Abstract
Background: High prevalence of EGFRm lung cancer was found in the Asian population. Preclinical data suggest that inflammatory cytokines activated by PM2.5 affected EGFRm clone expansion. Here, we explored the correlation between inflammatory markers and EGFRm NSCLC. Methods: Resected NSCLC patients (2016-2023) were enrolled. Tumor tissues and blood serum were retrieved from Ramathibodi tumor biobank. EGFR 19del and L858R mutations were performed by rt-PCR in cancerous tissue and dPCR in normal tissue in the same patient. NF-Kb and STAT3 protein signaling were measured by ELISA in both cancerous and normal tissue. Cytokines (IL-1ß, IL-6, IL-8, IL-10, IL-12 and TNF-α) were explored in serum by flow cytometry. Results: Among 140 patients, EGFRm prevalence was 58% in cancerous tissue but only 5% in normal tissue. NF-kB and STAT3 were statistically higher in cancerous tissue than normal tissue [NF-kB median O.D.=0.82 (IQR; 0.07-2.82) vs. 0.32 (IQR; 0.05-2.48), P < 0.001; STAT3 median O.D.=0.32 (IQR; 0.10-1.58) vs. 0.17 (IQR; 0.06-1.29, P < 0.001]. STAT3 was significantly increased in EGFRm compared to EGFRwt [median O.D.=0.36 (IQR; 0.234-0.592) vs. 0.23 (IQR; 0.158-0.409), OR = 11.09 (95% CI; 2.17-56.58), P = 0.004]. TNF-α, IL-10, and STAT3 in cancer cells were higher in EGFRm than EGFRwt (P = 0.003, 0.008, and < 0.001, respectively). None of cytokines was statistically different between EGFRm and EGFRwt patients. However, only STAT3 in cancer cells and non-smoker were associated with EGFRm NSCLC in multivariable analysis. Conclusion: Inflammation could be one of the pathogenesis of both NSCLC and EGFRm lung cancer as we demonstrated in our pilot study. STAT3 is a potentially inflammatory-predictive biomarkers. Larger cohort is needed.