Fig. 4

Associations between risk scores and clinical characteristics. (A–D) Boxplots showing the distribution of risk scores across clinical stage (I–IV), T stage (T2–T4), N stage (N0–N2), and M stage (M0 vs. M1). Statistically significant differences were observed among subgroups (ANOVA with Tukey’s post hoc test, *p < 0.05).