[Predicting response to non-small cell lung cancer immunotherapy using pre-treatment contrast-enhanced CT...

Area of Science:

• Radiology
• Oncology
• Medical Imaging Analysis

Background:

• Immunotherapy has revolutionized non-small cell lung cancer (NSCLC) treatment.
• Predicting patient response to immunotherapy remains a significant clinical challenge.
• Objective imaging biomarkers are needed to guide treatment decisions.

Purpose of the Study:

• To evaluate the predictive value of pre-treatment contrast-enhanced computed tomography (CT)-based texture analysis for immunotherapy response in NSCLC.
• To identify specific CT texture features associated with treatment outcomes.

Main Methods:

• Retrospective analysis of 51 lesions from 42 advanced NSCLC patients undergoing immunotherapy.
• Extraction and selection of optimal texture features from pre-treatment contrast-enhanced CT using MaZda software.
• Classification of lesions into non-progressive disease (non-PD) and progressive disease (PD) groups based on treatment efficacy.
• Application of principal component analysis (PCA), linear discriminant analysis (LDA), and nonlinear discriminant analysis (NDA) for predictive modeling.

Main Results:

• Significant differences in texture parameters (Perc.50%, Perc.90%, S(5, 5)SumEntrp, S(4, 4)SumEntrp) were observed between PD and non-PD groups (P<0.05).
• The best predictive model utilized texture features selected by POE+ ACC and analyzed by NDA, achieving an AUC of 0.802.
• This optimal model demonstrated high sensitivity (72%), specificity (88.5%), accuracy (80.4%), PPV (85.7%), and NPV (76.7%).

Conclusions:

• Pre-treatment contrast-enhanced CT texture characteristics show promise as non-invasive biomarkers for predicting immunotherapy response in NSCLC.
• Texture analysis may aid in stratifying patients and optimizing immunotherapy selection.
• Further validation in larger cohorts is warranted to confirm these findings.