Prognostic effects of previous cancer history on patients with major salivary gland cancer.

Abstract

To explore the prognostic effects of previous cancer history on patients with major salivary gland cancer (SGC). SGC patients with (sec-SGC) and without (one-SGC) a previous cancer from the SEER database were identified. Cox proportional hazards regression (CoxPH) models were used to compare the prognosis between sec-SGC and one-SGC patients. Subgroup analyses for sec-SGC patients by gender, previous cancer types, previous cancer histology, and cancer diagnosis interval (CDI) were performed. Two CoxPH models were constructed to distinguish sec-SGC patients with different prognostic risks. 9098 SGC patients were enrolled. Overall, sec-SGC patients (adjusted HR [aHR]=1.26, p < 0.001), especially those with a CDI ≤ 5 years (aHR=1.47, p < 0.001), had worse overall survival (OS) than one-SGC patients. In subgroup analysis, only sec-SGC patients with a previous head and neck cancer who were female (aHR=2.38, p=0.005), with a CDI ≤ 5 years (aHR=1.65, p=0.007) or with a previous squamous cell carcinoma (aHR=6.52, p < 0.001) had worse OS. Our models successfully differentiated all sec-SGC patients into high-, intermediate- and low-risk groups with different prognosis. Sec-SGC patients with different previous cancer types, gender, CDI and previous cancer histology had varied prognosis. The models we constructed could help differentiate the prognosis of sec-SGC patients with different risks.