Customized virus kills brain tumor stem cells that drive lethal cancer

Sep 12 2007

A tailored virus destroys brain tumor stem cells that resist other therapies and cause lethal re-growth of cancer after surgery, a research team led by scientists at the University of Texas M. D. Anderson Cancer Center reports in the Sept. 18 edition of the Journal of the National Cancer Institute.

"We have shown first in lab experiments and then in stem cell-derived human brain cancer in mice, that we have a tool that can target and eliminate the cells that drive brain tumors," says co-senior author Juan Fueyo, M.D., associate professor in M. D. Anderson's Department of Neuro-Oncology. A request to launch a clinical trial of the virus, called Delta-24-RGD, is expected to go to federal regulators this month.

The virus was tested against the most aggressive brain tumor - glioblastoma multiforme, which originates in the glial cells that surround and support neurons. It is highly resistant to radiation and chemotherapy and so invasive that surgery almost never eliminates it. Patients suffering from this malignant glioma live on average for about 14 months with treatment.

Fueyo and colleagues developed Delta-24-RGD to prey on a molecular weakness in tumors and altered the virus so it could not replicate in normal tissue. They showed in a JNCI paper in 2003 that the virus eliminated brain tumors in 60 percent of mice who received injections directly into their tumors. The virus spreads in a wave through the tumors until there are no cancer cells left, then it dies.

Since 2004 scientists have found that brain tumors are driven by haywire stem cells that replicate themselves, differentiate into other types of cells, and bear protein markers like normal stem cells.

"Research has shown that these cancer stem cells are the origin of the tumor, that they resist the chemotherapy and radiation that we give to our patients, and that they drive the renewed growth of the tumor after surgery," Fueyo said. "So we decided to test Delta-24-RGD against glioma stem cells and tumors grown from them."

The research team led by Fueyo, co-senior author Frederick Lang, M.D., professor in M. D. Anderson's Department of Neurosurgery, and first author Hong Jiang, Ph.D., instructor in neuro-oncology, derived four brain tumor stem cell lines from four specimens of glioblastoma multiforme. All four lines exhibited the characteristics and protein signatures of stem cells. Delta-24 succeeded in killing all four types in the lab.

Next, the researchers grafted the stem cell lines into the brains of mice and treated the resultant tumors with injections of Delta-24-RGD. Untreated mice had a mean survival time of 38.5 days, while treated mice had a mean survival of 66 days. Two of the eight treated mice survived for 92 days, until the end of the experiment, with no neurological symptoms.

"It's important in animal models to see improvement in survival in the majority of animals, but to have some be cured and survive a long time without neurological symptoms is very rare," Fueyo said. "We have to be cautious, because an animal model doesn't fully represent humans, but the tumors grown by these stem cells closely resemble the tumors we see in our patients, which is an exciting finding in itself."

Tumors in other mouse models tend to be round and self-contained, explains co-senior author and Frederick Lang, M.D., professor in M. D. Anderson's Department of Neurosurgery. Malignant tumors in patients are never round, they invade other tissues and delve deeply into the brain. The cancer stem cell-derived tumors in these experiments have the irregular shape and invasive characteristics of their human counterparts.

"That similarity to the human tumor is encouraging," Lang said. "And it's also encouraging that we got basically the same results with Delta-24-RGD in this experiment that we got in our earlier experiment using other tumor models."

A clinical quality version of Delta-24-RGD has been manufactured by the National Cancer Institute and an independent consultant has completed a toxicology assessment. An Investigational New Drug Application to proceed with a phase I clinical trial is expected to be filed with the U.S. Food and Drug Administration in September. A clinical trial could began as early as this fall.

Delta-24-RGD exploits the fact that a protein called retinoblastoma (Rb) is either missing or defective in brain tumors. Rb normally guards against both the proliferation of cancerous cells and against viral infection. So the virus has an easier time invading tumors and replicating in its cells. Adenoviruses attacking normal cells employ their own protein, E1A, to counteract retinoblastoma's defensive measures. To keep Delta-24-RGD out of normal cells, Fueyo and colleagues deleted a small part of the gene that produces E1A.

The JNCI paper shows that Delta-24-RGD forces tumor cells to devour themselves until they die. This self-cannibalization, called autophagy, occurs when a cell forms a membrane around part of its cytoplasm or an organelle and then digests the contents, leaving a cavity. A cell that dies from autophagy is riddled with cavities.

Cells normally employ autophagy temporarily to survive when nutrients are short, to recycle components to form new organelles, or to fend off viral or bacterial infection. In cancer research, there is evidence both that autophagy is a form of programmed cell death triggered to prevent the replication of damaged cells and that cancer cells in some instances employ it to survive attack.

"Our next experiments will address whether the cell kills itself or dies defending itself against the virus," Fueyo says. Sure, the cell dies either way, but the distinction is important, Fueyo says, because the virus could be redesigned to either fuel or block autophagy to make it more effective. The autophagic protein Atg5 is heavily expressed in the dead tumor cells, making it a potential biomarker of the virus' effectiveness.

http://www.mdanderson.org/