www.nature.

com/modpathol

REVIEW ARTICLE OPEN

Controversies in Pathology

Utility of ancillary studies in the diagnosis and risk assessment

of Barrett’s esophagus and dysplasia
1✉
Won-Tak Choi , Gregory Y. Lauwers2 and Elizabeth A. Montgomery3

© The Author(s) 2022

Barrett’s esophagus (BE) is a major risk factor for the development of esophageal adenocarcinoma (EAC). BE patients undergo
periodic endoscopic surveillance with biopsies to detect dysplasia and EAC, but this strategy is imperfect owing to sampling error
and inconsistencies in the diagnosis and grading of dysplasia, which may result in an inaccurate diagnosis or risk assessment for
progression to EAC. The desire for more accurate diagnosis and better risk stratification has prompted the investigation and
development of potential biomarkers that might assist pathologists and clinicians in the management of BE patients, allowing more
aggressive endoscopic surveillance and treatment options to be targeted to high-risk individuals, while avoiding frequent
1234567890();,:

surveillance or unnecessary interventions in those at lower risk. It is known that progression of BE to dysplasia and EAC is
accompanied by a host of genetic alterations, and that exploration of these markers could be potentially useful to diagnose/grade
dysplasia and/or to risk stratify BE patients. Several biomarkers have shown promise in identifying early neoplastic transformation
and thus may be useful adjuncts to histologic evaluation. This review provides an overview of some of the currently available
biomarkers and assays, including p53 immunostaining, Wide Area Transepithelial Sampling with Three-Dimensional Computer-
Assisted Analysis (WATS3D), TissueCypher, mutational load analysis (BarreGen), fluorescence in situ hybridization, and DNA content
abnormalities as detected by DNA flow cytometry.

Modern Pathology (2022) 35:1000–1012; https://doi.org/10.1038/s41379-022-01056-0

INTRODUCTION of BE patients with dysplasia (~20%) are resistant to endoscopic

Barrett’s esophagus (BE) is a pre-neoplastic condition that is therapy, and recurrences or progression to EAC during endoscopic
associated with an increased risk of developing dysplasia and therapy are not uncommon12,17–23. As for LGD, although
esophageal adenocarcinoma (EAC)1–3. It is defined as endoscopi- continued surveillance every 12 months is an acceptable
cally visible columnar epithelium containing goblet cells (intestinal approach, there has been a shift toward endoscopic ablation
metaplasia). Although the American Gastroenterological Associa- therapy in recent years4,14–16,24–26. Subsequently, there is greater
tion has not specified a length requirement1, the American emphasis on optimization of the diagnosis of dysplasia as well as
College of Gastroenterology requires extension at least 1 cm identification of patients who are more likely to progress to HGD/
proximal to the gastroesophageal junction4. BE is a genetically EAC and/or have a poor response to endoscopic therapy.
unstable metaplastic epithelium that accumulates an increasing However, considering the annual cancer risk for patients with
number of genetic and chromosomal abnormalities as it non-dysplastic BE (NDBE) is low (0.1–0.5% per year)27–29,
progresses to low-grade dysplasia (LGD), high-grade dysplasia identification of reliable biomarkers of low risk to allow prolonga-
(HGD), and eventually EAC5–10. Because dysplasia is currently the tion of surveillance intervals compared with the current recom-
primary clinical biomarker used to identify patients who are at an mendations of repeating endoscopy every 3–5 years in those with
increased risk for EAC, clinical guidelines recommend that BE NDBE4 remains an important goal of biomarker research.
patients undergo periodic endoscopic surveillance with biopsies Even though endoscopic therapy has revolutionized the
to detect dysplasia4. This allows for risk stratification and treatment of BE patients, the current surveillance protocols based
management of BE patients based upon the presence and grade on the histologic classification of BE have several shortcomings,
of dysplasia prior to the development of EAC. The detection of including limited sampling of the affected BE segment (leading to
HGD usually prompts endoscopic therapy, generally in the form of false negative biopsy results), sampling error of potentially
radiofrequency ablation (RFA) with or without endoscopic neoplastic lesions, and interobserver variability among patholo-
mucosal resection (EMR), due to its more frequent association gists in the diagnosis and grading of dysplasia, particularly for
with EAC compared with LGD11–16. However, a significant number LGD30–35. In fact, there is evidence that the current surveillance

1
University of California at San Francisco, Department of Pathology, San Francisco, CA 94143, USA. 2H. Lee Moffitt Cancer Center and Research Institute, Department of Pathology,
Tampa, FL 33612, USA. 3University of Miami Miller School of Medicine, Department of Pathology and Laboratory Medicine, Miami, FL 33136, USA. ✉email: [email protected]

Received: 15 December 2021 Revised: 9 February 2022 Accepted: 13 February 2022

Published online: 8 March 2022
 W.-T. Choi et al.
1001
protocols may not be effective in reducing mortality from EAC, Unfortunately, dysplasia has a number of limitations as a
with one study demonstrating that patients with fatal disease biomarker. First, dysplasia is often focal and may not be
were nearly as likely to have received surveillance (55%) as were endoscopically visible, so sampling error is a major issue as most
controls (60%)36. Furthermore, the rate of missed HGD/EAC (i.e., surveillance techniques sample only a minority of the BE segment.
diagnosed within 1 year of negative endoscopy) is high Although Reid et al. reported that four-quadrant biopsies taken
(19–24%)37, suggesting that early repeat endoscopy, ideally within every 1 cm in the BE segment (also known as the “Seattle
1 year of an initial BE diagnosis, may be crucial, although the cost- protocol”) can consistently detect early cancers arising in HGD92,
effectiveness of this approach remains to be determined. Notably, most endoscopists do not adhere to this protocol and take too
in a recent meta-analysis of 24 cohort studies of NDBE or LGD few biopsies, compounding the problem of sampling error.
patients followed for at least 3 years after index endoscopy, Second, consistent diagnosis and grading of dysplasia by
Visrodia et al. reported that ~25% of EACs and 27% of HGD/EACs histology is challenging, as exemplified by a relatively high
were classified as missed37. When only NDBE patients were degree of interobserver variability in the histologic classification of
considered, the rates of missed EACs and HGD/EACs were 24% BE among pathologists, particularly toward the lower end of the
and 19%, respectively37. spectrum30–32,34,93. The most pronounced variability is linked to
Consequently, there is an increased interest in ancillary tests that the diagnosis of LGD, with a recent study illustrating sub-optimal
could (1) improve the diagnostic accuracy of dysplasia (and its interobserver agreement for LGD (kappa = 0.11) even among
grading) in challenging situations to avoid a repeat endoscopic gastrointestinal (GI) pathologists32. In another study, up to 85% of
examination with biopsies (potentially more expensive than most LGD cases were downgraded to NDBE or IND following expert
ancillary tests); (2) predict which NDBE or LGD patients are at a higher pathology review91. Even though an excellent interobserver
risk for developing HGD/EAC (including missed lesions) so that such agreement for HGD among GI pathologists has been reported in
patients can be identified early and successfully treated with earlier studies30,31, a more recent study demonstrated that upon
endoscopic therapy to prevent progression to EAC; (3) identify review of 485 HGD samples from both academic and private
patients who are less likely to develop HGD/EAC so that the centers by experienced GI pathologists, up to 40% of these cases
surveillance of low-risk patients can be reduced; and/or (4) predict were reinterpreted as LGD, IND, NDBE, or no BE93. Consequently,
those more likely to have a poor response to endoscopic therapy. In both the American College of Gastroenterology and the American
this regard, a variety of biomarkers and assays, such as p53 Gastroenterological Association strongly recommend that all
immunostaining38–42, Wide Area Transepithelial Sampling with potential dysplasia cases be confirmed by at least one experi-
Three-Dimensional Computer-Assisted Analysis (WATS3D)43–50, Tissue- enced GI pathologist before embarking on a management
Cypher51–58, mutational load analysis (BarreGen)59–62, fluorescence plan4,94. This recommendation is further supported by several
in situ hybridization (FISH)7,63–67, and DNA content abnormalities as studies demonstrating a strong correlation between the number
detected by DNA flow cytometry68–77 have been extensively of pathologists who agree with a diagnosis of dysplasia and the
evaluated. Although none of these studies have comprehensively rate of neoplastic progression. For instance, Skacel et al. showed
evaluated the potential utility of these biomarkers in reducing that the rate of progression was 80% when three GI pathologists
mortality from EAC compared with the current surveillance standards, agreed on a diagnosis of LGD, while the rate was 41% when two
they have demonstrated a potential benefit when used in combina- GI pathologists agreed95. Finally, even if the issues stated above
tion with histologic findings to assist in the diagnosis and/or risk could be resolved, there are no observable histologic features in
stratification of BE and dysplasia. As such, this review provides an NDBE or LGD on hematoxylin and eosin (H&E) staining that can
overview of these biomarkers and tests that appear most promising accurately identify those patients most likely to develop HGD/EAC
based on the availability of multiple published results and/or on their versus remain stable for years. Indeed, recent studies have
commercial availability. suggested that many EACs develop through a more direct,
accelerated pathway in which TP53 mutation is followed by
doubling of the whole genome, rapidly resulting in genomic
DYSPLASIA AS A BIOMARKER FOR RISK STRATIFICATION instability, oncogenic amplifications, and EAC, rather than through
Currently, dysplasia is the primary clinical biomarker used for risk the stepwise accumulation of tumor suppressor alterations96,97.
assessment in the surveillance and management of BE patients. This accelerated pathway to EAC might explain in part why
Morphologically, dysplasia is defined as unequivocal neoplastic endoscopic surveillance is sometimes unsuccessful in detecting
epithelium that remains confined within the basement membrane dysplasia before the development of EAC in some BE patients36.
of the epithelium from which it developed, and it is classified as (1) Overall, these results suggest that additional or alternative
negative for dysplasia, (2) indefinite for dysplasia (IND), (3) LGD, or biomarker(s) may be useful to better risk stratify BE patients.
(4) HGD3,30,31,78. The rationale for its use as the primary clinical
biomarker is based on the premise that EAC in BE patients develops
through a sequence of molecular (i.e., loss of CDKN2A followed by P53 IMMUNOSTAINING AS A DIAGNOSTIC AND RISK
TP53 inactivation and aneuploidy) and morphologic changes that STRATIFICATION BIOMARKER
begin with intestinal metaplasia and then progress from LGD to Immunohistochemistry (IHC) for p53 to confirm a dysplasia
HGD, and ultimately to EAC3,5,68,71,78–82. This is also supported by diagnosis or predict likelihood of progression to EAC is of interest
multiple outcome studies demonstrating a strong correlation of but has limitations, as summarized by others38,98. The TP53 gene
higher EAC rates with increasing levels of dysplasia. While the annual encodes p53, which prevents mutations. Normal cells have low
cancer risk for NDBE patients is low (0.1–0.5% per year)27–29, HGD is levels of this protein in their nuclei, but the gene and protein are
considered a key premalignant step that is associated with a greater upregulated in the presence of DNA damage or stress, resulting in
risk of either already having EAC or developing it on follow-up DNA repair, growth attenuation, and apoptosis. In dysplastic cells
(16–100%)83–88. The natural history of LGD is more controversial, and EAC, mutations in TP53 lead to aberrant nuclear accumulation
with variable progression rates ranging from 0.4 to 13.4% per year89– of abnormal p53 protein (which has a long half-life) that can be
91
. It is worth emphasizing that it is not often possible to distinguish detected on immunostaining (Fig. 1A, B). Alternatively, truncating
true progression from missed lesions in these outcome studies. In mutations/bi-allelic inactivation of TP53 lead to complete loss of
other words, a patient with LGD may progress from an unsuspected nuclear expression of the protein, termed the “null” pattern
HGD in the same site or elsewhere in the esophagus. In such a case, (Fig. 1C, D). Light and patchy staining using p53 IHC reflects
HGD/EAC detected on follow-up may represent either true normal physiologic activity of the protein to maintain cell health
progression or a delayed/missed diagnosis. and is the pattern of cells that are TP53 wild-type (Fig. 1E, F).

Modern Pathology (2022) 35:1000 – 1012

 W.-T. Choi et al.
1002

Fig. 1 Different p53 expression patterns. A, B Strong and diffuse p53 overexpression is seen in a case of HGD. C, D This example shows
complete absence of p53 staining (null pattern). The base of the squamous epithelium shows normal positive staining (internal control).
E, F Wild-type pattern of p53 staining in NDBE shows scattered, faintly positive nuclei.

However, in one study of p53 staining, aberrant expression was resulted in kappa scores of 0.3, an unsurprising result since
detected in ~10% of cases regarded as non-dysplastic, ~40% of grouping cases into 2 categories versus 4 produces greater
LGD, ~85% of HGD, and all of EACs39. Strong nuclear staining observer variability ab initio41. In fact, when the authors grouped
aligns with TP53 mutations but can still be detected in cases of the morphologic interpretation into only two categories as they
LGD lacking TP53 mutations. Bian et al. reported that although had done for p53, namely “definite dysplasia” versus “no
95% of cases interpreted as LGD had p53 expression on IHC, TP53 dysplasia” on H&E, they achieved a comparable kappa score of
mutations were only detected in about a third40. 0.55 for morphology alone, diminishing their conclusions con-
Pathologists in many institutions, particularly in the UK and cerning p53 considerably. Nonetheless, the results of many studies
Europe, have advocated for the use of universal p53 IHC in BE have supported the use of p53 IHC as a marker of the likelihood of
cases to detect dysplasia that might be otherwise overlooked, to progression to HGD/EAC in patients whose biopsies show H&E
the point that the British Society of Gastroenterologists endorsed findings of negative for dysplasia, IND, or LGD. The latter studies
adding it reflexively in routine practice99. The recommendation are summarized by Srivastava et al.38.
seemed to reflect studies directed at predicting progression of More recently, Redston et al. studied “progressors” versus “non-
NDBE to HGD/EAC rather than establishing an initial diagnosis. In progressors” gleaned from a large commercial laboratory
one study, scoring p53 immunostaining as “significant” in the system42. The authors used a retrospective set of over 500 BE
presence of strong or absent staining versus “not significant” patients with or without known progression from negative for
resulted in kappa scores on the order of 0.6 (strong reproduci- dysplasia, IND, or LGD to HGD/EAC. To establish their IHC scoring
bility), whereas scoring morphologic features as “negative for system (Table 1), the authors obtained DNA for sequencing from
dysplasia” versus “IND” versus “LGD” versus “HGD” (4 categories) 92 BE samples derived from 28 progressors and 6 non-

Modern Pathology (2022) 35:1000 – 1012

 W.-T. Choi et al.
1003
Table 1. 42
Scoring method for p53 used by Redston et al. .
p53 expression pattern Criteria
Normal (wild-type) 2–3+ nuclear staining in ≤20% of epithelial cells in all individual gland bases or glandular profile, and all
contiguous stretches of surface epithelium.
Equivocal pattern (subsumed under 2–3+ nuclear staining in 21–50% of epithelial cells in at least one gland base or glandular profile, or
normal) within a contiguous focus of at least 20 surface epithelial cells.
Point mutation pattern 2–3+ nuclear staining in >50% of epithelial cells in at least one gland base or glandular profile, or within
a contiguous focus of at least 20 surface epithelial cells.
Absent (null) pattern Total absence of staining in all epithelial cells of at least one gland base or glandular profile. This
assumes presence of internal control staining (nuclear positivity of deep glands, base of squamous
epithelium, immune cells, or stromal cells).
Cytoplasmic pattern Total absence of staining in all epithelial cells of at least one gland base or glandular profile. This
assumes presence of internal control staining (nuclear positivity of deep glands, base of squamous
epithelium, immune cells, or stromal cells).

progressors. TP53 mutations were identified from 50 of the metaplasia. No cost analysis was provided by Redston et al.42,
samples, specifically from 21 patients who progressed and 3 who although pathologists might be motivated by payments to add
did not. In ~90% of cases, the TP53 mutational status correlated p53 staining to all BE samples that lack dysplasia or show LGD.
with p53 immunostaining results. The authors validated their Overinterpretation of normal staining, however, might result in
p53 staining criteria using 50 NDBE and 50 HGD biopsies. They unnecessary surveillance or ablation procedures.
found abnormal p53 staining in 4% of NDBE and 96% of HGD, Most laboratories have used p53 immunostaining for years in
thereby confirming their scoring criteria. In the testing phase, evaluation of samples from several organ systems. In esophageal
amongst 646 NDBE patients, 20 progressed to LGD, and 10 to biopsies, however, in expert hands, p53 staining is not necessary
HGD/EAC. Abnormal p53 immunostaining was detected in half of to diagnose HGD in BE, which is itself an excellent marker for high
the progressors, resulting in good specificity but poor sensitivity. risk for progression to EAC104, and many gastroenterologists
Essentially, amongst 646 NDBE patients, adding p53 staining request second opinions for diagnoses of LGD and HGD since
offered additional information for only 15, and arguably, progres- either is currently an indication for ablation of the affected
sion to LGD is not truly progression. The authors suggested that segment4,99. The updated 2021 Medicare reimbursement fee for
patients with abnormal p53 expression in NDBE have comparable the code for outside consultation (88321) is $102.49, which is
rates of progression to those who have LGD. They further slightly cheaper than a p53 stain. Adding p53 may have some
suggested annual endoscopy for such persons. However, the value in assessing LGD or adding diagnostic precision for cases
study was limited by the lack of uniformity of the screening and regarded as IND103. However, using positive p53 immunostaining
surveillance methods of the gastroenterologists submitting their to justify endoscopic therapy in IND or LGD patients, when the
materials to the commercial laboratory. Also, it is worth noting concordance between IHC and mutation analysis is less than
that although the risk of progression to HGD/EAC is reported to perfect (~90%), may mean overtreatment in ~10% of patients.
decline with an increased number of endoscopies showing
NDBE100,101, most studies on ancillary tests, including those of
p53, do not clarify the number of negative endoscopies prior to WATS3D AS A DIAGNOSTIC BIOMARKER
the development of HGD/EAC, complicating the interpretation of WATS3D or Wide Area Transepithelial Sampling with Three-
their outcome data. Dimensional Computer-Assisted Analysis (CDx Diagnostics, Suf-
In a 2018 study, Ten Kate et al. reported that simply refining fern, NY) is an adjunct test to targeted and random four-quadrant
histologic criteria for diagnosis of LGD identified BE patients likely esophageal biopsies using three-dimensional computer-assisted
to progress102. Similarly, other refined histologic criteria allowed tissue analysis. As discussed previously, the current screening and
another group that included one of us (EAM) to essentially surveillance guidelines for BE and associated dysplasia require
eliminate the IND category with excellent prediction of out- sampling of any visible mucosal abnormality followed by systemic
come103. Ten Kate et al. also used p53 staining alone and achieved random four-quadrant forceps biopsies obtained at 1–2 cm
similar results to those afforded by use of H&E alone, with some intervals (Seattle protocol). However, this recommended protocol
synergy for the two combined but probably not enough to is time-consuming, labor intensive, and subject to sampling error.
support reflex testing102. Years ago, two of us (GYL and EAM) were As such, the rationale for using WATS3D is to overcome these
part of a group that also achieved excellent prediction of outcome inherent problems associated with extensive blind sampling43. In
using H&E alone despite imperfect interobserver variability31,104. WATS3D, abrasive brushes are used to circumferentially sample the
We would also point out that the Kaplan–Meier curves for esophageal mucosa. The sampling consists of individual cells as
progression of NDBE with and without abnormal p53 staining well as mucosal strips reported to measure up to 150 μm in
from the study by Redston et al. do not differ dramatically because thickness. The material is first analyzed by an imaging system
so few patients without histologic dysplasia progress regardless of using a neural network optimized for evaluation of the esophageal
p53 immunostaining status42. mucosa. The computer system scans, analyzes, and integrates up
Incorporating reflex IHC for p53 is not terribly expensive in the to fifty 3-μm optical slices. Ultimately, the system builds three-
individual patient, and reimbursement is readily obtained. The dimensional images of esophageal glands, and flags goblet cells
2021 Medicare fee schedule listed a 2021 figure of $99.82 for a and dysplastic cells that are displayed for confirmation by a
global code of 88342 (immunostaining; technical only $67.41) and pathologist (Fig. 2). The 2019 guidelines of the American Society
modified it to $106.07 (technical only $70.82), whereas H&E global for Gastrointestinal Endoscopy conditionally endorsed the use of
code (88305) affords $66.76 (technical only $32.06), which was WATS3D based on low quality evidence for screening and
updated to $71.52 (technical only $33.84). This means that adding surveillance of BE, in addition to Seattle protocol biopsy sampling
a p53 stain increases the cost per biopsy by two and a half fold. for patients with known or suspected BE44.
This might be prohibitively expensive if p53 staining is added to Several studies have demonstrated a significant increase in the
every single esophageal biopsy demonstrating intestinal detection rates of BE and dysplasia when WATS3D was used

Modern Pathology (2022) 35:1000 – 1012

 W.-T. Choi et al.
1004

Fig. 2 Representative images of WATS3D. The images show (A) NDBE, (B) LGD, (C) HGD, and (D) EAC. The images were reproduced with
permission from Elsevier47.

adjunctively to the combination of both targeted and random that individuals without dysplasia on WATS3D had a very low risk
four-quadrant biopsies. For instance, in a large multicenter of progression to HGD/EAC (0.08% per patient-year), while those
prospective study of 12,899 patients undergoing BE screening with LGD had a higher rate of 5.79% per patient-year50. However,
and surveillance and evaluated by 58 community endoscopists, as noted by the authors, no comparison could be made to the
WATS3D was reported to detect additional 213 patients with progression rate of LGD detected by microscopy alone. Interest-
dysplasia (versus 88 cases detected on biopsy alone), increasing ingly, the authors also evaluated the category of crypt dysplasia
the overall detection of dysplasia by 242%46. WATS3D also and reported its risk of progression (1.42% per patient-year) higher
identified 2570 additional BE cases (versus 1684 BE cases by the than those with no dysplasia but lower than those with LGD. The
combination of targeted and random biopsies alone), increasing heterogeneous nature of this diagnostic category likely explains
the rate of BE detection from 13.1 to 33%. However, among the the results.
purported 213 “new” dysplasia cases, 128 (60%) were in fact There are limitations with WATS3D. First, although a study
classified as IND rather than as dysplasia by WATS3D, significantly examining the interobserver agreement among pathologists using
reducing the reported increased detection rate of dysplasia. WATS3D found substantial agreement for LGD, HGD, and no
Furthermore, the increased detection rate of BE was based on the dysplasia48, the diagnostic criteria used in WATS3D have not been
diagnosis of intestinal metaplasia, but the possibility that the independently tested. Also, the data used to construct the
cardia was sampled could not be excluded, further weakening the algorithm that creates three-dimensional images of esophageal
validity of the results. glands and adequacy criteria used for this test are not well
Only a small series involving 160 BE patients tackled the issue of delineated. Furthermore, regenerative epithelial changes in deep
HGD/EAC detection by WATS3D 47. In a multicenter, prospective, glands could be easily misinterpreted as HGD on a single plane of
randomized trial of referred BE patients at 16 medical centers, analysis as in WATS3D, an issue that merits additional evaluation49.
Vennalaganti et al. reported that the addition of WATS3D to biopsy Another hindrance to the full validation of this technology is that
sampling yielded additional 23 cases of HGD/EAC. Among these this commercial test is interpreted by a limited group of
23 patients, 11 were classified by biopsy as NDBE and 12 as LGD/ pathologists. Independent reproduction and evaluation of WATS3D
IND. However, the vast majority of these patients (91.3%) had diagnoses in academic settings with expert reviews of all forceps
been previously diagnosed with HGD/EAC on prior biopsies, and biopsies by specialized GI pathologists (by all means not infallible)
most (78%) were confirmed to have HGD/EAC on follow-up would go a long way to address some of these criticisms. Finally,
biopsies. Also, this study was performed in a high-risk BE as noted above, whether the progression rate of dysplasia
population at referral centers, and thus it is not representative detected by WATS3D differs from that of dysplasia identified by
of community GI practices and of the BE population at large. forceps biopsy and microscopy alone remains to be established.
Nonetheless, this series is important in that it demonstrates an The issue of additional cost of performing WATS3D has not been
increased detection rate of high-grade lesions by WATS3D, and the extensively evaluated. The cost of WATS3D has been reported to
biopsy diagnoses had been confirmed by GI pathologists. be in the range of $700–800 by our endoscopist colleagues, but
Long-term outcome studies of dysplasia diagnosed solely on whether using this commercial test as an adjunct to traditional
WATS3D are limited. In a study of over 4000 BE patients who had endoscopic surveillance is cost effective in the long-term manage-
two WATS3D separated by ≥12 months, Shaheen et al. reported ment of BE patients remains to be thoroughly evaluated. Also, the

Modern Pathology (2022) 35:1000 – 1012

 W.-T. Choi et al.
1005

Fig. 3 Representative images of TissueCypher. A–D show a NDBE biopsy from a 66-year-old man with 8-cm BE segment and who was
diagnosed with EAC at a surveillance endoscopy 3.9 years later (progressor). TissueCypher scored this specimen high risk. E–H show a NDBE
biopsy from a 69-year-old man with 11-cm BE segment with 5.6 years surveillance data showing no progression (non-progressor).
TissueCypher scored this specimen low risk. A and E show p16-green, AMACR-red, and p53-yellow; B and F show CD68-green and COX-2-red;
C and G show HIF-1α-green and CD45RO-red; D and H show HER2-green and CK20-red. Nuclei labeled by Hoechst are shown in blue in all
panels.

potential value of WATS3D in the era of ever improving advanced to identify nuclei as discrete objects in tissue. It also allows the
endoscopic imaging techniques has not been examined. As software to assess nuclear area, solidity, and DNA content. Some
endoscopists improve their ability to detect ever more subtle of the markers are combined on the same slide51,52. The slides are
lesions previously described as ‘invisible’, it may lessen the need then used to perform image analysis with an image analysis
for broad blind sampling, such as WATS3D. algorithm. The image analysis algorithm quantifies 15 different
“image features” (Table 2). The quantified image features are then
combined into a risk score. Samples are still reviewed in the typical
TISSUECYPHER AS A DIAGNOSTIC AND RISK STRATIFICATION manner (routine diagnosis by local pathologists) and then sections
BIOMARKER are prepared and subjected to the TissueCypher staining and
The objective of TissueCypher is to evaluate samples from BE patients algorithm.
diagnosed as negative for dysplasia, IND, or LGD on routine histologic This method offers the advantage of using a variety of markers
evaluation to identify those patients most likely to progress to HGD/ with a consistent interpretation, thus eliminating interobserver
EAC so that intensified screening or ablation can be offered to them. variability, although this does not necessarily mean an accurate
Similarly, the technique is intended to identify patients who are interpretation. Using the company’s platform, a risk score for
unlikely to progress such that their surveillance can be reduced. progression is stratified as low, intermediate, or high, but there is
TissueCypher uses immunofluorescent labeling of sections from some advantage to combining the intermediate and high risk
formalin-fixed paraffin-embedded (FFPE) samples for p16, AMACR, scores. Although similar data are reported in all studies from the
p53, HER2, CK20, CD68, COX-2, HIF-1α, and CD45RO, together with TissueCypher team51,52,54–56, the initial and some recent studies
Hoechst staining dye (Fig. 3)51–58. Hoechst dye allows fluorescent were performed in Europe, and in 2020, a US-based study from
detection of DNA105, thereby permitting image analysis software two institutions was added53. The latter study was a case-control

Modern Pathology (2022) 35:1000 – 1012

 W.-T. Choi et al.
1006
Table 2. Core features assessed by TissueCypher52.
Marker Image analysis feature
p53 p53 nuclear sum intensity
p53 p53 nuclear mean intensity
HER2/neu and CK20 Ratio of mean HER2/neu intensity:mean CK20 intensity in nuclear clustersb
HER2/neu and CK20 Ratio of 95th quantile HER2/neu intensity:95th quantile CK20 intensity in nucler clusters
COX-2 and CD68 Co-expression cellular COX-2 mean intensity and cellular CD68 mean intensity
p53 p53 mean intensity in nuclear clusters
p53, p16, and Hoechsta/nuclear morphology Nuclear solidity in p53+ p16– cells
CD45RO CD45RO plasma membrane sum intensity
AMACR AMACR microenvironment standard deviation
COX-2 COX-2 texture in cytoplasmc
HIF-1α HIF-1α microenvironment cell mean intensity
HIF-1α HIF-1α microenvironment cell moment (product of mean and standard deviation)
p16 p16 cytoplasm mean intensity
p53, p16, and Hoechst/nuclear morphology Nuclear area in p53+ p16– cells
Hoechst/nuclear morphology Hoechst nuclear 95th quantile intensity
a
Hoechst is used to stain DNA, which enables image analysis software to segment nuclei as individual objects in tissue images, and to measure nuclear area,
solidity, and DNA content.
b
Nuclear clusters are detected by the image analysis software, and biomarkers are measured within the image regions containing nuclear clusters.
c
Contrast textural feature is extracted from a co-occurrence matrix and is a measure of the COX-2 intensity contrast between a pixel and its neighbor over the
whole tissue image.

study from patients with biopsy diagnoses of negative for claimed that it would be cost-effective after 5 years by reducing the
dysplasia (n = 227), IND (n = 23), and LGD (n = 18). The samples number of patients requiring surveillance and reducing EAC-
were from 58 patients who progressed to HGD/EAC (median time associated deaths57. It does not change initial evaluation of patients
to progression of 2.7 years; 7/58 progressed after 5 years), and by routine histology. It has been assigned a CPT code by Medicare
from 210 patients who did not progress (median surveillance time (0108U). However, because of the high “up front” costs of
of 7 years). In this study, the prevalence-adjusted proportions of TissueCypher, most insurers do not cover the testing. Also, the
patients scoring low, intermediate, and high risk using the roughly similar performance characteristics of TissueCypher to
TissueCypher method were 84.2%, 9.4%, and 6.4%, respectively. histologic evaluation may not justify changing surveillance intervals
The sensitivity and specificity of the test at 5 years for the 3-tier based on the results, and thus the suggested cost benefit may not
TissueCypher classification (low, intermediate, and high risk) were materialize, especially given its high cost. As noted above, based on
29% and 86%, respectively, and 40% and 86%, respectively, for the the company’s data, a diagnosis of LGD by an expert pathologist
2-tier classification (low and intermediate/high risk combined). By offers more specificity than the test, and obtaining an expert
comparison, the sensitivity and specificity of an expert diagnosis opinion is certainly substantially cheaper. Regardless, the test is
of LGD were 19% and 88%, respectively, and the sensitivity and consistent, not subject to human observer variation, and outper-
specificity of the initial community diagnoses of LGD (i.e., forms pathologists in identifying patients without dysplasia who are
diagnosis recorded in the health records) were 26% and 66%, likely to progress to HGD/EAC.
respectively. Of 51 patients who progressed within 5 years,
14 scored high risk, 6 scored intermediate risk, and 31 scored low
risk. Among 210 patients who did not progress, 13 scored high MUTATIONAL LOAD (BARREGEN) AS A DIAGNOSTIC AND RISK
risk, 18 scored intermediate risk, and 179 scored low risk. Using STRATIFICATION BIOMARKER
the TissueCypher test, the prevalence-adjusted positive predictive Mutational load (ML) analysis provides a measure of cumulative
value (PPV) was 23%; i.e., 23% of patients who score high risk genetic aberrations and instability at 10 key genomic loci by
would progress to HGD/EAC within 5 years. The prevalence- assessing DNA damage around tumor suppressor genes asso-
adjusted negative predictive value (NPV) was 96.4%. The risk ciated with progression to HGD/EAC59–62. It can be assessed using
prediction test also showed improved risk stratification when a commercially available test (BarreGEN, Interpace Diagnostics,
compared to p53 alone using the automated scoring. Pittsburgh, PA), with its main objective being to detect dysplasia
Overall, expert pathologists’ diagnosis of LGD outperformed and risk stratify BE patients. To perform this assay, H&E-stained
TissueCypher in specificity and PPV, but TissueCypher was more slides are first evaluated to identify relevant histologic targets (e.g.,
sensitive. However, this was not the case for samples from patients LGD) that are micro-dissected from 1 to 3 unstained FFPE sections
with no dysplasia. Patients without dysplasia as confirmed by expert (4 μm in thickness)60–62,106. Greater than 90% of each micro-
pathologists who scored high risk were at about 5-fold increased risk dissected target should contain epithelial cells, from which
of progression as compared to patients without dysplasia who purified DNA is prepared. Polymerase chain reaction (PCR) and
scored low risk using TissueCypher. The adjusted PPV for the test in quantitative capillary electrophoresis methods are performed on
expert pathologist-confirmed NDBE was 26%, indicating that 26% of all micro-dissected areas. ML specifically assesses the presence
patients without dysplasia but with a high risk score using and extent (clonality) of loss of heterozygosity (LOH) and new
TissueCypher will progress within 5 years, a rate similar to that alleles consistent with microsatellite instability (MSI) for each
associated with an expert diagnosis of LGD. micro-dissected target. The following 10 loci are examined, with
TissueCypher is a send-out test (Cernostics, Pittsburgh, PA) and associated tumor suppressor genes in parentheses: 1p (CMM1,
costs about $5,000. A cost analysis sponsored by the company L-myc), 3p (VHL, HoGG1), 5q (MCC, APC), 9p (CDKN2A), 10q (PTEN,

Modern Pathology (2022) 35:1000 – 1012

 W.-T. Choi et al.
1007
(ML = 2.2) than non-progressors (ML = 0.4) (p < 0.001).61 No
progressor had a ML of 0 at baseline compared with 54% of
non-progressors. Sensitivity was 100% at ML ≥ 0.5, and specificity
was 96% at ML ≥ 1.5. Similarly, in a retrospective study of 28 IND
patients, patients who progressed to HGD had higher levels of
genomic instability (ML ≥ 1.5)62. At this threshold, the risk of
progression to HGD was 33% (versus 0% in those with an ML < 1.5;
p = 0.005), with a sensitivity of 100% and a specificity of 85%.
Overall, these results indicate that genomic alterations as
measured by ML often predate the development of HGD/EAC,
potentially allowing ML to be a useful biomarker for predicting
disease progression. In addition, Ellsworth et al. demonstrated a
signficant correlation of higher ML with increasingly severe
histologic grade of BE-associated lesions: ML = 1.1 for IND, 2.2
for LGD, and 3.3 for HGD (p < 0.001)59. These results suggest that
ML may serve as an adjunctive test in patients with equivocal
histology.
The biggest appeal of ML assessment is that it allows a direct
correlation with morphology and provides an objective quantitative
measure of the presence and extent of molecular alterations
associated with development of dysplasia and EAC, eliminating
human interobserver variability. However, similar to other PCR-based
tests, ML assessment may be hampered by insufficient amounts
and/or poor quality of DNA as is often the case in FFPE mucosal
biopsies. Also, micro-dissection of tiny histologic targets seen on
H&E slides is susceptible to sampling error. Another limitation is that
purified DNA rather than crude lysate should be used whenever
possible, as ML signal in crude lysate could be “muted”106. In fact,
using crude lysate, there was no difference in mean ML between
progressors (ML = 0.73) and non-progressors (ML = 0.74) (p = 0.93)
in a nested case-control study (involving 48 progressors and 101
non-progressors), failing to validate the previous finding that ML
could be useful in risk stratifying BE patients. Finally, as noted above,
BarreGEN is not fully validated for commercial use at this time, and it
is unclear when this test will be available for clinical use.

FISH AS A DIAGNOSTIC AND RISK STRATIFICATION

BIOMARKER
FISH (fluorescent in situ hybridization) is a technique that utilizes
fluorescently labeled DNA probes to detect chromosomal abnorm-
alities. To explore its potential utility in the diagnosis and risk
stratification of dysplasia in BE patients, several studies, all
conducted at Mayo Clinic, utilized a 4 locus-specific probe set
Fig. 4 Representative images of FISH signal patterns. A Normal targeting 8q24 (MYC), 9p21 (CDKN2A; alias P16), 17q12 (ERBB2; alias
FISH result shows 2 signals of each of the 4 probes. B Homozygous Her-2/neu), and 20q13 (ZNF217)7,63–67. Each cell is categorized as
loss of 9p21 shows no red signal. C Polysomic FISH result shows ≥ either normal (i.e., having 2 signals per probe) or abnormal (i.e.,
3 signals of ≥2 probes. The FISH probes are labeled with Spectrum having more or less than 2 signals per probe) (Fig. 4). Detectable
Aqua (8q24), Red (9p21), Green (17q12), or Gold (20q13) fluor- chromosomal alterations include polysomy (≥3 signals for ≥2 loci),
ophores. The images were reproduced with permission from single locus gain (3–9 signals of a single locus and two signals of
Elsevier63. other loci), amplification of a single locus (≥10 signals of a single
locus and two signals of other loci), and single locus loss (0–1 signal
MXI1), 17p (TP53), 17q (RNF43, NME1), 18q (SMAD4, DCC), 21q of a single locus and two signals of other loci). The test can be
(TFF1, PSEN2), and 22q (NF2). LOH is categorized as either high (> performed on either FFPE tissue7,66 or endoscopic brushing
75% of the DNA has LOH) or low clonality (50–75% of the DNA has specimens63–65,67.
LOH) and assigned values of 1 and 0.5, respectively. The value of In a FISH analysis of a range of histologic lesions from 10
the first MSI at a genomic locus is 0.75, and each additional MSI is esophagectomy specimens from BE patients, polysomy was found
assigned a value of 0.5. The highest weighted value at each locus to be more prevalent in HGD (88%) and EAC (100%) than in NDBE
is determined based on the values for low and high clonality LOH (<10%) and LGD (57%) (p < 0.001), whereas single locus gain was
and MSI at that locus (e.g., the weighted value of high clonality most commonly observed in LGD (28% versus 12% of NDBE versus
LOH is 1, which is the highest possible weighted value at each of 8% of HGD)7. Also, in a recent multicenter study, if ≥10% of cells
the 10 loci). The sum of all weighted values for all 10 genomic loci had polysomy in the specimen, FISH was able to differentiate
is defined as the ML for that micro-dissected target (range: 0–10). between HGD/EAC and the remaining histologic diagnoses with a
In a case-control study involving 69 BE patients (including 23 sensitivity of 80% and a specificity of 88%66. Furthermore, in a
progressors and 46 non-progressors), ML assessment was able to retrospective analysis of 245 BE patients with a history of HGD,
risk stratify patients with NDBE or LGD at baseline with respect to using a cutoff of more than 4 of 100 cells demonstrating
progression to HGD/EAC within a mean follow-up time of 4 polysomy, the risk of EAC was significantly higher within 2 years
years61. A mean ML score was significantly higher in progressors among patients with a polysomic FISH result (14.2%) compared

Modern Pathology (2022) 35:1000 – 1012

 W.-T. Choi et al.
1008
with those without a polysomic result (1.4%) (p < 0.001)65. In or IND could potentially enable clinicians to recommend
addition, Timmer et al. demonstrated that in BE patients with HGD endoscopic therapy, whereas continued surveillance may be an
or intramucosal adenocarcinoma (IMC) treated with ablation with acceptable approach in the setting of normal flow cytometric
or without preceding EMR, polysomy was associated with a lower results. Furthermore, Bowman et al. recently demonstrated that
probability of achieving complete eradication of HGD/IMC (HR = abnormal DNA content in baseline HGD/IMC can serve as a
0.57, p = 0.002) in a univariate analysis67. Given its high diagnostic predictive marker of persistent/recurrent neoplasia following
accuracy for identifying HGD/EAC and potential to detect HGD/ endoscopic therapy, with the univariate and multivariate HRs of
EAC on follow-up, polysomy in combination with histology may be 3.8 (p = 0.007) and 6.0 (p = 0.003), respectively76. This suggests
able to serve as a confirmatory marker of HGD/EAC and screening that the detection of DNA content abnormalities in baseline HGD/
tool to identify patients at highest risk for subsequent detection of IMC may help to identify high-risk BE patients who may benefit
HGD/EAC compared with those with non-polysomy. from alternative therapeutic strategies (e.g., different ablation
Despite these promising results, no reference laboratory currently technique, combined endoscopic modalities, or endoscopic
offers this test (apparently due to the lack of demand), although in submucosal dissection) as well as long-term follow-up with
the past it was available at Mayo Clinic and Neogenomics. However, shorter surveillance intervals following endoscopic therapy.
many academic centers and commercial laboratories routinely run There are some advantages of using DNA flow cytometry. First,
FISH, which can be completed in a few days, and the commercial DNA flow cytometry is an inexpensive send-out test ($350 at ARUP
availability of these probes (~$800 per probe, Abbott Molecular Inc., laboratories; CPT code: 88182) that can be completed within
Des Plaines, IL) allows these laboratories to validate and bring up the 2–3 days. Second, DNA flow cytometric markers of dysplasia or
same assay if needed. Also, FISH may be more sensitive than other progression (aneuploidy or elevated 4 N fraction) are usually absent
tests such as DNA flow cytometry (described below) by virtue of in NDBE72,75–77,109,110, and features potentially altering the histologic
having the low threshold for a positive polysomy result (e.g., 4 interpretation (i.e., increased acute inflammation or ulceration) do
polysomic cells)63,65,67. Yet, genetic alterations as detected by FISH not cause aneuploidy or elevated 4 N fraction, which can be very
are limited to specific gain/loss of genes targeted by the probe set, helpful in evaluating IND cases69,111. In fact, many genetic and
and thus other non-targeted chromosomal alterations would not be chromosomal abnormalities detected in BE (including 9p LOH [site
detected, potentially missing some high-risk individuals who could of CDKN2A], 17p LOH [site of TP53], and mutations of TP53 and
be identified by DNA flow cytometry. Furthermore, FISH often CDKN2A) tend to occur early and frequently throughout large areas
identifies cells with minimal DNA alterations, such as 9p21 (CDKN2A) of BE5–10,112–114, even before the first histologic sign of dysplasia,
loss, which often do not cause noticeable morphologic abnormal- limiting their utility as a diagnostic or prognostic marker of dysplasia
ities. Thus, a positive (especially non-polysomy) FISH result does not in BE patients.
always indicate the presence of dysplasia. In conclusion, as the current surveillance methods based on the
histologic diagnosis and classification of dysplasia imperfectly
assess the risk of BE patients, especially those with IND or NDBE
DNA CONTENT ABNORMALITIES AS DETECTED BY DNA FLOW histology, there is an increasing demand for ancillary tests to aid in
CYTOMETRY AS A DIAGNOSTIC AND RISK STRATIFICATION the diagnosis/grading of dysplasia and risk stratification of BE
BIOMARKER patients. In cases with equivocal histology, one may argue that a
Since the 1980s, a number of studies have consistently demon- repeat endoscopic examination with biopsies may provide the
strated the potential utility of DNA flow cytometry in the diagnosis answer without the need of an ancillary test. However, this
and risk stratification of dysplasia in BE patients68–77. Although its approach is likely to be more expensive than most ancillary tests.
availability has been limited to few medical centers due to In this regard, several biomarkers and assays, including p53 IHC,
perceived technical demands and use of fresh tissue in earlier WATS3D, TissueCypher, mutational load assessment (BarreGen),
studies68–71, subsequent studies have successfully employed FFPE FISH, and DNA content abnormalities as detected by DNA flow
tissue for DNA flow cytometric analysis to generate high-quality cytometry have been demonstrated as ways to support a
DNA content histograms, demonstrating the feasibility of this dysplasia diagnosis and aid in risk assessment for the develop-
methodology72–77. For optimal results, the computer program ment of HGD/EAC (Table 3). More importantly, many of these tests
Multicycle (De Novo software, Glendale, CA) should be used to are currently available in academic centers and commercial
analyze DNA content histograms68–71,76,77. The published con- laboratories, and often utilize FFPE, obviating the need to obtain
sensus guidelines for clinical DNA flow cytometry should be separate samples. Although none of these tools are widely used in
followed107,108. Most epithelial cells are normally in the G0/G1 practice, there is an increased interest among gastroenterologists
phase of the cell cycle and have diploid (2 N) DNA content, while to pursue ancillary tests in BE surveillance biopsies, as they have
less than 6% of cells have tetraploid (4 N) DNA content (G2) shown promising results in identifying early neoplasia and could
(Fig. 5A, B). Aneuploidy is defined as an extra G0/G1 peak that is potentially serve as adjuncts to histologic evaluation. By providing
bimodally separated from the normal diploid G0/G1 peak (Fig. 5C, information that cannot be assessed by morphology alone,
D). The presence of a G2/tetraploid (4 N) fraction greater than 6% especially if the cost is reasonable (i.e., cheaper than repeat
(with DNA index of 1.9–2.1) is also classified as abnormal due to its endoscopy with additional pathologic evaluation), these tests may
strong association with neoplasia (Fig. 5E, F)5,69,71,76,77. become attractive tools, especially for patients with inconsistent
A recent retrospective study analyzed 80 FFPE BE samples with diagnoses, IND, or LGD histology. Like many molecular tests (e.g.,
HGD, 38 LGD, 21 IND, and 14 NDBE and reported that the next-generation sequencing) currently used in the diagnosis and
frequency of DNA content abnormalities (aneuploidy or elevated management of many diseases, incorporating these tools in the
4 N fraction) increases with increasing histologic grade of management of BE patients, in conjunction with histologic
dysplasia: 0% of NDBE, 9.5% of IND, 21.1% of LGD, and 95% of evaluation, may allow for more precise surveillance and/or earlier
HGD77. As a diagnostic marker of HGD, the estimated sensitivity treatment in patients at higher risk of progression, while avoiding
and specificity of abnormal DNA content were 95% and 85%, unnecessary interventions or surveillance in those at lower risk.
respectively. Interestingly, DNA flow cytometry also identified a Prospective studies on these biomarkers (including assessment of
subset of LGD and IND patients who are at higher risk for their potential utility in reducing mortality from EAC) as well as
subsequent detection of HGD/EAC, with the univariate hazard cost-effective analysis compared with the current surveillance
ratios (HRs) of 7.0 and 20.0, respectively (p < 0.001)77. Considering methods are singularly missing. Until these comprehensive data
that endoscopic therapy is increasingly being recommended for exist, it is impossible to fully evaluate their potential impact and
LGD patients26, abnormal flow cytometric results at baseline LGD better tailor their potential roles in the care of BE patients.

Modern Pathology (2022) 35:1000 – 1012

 W.-T. Choi et al.
1009

Fig. 5 DNA content histograms of NDBE and HGD. A, B NDBE is characterized by the presence of intestinal metaplasia, but there is no
dysplasia. The DNA histogram shows a normal diploid population (green). C, D HGD is characterized by severe cytologic atypia with enlarged,
hyperchromatic, rounder nuclei. The DNA histogram demonstrates a discrete aneuploid peak (red) that is bimodally distinguishable from the
normal diploid peak (green). E, F Another example of HGD shows atypical glands lined by highly pleomorphic cells with enlarged nuclei.
There is an elevated 4 N fraction greater than 6% (with DNA index of 1.9–2.1) in the DNA histogram. No distinct aneuploid population is
present.

Modern Pathology (2022) 35:1000 – 1012

 W.-T. Choi et al.
1010
REFERENCES

Send-out (~$5,000)
No additional cost
~$100 per biopsy

Send-out (~$350)
~$800 per probe
1. Spechler, S. J., Sharma, P., Souza, R. F., Inadomi, J. M., Shaheen, N. J. & American

Send-out (cost
Gastroenterological Association. American Gastroenterological Association

Send-out (~
technical review on the management of Barrett’s esophagus. Gastroenterology

$700–800)

unknown)
140, e18–e52 (2011).
2. Wang, K. K., Sampliner, R. E. & Practice Parameters Committee of the American
Cost

College of Gastroenterology. Updated guidelines 2008 for the diagnosis, sur-

veillance and therapy of Barrett’s esophagus. Am. J. Gastroenterol 103, 788–797
(2008).

May be hampered by insufficient amounts and/

Testing algorithm unknown; not independently

High cost; performance characteristics roughly

3. Naini, B. V., Souza, R. F. & Odze, R. D. Barrett’s esophagus: a comprehensive and

or poor quality of DNA; sampling error; not

contemporary review for pathologists. Am. J. Surg. Pathol. 40, e45–e66 (2016).
Variable interpretation of staining results

4. Shaheen, N. J., Falk, G. W., Iyer, P. G., Gerson, L. B. & American College of
Sampling error; interobserver variability

Gastroenterology. ACG clinical guideline: diagnosis and management of Bar-

Not widely available; sampling error

rett’s esophagus. Am. J. Gastroenterol. 111, 30–50 (2016).

fully validated for commercial use

5. Galipeau, P. C. et al. 17p (p53) allelic losses, 4N (G2/tetraploid) populations, and
similar to histologic evaluation

progression to aneuploidy in Barrett’s esophagus. Proc. Natl. Acad. Sci. USA 93,
Not commercially available 7081–7084 (1996).
6. Galipeau, P. C., Prevo, L. J., Sanchez, C. A., Longton, G. M. & Reid, B. J. Clonal
expansion and loss of heterozygosity at chromosomes 9p and 17p in pre-
malignant esophageal (Barrett’s) tissue. J. Natl. Cancer Inst. 91, 2087–2095
(1999).
7. Brankley, S. M. et al. Fluorescence in situ hybridization mapping of esopha-
gectomy specimens from patients with Barrett’s esophagus with high-grade
Limitation

dysplasia or adenocarcinoma. Hum. Pathol 43, 172–179 (2012).

validated

8. Ross-Innes, C. S. et al. Whole-genome sequencing provides new insights into the

clonal architecture of Barrett’s esophagus and esophageal adenocarcinoma.
Nat. Genet. 47, 1038–1046 (2015).
9. Varghese, S. et al. Analysis of dysplasia in patients with Barrett’s esophagus
based on expression pattern of 90 genes. Gastroenterology 149, 1511–1518
May increase the rate of dysplasia or BE

Provides consistent interpretation of a

(2015).
10. Walch, A. K. et al. Chromosomal imbalances in Barrett’s adenocarcinoma and the
High sensitivity and specificity as a

High sensitivity and specificity as a

variety of biomarkers, eliminating

molecular alterations, eliminating

Provides quantitative measure of

metaplasia-dysplasia-carcinoma sequence. Am. J. Pathol 156, 555–566 (2000).

reimbursement readily obtained
Widely available; not expensive;

11. Pouw, R. E. et al. Efficacy of radiofrequency ablation combined with endoscopic

Established clinical biomarker

resection for Barrett’s esophagus with early neoplasia. Clin. Gastroenterol

Hepatol 8, 23–29 (2010).
interobserver variability

interobserver variability

12. Haidry, R. J. et al. Radiofrequency ablation and endoscopic mucosal resection for
dysplastic Barrett’s esophagus and early esophageal adenocarcinoma: out-
comes of the UK National Halo RFA Registry. Gastroenterology 145, 87–95
marker of HGD

marker of HGD

(2013).
13. Rouphael, C., Anil Kumar, M., Sanaka, M. R. & Thota, P. N. Indications, contra-
detection

indications and limitations of endoscopic therapy for Barrett’s esophagus and

Strength

early esophageal adenocarcinoma. Ther. Adv. Gastroenterol 13,

1756284820924209 (2020).
14. Sharma, P., Shaheen, N. J., Katzka, D. & Bergman, J. AGA clinical practice update
on endoscopic treatment of Barrett’s esophagus with dysplasia and/or early
cancer: expert review. Gastroenterology 158, 760–769 (2020).
FFPE or brushing

15. Standards of Practice Committee, Wani, S. et al. Endoscopic eradication therapy

for patients with Barrett’s esophagus-associated dysplasia and intramucosal
fresh tissue

cancer. Gastrointest Endosc 87, 907–931.e9 (2018).

Specimen

specimen
specimen
Brushing

16. Komanduri, S., Muthusamy, V. R. & Wani, S. Controversies in endoscopic eradi-

FFPE or

cation therapy for Barrett’s esophagus. Gastroenterology 154, 1861–1875.e1

FFPE
FFPE

FFPE
FFPE

(2018).
17. Shaheen, N. J. et al. Radiofrequency ablation in Barrett’s esophagus with dys-
plasia. N. Engl. J. Med. 60, 2277–2288 (2009).
18. Anders, M. et al. Long-term recurrence of neoplasia and Barrett’s epithelium
Summary of biomarkers or ancillary tests.

after complete endoscopic resection. Gut 63, 1535–1543 (2014).

19. Bulsiewicz, W. J. et al. Safety and efficacy of endoscopic mucosal therapy with
Diagnostic marker

predictive marker

predictive marker

predictive marker

predictive marker
predictive marker
Predictive marker

radiofrequency ablation for patients with neoplastic Barrett’s esophagus. Clin.

Potential utility

Diagnostic and

Diagnostic and

Diagnostic and
Diagnostic and

Diagnostic and

Gastroenterol Hepatol 11, 636–642 (2013).

20. Ell, C. et al. Endoscopic mucosal resection of early cancer and high-grade dys-
plasia in Barrett’s esophagus. Gastroenterology 118, 670–677 (2000).
21. Peters, F. P. et al. Endoscopic treatment of high-grade dysplasia and early stage
cancer in Barrett’s esophagus. Gastrointest Endosc. 61, 506–514 (2005).
22. Fujii-Lau, L. L. et al. Recurrence of intestinal metaplasia and early neoplasia after
endoscopic eradication therapy for Barrett’s esophagus: a systematic review and
meta-analysis. Endosc. Int. Open 5, E430–E449 (2017).
23. Cotton, C. C. et al. Late recurrence of Barrett’s esophagus after complete era-
p53 immunostaining

DNA flow cytometry

dication of intestinal metaplasia is rare: final report from ablation in intestinal

metaplasia containing dysplasia trial. Gastroenterology 153, 681–688 (2017).
Biomarker or
ancillary test

24. Wani, S., Rubenstein, J. H., Vieth, M. & Bergman, J. Diagnosis and management of
TissueCypher

low-grade dysplasia in Barrett’s esophagus: clinical practice updates expert

Dysplasia

BarreGen

review from the clinical guidelines committee of the American Gastro-

WATS3D
Table 3.

enterological Association. Gastroenterology 151, 822–835 (2016).

FISH

25. di Pietro, M., Fitzgerald, R. C. & BSG Barrett’s guidelines working group. Revised
British Society of Gastroenterology recommendation on the diagnosis and

Modern Pathology (2022) 35:1000 – 1012

 W.-T. Choi et al.
1011
management of Barrett’s oesophagus with low-grade dysplasia. Gut 67, sampling with 3-dimensional computer-assisted analysis: a retrospective ana-
392–393 (2018). lysis. Gastrointest Endosc. https://doi.org/10.1016/j.gie.2021.09.014 (2021).
26. Phoa, K. N. et al. Radiofrequency ablation vs endoscopic surveillance for patients 51. Critchley-Thorne, R. J. et al. A tissue systems pathology assay for high-risk Bar-
with Barrett esophagus and low-grade dysplasia: a randomized clinical trial. rett’s esophagus. Cancer Epidemiol. Biomark Prev. 25, 958–968 (2016).
JAMA 311, 1209–1217 (2014). 52. Critchley-Thorne, R. J. et al. A tissue systems pathology test detects abnormal-
27. Hvid-Jensen, F., Pedersen, L., Drewes, A. M., Sørensen, H. T. & Funch-Jensen, P. ities associated with prevalent high-grade dysplasia and esophageal cancer in
Incidence of adenocarcinoma among patients with Barrett’s esophagus. N. Engl. Barrett’s esophagus. Cancer Epidemiol Biomark Prev. 26, 240–248 (2017).
J. Med. 365, 1375–1383 (2011). 53. Davison, J. M. et al. Independent blinded validation of a tissue systems
28. Wani, S. et al. Patients with nondysplastic Barrett’s esophagus have low risks for pathology test to predict progression in patients with Barrett’s esophagus. Am.
developing dysplasia or esophageal adenocarcinoma. Clin. Gastroenterol Hepatol J. Gastroenterol 115, 843–852 (2020).
9, 220–227 (2011). 54. Diehl, D. L., Khara, H. S., Akhtar, N. & Critchley-Thorne, R. J. TissueCypher Barrett’s
29. Desai, T. K. et al. The incidence of oesophageal adenocarcinoma in non- esophagus assay impacts clinical decisions in the management of patients with
dysplastic Barrett’s oesophagus: a meta-analysis. Gut 61, 970–976 (2012). Barrett’s esophagus. Endosc Int. Open 9, E348–E355 (2021).
30. Reid, B. J. et al. Observer variation in the diagnosis of dysplasia in Barrett’s 55. Frei, N. F. et al. Tissue systems pathology test objectively risk stratifies Barrett’s
esophagus. Hum. Pathol 19, 166–178 (1988). esophagus patients with low-grade dysplasia. Am. J. Gastroenterol 116, 675–682
31. Montgomery, E. et al. Reproducibility of the diagnosis of dysplasia in Barrett (2021).
esophagus: a reaffirmation. Hum. Pathol 32, 368–378 (2001). 56. Frei, N. F. et al. Independent validation of a tissue systems pathology assay to
32. Vennalaganti, P. et al. Discordance Among Pathologists in the United States and predict future progression in nondysplastic Barrett’s esophagus: a spatial-
Europe in Diagnosis of Low-grade Dysplasia for Patients With Barrett’s Eso- temporal analysis. Clin. Transl. Gastroenterol 11, e00244 (2020).
phagus. Gastroenterology 152, 564–570.e4 (2016). 57. Hao, J., Critchley-Thorne, R., Diehl, D. L. & Snyder, S. R. A cost-effectiveness
33. Moole, H. et al. Progression from low-grade dysplasia to malignancy in patients analysis of an adenocarcinoma risk prediction multi-biomarker assay for
with Barrett’s esophagus diagnosed by two or more pathologists. World J. patients with Barrett’s esophagus. Clinicoecon Outcomes Res. 11, 623–635 (2019).
Gastroenterol 22, 8831–8843 (2016). 58. Prichard, J. W. et al. TissueCypher(™): A systems biology approach to anatomic
34. Sonwalkar, S. A. et al. A study of indefinite for dysplasia in Barrett’s oesophagus: pathology. J. Pathol. Inf. 6, 48 (2015).
reproducibility of diagnosis, clinical outcomes and predicting progression with 59. Ellsworth, E., Jackson, S. A., Thakkar, S. J., Smith, D. M. Jr & Finkelstein, S. Correlation
AMACR (alpha-methylacyl-CoA-racemase). Histopathology 56, 900–907 (2010). of the presence and extent of loss of heterozygosity mutations with histological
35. Kaye, P. V. et al. Barrett’s dysplasia and the Vienna classification: reproducibility, classifications of Barrett’s esophagus. BMC Gastroenterol 12, 181 (2012).
prediction of progression and impact of consensus reporting and p53 immu- 60. Khara, H. S. et al. Assessment of mutational load in biopsy tissue provides
nohistochemistry. Histopathology 54, 699–712 (2009). additional information about genomic instability to histological classifications of
36. Corley, D. A. et al. Impact of endoscopic surveillance on mortality from Barrett’s Barrett’s esophagus. J. Gastrointest Cancer 45, 137–145 (2014).
esophagus-associated esophageal adenocarcinomas. Gastroenterology 145, 61. Eluri, S. et al. The presence of genetic mutations at key loci predicts progression
312–9.e1 (2013). to esophageal adenocarcinoma in Barrett’s esophagus. Am. J. Gastroenterol 110,
37. Visrodia, K. et al. Magnitude of missed esophageal adenocarcinoma after Bar- 828–834 (2015).
rett’s esophagus diagnosis: a systematic review and meta-analysis. Gastro- 62. Trindade, A. J. et al. Mutational load may predict risk of progression in patients
enterology 150, 599–607.e7 (2016). with Barrett’s oesophagus and indefinite for dysplasia: a pilot study. BMJ Open
38. Srivastava, A. et al. The use of ancillary stains in the diagnosis of Barrett eso- Gastroenterol 6, e000268 (2019).
phagus and Barrett esophagus-associated dysplasia: recommendations from the 63. Barr Fritcher, E. G. et al. A comparison of conventional cytology, DNA ploidy
Rodger C. Haggitt gastrointestinal pathology society. Am. J. Surg. Pathol. 41, analysis, and fluorescence in situ hybridization for the detection of dysplasia
e8–e21 (2017). and adenocarcinoma in patients with Barrett’s esophagus. Hum Pathol 39,
39. Kastelein, F. et al. Aberrant p53 protein expression is associated with an 1128–1135 (2008).
increased risk of neoplastic progression in patients with Barrett’s oesophagus. 64. Brankley, S. M. et al. The development of a fluorescence in situ hybridization
Gut 62, 1676–1683 (2013). assay for the detection of dysplasia and adenocarcinoma in Barrett’s esophagus.
40. Bian, Y. S., Osterheld, M. C., Bosman, F. T., Benhattar, J. & Fontolliet, C. p53 gene J. Mol. Diagn. 8, 260–267 (2006).
mutation and protein accumulation during neoplastic progression in Barrett’s 65. Brankley, S. M. et al. Fluorescence in situ hybridization identifies high risk Bar-
esophagus. Mod. Pathol 14, 397–403 (2001). rett’s patients likely to develop esophageal adenocarcinoma. Dis. Esophagus 29,
41. Kaye, P. V. et al. Dysplasia in Barrett’s oesophagus: p53 immunostaining is more 513–519 (2016).
reproducible than haematoxylin and eosin diagnosis and improves overall 66. Poneros, J. M. et al. A multicenter study of a fluorescence in situ hybridization
reliability, while grading is poorly reproducible. Histopathology 69, 431–440 probe set for diagnosing high-grade dysplasia and adenocarcinoma in Barrett’s
(2016). Esophagus. Dig. Dis. Sci. 62, 1216–1222 (2017).
42. Redston, M. et al. Abnormal TP53 predicts risk of progression in patients with 67. Timmer, M. R. et al. Prediction of response to endoscopic therapy of Barrett’s
Barrett’s esophagus regardless of a diagnosis of dysplasia. Gastroenterology dysplasia by using genetic biomarkers. Gastrointest Endosc. 80, 984–991 (2014).
https://doi.org/10.1053/j.gastro.2021.10.038 (2021). 68. Reid, B. J., Levine, D. S., Longton, G., Blount, P. L. & Rabinovitch, P. S. Predictors of
43. Sharma, P. Review article: emerging techniques for screening and surveillance in progression to cancer in Barrett’s esophagus: baseline histology and flow
Barrett’s oesophagus. Aliment Pharm. Ther. 20, 63–70 (2004). cytometry identify low- and high-risk patient subsets. Am. J. Gastroenterol 95,
44. ASGE Standards of Practice Committee, Qumseya, B. et al. ASGE guideline on 1669–1676 (2000).
screening and surveillance of Barrett’s esophagus. Gastrointest Endosc. 90, 69. Reid, B. J., Haggitt, R. C., Rubin, C. E. & Rabinovitch, P. S. Barrett’s esophagus.
335–359.e2 (2019). Correlation between flow cytometry and histology in detection of patients at
45. Gross, S. A., Smith, M. S., Kaul, V. & US Collaborative WATS3D Study Group. risk for adenocarcinoma. Gastroenterology 93, 1–11 (1987).
Increased detection of Barrett’s esophagus and esophageal dysplasia with 70. Reid, B. J. et al. Flow-cytometric and histological progression to malignancy in
adjunctive use of wide-area transepithelial sample with three-dimensional Barrett’s esophagus: prospective endoscopic surveillance of a cohort. Gastro-
computer-assisted analysis (WATS). U Eur. Gastroenterol J. 6, 529–535 (2018). enterology 102, 1212–1219 (1992).
46. Smith, M. S. et al. Wide-area transepithelial sampling with computer-assisted 3- 71. Rabinovitch, P. S., Longton, G., Blount, P. L., Levine, D. S. & Reid, B. J. Predictors of
dimensional analysis (WATS) markedly improves detection of esophageal dys- progression in Barrett’s esophagus III: baseline flow cytometric variables. Am. J.
plasia and Barrett’s esophagus: analysis from a prospective multicenter Gastroenterol 96, 3071–3083 (2001).
community-based study. Dis. Esophagus 32, doy099 (2019). 72. Montgomery, E. A. et al. Barrett esophagus with dysplasia. Flow cytometric DNA
47. Vennalaganti, P. R. et al. Increased detection of Barrett’s esophagus-associated analysis of routine, paraffin-embedded mucosal biopsies. Am. J. Clin. Pathol.
neoplasia using wide-area trans-epithelial sampling: a multicenter, prospective, 106, 298–304 (1996).
randomized trial. Gastrointest Endosc. 87, 348–355 (2018). 73. Menke-Pluymers, M. B., Mulder, A. H., Hop, W. C., van Blankenstein, M. & Tilanus,
48. Vennalaganti, P. R. et al. Inter-Observer Agreement among Pathologists Using H. W. Dysplasia and aneuploidy as markers of malignant degeneration in Bar-
Wide-Area Transepithelial Sampling With Computer-Assisted Analysis in Patients rett’s oesophagus. The Rotterdam Oesophageal Tumour Study Group. Gut 35,
With Barrett’s Esophagus. Am. J. Gastroenterol. 110, 1257–1260 (2015). 1348–1351 (1994).
49. Canto, M. I. & Montgomery, E. Response. Gastrointest Endosc. 88, 202–203 (2018). 74. Huang, Q., Yu, C., Zhang, X. & Goyal, R. K. Comparison of DNA histograms by
50. Shaheen, N. J., Smith, M. S. & Odze, R. D. Progression of Barrett’s esophagus, standard flow cytometry and image cytometry on sections in Barrett’s adeno-
crypt dysplasia, and low-grade dysplasia diagnosed by wide-area transepithelial carcinoma. BMC Clin. Pathol 8, 1–9 (2008).

Modern Pathology (2022) 35:1000 – 1012

 W.-T. Choi et al.
1012
75. Kerkhof, M. et al. Aneuploidy and high expression of p53 and Ki67 is associated Reduce Equivocal Diagnoses and Improve Prediction of Patient Outcome: A 10-
with neoplastic progression in Barrett esophagus. Cancer Biomark 4, 1–10 Year Review. Am. J. Surg. Pathol 42, 1723–1729 (2018).
(2008). 104. Montgomery, E. et al. Dysplasia as a predictive marker for invasive carcinoma in
76. Bowman, C. J. et al. Persistent or recurrent Barrett’s neoplasia after an endo- Barrett esophagus: a follow-up study based on 138 cases from a diagnostic
scopic therapy session is associated with DNA content abnormality and can be variability study. Hum Pathol 32, 379–388 (2001).
detected by DNA flow cytometric analysis of paraffin-embedded tissue. Mod 105. Latt, S. A. & Stetten, G. Spectral studies on 33258 Hoechst and related bisben-
Pathol 34, 1889–1900 (2021). zimidazole dyes useful for fluorescent detection of deoxyribonucleic acid
77. Choi, W. T. et al. Diagnosis and risk stratification of Barrett’s dysplasia by flow synthesis. J. Histochem. Cytochem. 24, 24–33 (1976).
cytometric DNA analysis of paraffin-embedded tissue. Gut 67, 1229–1238 106. Eluri, S. et al. Validation of a biomarker panel in Barrett’s esophagus to predict
(2018). progression to esophageal adenocarcinoma. Dis Esophagus 31, doy026 (2018).
78. Goldblum, J. R. Current issues in Barrett’s esophagus and Barrett’s-related dys- 107. Shankey, T. V. et al. Guidelines for implementation of clinical DNA cytometry.
plasia. Mod. Pathol 28, S1–S6 (2015). International Society for Analytical Cytology. Cytometry 14, 472–477 (1993).
79. Wild, C. P. & Hardie, L. J. Reflux, Barrett’s oesophagus and adenocarcinoma: 108. Rabinovitch P. S. Practical considerations for DNA content and cell cycle ana-
burning questions. Nat. Rev. Cancer 3, 676–684 (2003). lysis. In: Bauer K. D., Duque R. E., and Shankey T. V., editors. Clinical Flow
80. Souza, R. F., Morales, C. P. & Spechler, S. J. Review article: a conceptual approach Cytometry: Principles and Applications. Baltimore, MD: Williams and Wilkins,
to understanding the molecular mechanisms of cancer development in Barrett’s 1992. p. 117-142.
oesophagus. Aliment Pharm Ther. 15, 1087–1100 (2001). 109. Sikkema, M. et al. Aneuploidy and overexpression of Ki67 and p53 as markers for
81. Reid, B. J. p53 and neoplastic progression in Barrett’s esophagus. Am. J. Gas- neoplastic progression in Barrett’s esophagus: a case-control study. Am. J.
troenterol 96, 1321–1323 (2001). Gastroenterol 104, 2673–2680 (2009).
82. Reid, B. J. et al. Predictors of progression in Barrett’s esophagus II: baseline 17p 110. Giménez, A. et al. Flow cytometric DNA analysis and p53 protein expression
(p53) loss of heterozygosity identifies a patient subset at increased risk for show a good correlation with histologic findings in patients with Barrett’s
neoplastic progression. Am. J. Gastroenterol 96, 2839–2848 (2001). esophagus. Cancer 83, 641–651 (1998).
83. Schnell, T. G. et al. Long-term nonsurgical management of Barrett’s esophagus 111. Lee, H., Rabinovitch, P. S., Mattis, A. N., Kakar, S. & Choi, W. T. DNA flow cyto-
with high-grade dysplasia. Gastroenterology 120, 1607–1619 (2001). metric analysis of paraffin-embedded tissue for the diagnosis of malignancy in
84. Skinner, D. B. et al. Barrett’s esophagus. Comparison of benign and malignant bile duct biopsies. Hum Pathol 99, 80–87 (2020).
cases. Ann. Surg. 198, 554–565 (1983). 112. Zeki, S. S. et al. Clonal selection and persistence in dysplastic Barrett’s eso-
85. Haggitt, R. C., Tryzelaar, J., Ellis, F. H. & Colcher, H. Adenocarcinoma complicating phagus and intramucosal cancers after failed radiofrequency ablation. Am. J.
columnar epithelium-lined (Barrett’s) esophagus. Am. J. Clin. Pathol. 70, 1–5 Gastroenterol 108, 1584–1592 (2013).
(1978). 113. Prasad, G. A. et al. Utility of biomarkers in prediction of response to ablative
86. Schmidt, H. G., Riddell, R. H., Walther, B., Skinner, D. B. & Riemann, J. F. Dysplasia therapy in Barrett’s esophagus. Gastroenterology 135, 370–379 (2008).
in Barrett’s esophagus. J. Cancer Res. Clin. Oncol. 110, 145–152 (1985). 114. Prasad, G. A. et al. Correlation of histology with biomarker status after photo-
87. Zhu, W. et al. A histologically defined subset of high-grade dysplasia in Barrett dynamic therapy in Barrett esophagus. Cancer 113, 470–476 (2008).
mucosa is predictive of associated carcinoma. Am. J. Clin. Pathol. 132, 94–100
(2009).
88. Williamson, W. A. et al. Barrett’s esophagus. Prevalence and incidence of ade- ACKNOWLEDGEMENTS
nocarcinoma. Arch. Intern. Med. 151, 2212–2216 (1991). We thank Dr. Rebecca Critchley-Thorne for providing representative images of
89. Singh, S. et al. Incidence of esophageal adenocarcinoma in Barrett’s esophagus TissueCypher.
with low-grade dysplasia: a systematic review and meta-analysis. Gastrointest
Endosc. 79, 897–909 (2014).
90. Qumseya, B. J. et al. Disease progression in Barrett’s low-grade dysplasia with
radiofrequency ablation compared with surveillance: systematic review and AUTHOR CONTRIBUTIONS
meta-analysis. Am. J. Gastroenterol 112, 849–865 (2017). W.-T.C., G.Y.L. and E.A.M. drafted the manuscript. All authors read and approved the
91. Curvers, W. L. et al. Low-grade dysplasia in Barrett’s esophagus: overdiagnosed final paper.
and underestimated. Am. J. Gastroenterol 105, 1523–1530 (2010).
92. Reid, B. J., Blount, P. L., Feng, Z. & Levine, D. S. Optimizing endoscopic biopsy
detection of early cancers in Barrett’s high-grade dysplasia. Am. J. Gastroenterol COMPETING INTERESTS
95, 3089–3096 (2000). The authors declare no competing interests.
93. Sangle, N. A. et al. Overdiagnosis of high-grade dysplasia in Barrett’s esophagus:
a multicenter, international study. Mod Pathol 28, 758–765 (2015).
94. American Gastroenterological Association, Spechler, S. J., Sharma, P., Souza, R. F., ADDITIONAL INFORMATION
Inadomi, J. M. & Shaheen, N. J. American Gastroenterological Association Correspondence and requests for materials should be addressed to Won-Tak Choi.
medical position statement on the management of Barrett’s esophagus. Gas-
troenterology 140, 1084–1091 (2011). Reprints and permission information is available at http://www.nature.com/
95. Skacel, M. et al. The diagnosis of low-grade dysplasia in Barrett’s esophagus and reprints
its implications for disease progression. Am. J. Gastroenterol. 95, 3383–3387
(2000). Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims
96. Stachler, M. D. et al. Paired exome analysis of Barrett’s esophagus and adeno- in published maps and institutional affiliations.
carcinoma. Nat Genet 47, 1047–1055 (2015).
97. Nones, K. et al. Genomic catastrophes frequently arise in esophageal adeno-
carcinoma and drive tumorigenesis. Nat. Commun 5, 5224 (2014).
98. Panarelli, N. C. & Yantiss, R. K. Do ancillary studies aid detection and classification Open Access This article is licensed under a Creative Commons
of Barrett esophagus? Am. J. Surg. Pathol. 40, e83–e93 (2016). Attribution 4.0 International License, which permits use, sharing,
99. Fitzgerald, R. C. et al. British Society of Gastroenterology guidelines on the adaptation, distribution and reproduction in any medium or format, as long as you give
diagnosis and management of Barrett’s oesophagus. Gut 63, 7–42 (2014). appropriate credit to the original author(s) and the source, provide a link to the Creative
100. Kunzmann, A. T. et al. Does risk of progression from Barrett’s esophagus to Commons license, and indicate if changes were made. The images or other third party
esophageal adenocarcinoma change based on the number of non-dysplastic material in this article are included in the article’s Creative Commons license, unless
endoscopies? Dig. Dis. Sci. 66, 1965–1973 (2021). indicated otherwise in a credit line to the material. If material is not included in the
101. Peters, Y. et al. Incidence of Progression of Persistent Nondysplastic Barrett’s article’s Creative Commons license and your intended use is not permitted by statutory
Esophagus to Malignancy. Clin. Gastroenterol Hepatol 17, 869–877.e5 (2019). regulation or exceeds the permitted use, you will need to obtain permission directly
102. Ten Kate, F. et al. Improved Progression Prediction in Barrett’s Esophagus With from the copyright holder. To view a copy of this license, visit http://creativecommons.
Low-grade Dysplasia Using Specific Histologic Criteria. Am. J. Surg. Pathol 42, org/licenses/by/4.0/.
918–926 (2018).
103. Waters, K. M., Salimian, K. J., Voltaggio, L. & Montgomery, E. A. Refined Criteria
for Separating Low-grade Dysplasia and Nondysplastic Barrett Esophagus © The Author(s) 2022

Modern Pathology (2022) 35:1000 – 1012