From Passive Sampling to Precision Intervention: A Standardized Radiomics-Driven Workflow for Molecularly Adequate...

Area of Science:

• Interventional Radiology
• Oncology
• Medical Imaging

Background:

• Molecular adequacy is crucial for lung cancer diagnosis and treatment planning.
• Computed tomography (CT)-guided transthoracic needle biopsy (TTNB) is a key diagnostic tool.
• Optimizing TTNB for molecular testing requires robust predictive frameworks.

Purpose of the Study:

• To develop and validate a predictive framework for molecular adequacy in CT-guided TTNB.
• To integrate procedural variables and radiomic features for enhanced prediction.
• To assess the framework's performance in a temporal validation cohort.

Main Methods:

• Observational derivation-validation study with retrospective and prospective cohorts (670 procedures).
• Development of a multivariable procedural model and an exploratory radiomics model.
• Evaluation of molecular adequacy for next-generation sequencing and PD-L1 testing.
• Performance assessment using AUC, calibration, and transportability analyses.

Main Results:

• Overall molecular adequacy was 83.9%, with higher rates in the prospective cohort (91.2%).
• The procedural model achieved AUCs of 0.86 (derivation) and 0.84 (validation).
• Radiomics integration improved discrimination (AUC 0.88), especially for subsolid lesions, without increasing complications.

Conclusions:

• Procedural standardization and radiomics integration significantly improve molecular adequacy in CT-guided lung biopsies.
• The developed framework serves as a basis for future decision-support strategies in interventional radiology.
• This approach enhances diagnostic yield for precision oncology.