This study is a multi-center, observational study aiming at developing a machine learning-based early detection model using prospectively collected liquid biopsy samples from clinically-annotated individuals.
Peripheral blood samples from participants with new diagnosis of pancreatic cancers will be collected to characterize the cancer-specific circulating signals by sequencing cell free DNA. A noninvasive test integrating machine learning algorithm will be trained and validated through a two-stage approach in recruited well-classified individuals, along with non-cancers without clinical diagnosis of cancer after routine medical screening. The performance of liquid biospy assays discovering cancer from non-cancer will be evaluated in participants with benign disease as well as average risk individuals.
Study Type
OBSERVATIONAL
Enrollment
276
Department of Pancreatic and Hepatobiliary Surgery, Fudan University Shanghai Cancer Center; Pancreatic Cancer Institute, Fudan University
Shanghai, Shanghai Municipality, China
RECRUITINGDepartment of Hepato-Biliary-Pancreatic-Splenic Surgery, Shanghai General Hospital
Shanghai, Shanghai Municipality, China
RECRUITINGDepartment of General Sugery, Huadong Hospital
Shanghai, Shanghai Municipality, China
Performance of cfDNA methylation-based model for discriminating pancreatic cancer versus non-cancer
Sensitivities of cfDNA methylation-based model at specificity of 99% and 95%, respectively.
Time frame: 12 months
Performance of models using multi-omic data for discriminating pancreatic cancer versus non-cancer
Sensitivities of model integrating multi-omics data at specificity of 99% and 95%,respectively.
Time frame: 12 months
Performance of pre-defined model in clinical sub-groups of interest
Sensitivity of pre-defined model in different pathological subtypes or different age groups or tumor marker-negative cases.
Time frame: 12 months
This platform is for informational purposes only and does not constitute medical advice. Always consult a qualified healthcare professional.