Ovarian endometrioma subtypes defined by the ratio of ciliated cells: implications for endometriosis origin and carcinoma histologic subtype

Read the full article See related articles

Discuss this preprint

Start a discussion What are Sciety discussions?

Listed in

This article is not in any list yet, why not save it to one of your lists.
Log in to save this article

Abstract

Background: The implantation (retrograde menstruation) theory hypothesizes that endometriosis arises when refluxed endometrial cells implant at ectopic sites; however, a fallopian tube epithelial origin has also been proposed. Because endometrium and fallopian tube epithelia differ markedly in their balance of ciliated and secretory cells, we asked whether ciliated-cell proportion defines ovarian endometrioma subtypes and associates with the histologic phenotype of endometriosis-associated ovarian carcinoma. Methods: We analyzed 67 specimens: paired eutopic endometrium and fallopian tube (17 pairs), ovarian endometrioma (n=30), deep (n=6) and peritoneal (n=3) endometriosis, and endometriosis-associated cancers (n=11). Ciliated and secretory differentiation were assessed by immunostaining for forkhead box J1 (FOXJ1) and ezrin (EZR), and 3-mercaptopyruvate sulfurtransferase (MPST), respectively; FOXJ1/paired-box gene 8 (PAX8) co-localization was evaluated by double immunofluorescence. Ovarian endometriomas were classified using an 18.71% FOXJ1-positive cut-off. Results: FOXJ1-positive ciliated cells were abundant in fallopian tube epithelium but scarce in eutopic endometrium, deep endometriosis, and peritoneal endometriosis. Ovarian endometriomas showed a bimodal distribution of FOXJ1-positive frequency and were classified as type 1 (<18.71%, endometrium-like, 19/30) or type 2 (≥18.71%, fallopian tube-like; 11/30). FOXJ1-positive frequency was significantly higher in fallopian tube and type 2 endometrioma than in endometrium, type 1 ovarian endometrioma, and deep and peritoneal endometriosis. In type 2 endometrioma, most FOXJ1-positive ciliated cells co-expressed PAX8, whereas FOXJ1-positive cells lacked PAX8 in fallopian tube epithelium. Clear cell carcinoma was EZR-positive and MPST-negative, whereas endometrioid carcinoma was predominantly MPST-positive with a minor FOXJ1/EZR-positive subset; a mixed carcinoma showed the same component-specific profiles. Conclusions: Ovarian endometriomas comprise two subtypes defined by ciliated-cell proportion, whereas deep and peritoneal endometriosis show low ciliated-cell content comparable to eutopic endometrium. The FOXJ1/PAX8 co-expressing ciliated cells in “fallopian tube-like” type 2 lesions indicate a transitional epithelial phenotype distinct from terminally differentiated fallopian tube epithelium and are compatible with an endometrial origin even for ciliated-cell-rich lesions. Along with the EZR/MPST profiles distinguishing clear cell and endometrioid carcinoma, these findings suggest that differentiation states of endometriotic epithelium may be linked to the histologic phenotype of endometriosis-associated ovarian cancer.

Article activity feed