Abstract 3900: A high-throughput platform identifies novel drug combinations towards acute myeloid leukemia therapy

Abstract Molecularly targeted therapies have reshaped oncology in recent decades, though not all patients are eligible for such treatments and resistance mechanisms limit clinical benefit. In response, combination therapies, which can induce potent synthetic lethalities, bypass drug resistance, and expand treatment options, are increasingly being developed. A major hurdle limiting novel combination therapy discovery is the sheer combinatorics associated with broadly testing two or more agents. Thus, there is a practical need for methodologies that prioritize test compounds and technologies for rapid and efficient screening. Towards these ends, we have developed a novel combination drug screening workflow that integrates machine learning-based drug synergy predictions with a high throughput droplet microfluidics-based screening platform. Our physical system, or FlowMatrix, consists of 9, 216 nano wells arrayed in a 96 × 96 grid in which cells are incubated and drug compounds delivered via emulsified droplets along both its rows and columns to facilitate combination screening. Following multi-day incubation, live cells are imaged with fluorescence microscopy and drug combination synergies are calculated from cellular viabilities. In our current setup, which includes four dose drug treatments with replicates and controls, we capture complete data (control, single agent, and 4 × 4 drug-drug combination viability measurements) for up to 100 unique drug combinations within the 9, 216 dual droplet loaded wells of one FlowMatrix. Our to-date screening efforts have focused on acute myeloid leukemia (AML) specifically, where we have profiled 12, 700 AML cell line-drug combinations in our system. This encompasses 3, 700 unique drug combinations profiled in 7 distinct AML cell lines with 183 total matrix runs. Among the strongest hits observed in our screens and confirmed with validation studies (~65% confirmation rate) are combinations currently being tested in clinical trials for other cancers and established therapies paired with non-cancer indicated drugs. In particular, we are interrogating combinations involving CDK12 inhibitors, which affect the transcription of DNA repair proteins, splicing, and cancer cell progression, including unexpected partner drugs that potentiate the effects of CDK12 blockade in AML cell lines. We have uncovered additional unexpected synergistic combinations with KIT inhibitors in RUNX1-RUNX1T1 altered AML lines, ATR inhibitors (including an established combination with Gemcitabine), and the BCL-2 inhibitor Venetoclax, including known partnerships with chemotherapies (e. g. Decitabine and Daunorubicin) and other targeted inhibitors. Deeper studies are underway to further validate and characterize these candidates towards expanding treatment options for AML. Ultimately, our platform may radically enhance drug combination screening potential. Citation Format: Anthony R. Soltis, Boryana Zhelyazkova, Pascal Drane, Efstathios Eleftheriadis, Andrew Ventresco, David A. Weitz, Arlinda Lee, Anthony J. Iafrate. A high-throughput platform identifies novel drug combinations towards acute myeloid leukemia therapy 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 (6Suppl): Abstract nr 3900.