Kristine M. Conner
University of Pennsylvania Cancer
Last Modified: May 13, 2001
During a panel presentation that ASCO members "are likely to be talking about ten years from now," to use the words of ASCO president- elect Dr. Larry Norton, four leading researchers in the molecular biology of cancer looked back, looked ahead, and also talked about some of the current exciting developments in the field. Harold Varmus, MD, of Memorial Sloan Kettering Cancer Center, David Botsein, PhD, of Stanford University, Brian Druker, MD, of Oregon Health Sciences University, and Michael Bishop, MD, of the University of California San Francisco, used the recent milestone reached by the Human Genome Project-that is, the completion of a draft sequence of the human genome-as the context for their talks, and in different ways they addressed the issue of how this will jumpstart the progress in further individualizing our approach to cancer.
In short, the ability to "map" the gene patterns associated with cancer holds forth great promise for the development of better targeted treatments. Instead of targeting the tumor as a whole, as the standard treatments of radiation therapy, chemotherapy, and surgery now do, more and more new treatments will attempt to target a step in the molecular process that actually gives rise to cancer. People who share that step, as indicated by genetic "fingerprinting" of their cancer cells, presumably would benefit from such targeted treatment.
"We've been using fairly nonspecific drug therapies for a long time now," noted Larry Norton, "but things really are going to change dramatically."
And the situation has already begun to change, noted Harold Varmus, with progress in this area occurring even before the milestone of the completion of the Human Genome Project. Just 25 years ago, he noted, researchers really had an "empty view" of cell signaling, and they were just beginning to test the hypothesis that it could be possible to implicate certain genes in the development of cancer. His own groundbreaking work on oncogenes, genes found to be associated with uncontrolled cell growth, was instrumental in prompting researchers to pursue this course. "Now there are many proteins that we can point to as important mediators in the process," Dr. Varmus added.
The new challenge brought by the Human Genome Project is to demonstrate more widely "which of the 35,000 genes are involved in cancer and how," he noted. Doing that will not only allow researchers to target therapy according to genetic differences, but also develop new strategies for early intervention and perhaps prevention.
Dr. David Botstein of Stanford described how the "DNA microarray" or "chip" technology is actually making it possible to observe and map the different patterns of gene expression that can be associated with the same type of cancer, such as breast or lung cancer. He showed the audience examples of how samples of different cancer subtypes, such as non-small cell, small cell, and squamous cell cancer, cluster together on this so-called genetic map when they are analyzed.
"We're now getting beyond cell growth and looking at cellular signaling pathways," said Dr. Botstein. "We can categorize the biological basis of tumors based on genes that are expressed within them." And this holds forth the promise of "aiming" new therapies accordingly.
One such targeted treatment that is being heralded as the shape of things to come at ASCO this year is STI-571, or Gleevec, which was recently approved by the FDA for chronic myeloid leukemia. It also has shown promise in two phase II trials for gastrointestinal stromal tumors, the results of which are being presented at this meeting. Dr. Brian Druker of the Oregon Health Sciences University described the preclinical testing process that was involved in identifying the molecular target for this treatment.
In short, 95 percent of patients with CML have been found to share an activated gene known as Bcr-abl, and this was subsequently identified as a "causative molecular abnormality" for this cancer. The agent known at STI-571 specifically inhibits the growth of cells that express this gene, said Dr. Druker, and the response rates have proven dramatic in people with CML.
Now, researchers at Oregon Health Sciences University conducting phase II trials are finding that STI-571 also works on a molecular target associated with solid gastrointestinal stromal tumors, the overexpression of a gene known as c- kit. Essentially it has been found to block this activity of this gene, which otherwise would lead to abnormal proliferation of cells. STI-571 has led to improvement or at least stabilization in a significant percentage of trial participants with these types of gastrointestinal tumors. And it may be that this treatment will be useful in other types of cancer, such as small cell lung cancer, in which overexpression of the c-kit gene is found.
"I think we will start to have huge numbers of possible molecular targets like this," said Dr. Druker. "The key is really identifying early pathogenetic events as targets for new therapies."
The accomplishments of the Human Genome Project should help, added final speaker Dr. Michael Bishop, but he tempered the optimism of the panel with a sense of caution about how much work remains to be done.
"Will the set of potential molecular targets be sufficiently limited, or will we be overwhelmed? Will we be able to develop an algorithm for preferable molecular targets? Can we achieve sufficient specificity of designer drugs?" Dr. Bishop raised these and several other questions to urge his audience to understand the amount of work that remains to be done on "targeted treatments." He also urged them to be "temperate" in making any broad claims to the public, since the translation of this approach from lab to bedside is likely to be a lengthy process.