Nature Biotechnology 23, 535 - 537 (2005)
The Human Cancer Genome Project—one more misstep in the war on cancer
Strap yourself in and get ready for some serious 'more of the same.' A recent proposal to sequence cancer genomes holds out the promise of personalized cures for each of 50 different cancers. The cost? A mere $12 billion at today's prices(1). This human cancer genome megaproject is the equivalent of 12,500 human genome projects and already has the backing of several prominent scientists. Harold Varmus believes that the project could "completely change how we view cancer"(1); Eric Lander argues that "knowing the defects of the cancer cell points you to the Achilles' heel of tumors"(1); and Francis Collins predicts that he "can confidently tell you that something will happen here"1. More pragmatically, Craig Venter points out that "...it's not clear what answer we'd get; there might be better ways to move cancer research forward"(1).
In a nutshell, the megaproject aims to catalog all somatic mutations from primary tumors as the basis for designer drugs to cure most cancers. Success is predicated on the assumption that drugs can be targeted to very specific mutated regions of gene products. However, most patients with a localized primary tumor are cured by surgery and local radiation. It is not the primary tumor, but the metastatic spread of a small population of deadly cells that ultimately compromises a normal tissue or organ, that kills in cancer(2) (for an excellent popular account, see ref. 3). Are primary tumors therefore the appropriate focus for such a massive project and are their bulk mutational spectra therapeutically useful?
The clinical track record Cancer research has consumed hundreds of billions of dollars to date(3) and yet the main killers—breast, prostate, lung and colorectal cancer—are essentially as deadly as ever(4). Despite the glacial progress in treatment and the advent of 'molecularly targeted' therapy, cancer research continues to focus myopically on individual oncogenes, tumor suppressors and repair genes(5), with little effort devoted to alternative mechanisms and targets (6-8).
Although conventional chemotherapeutic agents remain the first-line treatment of choice, newer molecularly targeted therapies are now reaching the market. Thus far, however, these therapies have had very limited success against solid tumors, which after all make up 90% of all cancers. Success has been largely restricted to rare leukemias; for example, imatinib mesylate (Gleevec; Novartis, Basel) has initially proven effective in patients with chronic myelogenous leukemia (CML).
Whereas the initial clinical success of imatinib in CML was spectacular, this has not been repeated in most of the succeeding cancer therapies against solid tumors. Gefitinib (Iressa; AstraZeneca, London)-based treatment shrinks tumors in only about 10% of advanced non-small cell lung cancer patients(9). A recent study in one very small (35) patient group indicated that trastuzumab (Herceptin; Genentech, S. San Francisco) induces a partial response in only 23% of individuals with advanced HER-2/neu-overexpressing breast cancers10; early indications are that bevacizumab (Avastin; Genentech, S. San Francisco) is not much better in colon cancer. All of these agents have serious associated toxicities (11-13), most extend patient survival only by a matter of months and there is a variable period of remission before a resistant form of the cancer returns, even in the case of imatinib(14).
In the light of these findings, the concept of intervening in cancer networks at a single 'oncoprotein' or 'tumor suppressor protein' has thus far had decidedly mixed results when translated to the clinic. This should not be surprising as the majority of solid tumors are characterized not by single gene-based events, but by multiple genomic alterations, which are specific to each tumor in an individual.
Thus, although a mutation-cataloging research megaproject may be a diverting occupation for sequencing centers and gene hunters, leading scientists should think carefully before they tout its therapeutic promise to patients and politicians. The simple truth is that the money would be much better spent if research priorities were reevaluated. A good place to start would be to dismiss the fallacious notion that single mutations in primary tumors are the optimal starting point for research that will lead to the discovery of new, more effective cancer drugs. The clinical reality is that it is not single genes, but rather the properties of aneuploid-based methylated networks that allow metastatic cancer cells to explore novel niches in different genetic backgrounds and to rapidly become resistant to drug-based therapies.
George Miklos is the director of MIKLOS-BOND Biomedical Information Services (Sydney, Australia). Dr. Miklos is a widely recognized expert in genomics, and was a pride of place author on the landmark publication describing the mapping of the human genome (2001, Science, 291,1304-1351).