Landscape of biomarkers and actionable gene alterations in adenocarcinoma of GEJ and stomach - A real-world data analysis

After a long period of therapeutic stagnation, a significant breakthrough in the treatment of adenocarcinomas of the gastroesophageal junction (GEJ) and stomach (GC) is now becoming evident with the implementation of Her2neu-based targeted therapies (trastuzumab and trastuzumab-deruxtecan) and the use of PDL1 checkpoint inhibitors (nivolumab, pembrolizumab).The required companion diagnostics regarding Her2neu overamplification or PDL1 expression are performed protein-based by immunohistochemistry (IHC) or additionally by in situ hybridization (FISH) in case of a Her2neu score of 2+.However, there are investigator-dependent differences in the assessment of Her2neu overamplification and in PDL1 scores obtained by IHC/FISH. The investigator-dependent differences could occur due to the quality of the tumor sample, the heterogeneous antigen expression of the biopsy, or the interpretation of the data. The use of high-throughput technologies such as next generation sequencing (NGS) has the potential to standardize the analyses and thus make them more comparable.In the presented study, we analyzed real-world multigene sequencing data from 75 patients diagnosed with GEJ and GC. We compared the results of conventional Her2neu diagnostics (IHC and FISH) with NGS findings of ErbB2 overamplification. Furthermore, we correlated the results of microsatellite instability (MSI) and tumor mutation burden (TMB) analyses by NGS with PDL1 protein expression (CPS) in IHC. In addition, several other potential therapeutic targets in GEJ and GC have been reported in the literature, such as the PI3K/Akt/mTOR pathway (potential drug of RADPAC trial: everolimus), c-MET gene variants (potential drug: c-MET inhibitor: tivantinib), EGFR family gene variants (ErbB-1/HER1, ErbB-2 (new, HER2), ErbB-3 (HER3) and ErbB-4 (HER4)). In our study we show the distribution of potentially actionable gene variants based on our real-world data.