Predicting Gastric Cancer Response to Chemo With Multimodal AI Model

A Radio-Pathomic Multimodal Machine Learning Model for Predicting Pathological Complete Response to Neoadjuvant Chemotherapy in Advanced Gastric Cancer: A Retrospective Observational Study

This study aims to develop a multimodal model combining radiomic and pathomic features to predict pathological complete response (pCR) in advanced gastric cancer patients undergoing neoadjuvant chemotherapy (NAC). The researchers intended to collected pre-intervention CT images and pathological slides from patients, extract radiomic and pathomic features, and build a prediction model using machine learning algorithms. The model will be validated using a separate cohort of patients. This research intend to build a radiomic-pathomic model that can outperform models based on either radiomic or pathomic features alone, aiming to improve the prediction of pCR in gastric cancer.

Inclusion Criteria:

• patients with histologically confirmed adenocarcinoma of the stomach or esophagogastric junction who received NAC and radical gastrectomy;
• patients who underwent abdominal multidetector computed tomography (CT) inspection, gastroscope, and tumor tissue biopsy before any intervention started;
• Lesions that are assessable according to The Response Evaluation Criteria in Solid Tumors Version 1.1

Exclusion Criteria:

• Patients with indistinguishable tumor lesions on the CT images due to insufficient filling of the stomach during the CT inspection;
• patients without indistinguishable tumor cell on the pathological slides due to inadequate sampling;
• patients with insufficient data.