Transcranial Ultrasound Stimulation for Tourette Syndrome

Objectives. The release and the suppression of tics in humans are behavioral outputs that reflect the activity of a complex brain network encompassing multiple deep and cortical brain regions, as delineated by structural and functional neuroimaging1-5. This network encompasses different frontal and parietal cortical areas, as well as the basal ganglia and the thalamus. Existing non-invasive neuromodulatory technologies, such as transcranial magnetic stimulation or transcranial direct current stimulation, are traditionally limited to cortical regions, and cannot reach deep structures, e.g., the basal ganglia or the thalamus6. This limitation can be overcome by the application of low energy transcranial ultrasound stimulation (TUS) integrated with MRI tracking-based neuronavigation and electronic steering of the focal stimulation spot. The TUS program at the University of Calgary and Hotchkiss Brain Institute is at the international forefront of clinical applications of this technology to people living with movement disorders and is therefore in an excellent position to test tolerability and early efficacy of TUS in tic disorders (Tourette syndrome, TS). The proposed research has the following aims: Aim 1: to explore the tolerability of TUS targeting deep and cortical brain regions when applied to older adolescents and adults with a diagnosis of TS. Aim 2: to explore if TUS over the intralaminar thalamic nuclei and the insular cortex is more effective than sham in improving tic frequency and severity, whether dual-target stimulation is more effective than single-target stimulation, and whether multiple session stimulation is more effective than single session stimulation protocols. Background and Novelty. Tourette syndrome (TS) is a highly prevalent neurodevelopmental condition characterized by tics, which are stereotyped movements and/or vocalisations. Tics often cause difficulties in daily life, and many with TS express a desire to reduce and/or gain control over them. A particular challenge is that no singular effective treatment exists for TS. Pharmacological and behavioural interventions can have variable effectiveness, with limited access and side effects being barriers to treatment7,8. Neuromodulation of tics has been developed to tackle these limitations, including functional surgery (deep brain stimulation, DBS) and non-invasive options. The most used target for DBS in severe, medically refractory tics is the centromedian-parafascicular/ventralis oralis internus (CM-Pf/Voi) complex, a crucial node within the brain's cortico-basal ganglia network, implicated in the pathophysiology of TS9. Prospective cohort studies and meta-analyses have shown an average improvement with DBS of 40%-50% in severity scores at long-term follow-up9. Stimulating this target is thought to modulate aberrant signaling in this circuit, potentially by normalizing connectivity with motor cortical areas like the insula and primary motor cortex. The procedure is generally considered safe, but potential adverse events include infection, hemorrhage, and stimulation-related side effects such as temporary dysarthria, paresthesia, and visual disturbances. More recently, non-invasive stimulation approaches, e.g. transcranial magnetic or direct current stimulation, suggested the SMA as the most promising cortical target, in both single and multiple session protocols, but effects have been variable, and these techniques are limited by the variability of the magnetic or electric field spread out, resulting in limited targeting precision6,10. Transcranial Ultrasound Stimulation (TUS) is an emerging non-invasive modality based on low-energy (\<100 J), pulsed, focused ultrasound emissions that can cause a transient neuromodulatory response in brain regions, lasting up to 60 minutes11,12. Following earlier animal studies13-17, recent human studies reported that TUS can modulate functional connectivity18,19 and alter motor-evoked potentials when targeting the primary motor cortex10,20. Work by several groups, including ours, has established the capability of TUS to produce either inhibitory11,21-23 or excitatory12,19,24-29 neuromodulatory responses. TUS can target deep structures such as the thalamus and the basal ganglia. TUS is often combined with frameless neuro-navigated systems, some of which using optical tracking, which is prone to targeting tracking errors. Our TUS program in Calgary tackled this limitation by integrating standard optical tracking-based neuronavigation with an increase in the focus size using a phased-array transducer, which allows for electronic steering of the focal spot. This steering is guided by a simulation software called BabelBrain that mitigates potential targeting errors. Using this technique and inhibitory TUS parameters, our group has successfully targeted the ventral intermediate nucleus (Vim) of the thalamus30, a well-established surgical target for tremor, leading to an acute response of significant decrease in tremor magnitude in patients with essential tremor, with a sham-controlled study underway. The application of this technology is very well tolerated, with rare and very mild undesired effects (headache, fatigue), although single session effects appear to be short-lived and not outlast the day of the stimulation, which is why testing multiple dosing is necessary and timely. The proposed study is the first study in humans to apply TUS to persons with a diagnosis of TS. A recent report employed low-intensity TUS of a secondary motor cortical area (M2) to an established, toxic (bicuculline) mouse model of tic disorder, demonstrating significant reduction of the frequency of tic-like movements31. Our study would provide the first evidence of safety and tolerability of repeated sessions of TUS in this specific population. This is also the first study designed to deliver inhibitory and excitatory TUS protocols to a deep and a cortical brain region independently. To date, TUS is the only technology capable of achieving this. Given the established connectivity between CM/PF thalamic nuclei and insula, this design explores the modulatory potential of TUS on tics at a 'network-level' rather than at the level of single brain target regions. Finally, by evaluating the effect on tic generation and suppression after 1, 2 and 3 sessions for each stimulation condition, the researchers' design would allow an initial assessment of the dose-response relationship of TUS on tic outcomes. The spread of 3 sessions across 5 days would also allow the measurement of tic frequency and intensity using both measures of acute effect (video-based tic rating) and medium-term effect (rating scale referred to a 7-days' time period). Methods. Participants and clinical assessment. This is a phase 1, double-blind, crossover trial in which participants will undergo the same intervention delivered in three modalities over three separate weeks, with at least 7 days of washout between conditions: a) TUS of the CM/Pf/Voi only; TUS of the insula only; sham TUS. Twenty participants aged 18 or older meeting DSM-5 criteria for TS will be recruited through the Calgary Tourette Syndrome adult and pediatric registries at the University of Calgary, which include \>600 participants (PI and Co-Investigator Dr. Pringsheim). Participants will need to have a total motor or vocal tic severity sub-score of at least 15, or a total tic severity score greater than 22 on the Yale Global Tic Severity Rating Scale (YGTSS), be on a stable medication regimen during the 3 months prior, clinically stable in any psychiatric comorbidities, and without any contraindications for MRI. The primary endpoint will be the percent change on the Rush Video-Based Tic Rating Scale (RVBTRS) score, the only validated instrument to measure current tics objectively through a videorecording as well as the ability of patients to actively inhibit tics, after a single session. Secondary endpoints will include percent changes on the total tic counts and on tic inhibition potency based on the RVBTRS videorecording protocol after a single session and after multiple sessions, percent change on the RVBTRS score after multiple sessions, and the percent change on the YGTSS total tic severity score and the self-report Premonitory Urges for Tics Scale (PUTS) after multiple sessions. Tic counts and RVBTRS scores will be measured independently by two trained evaluators (PI and Research Fellow Dr. Lambert), both blinded to the stimulation arm. The Yale-Brown Obsessive Compulsive Scale (Y-BOCS) will be used to measure obsessive-compulsive symptoms. All scales will be administered by the same trained evaluator, blinded to the stimulation arm. Intervention. A custom 128-element phased-array ultrasound transducer (H317, Sonic Concepts Inc, Bothwell, WA) will deliver TUS using an ultrasound frequency of 250 kHz and a spatial-peak pulse-average intensity (ISPPA) of 10 W/cm2 in situ. For inhibitory effects, we will use the timing parameters developed by our group11 with a pulse repetition frequency (PRF) of 100 Hz, a duty cycle (DC) of 10% and duration of 120s. For excitatory effects, we will use the timing parameters developed by the Toronto Western group with the so-called theta burst protocol32 with a PRF of 5 Hz, a DC of 10% and duration of 80s. A custom 3D-printed cone with a height of 13.9 mm will be filled with deionized and degassed water, and a custom retainer to hold the ultrasound gel between the subject's head and the cone will be used as a matching layer to improve ultrasound transmission. The focus will be electronically steered as required for different patients. Participants will undergo MRI to obtain 1 mm-isotropic T1-, T2-weighted, Zero Echo Time (ZTE), and diffusion tensor imaging images prior to the TUS session. In procedural planning, the bilateral CM/Pf/Voi and insula will be located using diffusion tensor tractography. Imaging will be performed using the 3T GE UHP scanner at the Seaman Family's MRI Imaging Center, UCalgary. To mitigate targeting errors, the researchers will employ the same successful multi-focus strategy with electronic steering, which was used in their recent study on tremor targeting the Vim nucleus of the thalamus, to probe the intended target and radially adjacent foci 5 mm from the centre30. Furthermore, the research team will expand this approach to cover the insular cortex. The electronic steering capabilities of their equipment enable this type of coverage without requiring the actuator to be repositioned. After pre-stimulation assessment, each TUS session will start with neuro-navigation registration using a Brainsight system (Rogue Research, Canada). After positioning the transducer on the target, sonications will start. The software BabelBrain33 will plan the multi-focus strategy. The acoustic simulation will be used to adjust for any offset due to aberrations by adding electronic steering. BabelBrain will also calculate the ISPPA in water required to achieve the intended ISPPA in tissue and to ensure that the mechanical Index (MI) is \< 1.9 and maximal thermal rise is \< 2 °C as recommended by the International Transcranial Ultrasonic Stimulation Safety and Standards34. Sham stimulation sessions will be unfocused, thereby providing the same sound produced by real stimulation but without sufficient intensity at any location. Sample size. Given the exploratory nature of our study, the research team will utilize a sample size of convenience, which is supported, however, by previously published data. In a recent randomized, sham-controlled, parallel group study from our team10, transcranial direct current stimulation of the supplementary motor area detected a significant 'treatment x visit' effect (F\[1,22\] = 3.9, p = 0.03) on YGTSS motor tic severity subscore with a sample size of 12 participants in each of the two treatment groups. The researchers also observed a trend towards increased tic inhibition potency (p = 0.07) with the same sample size. DBS targeting the CM/Pf/Voi in patients with severe, medically refractory TS shows a significant therapeutic effect, with studies reporting a mean reduction in tic severity (YGTSS score) ranging from approximately 50% to over 70%. In terms of standardized effect size measures, a 2024 meta-analysis of studies that investigated participant samples of similar size to the one proposed here35, encompassing various DBS targets (including CM/Pf) reported a large overall standardized mean difference of 1.88 (95% CI: 1.74 to 2.02) for the improvement in tic symptoms as measured by the YGTSS total score. Registry data do not support a significant difference between thalamic and pallidal targets using DBS36. Statistical Analysis and Anticipated Results. The primary endpoint will be analyzed using repeated-measures ANOVA, measuring the effect of 'Treatment' (real CM/Pf/Voi, real insula, sham). Bonferroni adjustments will be used to control physiological change in tic frequency as result of uncontrolled variables. Moreover, repeated-measures ANOVA will be used to analyse the effect of 'Treatment' on all secondary endpoints. Exploratory regression analyses will also be conducted to determine the association between demographic factors, clinical severity, and baseline pharmacological treatment and the effect of real TUS treatment conditions on tic frequency and severity. This is a phase 1, double-blind study in which the researchers aim to evaluate the effect of TUS over CM-Pf/Voi and insula upon tic frequency, severity and tic-suppressing capacity. First, the researchers anticipate target-specific effects. Specifically, like with DBS, they anticipate TUS-mediated inhibition of the CM-Pf/Voi to reduce tic frequency and severity. The research team anticipates TUS-mediated facilitation of the insula to improve tic inhibition potency and, secondarily, reduce tic frequency.