Description of the Evidence
Two histological types account for most malignant esophageal neoplasms: adenocarcinoma and squamous cell carcinoma. The epidemiology of these types varies markedly. In the 1960s, squamous cell carcinomas comprised over 90% of all esophageal tumors. The incidence of esophageal adenocarcinomas has risen markedly for the past 2 decades; it is now more prevalent than squamous cell carcinomas in the United States and Western Europe, with most tumors located in the distal esophagus.
Incidence and Mortality
In 2014, it is estimated that 18,170 Americans will be diagnosed with esophageal cancer and 15,450 will die of this malignancy. Of the new cases, it is estimated that 14,660 will occur in men and 3,510 will occur in women. Incidence rates generally increase with age in all racial/ethnic groups. In black men, however, the incidence rate for those aged 55 to 69 years is close to that of whites aged 70 years and older. In black women, aged 55 to 69 years, the incidence rate is slightly higher than that of white women aged 70 years and older.
Although the overall incidence of squamous cell carcinoma of the esophagus is declining, this histologic type remains six times more likely to occur in black males than in white males. In contrast, the incidence of adenocarcinoma of the esophagus rapidly increased from the 1970s to the mid-1990s.
Male gender is an important predictor of adenocarcinoma of the esophagus. The attributable risk is low enough in women that, although the risk from gender is not modifiable, other risk factors necessarily have limited impact.
Squamous Cell Carcinoma of the Esophagus
Factors with adequate evidence of increased risk of squamous cell carcinoma of the esophagus
Smoking cigarettes and drinking alcohol
In the United States, squamous cell carcinoma of the esophagus is strongly associated with tobacco and alcohol abuse. The relative risk associated with tobacco use is 2.4, and the population attributable risk is 54.2% (95% confidence interval [CI], 3.0–76.2).[5,6] Retrospective cohort studies adjusted for tobacco use have shown a twofold to sevenfold increase in risk of esophageal cancer in alcoholics compared with rates for the general population. Case-control studies have also suggested a significantly increased risk of cancer of the esophagus associated with alcohol abuse.
In a multicenter, population-based, case-control study of 221 patients with esophageal squamous cell carcinoma and 695 controls, ever-smoking, alcohol consumption, and low fruit and vegetable consumption accounted for 56.9% (95% CI, 36.6%–75.1%), 72.4% (95% CI, 53.3%–85.8%), and 28.7% (95% CI, 11.1%–56.5%) of esophageal squamous cell carcinomas, respectively, with a combined population attributable risk of 89.4% (95% CI, 79.1%–95.0%).
In China, where the overall prevalence of esophageal carcinoma is much higher than in the United States, esophageal cancer is associated with deficiencies of nutrients, such as retinol, riboflavin, alpha-carotene, beta-carotene, alpha-tocopherol, ascorbate and zinc, and with exposure to specific carcinogens (e.g., N-nitroso compounds).
Factors with adequate evidence of decreased risk of squamous cell carcinoma of the esophagus
A prospective, placebo-controlled, esophagus chemoprevention study randomly assigned 610 high-risk Chinese patients. Patients were aged 35 to 64 years and received either placebo or combined low-dose retinol (15 mg or 50,000 IU) plus riboflavin (200 mg) and zinc gluconate (50 mg) for 13.5 months. Standard histological evaluations (including two endoscopic biopsies) were conducted for 93% of all entered patients. Micronuclei from esophageal cells were obtained before therapy began and after the 13.5 months of treatment. Serum levels of vitamin A, beta-carotene, riboflavin, and zinc were obtained at 0, 2, and 13.5 months.
The second report of this study presented micronuclei frequency results. A statistically significant reduction in the mean percentage of micronucleated esophageal cells occurred in the active-treatment group compared with the placebo group. The pattern of cell proliferation, another potential intermediate endpoint marker, also improved.
Aspirin and nonsteroidal anti-inflammatory drug (NSAID) use
A systematic review and meta-analysis of the association between aspirin and NSAID use and esophageal cancer identified two cohort and seven case-control studies published between 1980 and 2001. Pooled results showed a protective association between aspirin/NSAID use and esophageal cancer (odds ratio [OR], 0.57; 95% CI, 0.47–0.71). The association with aspirin use was statistically significant (OR, 0.50; 95% CI, 0.38–0.66); the association with NSAIDs was borderline significant (OR, 0.75; 95% CI, 0.54–1.0). Aspirin/NSAID use was associated with lower risk of both adenocarcinoma (OR, 0.67; 95% CI, 0.51–0.87) and squamous cell carcinoma (OR, 0.58; 95% CI, 0.43–0.78).
Adenocarcinoma of the Esophagus
Factors associated with increased risk of adenocarcinoma of the esophagus
Gastroesophageal reflux disease (GERD)
The most important epidemiological difference between squamous cell carcinoma and adenocarcinoma is the strong association between GERD and adenocarcinoma. The results of a population-based case-controlled study suggest that symptomatic gastroesophageal reflux is a risk factor for adenocarcinoma of the esophagus. The frequency, severity, and duration of reflux symptoms were positively associated with increased risk of adenocarcinoma of the esophagus. In a case-control study from Sweden, the OR was 7.7 for patients with recurrent reflux symptoms, while the OR for patients with long-standing and severe symptoms was 43.5 (95% CI, 18.3–103.5). The probable mechanism is that long-standing GERD is associated with the development of Barrett esophagus, a condition in which an abnormal intestinal-type epithelium replaces the stratified squamous epithelium that normally lines the distal esophagus; Barrett esophagus is considered a precursor of esophageal adenocarcinoma. The intestinal-type epithelium of Barrett esophagus has a characteristic endoscopic appearance that differs from squamous epithelium. Dysplasia in Barrett epithelium represents a neoplastic alteration of the columnar epithelium that may progress to invasive adenocarcinoma.
A population-based cohort study in Sweden shows that patients with Barrett esophagus develop adenocarcinoma of the esophagus at about 1.2 cases per 1,000 person-years of follow-up monitoring, which is about 11.3 times higher than in the general population. Thus, while the relative risk may be elevated, the absolute risk is still not high. Furthermore, over half of the cases of adenocarcinoma of the esophagus are not associated with GERD symptoms.
An interesting hypothesis associates the rise in the incidence of adenocarcinoma of the esophagus to a declining prevalence of H. pylori infection in Western countries. Reports have suggested that gastric infection with H. pylori may protect the esophagus from GERD and its complications. According to this theory, H. pylori infections that cause pangastritis also cause a decrease in gastric acid production that protects against GERD. Patients whose duodenal ulcers were treated successfully with antibiotics developed reflux esophagitis twice as often as patients whose infections persisted. Other suggested factors to explain the increased risk of adenocarcinoma of the esophagus include obesity  and the use of medications such as anticholinergics that can predispose to GERD by relaxing the lower esophageal sphincter.
Interventions with adequate evidence of decreased risk of adenocarcinoma of the esophagus
Aspirin and NSAID use
A systematic review and meta-analysis of the association between aspirin and NSAID use and esophageal cancer identified two cohort and seven case-control studies published between 1980 and 2001. Pooled results show a protective association between aspirin/NSAID use and esophageal cancer (OR, 0.57; 95% CI, 0.47–0.71). The association with aspirin use was statistically significant (OR, 0.50; 95% CI, 0.38–0.66); the association with NSAIDs was borderline significant (OR, 0.75; 95% CI, 0.54–1.0). Aspirin/NSAID use was associated with lower risk of both adenocarcinoma (OR, 0.67; 95% CI, 0.51–0.87) and squamous cell carcinoma (OR, 0.58; 95% CI, 0.43–0.78).
Radiofrequency ablation in dysplastic Barrett esophagus
A randomized controlled trial  assessed whether radiofrequency ablation (vs. sham ablation) could eradicate dysplastic Barrett esophagus and decrease the rate of neoplastic progression in patients with Barrett esophagus and dysplasia. Among patients with low-grade dysplasia, eradication of dysplasia occurred in 90.5% of the treatment group compared with 22.7% in the control group; in the high-grade dysplasia group, rates were 81.0% in the treatment group compared with 19.0% in the control group. Additionally, 77.4% of patients in the ablation group had complete eradication of intestinal metaplasia, compared with 2.3% in the control group. Patients in the ablation group had less disease progression, and although cancer was not a primary outcome because expected numbers were small, there were fewer cancers in the ablation group (1.2% vs. 9.3%; P = .045). The complication rate was relatively low; among 84 treated patients, there was one upper gastrointestinal hemorrhage and five strictures that were easily treated.
This study suggests that the treatment of patients with Barrett esophagus and dysplasia may ablate Barrett esophagus and prevent disease progression, but the study provides only weak evidence about whether treatment reduces the outcome of esophageal cancer (because it was not designed to answer that question). Evidence from the study suggests that ablation does not simply coagulate and hide dangerous cells under the surface of the esophagus (those cells could later evolve to cancer). A question entirely separate from this study is whether patients should be screened for Barrett esophagus (this study focused on treatment of patients with Barrett who had been identified as having dysplasia). Furthermore, the study does not discuss the net benefits and harms of an overall program of screening (e.g., of screening patients with GERD or certain GERD symptoms) and the surveillance of patients with Barrett esophagus. The potential for overdiagnosis and overtreatment may be considerable if physicians used results of this study to treat patients with Barrett esophagus and no dysplasia.
- Holmes RS, Vaughan TL: Epidemiology and pathogenesis of esophageal cancer. Semin Radiat Oncol 17 (1): 2-9, 2007. [PUBMED Abstract]
- American Cancer Society: Cancer Facts and Figures 2014. Atlanta, Ga: American Cancer Society, 2014. Available online. Last accessed November 24, 2014.
- Devesa SS, Blot WJ, Fraumeni JF Jr: Changing patterns in the incidence of esophageal and gastric carcinoma in the United States. Cancer 83 (10): 2049-53, 1998. [PUBMED Abstract]
- Hur C, Miller M, Kong CY, et al.: Trends in esophageal adenocarcinoma incidence and mortality. Cancer 119 (6): 1149-58, 2013. [PUBMED Abstract]
- Oesophagus. In: World Cancer Research Fund, American Institute for Cancer Research: Food, Nutrition and the Prevention of Cancer: A Global Perspective. Washington, DC: The Institute, 1997, pp 118-129.
- Siemiatycki J, Krewski D, Franco E, et al.: Associations between cigarette smoking and each of 21 types of cancer: a multi-site case-control study. Int J Epidemiol 24 (3): 504-14, 1995. [PUBMED Abstract]
- Engel LS, Chow WH, Vaughan TL, et al.: Population attributable risks of esophageal and gastric cancers. J Natl Cancer Inst 95 (18): 1404-13, 2003. [PUBMED Abstract]
- Muñoz N, Wahrendorf J, Bang LJ, et al.: No effect of riboflavine, retinol, and zinc on prevalence of precancerous lesions of oesophagus. Randomised double-blind intervention study in high-risk population of China. Lancet 2 (8447): 111-4, 1985. [PUBMED Abstract]
- Muñoz N, Hayashi M, Bang LJ, et al.: Effect of riboflavin, retinol, and zinc on micronuclei of buccal mucosa and of esophagus: a randomized double-blind intervention study in China. J Natl Cancer Inst 79 (4): 687-91, 1987. [PUBMED Abstract]
- Yang GC, Lipkin M, Yang K, et al.: Proliferation of esophageal epithelial cells among residents of Linxian, People's Republic of China. J Natl Cancer Inst 79 (6): 1241-6, 1987. [PUBMED Abstract]
- Corley DA, Kerlikowske K, Verma R, et al.: Protective association of aspirin/NSAIDs and esophageal cancer: a systematic review and meta-analysis. Gastroenterology 124 (1): 47-56, 2003. [PUBMED Abstract]
- Lagergren J, Bergström R, Lindgren A, et al.: Symptomatic gastroesophageal reflux as a risk factor for esophageal adenocarcinoma. N Engl J Med 340 (11): 825-31, 1999. [PUBMED Abstract]
- Spechler SJ, Goyal RK: The columnar-lined esophagus, intestinal metaplasia, and Norman Barrett. Gastroenterology 110 (2): 614-21, 1996. [PUBMED Abstract]
- Van Dam J, Brugge WR: Endoscopy of the upper gastrointestinal tract. N Engl J Med 341 (23): 1738-48, 1999. [PUBMED Abstract]
- Reid BJ, Blount PL, Rabinovitch PS: Biomarkers in Barrett's esophagus. Gastrointest Endosc Clin N Am 13 (2): 369-97, 2003. [PUBMED Abstract]
- O'Connor HJ: Review article: Helicobacter pylori and gastro-oesophageal reflux disease-clinical implications and management. Aliment Pharmacol Ther 13 (2): 117-27, 1999. [PUBMED Abstract]
- Graham DY, Yamaoka Y: H. pylori and cagA: relationships with gastric cancer, duodenal ulcer, and reflux esophagitis and its complications. Helicobacter 3 (3): 145-51, 1998. [PUBMED Abstract]
- Labenz J, Blum AL, Bayerdörffer E, et al.: Curing Helicobacter pylori infection in patients with duodenal ulcer may provoke reflux esophagitis. Gastroenterology 112 (5): 1442-7, 1997. [PUBMED Abstract]
- Lagergren J: Controversies surrounding body mass, reflux, and risk of oesophageal adenocarcinoma. Lancet Oncol 7 (4): 347-9, 2006. [PUBMED Abstract]
- Lagergren J, Bergström R, Adami HO, et al.: Association between medications that relax the lower esophageal sphincter and risk for esophageal adenocarcinoma. Ann Intern Med 133 (3): 165-75, 2000. [PUBMED Abstract]
- Shaheen NJ, Sharma P, Overholt BF, et al.: Radiofrequency ablation in Barrett's esophagus with dysplasia. N Engl J Med 360 (22): 2277-88, 2009. [PUBMED Abstract]