CT-based decision tree model for predicting EGFR mutation status in synchronous multiple primary lung cancers

Area of Science:

• Radiology
• Oncology
• Medical Informatics

Background:

• Synchronous multiple primary lung cancers (SMPLCs) present diagnostic challenges.
• Epidermal growth factor receptor (EGFR) mutation status is crucial for targeted therapy in lung cancer.
• Predictive models can aid in clinical decision-making for SMPLC patients.

Purpose of the Study:

• To develop and validate a computed tomography (CT)-based decision tree algorithm (DTA) model.
• To predict the epidermal growth factor receptor (EGFR) mutation status in patients with SMPLCs.
• To assess the utility of the CT-DTA model in clinical practice.

Main Methods:

• Retrospective review of demographic and CT findings in 85 surgically resected SMPLC patients.
• Least Absolute Shrinkage and Selection Operator (LASSO) regression for predictor selection.
• Development and validation of a CT-DTA model using multivariate logistic regression and ROC analysis.

Main Results:

• The CT-DTA model incorporated eight key parameters, including bubble-like vacuole sign, air bronchogram sign, smoking status, lesion type, histology, pleural indentation sign, gender, and lobulation sign.
• The model achieved an Area Under the Curve (AUC) of 0.854 in receiver operating characteristic (ROC) analysis.
• Multivariate logistic regression confirmed the CT-DTA model as an independent predictor of EGFR mutation status (P<0.001).

Conclusions:

• The developed CT-DTA model offers a straightforward method for predicting EGFR mutation status in SMPLC patients.
• This CT-based predictive tool can assist in guiding treatment decisions.
• Further integration of imaging biomarkers can enhance personalized lung cancer management.