Stool host RNA turns exfoliated gut cells into a diagnostic signal
A stool sample can carry far more than microbial DNA. In this preprint, whole-transcriptome sequencing of host RNA recovered from stool captured a cellular readout of the gastrointestinal tract, with epithelial and immune transcripts together forming a noninvasive window into disease activity. The approach was tested across 285 libraries spanning colorectal cancer, inflammatory bowel disease, precancerous lesions, and normal samples, and the median recovery reached 6,300 protein-coding genes per library.
The biological premise is straightforward but clinically compelling: cells shed into the intestinal lumen leave behind RNA fragments that preserve signatures of tissue state. In healthy samples, epithelial cells accounted for 81% of stool host mRNA transcripts and immune cells for 11%. In colorectal cancer and inflammatory bowel disease, the immune fraction rose, and B- and T-cell receptor clonotypes increased alongside it. That shift points to more than passive shedding. It reflects a mucosal environment where epithelial turnover, immune recruitment, and local inflammation are all being recorded in real time.
Distinct disease signatures emerged across colorectal cancer, IBD, and normal stool
Investigators reported clearly separated transcriptome profiles for colorectal cancer, inflammatory bowel disease, and normal samples. A three-class model distinguished the groups with an area under the curve of 0.94 to 0.96, a range that suggests strong separation in this dataset. The signal is especially interesting because it comes from stool host RNA rather than invasive tissue sampling, and from a specimen that patients already provide routinely in noninvasive screening or monitoring workflows.
One sentence in this dataset carries particular clinical weight. The immune transcript fraction rose in disease, not just the epithelial fraction falling away. In practice, that means the assay is not only detecting loss of normal mucosal architecture; it is also sensing the inflammatory response that accompanies malignant and inflammatory injury. For gastrointestinal disorders that blur together symptomatically, a combined epithelial-immune signature could help triage whether the dominant biology is neoplastic, inflammatory, or within normal variation.
The inclusion of precancerous lesions adds another layer of interest, although the abstract does not provide separate performance metrics for that subgroup. Even so, the study design suggests the possibility that stool RNA may eventually detect biologic transitions before they become morphologic or symptomatic enough to force colonoscopy. Any such use would require prospective validation, but the pathophysiology is plausible: early mucosal remodeling should alter the transcript pool long before overt bleeding or mass effect appears.
Neutrophil transcripts tracked endoscopic activity better than fecal calprotectin in the same stool
Within inflammatory bowel disease, the most actionable result may be the 15-gene neutrophil score. This transcript-based measure tracked endoscopic activity better than fecal calprotectin in the same stool sample. For bedside care, that comparison matters because fecal calprotectin already serves as a widely used noninvasive marker of intestinal inflammation, yet it does not always align tightly with mucosal disease burden.
A transcript score built from neutrophil biology may have an advantage because it reads the local cellular program rather than a downstream protein spillover. Neutrophils are central to active mucosal inflammation, particularly in ulcerative and Crohn-like inflammatory states where epithelial injury recruits innate immune cells into the lumen. A host RNA readout could therefore capture both the cellular presence and the activation state of those cells, giving a more immediate measure of endoscopic inflammation than a single protein marker can provide.
The abstract does not supply the score’s calibration, external validation, or subgroup behavior across disease phenotypes. Those details will matter if the assay is to move toward clinical use. Even so, the direction of effect is clear: transcriptomic stool profiling may outperform a conventional biomarker when the clinical question is mucosal activity rather than generalized inflammation.
A noninvasive assay that reads both tissue state and immune dynamics
The main appeal of stool host transcriptomics lies in what it can unify. Colonoscopy visualizes mucosa directly, but it is invasive, requires preparation and anesthesia, and is not suited to frequent monitoring. A stool assay that profiles epithelial and immune cell states could, in theory, serve as a lower-burden way to monitor disease evolution between endoscopic examinations. For colorectal cancer surveillance, it may someday help identify abnormal epithelial programs. For inflammatory bowel disease, it may capture flares, response to treatment, or residual activity after symptoms begin to settle.
The study also reinforces a broader point in mucosal biology. The gut is not a static surface; it is a rapidly renewing barrier where epithelial differentiation, immune surveillance, and clonal expansion occur continuously. RNA recovered from stool appears to preserve that biology well enough to be measured at scale. The median recovery of 6,300 protein-coding genes across libraries is enough to suggest a rich molecular signal rather than a sparse or noisy remnant.
Validation remains the real hurdle before any clinical adoption
The preprint is promising, but it is still a preprint. No peer review has yet tested the analytic pipeline, the model transportability, or the robustness of the neutrophil score across independent cohorts and collection conditions. The abstract also does not report sensitivity, specificity, positive predictive value, or how the assay performs against other clinically relevant comparators beyond fecal calprotectin in inflammatory bowel disease.
Data availability may also limit immediate replication. Raw sequencing datasets are not publicly available because of patient confidentiality and institutional restrictions, although de-identified data may be requested through the corresponding author and review committee. That creates a practical barrier for outside groups that would want to reproduce the findings or compare them with other stool-based assays.
Still, the core message is concrete: stool host RNA can recover enough human transcript information to distinguish gastrointestinal disease states and to mirror inflammatory activity within the bowel lumen. If future studies confirm the signal in independent populations and show stable performance in real-world sampling, stool transcriptomics could move closer to a noninvasive complement to colonoscopy rather than a theoretical laboratory curiosity.