Inner Ear Gene Therapy: An Overview from Bench to Bedside

Abstract

Hearing loss represents a highly prevalent and debilitating sensory disorder affecting roughly one in five people worldwide. In a majority of patients with congenital hearing loss, genetic mutations cause the disease. Up until recently, therapeutic options for individuals with hearing loss were limited to hearing aids and different types of auditory implants. However, after numerous years of intensive basic and translational research, gene therapy strategies are now being investigated in clinical trials. First results show significant hearing improvement in treated patients, highlighting gene therapy’s role as a promising treatment for certain forms of genetic hearing loss. In this article, we provide an overview of genetic hearing loss and inner ear gene therapy research including relevant strategies that have been established in animal models and will likely be investigated in human patients soon. Furthermore, we summarize and contextualize the novel findings of recently completed and ongoing clinical trials, and discuss future hurdles needed to be overcome to allow for a broad and safe clinical application of inner ear gene therapy.

Key Points

Gene therapy of inner ear disorders represents a promising strategy to treat hereditary hearing loss.

First clinical results of inner ear gene therapy show a good tolerability and safety profile, as well as significant improvements in hearing levels after treatment of children with neurosensory non-syndromic autosomal recessive deafness 9 (DFNB9).

Further translational research efforts must be made to tackle remaining hurdles for a broad application of gene therapy to target inner ear disorders.

Hearing Loss

Hearing loss (HL) refers to any condition in which a person’s ability to perceive auditory stimuli is either impaired or completely lost. This condition represents the most common sensory deficit in humans affecting nearly one in five people totaling to around 1.5 billion people worldwide. Roughly a third of these individuals have disabling HL defined as HL of more than 35 dB in the better hearing ear and thus require some form of hearing aid or other treatment [1]. Nevertheless, HL is not a disease in itself but rather a symptom of existing pathology within the auditory system. Depending on the respective pathology’s location, HL can roughly be classified into (1) conductive HL, in which the transmission of sound waves to the cochlea is dampened or disrupted, (2) sensorineural HL, where those signals cannot be processed or transmitted properly by neural structures, or (3) mixed HL, which displays combined features of conductive and sensorineural HL.

Epidemiology

Even though HL is more prevalent in the elderly population and is in many cases exacerbated by exposure to noise, ototoxic substances, or infections, already about 2 per 1000 children are born with a clinically relevant degree of HL [2, 3]. In more than half of these children, genetic mutations can be determined as the underlying cause of HL [4]. In roughly a third of patients with inherited HL, accompanying clinical features in at least one other organ/functional system can be observed, resulting in a classification of these cases as syndromic HL. The most common syndromes associated with hearing impairment include Down syndrome, Treacher-Collins syndrome, Usher syndrome, Pendred syndrome, and Waardenburg syndrome [5]. In the remaining ~ 70% of patients, HL occurs as an isolated feature, and is thus classified as non-syndromic. These cases can further be distinguished by the mode of inheritance, with an autosomal recessive inheritance accounting for up to 75–80%, autosomal dominant for roughly 20%, and X-linked or mitochondrial for the remaining cases [6]. Over 150 genetic loci have been found to be associated with monogenic non-syndromic HL [7]. The majority of these genes affect cells within the cochlea, mostly hair cells, supporting cells, or cells of the stria vascularis. They frequently encode proteins such as gap or tight junctions (e.g., Connexin 26 via GJB2), ion channels (e.g., potassium channels via KCNQ4), proteins involved in neurotransmission (e.g., otoferlin via OTOF), motor proteins (e.g., Myosin VIIA via MYO7A), or transcription factors (e.g., POU4F3). Subsequently, mutations in these genes interrupt various molecular or cellular mechanisms including inner ear development, sound transduction, production and maintenance of endocochlear potential, and synaptic transmission between cochlear hair cells and spiral ganglion neurons [8]. Notably, in a small subset of patients, non-syndromic “mimics” may be observed. In these cases of syndromic HL, HL presents as an isolated clinical trait while other syndromic clinical features may have a mild/atypical phenotype or appear with a delayed onset, thus mimicking the clinical phenotype of non-syndromic HL. If not assessed and diagnosed appropriately at a young age, this may expose patients to significant risks of serious morbidity, for example, by overlooking QT prolongation in cases of Jervell and Lange-Nielsen syndrome [9, 10].

Consequences of HL

In any case of untreated debilitating hearing impairment, a serious impact on the person’s life can be expected. Hearing is, of course, more than just the physiological act of conveying and processing acoustic stimuli, as it plays a key role in communication, and therefore, provides the foundation for social interactions. Consequently, loss of hearing can result in social isolation, and may exacerbate or accelerate the development of cognitive decline, dementia, or depression, altogether severely impacting quality of life [11–14]. Moreover, in children, normal hearing is essential for proper speech acquisition and thus, HL in children can have detrimental effects on their speech, language, cognition, and education [15, 16]. Furthermore, the concomitant sensory deprivation associated with congenital HL can also negatively influence neuroplasticity in both auditory and non-auditory brain regions resulting in permanent structural and behavioral changes [17, 18].

Diagnosis

In order to allow early counseling and management of congenital HL to subsequently ensure proper language development, diagnosis of HL etiology should start as early as possible. The newborn hearing screening established in most developed countries around the world has immensely helped to comprehensively identify newborns with HL [19]. Nevertheless, definitive diagnosis of genetic HL represents a time- and resource-consuming process, usually including several assessments by medical professionals, comprehensive audiometric testing, and radiologic imaging, ultimately followed by genetic testing. Current guidelines of the American College of Medical Genetics and Genomics recommend a tiered genetic testing approach for non-syndromic HL [20]. First testing should constitute comprehensive HL gene panels using next-generation sequencing methods including recommended genes known to cause HL [21]. Next, if these panels return negative, genome-wide testing with exosome or genome sequencing can be considered. Finally, it must be noted that even if comprehensive genetic testing does not reveal a causative gene mutation or variation, a genetic etiology may still be the underlying cause for the clinically detected HL [22].

Currently Available Therapies

Recent advances in next-generation sequencing techniques have considerably reduced the costs and complexity of detecting genetic HL and thus, immensely improved diagnosis and clinical decision making [5, 23]. Nonetheless, treatment options for congenital genetic HL are still scarce as approved treatment options only encompass different hearing aids or implants. In cases of mild-to-moderate HL, conventional hearing aids, which amplify acoustic stimuli, are the treatment of choice. In patients with severe-to-profound HL, alternative options such as the cochlear implant (CI), which bypasses impaired sensory structures within the cochlea, must be considered. However, hearing through or with hearing aids and implants does not replicate hearing with healthy ears, especially in noisy environments or when trying to appreciate nuanced listening situations such as music [24, 25]. Moreover, because these are not causative treatments, progression of HL and further degeneration of cochlear and neural structures cannot be halted, subsequently diminishing the potential benefits of these hearing aids or sometimes even implants [26].

Over the last few years, numerous breakthroughs have been achieved by adopting various different gene therapy approaches to treat and cure genetic HL. For a long time, these strategies have been extensively studied in different animal models. Most recently, however, the translation of these findings into the first clinical trials in humans has yielded highly promising results.

Gene Therapy

Gene therapy is an umbrella term for different therapeutic approaches targeted at altering the gene expression or modifying the biological properties and functions of living cells. These objectives are typically achieved by using vectors to transfer genetic materials into the respective target tissue, in which the delivered nucleic acid sequences are transcribed and translated to subsequently exert the desired therapeutic effects [27]. These include the substitution of missing genes, the regulation of gene expression, as well as editing, replacing, or removing specific genetic sequences [28].

Gene Therapy Vectors

To achieve the above-mentioned goals of gene therapy, genetic material or tools must be transferred into the target tissue or cells. Several vectors have been identified, studied, and used for this exact purpose, all of which can broadly be categorized as either viral or non-viral vectors (Table 1).

Table 1.

Overview and description of vectors commonly used in inner ear gene therapy

	Viral vectors			Non-viral vectors, e.g., nanoparticles, liposomes, free DNA
	Adenovirus	Adeno-associated virus	Lentivirus
Genome	dsDNA	ssDNA	ssRNA	n/a
Packaging capacity	7.5–36 kB	~4.5 kB	8 kB	Variable
Host-genome interaction	Non-integrating	Non-integrating	Integrating	Non-integrating
Transgene expression	Transient	Transient and permanent	Transient and permanent	Transient and permanent
Advantages	Broad host range, large cargo, high transduction efficiency	Good safety profile, replication deficiency, persistent transgene expression, transduction of replicating and quiescent cells	Broad host range, large cargo, low immunogenicity, transgene integration resulting in persistent expression	Minimal immune response, low production costs, high packaging capacity
Disadvantages	High immunogenicity, pre-existing immunity, short transgene expression, pathogenicity	Small cargo, tissue specificity, possible immune responses	Insertional mutagenesis, pathogenicity, replication competence, risk of virus dissemination from patients	Low transduction efficiency, low tissue specificity, possible toxicity

dsDNA double-stranded DNA, kB kilobases, ssDNA single-stranded DNA, ssRNA single-stranded RNA

For viral vector-based gene therapy, the transgene is packaged into a viral capsid, thus facilitating the delivery into the target cells. Most commonly, viral vectors used in inner ear gene therapy are based on adenoviruses, adeno-associated viruses (AAVs), or lentiviruses (LVs). These vectors offer the ability to effectively transduce a broad range of target cells, have modifiable capsids, and display a high transduction efficiency over the short and long term [29, 30]. In most inner ear gene therapy strategies, AAVs have emerged as the vectors of choice because of their low immunogenicity, ability to permanently transduce both mitotic and quiescent cells, and their modifiable capsid offering tailored vectors for specific target cells [31, 32]. Nevertheless, to a varying extent, viral vectors also have the possible disadvantages of a limited cargo size, safety issues in clinical application, high costs associated with the production of clinical-grade viral vectors, and, especially in the case of LVs, the risk of viral shedding and insertional mutagenesis. Alternatively, non-viral vectors can be used for the transgene transfer. These vectors consist of synthetic particles, such as liposomes or polymers, inorganic particles, such as gold or silicone, or naked DNA being transported via physical methods, such as electroporation or sonoporation [33]. Non-viral vectors can offer the possibility of short-term transduction effects, which is especially desirable when transferring gene editing proteins. Moreover, they can be engineered to possess a higher transgene cargo size and are in principle easily and cheaply produced. Finally, these non-viral vectors may display a superior safety profile compared with viral vectors, although this often comes at the cost of significantly lower transduction efficiency and cell specificity [34].

Vector Delivery Routes to the Cochlea

The inner ear as part of the central nervous system is separated from the systemic circulation by the blood–labyrinth barrier, which limits the passage of large molecules from the blood into the inner ear [35]. Furthermore, it is located deep inside the temporal bone, the densest bone in humans, restricting physical access to the cochlea and limiting fluid volumes safely applicable into the inner ear. Hence, efficient drug and vector delivery to the cochlea is typically based on different local application techniques [36, 37].

Several techniques have been established for safe and reliable vector delivery in rodents and non-human primates (NHPs) alike (Fig. 1). Even though systemic delivery represents a technically simple approach with low invasiveness, its efficacy is severely hampered by the above-mentioned seclusion of the inner ear and it poses the risk of potential off-target effects after vector delivery [38]. Local application techniques, however, can effectively deliver large quantities of vectors directly into the inner ear. These include application onto the tympanic membrane (transtympanic) or into the middle ear (intratympanic), both of which rely on subsequent passive diffusion of vectors into and within the inner ear and therefore often suffer from distribution issues and do not reach their intended targets [39]. Alternatively, gene therapy vectors are routinely injected directly into the inner ear’s fluid compartments using either natural entry points, such as the stapes foot plate or round window membrane, or artificial surgical openings of the inner ear, for example, of the cochlea (cochleostomy), or of the vestibular system’s utricle or semicircular canals (canalostomy) [36]. In conjunction with these delivery methods, a venting hole, for example, in the stapes footplate or a semicircular canal after a round window membrane injection, can prevent an increase of intracochlear pressure while at the same time creating a concentration gradient that ensures even vector distribution along the whole cochlear length [40]. Finally, in rodents, primates, and likely the majority of human patients, injection of vectors into the cerebrospinal fluid can also result in the transduction of cochlear cells owing to the cochlear aqueduct connecting the subarachnoid space with the intracochlear peri-lymphatic space [41–43].

Fig. 1.

Overview of local vector delivery strategies to the inner ear. Local delivery techniques include transtympanic (TT) delivery to the tympanic membrane, intratympanic (IT) injection into the middle ear, intracochlear delivery via the round window membrane (RWM; yellow), a cochleostomy (CoS), or a canalostomy (CaS), and delivery through the cochlear aqueduct (CA) via the cerebrospinal fluid (CSF). The purpose and mechanism of a venting hole in the stapes footplate (blue) to allow even distribution along the cochlea is depicted using red and green arrows. This figure was created and adapted from Servier Medical Art (https://smart.servier.com/CC BY 4.0)

Inner Ear Gene Therapy

Because of the heterogeneity of affected genes and cells as well as the elicited disease mechanisms, various different vectors, delivery approaches, and therapeutic strategies have been established for treating genetic HL. Depending on the underlying pathogenic mechanism caused by the (monogenic) mutation, different gene therapy strategies may be adopted [28]. In cases of loss-of-function variants as seen in most of the autosomal recessive disorders, gene replacement/augmentation strategies can introduce an intact gene copy and thus restore the affected protein’s natural function. This strategy may also be used in cases of dominant disorders with haploinsufficient loss-of-function mutations. In dominant disorders with either dominant-negative or gain-of-function mutations resulting in an impaired or aberrant protein function, gene suppression strategies can inhibit translation of the mutated protein at the messenger RNA level and thus restore regular function. Finally, genetic mutations may be corrected at the DNA level by gene editing strategies, which can be employed in all of the above-mentioned pathogenic mechanisms.

Gene Replacement/Augmentation

Gene replacement or augmentation therapy aims at delivering a regular gene copy to introduce a functional protein in cases of loss-of-function mutations as seen in autosomal recessive or in haploinsufficient dominant disorders. Because of the high prevalence of autosomal recessive mutations in hereditary HL and this relatively straightforward therapeutic strategy, gene replacement strategies are most often used in inner ear gene therapy at the moment. An overview of studies investigating gene replacement strategies to treat genetic HL is provided in Table 2. From this compilation of articles, it becomes evident that the most common genetic targets for these strategies involve mutations of otoferlin (OTOF), connexin-26 (GJB2), transmembrane channel like 1 (TMC1), clarin-1 (CLRN1), and connexin-30 (GJB6).

Table 2.

Overview of studies employing a gene replacement/augmentation strategy to treat genetic HL

Gene replacement/augmentation
Gene	Expression site	Animal model	Treatment age	Vector	Delivery method	Results	Reference
OTOF	IHCs	Otof-KO	P6-P7	Dual AAV2/6	RWM injection	Hearing improvement (thresholds: 70 dB) up to 3 weeks after treatment	Al-Moyed et al. (2019) [44]
		Otof-KO	P10, P17, P30	Dual (AAV2 quadY-F)	RWM injection	Hearing improvement (thresholds of 35 dB similar to WT) up to 7.5 months after treatment	Akil et al. (2019) [45]
		Otof-KO	P5-P7	AAV9-PHP.B	RWM injection	Hearing improvement (thresholds: 60 dB) up to 1 month after treatment	Rankovic et al. (2021) [46]
		Otof-KO	P0-P2, P30	Dual AAV-PHP.eB	RWM injection	Hearing improvement (thresholds of 40 dB similar to WT) up to 6 months after treatment; hearing improvement in contralateral ear	Tang et al. (2023) [47]
		Otof-KO	P0-P2	Dual AAV1	RWM injection	Hearing improvement (thresholds: 60 dB) up to 6 months after treatment; AAV1 detected in the CNS and liver	Zhang et al. (2023) [48]
		Macaca fascicularis	5–7 years		RWM injection	Efficient gene expression, no hearing measurements
		Otof-KO	P1-P3, P30	Dual AAV with AAV1 and AAV2/Anc80L65	RWM injection	Hearing improvement (thresholds: 50 dB) up to 3 months after treatment; no off-target transduction; minimal (non-significant) adverse effects on WT mice	Qi et al. (2024) [49]
		Macaca fascicularis	3–7 years		RWM injection	Efficient gene expression, no change in hearing thresholds after injection
		Otof-KO	P0-P2	Dual AAV-PHP.eB	RWM injection	Hearing improvement (thresholds: 60 dB) up to 3–4 weeks after treatment; significantly reduced off-target expression in inner ear & CNS with Myo15 promoter	Wang et al. (2024) [50]
		Otof-KO	P0-P2	Dual AAV-PHP.eB	RWM injection	Hearing improvement (thresholds: 45 dB) up to 13 months after treatment	Hu et al. (2024) [51]
TMC1	HCs	Tmc1∆/∆	P0-P2	AAV2/1	RWM injection	Hearing improvement (thresholds: 90 dB) up to 1 month after treatment	Askew et al. (2015) [52]
		Tmc-Bth	P0-P2	AAV2/1	RWM injection	Hearing improvement (thresholds: 95 dB) up to 1 month after treatment
		Tmc1∆/∆	P0-P1, P4, P7, P14	sAAV	RWM injection	Hearing improvement (thresholds: 60 dB) up to 3 months after treatment	Nist-Lund et al. (2019) [53]
		Tmc1∆/∆	P1, P7	AAV9-PHP.B	Utricle injection	Hearing improvement (thresholds: 60 dB) up to 1 year after treatment	Wu et al. (2021) [54]
		Tmc1-Baringo	P1, P7	AAV9-PHP.B	Utricle injection	Hearing improvement (thresholds: 60 dB) up to 1 year after treatment
		Tmc1∆/∆	P1-P2	AAV9-PHP.B	Utricle injection	Hearing improvement (thresholds: 50 dB) up to 3 months after treatment; viral DNA and RNA found in the contralateral ear, brain, and liver, but no consequences observed	Marcovich et al. (2022) [55]
		Tmc1-p.N193I	P1-P2	AAV9-PHP.B	Utricle injection	Hearing improvement (thresholds equivalent to WT at 25 dB) up to 3 months after treatment; viral DNA and RNA found in the contralateral ear, brain, and liver, but no consequences observed
TMC1/TMC2	Vestibular HCs	Tmc1∆/∆ + Tmc2∆/∆	P0-P1, P14, P30	AAV2/Anc80L65	RWM injection	Almost full prevention of balance defects up to 1.5 months after treatment, hearing improvement see above	Nist-Lund et al. (2019) [53]
		Tmc1∆/∆ + Tmc2∆/∆	P1	AAV9-PHP.B	Utricle injection	Protein production and improvement of vestibular evoked potentials, no hearing tests reported	Ratzan et al. (2024) [56]
CLRN1	HCs	Clrn1-KO	P1-P3	AAV2, AAV8	RWM injection	Onset of progressive HL delayed by up to 3 months	Geng et al. (2017) [57]
		Clrn1ex4-/-, cKO	P1-P3	AAV2/8	RWM injection	Hearing improvement (thresholds: 40 dB) up to 1 month after treatment	Dulon et al. (2018) [58]
		Clrn1-KO	P0-P1	AAV9-PHP.B	RWM injection	Hearing improvement (thresholds: 40 dB) up to 1 month after treatment	György et al. (2019) [59]
		Macaca fascicularis	2.6–3.1 years		RWM injection	Efficient gene expression, similar ABR waveforms with some elevation of thresholds post-injection
	HCs / SGNs	Clrn1-KO	P1	AAV-S	RWM injection	Hearing improvement (thresholds of 35 dB similar to WT at low and middle frequencies) up to 5 months after treatment	Ivanchenko et al. (2021) [60]
		Macaca fascicularis	1–3 years		RWM injection	Efficient gene expression, no change in hearing thresholds after injection
CABP2	IHCs	Cabp2-KO	P5-P7	AAV2/1	RWM injection	Hearing improvement (thresholds: 50 dB) up to 4–7 weeks after treatment	Oestreicher et al. (2021) [61]
	IHCs	Cabp2-KO	P5-P7	AAV9-PHP.eB	RWM injection	Hearing improvement (thresholds: 50 dB) up to 4–7 weeks after treatment
GJB2	SCs	Gjb2-CKO	P0-P1	AAV2/1	Cochleostomy	Protein production and gap junction restoration, no hearing improvement reported	Yu et al. (2014) [62]
	SCs, fibrocytes of the spiral limbus and spiral ligament	Gjb2-CKO	P0, P42	AAV1	RWM injection	Hearing improvement (thresholds: 70 dB) up to 3 months after treatment	Iizuka et al. (2015) [63]
	HCs, SCs	Gjb2-iCKO	P28	AAV2/Anc80L65	RWM injection	Protein production, but no hearing improvement; off-target transduction to HCs; further hearing and HC loss in some animals	Guo et al. (2021) [64]
GJB6	SCs	Gjb6-KO	E11.5	Plasmid	Transuterine otocyst injection	Hearing improvement (thresholds: 50 dB) at P30 (treatment of embryo)	Miwa et al. (2013) [65]
		Gjb6-KO	P4	Bovine AAV	Canalostomy	Protein production up to 1 month after treatment, no hearing improvement reported	Crispino et al. (2017) [66]
		Gjb6-KO	P0-P2	AAV1	Scala media injection, Scala tympani injection	Hearing improvement (thresholds: 40 dB) up to 2 months after treatment	Zhang et al. (2022) [67]
MYO7A	HCs	Shaker-1	P0-P5	Dual AAV	PSCC injection	Improved vestibular function and HC survival, no hearing improvement	Lau et al. (2023) [68]
		Shaker-1	P4	Lentivirus	Canalostomy	Hearing improvement (thresholds: 75 dB) up to 2.5 months after treatment	Schott et al. (2023) [69]
TMPRSS3	Fetal cochlea	Tmprss3tm1/tm1	P1	AAV-KP1, AAV-DJ	PSCC injection	Hearing improvement (thresholds: 60 dB) up to 4 months after treatment	Aaron et al. (2023) [70]
		Tmprss3A306T/A306T	54–88 weeks	AAV2	RWM injection with canal fenestration	Hearing improvement (thresholds of 30 dB similar to WT) up to 5 months after treatment	Du et al. (2023) [71]
VGLUT3	IHCs	VGlut3-KO	P1-P3, P10-P12	AAV2/1	RWM injection, cochleostomy	Hearing improvement (thresholds: 40 dB) up to 1.5 years after treatment	Akil et al. (2012) [72]
		VGlut3-KO	P3; 5, 8, 20 weeks	AAV8	PSCC injection	Hearing improvement (thresholds: 30 dB) up to 3 months after treatment	Zhao et al. (2022) [73]
WHRN	HCs	Whrnwi/wi	P1-P5	AAV8	RWM injection	Restoration of stereocilium architecture and improved HC survival, no hearing improvement	Chien et al. (2015) [74]
		Whrnwi/wi	P1-P5	AAV8	PSCC injection	Hearing improvement (thresholds: 80 dB) up to 4 months after treatment	Isgrig et al. (2017) [75]
SLC26A4	Epithelial cells of the endolymphatic sac, sensory epithelia of utricle, saccule and SCC, putative spindle and root cells	Slc26a4-KO, Slc26a4tm1Dontuh/tm1Dontuh	E12.5	rAAV2/1	Transuterine otocyst injection	Hearing improvement (thresholds: 30 dB) 3–5 weeks after treatment, decline thereafter (60 dB at 16–21 weeks)	Kim et al. (2019) [76]
		Slc26a4-KO	E11.5	Plasmid	Transuterine otocyst injection	Hearing improvement (thresholds: 60 dB) at P30 (treatment of embryo)	Takeda et al. (2019) [77]
PJVK	HCs, SGNs	Pjvk-KO	P3	AAV2/8	Intracochlear injection	Hearing improvement (thresholds: 45 dB) up to 3 weeks after treatment	Delmaghani et al. (2015) [78]
		PjvkG292R/G292R	P0-P1	AAV2/Anc80L65	RWM injection	Hearing improvement (thresholds: 50 dB) up to 3 weeks after treatment	Lu et al. (2022) [79]
USH1G	HCs	Ush1g-KO	P2.5	AAV8	RWM injection	Hearing improvement (thresholds: 75 dB) up to 3 months after treatment	Emptoz et al. (2017) [80]
		Ush1g-KO	P13-P25	AAV2/Anc80L65	RWM injection	Hearing improvement (thresholds: 70 dB) up to 3 weeks after treatment	Lahlou et al. (2024) [81]
KCNQ1	SV marginal cells	kcnq1-KO	P0-P2	AAV1	Cochleostomy	Hearing improvement (thresholds: 45 dB) up to 4.5 months after treatment	Chang et al. (2015) [82]
MSRB3	HCs	Msrb3-KO	E12.5	AAV2/1	Transuterine otocyst injection	Hearing improvement (thresholds: 40 dB) at P28 (treatment of embryo)	Kim et al. (2016) [83]
USH1C	HCs	Ush1Cc.216G>A	P0-P1, P10-P12	AAV2/Anc80L65	RWM injection	Hearing improvement (thresholds: 40 dB) up to 6 months after treatment	Pan et al. (2017) [84]
CLRN2	HCs	Clrn2-KO	P1-P15	AAV-PHP.eB	RWM injection	Hearing improvement (thresholds: 35 dB) up to 6 months after treatment	Mendia et al. (2024) [85]
LHFPL5	HCs, SGNs	Lhfpl5-KO	P0-P2	Exo-AAV1	RWM injection, cochleostomy	Hearing improvement (thresholds: 80 dB) up to 1 month after treatment	György et al. (2017) [86]
SLC26A5	OHCs	Prestin-KO	P3	AAV-ie	RWM injection	Hearing improvement (thresholds: 70 dB) up to 1 month after treatment	Tao et al. (2022) [87]
STRC	OHCs	Strc-KO	P0-P1	Dual AAV9-PHP.B	Utricle injection	Hearing improvement (thresholds: 50 dB) up to 3 months after treatment	Shubina-Oleinik et al. (2021) [88]
KCNE1	SV marginal cells, vestibular dark cells	Kcne1-KO	P0-P2	AAV1	PSCC injection	Hearing improvement (thresholds: 60 dB) up to 5 months after treatment	Wu et al. (2021) [89]
SYNE4	OHCs	Syne4-KO	P0-P1.5	AAV9-PHP.B	PSCC injection	Hearing improvement (thresholds: 20 dB) up to 3 months after treatment	Taiber et al. (2021) [90]
PCDH15	HCs	Pcdh15-KO	P1	AAV9-PHP.B	RWM injection	Hearing improvement (thresholds: 40 dB) up to 2 months after treatment	Ivanchenko et al. (2023) [91]
NDP	HCs, cochlear microvasculature	Ndptm1Wbrg	P2, P21, P30	AAV9	Intravenous	Hearing improvement (thresholds: 30 dB) up to 3 months after treatment	Pauzuolyte et al. (2023) [92]
ILDR1	HCs, SCs, SV marginal cells	ILDR1-KO	P0-P5	Dual AAV2.7m8/AAV8BP2	PSCC injection	Hearing improvement (thresholds: 60 dB) up to 2 weeks after treatment	Isgrig et al. (2023) [93]
EPS8	HCs	Eps8-KO	P1-P3	AAV2/Anc80L65	RWM injection	Recovery of HCs, no hearing improvement	Jeng et al. (2022) [94]

ABR auditory brainstem response, CNS central nervous system, HCs hair cells, HL hearing loss, IHCs inner hair cells, (P)SCC (posterior) semicircular canal, SCs supporting cells, SGNs spiral ganglion neurons, SV stria vascularis, RWM round window membrane

Gene Suppression

In cases of heterozygous dominant mutations, which account for up to 80% of autosomal dominant forms of non-syndromic HL, suppression or silencing of the mutated allele can restore normal function through subsequent expression of the wild-type allele [95]. For this purpose, post-transcriptional inhibition of the aberrant mutation at the RNA level can be achieved by delivery of either antisense oligonucleotides or via RNA interference using small interfering RNA or microRNA. These subsequently inhibit the translation of the mutated dominant allele’s messenger RNA, thus allowing regular translation of the wild-type allele messenger RNA into functional protein. The use of antisense oligonucleotides has been shown to effectively suppress or correct an aberrant splice site in a mutation of harmonin (USH1C) resulting in significantly improved hearing levels following treatment [39, 96]. Suppressing dominant-negative mutations of GJB2 or TMC1 via delivery of small interfering RNA or microRNA, respectively, successfully inhibited translation of the aberrant gene sequence and subsequently improved hearing function in treated animals [97, 98]. Table 3 provides an overview of pre-clinical studies using gene suppression strategies as a treatment of HL.

Table 3.

Overview of studies employing a gene suppression strategy to treat genetic HL

Gene suppression
Gene	Expression site	Animal model	Treatment age	Vector	Delivery method	Results	References
USH1C	HCs	Ush1cc.216G>A	P5, P10	ASO-29	Intraperitoneal injection	Hearing improvement (thresholds: 50 dB) up to 3 months after treatment	Lentz et al. (2013) [96]
		Ush1cc.216G>A	P1, P5, P15	ASO-29	Intraperitoneal injection	Recovery of normal balance behaviors, no hearing improvement	Vijayakumar et al. (2017) [99]
		Ush1cc.216G>A	P1, P3, P5, P7	ASO-29	Intraperitoneal injection	Hearing improvement (thresholds of 25 dB similar to WT) up to 6 months after treatment	Ponnath et al. (2018) [100]
		Ush1cc.216G>A	P1, P5, P10	ASO-29	RWM, trans-tympanic membrane, topical tympanic membrane injection	Hearing improvement (thresholds: 50 dB) up to 6 months after treatment	Lentz et al. (2020) [39]
		Ush1cc.216G>A	E12.5	ASO-29	Transuterine otocyst injection	Hearing improvement (thresholds: 50 dB) up to P220 (treatment of embryo)	Wang et al. (2020) [101]
TMC1	HCs	Tmc1-Bth	P0-P2	rAAV2/9	RWM injection	Delay of progression of HL up to 35 weeks after treatment	Shibata et al. (2016) [102]
		Tmc1-Bth	P15-P16, P56-P60, P84-P90	AAV2/9	RWM injection with canal fenestration	Delay of progression of HL up to 12 weeks after treatment	Yoshimura et al. (2019) [103]
		Tmc1-Bth	P1-P2	AAV-PHP.eB	RWM injection	Hearing improvement (thresholds: 50 dB) up to 2 months after treatment	Zheng et al. (2022) [98]
		Tmc1-Bth	P18	AAV2	RWM injection with canal fenestration	Hearing improvement (thresholds: 30 dB) up to 5 months after treatment	Iwasa et al. (2023) [104]
GJB2	SCs	C57BL6 + GJB2R75W-eGFP-injection	P42-P45	Liposome	RWM injection	After induced transduction of GJB2R75W, siRNA treatment improved hearing thresholds to control WT levels	Maeda et al. (2005) [97]

ABR auditory brainstem response, ASO antisense oligonucleotide, HCs hair cells, HL hearing loss, miRNA microRNA, RWM round window membrane, SCs supporting cells, siRNA small interfering RNA

Gene Editing

The advent of gene editing technologies has opened up the possibility of correcting genetic mutations by directly adding, deleting, or replacing mutated DNA sequences. In general, two different methods are being used for gene editing. First, the CRISPR-Cas9 system induces targeted DNA double-strand breaks, subsequently triggering natural DNA repair mechanisms, which can be exploited to induce the desired DNA modification. Second, the method of base editing facilitates single nucleotide changes by targeted conversions of singular DNA or RNA bases via adenine, cytosine, or RNA base editors. In principle, both of these versatile approaches can treat any type of genetic mutation. In hearing research, these approaches of gene editing have been employed to correct various mutations including OTOF, KCNQ4, or Usher type IF [105–107]. Moreover, CRISPR-Cas9 technology has also been used to successfully generate a novel NHP model of Usher type IB (USH1B) [108]. Notable pre-clinical studies employing gene editing approaches for genetic HL are presented in Table 4.

Table 4.

Overview of studies employing a gene editing strategy to treat genetic hearing loss

Gene editing
Gene	Expression site	Animal model	Treatment age	Vector	Delivery method	Results	Reference
TMC1	HCs	Tmc1-Bth	P0-P2, P42	Lipofectamine 2000 (gene disruption via CRISPR-Cas9)	PSCC injection	Hearing improvement (thresholds: 55 dB) up to 1 month after treatment	Gao et al. (2018) [109]
		Tmc1-Bth	P1-P2	AAV2/Anc80L65 (gene disruption via CRISPR-Cas9)	Inner ear injection	Hearing improvement (thresholds: 60 dB) up to 1 year after treatment	György et al. (2019) [110]
		Tmc1-Baringo	P1	Dual AAV2/Anc80L65 (gene correction via CBE)	Inner ear injection	Hearing improvement (thresholds: 90 dB) up to 1 month after treatment	Yeh et al. (2020) [111]
		Tmc1-Bth	P1	Dual AAV9-PHP.B (gene disruption via CRISPR-Cas9)	Utricle injection	Hearing improvement (thresholds: 40 dB) up to 6 months after treatment	Wu et al. (2021) [54]
		Tmc1-Bth + ATP2b2Obl/+	P0-P2	Lipofectamine 2000 (gene disruption via CRISPR-Cas9)	Canalostomy	Hearing improvement (thresholds: 85 dB) up to 1 month after treatment	Tao et al. (2023) [112]
OTOF	IHCs	Otofc.2485C>T	P0-P3	AAV9-HGHC (gene correction via RBE)	Scala media injection	Hearing improvement (thresholds of 25 dB similar to WT) up to 7.5 months after treatment	Xue et al. (2023) [105]
		Otofc.2485C>T	P5-P7, P30	AAV9-HGHC (gene correction via RBE)	RWM injection	Hearing improvement (thresholds: 40 dB) up to 5 months after treatment
KCNQ4	OHCs	Kcnq4c.830G>C	P1-P3	AAV2/Anc80L65 (gene disruption via CRISPR-Cas9)	Utricle, PSCC, RWM, scala media injection	Hearing improvement (thresholds: 35 dB) up to 7 weeks after treatment	Noh et al. (2022) [113]
		Kcnq4c.683G>A	P1-P2	AAV-PHP.eB (gene disruption via CRISPR-Cas9)	Scala media injection	Hearing improvement (thresholds: 50 dB) up to 3 months after treatment	Cui et al. (2022) [106]
MYO6	HCs	Myo6p.C442Y	P0-P2	AAV-PHP.eB (gene disruption via CRISPR-Cas9)	Scala media injection	Hearing improvement (thresholds: 45 dB) up to 5 months after treatment	Xue et al. (2022) [114]
		Myo6p.C442Y	P0-P2	AAV-PHP.eB (gene correction via RBE)	Scala media injection	Hearing improvement (thresholds: 40 dB) up to 3 months after treatment	Xiao et al. (2022) [115]
PCDH15	HCs	Pcdh15av-3J	P0-P2	AAV2/9 (gene disruption via CRISPR-Cas9)	Scala media injection	Hearing improvement (thresholds: 90 dB) up to 1 month after treatment	Liu et al. (2022) [116]
MYO7A	HCs	Zygotes of Macaca mulatta	Zygote	None (gene disruption via naked CRISPR-Cas9 system)	Intra-zygote injection	No treatment, but generation of an USH1B-primate-model: mutant sequence in 40–50% of cells; no ABR/DPOAE response at birth, 26 kHz ABR thresholds bilaterally at 1 month	Ryu et al. (2022) [108]
		Shaker-1	P4	Extracellular vesicles (gene disruption via CRISPR-Cas9)	PSCC injection	Hearing improvement (thresholds: 40 dB) up to 6 months after treatment	Pan et al. (2023) [117]
ATP2B2	OHCs	Atp2b2Obl/+	P0-P2	Lipofectamine 2000 (gene disruption via naked CRISPR-Cas9)	Canalostomy	Hearing improvement (thresholds: 45 dB) up to 2 months after treatment	Tao et al. (2023) [112]
KLHL18	IHCs	Klhl18p.V55F	P1	Dual AAV with AAV9 and AAV-PHP.eB (gene correction via CRISPR/Cas9)	Inner ear injection	Hearing improvement (thresholds: 45 dB) up to 3 months after treatment	Gu et al. (2022) [118]
PCDH15	HCs	Pcdh15p.R245X	P0-P1	Dual AAV9-PHP.B (gene correction via ABE)	RWM injection	Hearing improvement (thresholds: 90 dB) up to 2 months after treatment; transduction of contralateral ear	Peters et al. (2023) [107]
MIR96	IHCs	Mir9614C>A/+	3, 6, 16 weeks	Dual AAV2 (gene disruption via CRISPR-Cas9)	Inner ear injection	Hearing improvement (thresholds: 40 dB) up to 5 months after treatment	Zhu et al. (2024) [119]

ABE adenine base editor, ABR auditory brainstem response, CBE cytosine base editor, DPOAE distortion-product otoacoustic emission, HCs hair cells, IHCs inner hair cells, OHCs outer hair cells, (P)SCC (posterior) semicircular canal, RBE RNA base editor, RWM round window membrane

Clinical Gene Therapy Trials

The steady progress and success made with gene therapy approaches in treating genetic HL in pre-clinical models has paved the way for translation into first clinical trials. So far, six clinical trials have been registered worldwide by different stakeholders including the Eye & ENT Hospital of Fudan University China (ChiCTR2200063181), Regeneron Pharmaceuticals (formerly Decibel Therapeutics; NCT05788536), the Eli Lilly subsidiary Akouos Inc. (NCT05821959), Otovia Therapeutics (NCT05901480), HuidaGene Therapeutics Co. Ltd. (NCT06025032), and Sensorion (NCT06370351) [120–125]. These trials all focus on otoferlin- (OTOF-) related mutations, which disrupt signal transduction at the sensory inner hair cell’s ribbon synapse and cause non-syndromic autosomal recessive deafness (DFNB9).

Five of said studies employ gene replacement delivering a functioning OTOF transgene using a dual AAV vector strategy. Because of the size of the OTOF gene amounting to roughly 6 kB, the complete transgene does not fit into a single AAV. To circumvent a single vector’s limited packaging capacity, the transgene is divided into two parts, packaged separately into two different vectors. Inside a cell transduced by both vectors, the two OTOF snippets can reconstitute a functional transgene for full OTOF expression [126, 127]. The sixth registered trial (Huida Gene Therapeutics, NCT06025032) aims to use an RNA-base editing approach in cases of HL caused by the p.Q829X mutation in the OTOF gene. In an animal model, this approach has successfully induced an adenosine-to-inosine conversion at the RNA level, leading to near-to-normal OTOF expression, and hearing improvement up to wild-type levels [105].

As of the time of writing, four of the trials have started patient recruitment and the results of two trials, NCT05901480 [128] and ChiCTR2200063181 [129, 130], have been published. In early 2024, the first preliminary data on two children (5 and 8 years of age; followed up for 3 and 2 months, respectively) successfully treated with dual AAV-OTOF within the NCT05901480 trial were published in Advanced Science by Qi et al. [128]. These results were the first published data to not only provide insights into the safety of cochlear gene therapy in humans, but also show that hearing restoration following cochlear gene therapy is achievable up to physiological levels (in the younger unilaterally injected patient) and to hearing levels, in which supply with conventional hearing aids (in the older bilaterally injected patient) is possible. Age is thought to play an important role as animal studies have shown overall that the earlier the injections are carried out, the better the long-term functional outcome [131].

Shortly afterwards, the first case series including six patients treated unilaterally with a dual AAV1-hOTOF was published by Lv et al. in The Lancet [129]. In comparison to the above-mentioned article, which had been published barely 3 weeks prior to The Lancet study, a total of six patients with a complete follow-up until 26 weeks after application were included since this trial (ChiCTR2200063181) had been initiated 9 months earlier than the clinical trial published in Advanced Science. As in the NCT05901480 trial, vectors were applied intracochlearly through the round window membrane with stapes fenestration in the ear of more severe HL (or without CI). Overall, hearing improved in five of the six enrolled patients beginning 4 weeks after application with a steady improvement up to 26 weeks post-application. On average, hearing thresholds determined by auditory brainstem response measurement improved by 40–57 dB at 0.5–4 kHz, which further resulted in improved speech perception. Interestingly, the one patient without a significant hearing improvement following cochlear gene therapy was found to bear the highest titers of anti-AAV neutralizing antibodies (NABs) at baseline (1:135 compared to <1:5 in the other five patients) and after treatment (1:1635 compared to 1:1215 in four other patients and 1:135 in one other patient), which may account for the limited therapeutic benefit. Of further note was the fact that no serious adverse effects or dose-limiting toxicity were observed in any of the patients treated.

Most recently, in June 2024, the first preliminary data on five bilaterally treated pediatric patients with DFNB9 were published by Wang et al. in Nature Medicine [130]. All patients were confirmed bilaterally deaf with averaged auditory brainstem response hearing thresholds > 95 dB at the time of recruitment. After bilateral vector application, hearing levels improved in every ear treated, though with varying benefit. In three patients followed up for 26 weeks, average auditory brainstem response levels were 63 dB at 0.5–4 kHz. Hearing levels of the other two patients followed for 13 weeks averaged at 70 dB. Notably, speech perception and sound localization capability improved in all five patients and again no serious adverse events were observed.

In addition to these published data, press releases and preliminary data presented at international conferences of the clinical trials conducted by Regeneron Pharmaceuticals (NCT05788536) and by Akouos Inc. (NCT05821959) have also shown the rescue of hearing in several patients. As such, hearing improvement has been observed in two children treated with Regeneron’s drug, DB-OTO, and in one following treatment with Akouos’ AK-OTOF [132, 133]. The latter patient was already 11 years of age when injected, with Akouos even including adults up to 44 years of age if the other criteria are met. Regeneron does not use an approach through the ear canal, but accesses the middle ear after a mastoidectomy, the same way CI surgeries are carried out. At least the first few children included in their study will receive contralateral implants, allowing for long-term intra-individual comparisons of the current gold standard and the novel gene therapy.

In conclusion, data on 13 children with OTOF-related autosomal recessive deafness (DFNB9) treated with AAV-hOTOF has been published so far. The published studies greatly helped to establish and confirm the safety and efficacy of intracochlear dual AAV-mediated gene therapy. The influence of anti-AAV NABs and other effect-limiting factors will have to be investigated in the long term to help screen for eligible patients.

Outlook

Gene therapy has revolutionized medicine in countless ways. After numerous years of extensive basic and translational research, genetic therapeutics for hereditary HL have just recently made the final leap into clinical trials. The first data on patients treated with gene therapy yield highly promising results with notable hearing improvements and excellent safety profiles as discussed above. Nevertheless, despite these convincing reports, there are still several barriers to be overcome for a broad application of inner ear gene therapy in clinical practice.

So far, only limited follow-up data up until half a year post-treatment have been published for the enrolled patients, although stable results up to 12 months were presented at conferences. However, it remains unclear at the moment if and how the patients’ hearing will evolve in the long term and whether it may further improve, remain stable, or worsen again owing to reduced transduction possibly necessitating repeated vector administration. However, previous experiences with AAV-mediated gene therapy for other diseases including spinal muscular atrophy, inherited retinal dystrophy, hemophilia A, and Leber hereditary optic neuropathy have already demonstrated robust transduction, efficacy, and safety over a follow-up of several years, which gives hope for similarly promising results in inner ear gene therapy [134–137].

In a majority of patients treated (n = 18 ears) with OTOF-inducing AAVs, a substantial improvement of hearing has been noted. However, in a single patient, only limited hearing rescue was detected, possibly correlating with increased titers of anti-AAV NABs [129]. Because of the natural occurrence of AAVs, a substantial part of the human population has been in contact with these viruses and thus, also possesses pre-existing circulating NABs against wild-type AAV — with the geographic location influencing exposure with and development of NABs against specific serotypes [138, 139]. Little is known about the presence of NABs within immune privileged tissues, and nothing so far about the antibodies’ role in the inner ear or its fluids, endolymph and perilymph. However, there is evidence showing that NAB titers within the central nervous system, in NHPs and humans alike, do not correlate with systemic titers, and thus may have less of a mitigating influence on AAV-mediated gene delivery after appropriate local delivery [140, 141]. Additionally, several strategies have been proposed, which could aid the reduction in immune responses following viral vector administration [142, 143]. First, the application of high vector dosages or decoy capsids can saturate NABs allowing a therapeutically sufficient dose of vectors to remain circulating; yet, this method may be associated with other drawbacks such as increasing manufacturing costs, the total immunogenic burden, and possible toxicity to dorsal root ganglia [144–146]. Second, alterations of the vector may help evade the neutralizing effect of NABs. This could be achieved through switching the vector serotype to one with a lower seroprevalence, chemically modifying the viral capsid, or artificially engineering novel serotypes through directed evolution [147–151]. However, these changes may reduce the vectors’ tropism and/or transduction efficiency, all while they could still be neutralized because of cross-reactivity of NABs [152, 153]. Last, a reduction in NABs or their function may be achieved through modes such as plasmapheresis, depleting antibody counts, or immunosuppression; all of which would require a meticulous risk-benefit evaluation of substantial possible side effects [154–157].

As briefly mentioned above, correctly timing gene therapy is another factor that may influence hearing outcomes in hereditary HL. In most pre-clinical mouse studies, treatment is usually performed within the first few days postnatally — if not even in utero (Tables 2, 3 and 4). Comparing the development of the inner ear of mice and humans, there are great disparities as mice start hearing at around 2 weeks after birth whereas human hearing fully develops at around gestational week 20 [158]. Consequently, data on outcomes following gene therapy in matured animals are rather scarce and partially suggest worse outcomes with advanced age [71, 73]. Moreover, several studies report age-related changes in NAB titers, which appear to decline after birth reaching a nadir between 6 and 12 months of age and subsequently rising again after 3 years of age [159, 160]. Finally, from studies in congenitally deaf children following cochlear implantation, it is known that the central auditory system progressively loses its plasticity after prolonged sensory deprivation [161]. As such, the central auditory pathways retain their full plasticity for roughly 3.5 years followed by another 3.5 years with mixed plasticity before neural plasticity is remarkably reduced. So far, only 5 of the 13 treated children (38.5%) would fall within the age gap of 6 months and 3 years. Nevertheless, as evident from the data published so far, it appears that also older children can greatly benefit from gene therapy prompting future considerations about possible age-limiting effects of gene therapy, which will have to be determined for all genetic mutations individually.

Additionally, one must consider that treatment of different genetic mutations will require targeting different cellular structures. For this purpose, specific vectors must be chosen or uniquely designed to allow targeted transduction of certain cells. In the case of OTOF mutations, appropriate vectors must transduce cochlear inner hair cells where otoferlin exerts its crucial role in synaptic transmission. While several vectors are capable of doing that, two vectors often used to transduce additional targets are AAV9-PHP.B and Anc80L65, both being synthetic AAV variants [31]. AAV9-PHP.B has formerly been investigated for its ability to cross the blood–brain barrier in C57BL/6 mice and was only later found to also efficiently transduce inner ear cells in other mouse strains and NHPs [162]. On the contrary, Anc80L65 was initially found to transduce both inner and outer hair cells in mice but application of this vector in NHPs only displayed effective transduction of inner hair cells [40]. These examples highlight the need for suitable vectors as well as of studies in appropriate large animal models. Furthermore, vectors must also be chosen according to their respective safety profile. While this is less of a problem in AAVs, other viruses—especially LVs—are usually associated with significant pathogenicity when applied in humans. Consequently, novel LVs are designed in ways to reduce risks including insertional mutagenesis while also improving intracochlear cell transduction [69].

Furthermore, the successes observed in treating OTOF mutations may not be translatable to other genetic HL diseases because of the different underlying pathomechanisms present such as hair cell degeneration. The inner ear is encased in one of the hardest bones of the body and opening the otic capsule surrounding the sensory cells usually leads to deafness, which makes it impossible to simply obtain a biopsy that reveals the current state of the cellular integrity. Thus, the most relevant information can be obtained from postmortem specimens of human temporal bones of patients with a given disease. Large collections of these samples are located at a handful of centers around the world and to interpolate the numbers of, for example, remaining hair cells at a certain age, is often not possible. This highlights the need for “natural history studies” regarding the progression of genetic forms of HL. As mature mammalian hair cells lack the ability to regenerate, treatment at birth may miss the appropriate window of opportunity in some diseases as no treatable cells would remain. Several pre-clinical studies have reported the possibility of inducing mitotic regeneration or transdifferentiation of cochlear supporting cells into hair cells [163, 164]. However, these regenerated hair cells seldomly mature fully and may lack re-innervation by spiral ganglion neurons, thus rendering these hair cells non-functional [165]. In humans, clinical trials investigating regenerative approaches have been initiated during the last few years. However, the trials failed to recover or improve hearing in treated patients [166, 167].

However, gene therapy could also be used to optimize the outcomes of currently available technology, such as the CI. Optogenetic cochlear implantation is a strategy that tries to express opsins, special light-sensitive proteins, on the surface of spiral ganglions neurons [168]. While early experiments focused on improved laser stimulation through a cochleostomy, more recent studies have demonstrated the feasibility of multi-channel CIs based on flexible µLED arrays in several different animal models [169–174]. In addition, neurotrophin gene therapy has also been hypothesized to potentially improve CI outcomes by facilitating the stimulation of spiral ganglion neurons [175, 176].

Finally, appropriate strategies for efficient vector delivery to the inner ear must also be taken into account. As stated above, local delivery methods have been established as the most efficient strategies for vector delivery to the inner ear but may also be associated with risks including the damage of sensible structures in the inner ear, for example, through increased intracochlear pressure or simply by opening of the cochlea [177]. Thus, several variations may be made including the surgical approach or the site and details of injection. Depending on the latter, different routes must be chosen: for example, a post-auricular approach is most suitable for applications to the lateral semicircular canal whereas a transcanal/endaural approach offers a less invasive access to the round and oval windows [178]. Last, the use of different application devices such as microneedles or catheter systems may help to minimize trauma during application and to optimize vector distribution along the cochlea [179–181].

What also still has to be determined in human studies are the most relevant parameters that evaluate therapeutic success. While the pure tone audiogram is the most widely used clinical tool, doubts regarding its validity when predicting the cellular damage in the inner ear have been expressed [182]. While adults can undergo a plethora of audiometric testing and a minimum speech test battery for adult cochlear implantation has been in use for many years, the tests that can be carried out with children are more limited and the above-mentioned intra-individual comparisons of current therapeutic gold standards and novel treatment options will be very relevant when these children are old enough to, for example, be assessed regarding more advanced patient-reported outcome and experience measures [183]. Overall, while many variables still have to be determined and optimized, the recent developments suggest a bright future for inner ear gene therapy.

Funding

Open access funding provided by Medical University of Vienna.

Declarations

Funding

Lukas D. Landegger is supported by a Career Development Award from the American Society of Gene & Cell Therapy and the Children’s Tumor Foundation. The content is solely the responsibility of the authors and does not necessarily represent the official views of the American Society of Gene & Cell Therapy. The financial support of the Austrian Federal Ministry for Digital and Economic Affairs, the National Foundation for Research, Technology and Development, and the Christian Doppler Research Association is gratefully acknowledged.

Conflicts of Interest/Competing Interests

Lukas D. Landegger has received research funding from Decibel Therapeutics/Regeneron Pharmaceuticals and Amgen and has worked as an independent consultant for Akouos/Eli Lilly and Company, Conclave Capital, and Gerson Lehrman Group. Anselm J. Gadenstaetter and Paul E. Krumpoeck have no conflicts of interest that are directly relevant to the content of this article.

Ethics Approval

Not applicable.

Consent to Participate

Not applicable.

Consent for Publication

Not applicable.

Availability of Data and Material

Data sharing is not applicable to this article as no original datasets were generated or analyzed during the current study. All data analyzed in this review are directly referenced to their original publications.

Code Availability

Not applicable.

Authors’ Contributions

AJG: methodology, validation, investigation, writing (original draft), visualization; PEK: methodology, validation, investigation, writing (review and editing); LDL: conceptualization, methodology, validation, investigation, writing (review and editing), supervision, project administration.