The evaluation of neck lymph node metastasis is critical for predicting survival after head and neck cancer treatment. However, traditional pathological N staging does not completely correlate with survival; the total number of lymph nodes resected during surgery affects staging, and a minimal number of nodes must be resected to achieve a superior outcome. Thus, the prognostic abilities of various lymph node staging systems for oral cavity squamous cell carcinoma (OSCC)-positive lymph nodes were compared.
Data for 639 patients with OSCC-positive nodes who were treated and monitored at the Changhua Christian Hospital were retrospectively analyzed. The different N staging systems were compared to evaluate their disease-free survival (DFS) predictability.
The areas under the receiver operating characteristic curve were as follows: 0.551 for the traditional American Joint Committee on Cancer (AJCC) N staging, 0.60 for lymph node density (LND), 0.596 for log odds of positive lymph nodes (LODDS), and 0.597 for the number of metastatic lymph nodes (nmLN). The LND, LODDS, and nmLN systems could predict DFS better than AJCC N staging. Multivariable analysis for DFS revealed that extranodal spread, level IV or V positive nodes, and tumor invasion deeper than 13 mm were independent prognostic factors in these four models. LND and LODDS predicted DFS better than pathological N staging.
LND and LODDS staging predicted DFS better than AJCC N staging for OSCC-positive nodes. In the future, the prognostic ability of AJCC staging may be strengthened by LND or LODDS staging.

Author