Abstract
PURPOSE: The coexistence of TERT promoter and BRAFV600E mutations is strongly linked to aggressive behavior and poor prognosis in papillary thyroid carcinoma (PTC). This study aimed to develop preoperative and postoperative predictive models for coexisting mutations based on ultrasound and clinicopathological characteristics to stratify prognostic risks and guide clinical decision-making. METHODS: Retrospective analysis of the ultrasound and clinicopathological characteristics of 120 patients with a surgical pathology of PTC with TERT promoter and BRAF(V600E) gene testing results in the Affiliated Hospital of Jining Medical University from December 2020 to December 2023. Univariate and multivariate logistic regression identified independent predictors, and nomograms were constructed. Model performance was evaluated using ROC curves, calibration curves, and decision curves, with internal validation via Bootstrap resampling. RESULTS: Age (OR: 1.24; 95% CI 1.12-1.37, P<0.001), tumor size (OR: 5.51; 95% CI 2.26-13.46, P<0.001), lateral lymph node metastasis (OR: 7.36; 95% CI 1.48-36.48, P=0.015), and irregular/ill-defined margins (OR: 6.06; 95% CI 1.19-30.75, P=0.030) were independent predictors of coexisting mutations. The cutoff values for age and tumor size were 44.5 years and 1.55 cm, respectively. Two models incorporating the four independent predictors were established to predict coexisting mutations in the preoperative and postoperative periods, achieving AUCs of 0.95 and 0.96, respectively, with both models demonstrating good calibration ability and clinical practicability through calibration and decision curve analyses. CONCLUSION: The predictive models enable clinicians to identify high-risk patients with coexisting mutations both preoperatively and postoperatively, supporting the development of individualized treatment strategies and potentially improving patient outcomes. However, the study is limited by its single-center design, and further external validation is needed to confirm the generalizability of the model.