AI Precision Medicine platform makes EV history for Cambridge-based Mursla Bio
Mursla Bio has announced the launch of its AI Precision Medicine platform, which enables non-invasive access to organ-specific molecular information from blood with high spatial and biological resolution.
The platform advances extracellular vesicle (EV) science as part of the Cambridge Science Park company’s mission to progress precision diagnostics and significantly improve cancer outcomes for at-risk patients.
Extracellular vesicle (EV) science explores the role of nano-sized vesicles released by cells in intercellular communication and disease. These vesicles, containing a variety of molecules like proteins and nucleic acids, are messengers, transferring cargo between cells and influencing recipient cellular behavior.
Mursla’s AI Precision Medicine is built on the ability to isolate and analyze EVs which are secreted by specific organs into the bloodstream. The platform reduces non-target background by up to five orders of magnitude compared to standard bulk EV isolation and significantly improves data structure, enhancing prediction model generalisation.
The technology developed by Mursla Bio incorporates optimised multi-omics workflows, embedded AI pipelines, and IVD translation processes which convert complex biological signals into practical, regulatory-grade assays. These assays are deployable on standard commercial instruments using patented, PCR (polymerase chain reaction) and ELISA-like (enzyme-linked immunosorbent assay) steps designed for clinical adoption.
The platform also underpins the development of EvoLiver, the company’s lead clinical program for liver cancer surveillance in cirrhotic patients. Built on multi-omics data from more than 300 patients, EvoLiver achieved 86 per cent sensitivity and 88 per cent specificity on a locked, IVD-compatible assay for early-stage hepatocellular carcinoma detection, and has received Breakthrough Device Designation from the FDA.
Mursla Bio, which was founded in 2017, is an emergent techbio company with a unique approach to delivering the next generation of liquid biopsies. Its model is underpinned by an understanding that the 21st century will be marked by the convergence of artificial intelligence, fuelled by Moore’s Law (‘Mursla’ pronounced in Japanese), and the increasing ability to decode the complex language of biology (‘Bio’).
The pre-print of Mursla Bio’s latest study presents the first multi-layered validation of hepatocyte EV isolation in humans, confirming liver origin through proteomic and nucleic acid markers. Key supporting proteomics data was generated in collaboration with Evotec International GmbH, with patient samples provided by University College London.
The research paper, co-authored by Professor Brian Davidson, a global leader in liver disease surgery and clinical research, shows that Mursla Bio’s method yields robust, organ-specific multi-omics data using minimal blood volume, offering a novel scalable approach for biomarker and AI development.
Pierre Arsène, founder and CEO of Mursla Bio, said: “EVs are nature’s native signal enrichment system for long-distance communication between cells and organs. Our platform harnesses this biology to non-invasively access organ-specific, multi-omics data from just a small blood sample. It produces structured, biologically labeled datasets that are optimized for AI and ready to support diagnostic translation.
“EvoLiver is our first proof point of real-world clinical impact, and we are now expanding access to the platform to advance our pipeline and enable partnerships across other disease areas.”
