Pathological metastatic lymph node density (ND) predicts early recurrence in papillary thyroid cancer patients after curative resection.

Abstract

Background: A growing body of research has recently suggested a high ratio of metastatic lymph nodes-to-harvested lymph nodes (LNR) as a robust prognostic factor in multiple solid cancers including papillary thyroid cancer (PTC). However, the clinical impact is still elusive. Accordingly, we assessed the clinical significance of LNR to select high-risk patients for recurrence. Here, LNR is described using the term "node density (ND)".

Methods: Clinicopathological analyses were retrospectively conducted including pathological ND on 936 PTC patients who underwent curative resection. The clinical impact of ND was assessed based on the area and the extent of lymph node dissection. The cutoffs for variables were determined using receiver operating characteristic (ROC) curves. The cutoffs for ND were also determined using the X-tile software. Prognostic analysis was performed using Kaplan-Meier and Cox's proportional hazard models. The correlation assay was performed using multivariable logistic regression analysis.

Results: Patients with recurrence had significantly higher ND than their counterparts. Patients with ND ≥17.4% showed significantly poor 10-year recurrence-free survival compared to those with lower ND (79.4% vs. 91.4%, P<0.001). Multivariate prognostic analysis revealed that ND ≥17.4% was an independent predictor of recurrence (hazard ratio =2.20, P=0.03). ND ≥17.4% was independently associated with younger age, male gender, gross extrathyroidal extension, and a large lymph node. Even in subgroup analysis by age or gender, ND ≥17.4% predicted a significantly poor prognosis, particularly in younger patients. ND levels were significantly higher in patients with recurrence in both the central (ND1a) and lateral lymph nodes (ND1b). The two peaks of best cutoffs (around 17% and 35%) were suggested as the optimal cutoffs; 17.4% was best across patients with high specificity, while 38.9% or 34.2% were best in patients with only central neck dissection, or patients with lateral neck dissection, respectively (both are around 35%). High ND than both cutoffs predicted recurrence despite the extent of lymph node dissection. Notably, recurrence occurred earlier in patients with high ND than in their counterparts in both cutoffs.

Conclusions: A high pathological ND strongly predicts an early recurrence in both early and locally advanced PTC. Thus, ND would be considered a legitimate high-risk factor in PTC patients after curative resection.