Development of fecal microbial diagnostic marker sets of colorectal cancer using natural language processing method.

BACKGROUND: Cancer screening and early detection greatly increase the chances of successful treatment. However, most cancer types lack effective early screening biomarkers. In recent years, natural language processing (NLP)-based text-mining methods have proven effective in searching the scientific literature and identifying promising associations between potential biomarkers and disease, but unfortunately few are widely used. METHODS: In this study, we used an NLP-enabled text-mining system, MarkerGenie, to identify potential stool bacterial markers for early detection and screening of colorectal cancer. After filtering markers based on text-mining results, we validated bacterial markers using multiplex digital droplet polymerase chain reaction (ddPCR). Classifiers were built based on ddPCR results, and sensitivity, specificity, and area under the curve (AUC) were used to evaluate the performance. RESULTS: A total of 7 of the 14 bacterial markers showed significantly increased abundance in the stools of colorectal cancer patients. A five-bacteria classifier for colorectal cancer diagnosis was built, and achieved an AUC of 0.852, with a sensitivity of 0.692 and specificity of 0.935. When combined with the fecal immunochemical test (FIT), our classifier achieved an AUC of 0.959 and increased the sensitivity of FIT (0.929 vs. 0.872) at a specificity of 0.900. CONCLUSIONS: Our study provides a valuable case example of the use of NLP-based marker mining for biomarker identification.

Postoperative circulating tumor DNA detection and CBLB mutations are prognostic biomarkers for gastric cancer.

BACKGROUND: Several studies have demonstrated that circulating tumor DNA (ctDNA) can be used to predict the postoperative recurrence of several cancers. However, there are few studies on the use of ctDNA as a prognosis tool for gastric cancer (GC) patients. OBJECTIVE: This study aims to determine whether ctDNA could be used as a prognostic biomarker in GC patients through multigene-panel sequencing. METHODS: Using next-generation sequencing (NGS) Multigene Panels, the mutational signatures associated with the prognosis of GC patients were identified. We calculated the survival probability with Kaplan-Meier and used the Log-rank test to compare survival curves between ctDNA-positive and ctDNA-negative groups. Potential application of radiology combined with tumor plasma biomarker analysis of ctDNA in GC patients was carried out. RESULTS: Disease progression is more likely in ctDNA-positive patients as characterized clinically by a generally higher T stage and a poorer therapeutic response (P < 0.05). ctDNA-positive patients also had worse overall-survival (OS: P = 0.203) and progression-free survival (PFS: P = 0.037). The combined analysis of ctDNA, radiological, and serum biomarkers in four patients indicated that ctDNA monitoring can be a good complement to radiological and plasma tumor markers for GC patients. Kaplan-Meier analysis using a cohort of GC patients in the TCGA database showed that patients with CBLB mutations had shorter OS and PFS than wild-type patients (OS: P = 0.0036; PFS: P = 0.0027). CONCLUSIONS: This study confirmed the utility and feasibility of ctDNA in the prognosis monitoring of gastric cancer.