Prediction model for occult lymph node metastasis of head and neck squamous cell carcinoma based on multi-dimensional
Yanjie Liu1, Yuanyu Wei2, Wei Jiang3
1Department of Stomatology, Liuzhou People's Hospital Liuzhou 545000, Guangxi Zhuang Autonomous Region, China.
Objective:
To develop and validate a nomogram based on multi-dimensional indicators for predicting occult lymph node metastasis (OLNM) in patients with head and neck squamous cell carcinoma (HNSCC), providing a quantitative tool for individualized clinical decision-making.
Methods:
Clinical data of 256 cN0 HNSCC patients treated in four centers from January 2021 to June 2025 were retrospectively analyzed. Patients were randomly assigned to a training set (n=179) and an internal validation set (n=77). An independent external set (n=134) was used for external validation. Multivariate logistic regression was performed to identify independent predictors, and a nomogram was constructed. Model performance was evaluated by area under the ROC curve (AUC) with 95% confidence interval (95% CI), calibration curves, and decision curve analysis (DCA).
Results:
The incidence of OLNM in the training set was 40.78%. Five independent predictors were identified: age <60.5 years (OR=0.037, 95% CI: 0.012-0.110), tumor size ≥3.05 cm (OR=4.200, 95% CI: 1.782-9.902), depth of invasion ≥7.15 mm (OR=12.812, 95% CI: 5.312-30.905), poor pathologic differentiation (OR=2.772, 95% CI: 1.276-6.020), and lymphovascular invasion (OR=1.693, 95% CI: 0.998-2.871). The nomogram showed good discrimination with AUC of 0.848 (95% CI: 0.791-0.906) in the training set, 0.708 (95% CI: 0.636-0.820) in the internal validation set, and 0.827 (95% CI: 0.755-0.899) in the external validation set. Calibration curves showed that the model's predictions followed the general trend of the observed risks.
Conclusions:
Age, tumor size, depth of invasion, pathologic differentiation, and lymphovascular invasion were key predictors of OLNM in HNSCC. The proposed nomogram provides an intuitive and reliable tool for preoperative OLNM risk assessment and may assist in identifying high-risk patients who could benefit from proactive management.
