A pooled analysis of the secondary resection for colorectal cancer liver metastases: Comparison of bevacizumab (Bmab) and anti-EGFRs.

Abstract

610 Background: Systemic chemotherapies can convert irresectable liver metastases (LM) to resectable in some cases of metastatic colorectal cancer (mCRC). Secondary surgical resection is now considered to be a worthwhile therapeutic aim in mCRC with LM. There is a clear correlation between radiological response and secondary resection rate (SRR) (Jones RP et al. Eur J Cancer 2014). However, it remains unclear which monoclonal antibodies contribute the improvement of SRR, Bmab or anti-EGFRs. Methods: We performed a systematic review of literature published between Jan 1998 and Aug 2015 to analyze the relationship between SRR and antibodies in mCRC treated with 1st-line chemotherapy. Phase II or III trials which obtained the information of response rate (RR) and resection of LM were included in this analysis. The RR, SRR, and R0 SRR were compared between Bmab- and anti-EGFR (cetuximab or panitumumab)-containing regimen groups using Mann–Whitney U test. The correlation between RR and R0 SRR was assessed by Spearman's rank correlation coefficient. Results: A total of 40 studies were enrolled: 33 treatment arms from 16 trials for liver-limited disease (LLD) and 31 arms from 24 trials for non-LLD. The result is shown below. No difference was observed in SRR and R0 SRR between Bmab and anti-EGFR groups. In addition, a preliminary analysis showed that RR had significantly positive association with R0 SRR in anti-EGFR group (R2 0.74, P=0.027 for non-LLD; R2 0.88, P=0.018 for LLD) but not in Bmab group (P=0.27 for non-LLD, P=0.12 for LLD). Conclusions: The SRR and R0 SRR were comparable between Bmab and anti-EGFRs, even in mCRC with LLD, while RR correlated with R0 SSR in anti-EGFRs but not in Bmab. These findings warrant validation in ongoing trials to compare the targeted-agents. [Table: see text]