Satisfactory Debulking Prediction Model for Advanced Ovarian Cancer Based on PET-CT Image Data and Its Clinical Application

Ovarian cancer is the gynecological malignant tumor with the highest fatality rate. More than 70% of patients are diagnosed with advanced stage, often involving various organs of the pelvis and abdomen, which increases the difficulty of surgical resection, and the 5-year survival rate is only 30%. Surgical treatment is the cornerstone of the treatment of ovarian cancer, and whether it can achieve satisfactory tumor reduction is an important factor affecting the prognosis of ovarian cancer. At present, the methods used to evaluate whether satisfactory tumor reduction can be achieved include Suidan score based on CT image and Fagotti score based on laparoscopic exploration, but there are problems such as low sensitivity, poor specificity or strong subjectivity, and the efficiency of predicting satisfactory tumor reduction is only about 60%. In recent years, PET-CT has been widely used in tumor diagnosis. Pet-ct combined with PET metabolic imaging technology and traditional CT scanning can help to distinguish the nature of tumors, assess the systemic tumor load, define the scope of the lesion, and provide the metabolic status of various parts of the body. The application value of PET-CT related imaging features and metabolic information in ovarian cancer needs to be clarified. Our team's previous study found that PET-CT related images and metabolic information showed certain advantages in predicting satisfactory resection of ovarian cancer, and the AUC reached 0.85, which was better than the current CT image score and laparoscopic score. Therefore, this project intends to conduct a multicenter retrospective study to evaluate the satisfactory tumor reduction rate of advanced ovarian cancer using PET-CT images to guide clinical practice and predict the prognosis of patients. At the same time, we will explore the molecular biological characteristics and clinical relevance of ovarian cancer through the combination of high-throughput sequencing genomics and radiomics.