Intermittent theta burst stimulation over the posterior superior temporal sulcus for children with autism spectrum disorder: A 4-week randomized blinded controlled trial followed by another 4-week open-label intervention

Abstract

The posterior superior temporal sulcus is a potential therapeutic target of brain stimulation for autism spectrum disorder. We conducted a 4-week randomized, single-blind parallel sham-controlled trial, followed by additional 4-week open-label intervention to evaluate the feasibility and efficacy regarding intermittent theta burst stimulation over the bilateral posterior superior temporal sulcus in autism spectrum disorder. In total, 78 intellectually able children and adolescents were randomized to the active (n = 40) and sham groups (n = 38). During the first 4 weeks, the active group received two-session/week intermittent theta burst stimulation, whereas the sham group received the same number of sham stimulation. After unblinding, both groups received eight-session real stimulation over the additional 4 weeks. In total, 91% participants completed the protocol with mild and transitory side-effects. There was no significant group-by-time interaction for active versus sham group on clinical symptoms and social cognitive performances in the first 4 weeks. The within-group analysis revealed 8 weeks (including a 4-week blind trial and a 4-week open-label intervention) of intermittent theta burst stimulation achieved greater efficacy than 4-week interventions. Participants with higher intelligence, better social cognitive performances, alongside less attention-deficit hyperactivity disorder severity at baseline, were more likely to be responders. Our study demonstrated the feasibility of long-term intermittent theta burst stimulation over the posterior superior temporal sulcus in children and adolescents with autism spectrum disorder. However, the findings from the first 4-week blind trial do not support the therapeutic efficacy of intermittent theta burst stimulation over the posterior superior temporal sulcus on the clinical symptoms and cognitive performance of social impairment, given the current stimulation protocol. The exploratory analyses suggest that the therapeutic efficacy might be moderated by several individual characteristics and more intermittent theta burst stimulation sessions.

Lay abstract

Intermittent theta burst stimulation is a varied form of repetitive transcranial magnetic non-invasive brain stimulation technique used to treat several neurological and psychiatric disorders. Its feasibility and therapeutic effects on the bilateral posterior superior temporal sulcus in children with autism are unknown. We conducted a single-blind, sham-controlled parallel randomized clinical trial in a hitherto largest sample of intellectually able children with autism (N = 78). Participants randomized to the active group received two-session/week intermittent theta burst stimulation for continuous 8 weeks. Those in the sham group received two-session/week sham stimulations in the first 4 weeks and then active intervention for the following 4 weeks after unblinding. First, we found that continuous 8-week intermittent theta burst stimulation on the bilateral posterior superior temporal sulcus in children with autism is safe and tolerable. Second, we found that 8-week intermittent theta burst stimulation produced greater therapeutic efficacy, although we did not find any significant effects of 4-week intermittent theta burst stimulation on core symptoms and social cognitive performances in autism. Further analysis revealed that participants with higher intelligence and better social cognitive performance, alongside less attention-deficit hyperactivity disorder severity at baseline, were more likely to be responders. This study identified that the factors contribute to responders and the results suggest that longer courses of non-invasive brain stimulation may be needed to produce therapeutic benefits in autism, with consideration of heterogeneous responses.

Keywords

autism spectrum disorder children and adolescents intervention posterior superior temporal sulcus theta burst stimulation

Autism spectrum disorder (ASD) is a neurodevelopmental condition affecting ~1% of children, with impairments in psychosocial functioning throughout life (Lord et al., 2020). But, there is no effective biomedical intervention targeting core autistic symptoms. Emerging evidence suggests that repetitive transcranial magnetic stimulation (rTMS) may be a potential option (Barahona-Correa et al., 2018; Oberman et al., 2016). rTMS delivers patterns of multiple focused electromagnetic pulses to induce durable changes in the activity of the stimulated brain region. rTMS can be used to investigate brain function and modify neuroplasticity (Klomjai et al., 2015) and is effective in treating several neurological and psychiatric disorders, especially treatment-resistant major depressive disorder (Berlim et al., 2013). However, the conventional rTMS is time-consuming (37.5 min/one session), limiting the capacity and accessibility (O’Reardon et al., 2007). Theta burst stimulation (TBS), a modified delivery of rTMS, mimics endogenous theta rhythms and could improve induction of synaptic long-term potentiation (Huang et al., 2005). Intermittent TBS (iTBS), which delivers TBS pulses for 2 s every 10 s (Huang et al., 2005), can facilitate cortical excitability and has a comparable performance to conventional rTMS in the treatment for depression (Blumberger et al., 2018), but with shorter stimulation duration (3 min) and lower total TMS pulses and intensity (Huang et al., 2005; Schwippel et al., 2019).

A recent meta-analysis (Barahona-Correa et al., 2018) of existing studies investigating efficacy of rTMS in ASD supports its effects on some domains of ASD-associated deficits. Specifically, rTMS treatment results in a significant improvement in repetitive behaviors and executive function, with a medium effect size. Effects on social interaction deficits are less robust with a small-to-medium effect size. Nonetheless, there is remarkable heterogeneity in designs across studies. Sokhadze et al. conducted a series of open-label waitlist-controlled trials on the dorsolateral prefrontal cortex (DLPFC), which showed that this protocol might improve the core symptoms and executive function in children and youth with ASD (Baruth et al., 2010; Casanova et al., 2012, 2014; E Sokhadze et al., 2010; EM Sokhadze et al., 2009, 2012, 2014, 2018). However, few (7 out of 21) clinical trials were conducted in a randomized, blinded, and sham-controlled approach (Ameis et al., 2020; Anninos et al., 2016; Enticott et al., 2012, 2014; Fecteau et al., 2011; Ni et al., 2017; Panerai et al., 2014). Among these low-risk-of-bias studies, sample sizes were not larger than 20 in each treatment group, limiting well-powered investigations. Most of these randomized controlled trials (RCTs) had treatment courses ⩽2 weeks, limiting the potential of advanced neuroplasticity. The stimulated areas also varied across studies (Ameis et al., 2020; Anninos et al., 2016; Enticott et al., 2012, 2014; Fecteau et al., 2011; Ni et al., 2017; Panerai et al., 2014). Among one of three promising target regions for ASD based on experts’ consensus (Cole et al., 2019), unlike the DLPFC, the posterior superior temporal sulcus (pSTS) receives limited attention in rTMS studies (Ni et al., 2017).

A recent review suggests that, in addition to processing social perception and action observation, the pSTS also plays a vital role in integrating this social relevant information to the computational process of the theory of mind (D. Y. Yang et al., 2015). Based on the resting-state fMRI literature, the pSTS is also one of the regions within the default-mode network (Power et al., 2011), which largely corresponds to the so-called social brain network (Padmanabhan et al., 2017). Interference with the right pSTS activity by inhibitory TMS leads to transitory disruption in the behavior of orienting toward the eyes (Saitovitch et al., 2016). Neuroimaging studies consistently highlight that reduced functional activation and altered neurodevelopment of the morphometry of the pSTS link to social deficits characteristic of ASD (Hotier et al., 2017; DY Yang et al., 2015). The pSTS and temporoparietal junction (TPJ) are synergistically involved in the complex processes of social cognition. Their anatomical locations are so close by that the effect of rTMS targeted on either region may spread to each other. Thus, a consensus statement regarding the potential targets of rTMS in ASD (Cole et al., 2019) used the expression of “pSTS/TPJ” to highlight this critical hub within the posterior part of social brain network. Notably, pSTS/TPJ may be the most targetable. Our pilot study showed that one-session iTBS over the bilateral pSTS improves compulsive behaviors in intellectually able adults with ASD (Ni et al., 2017). The feasibility and efficacy of a multisession treatment course of pSTS stimulation for children and adolescents with ASD remain unknown.

We conducted a randomized single-blind and sham-controlled clinical trial, with a biggest sample size ever published in literature in ASD, to investigate the efficacy on core autistic deficits as well as safety of iTBS over the bilateral pSTS in intellectually able children and adolescents with ASD over a 4-week course. Furthermore, after unblinding, we explored whether longer courses (another 4 weeks) of iTBS interventions could lead to greater benefits. Given excitatory nature of iTBS (Huang et al., 2005) and pSTS hypofunction in ASD (D. Y. Yang et al., 2015), we hypothesized that compared to those in the sham-controlled group, children and adolescents with ASD receiving iTBS over the pSTS would show improvement in social-communication symptoms and social cognition. Longer courses of iTBS intervention would result in greater improvement in social deficits.

Methods

Design

This was a 4-week randomized, parallel, single-blind, and sham-controlled trial, followed by another 4-week open-label intervention, to investigate the feasibility and efficacy of iTBS over the bilateral pSTS in children and adolescents with ASD at Chang Gung Memorial Hospital (CGMH, Linkou, Taiwan). As shown in Figure 1, there were two phases in this RCT, with 1-month follow-up. Specifically, after baseline assessments, participants were randomized to the active or sham group (Phase 1, Baseline–Week 4). Given limited human resources, the investigator (H.-C.N.) and his assistant assessed and delivered rTMS to all participants. Therefore, only participants and their caregivers were blind to the treatment condition (single-blind). Active versus sham iTBS was administered over the pSTS 2 days/week for 4 weeks. Entering Phase 2 (Week 5–8), participants were unblinded, and then active iTBS over the pSTS was administered to all participants 2 days/week for the other 4 weeks. This distinct design allowed for a conventional rigorous RCT of low risk of bias (Phase 1) and concomitantly resolved a potential inequity issue by providing access to iTBS for every participant (Phase 2) (Green, 2008). This also enabled investigating effects of longer treatment courses. Considering the feasibility based on their school activities, participants freely chose two intervention days, which are at least 48 h apart from each other, from Monday to Saturday in the beginning of their interventions. Once they made the choice, the schedule of sessions was fixed throughout the trial.

Figure 1.

Flow diagram.

Clinical assessments were completed in participants within 1 week of the first iTBS session of Phase 1 (Baseline) and Phase 2 (Week 5), following the last iTBS session (Week 8) as well as at follow-up (Week 12). With the same baseline and follow-up assessment schedule, social cognition was specifically measured within 1 h following the last iTBS session of Phases 1 and 2, respectively.

A minimum total sample size of 68 was estimated using G*Power 3.1 (Faul et al., 2009), being powered (90% power and two-sided 5% significance) to detect a standardized effect at 0.4 from within-between interaction of the repeated-measure analysis of variance (ANOVA) model. The estimated effect size was guided using the pool effect on social behavior deficits in existing studies (Barahona-Correa et al., 2018).

Before implementation, this study was approved by the Research Ethics Committee at CGMH (104-9413A) and registered with ClinicalTrials.gov (NCT03621189). The procedures and purpose of the study were explained face-to-face to participants and their parents, who then provided written informed consents.

Participants

We recruited participants, aged 8–17 years, with ASD from the psychiatry outpatient clinic of CGMH, Linkou, Taiwan. Diagnostic and Statistical Manual of Mental Disorders (4th ed.; DSM-IV) autistic disorder or Asperger’s disorder or Diagnostic and Statistical Manual of Mental Disorders (5th ed.; DSM-5) ASD was clinically diagnosed and corroborated using the Autism Diagnosis Objective Schedule (Lord et al., 2000). Exclusion criteria included: full-scale intelligence quotient (FIQ) < 70 based on the Wechsler Intelligence Scale for Children-III or Wechsler Adult Intelligence Scale-III (a cutoff at 16 years), any prior history of major neurological (especially epilepsy) or medical illness, mood and anxiety disorders, schizophrenia, and substance misuse. Participants with co-occurring attention-deficit hyperactivity disorder (ADHD) were included and assessed by experienced child psychiatrist (H.-C.N., Y.-Y.W., and H.-Y.L.). Since there are high psychiatric comorbidities in ASD (70% at least one comorbidity), we intended to design the current protocol to strike a balance between the generalizability, feasibility, as well as the heterogeneity resulted from the psychiatric comorbidities. Therefore, we decided to include those comorbid with ADHD, which is the most common co-occurring condition, has similar neurodevelopmental nature, and might share some etiologies with ASD, but exclude people with the co-occurring conditions which may happen in association with ASD but are not inherent in neurodevelopmental conditions (i.e. these disorders happen chronologically after ASD). Simultaneously, the effect of iTBS over the pSTS on other major psychiatric disorders are unknown, but is well tolerated in people with co-occurring ADHD (Ni et al., 2017). To minimize potential harms and heterogeneity, people with co-occurring mood and anxiety disorders, schizophrenia, and substance misuse were thus excluded from the current trial. All psychotropic medications were continued without change during the trial. All participants had been naïve to any non-invasive brain stimulation treatment.

Intervention

A 70-mm figure-of-eight coil connected to a Magstim Super Rapid² system (Magstim Company, Oxford, UK) was used. Initially, the coil was placed tangentially to the scalp over the contralateral motor cortex with the handle pointing backward. The location of motor “hot-spot” was determined where single-pulse TMS produced the largest motor-evoked potentials (MEPs) from the FDI at rest. We measured the active motor threshold (AMT) as the minimum stimulation intensity needed to elicit MEPs of no less than 200 µV in 5 out of 10 trials during 20% of maximum voluntary contraction of the FDI.

This study administered the iTBS protocol (Huang et al., 2005) as follows: each TBS train was comprised of a burst of three TMS pulses at 50 Hz, at 200 ms intervals, for 10 times. The TBS train was delivered every 10 s for 20 times to have 600 pulses in total for each iTBS course. In each iTBS session, we first delivered two iTBS with a 3-min break over the left pSTS. Five minutes later, we then delivered the other two iTBS over the right pSTS. The intervention pulses in each session were 1200 for each hemisphere (in total, 2400 pulses/session; 38,400 pulses/study). The stimulus intensity of iTBS over the pSTS was 80% of AMT for the active intervention but 60% AMT for the sham intervention. The sham stimulation was still targeted on the bilateral pSTS but delivered with the coil tilted one-wing 90° off the head (Lisanby et al., 2001), which is a valid sham condition commonly used in double- or single-blind, sham-controlled RCT across major psychiatric disorders (Lefaucheur et al., 2020). This one-wing 90° tilted coil sham manipulation is devoid of detectible biological effects (Lisanby et al., 2001) and produces haptic and auditory simulation. No participant assigned in the Sham group at Phase 1 actively disclosed or guessed that he or she received the sham stimulation.

The location of the bilateral pSTS was defined based on the meta-analysis of functional magnetic resonance imaging (MRI) (Van Overwalle & Baetens, 2009). These regions were registered to each individual’s native structural image in the Navigated Brain Stimulation system (Nexstim®, Helsinki, Finland). The details of localization process are reported elsewhere (Ni et al., 2017). Whether the lateralization exist in function of the pSTS remains elusive. Unlike the target site selection in depression, one recent study demonstrated 18 sessions of rTMS applied over bilateral DLPFC produce the most striking positive effects on improving symptoms in children and adolescence with ASD (EM Sokhadze et al., 2018). Furthermore, our prior preliminary study also shows the potential therapeutic effects of bilateral pSTS stimulation in adults with ASD (Ni et al., 2017). We thus followed this principle to set the target site at the bilateral pSTS.

Outcomes

The primary outcomes included the safety profiles as well as the social deficits as measured by the caregiver-rated social responsiveness scale (SRS) (Gau et al., 2013), since theoretically targeting the pSTS is aimed to modulate the activity within the social brain network. Atypical social cognition is often associated with the autistic symptoms. The secondary outcomes thus included two common tasks assessing the social cognition performances of people with ASD (Barahona-Correa et al., 2018), that is, the Reading the Mind in the Eyes test (RMET) (Baron-Cohen et al., 2001; TS Li et al., 2020) and Frith–Happe animations task (Barch et al., 2013; White et al., 2011). Notably, these two tasks, nonetheless, are not specific to the pSTS activities. Given our preliminary beneficial results in compulsory behaviors in adults (Ni et al., 2017), we also adopted the repetitive behavior scale-revised (RBS-R) (Bodfish et al., 2000 Y. C. Yang et al., 2019) as the exploratory outcome. Higher scores on the SRS and RBS-R represent greater severity of the two domains of autistic symptoms. RMET total scores represent how many correct mental states the participant has inferred from the eyes, indicating individual’s theory of mind capability. Frith–Happe animations also tap mentalizing capacity by asking the participant to infer whether interaction intents exist between two triangles. When the participant answered about the presence/absence of interactions, he or she was further asked to select words that best described how these triangles were feeling at the end of each video clip. Participants scored 1 point for correct answers to either presence/absence or exact description of mental states, which are summarized as the total categorical scores and feelings scores, respectively. To enhance a contrast and reduced assessment time, we adopted the revised Frith–Happe animations following those used in the Human Connectome Project (Barch et al., 2013), which only included social and random interactions.

Side-effects were assessed immediately after each session and at 1-month follow-up using open-ended questions inquiring any physical discomfort and then close-ended questions including “pain at application site,” “headache/dizziness,” “tinnitus,” and “anxiety” experienced during and after iTBS

Statistics

Herein, we only reported clinical and cognitive data during RCT. Independent t-test and chi-square test were used to evaluate the difference of baseline characteristics. To simultaneously examine the immediate effect of pSTS versus sham stimulation in different visits, the generalized estimating equation (GEE) model was conducted using the data of Baseline, Week 4, and Week 8. To account for correlations between individuals’ repeated measurements between visits, a working correlation matrix with a first-order autocorrelation was used with the robust estimator of standard error. Treatment, Time, and a two-way interaction (Active vs Sham × Time) effects were modeled in the GEE. The maximum likelihood method in the GEE was used to address the missing values from dropouts. In consideration of high inter-individual variability in the clinical symptoms and social cognitive function at baseline, we also implemented an additional analysis using the symmetrized percentage change (Ayers & Berry, 2006) to investigate the iTBS effects. Because there was significant difference for both groups at baseline in the RMET, all of the analyses in the GEE and symmetrized percentage change were adjusted.

To comprehensively examine effects across time for both conditions, exploratory within-group comparisons were also conducted in the same GEE regardless of whether the main effect was statistically significant. However, when the main effect of GEE is not statistically significant, the significant findings in these exploratory within-group comparisons should be interpreted conservatively. The correction for multiple comparisons was implemented to avoid type-I errors. Five pairwise within-group comparisons (i.e. Baseline vs Week 4, Baseline vs Week 8, Baseline vs Week 12, Week 4 vs Week 8, and Week 8 vs Week 12) were calculated separately in Active (8 week active TBS) and Sham (4 week sham followed by 4 week active TBS) groups.

Considering clinical heterogeneity in ASD, we defined responders based on the Reliable Change Index (RCI) calculated using the SRS total scores. The RCI takes into consideration the false positive outcome that could occur due to random error from repeated measurements alone (Jacobson & Truax, 1991). The RCI represents differences in individual’s scores between before and after intervention, divided by the standard error of the difference of the measure. Responders were those with RCI > 1.64, which corresponds to p < 0.05 with one-sided test (i.e. the null hypothesis was that iTBS would not help with social symptoms). Demographic and clinical features were compared between the responders and non-responders using non-parametric tests.

In addition to the stratified analysis based on responsiveness, we also tested a three-way interaction (Active vs Sham × Time × Modifier) to explore whether iTBS over the pSTS is more beneficial for individuals with ASD with certain characteristics. Effect modifiers included demographics, intelligence quotient, clinical traits, comorbidity, and medication.

Statistical analysis was performed using the SAS Version 9.4 software (SAS Institute, Cary, NC, USA). Considering multiple comparisons including the clinical symptoms and social cognitive function (five measurements), the GEE, within-group analyses, and three-way interaction were considered statistically significant at p value ⩽ 0.01. Despite contention, herein, alpha level in-between 0.01 and 0.05 were considered a nominal significance for hypothesis generating for the future study (Bays, 2019).

Community involvement

There were no community stakeholders involved in the development of research questions, study design, and outcome measurements. Taipei Parents Association of Autism and Foundation for Autistic Children and Adults in Taiwan helped us to disseminate the recruitment notice during the study implementation as well as provided the platform for knowledge translation and dissemination of the current findings to parent groups after the completion of the study. The progress report has been submitted to the funding agency, and the findings have been summarized in a user-friendly language and been feedbacked to the participants and their caregivers.

Results

Feasibility

Participant enrollment occurred between August 2016 and July 2019. Considering the attrition, eighty-three participants were screened, with 78 participants completing the baseline assessments and randomized to Active or Sham groups (n = 40 and 38, respectively; Figure 1). Three participants in the Sham group withdrew before the start of first iTBS session, resulting in 75 participants in the final analysis. Seventy-two remained in the trial until the end of Phase 1 (92% retention) and 71 remained after the Phase 2 (91% retention). At baseline, both groups were comparable in all demographic, clinical, and cognitive characteristics (Table 1), with the exception of the RMET, in which the Active performed worse than the Sham group (p = 0.022).

Table 1.

Demographics and baseline characteristics of participants with autism spectrum disorder.

	pSTS (n = 40)	Sham (n = 35)	p value
Age (mean (SD))	13.0 (2.8)	12.5 (2.9)	0.445
Male (n (%))	35 (88)	30 (86)	0.821
Intelligence quotient (IQ) (mean (SD))
Full intelligence	88.2 (16.7)	90.0 (14.7)	0.633
Verbal comprehension index	88.5 (17.4)	91.9 (17.4)	0.409
Perceptual reasoning index	98.3 (18.8)	92.3 (21.2)	0.204
Working memory index	92.9 (19.1)	91.1 (16.2)	0.669
Processing speed index	80.0 (17.4)	87.1 (21.8)	0.123
Social responsiveness scale (mean (SD))	107.3 (24.0)	102.1 (23.2)	0.348
Repetitive behavior scale-revised (mean (SD))	32.2 (18.9)	33.5 (20.6)	0.784
Eyes test (total correct) (mean (SD))	20.8 (8.0)	24.8 (6.9)	0.022*
Frith–Happe animations task (mean (SD))
Categorization (total correct)	5.5 (2.1)	5.6 (1.9)	0.781
Feelings	3.0 (2.1)	2.7 (2.2)	0.632
Comorbidity with ADHD (n (%))	22 (55)	14 (40)	0.195
Medication (n (%))
Methylphenidate	14 (35)	10 (29)	0.552
Atomoxetine	1 (3)	1 (3)	0.924
Antipsychotics	5 (13)	5(14)	0.821

pSTS: posterior superior temporal sulcus; SD: standard deviation; ADHD: attention-deficit hyperactivity disorder.

p < 0.05.

Side-effects during RCT included local pain during iTBS intervention (10% in the Active at both phases and 29% in the Sham group at Phase 2) as well as headache, dizziness, tinnitus, and anxiety (all incidence <5%; Supplementary Table 1). All of these discomforts disappeared soon after iTBS sessions. No seizure was reported during RCT or at follow-up.

Phase 1 (from Baseline to Week 4, blind)

Summary data for the clinical measures are reported in Table 2. Using the GEE, there was no significant treatment × time interaction for all measures at the end of Phase 1 (Table 3, Figure 2, and Supplementary Figure 1). Besides, there was also no significant group or time effect for all measures at Phase 1. In the analysis of symmetrized percentage change, there was no significant difference for both groups at Phase 1 (Figure 2, Supplementary Table 2, and Supplementary Figure 2). Within-group comparisons showed no significant change in social impairments and cognitive performance at Week 4 relative to Baseline across two groups (Table 2). The effect sizes of within and between group comparisons from Baseline to Week 4 ranged 0.07–0.26 and 0.12–0.23, respectively (Supplementary Table 3).

Table 2.

Behavioral and neuropsychological outcomes before and after intervention.

Mean (SD)	Active–Active group				Sham–Active group
	Baseline	Week 4	Week 8	Week 12	Baseline	Week 4	Week 8	Week 12
Clinical symptoms
SRS, total	107.3 (24.0)	103.4 (25.5)	98.5 (28.7)⁺	97.2 (25.6)^$$	102.1 (23.2)	100.4 (27.1)	95.1 (25.7)	94.7 (27.9)
RBS-R, total	32.2 (18.9)	28.9 (20.1)	25.9 (20.5)⁺	24.3 (17.6)^$$	33.5 (20.6)	28.1 (21.5)	31.5 (23.9)*	29.0 (23.9)
Social cognitive function
Reading the Mind in the Eyes test
Total correct	20.8 (8.0)	21.9 (8.1)	21.8 (9.0)	22.0 (8.6)	24.8 (6.9)	25.8 (7.0)	24.8 (6.3)	26.0 (7.1)
Frith–Happe animations task
Categorization, total	5.5 (2.1)	5.4 (1.8)	5.4 (2.0)	5.4 (2.1)	5.6 (1.9)	5.7 (2.2)	5.4 (2.3)	5.2 (2.6)
Feeling, total	3.0 (2.1)	2.7 (2.1)	2.3 (2.1)	3.0 (2.5)	2.7 (2.2)	2.8 (2.4)	3.3 (2.6)	3.4 (3.0)

SD: standard deviation; pSTS: posterior superior temporal sulcus; SRS: social responsiveness scale; RBS-R: repetitive behavior scale-revised.

Active group: 8-week active TBS; Sham group: 4-week sham intervention followed by 4-week active TBS.

Indicates significant within-group changes at Week 8 in comparison to baseline, ⁺p < 0.01.

Indicates significant within-group changes at Week 12 in comparison to baseline, ^$$p < 0.001.

Indicates significant within-group changes at Week 8 in comparison to Week 4, *p < 0.01.

Table 3.

Adjusted estimates of clinical and cognitive outcomes based on GEE model from baseline to week 4 for Active and Sham groups.

	Group effect		Time effect		Group × time effect
	Estimate	p value	Estimate	p value	Estimate	p value
Clinical symptoms
SRS, total	3.01 (5.72)	0.599	−1.47 (2.96)	0.620	−1.80 (4.04)	0.656
RBS-R, total	−2.87 (4.53)	0.526	−5.51 (2.58)	0.033	2.64 (3.27)	0.419
Social cognitive function
Reading the Mind in the Eyes test
Total correct	−4.08 (1.70)	0.017	0.61 (0.80)	0.444	0.49 (1.10)	0.655
Frith–Happe animations task
Categorization, total	0.19 (0.42)	0.642	0.29 (0.32)	0.362	−0.51 (0.42)	0.228
Feelings, total	0.47 (0.50)	0.341	−0.09 (0.28)	0.757	0.09 (0.54)	0.869

GEE: generalized estimation equation; SRS: social responsiveness scale; RBS-R: repetitive behavior scale-revised.

The Sham group serves as the reference group.

Figure 2.

The individual and averaged total scores of social responsiveness scale in (a) Active and (b) Sham iTBS groups across Baseline, Week 4, and Week 8 are presented. To give a better concept of the trend of changes, symmetrized percentage changes in the total scores of social responsiveness scale from the baseline of both groups are shown and compared (c). (error bars: standard errors).

Phase 2 (from Week 4 to Week 8, not blind)

The GEE model of Phase 2 showed significant time effect (p = 0.008) in the RBS-R (Supplementary Table 4), which means the total scores of RBS-R significantly increased at Week 8 than Week 4 for both groups. In addition, there was significant treatment × time interaction (p = 0.003) in the RBS-R. Within-group analyses showed that, in the 4-week only Active TBS group (Sham intervention at Phase 1) from Week 4 to Week 8, the RBS-R total scores showed significant increases (p = 0.008) (Table 2).

Phase 1 and Phase 2 (from Baseline to Week 8)

The GEE model from Baseline to Week 8 showed no significant time effect, treatment effect, and treatment × time interaction for all measures (Supplementary Table 5). In the analysis of symmetrized percentage change, there was also no significant difference for both groups from Baseline to Week 8 (Supplementary Table 2 and Supplementary Figure 2).

To examine the stimulation effect over time, despite no significant time effect from the GEE model, within-group pairwise analyses showed that, in the 8-week TBS group, the total scores of SRS and RBS-R were significantly lower at Week 8 in comparison to Baseline (p = 0.003 and 0.005, respectively; Table 2). Further analysis on the SRS subscales revealed that significant changes were noted in Social Communication and Autistic Mannerism (Supplementary Table 6).

In the 4-week TBS group (Sham group at Phase 1), there was no significant change in clinical and cognitive measures from Baseline to Week 8. The effect sizes of within and between group comparisons from Baseline to Week 8 ranged 0–0.70 and 0.1–0.49, respectively (Supplementary Table 3).

From Baseline to Week 12

The GEE model from Baseline to Week 12 showed no significant time effect, treatment effect, and treatment × time interaction for all measures (Supplementary Table 7). In the analysis of symmetrized percentage change, there was also no significant difference for both groups from Baseline to Week 8 (Supplementary Table 2 and Supplementary Figure 2).

To examine the stimulation effect over time, despite no significant time effect from the GEE model, within-group pairwise analyses showed that, in the 8-week Active TBS group (i.e. the Active group in Phase 1 of single-blind RCT, then followed by the other 4-week open-label intervention), the total scores of SRS and RBS-R were significantly lower at Week 12 in comparison to Baseline (both p < 0.001, Table 2). Further analysis on the SRS subscales revealed that significant changes were also noted in Social Communication and Autistic Mannerism (Supplementary Table 6). In the 4-week only Active TBS group (i.e. the Sham group in Phase 1, followed by 4-week active interventions), there was no significant change in clinical and cognitive measures from Baseline to Week 12. The effect sizes of within and between group comparisons from Baseline to Week 12 ranged 0.04–0.59 and 0.06–0.27, respectively (Supplementary Table 3).

Four-week follow-up (from Week 8 to Week 12)

The GEE model from Week 8 to Week 12 showed no significant time effect, treatment effect, and treatment × time interaction for all measures (Supplementary Table 8). In addition, there was no significant difference between Week 8 and Week 12 for all measures across two groups.

Characteristic of the responders

With the definition of responders having RCI > 1.64 in the SRS, in the 8-week active TBS group, there were six responders (15%) during the first 4 week intervention and 12 responders (30%) at the end of the total 8-week interventions (Supplementary Tables 9 and 10). There was no significant difference in demographic characteristics between responders and non-responders during the first 4 weeks. However, responders in the whole 8-week intervention had nominally better FIQ (p = 0.037) and performance on the animations task at baseline (p = 0.017) and nominally lower co-occurring ADHD than non-responders (p = 0.013). As for the 4-week only Active TBS group, there was no responder during Phase 1, but there were three responders (8.2%) at the end of the following 4-week open-label active TBS intervention (Supplementary Table 11).

The three-way interaction analysis of the GEE model was undertaken to explore effect moderators on the efficacy of Phase 1 trial (Supplementary Table 12). We found that FIQ (p = 0.024) nominally moderated the effects of iTBS on the total correct scores in the Frith–Happe animations. Specifically, participants with higher baseline FIQ might have higher improvement in animations in the Active versus Sham group.

Bias assessment

To assure that the treatment effects at Phase 2 were limitedly contributed by a placebo effect, the additional post hoc analysis showed that the effect size of changes in the SRS in the Active-Phase 1 (Cohen’s d = 0.227 from blind 4-week iTBS) was comparable to that in the Sham-Phase 2 (4-week only Active group) (Cohen’s d = 0.244 from unblinded 4 week intervention) (Supplementary Table 13). The ratio of SRS-responders (p = 0.392) and responder’s features (Supplementary Table 14) were similar between the Active-Phase 1 and Sham-Phase 2.

Discussion

As the world-first RCT on the iTBS (excitatory rTMS) applied over the pSTS in intellectually able children and adolescents with ASD, we found that 16-session iTBS (twice weekly for 8 weeks) is feasible, safe, and tolerable. With a well-powered sample (as twice large as the biggest-sized similarly-designed study ever published (Ameis et al., 2020)) in a randomized parallel single-blind, sham-controlled design, we did not find significant efficacy of 4-week iTBS over the pSTS on core symptoms and social cognitive performance in children with ASD from the blind trial phase. With a distinct design, we found that longer courses of iTBS (8 weeks in total, including the first 4-week phase of blind RCT and the second 4-week phase of unblinded universal open-label intervention) may produce greater efficacy, with a 30% responsive rate defined by the RCI. In addition, we found that the therapeutic effect of 8-week iTBS might maintain 4 weeks after the last intervention. Such exploration in the current trial reveals a number of factors, which may benefit planning future studies on rTMS/iTBS efficacy for ASD. Nonetheless, these results need to be interpreted in the context of the null findings in the blind RCT phase, as well as the caveat of a concatenation of the blind trial and open-label intervention.

With comparable attrition rates and side-effect profiles with earlier rTMS studies (Ameis et al., 2020; Enticott et al., 2014; EM Sokhadze et al., 2018), this study demonstrated that an at least 1-month course of regular and repeated iTBS for intellectually able children and adolescent with ASD is feasible and acceptable. The completion rate in our study (16 sessions/8 weeks) was 95% (38/40) in the active group, while that in a recent rTMS trial (20 sessions/4 weeks) was 90% (18/20) (Ameis et al., 2020). The side-effect profile was mild and transitory: The most common adverse effect herein was pain at the stimulated site (10%) and headache/dizziness (3%), compatible to a previous rTMS study in children with ASD (Ameis et al., 2020), alongside the TBS study in pediatric patients (Hong et al., 2015). Combined with our previous RCT (Ni et al., 2017), we demonstrated that iTBS/rTMS over the pSTS is practicable and safe in ASD, endorsing the endeavor of the earlier theoretical suggestion (Cole et al., 2019). Moreover, considering non-inferiority but shorter stimulation duration of iTBS relative to rTMS (Blumberger et al., 2018), the current iTBS protocol may favorably increase patients’ compliance and accessibility to its application.

Against our hypothesis, using the rigorous RCT design at Phase 1, we did not find any significant efficacy of 4-week (eight sessions in total) iTBS over the pSTS on the core symptoms and social cognitive functions in children and adolescents with ASD. We were only powered to detect a moderate effect size, limiting the ability to detect significances with small effects (e.g. effect 0.220 in SRS changes). A lack of efficacy, especially in results of social interaction deficits, was inconsistent with the only one published sham-controlled blind RCT study of 10-session 2-week deep rTMS over the dorsomedial PFC (Enticott et al., 2014) as well as several open-label RCTs over the DLPFC (Baruth et al., 2010; Casanova et al., 2012, 2014; E Sokhadze et al., 2010; EM Sokhadze et al., 2009, 2012, 2014, 2018). This discrepancy may be partially explained by the fact that this study with a larger sample size, compared to those low-powered studies (Barahona-Correa et al., 2018; Enticott et al., 2014), was less susceptible to exaggeration of effect size from inter-individual random variation (Button et al., 2013). Furthermore, the present design at Phase 1 could reduce possibilities of type I error due to biases from non-randomized open-label waitlist-controlled designs (Baruth et al., 2010; Casanova et al., 2012, 2014; E Sokhadze et al., 2010; EM Sokhadze et al., 2009, 2012, 2014, 2018), which might contribute part of inconsistency. However, we gave two sessions of iTBS every week, instead of daily session for five sessions per week in other protocols, could contribute to the slower and milder improvement due to the lower and less-accumulated dose effect in this study. Publication bias should also be considered in interpreting the inconsistency (Barahona-Correa et al., 2018).

Similar to the result of Phase 1, 4-week open-label active interventions at Phase 2 did not improve social symptoms in participants. Surprisingly, we observed a significant treatment by time interaction in the RBS-R at Phase 2, with an increase in the total scores (indicating symptomatic deterioration) across two groups. This result is at odds with our prior preliminary study showing beneficiary effects of pSTS stimulation on compulsion as measured by the Yale–Brown Obsessive Compulsive Scale in adults with ASD (Ni et al., 2017). The inconsistency highlights that the picture of rTMS effect on people with ASD is far from clear, which is discussed in detail as follows.

When the results of Phase 2, combined with those of Phase 1, are considered, a clearer picture emerges: longer courses and more sessions of iTBS over the pSTS might be needed to achieve the therapeutic efficacy for children with ASD. Specifically, our within-group analysis demonstrated that 8-week iTBS, but not 4-week stimulation, obtained significant improvement in the SRS and RBS-R in the 8-week active TBS group. This greater positive effect of longer iTBS courses was also supported by the result of a higher ratio of responder at Week 8 compared to that at Week 4 in the 8-week active TBS group. These results are consistent with a recent study exploring three different numbers of weekly 1 Hz rTMS session (6, 12, and 18 weeks) over the DLPFC, which also showed longer rTMS interventions (18 week > others) achieve greater improvement in clinical symptoms and executive function in children with ASD (EM Sokhadze et al., 2018). Similarly, a previous study (McClintock et al., 2018) found that at least 20–30 sessions are necessary for effective responses for treatment-refractory major depressive disorders, and some individuals may need even longer courses (Yip et al., 2017). The multi-session requirement has long been learned from animal studies showing repeated short sessions of training is better than a single-prolonged training for learning (Carew et al., 1972). Evidence largely suggests that rTMS/TBS produces therapeutic effects by inducing synaptic long-term depression/long-term potentiation through stimulations, resulting in following effects on regional neurotransmitters and synaptic plasticity that are considered as the basis of learning, memory, and recovery of the nervous system (Chervyakov et al., 2015). However, the intervention was blind in Phase 1 but not in Phase 2. Whether 8-week TBS is superior to 4-week TBS should be taken conservatively.

This “the more session the better” phenomenon can be explained in the following frameworks. First, a single session may change the synaptic transmission efficiency by modulating receptors and channels (Klomjai et al., 2015), while multiple sessions and long-term stimulation are required for changes in proteins and dendrite formation to consolidate the effects. Second, unlike most of other psychiatric disorders having diseased and state-related brain phenotypes, the ASD-associated brain appears to be natively idiosyncratically wired (Johnson, 2017). Thus, it is reasonable to assume that longer time courses or more sessions of rTMS/TBS may be needed to generate therapeutic impacts. Third, it is suggested that neuroplasticity is region- and condition-specific (Fuchs & Flugge, 2014). How local perturbations influence whole-brain functional dynamics is also region- and brain hierarchy-specific (Gollo et al., 2017). Some brain regions, for example, pSTS might be less plastic than the other locations with the common rTMS/TBS parameters, resulting in a protracted treatment response to the iTBS over the pSTS.

Consistent with prior studies (Enticott et al., 2011, 2014), our results demonstrated that the therapeutic effects of 8-week active iTBS (16 sessions) over the bilateral pSTS might persist for 4 weeks after the last TBS session. However, very limited studies on ASD reported or included the follow-up visit in the trial. Gomez et al. (2017) and Abujadi et al. (2018) showed that the therapeutic effects of rTMS over the DLPFC on autistic symptoms might last up to 3–6 months after the interventions. Similarly, the therapeutic effects of rTMS over the DLPFC in major depressive disorder also may persist from 2 weeks to 6 months following the end of trial (Health Quality Ontario, 2016). Whether there are extended effects of rTMS/TBS, which would be region specific and how long they could last have important clinical implications, and warrants the systemic investigations.

As the world-first to examine categorically who were the responders to rTMS/TBS for ASD based on the RCI (Yatawara et al., 2016), we found nominally significant results that those with higher FIQ, better social cognitive performance, and less ADHD comorbidity at baseline were more likely to be responders to iTBS over the pSTS. Furthermore, despite null treatment × time interactions, baseline FIQ might moderate the therapeutic effect of iTBS at Phase 1 (nominal significance; Bays, 2019). Notably, although non-significant (p = 0.052), the three-way moderating analysis showed that the concurrent use of methylphenidate might be associated with better responses to iTBS in the animation task. This phenomenon is consistent with that observed in patients with major depressive disorder users (Hunter et al., 2019). Moreover, preclinical studies suggest that by increasing the synaptic concentration of dopamine and norepinephrine (Leonard et al., 2004), methylphenidate may increase the magnitude of long-term potentiation induction in animal study (Urban et al., 2013), thereby modifying the synaptic plasticity synergistically with the rTMS/TBS (Meintzschel & Ziemann, 2006). This may be worthy of further investigation in subsequent more fully powered studies, given the common prescription of methylphenidate in people with ASD. We herein did not selectively recruit a subgroup of either more or less clinical severity. Furthermore, unlike the studies on depression (Health Quality Ontario, 2016), we did not find the baseline autistic symptoms moderating the treatment responses (null results from three-way interactions and responder vs nonresponder comparisons). However, we still observed some heterogeneous responses and potential factors moderating the efficacy, consistent with a similar attempt shown in a recent study, which showed baseline adaptive function modulates executive function improvement in rTMS over the DLPFC in ASD (Ameis et al., 2020). Our results support a notion that it is imperative to systemically identify the effect moderators and subgroup responders in future rTMS/TBS studies (Barahona-Correa et al., 2018; C Gross et al., 2015; Mottron & Bzdok, 2020).

The most effective protocol of TBS remains inconclusive. Although recent studies showed that the iTBS1800 (1800 pulses/session) have promising effects on major depressive disorder (Cole et al., 2020; CT Li et al., 2014), the physiological effects of iTBS1800 have never been well studied. Whether the therapeutic effect of iTBS1800 in major depressive disorder can be applied to ASD remains unknown. The results of a single session of iTBS with 1200 pulses on MEPs are mixed. Although Gamboa et al. (2010) showed inhibitory effects from the iTBS with 1200 pulses, Hsu et al. (2011) conversely found enhanced facilitation on MEPs. Learning from animal studies, a better way of enhancing the effect of rTMS protocol, for example, TBS, is to repeat standard blocks of rTMS with an interval in between (Goldsworthy et al., 2012; Maan et al., 2004; Mauelshagen et al., 1998), while the duration of interval could be critical for the final results. When we started our study in 2015, the appropriate intervals between two blocks of iTBS for enhancing the effects were not known. Considering the controversy results of the iTBS1200, as well as the theoretical model of TBS (Huang et al., 2011), we meant to designed our protocols to have a stronger facilitation by giving iTBS1200 with two 600 pulses separated with a 3-min break.

How many numbers of total sessions are sufficient to produce therapeutic effects of rTMS/TBS on ASD remains unknown. The protocol involving a specific 30 sessions as approved by US Food and Drug Administration (FDA) is for the major depressive disorder, but not for other major psychiatric disorders, including ASD. The total sessions of rTMS/TBS at one brain region ranged from 1 to 26, mostly between 10 and 20, in previous ASD studies (Barahona-Correa et al., 2018). The total sessions in two previous RCTs with the sham group were 10 (Enticott et al., 2014) and 20 (Ameis et al., 2020), respectively. Although 8 or 16 sessions in our study may not be enough to show the therapeutic efficacy in ASD, these numbers were largely comparable to previous studies on ASD. The future studies should address what the best rTMS/iTBS protocol for ASD is, such as the number of sessions, the pulses per session, the session interval, and the therapeutic duration.

There are several limitations in our pilot study. First, we could not totally exclude that the results at Phase 2 were somewhat biased by the unblinded process. However, a group-wise effect size of changes, as well as profiles of the responders, of the SRS were similar between the Sham-Phase 2 and Active-Phase 1 conditions. We thus considered that a placebo effect was limited, endorsing the validity in preceding inferences based on the results from 8-week courses. The present findings request replication in a shorter session interval (e.g. daily), long-enough therapeutic duration (i.e. 8 weeks or more), and blind design. Second, the selection of a fixed pSTS location may influence the results, especially given brain functional idiosyncrasy in ASD (Poulin-Lord et al., 2014). Future studies could be benefited by the localization guided by individual’s pSTS topography estimated by task-fMRI (D. Y. Yang et al., 2015). Moreover, previous evidence suggests a right-lateralized phenomenon of pSTS in social perception (Saitovitch et al., 2016). Whether stimulating bilateral pSTS would result in counteracting effects, as shown in rTMS over the bilateral DLPFC for depression (M Gross et al., 2007), remains unknown. Third, the clinical and social cognitive outcomes might not be sensitive enough to detect subtle changes. In the future, it is vital for the TMS field to move toward outcome measures which are specific to the stimulation neural target (Cole et al., 2019). Fourth, although the psychotropic medications for both groups were compatible and were required unchanged during the trial, we did not define the inclusion criteria that the psychotropic medication should be unchanged at least 4 weeks prior to the iTBS intervention. Per the post hoc chart review, medications in 88% (30/34) of the participants remained unchanged 4 weeks prior to the RCT. However, the information of other treatment such as behavioral therapy or social skill training was not collected during the trial. We acknowledged that the current observed results cannot be totally excluded from the confounding effects from this caveat in study design. Fifth, hitherto, there has been no study investigating the characteristics of rTMS/TBS responders and non-responders in people with ASD. We thus adopted the definition of the responder using RCI > 1.64, which may be considered lenient but actually corresponds to p < 0.05 with one-sided test. Future larger studies investigating the features of responders may benefit from a more stringent threshold, that is, RCT > 1.96. Sixth, despite the large sample size (n = 75) in our study, there are only 10 females enrolled. We did not have sufficient power to explore the sex/gender difference on the therapeutic response, which should be highlighted in the future study. Seventh, the heterogeneity resulted from co-occurring ADHD and associated medications may still impact our findings (potentially leading to a type II error), despite the fact that both treatment groups had comparable features in these factors, and we formally tested effects of these factors in the modulation analysis. Consistent with a recent consensus statement (Cole et al., 2019), future rTMS studies may benefit from recruiting participants with ASD based on the presentation of particular characteristics rather than simply having an ASD diagnosis. Finally, the age of the current sample ranged from 8 to 17 years. The literature regarding differential effects of developmental stage and puberty on responses to TMS/iTBS is still in the very early stage. Although we included the age in the three-way moderating analysis (Supplementary Table 10), this specific developmental factor warrants further investigation.

In conclusion, we demonstrated that 8-week repeated iTBS over the pSTS in children and adolescents with ASD is feasible and tolerable. The non-significant treatment × time interaction effect from the GEE model in the first 4-week single-blind RCT suggests that there is no treatment efficacy of facilitating pSTS on clinical symptoms and cognitive performance of social impairment given the current rTMS stimulation protocol. The exploratory within-group pairwise analysis suggests that longer course interventions (8 weeks) appear to produce greater therapeutic efficacy. Those responsive to iTBS over the pSTS had higher intellectual functioning and social cognition performance as well as lower ADHD comorbidity than the non-responders. Notably, the caveat in the current distinct design (i.e. combination of the blind trial and open-label intervention) weakens the level of the evidence, limiting the inferences of the present findings. If replicated in the future, our findings suggest that rTMS/TBS over the pSTS, despite theoretically a promising target (Cole et al., 2019), may need longer time courses or more sessions to generate therapeutic effects on social deficits in children with ASD. Moreover, neurobiological markers to predict iTBS responses, such as the factors favor responsiveness identified above, are needed for the future studies. However, clinical heterogeneity, several moderating effects regarding individual’s characteristics, as well as several methodological considerations in protocol designs need to be taken into account, when interpreting rTMS/TBS efficacy in ASD.

Supplemental Material

sj-pdf-1-aut-10.1177_1362361321990534 – Supplemental material for Intermittent theta burst stimulation over the posterior superior temporal sulcus for children with autism spectrum disorder: A 4-week randomized blinded controlled trial followed by another 4-week open-label intervention

Supplemental material, sj-pdf-1-aut-10.1177_1362361321990534 for Intermittent theta burst stimulation over the posterior superior temporal sulcus for children with autism spectrum disorder: A 4-week randomized blinded controlled trial followed by another 4-week open-label intervention by Hsing-Chang Ni, Yi-Lung Chen, Yi-Ping Chao, Chen-Te Wu, Yu-Yu Wu, Sophie Hsin-Yi Liang, Wei-Chih Chin, Tai-Li Chou, Susan Shur-Fen Gau, Ying-Zu Huang and Hsiang-Yuan Lin in Autism

Footnotes

Acknowledgements

The authors thank Ms Chiu Fen Lin for the assistance of executing TBS and all of the participants and their parents to participate in the study.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by grants from National Science Council of Taiwan (NSC 105-2628-B-182A-005-MY3, 105-2314-B-182-004-MY3, and 108-2628-B-182A-006) and Chang Gung Medical Foundation (BMRP844).

ORCID iDs

Yi-Lung Chen

Tai-Li Chou

Susan Shur-Fen Gau

Ying-Zu Huang

Supplemental material

Supplemental material for this article is available online.

References

Abujadi

Croarkin

P. E.

Bellini

B. B.

Brentani

Marcolin

M. A.

(2018). Intermittent theta-burst transcranial magnetic stimulation for autism spectrum disorder: An open-label pilot study. Brazilian Journal of Psychiatry, 40(3), 309–311. https://doi.org/10.1590/1516-4446-2017-2279

Ameis

S. H.

Blumberger

D. M.

Croarkin

P. E.

Mabbott

D. J.

Lai

M. C.

Desarkar

. . .Daskalakis

Z. J.

(2020). Treatment of executive function deficits in autism spectrum disorder with repetitive transcranial magnetic stimulation: A double-blind, sham-controlled, pilot trial. Brain Stimulation, 13(3), 539–547. https://doi.org/10.1016/j.brs.2020.01.007

Anninos

Chatzimichael

Adamopoulos

Kotini

Tsagas

(2016). A combined study of MEG and pico-Tesla TMS on children with autism disorder. Journal of Integrative Neuroscience, 15(4), 497–513. https://doi.org/10.1142/S0219635216500278

Ayers

G. D.

Berry

D. A.

(2006). Symmetrized percent change for treatment comparisons. The American Statistician, 60(1), 27–31.

Barahona-Correa

J. B.

Velosa

Chainho

Lopes

Oliveira-Maia

A. J.

(2018). Repetitive transcranial magnetic stimulation for treatment of autism spectrum disorder: A systematic review and meta-analysis. Frontiers in Integrative Neuroscience, 12, Article 27. https://doi.org/10.3389/fnint.2018.00027

Barch

D. M.

Burgess

G. C.

Harms

M. P.

Petersen

S. E.

Schlaggar

B. L.

Corbetta

. . .Consortium

W. U.-M. H

. (2013). Function in the human connectome: Task-fMRI and individual differences in behavior. NeuroImage, 80, 169–189. https://doi.org/10.1016/j.neuroimage.2013.05.033

Baron-Cohen

Wheelwright

Hill

Raste

Plumb

(2001). The “reading the mind in the eyes” test revised version: A study with normal adults, and adults with Asperger syndrome or high-functioning autism. Journal of Child Psychology and Psychiatry, 42(2), 241–251.

Baruth

J. M.

Casanova

M. F.

El-Baz

Horrell

Mathai

Sears

Sokhadze

(2010). Low-frequency repetitive transcranial magnetic stimulation (rTMS) modulates evoked-gamma frequency oscillations in autism spectrum disorder (ASD). Journal of Neurotherapy, 14(3), 179–194. https://doi.org/10.1080/10874208.2010.501500

Bays

H. E.

(2019). Alirocumab, decreased mortality, nominal significance, P values, Bayesian statistics, and the duplicity of multiplicity. Circulation, 140(2), 113–116. https://doi.org/10.1161/CIRCULATIONAHA.119.041496

10.

Berlim

M. T.

Van den Eynde

Jeff Daskalakis

(2013). Clinically meaningful efficacy and acceptability of low-frequency repetitive transcranial magnetic stimulation (rTMS) for treating primary major depression: A meta-analysis of randomized, double-blind and sham-controlled trials. Neuropsychopharmacology, 38(4), 543–551. https://doi.org/10.1038/npp.2012.237

11.

Blumberger

D. M.

Vila-Rodriguez

Thorpe

K. E.

Feffer

Noda

Giacobbe

. . .Downar

(2018). Effectiveness of theta burst versus high-frequency repetitive transcranial magnetic stimulation in patients with depression (THREE-D): A randomised non-inferiority trial. The Lancet, 391(10131), 1683–1692. https://doi.org/10.1016/S0140-6736(18)30295-2

12.

Bodfish

J. W.

Symons

F. J.

Parker

D. E.

Lewis

M. H.

(2000). Varieties of repetitive behavior in autism: Comparisons to mental retardation. Journal of Autism and Developmental Disorders, 30(3), 237–243. https://doi.org/10.1023/a:1005596502855

13.

Button

K. S.

Ioannidis

J. P.

Mokrysz

Nosek

B. A.

Flint

Robinson

E. S.

Munafo

M. R.

(2013). Power failure: Why small sample size undermines the reliability of neuroscience. Nature Reviews Neuroscience, 14(5), 365–376. https://doi.org/10.1038/nrn3475

14.

Carew

T. J.

Pinsker

H. M.

Kandel

E. R.

(1972). Long-term habituation of a defensive withdrawal reflex in aplysia. Science, 175(4020), 451–454. https://doi.org/10.1126/science.175.4020.451

15.

Casanova

M. F.

Baruth

J. M.

El-Baz

Tasman

Sears

Sokhadze

(2012). Repetitive transcranial magnetic stimulation (rTMS) modulates event-related potential (ERP) indices of attention in autism. Translational Neuroscience, 3(2), 170–180. https://doi.org/10.2478/s13380-012-0022-0

16.

Casanova

M. F.

Hensley

M. K.

Sokhadze

E. M.

El-Baz

A. S.

Wang

Sears

(2014). Effects of weekly low-frequency rTMS on autonomic measures in children with autism spectrum disorder. Frontiers in Human Neuroscience, 8, Article 851. https://doi.org/10.3389/fnhum.2014.00851

17.

Chervyakov

A. V.

Chernyavsky

A. Y.

Sinitsyn

D. O.

Piradov

M. A.

(2015). Possible mechanisms underlying the therapeutic effects of transcranial magnetic stimulation. Frontiers in Human Neuroscience, 9, Article 303. https://doi.org/10.3389/fnhum.2015.00303

18.

Cole

E. J.

Enticott

P. G.

Oberman

L. M.

Gwynette

M. F.

Casanova

M. F.

Jackson

S. L. J.

. . . rTMS in ASD Consensus Group. (2019). The potential of repetitive transcranial magnetic stimulation for autism spectrum disorder: A consensus statement. Biological Psychiatry, 85(4), e21–e22. https://doi.org/10.1016/j.biopsych.2018.06.003

19.

Cole

E. J.

Stimpson

K. H.

Bentzley

B. S.

Gulser

Cherian

Tischler

. . .Williams

N. R.

(2020). Stanford accelerated intelligent neuromodulation therapy for treatment-resistant depression. American Journal of Psychiatry, 177(8), 716–726. https://doi.org/10.1176/appi.ajp.2019.19070720

20.

Enticott

P. G.

Fitzgibbon

B. M.

Kennedy

H. A.

Arnold

S. L.

Elliot

Peachey

. . .Fitzgerald

P. B.

(2014). A double-blind, randomized trial of deep repetitive transcranial magnetic stimulation (rTMS) for autism spectrum disorder. Brain Stimulation, 7(2), 206–211. https://doi.org/10.1016/j.brs.2013.10.004

21.

Enticott

P. G.

Kennedy

H. A.

Zangen

Fitzgerald

P. B.

(2011). Deep repetitive transcranial magnetic stimulation associated with improved social functioning in a young woman with an autism spectrum disorder. The Journal of ECT, 27(1), 41–43. https://doi.org/10.1097/YCT.0b013e3181f07948

22.

Enticott

P. G.

Rinehart

N. J.

Tonge

B. J.

Bradshaw

J. L.

Fitzgerald

P. B.

(2012). Repetitive transcranial magnetic stimulation (rTMS) improves movement-related cortical potentials in autism spectrum disorders. Brain Stimulation, 5(1), 30–37. https://doi.org/10.1016/j.brs.2011.02.001

23.

Faul

Erdfelder

Buchner

Lang

A. G.

(2009). Statistical power analyses using G*Power 3.1: Tests for correlation and regression analyses. Behavior Research Methods, 41(4), 1149–1160. https://doi.org/10.3758/BRM.41.4.1149

24.

Fecteau

Agosta

Oberman

Pascual-Leone

(2011). Brain stimulation over Broca’s area differentially modulates naming skills in neurotypical adults and individuals with Asperger’s syndrome. European Journal of Neuroscience, 34(1), 158–164. https://doi.org/10.1111/j.1460-9568.2011.07726.x

25.

Fuchs

Flugge

(2014). Adult neuroplasticity: More than 40 years of research. Neural Plasticity, 2014, Article 541870. https://doi.org/10.1155/2014/541870

26.

Gamboa

O. L.

Antal

Moliadze

Paulus

(2010). Simply longer is not better: Reversal of theta burst after-effect with prolonged stimulation. Experimental Brain Research, 204(2), 181–187. https://doi.org/10.1007/s00221-010-2293-4

27.

Gau

S. S.-F.

Liu

L.-T.

Y.-Y.

Chiu

Y.-N.

Tsai

W.-C.

(2013). Psychometric properties of the Chinese version of the social responsiveness scale. Research in Autism Spectrum Disorders, 7(2), 349–360. https://doi.org/10.1016/j.rasd.2012.10.004

28.

Goldsworthy

M. R.

Pitcher

J. B.

Ridding

M. C.

(2012). The application of spaced theta burst protocols induces long-lasting neuroplastic changes in the human motor cortex. European Journal of Neuroscience, 35(1), 125–134. https://doi.org/10.1111/j.1460-9568.2011.07924.x

29.

Gollo

L. L.

Roberts

J. A.

Cocchi

(2017). Mapping how local perturbations influence systems-level brain dynamics. NeuroImage, 160, 97–112. https://doi.org/10.1016/j.neuroimage.2017.01.057

30.

Gomez

Vidal

Maragoto

Morales

L. M.

Berrillo

Vera Cuesta

. . .Robinson

(2017). Non-invasive brain stimulation for children with autism spectrum disorders: A short-term outcome study. Behavioral Sciences, 7(3), Article 63. https://doi.org/10.3390/bs7030063

31.

Green

J. P. H. S

. (2008). Cochrane handbook for systematic reviews of interventions. Cochrane Collaboration.

32.

Gross

Hoffmann

Bassell

G. J.

Berry-Kravis

E. M.

(2015). Therapeutic strategies in fragile X syndrome: From bench to bedside and back. Neurotherapeutics, 12(3), 584–608. https://doi.org/10.1007/s13311-015-0355-9

33.

Gross

Nakamura

Pascual-Leone

Fregni

, (2007). Has repetitive transcranial magnetic stimulation (rTMS) treatment for depression improved? A systematic review and meta-analysis comparing the recent vs. the earlier rTMS studies. Acta Psychiatrica Scandinavica, 116(3), 165–173. https://doi.org/10.1111/j.1600-0447.2007.01049.x

34.

Health Quality Ontario. (2016). Repetitive transcranial magnetic stimulation for treatment-resistant depression: A systematic review and meta-analysis of randomized controlled trials. Ontario Health Technology Assessment Series, 16(5), 1–66.

35.

Hong

Y. H.

S. W.

Pedapati

E. V.

Horn

P. S.

Huddleston

D. A.

Laue

C. S.

Gilbert

D. L.

(2015). Safety and tolerability of theta burst stimulation vs. single and paired pulse transcranial magnetic stimulation: A comparative study of 165 pediatric subjects. Frontiers in Human Neuroscience, 9, Article 29. https://doi.org/10.3389/fnhum.2015.00029

36.

Hotier

Leroy

Boisgontier

Laidi

Mangin

J. F.

Delorme

. . .Houenou

(2017). Social cognition in autism is associated with the neurodevelopment of the posterior superior temporal sulcus. Acta Psychiatrica Scandinavica, 136(5), 517–525. https://doi.org/10.1111/acps.12814

37.

Hsu

Y. F.

Liao

K. K.

Lee

P. L.

Tsai

Y. A.

Yeh

C. L.

Lai

K. L.

. . .Lee

I. H.

(2011). Intermittent theta burst stimulation over primary motor cortex enhances movement-related beta synchronisation. Clinical Neurophysiology, 122(11), 2260–2267. https://doi.org/10.1016/j.clinph.2011.03.027

38.

Huang

Y. Z.

Edwards

M. J.

Rounis

Bhatia

K. P.

Rothwell

J. C.

(2005). Theta burst stimulation of the human motor cortex. Neuron, 45(2), 201–206. https://doi.org/10.1016/j.neuron.2004.12.033

39.

Huang

Y. Z.

Rothwell

J. C.

Chen

R. S.

C. S.

Chuang

W. L.

(2011). The theoretical model of theta burst form of repetitive transcranial magnetic stimulation. Clinical Neurophysiology, 122(5), 1011–1018. https://doi.org/10.1016/j.clinph.2010.08.016

40.

Hunter

A. M.

Minzenberg

M. J.

Cook

I. A.

Krantz

D. E.

Levitt

J. G.

Rotstein

N. M.

. . .Leuchter

A. F.

(2019). Concomitant medication use and clinical outcome of repetitive transcranial magnetic stimulation (rTMS) treatment of major depressive disorder. Brain and Behavior, 9(5), Article e01275. https://doi.org/10.1002/brb3.1275

41.

Jacobson

N. S.

Truax

(1991). Clinical significance: A statistical approach to defining meaningful change in psychotherapy research. Journal of Consulting and Clinical Psychology, 59(1), 12–19. https://doi.org/10.1037//0022-006x.59.1.12

42.

Johnson

M. H.

(2017). Autism as an adaptive common variant pathway for human brain development. Developmental Cognitive Neuroscience, 25, 5–11. https://doi.org/10.1016/j.dcn.2017.02.004

43.

Klomjai

Katz

Lackmy-Vallee

(2015). Basic principles of transcranial magnetic stimulation (TMS) and repetitive TMS (rTMS). Annals of Physical and Rehabilitation Medicine, 58(4), 208–213. https://doi.org/10.1016/j.rehab.2015.05.005

44.

Lefaucheur

J. P.

Aleman

Baeken

Benninger

D. H.

Brunelin

Di Lazzaro

. . .Ziemann

(2020). Evidence-based guidelines on the therapeutic use of repetitive transcranial magnetic stimulation (rTMS): An update (2014-2018). Clinical Neurophysiology, 131(2), 474–528. https://doi.org/10.1016/j.clinph.2019.11.002

45.

Leonard

B. E.

McCartan

White

King

D. J.

(2004). Methylphenidate: A review of its neuropharmacological, neuropsychological and adverse clinical effects. Human Psychopharmacology: Clinical and Experimental, 19(3), 151–180. https://doi.org/10.1002/hup.579

46.

C. T.

Chen

M. H.

Juan

C. H.

Huang

H. H.

Chen

L. F.

Hsieh

J. C.

. . .Su

T. P.

(2014). Efficacy of prefrontal theta-burst stimulation in refractory depression: A randomized sham-controlled study. Brain, 137(Pt7), 2088–2098. https://doi.org/10.1093/brain/awu109

47.

T. S.

Liu

C. M.

Liu

C. C.

Hsieh

M. H.

Lin

Y. T.

Wang

E. N.

. . .Chou

T. L.

(2020). Social cognition in schizophrenia: A network-based approach to a Taiwanese version of the Reading the Mind in the Eyes test. Journal of the Formosan Medical Association, 119(1 Pt. 3), 439–448. https://doi.org/10.1016/j.jfma.2019.08.008

48.

Lisanby

S. H.

Gutman

Luber

Schroeder

Sackeim

H. A.

(2001). Sham TMS: Intracerebral measurement of the induced electrical field and the induction of motor-evoked potentials. Biological Psychiatry, 49(5), 460–463. https://doi.org/10.1016/s0006-3223(00)01110-0

49.

Lord

Brugha

T. S.

Charman

Cusack

Dumas

Frazier

. . .Veenstra-VanderWeele

(2020). Autism spectrum disorder. Nature Reviews Disease Primers, 6(1), Article 5. https://doi.org/10.1038/s41572-019-0138-4

50.

Lord

Risi

Lambrecht

Cook

E. H.

Jr Leventhal

B. L.

DiLavore

P. C.

. . .Rutter

(2000). The autism diagnostic observation schedule-generic: A standard measure of social and communication deficits associated with the spectrum of autism. Journal of Autism and Developmental Disorders, 30(3), 205–223.

51.

Maan

Al-Singary

Shergill

Joseph

J. V.

Arya

Patel

H. R.

(2004). Alternative use of botulinum toxin in urology. Expert Opinion on Pharmacotherapy, 5(5), 1015–1021. https://doi.org/10.1517/14656566.5.5.1015

52.

Mauelshagen

Sherff

C. M.

Carew

T. J.

(1998). Differential induction of long-term synaptic facilitation by spaced and massed applications of serotonin at sensory neuron synapses of Aplysia californica. Learning & Memory, 5(3), 246–256. https://doi.org/10.1101/lm.5.3.246

53.

McClintock

S. M.

Reti

I. M.

Carpenter

L. L.

McDonald

W. M.

Dubin

Taylor

S. F.

, . . . National Network of Depression Centers rTMS Task Group and the American Psychiatric Association Council on Research Task Force on Novel Biomarkers and Treatments. (2018). Consensus recommendations for the clinical application of repetitive transcranial magnetic stimulation (rTMS) in the treatment of depression. Journal of Clinical Psychiatry, 79(1), Article 16cs10905. https://doi.org/10.4088/JCP.16cs10905

54.

Meintzschel

Ziemann

(2006). Modification of practice-dependent plasticity in human motor cortex by neuromodulators. Cerebral Cortex, 16(8), 1106–1115. https://doi.org/10.1093/cercor/bhj052

55.

Mottron

Bzdok

(2020). Autism spectrum heterogeneity: Fact or artifact? Molecular Psychiatry, 25, 3178–3185. https://doi.org/10.1038/s41380-020-0748-y

56.

H. C.

Hung

C. T.

Y. Y.

Chang

C. J.

Chen

R. S.

Huang

Y. Z.

(2017). The impact of single session intermittent theta-burst stimulation over the dorsolateral prefrontal cortex and posterior superior temporal sulcus on adults with autism spectrum disorder. Frontiers in Neuroscience, 11, Article 255. https://doi.org/10.3389/fnins.2017.00255

57.

Oberman

L. M.

Enticott

P. G.

Casanova

M. F.

Rotenberg

Pascual-Leone

McCracken

J. T.

The TMS in ASD Consensus Group. (2016). Transcranial magnetic stimulation in autism spectrum disorder: Challenges, promise, and roadmap for future research. Autism Research, 9(2), 184–203. https://doi.org/10.1002/aur.1567

58.

O’Reardon

J. P.

Solvason

H. B.

Janicak

P. G.

Sampson

Isenberg

K. E.

Nahas

. . .Sackeim

H. A.

(2007). Efficacy and safety of transcranial magnetic stimulation in the acute treatment of major depression: A multisite randomized controlled trial. Biological Psychiatry, 62(11), 1208–1216. https://doi.org/10.1016/j.biopsych.2007.01.018

59.

Padmanabhan

Lynch

C. J.

Schaer

Menon

(2017). The default mode network in autism. Biological Psychiatry: Cognitive Neuroscience and Neuroimaging, 2(6), 476–486. https://doi.org/10.1016/j.bpsc.2017.04.004

60.

Panerai

Tasca

Lanuzza

Trubia

Ferri

Musso

. . .Elia

(2014). Effects of repetitive transcranial magnetic stimulation in performing eye-hand integration tasks: Four preliminary studies with children showing low-functioning autism. Autism, 18(6), 638–650. https://doi.org/10.1177/1362361313495717

61.

Poulin-Lord

M. P.

Barbeau

E. B.

Soulieres

Monchi

Doyon

Benali

Mottron

(2014). Increased topographical variability of task-related activation in perceptive and motor associative regions in adult autistics. NeuroImage: Clinical, 4, 444–453. https://doi.org/10.1016/j.nicl.2014.02.008

62.

Power

J. D.

Cohen

A. L.

Nelson

S. M.

Wig

G. S.

Barnes

K. A.

Church

J. A.

. . .Petersen

S. E.

(2011). Functional network organization of the human brain. Neuron, 72(4), 665–678. https://doi.org/10.1016/j.neuron.2011.09.006

63.

Saitovitch

Popa

Lemaitre

Rechtman

Lamy

J. C.

Grevent

. . .Zilbovicius

(2016). Tuning eye-gaze perception by transitory STS inhibition. Cerebral Cortex, 26(6), 2823–2831. https://doi.org/10.1093/cercor/bhw045

64.

Schwippel

Schroeder

P. A.

Fallgatter

A. J.

Plewnia

(2019). Clinical review: The therapeutic use of theta-burst stimulation in mental disorders and tinnitus. Progress in Neuro-Psychopharmacology & Biological Psychiatry, 92, 285–300. https://doi.org/10.1016/j.pnpbp.2019.01.014

65.

Sokhadze

Baruth

Tasman

Mansoor

Ramaswamy

Sears

. . .Casanova

M. F.

(2010). Low-frequency repetitive transcranial magnetic stimulation (rTMS) affects event-related potential measures of novelty processing in autism. Applied Psychophysiology and Biofeedback, 35(2), 147–161. https://doi.org/10.1007/s10484-009-9121-2

66.

Sokhadze

E. M.

Baruth

J. M.

Sears

Sokhadze

G. E.

El-Baz

A. S.

Casanova

M. F.

(2012). Prefrontal neuromodulation using rTMS improves error monitoring and correction function in autism. Applied Psychophysiology and Biofeedback, 37(2), 91–102. https://doi.org/10.1007/s10484-012-9182-5

67.

Sokhadze

E. M.

El-Baz

Baruth

Mathai

Sears

Casanova

M. F.

(2009). Effects of low frequency repetitive transcranial magnetic stimulation (rTMS) on gamma frequency oscillations and event-related potentials during processing of illusory figures in autism. Journal of Autism and Developmental Disorders, 39(4), 619–634. https://doi.org/10.1007/s10803-008-0662-7

68.

Sokhadze

E. M.

El-Baz

A. S.

Tasman

Sears

L. L.

Wang

Lamina

E. V.

Casanova

M. F.

(2014). Neuromodulation integrating rTMS and neurofeedback for the treatment of autism spectrum disorder: An exploratory study. Applied Psychophysiology and Biofeedback, 39(3–4), 237–257. https://doi.org/10.1007/s10484-014-9264-7

69.

Sokhadze

E. M.

Lamina

E. V.

Casanova

E. L.

Kelly

D. P.

Opris

Tasman

Casanova

M. F.

(2018). Exploratory study of rTMS neuromodulation effects on electrocortical functional measures of performance in an Oddball Test and behavioral symptoms in autism. Frontiers in Systems Neuroscience, 12, Article 20. https://doi.org/10.3389/fnsys.2018.00020

70.

Urban

K. R.

Y. C.

Gao

W. J.

(2013). Treatment with a clinically-relevant dose of methylphenidate alters NMDA receptor composition and synaptic plasticity in the juvenile rat prefrontal cortex. Neurobiology of Learning and Memory, 101, 65–74. https://doi.org/10.1016/j.nlm.2013.01.004

71.

Van Overwalle

Baetens

. (2009). Understanding others’ actions and goals by mirror and mentalizing systems: A meta-analysis. NeuroImage, 48(3), 564–584. https://doi.org/10.1016/j.neuroimage.2009.06.009

72.

White

S. J.

Coniston

Rogers

Frith

(2011). Developing the Frith–Happe animations: A quick and objective test of Theory of Mind for adults with autism. Autism Research, 4(2), 149–154. https://doi.org/10.1002/aur.174

73.

Yang

D. Y.

Rosenblau

Keifer

Pelphrey

K. A.

(2015). An integrative neural model of social perception, action observation, and theory of mind. Neuroscience & Biobehavioral Reviews, 51, 263–275. https://doi.org/10.1016/j.neubiorev.2015.01.020

74.

Yang

Y. C.

Jeng

S. F.

Tsao

P. N.

Cheong

P. L.

Y. J.

. . .Wu

Y. T.

(2019). Multidimensional developments and free-play movement tracking in 30- to 36-month-old toddlers with autism spectrum disorder who were full term. Physical Therapy, 99(11), 1535–1550. https://doi.org/10.1093/ptj/pzz114

75.

Yatawara

C. J.

Einfeld

S. L.

Hickie

I. B.

Davenport

T. A.

Guastella

A. J.

(2016). The effect of oxytocin nasal spray on social interaction deficits observed in young children with autism: A randomized clinical crossover trial. Molecular Psychiatry, 21(9), 1225–1231. https://doi.org/10.1038/mp.2015.162

76.

Yip

A. G.

George

M. S.

Tendler

Roth

Zangen

Carpenter

L. L.

(2017). 61% of unmedicated treatment resistant depression patients who did not respond to acute TMS treatment responded after four weeks of twice weekly deep TMS in the Brainsway pivotal trial. Brain Stimulation, 10(4), 847–849. https://doi.org/10.1016/j.brs.2017.02.013

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

1.06 MB