Experimental Drugs for Panic Disorder: An Updated Systematic Review

Figures & data

Figure 1 PRISMA flow diagram of study selection process.

Notes: PRISMA figure adapted from Liberati A, Altman D, Tetzlaff J, et al. The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaboration. Journal of Clinical Epidemiology. 2009;62(10). Creative Commons.³⁵

Table 1 Preclinical Studies with Orexin Receptor Antagonists

Authors, Year [Reference]	Compound(s)	Animal Model(s) of Panic	Procedure	Main Measures	Main Results	Other Results of Interest
Bonaventure et al, 2015^Citation45	Compound 56 (selective ORX1R antagonist)	NaLac panic provocation model in panic-prone rats*.	Five day following l-AG infusion onset, rats received, in a counterbalanced design and with 48 hours between crossover, a subcutaneous injection of either compound 56 (at 3, 10, or 30 mg/Kg) or vehicle as a control group 60 min prior to the 15-min NaLac challenge. SI was performed 7–8 days following radiotelemetry surgery recovery and repeated 2–3 days later during drug treatment crossover. On experimental drug testing days, the SI test was performed 5 min after the offset of the NaLac challenge, with different partners each time.	Cardiovascular responses (ie MAP and HR), as measured by a pressure transducer into the femoral artery. General motor activity and CBT, as measured by radiotelemetry probes implanted into the peritoneal cavity and sutured to the muscle wall. All these variables were recorded continuously in freely moving conscious rats, expressed as a 20-minute time course, and calculated as changes from the average of the baseline from each rat. Behavioral responses were measured by SI test.	Compound 56 at its highest dose (30mg/kg) attenuated the reduced SI (anxiety-like behavior), increased locomotor activity, and cardioexcitatory responses induced by the NaLac challenge. The lowest dose (3 mg/kg) was efficacious on locomotor and cardiovascular parameters, but not on SI. The intermediate dose was not efficacious on any parameters except the suppression of NaLac-induced locomotor activity. Compound56 had no significant sedative side effects at any of the doses, as assessed by monitoring baseline locomotion or autonomic activity.	Compound 56 was devoid of any hypnotic effect under basal condition. Compound 56 significantly attenuated the sleep-onset insomnia (ie, prolongation of NREM and REM sleep latencies) elicited by cage exchange, without impacting sleep duration. Compound 56 did not affect the cage-exchange stress-induced ACTH release, suggesting that ORX1Rs are not directly involved in HPA axis activation elicited by this stress model.
Johnson et al, 2015^Citation46	A dual OX1/2R antagonist (DORA-12), compound 56 (SORA1, selective ORX1R antagonist), SB334867 (ORX1R antagonist), JnJ10397049 (SORA2, selective ORX2R antagonist), lorazepam (positive control).	20% CO2-panic provocation model in naïve rats**	Enclosed flow cages with CO2 and O2 sensors were used. In a counter-balanced design (ie, all rats receive each drug treatment with at least 48 hr between treatments), rats were systemically treated with a control vehicle or different doses of compounds, then placed into the chamber where atmospheric air was being infused. All rats had infusions of the following: 5 min infusion of atmospheric gas (<1% CO2, 21% O2, 79% N2) for baseline measurements; then either the control gas or experimental normoxic, hypercarbic gas (20% CO2, 21% O2, 59% N2) for 5 min; and finally 5 min infusion of atmospheric gas. Following exposure to hypercarbic and atmospheric air gases, rats were immediately placed in the open field box for 5 min, then assessed in a social interaction (SI) test for 5 min.	Cardiovascular responses (ie MAP and HR), general motor activity and CBT, as measured by a pressure transducer implanted into the femoral artery and a radiotelemetry probe implanted into the peritoneal cavity. All these variables were recorded continuously in freely moving conscious rats. Behavioral responses were measured by placing rats in an open field box for 5 min and then carrying out a SI test in the same box.	Globally inhibiting orexin activity with DORA-12 attenuated CO2-induced anxiety-like behavior in SI test but did not modify CO2-induced cardiovascular and thermoregulatory responses or locomotor activity. Selectively inhibiting ORX1 receptor activity with SORA1 compound 56 attenuated CO2-induced anxiety-like behavior in SI (at the highest dose) and CO2-induced cardiovascular and thermoregulatory responses (at both highest and lower doses), while it did not modify locomotor activity; less selective SB334876 attenuated CO2-induced anxiety-like behavior in SI test and pressor and thermoregulatory responses, while it did not attenuate CO2-induced HR response or locomotor activity. Selectively inhibiting ORX2 receptor activity with SORA2 JnJ10397049 did not modify any CO2-induced behavioral or physiological response. Lorazepam at doses useful to attenuate CO2-induced behavioral and physiological response presented significant sedative effects.	ORXR antagonists did not present any sedative effects.
Bonaventure et al, 2017^Citation48	JNJ-54717793 (selective ORX1R antagonist)	NaLac panic provocation model in panic-prone rats* 20% CO2-panic provocation model in naïve rats**	Procedure of the experiment with the NaLac panic provocation model in panic-prone rats* was analogue to that used by Bonaventure et al, 2015, described above; procedure with the 20% CO2-panic provocation model in naïve rats** was analogue to that used by Johnson et al, 2015, described above. JNJ-54717793 at 3, 10, or 30 mg/kg (oral doses, 60 min prior the NaLac challenge) and a control vehicle were tested.	Main measures were analogue to those used by Bonaventure et al, 2015, and Johnson et al, 2015, described above.	JNJ-54717793 at its highest dose (30mg/kg) attenuated the reduced SI (anxiety-like behavior) induced by the NaLac challenge, while the intermediate dose (10 mg/kg) partly attenuated pressor response. No effects on NaLac challenge-induced tachycardia were found. The highest dose (30mg/kg) attenuated the CO2-induced pressor response at multiple time points (3mg/kg at one time point), while the 10 and 30 mg/kg doses attenuated HR response at multiple time points. The 30mg/kg dose attenuated/blocked the CO2-induced anxiety-like behavior in the SI test.	JNJ-54717793 had no significant sedative side effects at any of the doses, as assessed by monitoring baseline locomotion or autonomic activity, and behavior in the open field box, and had minimal effect on spontaneous sleep.
Salvadore et al, 2020^Citation49	JNJ-61393215 (selective ORX1R antagonist)	20% CO2-panic provocation model in naïve rats**	Procedure with the 20% CO2-panic provocation model in naïve rats** was analogue to that used by Johnson et al, 2015, described above. JNJ-61393215 was tested using oral doses of at 3, 10, or 30 mg/kg (30–50 min prior to the CO2 challenge); the less active enantiomer of JNJ-61393215, JNJ-63821238, was included as a negative control.	Main measures were analogue to those used by Johnson et al, 2015, described above.	JNJ-61393215 at dose of 10 and 30mg/kg blocked the reduced SI (anxiety-like behavior) induced by the CO2 challenge. No effects on pressor response, CBT, or locomotor activity were found, while JNJ-61393215 at 30 mg/kg attenuated CO2-induced bradycardia at two time points.	JNJ-61393215 had no significant sedative side effects at any of the doses, as assessed by monitoring baseline locomotion or autonomic activity, and behavior in the open field box, and had minimal effect on spontaneous sleep.

Notes: *Panic-prone rats= In this panic model, the inhibitory GABA-mediated activity is chronically disrupted by infusion of the GABA synthesis inhibitor l-allylglycine in the perifornical/dorsomedial (PeF/DMH) hypothalamic region of rats, resulting in a protracted glutamate-mediated activation of this area. As a consequence, rats display decreased social interaction (ie, an experimental anxiety-like behavior in rats) and increased panic-like cardio-respiratory responses (ie, tachycardia, tachypnea, hypertension) following sodium lactate infusion, which is an interoceptive stimulus that provokes panic attacks in humans.³⁹ **20% CO₂ panic-provocation model= In rats, the inhalation of hypercapnic (20% CO₂) gas mixture for 5 min produces decreased social interaction (ie, an experimental anxiety-like behavior in rats), increased locomotor activity, and marked cardiovascular (increased MAP, bradycardia) and thermoregulatory (decreased CBT) responses, partly replicating the well-established panicogenic effects of hypercapnia in humans³⁸. SI test (digitally video recorded) was performed as follows: the “experimental” rat and an “unfamiliar “partner” rat were placed together in the centre of a box, and the total duration (seconds) of nonaggressive physical contact (eg, grooming, sniffing, etc.) initiated by the experimental rat was quantified over a 5-minute duration. Changes in activity relative to the average of the baseline measurement from each rat were considered.

Abbreviations: ACTH, adrenocorticotropic hormone; CBT, core body temperature; CO₂, carbon dioxide; HR, heart rate; l-AG, l-allylglycine; MAP, mean arterial blood pressure; NaLac, sodium lactate; ORXR, orexin receptor; ORX1R, orexin-1 receptor; ORX2R, orexin-2 receptor; other results of interest, other findings relevant to the aims of this review; SI, social interaction.

Table 2 Phase I Clinical Studies with Orexin Receptor Antagonists

Authors, Year [Reference]	Compound(s)	Study Design	Participants	Laboratory Panic Provocation Method	Procedure	Main Measures	Main Results	Other Results of Interest	Side-Effects/Tolerability
Salvadore et al, 2020^Citation49	JNJ-61393215 (selective ORX1 antagonist)	Phase I study. Four-treatment 3-arm 2x2 cross-over design. According to this design, each participant was randomized to receive either placebo or one of the three active treatments [ie, JNJ-61393215 at 25 mg or 90 mg once daily; alprazolam (active comparator), 1 mg twice daily]. Single site, the Netherlands.	Thirty-nine healthy adult male volunteers who showed sensitivity to the 35% CO2 double-inhalation panicogenic challenge during screening	A protocolized administration of inhaled 35% CO2 was conducted using the CTT. Briefly, participants took a double VC breath of a 35% CO2 and 65% O2 (normoxic) gas mixture; they were motivated to inhale at least 80% of the previously measured VC. Blood pressure and heart rate was assessed through a finger cuff, connected to a cardiovascular monitor, fixed to the middle finger of the non-dominant hand (sampling rate 2 Hz). Sensitivity to the CO2 challenge at screening was defined as a change in PSL-IV total score ≥4 with ≥1-point increase for at least four symptoms, and an increase of at least 25mm on the VAS for fear-related symptoms.	Participants underwent a 35% CO2 double-inhalation challenge after 6 days of dosing with placebo or active treatment in each cross-over period; CO2-induced symptoms were measured immediately after CO2 inhalation using the PSL-IV. The CO2 challenge was performed 2.5 h after dosing with JNJ-61393215.	Primary outcome measure was the change of PSL-IV total score, measured as LS mean difference between each active treatment and placebo.	Both JNJ-61393215 at dose of 90 mg and alprazolam significantly decreased PSL-IV total scores compared to placebo [LE (SE)= −2.3 (0.9), p<0.02, and −3.4 (1.5), p<0.03, respectively]. Conversely, no significant effects were found with JNJ-61393215 at dose of 25 mg.	Both JNJ-61393215 and alprazolam did not have any significant effects on blood pressure or heart rate (data not shown in the article). JNJ-61393215 produced a decrease in most items on the PSL-IV, similar to the active comparator alprazolam.	-
Kaufmann et al, 2020^Citation50	ACT-539313 (selective ORX1 antagonist)	Phase I (proof-of-mechanism) study. Randomized two-way cross-over design. Participants received 200 mg ACT-539313 and matching placebo in two sequential treatment periods, each twice daily under fed conditions for 2.5 days to reach steady state. Participants were randomized to the sequence of the treatment periods, which were separated by a washout period of 10–11 days. Single site.	Thirty healthy volunteers (21 males, 9 females; age range: 22–60 years)	The CO2 challenge consisted of inhalation of air for 20 min, 10 min rest, inhalation of 7.5% CO2 for 20 min, 10 min rest, inhalation of 35% CO2 in a single full VC breath, and 30 min rest. Participants were blinded to the gas inhaled. Maximum concentration of CO2 was inhaled at the time of maximum plasma concentrations of ACT-539313. Subjective (ie, multiple VAS, GAD-C and PSI score) and objective (cardiovascular parameters, serum cortisol and plasma ACTH concentration) measures were recorded before, during, and after CO2 challenge	On Day 3 of each treatment period, participants underwent CO2 challenge. Data analysis was performed on peak effects calculated for parameters at the times of inhalation of air (110 min after dosing), inhalation of 7.5% CO2 (140 min after dosing), pre-inhalation of 35% CO2 (149 min after dosing), inhalation of 35% CO2 (150 min after dosing), and end of CO2 challenge (180–210 min after dosing).	Multiple VAS to measure the changes in subjective state (using the following adjectives: alert, fearful, relaxed, anxious, happy, feel like leaving the room, stressed, tense, nervous, irritable, and worried), GAD-C and PSI scores, serum cortisol and plasma ACTH concentrations, systolic and diastolic blood pressure, and pulse rate, were considered.	Overall, no significant differences between ACT-539313 and placebo were found on subjective measures except the following: ACT-539313 induced statistically significantly lower scores in VASalert and VAS anxious at time of peak air than placebo, and lower scores after inhalation of 7.5% CO2 in GAD-C scores in Period 2. No significant differences between ACT-539313 and placebo on any objective physiological measures during CO2 inhalation were found.	In the CO2 challenge, cortisol concentrations were lower during initial air inhalation after treatment with ACT-539313 compared to placebo.	Only few treatment-emergent adverse effects were reported, of which somnolence and headache were the most frequent. No effects on vital signs or laboratory/instrumental clinical measures were found.

Note: This research instrument was developed by Maastricht Instruments in collaboration with Maastricht University (details are available in Leibold et al, 2016).³⁸

Abbreviations: ACTH, adrenocorticotropic hormone; CO₂, carbon dioxide; CTT, CO₂ tolerance tester to induce panic attacks by 35% CO₂ inhalation, according to a protocolized administration; GAD-C, Generalized Anxiety Disorder Criteria; LS, least-square means; ORX1R, orexin-1 receptor; Other results of interest, other findings relevant to this review; PSI, Panic Symptom Inventory; PSL-IV, Panic Symptom List, assessing the 13 symptoms of a panic attack (ranging from 0, absent, to 4, very intense); SE, standard error; VAS, Visual Analogue Scale to assess fear (ranging from 0, not at all, to 100, the worst imaginable); VC, vital capacity.

Table 3 Clinical Studies in Patients with Panic Disorder

Authors, Year [Reference]	Study Design	Duration	Participants (n=Number)		Other Psychiatric Diagnoses (n= Number of Participants)			Sociodemographic/Clinical Characteristics of Participants	Treatments (n= Number of Participants)	Daily Doses (mg)	Additional Treatments
Reinecke et al, 2020^Citation52	Randomized, double-blind, placebo-controlled study: DCS augmented single-session CBT for PD	6 months, including, after screening, the following four study visits: day 1 (baseline assessment; intervention, ie single-session CBT), and 1 day, 1 month and 6 months after intervention	Participants (n=33, recruited from the community) with primary diagnosis of PD (21 with AG, with at least moderate agoraphobic avoidance, and 12 without AG). Diagnosis was assessed with the Structured Clinical Interview for DSM-IV Axis I Disorders; avoidance was measured with the Structured Panic Assessment Interview (“yes” response to>2 situations listed under “(2) Avoidance”). Exposure-based CBT for panic during the last 3 months was an exclusion criterion		Comorbid social phobia, n=4; comorbid specific phobia, n=5, comorbid obsessive-compulsive disorder, n=1. Current primary depressive disorder, lifetime bipolar or psychotic disorders, substance abuse/dependence, or epilepsy were exclusion criteria.			In the placebo group: F= 13 (81%), mean age= 41.9 years (SD= 13.7), AG= 10 (62%); PD duration 10.3 (SD= 12.8) years; in the DCS group: F= 12 (71%), mean age= 42.1 years (SD= 16.7), AG= 11 (65%); PD duration 9.8 (SD= 11.7) years. No significant differences in baseline sociodemographic/clinical characteristics, years of education, or verbal intelligence quotient were found between the two groups	DCS, n=17; Placebo, n=16; treatments were received 2 hours before the single-session CBT	Single administration of 250-mg DCS (or placebo)	Participants were free of CNS-active medications during the last 6 weeks prior to the study. Occasional benzodiazepine or beta-blockers were allowed until 48 hours before treatment and testing sessions.
Dobrovolsky et al, 2017^Citation53	Open-label (single site, Russian Federation)	6 months, including: 9–11 days- Xenon inhalation treatment, and follow-up assessment at 1 and 6 months after the treatment endpoint	Ninety outpatients with diagnosis of PD with or without other psychiatric disorders		In the completer sample (n= 81), 39 participants had other comorbid psychiatric disorders (Group 2) including major depression, bipolar disorder, obsessive-compulsive disorder, mixed anxiety-depressive disorder.			In the completer sample (n=81): F= 49 (60.5%); mean age= 35.2 years (SD= 12.52). Participants with current PD without comorbidities (Group 1, n= 42) and those with comorbidities (Group 2, n= 39) presented similar baseline social, demographic and clinical characteristics, except for higher scores in the subscale “Depression” of the HADS in Group 2	Inhalation of xenon-oxygen mixture (n=90). For each patient: 6–7 sessions; sessions 1,2,3: daily administration; from session 4: every other day administration. Administration of gas mixture was carried out via a face mask. During inhalation vital signs (ie, heart rate, blood pressure, oxygen saturation) were continuously monitored	Xenon-oxygen mixtures increased from 15%/85% to 30%/70% with titration increments of 5% per session. Upper limit of xenon consumption was 3 liters per procedure	Participants with PD without comorbidities (Group 1, n=42) received monotherapy with xenon; participants with comorbidities (Group 2, n=39) were allowed to take SSRIs/SSRNIs as long as with stable dose from at least 3–6 months before the start of the study (94.9% of patients with comorbidities took these additional medications). In Group 2, active symptoms of PD had to be present, despite the ongoing SSRIs/SSRNIs therapy.

Authors, year [Reference]	Completers (n= number of participants)	ITT population (n= number of participants)/Other analyses	Primary outcome measures	Primary results	Secondary outcome measures	Secondary results	Side-effects/tolerability
Reinecke et al, 2020^Citation52	Placebo, n= 13; DCS, n= 14	Data analyses included all randomised participants (n= 33) and were performed on the intention-to-treat basis (n= 16, placebo, n= 17, DCS), except for fMRI measures which were analysed per protocol (n=14, placebo, n= 13, DCS)	Face dot probe task (FDOT) to assess reaction-time based threat bias for fearful faces, performed 1 day after treatment visit (no baseline assessment was performed).	On the day after treatment (ie, single session CBT), threat bias for fearful faces was significantly lower in the DCS compared to the placebo group; however, threat bias did not predict any clinical change.	Measures: 1) emotion regulation task during fMRI, 1 day after treatment visit (no baseline assessment was performed); the task consisted of exposure to blocks of negatively valenced scenes of catastrophic expectations (alternated with grey fixation blocks), with instructions of naturally experiencing or reappraising (eg, reframing, rationalising) the emotional state (Maintain blocks, Reappraisal blocks, respectively).2) self-reported (ie, STAIS/T, BDI, BSQ, MIA, ACQ, PAS) and clinician-rated (PDSS) questionnaires, and a self-reported VAS (0–100) scale for fear referring to an in vivo behavioral stress test (ie being locked in an enclosed walk-in closet), were administered at all the four study visits, to assess clinical symptoms.	One day after treatment (ie, single session CBT), in Maintain blocks activation of right amygdala and prefrontal-cortical areas was significantly lower in DCS than placebo group. Lower amygdala responsivity predicted greater reduction in MIA scores at 1-month follow up across groups. Recovery rate (categorically defined as MIA scores falling within the range reported for healthy subjects) was significantly higher at 1-month follow-up in the DCS (12/17) than placebo group (4/16), whereas on the day after treatment and at 6-month follow-up no differences between groups were found. No differences were found in panic-specific continuous outcome measures (PDSS, PAS, BSQ, ACQ, MIA scores). STAIS/T and BDI scores, as well as VAS scores during behavioral stress test, were significantly lower in DCS than placebo group, at different time points.	No serious adverse events were reported. DCS caused no acute differential changes in blood pressure, heart rate and mood
Dobrovolsky et al, 2017^Citation53	n= 81 The analyses were performed on the completers only	-	SAS (measured BT, and 1 and 6 months after treatment); HADS-Anxiety subscale (measured BT, and at V3 and V6); CGI-S/-I (BT, and at V1-V6); number of monthly F-S and L-S PAs (measured BT and 6 months after treatment).	In both groups, HADS-Anxiety subscale significantly improved at V3 and V6 (at V6, the scores of both groups were in the “normal” category) and SAS significantly improved after 1 and 6 month, when compared to BT. CGI-I at V3: “marked improvement” in 40.5% of Group 1 and in 10.3% of Group 2; CGI-I at V6: a) Group 1: “ very marked improvement” in 52.4% and “marked improvement” in 47.6%; 2) Group 2: “very marked improvement” in 12.8% and “marked improvement” in 87.2%. CGI-S at BT: a) Group 1: “significantly expressed disease” in 90.5%, b) Group 2: “significantly expressed disease” in 87.2%; CGI-S at V3: a) Group 1: “significantly expressed disease” in 51.3%, b) Group 2: “significantly expressed disease in 88.1%; CGI-S at V6: in both groups, most patients reached the “borderline” level (82.1% in Group 1 and 88.1% in Group 2). Mean (SD) frequency of F-S PAs/month BT was 7.7 (7.8) and 11.7 (8.2) in Group 1 and 2, respectively, while that of L-S PAs/month was 44.8 (16.2) and 41.7 (15.2) in Group 1 and 2, respectively. Six months after treatment in both groups the mean (SD) frequency of F-S PAs/month was =0, while that of L-S PAs was 0.3 (0.5) in Group 1 and 1.0 (2.6) in Group 2.	HADS-Depression subscale (measured BT, and at V3 and V6)	In Group 2: HADS-Depression subscale scores corresponding to “clinical severe depression” were present in 92.3% of participants at BT, in 82.1% at V3, and in 46.2% at V6. In Group 1: the same category of scores was present in 33.3% of participants at BT, in 9.5% at V3, and in 2.4% at V6	Side effects, mainly headache and dizziness, were rare, leading to 5 drop-outs from the study (5.8%)

Abbreviations: ACQ, Agoraphobic Cognition Questionnaire; AG, agoraphobia; BDI, Beck Depression Inventory; BSQ, Body Sensation Questionnaire; BT, before treatment; CGI-S/I, Clinical Global Impression Scale, Severity and Improvement subscales; CBT, cognitive behavioral therapy; CNS, central nervous system; DCS, D-cycloserine; F, females; fMRI, functional magnetic resonance imaging; F-S, full-symptoms; HADS, Hospital Anxiety and Depression Scale; ITT, intention-to-treatment; L-S, limited symptoms; MIA, Mobility Inventory for Agoraphobia; PAS, Panic Attack Scale; PD, Panic disorder; PDSS, Panic Disorder Severity Scale; SAS, Zung Self-Rating Anxiety Scale; SD, standard deviation; SSRIs, selective serotonin reuptake inhibitors; SSRNIs, selective serotonin-norepinephrine reuptake inhibitors; V1/2/3/4/5/6, after the first/second/third/fourth/fifth/sixth xenon administration; VAS, Visual Analogue Scale; PA(s), panic attack(s).

Liberati A, Altman DG, Tetzlaff J, et al. The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaboration. J Clin Epidemiol. 2009;62(10):e1–e34. doi:10.1016/j.jclinepi.2009.06.006

 PubMed Web of Science ®Google Scholar

Bonaventure P, Yun S, Johnson PL, et al. A selective orexin-1 receptor antagonist attenuates stress-induced hyperarousal without hypnotic effects. J Pharmacol Exp Ther. 2015;352(3):590–601. doi:10.1124/jpet.114.220392

 PubMed Web of Science ®Google Scholar

Johnson PL, Federici LM, Fitz SD, et al. Orexin 1 and 2 Receptor Involvement in CO2-INDUCED panic-associated behavior and autonomic responses. Depress Anxiety. 2015;32(9):671–683. doi:10.1002/da.22403

 PubMed Web of Science ®Google Scholar

Bonaventure P, Dugovic C, Shireman B, et al. Evaluation of JNJ-54717793 a novel brain penetrant selective orexin 1 receptor antagonist in two rat models of panic attack provocation. Front Pharmacol. 2017;8(JUN):1–13. doi:10.3389/fphar.2017.00357

 Web of Science ®Google Scholar

Salvadore G, Bonaventure P, Shekhar A, et al. Translational evaluation of novel selective orexin-1 receptor antagonist JNJ-61393215 in an experimental model for panic in rodents and humans. Transl Psychiatry. 2020;10:1. doi:10.1038/s41398-020-00937-9

 Web of Science ®Google Scholar

Johnson PL, Shekhar A. An animal model of panic vulnerability with chronic disinhibition of the dorsomedial/perifornical hypothalamus. Physiol Behav. 2012;107(5):686–698. doi:10.1016/j.physbeh.2012.03.016

 PubMed Web of Science ®Google Scholar

Leibold NK, Van Den Hove DLAA, Viechtbauer W, et al. CO2 exposure as translational cross-species experimental model for panic. Transl Psychiatry. 2016;6(9):e885–e885. doi:10.1038/tp.2016.162

 Web of Science ®Google Scholar

Kaufmann P, Ort M, Golor G, Kornberger R, Dingemanse J. Multiple-dose clinical pharmacology of the selective orexin-1 receptor antagonist act-539313. Prog Neuropsychopharmacol Biol Psychiatry. 2020;(xxxx):110166. doi:10.1016/j.pnpbp.2020.110166

 Web of Science ®Google Scholar

Reinecke A, Nickless A, Browning M, Harmer CJ. Neurocognitive processes in D-cycloserine augmented single-session exposure therapy for anxiety: a randomized placebo-controlled trial. Behav Res Ther. 2020;129(September2019):103607. doi:10.1016/j.brat.2020.103607

 Web of Science ®Google Scholar

Dobrovolsky A, Ichim TE, Ma D, Kesari S, Bogin V. Xenon in the treatment of panic disorder: an open label study. J Transl Med. 2017;15(1):137. doi:10.1186/s12967-017-1237-1

 PubMed Web of Science ®Google Scholar