Screening strategies for prevalent esophageal adenocarcinoma may be different from those recommended for Barrett's esophagus

Feb. 13, 2021

The incidence of esophageal adenocarcinoma (EAC) has increased significantly within the Western Hemisphere over the past three to four decades. Despite some advances in diagnosis and treatment, EAC is a serious form of cancer with a five-year survival rate of only 10% to 20%.

Known risk factors for EAC include Barrett's esophagus (BE), gastroesophageal reflux disease (GERD), male sex, white race, central obesity and tobacco use. Characterized by a progression of metaplastic and dysplastic changes in the distal esophagus, BE is thought to be the sole known precursor for EAC. Patients with BE have a risk of EAC that is 30 to 125 times greater than that found in the general population. The current understanding is that BE can transition from nondysplastic BE to low-grade dysplasia to high-grade dysplasia to EAC.

While this association between BE and EAC has been well established, most patients with EAC present without a known history of BE. Researchers have thus begun to question the effectiveness of current screening strategies and the assumption that BE screening criteria apply to all cases of EAC. To address these issues, Mayo Clinic researchers recently conducted an age- and sex-matched case-control study comparing risk factors among patients diagnosed with EAC and BE, and the results were published in The American Journal of Gastroenterology in 2021.

Study methods and results

The researchers first collected data from an age- and sex-matched cohort that included 1,356 patients with a mean age of 63.6 years, including 678 with prevalent EAC, 678 with BE and 198 women (14.6%). Comparing these two groups, the researchers noted no significant differences in smoking status or body mass index (BMI). Analysis of the data yielded the following results:

  • Patients diagnosed with BE were more likely to report GERD (present for more than six months) than were patients with EAC (84.5% versus 52.7%).
  • Patients diagnosed with BE had a higher prevalence of the following conditions than did patients with EAC: diabetes (37.6% versus 29.2%), hyperlipidemia (82.2% versus 45.7%), hypertension (64.6% versus 48.2%), nonalcoholic steatohepatitis (18.4% versus 12%) and metabolic syndrome (44.1% versus 28.5%).
  • HbA1c and alanine aminotransferase levels also were significantly higher in patients with BE compared with patients with EAC.
  • Coronary artery disease was the only metabolic disease more common in patients with EAC than in patients with BE.

In a cohort of 532 patients that compared 266 patients with BE and 266 with early-stage (Stages I and II) EAC, the researchers observed the following:

  • No significant differences among the two groups in smoking status, BMI, nonalcoholic steatohepatitis, low-density lipoproteins, alanine aminotransferase levels and HbA1c
  • A significantly higher prevalence of GERD in the group with BE (84.5%) than in the group with early-stage EAC (65.9%)
  • Higher prevalence rates in the group with BE than in the group with early-stage EAC for diabetes (43.2% versus 33.5%), hyperlipidemia (85% versus 53%), hypertension (67.7% versus 67.5%) and metabolic syndrome (49.2% versus 35%)

According to David A. Katzka, M.D., a gastroenterologist at Mayo Clinic's campus in Rochester, Minnesota, who served as the article's corresponding author, the study findings suggest that there are two pathways to EAC.

"GERD symptoms were an unreliable predictor of EAC and early EAC because half the patients did not report reflux symptoms," explains Dr. Katzka.

Dr. Katzka and co-authors explain that the traditional pathway to EAC likely involves a slow progression from metaplasia to cancer. This pathway from BE to EAC provides time for metabolic diseases to develop and may allow inflammatory cytokines and growth factors to play a role.

"Our results suggest the existence of an alternative pathway — one in which metaplasia progresses rapidly to EAC, with or without intestinal metaplasia," says Dr. Katzka. "This accelerated pathway provides less time for developing metabolic consequences of central obesity."

Because the estimated presence of metabolic diseases is significantly higher in patients with EAC than in the general population, the researchers believe that this factor may still play a role in both pathways.

Dr. Katzka notes that the high percentage of patients with EAC who lack GERD symptoms may also mean that current screening and surveillance guidelines for this disease should be reevaluated.

"Our findings also suggest that the currently accepted criteria for risk factors that predict the presence of Barrett's esophagus on endoscopy — central obesity, age greater than 50, chronic heartburn, white race and male sex — are not necessarily those that predict the presence or development of early esophageal adenocarcinoma. This is important, as it is really the latter that we are searching for. Furthermore, patients with Barrett's who go on to develop cancer may be in a different group, in some respects, than those who never develop cancer. Overall, these findings suggest that we have to develop better guidelines. And, in the meantime, we should not view the society-endorsed parameters as absolute requirements for screening for esophageal adenocarcinoma."

Dr. Katzka says that further research is needed to classify these two groups of patients with EAC phenotypically and genetically.

"Our group is already finding that there are two distinct histologic phenotypes of esophageal adenocarcinoma — one with intestinal metaplasia and the other without — that behave differently with regard to prognosis, stage of presentation and response to therapy," explains Dr. Katzka."We are also examining the sensitivity and specificity of gastroenterology society guidelines more systematically in their accuracy to detect esophageal adenocarcinoma. In particular, we are looking at how they perform in the population of patients with prevalent cancers, a group in which more than 40% lack symptomatic heartburn and in which central obesity is commonly absent."

For more information

Sawas T, et al. Risk factor profiles can distinguish esophageal adenocarcinoma from Barrett's esophagus. The American Journal of Gastroenterology. 2021;116:198.