| Domain | Best-supported finding | Evidence type/sample | Certainty/limitations |
|---|---|---|---|
| Defining human phenotype | One consanguineous Pakistani family had eight affected relatives with bilateral severe-to-profound or profound hearing impairment and no consistent vestibular, facial, neurologic, or cardiac abnormality, supporting a predominantly nonsyndromic phenotype. (pqac-00000010, pqac-00000016) | Human pedigree; eight affected and five unaffected examined; pure-tone audiometry at 250–8000 Hz in affected individuals aged 10–25 years. | Strong within-family evidence, but only one DFNB123 family has been published; congenital onset, progression, sex effects, and population-level penetrance remain unestablished. |
| Gene and inheritance | Hearing impairment segregated as an autosomal-recessive trait with homozygous **STX4** NM_004604.5:c.232+6T>C; two-point LOD score was **5.9 at θ=0**. (pqac-00000009, pqac-00000010) | Exome sequencing, homozygosity mapping, linkage analysis, and Sanger segregation in a large consanguineous pedigree. | Compelling locus-level segregation, although independent DFNB123 families and additional alleles are needed for replication. |
| Variant rarity | c.232+6T>C had gnomAD v2 MAF **7.98×10⁻⁶**, observed heterozygously in two non-Finnish Europeans, and was absent from gnomAD v3, TOPMed Bravo, GME, and the examined All of Us release. (pqac-00000009, pqac-00000010) | Population-database analysis; ClinVar submission SCV002499562. (pqac-00000013) | Supports PM2 rarity; carrier frequency and disease prevalence cannot be estimated reliably from one exceptionally rare allele. |
| RNA consequence | A minigene assay demonstrated that c.232+6T>C disrupts splicing and causes **exon 3 skipping**, producing a frameshift predicted to trigger nonsense-mediated decay. (pqac-00000009, pqac-00000015) | In-vitro functional splicing assay plus computational prediction. | Direct evidence for aberrant splicing; nonsense-mediated decay and loss of STX4 protein were predicted rather than demonstrated in patient cochlear tissue. |
| Cochlear expression/localization | Murine Stx4a is broadly expressed in the developing and adult inner ear; STX4A localized to the stereocilia, plasma membrane, and cell body/cytoplasm of inner and outer hair cells, with expression also reported in spiral and vestibular ganglia. (pqac-00000009, pqac-00000018) | Mouse transcriptomic datasets and P12 cochlear immunofluorescence. | Strong anatomical plausibility, but mouse localization does not by itself identify the precise human pathogenic process. |
| Zebrafish functional evidence | Morpholino knockdown of **stx4** caused abnormal acoustic startle/ABER responses, absent or markedly impaired FM1-43 uptake in neuromast hair cells, developmental abnormalities, and impaired mechanotransduction. (pqac-00000011, pqac-00000019) | Zebrafish larvae; ATG- and splice-blocking morpholinos; behavioral, ABER, and FM1-43 assays at 5 dpf. | Supports conserved auditory function, but morpholino toxicity, multisystem developmental defects, lack of a stable auditory knockout/rescue model, and species-specific splicing limit mechanistic specificity. |
| Proposed mechanism | STX4 is a target-SNARE involved in membrane fusion and vesicle trafficking; loss is proposed to disturb apical recycling/stereocilia maintenance and/or basal synaptic trafficking, leading to defective hair-cell mechanotransduction and auditory signaling. (pqac-00000018, pqac-00000019) | Integration of protein function, cochlear localization, human splicing, and zebrafish assays. | Variant→aberrant splicing and knockdown→mechanotransduction deficit are demonstrated; the intervening vesicle-trafficking and synaptic steps remain inferred. |
| Phenotypic boundary | DFNB123 should be distinguished from broader biallelic STX4 disease: a separate homozygous p.Arg240Trp patient had congenital sensorineural hearing loss, developmental delay, hypotonia, myopathy, and severe dilated cardiomyopathy, while another compound-heterozygous patient had lethal multisystem fetal disease. (pqac-00000005, pqac-00000007) | Two unrelated human cases plus CRISPR zebrafish cardiac studies. | Establishes possible allelic pleiotropy but does not show that cardiac or neurologic disease is part of the c.232+6T>C DFNB123 phenotype; ECGs were normal in two members of the nonsyndromic family. |
| Treatment and trials | No DFNB123-specific drug, gene/RNA/cell therapy, or relevant clinical trial has been reported; the defining study recommends adding **STX4** to diagnostic hearing-loss panels. (pqac-00000011, pqac-00000026) | Literature and trial search; defining family report. | Evidence is absent rather than negative. A separate pleiotropic p.Arg240Trp patient received a cochlear implant at age six with improved communication, but this single case is not evidence of genotype-specific efficacy. (pqac-00000005) |


*Table: Concise evidence map for the human genetic association, phenotype, functional validation, mechanistic interpretation, phenotypic boundaries, and therapeutic status of STX4-related DFNB123 hearing loss.*