Autosomal Dominant Non-Syndromic Hearing Loss (DFNA): A Comprehensive Narrative Review

Autosomal dominant non-syndromic hearing loss (HL) typically occurs when only one dominant allele within the disease gene is sufficient to express the phenotype. Therefore, most patients diagnosed with autosomal dominant non-syndromic HL have a hearing-impaired parent, although de novo mutations should be considered in all cases of negative family history. To date, more than 50 genes and 80 loci have been identified for autosomal dominant non-syndromic HL. DFNA22 (MYO6 gene), DFNA8/12 (TECTA gene), DFNA20/26 (ACTG1 gene), DFNA6/14/38 (WFS1 gene), DFNA15 (POU4F3 gene), DFNA2A (KCNQ4 gene), and DFNA10 (EYA4 gene) are some of the most common forms of autosomal dominant non-syndromic HL. The characteristics of autosomal dominant non-syndromic HL are heterogenous. However, in most cases, HL tends to be bilateral, post-lingual in onset (childhood to early adulthood), high-frequency (sloping audiometric configuration), progressive, and variable in severity (mild to profound degree). DFNA1 (DIAPH1 gene) and DFNA6/14/38 (WFS1 gene) are the most common forms of autosomal dominant non-syndromic HL affecting low frequencies, while DFNA16 (unknown gene) is characterized by fluctuating HL. A long audiological follow-up is of paramount importance to identify hearing threshold deteriorations early and ensure prompt treatment with hearing aids or cochlear implants.


Introduction
The World Health Organization (WHO) estimates that approximately 34 million children worldwide have disabling hearing loss (HL), defined as HL greater than 35 dB in the better ear [1]. HL can be present at birth ("congenital HL") or appear sometime later in life ("acquired or delayed-onset HL") [2]. The prevalence of congenital sensorineural HL ranges from 1 to 3 per 1000 live births in term healthy newborns to 3-6 per 100 in children admitted to neonatal intensive care units (NICU) [3]. Overall, the prevalence of HL increases over time, ranging from 2.8 per 1000 in school-age children to 3.5 per 1000 in adolescents [4]. Non-hereditary HL can be caused by prenatal, perinatal, or postnatal factors. Prenatal risk factors for HL include prenatal exposure to teratogens (e.g., valproic acid, ethanol, and thalidomide), congenital infections (e.g., cytomegalovirus [CMV], toxoplasmosis, rubella, syphilis, and Zika), and malformations (e.g., Michel aplasia, enlarged vestibular aqueduct, Mondini malformation) [4,5].
Particularly, congenital CMV infection is considered the leading nongenetic cause of sensorineural HL in the developed world [6]. The characteristics of HL due to congenital CMV infection are extremely variable concerning onset (at birth/late onset), side (unilateral/bilateral), degree (mild/moderate/severe/profound), audiometric configuration (rising/flat/sloping), and threshold changes over time (stable, fluctuating, sudden, progressive) [6]. There are also several perinatal risk factors for HL, such as prematurity, very low birth weight, hyperbilirubinemia, asphyxia, and hypoxic-ischemic encephalopathy [3,5]. Postnatal risk factors for HL include infections (e.g., bacterial meningitis, Herpes Simplex Virus, and Epstein-Barr virus), use of ototoxic drugs (e.g., aminoglycosides, vancomycin, and furosemide), head trauma, chemotherapy, and anemia [5,7,8]. However, approximately 50-60% of HL in children is due to genetic causes, and a genetic etiology should be considered for every patient with a hearing problem, even in the presence of other environmental risk factors [9]. Hereditary HL can be syndromic (if other signs and symptoms are present) or non-syndromic (in the absence of other clinical manifestations) [7,10]. More than 70% of genetic HL is non-syndromic, with great clinical and genetic heterogeneity (more than 120 genes have been identified to date) [9,11]. Non-syndromic HL generally follows simple Mendelian inheritance and is predominantly transmitted as an autosomal recessive trait (75-80%), although autosomal dominant (20%), X-linked (2-5%), and mitochondrial mutations (1%) can also cause HL [12]. Children born to consanguineous parents have a higher incidence of autosomal recessive disorders, including HL [13]. The loci in inherited non-syndromic HL are designed as "DFN" (standing for "DeaFNess"); the letters "A", "B", and "X" indicate that the inheritance patterns are autosomal dominant (DFNA), autosomal recessive (DFNB), and X-linked (DFNX), respectively [7,10]. A Y-linked inheritance pattern has also been described for HL [14]. The most effective strategy for the diagnosis of non-syndromic genetic HL is to perform a multi-step approach based on next-generation sequencing technologies and copy number variations assays and a thorough clinical evaluation, including physical examination and audiometric tests [15]. The aim of this narrative review is to provide a comprehensive and critical overview of autosomal dominant nonsyndromic genetic HL. We screened titles, abstracts, and full texts from relevant literature to evaluate the content of the articles and extract valuable information.

Inheritance
Autosomal dominant inheritance occurs when only one dominant allele within the disease gene (located on one of the autosomal chromosomes) is sufficient to express the phenotype [16]. Therefore, a heterozygous parent with autosomal dominant non-syndromic HL (DFNA) has a 50% chance of passing it on to their children [7,16]. However, if one parent is homozygous, all offspring may inherit the disease. If both parents are heterozygous and affected by autosomal dominant non-syndromic HL, 75% of the offspring have the chance of inheriting the disease [16]. Males and females are equally likely to inherit the mutation [7,16]. Most patients diagnosed with autosomal dominant non-syndromic HL have a hearing-impaired parent [7]. However, although the family history is rarely negative, it may appear to be negative due to late-onset HL in a parent, reduced penetrance of the pathogenic variant in an asymptomatic parent, or a de novo variant [7]. In particular, de novo mutations are possible causes of genetic HL and should be considered in all cases of sporadic HL [17]. It is often difficult to distinguish between syndromic and non-syndromic HL, as symptoms can sometimes appear later. Furthermore, some genes (e.g., WFS1 and ACTG1) cause both syndromic and non-syndromic HL [11].
To date, more than 50 genes and 80 loci have been identified for autosomal dominant non-syndromic HL [11], and are summarized in Table 1.       Unlike autosomal recessive non-syndromic HL (in which the majority of cases are caused by mutations in the GJB2 gene), autosomal dominant non-syndromic HL does not have a single identifiable gene responsible for the majority of cases worldwide [7].

MYO6 Gene
Mutations in the MYO6 gene can cause either autosomal dominant non-syndromic HL (DFNA22) or autosomal recessive non-syndromic HL (DFNB37) [11]. DFNA22 is caused by a heterozygous mutation in the myosin VI gene (MYO6) on chromosome 6q14 [11]. Myosin VI is an actin-based motor protein which plays a key role in the endocytic and exocytic membrane trafficking pathways. In the inner and outer hair cells of the organ of Corti, myosin VI serves as an anchor and maintains the structure of the stereocilia [134]. Autosomal dominant HL associated with MYO6 mutations was reported in large Italian [135], Danish [136], Belgian [53,137], Dutch [138], German [139], and Austrian [140] families. However, several cases of DFNA22 were described in China [141][142][143], Japan [144,145], the Republic of Korea [146], and Brazil [147]. HL is typically post-lingual (often occurs during childhood), is slowly progressive, ranges from a mild to profound degree, and may be associated with mild cardiac hypertrophy [11,148]. Volk et al. suggested a favorable outcome of cochlear implantation in patients with DFNA22 [139].

How Knowledge of Genetic Mutations May Influence Treatment
All children diagnosed with sensorineural HL should be screened early for genetic mutations to ensure timely appropriate treatments (e.g., hearing aid or cochlear implant), personalized rehabilitation programs (e.g., in the presence of additional symptoms), prognosis (e.g., stable, progressive, or fluctuant HL), and family planning [7]. The team evaluating and treating these children should consist of an otolaryngologist with expertise in the management of pediatric otologic disorders, an audiologist experienced in the assessment of childhood HL, a clinical geneticist, a speech-language pathologist specializing in working with children affected by HL, and a pediatrician [7]. For children with severe-to-profound HL, hearing aids may be insufficient for HL rehabilitation, and cochlear implantation should be considered.
Cochlear implantation has a high probability of being effective if the mutated lesion is located in the hair cells or afferent synapses between hair cells and the auditory nerve, such as in patients with pathogenic variants in GJB2, COCH, MYO7A, ACTG1, or MYO6 genes. Conversely, cochlear implants are generally less effective if genetic mutations affect auditory nerve function [139,[300][301][302][303]. Moreover, genetic testing is useful not only for predicting performance after cochlear implantation but also for assessing residual hearing, estimating progression, and successful hearing preservation, leading to the most appropriate selection of candidates and electrodes [302].
As a matter of fact, better knowledge regarding genotype-phenotype correlation and cochlear implant outcome may provide effective auditory rehabilitation and would reduce unnecessary procedures, thereby limiting both surgical risks and healthcare costs [303].

Current Limitations and Future Trends
The genetics of non-syndromic HL are constantly evolving, and there are currently many limitations of knowledge in this field. The etiology of some patients with evident familial HL still remains unknown. Indeed, intra-familial variability in sensorineural HL is common not only from parent to child in dominant cases but also between siblings [12]. Many pathogenic variants affecting known deafness genes may go undetected using current diagnostic algorithms because they reside in non-coding (intronic and regulatory) sequences or unannotated exons [304]. Therefore, consideration should be given to implementing whole exome or whole genome sequencing with a virtual panel as the gold standard for genetic testing in HL instead of targeted gene sequencing panels [305].
Currently, many children with mild or progressive forms of HL remain undiagnosed during their critical period of speech development and neuroplasticity. Therefore, it appears to be a priority to develop a new cost-effective method of universal genetic screening that ensures early diagnosis of genetic HL in order to identify potential comorbid conditions and guide treatments [306].
In recent years, there have been major advances in the development of gene therapy vectors to treat sensorineural HL in animal models, representing a promising approach to prevent or slow down genetic HL. Interestingly, gene therapy is not limited to the addition of a healthy copy of the defective gene but may also involve gene silencing or editing through nucleic acid-based strategies, including antisense oligonucleotides, siRNA, microRNA, or nuclease-based gene editing [307]. However, many issues are still unresolved, such as the temporal window for therapeutic intervention, the need for viral vector optimization, the safety of surgery, and the type of immune response [308].

Conclusions
Patients diagnosed with autosomal dominant non-syndromic HL typically have a parent affected by HL, although de novo mutations should be considered in the case of negative family history. Overall, autosomal dominant non-syndromic HL tends to be bilateral, post-lingual in onset, high-frequency, progressive, and variable in severity. However, congenital, low-frequency, and stable forms of HL are also possible. A long and accurate audiological follow-up is of paramount importance to early identify hearing threshold deterioration and ensure prompt treatment with hearing aids or cochlear implants according to the degree of HL.
Despite the importance of the major findings, the study has many limitations, including that it does not show the mutations described in each gene and whether there are "hot spots" mutations or domains in which these mutations are localized.