Bevacizumab after cetuximab failure in Kras wild-type metastatic colorectal carcinoma

Abstract

Introduction: Bevacizumab and cetuximab both improve treatment efficacy when given with chemotherapy for metastatic colorectal carcinoma (mCRC) in both first- and second-line setting. Cetuximab has enhanced efficacy in Kras wild-type tumor. However, inferior outcomes were reported in two phase III randomised controlled trials for the concurrent use of bevacizumab and cetuximab with chemotherapy. There is an urgent need to define the optimal sequence of use of these two agents. With regard to the pre-clinical data that increased VEGF expression is associated with acquired resistance to anti-EGFR antibody, we performed a retrospective analysis on the outcomes of patients who received bevacizumab-containing regimens after cetuximab failure in Kras wild-type mCRC.

Methods: From January 2006 to December 2011, patients who received bevacizumab-containing regimens for mCRC in our institution were reviewed. Patients were eligible for further analysis if all the following criteria were met: [1] Kras wild-type mCRC, [2] received chemotherapy and cetuximab as immediate prior treatment, [3] received chemotherapy and bevacizumab as the index line of treatment and [4] had imaging for response evaluation. Outcome measures included median progression-free survival (mPFS) and objective response rate (ORR). Targeted adverse events were recorded in accordance with the two prospective observational cohort studies, the BRiTE and BEAT study. Adverse events include gastrointestinal perforation [perforation, intra-abdominal abscess, and fistula], arterial thromboembolic events (ATE) [myocardial infarction, cerebrovascular accident, transient ischemic attack, and unstable angina], postoperative bleeding or wound-healing complications (POWHC), grade 3/4 bleeding, and hypertension requiring additional anti-hypertensive.

Results: Fifty patients received bevacizumab-containing regimens were reviewed and 18 of them met all the criteria for further analysis. The median age was 56.5 and 10 patients had metastasis in multiple organs. After a median follow up of 12.1 months, the mPFS for the whole group of patients was 26.3 weeks (95% CI 19.5 – 33.0 weeks) with an ORR of 38.9%. For the 8 patients who received bevacizumab-containing regimens at second line, 1 complete response and 3 partial responses were observed, giving an ORR of 50%. The mPFS was 27.4 weeks (95% CI 2.0-52.8 weeks). For the 10 patients treated at third line, 3 PRs were observed, giving an ORR of 30%. The mPFS was 23.9 weeks (95% CI 19.7 – 28.1 weeks). Regarding previous exposure to oxaliplatin, irinotecan and 5FU, patients who exposed to only two of these agents, compared with those who exposed to all three agents, were more likely to respond to bevacizumab-containing regimen after cetuximab failure (Odds ratio 5.0, 95% CI 0.44-56.6). 2 patients (11.1%) had hypertension that required additional anti-hypertensive and 1 patient died of bowel perforation. The event of bowel perforation was unlikely attributed to bevacizumab since the event occurred 78 days after the last dose of bevacizumab while the patient was receiving hypofractionated palliative radiotherapy to the pelvis. There was no case of ATE, POWHC and grade 3/4 bleeding.

Conclusion: In patients with Kras wild-type mCRC, bevacizumab-containing regimens after cetuximab failure have promising activity and manageable toxicity. Further study is needed to compare switching to bevacizumab versus cetuximab beyond progression in this specific cohort of patients.