Risk of progression in Barrett’s esophagus indefinite for dysplasia: a systematic review and meta-analysis

SYSTEMATIC REVIEW AND META-ANALYSIS

Risk of progression in Barrett’s esophagus indefinite for dysplasia: a systematic review and meta-analysis Rajesh Krishnamoorthi, MD,1,* Babu P. Mohan, MD,2,* Mahendran Jayaraj, MD,3 Kenneth K. Wang, MD,5 David A. Katzka, MD,5 Andrew Ross, MD,1 Douglas G. Adler, MD,4 Prasad G. Iyer, MD, MSc5 Seattle, Washington; Tucson, Arizona; Las Vegas, Nevada; Salt Lake City, Utah; Rochester, Minnesota, USA

Background and Aims: Risk of progression in Barrett’s esophagus (BE) with low-grade dysplasia (LGD) and high-grade dysplasia (HGD) has been established. However, the natural history of BE with indefinite dysplasia (BE-IND) remains unclear. We performed a systematic review and meta-analysis to estimate the pooled risk of progression to HGD and/or esophageal adenocarcinoma (EAC) in BE-IND. Methods: We performed a systematic search of multiple databases to June 2018 to identify studies reporting the incidence of HGD, EAC, or HGD/EAC as an outcome in patients with BE-IND undergoing endoscopic surveillance. The pooled incidence rate of HGD and/or EAC and EAC alone was estimated. Results: We identified 8 studies reporting the incidence of HGD and/or EAC and 5 studies reporting the incidence of EAC in BE-IND. The pooled incidence of HGD and/or EAC (89 cases in 1441 patients over 5306.2 person-years) was 1.5 per 100 person-years (95% confidence interval [CI], 1.0-2.0). The pooled incidence of EAC (40 cases in 1266 patients over 4520.2 person-years) was 0.6 per 100 person-years (95% CI, 0.1-1.1). Substantial heterogeneity was noted in the analyses. On subgroup analysis, the incidence of EAC was higher in studies from Europe compared with North America (0.9% vs 0.1%, P Z .01). The pooled incidence of LGD was 11.4 per 100 person-years (95% CI, 0.06-0.2). Conclusion: The estimated incidence of HGD and/or EAC and EAC alone in BE-IND is similar to the previously reported progression risk in BE-LGD. Based on these risk estimates, patients with BE-IND should be placed on active endoscopic surveillance. (Gastrointest Endosc 2019;-:1-8.)

INTRODUCTION Barrett’s esophagus (BE) is a well-established precursor lesion to esophageal adenocarcinoma (EAC),1 a malignancy Abbreviations: ACG, American College of Gastroenterology; BE, Barrett’s esophagus; CI, confidence interval; EAC, esophageal adenocarcinoma; LGD, low-grade dysplasia; HGD, high-grade dysplasia; IND, indefinite for dysplasia; NDBE, nondysplastic Barrett’s esophagus; PPI, proton pump inhibitor. DISCLOSURE: Dr Iyer has received research funding from Exact Sciences, Medtronic, and Pentax Medical Nine Point Medical, and consulting fees from Medtronic, Pentax Medical, and Symple Surgical. Dr Wang has received funding from Abbott Diagnostics. Dr Katzka has received consulting fees from Shire and Celege. Dr Ross has received consulting fees from Boston Scientific. Dr Adler has received consulting fees from Boston Scientific. All authors disclosed no financial relationships relevant to this publication. *Drs Krishnamoorthi and Mohan contributed equally to this article. Copyright ª 2019 by the American Society for Gastrointestinal Endoscopy 0016-5107/$36.00 https://doi.org/10.1016/j.gie.2019.07.037 (footnotes continued on last page of article)

www.giejournal.org

with a dismal 5-year survival rate of less than 20%.2 Patients with BE undergo endoscopic surveillance with the goal of detecting dysplasia or early cancer. The presence of dysplasia increases the risk of EAC in BE.3-5 The degree of dysplasia is the most widely accepted predictor of progression, and hence the surveillance intervals recommended by GI societies are based on the dysplasia grade.1,6,7 It has been postulated that among most patients who develop neoplastic disease, intestinal metaplasia transforms from nondysplastic BE (NDBE) to low-grade dysplasia (LGD), to high-grade dysplasia (HGD) and eventually to intra-mucosal carcinoma.8,9 The reported risk of progression to EAC is estimated to be 0.3 per 100 person-years with NDBE, 0.5 per 100 person-years with LGD, and 6.6 per 100 person-years with HGD.3-5 An indeterminate category of dysplasia reported by pathologists is BE with indefinite for dysplasia (IND). There are no well-established diagnostic criteria for “indefinite for dysplasia.” Approximately 4.3% to 8.4% of BE biopsy specimens are diagnosed as BE-IND. This largely depends on the diagnostic thresholds used by pathologists.33 For a Volume

-,

No.

-

: 2019 GASTROINTESTINAL ENDOSCOPY 1

Progression in Barrett’s esophagus indefinite for dysplasia

Krishnamoorthi et al

definitive diagnosis of dysplasia, the epithelial abnormalities characteristic of dysplasia should not only be present in the crypts but also involve the surface epithelium.10 In BEIND, reactive changes in the nucleus and cytoplasm show similarities to dysplasia, but maturation toward the luminal surface and/or the presence of clonality are not always established.10 Studies have shown that the reproducibility and inter-reader reliability of BE-IND is poor among pathologists, resulting in variation of histologic categorization.11,12 The current recommendation for surveillance in patients with BE-IND is to treat inflammation with acid suppression for 8 to 12 weeks followed by endoscopy with repeat biopsies to reassess the histologic diagnosis.9 There is wide variation in the reported rates of malignant progression in BE-IND. A large population-based cohort study by Kestens et al13 reported the risk of progression to HGD and/or EAC in BE-IND as 1.4 per 100 personyears and EAC alone as 0.8 per 100 person-years. On the other hand, a well-conducted study by Sinh et al,14 using a multicenter database with a prior standardized definition for BE-IND, reported the incidence of HGD/EAC and EAC alone as 0.8 and 0.2 per 100 person-years, respectively. As the actual risk of malignant progression in BE-IND is unclear, patients with persistent BE-IND on repeat biopsies after optimization of anti-reflux measures are managed the same way as BE-LGD in clinical practice.4,15 As a result, surveillance endoscopies are performed at shorter intervals in patients with BE-IND compared with NDBE. Frequent endoscopies can lead to increased cumulative risk of adverse events and health care costs. Hence, reliably estimating the risk of progression in BE-IND is important for both patient care and cost effectiveness. We conducted a systematic review and meta-analysis to estimate the pooled risk of progression to HGD and/or EAC in patients with BE-IND.

METHODS

Study selection In this meta-analysis, we included cohort studies that met the following criteria: (1) study population consisted of patients with BE-IND undergoing endoscopic surveillance with biopsies without endoscopic therapy; (2) studies that reported progression to HGD and/or EAC as an outcome; and (3) studies that reported total personyears of follow-up to permit calculation of incidence rates of HGD and/or EAC. We excluded (1) studies with insufficient data to allow calculation of the incidence rate of HGD and/or EAC, (2) studies that included patients who had previously undergone endoscopic therapy or provided unclear information on whether some of the patients underwent endoscopic therapy, and (3) letters to the editor, editorials, and review articles. In the case of multiple publications from the same cohort, data from the most recent comprehensive report were included. This was the case with Ma et al17 and Sinh et al.14

Data abstraction and quality assessment

This systematic review was performed in accordance with the Cochrane Handbook for Systematic Reviews of Interventions.32 It is reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines.16

Search strategy We conducted a comprehensive search of several databases and conference proceedings, including PubMed, EMBASE, and Web of Science databases (earliest inception to June 2018) to identify studies that reported the incidence of HGD, EAC, HGD, and/or EAC as an outcome in patients with BE-IND undergoing endoscopic surveillance. An experienced medical librarian using input from the study authors helped with the literature search. The details of the search strategy and data sources are reported in Appendix 1 (available online at www.giejournal.org). 2 GASTROINTESTINAL ENDOSCOPY Volume

Key words used in the search included a combination of “Barrett’s esophagus,” “Barrett’s epithelium,” “indefinite dysplasia,” “high-grade dysplasia,” and “esophageal adenocarcinoma.” The search was restricted to studies with human subjects published in the English language. Two authors (M.B., K.R.) independently reviewed the title and abstract of studies identified in the primary search and excluded studies that did not address the research question, based on prespecified exclusion and inclusion criteria. The full text of remaining articles was reviewed to determine whether it contained relevant information. Any discrepancy in article selection was resolved by consensus and in discussion with a co-author. The bibliographic sections of the selected articles, as well as systematic and narrative articles on the topic, were also manually searched for additional relevant articles.

-,

No.

-

: 2019

After identifying relevant studies, data on the study characteristics, patient characteristics, and relevant study outcomes were abstracted onto a standardized form by 2 authors (M.B., K.R.). The quality of the individual studies was independently assessed by 2 authors (M.B., M.J.) using a scale modified from the Newcastle-Ottawa scale for cohort studies.18,31 The quality scale consisted of 7 questions, the details of which are provided in Supplementary Table 1 (available online at www.giejournal.org). A score of 5, 3 to 4, and 2 was considered suggestive of high-quality, medium-quality, and low-quality study, respectively.

Outcomes assessed The primary outcome of the current study was to estimate the incidence rate of combined HGD and/or EAC in patients with BE-IND. The secondary outcome was to estimate the incidence rate of EAC alone in patients with BEIND. We performed preplanned subgroup analyses based www.giejournal.org

Included

Eligibility

Screening

Identification

Krishnamoorthi et al

Progression in Barrett’s esophagus indefinite for dysplasia

Total studies found on search of PubMed, Embase, and others (N=5716}

Studies screened & removed (N=5676}

• Not relevant to study question= 5114 • Abstracts with inadequate data = 163 • Review = 294 • Erratum = 72 • Not in English = 33

Full text studies assessed for eligibility (N=40) Full-text studies excluded (n=32] • Overlapping cohorts=2 • Due to total study patients <15=15 • Unclear information on prior endoscopy treatment=12 • No follow-up information=3

Studies included in Meta-Analysis (N=8}

Figure 1. Study selection flow chart.

on geographic location (Europe vs North America) and study quality (high vs medium). In addition, to assess whether there was a difference in the reported incidence of progression based on the IND/BE ratio, we performed a stratified analysis using a prespecified cut-off IND/BE ratio of 0.05. Given the lack of IND/BE data from populationbased studies, an IND/BE cut-off of 0.05 was believed to be appropriate based on expert opinion (from gastroenterologists and GI pathologists) and nonpopulation-based studies.33

Statistical analysis The pooled incidence of HGD and/or EAC per personyear and 95% confidence intervals (CIs) were calculated www.giejournal.org

using the random-effects model described by DerSimonian and Laird,19 and our application can be seen to fit within their general approach (where effect is measured by the probability of risk). We assessed heterogeneity between study-specific estimates using the inconsistency index (I2 statistic), which estimates the proportion of total variances across studies because of heterogeneity rather than by chance. Values of <30%, 30% to 60%, 61% to 75%, and >75% were considered suggestive of low, moderate, substantial, and considerable heterogeneity, respectively.20 Once heterogeneity was noted, we investigated betweenstudy sources of heterogeneity using subgroup analyses by stratifying the original estimates according to the Volume

-,

No.

-

: 2019 GASTROINTESTINAL ENDOSCOPY 3

Progression in Barrett’s esophagus indefinite for dysplasia

Krishnamoorthi et al

TABLE 1. Study and population characteristics Study

Country

Study type

Age

Chan et al, 2016

Ireland

Retrospective cohort

Ma et al, 201717

USA

Retrospective cohort

Horvath et al, 2015

USA

Choi et al, 201525

USA

24

26

23

Male %

Total no. of patients with BE

Mean 63.1  11.4

70

1383

54 to 69

66

354

Retrospective cohort

54 to 86

73

NR

Retrospective cohort

39 to 86

78

NR

Callaway et al, 2015

USA

Retrospective cohort

NR

NR

848

Kestens et al, 201513

Holland

Retrospective cohort

Mean 60.9

70

NR

UK

Retrospective cohort

43 to 81

75

101

UK

Retrospective cohort

NR

NR

143

Sonwalkar et al, 2010

27

Abradu-Berchie et al, 201722

BE, Barrett’s esophagus; EAC, esophageal adenocarcinoma; IND, indefinite for dysplasia; HGD, high-grade dysplasia; NR, not reported.

study characteristics (as described earlier). A P value for differences between subgroups <.10 was considered statistically significant. This meant that stratifying based on those subgroups could potentially explain the heterogeneity observed in the overall analysis. If there were 10 studies included in a meta-analysis, we planned to assess for publication bias qualitatively, by visual inspection of a funnel plot, and quantitatively by the Egger test.21 All analyses and graphs were performed using Comprehensive Meta-Analysis (CMA) software, version 3 (BioStat, Englewood, NJ, USA).

RESULTS Search results and population characteristics From a total of 5716 studies identified by our search strategy, 8 studies13,17,22-27 were included in the metaanalysis. Forty full-text studies were reviewed for eligibility. Two studies were excluded due to overlapping cohorts with studies already included. Three studies were excluded because they did not report sufficient data to calculate follow-up duration in person-years. The schematic diagram of study selection is illustrated in Figure 1.

Vienna classification.9 Most of the patients received twice-daily proton pump inhibitor (PPI) therapy and underwent surveillance endoscopies at 3- to 6-month intervals with biopsy specimens obtained per the Seattle protocol.

HGD and/or EAC in BE-IND In the 8 studies13,17,22-27 that reported the incidence of HGD and/or EAC as an outcome, there were 89 incident cases in 1441 BE-IND patients over 5306.2 person-years of follow-up. The pooled annual incidence of HGD and/ or EAC was 1.5 per 100 person-years (95% CI, 1.0-2.0) (Fig. 2). Moderate heterogeneity (I2 Z 56.5%) was noted in the analysis.

EAC in BE-IND In the 5 studies13,17,24,26,27 that reported the incidence of EAC as an outcome, there were 40 incident cases in 1266 BE-IND patients over 4520.2 person-years of follow-up. The pooled annual incidence of EAC was 0.6 per 100 person-years (95% CI, 0.1-1.1) (Fig. 3). Considerable heterogeneity (I2Z 89.0%) was noted in the analysis.

LGD in BE-IND Characteristics and quality of the studies Table 1 describes the characteristics and quality of the studies included. Supplementary Table 1 describes the quality assessment of individual studies. Of the studies included, 1 study13 was population based, 1 study22 was multicenter, and the remaining were single-center studies. Four studies17,23,25,26 were from the United States, and 4 studies13,22,24,27 were from Europe. Overall, 3 studies13,24,26 were considered high quality, and 5 studies17, 22, 23, 25, 27 were considered medium quality. None were considered low quality.

BE-IND diagnosis and surveillance endoscopy In all of the studies included, esophageal biopsy specimens were reviewed by at least 2 expert GI pathologists, and histologic diagnosis was made using the revised 4 GASTROINTESTINAL ENDOSCOPY Volume

-,

No.

-

: 2019

Four studies (1165 BE-IND patients)13,17,24,26 reported specific data on the incidence of LGD as an outcome. The pooled incidence of LGD was 11.4 per 100 personyears (95% CI, 0.06-0.2; I2Z83.6). Only 1 study13 reported on the regression of BE-IND to NDBE, and therefore a pooled rate could not be calculated. The reported regression of BE-IND to NDBE was 58.4% in that study.

Subgroup analysis by country and study quality Table 2 summarizes the subgroup analysis for the pooled incidence rate of HGD and/or EAC. There was no difference in the incidence rates of HGD and/or EAC in the subgroup analysis based on study location (Europe vs North America: 1.4% vs 1.6%, P Z .83) and study quality (high vs medium quality: 1.3% vs 1.7%, P Z .48). www.giejournal.org

Krishnamoorthi et al

Progression in Barrett’s esophagus indefinite for dysplasia

TABLE 1. Continued Patients with BE-IND (total no.)

Follow-up time

Total person-years of follow-up

EAC

EAC and/or HGD

Quality

110

5.9 years

1000

6

7

High

106

2.3 years

222

0

2

Medium

107

5 years

417

2

5

High

96

3 years

174

NR

5

Medium

79

7.1 years

612

NR

13

Medium

842

2585 person- years

2585

30

51

High

41

3.2 years

131.2

2

4

Medium

60

165 person-years

165

NR

2

Medium

Figure 2. Forest plot, high-grade dysplasia and/or EAC in Barrett’s esophagus indefinite for dysplasia. Pooled rate: 1.5 per 100 person-years; I2 Z 56.5%. CI, Confidence interval.

analysis but the differences were not statistically significant. A mild decrease in heterogeneity (I2) was also noted in this analysis.

SENSITIVITY ANALYSIS AND PUBLICATION BIAS Figure 3. Forest plot, EAC in Barrett’s esophagus indefinite for dysplasia. Pooled rate: 0.6 per 100 person-years; I2 Z 89%. CI, Confidence interval.

Table 3 summarizes the subgroup analysis for the pooled incidence rate of EAC. The incidence rate of EAC was higher in the studies from Europe compared with North America, and the difference was statistically significant (0.9% vs 0.1%, P Z .01). The incidence rate of EAC was numerically higher in the high-quality studies compared with the medium-quality studies without a statistically significant difference (0.8% vs 0.1%, P Z .10). Both analyses were limited by the small number of studies in each subgroup. In the 5 studies that reported data to calculate IND/BE, the ratios were more than the 0.05 cut-off in all of them.17,22-24,27 On analysis of the studies with IND/BE >0.05, the pooled rate of HGD and/or EAC was 1.2% (95% CI, 0.5-1.9; I2 Z 41) and EAC alone was 0.3% (95% CI, 0.1-0.9; I2 Z 74). The pooled rates from this analysis were numerically lower than the estimates from the overall www.giejournal.org

One study at a time was excluded sequentially to assess for any potential dominant effect by an individual study. On this analysis, no single study significantly affected the summary estimates. Assessment for publication bias was not performed because there were <10 studies included in the analysis.

DISCUSSION Currently, there are no reliable estimates of progression risk to HGD/EAC in patients with BE-IND. Hence, management strategies for BE-IND are unclear, and endoscopy is not routinely recommended. In this systematic review and meta-analysis of 8 studies with 1441 patients with BE-IND, the estimated incidence of HGD and/or EAC in BE-IND was 1.5 per 100 person-years. The estimated incidence of EAC in BE-IND was 0.6 per 100 person-years. The current study is the first systematic review and meta-analysis evaluating the risk of progression in BEVolume

-,

No.

-

: 2019 GASTROINTESTINAL ENDOSCOPY 5

Progression in Barrett’s esophagus indefinite for dysplasia

Krishnamoorthi et al

TABLE 2. Subgroup analyses: incidence of HGD and/or EAC in BE-IND

Subgroup

No. of studies

Incidence of HGD and/or EAC per 100 person-years (95% CI)

I2

Country (P Z .82) North America17,23,25,26

4 studies, 388 patients

1.6 (0.6-2.5)

13.4

4 studies, 1053 patients,

1.4 (0.6-2.3)

75.8

High13,24,16

3 studies, 1059 patients

1.3 (0.5-2.1)

82

Medium17,22,23,25,27

5 studies, 382 patients

1.7 (0.8-2.7)

1

Europe13,22,24,27 Study quality (P Z .48)

EAC, Esophageal adenocarcinoma; HGD, high-grade dysplasia; BE-IND, Barrett’s esophagus indefinite for dysplasia.

TABLE 3. Subgroup analyses: incidence of EAC in Barrett’s esophagus indefinite for dysplasia

No. of studies

Incidence of EAC per 100 person-years (95% CI)

I2

North America17,26

2 studies, 213 patients

0.1 (0.01-0.5)

48.1

Europe13,24,27

3 studies, 993 patients

0.9 (0.5-1.4)

39.4

3 studies, 1059 patients

0.8 (0.3-1.3)

54.3

2 studies, 147 patients

0.1 (0.01-0.7)

49.6

Subgroup Country (P Z .01)

Study quality (P Z .10) High13,24,26 17,27

Medium

EAC, Esophageal adenocarcinoma.

IND. The American Gastroenterological Association guidelines do not recommend a specific surveillance interval for BE-IND.28 The more recent American College of Gastroenterology (ACG) guidelines recommend that surveillance in BE-IND should follow that of BE-LGD and hence, they suggest a surveillance interval of 12 months in patients with confirmed BE-IND.6 However, the ACG guidelines on BE-IND management are based on expert opinion with a low level of evidence. The current study attempts to fill this evidence gap. A previous meta-analysis of patients with BE-LGD estimated the annual risk of HGD/ EAC and EAC as 1.7% and 0.5%, respectively.5 The current study’s estimates of the risk of HGD/EAC and EAC in BE-IND are similar to the corresponding risk estimates reported for BE-LGD (1.5% vs 1.7% and 0.6% vs 0.5%, respectively). As the progression risks are similar between BE-IND and BE-LGD, surveillance intervals for BE-IND should be similar to BE-LGD. Also, there might be a role for endoscopic therapy in BE-IND similar to BELGD if these results are confirmed in larger prospective studies. In our subgroup analysis, the incidence rate of EAC in studies from Europe was higher compared with studies from North America (0.9% vs 0.1%, P Z .01), and this difference was statistically significant. Although the small number of studies in each subgroup weakens the statistical validity, differences in histologic criteria for the diagnosis of BE-IND between North America and Europe 6 GASTROINTESTINAL ENDOSCOPY Volume

-,

No.

-

: 2019

could be a contributing factor. Similar to current findings on BE-IND, European studies have reported a higher risk of malignant progression in BE-LGD compared with North American studies.29,30 The pooled incidence of LGD in patients with BE-IND was relatively high at 11.4 per 100 person-years. Our estimate supports active surveillance of patients with BE-IND, because recent ACG guidelines recommend endoscopic therapy in patients with confirmed LGD.6 Data on risk factors for progression in BE-IND, like BE segment length, unifocal versus multifocal dysplasia, presence of hiatal hernia, smoking status, body mass index, and use of nonsteroidal anti-inflammatory drugs, were limited in individual studies. This limited our ability to perform a meta-regression analysis to assess for their influence on progression rates. Four of the studies included reported information on previous use of PPI medications.17,25-27 Based on our meta-regression analysis, previous treatment with PPIs had no significant influence on progression risk in BE-IND (meta-regression 2-tailed P values: PPI treatment Z .32 and for no previous treatment with PPI Z .23, with no statistical significance to the intercept). Given the small number of studies in this analysis, this observation has to be interpreted with caution. The findings highlight the importance of surveillance in BE-IND irrespective of PPI use. This systematic review and meta-analysis has several strengths. These include a systematic literature search www.giejournal.org

Krishnamoorthi et al

with well-defined inclusion criteria, careful exclusion of redundant studies, rigorous evaluation of study quality, calculation of incidence rates by person-years, and subgroup analysis to assess the cause of heterogeneity. There were several limitations, most of which are inherent to any meta-analysis. The analyses were done assuming that the incidence rate was constant over time, which may not be accurate. The studies included reported wide variability in the incidence of IND in a cohort of patients with BE with potential for misclassification bias because of interobserver variability in the diagnosis of BE-IND. The confirmation of the IND diagnosis by expert GI pathologists mitigates this to some extent. Studies also did not consistently report endoscopic surveillance intervals, raising concern for interstudy differences. Data on risk factors for progression in BE-IND in individual studies were limited. The heterogeneity in the overall analysis was high. Subgroup analysis based on study location and study quality explained the heterogeneity to some extent. The current study provides the best available risk estimates that may be used in counseling patients with persistent BE-IND after optimization of anti-reflux treatment. The risk estimates of malignant progression in BE-IND are similar to previously reported risk estimates in BELGD. This may be due to the fact that a major proportion of patients who were diagnosed with BE-IND actually had underlying BE-LGD. The relatively high progression rate to LGD (11.4%) noted in our meta-analysis provides some support to this argument. Although endoscopic therapy is currently a management option in BE-LGD, its role in BE-IND is not clear given the limitations of existing data on progression risk in BE-IND. Based on the current study’s risk estimates of progression, patients with BEIND should be placed on active endoscopic surveillance after their anti-reflux regimen is optimized. Prospective studies to define the natural history of BE-IND are needed to confirm these data. REFERENCES 1. Spechler SJ, Sharma P, Souza RF, et al. American Gastroenterological Association technical review on the management of Barrett’s esophagus. Gastroenterology 2011;140:e18-52. 2. Pohl H, Welch HG. The role of overdiagnosis and reclassification in the marked increase of esophageal adenocarcinoma incidence. J Natl Cancer Inst 2005;97:142-6. 3. Desai TK, Krishnan K, Samala N, et al. The incidence of oesophageal adenocarcinoma in non-dysplastic Barrett’s oesophagus: a metaanalysis. Gut 2012;61:970-6. 4. Rastogi A, Puli S, El-Serag HB, et al. Incidence of esophageal adenocarcinoma in patients with Barrett’s esophagus and high-grade dysplasia: a meta-analysis. Gastrointest Endosc 2008;67:394-8. 5. Singh S, Manickam P, Amin AV, et al. Incidence of esophageal adenocarcinoma in Barrett’s esophagus with low-grade dysplasia: a systematic review and meta- analysis. Gastrointest Endosc 2014;79: 897-909.e4.

www.giejournal.org

Progression in Barrett’s esophagus indefinite for dysplasia 6. Shaheen NJ, Falk GW, Iyer PG, et al. ACG Clinical Guideline: diagnosis and management of Barrett’s esophagus. Am J Gastroenterol 2015;111:30. 7. Krishnamoorthi R, Singh S, Ragunathan K, et al. Factors associated with progression of Barrett’s esophagus: a systematic review and metaanalysis. Clin Gastroenterol Hepatol 2018;16:1046-55.e8. 8. Spechler SJ, Souza RF. Barrett’s esophagus. N Engl J Med 2014;371: 836-45. 9. Schlemper R, Riddell R, Kato Yea, et al. The Vienna classification of gastrointestinal epithelial neoplasia. Gut 2000;47:251-5. 10. Haggitt RC. Barrett’s esophagus, dysplasia, and adenocarcinoma. Hum Pathol 1994;25:982-93. 11. Coco DP, Goldblum JR, Hornick JL, et al. Interobserver variability in the diagnosis of crypt dysplasia in Barrett esophagus. Am J Surg Pathol 2011;35:45-54. 12. Montgomery E, Goldblum JR, Greenson JK, et al. Dysplasia as a predictive marker for invasive carcinoma in Barrett esophagus: a follow-up study based on 138 cases from a diagnostic variability study. Hum Pathol 2001;32:379-88. 13. Kestens C, Leenders M, Offerhaus GJA, et al. Risk of neoplastic progression in Barrett’s esophagus diagnosed as indefinite for dysplasia: a nationwide cohort study. Endoscopy 2015;47:409-14. 14. Sinh P, Anaparthy R, Young PE, et al. Clinical outcomes in patients with a diagnosis of “indefinite for dysplasia” in Barrett’s esophagus: a multicenter cohort study. Endoscopy 2015;47:669-74. 15. Wani S, Falk GW, Post J, et al. Risk factors for progression of low-grade dysplasia in patients with Barrett’s esophagus. Gastroenterology 2011;141:1179-86.e1. 16. Moher D, Liberati A, Tetzlaff J, et al. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. Ann Intern Med 2009;151:264-9. 17. Ma M, Shroff S, Feldman M, et al. Risk of malignant progression in Barrett’s esophagus indefinite for dysplasia. Dis Esophagus 2017;30: 1-5. 18. Stang A. Critical evaluation of the Newcastle-Ottawa scale for the assessment of the quality of nonrandomized studies in meta-analyses. Eur J Epidemiol 2010;25:603-5. 19. DerSimonian R, Laird N. Meta-analysis in clinical trials. Controlled Clinical Trials 1986;7:177-88. 20. Guyatt GH, Oxman AD, Kunz R, et al. GRADE guidelines: 7. Rating the quality of evidence–inconsistency. J Clin Epidemiol 2011;64: 1294-302. 21. Easterbrook PJ, Gopalan R, Berlin JA, et al. Publication bias in clinical research. Lancet 1991;337:867-72. 22. Abradu-Berchie B, Ragunath K, Card T, et al. PWE-116 Progression of Barrett’s oesophagus with low-grade dysplasia and indefinite for dysplasia. Gut 2017;66:A185. 23. Callaway JK, Kommineni VT, Kahn A, et al. Tu1564 Indefinite dysplasia in Barrett’s esophagus confers a similar risk of progression to high grade dysplasia or esophageal adenocarcinoma as low grade dysplasia [abstract]. Gastrointest Endosc 2015;81: AB511. 24. Chan G, Chin JL, O'Brien M, et al. Sa1253 Incidence rate and predictors of progression in patients with Barrett’s esophagus: experience from a large Irish tertiary centre. Gastroenterology 2016;150: S258. 25. Choi W-T, Emond MJ, Rabinovitch PS, et al. “Indefinite for dysplasia” in Barrett’s esophagus: inflammation and DNA content abnormality are significant predictors of early detection of neoplasia. Clin Transl Gastroenterol 2015;6:e81. 26. Horvath B, Singh P, Xie H, et al. Risk for esophageal neoplasia in Barrett’s esophagus patients with mucosal changes indefinite for dysplasia. J Gastroenterol Hepatol 2015;30:262-7. 27. 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 (a-methylacyl-CoAracemase). Histopathology 2010;56:900-7.

Volume

-,

No.

-

: 2019 GASTROINTESTINAL ENDOSCOPY 7

Progression in Barrett’s esophagus indefinite for dysplasia

Krishnamoorthi et al

28. Spechler SJ, Sharma P, Souza RF, et al. American Gastroenterological Association medical position statement on the management of Barrett’s esophagus. Gastroenterology 2011;140:1084-91. 29. Curvers WL, ten Kate FJ, Krishnadath KK, et al. Low-grade dysplasia in Barrett’s esophagus: overdiagnosed and underestimated. Am J Gastroenterol 2010;105:1523-30. 30. 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 2015;64:700-6. 31. Wells G, Shea B, O’Connell D, et al. The Newcastle-Ottawa Scale (NOS) for assessing the quality of nonrandomised studies in meta-analyses. Available at: http://www.ohri.ca/programs/clinical_epidemiology/ oxford.htm. 32. Higgins J, Green S. Cochrane handbook for systematic reviews of interventions version 5.1.0. Cochrane Collaboration 2011;5. Available from http://handbook.cochrane.org. 33. Waters KM, Salimian KJ, Voltaggio L, et al. Refined criteria for separating low-grade dysplasia and nondysplastic Barrett esophagus

8 GASTROINTESTINAL ENDOSCOPY Volume

-,

No.

-

: 2019

reduce equivocal diagnoses and improve prediction of patient outcome: a 10-year review. Am J Surg Pathol 2018;42:1723-9.

Received March 6, 2019. Accepted July 27, 2019. Current affiliations: Digestive Diseases Institute, Virginia Mason Medical Center, Seattle, Washington (1); Internal Medicine, Banner University Medical Center, University of Arizona, Tucson, Arizona (2); Department of Gastroenterology and Hepatology, University of Nevada, Las Vegas, Nevada (3); Gastroenterology and Hepatology, University of Utah School of Medicine, Huntsman Cancer Center, Salt Lake City, Utah (4); Division of Gastroenterology and Hepatology, Mayo Clinic, Rochester, Minnesota, USA (5). Reprint requests: Prasad G. Iyer, MD, MSc, Professor of Medicine, Barrett’s Esophagus Unit, Division of Gastroenterology and Hepatology, Mayo Clinic, Rochester, MN 55905.

www.giejournal.org

Krishnamoorthi et al

Progression in Barrett’s esophagus indefinite for dysplasia

APPENDIX 1 Epub ahead of print, in-process, and other non-indexed citations, Ovid MEDLINE Daily and Ovid MEDLINE 1946 to present

Searches 1 (barrett* adj (esophagus or oesophagus)).mp. [mpZtitle, abstract, original title, name of substance word, subject heading word, keyword heading word, protocol supplementary concept word, rare disease supplementary concept word, unique identifier] 2

esophageal stenosis/ or esophagitis, peptic/

3

1 or 2

4

3 and dysplas*.mp. [mpZtitle, abstract, original title, name of substance word, subject heading word, keyword heading word, protocol supplementary concept word, rare disease supplementary concept word, unique identifier]

5 3 and (ind or indefinite).mp. [mpZtitle, abstract, original title, name of substance word, subject heading word, keyword heading word, protocol supplementary concept word, rare disease supplementary concept word, unique identifier] 6 3 and atypia.mp. [mp=title, abstract, original title, name of substance word, subject heading word, keyword heading word, protocol supplementary concept word, rare disease supplementary concept word, unique identifier] 7

4 or 6

8

7 and progress*.mp. [mpZtitle, abstract, original title, name of substance word, subject heading word, keyword heading word, protocol supplementary concept word, rare disease supplementary concept word, unique identifier]

9 8 and (carcinoma* or malign* or invasive* or adenocarcinoma*).mp. [mpZtitle, abstract, original title, name of substance word, subject heading word, keyword heading word, protocol supplementary concept word, rare disease supplementary concept word, unique identifier] 10

5 or 6 or 9

11 10 and (surveillance or follow* or risk*).mp. [mpZtitle, abstract, original title, name of substance word, subject heading word, keyword heading word, protocol supplementary concept word, rare disease supplementary concept word, unique identifier] 12 10 and ((cohort* or incidence).mp. or follow-up studies/ or prospective*.mp. or retrospective*.mp. or random*.mp.) [mpZtitle, abstract, original title, name of substance word, subject heading word, keyword heading word, protocol supplementary concept word, rare disease supplementary concept word, unique identifier] 13

11 or 12

14

remove duplicates from 13

www.giejournal.org

Volume

-,

No.

-

: 2019 GASTROINTESTINAL ENDOSCOPY 8.e1

Progression in Barrett’s esophagus indefinite for dysplasia

Krishnamoorthi et al

EMBASE 1

(barrett* adj (esophagus or oesophagus)).mp. [mpZtitle, abstract, heading word, drug trade name, original title, device manufacturer, drug manufacturer, device trade name, keyword, floating subheading]

2

esophageal stenosis/ or esophagitis, peptic/

3

1 or 2

4

3 and dysplas*.mp. [mpZtitle, abstract, heading word, drug trade name, original title, device manufacturer, drug manufacturer, device trade name, keyword, floating subheading]

5

3 and (ind or indefinite).mp. [mpZtitle, abstract, heading word, drug trade name, original title, device manufacturer, drug manufacturer, device trade name, keyword, floating subheading]

6

3 and atypia.mp. [mpZtitle, abstract, heading word, drug trade name, original title, device manufacturer, drug manufacturer, device trade name, keyword, floating subheading]

7

4 or 6

8

7 and progress*.mp. [mpZtitle, abstract, heading word, drug trade name, original title, device manufacturer, drug manufacturer, device trade name, keyword, floating subheading]

9

8 and (carcinoma* or malign* or invasive* or adenocarcinoma*).mp. [mpZtitle, abstract, heading word, drug trade name, original title, device manufacturer, drug manufacturer, device trade name, keyword, floating subheading]

10

5 or 6 or 9

11

10 and (surveillance or follow* or risk*).mp. [mpZtitle, abstract, heading word, drug trade name, original title, device manufacturer, drug manufacturer, device trade name, keyword, floating subheading]

12 10 and ((cohort* or incidence).mp. or follow-up studies/ or prospective*.mp. or retrospective*.mp. or random*.mp.) [mpZtitle, abstract, heading word, drug trade name, original title, device manufacturer, drug manufacturer, device trade name, keyword, floating subheading] 13

11 or 12

14

remove duplicates from 13

15

limit 14 to human

16

15 not case report/

17

16 not (letter or note or comment*).pt.

18

*barrett esophagus/ and 17

Web of Science TOPIC: (barrett* SAME (eosphag* OR oesophag* OR metaplas*)) AND TOPIC: (dysplasia* SAME (ind OR indefinite OR "low grade")) AND TOPIC: (progress*) AND TOPIC: (outcome* OR carcinoma* OR adenocarcinoma* OR follow* OR surveill*)

Scopus (TITLE-ABS-KEY (barrett* W/2 (esophag* OR oesophag* OR metaplas*)) AND TITLE-ABS-KEY (dysplas* W/ 3 (ind OR indefinite OR "low grade")) AND TITLE-ABSKEY (progress* AND (malign* OR outcome* OR carcinoma* OR adenocarcinoma* OR risk* OR surveill* OR follow* OR incidence)))

8.e2 GASTROINTESTINAL ENDOSCOPY Volume

-,

No.

-

: 2019

www.giejournal.org

Krishnamoorthi et al

Progression in Barrett’s esophagus indefinite for dysplasia

SUPPLEMENTARY TABLE 1. Study quality assessment of individual studies

Question

Scoring scheme

Chan et al, 201624

Ma et al, 201717

Horvath et al, 201526

Choi et al, 201525

Callaway et al, 201523

Kestens et al, 201513

Sonwalkar et al, 201027

AbraduBerchie et al, 201722

Representative of the average adult in the community

1 point, population-based studies; 0.5 point, multicenter studies; 0 point, single-center hospital-based study

0

0

0

0

0

1

0

0.5

Cohort size

1 point, cohort size >2000; 5 points, cohort size 100-200; 0 point, cohort size <100

1

1

1

0

0

1

0

0

Histologic confirmation of BE

1 point, confirmed by consensus of 2 expert pathologists; 0.5 point, reviewed by 1 expert pathologist; 0 point, reviewed by community pathologist only or not reported in the study

1

1

1

1

1

1

1

1

Follow-up of cohort for outcome to occur

1 point, mean follow-up of entire cohort >5 years; 0.5 point, cohort follow-up 3-5 years; 0 point, mean follow-up of cohort <3 years

1

0

1

0

1

0

0.5

0.5

Information on duration of follow-up of patients with BE-IND

1 point, reported in study in total person-years; 0.5 point, reported as mean follow-up of BE-IND cohort; 0 point, imputed from entire BE cohort

1

1

1

0.5

0.5

1

0.5

0.5

Attrition rate

1 point, >80% of cohort followed up; 0.5 point, 60%-80% of cohort followed up; 0 point, >40% lost to follow-up

0

0

1

1

1

1

1

1

Information on progression of BE-IND

1 point, adequate information on rate of progression from BE-IND to BE-HGD and EAC separately; 0.5 point, only information on rate of progression to EAC, without information on BE-HGD

1

1

1

1

1

1

1

1

Total score (maximum Z 7; high quality 5; medium quality 3-4; low quality 2)

5

4

6

3.5

4.5

6

4

4.5

Total

BE, Barrett’s esophagus; IND, indefinite for dysplasia; HGD, high-grade dysplasia; EAC, esophageal adenocarcinoma.

www.giejournal.org

Volume

-,

No.

-

: 2019 GASTROINTESTINAL ENDOSCOPY 8.e3