Welcome to the Etzioni Lab: Modeling and Data Science for Cancer Diagnostics and Precision Oncology

We are a team of biostatisticians, computer scientists, and health economists with expertise in surveillance science, survival analysis, joint longitudinal and survival methods, deep learning, and simulation modeling. We develop models and methods to inform decisions about how best to use and act on results of cancer diagnostics including novel biomarkers, imaging tests, and AI-based predictions. For many years we focused on prostate cancer screening, as the co-ordinating center for the CISNET Prostate group, and our models have informed national screening guidelines from the AUA, ACS, and NCCN. Our group was the first to estimate overdiagnosis due to prostate cancer screening in the US and we have published studies on best practices for estimating overdiagnosis in prostate and breast cancer. We are actively developing new approaches for evaluating multi-cancer early detection tests and are involved with the leadership of the EDRN, the CSRN, BEACON, and CAIA.

  Multi-cancer early detection:

• Examining methods to estimate diagnostic performance across studies.
• Building models to support clinical recommendations for multi-cancer screening.
• Developing a new playbook for cancer screening trials.
• modeling multi-cancer early detection in breast cancer survivors

Models for precision diagnostics:

• Dynamic predictions to support salvage therapy predictions in recurrent prostate cancer using joint models
• PSMA-PET/CT imaging as a precision diagnostic at prostate cancer recurrence
• Circulating tumor DNA for predicting prostate cancer survival among patients on Pluvicto
• Digital pathology to predict tumor mutation status 
• Learning from PSA and treatment trajectories for trustworthy survival prediction in metastatic prostate cancer

Applied research:

• Leads the Biostatistics Core for the Pacific Northwest Prostate Cancer SPORE.
• Central consulting resource for prostate cancer investigators at Fred Hutch and the University of Washington.