New medicine means research rethink

The push for personalized medicine is forcing cancer researchers to re-evaluate how they approach the disease, and shifting the focus of research to diagnostic tools, a trio of leaders in the field of genomics said yesterday (June 2) at a panel discussion on personalized medicine marking the opening of a new cancer research center at the University of California, San Francisco.

J. Craig Venter, Susan Desmond-Hellmann, and Brook Byers Photo: J. Ryan Williams

Scientists are breaking out of the mold of typing cancer by the part of the body in which it originates, said Susan Desmond-Hellmann, who was recently appointed chancellor of UCSF. While with Genentech between 1995 and last month, Desmond-Hellmann oversaw the development of the drug Herceptin, which targets breast tumors formed by a genetic mutation that upregulates the gene HER2.

Herceptin, touted as an early success in personalized medicine, is "a very precise weapon not for breast cancer, it's a precise weapon for HER2," said Desmond-Hellmann. HER2-linked tumors are not the most common form of breast cancer, but are among the most aggressive, and Herceptin has shown some success in curbing them. Indeed, she said, an uncommon type of stomach cancer also contains the HER2 mutation and seems to respond to Herceptin. A similar example of genetics trumping location is a type of brain cancer and a type of skin cancer that share the hedgehog signaling pathway, potentially allowing the two diseases to be treated by a common therapy, she said.

This gene-specific thinking was echoed by the two other panelists, genomics pioneer J. Craig Venter (and member of The Scientist's editorial board) and biotech entrepreneur Brook Byers of the venture capital firm Kleiner Perkins Caufield and Byers in Menlo Park, Calif. "It's a shift happening right now," said Byers. "You are seeing it in real time."

"Cancers do have an awful lot in common," Venter said. "Understanding the genetics and the biochemistry of a tumor might be more important in the long run" than its origin, he said.

There is as yet little consensus on which tools will best target such genes, however. Frank McCormick, panel moderator and director of the new UCSF facility, extolled the promise of siRNA, a class of RNA that interrupts gene expression and can be used to knock down tumor-related genes. Desmond-Hellmann was less enthusiastic about this approach, noting that the most lethal cancers involve multiple genetic changes, and targeting all these genes simultaneously would be difficult.

However, the panelists did agree that one of the key stepping stones to personalized medicine will be highly focused diagnostic tools that can identify the correct target in a specific patient and eliminate treatments that are unlikely to work. Even in this tough economic climate, said Byers, he is investing heavily in startups that develop such genetic diagnostics.

"A recession is a great time to start companies," he said. "There are labs available, it's easy to hire, you can get equipment at a discount, and by the time you bring your product out, the economy has recovered."


Related stories:

Race-based medicine?
[19th November 2007]

Personalized Medicine: Seven Keys to Success
[January 2006]

Medicine gets personal
[25th April 2005]