The Role of Orexin in Human Panic Disorder

Orexins

Hypocretins (orexins), a more recently identified class of pro-arousal neuropeptides, are
synthesized by neurons in the lateral and posterior hypothalamus The main described orexins,
orexin A (ORX A) and orexin B (ORX B), are both cleaved from a common precursor peptide,
prepro-orexin. Orexin A, a 33 amino-acid residue peptide, appears to be conserved across many
mammalian species. Orexin B contains 28 amino acids. Orexins promote a variety of behaviors
including alertness, vigilance, locomotion, fight-flight responses, and feeding. The
physiological effects of the orexins are mediated via 2 G-protein coupled receptors, ORX1 and
ORX2. Orexin A binds with greater affinity to the ORX1 receptor, while orexin A and B bind
with similar affinity to the ORX2 receptor.

Orexins and Animal Fear Models

Orexins have been implicated in anxiogenesis in some animal fear models. For example, central
(icv) injection of orexin A in mice induced anxiety-like responses in the light-dark
exploration test and elevated plus maze.

Using an established, γ-aminobutyric acid (GABA)-deficit, rodent model of
panic-vulnerability, the investigators' Indiana University preclinical anxiety collaborators
provoked a panic-like response in rodents with an anxiogenic sodium lactate (NaLac) infusion,
which response was blunted following either site-specific orexin (ORX) gene silencing or
systemic pretreatment with an ORX1 antagonist.

In addition, ORX neurons (peptidergic neurons in the lateral hypothalamus) were shown, in
turn, to stimulate discrete efferent sites within an emotional network (e.g., bed nucleus of
the stria terminalis) to elicit specific behavioral components of the panic-response
following sodium lactate. Taken together, these results support the concept that ORX
hypersecretion or ORX neuronal overactivity could also be present in human panic disorder
(PD).

Orexins' Emerging Role in Human Anxiogenesis / Panicogenesis

Similar to the NaLac model animals, humans with PD have been reported to have cortical and
subcortical GABA deficits. If these GABA deficits also extend to impairment of GABAergic
inhibition of dorsomedial hypothalamus (DMH) ORX neurons in PD patients, as predicted by the
NaLac animal model, they may result in ORX hypersecretion, increased sympathetic activation,
and panicogenesis.

There have been few clinical studies of ORX metabolism or function in human anxiety
populations. However, recently generated human pilot data in the principal investigator's
(PI's) lab, studying the effects of a well-documented anxiogenic stimulus (35% CO2
inhalation) on behavioral, physiological, and biochemical (plasma ORX A; assayed by a
standard radioimmunoassays [RIA] kit) measures, in 1 PD patient and 2 healthy volunteers. In
this paradigm, the PD patient had a mild panic episode associated with marked early
elevations in plasma ORX levels, relative to the volunteers who had minimal anxiety,
consistent with a role for ORX in the initiation or elaboration of the human panic response.

It was also demonstrated that, in contrast to human subjects without any axis I psychiatric
disorder or with depression alone without panic, only subjects who had high panic scores but
no depression had significantly elevated ORX levels in the cerebrospinal fluid (CSF).

The ORX hyperactivity hypothesis of panic that has been evinced from this work is highly
innovative, and promises to broaden the understanding of the neurobiology of human panic
disorder, as well as provide new treatment directions.

While there are limitations with using plasma ORX A as a measure of central nervous system
(CNS) ORX function, one research group has recently published human data indicating that
resting state CSF and plasma ORX A levels are highly correlated.

Accordingly, the central hypothesis of this translational human pilot project, and a more
definitive project based on it, is as follows: PD is a human anxiety disorder associated with
specific cortical and subcortical GABA deficits that result in disruption of normal
inhibitory regulation of pro-arousal ORX neurons. This disruption promotes excessive ORX
release, sympathetic activation, and vulnerability to spontaneous or chemically induced
panic. Pretreatment with an ORX1 receptor antagonist prior to chemical challenge is therefore
expected to block the evoked panic response.

Rationale for the Use of CO2 Inhalation

The 35% CO2 challenge is well documented in the literature as being reliable, safe, and easy
to administer. The procedure has acceptable test-retest reliability, and may be used to
monitor improvement in clinical status following the administration of antipanic medications.

Approximately 70% of PD patients will have a panic attack in response to this challenge,
which closely resembles a real-life panic. Therefore, in addition to resting/baseline
measurement of plasma ORX A, CO2-evoked levels of plasma ORX A in PD patients will also be
examined, and these responses will be correlated with other behavioral and physiological
parameters recorded during the CO2 test. The PI has had considerable experience using PD
challenge paradigms in clinical research contexts, and he is very familiar with the
application of the 35% CO2 challenge.

Project Aims and Expected Results

The project is a study that gathers pilot data relating to the role of orexin in human panic
disorder. The effect sizes generated from this pilot work will permit planning and powering
of a larger-scale study. It is anticipated that the study will be completed over the course
of one year.

Specific Aim 1 will be to provide a preliminary demonstration that acute administration of
the first-in-class, FDA-approved insomnia agent, suvorexant, a mixed ORX1/2 receptor
antagonist, will block 35% CO2-induced panic symptoms in PD patients, via amelioration of
central ORX neuronal hyperactivity (as reflected in blunted plasma ORX responses to CO2
challenge).

To address Specific Aim 1, a prospective, parallel-group, repeated-measures design will be
used to compare behavioral, physiological, and biochemical (plasma ORX) responses in 2
independent, unmedicated groups of PD outpatients (n=6 in each group) at baseline/resting
state and after panic provocation due to brief (1 minute) inhalation of a 35% CO2 / 65% O2
gas mixture. PD patients will be randomized, in a double-blind manner, to receive either a
single, oral dose of the mixed ORX1/2 receptor antagonist, suvorexant (10 mg dose), or
identical placebo, 120 minutes before CO2 challenge.

Expected results: It is expected that, compared to placebo, suvorexant pretreatment will
blunt behavioral, physiological, and biochemical (plasma ORX) responses to 35% CO2 in PD, due
to suppression of CNS ORX hyperactivity. The effect sizes generated from the pilot work will
permit planning and powering of a larger-scale study, to definitively address Specific Aim 1.

Inclusion Criteria:

- They must be in stable physical health as determined by a medical evaluation,
including physical examination, electrocardiogram, laboratory findings (comprehensive
metabolic panel, complete blood count [CBC], free T4, urine pregnancy test,
urinalysis), urine toxicology screen, and a negative urine pregnancy test in women of
child-bearing potential.

- They must satisfy the new clinical criteria in the Diagnostic and Statistical Manual
for Mental Disorders, 5th edition (DSM-5) for a current principal diagnosis of PD as
confirmed by a semi-structured, diagnostic interview, the Mini International
Neuropsychiatric Interview (MINI), administered by the PI.

- Since clinical depression (MDD) is associated with CSF ORX abnormalities, only
patients with a current PD without MDD will be enrolled. They will also be required to
have a current Montgomery-Asberg Depression Rating Scale (MADRS) total score <12.

- They will be off all regular psychiatric medications and avoid drinking grapefruit
juice for at least 2 weeks prior to the 35% CO2 test.

- They must not be pregnant or breastfeeding a baby; and women of childbearing potential
must be using birth control while on this study.

Exclusion Criteria:

- any history of a psychotic disorder, bipolar disorder, MDD, depression not otherwise
specified (NOS), obsessive compulsive disorder, an eating disorder, post-traumatic
stress disorder, or generalized anxiety disorder

- medical conditions for which suvorexant could be contraindicated, such as narcolepsy

- any other sleep disorder

- a substance use disorder, as defined by the DSM-5, within 6 months of the screening
visit

- ongoing use of psychiatric medications in the 2 weeks prior to the 35% CO2 test

- current use of certain drugs, including

- strong cytochrome P450 3A (CYP3A) inhibitors (such as ketoconazole, itraconazole,
posaconazole, clarithromycin, nefazodone, ritonavir, saquinavir, nelfinavir,
indinavir, boceprevir, telaprevir, telithromycin, and conivaptan);

- moderate CYP3A inhibitors (such as amprenavir, aprepitant, atazanavir,
ciprofloxacin, diltiazem, erythromycin, fluconazole, fosamprenavir, imatinib,
verapamil);

- strong CYP3A inducers (such as rifampin, carbamazepine and phenytoin);

- digoxin

- history of any neurological disorder affecting the CNS

- history of uncontrolled or serious medical illness

- a history of hypersensitivity or allergy to suvorexant

- pregnancy or lactation status, or unwillingness to use birth control while on this
study, for women of child-bearing potential

- compromised lung function (e.g., chronic obstructive pulmonary disease [COPD],
emphysema, idiopathic pulmonary fibrosis, lung cancer)

- inability to fast the required amount of time prior to study visit 2

- a positive test for cannabinoids, opiates, benzodiazepines, amphetamines, cocaine and
metabolites

- out-of-range lab values

- an abnormal EKG

- a score > 12 on the Montgomery-Asberg Depression Rating Scale (MADRS)

- inability or unwillingness to avoid drinking grapefruit juice for two weeks prior to
the 35% CO2 challenge test

- a history of sudden onset of muscle weakness (cataplexy)