Development of a Simple Nodal Burden Score for Structural Recurrence in pN1a Papillary Thyroid Carcinoma

Background/Objectives: Quantitative nodal characteristics may refine recurrence risk assessment in papillary thyroid carcinoma (PTC). The aim of this study was to develop a simple Nodal Burden Score (NBS), combining metastatic lymph node number, lymph node ratio and extranodal extension, and to evaluate its association with structural recurrence and its discriminative performance in patients with low- and intermediate-risk pN1a PTC. Methods: This retrospective single-center cohort included 256 adults with American Thyroid Association (ATA) low- or intermediate-risk pN1a papillary thyroid carcinoma who underwent initial thyroid surgery with central neck dissection between 2018 and 2021. Postoperative radioactive iodine was administered to 165 patients (64.5%). NBS assigned one point each for >3 metastatic lymph nodes, lymph node ratio (LNR) ≥0.50, and extranodal extension (ENE). Patients were grouped as NBS 0, 1, and 2–3. The node-count threshold was selected through exploratory analyses in the development cohort. Cox regression assessed associations with structural recurrence. Discrimination was evaluated using Harrell’s concordance C-index and bootstrap internal validation. Results: During a median follow-up of 57 months, 27 patients (10.5%) developed structural recurrence. In the exploratory multivariable model, >3 metastatic nodes (Hazard ratio (HR) 2.74, 95% CI 1.25–6.02), LNR ≥ 0.50 (HR 2.51, 95% CI 1.13–5.57), and ENE (HR 2.38, 95% CI 1.04–5.45) were associated with recurrence. Observed recurrence proportions were 3.2%, 9.8%, and 30.6% for NBS 0, 1, and 2–3, respectively. The apparent C-index increased from 0.68 for the ATA-category-and-tumor-size model to 0.80 after adding NBS. The expanded model’s bootstrap optimism-corrected C-index was 0.78; the corrected difference between models was not established. Conclusions: NBS separated observed recurrence proportions in this development cohort. Its data-derived threshold, limited event count, and incomplete assessment of incremental performance warrant further internal evaluation and independent external validation before clinical use.