UvA-DARE is a service provided by the library of the University of Amsterdam (https://dare.uva.nl)
UvA-DARE (Digital Academic Repository)
What makes an expert Barrett’s pathologist?
Concordance and pathologist expertise within a digital review panel
van der Wel, M.J.
Publication date
2019
Document Version
Other version
License
Other
Link to publication
Citation for published version (APA):
van der Wel, M. J. (2019). What makes an expert Barrett’s pathologist? Concordance and
pathologist expertise within a digital review panel.
General rights
It is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons).
Disclaimer/Complaints regulations
If you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, stating your reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Ask the Library: https://uba.uva.nl/en/contact, or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam, The Netherlands. You will be contacted as soon as possible.
2
CHAPTER
WHAT MAKES AN EXPERT
BARRETT’S HISTOPATHOLOGIST?
M. J. van der Wel, M. Jansen, M. Vieth, S. L. Meijer
Adv Exp Med Biol. 2016;908:137-59 Published in ‘Stem Cells, Pre-neoplasia, and Early Cancer of the Upper Gastrointestinal Tract’, editors: M. Jansen, N. A. Wright, Springer International Publishing, Switzerland, 2016
INTRODUCTION
A strong increase in the incidence of oesophageal adenocarcinoma (OAC) over the last decades has prompted frequent and strict surveillance endoscopies in patients suffering from Barrett’s oesophagus (BO). BO is a known precursor lesion with the potential of developing into OAC through the sequence of inflammation to metaplasia and dysplasia. BO is a difficult entity sharing molecular characteristics with OAC, while the biological value of goblet cells as a prerequisite for the diagnosis of BO and as a precursor of OAC is up for debate. 1 2 This makes nomenclature confusing, since the
presence of goblet cells as a prerequisite for the diagnosis of BO is dependent on the continent and country of diagnosis. In this chapter, BO is equivalent to columnar lined oesophagus (CLO), which is diagnosed by the endoscopist. Intestinal metaplasia can either be present or absent (discussed in chapter: Definition, derivation & diagnosis of Barrett’s oesophagus). The risk of malignant transformation of BO appears to have been overestimated over the years 3-5 and only a small percentage of patients
with BO (0.5%) will eventually develop OAC. Some individuals undergo a rather fast malignant transformation without endoscopic abnormalities at the index endoscopy, whereas in others this process appears to take more time, the reasons of it still being unclear. Until now it has not been possible to create a reliable animal model, which complicates research efforts. Currently most guidelines and consensus meetings intend to decrease the endoscopy frequencies in patients with lower risk BO (i.e. non-dysplastic) compared to those with higher risk BO (indefinite and dysplasia). 6-8 In this
model of escalation and de-escalation, good quality pathology diagnosis regarding the presence or absence of dysplasia is crucial. Histopathological assessment of Barrett specimens has proven to be a challenge in the past, partially due to the use of different diagnostic criteria in Eastern and Western parts of the world9. In this chapter,
a diagnostic algorithm and examples of frequently encountered difficulties are used to discern dysplasia from non- dysplastic changes. Often, these diagnostic difficulties are due to the inflammatory nature of the disease, with continuous cycles of inflammation and epithelial regeneration. The implication of a high-grade dysplasia diagnosis has changed over the years, meaning that such a diagnosis on biopsy material does not automatically indicate the need for a radical surgical resection. Instead, endoscopic resections are now widely undertaken in the management of high-grade dysplasia and early (mucosal) malignancy, and are associated with an extensively reduced morbidity compared to surgical resections. A favourable development, since OAC remains a
highly aggressive and lethal cancer with a low curation rate. To improve survival, it is key to detect the disease in an early stage so it is still curable by endoscopic resections. Careful pathology assessment of endoscopic resection specimens is crucial for stratification of these patients, since potential subsequent invasive surgery depends on risk factors determined by the pathologist. In order to achieve high quality histological assessment of these risk factors, optimal handling and processing of the specimen during and after resection is indispensable. This requires dedication in both the endoscopy suite as well as in the receiving laboratory, and close collaboration between the endoscopist and pathologist in BO is essential.
Table 1: Vienna classification of gastrointestinal epithelial neoplasms Category Description
1 Negative for dysplasia / neoplasia 2 Indefinite for dysplasia / neoplasia
3 Non-invasive* low-grade neoplasia (low-grade adenoma / dysplasia) 4 Non-invasive high-grade neoplasia
4.1 High-grade adenoma/dysplasia
4.2 Non-invasive carcinoma (carcinoma in situ) 4.3 Suspicion of invasive carcinoma
5 Invasive neoplasia
5.1 Intramucosal carcinoma
5.2 Submucosal† carcinoma or beyond
*Non-invasive indicating absence of invasion, †Submucosal indicating invasion into lamina propria or
beyond
GENERAL ASSESSMENT
The classification of dysplasia in Barrett’s oesophagus (BO) has traditionally been based on the classification of dysplasia in inflammatory bowel disease (IBD)10. This
classification has evolved over the years and, in an attempt to minimize diagnostic difference between Western and Eastern pathologists worldwide, resulted in the Vienna classification, depicted in Table 1. 11 General features used to assess biopsies
from Barrett’s oesophagus have been extensively described 11-13 and include surface
maturation, architecture of the glands and cytonuclear abnormalities (Table
2). 13 14 These features are surveyed at low magnification and confirmed at higher
dysplasia are absolute, generally a combination of features is needed to arrive at a final diagnosis. An important feature when assessing dysplasia in Barrett’s oesophagus is whether or not the microscopy reflects ‘clonality’. In neoplasia, proliferation is driven by clonally accumulated mutations in the genome, and is therefore by definition a yes or no phenomenon. As a consequence in neoplastic mucosa, often a distinct circumscript region displays the specific cytonuclear or architectural abnormalities and these are sharply demarcated from the non-neoplastic mucosa. This so-called clonal step can often be appreciated in dysplastic Barrett’s mucosa at low magnification. This contrasts with reactive changes due to inflammation, where the proliferation and the accompanying microscopic alterations are a reaction to inflammatory stimuli and not governed by genetic mutations. These alterations are morphologically less uniform and circumscript and they gradually fade when further away from the noxious stimuli. The distinction between abrupt or gradual changes therefore is a useful and important feature for the distinction between dysplasia and inflammatory change.
Table 2: Diagnostic algorithm
1 Clonality Are changes observed within the biopsy abrupt or gradual? 2 Surface
maturation
Is surface maturation present within the biopsy and if not why not?
3 Architecture Is the architecture within normal/regenerative range or due to neoplastic proliferation?
4 Cytonuclear features
Can present changes be explained by metaplasia, regeneration or inflammation, or are they due to dysplasia?
5 Inflammation Are the alterations driven by inflammation or secondary to a clonal process?
Surface maturation is a phenomenon of zonation, with the deeper portion of the mucosa being darker in colour compared to the surface. This is due to the fact that in non-dysplastic Barrett’s oesophagus, proliferation takes place in the deeper parts of the glands. These glands are usually packed closer together, with higher nuclear/cytoplasmic (N/C) ratio and hyperchromatic nuclei. Towards the surface, the nuclei become smaller, less hyperchromatic and the cytoplasmic volume increas-es, resulting in a lighter tone. This phenomenon of zonation can disappear (loss of maturation) due to complete intestinal metaplasia, or to an increase in the number
of cells or nuclei at the mucosal surface. An increase cells or nuclei can be caused by either inflammatory stimuli or neoplastic proliferation or a combination of both. When surface maturation is not present when assessing a biopsy on overview, the underlying cause should be clarified. The architecture of the mucosa in a biopsy is determined by the relationship between the number and the shape of the glands and the surrounding lamina propria. Non-dysplastic glands tend to be round tubes surrounded by lamina propria. In the context of inflammation and regeneration, more variation in glandular shape and size is observed. Abnormal architecture include changes in gland’s shape, such as irregular contours or cystic dilatation, as well as changes in the number of glands compared to the surrounding stroma. In dysplasia, abnormal architecture is accompanied by cytonuclear abnormalities. Some architectural abnormalities are relatively specific for neoplasia, such as true back-to-back configuration or a cribriform growth pattern. These changes are best appreciated when compared to their neighboring gastric or intestinal metaplastic glands. A certain amount of cytonuclear atypia is inherent to the metaplastic and inflammatory nature of Barrett’s oesophagus. Size and shape of nuclei are assessed in relation to their location and compared to the surrounding glands or surface ep-ithelium. In general, larger nuclei are more readily accepted when they are located deeper in the glands rather than at the mucosal surface. At the surface, many con-ditions can give the impression of (pseudo-) stratification, including inflammatory, regenerative and dysplastic changes. Inflammation can cause severe architectural and worrisome cytonuclear abnormalities, mimicking dysplasia and the differential diagnosis encompasses both low and high-grade neoplastic lesions.
In summary, architectural and cytonuclear alterations in Barrett’s oesophagus can be caused by metaplastic properties of the deeper glands, reactive changes associated with inflammation or by dysplasia. The five features described above are integrated to attain a definitive diagnosis. The distinction between no dysplasia and dysplasia is the most important one and it is made first. In the paragraphs below, we first describe the discernment between dysplasia and regenerative changes, before discussing low- and high-grade dysplasia in biopsies. The category of ‘indefinite for dysplasia,’ is described separately.
Figure 1: Non-linear diff erential diagnostic possibilities in Barrett’s oesophagus
NDBO: non-dysplastic Barrett’s oesophagus; IND: indefi nite for dysplasia; LGD: low-grade dysplasia; HGD: high-grade dysplasia; OAC: oesophageal adenocarcinoma
Negative for dysplasia
The diagnosis ‘negative for dysplasia’ ranges from ‘normal’ metaplastic Barrett’s to epithelium with extreme regenerative features due to infl ammation. Because of this spectrum of epithelial changes, the diff erential diagnosis ranges from normal to dysplastic and sometimes even malignant transformation (Figure 1). With this in mind, it is important to consider the observed epithelial changes in their proper context and be aware of possible problematic areas.
In general, surface maturation and architecture remain intact in infl amed, non-dysplastic Barrett’s. Surface epithelium shows a regularly undulating or more villiform aspect, with the type of epithelium ranging from a columnar foveolar type to intestinal metaplastic type or a combination of both. The glands in the deeper part of the sample contain rather uniform and rounded glands of either cardiac, fundic or intestinal type mucosa and are regularly dispersed in ample intervening lamina propria. Junctional or cardiac type epithelium contains bland looking mucus-secreting columnar cells resembling cardiac or pyloric type glands. In fundic type epithelium both parietal and chief cells are observed, while specialized intestinal epithelium demonstrates the presence of goblet cells. Sometimes pancreatic metaplasia is observed. The most common forms of metaplasia are intestinal and junctional type epithelium, but all cell types and architectural patterns can coexist. Glands displaying intestinal metaplasia at the basis of non-dysplastic mucosa usually stand out more than gastric type
mucosa. Their nuclei are larger in size and more hyperchromatic than those located at the surface of the tissue, but nuclear contours remain smooth. This deeper area contains the proliferative compartment of the intestinal metaplastic gland as well and multiple mitoses will be present. Extension of mitotic figures to the surface is usually not observed.
Figure 2: Histological characteristics of non-dysplastic Barrett’s oesophagus
A) The heterogenous aspect at the squamo-columnar junction on overview gives the impression of stratification of nuclei on the surface and loss of maturation. B) Detail of Figure 2A, showing characteristic reactive changes depicting limited pseudo-stratification combined with basophilic changed cytoplasm and highlighted with an apical rim of clear cytoplasm. C) Multilayered epithelium in a specimen of non-dysplastic Barrett’s, displaying the transition from squamous epithelium on the left to pseudo-stratified columnar cells on the right. D) Extensive active inflammation present in a biopsy of non-dysplastic Barrett’s oesophagus resulting in diagnostic difficulties and the diagnosis of indefinite for dysplasia.
Most frequently encountered difficulties when discerning non-dysplastic from dysplastic tissue include loss of surface maturation, changes at the squamo-columnar junction, pseudo-stratification of nuclei at the surface epithelium and epithelial changes in the context of inflammation. Often a combination of these factors is observed. Due to the underlying repetitive inflammation and regeneration in Barrett’s oesophagus, metaplastic epithelial changes are most pronounced at the squamo-columnar junction (Figure 2A). Typical pseudo-stratification in reactive surface epithelium usually consists of two rows of nuclei embedded in basophilic cytoplasm and is highlighted with an apical rim of clear cytoplasm (Figure 2B). Cellular membranes of these pseudo-stratified cells are ill defined, giving it the appearance of confluent (syncytial) change and continuation throughout the surface epithelium. Strict application of diagnostic criteria is warranted at this junction before a diagnosis of dysplasia is justified. Not all crowding of nuclei is a sign of dysplasia. In Barrett’s oesophagus, several types of epithelium with multiple layers are recognized. Multilayered epithelium, probably representing an early stage of columnar metaplasia (Figure 2C) consists of multiple layers of squamous cells covered with columnar, mucinous epithelium. The number of squamous cell layers varies. They are typically multiple, but a single squamous cell layer covered with columnar, mucinous epithelium is also possible. Pseudo-stratification, usually located at the tip of villi, originates due to coalescence of migrating cells from multiple directions out of the lower proliferative compartment and is not associated with neoplastic change. In pseudo-stratification all nuclei are in contact with the basement membrane but due to orientation artifacts or crowding, nuclei may appear at different levels and look stratified. In true nuclear stratification, associated with dysplasia, only the bottom layer of nuclei is still in contact with the basement membrane. In daily practice the distinction between nuclear pseudo-stratification and real nuclear stratification can be challenging or even impossible.
Besides chronic inflammation, Barrett’s oesophagus often contains active inflammation, which may cause superficial mucin depletion, giving the impression of disturbed surface maturation. Moreover, active inflammation can induce more complex regenerative architectural configuration such as uneven distribution of glands, irregular glandular contours or cystic change and a protrusive villiform surface (Figure 2D). Intraluminal necrotic debris in dilated cystic glands is not a feature of inflammation. Cytologically, pseudo-stratification can be present, as well
as hyperchromasia and conspicuous nucleoli. An increase in mitotic activity is not unusual either. However, even though inflammation induced cytonuclear atypia can be severe, nuclear contours remain smooth and the N/C ratio is not enlarged. Overall, active inflammation can cause conspicuous architectural and cytological changes but these alterations tend to fade away when further from the noxious stimulus, while dysplastic changes are relatively uniform and circumscribed.
Figure 3: Histological characteristics of low-grade dysplasia
A) Overview of low-grade dysplasia with overall intact architecture, but loss of surface maturation due to increase of elongated nuclei in glands, which extends to the surface. B) Clonality in low-grade dysplasia, represented by a sharp demarcation in nuclear features (arrow) from normal columnar cells to low-grade dysplasia. C) Detail of a gland with dystrophic goblet cells that have lost their nuclear orientation.
Dysplasia
The biopsy is first examined at low magnification for features that stand out from the general pattern. Due to loss of goblet cell differentiation or mucin depletion at the surface together with an increase in the number of nuclei, the deeper and superficial parts of the mucosa may appear similar (Figure 3A). This loss of surface maturation can be caused by either inflammatory or neoplastic changes. The presence of active inflammation does not rule out the possibility of dysplasia, as both can occur simultaneously. If an abrupt change in architecture or nuclear features (either nuclear size or nuclear stratification) is present, this feature of clonality favors neoplastic change above inflammatory changes (Figure 3B). Dysplastic changes may be focal or extensive and occupy the whole mucosa. The severity of the grade of dysplasia in Barrett’s oesophagus is a continuous spectrum of architectural and cytonuclear abnormalities rather than sharply delineated categories. Few well-defined cut-off points can be used and in general, the architectural and cytonuclear changes in grade dysplasia are less marked than in high-grade dysplasia. The typical aspect of low-grade dysplasia is loss of surface maturation due to nuclear stratification at the surface, and the overall architecture remains intact (Figure 3A and 3B). The presence of goblet cells is diminished and some glandular crowding is observed, but some mesenchymal tissue is still present between these glands. These stratified nuclei are mildly enlarged, hyperchromatic and elongated, but still polarized, usually of uniform size and shape and without nuclear membrane irregularities. Loss of polarity of stratified nuclei is not a frequent observation in low-grade dysplasia. Dystrophic goblet cells are more frequently observed. These are goblet cells that have lost the polarity of their nucleus, displaying an upside down or otherwise mis-oriented mucin bubble instead of nicely at the top (Figure 3C). Other architectural disturbances associated with low-grade dysplasia are more extensive glandular crowding or back-to-back localization and some cystic change, as long as accompanying cytonuclear features remain within limits.
In high-grade dysplasia, architectural abnormalities are more pronounced and mitotic figures extend from the deeper compartments to the surface. Complex architectural changes are present and include glandular branching or budding, gland-in-gland formation or cribriform growth (Figure 4A). Cystic change in deeper parts of the mucosa with intraluminal necrotic debris is a distinct feature of high-grade dysplasia (Figure 4B). Surface maturation is generally absent but stratification of elongated
nuclei at the surface is not typical. Instead, a single layer of cells with marked cytonuclear atypia is more often observed (Figure 4A and 4B). Deeper in the tissue, the N/C ratio is disturbed and the nuclear size increased with nucleoli that are often but not necessarily prominent. 12 Nuclear shape is irregular and loss of polarity is frequently
observed (Figure 4C). A special type of high-grade dysplasia is represented by non-adenomatous or small cell type, 12 15 demonstrating a glandular lining with a monolayer
of small cuboidal to columnar cells with smooth nuclear contours. These cells contain mucinous cytoplasm with monotonous small rounded and hyperchromatic nuclei (Figure 4D). Some pathologists have tried to define a certain number of crypts that need to express high-grade characteristics in order to justify a diagnosis of high-grade dysplasia, 16 others accept focal high- grade features to diagnose the sample according
to the highest grade in the biopsy. To date no specific biomarker or panel of biomarkers is available for exact discrimination between low- and high-grade dysplasia.
Indefinite for dysplasia
When atypical features are observed but not sufficiently convincing for the diagnosis of (low-grade) dysplasia, the diagnosis ‘indefinite for dysplasia’ (IND) can be used. In the Vienna classification system, the IND category is placed between the ‘negative for dysplasia’ and ‘low-grade dysplasia’ category. However, the IND category does not represent a continuum between non-dysplastic Barrett’s and low-grade dysplasia (Figure 1). Indefinite for dysplasia is a category used when the pathologist cannot make a reliable distinction between the presence and absence of neoplasia. The differential diagnosis extends most frequently from non-dysplastic Barrett’s oesophagus to low-grade dysplasia, but also from reactive changes to high-grade dysplasia or adenocarcinoma. Epithelial changes caused by active inflammation and regenerative features at the (neo-) squamo-columnar junction most often necessitate diagnostic caution and may be over-diagnosed as low-grade dysplasia. Furthermore, technical problems with a biopsy such as poor orientation, tangential cutting, marked cautery artefacts or missing surface epithelium can hamper proper diagnostics, mostly due to the inability to appreciate surface maturation. If encountered, this should be addressed in the written report.
Figure 4: Histological characteristics of high-grade dysplasia
A) Overview of a biopsy with on the right, low grade dysplasia with elongated but regular nuclei, contrasting with high-grade dysplasia on the left, displaying a complex growth pattern, back-to-back glands, loss of polarity and nuclear polymorphisms. B) Detail of figure 4A showing high-grade dysplasia with a dilated gland containing intra-luminal necrotic debris. C) Detail of a dysplastic gland (high-grade) with complex architectural changes, nuclear polymorphisms and loss of polarity. D) Overview of small cell type high-grade dysplasia; glands lined with a monolayer of small cuboidal to columnar cells with smooth nuclear contours.
This broad spectrum of differential diagnostic possibilities in the ‘indefinite for dysplasia’ category can be confusing for the clinician. Even among pathologists, the use of this diagnostic category is controversial. Some avoid its use and advise acid suppression in order to decrease inflammatory changes before issuing a re-biopsy.
Others consider it a very important diagnostic category to use when a definite diagnosis cannot be made, but neoplastic changes cannot be ruled out either. The significance of this diagnosis mainly lies in the warning that this patient should not be lost from endoscopic surveillance. Nevertheless, such a diagnosis should be made with restriction and only after careful consideration to allow for proper risk stratification by the clinician. Few studies have tried to specifically address the implication of the diagnosis of IND in Barrett’s oesophagus. Horvath et al. 17 studied 82 ‘indefinite for
dysplasia’ patients without prevalent disease, of which 19 patients developed low-grade or high-low-grade dysplasia after at least one year of follow up. Similar results were shown by Sonwalkar et al. 18 who studied 41 patients from their archives with
a diagnosis of IND. Six patients (17%) showed progression to low-grade dysplasia or esophageal adenocarcinoma after a follow up of 31 months. In both studies, cases with concurrent dysplasia and patients developing neoplasia within 1 year were excluded, but uniform criteria for the diagnosis of IND were not described. A series of 276 patients with Barrett oesophagus was prospectively followed by Younes et al. 19
The definition of the diagnosis of IND used in this study was ‘crypts show dysplastic changes but the surface epithelium is uninvolved, denuded or replaced by squamous epithelium’. No significant difference in progression rate (development of high-grade dysplasia or esophageal adenocarcinoma) between patients with an initial diagnosis of non-dysplastic Barrett oesophagus (3 of 193 patients) or ‘indefinite for dysplasia’ (1 in 48 patients) was observed. However, patients with a multifocal diagnosis of ‘indefinite for dysplasia’ (1 of 7 patients, 14%) showed a similar progression rate as patients with low-grade dysplasia (2 of 21 patients, 10%). These results emphasise the lack of criteria for ‘indefinite for dysplasia,’ and frequently touch upon criteria being used for crypt dysplasia and dysplasia with preserved surface maturation.
20 Because of these limitations, ‘indefinite for dysplasia’ remains a challenging
diagnostic entity and a difficult and controversial subject to study. Multiple attempts to approach these diagnostic difficulties by establishing an algorithm have been made.
16 21 When the diagnosis of dysplasia is considered but no definite conclusion can be
reached, a second opinion by a colleague (gastro-oesophageal) pathologist outside the own institution is advocated. When thereafter no consensus on the presence or absence of dysplasia can be reached, Kaye et al. 21 suggest that additional p53
immunohistochemistry can be useful. When there is no agreement but an aberrant p53 immunohistochemical staining pattern is observed, a diagnosis of neoplasia can be considered. When there is no consensus and no aberrant pattern is observed, the
diagnosis of ‘indefinite for dysplasia’ is warranted, summoning for a new endoscopy and new biopsies. Both the added value of the use of biomarkers and a second opinion by a GI pathologist are described below.
Figure 5: Histological examples of p53 expression in Barrett’s oesophagus
A) Overexpression of p53 in low-grade dysplasia depicted by a strong brown nuclear staining pattern, contrasting with normal background staining, represented by an intermingling mix of light brown and blue nuclei. The aspect of clonality can also be observed with immunohistochemistry (arrow). B) Completely absent staining of p53 in low-grade dysplasia, discernible as strong blue staining which contrasts with normal background staining.
BIOMARKERS
To improve diagnostic reproducibility and risk stratification In Barrett’s oesophagus, considerable research has been conducted. Many biomarkers and also panels of biomarkers have been under investigation, including Ki-67, cyclin A and D1, β-catenin, COX-2, EGFR, α-Methylacyl Coenzyme A racemase (AMACR), p53, aneuploidy, DNA copy numbers, synthesis of Lewis antigens (sialyl Lea), abnormalities in wheat germ
agglutinin (WGA), Aspergillus oryzae lectin (AOL) and many others. 22-36 Even though
results have been promising, all markers have shown considerable limitations. Currently, the marker that may be of potential use in the routine diagnostic setting is expression of the tumour suppressor gene p53. Mutation of the p53 gene may result in a non-functioning protein with aberrant immunohistochemical staining properties. Due to an increased half-life of the truncated protein intense nuclear brown staining can be observed, i.e. there is overexpression due to the presence of more (truncated) protein (Figure 5A). However, a so-called null mutation results in a mutant protein configuration that prevents the antibody to bind the protein. This inability to bind to the p53 antibody results in dark blue stained nuclei and is observed as a completely absent staining pattern (Figure 5B). Interpretation of p53 staining is not always unequivocal, but both overexpression and complete negative staining should contrast evidently to the normal, background-staining pattern of adjacent non-dysplastic epithelium (clonality). When present in the biopsy material, the basal layer of squamous epithelium serves as an internal positive control for the quality of the staining. Aberrant overexpression as seen in neoplastic Barrett oesophagus should display significantly more intense staining of the nuclei. However, a normal background-staining pattern in morphologic dysplastic epithelium does not exclude dysplasia. Studies on the added value of immunohistochemical staining for p53 in general are nevertheless not unequivocal. Current British guidelines advocate the use of p53 as it improves inter-observer agreement, while American guidelines protest against the use of any additional molecular marker. 7 23 37 The field of molecular
pathology in Barrett’s is very dynamic future diagnostic algorithm will most likely incorporate a panel of immunohistochemical biomarkers as well as genomic profiling.
HISTOLOGIC REVIEW
Even though lengthy research has been conducted, no genetic biomarker has emerged yet that could replace histopathologic assessment. 38 39 Treatment and
dysplasia as determined by histological assessment of (surveillance) biopsies. It has long been recognised, however, that the histologic diagnosis of dysplasia in Barrett’s oesophagus is subject to considerable inter- and intra-observer variability. 14 The
variability between observers relates to the complete spectrum of dysplasia grades, but is especially pronounced in the distinction between regenerative changes and low-grade dysplasia. One of the most difficult categories in Barrett’s oesophagus has proven to be the ill-defined ‘indefinite for dysplasia’ (IND) category. Studies have reported inter-observer agreements for diagnosing low-grade dysplasia in Barrett oesophagus that can be classified as poor to fair (κ values ranging from 0.14-0.32). 14 40-44 Studies among gastro-oesophageal pathologists with a special interest in Barrett’s
oesophagus tend to demonstrate higher inter-observer agreements for low-grade dysplasia, with κ values ranging from 0.48-0.69. 21 45-48 These results imply that the
reliability of a low-grade dysplasia diagnosis improves when dedicated pathologists are consulted. Several European studies have reported that risk-stratification into low- and high-risk groups of neoplastic progression is improved if histological review is performed by dedicated BO pathologists. 45-47 49 These studies included histological
review, with agreement between panel pathologists in the upper range of reported inter-observer agreements. The majority of community low-grade dysplasia diagnoses could be downgraded, after expert consensus, to non-dysplastic Barrett’s and those patients subsequently demonstrated a low risk of neoplastic progression (0-0.9% per patient year of follow-up). On the other hand, when low-grade dysplasia was confirmed by expert consensus, the annual risk of progression to high-grade dysplasia and oesophageal adenocarcinoma markedly increased (9.1-13.4% per patient year of follow-up). 45 46 These data are, however, not in accordance with studies from the
United States, where considerably lower neoplastic progression rates have been reported for confirmed low-grade dysplasia. 40 50 The discrepancy is most likely
related to the method of histology review, since a consensus diagnosis was not always included 50 and inter-observer agreement between pathologists was low (κ value of
0.14). 40 In conclusion, it seems that histological review by a homogenous panel of
dedicated Barrett’s oesophagus pathologists can improve risk-stratification in low-grade dysplasia cases. Therefore, current international guidelines recommend review of all low-grade dysplasia cases by dedicated GI pathologists. 6 7 51
ENDOSCOPIC TREATMENT OF HIGH-GRADE LESIONS
The distinction between high-grade dysplasia and mucosal carcinoma on biopsy material is proven to be difficult and exact infiltration depth cannot reliably be determined on biopsy material. Despite the Vienna classification, global differences still exist in the threshold for diagnosing (early) carcinoma. Objective morphology is subjectively interpreted and histomorphologic features that define earliest signs of invasion are not agreed upon. A traditionally used `Western` infiltrative characteristic such as a desmoplastic stromal reaction is rarely observed in mucosal cancer and expansion in single tumour cell is not to be expected in well-differentiated mucosal carcinomas. Furthermore, the importance attributed to cytonuclear features to define an invasive lesion varies geographically and is used with greater emphasis by Eastern colleagues. 52 Diagnostic differences mostly apply to biopsy material
and to a lesser extent to endoscopic resection specimens in which features of well-differentiated mucosal carcinoma such as intra tubular lateral expansion and fusion of neighbouring glands can be more readily appreciated. With available endoscopic resection modalities however this distinction on biopsy material has become less clinical significant and, after careful staging, both will, at least in experienced centres, be treated in a similar way. The number of endoscopic resections is steadily increasing, calling for close collaboration between endoscopist and pathologist. For the pathologist, understanding of the clinical part, including the needs and problems in clinical diagnosis and patient care, deserves attention. The endoscopist on the other hand should be aware of diagnostic issues in histopathology, understand its limitations and deliver the specimen in a condition that allows for a high-quality, reproducible diagnosis to be made.
Handling of the endoscopic resection specimen
Endoscopic resection specimens are handled as one would a surgical specimen, loosely pinning it onto cork, wax or cardboard with the mucosa facing upwards. This is performed in the endoscopy suite immediately after endoscopic resection and before fixation, as it preserves tissue orientation and prevents the lateral margins from excessive curling. Avoid overstretching the specimen as it will shrink during fixation and pinning can cause tears of the mucosa, as well as artificial incorrect depth of invasion measurements. Preferably, the needles are not pinched through the macroscopically most abnormal areas. Fixation in 4% neutral buffered formalin
takes place overnight or for a minimum period of 12 hours. After fixation, the specimen is measured in three dimensions and the surface is scrutinised for visible lesions and the relation to closest margins appropriately assessed. EMR specimens are sometimes resected with a small tumour free area around it is and in some institutions microscope with reverse light is used for orientation and documentation of the specimen. The lateral and deep margins are inked and a photograph of the specimen is taken for orientation and mapping. If the resection consists of multiple specimens, each specimen is measured and photographed separately. The specimen is sliced into 2-3mm thick sections with slicing direction allowing for optimal assessment of the most threatened lateral margin. Depending on the size of the specimen, the slicing direction may be chosen parallel to its shortest axis, since long slices may curl during embedding and processing and can result in non-interpretable slides. If the resection margin of interest is not optimally visible in this way, another slicing direction may be chosen. In larger resection specimens (for instance endoscopic submucosal dissection (ESD) specimens) and depending on the distribution of visible lesions, divided perpendicular or en face sections of the resection margin may be chosen. 53
The advantage of perpendicular sections is the complete visualisation of the resection margins, especially in larger specimens. The disadvantage is that slides are often more difficult to interpret compared to en face sectioning and in a focal positive resection margin, multiple step sectioning will be necessary before determining the definite margin status. The slicing axis is documented on the photograph. If the resection is piecemeal, orientation between the specimens and the R0 status of the lateral margins cannot be adequately assessed. Consecutive slices of the specimen are placed into cassettes, with a maximum of three slices per cassette, the first and last slice being embedded separately. To allow for perfect orientated and optimal interpretable slides, slices are embedded and cut by a trained and dedicated technician.
A systematic assessment and description of features is important to ensure adequate follow-up of the patient. The pathology report includes the type(s) of surface epithelium as well as that of the deepest tissue layer. When present, dysplasia is graded and lateral margin status is described. If a(n) (mucosal) adenocarcinoma is present, the specimen is analysed for four factors to assess the risk for recurrent disease and lymph node metastasis and therefore potential subsequent oesophagectomy. These risk factors are: differentiation grade, depth of invasion, basal margin status, and presence or absence of (lympho-) vascular invasion.
Differentiation grade of (mucosal) adenocarcinoma is divided into well, moderate or poor, the latter correlating with a worse prognosis. The WHO classification of tumours of the digestive system 54 requires to take the most prominent differentiation pattern
for grading into account, without acknowledging minor parts of poor differentiation or the nature of the invasive tumour front. Foci of poor differentiation, however, influence patient outcome even if the majority of the tumour is well differentiated. 55 In other
organs such as the prostate, the heterogeneity of tumour differentiation is taken into account by combining both the most prominent and the worst differentiation patterns for final grading. Poor differentiation grade is considered to be a risk factor in EMR specimens; the significance of it being the predominant pattern or just a minor component is unknown. Until this is clear, we propose that when grading a(n) (mucosal) adenocarcinoma in an EMR specimen, to mention both the worst and the most prominent differentiation grade in the pathology report. This can be either done using descriptive terms, but is preferably noted as a percentage of total tumour volume. The most poorly differentiated part should be recognizable at low to medium magnification and includes solid growth pattern, expansion as single cells or signet cell differentiation.
For determination of infiltration depth, due to the frequent duplication of the muscularis mucosae, three classifications of T1a disease exists. One classification system is recommended by the American Joint Committee on Cancer (AJCC), as described by Hölscher et al. 56 Mucosal disease is divided into three levels by
high-grade dysplasia (m1), intramucosal carcinoma limited to the lamina propria (m2) and intramucosal carcinoma reaching within the muscularis mucosa (m3), whether duplicated or not. 56 The system by Westerterp et al is similar, except for invasion
in the double muscularis mucosae: into the superficial layer is defined as m2, into the deeper (pre-existing) muscularis mucosae is defined as m3. 57 The third system is
propagated by Vieth and Stolte and also takes duplicated muscularis mucosae into account. 58 This is a four-tiered system and classifies m1 as limited to the lamina propria,
m2 as invasion into the superficial layer of the duplicated muscularis mucosae, m3 as invasion into the space between the original and the duplicated muscularis mucosae and m4 as invasion into the deep (original) muscularis mucosae (Figure 6). This system is based on the histoanatomical structures of Barrett’s oesophagus and by providing a description of the exact histological layers underscores the significance of the presence of frequently duplicated muscularis mucosa. 59 Invasion of (mucosal) carcinoma in any
second present layer of muscularis in a mucosectomy specimen should not incorrectly be upstaged as invasion in the muscularis propria. The space between the duplicated and original muscularis mucosa consists of loosely connective tissue with thin walled blood and lymphatic vessels. The connective tissue of the submucosa is characterised by the presence of thick walled blood vessels, a feature frequently used to discriminate between the two layers of connective tissue. Sometimes some thicker walled blood vessels can be observed as well in the space between the duplicated and original muscularis mucosa. The clinician and pathologist need to agree on which classification scheme is used to avoid misunderstandings in reading the report, which could cause possible under-and overtreatment. The system of the AJCC 56 and the system by Vieth
and Stolte 58 are most frequently used, some authors even advocate to document
both systems. 60 In order to be able to compare studies that make use of different
classification systems, description of the exact depth of infiltration is crucial. This can be achieved by noting the exact histological layer instead of only mentioning the m-subclassification. However, it is of greater clinical importance to exclude submucosal adenocarcinoma, which is defined as invasion beyond the muscularis mucosa (T1b), since the risk of lymph node metastases or disease recurrence is significantly increased in this group of patients.
Figure 6: Different classification systems used to subdivide mucosal adenocarcinoma in relation
In all three classification systems mentioned above, the invasion depth in the submucosa is subdivided into sm1-3 and is based on the superficial, middle and deep 1/3 level of the submucosa in oesophagectomy specimens. Rarely, the complete submucosa up until the muscularis propria is present in a mucosectomy specimen. Invasion in the upper third of the submucosa (sm1) is defined as up to 500 microns below the deepest fibre of the original muscularis mucosa, with every lesion that reaches deeper being a strong indication for surgery. Invasion in the middle third of the submucosa (sm2) is defined as up to 1000 microns and invasion in the deeper third of the submucosa (sm3) as more than 1000 microns. Furthermore the vertical safety margin (VM) from the deepest point of infiltration to the basal resection margin is measured in microns.
The assessment of the R-status is complicated, since there are no clear definitions for local endoscopic resection margins. For the lateral margins, the presence and absence of dysplasia and (mucosal) carcinoma is described in the pathology report but only in case of a single specimen, radicality at the lateral margins can reliably be assessed. A positive lateral margin after local endoscopic resection is usually not a major clinical issue, as endoscopic resection is accompanied by endoscopic ablation therapy. A minimal lateral safety margin (HM) is therefore not defined. The ablation procedure can be repeated until all neoplastic tissue, and preferably the segment of intestinal or columnar metaplasia as well, is removed, or a second endoscopic resection can be performed. A positive deep margin in pT1b disease correlates with a worse prognosis and is a definite indicator for surgical resection. In many organs it has been shown that a safety margin of <1mm is associated with a greater risk of recurrence or is considered an R1 resection. In endoscopic resections, a minimal deep (vertical) safety margin is not defined and since only part of the submucosa is removed by endoscopic resection, the vertical margin of submucosal invasion (pT1b) is bound to be limited. The exact determination of infiltration depth and deep (vertical) margin can be a challenge further hampered by artefacts due to injection of fluid during the endoscopic procedure, handling and processing of the resection specimen and shrinkage of the tissue due to formalin fixation. Needle tracks can occur at the lateral border and cause artefacts that may mimic a positive basal resection margin. Curling up of the lateral border after fixation can resemble the deep margin upon microscopic view and cause unnecessary concerns about the radicality of the resection (Figure 7). Additional immunohistochemical stainings, such as a desmin stain to highlight muscle
fibres, are of use to visualise the (duplicated) muscularis mucosae and to highlight its absence in a curled lateral border. Given the problems and artefacts at the lateral border of an EMR specimen, the maximal infiltration depth or R-status of the basal resection margin should preferably not be measured at that location.
Vascular invasion or vessel permeation can involve small vessels like capillaries or lymphatic vessels, less frequently larger veins and arteries, 60 and its presence implies
a worse prognosis. Optional items for signing out are the width of tumour infiltration in the submucosa, perineural invasion and the total tumour thickness measured from the surface down to the deepest point of infiltration in microns.
Tumour differentiation grade, depth of invasion, basal margin status and presence of vascular invasion are assessed to stratify risk and the need for subsequent surgical oesophagectomy. The main goal of oesophagectomy in T1 adenocarcinoma is to benefit from local control in case of a positive basal margin in an endoscopic resection specimen and contain local recurrence by lymph node dissection. Series reporting lymph nodes metastases in T1 oesophageal adenocarcinoma vary from 16-44%, 57 61-66
most series showing a lower rate of lymph node metastases in sm1 adenocarcinoma (0- 20%) as compared to deeper infiltration. However, these series are based on retrospective studies on surgical resection specimens. Manner et al. 55 prospectively
studied 72 patients with endoscopically resected sm1 (T1b) adenocarcinoma (mean follow up 60 months) and patients were divided into low risk (LR) (n=49) and high risk (HR) (n=23) groups on basis of the presence of >1 risk factor. In the HR group, 9% of patients developed lymphe node metastases, in the LR group only 2%. These results suggest that selected patients with early submucosal cancers are also eligible for endoscopic management. It is not clear which (combination) of the individual risk factors has the highest prognostic value. Further research of larger series with longer follow up are necessary and are likely to be combined with a combination of immunohistochemical and genetic profiling. For the time being, however, controversy for endoscopic treatment of sm1 submucosal cancers remains. Management of these patients should not take place outside of specialised centres nor without consensus in multidisciplinary meetings.
Figure 7: Histological pitfalls when assessing the basal resection margin of an endoscopic
resection margin
A) Possible overdiagnosis of a positive basal resection margin of a mucosectomy due to curling of the lateral edge. B) Desmin stain of Figure 7A, highlighting the folded contours of the muscularis mucosa due to curling of the specimen.
ACKNOWLEDGEMENTS
We kindly thank G. Johan A. Offerhaus for his critical reading of the manuscript, and Kees A. Seldenrijk, Fiebo J.K. ten Kate and Mike Visser for their valuable contributions to the definitions and visual material.
REFERENCES
1. Gatenby PA, Ramus JR, Caygill CP, et al. Relevance of the detection of intestinal metaplasia in non-dysplastic columnar-lined oesophagus. Scandinavian journal of gastroenterology 2008;43(5):524-30. doi: 10.1080/00365520701879831
2. Barr H, Upton MP, Orlando RC, et al. Barrett’s esophagus: histology and immunohistology.
Ann N Y Acad Sci 2011;1232:76-92. doi: 10.1111/j.1749-6632.2011.06046.x
3. de Jonge PJ, van Blankenstein M, Looman CW, et al. Risk of malignant progression in patients with Barrett’s oesophagus: a Dutch nationwide cohort study. Gut 2010;59(8):1030-6. doi: 10.1136/gut.2009.176701
4. Hvid-Jensen F, Pedersen L, Drewes AM, et al. Incidence of adenocarcinoma among patients with Barrett’s esophagus. The New England journal of medicine 2011;365(15):1375-83. doi: 10.1056/NEJMoa1103042
5. Bhat S, Coleman HG, Yousef F, et al. Risk of malignant progression in Barrett’s esophagus patients: results from a large population-based study. Journal of the National Cancer Institute 2011;103(13):1049-57. doi: 10.1093/jnci/djr203
6. American Gastroenterological A, Spechler SJ, Sharma P, et al. American Gastroenterological Association medical position statement on the management of Barrett’s esophagus.
Gastroenterology 2011;140(3):1084-91. doi: 10.1053/j.gastro.2011.01.030
7. Fitzgerald RC, di Pietro M, Ragunath K, et al. British Society of Gastroenterology guidelines on the diagnosis and management of Barrett’s oesophagus. Gut 2014;63(1):7-42. doi: 10.1136/gutjnl-2013-305372
8. Bennett C, Moayyedi P, Corley DA, et al. BOB CAT: a Large-Scale Review and Delphi Consensus for Management of Barrett’s Esophagus With No Dysplasia, Indefinite for, or Low-Grade Dysplasia. The American journal of gastroenterology 2015;110(5):662-82; quiz 83. doi: 10.1038/ajg.2015.55
9. Schlemper RJ, Kato Y, Stolte M. Diagnostic criteria for gastrointestinal carcinomas in Japan and Western countries: proposal for a new classification system of gastrointestinal epithelial neoplasia. Journal of gastroenterology and hepatology 2000;15 Suppl:G49-57. 10. Riddell RH, Goldman H, Ransohoff DF, et al. Dysplasia in inflammatory bowel disease:
standardized classification with provisional clinical applications. Human pathology 1983;14(11):931-68.
11. Schlemper R, Riddell R, Kato Y, et al. The Vienna classification of gastrointestinal epithelial neoplasia. Gut 2000;47(2):251-5. doi: 10.1136/gut.47.2.251
12. Voltaggio L, Montgomery E, Lam-Himlin D. A clinical and histopathologic focus on Barrett esophagus and Barrett-related dysplasia. Arch Pathol Lab Med 2011;135:1249-60. doi: 10.5858/arpa.2011-0019-RA
13. Reid BJ, Haggitt RC, Rubin CE, et al. Observer variation in the diagnosis of dysplasia in Barrett’s esophagus. Human pathology 1988;19(2):166-78.
14. Montgomery E, Bronner MP, Goldblum JR, et al. Reproducibility of the diagnosis of dysplasia in Barrett esophagus: a reaffirmation. Human pathology 2001;32(4):368-78. doi: 10.1053/hupa.2001.23510
15. Mahajan D, Bennett AE, Liu X, et al. Grading of gastric foveolar-type dysplasia in Barrett’s esophagus. Modern pathology : an official journal of the United States and Canadian Academy
of Pathology, Inc 2010;23(1):1-11. doi: 10.1038/modpathol.2009.147
16. Odze RD. Diagnosis and grading of dysplasia in Barrett’s oesophagus. Journal of clinical
pathology 2006;59(10):1029-38. doi: 10.1136/jcp.2005.035337
17. Horvath B, Singh P, Xie H, et al. Risk for esophageal neoplasia in Barrett’s esophagus patients with mucosal changes indefinite for dysplasia. Journal of gastroenterology and
hepatology 2015;30(2):262-7. doi: 10.1111/jgh.12696
18. Sonwalkar SA, Rotimi O, Scott N, et al. A study of indefinite for dysplasia in Barrett’s oesophagus: reproducibility of diagnosis, clinical outcomes and predicting progression with AMACR (alpha-methylacyl-CoA-racemase). Histopathology 2010;56(7):900-7. doi: 10.1111/j.1365-2559.2010.03571.x
19. Younes M, Ertan A, Lechago LV, et al. p53 Protein accumulation is a specific marker of malignant potential in Barrett’s metaplasia. Digestive diseases and sciences 1997;42(4):697-701.
20. Lomo LC, Blount PL, Sanchez CA, et al. Crypt dysplasia with surface maturation: a clinical, pathologic, and molecular study of a Barrett’s esophagus cohort. The American journal of
surgical pathology 2006;30(4):423-35.
21. Kaye PV, Haider SA, Ilyas M, et al. Barrett’s dysplasia and the Vienna classification: reproducibility, prediction of progression and impact of consensus reporting and p53 immunohistochemistry. Histopathology 2009;54(6):699-712. doi: 10.1111/j.1365-2559.2009.03288.x
22. Bennett M, Mashimo H. Molecular markers and imaging tools to identify malignant potential in Barrett’s esophagus. World journal of gastrointestinal pathophysiology 2014;5(4):438-49. doi: 10.4291/wjgp.v5.i4.438
23. de Jonge PJ, van Blankenstein M, Grady WM, et al. Barrett’s oesophagus: epidemiology, cancer risk and implications for management. Gut 2014;63(1):191-202. doi: 10.1136/gutjnl-2013-305490
24. Weaver JMJ, Ross-Innes CS, Fitzgerald RC. The ‘-omics’ revolution and oesophageal adenocarcinoma. Nat Rev Gastro Hepat 2014;11(1):19-27. doi: DOI 10.1038/nrgastro.2013.150 25. Ong CA, Lao-Sirieix P, Fitzgerald RC. Biomarkers in Barrett’s esophagus and esophageal
adenocarcinoma: predictors of progression and prognosis. World journal of gastroenterology
: WJG 2010;16(45):5669-81.
26. Fitzgerald RC. Complex diseases in gastroenterology and hepatology: GERD, Barrett’s, and esophageal adenocarcinoma. Clinical gastroenterology and hepatology : the official clinical
practice journal of the American Gastroenterological Association 2005;3(6):529-37. [published
27. Bani-Hani K, Martin IG, Hardie LJ, et al. Prospective study of cyclin D1 overexpression in Barrett’s esophagus: association with increased risk of adenocarcinoma. Journal of the
National Cancer Institute 2000;92(16):1316-21.
28. Scheil-Bertram S, Lorenz D, Ell C, et al. Expression of alpha-methylacyl coenzyme A racemase in the dysplasia carcinoma sequence associated with Barrett’s esophagus.
Modern pathology : an official journal of the United States and Canadian Academy of Pathology, Inc 2008;21(8):961-7. doi: 10.1038/modpathol.2008.73
29. Dorer R, Odze RD. AMACR immunostaining is useful in detecting dysplastic epithelium in Barrett’s esophagus, ulcerative colitis, and Crohn’s disease. The American journal of surgical
pathology 2006;30(7):871-7. doi: 10.1097/01.pas.0000213268.30468.b4
30. Lisovsky M, Falkowski O, Bhuiya T. Expression of alpha-methylacyl-coenzyme A racemase in dysplastic Barrett’s epithelium. Human pathology 2006;37(12):1601-6. doi: 10.1016/j. humpath.2006.06.009
31. Kastelein F, Biermann K, Steyerberg EW, et al. Value of alpha-methylacyl-CoA racemase immunochemistry for predicting neoplastic progression in Barrett’s oesophagus.
Histopathology 2013;63(5):630-9. doi: 10.1111/his.12216
32. Reid BJ, Prevo LJ, Galipeau PC, et al. Predictors of progression in Barrett’s esophagus II: baseline 17p (p53) loss of heterozygosity identifies a patient subset at increased risk for neoplastic progression. The American journal of gastroenterology 2001;96(10):2839-48. doi: 10.1111/j.1572-0241.2001.04236.x
33. Reid BJ, Levine DS, Longton G, et al. Predictors of progression to cancer in Barrett’s esophagus: baseline histology and flow cytometry identify low- and high-risk patient subsets. The American journal of gastroenterology 2000;95(7):1669-76. doi: 10.1111/j.1572-0241.2000.02196.x
34. Galipeau PC, Li X, Blount PL, et al. NSAIDs modulate CDKN2A, TP53, and DNA content risk for progression to esophageal adenocarcinoma. PLoS medicine 2007;4(2):e67. doi: 10.1371/ journal.pmed.0040067
35. Paulson TG, Galipeau PC, Reid BJ. Loss of heterozygosity analysis using whole genome amplification, cell sorting, and fluorescence-based PCR. Genome research 1999;9(5):482-91. 36. Bird-Lieberman EL, Dunn JM, Coleman HG, et al. Population-based study reveals new risk-stratification biomarker panel for Barrett’s esophagus. Gastroenterology 2012;143(4):927-35 e3. doi: 10.1053/j.gastro.2012.06.041
37. Menezes A, Tierney A, Yang YX, et al. Adherence to the 2011 American Gastroenterological Association medical position statement for the diagnosis and management of Barrett’s esophagus. Diseases of the esophagus : official journal of the International Society for Diseases
of the Esophagus / ISDE 2014 doi: 10.1111/dote.12228
38. Timmer MR, Sun G, Gorospe EC, et al. Predictive biomarkers for Barrett’s esophagus: so near and yet so far. Diseases of the esophagus : official journal of the International Society for
Diseases of the Esophagus / ISDE 2013;26(6):574-81. doi: 10.1111/dote.12015 [published Online
39. Varghese S, Lao-Sirieix P, Fitzgerald RC. Identification and clinical implementation of biomarkers for Barrett’s esophagus. Gastroenterology 2012;142(3):435-41 e2. doi: 10.1053/j. gastro.2012.01.013
40. Wani S, Falk GW, Post J, et al. Risk factors for progression of low-grade dysplasia in patients with Barrett’s esophagus. Gastroenterology 2011;141(4):1179-86, 86 e1. doi: 10.1053/j. gastro.2011.06.055
41. Skacel M, Petras RE, Gramlich TL, et al. The diagnosis of low-grade dysplasia in Barrett’s esophagus and its implications for disease progression. The American journal of
gastroenterology 2000;95(12):3383-7. doi: 10.1111/j.1572-0241.2000.03348.x
42. Coco DP, Goldblum JR, Hornick JL, et al. Interobserver variability in the diagnosis of crypt dysplasia in Barrett esophagus. The American journal of surgical pathology 2011;35(1):45-54. doi: 10.1097/PAS.0b013e3181ffdd14
43. Kerkhof M, van Dekken H, Steyerberg EW, et al. Grading of dysplasia in Barrett’s oesophagus: substantial interobserver variation between general and gastrointestinal pathologists.
Histopathology 2007;50(7):920-7. doi: 10.1111/j.1365-2559.2007.02706.x
44. Wani S, Mathur SC, Curvers WL, et al. Greater interobserver agreement by endoscopic mucosal resection than biopsy samples in Barrett’s dysplasia. Clinical gastroenterology
and hepatology : the official clinical practice journal of the American Gastroenterological Association 2010;8(9):783-8. doi: 10.1016/j.cgh.2010.04.028
45. Curvers WL, ten Kate FJ, Krishnadath KK, et al. Low-grade dysplasia in Barrett’s esophagus: overdiagnosed and underestimated. The American journal of gastroenterology 2010;105(7):1523-30. doi: 10.1038/ajg.2010.171
46. Duits LC, Phoa KN, Curvers WL, et al. Barrett’s oesophagus patients with low-grade dysplasia can be accurately risk-stratified after histological review by an expert pathology panel. Gut 2014 doi: 10.1136/gutjnl-2014-307278
47. Lim CH, Treanor D, Dixon MF, et al. Low-grade dysplasia in Barrett’s esophagus has a high risk of progression. Endoscopy 2007;39(7):581-7. doi: 10.1055/s-2007-966592
48. Pech O, Vieth M, Schmitz D, et al. Conclusions from the histological diagnosis of low-grade intraepithelial neoplasia in Barrett’s oesophagus. Scandinavian journal of gastroenterology 2007;42(6):682-8. doi: 10.1080/00365520601075803
49. Vieth M, Schubert B, Lang-Schwarz K, et al. Frequency of Barrett’s neoplasia after initial negative endoscopy with biopsy: a long-term histopathological follow-up study. Endoscopy 2006;38(12):1201-5. doi: 10.1055/s-2006-944993
50. Jung KW, Talley NJ, Romero Y, et al. Epidemiology and natural history of intestinal metaplasia of the gastroesophageal junction and Barrett’s esophagus: a population-based study. The American journal of gastroenterology 2011;106(8):1447-55; quiz 56. doi: 10.1038/ ajg.2011.130
51. Whiteman DC, Appleyard M, Bahin FF, et al. Australian clinical practice guidelines for the diagnosis and management of Barrett’s Esophagus and Early Esophageal Adenocarcinoma.
52. Vieth M, Riddell RH, Montgomery EA. High-grade dysplasia versus carcinoma: east is east and west is west, but does it need to be that way? The American journal of surgical pathology 2014;38(11):1453-6. doi: 10.1097/PAS.0000000000000288
53. Vieth M, Langner C, Neumann H, et al. Barrett’s esophagus. Practical issues for daily routine diagnosis. Pathol Res Pract 2012;208(5):261-8. doi: 10.1016/j.prp.2012.03.001
54. Bosman FT, Carneiro, F., Hruban, R. H., Theise, N. D. . WHO classification of tumours of the digestive system. Geneva: World Health Organization 2010.
55. Manner H, Pech O, Heldmann Y, et al. The frequency of lymph node metastasis in early-stage adenocarcinoma of the esophagus with incipient submucosal invasion (pT1b sm1) depending on histological risk patterns. Surg Endosc 2015;29(7):1888-96. doi: 10.1007/ s00464-014-3881-3
56. Holscher AH, Vallbohmer D, Bollschweiler E. Early Barrett’s carcinoma of the esophagus.
Ann Thorac Cardiovasc Surg 2008;14(6):347-54.
57. Westerterp M, Koppert LB, Buskens CJ, et al. Outcome of surgical treatment for early adenocarcinoma of the esophagus or gastro-esophageal junction. Virchows Archiv : an
international journal of pathology 2005;446(5):497-504. doi: 10.1007/s00428-005-1243-1
58. Vieth M, Stolte M. Pathology of early upper GI cancers. Best practice & research Clinical
gastroenterology 2005;19(6):857-69. doi: 10.1016/j.bpg.2005.02.008
59. Kaneshiro DK, Post JC, Rybicki L, et al. Clinical significance of the duplicated muscularis mucosae in Barrett esophagus-related superficial adenocarcinoma. The American journal
of surgical pathology 2011;35(5):697-700. doi: 10.1097/PAS.0b013e3182159c4b
60. Kumarasinghe MP, Brown I, Raftopoulos S, et al. Standardised reporting protocol for endoscopic resection for Barrett oesophagus associated neoplasia: expert consensus recommendations. Pathology 2014;46(6):473-80. doi: 10.1097/PAT.0000000000000160 61. Stein HJ, Feith M, Bruecher BLDM, et al. Early esophageal cancer - Pattern of lymphatic
spread and prognostic factors for long-term survival after surgical resection. Annals of
surgery 2005;242(4):566-75. doi: DOI 10.1097/01.sla.0000184211.75970.85
62. Liu L, Hofstetter WL, Rashid A, et al. Significance of the depth of tumor invasion and lymph node metastasis in superficially invasive (T1) esophageal adenocarcinoma. The American
journal of surgical pathology 2005;29(8):1079-85.
63. Bollschweiler E, Baldus SE, Schroder W, et al. High rate of lymph-node metastasis in submucosal esophageal squamous-cell carcinomas and adenocarcinomas. Endoscopy 2006;38(2):149-56. doi: 10.1055/s-2006-924993
64. Ancona E, Rampado S, Cassaro M, et al. Prediction of lymph node status in superficial esophageal carcinoma. Annals of surgical oncology 2008;15(11):3278-88. doi: 10.1245/s10434-008-0065-1
65. Hagen JA, DeMeester SR, Peters JH, et al. Curative resection for esophageal adenocarcinoma: analysis of 100 en bloc esophagectomies. Annals of surgery 2001;234(4):520-30; discussion 30-1.
