A Phase 2 Trial of Deep Brain Stimulation for Spasmodic Dysphonia

Overview

Spasmodic Dysphonia (SD) is a focal, action-specific movement disorder with prominent effects on speech (1, 2). Patients with SD lose their ability to speak normally due to involuntary contractions of their laryngeal muscles. As a result, SD tremendously affects an individual's quality of life by limiting their ability to communicate effectively. The current standard of care for SD involves botulinum toxin (BTX) injections into the laryngeal muscles. BTX causes a weakness in the injected muscles thereby lessening the spasms (3). The primary neurological problem is not changed but weakening the muscles temporarily diminishes the symptoms. However, BTX therapy is associated with several limitations (3, 4). First, the clinical effect produced by BTX is temporary and repeated injections are required approximately every 3 months. Second, there is a delay in the onset of benefits provided by BTX injections; this delay results in a sinusoidal symptom curve where SD is optimally controlled for only a portion of each treatment cycle and patients' spasms return prior to the next injection cycle. Furthermore, the injections can be very painful and some patients develop antibodies to BTX (3, 4). Oral medications used in dystonia, such as anticholinergics, benzodiazepines, and baclofen, provide minimal relief and have numerous side effects at the doses required to influence a patient's voice. Thus, on basis of these limitations, we set out to explore new and innovative strategies to treat SD and provide patients with long-term benefit. Deep Brain Stimulation (DBS) is a neurosurgical procedure that involves the implantation of electrodes to deliver electrical stimuli to specific brain regions. It is the gold-standard surgical treatment for other movement disorders such as Parkinson's disease and generalized dystonia. During a DBS procedure, an electrode is inserted very precisely into the brain and is linked to a pacemaker implanted under the skin of the chest or abdominal wall. When the pacemaker is switched on, a very small electric current passes into the brain, blocking the damaging signals that cause the condition.

Methods \& Experimental Design: This study will combine a Phase 2 randomized clinical trial to assess the benefits of DBS therapy on SD with a neuroimaging study designed to elucidate the neural pathways for speech. The Surgical Study \& Timelines: A prospective, randomized, double-blinded, crossover, Phase 2 trial will be conducted in n=10 patients with adductor SD patients and n=10 patients with abductor SD. DBS electrodes will be placed in the left thalamus (ventral intermediate nucleus) using the same methods as reported in our Phase 1 trial (5). Patients will be randomized to either active stimulation (treatment arm) or no stimulation (control arm) for 3 months, then cross-over for another 3 months, and conclude with un-blinded open stimulation for 6 months. The purpose of the cross-over design is to evaluate the patients in a double-blinded manner. This will eliminate the placebo effect which is well known to influence clinical outcome and has been a confounding factor in all previous botox and surgical trials. Sample Size and Recruitment: We plan to recruit ten adductor SD patients for this trial. Our patient recruitment population will be from the province of British Columbia with a population of 5 million and approximately 500 patients with SD, many of whom have been receiving BTX treatment for \>10 years. Fourteen patients with adductor SD have already been referred for consideration. Patients will be invited to join the trial if they have been diagnosed with adductor SD by our experienced laryngology team. An additional ten patients with abductor SD will be recruited in a parallel trial to assess the efficacy of DBS for that rare condition. Due to the rarity of abductor SD, an international collaboration with the Indiana University Health Voice Centre may be required. Inclusion Criteria: We are studying spasmodic dysphonia and it will be important to be certain of the diagnosis. There is no one specific test for the diagnosis of SD. We will therefore have two separate teams evaluate the patients to ensure a consensus of agreement on the diagnosis. Phase 1: Patients will be recruited from the Vancouver SD Clinic and will be evaluated by a laryngologist (Dr. Hu), a speech language pathologist (Professor Rammage), and a psychiatrist (Dr. Howard). Those that are candidates for surgery will then be flown to the Indiana University Health Voice Centre for phase 2 assessment. Phase 2: Patients selected from phase 1 will be then evaluated by the team at the Indiana University (IU) Health Voice Centre. This will include a laryngologist (Dr. Halum), and Speech Language Pathologist (Dr. Patel). Based on standard clinical evaluation with speech tasks designed to promote characteristic voice breaks in adductor and abductor spasmodic dysphonia, Dr. Halum and Dr. Patel will assign the patient a clinical diagnosis. Those with agreed upon clinical diagnosis (100% inter-rater agreement) of adductor or abductor spasmodic dysphonia will then be evaluated with high-speed videostroboscopy (HSV) for more detailed evaluation of the patients' dysphonia, as Dr. Patel has previously described for the assessment of spasmodic dysphonia (6,7). HSV recordings will be performed with PENTAXMedical, model 9710, at 4000 frames per second and acoustic recordings will be performed with PENTAXMedical CSL Model 4500 with a Shur Beta 53 microphone at a fixed mouth-to-microphone distance of 4 cm in a quiet room with a sampling rate of 44KHz as previously described by Dr. Patel (8). Simultaneous acoustic and HSV recordings will be obtained for sustained vowel /a:/ and voice onset task of 'hehe' at habitual pitch and loudness (symptomatic task) for at least 3 seconds and whisper /a:/ for a total of 3 trials each (13). HSV examinations will be conducted using a standard flexible nasal fiberscope (Pentax FNL-10RP3) introduced unilaterally through the naris and/or a rigid seventy degrees endoscope without application of topical anesthetic to the nasal mucosa. Acoustic recordings before and after DBS implantation will be conducted in a quiet room, in a seated position at the IU Health Voice Centre. Three trials of sustained vowel /a:/, connected speech sample of reading a phonetically balanced 'Rainbow Passage,' and all-voiced sentence, 'Early one morning a man and a women were ambling along a one-mile lane running near rainy island avenue,' (9,10) will be used. If there is uncertainty regarding the diagnosis (for example, the breaks seen on high-speed videostroboscopy are not congruent with the clinical diagnosis) then additional testing with hooked wire electromyography (EMG) will be performed to localize the muscular breaks. Only patients with a confirmed diagnosis of spasmodic dysphonia by consensus with the two teams of laryngologists at the University of British Columbia and the Indiana University Health Voice Centre will then be entered into the study. Analysis: The primary endpoint will be the Voice-Related Quality of Life (V-RQOL) reported by the patients after three months of blinded DBS-ON or DBS-OFF. Results for the cohort will be statistically compared by a Wilcoxon analysis for paired non-parametric measures with the level of significance set at p\<0.05. There will be multiple secondary outcomes. During the blinded phase of the study, the patients' objective voice function will be measured by a speech-language pathologist blinded to the settings using the overall Unified Spasmodic Dysphonia Rating Scale (USDRS). During the open phase of the study, patients will be assessed before and 1-year after the trial with the Communicative Participation Item Bank (CPIB) (11), Beck Depression Inventory version II, and the Montreal Cognitive Assessment. The CPIT has been validated as an outcome measure for SD and its development has been supported by the NSDA. Patients will also be objectively compared pre-operatively and 1-year post-operatively using high-speed videostroboscopy measurements of voice onset time and vibratory breaks as well as acoustic analysis of voicing percentage and Cepstral peak prominence. Outcomes will be compared statistically with a Wilcoxon analysis for paired non-parametric measures. These analyses are designed to generate hypotheses (not answer our primary outcome) and will therefore not be subjected to statistical adjustment for multiple comparisons. Progress of Trial: Ethical approval for the original DEBUSSY trial was obtained from the University of British Columbia Clinical Research Ethics Board (H15-02535) and a new ethics proposal has been submitted for this current trial of twenty patients with a similar protocol. Fourteen patients with adductor SD and one patient with abductor SD have already been referred for consideration of participation in the trial and will be evaluated once the trial is initiated. Risks of Neurosurgical Intervention: The reported risk for DBS surgery varies in the literature but most centres quote a less than 1% chance of a stroke that can be potentially devastating or lethal and a 5% chance of infection or technical malfunction.11 At our centre, the last 1,000 DBS operations resulted in no deaths, 3 (0.3%) strokes within 30 days, and a 2% infection rate, the latter can be treated with antibiotics. The Neural Network for Speech in Spasmodic Dysphonia: All patients in this study will have detailed pre-operative neuroimaging with MRI. The anatomical images will be used for surgical planning. The functional imaging (fMRI) will be used to determine speech lateralization. Only those patients with highly lateralized language function in the left cerebrum and right cerebellum will enter the study. Those with bilateral or right hemisphere speech dominance will be included in a separate study using bilateral surgery and blinded outcome assessment of each electrode to determine speech lateralization in this cohort. Diffusion tensor imaging will look for anatomical differences in the brains of patients with SD compared to normal controls. The white matter tracts (WMT) around the motor thalamus have yet to be investigated as having implications in speech motor control (12). Our group has shown that motor thalamic DBS ameliorates the vocal dysfunction in SD. Using Diffusion Tensor Imaging (DTI), we hypothesize that the Dentato-Rubro-Thalamic tract is the WMT that is most likely modulated by DBS and therefore involved in speech. We will also map other WMT in the region to determine if they are involved in human speech production (13,14). Significance: The current treatment for SD includes repeated botulinum toxin injections in the throat. This standard of care is helpful for many patients - but not all. Some patients with adductor SD and most patients with abductor SD are not improved. What options do they have? With the help of the National Spasmodic Dysphonia Association, our goal is to evaluate the promising potential of DBS in treating all types of SD. We hope to provide an additional treatment option for those patients with SD who are not benefitting from the current therapies. We also hope to utilize this neurological technique to better understand the cause of SD and the pathways for speech. This new understanding may lead to novel medical (non-surgical) treatments of SD in the future.