Gut microbiome restructuring in laryngeal squamous cell carcinoma identifies stable microbial biomarkers with...

Alterations in the gut microbiota have been reported in various malignancies, but its role in laryngeal squamous cell carcinoma (LSCC) remains unclear.

This retrospective study included 101 patients undergoing laryngeal surgery (46 benign, 55 malignant). Preoperative fecal samples were collected and subjected to 5R 16S rRNA sequencing. Sequencing data were processed using DADA2 and QIIME2, followed by α/β diversity analysis, differential abundance detection (Wilcoxon test, LEfSe, random forest), and LASSO regression. Functional pathway differences were inferred using PICRUSt2.

There were no significant differences in α diversity metrics between groups, whereas β diversity analysis revealed significant separation between Benign and LSCC (PERMANOVA, P<0.01). Distinct community composition differences were observed: Malignant cases showed enrichment of genera such as Streptococcus and Lactobacillus, while Benign cases exhibited enrichment of genera including Bacteroides and Lachnospira. Multimethod integration identified 17 core bacterial genera, which were further refined via LASSO regression to select a stable set of genera (e.g., Streptococcus, Eubacterium, Lachnospira) capable of reliably distinguishing benign from malignant cases. The logistic regression model based on stable genera demonstrated excellent diagnostic performance (AUC > 0.8), particularly in distinguishing benign from LateLSCC. Functional prediction revealed pathway imbalances: malignant cases showed enrichment in cell wall and amino acid synthesis pathways, while benign cases favored vitamin and steroid metabolism pathways.

LSCC patients exhibit structural remodeling of their gut microbiota, characterized by distinct taxonomic and functional alterations. Stable microbial signatures holding potential as a foundation for the future development of non-invasive diagnostic and staging biomarkers, though their clinical translation necessitates further large-scale validation.