GEJ cancers: gastric or esophageal tumors? searching for the answer according to molecular identity

Abstract

The 7th edition of Union for International Cancer Control (UICC) staging system moved gastroesophageal junction (GEJ) cancers from gastric to esophageal group. Since clinical management is strongly influenced by this staging system, we looked at molecular fingerprints of GEJ tumors and compared to gastric and esophageal profiles. We aimed at elucidating whether GEJ cancers cluster with gastric or esophageal groups according to mRNA and microRNA expression pattern, since this might represent tumor identity. The clinical and expression data were downloaded from The Cancer Genome Atlas (TCGA) with 395 stomach, 184 esophagus and 521 colon samples for mRNA analyses and 392 stomach, 175 esophagus and 459 colon samples for microRNA comparisons. Both Principal Component Analysis (PCA) and Heat Map plots were performed in R platform, using Log2 transformation of RPKM normalized data. Differential Expression Analysis was also performed in R, using RAW data and the DESeq2 package. The mRNAs and microRNAs were tagged as differentially expressed if they met the following criteria: i) FDR adjusted p-value < 0.05; and ii) |Log2 (fold-change)| > 2. Esophagus squamous cell carcinoma (ESCC) clustered apart of the others tumors, while adenocarcinomas (AC) clustered all together according to both mRNAs and microRNAs expression patterns. The HMs of the differentially expressed mRNAs and microRNAs also demonstrated that ESCC belongs to a different group, while AC molecular signature of esophagus looks like AC of the cardia and non cardia regions. Even distal gastric cancers are quite similar to AC of the lower esophagus, demonstrating that esophagus AC relies much closer to gastric cancers than to esophagus cancers. By using robust molecular fingerprints, it was strongly demonstrated that GEJ tumors looks more like gastric cancers than esophageal cancers, despite of tumor heterogeneity.

Pan-cancer study detects genetic risk variants and shared genetic basis in two large cohorts

“…Fourth, we grouped esophageal and stomach cancers despite possible differences in their risk factor profiles. However, there is precedent for using a composite phenotype 54 , and analyses of stomach and esophageal tumors suggest that they have many overlapping molecular features 55,56 . In addition, sensitivity analyses for each cancer alone gave similar results, suggesting that they may have similar genetic bases despite potentially having different environmental risk factors.…”

Pan-Cancer Study Detects Novel Genetic Risk Variants and Shared Genetic Basis in Two Large Cohorts

“…However, there is precedent for using a composite phenotype, 60 and analyses of stomach and esophageal tumors suggest that they have many overlapping molecular features. 61,62 In addition, sensitivity analyses for each cancer alone gave similar results, suggesting that they may have similar genetic bases despite having different environmental risk factors. Finally, we focused solely on individuals of European ancestry.…”

Application of a Neural Network Whole Transcriptome–Based Pan-Cancer Method for Diagnosis of Primary and Metastatic Cancers

“…Our findings support observations in literature that physiologically proximal and morphologically similar cancer types, such as stomach adenocarcinomas and some esophageal adenocarcinomas, are highly similar at the whole-transcriptome level, in spite of having distinct clinical designations. 15 , 34 It also reflects the existing challenge with using glass-based pathology (even with the aid of immunohistochemistry) to discern these tumors.…”