Multi-Omic Biomarkers Improve Indeterminate Pulmonary Nodule Malignancy Risk Assessment.

Area of Science:

• Oncology
• Radiology
• Biomarker Discovery

Background:

• Indeterminate pulmonary nodules pose clinical management challenges, often leading to patient harm.
• Accurate lung cancer risk stratification is crucial for these patients.

Purpose of the Study:

• To develop and validate a novel risk prediction model for lung cancer in indeterminate pulmonary nodules.
• To integrate noninvasive imaging and molecular data for improved risk assessment.

Main Methods:

• Analysis of quantitative radiomic and qualitative semantic imaging features.
• Utilized molecular biomarkers from blood via antibody-based microarray.
• Developed a plasma, semantic, radiomic (PSR) risk prediction model using nine biomarkers.

Main Results:

• The initial PSR model achieved an AUROC of 0.964.
• In an independent cohort, the PSR model yielded an AUROC of 0.846.
• Incorporating clinical factors improved the PSR model's AUROC to 0.897, outperforming a clinical model (AUROC=0.802).

Conclusions:

• A multi-omics approach combining plasma, semantic, and radiomic data offers a powerful tool for lung cancer risk prediction.
• The PSR model demonstrates superior accuracy compared to traditional clinical risk models.
• This approach supports improved clinical management of indeterminate pulmonary nodules.