Clear-Brain: A partial randomised clinical trial investigating stimulation of the glymphatic system by either deepening sleep with lower-sodium oxybate or inhibiting cortical spreading depressions with non-invasive vagus nerve stimulation, or both, in patients with Cerebral Amyloid Angiopathy (CAA).

Spontaneous intracerebral haemorrhage (ICH) is a type of stroke caused by
rupture of a cerebral vessel within the brain parenchyma. Although ICH accounts
for only a minority of strokes (±20%), it is associated with a
disproportionately high rate of mortality and morbidity. Cerebral Amyloid
Angiopathy (CAA) is one of the major caauses of ICH and vascular dementia in
elderly. Approximately 60% of all lobar (cortical) ICHs are CAA related. In
CAA, vessel rupture is caused by an accumulation of Aβ, in the leptomeningeal
arteries, cortical arterioles and capillaries of the brain, which disrupts the
vessel wall integrity. Eventually this damage also leads to cognitive decline
and vascular dementia.

Confirmation of diagnosis of sporadic CAA (sCAA) is achieved through brain
autopsy, whilst a diagnosis of probable sCAA can be made by a combination of
clinical characteristics, brain imaging and, when available, pathology.
Dutch-type hereditary CAA (D-CAA; also referred to as Hereditary Cerebral
Haemorrhage With Amyloidosis-Dutch type or HCHWA-D) is a hereditary form of CAA
in which amyloid deposits are formed due to a mutation at codon 693 of the
amyloid precursor protein (APP) gene on chromosome 21. Patients with D-CAA have
a severely increased risk of ICH from a relatively young age. Since the
underlying pathologies are similar, D-CAA can be seen as a pure form of CAA
with an accelerated clinical course.

Currently, there is no treatment to cure or decelerate D-CAA or sCAA. One trial
examines the effectiveness of minocycline as a therapy, but is still in the
clinical phase. Other studies are exploring the clinical course and evaluation
of the disease, which hopefully will lead to target points for eventual
treatments. With Aβ-accumulation the vessel wall being the hallmark of CAA and
lack of proof of increased production of Aβ, it is hypothesized that CAA is in
essence a brain clearance disease.

The brain does not have a classic lymphatic system for the clearance of
interstitial fluid and neuronal waste products. Relatively recently the
glymphatic system was discovered as a macroscopic waste clearance system that
rids the brain of waste products. This system also clears Aβ via cerebrospinal
fluid (CSF) via perivascular spaces (PVS) into the subarachnoid space, where
dural lymphatic vessels and venous uptake via arachnoid granulations support
further egress out of the cranium. Details on CSF and ISF flow are still
debated with the original models of IPAD and glymphatics being updated
continuously; in this protocol we will be using the term *Glymphatics* to
describe the brain clearance system via perivascular spaces, but not in the
strict definition of the original paper, i.e. we explicitly include more recent
findings and adaptations under this term. The Radiology department of the LUMC
has recently developed a new way to depict CSF-mobility, in large CSF spaces
such as the ventricles but also in smaller ones like PVS. This can be done by
using the CSF-Selective T2-weighted Readout with Acceleration and Mobility
encoding (CSF-STREAM) technique at high-field (7T) MRI. The developed technique
is non-invasive and repeatable and therefore useable for disease monitoring and
for studying brain clearance in longitudinal fashion.

Several studies show that the glymphatic system is mostly active during sleep
and especially deep sleep. LXB is a naturally occurring neurotransmitter and a
psychoactive substance that deepens sleep (increased slow wave). It is safely
and effectively used in treatment for the primary sleep disorder narcolepsy in
adults and children and and Idiopathic Hypersomnia (IH) in adults. Sodium
oxybate (SXB) has the same active moiety as LXB. Several studies assessed the
efficacy and safety of SXB in various other disorders such as alcohol
withdrawal syndrome, fibromyalgia and Parkinson*s Disease. Thus, LXB can be
used safely for deepening sleep and we would also expect for stimulating the
glymphatic system.

Cortical Spreading Depolarisations (CSDs) are a mechanism that appears to
disrupt the glymphatic system. CSD*s are pathological waves of neuronal and
glial depolarisations throughout the cortex, the underlying cause of migraine
aura, transient focal neurologic episodes (TFNEs) and play a detrimental role
in secondary damage after stroke and microinfarcts. A recent study has shown
that CSD*s produce a reduction of outflow of interstitial fluid into the PVS as
a consequence of a rapid closure of the PVS*s. Therefore CSD*s may play a vital
role in brain dysfunction. The vagus nerve encompasses an intriguing network of
neuro-endocrine-immune modulating fibers with connections to multiple brain
regions. Research in animal models has shown that (invasive as well as
non-invasive) stimulation of the vagus nerve is very effective in inhibiting
spreading depolarisations (SD*s). Furthermore, vagus nerve stimulation (VNS)
has been tested safely for several purposes in humans. Although the effect of
nVNS on the glymphatic system in humans is unknown we expect an effect in CAA
patients specifically due to the high incidence of migraine with aura, TFNEs
and cerebral cortical microinfarcts in these patients.

Our hypothesis is that treatment of D-CAA and sCAA patients with LXB, nVNS or
both will improve clearance of Aβ by stimulating the glymphatic.

Objectives:
The main objective will be to assess whether treatment with nVNS, LXB or both
interventions will increase the clearance of Aβ from the brain, compared to
pre-treatment, in patients with CAA. Second objective is to study whether CAA
disease progression is modified by increasing Aβ clearance with disease
modification measured by assessing haemorrhagic and non-haemorrhagic imaging
markers on 7T MRI.

Exploratory objectives:
An exploratory objective is to measure the activity of the glymphatic system
before and after the interventions, compared to the same time period without
intervention, by using CSF mobility scans on ultra-high field 7 Tesla MRI.

Overall long-term goal:
To prevent CAA related ICH and cognitive decline with new disease modifying
treatment. Data from Clear-Brain! will be used to power a larger clinical
trial. Data and (left over) material will be stored in the LUMC Biobank
Neurological Diseases for use in future research.

Our study design is a randomised controlled proof-of-concept trial with
randomisation in three groups of 20 persons with CAA. Patients will receive
LXB, nVNS or a combination of both interventions. The study period will take
six months and the intervention will start at 3 months until 6 months.

Study period
- T0 = 0 months (baseline; MRI 7T, medical questions and several
questionnaires; in subjects who will receive LXB also screening for sleep apnea
by polygraphy)
- T1 = 3 months (start intervention; MRI 7T, lumbar puncture, collection of
blood samples and several questionnaires)
- T2 = 6 months (end intervention; MRI 7T, lumbar puncture, blood withdrawal
and several questionnaires)

Before inclusion participants will be screened for MRI safety by means of an
MRI checklist. To minimize the number of study visits, the screening - in
subjects who will receive LXB - for sleep apnea with a questionnaire and
polygraphy will be done at baseline. During the intervention with LXB
participants will be screened for developing sleep apnea with a validated
questionnaire.

To avoid unnecessary burden for patients, we will use LUMC Biobank Neurological
Diseases data from the placebo group from the current BATMAN trial as a control
group. Informed consent for further research was already obtained during the
start of the BATMAN trial (P19.110). This is a randomised controlled trial in
the LUMC with minocycline versus placebo as intervention for 3 months.

Medicinal product: Xywav* (LXB)
Investigational product: gammaCore Sapphire*

60 Patients will be randomised and receive either LXB, nVNS or both.
If patients are randomised to nVNS, two stimulations of two minutes each will
be applied in the neck twice per day, using the hand-held gammaCore Sapphire*.
One stimulation needs to be done before bedtime, within one hour before
sleeping. The second stimulation needs to be done within an hour after wake-up.
Except for the last day (T6), than the nVNS needs to be done within one hour
before the start of the MRI 7T. Depending on the preference of the participant
the amplitude of the gammaCore Sapphire* will be between 15 and 25. The
stimulation protocol was developed in close collaboration with electroCore*,
the manufacturer of the gammaCore Sapphire*. They have an extensive experience
with this device. Stimulation will be explained to the patient by a pool of
trained researchers.

LXB is used in treatment of narcolepsy with a safe and effective starting dose
of 4.5 g per night divided in two intakes. In treatment of narcolepsy the first
intake is administered immediately before bedtime and the second intake 2.5-4
hours later. The starting dosage can be changed with 1.5 g per night (divided
in two intakes) until the maximum of 9 g per night (divided in two intakes) is
reached. Between every change of dosage a period of at least 1-2 weeks is
necessary. Recently a study showed a safe and efficient treatment of idiopathic
hypersomnia with SXB not only twice-nightly with a maximum of 9 g per night but
also once-nightly with a maximum of 6 g per night. In this study LXB will be
administered orally in fluid form with a starting dose of 2 g before bedtime.
Because of a short half-life of 30-60 minutes the subjects need to take the LXB
immediately before going to bed. Initially patients will only take one dose per
night. The dose will be titrated with a weekly increase of 0.5 g until a dose
of 5 g is reached after 6 weeks. Therefore, each patient will have a stable
dose period of at least 6 weeks. If patients suffer from side effects a lower
dose than 5 g will be accepted and if patients suffer from LXB initiated
hypersomnia a second dose 2.5-4 hours after can be added. When a second dosage
is necessary a maximum of 9 g is allowed, preferably equally divided in two
doses but if needed unequally division is allowed. Furthermore, certain
anti-epileptica (valproic acid, phenytoin, ethosuximide and topiramate) can
interact with LXB and therefore a dosage reduction of 20% should be considered
when these anti-epileptica will be continued during the intervention period.

The burden mainly consists of three study visits, blood withdrawal, 7 Tesla MRI
and two lumbar punctures. Travel costs will be reimbursed for the study visits.
The measurements are routine procedures at the Department of Neurology. Lumbar
puncture will be performed by experienced physicians. We will use atraumatic
spinal needles to reduce the risk of post-lumbar puncture headache.
Participants will be informed extensively about the potential risks of these
procedures, after which written informed consent will be obtained. To avoid
unnecessary burden for patients, we will use the placebo group from the current
BATMAN trial as a control group in the analysis of amyloid-beta. The risks of
MRI are minimal and comparable to the risk of everyday life. Contra-indications
will be carefully examined per subject. Burden will be kept to a minimum by
using short protocols. The burden of respiratory polygraphic measurement will
be minimal because we will use sleep wearables, which can be used at home.
Invasive VNS (iVNS) has already been approved for treatment of depression and
refractory epilepsy and was safe in a small pilot study in ischemic stroke
patients. Non-invasive VNS (nVNS) has not yet been studied in CAA patients so
the safety profile for this condition is not fully known. However, based on
results in earlier trials, we don*t expect major risks to be associated with
nVNS. Since LXB is already an established treatment for cataplexy or excessive
daytime sleepiness (EDS) in patients 7 years of age and older with narcolepsy
and Idiopathic Hypersomnia (IH) in adults, we feel that the risks are small.
Benefit for the participants may be improved sleep when using LXB. If LXB and
nVNS proves to be effective for improving Aβ clearance, participants may
benefit from participating. Furthermore, participation in this study will
result in further insight in treatment options for CAA.