The short answer Biology Treatments Mechanism Feature by feature Receptors Spinal-fluid test Drug by drug History What it means Limits Method Sources

Deep dive · dogs

Narcolepsy in dogs and people

The dog led researchers to the cause of human narcolepsy. It also points two ways for drug development: the inherited dog form is broken at a different step from the human disease, and the dog form that matches the human cause has never been used in a drug trial.

3 of 11

features where the familial dog matches patients

4 of 11

features where the sporadic dog matches patients

6 of 17

drug comparisons where dog and human results agree

where they disagree

published sources, each linked

The short answer

Narcoleptic dogs have the same core syndrome as patients: cataplexy triggered by emotion, daytime sleepiness, and REM sleep that comes too early. Positional cloning in a Stanford Doberman colony found the gene in 1999. Within months, the same pathway was shown to be missing in patients.

But there are two canine diseases, and they fit the human one differently.

Familial narcolepsy (Dobermans, Labradors, dachshunds, Dogo Argentinos) is a broken receptor. The dogs make hypocretin normally.
Sporadic narcolepsy (many breeds, one dog at a time) is a missing peptide, as in patients.

Across 17 drug comparisons, dog and human results point the same way in 6 and opposite ways in 1. The dog evidence conflicts with itself in 2, and in 8 only one species was measured. The newest human drugs, including oveporexton, approved in August 2026, act on the receptor the familial dog lacks, and none of them has been tested in dogs.

How narcolepsy works

Narcolepsy type 1 is the loss of one small group of neurons. In 1998 two groups independently found a pair of peptides made only in the hypothalamus, and named them hypocretins (de Lecea et al. 1998) and orexins (Sakurai et al. 1998). The neurons that make them sit in the lateral hypothalamus but send fibers across the brain, most densely to the locus coeruleus (Peyron et al. 1998). The two peptides act on two receptors, HCRTR1 and HCRTR2 (also called OX1R and OX2R) (Mahoney et al. 2019). HCRTR1 prefers orexin-A; HCRTR2 responds to both peptides about equally (Sakurai et al. 1998).

A human brain holds tens of thousands of these neurons. In narcolepsy type 1 almost all are gone, while neighboring neurons survive (Thannickal et al. 2000).

Hypocretin neurons per brain. Hover a dot for its count.

Each dot is one brain, counted stereologically. Three further control brains could not be counted because part of the hypothalamus had been removed at autopsy. For scale, the rat brain holds about 1,100 of these neurons (Peyron et al. 1998). Source: Thannickal et al. 2000.

These neurons do two jobs. They keep the arousal centers active, which sustains long stretches of waking. Signaling through HCRTR2 in the histamine neurons of the tuberomammillary nucleus (TMN) and through HCRTR1 in the locus coeruleus each helps hold waking in mice (Mahoney et al. 2019). And they stabilize the brain's two switches, between waking and sleep and between REM and non-REM sleep. Without them the switches flip too easily, which is why patients are sleepy by day and wake often at night (Saper et al. 2010).

Cataplexy is REM sleep's muscle paralysis breaking into waking. Positive emotion activates the amygdala, which inhibits the brainstem neurons that keep REM paralysis switched off. Normally hypocretin offsets that inhibition; without it the paralysis circuit fires (Mahoney et al. 2019). This is why laughter can make a patient collapse while fully conscious.

Switch between the healthy brain and narcolepsy type 1, then pick a drug to see where it acts.

excites inhibits lost in narcolepsy type 1 where the selected drug acts

Where drugs act:

Hypocretin neurons excite the histamine, noradrenaline, serotonin, dopamine and acetylcholine centers that keep the cortex awake, and hold the REM-off neurons of the vlPAG–LPT active. During strong positive emotion the prefrontal cortex drives both the hypocretin neurons and the central amygdala; the amygdala's inhibition of the REM-off neurons is offset by hypocretin, so the atonia center stays silent and muscle tone is kept (Mahoney et al. 2019). Hypocretin also stabilizes the switches between waking and sleep, and between REM and non-REM sleep (Saper et al. 2010).

Narcolepsy type 2 is less severe and has no cataplexy. About 90% of patients have normal hypocretin levels, and its cause is not established (Mahoney et al. 2019).

Treatments, approved and in development

Until 2026 every approved drug worked downstream of the missing peptide. Stimulants and wake-promoting agents raise dopamine, noradrenaline or histamine to fight sleepiness. Oxybates, taken at night, and off-label antidepressants reduce cataplexy. Their labels mostly call the mechanism unknown or unclear (Provigil label; Xyrem label; Sunosi label).

HCRTR2 agonists are different: they stand in for the missing hypocretin. The first, TAK-994, worked but was stopped for liver injury (Dauvilliers et al. 2023). The second, oveporexton, was approved on 5 August 2026 as Orzeyful, for adults with narcolepsy type 1, on two trials in 273 adults (FDA 2026). Alixorexton added 22–26 minutes of wakefulness over placebo in a 92-patient phase 2 trial, though cataplexy fell significantly only at the middle of its three doses, and no liver problems appeared (Plazzi et al. 2026); it is now in phase 3. Intravenous danavorexton kept patients with narcolepsy type 2 awake as well, although their hypocretin is normal (Evans et al. 2022).

US approvals for narcolepsy by year. Filled: tested in narcoleptic dogs. Open: never tested in them. Amphetamine salts are approved for narcolepsy but their label gives no year.

10 drugs approved in the United States for narcolepsy, 3 of them ever tested in narcoleptic dogs, and the hypocretin receptor agonists in development. Status as of 2026-09-22.
Drug	How it works	Status and source	In narcoleptic dogs
Methylphenidate Ritalin	Dopamine and noradrenaline reuptake block and release	approved 1955 Narcolepsy · Ritalin label	tested Mignot et al. 1993
Amphetamine salts Adderall	Dopamine and noradrenaline reuptake block and release	approved Narcolepsy · Adderall label	tested Shelton et al. 1995
Modafinil Provigil	Unknown; stimulant-like	approved 1998 Sleepiness · Provigil label	tested Shelton et al. 1995
Sodium oxybate Xyrem	GABA-B, taken at night	approved 2002 Cataplexy and sleepiness · Xyrem label	not tested
Armodafinil Nuvigil	Unknown; the active enantiomer of modafinil	approved 2007 Sleepiness · Nuvigil label	not tested
Solriamfetol Sunosi	Dopamine and noradrenaline reuptake inhibitor	approved 2019 Sleepiness · Sunosi label	not tested
Pitolisant Wakix	Histamine H3 antagonist and inverse agonist	approved 2019 Sleepiness and cataplexy · Wakix label	not tested
Mixed-salt oxybates Xywav	GABA-B, lower sodium	approved 2020 Cataplexy and sleepiness · Drugs@FDA, Xywav	not tested
Sodium oxybate, once nightly Lumryz	GABA-B, extended release	approved 2023 Cataplexy and sleepiness · Drugs@FDA, Lumryz	not tested
Oveporexton Orzeyful	HCRTR2 agonist	approved 2026 Narcolepsy type 1 · FDA 2026	not tested
TAK-994	HCRTR2 agonist	stopped Stopped for liver injury in phase 2 · Dauvilliers et al. 2023	not tested
Danavorexton	HCRTR2 agonist, intravenous	phase 1 Wakefulness in type 1 and type 2 · Evans et al. 2022	not tested
Alixorexton	HCRTR2 agonist	phase 3 Narcolepsy type 1 and 2 · NCT07455383; Plazzi et al. 2026	not tested
Cleminorexton (ORX750)	HCRTR2 agonist	phase 2/3 Narcolepsy type 1 and 2 · NCT07598708	not tested
E2086	HCRTR2 agonist	phase 2 Narcolepsy · NCT07493265	not tested
TAK-360	HCRTR2 agonist	phase 2 Narcolepsy type 1 · NCT07633301	not tested

Of the 10 drugs approved for narcolepsy in the United States, 3 were ever tested in narcoleptic dogs, and all of those came before 2000. None of the 5 hypocretin agonists still in development has been, and the search behind that statement is recorded below.

Two diseases in dogs, one in people

Pick a model to see which step of the pathway is broken, and whether each kind of drug has something to act on. "Symptomatic drugs" are the antidepressants, stimulants, sodium oxybate and pitolisant, which act on arousal systems downstream of hypocretin.

works weak or brief conflicting fails or no target not tested · dashed box: lost · red box: broken

Patients have 85–95% fewer hypocretin neurons and spinal-fluid hypocretin averages about an eighth of normal. The receptors remain: in 5 patient brains HCRTR1 was 71% lower and HCRTR2 not significantly lower. That is why an HCRTR2 agonist works: TAK-994 and oveporexton restored wakefulness and cut cataplexy. Intranasal peptide had smaller effects on REM sleep in 14 patients.

Feature by feature

11 features, scored against patients for each form of canine narcolepsy. Mouse models are shown for reference, because that is where the newest human drugs were developed.

Click a row for the evidence behind each verdict.

Feature	In patients	Familial dog HCRTR2 mutation	Sporadic dog hypocretin deficient	Mouse for reference
Signs
Cataplexy triggered by emotion	A main feature, with daytime sleepiness (Nishino & Mignot 1997); 35–75% of patients report daily attacks, most often set off by laughter (Chemelli et al. 1999).	partly	partly	partly
Familial dog Present, but more frequent and severe than in patients, and set off by food and play rather than laughter (Chemelli et al. 1999). Emotional triggers in narcoleptic dogs are also described by Tonokura et al. 2007. Sporadic dog Present, and as with familial dogs set off by food rather than laughter: feeding triggered attacks in a hypocretin-deficient chihuahua (Tonokura et al. 2003). Sporadic cases were typically more severe than familial ones (Hungs et al. 2001). Mouse Episodes of behavioral arrest, 8 to 27 in four hours, set off by excited movement rather than emotion as such (Chemelli et al. 1999).
Daytime sleepiness and early REM sleep	Daytime sleepiness and abrupt transitions from waking into REM sleep (Lin et al. 1999).	matches	not tested	matches
Familial dog Narcoleptic dogs slept more in the daytime and needed high stimulant doses to reach control levels, which the authors call consistent with human data (Shelton et al. 1995). Sporadic dog No sleep recording of a dog known to be hypocretin deficient in the papers read. A 1974 poodle of unknown hypocretin status had sleep-onset REM over 41 hours of recording (Mitler et al. 1974). Mouse Direct transitions from waking into REM sleep in every knockout studied (Chemelli et al. 1999).
Cause
Hypocretin peptide	Below detection in 7 of 9 patients, though one of the other two was above the control range (Nishino et al. 2000); globally lost in 6 brains (Peyron et al. 2000).	differs	matches	matches
Familial dog Normal: 275 pg/mL in 32 narcoleptic Dobermans against 258 in 9 controls (Ripley et al. 2001). Sporadic dog Below detection in the spinal fluid of 2 of 2 dogs and absent from 3 of 3 brains (Ripley et al. 2001); below 40 pg/mL in a Schipperke (Fujiki et al. 2003) and below 80 in a chihuahua (Tonokura et al. 2003). Mouse Absent by construction in knockout mice (Chemelli et al. 1999).
Hypocretin neurons	85–95% fewer in 4 patient brains against 12 controls, with neighboring neurons spared (Thannickal et al. 2000).	differs	not tested	matches
Familial dog No difference from controls in number, shape or distribution in 4 Dobermans (Ripley et al. 2001). Sporadic dog The peptide is absent from the brain (Ripley et al. 2001); no neuron count appears in the abstracts read. Mouse Destroyed by design in the orexin/ataxin-3 and DTA models (Mitsukawa et al. 2024).
Receptors	Receptor-gene mutations were not found in 74 patients (Peyron et al. 2000). In 5 patient brains, cortical HCRTR1 was 71% lower than in controls and HCRTR2 19% lower, not significantly (Mishima et al. 2008).	differs	matches	matches
Familial dog HCRTR2 is broken: exon-skipping mutations in Dobermans and Labradors (Lin et al. 1999), a missense mutation in dachshunds whose receptor still responds weakly to very high hypocretin (Hungs et al. 2001), and a duplication in Dogo Argentinos whose effect is predicted, not measured (Mondino et al. 2025). Sporadic dog Same pattern as in patients, in 3 dogs: HCRTR1 at 0.49 of control (P=0.034), HCRTR2 at 0.66 (not significant) (Mishima et al. 2008). Mouse HCRTR1 fell to 0.67 of control in 27-week-old neuron-ablated mice; HCRTR2 not significantly (Mishima et al. 2008).
Inheritance	Mostly not familial; the paper cites a Finnish twin study for most identical twins being discordant (Thannickal et al. 2000). One hypocretin-gene mutation in 74 patients screened (Peyron et al. 2000).	differs	matches	not applicable
Familial dog Autosomal recessive; all 30 puppies of affected Doberman parents developed the disease (Foutz et al. 1979). Sporadic dog No hypocretin-gene mutation in 11 sporadic dogs (Hungs et al. 2001); onset at 1–3 years rather than before 6 months (Mishima et al. 2008). Mouse Engineered.
Immune involvement	HLA-DQB1*06:02 in 93% of patients with cataplexy against 17% of controls (Mignot et al. 2002); hypocretin-specific T cells in all 19 patients tested and in none of 13 healthy carriers of the same HLA type (Latorre et al. 2018).	differs	partly (weak)	partly
Familial dog No inflammation or lymphocyte infiltration in 7 narcoleptic Doberman brains; microglial MHC class II was higher at 3 and 8 months only, and the authors read this as against a localized autoimmune process (Tafti et al. 1996). No link to dog HLA, and bone marrow transplant neither transmitted nor relieved the disease (Wagner et al. 2000). Sporadic dog In 11 sporadic dogs no dog HLA allele tracked the disease, but in the 6 sporadic poodles one allele was less common than in control poodles (P=0.008) (Wagner et al. 2000). No study of T cells against hypocretin neurons in dogs was found (search recorded below). In 8 dogs with suspected acquired narcolepsy attacks fell on prednisolone, but hypocretin was not measured and there was no control group (Santifort et al. 2021). Mouse Knockout and ablation models are not immune-driven; immune-driven mouse models exist but reproduce the disease incompletely (Arias-Carrión & Ortega-Robles 2025).
Biomarker
Spinal-fluid hypocretin test	At or below 110 pg/mL is diagnostic; with definite cataplexy the test had 87% sensitivity and 99% specificity (Mignot et al. 2002).	differs	matches	not applicable
Familial dog Levels are normal (275 against 258 pg/mL in controls), so the test cannot identify these dogs (Ripley et al. 2001). Sporadic dog Undetectable in 2 of 2 sporadic dogs (Ripley et al. 2001); proposed for clinical diagnosis (Tonokura et al. 2007). A normal value in a dachshund pointed to a pituitary tumor instead (Schmid et al. 2017). Mouse Not used.
Drugs
Drugs acting downstream of hypocretin	Treated with amphetamine-like stimulants or modafinil and with antidepressants (Nishino & Mignot 1997), although trial evidence for antidepressants is scarce (Vignatelli et al. 2008).	matches	partly	matches
Familial dog Noradrenergic uptake inhibitors suppressed cataplexy (Mignot et al. 1993); modafinil and amphetamine increased wakefulness (Shelton et al. 1995); methamphetamine prevented cataplexy in a second laboratory (Wu et al. 2011). One exception, clomipramine, is in the drug table. Sporadic dog Imipramine reduced attacks in a hypocretin-deficient chihuahua (Tonokura et al. 2003); uncontrolled, one dog. Mouse An H3 inverse agonist promoted wakefulness in knockout mice, as in patients (Lin et al. 2008).
Prazosin worsens cataplexy	Near-continuous cataplexy within 3 days of starting prazosin, in one patient (Aldrich & Rogers 1989).	matches	not tested	not searched
Familial dog Worsened cataplexy in 6 Dobermans with controls unaffected (Mignot et al. 1988), and triggered cataplexy without any stimulus in a second laboratory (Wu et al. 2011). Sporadic dog Worsened cataplexy in all 4 narcoleptic poodles, a line that did not breed true, unlike the Dobermans (Mignot et al. 1988); their hypocretin was never measured. Mouse Not searched.
Drugs acting at the hypocretin defect	HCRTR2 agonists: TAK-994 improved wakefulness and cataplexy but injured the liver (Dauvilliers et al. 2023); oveporexton improved all three measures in two phase 3 trials (Dauvilliers et al. 2026).	conflicting	not tested	matches
Familial dog Hypocretin-1 reduced cataplexy at one dose in 6 Dobermans and worsened it at a higher one (John et al. 2000). Another laboratory found no effect in 6 Dobermans at 1–24 µg/kg, with repeated dosing, or injected into the brain, and ruled out peptide quality (Fujiki et al. 2003). No HCRTR2 agonist tested; in mice without the receptor, TAK-861 did nothing (Mitsukawa et al. 2024). Sporadic dog In one Schipperke, intravenous hypocretin-1 above 48 µg/kg suppressed cataplexy for under 30 minutes (Fujiki et al. 2003). No HCRTR2 agonist tested (search recorded below). Mouse TAK-861 suppressed cataplexy-like episodes in neuron-loss mice, and had no effect in mice lacking HCRTR2 (Mitsukawa et al. 2024).

Scorecard verdicts by model, excluding features that do not apply

Familial dog3 matches · 1 partly · 6 differs · 1 conflicting

Sporadic dog4 matches · 3 partly · 4 not tested

Mouse (reference)6 matches · 2 partly

matches partly differs conflicting not tested

The pattern is consistent. The familial dog matches patients on signs and on drugs acting downstream, and differs on cause: the gene, the peptide, the neurons, inheritance and the immune system. The sporadic dog matches on cause wherever it has been measured. Too few sporadic dogs have been studied for most of the other rows to be scored.

What happens to the receptors

A drug that stimulates the receptor needs the receptor to still be there. After the peptide is lost, patients, sporadic dogs and neuron-ablated mice all lose some HCRTR1, and none shows a significant loss of HCRTR2. In the familial dog it is the reverse: HCRTR1 is normal and HCRTR2 is the broken gene.

Receptor expression after hypocretin loss, relative to controls (1.0). Filled dot and * = significant (P < 0.05); bars are ± SEM. Hover for P values.

Receptor mRNA in cortex relative to each group's own controls (1.0 = control). Mean ± SEM from the results text. Groups are small: 3 sporadic dogs, 8 familial dogs, 5 patients, 27-week-old mice. Source: Mishima et al. 2008.

The spinal-fluid test

Measuring hypocretin in spinal fluid is a diagnostic test in people. It separates the two canine forms. It also exposes lookalikes: a dachshund with cataplexy and a normal level turned out to have a pituitary tumor.

Hypocretin-1 in spinal fluid, pg/mL. Dashed lines are the human thresholds. Hover a row for its source.

Dog and human values come from different assays and laboratories and are not calibrated against each other, so compare each group with its own species' normal values. Bars ending at a detection limit mean the level was below it.

Healthy dogs. Normal canine range quoted in Tonokura et al. 2003; Stanford's reference in 2017 was 200–350 (Schmid et al. 2017).
Control Dobermans. 258 ± 6.6 pg/mL, 9 dogs (Ripley et al. 2001).
Familial Dobermans. 275 ± 5.8 pg/mL, 32 dogs; not different from controls (Ripley et al. 2001).
Dachshund, pituitary tumor. 328.7 pg/mL; the cataplexy came from a tumor (Schmid et al. 2017).
Sporadic dog, chihuahua. Below 80 pg/mL, 1 dog (Tonokura et al. 2003).
Sporadic dog, Schipperke. Below 40 pg/mL, 1 dog (Fujiki et al. 2003).
Sporadic dogs. Below the 24 pg/mL detection limit in 2 of 2 (Ripley et al. 2001).
Healthy people. 363.2 ± 16.3 pg/mL, 47 daytime controls, direct assay (Mignot et al. 2002).
Patients, typical cataplexy. 47.5 ± 9.7 pg/mL, 101 patients, direct assay (Mignot et al. 2002).
Patients, small study. Below 40 pg/mL in 7 of 9; the other two had 255 and 638 (Nishino et al. 2000).

Drug by drug

Click a row for the sources. ▲ helps · ▼ worsens · ● no effect · ◆ conflicting · ○ not tested or not measured.

Drug	Symptom	Dogs	People	Agreement
Imipramine tricyclic antidepressant	cataplexy	▲ helpscase reports; 2 small series	▲ helpsconsensus; weak trials	agree
Dogs Reduced attacks in a poodle (Mitler et al. 1974) and in treated dogs of a 7-dog case series (Mitler et al. 1976), significantly shortened the food test in 4 poodles (Babcock et al. 1976), and reduced attacks in a hypocretin-deficient chihuahua (Tonokura et al. 2003). People Recommended by consensus; a Cochrane review found no good-quality trial evidence that antidepressants help, and no placebo-controlled imipramine trial among the five it included (Vignatelli et al. 2008). Acts downstream of hypocretin.
Clomipramine tricyclic antidepressant	cataplexy	● no effect4 poodles	▲ helps11 patients, no placebo	disagree
Dogs No clear effect at 0.5, 1 or 5 mg/kg on the time to finish the food test; attacks fell on the drug day at 1 mg/kg only (Babcock et al. 1976). The authors report large variability between dogs. People Over 50% fewer attacks in 7 of 11, in a trial against fluvoxamine with no placebo arm (Vignatelli et al. 2008). Acts downstream of hypocretin.
Venlafaxine serotonin and noradrenaline reuptake inhibitor	cataplexy	● no effect1 dog	○ not measured	cannot compare
Dogs Stopped after two doses for sedation and no perceived response, in a dachshund whose attacks came from a pituitary tumor (Schmid et al. 2017). People No human source read for venlafaxine. Acts downstream of hypocretin.
Noradrenaline reuptake inhibitors desipramine, viloxazine and others	cataplexy	▲ helps5–6 dogs per drug	▲ helpscrossover, 56 patients	agree
Dogs All five took attacks to zero at the highest dose; viloxazine from 2.9 to 0 per test (Mignot et al. 1993). People Viloxazine lowered the cataplexy score from 2.94 to 1.77 on a 0–9 scale; the reviewers doubt the clinical significance (Vignatelli et al. 2008). Acts downstream of hypocretin.
Serotonin reuptake inhibitors fluoxetine, femoxetine and others	cataplexy	◆ conflicting4 dogs; 5–6 per drug	▲ helpscrossover, 10 patients	dog evidence conflicts
Dogs Fluoxetine improved cataplexy in 4 dogs (Babcock et al. 1976) but not significantly in another laboratory (4.8 to 1.9 attacks, P=0.12), where paroxetine did (4.3 to 1.6, P=0.005) and zimelidine did not (Mignot et al. 1993). People Femoxetine reduced cataplexy (Vignatelli et al. 2008). Acts downstream of hypocretin.
Prazosin alpha-1 blocker	cataplexy	▼ worsens14 dogs, 2 laboratories	▼ worsens1 patient	agree
Dogs Worsened cataplexy in all 4 poodles and all 6 affected Dobermans, with carriers and controls unaffected (Mignot et al. 1988); triggered it without any stimulus in 4 Dobermans in a second laboratory (Wu et al. 2011). People Near-continuous cataplexy within 3 days; improved on stopping (Aldrich & Rogers 1989). Acts downstream of hypocretin.
Amphetamine stimulant	sleepiness	▲ helpscolony dogs	▲ helpsreview	agree
Dogs Increased wakefulness (Shelton et al. 1995). People A standard treatment for sleepiness (Nishino & Mignot 1997). Acts downstream of hypocretin.
Amphetamine stimulant	cataplexy	▲ helpscolony dogs	○ not measured	cannot compare
Dogs Significantly suppressed cataplexy (Shelton et al. 1995); dextroamphetamine took attacks to zero (Mignot et al. 1993). People No human measurement in the sources read. Acts downstream of hypocretin.
Modafinil wake-promoting agent	sleepiness	▲ helpscolony dogs	▲ helpstrial, 283 patients	agree
Dogs Increased wakefulness at 5 and 10 mg/kg (Shelton et al. 1995). People Reduced every measure of sleepiness over 9 weeks (US Modafinil Study Group 1998). Acts downstream of hypocretin.
Modafinil wake-promoting agent	cataplexy	● no effectcolony dogs	○ not measured	cannot compare
Dogs No effect on cataplexy from 0.125 to 8 mg/kg (Shelton et al. 1995). People The 283-patient trial measured sleepiness and excluded patients who needed cataplexy medication (US Modafinil Study Group 1998). Acts downstream of hypocretin.
Hypocretin-1 the missing peptide	cataplexy	◆ conflicting12 Dobermans in 2 labs; 1 sporadic dog	○ not measuredcrossover, 14 patients	dog evidence conflicts
Dogs In 6 Dobermans only the 3 µg/kg dose helped and 4 µg/kg made cataplexy worse (John et al. 2000). In 6 others, doses up to 24 µg/kg, repeated dosing and injection into the brain did nothing; one sporadic dog responded briefly above 48 µg/kg (Fujiki et al. 2003). People Intranasal peptide reduced wake-to-REM transitions; cataplexy is not reported in the abstract (Weinhold et al. 2014). Acts at the hypocretin defect.
HCRTR2 agonists TAK-994, oveporexton, alixorexton, danavorexton	sleepiness and cataplexy	○ not tested	▲ helpstrials, 73 to 273 patients	cannot compare
Dogs No study found (search recorded below). In mice lacking the receptor, TAK-861 had no effect (Mitsukawa et al. 2024), which is what the familial dog would predict. People TAK-994 worked but was stopped for liver injury that rat and monkey toxicology had not shown (Dauvilliers et al. 2023; Shinozawa et al. 2025). Oveporexton cut weekly cataplexy by a median of 79–89% against 28–39% on placebo (Dauvilliers et al. 2026) and was approved in August 2026 (FDA 2026). Alixorexton added 22–26 minutes of wakefulness over placebo in 92 patients, though cataplexy fell significantly only at the middle dose (Plazzi et al. 2026). Intravenous danavorexton worked in type 2 as well as type 1, where hypocretin is normal (Evans et al. 2022). Acts at the hypocretin defect.
Sodium oxybate GHB salt	cataplexy	○ not tested	▲ helpswithdrawal trial, 55 patients	cannot compare
Dogs No study in narcoleptic dogs found (search recorded below). People Stopping it raised the median attacks over two weeks from 0 to 21 (US Xyrem Study Group 2004). Acts downstream of hypocretin.
Pitolisant histamine H3 inverse agonist	sleepiness	○ not tested	▲ helpstrial, 95 patients	cannot compare
Dogs No study in narcoleptic dogs found (search recorded below). People Better than placebo on the Epworth sleepiness score by 3.0 points in 95 patients; cataplexy also fell as a secondary outcome (Dauvilliers et al. 2013). Acts downstream of hypocretin.
Almorexant dual hypocretin receptor blocker	sleep, in healthy subjects	▲ helpshealthy beagles, 8 per group	▲ helpshealthy volunteers	agree
Dogs Caused somnolence without loss of muscle tone (Brisbare-Roch et al. 2007). People Caused sleep without cataplexy (Brisbare-Roch et al. 2007). Acts at the hypocretin defect.
Methylphenidate stimulant	cataplexy	▲ helps5 dogs	○ not measured	cannot compare
Dogs Attacks fell from 5.9 to 0.3 per test at the highest dose (Mignot et al. 1993). People The label lists narcolepsy as an indication without naming a symptom (Ritalin label). Acts downstream of hypocretin.
Solriamfetol dopamine and noradrenaline reuptake inhibitor	sleepiness	○ not tested	▲ helpstrial, 236 patients	cannot compare
Dogs No study in dogs found (search recorded below). People Added 9.8 and 12.3 minutes of wakefulness at 150 and 300 mg, against 2.1 on placebo, over 12 weeks (Thorpy et al. 2019). Acts downstream of hypocretin.

Three things stand out.

Where both species were measured, the dog usually got the direction right. Noradrenergic drugs suppress cataplexy in both. Prazosin worsens it in both: in 10 dogs, again in a second laboratory, and in one patient. Modafinil wakes both.

The human side of the agreement is often thin. A Cochrane review found no good-quality trial evidence that antidepressants help narcolepsy, even though they are the standard treatment for cataplexy. Several of the "agree" rows rest on consensus or on one patient.

The dog was not always right, and not always consistent. Clomipramine had no clear effect in 4 dogs but halved attacks in 7 of 11 patients in a trial without a placebo arm. Fluoxetine helped in 4 dogs in one study and fell short of significance in another, where paroxetine helped. By contrast, every noradrenaline-selective drug in that study took attacks to zero. Hypocretin-1 itself helped at one dose and worsened cataplexy at a higher one in one laboratory, and did nothing in another.

The drugs approved or advanced for narcolepsy since 2000 (sodium oxybate, pitolisant and the HCRTR2 agonists) have no test in narcoleptic dogs in the literature searched. The HCRTR2 agonists were developed in neuron-loss mice and monkeys. In mice lacking the receptor, as the familial dog does, TAK-861 had no effect.

Animal toxicology did not foresee what stopped the first of those drugs. TAK-994 was halted after three patients met Hy's law criteria for liver injury. Its 26-week rat and 39-week monkey studies had shown no liver injury at many times the human exposure; mouse studies started during the human trials did. No dog study of TAK-994 is reported, so whether dogs would have shown it is unknown.

Fifty years of the model

Findings by year and species. Click a dot to find it in the list.

1973First veterinary report, in a dog and a cat · Knecht et al. 1973
1974A poodle with cataplexy and sleep-onset REM responds to imipramine, as patients do · Mitler et al. 1974
1976Stanford narcoleptic dog colony founded; 669 dogs born by 1995 · Cederberg et al. 1998
1979Doberman narcolepsy shown to be autosomal recessive · Foutz et al. 1979
1988Prazosin worsens canine cataplexy · Mignot et al. 1988
1989Prazosin worsens cataplexy in a patient · Aldrich & Rogers 1989
1993Noradrenergic, not serotonergic, drugs control canine cataplexy · Mignot et al. 1993
1995Modafinil wakes narcoleptic dogs but leaves cataplexy unchanged · Shelton et al. 1995
1998Hypocretins discovered in rodent hypothalamus by two groups · de Lecea et al. 1998
1998Modafinil reduces sleepiness in a 283-patient trial · US Modafinil in Narcolepsy Multicenter Study Group 1998
1999Canine narcolepsy traced to the HCRTR2 gene · Lin et al. 1999
1999Hypocretin knockout mice are narcoleptic · Chemelli et al. 1999
2000Hypocretin below detection in 7 of 9 patients · Nishino et al. 2000
2000Patients have lost 85–95% of hypocretin neurons · Thannickal et al. 2000
2000Hypocretin-1 reduces cataplexy in receptor-mutant dogs at one dose, and worsens it at a higher one · John et al. 2000
2001Sporadic dogs are hypocretin deficient, like patients; familial dogs are not · Ripley et al. 2001
2002Spinal-fluid hypocretin becomes a diagnostic test · Mignot et al. 2002
2003Hypocretin-1 has no effect in receptor-mutant dogs in a second laboratory · Fujiki et al. 2003
2004Sodium oxybate controls cataplexy in a withdrawal trial · US Xyrem Multicenter Study Group 2004
2007Blocking hypocretin receptors induces sleep in dogs and people alike · Brisbare-Roch et al. 2007
2013Pitolisant beats placebo in a 95-patient trial · Dauvilliers et al. 2013
2018T cells against hypocretin neurons found in all 19 patients tested · Latorre et al. 2018
2019Solriamfetol approved for sleepiness · Sunosi label
2019Pitolisant approved · Wakix label
2021Eight dogs with suspected acquired narcolepsy described · Santifort et al. 2021
2023First oral HCRTR2 agonist works, but injures the liver · Dauvilliers et al. 2023
2024TAK-861 tested in neuron-loss mouse models · Mitsukawa et al. 2024
2025TAK-994 liver injury traced to reactive metabolites; rat and monkey toxicology had missed it · Shinozawa et al. 2025
2025A fourth HCRTR2 mutation, in Dogo Argentinos · Mondino et al. 2025
2025Oveporexton phase 2 succeeds without liver injury · Dauvilliers et al. 2025
2026Oveporexton succeeds in two phase 3 trials · Dauvilliers et al. 2026
2026Oveporexton (Orzeyful) approved, the first HCRTR2 agonist · FDA 2026
2026Alixorexton phase 2 succeeds; phase 3 begins · Plazzi et al. 2026

Canine narcolepsy research peaked in the early 2000s and then gave way to mouse work. The count below is what PubMed returns. It is not a measure of quality: a review that mentions dogs once counts as a dog paper.

PubMed records per five years mentioning narcolepsy or cataplexy with each animal, retrieved 2026-09-22. Click a series to hide it.

PubMed records per five years whose title or abstract mentions narcolepsy or cataplexy and the animal. A paper counts once per animal it mentions, and a review that mentions dogs in passing counts as a dog paper. The last bin holds 2025 and 2026 to date only. Cat records include laboratory studies of cataplexy induced by drugs in cats; a title search found one report of a narcoleptic cat, a Siamese diagnosed in 1973 from its response to dextroamphetamine (Knecht et al. 1973).

Queries

dog (narcolepsy[tiab] OR cataplexy[tiab]) AND (dog[tiab] OR dogs[tiab] OR canine[tiab] OR Dogs[mh])
mouse (narcolepsy[tiab] OR cataplexy[tiab]) AND (mouse[tiab] OR mice[tiab] OR murine[tiab] OR Mice[mh])
cat (narcolepsy[tiab] OR cataplexy[tiab]) AND (cat[tiab] OR cats[tiab] OR feline[tiab] OR Cats[mh])

What this means for companion-animal trials

This deep dive supports three readings, and does not support a fourth.

Supported: dogs as a model of the syndrome. Signs, sleep and responses to drugs acting downstream of hypocretin match patients closely enough that the dog colony was used for decades to screen stimulant and anticataplectic drugs, and those results mostly held in people.

Supported: dogs as a route to causes. The receptor gene was found in dogs first. Dogs also showed that the receptor, not the peptide level, sets the drug response: drugs that change cataplexy shift hypocretin levels in normal dogs, but not in mutants whose cataplexy they still change.

Supported, with a caveat: the sporadic dog as a model of the human cause. Sporadic dogs lack the peptide, keep the receptor and start later in life, as patients do. The evidence rests on about six dogs, some measured in more than one study: the same three brains were used for peptide levels and for receptor expression, and spinal fluid comes from two more dogs in the first study, a Schipperke and a chihuahua. Whether they share the human immune cause is unknown: no study of T cells against hypocretin neurons in dogs was found.

Not supported: the familial dog as a test bed for today's human drugs. The approved HCRTR2 agonist (FDA 2026) and the ones in development all need a working HCRTR2. Most familial dogs have none, and the one exception, the dachshund mutation, responds only to abnormally high concentrations.

For a companion-animal program, the opportunity this literature points to is the sporadic, hypocretin-deficient dog, the one form that could respond the way patients do to an HCRTR2 agonist, a class that is now approved in people and has several more molecules in trials. How many such dogs there are is not established here. The only figure found is that sporadic cases occur in up to 17 breeds. The largest recent report is a case series of 8 dogs with suspected acquired narcolepsy, in which hypocretin was not measured.

What would change this reading

An HCRTR2 agonist tested in sporadic, hypocretin-deficient dogs.
A prevalence estimate for sporadic canine narcolepsy with hypocretin measured.
A search for T cells against hypocretin neurons in sporadic dogs.

What this page cannot tell you

The sources come from targeted searches, not a systematic review, so a study that contradicts this page could be missing. The searches behind each "not found" are listed below.
The drug table compares effects in direction only. Doses, endpoints and designs differ between species and between studies.
Most dog studies are small and come from two colonies, at Stanford and UCLA. The sporadic-dog evidence rests on about six animals, several of them reused across papers.
Spinal-fluid values from dogs and people were not calibrated against each other.
Market size, willingness to pay and the regulatory route for a canine narcolepsy drug are not assessed.

How this was checked

This review's own corpus holds 0 findings on this condition. Of 26,343 records retrieved for the review, 3 mention it (2 excluded, 1 not screened). This page is therefore a separate, targeted reading, not part of the systematic review.

Sources were found through PubMed searches. Each claim was checked against the source paper's results, tables and figure legends, and the numbers and design details behind it are listed under each source below. Last checked 2026-09-22.

PubMed searches behind each statement that something was not found. PubMed searches behind every statement of absence on the page. Run 2026-09-22 through the NCBI E-utilities. A count is what PubMed returned; the note says what reading the titles found.
Query	Records	What reading them found
`(oxybate OR "gamma-hydroxybutyrate" OR "gamma hydroxybutyrate" OR GHB) AND (narcolepsy OR cataplexy) AND (dog OR dogs OR canine)`	18	Reviews and one mouse study; no test of sodium oxybate in narcoleptic dogs.
`(pitolisant OR "H3 receptor" OR H3-receptor OR tiprolisant) AND (narcolepsy OR cataplexy) AND (dog OR dogs OR canine)`	4	Pharmacokinetics and chemistry papers; no test in narcoleptic dogs.
`("orexin receptor 2 agonist" OR "OX2R agonist" OR "orexin 2 receptor agonist" OR danavorexton OR TAK-994 OR TAK-861 OR oveporexton OR YNT-185 OR alixorexton OR ALKS-2680 OR ORX750 OR cleminorexton OR E2086 OR TAK-360) AND (dog OR dogs OR canine)`	0	No records.
`narcolepsy[ti] AND (cat[ti] OR cats[ti] OR feline[ti])`	1	One 1973 report, of a dachshund and a Siamese cat.
`(narcolepsy OR cataplexy) AND (dog OR dogs OR canine) AND (autoimmun* OR "T cell" OR "T cells")`	26	Human immunology and reviews, plus one canine microglia study; no search of dog T cells against hypocretin neurons.
`(solriamfetol OR JZP-110 OR ADX-N05) AND (dog OR dogs OR canine)`	1	One review; no test in dogs of any kind.
`armodafinil AND (narcolepsy OR cataplexy) AND (dog OR dogs OR canine)`	19	Reviews, mouse studies and the modafinil study in narcoleptic dogs; no separate test of armodafinil in dogs.

Sources

Click a source for the numbers and design details behind the claims that cite it.

Source	Title	Study
Ritalin label	Ritalin (methylphenidate) tablets prescribing information FDA label, DailyMed	regulatory record

Adderall label	Adderall (mixed amphetamine salts) tablets prescribing information FDA label, DailyMed	regulatory record

Knecht et al. 1973	Narcolepsy in a dog and a cat J Am Vet Med Assoc	case reports, 1 dog and 1 cat
Key details A 6-month-old dachshund, diagnosed by EEG, and a 1-year-old Siamese cat, diagnosed from its response to dextroamphetamine. Both animals were stabilized on dextroamphetamine.
Mitler et al. 1974	Narcolepsy-cataplexy in a female dog Exp Neurol	case report, 1 toy poodle
Key details One toy poodle; cataplexy from about 4 months of age; 41 hours of continuous EEG, EOG and EMG recording. Imipramine was given intravenously in two trials to confirm the diagnosis, because it was known to work in patients.
Babcock et al. 1976	Effects of imipramine, chlorimipramine, and fluoxetine on cataplexy in dogs Pharmacol Biochem Behav	experimental, 4 poodles
Key details Four narcoleptic poodles from an eight-dog colony; each drug given on one day between two placebo days. Imipramine 1 mg/kg and fluoxetine 1.5 and 3 mg/kg shortened the food test significantly. Clomipramine at 0.5, 1 and 5 mg/kg did not; at 1 mg/kg the number of attacks fell on the drug day (P<0.05) while the time to finish did not. The authors warn that variability within and between dogs is large and that general arousal alone reduces cataplexy.
Mitler et al. 1976	Narcolepsy in seven dogs J Am Vet Med Assoc	case series, 7 dogs
Key details Seven dogs, mostly poodles, with cataplexy beginning before or during young adulthood. Imipramine reduced cataplectic attacks in the dogs that received it; neostigmine had no effect, ruling out myasthenia gravis.
Foutz et al. 1979	Genetic factors in canine narcolepsy Sleep	breeding study

Mignot et al. 1988	Effect of alpha 1-adrenoceptors blockade with prazosin in canine narcolepsy Brain Res	experimental, 10 narcoleptic dogs
Key details Prazosin by mouth worsened cataplexy in all 4 narcoleptic poodles and all 6 affected Dobermans; heterozygous and control Dobermans were unaffected. Narcolepsy in poodles had not bred true in the colony, unlike in Dobermans; the poodles' hypocretin status was never measured.
Aldrich & Rogers 1989	Exacerbation of human cataplexy by prazosin Sleep	case report, 1 patient

Mignot et al. 1993	Canine cataplexy is preferentially controlled by adrenergic mechanisms Psychopharmacology	experimental, 5–6 dogs per compound, 19 compounds
Key details 5 or 6 colony dogs per compound, all older than one year with severe cataplexy. Every noradrenaline-selective compound took attacks to zero at the highest dose (desipramine, nisoxetine, nortriptyline, tomoxetine, viloxazine; P<0.0005). Serotonin-selective compounds were weaker and inconsistent: paroxetine 4.3 to 1.6 attacks (P=0.005), fluoxetine 4.8 to 1.9 (P=0.12), zimelidine 2.7 to 2.2 (P=0.72). Dextroamphetamine took attacks to zero.
Shelton et al. 1995	Comparative effects of modafinil and amphetamine on daytime sleepiness and cataplexy of narcoleptic dogs Sleep	experimental

Tafti et al. 1996	Major histocompatibility class II molecules in the CNS: increased microglial expression at the onset of narcolepsy in canine model J Neurosci	tissue study, dogs and humans
Key details 17 narcoleptic and 16 control Dobermans, all with the inherited form. No localized inflammation or lymphocyte infiltration anywhere in the brains of 4 adult and 3 young narcoleptic dogs; the authors say this argues against a localized autoimmune process. Microglial MHC class II rose at about 3 months in both groups, higher in narcoleptics at 3 and 8 months, and the difference was gone by 18 months.
Nishino & Mignot 1997	Pharmacological aspects of human and canine narcolepsy Prog Neurobiol	review
Key details Reviews two decades of drug testing in the Stanford dog colony, including a section on canine narcolepsy as a screening tool for anticataplectic and wake-promoting drugs.
Cederberg et al. 1998	Breeding history of the Stanford colony of narcoleptic dogs Vet Rec	colony records, 669 dogs
Key details Colony founded at Stanford in 1976; 669 dogs born by 1995, 487 surviving, 78% of them Dobermans. Affected Dobermans show signs at 4 to 12 weeks of age.
Peyron et al. 1998	Neurons containing hypocretin (orexin) project to multiple neuronal systems J Neurosci	anatomy, rat brain
Key details About 1,100 hypocretin neurons in the rat brain, in the perifornical nucleus and dorsal and lateral hypothalamus, distinct from MCH neurons. The densest projection outside the hypothalamus is to the locus coeruleus; fibers also reach the raphe, tuberomammillary nucleus, basal forebrain and brainstem.
Provigil label	Provigil (modafinil) prescribing information FDA label, DailyMed	regulatory record

Sakurai et al. 1998	Orexins and orexin receptors: a family of hypothalamic neuropeptides and G protein-coupled receptors that regulate feeding behavior Cell	discovery of the peptides and receptors
Key details Purified two peptides, orexin-A (33 amino acids) and orexin-B, as ligands of two orphan receptors, and mapped the neurons to the lateral and posterior hypothalamus. OX1R binds orexin-A with higher affinity than orexin-B; OX2R binds both with similar high affinity, which the authors call nonselective.
US Modafinil in Narcolepsy Multicenter Study Group 1998	Randomized trial of modafinil for the treatment of pathological somnolence in narcolepsy Ann Neurol	randomized trial, 283 patients
Key details 283 patients at 18 centers, 200 or 400 mg daily for 9 weeks. Patients needing medication for cataplexy were excluded; cataplexy appears only as an adverse event, in about 1% of each arm.
de Lecea et al. 1998	The hypocretins: hypothalamus-specific peptides with neuroexcitatory activity Proc Natl Acad Sci USA	discovery of the peptides
Key details A hypothalamus-specific mRNA encodes a 130-residue precursor, preprohypocretin, for two secretin-related peptides with excitatory effects on neurons.
Chemelli et al. 1999	Narcolepsy in orexin knockout mice Cell	knockout mice
Key details Cataplexy is less severe in patients than in dogs or knockout mice: 35–75% of patients report daily attacks, while narcoleptic dogs often have several in a single session. Triggers differ: laughter in people; food and play in dogs; excited movement in mice.
Lin et al. 1999	The sleep disorder canine narcolepsy is caused by a mutation in the hypocretin (orexin) receptor 2 gene Cell	positional cloning
Key details A ten-year positional cloning project over about 100 informative backcross dogs from the Stanford colony. Dobermans carry a SINE insertion causing exon 4 skipping; Labradors a different exon 6 skipping mutation. Both transcripts are grossly abnormal.
John et al. 2000	Systemic administration of hypocretin-1 reduces cataplexy and normalizes sleep and waking durations in narcoleptic dogs Sleep Res Online	experimental
Key details 6 Dobermans with the inherited HCRTR2 mutation. Only the 3 µg/kg intravenous dose reduced cataplexy; 1 and 2 µg/kg had no effect, and 4 µg/kg made cataplexy significantly worse. With repeated doses, 2 of 3 dogs went three or more days without cataplexy.
Nishino et al. 2000	Hypocretin (orexin) deficiency in human narcolepsy Lancet	case-control, 9 patients and 8 controls
Key details 9 patients and 8 controls, all patients HLA-DQB1*06:02 positive. 7 patients were below the 40 pg/mL detection limit; the other 2 had 255 and 638 pg/mL, the second above the control range, and were clinically indistinguishable from the rest.
Peyron et al. 2000	A mutation in a case of early onset narcolepsy and a generalized absence of hypocretin peptides in human narcoleptic brains Nat Med	6 brains; mutation screen of 74 patients
Key details Six narcoleptic brains against eight controls for peptide measurement. Screening HCRT, HCRTR1 and HCRTR2 in 74 patients found one HCRT mutation, in a patient whose cataplexy began at 6 months; no receptor mutation was found.
Thannickal et al. 2000	Reduced number of hypocretin neurons in human narcolepsy Neuron	post-mortem brains
Key details 4 patient brains against 12 controls. Hypocretin neurons reduced by 85–95%; melanin-concentrating hormone neurons unchanged (731 against 734). The statement that most identical twins are discordant cites a Finnish twin study (Hublin et al. 1994), not this paper's own data.
Wagner et al. 2000	DLA-DQB1 alleles and bone marrow transplantation experiments in narcoleptic dogs Tissue Antigens	DLA typing of 99 dogs, 11 sporadic; one transplant pair
Key details 31 narcoleptic and 68 control dogs typed; the 11 sporadic dogs were 6 poodles, 3 dachshunds, a Labrador and a Rhodesian-Labrador mix. In poodles, DLA-DQB1*02301 was less common in narcoleptic dogs (P=0.008), but one severely affected sporadic poodle carried two copies; the authors call it at most partly protective. After bone marrow transplant between identical-DLA littermates, the affected dog stayed affected and the carrier unaffected for 2 years, with full engraftment.
Hungs et al. 2001	Identification and functional analysis of mutations in the hypocretin (orexin) genes of narcoleptic canines Genome Res	sequencing, 11 sporadic dogs and 2 families
Key details 17 narcoleptic dogs sequenced; no hypocretin-gene mutation in 11 sporadic cases or in a poodle family. The dachshund E54K receptor reaches the cell surface and does not bind hypocretin at up to 4 nM, but responds to abnormally high concentrations above 1 µM. Sporadic cases were typically more severely affected than familial ones.
Ripley et al. 2001	Hypocretin levels in sporadic and familial cases of canine narcolepsy Neurobiol Dis	CSF from 32 familial and 2 sporadic dogs; 3 sporadic brains
Key details CSF hypocretin-1 (Table 1): 32 familial narcoleptic Dobermans 275 ± 5.8 pg/mL against 9 control Dobermans 258 ± 6.6 (P=0.17); puppies of both genotypes around 305–317. Two sporadic dogs, a dachshund aged 118 months and a mixed breed aged 49 months, were below the assay's detection limit of 24 pg/mL. The three sporadic brains (a dachshund, a poodle and a cocker-poodle mix) are the same three dogs later used for receptor expression by Mishima et al. 2008. Hypocretin neurons in 4 familial Dobermans did not differ from controls in number, morphology or distribution. Onset, citing earlier colony work: 1–4 months in familial dogs, 2.5–84 months in sporadic ones.
Mignot et al. 2002	The role of cerebrospinal fluid hypocretin measurement in the diagnosis of narcolepsy and other hypersomnias Arch Neurol	case series, 274 patients and 296 controls
Key details Narcolepsy with typical cataplexy: 101 patients, CSF hypocretin-1 47.5 ± 9.7 pg/mL (direct assay), against 363.2 ± 16.3 in 47 daytime healthy controls. The 110 pg/mL cut-off is 30% of the normal mean on the direct assay. Across all narcolepsy it had 60% sensitivity and 98% specificity; with definite cataplexy, 87% and 99%. HLA-DQB1*06:02 in 93% of patients with typical cataplexy against 17% of daytime controls.
Xyrem label	Xyrem (sodium oxybate) prescribing information FDA label, DailyMed	regulatory record

Fujiki et al. 2003	Effects of IV and ICV hypocretin-1 (orexin A) in hypocretin receptor-2 gene mutated narcoleptic dogs and IV hypocretin-1 replacement therapy in a hypocretin-ligand-deficient narcoleptic dog Sleep	experimental, 6 familial Dobermans and 1 sporadic Schipperke
Key details Six familial Dobermans: single intravenous doses of 1–6 µg/kg had no effect on cataplexy; 24 µg/kg in 4 dogs, none; 3 µg/kg on 3 consecutive days, none. Injection into the brain in 2 dogs, none. The sporadic dog was a 3-year-old Schipperke with CSF hypocretin below 40 pg/mL and onset at 10 months. Doses above 48 µg/kg suppressed cataplexy for under 30 minutes, completely in 2 of 3 sessions at 384 µg/kg. The authors tested peptide from two suppliers to rule out quality as the reason their results differ from John et al. 2000.
Tonokura et al. 2003	Narcolepsy in a hypocretin/orexin-deficient chihuahua Vet Rec	case report, 1 dog
Key details A 2-year-old male chihuahua in Japan, no family history; attacks set off by feeding and play, lasting seconds to minutes, with eyes open. Imipramine 1 mg/kg three times daily reduced attacks; no sleep recording was made. The authors note that abnormal sleep is not a serious problem in dogs, while cataplexy is disabling.
US Xyrem Multicenter Study Group 2004	Sodium oxybate demonstrates long-term efficacy for the treatment of cataplexy in patients with narcolepsy Sleep Med	randomized withdrawal trial, 55 patients
Key details 55 patients on sodium oxybate for 7 to 44 months were randomized to continue or switch to placebo for 2 weeks. Median attacks over 2 weeks: 21 on placebo, 0 on continued treatment; no rebound cataplexy on stopping.
Brisbare-Roch et al. 2007	Promotion of sleep by targeting the orexin system in rats, dogs and humans Nat Med	beagles, 8 per group; healthy volunteers
Key details Almorexant was given to healthy male beagles, 8 per group in a crossover design, and to healthy volunteers. It caused somnolence in dogs and sleep in people, with no abrupt loss of muscle tone in either.
Nuvigil label	Nuvigil (armodafinil) prescribing information FDA label, DailyMed	regulatory record

Tonokura et al. 2007	Review of pathophysiology and clinical management of narcolepsy in dogs Vet Rec	review
Key details Sporadic narcolepsy in over 17 breeds, citing colony records and unpublished data. Familial forms in Dobermans, Labradors and dachshunds; spinal-fluid hypocretin proposed for diagnosing ligand-deficient dogs in practice.
Lin et al. 2008	An inverse agonist of the histamine H3 receptor improves wakefulness in narcolepsy: studies in orexin-/- mice and patients Neurobiol Dis	mice; pilot trial, 22 patients
Key details Tiprolisant in orexin-knockout mice, and a single-blind pilot in 22 patients: Epworth score fell by 5.9 from 17.6 on the drug and by 1.0 on placebo.
Mishima et al. 2008	Hypocretin receptor expression in canine and murine narcolepsy models and in hypocretin-ligand deficient human narcolepsy Sleep	tissue study, mice, dogs, humans
Key details Sporadic dogs: 3 (a dachshund, a poodle and a cocker-poodle mix), onset at 1–3 years; familial Dobermans: 8, onset before 6 months. Cortex, relative to controls: sporadic dogs HCRTR1 0.49 (P=0.034), HCRTR2 0.66 (P=0.077); familial Dobermans HCRTR1 1.02; 5 patients HCRTR1 0.29 (P=0.047), HCRTR2 0.81 (not significant); 27-week neuron-ablated mice HCRTR1 0.67 (P=0.014), HCRTR2 0.74 (P=0.11). The discussion misstates one figure: it gives a 63% HCRTR2 decline in the pons, where the results report 63% for HCRTR1 and 3% for HCRTR2.
Vignatelli et al. 2008	Antidepressant drugs for narcolepsy Cochrane Database Syst Rev	systematic review, 5 trials, 246 participants
Key details Five trials, 246 participants, methodological quality of all unclear; no meta-analysis possible. Femoxetine (an SSRI), 10-patient crossover: reduced cataplexy, not sleepiness. Viloxazine, 56-patient crossover: cataplexy score 1.77 against 2.94 on a 0–9 scale (P=0.0001); the reviewers call the effect of doubtful clinical significance. Clomipramine against fluvoxamine, 18 patients, no placebo: over 50% fewer attacks in 7 of 11 on clomipramine and 5 of 11 on fluvoxamine.
Saper et al. 2010	Sleep state switching Neuron	review
Key details Mutually inhibitory wake–sleep and REM–non-REM circuits behave like flip-flop switches; hypocretin neurons stabilize both, and their loss makes states unstable.
Wu et al. 2011	Role of the hypocretin (orexin) receptor 2 (Hcrt-r2) in the regulation of hypocretin level and cataplexy J Neurosci	experimental, normal and mutant dogs
Key details 4 normal and 4 narcoleptic Dobermans, at UCLA rather than Stanford. Prazosin and physostigmine triggered strong cataplexy in the narcoleptic dogs without food or other stimuli; methamphetamine, labetalol and phenylephrine prevented it. The same drugs changed spinal-fluid hypocretin by 22–44% in normal dogs, but not in the mutants, and never produced cataplexy in normal dogs.
Dauvilliers et al. 2013	Pitolisant versus placebo or modafinil in patients with narcolepsy Lancet Neurol	randomized trial, 95 patients
Key details 95 patients randomized to pitolisant, modafinil or placebo for 8 weeks. Pitolisant beat placebo on the Epworth score by 3.0 points but was not shown non-inferior to modafinil; cataplexy rates fell as a secondary outcome.
Weinhold et al. 2014	The effect of intranasal orexin-A (hypocretin-1) on sleep, wakefulness and attention in narcolepsy with cataplexy Behav Brain Res	randomized crossover, 14 patients
Key details 14 patients, intranasal orexin-A 435 nmol against placebo, crossover. Fewer wake-to-REM transitions and less daytime REM sleep; cataplexy was not assessed, which the authors say would need an emotional-trigger design.
Schmid et al. 2017	Pituitary macrotumor causing narcolepsy-cataplexy in a dachshund J Vet Intern Med	case report, 1 dog
Key details Spinal-fluid hypocretin-1 was 328.7 pg/mL against a reference of 200–350, measured at Stanford. Venlafaxine was stopped after two doses for sedation and no perceived response; attacks resolved after radiotherapy shrank the tumor.
Latorre et al. 2018	T cells in patients with narcolepsy target self-antigens of hypocretin neurons Nature	cellular screens, 19 patients
Key details Hypocretin-specific CD4+ T cells in all 19 patients tested; T cells against another hypocretin-neuron protein in 8 of 13. The same assays found no such T cells in 13 healthy donors carrying HLA-DQB1*06:02.
Mahoney et al. 2019	The neurobiological basis of narcolepsy Nat Rev Neurosci	review
Key details Hypocretin neurons excite the TMN, locus coeruleus, raphe, PPT–LDT and basal forebrain; restoring HCRTR2 in the TMN, or HCRTR1 in the locus coeruleus, restores sustained waking in receptor-knockout mice. Cataplexy model: positive emotion drives the prefrontal cortex and central amygdala; without hypocretin to offset it, the amygdala inhibits the REM-off vlPAG–LPT, releasing the sublaterodorsal nucleus. Silencing the amygdala or prefrontal cortex reduces cataplexy in hypocretin-null mice. In narcolepsy type 2, about 90% of patients have normal CSF hypocretin.
Sunosi label	Sunosi (solriamfetol) prescribing information FDA label, DailyMed	regulatory record

Thorpy et al. 2019	A randomized study of solriamfetol for excessive sleepiness in narcolepsy Ann Neurol	randomized trial
Key details TONES 2: 239 randomized, 236 treated, 12 weeks, narcolepsy type 1 or 2. Maintenance of Wakefulness Test: +12.3 minutes at 300 mg and +9.8 at 150 mg against +2.1 on placebo; Epworth −6.4 and −5.4 against −1.6.
Wakix label	Wakix (pitolisant) prescribing information FDA label, DailyMed	regulatory record

Drugs@FDA, Xywav	NDA 212690, original approval 21 July 2020 FDA, Drugs@FDA	regulatory record

Santifort et al. 2021	Suspected acquired narcolepsy in 8 dogs J Vet Intern Med	retrospective case series, 8 dogs
Key details Seven French bulldogs and one chihuahua from four European referral centers; median age 2 years 3 months. No hypocretin measurement is reported; diagnosis rested on cataplexy observed by a board-certified neurologist. Six of 8 eventually remitted, 3 without long-term medication; the authors state that without a control group they cannot attribute improvement to treatment.
Evans et al. 2022	Orexin 2 receptor-selective agonist danavorexton improves narcolepsy phenotype in a mouse model and in human patients Proc Natl Acad Sci USA	mice; randomized crossover in patients
Key details Nine-hour intravenous infusions: placebo-adjusted gains in sleep latency of 18.8 to 36.7 minutes in narcolepsy type 1; in type 2, 22.4 minutes at 44 mg on day 1. Hypocretin-deficient mice were more sensitive to the agonist than normal mice.
Dauvilliers et al. 2023	Oral orexin receptor 2 agonist in narcolepsy type 1 N Engl J Med	phase 2 randomized trial, 73 patients
Key details Primary-endpoint data exist for 41 of the 73 patients (56%) because all TAK-994 trials were stopped early. Eight patients had raised liver enzymes; three met Hy's law criteria (two at 180 mg, one at 90 mg), and all TAK-994 trials were terminated.
Drugs@FDA, Lumryz	NDA 214755, original approval 1 May 2023 FDA, Drugs@FDA	regulatory record

Mitsukawa et al. 2024	TAK-861, a potent, orally available orexin receptor 2-selective agonist, produces wakefulness in monkeys and improves narcolepsy-like phenotypes in mouse models Sci Rep	mice and monkeys; sponsor-authored
Key details TAK-861 at 10 mg/kg, ten times its minimum effective dose, did not change wakefulness in mice lacking HCRTR2 (n=8), while it did in their normal littermates. It suppressed cataplexy-like episodes in orexin/ataxin-3 mice at 1 mg/kg. All authors were Takeda employees.
Arias-Carrión & Ortega-Robles 2025	Animal models of narcolepsy: from orexin deficiency to immune mechanisms and regenerative therapies Curr Issues Mol Biol	review
Key details Immune-driven mouse models exist (for example orexin-neuron loss after H1N1 infection in Rag1-deficient mice) but reproduce the phenotype incompletely.
Dauvilliers et al. 2025	Oveporexton, an oral orexin receptor 2-selective agonist, in narcolepsy type 1 N Engl J Med	phase 2 randomized trial, 112 participants
Key details 112 participants across four oveporexton doses and placebo for 8 weeks. Insomnia in 48%, urinary urgency in 33% and urinary frequency in 32%; no liver toxicity.
Mondino et al. 2025	Familial narcolepsy in Dogo Argentino dogs is caused by a tandem duplication mutation in HCRTR2 J Vet Intern Med	whole-genome sequencing, 10 dogs
Key details Three affected Dogo Argentinos in Uruguay; the duplication spans exons 3 and 4 and was absent from 2,766 other dogs. No RNA could be obtained, so the effect on the receptor is predicted from the sequence, not measured.
Shinozawa et al. 2025	TAK-994 mechanistic investigation into drug-induced liver injury Toxicol Sci	toxicology and cell studies; sponsor-authored
Key details 26-week rat and 39-week cynomolgus monkey toxicology studies showed no liver injury, at exposure margins of 19 to 198 times the human maximum. Liver injury appeared in mouse studies run for carcinogenicity testing while phase 1 was under way. No dog study is reported. The authors conclude the injury was likely idiosyncratic, from reactive metabolites, which standard animal studies rarely detect. Sponsor-authored.
Dauvilliers et al. 2026	Oveporexton for narcolepsy type 1: results from two phase 3 trials N Engl J Med	two phase 3 randomized trials, 273 participants
Key details Two 12-week trials, First Light (168 participants) and Radiant Light (105), aged 16 to 70. Adverse events in 86–89% on oveporexton against 43–54% on placebo; increased urinary frequency and transient insomnia occurred in a majority.
FDA 2026	FDA approves first drug to treat the full range of narcolepsy type 1 symptoms (Orzeyful, oveporexton), 5 August 2026 FDA press announcement	regulatory record

NCT07455383	Phase 3 trial of alixorexton (ALKS 2680) in narcolepsy type 1, Alkermes ClinicalTrials.gov	trial registration

NCT07493265	Phase 2 trial of E2086 in narcolepsy, Eisai ClinicalTrials.gov	trial registration

NCT07598708	Phase 2/3 trial of cleminorexton (ORX750) in narcolepsy types 1 and 2, Centessa ClinicalTrials.gov	trial registration

NCT07633301	Phase 2 trial of TAK-360 in narcolepsy type 1, Takeda ClinicalTrials.gov	trial registration

Plazzi et al. 2026	Safety, tolerability, and efficacy of alixorexton, a selective orexin 2 receptor agonist for narcolepsy type 1 (Vibrance-1) Lancet Neurol	phase 2 randomized trial, 92 participants
Key details 92 participants, 4, 6 or 8 mg once daily or placebo for 6 weeks; Alkermes-sponsored. Placebo-corrected gains on the Maintenance of Wakefulness Test of 22.2, 24.1 and 26.0 minutes. Weekly cataplexy fell significantly only at 6 mg (rate ratio 0.31, P=0.0099); at 4 mg and 8 mg the reductions were not significant (P=0.17 and 0.45). Most common adverse events were frequent urination, insomnia and excess salivation; no liver-function stopping criteria were met.

Narcolepsy in dogs and people

The short answer

How narcolepsy works

Treatments, approved and in development

Two diseases in dogs, one in people

Feature by feature

Familial dog

Sporadic dog

Mouse

Familial dog

Sporadic dog

Mouse

Familial dog

Sporadic dog

Mouse

Familial dog

Sporadic dog

Mouse

Familial dog

Sporadic dog

Mouse

Familial dog

Sporadic dog

Mouse

Familial dog

Sporadic dog

Mouse

Familial dog

Sporadic dog

Mouse

Familial dog

Sporadic dog

Mouse

Familial dog

Sporadic dog

Mouse

Familial dog

Sporadic dog

Mouse

What happens to the receptors

The spinal-fluid test

Drug by drug

Dogs

People

Dogs

People

Dogs

People

Dogs

People

Dogs

People

Dogs

People

Dogs

People

Dogs

People

Dogs

People

Dogs

People

Dogs

People

Dogs

People

Dogs

People

Dogs

People

Dogs

People

Dogs

People

Dogs

People

Fifty years of the model

What this means for companion-animal trials

What would change this reading

What this page cannot tell you