Quantitative Metastatic Lymph Node Regions on Magnetic Resonance Imaging Are Superior to AJCC N Classification for the Prognosis of Nasopharyngeal Carcinoma

J Oncol. 2018 Dec 2:2018:9172585. doi: 10.1155/2018/9172585. eCollection 2018.

Abstract

Purpose: Quantitative lymph node burden has been demonstrated to be a critical prognosticator in various malignancies, yet it was seldom explored in nasopharyngeal carcinoma (NPC). This study aimed to investigate the impact of the number of metastatic lymph node regions (LRN) on prognosis of NPC and to establish a new N classification system based on LRN.

Methods and materials: The magnetic resonance images (MRI) of 354 nondisseminated NPC patients before radical treatment were retrospectively evaluated. The regions with positive lymph nodes (LNs) were quantified according to 2013 updated guidelines for neck node levels. Prognostic value of LRN on distant metastasis-free survival (DMFS) was analyzed using multivariable Cox model after adjusting for other nodal characteristics and therapeutic factors.

Results: LRN strongly correlated with the size, laterality, level, extracapsular extension (ECE), and necrosis of LNs. Risk of distant metastasis significantly escalated with increased LRN (P<0.001). 5-year DMFS for LRN 0-1, 2-6, and 7 was 97.0%, 86.7%, and 69.7%, respectively. In multivariable Cox analysis, LRN (HR 2.45; 95% CI 1.55-3.88; P<0.001) and maximal LN diameter (MLD) >6cm (HR 4.11; 95% CI 2.23-7.56; P<0.001) were identified as independent predictors of DMFS. Laterality and level showed no prognostic significance when accounting for LRN. A novel N classification scheme was derived by recursive partitioning analysis based on LRN and MLD. Compared with the 7th and 8th edition of American Joint Committee on Cancer (AJCC) systems, the new stratification exhibited better accuracy in predicting survivals.

Conclusions: LRN is a promising quantitative predictor of survival in NPC, eclipsing other classic LN characteristics in prognostic value. The simplified N classification scheme with LRN and MLD is predictive and practical, thus warranting further validation in future.