Prognostic implications of extranodal extension in papillary thyroid carcinomas: A propensity score matching analysis and proposal for incorporation into current tumor, lymph node, metastasis staging

Abstract

BackgroundThe current American Joint Committee on Cancer tumor, lymph node, metastasis cancer staging system for papillary thyroid carcinoma places low weight on extranodal extension. This study examined the prognostic implications of extranodal extension in papillary thyroid carcinoma patients and attempted to design a new staging system incorporating extranodal extension. MethodsWe reviewed data from 6,165 consecutive papillary thyroid carcinoma patients from 2012 to 2018. Patients with extrathyroidal extension or extranodal extension were included and then divided into 3 groups: extrathyroidal extension (papillary thyroid carcinoma with extrathyroidal extension but without extranodal extension, N = 457); extranodal extension (papillary thyroid carcinoma with extranodal extension but without extrathyroidal extension, N = 116); and extrathyroidal extension and extranodal extension (papillary thyroid carcinoma with both extrathyroidal extension and extranodal extension, N = 116). Recurrence-free survival and cancer-specific survival were compared before and after adjusting for differences using propensity score matching owing to observed heterogeneity in baseline characteristics in the original cohort. Recurrence-free survival and cancer-specific survival were also compared between patients with and without extranodal extension after matching at a 1:1 ratio. Cox proportional hazards regression analyses were used to identify the relationships of factors associated with structural recurrent disease in the node-positive subset. Then a new staging system incorporating extranodal extension was established, and the discrimination of the new staging system for recurrence-free survival and cancer-specific survival was investigated. ResultsOf the 6,165 patients with papillary thyroid carcinoma, extrathyroidal extension was found in 573 (9.3%) patients, and extranodal extension was observed in 232 (3.8%) patients. The recurrence-free survival and cancer-specific survival rates of patients with extranodal extension were similar to those of patients with extrathyroidal extension (all P > .05). Patients with extrathyroidal extension and extranodal extension experienced worse recurrence-free survival than patients with extrathyroidal extension or extranodal extension and even worse cancer-specific survival than patients with extrathyroidal extension (all P < .05). The recurrence-free survival and cancer-specific survival rates of patients with extranodal extension were worse than those of patients without extranodal extension (P = .003; P = .048). Cox proportional hazards regression analysis demonstrated that after propensity score matching, extranodal extension (hazard ratio 1.911; 95% confidence interval 1.568–3.609; P < .001) remained an independent predictor of structural recurrent disease in patients with node-positive papillary thyroid carcinoma. After incorporating extranodal extension into the current tumor, lymph node, metastasis classification, the new staging system presented a better discrimination for recurrence-free survival and cancer-specific survival for those with lymph node metastasis. ConclusionPapillary thyroid carcinoma patients with extranodal extension present worse prognosis, and incorporating extranodal extension in tumor, lymph node, metastasis classification identifies poor-risk patients more accurately.