Computed tomography scan as a tool to predict tumor T category in resectable esophageal squamous cell carcinoma

Abstract

Background: The purpose of this study was to determine whether the degree of esophageal circumferential tumor involvement and tumor size of resectable esophageal squamous cell carcinoma (ESCC) assessed on computed tomography could predict T category.

Methods: One hundred eighty-five consecutive patients with ESCC underwent radical esophagectomy less than 3 weeks after contrast-enhanced computed tomography. The degree of esophageal circumferential tumor involvement and tumor size of ESCC expressed as tumor length, maximal thickness, and gross tumor volume were evaluated on computed tomography. Statistical analyses were performed to identify whether degree of esophageal circumferential tumor involvement and tumor size could predict T category.

Results: Esophageal squamous cell carcinoma with whole esophageal circumferential tumor involvement was more likely to be at T3 category, whereas tumor without this involvement was more likely to be at T1 or T2 category (p < 0.001). Degree of esophageal circumferential tumor involvement could distinguish ESCC at T1/T2 from ESCC at T3 category with a sensitivity of 77.4% and specificity of 74.8%. Tumor length, maximal thickness, and gross tumor volume increased with advancing T category (p < 0.001). Mann-Whitney tests showed that tumor size could distinguish T category (p < 0.001). Compared with degree of esophageal circumferential tumor involvement, tumor length, and maximal thickness, gross tumor volume could be a better differentiating indicator between T1 and T2 categories (cutoff, 5.15 cm(3)), between T1 and T3 categories (cutoff, 11.1 cm(3)), between T2 and T3 categories (cutoff, 17.75 cm(3)), and between T1/T2 and T3 categories (cutoff, 15.9 cm(3)), with sensitivity of 81.3%, 88.8%, 68.8%, and 78.8%, and specificity of 76%, 88%, 67.5%, and 75.4%, respectively.

Conclusions: Gross tumor volume of resectable ESCC measured with computed tomography could be a recommended indicator for predicting T category.