Study to Develop and Validate an AI System Predicting Response to Neoadjuvant Chemo-Immunotherapy for Locally Advanced Gastric Cancer

What this Trial Is About

Gastric cancer is a significant health challenge worldwide, and this research aims to develop a new Artificial Intelligence AI system to predict how well tumors respond to combined chemotherapy and immunotherapy before surgery. The study focuses on patients with locally advanced gastric cancer receiving standard neoadjuvant chemo-immunotherapy. The goal is to create a tool that helps doctors personalize treatment plans by predicting tumor response more accurately. Participants will receive standard neoadjuvant chemotherapy combined with approved PD-1 inhibitors, chosen by their treating physician, in a real-world clinical setting. Researchers will collect preoperative contrast-enhanced CT scans, pathological tissue slides, and biological samples like blood and tumor tissue. These will be analyzed using advanced deep learning methods to build and validate a multimodal AI model predicting pathological complete response. During the study, patients will undergo baseline assessments and postoperative pathological evaluations about five months apart. Researchers will monitor the accuracy of the AI model in predicting tumor response, including the rate of complete tumor disappearance after surgery. They will also track disease-free survival for up to three years. The study involves collecting imaging, tissue, and blood samples, as well as clinical data, to support AI development and patient monitoring throughout the treatment process.

Multimodal AI for Predicting Response to Neoadjuvant Immunotherapy in Gastric Cancer (PRISM-GC)