Deep Learning-Based Prediction of Gastric Cancer Response to Neoadjuvant Chemotherapy
试验速览
- 阶段
- 不适用
- 状态
- 招募中
- 入组人数
- 200
- 试验地点
- 22
- 主要终点
- Accuracy of TRG prediction by the AI model
研究概览
简要总结
This study seeks to develop a deep-learning-based intelligent predictive model for the efficacy of neoadjuvant chemotherapy in gastric cancer patients. By utilizing the patients' CT imaging data, biopsy pathology images, and clinical information, the intelligent model will predict the post-neoadjuvant chemotherapy efficacy and prognosis, offering assistance in personalized treatment decisions for gastric cancer patients.
详细描述
This study seeks to develop a deep learning model to predict the outcomes of neoadjuvant chemotherapy in patients with gastric cancer. Leveraging participants' CT scans, biopsy pathology images, and clinical profiles, this model aims to forecast the effectiveness of post-neoadjuvant chemotherapy and the subsequent prognosis, thereby aiding in individualized treatment choices for these participants.
Data Collection: The investigators will gather data from 1,800 retrospective cases and 200 prospective cases from multiple hospitals. The retrospective data will be divided into training and testing sets to train and validate the model, respectively. The model's performance will subsequently be evaluated using the prospective dataset.
Clinical Information: This encompasses the participant's gender, age, tumor markers, staging, type, specific treatment plans, pre and post-treatment lab results, etc.
Imaging Data: CT imaging data taken within one month prior to the neoadjuvant chemotherapy, with at least the venous phase CT imaging included.
Pathology Data: Pathology images from a gastric tumor biopsy stained with Hematoxylin and Eosin (HE) taken within one month prior to treatment.
研究设计
- 研究类型
- Observational
- 观察模型
- Cohort
- 时间视角
- Prospective
入排标准
- 年龄范围
- 18 Years 至 —（Adult, Older Adult）
- 性别
- All
- 接受健康志愿者
- 否
入选标准
- •Age 18 years or older;
- •Pathologically diagnosed with advanced gastric cancer in accordance with the American AJCC's TNM staging standards;
- •Have not undergone any systematic anti-cancer treatments before neoadjuvant chemotherapy and have not had surgery for local progression or distant metastasis;
- •Received standard neoadjuvant chemotherapy as recommended by the clinical guidelines, and have documented treatment details;
- •CT imaging and biopsy pathology images strictly taken within one month prior to starting neoadjuvant treatment;
- •Patients possess comprehensive preoperative clinical information and post-operative TRG grading.
排除标准
- •Patients whose CT or pathology images are unclear, making lesion assessment infeasible;
- •Patients diagnosed with other concurrent tumors.
研究组 & 干预措施
Gastric Cancer Patients Undergoing Neoadjuvant Chemotherapy
This group comprises participants diagnosed with advanced gastric cancer. The participants will be treated with standard neoadjuvant chemotherapy regimens recommended by clinical guidelines. Treatment details, including the generic name of the drugs, dosage form, dosage, frequency, and duration, will be recorded according to the specific regimen.
干预措施: Neoadjuvant Chemotherapy (Drug)
结局指标
主要结局
Accuracy of TRG prediction by the AI model
时间窗: two months
Accuracy measures the proportion of true positive and true negative predictions made by the AI model among all predictions. It indicates the capability of the model to correctly classify patients into their respective TRG gradings.
Area under the receiver operating characteristic curve (AUC) for TRG prediction by the AI model
时间窗: two months
The AUC will be used to evaluate the performance of the AI model in predicting TRG grading of gastric cancer patients after neoadjuvant chemotherapy. An AUC of 1 indicates perfect prediction, while an AUC of 0.5 indicates prediction no better than chance.
次要结局
- Overall Survival (OS) at 5 years(Five years)
- Progression-Free Survival (PFS) at 3 years(Three years)
研究者
Di Dong
Professor
Institute of Automation, Chinese Academy of Sciences