Three different subsite classification systems for carcinomas in the proximity of the GEJ, but is it all one disease?

BACKGROUND AND AIM

Currently there are three different subsite classification systems for carcinomas in the region of the gastroesophaeal junction (GEJ), namely ICD-O, Munich and Liverpool. The aim of the present study was to compare clinicoepidemiological, pathological and molecular features of adenocarcinomas in the proximity of the GEJ, classified according to their position with respect to the GEJ initially, and then classified according to ICD-O, Munich and Liverpool classifications.

METHODS

Forty-seven adenocarcinomas in the proximity of the GEJ were subdivided into groups 1 (exclusively within esophagus), 2 (mainly within esophagus but extending distally across GEJ), 3 (equally present in esophagus and stomach) or 4 (mainly in stomach but extending proximally across GEJ), and analysis of their clinicoepidemiological, pathological and molecular features was performed. Molecular characterization included loss of heterozygosity (LOH) and microsatellite instability analyses.

RESULTS

Group 3 carcinomas were younger than other carcinomas in the proximity of the GEJ, but otherwise these carcinomas were similar in their clinicoepidemiological, pathological and molecular features. There were no significant differences between esophageal (groups 1 and 2) and gastric (groups 3 and 4) carcinomas as classified by ICD-O. Munich types I (groups 1 and 2), II (group 3) and III (group 4) carcinomas of the GEJ were also similar. LOH at the site of the Rb tumor suppressor gene and at 17p11.1-p12 was more common in GEJ (groups 2, 3 and 4) than lower third (group 1) esophageal carcinomas classified according to the Liverpool system.

CONCLUSION

Overall, adenocarcinomas of the lower esophagus and adenocarcinomas involving the GEJ have similar clinicoepidemiological, pathological and molecular features no matter which subsite classification is used, adding further evidence that they represent the same disease.