Abstract Metastatic triple-negative breast cancer (TNBC) has a median survival of 13 months, and only 5 months if brain metastases are present. Advances in immunotherapy are promising, but only add weeks to a few months of progression-free survival to this guarded prognosis. Thus, there is a large unmet need for innovations that improve patient outcomes in the metastatic setting when patients are refractory to first-line treatment. Recently, we published an innovative strategy to identify multi-omic molecular correlates of drug response in TNBC. From this work, we now apply the strategy to 3D cell culture models and patient derived xenografts (PDXs) that are more representative of in vivo drug response in real patients. In brief, we have tested the dose response for a panel of 387 investigational drugs and plotted this readout against molecular characterization in 35 TNBC cell lines and PDXs. Using linear regression, we have identified statistically significant positively and negatively correlating DNA, RNA, and protein predictors of drug efficacy. From this expanded multi-omic set, we have identified a multi-omic signature for sensitivity to gemcitabine, a drug widely used in TNBC, but at present, without any a priori knowledge of who is likely to benefit. Moreover, we noted a significant correlation of response with MERIT40, an AKT substrate. MERIT40 is critical for localization of BRCA1 to facilitate DNA damage repair. MERIT40 promotes resolution of DNA damage caused by chemotherapy, and thus, based on our published work and supporting literature, PI3K/AKT inhibition is a rational synergistic partner with gemcitabine. Moreover, further supporting our hypothesis, preliminary results indicate Chou-Talalay synergy at therapeutically relevant doses when gemcitabine is combined with PI3K/AKT inhibition. In future studies, we will evaluate the ability of the gemcitabine biomarker to predict response, and the efficacy of the prioritized combinations using ex vivo patient models of metastatic breast cancer established in our laboratory. Citation Format: Nathan M. Merrill, Nathalie M. Vandecan, Athena M. Apfel, Peter J. Ulintz, Liwei Bao, Xu Cheng, Aki Morikawa, Sofia D. Merajver, Matthew B. Soellner. Optimizing gemcitabine in metastatic TNBC by rational PI3K/AKT combination [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 1187.