Background: Some 10,000 to 20,000 patients will be diagnosed with carcinoma of the esophagus or gastroesophageal junction each year in the United States. Carcinoma of the esophagus is the fifth-leading cause of cancer death in this country today. The primary microscopic patterns of esophageal cancer are adenocarcinomas (glandular in nature) or squamous cell (skin cell in nature) cancers. Squamous cell cancer is the most common esopha- geal neoplasm in the world and, until recently, it also was the most common microscopic variety in the United States.

During the past two decades, the incidence of esophageal adenocarcinoma has increased more rapidly than that of any other cancer. Currently, adenocarcinoma of the esophagus comprises 50 percent of all esophageal cancers in this country. The reason for the alarming increase in esophageal adenocarcinoma is unclear, but it is known to develop as a consequence of microscopic changes in the esophageal lining cells secondary to chronic gastroesophageal reflux disease (GERD). These microscopic findings are characterized by a transition from the normal squamous lining of the esophagus to the glandular lining cells of the stomach or intestinal tract. These changes in the esophageal lining cells have been termed Barrett's esophagus after the English physician who first noted this microscopic pattern.

The symptoms, diagnosis, staging and treatment of these common types of esophageal carcinoma are similar, apart from epidemiologic differences between squamous cell cancer and adenocarcinoma, which will be discussed below.

Epidemiology: Squamous cell cancer of the esophagus has a very high annual incidence in parts of Iran, China, Russia and South Africa. In the United States, African Americans have a fourfold to fivefold increased risk of squamous cell carcinoma of the esophagus compared to patients of Caucasian descent. It appears that the etiology of squamous cell cancer of the esophagus is related to environmental factors. In the United States, alcohol and tobacco appear to be the primary risk factors in the majority of cases. Although there is no known genetic predisposition for squamous cell cancer of the esophagus, a rare autosomal dominant condition known as "tylosis" (hyperkeratosis of the palms and soles, oral leukoplakia and esophageal squamous cell cancer) has been implicated in a small number of cases. Chronic esophageal injury from Lye stricture or as a result of the motor disorder of the esophagus, achalasia, may also increase the risk for developing squamous cell carcinoma of the esophagus.

Adenocarcinoma of the esophagus is a disease that predominantly affects male Caucasians older than 40. Unlike squamous cell carcinoma of the esophagus, adenocarcinoma has no association with environmental habits and is primarily a consequence of Barrett's esophagus. Normally, the esophagus is lined almost entirely by the stratified squamous epithelium similar to our skin. When the lining cells of the distal esophagus are replaced by the columnar epithelium characteristic of the stomach or the intestinal tract, this abnormality is termed Barrett's esophagus. This cellular change results from primary loss of the squamous cell lining from gastric acid ulceration of the lower esophagus. The columnar cells of the stomach then advance up the esophagus over the ulcer bed, as these cells tend to grow at a faster pace than the squamous cells that were there before.

For patients with endoscopically obvious Barrett's esophagus, the annual incidence of adenocarcinoma has been estimated at approximately 0.8 percent. This is equivalent to the risk of developing carcinoma of the lung in a heavy cigarette smoker.

Figure 1A -- Barium esophagram of obstructing distal esophageal cancer
Figures 1B and 1C -- CT scan images of the same patient

Clinical Presentation: The symptoms of esophageal adenocarcinoma and squamous cell cancer are identical. In early stages, esophageal cancer may be completely asymptomatic. As the tumor grows, dysphagia (difficulty swallowing) becomes the predominant symptom. Weight loss, anorexia, odynophagia (painful swallowing), gastrointestinal bleeding, chest pain and fatigue are other common presenting symptoms. Cough may result from pulmonary aspiration or tracheoesophageal fistula (an abnormal development of a communication between the esophagus and the airway related to tumor infiltration).

Physical findings are limited in early esophageal cancer, but the finding of enlargement of a supraclavicular node or the development of liver enlargement of fluid within the abdomen (ascites) suggests more advanced metastatic disease.

Diagnosis: The diagnosis of esophageal cancer can be suspected on examination through upper GI barium X-rays or direct endoscopic examination. Confirmation of the diagnosis of cancer must be made with microscopic analysis of an esophageal biopsy.

A barium esophagram is an insensitive test for uncomplicated Barrett's esophagus and early carcinoma. Only in advanced esophageal cancer does the esophagram reveal a lesion. The most common presentation of advanced esophageal cancer is that of a polypoid intraluminal mass. The mass may have the characteristic appearance of an "apple core" lesion with circumferential involvement by the cancer, leaving only a narrowed, irregular lumen (Figures 1 and 2).

Esophageal cancer (squamous or adenocarcinoma) may appear endoscopically as an intraluminal polypoid mass, a stricture (which, unlike peptic strictures, may be asymmetric) or a deep ulceration in the wall of the esophagus. The endoscopic application of toluidine blue, which stains Barrett's mucosa and columnar epithelium dark blue, and Lugol's iodine, which stains the squamous epithelium black, may be used to highlight Barrett's epithelium and cancer. Despite the enhanced visualization of abnormal esophageal epithelium with these dyes, the diagnostic sensitivity and specificity are only 89 percent and 93 percent, respectively.

At the time of endoscopic examination of a suspected cancer, multiple biopsies should be obtained from abnormal-appearing mucosa. The diagnostic accuracy of multiple biopsies is 80 percent to 95 percent. The role of endoscopic brush cytology in the detection of early esophageal cancer has not been documented. The use of cytology alone may not detect occult malignancy and, in one study, a 24 percent false negative rate for invasive carci- noma was reported. Combining cytology with endoscopic biopsies increases diagnostic accuracy to 90 to 100 percent. Therefore, cytologic brushing may be complementary to endoscopic surveillance but does not replace the need for biopsies. In patients with Barrett's esophagus, a systematic protocol of four-quadrant endoscopic biopsies at 2-cm intervals in the columnar lined esophagus can detect early neoplasia.

Staging: The staging of esophageal malignancies is based on the degree of wall penetration by the tumor (T), lymph node involvement (N) and metastases (M). The initial staging of esophageal involves abdominal and thoracic computer-assisted tomography (CT) scan to exclude distant metastasis. Assuming there are no metastatic lesions, local staging (T, N) with endoscopic ultrasound (EUS) or CT is recommended.

Endoscopic ultrasound is a relatively new device that allows standard endoscopic evaluation in combination with high-frequency ultrasound. EUS offers the most accurate imaging modality for locoregional staging. The advantage of EUS is that the ultrasound probe can be placed adjacent to a suspect lesion under direct visualization. EUS has been compared with surgical pathologic and CT staging of esophageal cancers. The accuracy of EUS in staging tumor depth of invasion varies from 87 to 92 percent, making it more accurate than CT (approximately 70 percent), especially in early tumors. The accuracy of EUS in detecting regional lymph node status ranges from 80 to 88 percent compared with CT, 51 to 74 percent. Overall, EUS appears to be better at determining resectability of cancer than CT. The limitations of EUS are that it does not detect micrometastatic disease, and some cancers associated with significant esophageal stricturing cannot be analyzed.

Surveillance: Because of the alarming increase in the number of cases of esophageal adenocarcinoma from Barrett's esophagus, careful endoscopic screening for esophageal mucosal dysplasia and adenocarcinoma within a segment of Barrett's esophagus becomes imperative. This endoscopic surveillance is aimed at identifying cancers at an early curable stage if they are present. With minor modifications by our group, the following management approach is that recommended by the 1990 Barrett's Esophagus Working Party of the World Congress of Gastroenterology:

1. A program of regular endoscopic surveillance for dysplasia and early carcinoma is recommended for patients with Barrett's esophagus unless contraindicated by comorbidity. For patients who have no dysplasia or cancer, endoscopy (with procurement of biopsy and brush cytology specimens from the Barrett's epithelium) is performed every other year.

2. If mucosal dysplasia is detected at biopsy, these findings should be confirmed by at least one other pathologist expert in the evaluation of Barrett's esophagus. If any doubt remains, the endoscopic examination is repeated immediately to obtain more biopsy and cytology specimens for analysis.

Figure 2: Partially obstructing mid-esophageal cancer associated with Barrett's esophagus

3. For patients confirmed to have multiple foci of high-grade dysplasia, surgery is advised to resect all of the esophagus lined by this abnormal columnar epithelium.

4. For patients confirmed to have low-grade dysplasia, intensive medical antireflux therapy (including omeprazole) should be given for eight to 12 weeks, at which time the endoscopic examination is repeated to obtain multiple esophageal biopsy and cytology specimens.

  a. For patients whose specimens show histologic improvement, intensive surveillance (e.g. endoscopic examinations every six months) is recommended until at least two consecutive examinations reveal no dysplastic epithelium.

  b. For patients with persistent low-grade dysplasia, continued intensive treatment and surveillance are recommended. Surgical anti-reflux surgery can also be considered for those patients refractory to medical therapy.

References

1. Spechler SJ. Barrett's Esophagus. Sem. Gastro Dis 1996;7:51-60.

2. Regueiro MD, Spechler SJ. Current Guidelines for Gastroesophageal Reflux Disease. Contemp Int Med. 1996;8:61-68.

3. Kim R, Weissfeld JL, Reynolds JC, Kuller LH. Etiology of Barrett's Metaplasia and Esophageal Adenocarcinoma. Cancer Epidem 1997;6:369-377.

4. Regueiro MD, Spechler SJ. Complications of Gastroesophageal Disease. In: Brandt LJ (ed), Clinical Practice of Gastroenterology 1997 (in press).

5. Rosch T. Endosonographic Staging of Gastric Cancer: A Review of Literature Results. Gastro Endosc Clin of North Am 1995;5:549-557.

6. Margulies C, Kim R, Reynolds JC. Early Detection and Management of Esophageal Cancer. Comprehensive Therapy 1996;22:565-578.

7. Dent J, Bremner CG, Collen MJ, et al. Working Party Report to the World Congresses of Gastroenterology, Sydney 1990: Barrett's esophagus. J Gastroenterol Hepatol 1991;6:1-22.

Back to Contents

HOME | WHAT'S NEW | PROGRAMS & SERVICES | PHYSICIAN NEWSLETTER
RELATED SITES | DOCTORS
Search | Visit the Library | Visitors | E-mail Comments