Clinical Study on the Evaluation of the Efficacy of Cervical Cancer

Official title

Study on the Application of Multi-omics in the Assessment of Efficacy and Prediction of Side Effects in Cervical Cancer

Overview

The main objectives of this study are to construct a multi-omics-based prognostic and side-effect prediction model for cervical cancer based on pre-treatment imaging, digital pathology, genomics, proteomics, molecular biology, metabolomics, and intestinal flora characteristics data of cervical cancer patients, combined with patients' clinical information, to guide the precise treatment of cervical cancer patients; and to deeply excavate the characteristics related to recurrent cervical cancer based on time-series multi-omics data. Construct an artificial intelligence auxiliary model for dynamic monitoring of cervical cancer recurrence based on longitudinal multi-omics. To provide a real-time and timely tool for clinical early prediction, early identification, early diagnosis and early intervention of cervical cancer, to prolong the survival time and improve the quality of patients' survival. 1. To realize multi-omics feature extraction of cervical cancer patients before treatment, and build a prognosis and side-effect prediction model of cervical cancer to guide accurate treatment; 2. To make iterative, comprehensive, real-time assessment of the risk of recurrence of cervical cancer based on time-series multi-omics data, and to build an early warning model for early identification and early diagnosis of recurrent cervical cancer; 3. To establish a prognostic and side-effect prediction and risk dynamic assessment model for cervical cancer, to build an intelligent decision support system, to implement the application of prognostic and side-effect prediction and dynamic monitoring model, to further assist in the precise diagnosis and treatment of cervical cancer, and to provide an accurate prognostic tool for identifying, diagnosing, and intervening in cervical cancer during the follow-up process.

Detailed description

1\. Construct a prognosis and side effect prediction model based on pre-treatment multi-omics features of cervical cancer patients.

1. Case selection: According to the overall experimental design, 2800 patients in the training group were used as the training data set, and 1200 patients in the validation group were used as the validation data collection. 2. Model training and tuning: a. Extract the multi-omics features of the training group, carry out self-learning of the features, and form a preliminary cervical cancer prognosis and side-effect prediction model; b. Input the multi-omics data of the validation group into the model, and carry out the structure of the model and the training parameters, and seek for the optimal model structure and training parameters; c. Determine the optimal cervical cancer prognosis prediction and side-effect model.

2.Mining recurrent tumor characteristics based on multi-omics data and constructing a comprehensive assessment model for recurrence risk .

1. Case selection: In accordance with the overall experimental design, 2800 patients in the training group were used as the training dataset, and 1200 patients in the validation group were collected as the validation data. 2. Model training and tuning: a. The multi-omics data features of the training group before the diagnosis of recurrence in previous follow-up visits are used to carry out self-learning of the features, assess the risk of tumor recurrence based on multi-omics features in the course of previous follow-up visits, form a dynamic, real-time recurrence risk assessment model, and derive a comprehensive risk value for the decision-making of recurrence intervention; b. Multi-omics features related to the previous follow-up visits of the validation group before the diagnosis of recurrence are inputted into the model, and the iterative time-series recurrence risk assessment is carried out on the patients. time-series recurrence risk iterative assessment of patients to assess the diagnostic performance of the model; c. Adjust the structure and training parameters of the model according to the segmentation accuracy of the validation group to seek the optimal model structure and training parameters; d. Use technical means such as data augmentation and other technical means to think of enlarging the sample size to improve the segmentation accuracy; e. Determine the optimal risk assessment model.

3\. Establish the prognosis and side-effect prediction and dynamic monitoring system of cervical cancer.

a. Docking the above constructed model with the outpatient system to construct a prognosis and side reaction prediction and dynamic monitoring system in the process of cervical cancer diagnosis and treatment; b. Constructing an intelligent decision support system through the prognosis and side reaction prediction and risk dynamic assessment model, implementing the application of recurrence prediction and dynamic monitoring system, and assisting the clinicians to make decisions on intervention measures.

Interventions

• Other Observational study
  Our study does not have any exposure factors.

Primary outcome measures

• Number of Circulating Tumor Cell Count [Time frame: From data of randomization until the date of first documented progression or date of death from any cause, whichever came first, assessed up to 5 years.]
• Concentration of Alpha-fetoprotein (AFP) [Time frame: From data of randomization until the date of first documented progression or date of death from any cause, whichever came first, assessed up to 5 years.]
• Concentration of Carcinoembryonic Antigen (CEA) [Time frame: From data of randomization until the date of first documented progression or date of death from any cause, whichever came first, assessed up to 5 years.]
• Concentration of carbohydrate antigen 199 (CA199) [Time frame: From data of randomization until the date of first documented progression or date of death from any cause, whichever came first, assessed up to 5 years.]
• Concentration of Squamous Epithelial Cell Carcinoma Antigen (SCC-Ag) [Time frame: From data of randomization until the date of first documented progression or date of death from any cause, whichever came first, assessed up to 5 years.]
• Concentration of carbohydrate antigen 125(CA125) [Time frame: From data of randomization until the date of first documented progression or date of death from any cause, whichever came first, assessed up to 5 years.]
• Count of Bacteria in urine [Time frame: From data of randomization until the date of first documented progression or date of death from any cause, whichever came first, assessed up to 5 years.]
• Count of bacteria in stool [Time frame: From data of randomization until the date of first documented progression or date of death from any cause, whichever came first, assessed up to 5 years.]
• 5-year overall survival rate of Participants [Time frame: From data of randomization until the date of first documented progression or date of death from any cause, whichever came first, assessed up to 5 years.]
• disease-free survival of Participants [Time frame: From data of randomization until the date of first documented progression or date of death from any cause, whichever came first, assessed up to 5 years.]

Eligibility criteria

Inclusion criteria

• Pathology: patients with pathologically confirmed cervical cancer
• Location: primary tumor of the cervix

Exclusion criteria

• Patients with no prior radiation therapy
• Patients without treatment
• Patients without regular follow-up

Criteria are shown verbatim from the registry (in English). Final eligibility is always assessed by the study center.

Healthy volunteers: Yes

Study design

Observational model

Cohort

Study locations

Center list to be confirmed — check the primary protocol.

Identifiers

NCT: NCT06254729 · YXJLRH2022039

Primary sources (government registries)

• ClinicalTrials.gov (US NIH)

View this study on ClinicalTrials.gov ↗