Endometrial Cells in Liquid-Based Cervical Cytology: A Diagnostic Pitfall Solved by Preparing Cytohistology from the Residual Thin Layer Sample

Abstract

Objective: It was our aim to assess the usefulness of cytohistology in cervical thin layer brush samples with problems in the differential diagnosis of endometrial cells. Study Design: This study reveals the cytological, cytohistological and immunohistochemistry findings of 8 cases suspicious of adenocarcinoma in situ (AIS)/adenocarcinoma (AC) in cervical liquid-based cytology (LBC) preparations that turned out to be normal endometrial cells. Results: All 8 cervical LBCs featured endometrial and atypical endocervical-like columnar cells with frequent ragged ‘feathered’ edge appearance and rosette formations. Overlapping atypical glandular cell groups were present on 2 ThinPrep slides as well. In cytohistology of 7 cases, the recognition of endometrial stroma with endometrial glands easily allowed the diagnosis of normal endometrium. In 1 case with very small loose tissue fragments without glands, the diagnosis could be established by positivity for CD10 marker (endometrial stroma) and without proliferative activity in the Ki-67 immunostaining. Conclusion: In cervical LBC preparations, nuclear hyperchromasia, pleomorphism and nucleoli in normal endometrial cells are more obvious than in conventional smears, and their arrangement is sometimes suggestive of AIS or AC. In the 8 cases presented, we could avoid a false-positive diagnosis of AIS or AC through cytohistology/immunohistochemistry, and in consequence, unnecessary colposcopical/histological examination.

Distribution and incidence of atypical glandular lesions in cervical cytology focusing on the association with high‑risk human papillomavirus subtypes

“…In addition, endometrial reactive changes were included, as well as premalignant and malignant changes of the endometrium. This was not only to monitor the endometrial changes in the cases assessed, but also as a control group for endocervical gland changes, to avoid misinterpretation and overinterpretation, which have been noticed in previous works as reported by Schnatz et al ( 11 ) and Risse et al ( 12 ) that have described the dysplastic changes in the endocervical epithelium and atypical glandular lesions. The present study also included the vaccination status of individuals against HPV, as until now, to the best of our knowledge, the importance of vaccination has not been well understood.…”

p16^INK4a and ProEx C Immunostains Facilitate Differential Diagnosis of Hyperchromatic Crowded Groups in Liquid-Based Papanicolaou Tests with Menstrual Contamination

“…The study group also did not include any cases of endocervical adenocarcinoma, another significant cervical lesion that may cause a diagnostic dilemma. However, it was recently demonstrated that cervical adenocarcinoma was successfully differentiated from endometrial cells by using cell block and an immunostain panel including p16 INK4a , Ki-67, and CD10 [3]. Furthermore, both p16 INK4a and ProEx C appear insensitive for detecting LSIL lesions, although the differential diagnosis is often straightforward based on cytologic features.…”

“…HCGs, are present. Several dysplasia-associated biomarkers have been used through immunohistochemistry in an attempt to distinguish precancerous or malignant HCGs from benign cell clusters [2,3]. Those markers include ProEx C, p16 INK4a , Ki-67, MCM2 (minichromosome maintenance protein 2), and TOP2A (DNA topoisomerase IIα) [reviewed in ref.…”

“…Previous studies from our group [14,15] demonstrated the successful applications of p16 INK4a and ProEx C immunostains in identifying cervical dysplastic/neoplastic lesions in cell block sections prepared from residual liquid-based cervicovaginal material. Recently, Risse et al [3] reported using a cell block and an immunostain panel (p16 INK4a , Ki-67, and CD10) to distinguish benign endometrial cells from cervical adenocarcinoma.…”