Microbiota-based Models Enhance Detection of Colorectal Cancer.

microbiota;microbiome;colorectal cancer;machine learning;Microbiology and Immunology;Health Sciences;Microbiology and Immunology

Baxter, Nielson T.Chen, Grace Y ;

University of Michigan

关键词: microbiota;    microbiome;    colorectal cancer;    machine learning;    Microbiology and Immunology;    Health Sciences;    Microbiology and Immunology;   

Others  :  https://deepblue.lib.umich.edu/bitstream/handle/2027.42/133364/ntbaxter_1.pdf?sequence=1&isAllowed=y

瑞士|英语

来源: The Illinois Digital Environment for Access to Learning and Scholarship

【 摘 要 】

Colorectal cancer (CRC) is the second leading cause of death among cancers in the United States. Although individuals diagnosed early have a greater than 90% chance of survival, more than one-third of individuals do not adhere to screening recommendations partly because the standard diagnostics, colonoscopy and sigmoidoscopy, are expensive and invasive. Thus, there is a great need to improve the sensitivity of non-invasive tests to detect early stage cancers and adenomas. Numerous studies have demonstrated a causal link between the formation of colonic lesions and the activity of the gut microbiota in tissue culture and animal models. These findings have been complemented by studies in human populations identifying shifts in the composition of the gut microbiota associated with the progression of colorectal cancer. These results suggest that the gut microbiota may represent a reservoir of biomarkers that would complement existing non-invasive methods such as the widely used fecal immunochemical test (FIT). Using stool samples from 490 patients we developed a cross-validated random forest classification model that detects colonic lesions using the relative abundance of gut microbiota and the concentration of hemoglobin in stool. The microbiota-based model had significantly higher sensitivity for lesions compared to FIT alone, detecting the majority of lesions that were missed by FIT. Furthermore, we demonstrated that microbial DNA isolated from the residual buffer of FIT cartridges could be used in place of stool samples for microbiota characterization. These findings demonstrate the potential for microbiota analysis to be combined with existing screening methods to improve detection of colonic lesions.