The highest risk of diagnosis with esophageal adenocarcinoma occurs in the first month after diagnosis with Barrett’s esophagus (BE).
Further, the risk of this type of esophageal cancer is lower than previously believed after that first year, according to a study of 7,932 Swedish residents who were diagnosed with BE from 2006 to 2013.
Massachusett's General Hospital website
Barrett's Esophagus: The red tissue is Barrett's mucosa and the white lining is normal, squamous mucosa.
Esophageal adenocarcinoma (EA) is the most common type of esophageal cancer in the United States, and more than half of patients present with late-stage disease. “The connection between BE and EA is well known, and risk factors for both conditions remain vastly the same, including male sex, age over 50 years, central obesity, and symptomatic GERD [gastroesophageal reflux disease],” Dr. Holmberg said in an interview.
“In order to reduce mortality from esophageal cancer through surveillance and early treatment, we need to improve on BE screening programs,” he said. He also noted that, of all EA cases in Sweden, where screening for BE itself is not recommended in national guidelines, an “astonishing” 97% were diagnosed in patients with previously unknown BE.
Screening programs urge regular endoscopic surveillance in nondysplastic BE patients, but Dr. Holmberg said that their cost-effectiveness is questionable. Complicating matters, there has been conflicting research about the risk of EA that these patients actually face.
As the new study notes, patients with BE were previously thought to face a 30-125 times greater relative risk of EA than the general population did and a mean annual risk of 0.5%. But, recent studies in Europe and the U.S. have suggested that the risk is lower, with an estimated annual risk range of 0.12%-0.18%.
Dr. Holmberg said he and his colleagues launched the new study to better understand the risk in the first year after BE diagnosis.
The researchers used the Swedish National Patient Register to track 7,932 Swedish residents who were diagnosed with BE from 2006 to 2013. About two-thirds were male, and the median age at BE diagnosis was 66 years.
The findings appear in the European Journal of Cancer (2017;75:41-6). Of the patients with BE, 89 new cases of EA were diagnosed 62% of the time within a year of BE diagnosis.
Over a median follow-up of 2.13 person-years, the overall incidence of EA in the patients with BE was 1.47 (95% CI, 0.91-2.02) per 1,000 person-years. That is the equivalent of an annual risk of 0.15%.
The highest incidence rate by far was in the period of 7 to 30 days after diagnosis at 15.53 (95% CI, 4.77-26.29). The incident rates varied from 1.44 to 4.10 (95% CI, 0.00-7.38) for other periods (31-100 days, 101 days to 6 months, over 6 months to 1 year, over 1 year to 3 years, over 3 years to end of follow-up).
How should physicians interpret the study findings and look forward? “Nondysplastic BE represents almost 80% of all BE,” Dr. Holmberg said, but “nondiscriminatory long-term surveillance in nondysplastic BE is not cost-efficient.”
However, while this is “the group with the least risk of EA development, it still generates a large chunk of all EA cases. It is likely that future management of nondysplastic BE will include more risk factors than just degree of dysplasia.”
He added that “individuals with multiple risk factors, such as Caucasian males, age over 50 years, obesity, symptomatic reflux, and with long-segment nondysplastic BE would likely benefit more from more frequent surveillance or even prophylactic treatment with radio-frequency ablation compared to individuals without those risk factors.”
The study was funded by the Swedish Cancer Society and the Swedish Research Council. The authors report no relevant disclosures.