Cancer-specific and overall survival in African American women with endometrial cancer: a SEER-Medicare study

Abstract

<h3>Objectives:</h3> To compare the overall survival (OS) and cancer specific survival (CSS) by race for women with stage I-III endometrial cancer (EC) by histologic subtype. <h3>Methods:</h3> This is a retrospective cohort study of women with stage I-III endometrioid, serous, clear cell, and carcinosarcoma who underwent hysterectomy as primary surgical staging as identified within the SEER-Medicare database. Endometrial cancer cases diagnosed between 2000 and 2015 with Medicare enrollment data from 1999-2016 were included. Demographics, cancer data, and survival outcomes were collected. Demographic and treatment factors by race were compared using descriptive analyses. OS and CSS outcomes were stratified by race, stage and histology, then analyzed using log rank tests and multivariate cox modeling (adjusting for age, geographic region, Charlson comorbidity index, post-operative treatment sequencing, FIGO stage, and grade). <h3>Results:</h3> 24,142 total patients were included in this dataset. 85.5% were White, and 8.5% were BlackBlack. 19,351 stage I, 1,484 stage II, and 3,307 stage III cases were identified. Black women were diagnosed at a younger age bracket (40-69yo) than White patients for endometrioid stage I (40% vs 35%), II (37% vs 31%), and III (31% vs 28%); serous stage I (45% vs 31%) and III (42% vs 28%), and carcinosarcoma stage I (39% vs 27%), p<0.05 all. Charlson comorbidity indexes were significantly higher for BlackBlack patients with endometrioid stages I (p<0.001), II (p=0.001), and III (p=0.02) and serous stage I (p<0.0001) and III (p=0.02). Receipt of adjuvant therapy did not differ by race for any stage or histology except for stage III endometrioid where Black women were less likely to receive adjuvant treatment (61.2% vs 70.1% White p=0.03). For stage I, BlackBlack patients had worse CSS and OS for all histologies except clear cell in unadjusted and adjusted analyses (Table I). For stage II, only CSS for endometrioid carcinoma differed by race in unadjusted analyses, whereas OS did not. For stage III, BlackBlack patients with endometrioid carcinoma had worse CSS and OS in unadjusted analyses, but significant survival differences only remained for CSS in the adjusted model. <h3>Conclusions:</h3> Prior studies have reported that BlackBlack women with endometrial cancer have worse survival consistently across stage and histology. However, we did not find this effect when evaluating by CSS and OS. For stage I, BlackBlack women had consistently worse CSS likely related to the disease itself and worse OS in adjusted analyses likely related to poorer overall health outcomes across all histologic subtypes except for clear cell. For stage II and III, racial disparities were seen only for CSS in endometrioid patients in the adjusted analyses, but not OS; and this may have been related to BlackBlack women with stage III cancer receiving less adjuvant therapy. When stratified by stage and histology, BlackBlack women with EC were diagnosed at younger age and with more comorbidities. Overall these findings emphasize the need to intervene in the early stage population, where a disparity exists and the largest population of women is affected.