Deafness poses a significant threat to the quality of life of newborns in China, where approximately 2 in 1,000 children are born with hearing impairment (Collaborators, 2021). Genetic factors are responsible for 60% of congenital or early-onset deafness (Sloan-Heggen et al., 2016), and most cases involve severe to profound sensorineural hearing loss (SNHL), Unfortunately, effective treatments remain unavailable. For those with severe to profound SNHL, cochlear implantation (CI) offers their only hope. However, clinical observations indicate that 7%–14% of eligible patients experience suboptimal auditory improvement after CI, leading to a substantial waste of medical resources and depriving children of future treatment opportunities. The factors and mechanisms influencing CI outcomes remain unclear, although studies suggest a critical role for genetic background in determining CI success (Park et al., 2017). Advances in molecular genetics have enabled the identification of over 180 genes associated with syndromic and non-syndromic hearing loss (https://hereditaryhearingloss.org/). Genetic defects can lead to various pathological changes along the auditory pathway, potentially affecting CI outcomes. Investigations have sought to establish correlations between specific gene mutations and post-implantation results. It was concluded that defects of the spiral ganglion were correlated with poor CI performance while defects of the membranous labyrinth and hair cells were correlated with favorable CI performance (Eppsteiner et al., 2012). For example, mutations in GJB2, SLC26A4, mtDNA12SRNA, and A1555G are associated with favorable CI outcomes due to their intra-cochlear pathology (Dahl et al., 2003; Wu et al., 2008; Tono et al., 1998; Rouillon et al., 2006). In contrast, when the causative gene leads to lesions in the spiral ganglion or retro-cochlear auditory pathway, CI tends to be less effective (Eppsteiner et al., 2012). To address this critical clinical challenge, we aim to further evaluate the auditory-linguistic performance and neural pathway development abnormalities resulting from deafness-related genes. Therefore, we systematically extend our search for molecular genetics from patients affected by deafness. Among these, the impact of auditory intervention on patients with PCDH15 gene mutation has drawn our attention.

In 1997, Wayne et al. discovered a novel USH1F locus (OMIM: 602083) for Usher syndrome (Wayne et al., 1996). In 2001, Alagramam et al. identified three clones of the human homologous protocadherin-15 (PCDH15) gene, localized on chromosome 10q11.2-q21 (Alagramam et al., 2001). PCDH15 belongs to the cadherin superfamily of calcium-dependent cell–cell adhesion molecules and is strongly expressed in the neurosensory epithelium of the cochlea, vestibular hair cells, retinal photoreceptors and brain (Ahmed et al., 2003). The PCDH15 gene encodes a calcium-binding protein essential for connections between stereocilia in the inner ear. Mutant alleles of PCDH15 are responsible for both Usher syndrome type 1F (USH1F) and autosomal recessive deafness (DFNB23). Patients with USH1F present with profound congenital deafness, vestibular areflexia, and pre-pubertal-onset retinitis pigmentosa (RP). Variants in PCDH15 are responsible for 11%–19% of USH1 cases (Tsang et al., 2018). Several reports have documented the expression of PCDH15 along with other Usher proteins, such as cadherin-23 (CDH23) and CLRN1, at the synapses of hair cells and photoreceptors. PCDH15 protein is believed to play an important role in synaptic maturation of the auditory pathway. Zallocchi et al. observed pre- and post-synaptic expression of PCDH15 in inner hair cells (IHCs) and outer hair cells during synaptic development, as well as in mature IHCs (Zallocchi et al., 2012). Therefore, mutations in PCDH15 can greatly affect auditory processing (https://www.proteinatlas.org/, accessed on 18 February 2024). Nishio et al. found that PCDH15 gene expression was higher in the spiral ganglion than in other parts of the cochlea (Nishio et al., 2017), suggesting a poor CI outcome for PCDH15-positive patients. Histological study in the rat model revealed severe defects in cochlear hair cell stereocilia, collapse of the organ of Corti, and marked reduction of ganglion cells in adult Pcdh15 (kci) mutants (Naoi et al., 2009).

A detailed review on navigating the Usher Syndrome genetic landscape by Micol B et al., indicate that patients with PCDH15 mutations are significantly less likely to have a “Positive” outcome, as only 3/12 cases were reported to acquire good or excellent CI performance for this gene, underscoring a potential link between PCDH15 gene and unfavorable outcomes (Liu et al., 2008; Brownstein et al., 2004; Wu et al., 2015; Grillet et al., 2009; Busi and Castiglione, 2024). These findings appear consistent with the results of a laser-microdissection-based gene expression study showing that PCDH15 is highly expressed outside the cochlea (Nishio et al., 2017). Adverse CI outcomes in PCDH15-affected individuals may be due to pathology in SGNs or brainstem auditory nuclei, which are essential for CI functionality (Eppsteiner et al., 2012). Therefore, the CI performance substantially varies among patients with defective PCDH15 (see Table 1) and can be suboptimal (Liu et al., 2008; Wu et al., 2015).

To evaluate the effect of CI and enable the early detection of ocular and vestibular dysfunction in patients with PCDH15 mutations, we combined electrophysiological assessments with routine clinical data to elucidate intrinsic relationships among genetic background, auditory neural pathway development, and auditory-linguistic outcomes. We aim to facilitate the prediction of CI outcomes through genetic testing.

We performed genetic analyses in six patients with PCDH15 mutations from five families. In addition to clinical examinations, hearing and speech performance, vision, fundus, and balance were assessed after CI. This approach provides a foundation for gene therapy in treating deafness and highlights the importance of incorporating genetic prognostic factors into clinical decision-making, thereby optimizing surgical outcomes.

PCDH15 is a member of the cadherin superfamily of calcium-dependent adhesion glycoproteins, highly expressed in the neurosensory epithelium of the cochlea, vestibular hair cells, and retinal photoreceptors. PCDH15 and CDH23 are components of kinociliary links, transient lateral links, and tip links (Siemens et al., 2002; Lagziel et al., 2005; Ahmed et al., 2006; Senften et al., 2006; Kazmierczak et al., 2007; Rzadzinska et al., 2005). While Protocadherin-15 is an essential component of tip-links in hair cells. Research suggests that CDH23 interacts with PCDH15 to form specialized tip links that gate the mechanoelectrical transduction (MET) channels in hair cells (Grillet et al., 2009), in addition to the well documented role for PCDH15 proteins in stereocilia development, PCDH15 protein likely function in synaptic maturation as well and is necessary for morphogenesis and polarity of the hair bundle. Variants in the PCDH15 gene can disrupt stereociliary tips and interactions with transmembrane proteins (Tmc1,Tmhs and Tmie) which are essential for targeting PCDH15 protein to plasma membrane and for MET, leading to functional impairment of the MET machinery (Pepermans and Petit, 2015). It may also compromise the synaptic connection between hair cells and spiral ganglion neurons (SGN), thereby impairing the transmission of auditory signals (Ahmed et al., 2006; Kazmierczak et al., 2007). Mutant alleles in PCDH15 may contribute to isolated forms of profound deafness (DFNB23, phenotype ID: 609533) and Usher syndrome subtype USH1F, characterized by congenital profound sensorineural hearing loss, progressive RP, and vestibular impairment.

At least 211 mutations in PCDH15 have been curated in the HGMD. The prevalent founder variant c.733C>T (p.Arg 245Ter) in PCDH15 is unique to Ashkenazi Jews, with an occurrence rate of 50%–60% (Ben-Yosef et al., 2003). Pathogenic mutations are distributed throughout the PCDH15 gene, without any apparent hot spots. The mutation spectrum in our data suggests that the nine mutations identified are dispersed, consistent with previous findings. In six PCDH15-affected individuals from five families, we identified 10 variants in PCDH15 and CDH23 (Table 3), six of which had not been previously reported: all the CNVs, two splicing variants in PCDH15, and a mutant allele c.209C>T in CDH23. The identification of these novel mutations broadens the spectrum of genetic deafness. Nearly all the mutations described are predicted to result in premature termination of the protocadherin protein. Several deletions were also identified. The PCDH15 gene spans nearly 1 Mb, with a corresponding 7,021 bp open reading frame (ORF), and the first three exons cover 0.42 Mb. The high rate of deletions is likely a consequence of this unusual gene structure. Patient 1 had compound mutations, including a large copy-number-loss variant spanning exons 14 to 21 and a novel splice variant. The CNV deletion is estimated to be 157 kb, resulting in the loss of 426 amino acids. The c.2869-2A>C variant is expected to disrupt RNA splicing within intron 21, leading to truncation of PCDH15 in the extracellular domains (EC5-EC9) (Baralle and Baralle, 2005). The literature suggests that this broad deletion contributes to the development of Usher syndrome (Vache et al., 2020). The patient developed USH1F at the age of 2 years. Patient 5 had homozygous deletions in exon 2, which is the critical first coding region for PCDH15 protein translation. The deletion is estimated to be 119 bp in size, altering the predicted translation-initiation codons and resulting in aberrant protein translation, leading to synthesis of truncated proteins. There is evidence that the presence of the first few exons is essential for adhesion and impacts the mechanical properties of tip links (Kazmierczak et al., 2007). The child, at 18 months of age, has not yet achieved normal ambulation and tends to fall frequently. Clinically, USH1 is strongly suspected. Due to the patient’s young age and the family’s limited understanding of Usher syndrome, an ophthalmologic evaluation under anesthesia cannot be conducted at this time. However, long-term follow-up should include funduscopic examination and ERG to detect pre-symptomatic RP.

Sequencing of the entire coding region of PCDH15 revealed that patient 2 carried compound heterozygous mutations (c.2367_2369del and c.1918-1G>A) in PCDH15 and a heterozygous missense mutation (c.209C>T) in CDH23, reflecting both autosomal recessive and digenic traits. The c.1918-1G>A mutation alters a splicing site and is predicted to cause exon 16 skipping, leading to a truncated protein. However, the significance of the novel CDH23 variant is unknown. CDH23 interacts with PCDH15 in the tip links of inner hair cells, and digenic interactions have been observed in both mice and humans, suggesting possible modification among genes associated with Usher syndrome. Digenic inheritance was suggested as an explanation for the cause of USH1 (Zheng et al., 2005). However, Ahmed et al. argued that the evidence supporting this hypothesis was not robust (Ahmed et al., 2008). Currently, patient 2 is 7 years old and exhibits no signs of visual disability or balance disorders, mimicking non-syndromic hearing loss. However, he is too young for a conclusive diagnosis. Most USH1 patients typically develop RP near puberty or even in the second or third decade of life. Therefore, close long-term follow-up is imperative. His mother subsequently had two pregnancies, and genetic analysis of amniotic fluid revealed that both fetuses carried the same compound heterozygous mutations in the PCDH15 gene. Consequently, the family decided to terminate both pregnancies. To enhance the likelihood of conceiving a child with optimal auditory capabilities, we recommend using third-generation in vitro fertilization (IVF) techniques combined with preimplantation genetic testing (PGT) to ensure the birth of a healthy child without hearing impairment. This case shows that an individual’s genetic profile can provide mechanistic insights while supporting personalized treatment planning, tailored genetic counseling, and prenatal diagnosis.

In family 3, the proband carried a c.4744del (p.Leu1582SerfsTer13) frameshift mutation and a deletion in exon 19. His sister exhibited the same mutations. The deletion of exon 19 in PCDH15 suggests splice variants with fewer cadherin repeats in the predicted calcium-binding domains (partial EC6 and EC7). The c.4744del frameshift mutation leads to the premature introduction of a stop codon, resulting in truncated proteins. The c.4744del mutation is predicted to affect only the PCDH15-CD2 isoform. There are three common human PCDH15 isoforms, which differ in their cytoplasmic domains (PCDH15-CD1, PCDH15-CD2, and PCDH15-CD3); each isoform has a specific role in the sensory hair bundle. In immature auditory hair cells, the various PCDH15 isoforms (CD1, CD2, and CD3) are interchangeable for tip-link formation, whereas PCDH15-CD2 is essential in mature auditory hair cells (Pepermans and Petit, 2015). The absence of PCDH15-CD2 leads to non-syndromic profound deafness in humans (Pepermans et al., 2014). Mice lacking pcdh15-CD1 and pcdh15-CD3 are capable of forming normal hair bundles and tip links; they also maintain hearing. However, pcdh15-CD2-deficient mice are deaf, with abnormally polarized hair bundles and no kinociliary links, although vestibular function remains intact (Webb et al., 2011; Le Guedard et al., 2007). A similar mechanism is likely involved in our case; our clinical observations are consistent with the pathogenic role observed in a pcdh15-CD2 knockout mouse model. The deaf siblings (patients 3 and 4) are characterized by a recessive form of deafness, DFNB23, with preserved peripheral vestibular function. Patient 6 has compound heterozygous variants, including c.5254_5280del and c.146A>G. The c.5254_5280del non-frameshift mutation in exon 30 results in the loss of nine amino acids, which is predicted to alter the proteins within the cytoplasmic region. The c.146A>G transversion predicts an amino acid substitution of glutamate with glycine at codon 49; a previously reported mutation in this region was linked to autosomal recessive non-syndromic deafness. This mutation is near EC1, a region critical for adhesion. HOPE server predictions highlighted a critical role of Glu49 in calcium ion (Ca²⁺) coordination. The loss of the negatively charged carboxyl group in the mutant (Gly49) likely destabilizes Ca²⁺ binding, potentially compromising the structural integrity of the EC1 domain essential for PCDH15-mediated adhesion. DFNB23 is partly caused by defects in adhesion. The patient exhibited isolated deafness without vestibular or retinal impairment, suggesting a strong genetic correlation with the clinical features. Functional null alleles in PCDH15 typically cause USH1, whereas missense mutations in conserved motifs or non-frameshift mutations that alter one or more amino acids, retaining some residual protein function, are commonly associated with recessive deafness.

In our series, six affected patients underwent successful CI before the age of 3. Patient 5 underwent CI only 5 months prior to this report. The child has shown normal latency and well-defined EABR waveforms III and V, with good postoperative auditory behavior. However, speech recognition depends on the development of hearing skills and the maturation of the auditory pathways. As she continues auditory speech development, she remains in speech rehabilitation training and is currently able to discriminate between “a” and “i” from the Ling Six sound check. Therefore, speech audiometry is not yet appropriate. Based on audiological and electrophysiological assessments, as well as clinical observations, no evidence of retro-cochlear lesions was found in our five Chinese families. More cases need to be analyzed to further explore these possibilities.

The poor CI outcomes in patients with PCDH15 mutations documented in the literature may be attributed to several factors, including the absence of previous exposure to auditory stimulation and language use for those implanted at older ages, lack of engagement in structured auditory-verbal rehabilitation, or discontinuation of device use. Although genetic etiology plays a critical role in CI outcomes, it is not the sole determinant. Many other factors influence outcomes, including age at implantation, duration of deafness, inner ear anatomy, surgical technique, device characteristics, rehabilitation training, and family support. It is essential to emphasize the importance of individualized patient care.

To improve outcomes for patients with PCDH15 mutations, early diagnosis is increasingly recognized as a pivotal step toward timely treatment, which can significantly impact disease management and results. PCDH15 mutant alleles can cause either USH1F or non-syndromic hearing loss. Timely CI can significantly enhance auditory function and improve communication skills. Early diagnosis of USH1F should prompt anticipatory interventions to address the progressive loss of vision, such as recommending light avoidance and vitamin A supplementation. There are ongoing efforts to use gene replacement and editing, antisense oligonucleotides, and small-molecule drugs to halt RP progression and restore vision; these approaches hold promise for improving patients’ quality of life.