What are the most useful non-histological markers of malignant potential?
A.N. Kingsnorth (Liverpool)
Relatively little is known of the molecular genetic alterations that take place during the development of Barrett's cancer from metaplasia . Nevertheless Barrett's esophagus (BE) is an excellent model in which to investigate intermediate events by serial biopsy over time, or an alternative approach is to examine non-histological markers in cohorts of patients having metaplasia, dysplasia and established carcinoma. This will potentially lead to an understanding of the order in which genetic abnormalities develop. Less than 10 oncogenes and tumor suppressor genes have been studied in Barrett's disease, yet it is estimated that 1000-2000 genetic abnormalities occur during the development of carcinoma.
It is convenient to discuss non-histological markers in terms of those that have been studied and found to be irrelevant, those that have been studied and whose role remains questionable and those in which a role has been established, (where expertise is available).
The bcl-2 proto-oncogene encodes a protein that blocks apoptosis and is involved in the development of follicular lymphoma by a chromosomal translocation. The bcl-2 protein is over-expressed during the dysplasia carcinoma sequence in ulcerative colitis. In 36 esophageal resection specimens no bcl-2 immuno-reactivity was detected in BE either in metaplastic tissue, low grade dysplasia, high grade dysplasia or carcinoma . In contrast to its role in colonic neoplasia, bcl -2 alterations are not important molecular markers in neoplastic progression in BE. E-cadherin is a prime mediator of cell-cell interactions in epithelial cells. Down regulation of E-cadherin favors the process of invasion and metastasis . A comparison of biopsy specimens of esophagitis (n = 6), Barrett's metaplasia with or without dysplasia (n = 16), and esophageal adenocarcinoma, showed that normal mRNA E-cadherin levels were found in patients with esophagitis . Reduced or absent expression of E-adherin was found in patients with adenocarcinoma. However, the number of patients was small and confirmation of this study is required.
The ras genes (H-, K- and N-) code for highly related proteins of 189 amino acid residues generically known as p21 . Ras proteins bind GTP and GDP and possess intrinsic GTPase activity and are involved in signal transduction inducing manifestations of the malignant phenotype. Ras genes are the most frequent group of oncogenes so far identified in human cancers and are thought to participate in the early stages of tumour development. Two series, in which a small number of patients were studied, suggest over-expression of H-ras genes in Barrett's cancers with a predisposition for patients that develop metastatic disease [6, 7]. These studies need to be confirmed in larger series. Microsatellites are short-segment repeats of 2-4 nucleotides (tandem repeat DNA sequences) common in introns and show pronounced polymorphism. Microsatellites have been used for gene mapping and linkage analysis, markers of loss of heterozygosity and are a diagnostic tool for detection of tumor cells in histologically unremarkable specimens . Loss of heterozygosity has been shown in the adenomatous polyposis coli and mutated colon cancer and retinoblastoma locus in Barrett's cancers [9, 10]. In an analysis of a specific area on chromosome 17 at 17q11.2-q12, the highest loss on 17q was found at the TCF-2 locus which may represent the site of a predisposing gene in Barrett's adenocarcinoma which is a novel tumor suppressor gene . At the present time these results are difficult to explain although microsatellites might be sensitive indicators of disrupted mechanisms and indicate a propensity to mutagenesis.
The prevalence of c-erbB-2 over-expression in the development of Barrett's cancer is controversial. c-erbB-2 is a growth factor receptor with many similarities to the epidermal growth factor receptor. A transmembrane region connects the glycosylated extracellular domain with an intracellular domain which has inherent tyrosine kinase activity. c-erbB-2 has multiple ligands and is over-expressed in various types of human adenocarcinomas . Although there is consensus that c-erbB-2 over-expression can be detected in Barrett's cancers but not in dysplasia nor in metaplasia [13-16], the findings differ with regard to prognosis. Some studies find that over-expression imparts a poorer prognosis and others that it predicts improved survival and a favorable response to therapy.
Flow cytometry in the measurement of DNA ploidy is a controversial tool. Although reliable and reproducible in expert hands it remains a research modality . Once again, there has been controversy over its utility [18, 19]. Initial studies suggested good correlation between aberrant flow cytometric finding (IEG2* tetraploid fractions greater than 6%) and conventional histological diagnosis of dysplasia and carcinoma. It was suggested that the subset of patients with no dysplasia, but aneuploid cell populations may be at risk of the development of carcinoma. Subsequent studies failed to confirm this finding: Barrett's epithelium without dysplasia having aneuploid cell populations and Barrett's epithelium with dysplasia having aneuploidy. A later study involving a prospective endoscopic surveillance of a cohort of patients showed that a high proportion of those entering this study with aneuploidy or increased G2 or tetraploid patterns were at high risk of developing high grade dysplasia or carcinoma  . The real value of flow cytometry may be in its indicating low or no risk of cancer in patients with no aneuploidy and no dysplasia. Conversely high risk patients are those with both high grade dysplasia and aneuploidy particularly in younger patients .
Many investigators have studied p53 in BE . The reader is referred to Levine's review for an analysis of the literature. p53 functions as a tumour suppressor gene by mediating a cell cycle dependent check point which allows proliferating cells to repair DNA damage and prevents propagation of genetic errors that may lead to cancer. Immunohistochemical staining for p53 increases in frequency through metaplastic, dysplastic to adenocarcinoma. Positive p53 staining occurs in 0%, 9%, 55% and 87% of specimens interpreted as negative for dysplasia, indefinite-low grade dysplasia, high grade dysplasia, or adenocarcinoma, respectively. Moreover during follow-up patients with p53 staining progress to higher grades of dysplasia. However the role of assays for p53 gene mutations or p53 protein over-expression in the clinical management of patients with BE is unclear. This is because p53 immuno staining is not as sensitive as molecular genetic assays. This could also result in a high frequency of false positive and false negative results. A further refinement of measurement of p53 abnormalities would be to combine this with DNA ploidy abnormalities. Research to date indicates that p53 abnormalities precede the development of aneuploidy and are an early development in the genetic instability observed in the multistep progression to Barrett's adenocarcinoma.
In conclusion molecular genetics is on the threshold of delivering non-histological markers of malignant potential, the closest to realization in clinical practice are p53 and DNA ploidy.
4. Swami S, Kumble S, Triadafilopoulos G. E-cadherin expression in gastroesophageal reflux disease, Barrett's esophagus, and esophageal adenocarcinoma: an immunohistochemical and immunoblot study. Am J Gastroenterol 1995;90:1808-1813.
6. Meltzer SJ, Mane SM, Wood PK, Resau JH, Newkirk C, Terzakis JA, Karelitzs Bl, Weinstein WM, Needleman SW. Activation of c-Ki-ras in human gastrointestinal dysplasias determined by direct sequencing of polymerase chain reaction products. Cancer Res 1990;50:3627-3630.
7. Sorsdahl K, Casson AG, Troster M, Van Meyei D, Inculet R, Chambers AF. p53 and ras gene expression in human esophageal cancer and Barrett's epithelium; a prospective study. Cancer Detect Prev 1994;18:179-815.
9. Boynton RF, Blount PL, Yin J, Brown VL, Huang Y, Tong Y, McDaniel T, Newkirk C, Resau JH, Raskind WH, Haggitt RC, Reid BJ, Meltzer SJ. Loss of heterozygosity involving the APC and MCC genetic loci occurs in the majority of human esophageal cancers. Proc Natl Acad Sci USA 1992;89:3385-3388.
10. Boynton RF, Huang Y, Blount PL, Reid BJ, Raskind WH, Haggitt RC, Newkirk C, Resau JH, Yin J, McDaniel T, Meltzer SJ. Frequent loss of heterozygosity at the retinoblastoma locus in human esophageal cancers. Cancer Res 1992;52:5766-5769.
13. Fléjou JF, Paraf F, Muzeau F, Henin D, Jothy S, Potet F. Expression of c-erbB-2 oncogene product in Barrett's adenocarcinoma; pathological and prognostic correlations. J Clin Pathol 1994;47:23-26.
14. Nakamura T, Nekarda H, Hoelscher AH, Bollschweiler E, Harbec N Becker K, Siewert JR. Prognostic value of DNA ploidy and c-erbB2 oncoprotein overexpression in adenocarcinoma of Barrett's esophagus. Cancer 1994;73:1785-1794.
18. Reid BJ, Haggitt RC, Rubin CE, Rabinovitch PS. Barrett's esophagus; correlation between flow cytometry and histology in detection of patients at risk for adenocarcinoma. Gastroenterology 1987;93:1-11.
20. Reid BJ, Blount PL, Rubin CE, Levine DS, Haggitt RC, Rabinovitch PS. Flow-cytometric and histological progression to malignancy in Barrett's esophagus: prospective endoscopic surveillance of a cohort. Gastroenterology 1992;102:1212-1219.