• Yun-Suhk Suh, Deukchae Na, Ju-Seog Lee, Jeesoo Chae, EuiHyun Kim, Giyong Jang, Jieun Lee, Jimin Min, Chan-Young Ock, Seong-Ho Kong, Joshy George, Chengsheng Zhang, Hyuk-Joon Lee, Jong-Il Kim, Seong-Jin Kim, Woo Ho Kim, Charles Lee, and Han-Kwang Yang.
• Department of Surgery, Seoul National University College of Medicine, Seoul, Korea.
• Ann. Surg. 2022 Apr 1; 275 (4): 706-717.

ObjectiveTo investigate the molecular characteristics of AGEJ compared with EAC and gastric adenocarcinoma.Summary Of Background DataClassification of AGEJ based on differential molecular characteristics between EAC and gastric adenocarcinoma has been long-standing controversy but rarely conducted due to anatomical ambiguity and epidemiologic difference.MethodsThe molecular classification model with Bayesian compound covariate predictor was developed based on differential mRNA expression of EAC (N = 78) and GCFB (N = 102) from the Cancer Genome Atlas (TCGA) cohort. AGEJ/cardia (N = 48) in TCGA cohort and AGEJ/upper third GC (N = 46 pairs) in Seoul National University cohort were classified into the EAC-like or GCFB-like groups whose genomic, transcriptomic, and proteomic characteristics were compared.ResultsAGEJ in both cohorts was similarly classified as EAC-like (31.2%) or GCFB-like (68.8%) based on the 400-gene classifier. The GCFB-like group showed significantly activated phosphoinositide 3-kinase-AKT signaling with decreased expression of ERBB2. The EAC-like group presented significantly different alternative splicing including the skipped exon of RPS24, a significantly higher copy number amplification including ERBB2 amplification, and increased protein expression of ERBB2 and EGFR compared with GCFB-like group. High-throughput 3D drug test using independent cell lines revealed that the EAC-like group showed a significantly better response to lapatinib than the GCFB-like group (P = 0.015).ConclusionsAGEJ was the combined entity of the EAC-like and GCFB-like groups with consistently different molecular characteristics in both Seoul National University and TCGA cohorts. The EAC-like group with a high Bayesian compound covariate predictor score could be effectively targeted by dual inhibition of ERBB2 and EGFR.Copyright © 2020 Wolters Kluwer Health, Inc. All rights reserved.

15 keywords...

hide…