Geographic and demographic disparities in breast cancer outcomes: A population-based study.

Abstract

7028 Background: Breast cancer (BC) is one of the leading causes of cancer-related mortality worldwide, with over 2 million new cases diagnosed each year. Area of residence has been shown to affect survival, but its impact on BC is unknown. The purpose of this study is to address the potential disparities in BC outcomes between the urban area (UA) and rural area (RA). Methods: We obtained data based on BC patients’ (pts) area of residence; UA vs RA, using the Surveillance, Epidemiology, and End Results (SEER) database treated from 1975-2016. Different socio-demographic variables and tumor characteristics (cts) from both groups were analyzed. Disease specific survival (DSS) and overall survival (OS) were analyzed by Kaplan-Meier methods. Multivariable analyses were conducted using Cox regression models. Results: A total of 99,339 and 872,742 pts were analyzed from RA and UA respectively. Pt cts, including age, sex, marital status, insurance status were similar across both RA and UA. There were a higher number of Black and Hispanic pts in the UA compared to RA. A total of 264,191 pts had receptor subtype data available. A slightly higher percentage of triple negative BC (TNBC) pts was seen in the RA compared to UA (13.5% v 12.7%) whereas UA had a higher percentage of HER2+ BC (18% v 17.3%). There were 95.4% early/locally advanced (stage I-III) and 4.6% metastatic pts. UA pts had significantly improved DSS and OS regardless of stage, grade or other tumor cts compared to RA pts. When stratified by receptor subtype, UA had significantly improved DSS and OS in hormone positive (HR+) (OS HR of 0.86; DSS HR of 0.87; p < 0.05) and HER2+ pts (OS HR of 0.87; DSS HR of 0.83; p < 0.05). However, there were no significant differences seen between UA and RA pts with TNBC. Conclusions: To our knowledge, this is the first study to report BC outcomes based on geographic disparities. Interestingly, HR+ and HER2+ UA pts had better outcomes than RA pts, whereas TNBC pts had similar survival regardless of residence area. Perhaps this disparity is partly due to the longer duration of treatment needed with hormonal therapy in the adjuvant setting making compliance an important factor to take into account in RA pts, in addition to complicated decision making with upcoming data in HER2 + cancer in the early stage setting. Access to healthcare, differences in surgical and systemic approaches of pt care, management of treatment-related toxicities and environmental factors are all potential factors that may affect outcomes in BC pts. Based on the above data, further studies are needed to further evaluate and reduce such disparities.