From Signal to Action: Functional Microbiome Science for Clinical Decision-Making

Microbiome research has produced many signals; the next challenge is to establish which ones can guide a clinical decision, for whom, and with what confidence.Microbial communities influence human health through immune regulation, metabolic activity, epithelial-barrier function and drug processing. Across cancer, inflammatory disease, metabolic disorders, transplantation, infection and oral health, studies have linked microbial composition and function with disease activity, treatment response, adverse effects and recovery.The clinical promise is now clear, but translation remains incomplete. In 2026, three early studies combining faecal microbiota transplantation with cancer immunotherapy brought microbiome interventions closer to clinical testing. They produced encouraging signals, but also showed the limits of the current evidence base. In the placebo-controlled TACITO trial in metastatic renal-cell carcinoma, donor FMT improved median progression-free survival but did not meet the primary 12-month progression-free-survival endpoint. Few microbiome-informed approaches currently guide routine treatment selection, toxicity prevention or disease management outside well-established indications.The problem is not a lack of microbiome associations. It is the need to identify which signals are reproducible, biologically meaningful and useful in clinical practice. Results may vary across populations, diet, medication exposure, sampling approaches and analytical pipelines. Many studies describe taxonomic differences without linking them to microbial function, host physiology or a practical action.This Research Topic welcomes the full range of research needed to close that gap. It brings together microbial mechanisms, host physiology, biomarkers, therapeutics, clinical studies, multi-omics, data science, artificial intelligence and implementation research. It is inclusive by design: longitudinal studies, machine learning, clinical language models and digital twins are encouraged where they add insight, but they are not required.The shared question is simple: what evidence is needed for a microbiome finding to change patient care?We welcome submissions across human disease areas, body sites and microbial kingdoms, including work on the gut, oral, tumour, skin, respiratory and other clinically relevant microbiomes.1. Microbial function, mechanisms and host physiology-Microbial metabolites, enzymes and pathways that influence immunity, metabolism, barrier integrity, inflammation or drug response.-Functional studies using metagenomics, metabolomics, metaproteomics, metatranscriptomics, culturomics and experimental assays.-Host–microbe interactions involving bile acids, short-chain fatty acids, enterohepatic circulation, mucosal physiology, pharmacokinetics and oral–systemic pathways.-Taxonomic and strain-level studies that advance understanding of clinically relevant biology.2. Clinical associations, biomarkers and patient outcomes-Microbiome links with diagnosis, prognosis, disease activity, treatment response, treatment toxicity or supportive-care outcomes.-Clinical cohorts, prospective or retrospective studies, case series and studies in under-represented populations.-Longitudinal sampling, replication studies, failed replications and negative findings, where available.-Biomarkers intended for diagnosis, risk stratification, monitoring or treatment selection.3. Microbiome therapeutics and intervention studies-Faecal microbiota transplantation, defined microbial consortia, live biotherapeutics, bacteriophages, probiotics, prebiotics, postbiotics and dietary interventions.-Clinical trials, pilot studies, protocols and preclinical work that supports therapeutic development.-Microbial drug metabolism, pharmacomicrobiomics and host–microbe–drug interactions.-Safety, donor selection, engraftment, dose, durability and response heterogeneity.4. Data, methods and clinical decision support-Standardisation of sampling, storage, extraction, sequencing, absolute quantification and reporting.-Multi-omic integration, statistical analysis, reproducible workflows and benchmarking.-Predictive models, machine learning, clinical language models, patient-state models and digital twins that use microbiome data to address a clinically meaningful question.-Studies that integrate microbial measures with clinical data, physiology, imaging, laboratory measurements or patient-reported outcomes.5. Evidence, implementation and policy-Systematic reviews, meta-analyses and evidence syntheses.-Study quality, reproducibility, clinical utility and reporting standards.-Regulation, manufacturing, donor screening, ethics, data governance, health economics, equity and adoption.-Perspectives from clinicians, patients, trialists, regulators and industry. Topic editors: Pradeep Kumar Yadalam Source: https://www.frontiersin.org/research-topics/86933/from-signal-to-action-functional-microbiome-science-for-clinical-decision-making