| Study | Representative human variant(s) and transcript | Patient phenotype | Evidence and uncertainty |
|---|---|---|---|
| Delmaghani et al., 2016 | **CDC14A** c.1126C>T, p.Arg376\*; c.1015C>T, p.Arg339\*. Exact RefSeq version was not confirmed in the retrieved text. | c.1126C>T segregated in a consanguineous Iranian family with 11 individuals affected by congenital/prelingual severe-to-profound cochlear hearing loss; c.1015C>T was homozygous in one Mauritanian patient with severe/profound congenital deafness. No major extra-auditory features were reported. | Linkage defined a 2.8-Mb interval at 1p21.2–p21.1. Segregation, rarity, predicted truncation/NMD, mouse kinocilium localization, and zebrafish morpholino data supported causality. The proposed short-kinocilium mechanism was later challenged by germline mouse and zebrafish mutants with normal-length kinocilia. (pqac-00000054, pqac-00000055, pqac-00000056) |
| Imtiaz et al., 2018 | c.376delT, p.Tyr126Ilefs\*64; c.417C>G, p.Tyr139\*; c.934C>G, p.Arg312Gly; c.959A>C, p.Gln320Pro. Exact RefSeq version was not confirmed in the retrieved text. | Homozygous variants segregated with progressive moderate-to-profound or severe-to-profound hearing loss in Pakistani and Iranian families. Deaf males in several families were infertile, whereas deaf females remained fertile; the p.Gln320Pro family had profound deafness. | Human segregation plus mouse null and phosphatase-dead p.Cys278Ser models supported a recessive phosphatase-loss mechanism. Mice developed normally patterned hair bundles followed by hair-cell degeneration. Individual variant effects and residual activity were not uniformly assayed. (pqac-00000011, pqac-00000018, pqac-00000044, pqac-00000045) |
| Mohseni et al., 2020 | **NM_033313.2:** c.1033C>T, p.Arg345\*; c.1126C>T, p.Arg376\*. **NM_033312.2:** c.1351_1352del, p.Ala451Thrfs\*43. | Six Iranian families had isolated bilateral moderate-to-profound hearing loss, usually beginning at ages 1–3 years; p.Arg345\* produced postlingual progressive loss and p.Ala451Thrfs\*43 prelingual severe-to-profound loss. Evaluated males had preserved fertility, although two p.Arg376\* homozygotes had borderline sperm morphology. | Segregation, rarity, founder-haplotype evidence for p.Arg376\*, and semen analyses supported DFNB32 rather than HIIMS. Transcript-specific preservation of NM_033313.2 was proposed to maintain fertility, but direct protein-function assays were not reported for every allele. (pqac-00000008, pqac-00000009, pqac-00000014, pqac-00000032) |
| Doll et al., 2020 | **NM_033312.2:** c.1421+2T>C, producing c.1414_1421del and p.Val472Leufs\*20; c.1041dup, p.Ser348Glnfs\*2. | Two consanguineous families—one Iranian and one Pakistani—each included two individuals with congenital bilateral sensorineural hearing loss ranging from severe-to-profound to profound; progression was not reported. | A minigene assay demonstrated cryptic splice-site activation for c.1421+2T>C. Blood RT-qPCR showed approximately 99% lower CDC14A expression in a c.1041dup homozygote, consistent with NMD. Reproductive phenotyping was insufficient to define DFNB32 versus HIIMS for every affected male. (pqac-00000012, pqac-00000017, pqac-00000042) |
| Zehri et al., 2024 | **NM_003672.4:** c.1000C>T, p.Gln334\*; c.684C>A, p.Asn228Lys. | p.Gln334\* segregated in family A with congenital bilateral profound sensorineural hearing loss and normal semen parameters (85 and 82 million sperm/mL). p.Asn228Lys segregated in family B with the same hearing phenotype plus severe oligozoospermia (below 0.1 million sperm/mL) and normal hormone profiles. | Both variants were absent from the queried population resources and matched controls and segregated in consanguineous families, but **both were classified as ACMG VUS** (PP1, PM2, PP3). No variant-specific functional assay was performed; p.Gln334\* should not automatically be treated as proven loss-of-function, and p.Asn228Lys/HIIMS causality remains provisional. (pqac-00000019, pqac-00000021, pqac-00000022, pqac-00000046, pqac-00000047) |


*Table: Disease-specific human CDC14A variants reported in DFNB32 and HIIMS, with transcript information, associated phenotypes, and the strength or limitations of supporting evidence. The table preserves the 2024 authors’ VUS classifications and avoids treating untested variants as definitively pathogenic.*