Qualitative Computed Tomographic Features Predict Epidermal Growth Factor Receptor Mutations in Advanced Lung Adenocarcinoma

• Main Author: Chai, Chee Shee
• Format: Article
• Language: English
• Published: Elsevier 2021
• Subjects: R Medicine (General); RC Internal medicine; RC0254 Neoplasms. Tumors. Oncology (including Cancer)
• Online Access: http://ir.unimas.my/id/eprint/34903/1/abstract.pdf; http://ir.unimas.my/id/eprint/34903/; https://doi.org/10.1016/j.jtho.2021.01.1005

Introduction: Current guidelines recommend reflex testing for EGFR mutations in patients with advanced lung adenocarcinoma. Although tissue biopsy is considered the “gold standard” for EGFR mutation testing, it may not always be feasible in a small group of patients. Hence, this study aims to determine the computed tomography (CT) features that may predict the presence of common sensitising EGFR mutation in newly diagnosed advanced lung adenocarcinoma patients. Methods: 182 diagnostic contrast enhanced CT-images of newly diagnosed advanced lung adenocarcinoma patients who attended University Malaya Medical Center from 2010 to 2014 were analysed. For EGFR mutation testing, tissue biopsy specimens of all patients were tested for exon 19 deletion and exon 21 L858R point mutation. Results: Of the 182 patients, 158 patients with measurable primary tumour, were included in the analysis. 67 (42.4%) patients had common sensitising EGFR mutations, of which 43 (64.1%) were exon 19 deletion and 24 (35.9%) were exon 21 L858R point mutation. CT features of these patients are stated in Table 1. Common sensitising EGFR mutations were significantly more common in primary tumours with round contour (p ¼ 0.029), homogenous enhancement (p ¼ 0.029) and presence of air-bronchogram (p ¼ 0.029). On the other hand, presence of emphysema (p ¼ 0.005) and intrathoracic lymph nodes enlargement (p ¼ 0.001) were significantly against the presence of common sensitising EGFR mutations. These findings were further supported by the multivariate analysis: round contour [odd ratio (OR): 13.6, 95% confidence interval (CI): 1.40 - 13.2, p ¼ 0.024], homogenous enhancement (OR: 2.5, 95% CI: 1.00 - 6.47, p ¼ 0.049), presence of air bronchogram (OR: 2.7, 95% CI: 1.28 e 5.53, p ¼ 0.009) and intrathoracic lymph nodes enlargement (OR: 0.2, 95% CI: 0.05 - 0.52, p ¼ 0.002). None of the CT features could differentiate exon 19 deletion from exon 21 L858R point mutation. Conclusion: Tumours with round contour, homogenous enhancement and air bronchogram as well as absence of intrathoracic lymph nodes enlargement were independent predictors of the presence of common sensitising EGFR mutations. This information is helpful to clinicians in selecting best empirical treatment for patients who are not eligible for tissue biopsy and in those who need urgent treatment because of rapid clinical deterioration.