The Cancer Genome Atlas: Generating a "Parts List" for Cancer
April 2, 2015, by Jean-Claude Zenklusen, Ph.D.
When I was working in a genomics laboratory in the 1990s, I sequenced a section of human chromosome 7. The part I focused on coded for a tumor-suppressor gene. At that time, we had to read the DNA sequence letter by letter in short stretches and then figure out where the genes resided. It was tedious and required extreme care and attention to detail, plus a fair amount of educated guesses.
Today, scientists no longer need to read DNA letter by letter to find the sequence of a gene. Thanks to the Human Genome Project, researchers can quickly and easily access genomic sequence through many data portals, including the well-known UCSC Genome Browser.
The Cancer Genome Atlas (TCGA)—featured in the last installment of the PBS documentary Cancer: The Emperor of All Maladies—is on the same scale as the Human Genome Project, with a focus on tumor genome characterization. For each of the 34 selected cancer types, TCGA generates a “parts list,” or catalog, of its genomic changes and provides a detailed analysis of mutations that contribute to the beginnings and growth of cancer. These data are freely available to researchers on the TCGA Data Portal and CGHub.
By analyzing this list, TCGA researchers have found exciting new connections across cancer types. For example, TCGA analyzed samples from more than 800 breast cancer patients. TCGA’s 2012 publication in Nature showed that one type of breast cancer exhibited greater genomic similarity to a subtype of ovarian cancer than to the other types of breast cancer. This research suggests that these two cancer subtypes may share a common starting point and, even better, may be susceptible to the same targeted therapies. This finding is of particular importance since, as the recently released Annual Report to the Nation shows, breast cancer is the most commonly diagnosed cancer in women.
Inspired by this finding, TCGA researchers used the list find out if these kinds of similarities exist in 12 other cancers TCGA had characterized. They identified more mutations common across cancers. For example, they found that lung, head and neck, and bladder cancer have similar mutation profiles.
TCGA and its parent center, the Center for Cancer Genomics, have launched two even more ambitious projects in 2014: the PanCanAtlas and Pan-Cancer Analysis of Whole Genomes (PCAWGs).
The PanCanAtlas will use all of the data generated by TCGA to analyze different cancers, allowing researchers to discover even more complex relationships between cancer types. PCAWGs is a collaboration with the International Cancer Genome Consortium to analyze whole-genome data from more than 2,000 patients. That sequence will be integrated with clinical and other molecular data to yield a more complete portrait of cancer. We hope findings from these projects will be published in 2016.
Because nearly 4,000 journal articles have cited TCGA data, we know that TCGA has become a key resource for researchers. Scientists can analyze entire genomes without needing to generate even a single nucleotide like we had to some years back. Our hope is that the availability of these data will speed the research process, enabling TCGA to achieve its overarching goal: improving our ability to prevent, diagnose, and treat cancer.
Jean-Claude Zenklusen, Ph.D.
Director, The Cancer Genome Atlas Program Office
This is the last in a series of posts authored by NCI cancer research leaders, inspired by topics covered in the documentary film, Cancer: The Emperor of All Maladies, which aired this week on PBS nationwide.
Other Posts in This Series
Looking to the Future of Childhood Cancer Research, by Malcolm Smith, M.D., Ph.D
The Importance of Cancer Prevention Research and Its Challenges, by Barry Kramer, M.D., M.P.H.