Trial design

This is a phase II, single-centre, randomised, parallel group, sham-controlled clinical trial that will be held at the Alberta Children’s Hospital (ACH) in Calgary, Alberta, Canada (figure 1). The trial will follow the Consolidated Standards of Reporting Trials guidelines34 (figure 2). Participants will be randomised to either an (1) active rTMS +CBIT arm or (2) sham rTMS +CBIT arm in a 1:1 ratio using an algorithm that will stratify each participant by sex and age group (6–10, 11–14, 15–18 years). Sham-controlled TMS research is necessary to understand the neurobiological effects of TMS separate from placebo response. rTMS will be set up for either active or sham delivery in such a way that the participants, their families and all clinical evaluation raters will be blind to which group they are in: the two coils will be identical in appearance and produce similar auditory and scalp sensations. At week 36, there will be a brief follow-up consisting of TS symptom assessment to determine the durability of treatment effects.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

TITANS trial timeline. Outline of participant progression through the TITANS trial. At baseline (week 0) and post-treatments (weeks 7 and 11), participants undergo MRI, a battery of clinical assessments, symptom monitoring (Yale Global Tic Severity Scale and Modified Rush Video Rating Scale), transcranial magnetic stimulation (TMS) neurophysiology measures, motor assessment (Purdue Pegboard Task), and TMS motor mapping. Treatment consists of active or sham repetitive TMS (rTMS) for 5 weeks overlapped with simultaneous Comprehensive Behavioural Intervention for Tics (CBIT) therapy for 10 weeks. CBIT occurs on Mondays, weekly for the first 6 weeks then bi-weekly for a total of eight therapy sessions. rTMS treatment occurs 4 days a week (Tuesday to Friday) for 5 weeks (weeks 2–6). Symptom monitoring occurs every Friday (weekly). TMS tolerability is recorded with each TMS neurophysiology and motor mapping session, and weekly for rTMS sessions. Follow-up (Tourette’s and mental health symptoms) occurs at week 36.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2

Study flow diagram. Consolidated Standards of Reporting Trials diagram showing the flow of participants through each stage of the TITANS clinical trial: phase II single centre, randomised, parallel group, double-blind sham controlled rTMS trial in children with moderate-to-severe Tourette’s syndrome. CBIT, comprehensive intervention for tics; HRT, habit reversal therapy; rTMS, repetitive; TMS, transcranial magnetic stimulation; YGTSS, Yale Global Tic Severity Scale.

Inclusion criteria

1. Between the ages of 6–18 years.

2. Primary diagnosis of TS (with possible comorbid conditions), measured by the Mini-International Neuropsychiatric Interview for children and adolescents (MINI-KID).35

3. Tourette’s syndrome of moderate or greater severity at baseline, as measured by the YGTSS: total tic severity scores greater than 22 (or >11 if only motor or phonic tics).36

4. IQ greater than 80, as measured using the Wechsler Abbreviated Scale of Intelligence second Edition.

5. English fluency (to enable consent).

6. Medications for tics or psychiatric disorders are allowed if the dose has been stable for 6 weeks with adequate compliance, with a commitment to not change medication or dosage during the trial period.

Exclusion criteria

1. Diagnosis of mania or psychosis, as measured by the MINI-KID.35

2. Impediments to TMS or MRI (eg, implanted device, metal in the brain, pregnancy).

3. More than four previous HRT or CBIT sessions for Tourette’s syndrome.

Informed consent

Parents and adolescents will provide written informed consent before participating in accordance with the Conjoint Health Research Ethics Board and Tri-Council Policy Statement 2. Study staff will ask permission to share anonymised study data with collaborators for larger imaging analysis projects: this consent is voluntary and separate from the TITANS trial participation consent. No biological data (eg, blood, saliva or tissue samples) will be collected in this study. Consent will be obtained by the study principal investigator (PI) or research coordinator.

Patient and public involvement

Patients and the public were not involved in the design of the study.

Interventions

Eight sessions of CBIT will be delivered over 10 weeks, once a week for 6 weeks then biweekly.37 The first two CBIT sessions will take approximately 90 min, with the following six sessions taking approximately 60 min. A trained CBIT therapist will use a private room to deliver the CBIT therapy to a participant, with or without a caregiver in the room (participant choice). The CBIT sessions will be video recorded to ensure inter-rater reliability between CBIT therapists.

Twenty sessions of rTMS (active or sham condition) will be delivered four times a week for 5 weeks: low-frequency (1 Hz), 100% resting motor threshold (RMT), fMRI and neuronavigated TMS to the participant’s bilateral SMA, 1800 stimulations (active) or clicks (sham) per session. Participants will take breaks throughout the TMS sessions as needed, and therefore each TMS session will take 30–60 min. A weighted blanket, footrest, earplugs, neck, back and lap pillows will be available for participant comfort. TMS sessions will be video recorded once a week to assess tic expression during TMS.

Figure 1 outlines the time schedule of participation including baseline assessments (week 0), CBIT treatments (weeks 1–6, 8 and 10), rTMS treatments (weeks 2–6), post-TMS assessments (week 7), post-CBIT assessments (week 11) and follow-up (week 36).

Criteria for discontinuing or modifying interventions

Non-compliance, defined as missing ≥2 CBIT sessions or ≥2 rTMS sessions, without a desire/ability to make-up the missed sessions, may result in removal from the study. In addition, participation will be terminated if a participant experiences (1) a serious adverse reaction (eg, seizure), or (2) complications or severe worsening of symptoms (clinical judgement of TP and AK). Participants may withdraw from the study at any time.

Strategies to improve adherence

This study includes 28 intervention visits over 10 weeks of time: every effort will be made to book appointments at times that work best for participant families (eg, before or after school). Participants can choose to watch movies or listen to music during the TMS visits, and they can choose to do an activity (eg, draw, colour, build puzzles) during CBIT visits. Study staff will ask participants about their CBIT homework during TMS sessions to improve adherence of CBIT practice between therapy sessions.

Relevant concomitant care

Medications for tics or psychiatric disorders are allowed if the dose has been stable for 6 weeks with adequate compliance and a commitment to not change the medication dosage during the trial period. Therapy other than HRT or CBIT for tics or psychiatric disorders is also allowed. Participants cannot participate in other neurostimulation or habit reversal type therapies during the trial period. These will be recorded and reported in any subsequent publications.

No direct post-trial care is offered, but participants can obtain their personal data from study staff to share with other health providers and/or educators if they choose. If a participant suffers harm during the study, they have legal rights to seek damages: no direct compensation will be provided by the researchers, funding bodies, the University of Calgary or Alberta Health Services.

Outcomes

The primary outcome is Tourette’s symptom change (YGTSS) from baseline to post-CBIT treatment. Secondary outcomes include changes in (1) imaging, (2) neurophysiological and (3) behavioural markers (tables 1 and 2). First, magnetic resonance images will be obtained at baseline, post-TMS and post-CBIT on a GE 3T Discovery 750 W MRI scanner. Proton magnetic resonance imaging (MRS) will be used to determine metabolite concentrations in voxels placed in the SMA (2.5×2.5×2.5 cm³), left and right primary motor areas (2×3×4 cm³) using anatomical markers and an fMRI finger-tapping task as a guide. Changes in glutamate and GABA will be assessed from baseline to post-TMS and post-CBIT treatments: as an inhibitory neurotransmitter,12 we expect an increase in GABA concentration to correlate with reduction in tic symptom severity. Second, neurophysiological testing will be performed using a handheld figure-eight flat iron TMS coil (Magstim) to determine stimulus response curve,38 cortical silent period,39 40 short interval intracortical inhibition and facilitation41 42 and long interval intracortical inhibition43 at baseline, post-TMS and post-CBIT to measure changes in excitability and inhibition of corticospinal, intracortical and interhemispheric motor networks.44 Resting and active motor maps of the dominant hand area using robotic TMS (figure-eight airfilm coil, Magstim) will measure two-dimensional area, three-dimensional volume, hotspot location and size and centre of gravity of the dominant hand motor area45–48 at baseline, post-TMS and post-CBIT as a potential measure of motor control change. Motor mapping has been shown to be feasible in Tourette’s syndrome.49 Given, its novelty, motor mapping will be exploratory. Third, behavioural symptoms will be measured with the Conners-3 parent rating scale,50 Children’s Yale-Brown Obsessive-Compulsive Scale,51 Multidimensional Anxiety Scale for Children52 and Paediatric Quality of Life (PedsQL)53 at baseline, post-TMS and post-CBIT treatments. Tic symptom severity will be measured once a week during the TMS treatment weeks using the YGTSS and a video assessment. YGTSS and the Modified Rush Video Scale54 will be conducted at baseline, post-TMS and post-CBIT treatment to thoroughly quantify tic expression change with treatment. We hypothesise greater tic reduction in the active rTMS +CBIT treatment arm compared with sham rTMS +CBIT treatment.

Sample size

The sample size was estimated with G*Power (V.3.1.9.4).55 Prior pilot rTMS data (n=10) and a previous randomised controlled CBIT study (n=126) informed the estimates of effect size. Piacentini et al37 reported a mean difference of 4.1 (95% CI, 2.0 to 6.2) and a standardised mean difference of 0.7 in total YGTSS scores between a CBIT and control group.37 We expect the added effect of rTMS to be of the same magnitude, and thus we expect to see a difference in change of at least 4.1 between rTMS plus CBIT compared with CBIT alone. We estimate the SD of change to be 5.85 based on the reported Cohen’s d of 0.7. We will use the Benjamini-Hochberg approach to adjust alpha, with a false discovery rate of 5% and a minimum power of 80%. We assume a potential data loss of 20% due to withdrawal, loss of follow-up, data acquisition failures/errors and so on. With this difference of 4.1 and SD of 5.85 our ideal sample size calculation is 31 per group (62 total), 80% power. However, the funding received for this study allows for n=25 per group (50 total): with 25 per group, as long as the SD of pre to post change is not bigger than 5.31 (alpha (α)=0.033 as per Benjamini-Hochberg), we will still be powered to detect a desired 4.1 change. In addition, 20 participants per group (due to potential data loss/drop out) is still powered to detect a 4.1 difference as long as the SD of the pre to post change is not bigger than 4.73 (α=0.033).

Participants will be recruited through the Calgary Tourette’s Syndrome Clinic (ACH, Calgary, Alberta—led by TP), a network of paediatricians, TS families and advertisements.

Assignment of interventions: allocation

Participants will be randomly assigned into either active or sham rTMS plus CBIT in a 1:1 ratio. The sequence will be generated with random sized blocks of 2, 4 and 6, and use the Research Electronic Data Capture (REDCap) randomisation module for recruiters to access subject allocation. The algorithm will stratify by sex and age group (7–10, 11–14, 15–18 years). On allocation, a participant ID will be generated for each participant to allow for blinded analysis. An unblinded TMS specialist that is not involved in clinical assessments or symptom monitoring will set up the rTMS coil for active or sham delivery.

Assignment of interventions: blinding

Participants, their parents and all raters responsible for clinical evaluation will be blind to group status. Blinding will remain in effect until study completion through the anonymised participant ID, and separation of clinical and randomisation data: only AN-A and the TMS specialists responsible for TMS treatment coil set-up will be unblind to treatment condition.

The only time unblinding may occur is if a safety concern has arisen. This includes any adverse event or significant intolerability that results in participant drop-out. AN-A would be responsible for unblinding and communicating allocation with necessary providers.

Data collection and management

Assessment data will be collected by trained study staff (table 2). On trial initiation, or addition of new staff members, two raters will complete clinical assessments simultaneously to determine inter-rater reliability (minimum n=2). Data from TMS neurophysiology and motor mapping measures will be analysed throughout the study period to ensure quality data collection.

To promote participant retention, participants and their families will receive an educational tour of research space, including the opportunity to go inside a mock MRI scanner and feel TMS stimulations. The schedule and reasoning of study visits will be thoroughly explained prior to study enrolment, and therefore we expect very high retention rates. In fact, we expect a compliance of 99% based on a recently completed open trial of rTMS in youth with treatment-resistant depression conducted by our laboratory.56 Deviations from the standard study protocol (eg, rescheduling a CBIT or rTMS visit due to illness) will be thoroughly documented in paper and electronic databases.

Clinical assessment data will be entered and stored in REDCap, with data exported to excel after participant completion for secondary electronic data storage. REDCap requires two-factor authentication and Excel databases will be password protected. All assessments can be completed electronically by participants and parents for self-report and parent-report respectively through the REDCap survey feature. Paper assessment data will be stored in locked filing cabinets, and only the informed consent form will contain the participant’s name: all other assessments are anonymised by participant ID, including REDCap and excel databases. Imaging data are labelled with participant ID when collected (not name) and stored on secure servers (password protected). TMS assessment data are also labelled with participant ID when collected and stored on secure servers. All assessment data are anonymised by participant ID and is only accessible by study staff involved in data collection or analysis.

Statistical methods

Our primary outcome is Tourette’s symptom change (total YGTSS score) from baseline (week 0) to post-treatment (week 11): within-subject changes will be calculated using a paired t-test to determine changes in YGTSS scores within each group (α=0.017; determined using a Benjamini-Hochberg approach). With these change variables we will conduct a one-tailed two-sample t-test of within subject changes between the active and sham rTMS treatment arms (α=0.033). If data violate assumptions, we will use the Wilcoxon rank-sum test (non-parametric approach). Our secondary outcomes look at metabolite concentration and neurophysiology measure changes from baseline (week 0) to post-treatment (week 11): two-tailed two-sample t-tests of the within subject changes will be conducted for imaging and neurophysiology measures (weeks 0–11). Change across time (comparing weeks 0, 7 and 11) is an exploratory outcome of this study that will use a linear mixed effects model with time, group and time-by-group interaction as fixed effects and participant as random effect.

Data from participants who withdraw from the study, voluntarily or due to non-compliance, will not be included in statistical analyses. The Last Observation Carried Forward Method will be used if data are missing between weeks 1 and 10. Finally, no interim analyses are planned: stopping the entire trial will be a consideration if two or more significant adverse events (eg, seizure) occur, and the study neurologists (AK and TP) will advise such a decision.

Participant safety

MRI technologists will assist in ensuring safe MRI protocols (ie, no metal in scan area, ear protection during scan). TMS procedures will be performed by experienced staff trained in management of potential adverse events. There is a very small risk of seizure associated with TMS, reported primarily in people with pre-existing seizure disorders: the seizure incidence with TMS is estimated at ~0.01%–0.1% compared with 0.07%–0.09% spontaneous incidence in the general population.57 Safety and tolerability scales will be used to monitor TMS safety. Brief discomforts such as headache, neck pain or tingling in the hands are sometimes reported: these are usually rated as mild and typically disappear on their own within an hour of TMS completion. Our academic paediatric centre has obtained standardised safety and tolerability data from paediatric participants using both hand-held and robotic TMS, and over 3.5 million TMS stimulations were shown to be safe and well-tolerated in children.58

If an adverse event of sufficient magnitude occurs, the parent will be offered support to take the child to the emergency department or home. The PI will be responsible for reporting serious adverse events and any unanticipated problems to the local research ethics board. The study statistician (AN-A) will prepare reports that list adverse events, serious adverse events and disorder-specific or treatment-specific events required to ensure good clinical care, and to identify emerging trends.

Participation will be terminated under the following conditions: (1) if serious adverse reactions occur (eg, seizure), (2) serious complications or severe worsening of symptoms (clinical judgement of TP and AK), (3) if a participant needs or wants to withdraw from the study and (4) if the participant is not cooperative or physically unable to take part in the study (eg, RMT is greater than TMS machine output ability).

Oversight and monitoring

In accordance with Good Clinical Practice Guidelines, the Trial Steering Committee will manage the trial: all lead investigators of this paper (FM, AK, TP and GW) are steering committee members. The Trial Management Committee (CKK, RS, SC and FM) are responsible for day-to-day running of the trial, and the Data Safety Monitoring Board are responsible for safeguarding the interests of trial participants, potential participants and investigators (EZ and AN-A).

This study does not have a data monitoring committee. The investigative team, particularly the study coordinator, will be responsible for ensuring data are complete and accurately entered. Inter-rater reliability will be performed for assessment protocols, and integrity ratings will be made on 10% of CBIT sessions via video recording akin to Piacentini et al.37 Protocol modifications will be approved by the local research ethics board and logged on ClinicalTrials.gov. When appropriate, changes will be communicated to participants (eg, a change in assessment tool or study timeline). The study is supported by grant funding, which plays no role in data monitoring or auditing.