RESUMEN
BACKGROUND: Bone metastases are rare in oral squamous cell carcinoma (OSCC). It has not been defined on the risk and prognosis of OSCC patients with bone metastases. The purpose of this study was to assess the factors associated with the development and prognosis of bone metastases among OSCC patients. METHODS: Demographic and clinicopathological characteristics of OSCC patients diagnosed between 2010 and 2019 was retrieved from the Surveillance, Epidemiology, and End Results (SEER) database. To explore risk factors for developing bone metastases and prognosis, the univariate and multivariate logistic and Cox regression analysis were performed, further the predictive nomogram models were constructed. RESULTS: The incidence rate of bone metastases in newly diagnosed OSCC patients was 0.91 % (95 %CI 0.81% -1.02 %). Ultimately, 137 OSCC patients with bone metastases and 19,469 OSCC patients without bone metastases were included in the present study. Pathological grade, primary site, T/N stage and distant organ metastases (liver/lung/brain) were independently associated with the risk of developing bone metastases among OSCC patients. The C-index of a constructed risk-predicting nomogram was 0.86 (95 %CI 0.83-0.89). Multivariate Cox regression analysis indicated that lung metastases, the use of surgery as well as chemotherapy were three independent prognostic factors. The C-indexes of constructed risk-predicting nomograms were 0.70 (95 %CI 0.65-0.75), 0.68 (95 %CI 0.63-0.73) for OS and CSS, respectively. Calibration plots demonstrated an agreementbetween the established nomogram's predicted survival and actual survival. In addition, decision curve analysis (DCA) indicated these established nomograms had considerable net benefits and clinical utilities. CONCLUSION: This study defined the risk and prognostic factors for bone metastases among OSCC patients and the established nomograms were well calibrated for discrimination to predict bone metastasis development and prognosis.