(076) Real-World utilization of 1L systemic therapy amongst Black and Non-Hispanic White patients with advanced or recurrent endometrial cancer

Background: Endometrial cancer (EC) is the most frequent gynecological malignancy in the US. Black women are more likely to be diagnosed with aggressive subtypes of the disease at later stages, which affects their prognosis. Previous research has found that non-White women are less likely to receive treatment for endometrial cancer that follows National Comprehensive Cancer Network guidelines (NCCN). Such observed gaps are complex and may be driven by factors including tumor characteristics, practice patterns, access to care, and socioeconomic factors.

Objectives: To evaluate real-world utilization of first-line (1L) systemic therapy amongst patients with advanced or recurrent endometrial cancer (aEC).

Methods: This cohort study used a nationwide Flatiron Health electronic health record-derived de-identified database to identify newly diagnosed cases of aEC (initial diagnosis of stage III or IV EC or an initial diagnosis of stage I or II EC with subsequent locoregional or distant recurrence) that initiated 1L systemic therapy between 2013-2022.

Results: We identified 2053 subjects who met the inclusion criteria. The median age at 1L therapy initiation was 68 years [IQR, 62–74 years]. Most patients (99%) diagnosed with stage I-II EC received surgery as part of their primary treatment. In contrast, only 19% of those diagnosed with stage III-IV received radiotherapy and there was no difference by race (p=0.54). Most patients received a platinum combination in the 1L, with 53% receiving carboplatin-paclitaxel with no difference in receipt of treatment by race (p=0.18). Black patients represented 22% of the cohort, of whom 40% had serous carcinoma histology, compared to 24% of non-Hispanic Whites. A higher proportion of Black patients presented with advanced cancer staging (III or higher) at their initial diagnosis (79% vs 62%; p< 0.01). Similarly, a higher proportion of Black were in the lowest socioeconomic quintile (45% vs 14%; p< 0.01). This observation was made using an area-level socioeconomic status measurement that relied on U.S. census block group data.

Conclusions: As disparities are complex, future studies should evaluate race as a biological construct when assessing survival differences and quantify the impact of causative risk factors on survival endpoints by utilizing causal mediation analyses.