Esophageal Cancer: An Updated Review

Michael DiSiena, MD; Alexander Perelman, DO; John Birk, MD; Houman Rezaizadeh, MD


South Med J. 2021;114(3):161-168. 

In This Article

Risk Factors

On examining the epidemiology of AC and SCC in North America, unique risk factors can be seen. In the United States and other Western countries, SCC risk is three to five times higher in people who consume alcohol.[8] Hard liquor may pose a greater risk than beer or wine, but cumulative alcohol intake is more important than the type.[8] Cigarette smoking also leads to a three- to fivefold increase in SCC risk, and cigar and pipe smokers are at even higher risk. Fortunately, there is significant risk reduction (RR 0.59, 95% confidence interval [CI] 0.47–0.75) of SCC following 5 years of smoking cessation.[9] In smokers who also have heavy alcohol consumption, their risk of SCC is 20 times higher than those who never drank, highlighting the synergistic effect of cigarette smoking and alcohol.[8] These synergistic risk factors are present in 80% of SCC in men and 40% in women (Table 1).[8]

There is a correlation between SCC and head and neck cancer, and because the human papillomavirus (HPV) is associated with head and neck cancer, investigations of SCC and HPV have been conducted. A systemic review and meta-analysis evaluated 124 studies from around the world, totaling 13,832 cases of SCC, and found that the average prevalence of HPV was <40%, with HPV-16 being the predominant subtype.[19] It seems, therefore, that HPV infection may play a role in a subset of patents, but it is not as significant a risk factor as are alcohol and smoking.

In the highest prevalence areas of SCC, which include the EC belt region of Asia, eastern Asia, and southern and eastern Africa, studies have investigated dietary risk factors. Certain popular foods in these areas, including pickled vegetables (these contain N-nitroso compounds that induce carcinogenic alkyl to adduct to DNA), areca nuts and betel quid (these release copper, which induces collagen synthesis by fibroblasts), and hot tea and other high-temperature foods and beverages (that cause thermal injury) may play a minor role in the higher prevalence of SCC in these areas.[20–22] Other general dietary risk factors for SCC include red meat consumption, low selenium levels, and low dietary intake of fruits and vegetables.[23–25]

With specific geographic regions having significantly higher rates of SCC, the question of a hereditary component to SCC has been suggested; however, it is not clear whether this truly represents an inherited predisposition or lifestyle/environmental risk factors.[26] With this being said, there are inherited diseases that can predispose individuals to EC: Peutz-Jeghers syndrome is an autosomal-dominant syndrome consisting of multiple hamartomatous polyps in the gastrointestinal tract, mucocutaneous pigmentation and an increased risk of all cancers (RR 15.2), and especially gastrointestinal malignancies (including EC, RR 57);[27] tylosis (or Howell-Evans syndrome), a significant risk factor for SCC, is autosomal dominant, associated with a germline RHBDF2 mutation, and 90% of patients with the disorder will contract SCC before age 65;[28] Cowden syndrome, yet another autosomal dominant germline mutation (PTEN tumor suppressor gene), can cause a variety of phenotypic presentations, including hamartomatous gastrointestinal tumors and an increased cancer risk (although the major gastrointestinal malignancy to screen for in Cowden syndrome is CRC, there have been rare cases of EC reported; Table 1).[29]

There are also a number of esophageal and gastric diseases noted to have an increased risk of SCC. Patients with achalasia have a 16-fold increase in SCC within the first 24 years after diagnosis, with a mean time to diagnosis at 11 years after presentation.[30] Caustic (ie, lye) ingestion injury increases one's chances of SCC 1000-fold with a mean time to diagnosis of 41 years after ingestion.[31,32] Other esophageal diseases that increase one's risk of SCC include Plummer-Vinson syndrome (lifetime incidence rate 3% to 15% to develop cancer of the esophagus or pharynx), Fanconi anemia (particularly after bone marrow transplant, with a 2000-fold increase), celiac disease (odds ratio [OR] 3.72), and scleroderma with esophageal involvement (standardized incidence ratio [SIR] 2.86–35.0; Table 1).[33–38]

Although AC shares some similar risk factors with SCC (eg, cigarette smoking), the majority of AC is induced by gastroesophageal reflux disease (GERD), often causing Barrett's metaplasia. In fact, GERD for >20 years and severe symptoms of reflux have an AC development OR of 43.5.[39] It should be noted, however, that 50% of esophageal AC cases have no history of symptomatic reflux disease and antireflux surgery does not seem to decrease the risk of developing AC.[39–41] Patients with Barrett's esophagus have a 10-fold higher risk of developing AC, but overall, the absolute risk of developing AC in nondysplastic Barrett's is only 0.12% to 0.33%/year (Table 2).[45,46]

Obesity also has been associated with AC. This has been confirmed in studies of the International Barrett's and Esophageal Adenocarcinoma Consortium (BEACON).[47] There was initially a question of whether these were concomitant factors (increased acid reflux and obesity), and obesity itself was not a risk factor. A study using Mendelian randomization as an instrumental variables approach to causal inference using the data from studies participating in BEACON to quantify nonreflux effects and obesity, however, found obesity to be an independent risk factor (OR 1.16).[48] There is an inverse risk factor association of SCC with weight and smoking, initially attributed to smoking being a concomitant factor causing lower weight.

As noted above, cigarette smoking is a major risk factor for AC, but when it comes to alcohol use, BEACON showed no relation between AC and the amount, duration, or frequency of alcohol use.[49]

In terms of protective effects, nonsteroidal anti-inflammatory drug use has been shown to be protective for both SCC and AC; it is theorized that inhibiting the cyclooxygenase-2 enzyme production impedes tumor growth promoted by prostaglandins.[50–52] Another protective effect, but exclusive to AC, is Helicobacter pylori. Infected patients have been shown to have a lower risk of developing AC (this is theorized to be related to decreased gastric acid secretion).[53] When it comes to the protective effects of proton pump inhibitors (PPIs), the data are mixed on the extent of the protective effects of PPIs in developing AC, but there is evidence that the use of PPIs in Barrett's may improve the survival of patients who subsequently contract AC.[54]