Correlation of solid proportion and lymph node metastasis in lung cancers ≤30 mm in diameter

Area of Science:

• Radiology
• Oncology
• Medical Imaging

Background:

• Lymph node metastasis (LNM) is a common and challenging aspect of lung cancer diagnosis.
• Accurate preoperative prediction of LNM is crucial for effective treatment planning.
• Lung cancers ≤30 mm present unique diagnostic challenges based on solid proportions (SPs).

Purpose of the Study:

• To investigate the correlation between clinical computed tomography (CT) imaging features and LNM in lung cancers ≤30 mm.
• To develop and compare predictive models for LNM based on different solid proportions (SPs) of lung nodules.
• To assess the diagnostic value of CT features for predicting LNM in various lung cancer subtypes.

Main Methods:

• A retrospective analysis of 2,074 lung cancer patients with confirmed pathology and lymph node dissection.
• Categorization of lung cancers into ground-glass nodule (GGN), solid nodule (SN), and GGN + SN groups.
• Development of predictive models using logistic regression and comparison via the Delong test.

Main Results:

• Model 1 (GGN) showed solid proportion (SP) as the sole predictor (AUC: 0.929, accuracy: 0.850).
• Model 2 (SN) identified high blood pressure, ventilation/perfusion status, short diameter (SD), and CT mean density (CTmean) as risk factors (AUC: 0.733, accuracy: 0.735).
• Model 3 (GGN + SN) found age, SP, spiculation, pleural tag, and rim sign to be significant predictors (AUC: 0.904, accuracy: 0.751).

Conclusions:

• Predictive models for GGN (Model 1) and mixed GGN + SN (Model 3) demonstrate high diagnostic value for preoperative LNM prediction.
• CT imaging features are valuable for predicting LNM in lung cancers, with varying effectiveness based on nodule composition.
• Further research may refine CT-based LNM prediction models for improved lung cancer management.