Radiogenomic-based prediction of genetic alterations in oncogenic signaling pathways in lung cancer

Abstract Rationale and objectives Lung adenocarcinoma, a type of non-small cell lung cancer (NSCLC), originates from the peripheral lung tissue and is the most prevalent histologic subtype, constituting around 40% of cases. Understanding the intricate relationships among the signaling pathways and their changes in lung adenocarcinoma is crucial for developing targeted therapies and personalized treatment. This study aimed to analyze the utility of radiomics in assessing alterations in oncogenic signaling pathways in lung cancer. Materials and methods We investigated the use of radiomic features extracted from computed topography (CT) scan images of four lung adenocarcinoma patients along with alteration data retrieved from online cancer resource (cBioPortal) to generate radiomic signatures associated with oncogenic specific pathways. The combined dataset was used to identify significant features from the CT images. Results A total of 3408 radiomics features were extracted from CT images, and most significant features among the five pathways TP53, Hippo, Cell cycle, Myc, and Notch pathways were determined using PyRadiomics. Hypothesis testing identified 161, 91, 105, 89, and 9 radiomics features that significantly differed between alterations and no alterations in the TP53, Hippo, Cell cycle, Myc, and Notch pathways with ROC values of 0.82, 0.78, 0.75, 0.75, and 0.83, respectively. The Hippo pathway radiomic signatures exhibited the strong correlation with Myc and Cell cycle pathways. Conclusion This study highlights the potential of integrating radiomic features with genetic alteration data to identify significant oncogenic pathways. These findings provide insights that could aid in the development of more precise and effective treatment strategies in precision medicine to enhance our understanding of lung adenocarcinoma.