A team of researchers from Columbia University has discovered two genes that appear to be responsible for one of the most aggressive forms of brain cancer, called glioblastoma multiforme.

The researchers say the genes are what make glioblastoma multiforme invade the normal brain so rapidly, producing inoperable and often fatal tumours.

Recently, glioblastoma claimed the life of U.S. Sen. Edward Kennedy, 16 months after diagnosis.

The discovered genes - C/EPB and Stat3 - appear to be active in about 60 per cent of glioblastoma patients, report the researchers in the journal Nature.

The study found that the genes do not fuel tumours unless they are both "turned on." When both are activated, they drive hundreds of other genes to transform brain cells into highly aggressive, migratory cancer cells.

The study found that patients whose tumours showed evidence of both genes being active died within 140 weeks of diagnosis. In contrast, half of those patients without activity from these genes were alive after that time.

The researchers also looked at human glioblastoma tumour cells injected into the brains of mice. When the researchers silenced both genes in the human glioblastoma cells, it completely blocked their ability to form tumours in the mice.

Lead researcher Dr. Antonio Iavarone, an associate professor of neurology in the Herbert Irving Comprehensive Cancer Center, described the two genes as the disease's "ruling officers."

"When the genes are turned off, the tumour cells are unable to form tumours in the brains of mice," he explained to Canada AM.

"And in the few cases when they do, these tumours are much less malignant than the original tumour that came from the patient," he said, explaining that the tumours are unable to form the new blood vessels they need to feed themselves.

To conduct the research, scientists first had to map out the comprehensive and highly complex network of molecular interactions driving the behaviour of glioblastoma cells.

"We created a fairly complex map of all the molecular interactions within the cancer cell," explained co-author Dr. Andrea Califano, director of the Columbia Initiative in Systems Biology.

"Once we had the logic of the cancer cell, we could ask questions, like for instance: if this is the end result of the tumour and this is what we observe, then who is regulating that endpoint?"

When they identified the genes responsible, they were surprised, since C/EPB and Stat3 had never been implicated before in brain cancer.

"These two genes were very surprising because they came out of the blue, if you want, and they don't work in isolation; neither one by itself is able to do the job," said Califano.

Having now identified the "ruling officer" genes leading the tumours' growth, researchers will no longer focus on developing drugs against "minor actor" genes fuelling the cancer.

"We hope that by attacking these two molecules in the future, we can develop targeted therapies that might be used for this terrible form of cancer," said Iavarone.

Califano noted that treatments will need to focus on shutting down both genes simultaneously.

And in fact, his team has just received funding to research new therapeutic compounds that could potentially attack each one of the genes at the same time.