Treatment With Auricular Vagus Nerve Stimulation in Rheumatoid Arthritis - the TRAVAGA Study

Purpose and aim:

The overall aim is to investigate efficacy and safety of a newly developed non-invasive auricular investigational device that electrically stimulates the auricular branch of the vagus nerve to activate the cholinergic anti-inflammatory mechanism to treat rheumatoid arthritis patients not responding to standard therapy. The mode of treatment is termed transcutaneous auricular vagus nerve stimulation (taVNS). Specifically, the investigators will address the following research questions:

1. What is the effect of taVNS on disease activity (ACR20, primary outcome) in rheumatoid arthritis?
2. What are the effects of taVNS on pain, fatigue and functional status in rheumatoid arthritis?
3. What is the safety profile of taVNS in rheumatoid arthritis? Study Design

Survey of the field There is a clinical need for improved and less expensive therapy with less serious adverse effects in rheumatoid arthritis (RA). The aim of the project is to work out such treatment using external, non-invasive activation of the endogenous cholinergic anti-inflammatory pathway, a mechanism discovered by Kevin J Tracey, one of the founders of taVNS AB, the sponsoring company of this clinical investigation. This mechanism is the efferent part of the inflammatory reflex, a neural circuit that counteracts exaggerated dysfunctional inflammatory responses. The anti-inflammatory effects are mediated via acetylcholine released via the vagus system and a subset of circulating T lymphocytes capable of acetylcholine synthesis. The mobile anti-inflammatory T cells operate both within and outside compartments innervated by the vagus system. Alpha-7 nicotinic acetylcholine receptors (a7nAChR) respond to acetylcholine by guiding activities downregulating proinflammatory cytokine synthesis and redirecting the traffic of mobile inflammatory cells. Electrical vagus nerve stimulation (VNS) as well as administration of a7nAChR agonists have mediated promising therapeutic results in experimental arthritis models and in rheumatoid arthritis patients. Using a surgically implanted vagus stimulator Kevin J Tracey and Ulf Andersson (the author of this document) provided the original clinical report of successful VNS treatment in RA. Three subsequent pilot studies in RA cohorts based on similar invasive VNS technology later confirmed the encouraging therapeutic outcome of the original study.

The auricular branch of the vagus nerve is a sensory nerve to the external ear including the cymba conchae region. Stimulation of this auricular nerve branch delivers afferent neuronal impulses, whereas cervically implanted devices deliver both efferent and afferent nerve stimulation. A long experience from therapeutic epilepsy studies has provided reassuring safety results regarding taVNS therapy. Functional magnetic resonance imaging demonstrates that taVNS activates the main vagal afferent pathway through the brainstem to upstream cortical projections in a similar way to implanted cervical VNS electrodes, confirming this nerve as a suitable non-invasive target to administer vagus nerve stimulation. Beneficial therapeutic taVNS effects in joint inflammation and inflammatory bowel diseases (IBD) have been demonstrated in pilot studies. Furthermore, use of taVNS has recently been successful in a proof-of-concept study of RA patients. Sham-controlled therapeutic taVNS studies in RA and IBD patients failing standard treatment are now justified and such studies in RA form the aim of our project.

Study design

Study outline and Patients/Intervention/Control/Outcome (PICO) The study will comprise a multicenter, placebo-controlled, randomized, double-blinded, superiority, clinical investigation in patients with moderate to severe RA (defined as DAS28 - CRP > 3.2) who have failed one conventional disease modifying anti-rheumatic drug (csDMARDs) (commonly methotrexate) and in addition at least one biologic disease-modifying anti-rheumatic drug (bDMARD) (commonly a TNF-blocking agent) or one targeted synthetic DMARD (tsDMARD) (P).

The therapeutic intervention (I) is electrical taVNS in the auricular cymba conchae/cavum region of the left ear during 5 minutes/twice daily for 12 weeks. The sensory innervation of the cymba conchae region is to 100% mediated via the auricular branch of the vagus system. The sham control (C) will be electrical stimulation of the earlobe during 5 minutes/twice daily for 12 weeks. The earlobe is not innervated by the vagus system.

Primary outcome (O) is ACR2019. Primary outcome measure Outcome Measure Measure Description The American College of Rheumatology (ACR) 20 response Difference between cymba conchae treatment and earlobe control groups in the proportion of patients who achieve at least 20% improvement from baseline to day 84 in tender and swollen joint counts (scale 0=best to 28=worst). In addition, 3 out of the following 5 measures: • Health Assessment Questionnaire Disability Index (HAQ-DI) score (scale 0=no difficulties to 3=unable to do) • Patient global assessment (0=best to 10=worst) • Patient pain (0=no pain to 10=worst) • Evaluator´s global assessment (0=best to 10=worst) • High-sensitive C-reactive protein concentration with higher values representing worse outcome

Secondary outcome measure Outcome Measure Measure Description Safety ACR50/ACR70 DAS28-CRP EULAR response rate

Patient-reported outcomes:

Pain Fatigue Assessments of functional status

Adverse events will be collected during the entire study period following MDCG guidance MDCG 2020-10/1 and /2 and will be presented by MDCG term as frequency by study arm. Difference between cymba conchae treatment and earlobe stimulation control groups in the proportion of patients who achieve at least 50 or 70% improvement from baseline to day 84 in tender and swollen joint counts (scale 0=best to 28=worst). In addition, 3 out of the following 5 measures: • Health Assessment Questionnaire Disability Index (HAQ-DI) score (scale 0=no difficulties to 3=unable to do) • Patient global assessment (0=best to 10=worst) • Patient pain (0=no pain to 10=worst) • Evaluator´s global assessment (0=best to 10=worst) • High-sensitive C-reactive protein concentration with higher values representing worse outcome. Mean change of DAS28-CRP between the visit on baseline and day 84. DAS28-CRP provides another composite RA evaluation assessment to combine different basic measures of RA disease activity into a single score. The algorithm takes into account the patient's and physician's overall assessment of disease activity over the last 48 h, the number of swollen and painful joints, and serum level of CRP. The DAS28-CRP measures the overall disease activity on a scale of 0-10.

The EULAR response rate criteria classify individual patients as non-, moderate, or good responders, depending on the extent of change and the level of disease activity reached.

Longitudinal follow-up of pain (VAS scale) at baseline and after 7, 14, 21, 28, 42, 56, 70, and 84 days compared to baseline. Analogous measurements to above and Multidimensional Fatigue Inventory (MFI-20, ordinal). And VAS-scale assessment. Health assessment questionnaire (HAQ1, ordinal). Fatigue (Multidimensional Fatigue Inventory (MFI-20, ordinal), SF-36 (ordinal), EQ 5D (ordinal). Hospital Anxiety and depression scale (HADS, ordinal). PSQI sleep quality questionnaire (ordinal). Brief Pain Inventory (ordinal).

Arm A: Baseline - 12 weeks: Electrical taVNS in the cymba conchae/cavum region of the left ear via TRAVAGUS headset electrodes for 5 minutes twice daily for 12 weeks. Arm B: Baseline - 12 weeks: Electrical sham taVNS stimulation in the left earlobe via two ear-clip electrodes for 5 minutes twice daily for 12 weeks.

All subjects treated with conventional synthetic DMARD or low-dose glucocorticoids will continue at the same stable dose as for 4 weeks prior to the trial. Study patients will be instructed not to discuss stimulation modes with each other.

Patient selection

Inclusion criteria • > 18 years of age on day of signing informed consent • Active RA, DAS28>3.2, at least 4/28 tender and 4/28 swollen joints • Demonstrated an inadequate response, loss of response, or intolerance to one or more approved for rheumatoid arthritis biologic or targeted synthetic Disease-Modifying Anti-Rheumatic Drugs (DMARDs), including Janus kinase inhibitors • Stable dose of glucocorticoids or conventional disease-modifying agents in RA (csDMARDs) at 4 weeks before screening. No intraarticular steroids will be allowed later than 2 weeks before trial start or during the trial period.

Exclusion criteria • History of vagotomy • Partial or complete splenectomy • Recurrent vasovagal syncope episodes • Untreated or poorly controlled psychiatric illness • Significant immunodeficiency due to underlying illness • History of cerebrovascular insult • Clinically significant cardiovascular disease • Uncontrolled fibromyalgia • Pregnancy (if sexually active, using reliable form of birth control or being at least 2 yrs post-menopausal)

Description of the investigational device and the comparator

The TRAVAGUS headset (class I) is manufactured by taVNS AB. The therapeutic taVNS system includes a tens unit device that sends electrical pulse to external transcutaneous auricular electrodes to stimulate the auricular vagus system to activate the cholinergic anti-inflammatory mechanism. This physiological system has the potential to alleviate inflammation and pain and to provide symptomatic relief to subjects suffering from moderately to severely active rheumatoid arthritis. The TRAVAGUS System includes a class IIa, CE-marked (2460) battery-powered, wearable, nerve stimulator, named AXION STIM-PRO X9+, manufactured by Axion GmbH, Leonberg, Germany, that mediates unilateral, transcutaneous auricular electrical nerve stimulation (tens) to nerves located in the cymba conchae and the cavum conchae ear regions. The TRAVAGUS System contains two connected device components. The electrical stimulator is connected via a cable to an external, the wearable, neural interface earpiece headset, designed by the sponsoring taVNS AB company and manufactured by OIM Sweden AB in Malmö, Sweden. The TRAVAGUS System delivers safe, charge-balanced, current-controlled, asymmetrical, bi-phasic, square waves via two aluminium electrodes with adequate electrical conductivity without a need for electrode gel application to cutaneous areas in the cymba and cavum conchae region of the left ear. A second advantage with the design of the TRAVAGUS electrode headset is that the diadem provides a pushing force on the electrodes to optimize conductivity. Most other taVNS devices use electrodes that are just positioned in the auricle without any pressure against the skin. That introduces a risk for intermittent or poor conductivity. A third advantage with the TRAVAGUS design is the positioning of the two electrodes. It is most essential that the upper electrode gets into the cymba conchae area, which is exclusively innervated by sensory vagal fibers. The electrodes are fixed in the diadem to provide a user-friendly positioning of the upper electrode. The mission of the lower electrode is to create a circuit of electricity flow by providing a second cutaneous contact in the ear within a wide area of the cavum conchae region. This electrode can be adjusted in length via a threaded montage to enable a good cutaneous contact, since the shape of an auricle varies substantially between individuals.