文献速递头颈外科

早期口腔癌何时需行选择性颈清扫?1109例多中心研究及多参数预测模型

When Should Elective Neck Dissection Be Considered for Early-Stage Oral Cavity Tumors? Insights From a Multicenter Study of 1109 Patients and Development of a Multiparametric Predictive Model.
PubMed 文献 · Head Neck · 2025-10-13

该多中心回顾性研究纳入1109例cT1-T2 cN0口腔鳞癌患者,分析隐匿性淋巴结转移的预测因素。结果显示DOI、LVI、PNI、WPOI和分级为显著预测因子,整合这些变量的模型优于仅用DOI的模型(AUROC比较,p<0.01),可个体化指导选择性颈清扫。

要点

  • cT1-T2 cN0口腔鳞癌中隐匿性淋巴结转移率为28%。
  • DOI、LVI、PNI、WPOI和分级是隐匿性转移的显著预测因子。
  • 多参数模型预测准确性优于仅用DOI的模型(p<0.01)。

核心结论

  • 整合多个组织病理学参数的预测模型可更精准指导早期口腔鳞癌的选择性颈清扫,实现个体化风险评估。

研究设计与样本

  • 通过REDCap平台收集组织病理学参数、肿瘤特征和患者结局数据。
  • 采用简单和多变量逻辑回归模型评估隐匿性淋巴结转移的预测因素。

主要发现

  • 585例(53%)为cT1,503例(45%)为cT2,cTis占1.9%。
  • 1109例中193例(28%)存在隐匿性淋巴结转移。
  • DOI、LVI、PNI、WPOI和分级为隐匿性淋巴结转移的显著预测因子。
  • 整合上述变量的预测模型准确性优于仅用DOI的模型(AUROC比较,p<0.01)。

关键数值与推荐

  • 隐匿性淋巴结转移率28%(193/1109例cT1-T2 cN0口腔鳞癌患者)。
  • 多参数模型与仅用DOI模型AUROC比较p<0.01。

临床怎么用

  • 对早期口腔鳞癌cN0患者,应综合评估DOI、LVI、PNI、WPOI和分级等多参数,而非仅依赖固定DOI cutoff。
  • 所提出的预测模型可用于个体化评估隐匿性淋巴结转移风险,辅助选择性颈清扫决策。

以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。

原文

BACKGROUND: Nodal metastases significantly affect prognosis in oral cavity squamous cell carcinoma (OCSCC). In early-stage OCSCC (cT1-T2 cN0), management of the clinically negative neck (cN0) remains controversial. Depth of invasion (DOI) is a key determinant for END, but other histopathological parameters, such as grading, perineural invasion (PNI), lymphovascular invasion (LVI), and worst pattern of invasion (WPOI), are emerging prognostic factors. METHODS: This multicenter retrospective study analyzed 1109 patients with cT1-T2 cN0 OCSCC treated at 30 Italian hospitals since 2017. Data on histopathological parameters, tumor characteristics, and patient outcomes were collected via the REDCap platform. Simple and multivariable logistic regression models were developed to assess predictors of occult nodal metastases. RESULTS: A total of 585 patients were clinically classified as cT1 tumors (53%), 503 as cT2 (45%), and cTis (1.9%). Of the 1109 patients, 193 (28%) had occult nodal metastases, with DOI, LVI, PNI, WPOI, and grading emerging as significant predictors. A predictive model integrating these variables demonstrated superior accuracy compared to a DOI-only model (AUROC comparison, p < 0.01). CONCLUSION: This study highlights the importance of incorporating multiple histopathological parameters into risk models for occult nodal metastases, overcoming the fixed DOI as a cutoff. The proposed predictive model offers a more precise method for guiding END in early-stage OCSCC, allowing individualized risk estimation.

来源:Head Neck,查看原文。版权归原作者所有。

引用

  1. Calabrese L, Abousiam M, Tagliabue M, et al. When Should Elective Neck Dissection Be Considered for Early-Stage Oral Cavity Tumors? Insights From a Multicenter Study of 1109 Patients and Development of a Multiparametric Predictive Model. Head Neck. 2025. PMID: 41078311. doi:10.1002/hed.70064

供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。