Applying simple linear combination, multiple logistic and factor analysis methods for candidate fecal bacteria as novel biomarkers for early detection of adenomatous polyps and colon cancer

• Published in: Journal of microbiological methods 2018-12, Vol.155, p.82-88
• Main Authors: Rezasoltani, Sama, Sharafkhah, Maryam, Asadzadeh Aghdaei, Hamid, Nazemalhosseini Mojarad, Ehsan, Dabiri, Hossein, Akhavan Sepahi, Abbas, Modarressi, Mohammad Hossein, Feizabadi, Mohammad Mehdi, Zali, Mohammad Reza
• Format: Article
• Language: English
• Subjects: Adenomatous polyps; Adenomatous Polyps - diagnosis; Adenomatous Polyps - microbiology; Adult; Aged; Bacteria - classification; Bacteria - isolation & purification; Biomarkers; Case-Control Studies; Colonic Neoplasms; Colonoscopy; Colorectal cancer; Colorectal Neoplasms - diagnosis; Colorectal Neoplasms - microbiology; Early Detection of Cancer - methods; Factor Analysis, Statistical; Fecal bacteria; Feces - microbiology; Female; Gastrointestinal Microbiome; Humans; Iran; Male; Mass Screening; Middle Aged; Real-Time Polymerase Chain Reaction; ROC Curve; Screening test; Sensitivity and Specificity
• ISSN: 0167-7012; 1872-8359
• DOI: 10.1016/j.mimet.2018.11.007

Colorectal cancer (CRC) is the third leading cause of cancer, and presents a considerable disease burden, worldwide. Recently, the gut microbiota has been proposed as a potential risk factor for CRC, and even adenomatous polyps (AP). Here, the aim of this study was to investigate the role of selected gut bacteria as fecal bacterial biomarkers, in early detection of CRC and AP. Fecal samples (n = 93) were collected from Taleghani Hospital, Tehran, Iran, between 2015 and 2017, from normal controls (NC), AP cases and CRC stage I patients, who were undergoing screening for colonoscopy. Absolute quantitative real time PCR (qPCR) assays were established for the quantification of bacterial marker candidates, in all cases and control groups. In order to evaluate the diagnostic value of bacterial candidates in distinguishing CRC from a polyp, receiver operating characteristic curve (ROC) was performed. Multiple logistic regressions were used to find the best combinations of the bacterial candidates, then, combinations were analyzed based on three methods, including linear combination, multiple logistic and factor analysis models. According to the logistic model, combination of Fusobacterium nucleatum, Enterococcus feacalis, Streptococcus bovis, Enterotoxigenic Bacteroides fragilis (ETBF) and Porphyromonas spp. showed improved diagnostic performance, compared to each bacterium alone, as area under the receiver operating characteristic (AUROC) increases to 0.97, with 95% confidence interval. It was found that a simple linear combination was an appropriate model for discriminating AP and CRC cases, compared to the NC, with a sensitivity of 91.4% and specificity of 93.5%. Our results indicated that based on fecal bacterial candidates, statistical simple linear combination model and ROC curve analysis, early detection of AP and CRC might be possible. •Fecal bacteria was evaluated regarding colon cancer (CRC) early detection.•Quantitative real time PCR (qPCR) assays were established for quantification.•Multiple logistic regressions were applied to find the best combinations.•The combinations were analyzed to find best sensitivity and specificity.•Based on fecal bacteria, CRC early detection might be possible.