|Klaus Gerwert, Stephanie Schörner and Frederik Großerüschkamp (from left) want to improve the diagnosis of colon cancer with the help of artificial intelligence. |
Photo Credit: © RUB, Marquard
Artificial intelligence and infrared imaging automatically classify tumors and are faster than previous methods.
The immense progress in the area of therapy options over the past few years has significantly improved the chances of recovery for patients with colon cancer. However, these new approaches, such as immunotherapy, require a precise diagnosis so that they can be tailored to the respective person. Researchers at the Center for Protein Diagnostics PRODI at the Ruhr University Bochum use artificial intelligence in combination with infrared imaging to optimally coordinate the therapy of colon cancer with the individual patient. The label-free and automatable method can complement existing pathological analyzes. The team around Prof. Dr. Klaus Gerwert reports in the journal "European Journal of Cancer" on January 28, 2023.
Deep insights into human tissue within an hour
The PRODI team has been developing a new method of digital imaging for several years: The so-called label-free infrared (IR) imaging measures the genomic and proteomic composition of the tissue examined, i.e. provides molecular information based on the infrared spectra. This information is decoded using artificial intelligence and displayed as false color images. For this purpose, the researchers use image analysis methods from the field of deep learning.
The PRODI team, in collaboration with clinical partners, was able to show that the use of deep neural networks makes it possible to reliably determine the so-called microsatellite status, a prognostic and therapeutically relevant parameter, in colon cancer. The tissue sample goes through a standardized, user-independent, automated process and enables a location-based differential classification of the tumor within one hour.
Indication of the effectiveness of therapies
In classic diagnostics, the microsatellite status is determined either by a complex immune coloring of various proteins or by a DNA analysis. "15 to 20 percent of colon cancer patients have instability in the microsatellites in the tumor tissue," said Prof. Dr. Andrea Tannapfel, head of the Institute for Pathology at the Ruhr University. “This instability is a positive biomarker that suggests that immunotherapy will be effective."
With the ever-better therapy options, the quick and uncomplicated determination of such biomarkers is becoming increasingly important. Based on IR microscopic data, neural networks were modified, optimized and trained at the PRODI in order to establish label-free diagnostics. Unlike immunodeficiency, this approach does not require dyes and is significantly faster than DNA analysis. "We were able to show that the accuracy of IR imaging for determining microsatellite status comes close to the most common method in the clinic, immunofoloration," said doctoral student Stephanie Schörner. "By constantly developing and optimizing the method, we expect a further increase in accuracy," adds Dr. Frederik Großerüschkamp.
This project was made possible by a long-term, intensive cooperation between the Institute for Pathology at Ruhr University (Prof. Dr. Andrea Tannapfel), the Clinic for Hematology and Oncology of St. Josef Hospitals, Clinic of the Ruhr University (Prof. Dr. Anke Reinacher-Schick) and the Center for Protein Diagnostics (Prof. Dr. Klaus Gerwert).
PRODI researchers were able to access the ColoPredict Plus 2.0 molecular register for the development of the diagnostic approach, a non-interventional, multicenter registry study for patients with early colon cancer. “The Colopredict Register also enables more targeted therapy for patients through the targeted analysis of biomarkers. The register has recently served as a study platform for precision soncological approaches,” says Anke Reinacher-Schick. In addition to the provision of tissue samples, the register offers a well-founded database of prognostically and therapeutically relevant basic characteristics. "In such a project, it is extremely important to have an excellent cohort and pathological expertise," emphasizes Klaus Gerwert. "Our work on the classification of microsatellite status in colon cancer patients is based on one of the largest cohorts we have published so far and clearly shows the possibility of use in translational cancer research," said Andrea Tannapfel.
Funding: The work was funded by the State of North Rhine-Westphalia, Ministry of Culture and Science, in the Research Center for Protein Diagnostics (PRODI) (funding code: 111.08.03.05-133974). The register study was funded by Roche Pharma AG. Parts of the project were funded by the Slide2Mol project through the Computational Life Science program from the Federal Ministry of Education and Research.
Published in journal: European Journal of Cancer
Source/Credit: Ruhr University Bochum
Reference Number: msd021523_01