Interaction of race and socioeconomic status as risk modulators of treatment delay and cancer-specific mortality in uterine cancer.

Abstract

5595 Background: The majority of studies of uterine cancer combine high and low-grade histologies and do not sample a diverse cohort of patients. In many studies race is treated as biologic construct, when it may be better thought of as a proxy for socioeconomic inequity and deprivation. Socioeconomic (SE) deprivation may play a significant role in the disease trajectory of women with uterine cancer. Methods: Data were drawn from the Metropolitan Detroit Cancer Surveillance System which covers a tri-county area of approximately 4 million people. We included non-Hispanic Black (NHB) and White (NHW) women diagnosed with uterine cancer between 2010 and 2018. Poorly differentiated and undifferentiated endometrioid, serous, clear cell, mixed, carcinosarcoma and mucinous histologies were considered high grade. Patients diagnosed by death certificate, or with unknown stage or histology were excluded. Socioeconomic status was assessed using the Yost Score, an area-level composite measure of socioeconomic deprivation derived from census-tract data at cancer diagnosis. Lower Yost quintile indicates higher deprivation. Competing risk analysis was used to determine risk of uterine cancer specific mortality (reported as subdistribution hazard ratio [SHR]) and to assess statistical interaction between race and Yost score. Results: A total of 4,840 patients were identified. Race conferred significant increased risk of cancer-specific mortality (SHR 2.11, p < 0.0001). Race and Yost score interacted to increase risk of cancer-specific mortality in NHB women in the lowest Yost quintile (SHR 2.23, p < 0.0001) compared to NHW and NHB women in the highest quintiles. The interaction between race and Yost score persisted only among women with low grade cancers (SHR 1.7, p = 0.04). Time from diagnosis to surgery increased as Yost score decreased. Women in the lowest Yost quintile had lower likelihood of receiving surgery within 6 weeks of diagnosis (OR 0.74, p = 0.001). This effect persisted among women with low grade cancer (NHB OR 0.75, p = 0.014; lowest Yost quintile OR 0.68, p < 0.0001). An association between race, Yost score and delays in time to surgery was not seen among women with high grade cancers. Conclusions: Race and Yost score, an area-based measure of socioeconomic deprivation, are associated with increased cancer-specific mortality risk among women with low grade cancer. NHB race and high socioeconomic deprivation are associated with delayed primary surgery. The interaction between race and socioeconomic deprivation may underlie known disparities in uterine cancer survival, particularly in low grade disease where there is the greatest opportunity for timely curative surgery.