Development and validation of a clinic-radiological model to predict tumor spread through air spaces in stage I lung adenocarcinoma

Zhaisong Gao ^1,2, Pingping An ³, Runze Li ²

⁰Department of Nuclear Medicine, Affiliated Hospital of Qingdao University, Qingdao, Shandong, China.

Cancer Imaging : the Official Publication of the International Cancer Imaging Society +

|

February 9, 2024

View abstract on PubMed

Summary

Tumor spread through air spaces (STAS) in lung adenocarcinoma is linked to older age, solid lesions, and high SUVmax. A new PET/CT model accurately predicts STAS, aiding surgical decisions.

Area Of Science

Oncology
Radiology
Nuclear Medicine

Background

Tumor spread through air spaces (STAS) is a significant prognostic factor in lung adenocarcinoma (LAC).
STAS impacts surgical treatment strategies and patient outcomes.
Predicting STAS preoperatively is crucial for personalized treatment planning.

Purpose Of The Study

To develop and validate a user-friendly prediction model for STAS in stage I LAC.
To utilize 2-[18F] FDG PET/CT imaging features and clinical data for STAS prediction.
To assess the reliability of the developed model in independent patient cohorts.

Main Methods

Retrospective analysis of 466 stage I LAC patients who underwent 2-[18F] FDG PET/CT and surgery.
Patients were divided into training, validation, and test cohorts based on chronological order.
Independent predictors of STAS, including age, lesion density, and SUVmax, were identified using logistic regression to build the A-D-S model.

Main Results

Age ≥ 56 years, solid lesion density, and SUVmax ≥ 2.5 g/mL were identified as independent predictors of STAS.
The A-D-S (Age-Density-SUVmax) prediction model demonstrated high performance.
Area Under the Curve (AUC) values for the model were 0.808 (training), 0.786 (validation), and 0.806 (test) cohorts.

Conclusions

STAS in stage I LAC is associated with older patient age, solid tumor lesions, and elevated SUVmax.
The PET/CT-based A-D-S model is a reliable and user-friendly tool for diagnosing STAS.
This model can assist in preoperative assessment and surgical planning for lung adenocarcinoma.

Keywords:

Adenocarcinoma Computed tomography Invasion Lung Positron emission tomography