Primary Motility  Disorders of the  Esophagus
 The Esophageal
 Esophagogastric  Junction

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

Should Barrett's esophagus be treated at all?

S.J. Sontag (Hines)

The management of Barrett's esophagus is controversial. Current surveillance recommendations for patients with documented Barrett's esophagus are predicated on the hope that Barrett's cancer either can be prevented or can be detected early enough so that therapy prevents death from the cancer. The frequency of surveillance endoscopies needed to accomplish this goal is not known. Frequency intervals vary from every 3 months for Barrett's with high-grade dysplasia (HGD) to every 3-5 years for Barrett's without dysplasia. Thus, if the goal of treatment is early detection of esophageal cancer, it is imperative to know something about the natural history of the disease.

It is generally accepted that the development of Barrett's cancer follows a dysplasiacarcinoma sequence: intestinal metaplasia (IM) without dysplasia progresses to IM with low-grade dysplasia (LGD) to IM with HGD to cancer [1]. The factors that encourage or discourage the leap to cancer remain unidentified. A few facts, however, are generally accepted:
- the vast majority of patients with Barrett's esophagus are rather healthy, and thus remain undiagnosed;
- the vast majority of patients with Barrett's cancer come to endoscopy when it is too late.

Based on our 20-year Hines VA database, the prevalence and incidence of Barrett's esophagus and Barrett's cancer as follows

- Prevalence:
  - 15% of all adults with gastroesophageal reflux (GER) symptoms have Barrett's esophagus;
  - 0.7% of all adults with GER symptoms will have Barrett's cancer at first endoscopy.

- Incidence:
  - 2% of all Barrett's esophagus patients will develop Barrett's cancer during a mean of 7 years (0.3% per year);
  - 15% of all Barrett's HGD patients will develop Barrett's cancer during a mean of 7 years (2.2% per year).

Based on national statistics, the importance of esophageal cancer as the percentage of all cancers, the percentage of all cancer deaths and the esophageal cancer death rate can be placed into perspective

- Figure 1a shows that esophageal cancer is uncommon. Esophageal cancer in 1998 represented only 1% of all new cancers [1].
  - # of new esophageal cancers: 12,200
  - total # of cancers: 1,220,000
  - percentage: 1%

- Figure 1b shows that esophageal cancer in 1998 represented only 2% of all cancer deaths[1].
  - # of esophageal cancer deaths: 11,800
  - total # of cancer deaths: 578,200
  - percentage: 2%

- Figure 1c shows the percentage of patients with esophageal cancer that died.
  - 97% of esophageal cancer patients died (11,800 of 12,200)
  - 3% did not die (400 of 12,200)

From these statistics, it appears that only a small minority of patients with Barrett's esophagus will go on to develop HGD and eventually cancer. Indeed, only 12,200 new cases of esophageal cancer are diagnosed each year, and about half of those are squamous cell carcinomas. Thus, about 6,000 cases of Barrett's cancer occur each year, and almost all are diagnosed when it is too late.

It seems, therefore, that if any major decrease in the mortality of Barrett's cancer is to be achieved, it will be brought about only by the large-scale identification of patients with Barrett's esophagus who do not yet have cancer. At the present time, such identification could occur only by massive screening of individuals at high risk for Barrett's and cancer.

Figure 1. A. In 1998, esophageal cancer represented only 1% of all new cancers in the US. B. In 1998, esophageal cancer represented only 2% of all cancer deaths in the US. C. In 1998, 97% of patients diagnosed with esophageal cancer died.

Such individuals are otherwise healthy, although heartburn symptoms may make them easily identifiable. Indeed, a recent study from Sweden showed that the presence of heartburn, regurgitation or both at least one time per week increases the risk of esophageal cancer 8-fold [2]. If the symptoms are frequent, severe and of 20 years duration, the cancer risk increases 44-fold. In order to substantially reduce the esophageal cancer death rate, this group would need to be targeted for early endoscopic examination. Anything short of a fullblown attack on this group with symptomatic GER will necessarily fail to achieve that goal. On the other hand, a full-blown attack on this group would require massive investment of time and energy by currently available endoscopists.

Figure 2 is a decade-old poll demonstrating that 44% of adult Americans have heartburn at least once monthly and 18% take indigestion medication at least twice a week [3]. These numbers indicate that 18 million - at minimum - adult Americans would need screening endoscopy to detect Barrett's esophagus, so that they could be enrolled in a surveillance program of additional endoscopies.

Figure 3 compares the incidence of esophageal cancer and colon cancer in the United States. Colon cancer occurs 13 times as often as Barrett's cancer, raising the important question of whether we should divert our resources to the other end of the intestine. Indeed, every 90-minutes in the United States, one individual dies from esophageal cancer, while 13 die from colon cancer.

Given the high prevalence of symptomatic GER in the general population, the low prevalence of Barrett's esophagus in the symptomatic GER population, the very low Barrett's esophagus diagnosis rate in the total Barrett's population, the minuscule incidence of cancer in the Barrett's population, and the dismal survival rate of patients with Barrett's cancer, it is unlikely that a major decrease in mortality from Barrett's cancer will occur in the near future. There is, however, a possibility that a completely different form of therapy may have an impact in the morbidity and mortality of esophageal cancer.

Figure 2. Gallop survey showing that 44% of adult Americans have heartburn at least once monthly and 18% take indigestionmedication at least twice a week.

Figure 3. The incidences of esophageal cancer andcolon cancer.

A different approach

If we must question the wisdom of spending a hunk of our resources on a disease in which 97% of its victims are "goners" from the time of first diagnosis, and less than 3% are potentially treatable or curable, then perhaps we might find some hope in the concept of chemoprevention. Although still untested and unproven, chemoprevention - in the form of COX-2 inhibitors - are now being studied for the prevention and regression of colon polyps and the prevention and regression of Barrett's esophageal dysplasia.


The theoretical goal of cancer chemoprevention is to inhibit or reverse pre-invasive carcinogenesis before the development of clinical cancer. The goals of current clinical studies are to demonstrate regression of existing neoplasia, prevention of recurrent neoplasia, and prevention of first-time neoplasia. The genetic targets for chemoprevention are genetic damage and genetic defects. If chemoprevention, whether through COX-2 inhibitors or some other agents, eventually proves to be a reliable discipline in the reversal of dysplasia or the prevention of the ultimate dysplasia - cancer - its eventual impact would depend completely on how the medical community responded to the following question: "If physicians - and their patients with Barrett's esophagus - could be relatively certain that they would not develop cancer - regardless of the current grade of dysplasia - would the medical community accept in lieu of repeated endoscopies and surgical resection a lifetime daily dose of a chemopreventive agent, such as a COX-2 inhibitor?"


1. Cancer Statistics. CA. A cancer journal for physicians. Am Cancer Soc 1998;48:6-30.

2. Lagergren J, Bergstrom R, Lindgren A, et al. Symptomatic gastroesophageal reflux as a risk factor for esophageal adenocarcinoma. N Engl J Med 1999;340:825-831.

3. Gallup Survey. Heartburn across America. Princeton, NJ: Gallup Orgainzation, 1988:Mar 24.

Publication date: August 2003 OESO©2015