PMILACG Model: A Predictive Model for Identifying Invasiveness of Lung Adenocarcinoma Based on High-Resolution CT-Determined Ground Glass Nodule Features

PMILACG Model: A Predictive Model for Identifying Invasiveness of Lung Adenocarcinoma Based on High-Resolution CT-Determined Ground Glass Nodule Features

Bo Yan,^1,2 Yifeng Jiang,³ Shijie Fu⁴ and Rong Li^1,2

¹Clinical Research Unit, Shanghai Chest Hospital, School of Medicine, Shanghai Jiao Tong University, Shanghai, China
²Department of Pulmonary Medicine, Shanghai Chest Hospital, Shanghai Jiao Tong University, Shanghai, China
³Department of Radiology, Shanghai Chest Hospital, Shanghai Jiao Tong University, Shanghai, China
⁴Department of Thoracic Surgery, Shanghai Chest Hospital, School of Medicine, Shanghai Jiao Tong University, Shanghai, China

The morphology of ground-glass nodule (GGN) under high-resolution computed tomography (HRCT) has been suggested to indicate different histological subtypes of lung adenocarcinoma (LUAD); however, existing studies only include the limited number of GGN characteristics, which lacks a systematic model for predicting invasive LUAD. This study aimed to construct a predictive model based on GGN features under HRCT for LUAD. A total of 1,189 surgical LUAD patients were enrolled, and their GGN-related features were assessed by 2 individual radiologists. The pathological diagnosis of the invasive LUAD was established by pathologic examination following surgery (including 1,073 invasive and 526 non-invasive LUAD). After adjustment by multivariate logistic regression, GGN diameter (OR = 1.382, 95% CI: 1.300-1.469), mean CT attenuation (OR = 1.007, 95% CI: 1.006-1.009), heterogeneous uniformity of density (OR = 2.151, 95% CI: 1.587-2.915), not defined nodule-lung interface (OR = 1.915, 95% CI: 1.384-2.651), GGN with spiculation (OR = 2.097, 95% CI: 1.519-2.896), type I (OR = 1.678, 95% CI: 1.216-2.371), and type II (OR = 3.577, 95% CI: 1.153-11.097) vessel changes were independent risk factors for invasive LUAD. In addition, a predictive model integrating these six independent GGN features was established (named as invasion of lung adenocarcinoma by GGN features (ILAG)), and receiver-operating characteristic curve illustrated that the ILAG model presented good predictive value for invasive LUAD (AUC: 0.905, 95% CI: 0.890-0.919). In conclusion, The ILAG predictive model, which integrates imaging features of GGN via HRCT, including diameter, mean CT attenuation, heterogeneous uniformity of density, not defined nodule-lung interface, GGN with spiculation, type I, and type II vessel changes, shows great potential for early estimation of LUAD invasiveness.

Key words —— ground-glass nodule features; high-resolution computed tomography; invasion of lung adenocarcinoma by GGN features predictive model; invasiveness; lung adenocarcinoma

===============================

Tohoku J. Exp. Med., 2025 January, 265(1), 13-20.

Correspondence: Rong Li, Clinical Research Unit, Department of Pulmonary Medicine, Shanghai Chest Hospital, Shanghai Jiao Tong University, 241 West Huaihai Road, Xuhui Distinct, Shanghai 200030 China.

e-mail:xkyylirong@163.com