News
CISNET 2024 Annual Meeting
Experts and leaders in cancer research across all cancer sites gathered at the National Cancer Institute (NCI) Shady Grove Campus for the CISNET 2024 Annual Meeting.
We wish to highlight and recognize our newest member of the team, Jonathan Shuhendler, for his very first poster presentation! Jonathan shared his work which analyzed at a county-level whether increased availability of screening reduced disparities in prostate cancer mortality between Black and White men. He found that a greater availability of screening is associated with higher mortaility disparities, though unclear as to the reasons why. On the left is a picture of Jonathan posing with his poster during the Common Day poster session. In the middle, Jonathan is seen sharing his work with John Wong from the U.S. Preventative Services Task Force (USPSTF). And on the right, two Jonathans are spotted together (Jonathan Shuhendler and Jonathan Shoag)!
Another acknowledgement to Lukas Owens for his presentation "Counterfactual Predictions of Cancer Progression for Tailoring Secondary Treatment". He spoke about using the framework of joint longitudinal and failure-time models to tackle the problem of making counterfactual predictions of patient outcomes in the presence of a time-varying biomarker and tim-varying treatment. This framework is applied to the case of recurrent prostate cancer, where he and his collaborators developed a model to inform the decision of whetehr or not to start salvage therapy. Great job Lukas!
NCPF Multicancer Detection (MCD) Tests Workshop
The National Cancer Policy Forum (NCPF) hosted a 1.5-day workshop examining the state of the science for clinical use of multicancer early detection tests. The event included presentation and panel disussion on topics such as:
- Examples of current and emerging MCD tests.
- Challenges and opportunities to validate MCD tests and determine their clinical utility for detecting cancer and reducing cancer-specific mortality.
- Strategies for cancer care downstream of MCD testing, such as follow-up diagnostic testing and treatment decision-making.
- Limitations of MCD tests, including the burden on patients and health care systems from false-positive test results, overdiagnosis, and overtreatment.
- Research and policy gaps for assessing MCD tests and their impact on cancer care and outcomes and health equity.
We would like to give a big shoutout to Jane Lange for presenting her work, "Modeling multi cancer early detection: a tool to move between performance and outcomes", and our very own Ruth Etzioni for her presentation on "Diagnostic Performance The Many Faces of Screening Test Sensitivity", and fulfilling her critical role as a member of the Planning Committee.
Objectives and Key Results (OKR) Training
The Etzioni Lab would like to extend a very sincere Thank You to Hannah Berson, CEO of SALT Collaboratory, for leading our group through our very first Objectives and Key Results (OKR) Training! Through this workshop, we were able to learn how to set attainable goals for ourselves, learn what the necessary components and key features are for a good OKR, and begin thinking about how we can collaborate better to set ourselves up for success over the next year. Hannah guided us through a "Rose, Thorn, and Bud" exercise, in which each member listed things that were going well for them (Rose), things that weren't working for them (Thorn), and things that could be improved (Bud). We were able to visualize as a group what areas are going well for us, and what areas we need to improve.
The Etzioni Lab goes to JSM!
Members of our Etzioni Lab had the oppurtunity to travel to Portland, Oregon to attend the 2024 Joint Statistical Meeting (JSM), one of the largest statistical events in the world! JSM had over 5,000 attendees from over 52 countries, 600+ sessions, 100+ exhibitors on display, and more than 40 professional development short courses and workshops. Over the course of five days, our team was able to learn about a wide variety of topics ranging from statistical applications to methodology and theory to the expanding boundaries of statistics, exchange ideas with other statisticians in academia, industry, and government, and explore ideas for collaboration. A few of our own members even gave presentations on their outstanding work!
Yibai Zhao shared her poster titled "A Taxonomy of Sensitivity Measures in Studies of Cancer Early Detection Biomarkers.”
Lucas Liu presented his poster, "A Deep Learning-Statistical Framework for Uncertainty Mitigation of Noisy Labels on Pathology Image", discussing AI algorithms for predicting cellular composition from pathology images.
Lucas also gave a talk titled "Trustworthy AI for Temporal Electronic Health Records: Ensuring Fair Patient Risk Prediction."
Lukas Owens presented his talk, "Predictimands with Time-varying Treatments with Joint Longitudinal and Failure-time Models."
Our very own, Ruth Etzioni, spoke on a panel titled "Policy Evaluation: The Role of the Statistician."
Welcome our students from South Africa!
The Etzioni Lab hosted a Welcome Luncheon for our two interns who joined us from South Africa, Sarah Ogutu and Tshepo Maja. Our group enjoyed food from Ba Bar Green while listening to Sarah and Tshepo give presentations about growing up in South Africa and how they developed their interest in Biostatistics.
Amidst working hard on their projects, Sarah and Tshepo have been enjoying their time in Seattle!
Lucas Liu shares about AI in Cancer Research
The cancer research field is most impressed by AI's ability to learn and identify complex patterns from high-dimensional data which are usually imperceptible to human eyes. Oncological applications involving early cancer detection, precise tumor characterization, accurate prognostication and personalized treatment response prediction have been facilitated by AI models through the development of innovative assays that utilize patient genomic, molecular, and clinical data. These models need to be implemented in real-world clinical settings for the next step in oncology with AI. There is a need for interpretable and trustworthy AI systems that can construct comprehensive patient profiles by integrating multiple data sources. By leveraging AI's capacity to analyze vast amounts of complex data and generate actionable insights, oncologists can make more informed decisions, potentially leading to more effective treatments.
Lucas Liu is a Postdoctoral Fellow in the Etzioni lab whose research focuses on developing AI models for cancer research and precision oncology. One key part of his work is developing interpretable and trustworthy deep learning models to predict molecular alterations, survival outcomes, and recurrence from diverse data sources such as pathology images, radiology images, imaging reports, and omics data. He is dedicated to leveraging AI technologies to enhance cancer diagnosis and prognosis, optimize treatment strategies, and ultimately improve personalized cancer care and patient outcomes.
Lucas Liu presented posters at NAACCR, discussing AI algorithms for detecting recurrence and metastasis in population-based electronic radiology reports.
Lukas Owens presents at NAACCR
Lukas Owens gave his presentation "Trends in Age and PSA at Diagnosis in US Prostate Cancer Patients Suggest an Explanation for Increases in Metastatic Disease Incidence" at the 2024 North American Association of Central Cancer Registries (NAACCR) Annual Conference. This study examined features of prostate cancer at diagnosis during the 2010s to answer questions about trends in incidence of de novo metastatic disease.
Welcome our Summer Interns!
Deeana Nasrulai
Deeana Nasrulai is an undergraduate student at the University of Washington working towards a B.S. in Biochemistry and a minor in Data Science. She is a part of Fred Hutch’s 2024 SeattleStatGrow cohort. In Dr. Etizoni’s lab she is looking at the sensitivity of cancer screening in ovarian cancer cases.
Bruk Tefera
Born and raised in Seattle, Bruk Tefera is a current sophomore at the University of Southern California studying Computational Biology with an interest in pursuing a Masters in Biostatistics. He is currently working under Dr. Etzioni in examining racial and socio-economic disparities in prostate cancer treatment, accessibility, and outcome.
Jonathan Shuhendler
Seattle native Jonathan Shuhendler completed his BA in Mathematics & Economics at Western Washington University. He is now an intern at the Etzioni Lab, where he focuses on data analysis to better understand disparities in cancer mortality. He uses machine learning to identify the drivers of these disparities and conducts statistical simulations to inform patient decision-making on treatment options.
Dr. Caroline Thompson speaks on Cancer Diagnosis
Thank you to Dr. Caroline Thompson for presenting her talk "Decoding Cancer Diagnosis: Data-Driven Insights on Circumstances, Timeliness, Outcomes and Disparities." Her talk focused on the use of cancer registry linked healthcare data to examine barriers to early cancer diagnosis following symptomatic presentation, such as diagnostic delays, missed opportunities for diagnosis, and emergency department presentations, and the association of these barriers with downstream outcomes and disparities.
Dr. Thompson earned a PhD in Epidemiology at UCLA and was an Academy Health Delivery Systems Science Fellow at Palo Alto Medical Foundation Research Institute and Stanford University. Her research program in cancer data science and analytical epidemiology is primarily focused on patterns of healthcare delivery for cancer, especially around the period of cancer diagnosis.
Liu wins TDS IRC Poster Competition
Lucas Liu's poster titled "A Machine Learning Approach for Identifying Breast Cancer Recurrence Events in Population-based Claims Data" was awarded the Best AI/ML Poster at the 2023 Translational Data Science Integrated Research Center (TDS IRC) Annual Retreat. Congratulations, Lucas!
Meet the Group: Felipe Montano-Campos
Felipe is an incoming third-year PhD student at the Univeristy of Washington CHOICE Institute, specializing in health economics and outcomes research. Prior to joining UW, he earned a master's degree in Economics and Computer Science. During his time at the Duke Clinic Research Institute, he conducted research on patient preferences, specifically examining the trade-offs between risks and benefits associated with different treatments.
Currently, Felipe is focusing on precision screening and precision oncology. Leveraging his extensive knowledge of decision modeling, he aims to shed light on critical issues within the rapidly evolving field of public health and health decision making. By utilizing his expertise, Felipe hopes to contribute valuable insights to address pressing challenges in these areas.
Meet the Group: Kemal Caglar Gogebakan
Kemal is a brand-new Research Associate in the Etzioni Lab. Coming from Turkey with a PhD in econometrics, he has deeply immersed himself into the world of modeling and lead-time analysis for a variety of projects. Previously, he has worked with our collaborators at CEDAR , OHSU Knight Cancer Institute, as a postdoc for 4 years. He brings a lot of experience to the Etzioni lab, and we are glad to have him.
Lukas Owens, Yibai Zhao, and Mohammed Ahmed present at CISNET
Lukas, Yibai, and Mohammed presented at CISNET. Lukas’ presentation discussing prostate cancer recurrence and time to salvage therapy depending on various methods of treatment. Yibai’s presentation discusses the idea of whether Gleason 6 prostate cancer should be redefined as “non-cancer.” Mohammed’s presentation discusses the idea of various screening strategies to limit adverse risk and mortality for African American men. These thought-provoking topics were a highlight of the Mid-Year CISNET prostate cancer group meeting.
Stat News: Etzioni comments on MCD test
Stat News reports on Grail's efforts to get Medicare to pay for its cancer-screening test, Galleri.
CISNET semi-annual meeting in Seattle
CISNET semi-annual meeting was held in Seattle from June 26-30. The prostate group is shown here with Rocky Feuer, Tasha Stout, and Angela Mariotto (NCI) after a very successful meeting that showcased precision oncology, PSMA-PET/CT as a game-changing technology in prostate cancer management, predictive modeling of events in post-treatment progression, and the making of prostate cancer screening guidelines. Missing from the photo is Roman Gulati, who plays an key role in the CISNET modeling and co-ordinating center.
Yibai Zhao wins the JUICE Poster competition at CISNET midyear conference 2023
In the recent CISNET mid-year conference, Yibai Zhao represented her project “Degradation of screening test sensitivity estimated in a retrospective study when utilized in a prospective study” in the JUICE poster competition. She and her collaborators won one of the best poster awards.
The premise of the poster is to model the degradation of sensitivity of screening tests which can lead to the progression of higher rates of detection of more indolent cancers. As a result, it is important to look at the results of retrospective studies, which look at more symptomatic diseases meaning they are likely to be more advanced that are clinically significant. When imposed on prospective studies, the degradation increases with a longer mean sojourn time, meaning a higher prevalence of clinically insignificant cancer. This means cancers with longer sojourn times, the sensitivity confirmation is lowered; thus there is a higher fraction of clinically insignificant cancers.
New Gulati editorial
Roman Gulati wrote an editorial for the New England Journal of Medicine on the GÖTEBORG-2 trial.
Stat News: Etzioni comments on NEJM colorectal screening paper
Dr. Ruth Etzioni receives Outstanding Investigator Award
The seven-year, $7.4M Outstanding Investigator Award (OIA) from the National Cancer Institute will will provide long-term support for Ruth's work to investigate accuracy of multi-cancer liquid biopsies and other new technology. Details here.
Roman Gulati celebrates 17 years with the Etzioni Lab
A biostatistician, a disease modeler, a cancer epidemiologist, a computer programmer, a policy researcher, and now a Lab Elder. Congratulations and thanks, Roman! Your contributions cannot be overstated.
New Go-to Estimate for Overdiagnosis of Breast Cancer Screening in US
A study my colleagues and I published March 1 in the Annals of Internal Medicine offers a new estimate of the extent of overdiagnosis in contemporary U.S. practice and concludes that it might not be as frequent as we thought.
Media Coverage
- US News and World Report
- Wall Street Journal
- Stat News
- Center News
- Smithsonian Magazine
© 2025 Fred Hutchinson Cancer Center, a 501(c)(3) nonprofit organization.