Potential Metastatic Distance of Lymph Nodes in Different Metastatic Patterns for Nasopharyngeal Carcinoma: Implications for Caudal Border of Elective Nodal Target Volume Delineation

Abstract

<h3>Purpose/Objective(s)</h3> To investigate the potential metastatic distance of involved lymph nodes (LNs) which can reveal possible pathways of microscopic spread and determine the delineation of caudal border of clinical target volume (CTV) for elective nodal irradiation (ENI) in nasopharyngeal carcinoma (NPC). <h3>Materials/Methods</h3> A total of 1412 patients with pathologically-proven NPC between January 2010 and December 2020 were enrolled. The center points of caudal edge of the most caudally involved LN, lateral process of C1, hyoid bone, cricoid cartilage and cranial edge of sternal manubrium were marked on treatment planning CT scans of each patient. The metastatic distance of the most caudally involved LN, lengths of entire neck, level II, III and IV were determined. Metastatic patterns and their respective metastatic distances covering 95% of patients were analyzed. The metastatic distance covering 95% of patients with LN involvement in lower level is recommended as the caudal border of CTV for patients with involved LN in the upper level. Recommendations in this study were compared with consensus guidelines to propose modifications to caudal border of CTV. <h3>Results</h3> In 1412 patients, lengths of level II, III, IV and II + III ranged widely from 2.9-8.4, 1.6-5.5, 1.2-9.6 and 5.5-12.6 cm, accounting for 21%-62%, 9%-41%, 9%-59% and 41%-91% of the entire neck, respectively. A total of 1933 LNs were delineated, LN metastasis was step by step with 1% skipping rate. Six main metastatic patterns were Level II (498, 25.8%), RPLN + II (801, 41.4%), Level II + III/Va (111, 5.7%), RPLN + Level II + III/Va (327, 16.9%), Level II + III/Va + IV/Vb (58, 3.0%) and RPLN + Level II + III/Va + IV/Vb (119, 6.2%), with significantly different mean metastatic distances. Their maximum distance covering 95% patients were 6.0, 6.3, 8.8, 9.8, 12.8, and 13.7 cm, respectively, based on which, caudal borders for metastatic patterns N0, RPLN, Level II, RPLN + Level II, RPLN + Level II, Level II + III/Va, RPLN + Level II + III/Va were recommended. Compared with current recommendations based on anatomic landmarks, the caudal borders of CTV shifted upward in 175 (99.4%) N0 patients, 88 (98.9%) with metastatic RPLN only and all patients with Level II and RPLN + Level II node involvement by 3.2, 2.9, 6.1, and 5.1 cm. However, 9(15.5%) and 6(5.0%) of patients with level III node involvement had lower recommended caudal borders of ENI than cranial edge of sternal manubrium (Table 1). <h3>Conclusion</h3> Delineation of the caudal border of CTV based on anatomical markers is highly heterogeneous and unstable. This study suggests using the potential metastatic distance (representing the metastatic capacity) of LNs in the lower level as the caudal border of CTV when with involved LN in the upper level. It could reduce the CTV volume in most patients, leading to an expected significant reduction in acute and late toxicity. However, potential metastatic distance in some patients suggests a lower CTV caudal border, indicating the anatomical marker-cased recommendation are insufficient.