Study protocol for design of a personalized dietary supplement based on the gut microbiota of Alzheimer's patients and evaluation of its effects in a pilot randomized controlled trial

Abstract

BACKGROUND: Increasing evidence links gut microbiota composition to neurological disorders, including Alzheimer's disease (AD), through the gut-brain axis. Microbial metabolites such as lipopolysaccharide (LPS) and butyrate are associated with AD progression. Among modifiable factors, diet plays a central role in shaping gut microbiota and may influence disease-related microbial patterns. Personalized nutrition based on individual microbiota profiles offers a promising strategy to modulate these biomarkers. This protocol describes a study designed to develop a personalized dietary supplement for patients with AD by integrating microbiota, clinical, and dietary data using artificial intelligence (AI) and network analysis. A secondary objective is to assess in a pilot study the short-term effects of the supplement on microbiota composition, function, and plasma metabolomics to identify modifiable biomarkers. DESIGN: This two-phase study will begin with baseline data collection from 60 patients with AD and 60 healthy controls. AI and network-based analyses will identify dietary and microbial variables predictive of disease status. Based on these findings, a personalized supplement will be formulated using dietary components-such as fibers, polyphenols, or fatty acids-targeting microbial taxa and metabolic pathways associated with AD. In the second phase, 60 patients with AD will be randomized (1:1) to receive either the personalized supplement or a standard product for 3?months. Effects on microbial taxa, LPS, short-chain fatty acids (SCFAs), and plasma metabolites will be evaluated before and after the intervention. METHODS: Participants will undergo assessments of lifestyle factors, including diet and physical activity, as well as measurements of blood LPS and fecal SCFAs. Machine learning and network analyses will explore links among microbiota, diet, and clinical features. Predictive variables will guide supplement design. The intervention will evaluate changes in LPS, butyrate, and microbial markers. Plasma metabolomic profiles will also be analyzed. Data integration using AI and network approaches will help identify biomarkers and assess intervention outcomes. DISCUSSION: Combining AI and network analysis in microbiota research supports the development of personalized nutrition strategies in AD. This approach may modulate disease-related microbiota and systemic markers, contributing to innovative therapeutic tools and advancing both Alzheimer's care and nutritional science. CLINICAL TRIAL REGISTRATION: ClinicalTrials.gov, identifier NCT06199193.