Leveraging Fecal Microbial Markers to Improve the Diagnostic Accuracy of the Fecal Immunochemical Test for Advanced Colorectal Adenoma

Abstract

Background: Fecal immunochemical tests (FIT) detect colorectal adenoma inefficiently. Alterations in the gut microbiota were associated with the colorectal cancer (CRC) development. We aimed to explore fecal microbial signatures for advanced colorectal adenomas and evaluate their individual diagnostic value and complementary capacity to FIT. Methods: We recruited 1,546 subjects from a CRC screening program, including 268 advanced adenomas, 490 non-advanced adenomas, and 788 healthy subjects. We used a 16S rRNA sequencing based microbiome approach to profile the global microbial composition in fecal samples. Three approaches were applied to feature selections. Receiver operating characteristic (ROC) curves were used to evaluate diagnostic values of microbial signatures and their auxiliary role to FIT and risk-stratified Asia-Pacific Colorectal Screening (APCS) score. We applied ·632+ bootstrapping to adjust the potential overfitting of the prediction models. Findings: We identified 13 microbial signatures showing their joint diagnostic value for advanced adenoma, with genus Tyzzerella 4 demonstrating the highest adjusted area under the curve (AUC) of 0 · 545 (95% CI, 0·520-0·610). A 13-marker classifier was constructed and increased the adjusted AUC to 0·607 (95% CI, 0·548-0·660). Compared with FIT solely, the combination of 13-bacteria panel and FIT reached a promising adjusted AUC of 0·641 (95% CI, 0·579-0·691). At cut-off values yielding specificities of 90% and 80%, the adjusted sensitivities were 28·4% (95% CI, 19·3%-36·8%) and 41.1% (95% CI, 29·9%-49·4%), respectively . APCS score further boosted the adjusted AUC to 0·706 (95% CI, 0·648-0·750). Interpretation: The identified fecal microbial signature could complement the diagnostic capacity of FIT for detecting advanced adenomas. Funding Statement: This work was supported by the CAMS Innovation Fund for Medical Sciences (2017-I2M-1-006, 2019-I2M-2-002), the National Natural Science Foundation of China (81703309). Beijing Nova Program of Science and Technology (Z191100001119065), and Natural Science Foundation of Beijing Municipality (7202169). Declaration of Interests: The authors declare no potential conflicts of interest. Ethics Approval Statement: This study was approved by the Ethics Committee of the National Cancer Center/Cancer Hospital, Chinese Academy of Medical Sciences, and Peking Union Medical College (18-013/1615).