Volume 60, October 2022, 152025Abstract

Stratified mucin-producing intraepithelial lesion (SMILE) is a histologic subtype of HPV-associated endocervical adenocarcinoma in situ. We have observed benign endocervical changes resembling SMILE. We aim to characterize this pattern and explore its potential association with dysplasia. We retrospectively retrieved all 296 consecutive cases accessioned as endocervical biopsies. Some included multiple specimens, totaling 483 biopsies and 219 endocervical curettages (ECC), n = 702. We included cases showing endocervical epithelial stratification often with pencillate (triangular-shaped) nuclei. We rejected cases in which layering represented tangential sectioning, metaplasia, microglandular hyperplasia, gastric type epithelial changes, and dysplasia. We found benign stratified intraepithelial mucinous proliferation in 51 patients, either with a multilayered (n = 27) or a two-layered appearance (n = 24). Overall, multilayered proliferation occurred in 6 % (29/483) of biopsies and in 0.9 % of ECCs (2/219). Two-layering was identified in 4 % of all biopsies (20/482) and was not seen in ECCs. Histologic findings included stratification, intracytoplasmic mucin, paler cytoplasm, low nuclear-to-cytoplasmic ratio, often pencillate nuclei, rare mitoses, and no apoptotic bodies. P16 immunohistochemistry (n = 12) was negative, suggesting absence of underlying high-risk HPV infection. HSIL was concomitant in 29.6 % (8/27) of patients with multilayered proliferation. Concurrent SMILE was not observed. We also reviewed 13 SMILE cases. Concurrent multilayered benign proliferation was identified in 54 % (7/13) of cases. We describe benign stratified intraepithelial mucinous proliferation of the cervix, which morphologically may overlap with SMILE. Its presence in most SMILE cases suggests a potential relationship. The multilayered form represents a diagnostic pitfall when mitotically active. Because of the often-coexistent HSIL, we propose that its presence should prompt scrutiny to rule out any associated dysplasia.

Introduction

Stratified mucin-producing intraepithelial lesion (SMILE) is a histologic subtype of HPV-associated cervical adenocarcinoma in situ (AIS). It was described in 2000 [1]. SMILE displays stratification and intracytoplasmic mucin, whereas typical AIS is mucin-depleted and pseudostratified. It can resemble high-grade squamous intraepithelial lesion (HSIL) in cases where diffuse intracytoplasmic mucin is not appreciated at low-power examination. HSIL can also produce mucin, however this finding should be focal [2], [3]. Ancillary studies are not required for the diagnosis. A block-type p16 staining however suggests the underlying causal high-risk HPV infection. High-risk HPV in situ hybridization can also be performed. Ki-67 proliferation index increases in SMILE compared with benign endocervical epithelium. Mucicarmine highlights intracytoplasmic mucin throughout the lesion's thickness. Markers of squamous differentiation p63 and p40 show decreased expression in comparison with HSIL [1], [2], [3]. SMILE is associated with other dysplastic processes, namely HSIL, AIS and invasive carcinoma [4]. Invasive stratified mucin-producing carcinoma can arise in association with SMILE [3], [5]. For those reasons, management options for SMILE include excision and resection, depending on the patient's desire for future childbearing [2].

A gynecologic pathologist (MRQ) at our academic hospital center has observed endocervical benign epithelial changes that share some similarities with SMILE. They consist in epithelial stratification, often with pencillate nuclei and pale mucinous cytoplasm. Features fall short of criteria for dysplasia, with the paucity of mitoses, absence of apoptotic figures and absence of nuclear atypia. In the rare instances where p16 immunohistochemistry is performed for neighboring dysplasia, it is interpreted as negative.

We aimed to characterize benign SMILE-like changes, and to determine its prevalence in a SMILE cohort. We first retrospectively reviewed 702 endocervical samples to identify and study cases with SMILE-like changes. We then retrospectively retrieved 13 SMILE cases to determine the prevalence of SMILE-like benign changes in these specimens.

Section snippetsDescription of the benign pattern mimicking SMILE in routine cervical biopsy samples

We retrospectively retrieved all 296 consecutive cases accessioned as cervical biopsies irrespective of the indication of biopsy at our academic hospital during a two-and-a-half-month period in 2021. Some included multiple specimens, totaling 483 biopsies and 219 endocervical curettages, n = 702. Only material already available from the original diagnostic work-up was reviewed. No additional levels or ancillary studies were performed for the sole purpose of this study. Our cervical biopsies

Pathologic characteristics

We found multilayered proliferation in 6 % of biopsies (29/483) and in 0.9 % of ECC (2/219). Two-layering was identified in 4 % of all biopsies (20/482) and was not seen in ECCs. Overall, 51 of 296 patients (17 %) had at least one sample showing proliferative changes. It was present in the multilayered form in 27 patients (27/296 = 9 %) and in the two-layered form in 24 of them (24/296 = 8 %). Histologic findings included epithelial stratification, intracytoplasmic apical mucin visible on

Previously unreported changes

We believe it is likely that other gynecologic pathologists have noticed the presence of intraepithelial stratification of the endocervix. However, we have not found any reports or analyses of such changes in the English language literature. Although it does not seem to be a dysplastic or neoplastic process on its own, its importance lies in the high prevalence of associated lesions.

Conclusions/concluding remarks

We report a benign pattern of intraepithelial proliferation of endocervical cells. It morphologically mimics SMILE. However, it is p16 negative. Frequent co-occurrence of dysplasia (HSIL and SMILE) warrants caution when encountering this benign proliferation. This finding could potentially be used as an indicator to scrutinize the tissue more carefully, and even consider deeper levels into the block. Examining deeper levels also would eliminate potential tangential multilayering artifacts.

Funding disclosure

This research did not receive any specific grant from funding agencies in the public, commercial or not-for-profit sectors.

Declaration of competing interest

None.

References (5)

There are more references available in the full text version of this article.

View full text

© 2022 Elsevier Inc. All rights reserved.