Defining colon cancer biomarkers by using deep learning

Refined approaches are needed to better risk-stratify patients with colorectal cancer for prognosis. No predictive biomarkers of treatment efficacy have yet been identified in patients with non-metastatic disease. Ole-Johan Skrede and colleagues 1 Skrede O-J De Raedt S Kleppe A et al. Deep learning for prediction of colorectal cancer outcome: a discovery and validation study. Lancet. 2020; 395: 350-360 Summary Full Text Full Text PDF PubMed Scopus (191) Google Scholar in The Lancet report on a computer-generated biomarker, the DoMore-v1-colorectal cancer (DoMore-v1-CRC) classifier, which was derived from conventionally stained histopathological images by using deep learning methods. This study adds value to the application of deep learning methods in cancer research as it stimulates a discussion on the potential use of automated methods to generate new information from existing pathological data. Deep learning for prediction of colorectal cancer outcome: a discovery and validation studyA clinically useful prognostic marker was developed using deep learning allied to digital scanning of conventional haematoxylin and eosin stained tumour tissue sections. The assay has been extensively evaluated in large, independent patient populations, correlates with and outperforms established molecular and morphological prognostic markers, and gives consistent results across tumour and nodal stage. The biomarker stratified stage II and III patients into sufficiently distinct prognostic groups that potentially could be used to guide selection of adjuvant treatment by avoiding therapy in very low risk groups and identifying patients who would benefit from more intensive treatment regimes. Full-Text PDF