Contact

Duke Health News

919-660-1306

DURHAM, N.C. -- In a daring yet successful experiment to
cure deadly brain tumors, researchers have combined the
cancer-killing properties of poliovirus together with a
harmless genetic coding element from the common cold.

The resulting modified virus created a remarkably strong
anti-cancer agent that rapidly killed cancer cells in
laboratory cell cultures and in animals -- and without causing
polio, said Matthias
Gromeier, M.D., assistant professor of molecular genetics
and microbiology at the Duke
Comprehensive Cancer Center. Testing of the new viral agent
in humans should begin within two years, he said.

In the study, the modified poliovirus rapidly killed cancer
cells derived from primary brain tumors as well as cells
derived from breast and colon cancer metastases -- all within a
matter of four to six hours. In fact, polio is known to be one
of the quickest killers of infected host cells, producing
approximately a thousand additional infectious viral units per
infected cell, he said.

"We made a drug out of a virus by engineering its
destructive abilities from a foe into a friend," said Gromeier.
His most recent results -- a collaborative effort with Darrell
Bigner, M.D., Henry Friedman, M.D., Allan Friedman, M.D., and
John Sampson, M.D., of the Brain Tumor Center at Duke -- will
be published in the Dec. 9, 2003, issue of the Proceedings of the National Academy of
Sciences, which is currently available online.

The key to Gromeier's success has been disabling the
poliovirus' ability to kill normal brain cells while retaining
its ability to kill cancer cells in the brain. To do so,
Gromeier's team swapped a critical genetic element from the
common cold "rhinovirus" with the corresponding genetic element
from the poliovirus. The genetic element, called an "IRES"
(internal ribosomal entry site), enables a virus to express its
own genetic information inside the host cell it has invaded,
said Gromeier.

Gromeier selected the IRES from a rhinovirus because it does
not typically infect the human brain. Normal brain cells lack
the appropriate environment required for the rhinovirus IRES to
begin translating the poliovirus's genetic information, his
study demonstrated.

Cancer cells, however, regulate gene expression very
differently than normal cells do. They grow faster, lack growth
inhibitors and generally provide a supportive environment that
is highly susceptible to viruses of all sorts, making viruses
an excellent invader to disrupt cancer's growth.

"In cancer cells, the IRES from rhinovirus acts as the
trigger that activates gene expression, but the genes being
expressed -- the silver bullets in the gun, so to speak -- are
all from the poliovirus," said Gromeier. "The polio proteins
kill the cancer cells quickly and efficiently."

In fact, polio is the perfect virus to attack brain cancer
cells because it has a natural affinity for invading the brain,
said Gromeier. Polio infects brain cells by binding to a
receptor or "docking site" called CD155 on the outside of motor
neurons. Gromeier showed that brain tumors over-produce this
CD155 receptor, making the cancer cells particularly vulnerable
to infection with poliovirus. The modified poliovirus still
enters normal motor neurons because it shares the same CD155
receptor as brain tumor cells, but it can no longer grow in
normal cells.

"We have a virus that naturally targets brain cells, but we
have replaced the genetic coding element that makes the virus
so dangerous," said Gromeier. "The virus has lost its ability
to grow in normal neurons."

Tests in mice and in non-human primates have confirmed that
the modified poliovirus does, indeed, kill brain tumor cells
but does not affect normal motor neurons. Moreover, viruses
don't carry the toxic side effects of chemotherapy and
radiation, and viruses can be introduced directly into the
tumor.

"The brain is a very common site of cancer metastasis, but
cancer in the brain is extremely difficult to treat," said
Gromeier. "Cancer cells are often interspersed throughout
normal brain tissue, and most chemotherapy does not cross the
blood-brain barrier, so getting the drug to the target site of
treatment is a huge problem clinically."

To combat that problem, the modified poliovirus is directly
injected into the brain tumor. Once there, the virus seeks out
and destroys cancerous cells without detection by the body's
immune system. The immune system would normally neutralize the
poliovirus in vaccinated individuals because they have built up
antibodies against polio. But the brain does not have immunity
against polio because antibodies do not cross the blood-brain
barrier. Hence, infusing the brain directly with modified polio
is the most effective method of killing cancer cells.

But giving the modified poliovirus to humans -- even to
terminal brain cancer patients -- requires rigorous testing to
understand the mechanism behind its virulence in cancer cells
and its impotence in normal brain cells.

So, Gromeier's team embarked on a mission to elucidate the
molecular mechanism that causes the rhinovirus IRES to function
in cancer cells but to malfunction in normal neurons.

After extensive testing, he discovered that the IRES from
the rhinovirus communicates with the opposite end of the
poliovirus genome. That distant region, called the 3-prime
non-translated region, drives how the virus transmits its
genetic instructions inside the host cell. Gromeier's data
suggest that -- in cancer cells -- the rhinovirus IRES and the
3-prime communicate via a set of proteins, called co-factors,
which ignite the IRES to begin functioning.

Normal motor neurons, however, may not provide the
appropriate set of co-factors to stimulate rhinovirus IRES
function, said Gromeier. Hence, the modified virus cannot grow
in normal motor neurons.

"Cells differ in terms of how well-suited they are to a
particular virus," said Gromeier. "Every cell type has unique
cellular proteins that can either support or block viral
function, and we believe differences in these proteins account
for the modified virus' inability to infect normal brain
cells."

The research was funded by the National Institutes of
Health, the National Cancer Institute, The Burroughs Wellcome
Fund, ABC2 Foundation, and the Brain Tumor Society.