2024 World Conference on Lung Cancer (WCLC) - ePosters-EP.08C.03 To Construct Multi-Omics Modelfor Predicting Prognosis of Definitive Chemoradiotherapyin Locally Advanced Non-Small Cell Lung Cancer

EP.08C.03 To Construct Multi-Omics Modelfor Predicting Prognosis of Definitive Chemoradiotherapyin Locally Advanced Non-Small Cell Lung Cancer

Back to course

Pdf Summary

The study by Li Li and Shuanghu Yuan explores the development of a multi-omics predictive model for assessing the prognosis of definitive chemoradiotherapy (dCRT) in patients with locally advanced non-small cell lung cancer (LA-NSCLC). Conducted at the Department of Radiation Oncology in Shandong Cancer Hospital and The First Affiliated Hospital of USTC, the research integrates clinical data, CT images, and genetic information.<br /><br />The retrospective study involved 105 stage III LA-NSCLC patients who had undergone dCRT. The research utilized 3D-Slicer software to delineate the primary tumor on CT images and extract radiomics features, followed by the Least Absolute Shrinkage and Selection Operator (LASSO) for feature selection. Genomic data were derived from baseline tumor tissue samples. <br /><br />The performance of this comprehensive model was evaluated using the area under the curve (AUC) of the Receiver Operating Characteristic (ROC) and the concordance index (C-index). The median follow-up period was 30.1 months, and the median progression-free survival (PFS) was noted as 10.60 months. Four features were key to constructing the radiomics model. The study identified Rad-score, along with KEAP1 and MET mutations, as significant prognostic factors for PFS. <br /><br />The results indicated that the C-index of the radiomics, genomics, and combined radiogenomics models performed favorably in both training and validation sets, suggesting that the radiogenomics model may surpass the other two in predicting dCRT prognosis for LA-NSCLC patients. The study concludes that such integrated models could significantly enhance predictive accuracy, aiding in personalized cancer treatment strategies.

Asset Subtitle

Li Li

Meta Tag

Speaker Li Li

Topic Local-Regional Non-small Cell Lung Cancer

Keywords

multi-omics predictive model

definitive chemoradiotherapy

locally advanced non-small cell lung cancer

radiomics features

LASSO feature selection

genomic data

Receiver Operating Characteristic

concordance index

Rad-score

personalized cancer treatment