Auditory and vestibular disorders comprise a heterogeneous group of conditions that have different pathophysiological mechanisms. Depending on the disease, they may affect different structures of the inner ear, cochleovestibular nerve, central signaling or processing nuclei. At the same time, these disorders can either increase or decrease sensory and neural excitability. For example, increased sensory and neural excitability occurs during an acute vestibular crisis (e.g., vestibular neuritis, migraine, and labyrinthitis) or during tinnitus. By contrast, in other disorders such as chronic vestibular hypofunction or age-related hearing loss, there is a deterioration of sensory and neural activity (Peelle and Wingfield, 2016). The neurochemistry of these audiovestibular disorders includes diverse neurotransmitters with excitatory (glutamate, acetylcholine), inhibitory (GABA, glycine), modulatory (histamine, noradrenaline) and other actions (enkephalin, motilin, somatostatin, dopamine and serotonin) (Guerra and Cacabelos, 2019). As a result, and depending on the case, therapeutic approaches differ, with depressive drugs administered in cases of increased neuronal excitability, and stimulants prescribed in cases of decreased neuronal activity.

Regardless of their etiology, acute vestibular disorders are treated with vestibular sedatives. Most of the medications used to treat audiovestibular syndromes are also used to treat central nervous system (CNS) disorders. Anticholinergics, antihistamines, benzodiazepines, calcium-channel blockers and dopamine receptor antagonists minimize nystagmus and vegetative symptoms that include tachycardia, sweating, nausea, and vomiting. To avoid delaying vestibular compensation, prescribing those drugs that have sedative effects should be used in the short term (Hain and Uddin, 2003). Drugs that decrease oxidative stress and inflammation and increase cochlear vascularization can be used to slow hearing loss. Antioxidants, neurotrophic factors, calcium-channel blockers, vasodilators, antiglutamatergic, steroid, and other anti-inflammatory and antiapoptotic agents are also commonly used to treat auditory disorders. However, the majority of these drugs are only protective and there is weak evidence of their efficacy (Sha and Schacht, 2016).

Knowledge of the patient’s pharmacogenetic profile is essential for reducing the risk of adverse effects and for optimizing treatment response; this prescription of pharmacogenetic profile-based strategies improves the therapeutic response. Single nucleotide polymorphisms (SNPs) in genes encoding drug-metabolizing enzymes, drug transporters and drug targets are primarily responsible for the variations in individual drug response. One of the potential consequences of using pharmacogenomics in clinical trials and molecular therapeutics is that a given disease may be treated based on genomic and biological markers, selecting drugs optimized for individual patients or for groups of patients with similar genomic profiles. This aspect has been little explored in the scientific literature (Guerra and Cacabelos, 2019).

The endocannabinoid (EC) system is distributed widely in the brain and is also expressed in the audiovestibular pathway (Gong et al., 2006; Chi and Kandler, 2012; Trattner, Berner, Grothe and Kunz, 2013). There has been increased interest in the development of drugs that act either as an agonist or antagonist on the EC system. This review summarizes current understanding on the role of the EC system for the future use and development of natural- or synthetically-derived compounds that act on the audiovestibular pathway. To pursue those objectives, it is necessary to fully understand the properties of the EC system and the pharmacogenetic profiles of drugs that act on the EC system.

CB1 (encoded by CNR1 gene, locus 6q15) and CB2 (encoded by CNR2 gene, locus 1p36.11) are the 2 G protein-coupled cannabinoid (CB) receptors. Activation of the Gi/Goα subunits reduces intracellular cAMP and increases MAPK levels. CB1 can occasionally activate the G-protein Gs thereby enhancing cAMP levels and the CB2 Gβγ subunits thus activating the MAPK-ERK signaling pathway. Most of the CB receptors in the CNS are CB1 receptors whose function is primarily neuromodulatory through inhibiting GABAergic and glutamatergic neurotransmission. However, CB2 receptors are functionally anti-inflammatory. Other clinically cannabinoid orphan receptors of interest in this review are GPR18 (13q32.3), GPR55 (2q37.1 locus) and GPR119 (Xq26.1 locus), that also encode G protein-coupled receptors (Lu and Mackie, 2016; Zou and Kumar, 2018; Kilaru and Chapman, 2020). Regarding endogenous cannabinoids (CBs), six ligands that act as neurotransmitters have been identified: arachidonoylethanolamine (anandamide/AEA), 2-arachidonoylglycerol (2-AG), 2-arachidonoylglyceryl ether (noladin ether) N-arachidonoyl-dopamine (NADA), Virodhamine, and lysophosphatidylinositol (LPI) (Gallelli et al., 2018). Some of these molecules have anti-inflammatory, sedative, or analgesic properties that are of interest in several audiovestibular disorders, with AEA and 2-AG being the most clinically relevant (Zogopoulos, Vasileiou, Patsouris and Theocharis, 2013). Fatty acid amide hydrolase (FAAH, encoded by FAAH gene, locus 1p33) and monoacylglycerol lipase (MAGL, encoded by MGLL gene, locus 3q21.3) are the enzymes responsible for the degradation of the endocannabinoids (Lu and Mackie, 2016; Zou and Kumar, 2018; Kilaru and Chapman, 2020).

More than one hundred CBs derived from the Cannabis sativa plant have been identified. Of these, tetrahydrocannabinol (THC), cannabidiol (CBD) and cannabinol (CBN) are the most studied and for which the most clinical data are available (Chandra et al., 2020).

In addition to the commercialization of cannabinoid-containing medications and supplements, several clinical studies are currently investigating the effects of 1) synthetic cannabinoid-derived drugs (CB1/CB2 receptor agonists/antagonists), and 2) non-cannabinoid-derived compounds (MAGL/FAAH inhibitors) that belong to a part of a different class of biologically active molecules, that regulate the EC system. These compounds have been used in clinical trials and observational studies to manage treatment-resistant epilepsy (Devinsky et al., 2016), ALS (amyotrophic lateral sclerosis) (Weber et al., 2010), MS (multiple sclerosis) (Collin, Davies, Mutiboko and Ratcliffe, 2007; Zajicek et al., 2012), Alzheimer’s disease (Herrmann et al., 2019), schizophrenia (Leweke et al., 2012; McGuire et al., 2018), acute and chronic pain (van de Donk et al., 2018; Weizman et al., 2018), and several other conditions (Kogan and Mechoulam, 2007; Perin et al., 2020; Khalsa et al., 2022). Since many of the above-mentioned and other nervous system disorders also affect the auditory and vestibular pathway, and are associated with dizziness, nystagmus and auditory symptoms, these molecules may help address several types of audiovestibular disorders. Moreover, symptoms that show these neurological diseases appear in peripheral audiovestibular pathology, such as nystagmus, tinnitus or hearing loss (Baguley et al., 2013; Zwergal and Dieterich, 2020).

THC and its synthetic derivative dronabinol are CB1/CB2 receptor agonists. They mimick the effects of ECs in the CNS and modulate the excitatory effect of other neurotransmitters including glutamate. These agents may adversely enhance the tachycardic response of anticholinergics and sympathomimetics and can cause tinnitus, ataxia, and seizures as side effects. Dronabinol’s half-life is 25–36 h, while its metabolites can last up to 59 h. Nabilone, another synthetic cannabinoid that mimics THC, has a more variable half-life, ranging from 2 to 35 h (Cacabelos, 2012). The analogous use of THC and its synthetic derived drugs with opioid derived drugs, antidepressants and anticonvulsants requires a dose reduction due to greater analgesic efficacy and/or higher drug toxicity caused by down-regulation of transporter efflux or enzymatic inhibition (Vázquez et al., 2020). The pharmacogenetic profile of cannabinoid derivatives is still under development, especially with regard to new synthetic derivatives. Dronabinol’s mechanistic genes are CNR1 and CNR2, and it acts as an agonist. Its metabolic genes are CYP3A4, CYP2C9, CYP2C19, CYP1A2, CYP2B6, CYP2D6, CYP1A1, CYP2J2, CYP3A5, CYP3A7, CYP1B1, CYP2A6, CES1 and PTGS1, and the transporter genes include ABCG2, ABCB1 and ABCC1. Nabilone displays the same mechanistic genes as Dronabinol, but it binds partially to its receptors; CYP2C9, CYP3A4, CYP2C8, CYP2E1 and CYP2J2 are its metabolic genes.

Cannabidiol, however, is a non-competitive CB1/CB2 receptor antagonist. It reduces neuronal hyperexcitability by modulating Ca²⁺ entry via GPR55 and TRPV1 receptors, and by inhibiting ENT-1-receptor-mediated adenosine reuptake (Ibeas Bih et al., 2015). It has a half-life of 18–32 h and causes somnolence, sedation, diarrhea, and changes in appetite as its side effects. It may reduce the adverse effects caused by THC and its derivatives. Drug-drug interactions include regular medications such as anticonvulsants, statins or common analgesics (Vázquez et al., 2020; Davis et al., 2021). The mechanistic genes of cannabidiol are much broader, with diverse targets, as seen in Table 1. CYP3A4, CYP2C19, CYP1A2, CYP2B6, CYP2C9, UGT1A7, UGT1A9, UGT2B7, UGT2B17, CYP1A1, CYP1B1, CYP2C8, CYP2D6, CYP2E1, CYP3A5, CYP2A6, CYP3A7, CYP2J2, CES1, AANAT, FAAH, ALOX5, ALOX15, and CRYZ are its metabolic genes. Transporter genes associated with cannabidiol include ABCC1, ABCG2, SLC29A1, and ABCB1 (Cacabelos, 2012; Wishart et al., 2017).

CNR1 and CNR2 variants show mixed results in terms of cannabis dependency. CNR1, -64 + 1046T>G and c.-63-9c.1359G>A variants show an association with drug dependency, while 597T>C -64 + 1046T>G c.*3475A>G are also linked to drug dependency in single studies, although with no replication of results. The CNR2 188A>G and 946C>T variants alter its CB2 receptor function in human embryonic kidney-293 cells. Because the COMT rs4680 472A>G Val158Met variant is linked to memory impairment, alterations in executive function, and psychosis, the use of cannabis derivatives should be approached with caution with corresponding adjustments in prescribed doses. The FAAH 385C>APro129Thr variant may be involved in excessive consumption of cannabis, although other authors did not find any association between FAAH variants. The ABCB1 rs10456423455C>T Ile1145Ile variant is associated with cannabis dependency (Hryhorowicz et al., 2018).

CB1 antagonists such as rimonabant, surinabant and other diarylpyrazole derivatives have been experimentally and clinically evaluated for treating conditions unrelated to audiovestibular symptoms such as obesity or addiction disorders. However, their use has raised safety concerns due to CNS or cardiac side effects (Christopoulou and Kiortsis, 2011; Klumpers et al., 2013). Therefore, its role in treating the conditions described in this review is not defined.

Unfortunately, little is known regarding the half-life, adverse effects and pharmacogenetic profiles of other drugs of interest described in this review. One of the primary reasons is that those drugs have been used solely for research purposes. These include CB1/CB2 agonists (CP 55,940, JWH-015 and WIN55,212–2) CB1 agonists (JWH-133), CB1 antagonists (Dietressa^© and Brizantin^©), FAAH inhibitors (URB597, URB937, and AM374) and FAAH/MAGL inhibitors (JZL195). The targets of these compounds are listed in Table 1 (Cacabelos, 2012; Wishart et al., 2017).

The association of the EC system in the vestibular pathway has been previously demonstrated by early reports that indicated that the audiovestibular pathway expresses CB1 receptors. The density of these receptors in this pathway was initially reported as modest, but functionally significant (Newsham-West et al., 1998). Furthermore, CB2 receptors are expressed in the vestibular and cochlear nuclei (Gong et al., 2006; Baek et al., 2008), with lower distribution and relevance than CB1 receptors. The EC system interacts with the different vestibular pathways such as the vestibulocerebellar tract (Suárez et al., 2008). Optogenetically-induced plasticity and motor adaptation in the cerebellum are dependent on EC signaling. When vestibular stimuli are paired with relatively small amplitude Purkinje cell calcium responses, this Purkinje cell activation induces decreased vestibulo-ocular reflex gain. By contrast, pairing with large magnitude calcium responses promotes vestibulo-ocular reflex gain. This indicates that EC signaling acts downstream of Purkinje cell calcium elevation (Bonnan et al., 2021). CB1 receptors modulate the vestibular reflex, as indicated by their activation following unilateral vestibular deafferentiation in rats. However, their expression does not change during vestibular compensation (Ashton et al., 2004). In rats, bilateral vestibular deafferentation causes lifelong spatial memory impairments and hippocampal dysfunction (Baek et al., 2010), that is associated with CB1 receptor down-regulation in the CA3 area of the rat hippocampus (Baek et al., 2011).

EC signaling modulates neurotransmission within auditory circuits and contributes to their development (Chi and Kandler, 2012; Trattner, et al., 2013). CB1-receptor knockout mice have lower auditory thresholds at hearing frequencies greater than 8 kHz in their audiograms, than wild type mice. In terms of central auditory processing, these knockout mice outperform wild-type mice in identifying gaps in low-pass noise bursts (Toal et al., 2016). CB2 receptors are found in several areas of the inner ear, including the Organ of Corti, stria vascularis, spiral ligament, and spiral ganglion cells (Ghosh et al., 2018). CB2 receptor activation is involved in protecting against drug-induced hearing loss (Martín-Saldaña et al., 2016; Dhukhwa et al., 2021).

The present review shows that the EC system regulates audiovestibular function. The few but significant references to the use of CB agonists/antagonists, and MAGL/FAAH inhibitors that act on the EC system indicate that they may be effective for treating audiovestibular disorders. The neuromodulatory effects of those compounds would act on symptoms as diverse as dizziness, nystagmus, nausea, vomiting, tinnitus or headache across different vestibular syndromes of central and peripheral origin (linked or not to neurological pathology), in some cases with inhibitory effects, particularly CB agonists, with sedative action and excitatory effects in others, and with compensatory or protective functions, as is typical in the case of CB antagonists. A critical aspect concerning the use of drugs that act on the EC system is the absence and/or variability of doses used in all studies, even in those studies performed in animal models. This severely limits the application of EC system-targeting drugs and standardization in humans. A further issue that requires discussion is the lack of pharmacogenetic profiles for most cannabinoid derivatives, which affects all conditions that may benefit from their use. However, an understanding of the pharmacogenetics of the cannabinoid derivatives dronabinol, nabilone and cannabidiol is undoubtedly invaluable. Genes related to the metabolism and/or to the mechanisms of action of these drugs contribute to problems with therapeutic efficacy and safety of those agents. A pharmacogenomic strategy to optimize the use of drugs that interact with the EC system is required for successful, personalized therapy. Finally, whereas systemic drug administration allows for wider drug distribution to exert central and peripheral effects, the effects of local delivery of these agents to the inner ear via more invasive direct approaches are unknown. This application would be particularly appealing for those drugs with peripheral action or those with minimal CNS effects.

JG wants to acknowledge Adria Pijuan Monroy and Iria Rodriguez Lopez for being the inspiration to prepare this article.

Made substantial contributions to conception and design of the review and interpretation, JG, read, adjusted and approved the final manuscript, VN and RC.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.