Lymph node density as prognostic factor in regional recurrent or residual head and neck cancer.

Background: Prognostic factors from neck dissection for regional recurrent or residual head and neck cancer remain undetermined.

Aims/objectives: We studied whether continuous variables from pathologic lymph nodes of salvage neck dissection could estimate survival outcomes in patients with head and neck squamous cell carcinomas (HNSCC).

Materials and methods: Pathologic continuous variables from neck dissection without primary tumor resection in 184 patients with HNSCC were retrospectively evaluated. Cutoff values from both lymph node density (LND) and log odds of positive lymph nodes (LODDS) as the variables were applied to death estimation. Uni-/multivariate survival analyses adjusting for pathologic N classification and extranodal extension were evaluated by log-rank test or Cox's proportional hazard model, respectively.

Results: LND ≥ 0.14 and LODDS ≥ -1.72 were significantly related to univariate shorter survival outcomes. Multivariate analysis indicated a significant association between LND ≥ 0.14 and shorter overall (hazard ratio [HR] = 2.16, 95% confidence interval [CI] = 1.30-3.58), local recurrence-free (HR = 2.34, 95% CI = 1.06-5.16), regional recurrence-free (HR = 2.32, 95% CI = 1.27-4.25), and distant metastasis-free survival (HR = 2.17, 95% CI = 1.31-3.61).

Conclusions and significance: Based on the obtained results, increased LND was a poor prognostic factor following salvage neck dissection.