SAT-254 - The Updated Androgen Receptor Variant Database: Domain-Level Genotype-Phenotype Architecture in Androgen Insensitivity Syndrome

Background: Pathogenic variants in the androgen receptor (AR) gene cause the clinical
spectrum of androgen insensitivity syndrome (AIS), ranging from complete (CAIS) to partial
(PAIS) and mild (MAIS) forms. Despite extensive case reporting, a quantitative, integrated
assessment of the AR mutational landscape and its relationship to clinical severity has
remained limited.

Methods: We systematically curated 632 pathogenic/likely pathogenic AR variants affecting
382 unique amino acid residues, harmonized to HGVS nomenclature (NM_000044.6), and
annotated by protein domain, mutation type, and phenotype category. Domain–phenotype
and mutation type–phenotype associations were tested using chi-square analyses. Mutational
hotspots and report recurrence were assessed at the residue level (hotspots defined as
residues harboring ≥3 distinct variants), complemented by a 10–amino-acid sliding-window
density analysis (step size = 1). Phenotypic plasticity was evaluated by identifying residues
linked to more than one AIS phenotype category.

Results: AR variants showed a significant association between protein domain and phenotype
(χ² = 38.61, df = 9, p = 1.35 × 10⁻⁵). CAIS variants predominated in the ligand-binding domain
(LBD, 56%) and N-terminal domain (NTD, 30%), whereas PAIS showed higher relative
representation of DNA-binding domain (DBD, 19%) variants; MAIS was enriched for NTD
variants (61%) with fewer LBD variants (29%). Mutation type also correlated with phenotype
(χ² = 59.32, df = 9, p = 2.1 × 10⁻⁹): CAIS included a substantial proportion of truncating variants
(nonsense 19.5%, small indels 27.6%), while missense variants predominated in PAIS (89.4%)
and multiple-phenotype cases (91.9%); MAIS was enriched for missense variants (78.6%) with
synonymous variants also observed. Hotspot mapping identified 65 hotspot residues (17% of
mutated residues), with 29 residues harboring ≥4 variants and 7 residues ≥5, clustering mainly
within the DBD and LBD. Recurrent positions included Arg616/Arg609 (DBD) and Trp752,
Met773, Arg841, Thr871 (LBD). The number of literature reports per variant differed markedly
across phenotypic categories (Kruskal–Wallis χ² = 198.2, p < 1 × 10⁻⁴²). Phenotypic co-
occurrence was observed in 47/382 residues (12.3%), predominantly within the LBD.

Conclusions: This updated AR variant resource delineates a structured, domain-dependent
genotype–phenotype landscape in AIS, identifying recurrent hotspots and a restricted set of
phenotypically plastic residues—especially in the LBD—that likely act as molecular “switches”
of androgen responsiveness. The curated dataset is being deposited in Zenodo for broad reuse.

*Unless otherwise noted, all abstracts presented at ENDO must not be released to the press or the public until the date and time of presentation. For oral presentations, the abstracts are embargoed until the session begins. The Endocrine Society reserves the right to lift the embargo on specific abstracts that are selected for promotion prior to or during ENDO.*