A Prospective, Multicenter, Real-World Cohort Study for the Development and Validation of a Multimodal Artificial Intelligence System to Predict Response to Neoadjuvant Chemo-Immunotherapy in Locally Advanced Gastric Cancer (The PRISM-GC Study)

Inclusion Criteria

Inclusion Criteria: Age ≥ 18 years. Histologically confirmed gastric or gastroesophageal junction adenocarcinoma. Clinical stage cT3-4a, N+, M0 (locally advanced) assessed by CT/MRI and endoscopic ultrasound. Scheduled to receive neoadjuvant chemotherapy combined with PD-1 inhibitors (regimens including but not limited to SOX/XELOX + Sintilimab/Tislelizumab/Camrelizumab, etc.) as standard of care. Availability of standard pre-treatment contrast-enhanced abdominal CT images. Willingness to provide peripheral blood samples and tumor tissue (biopsy/surgical) for sequencing and analysis. ECOG performance status 0-1. Adequate organ function to tolerate systemic chemotherapy. Exclusion Criteria: Evidence of distant metastasis (Stage IV) or unresectable disease. Previous systemic anti-tumor therapy for gastric cancer (chemotherapy, radiotherapy, or immunotherapy). History of other malignancies within the past 5 years. Active autoimmune diseases requiring systemic immunosuppressive treatment (contraindication for PD-1 inhibitors). Emergency surgery due to obstruction, perforation, or uncontrolled bleeding. Severe metallic artifacts on CT images that interfere with radiomic feature extraction. Pregnancy or lactation.