The Impact of Radiation Dose on Survival of Salivary Gland Tumors Managed With Definitive Radiation Therapy

Abstract

Cancers of the major salivary gland cancers (MSGCs) are comprised of heterogeneous histologic subtypes and diverse natural histories. Surgical resection with or without adjuvant radiation therapy (RT) is the preferred primary management approach. If surgery is not an option, primary RT is the mainstay of treatment, but optimal doses and techniques are not well-defined. We aimed to identify the association between RT doses and techniques and the survival of MSGCs treated with definitive RT. We hypothesized that higher RT dose would be associated with better overall survival (OS). We queried the National Cancer Database for cases of MSGC patients diagnosed between 2004 and 2013 including only those presenting with single primary non-metastatic cases that were managed with definitive radiotherapy to at least 40 Gy or higher. Cases were grouped by RT dose (<66 Gy vs. ≥66 Gy) and by modality (IMRT vs. non-IMRT). Kaplan Meier of OS was performed and log-rank compared OS among subgroups. Multivariate Cox proportionate hazards regression was performed to identify factors associated with increased risk of death. Propensity score matching was performed to control for imbalance in covariates such as stage, anatomic site, and chemotherapy use. The cohort included 730 patients, predominantly males (68%), aged 60 or older (81%) with tumors arising from the parotid (79%). The most common histology was squamous cell carcinoma (41.6%), and the most common clinical stage was 4A (31.8%). IMRT was employed in 45% of the cases and 51.6% of patients also received chemotherapy. At median follow-up of 20.2 months, the 5-year OS for the cohort was 27% (95% confidence interval [CI]: 22.9%-30.9%). Multivariate analysius identified older age, higher grade, non-parotid sites and lower RT dose as unfavorable prognostic factors. After propensity score matching, patients treated to ≥66 Gy had significantly higher OS (p=0.029) and median survival (30.3 months [95% CI 25.6-35.1] vs. 19.75 months [95% CI 16.4-23.1]). RT technique was not associated with signficant improvement of survival. No differences in OS by histological subgroups were observed with regard to either dose or modality. In MSGC managed with definitive RT, treatment to higher doses (≥66 Gy) is associated with better survival outcomes. Our findings suggest that patients with MSGC who require nonsurgical treatment should be managed with high-dose RT to maximize survival.