Dynamic changes and clinical significance of the gut microbiota and serum metabolites in breast cancer onset, progression and chemotherapy intervention.

OBJECTIVE: Alterations of the gut microbiota and host metabolic reprogramming are closely associated with the development of breast cancer and the treatment response; however, integrated studies of the gut microbiota and metabolome spanning the transition from benign breast disease (BBD) to malignancy and the postchemotherapy phase remain limited. This study aims to systematically characterize the dynamic changes in the "gut microbiota-serum metabolome-breast tumor" axis from benign breast disease (BBD) to breast cancer (BC) and postchemotherapy breast cancer (PCBC) and to evaluate its potential value in diagnosis and disease monitoring.

METHODS: We enrolled 295 female participants, who were divided into a BBD group (n = 83), a BC group (n = 100), and a PCBC group (n = 88), and included 24 paired fecal samples from the same patients that were collected before and after chemotherapy. Fecal samples underwent 16S ribosomal RNA (rRNA) sequencing, while serum samples underwent an liquid chromatography-high-resolution mass spectrometry (LC-MS/MS) based nontargeted metabolomic analysis; we compared differences in gut microbiota diversity, taxonomic composition, and functional predictions across groups and screened for differentially abundant metabolites and enriched metabolic pathways. In a subset of patients with paired multiomics data (BBD n=19, BC n=31, and PCBC n=34), Spearman's correlation analysis, multiomics principle component analysis (PCA)/partial least squares-discriminant analysis (PLS-DA), and random forest models were employed to integrate the microbiota and metabolic features.

RESULTS: Cross-sectional remodeling of the gut microbiome structure occurred. The α diversity of the gut microbiota was similar across the three groups (BBD, BC, and PCBC); however, the β diversity analysis based on the weighted UniFrac distance revealed the significant separation of the microbial community structure among the three groups. At the taxonomic level, the BBD group was significantly enriched with beneficial commensal bacteria that produce short-chain fatty acids (e.g., Faecalibacterium and Roseburia); in contrast, the BC group shifted toward the enrichment of inflammation- or tumor-associated genera (e.g., Blautia, Fusobacterium, Sneathia, and Prevotella), while the PCBC group further accumulated various opportunistic pathogens (such as Phocaeicola, Sutterella, Enterococcus, and Chlamydia). As the disease progressed and chemotherapy was administered, the microbiome gradually shifted from a "metabolic protective" state to an "inflammatory/pathogenic" state. The characteristics of the dynamic remodeling of serum metabolomic profiles were identified. Nontargeted metabolomics revealed more than 3,000 metabolites, and the multivariate analysis indicated significant heterogeneity in the metabolomic profiles associated with malignant transformation and those measured before and after chemotherapy. In the BC group, energy, amino acid, and lipid metabolism were significantly disrupted, and widespread metabolite depletion was observed; however, in the postchemotherapy PCBC group, adaptive pathways such as estrogen, bile acid, and drug metabolism were activated, and persistent abnormalities in purine/nucleotide, carbon, and multiple amino acid metabolism were detected. Overall, the serum metabolic network underwent a dynamic remodeling process, transitioning from homeostasis to severe disruption and then to partial reconstruction following treatment. Among these changes, the differentially abundant metabolites torsemide, cortolone-3-glucuronide, and trimethylselenonium all had area under the curve (AUC) values greater than 0.75 in distinguishing between different disease stages and chemotherapy statuses, demonstrating their good potential as biomarkers. Interaction networks and multiomics predictive models of the "gut-metabolism-tumor" axis were established. Multiomics association networks revealed a systemic shift in gut-metabolism interaction patterns from a steady state characterized by "probiotic-energy/amino acid and polyphenol metabolism coupling" during the benign phase to a new steady state dominated by "proinflammatory/opportunistic pathogens-lipid reprogramming, exogenous metabolite metabolism, and oxidative stress." A multiomics classification model based on random forests demonstrated that the combined analysis of the gut microbiota and serum metabolite profiles exhibited exceptional efficacy in distinguishing between the BBD, BC, and PCBC groups. It identified a cluster of strongly associated features characterized by anaerobic gram-positive cocci, Lactobacillus-associated microbiota, and their paired metabolites, providing an important molecular fingerprint for clinical assessment.

CONCLUSIONS: This study integrates gut microbiome and serum metabolomic data to reveal that during the progression of benign breast lesions to breast cancer and throughout chemotherapy, the "gut microbiota-host metabolism-breast tumor" axis transforms from a state characterized by commensal depletion, the expansion of opportunistic pathogens, and reprogramming of energy/lipid metabolism to a persistent metabolic signature associated with drug metabolism and the activation of oxidative stress pathways. The integrated multiomics model helps characterize the biological differences across various stages of breast disease. The identified characteristic bacterial genera and metabolite combinations provide preliminary theoretical clues for the future exploration of microbiome-related mechanisms in breast cancer and for auxiliary assessments.