Characterization of cervical microbiota in cervical intraepithelial neoplasia and cervical cancer using low-coverage whole genome sequencing.

This study characterized compositional shifts in cervical microbiota across disease stages from benign conditions through cervical intraepithelial neoplasia (CIN) to cervical cancer (CC) and investigated interactions with high-risk HPV (hr-HPV) infection using species-resolution profiling to identify severity-associated biomarkers. Cervical exfoliated epithelial cells from 50 patients (eight normal/CIN1, 15 CIN2, 19 CIN3, 5 CC) were analyzed using Low-Coverage Whole Genome Sequencing combined with the Ultrasensitive Chromosomal Aneuploidy Detector (UCAD), a technology featuring a two-step normalization framework that systematically converts raw microbial reads into statistically validated abundance deviations. This enables quantitative identification of pathologically relevant microbiota through cohort-wide Z-score benchmarking. Microbial diversity, differential biomarkers, and HPV-microbiota interactions were assessed using Kruskal-Wallis tests, LEfSe, and Random Forest modeling. Results revealed progressive Lactobacillus depletion (e.g., Lactobacillus crispatus: 32.9% in ≤CIN2 vs. 8.8% in CC) and enrichment of pathobionts like Gardnerella and Bacteroides with lesion severity. CC exhibited the highest microbial diversity (Shannon index: CC vs. CIN2, P=0.045), dominated by HPV16 (11.8%), Bacteroides (55.4%), and Porphyromonas (25.2%). LEfSe identified HPV16, HPV35, Parvimonas micra, and Anaerococcus lactolyticus as CC-specific markers, while Random Forest highlighted Mobiluncus curtisii (importance score=2.0) and HPV16 as key discriminators. CC microbiota showed significant Bacteroidetes enrichment (82% at class level) and reduced Firmicutes abundance. These findings suggest carcinogenesis-associated microbial restructuring, marked by Lactobacillus loss, anaerobic proliferation, and HPV16/35 dominance, potentially modulating disease progression. The identified signatures may inform diagnostic development and microbiome-targeted therapies.IMPORTANCEOur study pioneers an LC-WGS/UCAD approach to characterize microbial across the spectrum from benign lesions through precancerous cervical intraepithelial neoplasia to invasive cervical carcinoma. By identifying lesion-specific microbial biomarkers and HPV-associated cofactors, this work advances mechanistic understanding of microbiota-driven oncogenesis and informs future strategies for microbiota-targeted cervical cancer prevention.