AUTHOR=Jiang Meilin , Yang Pei , Li Jing , Peng Wenying , Pu Xingxiang , Chen Bolin , Li Jia , Wang Jingyi , Wu Lin 

TITLE=Computed tomography-based radiomics quantification predicts epidermal growth factor receptor mutation status and efficacy of first-line targeted therapy in lung adenocarcinoma

JOURNAL=Frontiers in Oncology

VOLUME=Volume 12 - 2022

YEAR=2022

URL=https://www.frontiersin.org/journals/oncology/articles/10.3389/fonc.2022.985284

DOI=10.3389/fonc.2022.985284

ISSN=2234-943X

ABSTRACT=Background: Biomarkers that predict efficacy of first-line tyrosine kinase inhibitors (TKI) is pivotal in epidermal growth factor receptor (EGFR) mutant advanced lung adenocarcinoma. Imaging-based biomarkers have attracted much attention in anticancer therapy. This study aims to use machine learning method to distinguish EGFR mutation status and further explores the predictive role of EGFR mutation-related radiomic features in response to first-line TKIs. 

Methods: We retrospectively analyzed pretreatment CT images and clinical information from a cohort of lung adenocarcinomas. The top-ranked features were entered into support vector machine (SVM) classifier to establish a radiomic signature that predicted EGFR mutations status. Furthermore, we identified the best response-related features based on EGFR mutant-related features in first-line TKI therapy patients. Then we test and validate the predictive effect of the best response-related features for progression-free survival (PFS). 

Results: Six hundred ninety-two patients were enrolled in building radiomic signature. The 13 top-ranked features were input into a SVM classifier to establish radiomic signature of the training cohort (n=514), and the predictive score of the radiomic signature was assessed on an independent validation group with 178 patients and obtained an AUC of 74.13%, a F1-score of 68.29%, a specificity of 79.55%, an accuracy of 70.79%, and a sensitivity of 62.22%. More importantly, the skewness-Low (≤ 0.882) or 10th percentile-Low group (≤21.132) had a superior partial response (PR) rate than the skewness-High or 10th percentile-High group (p＜0.01). We also found that higher skewness (HR=1.722, p=0.001) were significantly associated with worse PFS. 

Conclusions: The radiomic signature can be used to predict EGFR mutations status. Skewness may contribute to the stratification of disease progression in lung cancer patients treated with first-line TKIs.