Abstract
BACKGROUND: The cervicovaginal microbiome-spanning from the vagina to endometrium-remains undercharacterized in endometriosis-associated infertility. Objective: To determine whether combined vaginal and cervical microbial profiles predict frozen embryo transfer (FET) outcomes. METHODS: In 22 endometriosis patients undergoing FET, paired vaginal and cervical samples were collected on transfer day. 16S rDNA sequencing quantified microbial composition; alpha/beta diversity, PCoA, LEfSe, and PICRUSt analyses identified taxonomic and functional signatures linked to implantation success. Conduct a differential analysis of microorganisms in different body parts through DMI. RESULTS: Microbial profiles associated with successful pregnancies featured a higher relative abundance of Lactobacillus and Bifidobacterium, whereas Gardnerella, Streptococcus, and Atopobium were more enriched in failures. Cervical alpha diversity was significantly lower in successful transfers. LEfSe highlighted differential taxa including Peptostreptococcales in successes and Pseudomonadaceae in failures. Functional inference predicted dysregulated metabolic pathways in failure-associated communities. Furthermore, the cervical microbiota exhibited higher DMI, indicating greater individual specificity. CONCLUSIONS: Our pilot findings suggest that a continuous cervicovaginal microbial ecosystem presents distinct taxonomic and functional patterns associated with FET success in endometriosis. Specifically, cervical microbiota profiling emerges as a promising, minimally invasive approach worthy of further investigation to potentially personalize ART strategies.