The Cancer Genome Atlas (TCGA), a collaborative project between NCI and the National Human Genome Research Institute (NHGRI), is at the mid-point of its 3-year pilot phase. The TCGA pilot is a truly integrative, multidisciplinary effort to develop and assess a framework for systematically identifying and characterizing the genomic changes associated with three cancer types: brain cancer (glioblastoma multiforme, or GBM), lung cancer (squamous cell carcinoma of the lung), and ovarian cancer (serous cystadenocarcinoma of the ovary).

Already we are beginning to see the value of this project. Not only are new data being developed and shared with researchers around the world, but new technologies and tools are being developed that are allowing researchers to delve further into the molecular machinery of cancer with greater precision and efficiency.

For example, multiple technology platforms are being used at TCGA for molecular characterization and sequencing to interrogate tumor samples and their corresponding normal samples, ensuring that the data for each case are incredibly rich and as complete as possible. As the centers gain experience with some of the newer platforms, they learn how best to use them, both in terms of producing quality data as well as cost efficiency.

The different centers at the core of TCGA - which focus on characterization, sequencing, data collection and informatics, and technology development - are working in close concert integrating different data types, such as gene-expression signatures, microRNA-expression signatures, and methylation patterns, looking for new and more powerful ways to stratify tumors by subtypes, each of which may have different prognoses and require different treatment approaches. These data are then used to identify specific genes to sequence. The project team has designed TCGA to provide biomolecules (RNA and DNA) from a central facility to all TCGA centers to ensure uniform quality of the material.

To date, efforts with GBM are the furthest along, with more than 234 tumor cases (out of a planned 500) having undergone comprehensive characterization and a subset sequencing of 1,300 genes. This work is providing valuable insights into the biology of GBM and identifying potential new treatment avenues.

One example is an expanded understanding of the key signaling pathways involved in GBM. The integration of characterization data from the different TCGA labs has shown that not only is p53 - an important tumor suppressor gene - mutated more often in GBM than previously believed, but also that other key genes in the p53 pathway are often mutated, such that as many as 80 percent of GBM cases appear to have potentially important mutations in this pathway.

These findings suggest that other genes "downstream" of p53 could be valuable therapeutic targets. And this is just the first signaling pathway in GBM to be better elucidated.

It's important to stress that all TCGA centers have to immediately submit their raw data to the Data Coordinating Center, a database that is part of the cancer Biomedical Informatics Grid. The data are available to the scientific community, and all interested parties must sign agreements meant to strictly protect patient confidentiality.

The public spotlight is once again heavily focused on cancer, particularly brain cancer, following the malignant glioma diagnosis given to Senator Edward Kennedy, a long-time advocate for cancer research. Unfortunately, little progress has been made over the past several decades in the early detection and treatment of GBM. Our hope is that TCGA will help change that, allowing researchers to identify new methods for early detection, diagnosis, and treatment of GBM and the other cancers, and, in the process, recognize the true promise of molecular oncology.

Dr. Daniela S. Gerhard
Director, NCI Office of Cancer Genomics