甲状腺乳头状癌颈部淋巴结转移和复发风险列线图预测模型的构建与验证

Prediction and validation of anomogram model for cervical lymph node metastasis and recurrence risk in papillary thyroid carcinoma

摘要

目的：构建甲状腺癌乳头状癌（PTC）颈部淋巴结转移和复发风险的列线图预测模型，并验证其预测性能。方法：回顾性收集2020年1月至2024年1月扬州大学临床学院盱眙人民医院收治的PTC患者350例作为研究对象，按照4:1比例将患者随机分为建模组（n=280）和验证组（n=70）。收集患者临床资料，通过单因素和多因素分析，筛选PTC颈部淋巴结转移和复发的危险因素，构建列线图预测模型，并验证其预测性能。结果：患者颈部淋巴结转移率为47.14%（165/350），复发率为11.43%（40/350）。颈部淋巴结转移组患者的年龄、其他组织学亚型占比、甲状腺分化评分均低于未转移组，而肿瘤直径、多发肿瘤占比、双侧病灶占比、甲状腺外侵犯占比、淋巴结清扫数目均高于未转移组（P＜0.05）。复发组患者年龄、甲状腺外侵犯占比、侧方区淋巴结转移占比均高于未复发组，而甲状腺分化评分低于未复发组（P＜0.05）。多因素分析结果显示，年龄、肿瘤直径、甲状腺外侵犯、组织学分型、甲状腺分化评分均为PTC颈部淋巴结转移的危险因素，年龄、甲状腺分化评分、侧方区淋巴结转移均为PTC复发的危险因素（P＜0.05）。建模组和验证组的转移预测模型曲线下面积（AUC）分别为0.807（95%CI：0.757～0.857）和0.817（95%CI：0.762～0.872），复发预测模型AUC分别为0.753（95%CI：0.672～0.833）和0.784（95%CI：0.705～0.863）。转移预测模型行Hosmer-Lemeshow检验，建模组χ ²=9.463、P=0.305，验证组χ ²=8.007、P=0.433。复发预测模型行Hosmer-Lemeshow检验，建模组χ ²=5.065、P=0.751，验证组χ ²=9.355、P=0.313。决策曲线分析结果显示，采用列线图预测模型进行预防性干预，可获得更高标准化净效益。结论：基于临床资料构建的PTC颈部淋巴结转移和复发风险列线图预测模型，具有较高的预测性能，能够识别高危患者，并为制定个体化治疗方案和疾病管理策略提供参考。

Abstract

Objective: Construct a nomogram prediction model for cervical lymph node metastasis and recurrence of papillary thyroid carcinoma (PTC) and validate its predictive performance. Methods: A retrospective study was conducted on 350 PTC patients admitted to Xuyi People's Hospital, Clinical College of Yangzhou University from January 2020 to January 2024. The patients were randomly divided into a modeling group(n=280) and a validation group(n=70) in a 4:1 ratio. Collect clinical data from patients, screen risk factors for PTC cervical lymph node metastasis and recurrence through univariate and multivariate analysis, construct a column chart prediction model, and verify its predictive performance. Results: The cervical lymph node metastasis rate of patients is 47.14%(165/350), and the recurrence rate is 11.43%(40/350). The age, proportion of other histological subtypes, and thyroid differentiation score of patients with cervical lymph node metastasis were lower than those of the non metastatic group, while the tumor diameter, proportion of multiple tumors, proportion of bilateral lesions, proportion of extrathyroid invasion, and number of lymph node dissection were higher than those of the non metastatic group(P＜0.05). The age, proportion of extrathyroid invasion, and proportion of lateral lymph node metastasis in the recurrent group were all higher than those in the non recurrent group, while the thyroid differentiation score was lower than that in the non recurrent group(P＜0.05). The results of multivariate analysis showed that age, tumor diameter, extrathyroid invasion, histological classification, and thyroid differentiation score were all risk factors for PTC cervical lymph node metastasis, while age, thyroid differentiation score, and lateral lymph node metastasis were all risk factors for PTC recurrence(P＜0.05). The area under the curve (AUC) of the metastasis prediction model for the modeling group and the validation group were 0.807(95% CI: 0.757~0.857) and 0.817(95% CI: 0.762~0.872), respectively. The AUC of the recurrence prediction model were 0.753(95% CI: 0.672~0.833) and 0.784(95% CI: 0.705~0.863), respectively. The Hosmer Lemeshow test was performed on the transfer prediction model, and the modeling group had a chi square value of 9.463 and a P-value of 0.305, while the validation group had a chi square value of 8.007 and a P-value of 0.433. The Hosmer Lemeshow test was performed on the recurrence prediction model, and the modeling group had a chi square value of 5.065 and a P-value of 0.751, while the validation group had a chi square value of 9.355 and a P-value of 0.313. The decision curve analysis results show that using a column chart prediction model for preventive intervention can achieve higher standardized net benefits. Conclusion: The nomogram model established based on clinical data for predicting cervical lymph node metastasis and recurrence risk of PTC exhibits excellent predictive performance. It can efficiently identify high-risk patients and provide evidence for formulating individualized treatment regimens and disease management strategies.

关键词

甲状腺乳头状癌 / 淋巴结转移 / 复发 / 列线图预测模型

Key words

Papillary thyroid carcinoma / Lymph node metastasis / Recurrence / Nomograms prediction model

参考文献

[1] Londero SC , Krogdahl A , Bastholt L , et al. Papillary thyroid carcinoma in Denmark 1996—2008: an investigation of changes in incidence[J]. Cancer Epidemiol, 2013, 37(1): e1-6.

[2] Siegel RL , Miller KD , Fuchs HE , et al. Cancer Statistics, 2021[J/OL]. CA Cancer J Clin, 2021, 71(1): 7-33. DOI: 10.3322/caac.21654.

[3] 郭星， 柴健， 张建阳， 等 . 基于PET—CT和超声造影构建甲状腺癌术后复发转移的预测模型[J/OL]. 中国实验诊断学， 2024， 28（8）： 883-888. DOI: 10.3969/j.issn.1007—4287.2024.08.001.

[4] 胡荣毅， 屈倩倩， 杨尧禹， 等 . 甲状腺乳头状癌颈淋巴结转移高危因素及预测模型构建研究进展[J/OL]. 国际耳鼻咽喉头颈外科杂志， 2025， 49（1）： 36-40. DOI: 10.3760/cma.j.issn.1673—4106.2025.01.008.

[5] 中国抗癌协会甲状腺癌专业委员会 . 中国抗癌协会甲状腺癌整合诊治指南（2022精简版）[J/OL]. 中国肿瘤临床， 2023， 50（7）： 325-330. DOI: 10.12354/j.issn.1000—8179.2023.20221410.

[6] World Medical Association . World Medical Association Declaration of Helsinki: Ethical Principles for Medical Research Involving Human Participants[J/OL]. JAMA, 2025, 333(1): 71-74. DOI: 10.1001/jama.2024.21972.

[7] Cancer Genome Atlas Research Network . Integrated genomic characterization of papillary thyroid carcinoma[J/OL]. Cell, 2014, 159(3): 676-690. DOI: 10.1016/j.cell.2014.09.050.

[8] 国家卫生健康委员会 . 甲状腺癌诊疗指南（2022年版）[J/OL]. 中国实用外科杂志， 2022， 42（12）： 1343—1357, 1363. DOI: 10.19538/j.cjps.issn1005—2208.2022.12.02.

[9] Feng JW , Yang XH , Wu BQ , et al. Predictive factors for central lymph node and lateral cervical lymph node metastases in papillary thyroid carcinoma[J]. Clin Transl Oncol, 2019, 21(11): 1482-1491.

[10] Wang W , Zhang Z , Zhao Y , et al. Management of Lateral Multiple—Level Metastasis in N1b Papillary Thyroid Microcarcinoma[J/OL]. Front Oncol, 2020, 10: 1586. DOI: 10.3389/fonc.2020.01586.

[11] Liu Y , Wang X , Sun X , et al. Nomogram for predicting the risk of cervical lymph node metastases and recurrence in papillary thyroid carcinoma based on the thyroid differentiation score system and clinical characteristics[J/OL]. BMC Endocr Disord, 2025, 25(1): 39. DOI: 10.1186/s12902—025—01867—7.

[12] Gong J , Zhu B , Liu W , et al. Risk Factors for Lymph Node Metastasis in Papillary Thyroid Carcinoma: A Retrospective Study[J]. Horm Metab Res, 2023, 55(5): 315-322.

[13] Jiang LH , Yin KX , Wen QL , et al. Predictive Risk—scoring Model For Central Lymph Node Metastasis and Predictors of Recurrence in Papillary Thyroid Carcinoma[J/OL]. Sci Rep, 2020, 10(1): 710. DOI: 10.1038/s41598—019—55991—1.

[14] Seifert R , Schäfers MA , Heitplatz B , et al. Minimal extrathyroid extension in papillary micro carcinoma of the thyroid is an independent risk factor for relapse through lymph node and distant metastases[J/OL]. J Nucl Med, 2021, 62(12): 1702-1709. DOI: 10.2967/JNUMED.121.261898.

[15] Lin S , Chen D , Pan CW , et al. Thyroid differentiation score—related genes and prognostic model for thyroid cancer[J/OL]. Transl Cancer Res, 2025, 14(8): 4662-4678. DOI: 10.21037/tcr—2025—460.

基金资助

江苏省卫生健康委2023年度医学科研立项项目(Z2023065)