TU Delft researchers have developed a tool that offers a simple means of comparing cancer patients in order to gain increased insight into various forms of cancer and help develop tailor-made treatments. The researchers have published their findings in the online open-access journal scientific-reports/” title=”View all articles about Scientific Reports here”>Scientific Reports.
Cancer Genome Atlas
Cancer is a complex disease that can be caused when various elements are disordered, such as DNA mutation. A detailed profile of cancer patients will help medical professionals offer tailor-made treatment. The Cancer Genome Atlas is an American initiative that is assisting in this process by conducting various molecular measurements on a large number of cancer patients, in order to gain better insight into the various forms of cancer. Headed by Dr Erdogan Taskesen, TU Delft researchers have now developed a tool that offers a simple means of comparing patients. This will allow the researchers to gain improved insight into the various forms of cancer, each of which progresses in a different way.
19 types of cancer
The TU Delft researchers applied their analyses to 4,434 patients and 19 types of cancer, examining four molecular aspects. Their ‘trick’ was to intelligently simplify huge amounts of data so that patients can be displayed as dots on a screen. The distance between the patients is based on four different types of molecular information. Colouring the dots based on information about the patients – such as progression of illness and gender – enables cancer specialists to conduct more detailed research into the groups. Specific molecular information can also be removed in order to examine different groupings. This provides the specialists with insight into the molecular mechanisms that affect specific groups of patients. The researchers have named the tool MEREDITH and made it available on the website: http://pancancer-map.ewi.tudelft.nl/
Leukaemia and breast cancer
The researchers used MEREDITH to identify a new group of leukaemia patients, which can now be researched in more detail by cancer specialists. In the case of breast cancer, they have demonstrated how the different molecular information contributes to the classification of known sub-groups. Another intriguing finding is that some patients proved to have an extremely divergent molecular profile and a remarkably poor prognosis. With this tool, the TU Delft researchers offer cancer specialists a powerful resource with which to take the next steps towards introducing more tailor-made solutions into treatment for cancer patients.
This research was led by Prof. Marcel Reinders from The Delft Bioinformatics Lab.
‘Pan-cancer subtyping in a 2D-map shows substructures that are driven by specific combinations of molecular characteristics’ as published in scientific-reports/” title=”View all articles about Scientific Reports here”>Scientific Reports.
The TU Delft Bioinformatics Department website.