Epigenomics to Enhance Tumor Classification
News Mar 19, 2018 | Original Story from University Hospital Heidelberg
Scientists from the Hopp Children's Cancer Center (KiTZ) and the Neuropathology Department at Heidelberg University Hospital have substantially enhanced the classification of tumors of the central nervous system (CNS). The study, published in Nature, means physicians will now be able to categorize CNS tumors more precisely into specific risk groups and make therapy decisions on this basis.
To be able to treat cancer of the central nervous system (CNS) successfully, it is important to have very precise knowledge about the molecular characteristics of the tumors in order to “give them the right name”. It is presently possible to differentiate about 100 types of CNS tumors based on tissue characteristics. These tumor types show widely varying responses to radiotherapy and chemotherapy. In some cases, methods of molecular diagnostics are used to further classify tumors, for example based on certain gene mutations. Nevertheless, their variability is large, which makes it difficult to standardize diagnostic methods.
To enhance the diagnosis of CNS tumors, a team led by Professor Stefan Pfister, KiTZ director and department head of “Pediatric Neurooncology” at the DKFZ, in collaboration with colleagues from the Neuropathology Department at Heidelberg University Hospital led by Professor Andreas von Deimling, have developed a new computer-based method. “We hope that our new molecular classification method will help improve diagnostic accuracy in CNS tumors and, thus, also improve the chances for successful treatment,” said von Deimling.
The researchers analyzed specific chemical tags in the tumor genomes called DNA methylations. Different cell types exhibit characteristic patterns of DNA methylation which enable scientists to draw conclusions about a tumor’s cellular origin. "We have developed computer-based algorithms that reliably differentiate 82 types of CNS tumors based on their methylation patterns," said Professor David Capper, who is one of the four first authors of the study. Capper is a faculty member of the DKTK partner site in Berlin and has recently accepted a professorship for Molecular Neuropathology at Charité University Medicine Berlin. “Particularly in tumors which we cannot easily assign to a diagnostic category based solely on microscopic examination, methylation analysis is often helpful to make a precise diagnosis. The analysis of approximately 2,800 reference tumor samples additionally made it possible to classify tumors into specific subgroups that are not yet included in the classifications that have been used so far.”
In order to test whether the method is suitable for use in clinical routine diagnostics, the scientists analyzed more than 1,100 additional tumor samples. In about twelve percent of the cases, they were able to correct the initial diagnosis using the methylation patterns. In almost all cases where it was possible, further molecular-diagnostic examinations showed that molecular classification characterized the tumors even better than the initial microscopic diagnosis.
“We are convinced that our new method is well suited to be used in the clinic,” said Pfister. He added: “We have made our classification system available online in order to enable researchers to analyze their data at our platform.” The information that will come in this way will at the same time help achieve more precise diagnoses and, thus, better treatment of rarer cancer types.
The online platform on DNA methylation analysis is available at: www.molecularneuropathology.org.
This article has been republished from materials provided by University Hospital Heidelberg. Note: material may have been edited for length and content. For further information, please contact the cited source.
Reference: Capper, D., Jones, D. T. W., Sill, M., Hovestadt, V., Schrimpf, D., Sturm, D., … Pfister, S. M. (2018). DNA methylation-based classification of central nervous system tumours. Nature. https://doi.org/10.1038/nature26000
Machine Learning Helps Predict Schizophrenia Treatment OutcomesNews
New research that used a machine-learning algorithm to investigate MRI images of schizophrenia patients has successfully identified response to antipsychotic treatment in 4 out of 5 patients.READ MORE