Body composition as a biomarker for assessing future lung cancer risk

Area of Science:

• Oncology
• Radiology
• Biomarkers

Background:

• Lung cancer screening with low-dose computed tomography (LDCT) is crucial for early detection.
• Identifying individuals at high risk for lung cancer is essential for effective screening strategies.
• The role of body composition as a potential biomarker for lung cancer risk requires further investigation.

Purpose of the Study:

• To determine if body composition, assessed via LDCT, serves as a biomarker for predicting lung cancer risk.
• To explore the utility of body composition metrics in enhancing the effectiveness of lung cancer screening programs.

Main Methods:

• Utilized LDCT scans from the Pittsburgh Lung Screening Study (PLuSS) and NLST-ACRIN cohorts.
• Developed artificial intelligence (AI) algorithms for automated segmentation and quantification of key body tissues: subcutaneous adipose tissue (SAT), visceral adipose tissue (VAT), intramuscular adipose tissue (IMAT), skeletal muscle (SM), and bone.
• Employed cause-specific Cox proportional hazards models and time-dependent receiver operating characteristic (ROC) analysis to assess risk and prognostic ability.

Main Results:

• A composite model incorporating age, smoking status, bone volume, SM density, IMAT ratio, IMAT density, and SAT volume demonstrated predictive capability.
• Models trained on the PLuSS cohort achieved an area under the curve (AUC) of 0.76 over 21 years and 0.70 over 7 years for lung cancer prediction.
• Models validated on the NLST cohort showed AUCs ranging from 0.61 to 0.68 over a 7-year follow-up.

Conclusions:

• Body composition metrics derived from LDCT are significant predictors of lung cancer risk.
• Integrating body composition analysis into LDCT screening can optimize eligibility criteria and screening frequency, thereby improving screening effectiveness.
• Prediction models combining demographic factors and body composition features can effectively identify individuals at elevated risk for lung cancer.