Background: Parotid cancer (PC) is a heterogeneous group of malignant tumors whose management mainly focuses on locoregional control. However, distant metastases (DM) can represent the most common cause of treatment failure. We have defined the predictors and developed a model that can predict a single patient's risk to develop DM.
Methods: We have analyzed our institutional database of 259 patients with PC and use it to develop a nomogram. C-index and calibration curves were used to assess performance of our model.
Results: DM appeared in 18.9% of patients. Age, cT, pN, perineural invasion, and adenoid cystic carcinoma were significantly associated with distant failure.
Conclusions: We here present the first model to identify patients with PC at high risk of DM. Such tool can be of great value in managing these rare cancers in terms of a more precise prognosis and follow-up while waiting for better systemic therapies to come in the future.
Keywords: head and neck cancer; metastases; nomogram; parotid cancer; risk factors.
© 2019 Wiley Periodicals, Inc.