Differences in cancer presentation, treatment, and mortality between rural and urban patients diagnosed with kidney cancer in the United States.

Introduction: Rural-urban discrepancies in care and outcomes for kidney cancer (KCa) in the United States remains poorly understood. Our study aims to improve our understanding of the influence of rurality on KCa outcomes in the United States by analyzing differences in presentation, treatment, and mortality between urban areas (UAs) and rural areas (RAs) in the Surveillance, Epidemiology, and End Results (SEERs) database.

Methods: SEERs data was queried from 2000 to 2019 for KCa patients. Patient counties were classified as UAs, rural adjacent areas (RAAs), or rural nonadjacent areas (RNAs) using Rural Urban Continuum Codes. Demographic, tumor characteristics, and treatment variables were compared. Propensity score matching was performed to create matched UA-RAA and UA-RNA cohorts. Multivariate regression evaluated rural-urban status as a predictor of treatment selection. Multivariate cox regression assessed the predictive value of rural-urban status for overall survival (OS) and cancer-specific survival (CSS). Kaplan-Meier analysis was used to generate survival curves for OS and CSS.

Results: 179,509 KCa patients were identified (UA = 87.0%, RAA = 7.7%, RNA = 5.3%). Patients in RAs were more likely to present with tumors of higher grade and stage than UAs. Following multivariate analysis, rural residency predicted undergoing nephrectomy (RAA: OR = 1.177, RNA: OR = 1.210) but was a negative predictor of receiving partial nephrectomy (RAA: OR = 0.744, RNA: OR = 0.717), all P < 0.001. Multivariate cox regression demonstrated that RAA or RNA residency was predictive of overall and cause-specific mortality. After matching, median OS was 151, 124, and 118 months for UA, RAA, and RNA cohorts respectively; mean CSS was 152, 147, and 144 months for UA, RAA, and RNA cohorts, respectively, all P < 0.001. Stage-specific analysis of CSS demonstrated significantly poorer CSS among RNA patients for localized, regionalized, and distant KCa after matching. Only RAA patients with localized KCa experienced significantly lower CSS than UA patients.

Conclusions: Patients in RAs are more likely to present with advanced KCa at diagnosis compared to those in UAs and may also experience different treatment options including a lesser likelihood of undergoing partial nephrectomy. Rural patients with KCa also demonstrated significantly worse OS and CSS compared to their urban counterparts. Further patient-level studies are required to better understand the discrepancy in CSS between urban and rural patients diagnosed with KCa.