Radiomics and Delta-Radiomics Signatures to Predict Response and Survival in Patients with Non-Small-Cell Lung Cancer...

Area of Science:

• Oncology
• Radiology
• Medical Imaging

Background:

• Immune checkpoint inhibitors (ICIs) have revolutionized non-small cell lung cancer (NSCLC) treatment.
• Predicting treatment response to ICIs remains a challenge.
• Novel imaging biomarkers are needed to guide patient selection for ICI therapy.

Purpose of the Study:

• To evaluate the role of CT-based radiomics in predicting treatment response and survival in advanced NSCLC patients treated with ICIs.
• To compare the predictive performance of baseline radiomics, delta-radiomics, and clinical parameters.

Main Methods:

• Retrospective analysis of 188 advanced NSCLC patients treated with PD-1/PD-L1 inhibitors from two centers.
• Radiomics features extracted from pre-treatment and follow-up contrast-enhanced CT scans.
• Delta-radiomics analysis performed on a subset of 160 patients.
• Linear and random forest (RF) models used for response and survival prediction.

Main Results:

• The RF delta-radiomics model achieved the best response prediction (AUC 0.8) on the external test set.
• The Cox regression delta-radiomics model showed the highest accuracy in survival prediction (concordance index 0.68, p=0.02).
• Baseline CT radiomics did not significantly predict treatment response or survival.

Conclusions:

• CT-based delta-radiomics is a promising tool for early identification of NSCLC patients who will benefit from immunotherapy.
• Delta-radiomics outperforms baseline radiomics and clinical parameters in predicting treatment outcomes.
• This imaging biomarker can aid in optimizing ICI treatment strategies for NSCLC.