> Our Work > The Work We Fund

2019 Movember-PCF Challenge Award

Eliezer Van Allen, MD
Eliezer Van Allen, MD

A Genomics-Guided Clinical Interpretation and Translational Discovery Engine for Prostate Cancer

Principal Investigators: Eliezer Van Allen, MD (Harvard: Dana Farber Cancer Institute), Nikolaus Schultz, PhD (Memorial Sloan Kettering Cancer Center), Wassim Abida, MD, PhD (Memorial Sloan Kettering Cancer Center), Deborah Schrag, MD, MPH (Harvard: Dana Farber Cancer Institute)

Co-Investigators: Haitham Elmarakeby, PhD (Harvard: Dana Farber Cancer Institute)

Description:

  • Prostate cancer is a highly variable disease, which differs across and even within patients. Understanding the biological and clinical variations and how best to treat individual patients is of great need. The ability to harness and learn from the vast amounts of existing multi-dimensional patient data will enable researchers and clinicians to identify and apply optimal precision medicine treatment strategies for all patients.
  • Eliezer Van Allen and team will develop a prostate cancer patient registry that will capture clinical data (which includes treatments and outcomes), pathology and imaging data, and data on tumor mutations, from over 10,000 patients spanning all stages of disease.
  • The team will develop artificial intelligence algorithms to identify clinical and/or molecular features that predict patient outcomes and responses to treatments. Relationships that are identified between genomic alterations and prostate cancer outcomes will be studied and validated in experimental models.
  • A significant amount of clinical information is included in medical reports in a non-standardized fashion. The team will develop biomedical natural language models that can extract information from medical reports including disease classification, regression, and survival prediction. The team will also build algorithms to read radiographic images and convert this information into standardized data.
  • All of these data will be combined with other patient clinical and molecular data in the registry and used to create a single model that can predict patient outcomes and treatment responses.
  • The team will also develop a digital clinical trials framework and an algorithm that will enable clinicians to match patients to precision medicine clinical trials based on their clinical, genomic, and other features.
  • If successful, this project will develop a rich data resource for the research and clinical community, as well as algorithms to learn from this data and to apply it directly toward making immediate optimal precision medicine treatment decisions for patients with prostate cancer.

What this means to patients: Precision medicine is an emergent clinical approach in which patients’ treatment plans are made based on their unique personal and tumor features. Dr. Van Allen and team will develop a prostate cancer registry that contains multi-dimensional patient data, and create algorithms that can predict patient outcomes. The team will also build a digital resource to match patients with optimal precision medicine clinical trials. This project will significantly accelerate precision medicine for all men with prostate cancer.