Submitted:
04 November 2026
Posted:
05 November 2026
You are already at the latest version
Abstract
Claudin-1, encoded by CLDN1, is the principal epidermal tight-junction protein, and four tagging variants have been linked to atopic dermatitis in population- and exposure-specific studies. Because transfer of these associations is uncertain, we tested whether they mark atopic dermatitis in Vietnamese adults and quantified the precision of the null. We genotyped rs17501010, rs893051, rs9290927, and rs9290929 by Sanger sequencing in 82 adults with atopic dermatitis, 82 with other chronic dermatoses, and 83 without chronic skin disease, and estimated odds ratios by age- and sex-adjusted Firth-penalized logistic regression. The null was evaluated by equivalence testing, replication Bayes factors, and exclusion bounds, with robustness checked by propensity weighting, onset restriction, and SKAT-O. No variant was associated under any inheritance model; against pooled controls, adjusted per-allele odds ratios were 1.65, 1.11, 0.88, and 0.63 (all p ≥ 0.18). Joint four-marker and SKAT-O tests were likewise null, and results held in the 71 genuinely childhood-onset cases. Equivalence testing excluded odds ratios ≥ 2.0 at the two common variants, and at rs9290927 a Bayes factor of 183 favored no effect over the published odds ratio of 1.8. Linkage disequilibrium differed markedly across ancestries, this panel capturing only 15% of common CLDN1 variation in Vietnamese. These four variants therefore do not mark atopic dermatitis susceptibility in Vietnamese adults: the null is tightly bounded at the two common variants, underpowered at the two rare ones, and cross-ancestry transfer of this tag-SNP panel cannot be assumed.
Keywords:
CLDN1
; claudin-1
; atopic dermatitis
; tagging single-nucleotide polymorphism
; case–control study
; Vietnam
; equivalence testing
; linkage disequilibrium
Copyright: This open access article is published under a Creative Commons CC BY 4.0 license, which permit the free download, distribution, and reuse, provided that the author and preprint are cited in any reuse.