIMS: Pathological response has been shown to be a predictor for survival after preoperative chemotherapy and surgical resection of colorectal cancer liver metastases. This retrospective analysis evaluated the effect on pathological response of adding bevacizumab to standard neoadjuvant chemotherapy in patients with metastatic colorectal cancer (mCRC) and liver metastases. METHODS: Patient records from two Spanish centres were retrospectively examined for this analysis. Patients were included if they had stage IV mCRC with liver metastases, were unresectable or marginally resectable tumour before chemotherapy, and had oxaliplatin- or irinotecan-based chemotherapy, with or without bevacizumab, before resection. Tumour response was evaluated using response evaluation criteria in solid tumours (RECIST). Pathological response was assessed by pathologists blinded to treatment. RESULTS: Ninety-five patients were included. Good pathological responses (PR0/PR1) were observed in 37 patients (39 %). The RECIST response rate was 51 %. Only 42 % of patients with a good pathological response had a complete or partial response according to RECIST, while 57 % of those with a poor pathological response had a complete or partial response according to RECIST. RECIST response rates were similar with and without bevacizumab, although 49 % of bevacizumab-treated patients had a good pathological response versus 27 % of those receiving chemotherapy alone (χ (2) P = 0.0302). CONCLUSION: Pathological response may be a better indicator of treatment efficacy than RECIST for patients with mCRC receiving bevacizumab in the neoadjuvant setting. Adding bevacizumab to chemotherapy has the potential to increase pathological response rates. Well-designed prospective clinical studies are required to establish the efficacy and tolerability of this approach.