2155 - CT-Based Radiomics and Genomics Analysis for Survival Prediction in Stage III Unresectable Non-Small Cell Lung Cancer Treated with Definitive Chemoradiotherapy Following with Immunotherapy

Y. Geng¹, T. Yin^1,2, Y. Li¹, J. Yu³, and F. Teng¹; ¹Department of Radiation Oncology, Shandong Cancer Hospital and Institute, Shandong First Medical University and Shandong Academy of Medical Sciences, Jinan, Shandong, China, ²Cancer Center, Union Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, Hubei, China, ³Department of Radiation Oncology, Shandong Cancer Hospital and Institute, Shandong First Medical University and Shandong Academy of Medical Sciences, Jinan, China

Purpose/Objective(s): We aim to construct a radiogenomic map linking DNA-NGS sequencing-derived gene expression patterns with computed tomographic (CT) image signatures in patients with unresectable stage III non-small cell lung cancer (NSCLC), assessing its clinical utility. Materials/

Methods: A total of 134 patients with unresectable stage III NSCLC who underwent immunotherapy (IO) after chemoradiotherapy (CRT) or CRT alone were included to build radiomics model. Patients treated with IO after CRT were randomly split into cohort 1 (training set, n = 75, validation set, n = 26), and patients who were treated with CRT alone were enrolled in cohort 2 (testing set, n = 33). Radiomic features were derived from the region of the primary tumor on CT images. Least absolute shrinkage and selection operator (LASSO) regression was conducted to select optimal radiomic features, and Rad-score was calculated to predict progression-free survival (PFS). The C-index was used to evaluate the predictive performance and discriminatory ability of the prediction models. Based on the median Rad-score, patients were divided into high-risk and low-risk group separately. Baseline tumor tissue and peripheral blood samples were collected from patients in both the validation and testing sets prior to treatment initiation for NGS sequencing.
Results: The Radiomics model, integrated with 5 radiomics features, demonstrated good performance in predicting PFS, with C-index values of 0.66 and 0.69. The low-risk group exhibited significantly longer PFS than the high-risk group. Survival analysis based on NGS sequencing revealed that patients with wild-type Phosphatidylinositol 3-kinase (PI3K) signaling pathways (HR=0.33, 95% CI 0.15 to 0.72, p=0.004) and circulating tumor DNA (ctDNA) having a maximum somatic allele frequency (MSAF) of =1% (HR=0.22, 95% CI 0.08 to 0.60,p=0.001) showed significant survival benefits from IO. To investigate the association between imaging characteristics and gene signaling pathways, we analyzed and discovered significant enrichment of both the Hippo and the Yeast mating-type switching/sucrose non-fermenting (SWI-SNF) signaling pathways in the low-risk group compared to the high-risk group (p < 0.05).
Conclusion: This study introduced a radiomic model to predict PFS in stage III unresectable NSCLC patients undergoing definitive chemoradiotherapy followed by consolidation immunotherapy. Our analysis identifies potential biomarkers and therapeutic targets, offering valuable management strategies and novel insights into cancer biology.