Developing machine learning algorithms for early detection of hepatocellular carcinoma
Prof Sara Zanivan, Prof Crispin Miller
Follow LINK to apply
APPLY HERE. Application closing date: 9 January 2022.
Primary liver cancer is the third leading cause of cancer-related death worldwide. The most common type of liver cancer is hepatocellular carcinoma (HCC), which develops in the background of liver diseases, such as cirrhosis. HCC incidence is rising faster than other cancers worldwide. Given the only cure for HCC is surgery, which can only be used in patients with early-stage disease, there is urgent need to develop new methods for early HCC detection to reduce mortality by increasing the number of patients who can access curative interventions. Current HCC surveillance methodologies in routine clinical practice lack sensitivity and are of doubtful cost-effectiveness, posing a substantial problem for healthcare systems.
Recent breakthroughs in mass spectrometry (MS) technology have dramatically advanced the proteomic field, such that MS proteomics is considered the next massive investment opportunity in biomedical research (https://www.forbes.com/sites/stephenmcbride1/2021/06/23/proteomics-the-next-truly-massive-investing-opportunity/). Liver diseases are particularly well suited for plasma biomarker studies, because the liver is a highly secretory organ. Consequently, the majority of blood-secreted proteins comes from the liver, of which some are routinely assessed in the clinics as biomarkers. Single blood biomarkers used in the clinics to detect HCC lack sensitivity and specificity. However, recent mass spectrometry (MS)-based plasma proteomic studies have shown that subsets of plasma proteins can be used for predictive models for diagnosis, staging and to predict progression of liver disease and performed as well as existing diagnostic strategies or better.
This project aims at developing a proteomic classifier for risk of developing HCC for patients with liver diseases. To achieve this, the student will analyse proteomic data that will be generated at the Beatson using novel MS technology and will develop machine learning approaches for the analysis of proteomic data of plasma samples obtained from large cohorts of patients at risk of developing HCC and with early-stage HCC. This project is part of an interdisciplinary project that has recently been funded by CRUK Early Detection and Diagnosis Research Committee, which involves the Beatson Institute (Zanivan, Bird and Miller groups), the University of Oxford and University of Nottingham.
Keywords: proteomics; mass spectrometry; liver cancer; early detection; plasma biomarkers; machine learning
Emily J Kay, Karla Paterson, David Sumpton, Ekaterina Stepanova, Claudia Boldrini, Juan R Hernandez-Fernaud, Sandeep Dhayade, Enio Gjerga, Robin Shaw, Lisa J Neilson, Grigorios Koulouras, Grace McGregor, Sergio Lilla, Craig Jamieson, Ann Hedley, Radia Marie Johnson, Morag Park, Crispin Miller, Jurre J Kamphorst, Fabricio Loayza-Puch, Julio Saez-Rodriguez, Karen Blyth, Michele Zagnoni, Sara Zanivan. Metabolic control of tumour extracellular matrix production in cancer-associated fibroblasts. BioRxiv https://doi.org/10.1101/2020.05.30.125237
Kugeratski FG, Hodge K, Lilla S, McAndrews KM, Zhou X, Hwang RF, Zanivan S, Kalluri R. Quantitative Proteomics Identifies the Core Proteome of Exosomes with Syntenin-1 as the highest abundant protein and a Putative Universal Biomarker. Nat Cell Biol. 2021 Jun;23(6):631-641.
Kugeratski FG, Atkinson SJ, Neilson LJ, Lilla S, Knight JRP, Serneels J, Juin A, Ismail S, Bryant DM, Markert EK, Machesky LM, Mazzone M, Sansom OJ, and Zanivan S. Hypoxic cancer–associated fibroblasts increase NCBP2-AS2/HIAR to promote endothelial sprouting through enhanced VEGF signaling. Science Signaling 2019 12(567), eaan8247. Cover Story. Editor's choice in Science Signaling 12(569), eaax0155.
van der Reest J , Lilla S, Zheng L*, Zanivan S* and Gottlieb E*. Proteome-wide analysis of cysteine oxidation reveals metabolic sensitivity to redox stress. Nat Commun. 2018 9(1):1581. *Corresponding author.
Reid SE, Kay EJ, Neilson LJ, Henze AT, Serneels J, McGhee EJ, Dhayade S, Nixon C, Mackey JB, Santi A, Swaminathan K, Athineos D, Papalazarou V, Patella F, Román-Fernández Á, ElMaghloob Y, Hernandez-Fernaud JR, Adams RH, Ismail S, Bryant DM, Salmeron-Sanchez M, Machesky LM, Carlin LM, Blyth K, Mazzone M, Zanivan S. Tumor matrix stiffness promotes metastatic cancer cell interaction with the endothelium. EMBO J. 2017 36(16):2373-2389.
For informal enquiries or further details on the project, please email Prof Sara Zanivan (firstname.lastname@example.org).