Primary Motility  Disorders of the  Esophagus
 The Esophageal
 Esophagogastric  Junction

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

What is the annual incidence of adenocarcinoma in patients with endoscopically evident Barrett's esophagus?

S.J. Spechler (Dallas)

A recent study by Shaheen et al. has suggested that the cancer risk in Barrett's esophagus had been overestimated for years because of publication bias, the selective reporting of studies that have positive or extreme results [1]. The investigators used the premise that cancer risk estimates derived from small studies should be less accurate than those from large studies, because sampling error is inversely related to sample size. Compared to a group of large studies, therefore, a group of small studies should describe a wider range of risk estimates, and those estimates should be symmetrically distributed around the midpoint value (similar in pattern to a bell-shaped curve). If there is publication bias, however, small studies that find low risks either will not be submitted to journals at all, or the submitted manuscripts will be rejected by editors who deem the reports uninteresting. In this situation, the preponderance of small studies published will describe high risks. To seek evidence of publication bias in the reporting of cancer risk in Barrett's esophagus, Shaheen et al. searched the available literature on this subject and identified 24 informative studies. The annual incidence of cancer in Barrett's esophagus described in those reports ranged from 0.2% to 2.9%. As would be expected in the case of publication bias, the investigators found a strong, inverse correlation between the magnitude of the cancer risk and the size of the study population. No significant associations were found between cancer risk and a number of factors other than study size that might have influenced the observed outcome (e.g. country of origin, retrospective study design, year of publication).

Some authors have pooled data from published studies to derive a composite estimate of cancer risk in Barrett's esophagus [2, 3]. Using such pooled data, one widely quoted report has estimated the cancer risk at 1% per year [2]. However, data pooling does not obviate the problem of publication bias, because the derived estimates are based on data that are skewed to exaggerate the incidence of cancer. Furthermore, available studies on cancer risk in Barrett's esophagus are comprised largely of patients referred to tertiary care centers. The incidence of cancer for such patients may exceed that for patients treated in primary care practices. With these biases in the published studies, it is likely that the risk of esophageal cancer for patients with Barrett's esophagus in the general population is substantially lower than the pooled literature would suggest. Although the precise risk remains unclear, Shaheen et al. suggest that an esophageal cancer incidence rate of 0.5% per year might be a reasonable estimate [1]. This estimate is in keeping with the results of three modern, large, prospective studies on cancer incidence in Barrett's esophagus [2-4].

It seems paradoxical that the risk of adenocarcinoma in Barrett's esophagus appears to have been overestimated for the past several decades, a period during which the frequency of this tumor has increased dramatically in the United States [5]. However, it is important to put this trend into perspective. The overall incidence of esophageal cancer in this country has not changed substantially over the past 25 years [6]. What has changed during this period is the proportion of esophageal cancers that are adenocarcinomas. For most of the twentieth century, squamous cell carcinoma was the major histologic type of esophageal cancer, and adenocarcinoma of the esophagus was rare [7]. The incidence of esophageal adenocarcinoma has nearly quadrupled over the past 25 years, but 25 years ago the incidence of this tumor was extremely low (< 1 case per 100,000) [5-8]. Despite the impressive relative rise in the frequency of esophageal adenocarcinoma, the total number of new cases remains low. In 1997, there were an estimated 12,500 new cases of esophageal cancer in the United States, of which approximately 5,750 (46%) were adenocarcinomas [8]. For comparison, there were an estimated 131,200 new cases of colorectal cancer that same year [6]. Further supporting the notion that the risk of cancer in Barrett's esophagus has been exaggerated are data from studies showing that Barrett's esophagus does not adversely influence long-term survival [9].

The data discussed above primarily pertain to patients who have Barrett's esophagus without dysplasia. Dysplasia in Barrett's esophagus can be an ominous finding. For patients with high-grade dysplasia, the risk of developing esophageal adenocarcinoma may be orders of magnitude greater than that for patients who have no dysplasia. For the latter, however, the risk of developing esophageal cancer appears to have been exaggerated for years. A reasonable estimate of adenocarcinoma incidence in patients with endoscopicallyevident Barrett's esophagus is 0.5% per year.


1. Shaheen NJ, Crosby MA, Bozymski EM, Sandler RS. Is there publication bias in the reporting of cancer risk in Barrett's esophagus? Gastroenterology 2000;119:333-338.

2. Drewitz DJ, Sampliner RE, Garewal HS. The incidence of adenocarcinoma in Barrett's esophagus: a prospective study of 170 patients followed 4.8 years. Am J Gastroenterol 1997;92:212-215.

3. O'Connor JB, Falk GW, Richter JE. The incidence of adenocarcinoma and dysplasia in Barrett's esophagus. Report on the Cleveland Clinic Barrett's esophagus registry. Am J Gastroenterol 1999;94:2037-2042.

4. Spechler SJ, Lee E, Ahnen D, Goyal RK, Hirano I, Ramirez F, Raufman JP, Sampliner R, Schnell T, Sontag S, Vlahcevic ZR, Young R, Williford W. Long-term outcome of medical and surgical treatments for gastroesophageal reflux disease. Follow-up of a randomized controlled trial. JAMA 2001;285:2331-2338.

5. Devesa SS, Blot WJ, Fraumeni JF Jr. Changing patterns in the incidence of esophageal and gastric carcinoma in the United States. Cancer 1998;15;83:2049-2053.

6. Ries LAG, Eisner MP, Kosary CL, Hankey BF, Miller BA, Clegg L, Edwards BK. SEER cancer statistics review, 19731997. Bethesda: National Cancer Institute, 2000.

7. Hesketh PJ, Clapp RW, Doos WG, Spechler SJ. The increasing frequency of adenocarcinoma of the esophagus. Cancer 1989;64:526-530.

8. Data provided by SEER Program, National Cancer Institute, Bethesda, MD, August 199

9. submission. 9. Eckardt VF, Kanzler G, Bernhard G. Life expectancy and cancer risk in patients with Barrett's esophagus:a prospective controlled investigation. Am J Med 2001;111:33-37.

Publication date: August 2003 OESO©2015