Integrative radiomics and habitat imaging models for predicting PD-L1 expression in non-small cell lung cancer - Scorecard - 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 Scorecard: Radiomics and Habitat Imaging Approaches for Noninvasive Prediction of PD-L1 Expression in Non-Small Cell Lung Cancer

At a Glance

CategoryDetail
ConditionNon-Small Cell Lung Cancer (NSCLC)
Key MechanismsArterial-phase CT-based radiomics and habitat imaging for predicting PD-L1 expression.
Target PopulationPatients with pathologically confirmed NSCLC.
Care SettingRetrospective study in a clinical imaging context.

Key Highlights

  • Study included 246 patients with NSCLC undergoing arterial-phase CT.
  • Habitat imaging model outperformed whole-tumor radiomics model in predicting PD-L1 expression.
  • Combined model achieved the highest AUC of 0.840 in the training cohort.
  • Tumor maximum diameter and intratumoral necrosis identified as independent predictors of PD-L1 expression.
  • Study supports noninvasive imaging for pre-immunotherapy evaluation.

Guideline-Based Recommendations

Diagnosis

  • Assessment of PD-L1 expression should be performed using immunohistochemical analysis.

Management

  • Consider noninvasive imaging approaches for predicting PD-L1 expression prior to immunotherapy.

Monitoring & Follow-up

  • Monitor PD-L1 expression using validated imaging techniques to inform treatment decisions.

Risks

  • Invasive biopsy methods may introduce sampling bias and delays in clinical decision-making.

Patient & Prescribing Data

Patients with NSCLC who have not received prior antitumor therapy.

High PD-L1 expression is associated with better response to immune-based treatments.

Clinical Best Practices

  • Utilize arterial-phase CT for improved tumor characterization.
  • Incorporate radiomics and habitat imaging for enhanced predictive modeling.
  • Ensure comprehensive assessment of tumor heterogeneity in treatment planning.

Related Resources & Content

Original Source(s)

Related Content