2023 World Conference on Lung Cancer (Posters)-P2.09. MRI Radiomics Approach to Predict the Intracranial Efficacy of the Third Generation EGFR-TKI Osimertinib in NSCLC Patients with Brain Metastases

P2.09. MRI Radiomics Approach to Predict the Intracranial Efficacy of the Third Generation EGFR-TKI Osimertinib in NSCLC Patients with Brain Metastases - PDF(Abstract)

Back to course

Pdf Summary

In this study, the researchers aimed to investigate the use of MRI radiomics in predicting the intracranial efficacy of osimertinib, a third-generation EGFR-TKI, in patients with non-small cell lung cancer (NSCLC) and brain metastases. The researchers used a 5-fold cross-validation strategy and applied eight feature selectors and eight machine learning classifiers to construct radiomic models. They evaluated the models using the area under the receiver operating characteristic curve (AUC), calibration curve, and decision curve.<br /><br />The study included data from 60 NSCLC patients with EGFR-T790M-positive mutations who were receiving second-line osimertinib. After analyzing the data, the researchers found that the radiomic model constructed using the mRMR feature selector and stepwise logistic regression classifier achieved the highest predictive accuracy, with AUC values of 0.879 for the training cohort and 0.786 for the validation cohort. This was higher than the clinical-MRI morphological model, which only included age, ring enhancement, and peritumoral edema, with AUC values of 0.794 and 0.697 for the training and validation cohorts, respectively. The radiomic model also performed well in calibration and decision curve analyses.<br /><br />Based on the radiomic score, the researchers divided the patients into two groups with significantly different median intracranial progression-free survival: 3.0 months for the low-score group and 15.4 months for the high-score group.<br /><br />Overall, this proof-of-principle study suggests that the MRI radiomic model could be a promising tool for predicting the intracranial efficacy of osimertinib in NSCLC patients with brain metastases. It has the potential to assist clinicians in making personalized treatment strategies.

Asset Subtitle

Xin Tang

Meta Tag

Speaker Xin Tang

Topic Metastatic NSCLC: Targeted Therapy - EGFR/HER2

Keywords

MRI radiomics

osimertinib

EGFR-TKI

non-small cell lung cancer

brain metastases

AUC

feature selectors

machine learning classifiers

clinical-MRI morphological model

intracranial progression-free survival