What is endoscopic biopsy surveillance?

If you have Barrett's esophagus, periodic endoscopic biopsy surveillance (EGD with biopsy of the Barrett's esophagus lining to check for cancer) is the ONLY way to know for sure whether you are at increased risk for developing cancer. It can also detect an early cancer that has not yet grown large enough to cause symptoms, such as bleeding, food sticking in the esophagus or weight loss. Endoscopic biopsy surveillance instead of esophagectomy (surgical removal of the esophagus) is recommended for most patients who have Barrett's esophagus. This is because most patients who have Barrett's esophagus do not develop cancer. Also, esophagectomy is a surgery that has a much higher rate of complications and mortality (death) as compared to most gastrointestinal surgeries. Therefore, esophagectomy is typically performed only in those Barrett's esophagus patients who have high-grade dysplasia or cancer.

Periodic endoscopic biopsy surveillance has been shown to be safe and there is evidence that it is successful in the detection of esophageal adenocarcinoma when the cancer is early and curable with surgery. Unfortunately, because most patients who develop esophageal adenocarcinoma are not in a cancer surveillance program, endoscopic biopsy surveillance has not favorably impacted the mortality (rate of death) of this disease. Esophageal adenocarcinoma must be detected early, before the patient develops symptoms, for the patient to have a high likelihood of a cure.

Endoscopic biopsy surveillance is just like having an EGD with biopsy except that many more biopsies are taken from the Barrett's lining, in a systematic manner, to screen for dysplasia (atypical or abnormal changes in the Barrett's tissue) and cancer. Although many more biopsies are taken compared to a standard EGD with biopsy, this practice has been shown to be safe. In the past, there have been no standardized guidelines for how biopsies are taken. It is now widely recommended that the patient with Barrett's esophagus have four quadrant biopsies at every two centimeter intervals along the entire length of the Barrett's lining. A jumbo forceps may be useful in obtaining  larger biopsies. The reason for taking four quadrant biopsies is that sampling all four walls of the esophagus decreases the odds of missing a small area of abnormal cells that can only be seen by histologic analysis (examination of tissue under the microscope) and not through the endoscope. Because high-grade dysplasia and early cancer can be invisible to the doctor performing the endoscopy and seen only under a microscope, biopsies taken in a systematic manner may increase the odds of hitting an endoscopically invisible abnormality. Small abnormalities that CAN be seen through the endoscope, such as nodules, polyps, ulcers, reddened areas and erosions are sometimes the only areas in which there is high-grade dysplasia or cancer so these should have additional biopsies. Future endoscopic technologies, including increasing the resolution of the endoscopic image (Like high definition TV) offers the possibility of endoscopic detection of areas of high-grade dysplasia and early cancer that cannot be currently seen with the current endoscopic technology.

The four quadrant biopsy pattern as seen through the endoscope		Illustration of an every two centimeter endoscopic biopsy surveillance protocol
The "x" marks represent the positions of the biopsies obtained in a four quadrant biopsy pattern Photo courtesy of the Seattle Barrett's Esophagus Research Program		Four quadrant biopsies are taken at two centimeter intervals throughout the entire length of the Barrett's esophagus segment.

Problems with endoscopic biopsy surveillance

There are problems with endoscopic biopsy surveillance. The main problem is sampling error. Sampling error means that there can be abnormal areas in the esophagus that are so small that they cannot be seen through the endoscope and therefore missed with the biopsy forceps even when a systematic biopsy protocol is used. Presumably, the sampling error gets lower when the doctor takes more biopsies and follows a systematic protocol, but sampling error is never completely eliminated.

There are studies looking at different ways to minimize sampling error. One group of investigators has shown that brush cytology (collecting cells that are brushed off the surface of the Barrett's lining), when combined with endoscopic biopsy, improves the detection of dysplasia and cancer as compared to biopsy alone.

Some studies have looked at applying stains or dyes to the Barrett's lining, or investigated new endoscopic technologies that direct the doctor to the abnormal areas in the esophagus that need to be biopsied. Some of these studies show the results to be good while others do not. Because it remains unclear whether any of these techniques improve surveillance, brush cytology, stains and dyes are not routinely used in most medical centers in the cancer surveillance of patients who have Barrett's esophagus. New endoscopic technologies to enhance the detection of dysplasia such as narrow band imaging, autofluorescence endoscopy, optical coherence tomography, magnification endoscopy and confocal microscopy are currently under study in an effort to enhance the detection of dysplasia. However, none are as yet ready for routine clinical practice.

How often do patients with Barrett's esophagus need to have endoscopic biopsy surveillance?

Biopsies obtained during endoscopic biopsy surveillance are sent to a pathology lab for histologic analysis. How often a patients returns for endoscopy is usually based on their histologic grade of dysplasia. Although high-grade dysplasia identifies patients who are at increased risk for cancer, readings of low-grade dysplasia are much less useful in predicting who will develop cancer. Other tests, such as flow cytometry, genetic, and protein markers, are being investigated to better separate patients into low-risk and high-risk groups for the development of cancer. None of these other tests are routinely used, in most medical centers, in the clinical care of patients who have Barrett's esophagus.

American College of Gastroenterology (ACG) Guidelines

In most medical centers, how often a patient returns for endoscopic biopsy surveillance is based solely on the biopsy readings of dysplasia. The American College of Gastroenterology (ACG) has developed guidelines for physicians on how often a patient with Barrett's esophagus should undergo endoscopic biopsy surveillance based on readings of dysplasia. The current recommendation for patients who have a stable diagnosis of negative for dysplasia, confirmed by two endoscopic biopsy surveillance procedures, is that they come back every 3 years for follow-up endoscopic biopsy surveillance. For patients who have a stable diagnosis of low-grade dysplasia, confirmed by two endoscopic biopsy surveillance procedures, it is recommended that they return yearly for endoscopic biopsy surveillance until they have a set of biopsies in which no dysplasia is detected. Endoscopic biopsy surveillance of high-grade dysplasia should be individualized in each patient who has that diagnosis and who elects to continue in a cancer surveillance program or have an endoscopic therapy rather than undergoing esophagectomy as treatment for their high-grade dysplasia. The most recent ACG guidelines recommend that patients who have a diagnosis of high-grade dysplasia undergo a second endoscopic biopsy surveillance procedure to increase the chance of detecting an early cancer if it is already there, and that all biopsies be reviewed by a second expert GI pathologist. If no cancer is detected, then patients have several options including: remaining in endoscopic biopsy surveillance and returning every few months for their procedures; esophagectomy (surgical removal of the esophagus); or endoscopic therapy followed by endoscopic biopsy surveillance, indefinitely.  These ACG guidelines are based on the use of a four-quadrant every two centimeter biopsy protocol with additional biopsies of any abnormality in the Barrett's lining seen through the endoscope. Several centers now perform (and some Barrett's experts recommend) four-quadrant biopsies every one centimeter in patients who have high-grade dysplasia to further decrease the chances of missing a very small cancer if present..

The current guidelines for how often a patient should return for endoscopic biopsy surveillance are based on readings of dysplasia and the hypothesis that cancer develops in Barrett's esophagus in an orderly fashion from metaplasia (no dysplasia) to dysplasia to cancer. The current endoscopic biopsy surveillance guidelines have not been tested in patient studies or in the community practice. This is an area of active research. However, it is important to note that your physician may recommend different endoscopic biopsy surveillance intervals for you, depending on a variety of factors. The ACG guidelines emphasize that "physicians must always choose the course best suited to the individual patient". At this point in time, it is impossible to adequately address the management of each individual patient who has Barrett's esophagus in a set of published guidelines.

For most patients who have Barrett's esophagus and who will never develop cancer, endoscopies are performed more frequently than necessary. Some guidelines are even suggesting that most patients who have Barrett's esophagus can safely go up to five years between endoscopies. The patient's risk of developing cancer must be weighed against the cost of endoscopic biopsy surveillance as well as the risk of complications from EGD and discomfort to the patient. In the future, endoscopic surveillance intervals will likely be lengthened out for most patients who have Barrett's esophagus, but to do this safely, we must be able to better predict who will and won't develop cancer in Barrett's esophagus.

Endoscopic biopsy surveillance of high-grade dysplasia

A standard recommendation for the treatment of high-grade dysplasia in Barrett's esophagus is esophagectomy, surgical removal of the esophagus. This recommendation has been largely based on studies showing that 30-50% or more of patients who had a diagnosis of high-grade dysplasia and then underwent esophagectomy for high-grade dysplasia, actually had cancer in their esophagectomy specimen (surgically removed esophagus). In other words, endoscopic biopsy surveillance missed a cancer that was already present in their esophagus. The problem with most of these studies is that they did not report if a systematic biopsy protocol was used, the number of endoscopic biopsies taken from the Barrett's lining, how many endoscopies were preformed prior to the surgery or whether there was an abnormality in the Barrett's lining, such as an ulcer or nodule, that could be seen endoscopically prior to surgery.

Three recent studies performed endoscopic biopsy surveillance in patients who had a diagnosis of high-grade dysplasia to determine how many cancers would be missed using a jumbo forceps and a four-quadrant, two centimeter protocol. Endoscopic biopsy evaluation was performed once in each patient, prior to surgery. The patients then underwent esophagectomy and their esophageal specimens were checked for cancer. In the largest study, two of 19 patients had cancer in their esophagectomy specimens (10.5%). In the two other studies, four of 12 (33%) patients and 5 of 9 (56%) patients had cancer in their esophagectomy specimens.

The argument for continuing endoscopic biopsy in patients who have high-grade dysplasia is that a substantial number of patients who have high-grade dysplasia (range 40% to 85%) do not progress to cancer during long-term follow-up. Also it has been reported that in some patients who have a diagnosis of high-grade dysplasia, the high-grade dysplasia is not detected in follow-up endoscopies. At the present time, it is unclear whether the high-grade dysplasia truly goes away in some patients, whether it is not detected due to sampling error (missing a small area of high-grade dysplasia in the set of biopsies taken) or the initial diagnosis of high-grade dysplasia was incorrect. Therefore, some patients may elect to remain in surveillance, unless cancer is detected.

It has been recommended that sampling error can be greatly minimized and cancer detected when it is still early and curable if patients who have high-grade dysplasia return for three consecutive (back-to-back) and closely spaced endoscopies to detect cancer that may already be there. This allows a diagnosis to be made based on many more biopsies obtained in a short period of time as compared to declaring the patient free of cancer based on one endoscopic surveillance procedure. The American College of Gastroenterology (ACG) guidelines now include a recommendation that patients with a diagnosis of high-grade dysplasia undergo repeat EGD with biopsy to minimize sampling error and maximize detection of cancer, if present. If no cancer is detected and the patient elects to remain in endoscopic biopsy surveillance, then it is recommended that the patient undergo endoscopic biopsy surveillance every 3 months.

In patients with high-grade dysplasia, it is also recommended by one Barrett's research center that four quadrant biopsies be taken, using a jumbo forceps, at every one centimeter interval rather than every two centimeters. At this center, if no cancer is detected, the patient returns in one month for repeat endoscopic biopsy surveillance. If again no cancer is detected, the patient returns in three months for their next surveillance procedure. If after three closely spaced procedures, no cancer is detected, the patient undergoes endoscopic biopsy surveillance at least every 6 months to detect cancer, if it develops, when it is early and curable with surgery. The basis for the recommendation of a one centimeter biopsy protocol in patients who have high-grade dysplasia is a study that demonstrated that half of all cancers detected using the 1 centimeter protocol would be missed using a 2 centimeter protocol. This biopsy protocol is not in widespread use at the present time and even though cancers are typically detected early in experienced centers, cancers can still be missed until they are larger and less curable, even when such a rigorous protocol is used. 

Sampling error, and the concern that a patient who has a diagnosis of high-grade dysplasia already has a cancer that will not be detected until it is advanced and incurable with surgery, remains a big concern. As with any medical practice, the recommendation of endoscopic biopsy surveillance or endoscopic therapies versus surgery in the management of high-grade dysplasia must be weighed with respect to the risk of developing advanced and incurable cancer while in surveillance versus the mortality (rate of death) from the surgery. Because high-grade dysplasia is rare in patients who have Barrett's esophagus, many gastroenterologists do not have the opportunity to care for many of these patients or have routine access to an expert GI pathologist. Further, intensive and frequent endoscopic biopsy surveillance of patients with high-grade dysplasia, using a one centimeter biopsy protocol, may require increased staffing and advanced planning in the endoscopy unit and the cost of such a complex procedure may not be fully reimbursed by insurance companies. For these reasons, and because the mortality (rate of death) of esophagectomy is lower in institutions where a high number of esophagectomies are performed, consideration should be given to referral of these patients to specialty centers that follow large numbers of patients with high-grade dysplasia.

Care of the patient who has high-grade dysplasia should be individualized, taking into account the patient's desires, medical fitness to undergo esophagectomy, and willingness to come back for frequent endoscopies. Any patient who will NOT return at the recommended endoscopic intervals and who has a diagnosis of high-grade dysplasia, CONFIRMED by an expert GI pathologist and a repeat endoscopic biopsy surveillance procedure, should undergo esophagectomy, performed by an experienced esophageal surgeon. Esophagectomy is the ONLY treatment that allows a patient to safely stop periodic endoscopic biopsy surveillance. Even the patient who chooses endoscopic therapy, must be willing to undergo endoscopic biopsy surveillance indefinitely, even if appears that the high-grade dysplasia has disappeared. A recent randomized controlled trial of photodynamic ablation therapy in patients who had high-grade dysplasia, reported that although the numbers of cancers were cut in half as compared to those patients who did not have ablation, it did not completely eliminate the cancer risk.