The Development of Brain Metastases in Patients with Renal Cell Carcinoma: Epidemiologic Trends, Survival, and Clinical Risk Factors Using a Population-based Cohort

Abstract

BackgroundThe incidence of brain metastases (BM) in patients with renal cell carcinoma (RCC) is hypothesized to have increased in the last 2 decades. ObjectiveTo define incidence trends according to patient and clinical characteristics, to identify risk factors, and to describe outcomes of patients with BM for RCC. Design, setting, and participantsPatients diagnosed with RCC between the years 2010 and 2013 within the Surveillance, Epidemiology, and End Results database. An external validation was also considered using patients diagnosed with RCC between 2010 and 2012 within the National Cancer Database. Outcome measurements and statistical analysisIncidence proportions of BM were calculated. Risk factors correlated with BM at diagnosis were identified via a 1000-bootstrap corrected multivariable logistic regression model. A risk model was then developed and evaluated using measures of predictive accuracy. Overall survival was examined using Cox regression analyses. Results and limitationsThe overall incidence proportions of BM at RCC diagnosis was 1.51% (95% confidence interval: 1.39–1.64%). White/other race, clear cell histology, and sarcomatoid differentiation, T2–4 disease, tumor dimension >10 cm, and N+ disease were significantly associated with BM at RCC diagnosis, and retained within the final prediction model. A risk score was created based on these variables (c-index: 0.803). BM at RCC diagnosis occurred in 0.5%, 3.6%, and 7.7% of patients categorized as low risk, intermediate risk, and high risk. Patients with BM were more likely to succumb to any death than those without BM at diagnosis (median overall survival: 6.4 mo vs not reached, respectively, adjusted hazard ratio: 1.87, 95% confidence interval: 1.67–2.08, p < 0.001). The real incidence of BM at RCC diagnosis is likely underestimated given that the observed rate likely reflects patients who presented with symptoms. ConclusionsPatients with BM at RCC have poor oncological outcomes. We have characterized the epidemiology of BM at RCC diagnosis and developed a clinical risk model for the purpose of predicting the development of BMs in patients diagnosed with a cortical renal mass. Patient summaryIn this report we examined recent proportions of patients with brain metastases at kidney cancer diagnosis in a large community database originating from the US. We developed a model that may be used during routine clinical practice to predict brain metastases. The urologic-oncological community may consider baseline imaging for brain metastases in patients without any symptoms but at high risk of having brain metastases according to the risk model. However, the proposed model certainly needs further testing and validation in the clinical setting. Future studies on brain metastases survival and treatment options are also needed.