Context-Dependent Alterations of E-Cadherin, β-Catenin, and Vimentin in Endometrial Ciliated Epithelial Change: An Exploratory Immunohistochemical Study.

Background/

Objectives

Ciliated epithelial change in endometrial lesions is a recognized morphologic finding, but its immunophenotypic correlates and biological significance remain insufficiently defined. We investigated whether endometrial lesions with ciliated epithelial change show reproducible immunohistochemical alterations across benign, premalignant, and malignant diagnostic categories.

Methods

We performed a retrospective immunohistochemical study of 315 formalin-fixed paraffin-embedded eutopic uterine endometrial specimens (no endometriotic/ectopic lesions included) collected between 2019 and 2024 and distributed equally across seven diagnostic categories (n = 45 each): normal endometrium, endometrial polyp, hyperplasia with cystic/disordered glands, hyperplasia with crowded glands, atypical hyperplasia/EIN, endometrioid carcinoma, and serous carcinoma. Marker expression was quantified by digital image analysis and compared between lesions with and without ciliated epithelial change, including lesions with ciliated epithelial change showing cytological atypia.

Results

Ciliated epithelial change (CEC) was identified in 86/315 cases (27.3%), including 41 cases (13.0%) with atypical CEC. In benign categories, lesions with CEC showed lower E-cadherin expression and higher β-catenin expression, including more frequent nuclear β-catenin localization. In carcinomas, these patterns were not recapitulated and instead showed an opposite or attenuated profile, supporting a context-dependent rather than linear model. Vimentin was consistently reduced in lesions with CEC across diagnostic categories. p53 and CD44 showed heterogeneous findings and were less informative than the adhesion- and phenotype-related markers.

Conclusions

Endometrial lesions with CEC show reproducible, context-dependent immunohistochemical alterations, most consistently involving E-cadherin, β-catenin, and vimentin. In particular, nuclear β-catenin reactivity in this setting should not be interpreted as evidence of canonical Wnt-pathway activation in the absence of CTNNB1 sequencing or validated downstream readouts, and the carcinoma findings cannot be assigned to a specific TCGA/ProMisE molecular subgroup using immunohistochemistry alone. The observations should therefore be regarded as exploratory and warrant validation in studies incorporating molecular classification, direct ciliogenesis markers (FOXJ1, acetylated α-tubulin, basal body markers), and outcome data.