Exscientia’s precision medicine AI platform is being further assessed by Blood Cancer Discovery – Journal to enhance patient outcomes


Exscientia, ETH Zurich, Medical University of Vienna, and the Center for Molecular Medicine (CeMM) of the Austrian Academy of Sciences today announced a new publication from Prof. Berend Snijder’s lab entitled “Deep Morphology Learning Enhances Precision Medicine by Image-Based Ex Vivo Drug Testing” in Blood Cancer Discovery, a journal of the American Association for Cancer Research. The post-hoc analysis expands on the groundbreaking work of the EXALT-1 study, which used deep learning algorithms to categorise complicated cell morphologies in tissue samples from cancer patients into disease “morphotypes.” This study was published in Cancer Discovery.

Using an AI-powered precision medicine platform to guide individualised therapy recommendations—instead of the doctor’s recommended course of treatment—EXALT-1 was the first prospective research to demonstrate significantly superior outcomes for patients with late-stage hematologic malignancy. For at least three times longer than anticipated for their condition, 40% of patients in EXALT-1 experienced an outstanding response. The post-hoc analysis, which was published today in Blood Cancer Discovery, demonstrates the potential for even better patient outcomes when the EXALT-1 technology is combined with recent developments in deep learning that take advantage of cell-specific properties in high-resolution images.

“Following the results of the EXALT-1 study, these results further validate our AI-powered precision medicine platform’s ability to identify highly actionable clinical treatment recommendations for blood cancers. This deepens our insights and increases the clinical predictive power of the platform to help patients”, “Cell morphology, i.e. the assessment of the properties of cells, is of fundamental importance for the diagnosis of cancer. As part of this research, we were able to use deep learning within the platform to improve our ability to identify personalized cancer therapies, leading to better clinical outcomes for patients.

Dr. Gregory Vladimer, VP Translational Research at Exscientia and co-inventor of the platform technolog


“We believe that performing drug screening directly in tumor tissues of cancer patients is a major advance in understanding the complexity of tumors compared to traditional cell model systems. The fact that we can now use the power of deep learning to convert these terabytes of images into actionable insights is very exciting indeed”

Prof. Berend Snijder, Principal Investigator at the Institute of Molecular Systems Biology at ETH Zurich

In a post hoc analysis of 66 patients over a three-year period, 1.3 billion patient cells from 136 ex vivo-assessed tested agents for haematological diagnoses, including acute myeloid leukaemia, T-cell lymphoma, diffuse large B-cell lymphoma, chronic lymphocytic leukaemia, and multiple myeloma, were combined. The impact of deep learning on the clinical predictive power of ex vivo drug screening was assessed Patients who followed the recommended course of treatment based on the platform’s immunofluorescence analysis or deep learning of cell morphologies showed a higher rate of exceptional clinical response, which was measured as a progression-free survival time that lasted three times longer than anticipated for the patient’s particular disease. When the drug’s toxicity to the normal cells in the patient sample analysed was also taken into account, post-hoc analysis proved that the clinical predictions were more accurate.

The precision medicine platform from Exscientia extracts valuable single-cell data from high-resolution photographs of patient tissue samples using specialised deep learning and computer vision techniques. This research offers therapeutically useful insights on the therapies that are most advantageous for a certain patient. Exscientia can better understand which individuals would profit from comparable therapies by examining the outcomes of specific patients using its genomes and transcriptomics capabilities. Dr. Gregory Vladimer and Prof. Berend Snijder created the underlying technology while working in Giulio Superti-group Furga’s at the CeMM Research Center for Molecular Medicine in Austria.

Next Post

SCW Health & Fitness Business Summit

“Further your knowledge and climb higher in your profession with SCW Health & Fitness Business Summit” If you want to grow your business and build your career, the 3-DAY SCW Health & Fitness Business Summit is a must-attend event for you and your team. 30 Sep – 02 Oct 2022 […]