Knowledge Connector

© dkfz.de

We know today that the biological characteristics do not only differ between cancers, but also significantly between individual patients. This insight can be transformed into a benefit for the patients. Personalized oncology pursues this goal: Each cancer patient should be offered an individualized treatment based on a comprehensive molecular, cellular and functional analysis of his tumor. For this purpose, the MASTER (Molecular Aided Stratification for Tumor Eradication Research) program was launched in 2013, which investigates the value of a comprehensive characterization of individual tumors in young adults with advanced cancer, as well as patients with rare tumors. This includes whole-genome and RNA sequencing as well as analyzes of DNA methylation. A molecular tumor board evaluates the collected data and identifies a patient-tailored therapy option.

"This process requires a great deal of time and, unfortunately, is not yet available for all cancer patients," says Alexander Knurr. He and his colleagues are developing an analysis tool that addresses this issue. Katrin Glocker, a member of Knurr's team, continues, "We want to help physicians by automating the entire process, from preparation, through the execution of the molecular tumor board, to creating the tumor board report. With our tool, as many patients as possible should have access to "Precision Oncology". The presented tool is the so-called Knowledge Connector.

"The Knowledge Connector links different external databases with each other and evaluates the collected information with regard to the individual characteristics, i.e. the molecular findings as well as medical reports of a patient," explains Glocker, who is responsible for the scientific components of the project. These include gene variants, drugs, clinical trials and publications. From this information, the tool then creates a ranking of variants whose targeted treatment suggests a therapeutic success.

The team around Knurr has recently completed the conceptual phase and is now starting a test run together with physicians at the Heidelberg University Hospital. "We have the distinct advantage of being able to test our work in practice after each development step. Thanks to the close cooperation with the Heidelberg University Hospital, the NCT Heidelberg and the NCT Dresden, we have leading centers at our disposal that evaluate our tools in routine operation and provide us with the necessary clinical feedback," reports Knurr.

In addition to the standard therapies, cancer patients can participate in individual treatment methods, such as early drug trials. "That's exactly the goal of translational research: Studies as well as active ingredients, which have arisen from basic research, should be conducted out of the laboratory. This can create added value for the patient," Knurr emphasizes.

The origin or idea of ​​the Knowledge Connector originated a few years ago. "In larger medical institutions or research centers, clinical and scientific data are documented for various reasons in different systems, which are usually not optimized to exchange these data as needed or to analyze them in an integrated way," explains Prof. Ückert, the head of the department. The various medical data, which occur every day, are a valuable asset for physicians and researchers, which, however, is difficult to access without appropriate tools. The employees of the department have addressed this problem in cooperation with a well-known company in the region. Together they established a data warehouse that contains the clinical and scientific data of the oncological patients of the National Center for Tumor Diseases (NCT) and research data of the German Cancer Research Center (DKFZ). "The IT system, which we call the Electronic Science Record (ESR), integrates data from different sources to the same patient," says Knurr. The Knowledge Connector is taking advantage of this preliminary work: "We combine the information gathered in the ESR with information from publications and from a variety of third-party databases," reports Glocker. Sharing this information is a critical step towards personalized oncology.

 

© dkfz.de

to top