Primary Motility  Disorders of the  Esophagus
 The Esophageal
 Esophagogastric  Junction

  Browse by Author
  Browse by Movies
Volume: Barrett's Esophagus
Chapter: Screening and surveillance

Do the current data support that an endoscopic-biopsy surveillance program prevents or does not prevent cancer in Barrett's mucosa?

H.D. Appelman (Ann Arbor)

There is little doubt that:
- Barrett's mucosa has an increased risk for esophageal carcinoma compared to the squamous mucosa it replaced;
- the survival of patients with esophageal carcinoma is directly related to the stage of the carcinoma when it is first detected;
- adenocarcinomas are detected during surveillance programs that entail endoscopy and biopsy sampling;
- the carcinomas detected during surveillance for Barrett's mucosa are, as a group, of a lower stage than are the carcinomas that occur in patients whose Barrett's mucosa was never known prior to the development of the carcinoma [1];
- dysplasias, especially high-grade dysplasias (HGD), are the precursors of invasive carcinoma, or, if not the immediate precursors, then at least markers of esophagi at great risk for cancer development.

Although the studies of dysplasias in Barrett's mucosa often supply a reference for diagnostic criteria for both HGD and low-grade dysplasias, it is obvious to experienced pathologists that the published criteria are insufficient, and not all pathologists diagnoze these dysplasias in exactly the same ways using the same sets of criteria. Neither grade of dysplasia is a single histologic diagnostic entity, but each grade contains a group of mucosal abnormalities with slightly different individual histologic criteria. As a result, all the published studies of HGD and incipient cancers developing within such dysplasias may not be talking about the same patients. This is a critical realization.

With this last limitation in mind, the following information can be accepted as helpful, but the results of the published studies must be cautiously analyzed and accepted with even more caution.

If HGD is truly the cancer precursor, then treatment of HGD theoretically prevents cancers from developing. However, when HGD is the indication for esophagectomy, invasive carcinomas are discovered in about 40% of the resection specimens. (Table I) [19]. These unsuspected cancers are detected mainly at a low stage when they are more likely to be curable, but, obviously, surveillance did not prevent their development. As a result of finding these incidental cancers, the diagnosis of HGD, not invasive carcinoma, has

Table I. Prevalence of carcinoma in resections for biopsy proven high-grade dysplasia.

become the end-point of surveillance and the usual indication for treatment in many centers, whether that treatment is resection or ablation.

Therefore, it has seemed reasonable to suggest that surveillance in Barrett's mucosa is the best way not only to detect cancer precursors, but to detect cancers when they are likely to be of low stage and curable [10, 11].

To complicate matters, in some centers, the indication for esophagectomy is not HGD, but carcinoma invading the lamina propria [12, 13]. Those cases with only HGD are continued in the surveillance programs until cancer is found in biopsies. This approach was introduced because not all HGD progressed to invasive carcinoma. In such centers, it is obvious that surveillance does not prevent cancer. Instead it demands the presence of cancer for definitive treatment.

We do not know the risk of invasive carcinoma developing in a field of HGD. We supect that this risk is high, because of the large number of cases of coexisting cancer with HGD (Table II) lists the most recent studies that followed patients with HGD [12-17]. The followup periods are short, but even in such short periods, the overall incidence of adenocarcinoma developing after HGD was first discovered is about 25%. However some cancers are detected during surveillance without a pre-existing biopsy of HGD. It is possible that this resulted from an insufficiently extensive biopsy sampling of the Barrett's segment so that a pre-existing focus of HGD was missed.

We have a great deal of data concerning the incidence of carcinoma developing in Barrett's mucosa during surveillance, invariably given as the number of patient years of follow-up needed to detect a single cancer [14, 15, 18, 19]. However, we have much less information about the incidence of the different grades of dysplasia in the same setting.

The only way to answer the question about whether surveillance prevents cancer is to conduct a controlled study of Barrett's patients. As soon as the Barrett's is identified, half the patients would be entered into an endoscopic surveillance program with a fastidious biopsy protocol, and the other half would not be surveyed but would be contacted periodically to see how they are doing. The end point of surveillance and the indication for treatment would have to be something short of carcinoma, presumably HGD. The

Table II. Incidence of carcinoma developing during surveillance after biopsy proven high-grade dysplasia.

difference in the cancers developing in the non-surveyed group compared with the surveyed group would give us an indication of the value of surveillance in preventing cancers. At the moment, no such studies have been published. Instituting them would require huge numbers of patients followed for many years and it would be a problem to deny half these patients access to surveillance that may prevent cancer death [11]. Since we know that Barrett's mucosa has a cancer risk, we feel bound to put all patients on surveillance.

Perhaps we can get data from Barrett's mucosa comparable to the ulcerative colitis data. Most such ulcerative colitis studies have compared survival of colitic cancer patients on surveillance with those who were not on surveillance, but none of these studies were controlled. Most such studies indicate that surveillance of long standing colitics with regular colonoscopy and extensive biopsy sampling of flat mucosa and unusual polyps may not prevent colitic cancer, but it seems to diminish colitic cancer mortality [20, 21]. There have been several studies of Barrett's patients comparing the cancer stage and patient survival of those patients whose cancers presented at the time the Barrett's was found with those whose cancers are found during surveillance. Not surprisingly, the cancers detected during surveillance were of lower stage and were associated with far better survival than were the de novo cancers. However, this comparison is different than comparing cancers that develop with or without surveillance after the nondysplastic Barrett's is first found [1, 4].

It is problematic to use the colitic model in Barrett's mucosa for several reasons. First, ulcerative colitis, although a relatively common gastrointestinal disease, is not common in the general population, so the pool of patients is small. In contrast, Barrett's mucosa seems to be a common disease in the general population, so the pool of patients may be hugs. Second, the colitics are known because of the signs and symptoms of the disease. In contrast, we do not know who has Barrett's, since there are no signs or symptoms. They have reflux symptoms, just like many other people who don't have Barrett's. Third, it is rare for colitics to present for the first time with carcinoma. In contrast, most Barrett's cancers present with signs and symptoms of the cancers; the Barrett's is discovered at the same time when biopsies are taken from mucosa next to the cancers.

This question avoids other issues pertaining to surveillance, most important of which is cost effectiveness. The pool of Barrett's patients is very large, but the true extent is unknown. The risk of cancer is also unknown, but the reported incidence figures for surveillance vary greatly from 1 in about 50 years of follow-up to 1 in about 440 years, indicating that there are likely to be population differences. Furthermore, there is some data that suggests that the cancer risk is over-estimated, especially in the studies with smaller numbers of patients and fewer patient years of follow-up.

The original question that this brief discussion attempted to answer was: do the current data support that an endoscopic biopsy program prevents or does not prevent cancer in Barrett's mucosa? The answer is that we think we are preventing cancer through surveillance, but we have no facts to support this contention. When we resect for HGD discovered during surveillance, we find unsuspected invasive carcinomas about 40% of the time, so these were not prevented by surveillance. We have no idea how many of the other 60% of esophagi resected HGD would have developed invasive carcinomas had we left them in situ. At the centers that don't resect until invasive carcinoma is found, then it is obvious that surveillance didn't prevent cancer; surveillance only ended when the cancers developed.


1. Peters JH, Clark GWB, Ireland AP, et al. Outcome of adenocarcinoma arising in Barrett's esophagus in endoscopically surveyed and nonsurveyed patients. J Thorac Cardiovasc Surg. 1994;108:813-822.

2. Pera M, Trastek VF, Carpenter HA, et al. Barrett's esophagus with high-grade dysplasia: an indication for esophagectomy? Ann Thorac Surg 1992;54:199-204.

3. Cameron AJ, Carpenter HA. Barrett's esophagus, high-grade dysplasia, and early adenocarcinoma:a pathologic study. Am J Gastroenterol 1997;92:586-591.

4. Van Sandick JW, van Lanschot JJB, Kuiken BW, et al. Impact of endoscopic biopsy surveillance of Barrett's oesophagus on pathological stage and clinical outcome of Barrett's carcinoma. Gut 1998;43:216-222.

5. Heitmiller RF, Redmond M, Hamilton SR. Barrett's esophagus with high-grade dysplasia. An indication for prophylactic esophagectomy. Ann Surg 1996;224:66-71.

6. Altorki NK, Sunagawa M, Little AG, Skinner DB. High-grade dysplasia in the columnar-lined esophagus. Am J Surg 1991;161:97-99.

7. Edwards MJ, Gable DR, Lentsch AB, Richardson JD. The rationale for esophagectomy as the optimal therapy for Barrett's esophagus. Ann Surg 1996;223:585-589.

8. Rice TW, Falk GW, Achkar E, Petras RE. Surgical management of high-grade dysplasia in Barrett's esophagus. Am J Gastroenterol 1993;88:1832-1836.

9. Ferguson MK, Naunheim KS. Resection for Barrett's mucosa with high-grade dysplasia: implications for prophylactic photodynamic therapy. J Thorac Cardiovasc Surg 1997;114:824-829.

10. Ackroyd R, Wakefield SE, Williams JL, et al. Surveillance of Barrett's esophagus:a need for guidelines. Dis Esophagus 1997;10:185-189.

11. Tytgat GNJ. Barrett's esophagus: is it all that bad? Can J Gastroenterology 1999;13:385-388.

12. Levine DS, Haggitt RC, Irvine S, Reid BJ. Natural history of high-grade dysplasia in Barrett's esophagus. Gastroenterology 1996;110:A550.

13. Schnell T, Sontag SJ, Chejfec G, et al. High grade dysplasia is not an indication for surgery in patients with Barrett's esophagus. Gastroenterology 1996;110:A590.

14. 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.

15. Katz D, Rothstein R, Schned A, et al. The development of dysplasia and adenocarcinoma during endoscopic surveillance of Barrett's esophagus. Am J Gastroenterol 1998;93:536-541.

16. Iftikhar SY, James PD, Steele RJC, et al. Length of Barrett's oesophagus:an important factor in the development of dysplasia and adenocarcinoma. Gut 1992;33:1155-1158.

17. Weston AP, Sharma P, Topalovski M, et al. Long-term follow-up of Barrett's high-grade dsyplasia. Am J Gastroenterol 2000;95:1888-1893.

18. 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.

19. Streitz JM Jr, Ellis FH Jr, Tilden RL, Erickson RV. Endoscopic surveillance of Barrett's esophagus: a cost-effective comparison with mammographic surveillance for breast cancer. Am J Gastroenterol 1998;93:911-915.

20. Choi PM, Nugent FW, Schoetz DJ Jr, et al. Colonoscopic surveillance reduces mortality from colorectal cancer in ulcerative colitis. Gastroenterology 1993;105:418-425.

21. Giardiello FM, Gurbuz AK, Bayless TM, et al. Colorectal cancer in ulcerative colitis:survival in patients with and without colorectal cancer symptoms. Inflam Bowel Dis 1996;2:6-10.

Publication date: August 2003 OESO©2015