Introduction: The intratumoral heterogeneity of Acute Myeloid Leukemia (AML) and other hematologic malignancies presents a challenge in developing effective single-agent targeted treatments. Furthermore, the emergence of genetically heterogeneous subclones leading to relapse suggests that effective therapies associated with discrete genotypes may require drug combinations, each of which modulates distinct pathways. In addition, microenvironmental rescue signals as well as tumor-intrinsic feedback pathways in AML and other hematologic malignancy subsets will necessitate combinatorial therapy approaches. Towards the goal of identifying new therapeutic combinations for AML and other hematologic malignancies, we assessed the sensitivity of primary patient samples to various drug combinations using an ex vivo functional platform. Methods: We have previously screened over 1000 primary patient specimens against a panel of single-agent small-molecule inhibitors. Using these historical drug sensitivity data, we ranked drugs by their IC50, and used these rankings to assemble an initial panel (1) of 44 drug combinations consisting primarily of kinase inhibitors with non-overlapping pathways. Primary patient samples (n = 74) with various hematologic malignancies were assessed for sensitivities to these combinations by culturing cells in the presence of fixed molar concentrations of the drugs over a dose series. Sensitivity was assessed by a viability assay on day 3 using a tetrazolium reagent. IC50 values for samples sensitive to a combination were sorted according to disease type and compared to those for each single agent to derive an index of effectiveness. Based on data from panel 1, we generated a second panel (2) consisting of 44 drug combinations, including new combinations of kinase inhibitors as well as combinations of drugs from different classes, such as bromodomain inhibitors, BH3 mimetics, proteasome inhibitors, IDH1/2 inhibitors coupled with kinase inhibitors. Primary patient samples (n = 78) were assessed for sensitivities to these combinations. Results: The performance of drug combinations across AML, ALL, CLL, CML or other MDS/MPN specimens are displayed in a heat map (Figure 1) representing the sensitivities of each drug combination relative to either of the single agents comprising that combination (the combination IC50 divided by the lowest single agent IC50 is our combination ratio). For each combination, we then compared the combination ratio of each individual specimen to the median combination ratio across all specimens tested, and cases with a combination ratio value less than 20% of the median were considered hypersensitive to that combination. We calculated the percentage of cases that were sensitive to each combination within the diagnostic subsets of AML, ALL, CLL, CML, and MDS/MPN and subsets with the most frequent sensitivity to a drug combination are indicated on the heat map (80%, white). Combinations of two kinase inhibitors that included the p38MAPK inhibitor, doramapimod, were generally more effective on AML and CLL samples than other diagnostic subsets (panel 1). For CLL sample, combinations including midostaurin and either alisertib, ruxolitinib or sorafenib were particularly effective. Among combinations on panel 2, doramapimod coupled with an apoptosis inducer (ABT-199) exhibited broad efficacy on AML samples. In addition, combinations with the bromodomain inhibitor, JQ1, or the BH3 mimetic, ABT-199, were more broadly effective across diagnostic subsets than many of the kinase-kinase pairs tested. To validate the apparent synergies observed with patient samples, we tested selected combinations on AML-derived cell lines and observed synergies, which were supported with combination indices derived by the Chou-Talalay method. Conclusions: These data suggest that specific drug combinations formed either with two kinase inhibitors or with two compounds from different drug classes are effective in a patient-specific manner with enrichment for certain drug pairs within specific diagnostic subsets. While a secondary evaluation is necessary to validate the initial observation of sensitivity, linking this methodology with genetic attributes for patient samples will identify effective combinations of targeted agents and add therapeutic options for AML treatment. Figure 1. Figure 1. Disclosures Pandya: Microsoft: Employment, Equity Ownership. Bolosky:Microsoft: Employment, Equity Ownership. Druker:Oregon Health & Science University: Patents & Royalties; Henry Stewart Talks: Patents & Royalties; CTI Biosciences: Consultancy, Equity Ownership, Membership on an entity's Board of Directors or advisory committees; Novartis Pharmaceuticals: Research Funding; Aptose Therapeutics, Inc (formerly Lorus): Consultancy, Equity Ownership, Membership on an entity's Board of Directors or advisory committees; McGraw Hill: Patents & Royalties; Leukemia & Lymphoma Society: Membership on an entity's Board of Directors or advisory committees, Research Funding; MolecularMD: Consultancy, Equity Ownership, Membership on an entity's Board of Directors or advisory committees; Roche TCRC, Inc.: Consultancy, Membership on an entity's Board of Directors or advisory committees; Blueprint Medicines: Consultancy, Equity Ownership, Membership on an entity's Board of Directors or advisory committees; Bristol-Myers Squibb: Research Funding; Millipore: Patents & Royalties; AstraZeneca: Consultancy; Oncotide Pharmaceuticals: Research Funding; Cylene Pharmaceuticals: Consultancy, Equity Ownership, Membership on an entity's Board of Directors or advisory committees; Fred Hutchinson Cancer Research Center: Research Funding; ARIAD: Research Funding; Gilead Sciences: Consultancy, Membership on an entity's Board of Directors or advisory committees; Sage Bionetworks: Research Funding. Tyner:Incyte: Research Funding; Janssen Pharmaceuticals: Research Funding; Constellation Pharmaceuticals: Research Funding; Array Biopharma: Research Funding; Aptose Biosciences: Research Funding.