FISH identifies high-grade dysplasia and adenocarcinoma in BE

In Barrett's esophagus (BE), an abnormal columnar epithelium replaces the stratified squamous epithelium that normally lines the distal esophagus. The change in mucosal tissue, which represents an adaptation to reflux-induced injury, is the strongest risk factor for esophageal adenocarcinoma.

The vast majority of people with BE don't develop cancer — the most recent estimated rate of progression is 1.8 per 1,000 patient years of follow-up. Furthermore, population-based data from Mayo Clinic, the Netherlands and elsewhere indicate that a diagnosis of BE does not affect overall life expectancy.

Yet esophageal adenocarcinoma is one of the fastest growing cancers in Caucasian men and has an unusually poor prognosis, with early metastatic disease developing in many patients. The five-year survival rate after surgical treatment of resectable tumors is 5 to 30 percent, with higher survival rates in patients with early-stage cancers.

"This is a big problem for clinicians," says Kenneth K. Wang, M.D., a gastroenterologist at Mayo Clinic's campus in Rochester, Minnesota. "There are so many people with Barrett's who are worried about cancer, but most will never develop it, so there is great debate about the value of surveillance in the average patient."

Furthermore, the current strategy of periodic endoscopic biopsies, which are subject to sampling error and interobserver variation, is unreliable. It is also costly because patients at low risk are surveyed too often. Thus, the goal has been to develop a cost-effective, noninvasive test that would predict progression to cancer with better accuracy than histology.

"What we are trying to do is develop tests like cytology that are easy for anyone to do. Tests and treatments that can only be performed in tertiary centers don't benefit patients or health care in general," Dr. Wang notes.

One option, currently part of clinical practice at Mayo Clinic, is the evaluation of chromosomal instability — high rates of chromosome missegregation — in cytology specimens. Chromosomal instability has been shown to be the most predictive biomarker of cancer development in BE.

Chromosomal instability

Nearly 15 years ago, University of Washington researchers found that BE patients with a loss of heterozygosity (LOH) for p53, a tumor suppressor gene, were at increased risk of progression to high-grade dysplasia and adenocarcinoma. Mutations in p53 have since been reported in more than 90 percent of esophageal adenocarcinomas, and inactivation of the gene may occur early in neoplastic progression. Studies also showed that numerical or structural changes in chromosomes, such as aneuploidy and tetraploidy (4n), were highly predictive of cancer development.

Chromosomal instability can be detected in cytology brushings using fluorescence in situ hybridization (FISH), a technique that utilizes fluorescent-tagged DNA probes to find, among other things, gene copy number abnormalities. Mayo Medical Laboratories uses a multicolor, multitargeted FISH assay with probes to 8q24 (MYC), 9p21 (CDKN2A), 17q12 (ERBB2 or HER2) and 20q13.2 (ZNF217).

A Mayo Clinic study, published in Human Pathology in 2008, found that this probe set identified high-grade dysplasia and adenocarcinoma with 84 percent sensitivity and 93 percent specificity — far better than routine cytology and digital image analysis.

"We looked at many different FISH probes and at basic combinations of all the usual suspects and found this was the most robust set to use," Dr. Wang explains. "A good clinical test doesn't find every genetic abnormality, but it is reproducible and it shows good characteristics day to day and week to week. FISH, a CLIA-approved test, gives that stability."

Still, the test is not used indiscriminately in clinical practice. "There is no sense in using it in people with cancer and high-grade dysplasia. I use it in patients where I'm not sure what to do," Dr. Wang explains. "When biopsies are indefinite for dysplasia, those patients more often than not don't have it. But they're worried and we're worried, so when histology doesn't seem clear, we use the molecular test. We also use it for patients who have no dysplasia but a strong family history of esophageal cancer. If we detect genetic abnormalities, those patients may need to be treated or undergo a more intensive surveillance regimen."

The test may also be useful in predicting response to treatment. Abnormalities in p16, for instance, seem to predict response to ablative therapies.

FISH is expensive — about $800 — "because it's a test that our lab feels needs to be run manually by a highly skilled cytotechnologist rather than a machine. That takes a lot of person-hours of work," Dr. Wang points out. "In the future, as personalized testing becomes more common and the use of such tests more widespread, the cost will likely decrease."

For more information

Fritcher EG, et al. A comparison of conventional cytology, DNA ploidy analysis, and fluorescence in situ hybridization for the detection of dysplasia and adenocarcinoma in patients with Barrett's esophagus. Human Pathology. 2008;39:1128.