Association of smoking status with non-small cell lung cancer patients harboring uncommon epidermal growth factor receptor mutation.

Introduction: Uncommon epidermal growth factor receptor (EGFR) mutations include single and complex mutations. However, the association of the smoking status of patients with uncommon and complex EGFR mutations remains unclear. Methods: This retrospective study evaluates the spectrum of uncommon EGFR mutations and investigates the influence of smoking status on the frequency of various uncommon EGFR mutations using a multi-institutional medical database. Results: Between 2010 and 2019, 5,608 non-small cell lung cancer (NSCLC) patients were analyzed. EGFR mutations were detected in 3,155 (56.3%) patients. Among the 399 (12.6%) patients with uncommon mutations, 198 had single uncommon and 201 complex mutations, including 87 exon 20 insertions, 79 de novo T790M, 70 complex common, and 52 complex uncommon mutations. For comparison, we also included 402 patients with common EGFR mutations. The percentage of ever-smokers was significantly higher in patients with uncommon EGFR mutations than in patients with common EGFR mutations (25.8% vs. 17.4%, p = 0.005). Furthermore, the percentage of ever-smokers was higher in those with a complex mutation than in those with a single uncommon mutation (30.3% vs. 21.2%, p = 0.040). Among patients carrying uncommon EGFR mutations, ever-smokers had significantly more complex uncommon EGFR mutations than never-smokers (22.3% vs. 9.8%, p = 0.002). Among patients carrying G719X, L861Q, and S768I, ever-smokers tended to have complex EGFR mutations more frequently than never-smokers (64.7% vs. 28.7%, 50.0% vs. 18.7%, 88.9% vs. 81.2%, respectively). Conclusions: Our study demonstrates not only a comprehensive spectrum of uncommon EGFR mutations, but also a positive relationship between smoking status and uncommon EGFR mutation frequency, especially complex uncommon EGFR mutations. The results suggest that smoking contributes to the development of complex EGFR mutations.