Abstract 2689: High-content imaging and single-cell analysis of drug response ex vivo is predictive of clinical outcome for hematologic cancer patients

Abstract

Abstract The ability to predict clinical drug response is a fundamental goal of precision cancer medicine. Genetics has performed well in identifying the presence of targets and stratifying patients into clinical trials. However, particularly for late-stage and relapsed cancer patients, many genetic driven trials have yet to report improvements in clinical response. In parallel, directly measuring cytotoxicity of drugs ex vivo in patient samples with functional technologies has been gaining traction during the past five years, with some programs showing translational success. In our approach, we use high throughput and high content microscopy to quantify single-cell phenotypic biomarkers of cancer cells, and healthy cells, that are indicative of cell death in order to determine a differential drug response. Cancer cells and healthy cells are distinguished using fluorescently-labeled extracellular-marker targeted diagnostic antibodies, and no long term ex vivo culture is necessary as the assay lasts only 24 hours. The cytotoxicity of cancer cells under drug treatment can be scored and compared to any cell death occurring in healthy cells within the same patient sample. As all is assessed by imaging of each individual cell within the same sample and set-up, the healthy cells of the patients can be used as internal controls, statistical analysis covers millions of events and images can be re-mined and re-analysed in future. In a prospective basket trial testing 140 drugs in triplicate at two concentrations (768 tests, yielding 15.000 images per patient), we ranked drugs prospectively for late-stage patients diagnosed with hematological cancers. In this interim analysis, 15 (88%) of 17 patients receiving guided treatment had an overall response compared with four (24%) of 17 patients with their most recent regimen. 12 (71%) of 17 patients had a progression-free survival ratio of 1·3 or higher, and median progression-free survival increased by four times, from 5.7 weeks to 22.6 weeks. While continued follow-up of the study is warranted, this clinical use case foreshadows the success of functional drug screening using single-cell imaging, and further, big data medicine. A unique byproduct of this big data medicine approach, is large pan-indication off-labeling drug-to-patient maps. These maps describe both on- and off-target drug responses, but also indication areas that lack ex vivo drug response - potentially enabling clinical re-use of drugs, and elucidating mechanism of action of drugs where it is unknown. The functional screening data over many patients, indications, and drugs, combined with clinical response data, can eventually be used for larger predictive studies. Citation Format: Gregory I. Vladimer, Berendd Snijder, Nikolaus Krall, Christoph Kornauth, Stefan Kubicek, Ulrich Jäger, Philipp B. Staber, Giulio Superti-Furga. High-content imaging and single-cell analysis of drug response ex vivo is predictive of clinical outcome for hematologic cancer patients [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 2689.