Construction of nomogram model for predicting the risk of lymph node metastasis in lung malignancies based on imaging parameters of lymph nodes
10.3760/cma.j.cn321828-20240617-00214
- VernacularTitle:基于淋巴结自身影像参数构建预测肺部恶性肿瘤淋巴结转移风险的列线图模型
- Author:
Hao SUN
1
;
Xuemei WANG
1
;
Guojian ZHANG
1
Author Information
1. 内蒙古医科大学附属医院核医学科、内蒙古自治区分子影像学重点实验室，呼和浩特 010050
- Publication Type:Journal Article
- Keywords:
Lung neoplasms;
Lymphatic metastasis;
Peptides, cyclic;
Technetium;
Tomography, emission-computed, single-photon;
Tomography, X-ray computed;
Nomograms
- From:
Chinese Journal of Nuclear Medicine and Molecular Imaging
2025;45(5):269-275
- CountryChina
- Language:Chinese
-
Abstract:
Objective:To construct nomogram model based on lymph node imaging parameters for predicting the risk of lymph node metastasis in lung malignancies.Methods:From November 2020 to September 2022, 34 patients (23 males, 11 females, age (65.8±6.8) years) diagnosed with lung malignancies by pathology at the Affiliated Hospital of Inner Mongolia Medical University were prospectively collected. Based on enhanced CT and 99Tc m-hydrazinonicotinamide-(polyethylene glycol) 4-E((polyethylene glycol) 4-c((Arg-Gly-Asp)fK)) 2 (HYNIC-PEG 4-E(PEG 4-c(RGDfK)) 2; 3PRGD 2)SPECT/CT imaging, referring to the mediastinal lymph node zoning standards formulated by the International Association for the Study of Lung Cancer, lymph nodes with clear pathological properties and imaging locations were included in the study. Lymph nodes were randomly divide into a training group and a validation group at a ratio of 7∶3. Differences of imaging parameters between positive and negative lymph node metastasis were compared by independent-sample t test or Mann-Whitney U test or χ2 test. Parameters with statistical differences were incorporate into the multivariate logistic regression equation, and a joint variable diagnostic model for predicting lymph node metastasis was generated. The potential of the model was evaluate by ROC curve analysis, calibration, and decision curve analysis (DCA). Results:Among 34 patients with malignant lung tumors, 11 had lymph node metastasis. A total of 174 lymph nodes met the inclusion criteria were randomly divided into a training group of 114 nodes and a validation group of 60 nodes. In the training group and validation group, there were statistically significant differences in lymph node length, lymph node short diameter, lymph node length/short diameter, necrosis, lymph node to mediastinal blood pool radioactive count ratio (T/B), lymph node to liver radioactive count ratio (T/L), lymph node to muscle radioactive count ratio (T/M), and lymph node enhancement mode between patients with positive and negative lymph node metastasis ( χ2 values: 3.89-34.06, t values: 2.31-3.87, Z values: from -3.63 to -2.30, all P<0.05). The lymph node edge was different only in the training group ( χ2=5.62, P=0.018). Finally, the lymph node length/short diameter, edge, necrosis, T/B, T/L, T/M, and lymph node enhancement modes were included in the multivariate logistic regression prediction model, with the AUCs of 0.878 and 0.949 in the training and validation groups, respectively. The calibration curve showed good consistency between the predicted results and the actual results, and DCA showed that the nomogram had clinical practicality. Conclusion:The nomogram model constructed based on the imaging parameters of lymph nodes can evaluate the risk of lymph node metastasis in patients with lung malignancies, providing a convenient and objective tool for determining staging and developing treatment plans.
