Endometrioid intraepithelial neoplasia (EIN) is a preinvasive precursor of endometrial carcinoma, yet its recognition in secretory-phase endometrium remains diagnostically challenging. Current criteria for EIN diagnosis do not account for female reproductive physiology, including the effects of endogenous progestins on the endometrium or EIN. Because women spend a substantial proportion of reproductive life in the secretory phase, failure to recognize EIN in this context represents an important gap in early detection. We systematically analyzed 40 cases of EIN arising in unequivocal physiological secretory endometrium, excluding patients with exogenous hormone exposure or other confounding factors. All cases exhibited architectural and cytologic distinctiveness relative to the background endometrium. Paradoxically, most lesions demonstrated diminished or absent secretory differentiation, challenging prevailing assumptions. Morules were identified in 38% of cases and represented a useful diagnostic clue. Additional features, including eosinophilic cytoplasm, epithelial stratification, mitotic activity, and apoptotic bodies, were variably present and functioned as supportive but nonessential findings. Immunohistochemistry demonstrated aberrancy for at least 1 of 3 established biomarkers (PAX2, PTEN, or β-catenin) in 95% of cases, underscoring the diagnostic value of this panel in the secretory phase, whereas Ki-67 proved unreliable. Collectively, these findings delineate morphologic and immunophenotypic features of EIN in secretory phase endometrium. Because most lesions have diminished secretory differentiation, the term “secretory EIN” is diagnostically problematic. We recommend adoption of the term “EIN in the secretory phase” and emphasize that improved recognition of this entity has direct implications for diagnostic accuracy and cancer prevention.
Means et al. (Mon,) studied this question.