Altered Fecal Small RNA Profiles in Colorectal Cancer Reflect Gut Microbiome Composition in Stool SamplesOriginal paper
What was studied?
This study examined whether small RNA sequencing and shotgun metagenomic sequencing of stool could be combined to detect colorectal cancer (CRC). The researchers profiled both human small noncoding RNAs and bacterial small RNAs (bsRNAs) in stool, alongside DNA-based microbiome taxonomic data. They evaluated whether these small RNA profiles reflect gut microbiome composition and whether dysbiosis in CRC is detectable through altered small RNA patterns. The overall goal was to test the combined use of these data types as a predictive tool for disease detection.
Who was studied?
The analysis used 80 stool specimens collected in a cross-sectional study. Samples came from patients with colorectal cancer, patients with adenomas, and healthy subjects. No further demographic or clinical details of the cohort are given in the abstract.
What were the most important findings?
The researchers found considerable overlap and a correlation between metagenomic and bacterial small RNA (bsRNA) quantitative taxonomic profiles derived from the two sequencing approaches. They identified a combined predictive signature of 32 features drawn from human small RNAs, microbial small RNAs, and DNA-based microbiome data. This signature accurately classified CRC samples as distinct from healthy and adenoma samples, achieving an area under the curve (AUC) of 0.87. The findings show that host-microbiome dysbiosis in CRC can be observed through altered small RNA stool profiles, not just through standard microbiome sequencing.
What are the greatest implications of this study?
The results suggest that integrating small RNA data (both human and bacterial) with microbiome DNA sequencing can improve noninvasive stool-based detection of colorectal cancer. Because bsRNA and metagenomic taxonomic profiles correlate, small RNA sequencing may serve as an additional or complementary readout of gut microbiome composition. The 32-feature combined signature points toward multi-omic stool panels as a promising direction for distinguishing CRC from adenomas and healthy states. This integrated approach may inform future noninvasive screening tools designed for earlier CRC detection.