Q&A: Christina Curtis on Computational and Systems Biology

Computational and systems biology is a fast-moving field of cancer research in which scientists use novel technologies and modeling approaches to quantify biological aspects of tumors, generating data that might point to methods of improving cancer detection, diagnosis, and treatment. “There is a huge degree of heterogeneity within tumors, and we're interested in moving from those observations to quantitative and mechanistic insights,” says Christina Curtis, PhD, who leads the Cancer Computational and Systems Biology Group at Stanford University School of Medicine in California.

Curtis discussed her research with Cancer Discovery's Catherine Caruso, some of which she will share at the American Association for Cancer Research (AACR) Annual Meeting 2020 in San Diego, CA, April 24–29.

What's the focus of your research?

We're interested in understanding tumor origins and the molecular basis of disease progression, including therapy resistance and metastasis. We characterize these processes in both primary patient samples and model systems—including patient-derived organoid models of human tumors—using all flavors and varieties of “omic” profiling, ranging from whole-genome sequencing to single-cell DNA, RNA, and ATAC sequencing, as well as spatial transcriptomic and proteomic techniques. We then use computational approaches to assimilate these diverse multidimensional data toward a systematic interpretation of the data and the development of prognostic and predictive biomarkers.

Where does your work fit on the basic-to-applied science spectrum?

Our work is highly interdisciplinary and spans the continuum of basic to translational science while bridging both computational and experimental techniques. I'm a basic scientist at heart and find myself coming back to ask, “Can we understand the mechanism? Can we define the molecular determinants? How do we interpret and integrate the measurements made with new technologies?” Quantification is an essential first step toward developing predictive models, so we often start there, but it's been rewarding to see some of our recent findings have potential for clinical translation.

You've been studying how tumors metastasize. What have you learned so far?

We've been investigating the timing and patterns of metastatic spread—including monoclonal versus polyclonal seeding of metastases and how they differ across common tumor types, such as breast, colon, and lung. We've focused on these tumors because they are relatively common and there are sizeable numbers of paired primary and metastatic tumor samples. For example, to delineate the timing of metastatic dissemination in colorectal cancer, we analyzed genomic sequencing data from paired primary and metastatic tumors and built a spatial computational model of tumor progression. We found that colorectal cancers often metastasize before the primary tumor is clinically detectable [Nat Genet 2019; 51:1113–22]. Next, we asked, “What are the molecular determinants of tumors that invade and metastasize early—tumors that are effectively ‘born to be bad’?” Extending this type of modeling to other tumor types revealed that early dissemination is not unique to colorectal cancer. We've also sought to characterize the dynamics of breast cancer relapse and the associated molecular drivers [Nature 2019;567:399–404]. Other recent work has highlighted the profound effect of treatment on cancer cell genomes. It will be important to better understand those mutational footprints, as well as how the tumor-immune microenvironment differs at metastatic sites and in the context of different treatments.

How does your approach vary when you study other tumor types or disease stages?

To understand the impact of therapy and identify determinants of response in breast cancer, we leverage the neoadjuvant paradigm, which represents a powerful model because it's possible to obtain repeat biopsies. In particular, we can obtain a biopsy prior to therapy, on therapy, and at the time of surgery—such longitudinal samples can help us to understand how the tumor and immune compartment change with treatment. To this end, we're now using spatially resolved proteomic technologies to study the tumor and immune compartment in situ, without dissociating the cells. We want to know, “Can we, with an additional biopsy, look at what has happened to the tumor during and after therapy?” This spatial proteomics work will likely be the topic of my talk at the AACR meeting.

Speaking of the AACR meeting, you will lead a session delving into some of the research approaches we've discussed. What's the goal of that session?

We're bringing together a panel of experts who use cutting-edge molecular technologies to dive deep into tumor-immune biology and whose work has both illuminated and helped to address some of the open questions in the field. I expect that this session will showcase some of these new technologies—and how they are being used to reveal new biological insights and to address pressing clinical questions ranging from disease etiology to response to immunotherapy.

For more news on cancer research, visit Cancer Discovery online at http://cancerdiscovery.aacrjournals.org/CDNews.