Endometrial cancer is the third leading cause of cancer-related deaths in women, with 75,000 deaths in 2012. But a disproportionate number of these deaths are due to a particular, rare subtype: carcinosarcoma.
Most tumors are made up of cells that largely stick to a single pattern of growth, however abnormal that pattern may be. But in a uterine carcinosarcoma (UCS), cells suffer from an “identity crisis” – they switch off certain genes specific to the tissue they came from, helping them fit in more easily in different tissues, spreading the cancer. Within a single tumor, one can find several different cell types, and even elements that would never occur on their own within the uterus.
We wanted to understand why these tumor cells are so good at spreading to other parts of the body, and how we could stop that from happening and reduce the number of deaths from endometrial cancer. We hypothesized that the ability of these cells to adopt different guises may explain why they can spread so readily to different sites of the body.
In our study in Neoplasia, we created a map showing which genes were switched on and off in different parts of the tumor to create the cells’ disguises, providing a “signature” of these switches throughout the genome. Switching these genes back on could lead to effective treatments in the future.
Studying separate components
Carcinosarcoma cells are fascinating, as they show a unique ability to jump horses in mid-stream, switching from one cell type to another. In designing this project, we hoped to get at the basis for this behavior. Since epigenetic marks are partly responsible for enforcing cell types, we wanted to determine if there were consistent epigenetic aberrations in carcinosarcoma.
What we learned was that this was much harder to study than we expected. Carcinosarcoma always consists of multiple components, and we wanted to study these separately. To determine how these cells manage to switch from one type to another, we started from three human UCS samples, which we microdissected into their two component parts: carcinoma and sarcoma. This required us to learn to work with very small amounts of material, and to find creative ways to use large public databases to validate our findings.
While others had studied the genetic mutations in these tumors, we were interested in studying epimutations – changes in the way the DNA is decorated that can turn genes on or off. In the past, epigenetic changes were difficult to study on a genome-wide basis. Our laboratory has pioneered several methods that make it possible to construct whole-genome methylation maps at single-nucleotide resolution. With these improved tools, we can now reveal epigenetic changes in cancers, which may well be just as significant as genetic mutations.
We learned that some regions of the DNA are consistently hypermethylated (have more decorations) in UCS, and other regions are consistently hypomethylated (have fewer decorations). These changes switched off some tumor suppressor genes (KLF4, NDN, and WT1), giving us potential targets to switch back on and suppress tumor growth.
The value of sharing knowledge
Sharing information, data and knowledge has been very valuable to us and this research. Having access to electronic data sharing was vital: we used a collaboration tool – an electronic lab notebook where members of the group record their data. It’s lightyears ahead of the old model, in which lab notebooks are something you keep on a shelf.
A tool for research data management
Researchers have their own ways of managing and explaining the data they collect, and if that information is hidden away in a lab book, it’s unlikely to be useful to anyone else. That’s particularly unfortunate if the researcher changes institutions, leaving no usable record of their work. In 2016, Elsevier acquired Hivebench – a lab notebook tool that puts the essential first step in research data management at the researcher’s fingertips.
“Saving researchers time by providing them with a user-friendly way to store and manage their data has been our focus until now,” said Dr. Julien Thérier, Software Engineering Lead at Elsevier and CEO and founder of Shazino, the Lyon, France-based company that developed Hivebench. “But we knew that if we wanted to scale up our activities and create additional added value, our product would need to be integrated with a chain of tools that catered to the need of researchers to share and reuse data sets as well. Elsevier’s Research Data Management portfolio does exactly that, and a lab notebook like Hivebench is a key asset to that portfolio.
“The integration with Elsevier also enables us to make the Hivebench service available to many more researchers, making sharing and reuse possible on an unprecedented scale.”
We also needed access to previously published research, which we got through the Bernard Becker Medical Library at our institution; we could not have done this work without consulting the literature. The library’s 3,700 journal subscriptions made it possible for us to do this work at Washington University in St. Louis.
But we’re aware that not every university can support such a rich library. We therefore chose an open-access model for publishing in Neoplasia to maximize the impact of our work. Open access is a wonderful way to lower barriers to scientific communication, when coupled with robust editorial standards and peer review.
Having the article on ScienceDirect is important to us because it’s a durable and visible repository. We can count on ScienceDirect to provide high-quality content, carefully edited, with correct standards of proofreading and publishing.
Read the study
Elsevier has published this article open access:
- Li, Jing et al: “Whole-Genome DNA Methylation Profiling Identifies Epigenetic Signatures of Uterine Carcinosarcoma,” Neoplasia (February 2017)
Neoplasia publishes the results of novel investigations in all areas of oncology research. The title Neoplasia was chosen to convey the journal’s breadth, which encompasses the traditional disciplines of cancer research as well as emerging fields and interdisciplinary investigations. Neoplasia is interested in studies describing new molecular and genetic findings relating to the neoplastic phenotype and in laboratory and clinical studies demonstrating creative applications of advances in the basic sciences to risk assessment, prognostic indications, detection, diagnosis, and treatment. Read more.