Primary Motility  Disorders of the  Esophagus
 The Esophageal
 Esophagogastric  Junction

  Browse by Author
  Browse by Movies
Volume: Barrett's Esophagus
Chapter: Dysplasia

What is the value of the sucrase-isomaltase expression in detecting dysplastic changes in Barrett's esophagus?

P. Chaves, J. Soares (Lisbon)

Barrett's esophagus is the premalignant condition related to esophageal adenocarcinoma [1, 2] and, probably, to cardia adenocarcinoma [3, 4]. It results from the replacement of normal squamous epithelium by a metaplastic columnar lining [5, 6] associated to prolonged gastroesophageal or bile reflux [7, 8]. Three histological subtypes of Barrett's mucosa are recognized [9], the columnar epithelium with specialized intestinal metaplasia (IM), characterized by the presence of goblet cells as part of the metaplastic elements [9], being the only associated with increased risk for malignant transformation [10].

The recognition of an unequivocal metaplastic esophageal lining in endoscopic biopsies is a critical practical issue for the identification of Barrett's esophagus patients [6], goblet cells being a precious tool for the morphological diagnosis [11]. Goblet elements present on the distal esophagus, are unequivocally metaplastic [11], and also frequently contained on segments adjacent to Barrett's adenocarcinoma [10], being considered as diagnostic [11] and prognostic [10] hallmarks for management of patients with a columnar-lined esophagus.

The biopathological mechanisms underlying the metaplastic process in esophageal mucosa are mostly unknown, as it occurs in many other type of tissues, and the controversies surrounding the surveillance of Barrett's esophagus patients really stem from this limited knowledge. The recognition of intestinal differentiation of the columnar epithelium lies on the identification of goblet elements assumed as the risk factor for malignant transformation [1, 6, 10, 11]. Nevertheless, the metaplastic epithelium is formed by an heterogeneous cell population and the role of the columnar non-goblet elements on neoplastic transformation is unclearly understood. Based on ultrastructural studies it was recognized that columnar metaplastic cells are a complex mixture of cells with apical microvilli of variable size and development [12] and with precursor mucous vesicles [13] suggesting they share features of both gastric and intestinal differentiation.

Recently, Wu et al. [14], using molecular and immunohistochemical methods, demonstrated that columnar cells of Barrett's esophagus possess sucrase-isomaltase, an enterocytic disaccharidase also expressed on Barrett's adenocarcinomas. Using PCR, they detected sucrase-isomaltase in 76% of the Barrett patients and in 82% of those with esophageal adenocarcinomas, while they found no sucrase-isomaltase in squamous cell carcinomas (SCC). By immunohistochemical methods the authors confirmed the absence of sucrase-isomaltase in squamous cell carcinomas and found sucrase-isomaltase to be expressed in 88% of the Barrett's esophagus and in 81% the adenocarcinomas. Wu et al. study proved the intestinal phenotype of the columnar elements of Barrett's epithelium as well as that of the associated neoplasia. The distinct patterns of staining, mostly apical in metaplasia and predominantly cytoplasmatic in neoplasia, allowed the authors to admit the disruption of post-transcriptional sucrase-isomaltase machinery during neoplastic transformation. These results, suggesting a continuum from metaplasia to neoplasia, called the attention to the role of the columnar non-goblet cells in the biologic process of malignancy, the authors stating that columnar cells expressing sucrase-isomaltase may become dysplastic leading to adenocarcinoma. The methodology used by Wu et al. [14] did not allow the evaluation of sucrase-isomaltase expression in the different types of Barrett's mucosa and they assumed that only specialized Barrett's epithelium would express sucraseisomaltase.

The study undertaken by Almeida et al. [3, 4] supports the intestinal differentiation not only of esophageal but also of most cardia adenocarcinomas, since they expressed enterocytic markers, namely sucrase-isomaltase, and sustains the importance of Barrett's esophagus as the precursor lesion for both neoplasia. They found sucrase-isomaltase to be present in 93% and 96% of the metaplasia and the neoplasia, respectively. The glycoprotein expression in metaplasia was independent of the epithelial type, the existence of dysplasia, or the presence of adenocarcinoma. In neoplasia sucrase-isomaltase expression was independent of the presence of Barrett's esophagus, or tumour location (esophagus versus cardia).

Similar results were obtained by Iannettoni et al. [15], who found sucrase-isomaltase expression in 55% of Barrett's adenocarcinomas and 44% of cardia adenocarcinomas. In 100% of the cardia sucrase-isomaltase positive tumors there was no associated Barrett's esophagus, while 31% of the cardia adenocarcinomas with no recognized Barrett's esophagus expressed sucrase-isomaltase. The methodology used by Almeida et al. [3, 4] and by Iannettoni et al. [15], allowed to demonstrate the intestinal phenotype of esophageal and cardia adenocarcinomas supporting their relationship with Barrett's esophagus, but did not make the individual assessment of sucrase-isomaltase expression at the different subtypes of Barrett's epithelium.

In a recent study using markers of enterocytic differentiation, namely sucrase-isomaltase and dipeptidilpeptidase i.v. (DPP), we [16] explored the intestinal phenotype of Barrett's esophagus columnar cells and Barrett's adenocarcinoma cells in paraffin-embedded samples from 12 surgical esophagectomy specimens. We studied separate areas of columnar epithelium with and without specialized IM, sucrase-isomaltase and DPP being exclusively expressed by the columnar cells. We found sucrase-isomaltase in 58.3% of the Barrett's adenocarcinomas and in 66.6% of the Barrett's esophagus with and without specialized IM (Figure 1). DPP was identified in 66.6% of the carcinomas, in 50.0% of the metaplastic areas with specialized IM and in 75.0% of the areas without specialized IM. These results support the involvement of the Barrett's esophagus columnar elements on Barrett's malignant transformation and suggest the presence of non-goblet elements with an intestinal phenotype in Barrett's esophagus areas with or without specialized IM. The demonstration of intestinal differentiation in areas of Barrett's esophagus without specialized IM contradicts Weinstein's view point, that advocate Barrett's esophagus diagnosis based, exclusively, on the identification of goblet cells on the metaplastic segment.

Figure 1. Sucrase-isolmatase and dipeptidilpeptidase expression in Barrett's esophagus without (I) and with (II) specialized intestinal metaplasia and in Barrett's adenocarcinoma (III).

The question about the value of sucrase-isomaltase for the detection of dysplastic changes stems on the incomplete understanding of the mechanisms of sucrase-isomaltase synthesis and accumulation during Barrett's esophagus progression towards malignancy. Aiming at investigating the enterocytic phenotype of the columnar lined esophageal (CLE) segments with and without associated adenocarcinoma, we evaluated sucrase-isomaltase immunoexpression in three different groups of CLE segments [17] (Figure 2). Group I included long segments of CLE segments without specialized IM. Group II was formed by short and long segments of CLE segments with specialized IM (Barrett's esophagus) [11] and was divided into the subgroups IIA (areas without specialized IM) and IIB (areas with specialized IM). Group I and Group II had no associated dysplasia or carcinoma. Group III was composed by columnar metaplastic segments adjacent to Barrett's adenocarcinoma and was also divided in areas without specialized IM (IIIA) and areas with specialized IM (IIIB). In metaplasia areas, we found sucrase-isomaltase to be expressed exclusively by the columnar elements, the apical staining being associated to the presence of specialized IM. In CLE segments without specialized IM (Group I) sucrase-isomaltase was only detected at the cell cytoplasm in 21.4% of the cases. Sucrase-isomaltase apical expression was observed in Barrett's esophagus with and without carcinoma either in areas with (Figure 3) or without specialized IM (Figure 4). In Barrett's esophagus without carcinoma sucraseisomaltase was present at the apical membrane more frequently in specialized IM areas (p = 0.041). The apical staining predominated at specialized IM areas of Barrett's esophagus adjacent to adenocarcinoma but the quantitative difference in relation to areas without specialized IM was not statistically significant [17]. The apical expression of sucrase-isomaltase was more frequently observed in Barrett's esophagus adjacent to adenocarcinoma either in areas with (p = 0.018) and without specialized IM (p = 0.003). There was lower expression of apical sucrase-isomaltase in neoplasia than in columnar cells of adjacent Barrett's esophagus, but the quantitative differences had no statistical significance. These results favour the relationship between the enterocytic phenotype of Barrett's esophagus and malignancy, supporting the use of Barrett's esophagus intestinal differentiation as a morphological marker for cancer risk. They also confirm previous observation [16] that intestinal features are underestimated if solely relying on the identification of goblet elements and therefore they highlight the utility of immunocytochemical methods to recognize enterocytic differentiation.

Figure 2. Sucrase-isomaltase expression in Barrett'sesophagus without (I, IIA and IIB) and with associated Barrett's adecarcinoma (IIIA and IIIB). I. Columnar lined esophageal segment without specialized IM; IIA. Barrett's esophagus areas without specialized IM; IIB. Barrett's esophagus areas with specialized IM; IIIA. Barrett's esophagus without specialized IM adjacent to neoplasia ; IIIB. Barrett's esophagus with specialized IM adjacent to neoplasia.

The localization of sucrase-isomaltase was investigated at cellular level, using preembedding immunoelectromicroscopy technical procedure [18]. The glycoprotein was focally expressed by some columnar cells (Figure 5) at the membrane of the microvilli. Two patterns were observed: a strong immunostaining in cells with abundant, long and thin microvilli with well developed core rootlets and a weak expression in cells with sparse,

Figure 3. Sucrase-isolmatase apical immunoexpression on Barrett's esophagus areas with specialized IM ; x 300.

Figure 4. Sucrase-isolmatase apical immunoexpression on Barrett's esophagus areas without specialized IM ; x 300.

Figure 5. Sucrase-isolmatase immunoelectromicroscopic expression at the columnar cells membrane microvilli; x 45,000.

short and thicker microvilli without or with ill-defined core rootlets. Immunoelectromicroscopy results confirm previous immunocytochemical studies and suggest that some columnar cells of Barrett's esophagus may either exhibit a mature or an immature enterocytic phenotype.


The demonstration of sucrase-isomaltase in Barrett's esophagus columnar cells in areas with and without specialized IM suggests the existence of an enterocytic differentiation pathway that can be underestimated on routine hematoxilin-eosin histology. This fact highlights the use of immunohistochemistry to solve inconclusive daily practice cases.

Sucrase-isomaltase expressed by either metaplastic or neoplastic cells in the setting of Barrett's esophagus allows to admit that the enterocytic phenotype, even incompletely expressed, might constitute a marker for cancer risk to be considered on surveillance of Barrett's esophagus patients. On the other hand, the pattern of sucrase-isomaltase expression on CLE segments without and with Barrett's adenocarcinoma emphasize the role of columnar cells with enterocytic differentiation on Barrett's esophagus malignant transformation.


1. Spechler JS, Robbins AH, Rubins HB, et al. Adenocarcinoma and Barrett's oesophagus. An overrated risk? Gastroentology 1984;87:927-933.

2. Cameron AJ, Ott BJ, Payne WS. The incidence of adenocarcinoma in columnar-lined (Barrett's) oesophagus. N Engl J Med 1985;313:857-859.

3. Almeida JCM, Rajasekaran, Godwin T, Quaroni A,Rodriguez-Boulan E, Altorki NK. Carcinoma of the cardia, Barrett's oesophagus and adenocarcinoma of the oesophagus and cardia ubiquitously express Sucrase-isomaltase and crypt cell antigen. Dis Esophagus 1996;9:191-197.

4. Almeida JCM, Chaves P, Dias Pereira A, Altorki NK. Is Barrett's esophagus the precursor of most adenocarcinomas of the esophagus and cardia? A biochemical study. Ann Surg 1997;226:725-733.

5. Barrett NR. The lower oesophagus lined by columnar epithelium. Surgery 1957;41:881-894.

6. Spechler JS, Goyal RK. The columnar-lined esophagus, intestinal metaplasia and Norman Barrett. Gastroenterology 1996;110:614-621.

7. Chow WH, Finkle HD, McLaughlin JK, Ziel HK, Fraumeni JF Jr. The relationship of gastroesophageal reflux disease and its treatment to adenocarcinomas of the oesophagus and gastric cardia. JAMA 1995;274:474-477.

8. Meyer W, Vollmar F, Barr W. Barrett's esophagus following total gastrectomy. Endoscopy 1979;11:121-126.

9. Paull A, Trier JS, Dalton MD, Camp RC, Loeb P, Goyal RK. The histologic spectrum of Barrett's oesophagus. N Engl J Med 1976;295:476-480.

10. Haggit RC. Barrett's esophagus, dysplasia and adenocarcinoma. Human Pathol 1994;25:982-993.

11. Weinstein WM, Ippoliti AF. The diagnosis of Barrett's esophagus:goblets, goblets, goblets. Gastrointest Endosc 1996;44:91-95.

12. Levine DS, Rubin CE, Reid BJ, Haggitt RC. Specialized metaplastic columnar epithelium in Barrett's esophagus. Lab Invest 1989;60:418-432.

13. Trier JS. Morphology of the columnar cell-lined (Barrett's) esophagus. In: Goyal RK, Spechler SJ, eds. Barrett's esophagus: pathophysiology, diagnosis and management. New York: Elseveir, 1985:19-28.

14. Wu GD, Beer DG, Moore JH, Orringer MB, Appelman HD, Traber PG. Sucrase-isomaltase gene expression in Barrett's esophagus and adenocarcinoma. Gastroenterology 1993;105:837-844.

15. Iannetttoni MD, Lee SS, Bonell MR, Sell TL, Whyte RI, Orringer MB, Beer DG. Detection of Barrett's adenocarcinoma of the gastric cardia with sucrase-isomaltase and p53. Ann Surg 1996;62:1460-1465.

16. Chaves P, Cardoso P, Almeida JCM, Pereira AD, Leitão CN, Soares J. Non-goblet cell population of Barrett's oesophagus: an immunohistochemical demonstration of intestinal differentiation. Human Pathol 1999;30:1291-1295.

17. Chaves P, Cruz C, Cardoso P, Suspiro A, Pereira AD, Mendes de Almeida JC, Leitão CN, Soares JS. The enterocytic differentiation of Barrett's epithelium with and without associated cancer. 6th OESO World Congress 2000 (Abstract p. 67).

18. Chaves P, Moura Nunes J, Cruz C, Pereira AD, Leitão CN, Soares J. The enterocytic differentiation of columnar nongoblet cells on Barrett's oesophagus. 6th OESO World Congress 2000 (Abstract p. 66).

Publication date: August 2003 OESO©2015