Construction and validation of a nomogram for predicting cervical lymph node metastasis in classic papillary thyroid carcinoma.

ABSTRACT

PURPOSE: Patients with papillary thyroid carcinoma (PTC) frequently present a relatively poor prognosis when they coexist with cervical lymph node metastasis (LNM). Moreover, it remains controversial whether prophylactic lymph node dissection (LND) should be performed for patients without clinically lymph node metastasis. Thus, we hereby develop a nomogram for predicting the cervical LNM (including central and lateral LNM) in patients with PTC. METHODS: We retrospectively reviewed the clinical characteristics of adult patients with PTC in the surveillance, epidemiology, and end results (SEER) database between 2010 and 2015 and in our Department of Breast and Thyroid Surgery in the Second Affiliated Hospital of Chongqing Medical University between 2019 and 2020. RESULT: A total of 21,972 patients in the SEER database and 747 patients in our department who met the inclusion criteria were enrolled in this study. Ultimately, six clinical features including age, gender, race, extrathyroidal invasion, multifocality, and tumor size were identified to be associated with cervical LNM in patients with PTC, which were screened to develop a nomogram. This model had satisfied discrimination with a concordance index (C-index) of 0.733, supported by both internal and external validation with a C-index of 0.731 and 0.716, respectively. A decision curve analysis was subsequently made to evaluate the feasibility of this nomogram for predicting cervical LNM. Besides, a positive correlation between nomogram score and the average number of lymph node metastases was observed in all groups. CONCLUSION: This visualized multipopulational-based nomogram model was successfully established. We determined that various clinical characteristics were significantly associated with cervical LNM, which would be better helping clinicians make individualized clinical decisions for PTC patients.