Accurate Molecular Classification of mCRPC for Precision Treatment Through Multiomic Analysis of Circulating Tumor DNA
Abstract
Metastatic castration-resistant prostate cancer is the advanced stage of the disease that is lethal with no cure. The tumors of advanced prostate cancer can be classified into different groups, or subtypes, each having different properties that make up its phenotype. These subtypes harbor differences in the cancer cell’s molecules that can change as the disease worsens or no longer responds to the right treatments. Therefore, to direct the proper treatments to the right patients, we need to accurately classify prostate tumors into subtypes using “precision medicine.” Tumor tissue is required for making clinical decisions, but it is often difficult to obtain, which is a major limitation of current treatment strategies for patients with advanced prostate cancer. To inform clinical care more rapidly, we need to be able to determine the tumor subtype during therapy in “real-time.” An exciting method that is gaining popularity is to use a simple blood draw, a “liquid biopsy,” to search for DNA that is shed from tumor cells called circulating tumor DNA (ctDNA). This strategy has been useful to find mutations in DNA, or the genotype, of prostate tumors. However, the genotype alone does not always reliably provide the full picture of what underlies the phenotype of the tumor, and ultimately, its ability to respond to treatment. In this proposal, we will use an innovative approach to predict the phenotype of the tumor directly from ctDNA. We will develop new computer algorithms to study different layers of the cellular machinery that regulates genes using standard sequencing of ctDNA molecules. By employing machine learning techniques, we will combine different sources of data to classify the tumor subtypes and to identify tumor changes when treatments fail. To develop and test the utility of our approach, we will study ctDNA from the blood of mice grafted with human tumors and from the blood of prostate cancer patients receiving next-generation and experimental therapies. We believe that our new approach will lead to a new paradigm of diagnostics that will accurately select patients for the appropriate life-prolonging therapy in a rapid and non-invasive way. Because our approach does not require specialized experiments, its adoption into clinical platforms can be achieved more readily. In the near term, this research has the potential to critically advance precision medicine and transform the standard of care to improve the lives of patients with lethal prostate cancer.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Dec 05, 2021
- Source ID
- W81XWH2110513
Entities
People
- Gavin Ha
Organizations
- Fred Hutchinson Cancer Center
- United States Army