BMC Medical Imaging (Jan 2024)

A diagnostic model based on 18F-FDG PET/CT parameters in improving the differential diagnosis of invasive thymic epithelial tumors and anterior mediastinal lymphomas

  • Shuo Zhou,
  • Pokwan Tsui,
  • Meifu Lin,
  • Guobao Chen,
  • Wenxin Chen,
  • Xiangran Cai

DOI
https://doi.org/10.1186/s12880-023-01185-1
Journal volume & issue
Vol. 24, no. 1
pp. 1 – 8

Abstract

Read online

Abstract Background Accurately distinguishing between invasive thymic epithelial tumors (TETs) and anterior mediastinal lymphoma before surgery is crucial for subsequent treatment choices. But currently, the diagnosis of invasive TET is sometimes difficult to distinguish from anterior mediastinal lymphoma. Objective To assess the application of fluorine-18-fluorodeoxyglucose (18F-FDG) positron emission tomography/computer tomography (PET/CT) in the differential diagnosis of TETs and anterior mediastinal lymphomas. Methods 18F-FDG PET/CT images of 133 invasive TETs and anterior mediastinal lymphomas patients were retrospectively analyzed. In particular, the tumor’s longest diameter and maximum standardized uptake value (SUVmax) were evaluated. The SUVmax and longest diameter values of the two groups were analyzed by using the receiver operating characteristic (ROC) curve to determine the optimal threshold and diagnostic efficiency. Results Age, myasthenia gravis, SUVmax and tumor longest diameter differed significantly between invasive TETs and anterior mediastinal lymphomas patients. The tumor location, calcification, relationship with adjacent vessels and distant metastasis differed significantly between the groups. The ROC analysis showed an AUC for SUVmax and tumor longest diameter of 0.841 and 0.737. Respectively, the cutoff values with the best diagnostic performance were 9.65 (sensitivity: 77.78%, specificity: 81.97%) and 6.65 (sensitivity: 80.56%, specificity: 62.30%) for SUVmax and tumor longest diameter. The diagnostic model of SUVmax, calcification, relationship with surrounding blood vessels, lymph node metastasis and lung metastasis in the highest AUC of 0.935 (sensitivity: 90.16%, specificity: 88.89%). In addition, we incorporated splenic involvement and metastatic sub-diaphragmatic lymph node into Model 2 as a new predictive model 3 for differential diagnosis and found a significant improvement in the diagnostic performance of Model 3. Conclusion The diagnostic model composed of 18F-FDG PET parameters is improving the differential diagnosis of invasive TETs and anterior mediastinal lymphomas.

Keywords