Background: Predicting recurrence in low grade, early stage endometrial cancer (EC) is both difficult and important, as recurrence outside the pelvis is typically incurable. We sought to determine whether somatic mutations from NGS panels could help identify at-risk patients. Methods: This was a single-institution, retrospective cohort study of 342 patients. Clinical data were obtained by review of the electronic medical record. Low grade, early stage EC was defined as FIGO grade 1 or 2 endometrioid histology with stage I or II (EC confined to the uterus) at the time of hysterectomy. Molecular testing was performed using NGS panels of 46 or 50 genes (clinical assay) or 200 genes (research assay). Univariate, multivariate, and Kaplan-Meier statistical analyses were performed to identify variables that were associated with recurrence-free survival (RFS). Results: In the cohort, 72% of patients had endometrioid histology. The most frequent mutations in these tumors were PTEN, PIK3CA, ARID1A, CTNNB1, and KRAS. Among 137 patients with low grade, early stage endometrioid EC, CTNNB1 mutation and TP53 mutation were significantly associated with decreased RFS. On multivariate analysis, CTNNB1 mutation, TP53 mutation, and age were the only significant predictors of worse RFS. Multivariate analysis using a combination of CTNNB1 mutation or TP53 mutation showed a hazards ratio of 4.65 for the combined mutation group in low grade, early stage endometrioid tumors (Table 1). Only one patient in the low grade, early stage group had a mutation in both TP53 and CTNNB1. Conclusions:CTNNB1 and TP53 mutations in low grade, early stage, endometrioid EC predicted a subset of patients with worse RFS in a multivariate model. This information could be incorporated into updated adjuvant treatment strategies.