|Home | About | Journals | Submit | Contact Us | Français|
Our current management of cancer risk in Barrett’s esophagus is to perform endoscopic surveillance for the detection of dysplasia. However, dysplasia is an imperfect predictor of cancer risk for a variety of reasons including biopsy sampling error, poor intra- and inter-observer reproducibility of dysplasia interpretations and the poor predictive value for negative, indefinite, low-grade, and even high-grade dysplasia.1–3 Dysplasia is a conglomerate of histologic abnormalities that suggest that clones of cells have acquired neoplastic properties that predispose them to cancer formation. Therefore, dysplasia is a surrogate marker for cells that have accumulated enough genetic damage that they now possess some of the physiologic properties of cancer cells. Therefore, a better indicator of cancer risk would be detection of the genetic damage itself before the histologic manifestations of dysplasia are even apparent. In addition, the identification of molecular biomarkers may offer easy reproducibility and standardization in addition to the truly early detection of neoplastic progression.
In the traditional phenotypic model, carcinogenesis in Barrett’s esophagus is viewed as occurring in discrete steps from metaplasia to dysplasia and finally to carcinoma. In the genetic model, neoplastic progression is envisioned as a continuum over which cells progressively accumulate genetic abnormalities until they acquire the six essential physiologic hallmarks of cancer.4 These cancer hallmarks include the ability of cells (1) to provide their own growth signals, (2) to avoid growth inhibitory signals, (3) to avoid apoptosis, (4) to replicate without limit, (5) to sustain angiogenesis (the formation of new blood vessels), and (6) to invade and metastasize. These hallmarks represent the physiologic traits that must be acquired by cells during the genesis of all human tumors and, therefore, are not specific for neoplastic progression of Barrett’s esophagus. However, there are differences among human tumors regarding the specific genetic alterations acquired that endow the cell with each of these physiologic hallmarks and we will highlight some of the genetic alterations that occur in Barrett’s esophagus which allow the cell to acquire each of the hallmarks.
In general, this occurs by the activation of oncogenes. Genes that stimulate cell growth in normal cells are termed protooncogenes. When these same genes become overactive as a result of certain types of mutations, they are called oncogenes. Thus, oncogene activation leads to uncontrolled cell growth. Examples of oncogenes implicated in Barrett’s carcinogenesis include cyclins D1 and E, transforming growth factor-α, epidermal growth factor, and the epidermal growth factor receptor.5
In general, growth inhibitory signals are transmitted by tumor suppressor genes. Tumor suppressor genes are normal genes that restrain cell proliferation. When tumor suppressor genes are inactivated, the cells are able to avoid growth inhibitory signals allowing for uncontrolled proliferation. Mutation, deletion of the chromosomal region containing the gene (called loss of heterozygosity (LOH)), and attachment of methyl groups to the promoter region of genes (called promoter hypermethylation) are ways in which tumor cells can inactivate tumor suppressor genes. Examples of tumor suppressor genes inactivated during neoplastic progression of Barrett’s esophagus include p53, p16, and the adenomatous polyposis coli (APC) gene.5
Apoptosis is a pre-programmed mechanism for normal cells to self-destruct. This is beneficial to normal cells, in that it prevents cells with damaged, mutated DNA from undergoing replication. However, to cancer cells, apoptosis is detrimental, and cancer cells must find ways to avoid self-destruction. Barrett’s cells have found a variety of ways to overcome triggering apoptosis. For example, as already discussed, inactivation of p53 is one way in which Barrett’s cells avoid inducing apoptosis in response to DNA damage or mutation. Another way Barrett’s cells avoid apoptosis is by the upregulation of cycloxygenase-2, a gene whose protein product exerts antiapoptotic effects. Finally, Barrett’s cancer cells have been found to express Fas-ligand, a death-promoting ligand that can activate the apoptotic cascade within the tumor killing immune cells resulting in their destruction.6,7
Normally, as cells undergo successive cell divisions, they reach senescence. Senescence is an intrinsic mechanism of cells that limits their normal proliferative capacity. The triggering of senescence involves the loss of telomeres which are repetitive pieces of DNA located at the ends of chromosomes. When telomeres become too short, senescence ensues. Telomerase is the enzyme that allows for the synthesis and stabilization of telomeres.8 Stable telomeres confer immortality to the cell. In contrast to normal esophageal tissues, benign Barrett’s esophagus expresses low levels of telomerase, which appears to increase as the metaplastic cells progress to high-grade dysplasia and cancer.9
In order for a tumor to increase in size, it must maintain an adequate blood supply. The synthesis of new blood vessels is termed angiogenesis. One way in which tumor cells synthesize new blood vessels is by secreting vascular endothelial growth factors (VEGFs) which promote the proliferation and migration of endothelial cells upon binding to their receptors, the vascular endothelial growth factor receptors (VEGFRs). The expression of VEGFs and VEGFRs has been found in metaplastic Barrett’s esophagus as well as in neoplastic Barrett’s tissues.10,11
Although the mechanisms of cancer cell invasion and metastasis are poorly understood, abnormalities in cell–cell interaction mediated by cadherins and catenins are thought to play a role.12 In the neoplastic progression of Barrett’s esophagus, the normal membraneous location of E-cadherin and β-catenin decreases, and the cytoplasmic and nuclear staining for these proteins increases as the degree of dysplasia increases.13 In addition, Barrett’s cancers have been found to express matrix metalloproteases which degrade the extracellular matrix and facilitate invasion.14
Some of the individual abnormalities described above have been proposed as biomarkers for cancer risk in Barrett’s esophagus, and it is likely that a few of these will eventually become clinically useful. The National Cancer Institute’s Early Detection Network has proposed five phases of study that biomarkers must undergo for validation.15 It is only in the latter three phases that clinical studies are carried out to (1) evaluate retrospectively the predictive ability of the biomarker and to define a “positive” test (phase 3), (2) prospectively determine the predictive ability of the biomarker (phase 4), and (3) estimate the reduction in mortality by action taken based on the biomarker assay (phase 5).16 The majority of the biomarkers proposed for Barrett’s esophagus have been evaluated in phase 3 studies, and none of these potential biomarkers have been evaluated in phase 5 studies. The Barrett’s biomarkers that have shown the most promise thus far include aneuploidy and increased tetraploidy, 17p LOH, and 9p LOH.
Aneuploidy does not indicate a single genetic abnormality but rather refers to an alteration in the normal diploid (2n) or tetraploid (4n) (where n refers to the number of chromosomes) DNA content of a cell. Thus, aneuploidy reflects the accumulation of multiple genomic abnormalities like the ones discussed above. Aneuploidy can be detected by flow cytometry or by fluorescence in situ hybridization (FISH); however, FISH may be less sensitive than flow cytometry in the detection of chromosomal abnormalities.17,18 A number of large prospective studies have found that aneuploidy and/or increased tetraploidy (>6% of cells within a tissue with 4n) are significant predictors of cancer risk in Barrett’s esophagus.3,19,20
17p is the chromosomal locus for p53, and a number of studies have investigated the ability of 17p LOH to predict neoplastic progression of Barrett’ esophagus.17,21–23 In two large prospective studies 17p LOH, as detected by flow cytometry, in baseline biopsy specimens of Barrett’s esophagus was found to be a significant predictor of neoplastic progression regardless of the degree of dysplasia.21,22 More recently, a number of cross-sectional studies have reported promising results on the ability of 17p LOH, detected by FISH in biopsy tissues and in brush cytology specimens of Barrett’s esophagus, to predict neoplastic progression.17,23
In a large, prospective study, the ability of this combination of biomarkers to predict neoplastic progression in Barrett’s esophagus was found to be better than any of the biomarkers used alone.22 The incidence of cancer was 80% at 6 years in those patients whose biopsies contained all three abnormalities, whereas the incidence of cancer was 12% at 10 years in those patients whose biopsies did not demonstrate any of these abnormalities.22 Thus, it is likely that a panel of biomarkers will be better predictors of neoplastic progression in Barrett’s esophagus than a single individual genetic abnormality.
Overall, although the results of these studies are promising, the use of these biomarkers in routine clinical practice is not yet recommended. In light of the recent advances in biomarker discovery, it is likely that combinations of molecular biomarkers will eventually be better predictors of neoplastic progression than dysplasia. Large, prospective clinical trials of candidate biomarkers (phases 4 and 5 studies) for detecting cancer arising in Barrett’s esophagus are eagerly awaited.
This work was supported by the Office of Medical Research, Department of Veterans Affairs, Dallas, TX (R.F.S.) and the National Institutes of Health (R01-DK63621 to R.F.S).
This paper was originally presented as part of the SSAT/AGA/ASGE State-of-the-Art Conference, Barrett’s Esophagus, Dysplasia, and Early Esophageal Adenocarcinoma: Managing the Transition, at the SSAT 50th Annual Meeting, June 2009, in Chicago, Illinois. The other articles presented in the conference were Sarosi GA, Introduction: Barrett’s Esophagus, Dysplasia, and Early Esophageal Adenocarcinoma: Managing the Transition; DeMeester SR, Reflux, Barrett’s and Adenocarcinoma of the Esophagus: Can We Disrupt the Pathway?; Wang KK, Endoscopic Treatment for Barrett’s Esophagus and Early Esophageal Cancer; and Pennathur A and Luketich JD, Minimally Invasive Esophagectomy for Barrett’s with High Grade Dysplasia and Early Adenocarcinoma of the Esophagus.