Beyond Morphology: A Nomogram Integrating CT Net Arterial Enhancement and Serum Biomarkers for Preoperative N-Staging in Colorectal Cancer

Background: Accurate preoperative assessment of lymph node (LN) metastasis dictates treatment strategies in colorectal cancer (CRC), yet traditional morphological criteria (e.g., size) lack sensitivity. We aimed to develop a diagnostic nomogram integrating functional hemodynamics and serum biomarkers to overcome these limitations. Methods: Data from 130 CRC patients undergoing radical resection were retrospectively analyzed. Preoperative multi-phase computed tomography (CT) features, specifically Net Arterial Enhancement (ΔA) and morphological metrics, along with serum carcinoembryonic antigen (CEA) and carbohydrate antigen 19-9 (CA19-9) levels, were evaluated. A diagnostic nomogram was constructed based on multivariate logistic regression, with performance assessed using Area Under the Curve (AUC), bootstrap validation, and Decision Curve Analysis (DCA). Results: Multivariate analysis revealed that traditional morphological features (size, shape, margin) were not independent predictors of metastasis. In contrast, ΔA (p = 0.001), CEA (p < 0.001), and CA19-9 (p = 0.024) were identified as significant independent predictors. The novel nomogram integrating these three variables achieved an AUC of 0.925 (95% CI 0.877–0.972), significantly outperforming both the conventional CT model incorporating short-axis diameter, short/long axis ratio, and ΔA (AUC 0.778, 95% CI 0.699–0.857; p < 0.001) and a pure morphology model combining short-axis diameter, short/long axis ratio, and margin status (AUC 0.734, 95% CI 0.649–0.819; p < 0.001). The model demonstrated excellent calibration (bias-corrected C-index: 0.917) and provided superior clinical net benefit across a wide range of threshold probabilities compared to standard strategies. Conclusion: Within an integrated multivariable framework, functional hemodynamic parameters (ΔA) and biological markers offer preoperative predictive information that is not fully captured by size-based criteria alone, supporting their complementary role in nodal assessment. This nomogram provides a non-invasive and quantitative tool for patient-level preoperative N-staging based on index-node imaging surrogates, facilitating personalized surgical decision-making. Given that the model was developed in a modest single-center retrospective cohort and evaluated only through internal bootstrap validation, external validation in independent, preferably prospective, multi-center cohorts is required before the nomogram can be considered robust or adopted in routine clinical practice.