Background: Cetuximab plus chemotherapy is a first-line treatment option for metastatic RAS wild type colorectal cancer patients. Currently, no data are available on continuing cetuximab or changing bevacizumab as second-line therapy beyond first-line cetuximab-based chemotherapy. Methods: Patients (aged ≥18 years) with metastatic, histologically and genetically confirmed wild-type KRAS, NRAS and BRAF colorectal cancer progressing after first-line cetuximab plus chemotherapy were randomly assigned (1:1 ratio) to second-line chemotherapy with cetuximab (arm A) or with bevacizumab (arm B) 2·5 mg/kg per week equivalently. The choice between oxaliplatin-based or irinotecan-based second-line chemotherapy depended on the first-line regimen (switch of chemotherapy). The primary endpoints were progression-free survival (PFS) and objective response rate (ORR). The second endpoint was overall survival (OS). Results: 77 Patients (from July 1, 2016 to Sept 20, 2019, 77) were randomized (41 in arm A and 36 in arm B). ORR was 29.3% and 19.4% in Arm A and Arm B (p= 0.31). PFS was 7.2 months (95% CI 5.2–9.2) for Arm A and 5.9 months (95% CI 5.1–6.7) for Arm B (p= 0.677). OS was 18.5 months (95% CI 15.1–21.8) for Arm A and 17.5 months (95% CI 15.4–19·7) for Arm B (p= 0.444). Patients with ECOG PS 0 had significantly longer PFS and OS than ECOG PS 1 in second-line therapy whether cetuximab or bevacizumab combined with chemotherapy. ECOG 0 group vs ECOG 1 group, PFS was 8.7 months vs 4.6 months (p = 0.00) and OS was 21.2 months vs 12.3 months (p = 0.00). Moreover, ETS may predict efficacy of second-line continued cetuximab. The most frequently grade 3–4 adverse events in both arms were neutropenia (19.4% VS 16.7%), diarrhea (7.5% vs 11.1%), and nausea(10% vs 13.9%). Conclusions: Continuing cetuximab or changing bevacizumab plus standard second-line chemotherapy in patients with metastatic wild-type KRAS, NRAS and BRAF colorectal cancer after first-line cetuximab plus chemotherapy have similar clinical benefits. ECOG score is an independent predictor of prognosis and second-line treatment efficacy for colorectal cancer.