Positron Emission Tomography in Thymic Tumors: Analysis Using a Prospective Research Database

Abstract

Background: Positron emission tomography may have a role in the pretreatment workup of patients with thymic malignancies. This study was undertaken to determine the utility of the maximum standardized uptake value (SUV_max) in predicting histologic type and tumor stage in a large cohort of thymic epithelial tumors.

Methods: The large, multiinstitutional, prospective database of The International Thymic Malignancy Interest Group (ITMIG) was queried for the use of positron emission tomography in the pretreatment workup of patients with thymic tumors. Data analyzed included demographics, SUV_max, histologic tumor type, and tumor stage. The distribution of SUV_max according to histologic type and Masaoka-Koga pathologic stage was determined, and the ability of SUV_max to predict these two variables was calculated using analysis of receiver operating characteristic curves.

Results: Since 2012, data from 926 patients with thymic malignancies were entered into the ITMIG prospective database, of which 154 had a reported value for SUV_max. The area under the receiver operating characteristic curve for SUV_max in predicting histologic type and pathologic stage was 0.79 (95% confidence interval, 0.70 to 0.88; p < 0.001) and 0.81 (95% confidence interval, 0.73 to 0.88; p < 0.001), respectively. In addition, there was a significant relationship between SUV_max and histologic type (p < 0.001) as well as Masaoka-Koga pathologic stage (p < 0.001).

Conclusions: Positron emission tomography has utility in predicting clinicopathologic features of thymic malignancies. These results may have clinical application in the pretreatment workup of patients with these rare tumors.

Figure 1

Distribution of SUV_max in a cohort of 154 thymic epithelial tumors. A. Histogram showing the distribution of SUV_max in the entire cohort. B. Box and whisker plot showing the distribution of SUV_max according to WHO histologic type. The upper and lower ends of each box represent the upper and lower quartiles, the thick horizontal bar represents the median, and the extreme horizontal bars represent the maximum and minimum values. For the statistical analysis, neuroendocrine tumors were grouped with thymic carcinomas and mixed B types were grouped with the highest B histology in the tumor; (e.g. B₁/B₂/B₃ and B₂/B₃ were grouped with B₃). C. Box and whisker plot showing the distribution of SUV_max according to Masaoka-Koga pathologic stage. The upper and lower ends of each box represent the upper and lower quartiles, the thick horizontal bar represents the median, and the extreme horizontal bars represent the maximum and minimum values. For the statistical analysis, all stage II tumors were grouped together. For panels B and C, the number of data points for each group is listed above the upper horizontal bar.

Figure 2

The relationship between SUV_max and maximum tumor dimension as measured on CT scan. Each data point represents an individual patient.

Figure 3

Ability of SUV_max to predict WHO histologic type and Masaoka-Koga pathologic stage in patients with TET. A. ROC curve for WHO histologic type. B. ROC curve for Masaoka-Koga pathologic stage.

Figure 4

Box and whisker plot showing the effect of tumor invasiveness on SUV_max according to the histologic type of TET. The upper and lower ends of each box represent the upper and lower quartiles, the thick horizontal bar represents the median, and the extreme horizontal bars represent the maximum and minimum values. The number of data points for each group is listed above the upper horizontal bar. A. Stage I and II TET. B. Stage III and IV TET.