Probiotic supplementation on cognitive and other aging-related physiological functions in middle-aged and older adults with mild cognitive impairment (PCAMCI): protocol for a randomized, triple-blinded, placebo-controlled trial Authors: Jiani Huang, Quan Zou, Yue Chen, Yutong Zuo, Berty Ruping Song, Shuning Li, Chuang Han, Haiyang Dong, Gongshang Liu, Zidong Zhao, Wei Wei, Zhongren Liu, Jinglei Chai, Sixin Xie, Yunxin Wang, Piao Hou, Xiang Gao, Liang Sun Published: 2025 ABSTRACT BackgroundEmerging evidence suggests gut microbiota modulation may influence neurocognitive function through the gut-brain axis. Although preliminary studies indicate probiotics’ potential benefits for mild cognitive impairment (MCI), well-designed randomized controlled trials remain limited. This protocol paper describes a rigorously designed, registered clinical trial investigating the effects of targeted probiotic supplementation on cognitive and physiological outcomes in MCI participants.MethodsA total of 110 middle-aged and older participants aged 55–80 years with MCI were scheduled to be included in the study, and randomized in a 1:1 ratio to either a probiotic group receiving supplementation consisting of Lactiplantibacillus plantarum ST-III, Lacticaseibacillus rhamnosus KF7, and Lacticaseibacillus paracasei BD5115, or a placebo group with maltodextrin for 12 months. All the participants, researchers, and analysts will remain blinded to the information regarding group allocation in the study. The primary outcome will be the effect of probiotic supplementation on cognitive function, measured by Montreal Cognitive Assessment (the Chinese Beijing Version). The secondary outcomes will include the impact of probiotic supplementation on digestive health, sleep health, facial aging, fundus conditions, olfactory and auditory function, body composition, bone density, and muscle function. Brain magnetic resonance imaging and wearable device including continuous glucose monitor and smart band will also be employed.DiscussionThis study will provide some new insights on how probiotic supplementation could impact cognitive function and other aging-related physiological functions in MCI adults and explore the potential underlying mechanisms. The findings may inform the development of strategies to delay cognitive decline by modulating the gut-brain axis in this high-risk population.Trial registrationChiCTR2400084594. Introduction Mild cognitive impairment (MCI), a stage where cognitive function declines beyond the expected level according to a person’s age and education level, often serves as a precursor to the clinical diagnosis of Alzheimer’s disease (AD) and other forms of dementia [1]. A recent meta-analysis estimated the global prevalence of MCI to be 15% among community dwellers aged 55 and older, which increased with age and was higher in lower educational level [2]. As life expectancy rises globally, MCI progression to dementia poses a major public health challenge. Current pharmacologic MCI treatments show limited or inconclusive efficacy in delaying cognitive decline or AD conversion [3, 4]. Evidence indicated the gut-brain axis mediates interactions between the gastrointestinal tract and central nervous system. This axis influences metabolic, neuroimmune, and sensory neuronal pathways, and may thus represents a potential target for preventing dementia [5]. Probiotics are live microorganisms that confer a health benefit on the host, and may modulate this axis [6]. Animal studies showed that the use of Lactobacillaceae strains could improve motor, cognitive, and psychiatric symptoms associated with neurodegeneration [7]. Human studies revealed cognitive enhancements, especially in attention, following 12-week Lactiplantibacillus plantarum supplementation in older adults [8, 9]. Another study with a three-month intervention of Lacticaseibacillus rhamnosus identified specific gut microbiota species (Prevotella ruminicola, Bacteroides thetaiotaomicron, and Bacteroides xylanisolvens) correlated with cognitive performance in MCI people [10]. However, most previous studies had short intervention durations (< six months) [8, 9, 11, 12] and were often designed without blinding or randomized controls, which limited the robustness of the findings. Mechanistic studies suggested that gut microbiota may affect cognition via aging-related physiological health [13]. Regarding digestive health, microbiota-mediated intestinal aberrations was found to contribute to AD’s progress in mice, indicating that restoring gut microbial homeostasis could benefit AD treatment [14]. Evidence showed that sleep disturbance was also a candidate risk factor for AD [15]. Sleep and circadian rhythm disruption can lead to dysbiosis in gut microbiota, which compromises both the gut and blood-brain barriers, resulting in neuroinflammation, β-amyloid (Aβ) deposition, and the onset of AD [16]. A recent randomized trial suggested that probiotic supplementation could effectively alleviate insomnia symptom by regulating gut microbiota and promoting the production of beneficial fecal metabolites [17]. Targeting this, probiotics may improve digestive and sleep health and enhance cognitive function by reducing the aging-related damage of the intestinal barrier [18]. However, the studies mentioned above were mostly conducted in animal models, while the effect on humans has not been conclusively established. Beyond digestive function and sleep health, multisystem degeneration may synergistically accelerate pathological aging. Accumulating evidence demonstrates that probiotic interventions can ameliorate musculoskeletal aging, with randomized controlled trials reporting improvements in facial collagen integrity [19], body composition [20], bone mineral density [21], and muscle strength [22]. In contrast, the effects of probiotics on other hallmarks of systemic aging, such as retinal microvascular health (fundus) and sensory function (olfaction/audition), remain unexplored. To address these gaps, we conduct a rigorous RCT with a 12-month probiotic supplementation intervention. This is a critical improvement extending beyond the typical durations (≤ 6 months) of most comparable work. We evaluate the effects of a novel multi-strain probiotic (L. plantarum ST-III, L. rhamnosus KF7, L. paracasei BD5115) compared to placebo in middle-aged/older adults with MCI. Previous human probiotic intervention trials on cognitive or other health outcomes have mostly used a limited set of well‑established strains, such as Bifidobacterium lactis, B. bifidum, Lactobacillus acidophilus, and L. casei [23–26]. In contrast, our trial employed a specific combination of Lactiplantibacillus plantarum ST‑III, Lacticaseibacillus rhamnosus KF7, and Lacticaseibacillus paracasei BD5115. L. plantarum ST-III was shown to improve behavioral symptoms in rodent models of neurological disorders [27]. L. rhamnosus strains were found to have protective effects targeting intestinal microbiota for abnormal brain activity [28]. L. paracasei stains modulated gut microbiota composition and inhibited pathogenic bacteria [29]. While many previous human trials have used common strains such as B. lactis, L. casei, B. bifidum, and L. acidophilus, our selected combination includes strains that are relatively less common in this research context, supported by multi-targeted preclinical evidence for safety and efficacy, and holds potential to expand probiotic options in cognitive health research. Moreover, unlike previous RCTs concentrating solely on one major health domain [8, 9, 11, 12], this study integrates comprehensive functional, biochemical, and imaging assessments spanning multiple interconnected aging domains. Assessments encompass cognitive function, gut microbiota, digestive health, sleep health, facial aging, fundus conditions, olfactory and auditory function, body composition, bone density, muscle function, brain structural MRI, and continuous glucose monitoring (CGM) via standardized methodologies. We hypothesize that 12-month supplementation will significantly benefit cognitive function, intestinal health, and sleep health compared to placebo. Methods Study aims This protocol paper describes the methodology of a randomized controlled trial investigating probiotic supplementation in MCI participants. The RCT aims to: (1) evaluate the effect of probiotic supplementation on cognitive function in middle-aged and older adults (55–80 years) with MCI; (2) assess the impact on digestive health, sleep health, facial aging, fundus conditions, olfactory and auditory function, body composition, bone density, and muscle function. Study design This trial is a 12-month, single-center, randomized, triple-blind, placebo-controlled, parallel-group study that has been registered at the Chinese Clinical Trial Registry (ChiCTR2400084594) and approved by Medical Research Ethics Committee, School of Public Health, Fudan University (IRB#2024-02−1097). The study protocol (Version 1.0, 20 February 2024) was developed in compliance with the Standard Protocol Items: Recommendation for Interventional Trials (SPIRIT) 2013 Statement. The study involves four phases (Fig. 1): (1) Enrollment (Completed): Community-dwelling participants were recruited in Shanghai (June-September 2024) using MoCA screening (cut-off < 26; detailed in section "Data collection"); (2) Run-In (Completed): Two-week placebo lead-in with baseline assessments; (3) Intervention (Ongoing): 12-month supplementation probiotic/placebo) with assessments at 3, 6, and 12 months; (4) Follow-Up (Scheduled): 12-month post-intervention monitoring. Fig. 1 showed the study schema and timeline. The SPIRIT diagram was displayed in Fig. 2. Fig. 1Overview of the study design and timeline Overview of the study design and timeline Fig. 2Schedule of enrolment, interventions, and assessments Schedule of enrolment, interventions, and assessments Study population A total of 110 participants aged 55 to 80 years with MCI were scheduled to be recruited for this study. To be eligible for the trial, participants must meet all the inclusion criteria and none of the exclusion criteria, as stated in Table 1. In addition, reasons for participants to be discontinued from the study are as follows: Table 1Inclusion criteria and exclusion criteriaCriteriaContentsInclusion1) Aged 55 to 80 years at the time of screening;2) MoCA (the Chinese Beijing Version) score < 26 points;3) Sufficient visual and auditory capacities to participate in the entire trial.Exclusion1) Allergic to probiotic supplementation;2) A history of diseases that can affect cognitive function (e.g., dementia, Parkinson’s disease, stroke, hydrocephalus, and epilepsy);3) Receiving medication for organic psychosis or chronic psychosis that can affect cognitive function (e.g., antipsychotic drugs and tricyclic antidepressants) in the past three months;4) A history of serious diseases, including metabolic diseases, cancer, or major diseases of the heart, liver, kidney, lungs, gastrointestinal tract, blood, hematopoietic system, or endocrine system;5) A history of major gastrointestinal surgeries (e.g., gastrectomy, intestinal resection, or gastrointestinal suture surgery);6) A diagnosis of serious anxiety disorder or depression;7) A history of alcoholism or drug dependence;8) Participation in other clinical studies within the past three months;9) Regular use of the test-related products (other probiotic or prebiotic products) in the past month;10)Long-term absence from Shanghai (longer than 1 month) during the intervention period;11)Deemed ineligible based on screening data or by the researchers. Table 1Inclusion criteria and exclusion criteriaCriteriaContentsInclusion1) Aged 55 to 80 years at the time of screening;2) MoCA (the Chinese Beijing Version) score < 26 points;3) Sufficient visual and auditory capacities to participate in the entire trial.Exclusion1) Allergic to probiotic supplementation;2) A history of diseases that can affect cognitive function (e.g., dementia, Parkinson’s disease, stroke, hydrocephalus, and epilepsy);3) Receiving medication for organic psychosis or chronic psychosis that can affect cognitive function (e.g., antipsychotic drugs and tricyclic antidepressants) in the past three months;4) A history of serious diseases, including metabolic diseases, cancer, or major diseases of the heart, liver, kidney, lungs, gastrointestinal tract, blood, hematopoietic system, or endocrine system;5) A history of major gastrointestinal surgeries (e.g., gastrectomy, intestinal resection, or gastrointestinal suture surgery);6) A diagnosis of serious anxiety disorder or depression;7) A history of alcoholism or drug dependence;8) Participation in other clinical studies within the past three months;9) Regular use of the test-related products (other probiotic or prebiotic products) in the past month;10)Long-term absence from Shanghai (longer than 1 month) during the intervention period;11)Deemed ineligible based on screening data or by the researchers. Inclusion criteria and exclusion criteria 1) Aged 55 to 80 years at the time of screening; 2) MoCA (the Chinese Beijing Version) score < 26 points; 3) Sufficient visual and auditory capacities to participate in the entire trial. 1) Allergic to probiotic supplementation; 2) A history of diseases that can affect cognitive function (e.g., dementia, Parkinson’s disease, stroke, hydrocephalus, and epilepsy); 3) Receiving medication for organic psychosis or chronic psychosis that can affect cognitive function (e.g., antipsychotic drugs and tricyclic antidepressants) in the past three months; 4) A history of serious diseases, including metabolic diseases, cancer, or major diseases of the heart, liver, kidney, lungs, gastrointestinal tract, blood, hematopoietic system, or endocrine system; 5) A history of major gastrointestinal surgeries (e.g., gastrectomy, intestinal resection, or gastrointestinal suture surgery); 6) A diagnosis of serious anxiety disorder or depression; 7) A history of alcoholism or drug dependence; 8) Participation in other clinical studies within the past three months; 9) Regular use of the test-related products (other probiotic or prebiotic products) in the past month; 10)Long-term absence from Shanghai (longer than 1 month) during the intervention period; 11)Deemed ineligible based on screening data or by the researchers. Any serious adverse event occurring during the intervention period, including any event that results in death, is life-threatening, requires inpatient hospitalization or prolongation of existing hospitalization, or results in persistent or significant disability/incapacity according to the International Conference on Harmonization-Good Clinical Practice guideline (ICH-GCP 1997);The development of any disease, or the use of therapy or medication that can affect the effectiveness of probiotic intervention during the intervention period;Non-compliance with the intervention (probiotics or placebo) for more than one month;Withdrawal consents from participants. Any serious adverse event occurring during the intervention period, including any event that results in death, is life-threatening, requires inpatient hospitalization or prolongation of existing hospitalization, or results in persistent or significant disability/incapacity according to the International Conference on Harmonization-Good Clinical Practice guideline (ICH-GCP 1997); Any serious adverse event occurring during the intervention period, including any event that results in death, is life-threatening, requires inpatient hospitalization or prolongation of existing hospitalization, or results in persistent or significant disability/incapacity according to the International Conference on Harmonization-Good Clinical Practice guideline (ICH-GCP 1997); The development of any disease, or the use of therapy or medication that can affect the effectiveness of probiotic intervention during the intervention period; The development of any disease, or the use of therapy or medication that can affect the effectiveness of probiotic intervention during the intervention period; Non-compliance with the intervention (probiotics or placebo) for more than one month; Non-compliance with the intervention (probiotics or placebo) for more than one month; Withdrawal consents from participants. Withdrawal consents from participants. Sample size calculation The sample size calculation was based on the primary outcome measure of a change in the mean score of MoCA within groups. Previous studies showed that for a similar intervention using probiotic supplementation in adults with MCI, a standard deviation of 0.90 between intervention and placebo groups was observed [30]. Assuming a mean score change of 0.65 between the two arms with a probability (power) of 0.95, a Type-I error probability of 0.05, and a dropout rate of 20%, a total of 110 participants will be sufficient for the study. The calculation was conducted using PASS version 21.0.3 software based on the following formula:\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:n=\frac{(2\times\:({Z}_{1-\alpha\:/2}+{Z}_{1-\beta\:})\times\:{\sigma\:}^{2}}{{\varDelta\:}^{2}}$$\end{document} Where: n = required sample size per group. Z1 − a/2 = critical value for a two-tailed test (1.96 for a = 0.05). Z1−β= critical value for statistical power (1.645 for 95% power). σ = standard deviation (SD) of the outcome within each group. ∆= expected mean difference between the two groups. Randomization, blinding, and unblinding Participants were randomly assigned in a 1:1 ratio according to a computer-generated randomization scheme with either probiotics or placebo stratified by age, sex, and MoCA score. An independent statistician, who is not involved in participant recruitment, data collection, or outcome assessment, performed the randomization stratified by age (≤ 65 vs. >65 years), sex (men vs. women), and baseline MoCA score (≤ median vs. >median) via the block randomization (block size = 4) using SAS 9.4. Variables not included in stratification (e.g., education level, comorbidities) were compared between groups using the Mann-Whitney U test to confirm baseline balance. The allocation sequence will remain unavailable to participants or any member of the research team until the completion of the study. Unblinding events (e.g., accidental unblinding, unblinding due to a serious adverse event) should be documented by researchers, including the time of unblinding, reasons for unblinding, and the responsible researchers. Intervention During the intervention period, participants take one sachet of supplementation (probiotics or placebo) daily for 12 months. They were instructed to dissolve and mix the powder in room-temperature water and consume it 1 to 2 h after a meal. If any adverse events or symptoms of discomfort occur, participants should stop taking the supplementation and report the issue as soon as possible. Probiotic supplementation and placebo are identical in size, appearance, taste, and quality, except that the placebo does not contain probiotics. The probiotic used in this study is provided by Bright Dairy Co., Ltd with no known side effects [31, 32]. Each probiotic supplementation (2 g/sachet) contains no less than 3 × 1010 colony-forming units (CFUs) of probiotics (including 5 × 109 CFU of Lactiplantibacillus plantarum ST-III, 2 × 1010 CFU of Lacticaseibacillus rhamnosus KF7, and 5 × 109 of CFU Lacticaseibacillus paracasei BD5115), 1.8 g to 1.9 g of maltodextrin, and 0.001 to 0.01 g of food flavoring. In addition, biweekly face-to-face visits will be conducted by trained investigators following a standardized procedure limited to the intervention protocol. At each session, participants will receive probiotics for the subsequent two weeks, and their adherence, medication use in the past two weeks, and any adverse events will be recorded. To avoid potential bias in outcomes, no general health or lifestyle advice beyond the study protocol will be provided. Data collection Participants will complete four surveys including questionnaires, physical examinations, and sample collections during the intervention period: at baseline, 3 months, 6 months, and 12 months. During the post-intervention period, participants will undergo two additional rounds of surveys at 6 months and 12 months after the intervention. An overview of the data collection and timeline is presented in Fig. 1. The primary outcome of this study is cognitive function assessed by MoCA. MoCA is mainly used to assess cognitive function, which was developed as a cognitive screening tool that covers eight cognitive domains to detect MCI from healthy individuals [32], presenting excellent reliability (80–100%) and validity (50–87%) in middle-aged and older Chinese population [33, 34]. In this study, we use the Chinese Beijing Version of MoCA, with a cut-off score of less than 26 points to identify MCI in the screening survey. MoCA will be repeatedly assessed at 3-month, 6-month, 12-month during the intervention, and 6-month and 12-month post-intervention. Secondary outcomes as potential markers of probiotic effects included digestive health and sleep health. Exploratory outcomes included facial aging, fundus conditions, olfactory and auditory function, body composition, bone density, and muscle function, aiming to provide novel insights into the broader systemic effects of probiotic supplementation. For secondary outcomes including digestive health, sleep health, facial aging, fundus conditions, olfactory and auditory function, body composition, bone density, and muscle function and other covariates, information will be collected by standardized questionnaires and physical examinations described in Table 2. Table 2Data collection content during the intervention periodCategoryItemContentsTimepointGeneral QuestionnaireDemographic InformationAge, Sex, Marital Status, Occupation, Date of BirthMonths 0, 3, 6, 12Socioeconomic StatusYears of Education, Personal and Household IncomeMonths 0, 3, 6, 12Disease HistoryHistory of hyperglycemia, hyperlipidemia, hypertension, etc.Months 0, 3, 6, 12Smoking and Alcohol UseCurrent smoking/drinking status (quantity, type, etc.)Months 0, 3, 6, 12Physical Activity LevelOccupational and leisure-time physical activity/exerciseMonths 0, 3, 6, 12Dietary QuestionnaireDietary Survey3-day 24-hour Dietary Recall; Food Frequency QuestionnaireMonths 0, 3, 6, 12 for Dietary Recall; Months 0, 12 for Food Frequency QuestionnaireOther QuestionnairesCognitive FunctionMontreal Cognitive AssessmentMonths 0, 3, 6, 12Mental StatusGeriatric Depression Scale; Self-Rating Anxiety ScaleMonths 0, 3, 6, 12Activities of Daily LivingActivities of Daily Living ScaleMonths 0, 3, 6, 12Sleep HealthPittsburgh Sleep Quality Index; Epworth Sleepiness Scale; Athens Insomnia ScaleMonths 0, 3, 6, 12Digestive FunctionOlfaction, taste, mastication, oral health, gastrointestinal function, constipationMonths 0, 3, 6, 12ExaminationsPhysical ExaminationHeight (knee height), weight, waist/hip circumference, blood pressure, pulseMonths 0, 3, 6, 12Body composition (InBody 770), bone mineral density (OSTEO KJ3000, Kejin)Functional ExaminationMuscle function, motor capabilityMonths 0, 3, 6, 12Head Magnetic Resonance Imaging (pioneer 3.0T scanner, Signa Pioneer)Months 0, 12Visual (macular pigment densitometer, MPSII, Electron technology; Non-mydriatic CANON CR-2AF Retinal Image Camera), olfactory (ODOFIN Sniffin Sticks), auditory (hearWHO)Months 0, 3, 6, 12Facial aging assessment (Meitu eve MS20A15, https://eve.meitu.com/)Months 0, 3, 6, 12Wearable DevicesContinuous Glucose Monitor (Abbott Laboratories, Chicago, IL, USA);Activity tracker (Huawei Band 9)Months 0, 3, 6, 12 Table 2Data collection content during the intervention periodCategoryItemContentsTimepointGeneral QuestionnaireDemographic InformationAge, Sex, Marital Status, Occupation, Date of BirthMonths 0, 3, 6, 12Socioeconomic StatusYears of Education, Personal and Household IncomeMonths 0, 3, 6, 12Disease HistoryHistory of hyperglycemia, hyperlipidemia, hypertension, etc.Months 0, 3, 6, 12Smoking and Alcohol UseCurrent smoking/drinking status (quantity, type, etc.)Months 0, 3, 6, 12Physical Activity LevelOccupational and leisure-time physical activity/exerciseMonths 0, 3, 6, 12Dietary QuestionnaireDietary Survey3-day 24-hour Dietary Recall; Food Frequency QuestionnaireMonths 0, 3, 6, 12 for Dietary Recall; Months 0, 12 for Food Frequency QuestionnaireOther QuestionnairesCognitive FunctionMontreal Cognitive AssessmentMonths 0, 3, 6, 12Mental StatusGeriatric Depression Scale; Self-Rating Anxiety ScaleMonths 0, 3, 6, 12Activities of Daily LivingActivities of Daily Living ScaleMonths 0, 3, 6, 12Sleep HealthPittsburgh Sleep Quality Index; Epworth Sleepiness Scale; Athens Insomnia ScaleMonths 0, 3, 6, 12Digestive FunctionOlfaction, taste, mastication, oral health, gastrointestinal function, constipationMonths 0, 3, 6, 12ExaminationsPhysical ExaminationHeight (knee height), weight, waist/hip circumference, blood pressure, pulseMonths 0, 3, 6, 12Body composition (InBody 770), bone mineral density (OSTEO KJ3000, Kejin)Functional ExaminationMuscle function, motor capabilityMonths 0, 3, 6, 12Head Magnetic Resonance Imaging (pioneer 3.0T scanner, Signa Pioneer)Months 0, 12Visual (macular pigment densitometer, MPSII, Electron technology; Non-mydriatic CANON CR-2AF Retinal Image Camera), olfactory (ODOFIN Sniffin Sticks), auditory (hearWHO)Months 0, 3, 6, 12Facial aging assessment (Meitu eve MS20A15, https://eve.meitu.com/)Months 0, 3, 6, 12Wearable DevicesContinuous Glucose Monitor (Abbott Laboratories, Chicago, IL, USA);Activity tracker (Huawei Band 9)Months 0, 3, 6, 12 Data collection content during the intervention period Continuous Glucose Monitor (Abbott Laboratories, Chicago, IL, USA); Activity tracker (Huawei Band 9) Questionnaires A series of comprehensive questionnaires were administered to collect information on demographic characteristics, medical history, medication use, health status, lifestyle factors, and dietary intakes (Table 2). Baseline dietary intake was assessed with a 74-item food frequency questionnaire modified from the food frequency questionnaire used in the national survey on the Status of Nutrition and Health of the Chinese People in 2002 [35]. Meanwhile, three consecutive 24-hour dietary recalls, including two weekdays and one weekend day, will be performed at each round of survey. The nutrient calculation will be performed using the China Food Composition Table [36], with the Hospital Nutrition Therapy System (Shanghai WINCOME Technology, Shanghai, China) serving as the calculation platform. Examinations Anthropometric measures, functional measurement, brain magnetic resonance imaging, and facial aging test will be conducted to collect detailed health related status (Table 2). Sample collection Participants were instructed to collect urine and fecal samples at the morning of the physical examination day and bring them to the research site. Fasting blood samples were collected by professional nurses after fasting overnight. All the samples were labeled with a unique code and stored in a cryogenic storage box until transported to the laboratory. Collected samples were stored at − 80 °C until all samples are gathered. See Table 3 for the contents of biomarkers that will be measured. Table 3Measurement of biomarkersSampleContentsBiomarkers/AnalytesTimepoint*BloodFasting venous blood (12 mL)Clinical: Glucose, lipids, ALT/AST/GGT, creatinine, etc.Nutritional: Iron, Ca, vit A/B12/D, etc.Cognitive: Aβ, p-tau, etc.Metabolomics: Sugars/lipids/amino acids, etc.Genomics: APOE ε4, SNPs, etc.Months 0, 3, 6, 12UrineMorning fasting urine (10 mL)Cortisol, melatonin, environmental pollutants (phthalates), metabolomic profiles, etc.Months 0, 3, 6, 12StoolSelf-collected (researcher-preserved)Microbiome: Taxonomic/functional profilingMetabolomics: SCFAs, bile acidsMonths 0, 3, 6, 12*Timepoints for sample collection, biomarker test and analytes will be conducted after intervention period Table 3Measurement of biomarkersSampleContentsBiomarkers/AnalytesTimepoint*BloodFasting venous blood (12 mL)Clinical: Glucose, lipids, ALT/AST/GGT, creatinine, etc.Nutritional: Iron, Ca, vit A/B12/D, etc.Cognitive: Aβ, p-tau, etc.Metabolomics: Sugars/lipids/amino acids, etc.Genomics: APOE ε4, SNPs, etc.Months 0, 3, 6, 12UrineMorning fasting urine (10 mL)Cortisol, melatonin, environmental pollutants (phthalates), metabolomic profiles, etc.Months 0, 3, 6, 12StoolSelf-collected (researcher-preserved)Microbiome: Taxonomic/functional profilingMetabolomics: SCFAs, bile acidsMonths 0, 3, 6, 12*Timepoints for sample collection, biomarker test and analytes will be conducted after intervention period Measurement of biomarkers Clinical: Glucose, lipids, ALT/AST/GGT, creatinine, etc. Nutritional: Iron, Ca, vit A/B12/D, etc. Cognitive: Aβ, p-tau, etc. Metabolomics: Sugars/lipids/amino acids, etc. Genomics: APOE ε4, SNPs, etc. Microbiome: Taxonomic/functional profiling Metabolomics: SCFAs, bile acids *Timepoints for sample collection, biomarker test and analytes will be conducted after intervention period Measurement of compliance To ensure volunteer safety and trial integrity, new intervention powder was distributed and unused powder was collected every two weeks, accompanied by detailed recording of adverse events, medication, lifestyle, and physical activity changes. Comprehensive assessments and sample collections occurred at baseline, 3, 6, and 12 months. Compliance was primarily calculated using the formula: [(Powder Distributed - Powder Returned)/Theoretical Intake] × 100%, with a threshold of ≥ 80% required for inclusion in per-protocol analyses; stool sample results provided supplementary verification. Adherence was reinforced through a personalized reminder approach developed during supervised visits based on individual schedules, coupled with unscheduled phone follow-ups to address any issues related to study product consumption, monitor adherence, and respond to protocol-related queries, while maintaining rapport with participants. Data management and quality control Data management mainly involves three parties: researchers, the quality control team, and data administrators. During the collection phase, data management forms including participant recruitment, follow-up information, and quality control check, are pre-designed. Researchers have undergone training before the trial begins. At each visit, researchers are responsible for correctly conducting surveys, accurately filling in the forms with participants information, and completing quality control check sheets. The quality control team reviewed data collection and storage procedures biweekly throughout the trial. Errors will be corrected by the researchers and signed off, and adverse events are verified and then assessed and handled by the principal investigator. After the study ends, the collected data and biological samples will be stored in the School of Public Health, Fudan University, for 10 years before destruction. The principal investigator of this trial will be responsible for the data retention, which is only accessible to the specified members of the research team to ensure the security and confidentiality of the database and all materials. Statistical analysis Participant characteristics will be summarized using means ± standard deviations for normally distributed continuous variables, medians with interquartile ranges [25th, 75th percentiles] for skewed variables, and frequencies (percentages) for categorical variables, with baseline group comparisons performed via independent t-tests or χ² tests as appropriate. Primary intention-to-treat analyses will include all randomized participants, supplemented by per-protocol sensitivity analyses restricted to participants achieving ≥ 80% intervention compliance. Two sensitivity analysis will also be conducted: (1) Missing data will be addressed through last observation carried forward, with robustness verified using multiple imputation by chained equations; (2) Worst-case scenario sensitivity analyses. Longitudinal outcomes will be analyzed using linear mixed-effects models with unstructured covariance matrices. For the primary outcome, MoCA score, the model will incorporate random participant intercepts and fixed effects for intervention group, categorical time points (baseline, 3, 6, and 12 months), group-by-time interaction, along with adjustment for the following prespecified covariates: age, sex, education level, baseline MoCA score, BMI, APOE ε4 carrier status, smoking history, physical activity (IPAQ-derived MET-minutes/week), alcohol consumption, and total energy intake. Analysis of secondary and exploratory outcomes will follow a similar approach, with model specifications adapted as appropriate to the type of outcome variable. For microbiome analyses, community richness and diversity of gut microbiota were estimated by Chao1 and Shannon indices. PCoA based on Bray-Curtis distance and PERMANOVA (10,000 permutations) was employed to compare global microbiota composition. Enterotypes were determined using the abundances of genera by Jensen‒Shannon distance (using the vegan and ape packages in R version 4.3.1). Causal mediation testing of microbial species and enterotypes on cognitive outcomes were conducted through structural equation modeling. All analyses will maintain two-tailed significance thresholds at p < 0.05 except whenever necessary multiple comparisons were applied using Benjamini-Hochberg method. Discussion This triple-blinded, placebo-controlled RCT will systematically evaluates the multidimensional effects of a 12-month probiotic intervention on cognitive outcomes and ageing-related features such as gastrointestinal health, sleep, sensory function, and body composition in middle-old aged adults with MCI. To our knowledge, this is one of the most comprehensive RCTs to date, distinguished by its triple-blind design, 12-month intervention period which was relatively longer than most probiotic trials in MCI, and the integration of multidimensional aging-related biomarkers across cognitive, sensory, metabolic, and structural domains to concurrently assess neurocognitive and systematic aging biomarkers in MCI populations with well-designed procedures. To date, numerous studies have explored the association between gut microbiota and cognitive function, postulating various underlying mechanisms [37]. A compromised intestinal barrier allows gut microbes and their metabolites (e.g. isoamylamine) [38] to enter the circulation [39], which may influence microglial cells [40] and astrocyte activation [41], facilitating the Aβ deposition [42]. Moreover, pathogens could damage the gut barrier and increase blood-brain barrier permeability, allowing bacteria-derived Aβ to invade the central nervous system [43]. Therefore, an intact intestinal barrier is crucial for brain health and probiotics could help restore gut barrier function, reduce Aβ plaque formation [44], and reduce oxidative stress in the brain [45]. Previous studies have explored potential link between sleep health and cognitive function, suggesting it to be a modifiable risk factor for cognitive performance. Dysbiosis in gut microbiota and poor sleep quality could synergistically accelerate cognitive decline [16]. Sleep disturbance, such as sleep deprivation, sleep fragmentation, and jet lag, were found to alter the composition of gut microbiota, with increased pathobionts and decreased beneficial bacteria [46]. Probiotics may help modulate immune responses by promoting anti-inflammatory cytokines and influencing gut microbiota composition, thereby impacting stress levels and enhancing sleep quality [47]. Additionally, the microbial production of tryptophan may enhance serotonin and melatonin levels, further promoting healthy sleep [48]. Beyond cognitive function and sleep health, emerging evidence suggests that probiotics may influence multidimensional markers related to aging, such as facial aging, body composition, bone density, and muscle strength through interconnected metabolic and anti-inflammatory pathways. Clinical trials indicated that a probiotic product (a combination of B. lactis Bi-07, B. lactis HN019, L. acidophilus NCFM, L. rhamnosus HN001, and L. paracasei Lpc37) could improve skin barrier function by enhancing hydration and reducing trans-epidermal water loss [19], potentially mitigating facial aging via collagen preservation. For musculoskeletal health, meta-analyses demonstrated that probiotics could increase muscle mass and strength [22], likely through modulating gut-derived inflammatory markers that contribute to sarcopenia. While human studies on bone density remained inconclusive [21], animal models suggested bile acid metabolism as a potential mediator of osteoblast activity [49]. Additionally, probiotic interventions also improve body composition by reducing visceral adiposity and enhancing lipid profiles [20], supporting their role in metabolic regulation. Although there was limited evidence regarding retinal microvascular health and sensory function, the probiotics may have beneficial effects due to their ability to attenuate oxidative stress and systematic inflammation. The current study protocol has several strengths: (1) Exclusion criteria has been carefully selected to minimize confounding factors (e.g., excluding individuals with known neurological disorders or medications affecting cognitive function); (2) Brain MRI, CGM, and activity band, and multi-omics profiling (metagenomics/metabolomics) will be implemented to capture multidimensional biomarkers, building upon established techniques reported in prior mechanistic studies; (3) Relatively long intervention period (12 months) may allow for detection of cumulative cognitive and neurobiological changes that may not be apparent in shorter trials. There are also several limitations. The trial will probably have a selection bias, as participants willing to join trials might be more health-conscious or motivated than the general MCI population, which may limit its generalizability. However, Participants will be recruited from both community clinics and residential community to capture individuals with varying healthcare access and engagement levels. In addition to face-to-face recruitment, we will use public advertisements, social media campaigns, and collaborations with local volunteer organizations to reach less health-motivated populations. Variables associated with health behaviors (e.g., level of physical activity, dietary patterns, and adherence to intervention) will be adjusted as potential confounders. Additionally, certain self-reported measures, like physical activities and sleep quality, may be influenced by recall bias. However, these limitations are inevitable in self-reported measures and are not exclusive to this study. The usage of objective assessments in our trial could help reduce reliance on delayed self-reports. The sample size has been determined based on the primary outcome, which may limit our ability to detect smaller effect sizes in secondary outcomes, which were considered as exploratory outcomes for this trial. Furthermore, the study solely employs MoCA (total score < 26) for defining MCI. Full neuropsychological batteries were not administered due to the community-based setting and the broad range of secondary outcomes. Nevertheless, its high diagnostic accuracy (sensitivity 95.8%, specificity 87.1%) has been validated in elderly Chinese populations [50, 51], and has been widely used in community-based intervention trials [52–54]. In the Chinese MoCA-BJ, delayed recall and verbal fluency tasks have shown high sensitivity (89%) for distinguishing between MCI and AD stages [55]. To minimize educational bias, we apply a one‑point adjustment for participants with fewer than 12 years of education [50]. Furthermore, to complement cognitive assessments and provide additional insights into the neurocognitive effects of probiotic supplementation, we will explore relevant cognitive biomarkers from subsequent bio-sample measurements. In conclusion, this study will provide some new insights on how probiotic supplementation can impact cognitive function and other aging-related physiological functions in MCI adults and explore the potential underlying mechanisms. The findings may inform the development of strategies to delay cognitive decline by modulating the gut-brain axis in this high-risk population.