Posted: May 9, 2011

Dealing with Chaos

Image: Dr. George Sledge

George W. Sledge, Jr., President, American Society of Clinical Oncology (ASCO)

TCGA is an ongoing project of the National Cancer Institute and the National Human Genome Research Institute (NHGRI) whose modest goal is to create a genomic atlas of human cancer. Even five years ago this would have seemed preposterous--way too expensive, way too scientifically complicated to analyze. Now it is well within striking distance. The price of sequencing genomes has plummeted, and the hardware and software tools required to analyze large data sets are improving rapidly. TCGA sequenced its thousandth case within the last month with money received from ARRA (American Recovery and Reinvestment Act), and many more are in process. While the tumor types being analyzed are not a comprehensive list of existing cancer types, there is excellent coverage of the most common cancers. 

TCGA and its non-U.S. equivalents are poised to transform our understanding of human cancer. Bear in mind that the first human cancer genomes are quite recent. Last year the release of a single cancer sequence was sufficient to publish a paper in Nature. The next year or two will see the publication of papers containing the genomes of hundreds of patients within specific tumor types. And not far beyond that is what the molecular biologists refer to as the $1000 Genome, the point at which it becomes easy for us to start thinking about a patient walking into a doctor's office with a memory stick containing a full genome's worth of data. Analyze that, Doc!

What will we learn when the dust settles? Without discussing specific data, my read of all this is that we will all be impressed with the genetic complexity of common human cancers. We have spent the past decade or two thinking in terms of this kinase or that receptor being turned on or off in this or that cancer. Those days will be pretty much over: genomic analysis is suggesting a world in which quantitative alterations (defined in terms of mutations per megabase) may be as or more important than specific qualitative alterations (e.g., which kinase is mutated?). There will, no doubt, be some pleasant surprises out there, as a recent JAMA article (Welch, JS et al. JAMA. 2011;305(15):1577-1584) suggests: we may find individual patients with finite numbers of dominant driving mutations and easily drugable targets. But these may turn out to be the rare exceptions, particularly in cancers generated by a lifetime of carcinogenic assault (think cigarettes and hamburgers).

I will be discussing what this means for cancer doctors and their patients in my ASCO Presidential Address in Chicago. But the bottom line is that genomic chaos is common, and we must learn to deal with chaos if we are to beat this disease.

This post has been repurposed from the American Society of Clinical Oncology (ASCO) Connection, with authorization from ASCO.