Novel and more effective way of testing which gene mutations cause cancer

Jun 12 2005

Often cancer research goes like this: study cancerous cells in a lab dish, find mutations that appear in many of the samples, develop drugs to target proteins made by the mutated genes, and voila, new chemotherapy drugs are born. Or at least that's the hope.

The problem with this approach is that while some mutations do lead to cancer, others just tag along, frequently occurring but not driving the cancer to spread. New research under the direction of Paul Khavari, MD, PhD, professor in the Program in Epithelial Biology at the Stanford University School of Medicine and chief of the dermatology service at the Veterans Affairs Palo Alto Healthcare System, shows a novel and more effective way of testing which mutations cause cancer and which are mere research distractions. The work is published in the June issue of Nature Genetics.

"If you have multiple suspects at the scene of a crime, you don't know who committed the offense," said Khavari. Without any way of differentiating the criminal from the bystander, researchers and drug companies spend time and money investigating all suspects. But Khavari and his colleagues have developed a technique that allows scientists to distinguish more quickly between possible perpetrators: they grew human skin cells on the skin of mice where the researchers could see the effects of mutations they induced.

"The surprise in this study was that what is perhaps the most famous mutation in this cancer didn't cause melanoma," Khavari said. A mutation in a gene for the protein B-Raf shows up in the majority of all melanoma cases. "This suggests that it must be doing something," Khavari said. But when co-first authors graduate student Yakov Chudnovsky and postdoctoral scholar Amy Adams, MD, PhD, made that mutation in their melanoma model, the cells did not become cancerous.

Although these results came as a surprise, several cancer trials under way to target the B-Raf protein with chemotherapy haven't been successful in treating melanoma. If people developing chemotherapy drugs had access to Khavari's mouse model, they might have suspected that their trials would be in vain.

B-Raf turned out to be incapable of creating cancer, but another commonly mutated gene appeared to be a true criminal. Mutations in the gene that makes a protein called PI3K do cause melanoma in the mouse model.

"These studies highlight which mutations are primary drivers of cancer and allow us to focus in on that pathway," Chudnovsky said.

This work could be a breakthrough for studying one of the hardest cancers to treat. Melanoma first appears as a mole or skin discoloration. The best way to beat the disease is to remove this early cancer before it spreads. "Once it spreads, there are no therapies that are universally effective," Khavari said.

The key to this study was developing a way to induce mutations in the human skin cells and transplant them onto the mice. In the past, researchers studied melanoma in a lab dish where the cells don't behave like normal cancers. And although mice can develop melanoma, it is quite different than the human disease and is also not an ideal way of studying human cancer.

Chudnovsky said that tricking the human cells to grow on mice was no small feat. "Normally the changes leading to cancer take 60 or so years to develop. We telescoped that process into a matter of days," he said. This allowed Chudnovsky and Adams to test several different criminal suspects for their ability to cause cancer.

In future work the team hopes to test additional mutations and develop similar models for other forms of human cancers.