Integrative radiomics and habitat imaging models for predicting PD-L1 expression in non-small cell lung cancer - Report - MDSpire

Integrative radiomics and habitat imaging models for predicting PD-L1 expression in non-small cell lung cancer

  • By

  • Hao Fang

  • Huadong Chen

  • Wei Tan

  • Peijun Liu

  • July 6, 2026

  • 0 min

Share

Clinical Report: Radiomics and Habitat Imaging for PD-L1 Prediction in NSCLC

Overview

This study evaluates an arterial-phase CT-based model for noninvasive prediction of PD-L1 expression in non-small cell lung cancer (NSCLC).

Background

Non-small cell lung cancer (NSCLC) is a major cause of cancer-related mortality, with PD-L1 expression serving as a critical biomarker for immunotherapy response. Traditional assessment methods for PD-L1 are invasive and may not accurately reflect tumor heterogeneity. This study explores noninvasive imaging techniques to improve the prediction of PD-L1 expression.

Data Highlights

ModelAUC (Training Cohort)AUC (Validation Cohort)
Whole-tumor Radiomics0.7580.758
Habitat Imaging0.7740.758
Combined Model0.8400.828

Key Findings

  • Independent predictors of PD-L1 expression include tumor maximum diameter and intratumoral necrosis.
  • The habitat imaging model outperformed the whole-tumor radiomics model in predicting PD-L1 expression.
  • The combined model integrating radiomics and clinical variables showed the highest predictive performance.
  • Decision curve analysis indicated superior net clinical benefit for the combined model.

Clinical Implications

The findings suggest that integrating habitat imaging with clinical features can enhance the accuracy of PD-L1 expression predictions in NSCLC.

Conclusion

The arterial-phase CT-based habitat imaging model effectively predicts PD-L1 expression in NSCLC, with improved performance when combined with clinical variables.

Related Resources & Content

  1. Journal of Neuro-Oncology, 2023 -- Utilizing Radiomics to Predict PD-L1 Expression Non-Invasively in Patients with Brain Metastases from Non-Small Cell Lung Cancer
  2. Frontiers in Medicine, 2026 -- The predictive value of 18F-FDG PET/CT habitat radiomics combined model in evaluating EGFR gene mutations in lung adenocarcinoma
  3. asco ai in oncology, 2026 -- Improved Immunotherapy Response Prediction in NSCLC With Deep-Learning Radiomic Biomarker
  4. Frontiers in Immunology, 2026 -- Early prediction of immune checkpoint inhibitor-related pneumonitis in advanced non-small cell lung cancer based on primary tumor Delta-radiomics features
  5. CAP Publishes Guideline For PD-L1 Testing Of Patients With Lung Cancer - CAP
  6. Are 18F-FDG PET/CT based radiomics features useful for prediction of PD-L1 expression in non-small cell lung cancer? - PubMed
  7. CAP Guidelines for PD-L1 Testing
  8. CAP Publishes Guideline For PD-L1 Testing Of Patients With Lung Cancer - CAP
  9. Are 18F-FDG PET/CT based radiomics features useful for prediction of PD-L1 expression in non-small cell lung cancer? - PubMed

Original Source(s)

Related Content