Screening and Surveillance for Esophageal Cancer
Early detection is the key to the treatment of any gastrointestinal malignancy, and this is particularly true of esophageal adenocarcinoma. Intramucosal carcinomas have virtually no risk of metastasis, while cancers that penetrate into the deep mucosa or submucosa can have significant risks of dissemination. At the current time, screening and surveillance for esophageal cancer is still controversial. Screening asymptomatic populations for adenocarcinoma cannot be recommended because current studies suggest that cancers and predisposing conditions such as long-segment Barrett's esophagus are uncommon without symptoms. Screening patients with symptoms of heartburn who are older than 50 years of age may be of value because this is a higher-risk group, but this has not been proven in a prospective study. Squamous cell cancers of the esophagus are less common in the United States, and screening is not recommended except for very select subgroups such as patients with tylosis, Fanconi's anemia, and lye-induced strictures. The presence of risk factors such as long-term tobacco or alcohol use, achalasia, or squamous head and neck cancers may identify selected groups for screening.
Surveillance endoscopy has been used the most in the setting of Barrett's esophagus. Surveillance endoscopy is complicated by 2 major problems: sampling error (missing significant neoplastic lesions) and interpretation errors (histologic uncertainty regarding the presence of dysplasia). Sampling errors will be hopefully reduced in the future by the development of novel technologies that allow the endoscopist to visualize neoplastic lesions at the time of endoscopy, such as optical biopsy techniques or magnification techniques. Most have not been proven to be effective in trials, and some techniques such as methylene blue–assisted chromoendoscopy have been suggested to be potentially deleterious to the mucosa. The use of molecular markers to predict the occurrence of cancer has promise to improve the yield of surveillance biopsies and potentially augment histology.
At this time, surveillance endoscopy should be performed with random biopsies with 4 biopsy specimens taken at least every 1-2 cm of esophageal mucosa with additional biopsy specimens taken of any mucosal abnormality. Patients without dysplasia or mucosal abnormalities on their initial evaluation should be examined again in 1 year with surveillance biopsies to decrease the chances of sampling error. If no dysplasia is found again, surveillance can reasonably be deferred for another 5 years until the patient reaches a point at which cancer therapy is not possible or life expectancy is limited. Surveillance should be practiced only if the patient is anticipated to have a reasonable life expectancy and can tolerate treatment for esophageal cancer. In patients in whom low-grade dysplasia is found without mucosal abnormalities, it is recommended that endoscopy be repeated again in 1 year to be certain that there is no evidence of high-grade dysplasia or cancer. If low-grade dysplasia is confirmed by 2 pathologists, then the patient should be reexamined on a yearly basis because this group tends to have an increased risk of cancer. If there is disagreement about the presence of any dysplasia, surveillance can be deferred for 2 years. If high-grade dysplasia is detected and confirmed by 2 experienced pathologists, treatment with either surgical resection or endoscopic therapy can be recommended depending on the presence and nature of any mucosal abnormalities. The presence of multifocal high-grade dysplasia appears to be associated with an increased risk of development of cancer, and these patients might be more suitable for more aggressive treatment. Surveillance can be offered if both the patient and the physician are willing to follow a careful regimen of endoscopy every 3 months with at least 8 random biopsy specimens taken every 2 cm of involved esophagus. All mucosal abnormalities should be investigated with endoscopic ultrasonography and mucosal resection if these techniques are available to be certain that there is no underlying cancer. If ablative therapy has been performed, surveillance is still needed in the same area of involvement at least as frequently as if ablation had not been performed. Efforts should be made to examine the new squamous mucosa to determine if submucosal lesions might be present.
Chemoprevention for Esophageal Cancer
Chemoprevention would seem to be ideally suited to the problem of esophageal cancer, but studies have not proven that there is a specific treatment that would decrease cancer risk. There is strong epidemiologic and preclinical data that cyclooxygenase-2 inhibition might be of value as a chemoprevention agent, but no prospective clinical studies have been performed to prove that there is a reduction in risk for cancer. Control of acid or gastroesophageal refluxate should decrease inflammation and therefore decrease cancer formation; however, epidemiologic studies do not support the use of fundoplication for prevention of cancer. Similarly, there is no clear rationale for the use of high-dose proton pump inhibitors solely as a chemoprevention agent. Lifestyle modifications such as weight loss, stopping tobacco use, and eating fresh fruits and vegetables can be recommended based on epidemiologic evidence to decrease cancer risk.
Diagnosis and Staging of Esophageal Cancer
The diagnosis of esophageal cancer is established by flexible endoscopy with biopsy to confirm histologic evidence of disease. Computed tomography is recommended as the first staging study because detection of metastasis would clearly alter therapy. Endoscopic ultrasonography can also be recommended if available to evaluate esophageal cancer if computed tomography does not find any evidence of distant metastasis or unresectable cancer. Fine needle aspirates should be performed on regional lymph nodes if present and suspicious on endoscopic ultrasonography. If the cancer was diagnosed as mucosal based on ultrasonography, endoscopic mucosal resection could then be performed to stage and potentially treat early-stage cancer if it is available. In localities where endoscopic ultrasonography and endoscopic mucosal resection are unavailable, surgical resection should be performed for early-stage cancers. Positron emission tomography scans can be performed if all of the previous studies do not show evidence of metastatic disease because it may be more sensitive for distant metastasis, which would alter the therapeutic approach.
Treatment of Esophageal Cancer
Treatment of cancer is dependent on the stage of the cancer. Early cancers T1, N0, M0 by the American Joint Commission on Cancer are the most likely to be potentially curable. If the cancer is confined to the mucosa, these cancers are usually treated with esophagectomy, although endoscopic therapy has been shown to be effective in early squamous cell cancers of the esophagus treated in Japan using endoscopic mucosal resection. There is also some evidence that this approach may be successful in early adenocarcinoma. The risk of metastasis is very low if the cancer is confined to the mucosal layers. If there is penetration into the submucosa, the risk of metastasis becomes significant and esophagectomy would be recommended if there were no signs of distant metastasis or invasion of adjacent structures. If there is definite evidence of metastasis to regional lymph nodes, neoadjuvant chemotherapy in combination with radiation therapy administered before surgical resection may improve survival. Esophagectomy can be practiced by transhiatal or transthoracic routes. Minimally invasive esophagectomy has been advocated but is still associated with substantial morbidity and mortality. More advanced disease with metastasis to other organs or distant lymph node groups should be considered for palliative therapy with chemotherapy. Most commonly, combined therapy consists of chemotherapy with multiple courses of cis-platinum and 5-fluorouracil concomitantly given with ionizing radiation.
Palliation of advanced cancer can be achieved endoscopically with dilation, although relief is short-lived. Esophageal stenting is recommended for long-term palliation of cancers that are long and located at least 2 cm from the cricopharyngeal muscle. Expandable metal stents are preferred to plastic semirigid stents and are available in a coated form to decrease tumor ingrowth. Expandable plastic stents are reasonable for intermediate-term use when removal of stents is planned. Endoscopic tumor ablation for palliation of dysphagia can be performed with photodynamic therapy, alcohol injection, or laser therapy with similar efficacy in appropriate patients, although stents are clearly the preferred method of palliation. Nutritional support should be considered before chemotherapy or radiation therapy with enteral nutrition being the preferred method. Physicians must also keep in mind the emotional support required of these patients because esophageal cancer tends to be socially isolating as well as physically disabling.