Intratumoral and peritumoral CT radiomics combined with clinical and hematologic inflammatory markers for predicting lymph node metastasis in esophageal squamous cell carcinoma: a retrospective single-center study.

OBJECTIVE

To develop and validate a hybrid model integrating intratumoral (TR) and peritumoral (PTR) CT radiomics with clinical and hematologic inflammatory markers for predicting lymph node metastasis (LNM) in esophageal squamous cell carcinoma (ESCC).

METHODS

This retrospective single-center cohort study included 304 patients with pathologically confirmed ESCC. Patients were randomly divided into a training cohort (n = 243) and a test cohort (n = 61) in an 8:2 ratio. Radiomics features were extracted from manually segmented three-dimensional volumes of interest (VOIs) on venous-phase thin-slice contrast-enhanced CT images. The PTR VOIs were defined as 1-, 2-, and 3-mm ring-shaped regions surrounding the TR VOI. Clinical data, hematologic inflammatory markers, and TR and PTR CT radiomics features were used to construct six categories of models: a clinical model, a hematologic model, TR + PTR radiomics models, a clinical + TR + PTR radiomics model, a hematologic + TR + PTR radiomics model, and a clinical + hematologic + TR + PTR hybrid model. Model performance was evaluated using receiver operating characteristic (ROC) analysis, calibration curves, and decision curve analysis (DCA).

RESULTS

The overall LNM prevalence was 37.5% (114/304), including 37.9% (92/243) in the training cohort and 36.1% (22/61) in the test cohort. Multivariate analysis identified cN stage as an independent predictor for LNM (p < 0.05). Among the TR + PTR radiomics models, the TR + PTR1mm model showed the strong performance. The hybrid model exhibited strong performance, with AUC values of 0.898 (95% CI 0.859-0.937) and 0.867 (95% CI 0.766-0.968) for the training and test cohorts, respectively. At the optimal threshold, the hybrid model achieved a sensitivity of 0.818 and a specificity of 0.769 in the test cohort. The hybrid model demonstrated favorable net benefit in DCA compared with all other models.

CONCLUSION

The hybrid model based on TR and 1-mm PTR CT radiomics combined with clinical and hematologic inflammatory markers may serve as a promising noninvasive tool for preoperative individualized risk assessment of LNM in ESCC. However, this retrospective single-center study had a relatively small internal validation cohort and lacked external validation. Therefore, further multicenter studies with external validation are warranted to confirm the robustness and generalizability of the proposed model.