Research
I often think of our research portfolio on a "knowledge vs clinical application" plot, describing how far away a research topic is from widespread clinical application (in my opinion) and what's the current knowledge status on the topic.Â
We have a range of projects covering a wide space of this graph, with the exception of the top left corner which is not so much our purpose as a research group in a clinical department.
Different people are excited by different types of projects: some love the exploration of data in a topic where little is known, others have limited amounts of time and want to learn how to use established methods on a well defined problem - we have something for everyone.
FLASH & ultra-high dose rate radiation
Using pre-clinical experiments we are studying the impact of FLASH on various endpoints. UW has two different FLASH capable proton delivery systems with different properties, a very unique setup. FLASH is currently a high impact topic in radiation oncology, and given our unique experimental capabilities and collaborators at Fred Hutch Cancer Center we can explore novel questions in this field.
We also study FLASH in the context of immune response and possible implications for combining FLASH with immunotherapeutic approaches. Projects in this area usually touch on both biological and physical questions.
Re-irradiation
Re-Irradiation is an emerging problem of high clinical relevance in radiation oncology. Better systemic agents and the emerging role of radiation in metastatic disease are leading to increasing numbers of re-irradiation, with little clinical data to guide RT dose constraints and patient selection.
Very little is know about the biological underpinnings of toxicity after re-irradiation. We need quantitative guidelines in the clinic, i.e. guidance how to actually treat our current patients. We believe that this will be best achieved by thorough analysis of patient data, and exploration of different models to fit the outcomes and toxicities we observe in the clinic now.
Immune Response to Radiotherapy
While animal experiments demonstrated that RT can synergize with immunotherapy, clinical trials have been negative up to now. We're studying the immune response in patients to understand how to make RT less immuno-suppressive (lymphocyte-sparing RT) and investigate biomarkers to personalize RT in the context of immunotherapy.
Traditional Biomarkers & Outcome Modeling
The first analysis to be done on any new clinical dataset involves traditional statistical methodologies. We have a range of projects in this space, from defining novel biomarkers of liver toxicity after radiotherapy to dosimetric predictors of cardiac toxicity.
Data-Driven Prediction Models & Digital Twins
We're not an AI lab, that is we are not looking at everything through that lens, but sometimes we encounter an impactful clinical challenge and data-driven techniques are just the best tool to solve the problem. We have used different types of prediction models and even tailor-made neural networks to predict clinical outcomes and understand relationships in complex datasets. See publication from Yejin Kim & Ibrahim Chamseddine.
