Prof. Michal Holčapek, University of Pardubice; Dr Eva Cífková, University of Pardubice; Dr Miroslav Lísa, University of Pardubice; Dr Denise Wolrab, University of Pardubice; Dr Robert Jirásko, University of Pardubice
University of Pardubice
Pancreatic cancer belongs among the most lethal tumors with the lowest survival rate of all cancers. It is expected to become the second leading cause of cancer-related death in the US as well as Europe by the year 2020. Pancreatic cancer is very hard to be diagnosed at early stages. Screening tests or examinations may be used to look for a disease in people who have no symptoms, and no genetic predispositions. However, neither the imaging tests (CT, MRI, etc.) nor the invasive methods (percutaneous, endoscopic, or surgical biopsy) are applicable for large-scale population screening and demonstrate sensitivity limitations in the detection of tumors at an early stage. There is an urgent need for an effective non-invasive method with sufficiently high sensitivity and specificity.
The mass spectrometry (MS) based analysis of the specific metabolites dysregulation of non-invasively collected samples of pancreatic cancer patients and healthy volunteers enables to build up the statistical models, which may be used to determine the level of probability of the patient suffering from pancreatic cancer. The whole methodology is based on accurate performance of multiple steps, including mainly the sample collection, storage, transport, and processing, followed by analytical qualitative and quantitative analysis using one of three mass spectrometry based methods, followed by multivariate data analysis (MDA) of obtained absolute quantitative data. The sensitivity and specificity of this determination is very high, typically 95-100%. The method can be used for population screening of the whole population or selected population groups based on risk factors such as age, gender, body-mass-index, genetic predispositions, risk behavior, etc.
The methodology is available for licensing.
The final methodology provides high sensitivity, specificity and sample throughput for pancreatic cancer classification, which is 100% for samples with known classification (292 subjects) and ca. 95% for samples with unknown classification (73 subjects). The sample throughput of our methodology is at least 4000 samples per year and one MS system with possible improvements using automation and multiplexing. The methodology is fully validated in line with recommendations of authoritative organizations, such Food and Drug Administration or European Medicines Agency (EMEA).
EP-application was filed in January 2018.
Some of the previous publications of the group on the related topic:
E. Cífková, M. Lísa, R. Hrstka, D. Vrána, J. Gatěk, B. Melichar, M. Holčapek, Rapid Commun. Mass Spectrom. 31 (2017) 253-263.
E. Cífková, M. Holčapek, M. Lísa, D. Vrána, J. Gatěk, B. Melichar, Anal. Bioanal. Chem. 407 (2015) 991-1002.