Abstract Background: Endometrial cancer (EC) is the most common gynecological cancer in the US, with increasing incidence and mortality rates. The survival rate lags behind four decades ago (81% vs 87%), underscoring the critical need for more effective treatment strategies. The heterogeneous nature of EC contributes to varied outcomes with current treatments. This project aims to establish reliable EC models for characterizing each individual tumor, distinguish optimal drug treatment, and determine specific drug effect mechanisms for personalized therapy. Methods: Freshly collected tumors were used for patient-derived xenograft models (PDXs) and patient-derived primary cancer cells (PDCs). Immunohistochemistry (IHC) staining was used to confirm the tumors grown in mice faithfully replicate the characteristics of patient tumor tissues. FDA-approved anticancer drugs were used to screen the optimal drugs to inhibit tumor cell proliferation and further utilized to inhibit tumor growth using PDXs. Results: We have collected 26 EC tumors and characterize them by DNA mutation and RNA-seq analysis. We implanted them in NSG mice, and 16 tumor samples have successfully grown in mice with a success rate of 61%. Most PDXs can be faithfully passaged to three generations. From the 16 PDXs, we created six novel PDCs. Using 2D and 3D spheroid cell culture, we screened 133 FDA-approved oncology drugs in the EC-PDC models. The drug screening results clearly highlighted several standout drugs, including CUDC-907 (a dual PI3K/HDAC inhibitor), two histone deacetylase (HDAC) inhibitors including romidepsin and panobinostat, four topoisomerase II (TOP II) inhibitors including mitoxantrone, daunorubicin, doxorubicin, and epirubicin, two proteasome inhibitors including carfilzomib and bortezomib, 2 DNA-directed RNA synthesis inhibitor dactinomycin and plicamycin, omacetaxine mepesuccinate (protein synthesis inhibitor), and valrubicin (DNA synthesis inhibitor). These drugs have demonstrated a significant capacity to inhibit tumor cell proliferation. Our pilot studies using single drugs, CUDC-907, romidepsin and mitoxantrone, achieved moderate drug effects. Combine mitoxantrone with romidepsin or CUDC-907 show surprisingly synergistic effects in multiple PDC models. The combination strategy will be tested in PDXs in the near future. Conclusion: Our unique PDXs and PDCs are excellent models for representing various characteristics of EC and testing novel therapeutics. This current study presents a promising direction for developing personalized therapy options for EC patients and provides a platform for further investigation of drug mechanisms and tumor development. Future studies will also involve etiology, such as chronic psychological stress, DNAm age and intratumoral microbiome. This information will help with endometrial cancer prevention, diagnosis, and prognosis. Citation Format: Tianyue Li, Xiaohao Huang, Riley Rosenmeyer, Samuel Robinson, Kazi Salsabil, Yiqin Xiong, Xiangbing Meng, Shujie Yang. Patient-derived model systems of endometrial cancers for disease modeling and drug sensitivity testing [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 941.