Background: In the GERCOR-DREAM trial, maintenance therapy (MT) with bev + EGFR tyrosine kinase inhibitor erlotinib (E) after a first-line Bev-based induction therapy (IT) in pts with mCRC significantly improved PFS compared with bev alone. Here we explore the influence of KRAS status on erlotinib efficacy. Methods: Pts with previously untreated and unresectable mCRC were eligible. After a Bev-based IT with FOLFOX or XELOX or FOLFIRI, pts without disease progression were randomized to MT between Bev alone (Bev 7.5 mg/kg q3w; arm A) or Bev+E (Bev 7.5 mg/kg q3w, E 150 mg/day continuously; arm B). KRAS determination was established by local assessment in each center. Results: Among the 452 randomized patients, KRAS status was available in 403 pts (89%): 234 pts (58%) KRAS wt and 169 pts (42%) KRAS mut. Clinical characteristics were similar between both populations. For the whole population of randomized patients (n=452), median PFS from inclusion were 9.33m and 10.55m in arm A and B, respectively (HR=0.76 [0.61-0.94], p=0.011). For KRAS wt population, median PFS from inclusion was 9.66 m and 10.94 m in arm A and B, respectively (HR=0.80 [0.59-1.08], p=0.141). For KRAS mut population, median PFS from inclusion was 9.79 m and 9.79 m in arm A and B, respectively (HR=0.86 [0.61-1.22], p=0.393). In KRAS wt pts treated with erlotinib, cutaneous toxicity was predictive of PFS: mPFS was 9.66m in pts with grade 0 (n=101) and 10.91m in pts with grade ≥1 (n=114) (HR=0.69 [0.51-0.95], p=0.0186). Conclusions: The addition of erlotinib to bevacizumab as maintenance treatment in first-line metastatic colorectal cancer significantly improves progression-free survival from inclusion. However, in both wt and mut KRAS pts, difference was not statistically significant. Unlike anti-EGFR monoclonal antibodies, the addition of erlotinib to bevacizumab does not appear to be antagonist in KRAS mutant patients. Clinical trial information: NCT00265824.