Introduction/Background Endometriosis is a benign pathological condition characterized by the ectopic presence of endometrial tissue. Whether endometriosis predisposes the pathogenesis of endometrial cancer (EC) is still debated. This study uses realworld data (RWD) from the network of TriNetX healthcare organization (HCO) networks in the US (TNX-US) and EMEA (TNX-EMEA) to analyze the impact of endometriosis as a risk factor for the development of EC.

Methodology Using TriNetX Platform, we defined a cohort of 284,287 patients with endometriosis and at least 6 months of follow up at the HCO, 254,726 from TNX-US and 29,561 TNX-EMEA. Propensity score matching between these cohorts and the female control cohorts in each regional network was used to remove the possible confounding effects of age, body mass index (BMI), previous diagnosis of pelvic inflammatory disease, breast cancer, other cancer of female genital organs or genetic susceptibility to cancer. Hazard ratio (HR) was used to compare the incidence of EC between the matched cohorts. Kaplan Meier analysis was used to compare the overall survival (OS) of EC patients with previous endometriosis vs those without endometriosis patients after propensity score matching. The time window of observation in both analyses was 10 years.

Results Patients with endometriosis diagnosis had a higher risk of developing EC in both TNX-US (2,151/237,034 vs 620/238,837, HR 3.49, 95% CI 3.19–3.82) and TNX-EMEA (319/28,241 vs 41/28,282, HR 7.58, 95% CI 5.48–10.50). The OS of EC patients with endometriosis was demonstrated to be significantly better than those without endometriosis: the 10-year OS probability was 78.37% vs 62.41% (p<0.01) and 73.11% vs 49.61% (p<0.01), in TNX-US and TNX-EMEA, respectively.

Conclusion Our RWD supports the association between endometriosis and an increased risk of developing EC. Endometriosis-associated tumors appear to have a better prognosis.

Disclosures U.D.G. has received advisory board or consultant fees from Merck Sharp & Dohme, Bristol My-ers Squibb, Janssen, Astellas, Sanofi, Bayer, Pfizer, Ipsen, Novartis, and Pharmamar and institutional research grants from Astrazeneca, Sanofi, and Roche. A.F. has received personal honoraria for lectures from Astrazeneca, GSK-Tesaro, Clovis, and advisory board from Jannsen, Astrazeneca, GSK-Tesaro. The other authors declare no conflict of interest.

TriNetX contributed in the collection and analyses of the data, but had no role in interpretation of data, in the writing of the manuscript, or in the decision to present the results.