2023 World Conference on Lung Cancer (Posters)-EP08.01. Subtyping EGFR-mutated NSCLC Patients with Adjuvant TKIs or Chemotherapy Using Deep Radiomic Signatures

EP08.01. Subtyping EGFR-mutated NSCLC Patients with Adjuvant TKIs or Chemotherapy Using Deep Radiomic Signatures - PDF(Abstract)

Back to course

Pdf Summary

Researchers presented findings on using deep radiomic signatures to subtype patients with epidermal growth factor receptor (EGFR)-mutated non-small cell lung cancer (NSCLC) and determine their response to adjuvant tyrosine kinase inhibitors (TKIs) or chemotherapy. The study included 194 patients with completely resected stage IB-III EGFR-mutated NSCLC. A deep learning model was developed to predict disease-free survival (DFS) for each type of treatment, and the deep hazard indexes dhT and dhC were used to cluster patients into four groups based on their DFS outcomes with TKIs and chemotherapy. <br /><br />The study found that 92.27% of patients had either 19del or L858R EGFR mutations, while 7.73% had other mutations. Median follow-up was 1540 days. Overall, there was no significant difference in median DFS between the TKI and chemotherapy groups. However, the clustering analysis revealed distinct subgroups with varying clinicopathological features. Cluster 1 had larger tumor diameters, higher T stage, more advanced TNM stage, and higher pathological grade. Cluster 4 had a higher frequency of unclassical EGFR mutations compared to cluster 3.<br /><br />In terms of DFS, cluster 3 patients had significantly better outcomes with TKIs compared to chemotherapy, while cluster 4 patients had better outcomes with chemotherapy. The study concludes that CT-based deep radiomic signatures can help identify EGFR-mutated NSCLC patients who are more likely to benefit from adjuvant chemotherapy or TKIs. These findings have implications for guiding personalized treatment decisions in this patient population.<br /><br />The presented figure illustrated the workflow of the study and showed the DFS of patients in each cluster treated with TKIs and chemotherapy. The keywords associated with this research are: non-small cell lung cancer, epidermal growth factor receptor, and radiomics. The study falls under the track of Local-Regional Non-small Cell Lung Cancer and is classified as a regular abstract.

Asset Subtitle

Yi-Fan Qi

Meta Tag

Speaker Yi-Fan Qi

Topic Local-Regional NSCLC: Biomarkers

Keywords

deep radiomic signatures

epidermal growth factor receptor

EGFR-mutated non-small cell lung cancer

adjuvant tyrosine kinase inhibitors

chemotherapy

disease-free survival

deep hazard indexes

cluster analysis

CT-based radiomic signatures

personalized treatment decisions