Computed Tomography Texture Analysis of Lymph Nodes for Improved Staging Accuracy in Patients with Colon Cancer

Texture analysis can provide quantitative imaging markers and better characterize tumor tissue in oncological imaging. The present analysis investigated the diagnostic benefit of computed tomography (CT)-derived texture analysis to categorize and stage lymph nodes in patients with colon cancer.

In this study, 85 patients were included (n = 39 females, 45.9%) with a mean age of 70.3 ± 14.8 years. All patients were surgically resected, and the lymph nodes were histopathologically analyzed. All investigated lymph nodes were further investigated with texture analysis using the MaZda package.

Out of a total of 279 extracted CT texture features, 7 parameters independently showed statistically significant differences between lymph node positive to negative ones. For instance, the texture parameter S(1,0)AngScMom showed statistically significant differences regarding lymph node metastasis status (0.007 ± 0.004 for N0 vs. 0.005 ± 0.001 for N1-2, p = 0.001). A multivariate model was developed based on n = 7 independent texture parameters. The diagnostic accuracy reached an area under the curve of 0.79 (95% CI: 0.69-0.89) with a sensitivity of 0.77 and a specificity of 0.70, resulting in an accuracy of 0.73.

Texture analysis can improve the diagnostic accuracy for nodal CT staging in patients with colon cancer. Further validation studies are needed to confirm the present results.