Study design
We are planning a monocentric, double-blind, sham-controlled, randomized clinical trial to evaluate the efficacy and safety of rTMS in in children and adolescents diagnosed with ASD. The study will involve 40 patients randomized into active rTMS and sham stimulation groups. The duration of the study will be 36 months. Participants will be assessed before rTMS (T0), immediately post-treatment (T1), and at a 1-month follow-up (T2) (Fig. 1). Assessments will evaluate neuropsychological function, mainly executive functions; the severity of ASD clinical symptoms; and safety and tolerability. Additionally, biological samples, including urine and blood, will be collected at each assessment to measure biomarker changes.
Ethical issues
A participant information sheet will be provided, and informed consent will be obtained from the parents or caregivers of study participants (see Supplementary File 1). The study protocol has been reviewed and approved by the Independent Ethics Committee of Policlinico “Riuniti” of Foggia (reference number 50/CE/2023). The trial has been registered with ClinicalTrials.gov under the identifier NCT06069323. The study will be conducted in accordance with the Declaration of Helsinki, Good Clinical Practice, and applicable regulatory requirements. To ensure appropriate oversight, key study staff will meet monthly to review trial conduct, participant recruitment, protocol adherence, data quality, and any adverse events. Given the low-risk nature of the rTMS intervention, a formal independent Data Monitoring Committee was not established. Instead, trial oversight, including safety and data integrity, will be managed internally by the study team. Any serious adverse events or protocol deviations will be reported to the Ethics Committee in accordance with regulatory requirements. No formal interim analyses or independent audits are planned. Any significant amendments to the protocol will be reviewed by the Principal Investigator and submitted to the Ethics Committee for approval. Once approved, all relevant members of the study team will be informed, and the revised protocol will be stored with the local study documentation. Any deviations from the approved protocol will be documented using a breach report form, and the Clinical Trial Registry will be updated as required.
Study setting
Participants will be recruited through the Child and Adolescent Neuropsychiatry Unit at the General Hospital “Riuniti of Foggia”, community health clinics, and family associations supporting individuals with neurodevelopmental disorders. Initially, interested participants will receive detailed study information and undergo preliminary screening to assess eligibility based on age, confirmed ASD diagnosis, and absence of a personal history of seizures. Following this initial screening, eligible participants will be invited for an in-person assessment to further verify inclusion criteria. Those meeting the criteria will subsequently receive neurostimulation treatments as outpatients at the Department of Clinical and Experimental Medicine, University of Foggia. A detailed timeline of data collection is shown in Table 1. We used the SPIRIT checklist when writing our protocol (see Supplementary File 2) [27].
Eligibility criteria, sample size, and data collection
Participants will be eligible for inclusion in the study if they are between 7 and 18 years old and have a confirmed diagnosis of ASD based on DSM-5 criteria. Exclusion criteria for this study include any history of seizures, epilepsy, or repeated febrile seizures, as well as any severe or traumatic brain injury. Participants with comorbid neurological or genetic conditions that impact brain function or structure, such as brain tumors, fragile X syndrome, or tuberous sclerosis, will also be excluded. Additionally, those with known endocrine, cardiovascular, pulmonary, liver, kidney, or other significant medical diseases, or with any unstable medical condition, will not be eligible to participate. Participants will also be excluded if they are on an unstable medication regimen or using medications contraindicated for TMS. Vision or auditory impairments that could interfere with study participation will preclude eligibility. Participants will also be excluded if they demonstrate significant epileptiform activity on electroencephalogram (EEG), such as seizures or continuous epileptiform discharges. Furthermore, individuals with psychosis disorder and diagnosed chronic or acute inflammation or infection, or those unable to provide informed consent, will not be eligible for the study. To detect significant differences in outcomes between the active and sham groups with 80% statistical power and a significance level (α) of 0.05, the sample size calculation indicated that 20 participants per group (n = 40 total) would be required. We will use REDCap (Research Electronic Data Capture) as our clinical data management system. Personal information will be stored separately from study data, secured on password-protected systems, and pseudo-anonymized with unique numeric codes accessible only to authorized personnel.
Blinding and randomization
Participants, care providers, and clinical raters will be blinded to treatment assignment. Only the clinician who generates the treatment allocation and delivers the pulses will remain unblinded, without participating in any other study activities. Data analysis will be conducted independently by two statisticians who are not otherwise involved in the trial procedures. A stratified randomization approach with permuted blocks will be used. Clinical severity will be classified based on scores from the Autism Diagnostic Observation Schedule-2 (ADOS-2) and the Childhood Autism Rating Scale Second Edition (CARS-2). Participants will be stratified into mild-to-moderate and severe groups before randomization to ensure balanced allocation between treatment arms. This approach is intended to minimize baseline differences in symptom severity across groups. Randomization will be performed using a computer-generated allocation sequence (a function available in SAS software), with permuted blocks employed to prevent predictability of group assignments and maintain balance within each stratum.
Neuropsychological assessment and primary outcomes
A comprehensive neuropsychological assessment will be conducted using a battery of tests to confirm the ASD diagnosis and evaluate clinical severity. These assessments will be performed at baseline, post-treatment, and at a 1-month follow-up, representing the primary outcomes of the study. The ADOS-2 and Autism Diagnostic Interview-Revised (ADI-R) will be administered to evaluate social interaction, communication, play, and the imaginative use of materials across different age groups [28, 29]. The Wechsler Intelligence Scale for Children, Fourth Edition (WISC-IV) will be used to evaluate intellectual functioning in verbal children, whereas non-verbal children will be assessed with the Leiter International Performance Scale-Revised (Leiter-R) [30, 31]. To measure neuropsychological functions, particularly executive functions, the NEPSY Second Edition (NEPSY-II) will be used. The Movement Assessment Battery for Children Second Edition (MABC-2) will assess motor skills in everyday activities [32]. The CARS2 will be administered to evaluate autistic symptoms, and the Vineland Adaptive Behavior Scales Second Edition (Vineland-II) will assess a range of adaptive behaviors [33,34,35]. Additionally, parents will complete the Conners Third Edition (Conners 3) to report attention-deficit/hyperactivity disorder (ADHD) symptoms in their children, the Child Behavior Checklist for Ages 6–18 (CBCL/6–18) to identify behavioral and emotional problems, and the Social Communication Questionnaire (SCQ), a tool used to evaluate communication skills and social functioning [36,37,38]. The Children’s Depression Inventory 2 for parents (CDI-2) and the Multidimensional Anxiety Scale for Children (MASC) will also be used to assess depressive and anxiety symptoms [39, 40]. A detailed patient and medication history will be collected to assess any potential impact on the efficacy of rTMS.
Electroencephalogram recording
A standard EEG, following the international 10–20 system, will be recorded at each study time point: at baseline (T0), after rTMS treatment completion (T1), and at a 1-month post-treatment follow-up (T2). This procedure will identify pre-neurostimulation epileptic discharges that may exclude participants from the study and monitor any electrical changes induced by rTMS.
Intervention and adherence to protocol
The intervention will consist of 18 sessions of 2 Hz rTMS over 9 weeks, targeting the DLPFC. Stimulation intensity will be set at 90% of the motor threshold, delivering 180 pulses per session consisting of 9 trains of 20 pulses each, with a 20-s inter-train interval. The selection of 2 Hz frequency and 180 pulses per session was based on prior studies demonstrating that low-frequency stimulation over the DLPFC can reduce gamma activity and enhance executive function in individuals with ASD [4, 5, 19]. These parameters were selected to ensure both safety and potential clinical benefit in a pediatric population, consistent with previous trials that reported good tolerability and modulation of cortical excitability using low-frequency, low-dose protocols [4, 18]. The first six sessions will target the left DLPFC, the next six the right DLPFC, and the final six both hemispheres. rTMS will be administered using a Magstim R2 stimulator (Magstim, Whitland, UK) with a 70-mm figure-eight coil positioned at a 45° angle from the midline. Anatomical landmarks corresponding to EEG sites F3 and F4 (10–20 system) will be used to accurately target the DLPFC and to minimize discomfort in pediatric patients with ASD. Motor thresholds for each hemisphere will be established by incrementally raising machine output by 5% until a visible twitch in the first dorsal interosseous (FDI) muscle is observed in 2 out of 3 trials [41]. A sham group will undergo identical procedures without magnetic field application. The sham stimulation will replicate the auditory and tactile sensations of active rTMS by using the same coil orientation and clicking sounds. The coil will be positioned against the scalp at the same angle and location, and the stimulator will be activated to produce the characteristic clicking noise and slight vibration. This is to ensure that the sensory experience remains comparable between active and sham conditions, thereby supporting effective blinding. Throughout and immediately after each session, participants will be monitored for well-being, with any reported side effects documented and reviewed post-study. All sessions will be conducted under standardized conditions, with participants seated comfortably in an armchair in a quiet room, and their elbows positioned at a 90° flexion angle.
All rTMS sessions will be scheduled at consistent times and on the same days to establish a routine for participants and their parents or caregivers. Reminders via phone calls or text messages will be sent to parents or caregivers before each session. Missed sessions will be promptly rescheduled within the same week whenever possible. Study staff will educate families on the importance of session attendance and protocol compliance. Reasons for missed sessions will be documented, and efforts will be made to re-engage participants. The rTMS clinician will complete a checklist at each session to confirm that all protocol steps (e.g., stimulation parameters, coil positioning) are followed.
Biochemical measures and secondary outcomes
Biochemical measures will be assessed as secondary outcomes to provide a more comprehensive understanding of the effects of rTMS on ASD. Tryptophan metabolites, including 3-hydroxykynurenine (3-HKYN), KYNA, and QUIN, will be measured in urine using
high-performance liquid chromatography with electrochemical detection (HPLC-ECD) (Ultimate ECD, Dionex Scientific, Milan, Italy) [42]. HPLC-ECD will be also used to measure systemic levels of neurotransmitters, such as glutamate, GABA, serotonin, and dopamine, in urine samples [43]. To measure circulating BDNF, serum samples will be collected, and an enzyme-linked immunosorbent assay (ELISA) (DBA Italia, Segrate, Italy) will be performed [44]. Additionally, BDNF gene polymorphism (Val66Met) will be genotyped using Polymerase Chain Reaction-Restriction Fragment Length Polymorphism (PCR–RFLP). This approach amplifies the BDNF gene region containing the Val66Met variant, followed by restriction enzyme digestion to distinguish the Val and Met alleles based on fragment length analysis [45]. Inflammatory mediators, including IL-6, IL-10, IL-1β, TNF-α, and CRP, will be measured in serum using ELISA [46].
Follow-up
A follow-up visit will be scheduled 1 month after the final rTMS session to evaluate the persistence of rTMS effects and the need for possible booster sessions. Accordingly, neuropsychological assessments and biochemical sampling will be performed again at the follow-up visit. To enhance the response rate, the study coordinator will request contact information from both parents.
Safety
Before the enrollment, participants will undergo comprehensive screening to confirm they meet eligibility and rTMS safety criteria. Additionally, a qualified physician will review medical history and perform a general physical examination to confirm the absence of risk factors. After enrollment, a standard clinical EEG, reviewed by a neurologist, will be conducted to exclude participants with epileptiform discharges. To ensure safety, rTMS will be administered by a trained neuropsychiatrist equipped to promptly manage any seizures or adverse events. A stimulation intensity of 90% of the motor threshold will be used for ASD participants to minimize seizure risk. Before each rTMS session, participants will be asked about their current health status and any adverse events experienced since the previous session. All adverse events will be documented and reported. In the case of a serious adverse event, treatment will be suspended.
Statistical methods
Statistical analyses will be performed using SPSS (IBM Corp., Armonk, NY). Primary analyses will focus on evaluating differences between the active and sham rTMS groups across three time points: baseline (T0), immediately post-treatment (T1), and 1-month follow-up (T2). Changes in neuropsychological and clinical measures across time points will be assessed using repeated-measures analysis of variance (ANOVA), with group (active vs. sham) as the between-subjects factor. If assumptions of normality are not met, the Wilcoxon signed-rank test will be applied to evaluate within-group differences over time, while the Mann–Whitney U test will be used to compare between-group differences at each time point. Biomarker levels will be compared using generalized linear models to account for potential covariates, including age and baseline severity. Chi-square tests will evaluate categorical variables, and t-tests will be applied for continuous variables where applicable. Spearman’s rank or Pearson correlation coefficients, depending on data distribution, will be employed to examine relationships between neuropsychological scores and biochemical measures. All statistical tests will be two-sided, with p-values below 0.05 considered statistically significant.