Characterizations of the multi-kingdom gut microbiota in elderly individuals with end-stage renal disease

End-stage renal disease (ESRD) is common in older adults and has been linked to gut dysbiosis, yet multi-kingdom alterations and cross-kingdom interactions in elderly patients remain incompletely defined. We characterized the gut bacteriome, mycobiome, and virome in elderly individuals with ESRD and assessed their diagnostic potential. Fecal samples from 42 patients with ESRD and 50 age-, sex-, and body mass index–matched healthy controls (all ≥ 65 years) underwent deep whole-metagenome shotgun sequencing. Species-level profiles were generated for bacterial and fungal communities using reference genome catalogues, while viral genomes were identified using a contig-based pipeline with stringent quality control. Microbial functional potential was characterized by KEGG Orthology (KO) profiling. The bacterial community structure differed between groups, with higher species richness in ESRD and 41 differential bacterial species identified; Faecalibacterium prausnitzii was depleted, whereas Ruminococcus gnavus was enriched. The fungal mycobiome showed increased richness and diversity in patients with ESRD, with enrichment of Cunninghamella , Lodderomyces , Barnettozyma , and Pseudopithomyces and 11 differential fungal species identified. Viral alpha diversity was comparable between groups, but virome composition differed significantly, with 151 differential viral species and enrichment of Siphoviridae in ESRD patients. Functional profiling revealed selective alterations in bacterial and viral Kos despite limited differences in overall functional diversity. Cross-kingdom correlation networks demonstrated more extensive bacteria–virus associations than fungus-related interactions. Random forest models based on microbial features showed potential for discriminating ESRD from healthy controls, with areas under the curve of 0.822 (bacteria), 0.739 (fungi), 0.904 (viruses), and 0.898 (combined model). Elderly patients with ESRD exhibit extensive multi-kingdom alterations in gut microbial composition and function. Differential bacterial, fungal, and viral taxa, together with altered bacteria–virus associations and selective microbial functional changes, characterize ESRD-associated gut microbiome remodeling. Microbial features identified by random forest models showed potential for distinguishing ESRD patients from healthy individuals. Further validation in independent cohorts is required to determine their diagnostic utility and biological relevance.