RSS Feeds
Adenocarcinoma of the Esophagus and Gastric Cardia: Is There Progress in the Face of Increasing Cancer Incidence?
From the late 1970s to the mid-1980s, the incidence of adenocarcinoma of the esophagus and gastric cardia increased more rapidly than that of any other cancer in the United States (1). Adenocarcinoma of the esophagus has continued to increase in frequency in the early 1990s. An estimated 25 000 people die every year of these cancers in the United States, and a similar trend has been noted in western Europe (2).
Adenocarcinoma of the esophagus itself usually occurs in patients who have Barrett esophagus (3). Adenocarcinoma of the gastric cardia and esophagogastric junction is associated with Barrett esophagus about 40% of the time (4). Unfortunately, the terminology for defining adenocarcinoma of the proximal stomach is not standardized. The International Classification of Diseases for oncology lumps together cancers of the cardia and the gastroesophageal junction. For example, the origin of a tumor may be described as the gastric cardia “when the epicenter is at the gastroesophageal junction” (4) or the esophagogastric junction “when the vertical midpoint of the tumor was ≥ 2 cm above or below the esophagogastric junction” (5).
Barrett esophagus is, in turn, a complication of gastroesophageal reflux disease (GERD). Most patients with Barrett esophagus have severe GERD, as judged by predisposing pathophysiologic factors: weak lower-esophageal sphincter pressure, decreased amplitude of contractions in the distal esophagus, increased acid exposure, increased bile acid exposure, and presence of a hiatal hernia.
Recognition of Barrett esophagus has increased with the widespread application of flexible upper endoscopy, the most sensitive method of detecting Barrett esophagus (6). Most investigators would agree that intestinal metaplasia documented on directed biopsy specimens obtained from an abnormal-appearing esophageal lining is also necessary for the diagnosis of Barrett esophagus (7).
Why Is the Incidence of Adenocarcinoma Increasing?
The frequency of heartburn, the leading symptom of GERD, has not changed in the past two decades (8). This makes it hard to postulate that the frequency of GERD has increased. Familial aggregation of GERD symptoms has recently been documented in patients with Barrett esophagus and esophageal adenocarcinoma (9), offering an opportunity to study the genetics of this disease. We do not know whether the incidence of Barrett esophagus is increasing. Although Helicobacter pylori infection is no more common in patients with Barrett esophagus than in controls, cagA-positive strains may protect against the development of Barrett esophagus and adenocarcinoma of the esophagus (10). If this relation is confirmed, H. pylori eradication may need to be more selectively applied. The risk factors for Barrett esophagus are difficult to identify because the condition is defined at the time of endoscopy and biopsy for symptoms unrelated to Barrett esophagus itself. In contrast, because of dysphagia related to the neoplasm, adenocarcinoma becomes clinically apparent in a short time.
Information about the epidemiology of adenocarcinoma of the esophagus is emerging. High body mass index and smoking are risk factors for adenocarcinoma of the esophagus (11); evidence for alcohol is mixed (12, 13). A recent review of large numbers of surgeries involving resection of adenocarcinoma associated with Barrett esophagus highlights the point that most patients with Barrett esophagus are male and white (14). The concern that certain therapeutic interventions—specifically, H2-receptor antagonists—may account for the increase in incidence of adenocarcinoma was not supported by a case–control study of 196 patients with adenocarcinoma of the esophagus and gastric cardia (15).
Thus, although demographic features of patients with adenocarcinoma are now becoming clearer, risk factors for this disease are not yet well defined, and it is still too early to develop specific interventions that might decrease its incidence.
Screening for Barrett Esophagus
Accumulating data show that patients with Barrett esophagus have an earlier age of heartburn onset and a longer duration of GERD symptoms than patients with GERD who do not have Barrett esophagus (16). Opinion still outweighs evidence, however, in developing screening guidelines to detect patients with Barrett esophagus. Because most adenocarcinomas occur in white men, intuitively it would make sense to screen older white men with a longer duration of GERD symptoms (7). This would, of course, define Barrett esophagus only in symptomatic patients presenting for medical care. Asymptomatic persons can have Barrett esophagus, but the frequency of this occurrence is unknown and can be defined only through population studies.
Surveillance of Patients with Barrett Esophagus
A randomized trial to document the efficacy of surveillance as a means of decreasing death caused by adenocarcinoma in patients with Barrett esophagus would require 360 to 5000 patients followed for 10 years (17). Such a trial would be difficult and expensive to perform. In the absence of such evidence, decision analysis has been used to estimate the benefits of surveillance strategies. Annual endoscopic surveillance for patients with Barrett esophagus and esophagectomy for those with high-grade dysplasia is estimated to have the greatest survival benefit. Endoscopy every 2 to 3 years provides the greatest quality-adjusted life expectancy; endoscopy every 5 years provides the greatest incremental cost-effectiveness. Such estimates are sensitive to the incidence of adenocarcinoma in patients with Barrett esophagus. The most recent prospective series has the lowest incidence yet described—0.5% per year (18). This incidence is at the margin at which the risks of esophagectomy outweigh the benefits of surveillance in the resulting length and quality of life.
Although evidence for the efficacy of surveillance is lacking, the need to deal with individual patients remains. Thus, data from the few prospective series of dysplasia have been used to develop guidelines. After dysplasia is determined not to be present in patients with Barrett esophagus, extension of surveillance to 2- to 3-year intervals is recommended. If low-grade dysplasia is found and concomitant cancer is ruled out, surveillance every 6 months for 1 year and then annually is recommended for patients with persistent low-grade dysplasia. The approach to a patient with high-grade dysplasia confirmed by an expert pathologist is controversial, ranging from more frequent surveillance (every 3 months) to esophagectomy in patients who are considered candidates for surgery. Documentation of the long-term effectiveness of promising endoscopic therapy to eliminate high-grade dysplasia and early cancer would offer an alternative to surgical therapy and may expand the number of patients eligible for surveillance (19). A technique less invasive than endoscopy and biopsy for surveying patients with Barrett esophagus would have great potential to decrease cost. Early data on balloon cytology without endoscopy are promising but need validation (20).
Treatment of Barrett Esophagus
Barrett esophagus therapy consists of pharmacologic acid control or antireflux surgery to treat the underlying GERD. Unfortunately, neither of these therapies has been documented to reliably reverse Barrett esophagus, even when esophageal acid is reduced to normal levels. In addition, both medical therapy with proton-pump inhibition and surgical therapy with fundoplication are effective in controlling GERD symptoms and healing esophagitis, but there is no evidence that these therapies prevent cancer. Antireflux surgery should in theory reduce all refluxate into the esophagus, but because surgery fails to eliminate the risk for cancer, it is unlikely that pharmacologic control of acid will be able to prevent adenocarcinoma in patients with Barrett esophagus.
The Future
Our understanding of the molecular biology of adenocarcinoma of the esophagus and gastric cardia and of Barrett esophagus is increasing. The particular molecular events that occur during progression of Barrett esophagus from metaplasia to dysplasia to adenocarcinoma are being defined. However, the cause of the increasing incidence of adenocarcinoma is not yet known; this limits our ability to intervene to decrease its incidence.
The important undefined issues are clear. These include an understanding of risk factors for the development of adenocarcinoma, a less invasive method to screen for Barrett esophagus, a better definition of whom to screen for Barrett esophagus, a less invasive method to survey for dysplasia, validation of biological markers to identify the risk for adenocarcinoma in an individual patient with Barrett esophagus, and use of chemoprevention agents to alter biological markers of and reduce the risk for adenocarcinoma of the esophagus and gastric cardia. In an era when many other cancers are being prevented or cured and overall cancer mortality rates are beginning to decrease for the first time, it seems ironic that the incidence of this particular cancer should be increasing. A new “front” has thus opened up in the “war” on cancer, calling for a new mobilization of forces, better tactics, and the most sophisticated of strategies. Let's hope that we are equal to the challenge.
Richard E. Sampliner, MD
Arizona Health Sciences Center
Tucson Veterans Affairs Medical Center
Tucson, AZ 85723
Article and Author Information
-
Requests for Reprints: Richard E. Sampliner, MD, Tucson Veterans Affairs Medical Center (111G-1), 3601 South 6th Avenue, Tucson, AZ 85723.
- Copyright ©2004 by the American College of Physicians
