© 2018 Elsevier Ltd Background EGFR antibodies have shown promise in patients with advanced non-small-cell lung cancer (NSCLC), particularly with squamous cell histology. We hypothesised that EGFR copy number by fluorescence in-situ hybridisation (FISH) can identify patients most likely to benefit from these drugs combined with chemotherapy and we aimed to explore the activity of cetuximab with chemotherapy in patients with advanced NSCLC who are EGFR FISH-positive. Methods We did this open-label, phase 3 study (SWOG S0819) at 277 sites in the USA and Mexico. We randomly assigned (1:1) eligible patients with treatment-naive stage IV NSCLC to receive paclitaxel (200 mg/m2; every 21 days) plus carboplatin (area under the curve of 6 by modified Calvert formula; every 21 days) or carboplatin plus paclitaxel and bevacizumab (15 mg/kg; every 21 days), either with cetuximab (250 mg/m2weekly after loading dose; cetuximab group) or without (control group), stratified by bevacizumab treatment, smoking status, and M-substage using a dynamic-balancing algorithm. Co-primary endpoints were progression-free survival in patients with EGFR FISH-positive cancer and overall survival in the entire study population. We analysed clinical outcomes with the intention-to-treat principle and analysis of safety outcomes included patients who received at least one dose of study drug. This study is registered with ClinicalTrials.gov (number NCT00946712). Findings Between Aug 13, 2009, and May 30, 2014, we randomly assigned 1313 patients to the control group (n=657; 277 with bevacizumab and 380 without bevacizumab in the intention-to-treat population) or the cetuximab group (n=656; 283 with bevacizumab and 373 without bevacizumab in the intention-to-treat population). EGFR FISH was assessable in 976 patients and 400 patients (41%) were EGFR FISH-positive. The median follow-up for patients last known to be alive was 35·2 months (IQR 22·9–39·9). After 194 progression-free survival events in the cetuximab group and 198 in the control group in the EGFR FISH-positive subpopulation, progression-free survival did not differ between treatment groups (hazard ratio [HR] 0·92, 95% CI 0·75–1·12; p=0·40; median 5·4 months [95% CI 4·5–5·7] vs 4·8 months [3·9–5·5]). After 570 deaths in the cetuximab group and 593 in the control group, overall survival did not differ between the treatment groups in the entire study population (HR 0·93, 95% CI 0·83–1·04; p=0·22; median 10·9 months [95% CI 9·5–12·0] vs 9·2 months [8·7–10·3]). In the prespecified analysis of EGFR FISH-positive subpopulation with squamous cell histology, overall survival was significantly longer in the cetuximab group than in the control group (HR 0·58, 95% CI 0·36–0·86; p=0·0071), although progression-free survival did not differ between treatment groups in this subgroup (0·68, 0·46–1·01; p=0·055). Overall survival and progression-free survival did not differ among patients who were EGFR FISH non-positive with squamous cell histology (HR 1·04, 95% CI 0·78–1·40; p=0·77; and 1·02, 0·77–1·36; p=0·88 respectively) or patients with non-squamous histology regardless of EGFR FISH status (for EGFR FISH-positive 0·88, 0·68–1·14; p=0·34; and 0·99, 0·78–1·27; p=0·96; respectively; and for EGFR FISH non-positive 1·00, 0·85–1·17; p=0·97; and 1·03, 0·88–1·20; p=0·69; respectively). The most common grade 3–4 adverse events were decreased neutrophil count (210 [37%] in the cetuximab group vs 158 [25%] in the control group), decreased leucocyte count (103 [16%] vs 74 [20%]), fatigue (81 [13%] vs 74 [20%]), and acne or rash (52 [8%] vs one [<1%]). 59 (9%) patients in the cetuximab group and 31 (5%) patients in the control group had severe adverse events. Deaths related to treatment occurred in 32 (6%) patients in the cetuximab group and 13 (2%) patients in the control group. Interpretation Although this study did not meet its primary endpoints, prespecified subgroup analyses of patients with EGFR FISH-positive squamous-cell carcinoma cancers are encouraging and support continued evaluation of anti-EGFR antibodies in this subpopulation. Funding National Cancer Institute and Eli Lilly and Company.