Background: IO holds promise for EA with mismatch repair (MMR) or MSI-H phenotype, while trials of targeted agents matched by histology/single biomarkers have yet to meet clinical endpoints. We hypothesized a single comprehensive genomic profiling (CGP) assay could better match EA subsets to treatment modalities by simultaneously assessing MSI status, TMB, and targetable GAs. Methods: CGP of 717 FFPE EA clinical specimens by hybridization-capture of up to 315 cancer-related genes (FoundationOne) provided GA (SV, indels, CNA, rearr), TMB, and MSI status. TMB was calculated by counting mutations across a 1.25Mb region and classified as high (TMB-H; range 10.4-541 mut/Mb) or low (TMB-L; 0-10.3 mut/Mb) using the top quartile threshold. MSI high (MSI-H) or stable (MSS) status was assigned by a computational algorithm examining 114 intronic homopolymer loci. Results: Median patient age for 717 advanced or recurrent, treatment refractory EAs was 65 yrs (range 24-92 yrs). 16.9% were MSI-H, which correlated with non-serous/clear cell histology (representing 28.9% of total endometrioid, 18.8% mixed/NOS, 7.1% clear cell, and 1.5% serous cases; p < 0.0001). MMR GA (in MLH1, MSH2, MSH6, PMS2) were identified in 33% of MSI-H cases, compared to 1.9% MSS cases (p < 0.0001). Of the 173 TMB-H cases, 73% were MSI-H while 27% had a hypermutated, non-indel MSS signature (POLE GA in 16.3%). 7.4% of all MSS (43/481) samples were TMB-H and were distributed across histologies: 35% endometrioid, 40% mixed/NOS, 19% serous, 7% clear cell. GA co-occurrence frequencies differed significantly between MSI-H and TMB-H MSS EA for PTEN, MLL2, ARID1A, TP53, and KRAS (p < 0.0001). Patients with MSI-H and MSS TMB-H EA responding to IO, including 2 cases initially excluded by standard testing, will be presented. Conclusions: MSI status by CGP identified 3 times as many potentially IO-relevant cases as MMR GA, and TMB identified an additional 35% of MSS EA, including non-endometrioid subtypes. Co-occurring GA differ between TMB-L, MSI-H TMB-H, and MSS TMB-H, suggesting molecular subgroups relevant to targeted agent efficacy.