Predictive Value of 18 F-FDG PET/MRI for Pleural Invasion in Solid and Subsolid Lung Adenocarcinomas Smaller Than

Annan Zhang1, Xiangxi Meng1, Yuan Yao1

  • 1Key Laboratory of Carcinogenesis and Translational Research (Ministry of Education/Beijing), Department of Nuclear Medicine, Peking University Cancer Hospital & Institute, Haidian, Beijing, China.

Abstract

Insights

Positron emission tomography (PET)/MRI shows high predictive value for assessing pleural invasion in lung cancer. This imaging technique aids in diagnosing smaller lung adenocarcinomas with potential pleural involvement.

Area of Science:

  • Oncology
  • Radiology
  • Medical Imaging

Background:

  • Positron emission tomography (PET)/MRI integrates metabolic imaging with high soft tissue resolution.
  • This combination offers potential for high diagnostic efficacy in assessing pleural invasion (PI) of lung cancer.

Purpose of the Study:

  • To evaluate the utility of 18F-fluorodeoxyglucose (FDG) PET/MRI in predicting pleural invasion (PI) for lung cancers with a maximum diameter of 3 cm or less.

Main Methods:

  • A prospective study involving 44 non-small cell lung cancer (NSCLC) patients.
  • Utilized a 3-T hybrid PET/MRI scanner with T2 fat-suppressed imaging (T2FS) and diffusion-weighted imaging (DWI).
  • Developed three predictive models incorporating CT, PET, and MRI features, using Lasso regression for feature selection and ROC analysis for diagnostic performance.

Main Results:

  • The developed models (Model 1: CT, Model 2: CT+PET, Model 3: PET+MRI) achieved Areas Under the Curve (AUC) of 0.762, 0.829, and 0.915, respectively.
  • Model 3 (PET+MRI) demonstrated the highest AUC, indicating superior predictive value.
  • Statistical analysis showed a significant difference between Model 1 and Model 3, suggesting the added value of PET/MRI features.

Conclusions:

  • 18F-FDG PET/MRI demonstrates significant potential for predicting pleural invasion in lung adenocarcinomas measuring less than 3 cm.
  • The integration of PET and MRI features in predictive models enhances diagnostic accuracy for pleural invasion.