Epidermal growth factor receptor (EGFR)-tyrosine kinase inhibitors (TKIs) have been shown to improve the prognosis of EGFR-mutated (exon 19/21) non-small cell lung carcinoma (NSCLC). Positive EGFR mutation status is associated with NSCLC in non-smokers. Genetic and environmental factors have been linked to the etiology of EGFR mutations and sensitivity to EGFR-TKIs in non-smoking NSCLC patients. Cooking fume exposure (CFE) has also been proposed as an etiologic factor for NSCLC in non-smokers; however, the association of CFE with EGFR mutation status and EGFR-TKI response is unclear. The objective of this study was to determine the association between CFE and clinical response to EGFR-TKI therapy in NSCLC. The association of CFE, smoking history, occupational hazard exposure, tumor pathological type, EGFR mutation status, environmental exposure, living environment, and performance status with EGFR-TKI efficacy was determined in metastatic NSCLC patients who were treated with EGFR-TKIs (gefitinib or erlotinib). Objective response rate (ORR) and progression-free survival (PFS) were used to evaluate EGFR-TKI response. A total of 273 patients with a median age of 60.97 years (range 27-86 years) were included in this study. The proportion of patients receiving gefitinib and erlotinib was 72.53% (198/273) and 27.47% (75/273), respectively. ORRs (complete+partial responses) to gefitinib and erlotinib treatment were 20.70% (41/198) and 14.67% (11/75), respectively. Of the 273 patients, 98 (36.03%) had CFE and 112 (44.69%) had exposed to tobacco smoke. EGFR mutations were present in 55 patients, including exon 19 deletion (n=43) and exon 21 point mutations (n=12). Of the 55 EGFR mutation-positive patients, 52 (94.5%) had CFE. In the multivariate conditional logistic analysis, clinical response to EGFR-TKI was associated with non-smoking status, EGFR mutation, and CFE. Among these factors, CFE was the strongest predictor of EGFR-TKI response (odds ratio 13.66; 95% confidence interval (CI) 5.66-32.98; P<0.001). PFS was associated with a performance status of 0/1, adenocarcinoma pathological type, non-smoking status, EGFR mutation, and CFE. Among these, CFE was the most important factor for longer PFS (hazard ratio 0.37; 95% CI 0.26-0.52; P<0.001). The median PFS was 15.15 months in patients with CFE and 4.37 months in those without (P<0.0001). Knowledge of CFE history might be useful as a response predictor to EGFR-TKI treatment in NSCLC. Furthermore, CFE history might help to assess EGFR mutation status when genetic testing is not available.