Biomarker-guided immunotherapy in gastric cancer: current insights and future perspectives

<h2>Abstract</h2> Gastric and gastroesophageal junction adenocarcinoma (GC) is a biologically challenging malignancy associated with suboptimal clinical outcomes due to limited effective treatment options. The recent incorporation of immune checkpoint inhibitors (ICIs) into therapeutic algorithms has improved the clinical prospects of subsets of GC patients. However, responses to anti-PD-1/PD-L1 agents remain highly heterogeneous, with only some patients deriving long-term benefits. This variability highlights the importance of identifying optimal biomarkers to enhance patient selection, thereby enabling tailored immunotherapy strategies. Whereas microsatellite instability has demonstrated a potent capacity for predicting immunotherapy benefits in GC, other predictive biomarkers, such as PD-L1 expression, remain suboptimal. Advances in gene expression and epigenetic profiling, liquid biopsy approaches, gut microbiome characterization, and artificial intelligence-driven multimodal algorithms applied to multi-omics or digital pathology are key drivers for the comprehensive characterization of the GC tumour microenvironment (TME), which could be used for better treatment selection. Similarly, elucidating the complex tumour-immune interplay with these technologies will be crucial for the success of novel immunotherapeutic approaches under clinical development, by evaluating alternative immune pathways alone or in combination with current actionable targets of GC. The current review aims to give an overview of the current immunotherapeutic landscape in GC, evaluate standard-of-care and emerging biomarkers of immunotherapy response, and discuss the translational potential of incorporating multi-omic and AI-derived biomarkers into biomarker-enriched clinical decision-making frameworks.