Epidermal Growth Factor Receptor Mutations in Plasma DNA Samples Predict Tumor Response in Chinese Patients With Stages IIIB to IV Non–Small-Cell Lung Cancer

<jats:sec><jats:title>Purpose</jats:title><jats:p> Mutations in the epidermal growth factor receptor (EGFR) kinase domain can predict tumor response to tyrosine kinase inhibitors (TKIs) in non–small-cell lung cancer (NSCLC). However, obtaining tumor tissues for mutation analysis is challenging. We hypothesized that plasma-based EGFR mutation analysis is feasible and has value in predicting tumor response in patients with NSCLC. </jats:p></jats:sec><jats:sec><jats:title>Patients and Methods</jats:title><jats:p> Plasma DNA samples and matched tumors from 230 patients with stages IIIB to IV NSCLC were analyzed for EGFR mutations in exons 19 and 21 by using denaturing high-performance liquid chromatography. We compared the mutations in the plasma samples and the matched tumors and determined an association between EGFR mutation status and the patients' clinical outcomes prospectively. </jats:p></jats:sec><jats:sec><jats:title>Results</jats:title><jats:p> In 230 patients, we detected 81 EGFR mutations in 79 (34.3%) of the patients' plasma samples. We detected the same mutations in 63 (79.7%) of the matched tumors. Sixteen plasma (7.0%) and fourteen tumor (6.1%) samples showed unique mutations. The mutation frequencies were significantly higher in never-smokers and in patients with adenocarcinomas (P = .012 and P = .009, respectively). In the 102 patients who failed platinum-based treatment and who were treated with gefitinib, 22 (59.5%) of the 37 with EGFR mutations in the plasma samples, whereas only 15 (23.1%) of the 65 without EGFR mutations, achieved an objective response (P = .002). Patients with EGFR mutations had a significantly longer progression-free survival time than those without mutations (P = .044) in plasma. </jats:p></jats:sec><jats:sec><jats:title>Conclusion</jats:title><jats:p> EGFR mutations can be reliably detected in plasma DNA of patients with stages IIIB to IV NSCLC and can be used as a biomarker to predict tumor response to TKIs. </jats:p></jats:sec>