A nested case-control study found that gut microbiota alterations may serve as potential early biomarkers for incident NAFLD, with a CatBoost model integrating multi-omics and clinical data achieving an AUC of 0.762 for predicting incident NAFLD.
The NAFLD incident group had significantly lower species-level beneficial gut microbiota compared to controls.
A CatBoost model integrating gut microbiota, functional pathways, and clinical indicators achieved moderate predictive performance for incident NAFLD.
Coprobacillus unclassified was identified as a key microbial species distinguishing NAFLD incident cases from controls within the diagnostic model.
The study used a nested case-control design drawn from an initial cohort of 1,772 participants, with 404 participants included in the final analysis.
Gut microbiota-based approaches show potential for non-invasive risk stratification of NAFLD.
This research used a nested case-control study design to investigate whether the bacteria living in the gut (gut microbiota) could help predict who will develop non-alcoholic fatty liver disease (NAFLD), a condition where fat accumulates in the liver in people who drink little or no alcohol. Researchers followed a large group of over 1,700 people and compared 101 who later developed NAFLD to 303 matched individuals who did not, analyzing their gut bacteria using advanced DNA sequencing. They found that people who went on to develop NAFLD already had lower levels of beneficial gut bacteria and distinct differences in how their gut bacteria functioned metabolically, even before the disease was formally diagnosed. The researchers built a machine learning model called CatBoost that combined gut bacteria data, metabolic pathway information, and standard clinical measurements. This model was able to predict who would develop NAFLD with moderate accuracy (AUC of 0.762, where 1.0 would be perfect prediction). A specific bacterial species called Coprobacillus unclassified and a metabolic pathway called PWY-5030 were identified as particularly important distinguishing features between those who did and did not develop NAFLD. This research suggests that the composition of gut bacteria may serve as an early warning signal for NAFLD development and could potentially be used as a non-invasive tool to identify people at higher risk before the disease fully develops. Since current NAFLD diagnosis often relies on liver biopsies or imaging, a gut microbiota-based approach could offer a less invasive alternative for risk screening, though further research in larger and more diverse populations would be needed to validate these findings.
Yang X, Chen J, Song J, Liu H, Wang L, Li T, et al.. (2026). A nested case-control study on the relationship between gut microbiota and non-alcoholic fatty liver disease.. Frontiers in endocrinology. https://doi.org/10.3389/fendo.2026.1917815