Objective:

To analyze the tumor microenvironment (TME) in locally advanced gastric cancer (LAGC) and identify spatial predictive biomarkers associated with neoadjuvant immunochemotherapy (nICT) response, emphasizing the need for reliable predictive markers.

Key Findings:

• Unique transcriptional patterns identified across tumor-core regions, immune cell infiltration zones, and stromal areas, with implications for targeted therapies.
• High expression levels of NOTUM, NKD1, and SERPINA3 correlated with major pathological response (MPR), indicating their potential as predictive biomarkers.
• Increased CD8+ T cell infiltration and a lower Treg/CD3+ ratio were associated with better treatment outcomes, suggesting immune landscape's role in therapy response.
• A six-gene signature linked to survival and immune infiltration was identified in TCGA-STAD data, warranting further exploration in clinical settings.

Interpretation:

NOTUM, SERPINA3, and CD8+ T cell density serve as spatially defined biomarkers for predicting nICT response in LAGC, highlighting the importance of TME analysis in personalized immunotherapy and potential clinical applications.

Limitations:

• Study limited to a small cohort size (19 patients for profiling, 20 for validation), which may affect the generalizability of the findings.
• Findings may not be generalizable to all gastric cancer populations, and potential biases or confounding factors should be considered.

Conclusion:

The study underscores the significance of spatial TME analysis in identifying biomarkers that can inform individualized immunotherapeutic strategies for LAGC, with implications for future research and clinical practice.