Clinical Report: Utilizing 18F-FDG PET/CT to Assess Risk Factors in LUAD
Overview
This study developed an 18F-FDG PET/CT-based predictive model to differentiate high-risk invasive lung adenocarcinoma (LUAD) subtypes associated with ground-glass nodules (GGNs). Key independent predictors identified include SUVmax, nodule diameter, and lesion location.
Background
Lung adenocarcinoma (LUAD) is the most common subtype of lung cancer, accounting for about 40% of cases. Micropapillary- and solid-predominant LUAD subtypes are associated with poor prognosis and high recurrence rates. Accurate preoperative risk assessment is crucial for optimizing clinical management.
Data Highlights
Model
AUC (Training)
AUC (Test)
Original Model
0.921
0.855
Optimized Model
0.934
0.873
Key Findings
MPA and SPA subtypes of LUAD are high-risk with unfavorable prognosis.
Independent predictors for high-risk LUAD include SUVmax, nodule diameter, and lesion location.
The original three-variable model achieved AUCs of 0.921 and 0.855 for training and test cohorts, respectively.
The optimized model incorporating CT attenuation and vacuole sign improved AUCs to 0.934 and 0.873.
No statistical intergroup differences were found between the original and optimized models (all P > 0.05).
Clinical Implications
The findings support the use of 18F-FDG PET/CT parameters for non-invasive preoperative risk evaluation of high-risk LUAD associated with GGNs.
Conclusion
The study demonstrates that PET/CT-derived metrics can effectively stratify high-risk LUAD.
