Perceiving sound is the initial step and a prerequisite for hearing rehabilitation with CI. The ability to perceive sound after CI was evaluated by audiometry. Loundon et al. (2005) and Rouillon et al. (2006) found that mean pure tone thresholds of 250, 500, 1000 and 2000 Hz in the patients with OTOF-related auditory neuropathy improved from 75 dB with HA to 37 and 45 dB with the cochlear implants after 1–1.5 years of rehabilitation. Similarly, Zhang et al. (2013), Zhang Q. et al. (2016), and Chen et al. (2018) found that the mean pure tone thresholds of 500, 1000, 2000 and 4000 Hz received 25 to 37.5 dB with cochlear implants after more than 2-year of rehabilitation. Although it is still unknown whether the audiometric thresholds would continue to improve with the extension of the rehabilitation time, the available data have shown that the sound perception in patients with OTOF mutations can be significantly improved by CI.

Being able to understand speech is one of the main purposes of CI rehabilitation. Thus, speech recognition is a direct indicator of CI outcomes evaluation. Objective indicators, such as speech perception testing and speech recognition thresholds, showed that all the patients with OTOF mutations benefited from CI. In terms of speech perception, most of the patients with OTOF mutations got a ≥90% score in closed-set or open-set perception (detailed in Table 1; Loundon et al., 2005; Rouillon et al., 2006; Santarelli et al., 2015; Zhang Q. et al., 2016; Chen et al., 2018). More recently, Wu et al. (2018) reviewed 10 cases of OTOF-related auditory neuropathy and found that the speech discrimination score received 77.5 ± 37.1% at 3 years. In speech recognition thresholds, Runge et al. (2013) reported 2 cases (siblings) of OTOF-related auditory neuropathy and found that the thresholds were 44 and 65 dB in quiet, 52 and 70 dB in noise, respectively.

In addition, scales, such as meaningful auditory integration scale/infant-toddler meaningful auditory integration scale (MAIS/IT-MAIS), categories of auditory performance (CAP) and speech intelligibility rating (SIR) are widely used to evaluate the CI performances. In the case reports by Loundon et al. (2005) and Rouillon et al. (2006), the MAIS/IT-MAIS scores increased from 4/40 and 4/40 with HA to 40/40 and 31/40 with cochlear implants. In terms of CAP and SIR, patients with OTOF mutations showed rapid improvement of scores after CI (Wu et al., 2018). The CAP scores reached 6/6–7/7 during the 2–3 year follow up (Wu et al., 2011; Park et al., 2017; Kim et al., 2018). Moreover, studies by Hosoya et al. (2018) and Wu et al. (2018) showed that there was no significant difference in CAP or SIR scores among patients with OTOF, GJB2, SLC26A4 mutations or cytomegalovirus infections.

According to the literature, OTOF-related auditory neuropathy is associated with excellent CI outcomes. Patients with this type of auditory neuropathy can not only “hear” the sound, but also “understand” the speech well with the help of cochlear implants. Unlike other types of auditory neuropathy, the CI performances of the patients with biallelic OTOF mutations are predictable and comparable to those of “typical” SNHL. A detection of OTOF mutations can help in accurately localizing the site of lesion and informing therapy-related clinical decision making in patients with auditory neuropathy.

Earlier implantation is associated with better CI outcomes in patients with OTOF mutations. It has been widely accepted that early implantation is good for CI outcomes in patients with typical SNHL (Niparko et al., 2010; Black et al., 2011; Panda et al., 2019). 0 to 3.5 years of age is considered a critical period for first language acquisition, and implantation after that period tends to have poorer outcomes (Sharma et al., 2002; Kral and Sharma, 2012). For patients with auditory neuropathy, the critical period seems to be narrower than those with typical severe to profound SNHL. That is, patients with auditory neuropathy who undergo cochlear implantation before the age of 2 years may have better auditory outcomes than those after the age of 2 years (Cardon and Sharma, 2013; Liu et al., 2014). This could be explained by that the disordered pattern of neural input to the cortex, as a result of auditory nerve dys-synchrony in patients with auditory neuropathy have negative effects on central auditory maturation (Cardon and Sharma, 2013; Sharma and Cardon, 2015).

As mentioned before, the site of lesion in patients with OTOF mutations is assumed to be presynaptic (Roux et al., 2006; Pangrsic et al., 2010). Thus, the pathological mechanism of OTOF-related auditory neuropathy is thought to be more like a typical SNHL but not an auditory neuropathy. However, Park et al. (2017) found that early (<24 months) implantees experienced notably better outcomes than late (>24 months) implantees. Similarly, Kim et al. (2018) found that the early (≤18 months) implantees had better outcomes than the late (>18 months) implantees at 6 months after CI. Furthermore, listening skills improved more rapidly in early implantees than late implantees (Kim et al., 2018). All the above suggest that patients with auditory neuropathy caused by mutations of OTOF seemed to be more affected by delayed implantation than those with typical SNHL that caused by mutations of SLC26A4 and GJB2 (Park et al., 2017).

The above results still require further confirmation as the sample size of existing comparative studies were small (n ≤ 10). Besides, late implantation (6 years of age) reported by Wu et al. (2011) also showed with a “good” CI outcome, which implied that late implantation was still beneficial for some of the patients. However, from a clinicians’ point of view, most of the patients with OTOF mutations did not experience spontaneous recovery of auditory performance, it was more likely that early implantation (<24 years) could achieve optimal CI performances (Wu et al., 2018).

The integrity of auditory nerve function is a key determinant of CI performances, especially for patients with auditory neuropathy. The auditory nerve function in OTOF-related auditory neuropathy is usually presumed intact. This assertion was supported by testing the neural responses of SGNs. Neural responses eluted by CI are objective indicators for evaluating the ability of the auditory pathway to receive, transmit and process complex electrical signals. By testing electrically evoked compound action potentials (ECAPs) or electrically evoked auditory brainstem responses (EABRs), the auditory nerve has been proved to have “good response” to cochlear implant stimuli (Loundon et al., 2005; Rouillon et al., 2006; Runge et al., 2013; Zhang et al., 2013; Santarelli et al., 2015; Wu et al., 2018).

However, some other studies might challenge the above view. Runge et al. (2013) quantitatively analyzed ECAP recovery rates in the sibling pair with OTOF-related auditory neuropathy. Though with the same genotype, one sibling had a recovery exponent within the average range of SNHL, while the other one had a more than one standard deviation (SD) higher recovery exponent than the average range of SNHL. Hosoya et al. (2018) tested the EABRs in patients with congenital hearing loss and found that the Wave III, Wave V, and Wave III–Wave V latencies were significantly longer in patients with OTOF mutations than those in SNHL. These two studies implicated that OTOF mutations might affect the more central auditory pathway beyond the synapse between the IHCs and SGNs. In addition, neurological and/or central pathologies should not necessarily be ruled out even when a patient was diagnosed with OTOF-related auditory neuropathy.

As mutations of the OTOF gene are the cause of the disease, it is reasonable to speculate whether CI outcomes are associated with distinct genotypes. To date, more than 130 variants in the OTOF gene have been implicated pathogenic or likely pathogenic¹. Although patients who underwent CI showed a high frequency of p.Gln829^∗ in European (Rodriguez-Ballesteros et al., 2003; Loundon et al., 2005; Rouillon et al., 2006), p.Glu1700Gln in Chinese (Chiu et al., 2010; Wu et al., 2011; Wu et al., 2018) and p.Arg1939Glu in Korean (Park et al., 2017; Kim et al., 2018) population, more than 30 different OTOF mutations have been detected in the CI recipients (listed in Supplementary Table 1). Due to the different methods of CI performance evaluation and the variety of genotypes, it is impracticable to compare CI outcomes among patients with different genotypes directly.

Nevertheless, it is still viable to investigate whether there is a correlation between mutation types and CI outcomes. Otoferlin has six C2 domains, one Fer-like structure and one transmembrane domain (TMD). Nonsense mutations (like p.Gln829^∗) usually lead to the loss of C2 domain(s) and the TMD, and result in a complete loss of otoferlin function (Migliosi et al., 2002), while missense mutations (like p.Glu1700Gln and p.Arg1939Glu) might affect only one C2 domain, the TMD, or neither of them (Varga et al., 2003; Chiu et al., 2010; Zhang Q. J. et al., 2016). Based on the available data, all patients with homozygous p.Gln829^∗, p.Glu1700Gln or p.Arg1939Glu revealed excellent outcomes in sound perception and/or speech recognition (Rodriguez-Ballesteros et al., 2003; Chiu et al., 2010; Wu et al., 2011; Kim et al., 2018). Therefore, there is insufficient evidence that CI outcomes are correlated with distinct OTOF genotypes.

As most of the patients with OTOF-related auditory neuropathy presented a phenotype of stable, severe to profound non-syndromic hearing loss (Zhang Q. J. et al., 2016), it seems that CI is the optimal and the only defective treatment option. However, CI may not be suitable for all patients with OTOF mutations. Firstly, the confirmation of OTOF-related auditory neuropathy could be challenging. There is no well-accepted hotspot mutation in OTOF except p.Gln829^∗ in Spanish (Migliosi et al., 2002), and the number of novel OTOF mutations is growing, but most of the mutations lack functional studies. In a patient with OTOF mutations, it is difficult to confirm that the hearing loss is caused by otoferlin (OTOF) deficiency and rule out other causes. Secondly, some of the patients with OTOF mutations manifested as temperature sensitive auditory neuropathy (TS-AN), i.e. the hearing thresholds fluctuate with a variation of core body temperature and may improve with age (Zhang Q. et al., 2016). In these cases, HA could be an effective treatment. Thus, HA trials are still recommended in the patients with mild to moderate fluctuating hearing loss, but once a patient was identified with severe to profound hearing loss or fail to develop age-appropriate language skills, CI would be considered. Due to a possible narrower critical period for CI, earlier (<24 months) implantation is recommended (Cardon and Sharma, 2013). Before the operation, clinical manifestations, molecular test results, and the auditory nerve functions should be comprehensively assessed to exclude TS-AN and auditory neuropathy caused by other reasons.