Today's Clinical Lab - News, Editorial and Products for the Clinical Laboratory
Gloved hands point to a vector illustration of human intestines.
AI predictions can speed up the analysis of tissue samples, resulting in faster treatment decisions for colorectal cancer.

AI-Based Biomarker Prediction Accelerates CRC Prognosis

New transformer network-based approach substantially improves the data efficiency and interpretability of colorectal cancer biopsies

Helmholtz Zentrum München
Published:Aug 31, 2023
|2 min read
Register for free to listen to this article
Listen with Speechify

Colorectal cancer (CRC) ranks second in leading causes of cancer-related deaths globally, according to the WHO. For the first time, researchers from Helmholtz Munich and the University of Technology Dresden (TU Dresden) show that artificial intelligence (AI)-based predictions can deliver comparable results to clinical tests on biopsies of patients with CRC. 

AI predictions can speed up the analysis of tissue samples, resulting in faster treatment decisions. This novel model for biomarker detection represents a significant stride toward the realization of precision therapy approaches in the field of oncology. The method is published in Cancer Cell.

A team of scientists around Tingying Peng, PhD, from Helmholtz Munich and Jakob N. Kather, MD, MSc, from TU Dresden, shows that AI can predict specific biomarkers in stained tissue samples of patients with CRC. They used “transformer networks,” a recent deep learning (DL) approach, to identify patterns and support diagnostic decisions in cancer management. The new method significantly improves previous approaches for biomarker detection.

Large-scale evaluation proves better generalization, data efficiency

The team of researchers developed software that uses the new technology of transformer neural networks throughout the analysis process. They show that their approach substantially improves the performance, generalizability, data efficiency, and interpretability by evaluating it on a large multicentric cohort of over 13,000 patients from 16 cohorts from seven countries, part of which was contributed by researchers at the German Cancer Research Center (DKFZ) Heidelberg and the network of the National Centers for Tumor Diseases (NCT).

The algorithm achieves a very high sensitivity on resection tissue samples obtained during surgery. Strikingly, even though the model has only been trained on tissue samples from resections, the results can also reach a high performance on biopsy tissue obtained during colonoscopy. Sophia J. Wagner, a PhD candidate and first author of the study, emphasizes that “the generalization to biopsy tissue increases the algorithm’s benefit for the patient when ultimately implemented in clinical routine.”

AI-based prescreening for biopsies accelerates diagnosis

Because of its high sensitivity to biopsy tissue, the algorithm could serve as a prescreening tool followed by affirmative testing for cases that received a positive result during AI testing. Applying AI-based biomarker prediction could reduce the testing burden and, therefore, speed up the step between taking the biopsy and the molecular determination of the genetic risk status, thus enabling an earlier patient treatment with immunotherapy if indicated.

- This press release was originally published on the Helmholtz Munich website