A new clinical sign or just fancy apparel? Determining the significance of a skirt

As endoscopic technology advances, the mucosal pathology that can be seen in real time has expanded. High-definition endoscopes, magnification, manipulation of light wavelengths, and image processing techniques have allowed us to interrogate the surface patterns, morphology, and vascular patterns of the gastrointestinal mucosa in ever greater detail. Many of these technologies have been available for some time, yet new endoscopic findings are still regularly encountered. Many features must have been in plain sight for several years. Some of these findings were likely to have been disregarded by endoscopists, either consciously or unconsciously, as “noise” as the features did not yet fit into the contemporary framework of understanding for interpretation of mucosal pathology.

This failure to “see” is essentially not a technological hurdle but rather a research and comprehension challenge. All such new observations require detailed scrutiny and independent validation to confirm their nature and relevance. Any new endoscopic finding or sign should be carefully interpreted in light of the underlying histopathological and molecular changes, guided by the established literature and working understanding of the natural history of the lesion. Once recognized and validated, these “signs” may become the targeted features of lesions that endoscopists specifically search for.

The most visible example of this phenomenon of endoscopist’s understanding influencing detection and recognition surrounds sessile serrated adenomas/polyps (SSA/Ps). These lesions were once thought to be rare or inconsequential, and were often missed or disregarded. Now, a large body of research has highlighted the endoscopic characteristics of SSA/Ps and their strong links to colorectal cancer (CRC) [1]. The majority of appropriately trained and informed endoscopists now actively examine the mucosa for these lesions, washing carefully and looking for their homogeneous, cloud-like features, indistinct borders, and minimally elevated morphology [2]. Chromoendoscopy and magnification endoscopy has also identified that SSA/Ps are associated with a Kudo II pit pattern, often with an open conformation (Kudo II-O) [3]. SSA/Ps can also develop focal dysplastic change, which may correlate with molecular alterations predisposing to rapid progression to CRC [4] [5] [6]. It has only recently been recognised that these changes are often endoscopically visible and may manifest as a nodule resembling a conventional adenoma within an SSA/P or a transition area where the surface pattern changes [7] [8]. Previously, endoscopists and histopathologists may have regarded this finding as a co-located hyperplastic polyp and conventional adenoma, or a “mixed polyp.” It is now clear that this is incorrect and seriously underestimates the clinical relevance of the lesion and the associated risk.

In this issue of Endoscopy, Osera et al. present a novel finding associated with laterally spreading tumors (LSTs) that may have been present but not “seen” by endoscopists [9]. This feature was first noted in an analysis of LSTs by the same group in 2014 [10]. They describe a “skirt” noted at the margins of colonic LSTs. The skirt was defined by the three study endoscopists as spreading across the margin of the LST, slightly elevated or flat, and containing wide pits. Over an 8-year period at this single center, 35 lesions with skirts were retrospectively identified from 1023 LSTs. On univariate analysis, lesions with a skirt had a higher rate of high-grade dysplasia (HGD) (60.0 %) or submucosal invasive cancer (SMIC) (20.0 %) compared with lesions without a skirt (HGD 30.6 %, SMIC 14.9 %). Lesions with a skirt were also larger (47.0 ± 19.1 mm vs. 26.9 ± 17.1 mm), and the majority (74.3 %) were located in the rectum. However, on multivariate analysis, skirts were only associated with increasing size and rectal location rather than with HGD or SMIC. This probably indicates that large rectal lesions are high risk, regardless of whether they have an associated skirt.

Only five of the lesions had detailed histopathological and molecular examination of the described skirt. These five specimens were compared with five examples of hyperplastic polyp, five tubular adenomas with low-grade dysplasia (LGD), and five examples of normal colon mucosa. Approximately 100 crypts were examined in total from each group, although the selection process for these crypts was not clear. As the crypts in the skirt group appeared different from those in the other groups, a novel and unvalidated method was used to describe the diameter of the crypt openings (DCO) and width of the individual glands (WIG). These measurements appeared to be larger in the skirt group, although the clinical relevance of this is unclear. Analysis of microvessel sizes did not determine differences when compared with hyperplastic polyps and normal mucosa, although lesions with LGD had slightly larger diameters. Cytological differences were not reported. Tissue from the skirt lesions was examined for a panel of mutations commonly associated with CRC precursors; however, only a single KRAS mutation was noted in a dysplastic portion of one of the lesions.

Limitations to the study include its retrospective and single-center nature, and the small number of lesions that were subjected to histopathological and molecular scrutiny. There was no independent or blinded examination of the histopathology by pathologists and no comparison with serrated lesions (SSA/P or traditional serrated adenoma). There was also no description of the number or experience of the pathologists involved and what methods they used to distinguish these lesions from lesions described in the World Health Organisation guidelines. The statistical comparisons of crypt openings were made using unvalidated tools, and it was unclear whether the measurements were blinded, rendering them open to bias. In the absence of validated tools, the authors may have considered a blinded assessment by a panel of independent expert pathologists in order to determine differences. The conclusions of the study may be reaching a little too far at this early stage, as the authors describe a skirt as associated with HGD or CRC, which it is not reflected in their own multivariate analysis. They also postulate that it may represent a novel precursor lesion to CRC, which is also not supported by their data.

Overall the endoscopists in this study should be applauded for their attention in noting the skirt surrounding rectal LSTs. It is a provocative finding and it requires further robustly designed study. Despite this, all that can be reported from the study at this stage is that it is a feature of larger, rectally located lesions. It may represent a phenotypic change associated with increasing size in this location. Similar phenotypic changes such as “chicken-skin” mucosa have been noted around colonic lesions, a feature that is thought to represent mucosal edema and macrophage infiltration, possibly associated with advanced histological features [11]. Prospective studies are required to report on the presence of the skirt feature in other centers, including careful comparison with existing lesions such as SSA/Ps and traditional serrated adenomas to ensure that this truly is a new finding. Its low prevalence means that collaborative efforts may be required, and it should be subjected to intense histopathological and molecular scrutiny before being accepted as a novel pathway to CRC.

• References

• 1 Rex DK, Ahnen DJ, Baron JA et al. Serrated lesions of the colorectum: review and recommendations from an expert panel. Am J Gastroenterol 2012; 107: 1315-1329
• 2 Hazewinkel Y, López-Cerón M, East JE et al. Endoscopic features of sessile serrated adenomas: validation by international experts using high-resolution white-light endoscopy and narrow-band imaging. Gastrointest Endosc 2013; 77: 916-924
• 3 Ishigooka S, Nomoto M, Obinata N et al. Evaluation of magnifying colonoscopy in the diagnosis of serrated polyps. World J Gastroenterol 2012; 18: 4308-4316
• 4 Snover DC. Sessile serrated adenoma/polyp of the large intestine: a potentially aggressive lesion in need of a new screening strategy. Dis Colon Rectum 2011; 54: 1205-1206
• 5 Burgess NG, Tutticci NJ, Pellise M et al. Sessile serrated adenomas/polyps with cytologic dysplasia: a triple threat for interval cancer. Gastrointest Endosc 2014; 80: 307-310
• 6 Erichsen R, Baron JA, Hamilton-Dutoit SJ et al. Increased risk of colorectal cancer development among patients with serrated polyps. Gastroenterology 2015; pii 0016-5085(15)01732-1 DOI: 10.1053/j.gastro.2015.11.046.
• 7 Burgess NG, Pellise M, Nanda KS et al. Clinical and endoscopic predictors of cytological dysplasia or cancer in a prospective multicentre study of large sessile serrated adenomas/polyps. Gut 2016; 65: 437-446
• 8 Nanda KS, Tutticci N, Burgess N et al. Caught in the act: endoscopic characterization of sessile serrated adenomas with dysplasia. Gastrointest Endosc 2014; 79: 864-870
• 9 Osera S, Fujii S, Ikematsu H et al. Clinicopathological, endoscopic, and molecular characteristics of the “skirt” – a new entity of lesions at the margin of laterally spreading tumors. Endoscopy 2016; 48: 448-455
• 10 Miyamoto H, Ikematsu H, Fujii S et al. Clinicopathological differences of laterally spreading tumors arising in the colon and rectum. Int J Colorectal Dis 2014; 29: 1069-1075
• 11 Chung EJ, Lee JY, Choe J et al. Colonic chicken skin mucosa is an independent endoscopic predictor of advanced colorectal adenoma. Intest Res 2015; 13: 318