CHICAGO (GenomeWeb News) – As members of The Cancer Genome Atlas continue to characterize tumors, those involved in the effort are also looking beyond an already large catalog of cancer-associated genomic glitches, considering additional cancer features for profiling as well as ways of applying data that are available already.
During an educational session at the American Society of Clinical Oncology annual meeting here this past weekend, TCGA members discussed the progress made by the consortium to date, as well as future schemes for building on and translating TCGA data into clinical care.
Participating in the presentations and subsequent panel discussion were Dana-Farber Cancer Institute's Levi Garraway, Columbia University researcher Andrea Califano, Luis Diaz of Johns Hopkins University's Sidney Kimmel Comprehensive Cancer Center, and Matthew Ellis, a researcher based at Washington University's Siteman Cancer Center.
TCGA researchers have been steadily tallying up point mutations, copy number changes, and translocations across cancer types, and advances have been made on several fronts as that data becomes available, the panelists said.
In some cases, the genomic profiles have led to a better classification of tumors, in others a clearer picture has been provided of the biological processes at play in cancer or potential vulnerabilities of tricky-to-treat tumors.
For instance, Garraway provided examples of biological insights that have hinged on the availability of TCGA data and data from similar sequencing-based studies of cancer. Among them: the identification of recurrently mutated genes in pathways not previously linked to cancer and a growing appreciation of the role that epigenetics and messenger RNA splicing play in certain cancers.
Somewhat fewer TCGA-inspired advances have occurred on the precision medicine front so far, Garraway noted, in part owing to the massive amounts of data that researchers have to sort through for each patient.
There are also gaps in the information available for advanced cancers, noted Garraway, who touched on the importance of moving beyond primary tumors in the future to generate similar mutation catalogs using samples from metastatic disease.
Nearly a decade after its inception, TCGA has produced a catalog of genomic alterations across multiple cancer types that Columbia's Califano called a "large collection of broken parts in cancer."
Even so, Califano noted that there's still an incomplete understanding of how these parts work together. In addition, he said, roughly half of tumors still contain few, if any, mutations that are clinically actionable, recurrent in other patients, and conclusively linked to cancer.
For his part, Califano argued that additional systems biology-based studies are needed to bring together different layers of information on tumors and to work back to shared regulatory processes that go awry in various cancer cells.
Proving examples from colon cancer, meanwhile, Diaz discussed ways in which tumor mutation catalogs in cancer could feasibly be exploited for diagnostic purposes, resistance tracking, and treatment targeting, development, or monitoring.
Beyond identifying biomarkers and/or treatment targets in existing TCGA data, Diaz noted that there will likely be a benefit to doing more extensive whole-genome sequencing studies on tumors from various cancer types.
Despite the studies needed to further flesh out mutation patterns in cancer, though, Washington University's Ellis noted that catalogs of the most prevalent mutations in cancer are "largely done," leaving researchers with the daunting tasks of interpreting the data, extracting the relevant information, and finding associations that might impact treatment decisions.
He highlighted the importance of not only harvesting therapeutic insights from existing data, but also of working on more comprehensive studies of the "missing –ome," the cancer proteome.
Ellis, a member of one of TCGA's Proteome Characterization Centers, argued that the effort would benefit from both systems biology-focused studies and proteomics profiling. According to TCGA's web site, the project's Proteome Characterization Centers "add further depth to TCGA's integrative process."
"TCGA has analyzed gene expression, but the proteome is an additional layer of data for researchers," the site states. "These data will provide valuable information linking genotype to proteotype and ultimately to phenotype."