top of page

Case Organized By:


Physicist and Bioinformatician


Metastatic Ovarian Cancer


Case Concluded


We worked on Shirley's case from Fall 2021 through Summer 2022. Analysis for her case was featured in Research to the People @ Stanford's First Annual Symposium. A full retrospective of Shirley's case will be available through publication.



I was diagnosed with ovarian cancer stage IIIc in September, 2013. The diagnosis followed several months of vague, intermittent symptoms, starting with GI issues, then pelvic pain, bloating, and finally difficulty breathing. I visited my primary care physician several times and my ob/gyn once, before more or less diagnosing myself prior to a second urgent visit to the ob/gyn. A whirlwind of scans, consultations, and lab tests followed, leading up to a tumor debulking surgery less than a week after my realization that something was terribly wrong. This is a sadly typical course for ovarian cancer, which is diagnosed at advanced stage in >75% of cases. Five-year survival at this stage was approximately 30% and had barely improved since the advent of standard of care surgery plus platinum/taxol treatment in the 1990s. 


I am a computational biologist and was highly motivated to survive to raise my children, who were 9 and 3 years old at the time of diagnosis. I had worked at iconic institutions such as NASA and Caltech, where solving difficult science and engineering problems was a part of daily life. Thus, even before my surgery, I began researching my disease and, in particular, the most aggressive treatments that might give me more time. Intraperitoneal chemotherapy, though difficult to withstand, was linked to longer survival in certain patients. Thus, soon after beginning chemotherapy, I switched hospitals to be under the care of an oncologist who had experience getting women through that grueling regimen. I pursued all genetic testing options available to me, including getting my tumor DNA and RNA sequenced under a research protocol at a translational genomics institute. Analysis of the tumor DNA revealed a mutation that indicated a great response to chemotherapy. What a relief! 


But that relief proved short-lived. Less than a year after enduring eight cycles of high dose chemotherapy, including twelve intraperitoneal infusions, I recurred in my abdomen. No one could tell me why I had recurred so soon, but it did not bode well. 


During the prior year, I had leveraged my background in genomics research and begun studying patterns of gene expression in ovarian cancer. Through a friend, I connected with a machine learning researcher who was working on a new type of algorithm that he thought might be applied to the ovarian cancer data I was interested in. With the help of this algorithm and data from The Cancer Genome Atlas, I began to understand why certain women lived longer than others. I used this to inform my treatment choices and had a surprisingly long remission following my early recurrence. I shared my story publicly and became an advocate for personalized treatment based upon patient-centered research. In the seven plus years since my diagnosis, I have continued to pursue out of the box treatment options based upon my tumor’s molecular characteristics and help others do the same. I endeavored to be guided by the science, hoping it could keep me one step ahead of the disease. Unfortunately, in the fall of 2020, my disease entered a new, yet more aggressive phase. I continue to believe the answer for me lies in science, the only question being whether that answer will come quickly enough. 



  • ​Biologist recruits USC researcher to study cancer — her own - USC News

  • Cancer in the time of Algorithms - USC Viterbi Magazine

  • Keynote at Precision Medicine 2017 - Harvard Medical School Institute of Biomedical Informatics

  • Diagnosed With Ovarian Cancer, a Researcher Mined TCGA Data to Study Her Own Disease - NIH NCI​



  • Medical Records.

  • Labs.

  • Imaging.

  • Whole Genome Sequencing.

  • Transcriptomics.

  • Immune Cell Profiling.

  • Longitudinal Multi-omics.

  • More.

Research Objectives

Drafted by Shirley.


What significant change occurred to allow aggressive invasiveness, i.e. extensive infiltration of vital organs in 2020, whereas for prior seven years, it had existed apparently only in abdominal cavity surfaces and lymph nodes.

  • This should be discernible from recent samples vs. original tissue or prior recurrences. Blood analyses may be informative? Microenvironment/terrain? 

  • How can such a change be reversed/inhibited to cause metastasis regression?



Is the immune system playing a role? How can this be exploited for therapeutic purposes? 

  • Can markers of immune recognition of tumor be measured?

  • Can markers of immune environment, e.g. systemic immune dysregulation, systemic or tumor-specific immunosuppression be usefully assessed? E.g. tumor-associated macrophages, tumor-promoting NK cells…


What it driving parp inhibitor resistance and how?

  • Is there a way to determine functional/dynamic HRD, as opposed to genomic scarring

  • Are mutations in DNA damage response pathways driving resistance? Epigenetic changes? Efflux pumps? Other?

  • Can this particular resistance be modified to restore/increase parp inhibitor sensitivity?


These generally play a significant role in ovarian cancer progression. 

  • Are there specific structural variants in my tumor that are informative? 

  • How does copy number of genes relate to resistance or tumor phenotype?



PARPi drugs are synthetic lethal to HRD tumors, but resistance developed relatively quickly in my case. Are there other synthetic lethal targets that can be exploited?



In solid tumors generally, chemoresistance increases, likely due to increased cellular plasticity.

  • Can this plasticity be characterized in my tumor?

  • Is there a way to reduce the cellular “options” either pharmacologically or via supplements?



Are non-cancerous stromal/secretory/tissue/immune cells promoting the tumor?

  • Can we assess this from tumor tissue, non-tumor tissue, or blood?

  • What are strategies to reduce pro-tumor TME? 



Can we measure this for my tumor? How does this relate to other factors and therapeutic options?



What is the comprehensive list of potentially targetable genes with altered genetics or expression that may indicate targeted therapy, including those in clinical trials? Especially perhaps antibody drug conjugates that have fewer side effects and are also a modality I have not tried?



What role are noncoding RNAs and non-exonic DNA changes playing? There has already been a splicing variant detected in my tumor by Tempus that was useful for therapy. 

  • How can we analyze whole genome be examined for alterations to regulatory regions? 

Shirley with Blue Background.png
bottom of page