This is a single-center, randomized, double-dummy, triple-blind, placebo-controlled, three-arm parallel-group superiority trial. The study aims to compare the efficacy and safety of Fecal Microbiota Transplantation (FMT) administered via two different invasive routes-nasojejunal tube (NJT) and colonoscopy-versus a placebo control in children aged 3-16 years with moderate-to-severe Autism Spectrum Disorder (ASD). A total of 75 participants will be randomized in a 1:1:1 ratio to receive either active FMT via NJT with sham colonoscopy, active FMT via colonoscopy with sham NJT, or placebo via both routes. All participants will continue their stable behavioral interventions throughout the study. The primary outcome is the change from baseline to Week 24 in the total score of the Childhood Autism Rating Scale (CARS). Secondary outcomes include changes in other behavioral and gastrointestinal symptom scores, gut microbiota profiling, and safety assessments over 48 weeks.

Background:

Autism Spectrum Disorder (ASD) is a neurodevelopmental disorder often accompanied by gastrointestinal (GI) symptoms and gut microbiota dysbiosis. Fecal Microbiota Transplantation (FMT) has shown promise in modulating the gut-brain axis and improving both behavioral and GI symptoms in preliminary ASD studies. However, the optimal route of FMT administration remains unclear, and high-quality comparative evidence is lacking.

Objectives:

Primary: To compare the efficacy of FMT delivered via NJT versus colonoscopy versus placebo in improving social interaction and communication, as measured by the change in CARS total score from baseline to Week 24.

Secondary: To evaluate effects on social responsiveness (SRS), aberrant behaviors (ABC), sensory processing (SSP), sleep quality (CSHQ), GI symptoms (GSRS), gut microbiota engraftment dynamics, and safety/tolerability.

Methods:

Design: Single-center, randomized, double-dummy, triple-blind, placebo-controlled, three-arm parallel-group trial.

Participants: 75 children aged 3-16 years with DSM-5-confirmed moderate-to-severe ASD (CARS ≥36) and stable behavioral intervention.

Interventions:

Group 1 (FMT-NJT): Active FMT via NJT + sham colonoscopy. Group 2 (FMT-C): Active FMT via colonoscopy with placement of a transendoscopic enteral tube (TET) in the cecum for subsequent infusions + sham NJT.

Group 3 (Control): Placebo via both NJT and colonoscopy (sham procedures). Dosage: 5 mL/kg (max 100 mL) per infusion, administered every other day for three sessions.

Blinding: Triple-blind-participants/guardians, outcome assessors, and data analysts are blinded. An independent pharmacy unit prepares identical active and placebo preparations.

Assessments: Behavioral scales (CARS-2, SRS-2, ABC, SSP, CSHQ), GI symptoms (GSRS), stool metagenomics, and safety monitoring at baseline, Weeks 2, 6, 12, 24, and 48.

Sample Size: 25 per group (total N=75), calculated to detect a 2.5-point difference in CARS change with 80% power, accounting for 15% dropout.

Randomization: Centralized block randomization stratified by age and baseline CARS severity.

Statistical Analysis: ANCOVA for primary outcome with baseline adjustment; mixed models for repeated measures; descriptive and inferential methods for secondary and safety outcomes.

Outcomes:

Primary: Change in CARS-2 total score from baseline to Week 24. Secondary: Changes in SRS-2, ABC, GSRS, SSP, CSHQ scores; microbiota composition/function; incidence and severity of adverse events (CTCAE v5.0).

Significance: This trial will provide high-level evidence on whether the therapeutic effect of FMT in ASD depends on the gastrointestinal delivery site. The novel TET-based protocol for repeated cecal delivery allows for a rigorous comparison of microbial engraftment in the lower versus upper GI tract. The results will guide the optimization of microbiota-targeted therapies for ASD and other conditions linked to the gut-brain axis.

Inclusion Criteria:

• Aged 3-16 years.
• Diagnosed with ASD according to the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5), with a Childhood Autism Rating Scale (CARS) total score ≥36 (moderate-to-severe autism).
• Legal guardians fully comprehend the trial's informed consent and voluntarily provide written consent.
• Compliance with follow-up visits, examinations, and specimen collection.
• No probiotic supplements consumed within the preceding 3 months.

Exclusion Criteria:

• Use of probiotics or prebiotics within 3 months prior to enrollment.
• Antibiotic usage within 1 month prior to enrollment.
• Presence of fever (axillary temperature ≥37.5°C).
• Dependency on tube feeding.
• Severe gastrointestinal conditions requiring immediate intervention (e.g., life-threatening intestinal obstruction, perforation, hemorrhage, ulcerative colitis, Crohn's disease, celiac disease, or eosinophilic esophagitis).
• Diagnosis of severe malnutrition, underweight status (BMI-for-age <3rd percentile), or severe immunodeficiency disorders.
• History of severe allergic reactions (e.g., anaphylaxis).
• Monogenic disorders (e.g., Fragile X syndrome, Rett syndrome).
• Comorbid psychiatric diagnoses, including depression, developmental speech/language disorders, intellectual disability, attention-deficit/hyperactivity disorder (ADHD), selective mutism, reactive attachment disorder, or childhood schizophrenia.

Date for public release of raw data: within no later than 3 years after the publication of research results.

To download data, contact must be made with the researcher.