Sensitivity and Specificity of Extranodal Extension: Unlocking One of the Strongest Prognostic Factors in Head and Neck Cancer

Springer eBooks(2023)

引用 0|浏览3
暂无评分
摘要
Abstract Extranodal extension (ENE) represents a spectrum of tumor invasion beyond the nodal capsule. The earliest stages of ENE can only be detected under the microscope (pathologic-ENE, pENE). As ENE progresses, it can eventually become visible on imaging (radiologic-ENE, rENE). When ENE further advances to invade skin and/or underlining structures causing fixation and neurovascular impairment, it becomes clinically evident (clinical-ENE, cENE). pENE is the most objective and sensitive way of identifying ENE while subjectivity exists for rENE and cENE detection. Hence, pENE often serves as a gold standard for assessing the accuracy of rENE and cENE. The sensitivity and specificity of rENE for pENE depends on the level of certainty that a radiologist has adopted for declaration. If unequivocal radiologic signs are used for declaration, the specificity of rENE for pENE is very high. Unequivocal rENE carries prognostic significance beyond traditional cN classification for both viral-related and unrelated head and neck cancer, and can serve an important role for clinical care and risk stratification. For clinical care, such as triaging HPV-positive oropharyngeal cancer to surgery vs radiotherapy, a relatively modest level of certainty (>50%) may be used for rENE declaration before treatment assignment to achieve high sensitivity and avoid potential triple-modality treatment. For staging, a high level of certainty (>90%) should be used for rENE declaration to preserve its prognostic importance and avoid dilution due to equivocal cases, or the inclusion of minimal ENE lacking importance due to mitigation by contemporary treatments. Standardization of definitions and radiology reporting templates should facilitate the adoption of rENE into clinical care and staging.
更多
查看译文
关键词
neck cancer,extranodal extension,head
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要