Statistical analysis

Student’s t-test was used to analyze whether the signatures of glucose metabolism were significantly different according to the genetic mutation status of BRAFV600E. Pearson’s correlation test was performed to evaluate the associations between TDS and signatures of glucose metabolism. The difference between two correlation coefficients regarding BRAFV600E was calculated by Fisher z-transformation. In addition, the difference between two correlation coefficients according to the histologic cell types was calculated by Fisher z-transformation. The effect of the signatures of glucose metabolism on patients’ recurrence-free survival was analyzed using the Cox proportional hazard model. Patients were divided into two groups; lower or higher than the median value of GLUT, and glycolysis unit. Kaplan-Meier survival curves were demonstrated to compare recurrence-free survival between two groups. The statistical significance was evaluated using the Log rank test. The differences between groups were considered statistically significant at p-value < 0.05. All statistical analyses were performed with R 3.4.1 (https://www.r-project.org/).