Abstract Pancreatic cancer has the worst prognosis of all cancers (5-year survival rate of 7%). There is a critical need for a tool to match individual patients with optimal drugs for their cancer. The goal of this work is to validate Optical Metabolic Imaging (OMI) of patient-derived pancreatic tumor organoids as a high-throughput predictive drug screen for patients. Drug induced changes in cell metabolism precede changes in tumor volume and thus present an earlier marker of treatment response. OMI is sensitive to cellular metabolism because it measures the intrinsic fluorescent properties of NAD(P)H and FAD, coenzymes of metabolic reactions. OMI endpoints include the optical redox ratio (the fluorescence intensity of NAD(P)H divided by the fluorescence intensity of FAD), and the fluorescence lifetimes of NAD(P)H and FAD. The redox ratio reflects the cellular redox balance, and the fluorescence lifetimes report on the binding activity of these coenzymes. OMI, performed using high-resolution multiphoton microscopy, has the unique ability to non-invasively monitor metabolism in living, intact 3D samples on the single-cell level, and can thus quantify inter-cellular heterogeneity in drug response. OMI also allows for high-throughput screening of potential cancer drugs and drug combinations on patient biopsy samples cultured ex vivo. These patient-specific organoids are grown in a 3D matrix that mimics the natural tumor environment. Drug response can be quantified separately in tumor cells and tumor-associated fibroblasts. Organoids were successfully generated from 9 of 11 surgically resected pancreatic tumors and 2 of 2 pancreatic intraepithelial neoplasias. These organoids were treated with a panel of standard and experimental therapies, and early metabolic changes were measured using OMI. Changes at the single-cell level were quantified using the OMI Index, a linear combination of the optical redox ratio and the mean NAD(P)H and FAD fluorescence lifetimes. Patient survival data after surgery and drug treatment was compared to drug response in organoids to that same treatment. Early results have focused on the standard gemcitabine + 5-fluorouracil (G+F) adjuvant treatment. Organoids generated from a patient whose CA19-9 biomarker levels increased while on G+F adjuvant therapy contained a subpopulation of 68% of the cells which were resistant to G+F therapy. Moreover, no G+F resistant cell subpopulations were found in organoids generated from a patient with no signs of disease recurrence over 14 months of G+F adjuvant therapy. Follow-up with the remaining patients is on-going. The ability of OMI to capture heterogeneous cell-level drug responses shows promise for predicting treatment response of individual pancreatic cancer patients. Citation Format: Joe T. Sharick, Tongcheng Qian, Jillian K. Johnson, Lingjun Li, Cheri A. Pasch, Dustin A. Deming, Alexander A. Parikh, Melissa C. Skala. Personalized medicine in pancreatic cancer using cellular-resolution optical metabolic imaging [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 4100.