In this study, we investigated the clinical characteristics of 65 patients who were diagnosed with OTOF-related hearing loss. One of the main aims was to identify genotype-phenotype correlations. As reported previously, most of the patients (90.8%) participating in this study show a “typical” phenotype, congenital or prelingual onset and severe-to-profound hearing loss (Fig. 1). All of the patients who revealed a homozygous c.5816G > A (p.R1939Q) and compound heterozygote with p.R1939Q and a truncating mutation, and 1 patient with two truncating mutations (c.1422T > A (p.Y474X) and c.885 + 5G > A) showed profound hearing loss. Therefore, it is plausible that p.R1939Q and a truncating mutation are related to severe-to-profound hearing loss based on these results. On the other hand, the phenotype of patients with non-truncating mutations is more complicated. Figure 5a shows a summary of genotype-phenotype correlations in this and previous studies, in which the patients’ hearing level and mutation information were available (Tekin et al. 2005; Varga et al. 2006; Rodríguez-Ballesteros et al. 2008; Santarelli et al. 2009; Chiu et al. 2010; Zadro et al. 2010; Mahdieh et al. 2012; Matsunaga et al. 2012; Yildirim-Baylan et al. 2014a; Zhang et al. 2016; Kim et al. 2018b; Wang et al. 2018). There is a clear relationship between p.R1939Q and a severe phenotype, as well as truncating mutations including c.2485C > T (p.Q829X). On the other hand, when patients have one or more non-truncating mutations, almost half of them showed mild-to-moderate hearing loss (Fig. 5c, d). Based on the results, patients with non-truncating mutations could show a milder phenotype even when they have p.R1939Q or a truncating mutation.
Among the moderate cases in our cohort, there were representative 3 cases who showed “true” auditory neuropathy-like clinical characteristics (AK6640, AL8610 and AH0083 were compound heterozygote with c.[5816G > A]; [1538A > G] (p.[R1939Q];[H513R], c.[5815C > T];[5728G > A](p.[R1939W];[E1910K]) and c.[4718T > C];[4129_4138del](p.[I1573T];[A1377Rfs*142], respectively)); the PTA of all patients showed moderate hearing loss while their ABR showed no response. p.H513R is located in the C2C domain, which also contains the p.I515T and p.G541S mutations reported as mutations related to temperature-sensitive auditory neuropathy (TS-AN) and milder hearing loss in a non-febrile state (Varga et al. 2006; Matsunaga et al. 2012). p.I1573T was reported as the cause of ANSD with mild-to-moderate hearing loss in a previous report, which is compatible to the present case (Yildirim-Baylan et al. 2014b). We consider that p.H513R and p.I1573T are related to a milder phenotype. p.R1939W was reported previously, and the patient with homozygous p.R1939W showed severe-to-profound hearing loss (Choi et al. 2009). p.E1910K is novel mutation. As p.R1939W causes the same amino acid change as p.R1939Q, p.E1910K might cause moderate hearing loss. Other than these mutations, it has been reported that p.E1700Q was related to a progressive and milder phenotype (Chiu et al. 2010), and p.R1607W and p.G1804del were related to TS-AN (Marlin et al. 2010; Wang et al. 2010). Interestingly, all of the patients with the aforementioned mutations related to a milder and non-typical phenotype showed no ABR response, just like “true” auditory neuropathy. Electrophysiological testing, such as ABR, is important, especially when estimating the hearing threshold of very young children. However, when determining the hearing level for OTOF-related hearing loss, the results of audiological testing should be carefully interpreted, considering that there is a discrepancy between PTA threshold and ABR response, especially when the PTA thresholds are mild-to-moderate.
This study revealed that hearing loss was detected by NBHS in only 43.1% of the patients. The presence of OAE is main reason for missing OTOF-related ANSD at NBHS. Similar to our result, it was reported that 75% of OTOF-related patients improperly passed NBHS in a previous study (Wu et al. 2019). The risk of using OAE for NBHS has been discussed from the point of view of its inability in detecting ANSD. Nevertheless, OAE is still used for NBHS in most countries and there are only a few countries where both OAE and AABR are used for NBHS (Wroblewska-Seniuk et al. 2017). In Japan, the screening method differs depending on the clinic and prefecture. All of the patients in this study screened by OAE passed NBHS improperly, and their hearing loss was detected by their unawareness to sound or delayed language development. For better language development, early detection of HL is important. We strongly believe that NBHS should be performed by AABR or combination of OAE and AABR for reliable detection of ANSD.
The efficacy of cochlear implantation for ANSD has been controversial due to the heterogenicity of ANSD in terms of its etiology: ANSD is a disorder which includes auditory synaptopathy and neuropathy. Generally, prior to cochlear implantation, genetic testing should be considered in order to clarify which part of the auditory pathway is impaired and in order to be able to predict the outcomes of CI (Miyagawa et al. 2016). Once a patient is diagnosed with OTOF-related ANSD, a good CI outcome is expected because the auditory nerve remains intact. Indeed, past reports suggested that patients with OTOF-related ANSD are good candidates for CI (Zheng and Liu 2020). In this study, we investigated the efficacy of CI for OTOF-related hearing loss in the biggest cohort studied to date. Most of the patients who underwent cochlear implantation showed successful outcomes: approximately 85–90% of patients showed a hearing level of 20-39dB with CI and a CAP scale level 6 or better, which means the OTOF-related hearing loss patients who underwent CI can understand conversation without lip-reading. This data will support preoperative counseling for OTOF-related ANSD patients who are considered to be CI candidates.
In this study, we also investigated the timing of OAE disappearance. Although the exact timing of OAE disappearance was not available for most of the patients, some patients showed a positive OAE even at 3–4 years of age (Fig. 3b). Given no patients showed a positive OAE passed 5 years of age, we presume that the OAE response in OTOF-related ANSD disappears by 4–5 years of age at most, a little longer than previously thought. Related to the disappearance of OAE, some reports suggest that the disappearance of OAE could occur due to OHC damage originating from hearing aid use (Rouillon et al. 2006; Vona et al. 2020). On the other hand, it is also suggested that the use of hearing aids is not a definitive factor for deterioration of OAE (Kitao et al. 2019). In our study, some patients lost OAE regardless of HA use, and some kept their OAE responses even after wearing HAs until CI surgery; therefore, it is unclear whether the disappearance of OAE is related to wearing HAs or is part of the natural course of OTOF-related hearing loss. As early intervention of HA for deaf patients is important for language development, there is no evidence to not recommend HAs for patients with OTOF-related ANSD with the aim of maintaining OAE response.
Although CIs provide excellent hearing performance in cases of OTOF-related ANSD, it does not reach normal hearing level. Therefore, there has been a strong desire for a curative therapy, such as gene therapy. Recently, two reports have shown that cochlear gene therapy mediated by adeno-associated virus (AAV) successfully improved the prognosis of hearing impairment in Otof−/− mice (Al-Moyed et al. 2018; Akil et al. 2019). AAV-based gene therapy is considered to be a promising tool for actual human gene therapy due to its low immunogenicity, long-lasting transgene expression and ability to infect both dividing and non-dividing cells (Chandran et al. 2017). At present, a few companies around the world are preparing for clinical trials using AAV (Akouos, Boston, MA, USA; Decibel Therapeutics, Boston, MA, USA; Sensorion, Montpellier, France) (Vona et al. 2020). A new era of next-generation treatments for hereditary hearing loss, such as gene therapy, is certainly approaching; however, we should not forget that accurate treatment is based on accurate diagnosis and understanding of the clinical course. With the improvements in molecular biological diagnostic method, the etiologies of hereditary hearing loss have gradually become better understood. Although genotype-phenotype correlations have been obscure, the accumulation of patient data is gradually revealing the clinical characteristics of OTOF-related hearing loss. We believe that the clinical characteristics and genotype-phenotype correlation found in this study will support appropriate intervention and future treatment for OTOF-related hearing loss patients.