The Combination of Maximum Standardized Uptake Value and Clinical Parameters for Improving the Accuracy in Distinguishing Primary Mediastinal Lymphomas from Thymic Epithelial Tumors

Hui Yan,Lihua Wang,Bei Lei,Maomei Ruan,Cheng Chang,Mingge Zhou,Liu,Wenhui Xie,Yuetao Wang
DOI: https://doi.org/10.21037/qims-23-496
2024-01-01
Quantitative Imaging in Medicine and Surgery
Abstract:Background: Anterior mediastinal masses are relatively uncommon, and mediastinal lymphomas are the malignancies most likely to be confused with thymic epithelial tumors (TETs). The aim of this study was to investigate whether the combination of (18)fluorine-fluorodeoxyglucose positron emission tomography-computed tomography (F-18-FDG PET-CT) findings and clinical parameters is useful in differentiating lymphoma from TETs in anterior mediastinal masses.Methods: This retrospective study consecutively included 304 patients with anterior mediastinal masses (244 TETs and 60 lymphomas) who underwent(18)F-FDG PET-CT 1 to 2 weeks before tumor resection or biopsy between August 2016 and March 2022. The correlations between the maximum standardized uptake value (SUVmax) of tumors and clinical parameters of patients with histology subtypes were analyzed. Receiver operating characteristic curve analysis was used to obtain the optimal cutoff values of age, lactate dehydrogenase (LDH), tumor size, and SUVmax to predict lymphoma. Logistic regression analysis was used to identify potential predictive factors for lymphoma.Results: Lymphoma was significantly associated with younger patient age, higher LDH level, larger tumor size, and higher SUVmax compared to TETs (P<0.001). In the modeling cohort, age <= 40.5 years, LDH level >= 197 U/I, tumor size >= 10.72 cm, and SUVmax >= 11.95 were identified as independent predictors for lymphoma with odds ratios of 20.14 [95% confidence interval (CI): 6.02-67.40; P<0.001], 4.89 (95% CI: 1.27-18.89; P=0.021), 8.82 (95% CI: 2.31-33.69; P=0.001), and 30.01 (95% CI: 6.59-136.72; P<0.001), respectively. The accuracy of age, LDH, tumor size, and SUVmax in predicting lymphoma was 84.8%, 67.8%, 85.2%, and 78.3% respectively. The combination of the four above parameters could improve the predictive accuracy to 89.1%, and in the validation cohort, this combination increased the predictive accuracy to 87.8%. Conclusions: SUVmax on F-18-FDG PET-CT has the potential ability to discriminate lymphomas from TETs in the diagnosis of anterior mediastinal masses, and the combination of SUVmax with clinical parameters can improve the diagnostic accuracy. This combination may therefore may be helpful in avoiding unnecessary operation in patients with anterior mediastinal lymphomas.
What problem does this paper attempt to address?