Drugs May Turn Cancer Into Manageable Disease
June 6, 2004
By ANDREW POLLACK
NEW ORLEANS, June 5 - Brett Smith, the father of two young
children, was only 26 three years ago when he was found to
have advanced melanoma, a deadly skin cancer. Several drugs
failed to stop the cancer, while leaving him frail,
depleted and ill.
But two years ago, Mr. Smith began taking an experimental
pill along with chemotherapy, and his tumors disappeared.
He dropped the chemotherapy nearly a year ago but still
takes the pill twice a day. And his disease, though it may
return one day, is still at bay.
"No one could take chemotherapy for two years straight; it
would just wear you out," said Mr. Smith, a designer for an
auto parts company who lives in Cherry Hill, N.J. "But this
is simple." The main side effect, he said, is a hoarse
The pill Mr. Smith takes, known by the awkward code name
BAY 43-9006, could reach the market in one to three years.
It is one of a new generation of "targeted" therapies that
are transforming cancer treatment by attacking the
underlying molecular mechanisms of the disease.
Some experts see Mr. Smith's experience as a harbinger of a
future in which cancer, while not cured, will be held in
check for years by drugs tolerable enough to take on a
"Cancer will become a chronic disease that we will manage
much the same way we manage high blood pressure or
diabetes," said Dr. Andrew C. von Eschenbach, the director
of the National Cancer Institute.
Dr. Harold Varmus, president of the Memorial
Sloan-Kettering Cancer Center in New York, warns against
setting strict timetables that would create "false
expectations." But he agreed that it was now a "reasonable
goal to dramatically reduce death from cancer, making it a
Targeted therapy is the major theme at the American Society
of Clinical Oncology meeting now taking place here. There
is a palpable sense of excitement that progress, although
limited, is being made against even normally stubborn
cancers like those of the lungs and kidneys.
There have already been some heralded successes with these
newer cancer drugs. Gleevec, a Novartis drug, has had
striking results in chronic myelogenous leukemia and a rare
gastric cancer. Avastin, a recently approved drug from
Genentech that blocks the flow of blood to tumors, extended
lives of colon cancer patients by about five months in a
The enthusiasm has also spread to the pharmaceutical
industry, where cancer, once neglected, has become by some
measures the most popular disease target. Biotechnology
companies in particular view cancer, a disease caused by
genes gone awry, as a good match for their techniques of
genetic engineering and molecular biology.
Surveys over the last couple of years by the Pharmaceutical
Research and Manufacturers of America found 395 drugs in
clinical trials for cancer, compared with 122 for heart
disease and stroke combined and 176 for neurological
disorders. Another survey of only biotechnology-related
drugs in clinical trials found nearly half were for cancer,
far more than for any other disease.
Indeed, so many cancer drugs are in development that
oncologists say it is becoming difficult to run all the
clinical trials needed to test them and to determine the
best sequences and combinations in which to use them. One
session here is titled "Therapy for Metastatic Colorectal
Cancer: What Do We Do with So Many Options?"
And when more of these drugs come to market, paying for
them could strain the health care system, especially since
some of the newer ones cost tens of thousands of dollars
for a course of treatment. Bain & Company, the management
consulting firm, has estimated that paying for all the
cancer drugs likely to come to market would require $60
billion a year, up from $10 billion now.
"The whole U.S. drug market is $150 billion," said Elgar
Peerschke, head of the North American health care practice
at Bain. "Who's going to pay for that? It's just going to
Still, there is a long way to go to make cancer a chronic
disease. One of the biggest pieces of news at the
conference here is that a targeted drug, Tarceva, extended
lives of patients with advanced lung cancer. But when the
results are announced, the extension is expected to be only
a couple of months.
"If I saw a couple more drugs with the impact Gleevec has
had I'd be much more excited," said Dr. Otis Brawley, an
oncologist and professor at Emory University in Atlanta. "I
still think the targeted therapies are the right direction
to go, but I think it's important we don't promise the
American people something we don't have."
Virtually everyone agrees, too, that any progress will come
not only from drugs, but from better means of detecting
cancer early, when it is most easily treated, and by
preventive steps like getting people to quit smoking.
All cancer therapies are targeted in some sense. Since
cancer is characterized by runaway growth of cells, many
conventional chemotherapy drugs disrupt cell growth. They
also, therefore, attack other fast-growing cells like those
in bone marrow and the gastrointestinal tract, leading to
such side effects as anemia, infections, nausea and
While chemotherapy has prolonged lives, it is running up
against limits. Lung, colon, breast and prostate cancers,
the four main killers, essentially "remain incurable by
standard chemotherapy" once they have spread to other parts
of the body, said Dr. Lawrence N. Shulman, chief medical
officer of the Dana-Farber Cancer Institute in Boston.
Targeted therapies attack molecular mechanisms that spur
tumor growth, or even cause the cancer. Cancer arises after
a series of genetic mutations remove the normal checks on
Ideally, targeted therapies would be tailored to the
genetic mechanisms responsible for a particular patient's
tumor. In the future, scientists say, the genes in tumors
will be routinely checked when cancer is diagnosed. And
cancers will be classified mainly by their genetic
characteristics, not by where in the body they arise or how
they look under a microscope.
But knowledge of how to do that is still lacking in most
cases. So the drugs are being tried on many patients, but
work for only a minority.
Moreover, the targeted therapies are not completely free of
side effects. And in some cases, the targeted therapy works
best when used with conventional chemotherapy. So toxic
chemotherapy is not likely to be eliminated so quickly. The
term targeted therapy implies that scientists are
rationally targeting the Achilles' heel of cancer. But in
truth, in most cases scientists do not really know what the
Achilles' heel is, or understand very well the targets they
are shooting at.
Tumors grow by multiple mechanisms. So stopping just one of
them might not be enough, just as blocking a single road
probably would not stop cars from getting to their
destinations. Cancer cells also mutate rapidly, so tumors
can evolve resistance and neighboring cells in a tumor
might be different and not susceptible to the same drug.
"The idea of reforming your tumor cell so it will behave
properly, so it's a behaved little guy that you can leave
in your body, you just don't want to do that," said Dr.
Garth Anderson, a cancer genetics expert at the Roswell
Park Cancer Institute in Buffalo. "Over all, if you're
trying to cure people with cancer, the way we're curing
them now and the way we'll be curing them 10 years from now
and the way we'll be curing them 20 years from now is to
physically separate them from the tumor by cutting it out
with a knife."
Scientists say the mutation problems and complexity might
be overcome if targeted drugs are used early, when the
tumor cells have not mutated as much. Drugs aiming at
different targets might be used together. And the latest
trend is to develop drugs that hit more than one target.
The lack of understanding is shown in the case of drugs
that block the epidermal growth factor receptor. These
include AstraZeneca's Iressa, approved last year for lung
cancer, and ImClone Systems' Erbitux, approved in February
for colorectal cancer.
The receptor sits on the surface of a cell. When the right
growth factor binds to it, it is like a key going into an
ignition. It sets off a chain of events inside the cell
that leads to cell growth and division.
The receptor is far more abundant on some cancer cells than
on healthy cells, so it was thought that blocking the
ignition would stem the runaway growth of tumor cells. But
it has since been found that there is no clear correlation
between the number of receptors on a patient's tumor cells
and response to the therapy, suggesting that the premise
behind drugs' development was flawed.
"There's a difference between targeting something that's
present in the tumor and targeting something that causes
the tumor," said Dr. Brian Druker, a professor at Oregon
Health and Science University and an important developer of
Only in April did scientists suggest a possible
explanation: the relatively few patients who respond very
well to Iressa have a genetic mutation that makes their
tumors especially dependent on that receptor for growth.
The development of BAY 43-9006, the drug Mr. Smith takes,
has also been marked by imperfect knowledge and
serendipity. The drug is being developed by Bayer, the
German company, and by Onyx Pharmaceuticals, a
biotechnology company in Richmond, Calif.
The drug is the first to block a protein called RAF, one of
a family of enzymes called kinases that relay signals
When certain receptors on the cell surface are activated,
the chain reaction of signals that leads to cell growth,
with one protein switching on another, is likely to pass
through RAF. "One way or the other most cancers turn this
pathway on," said Dr. Leonard E. Post, senior vice
president for research and development at Onyx.
But at least in kidney cancer, where it has shown most
effectiveness so far, BAY 43-9006 might not be working
mainly by blocking RAF. Because of "dumb luck," in the
words of Dr. Frank McCormick, founder of Onyx and director
of the cancer center at the University of California at San
Francisco, it turns out that the drug also blocks a protein
involved in the flow of blood to the tumor.
Two years ago, in another discovery unforeseen by Bayer and
Onyx, scientists in Britain found that RAF was mutated in
about 70 percent of melanomas. That immediately raised
hopes the drug would be remarkably effective against
"We were hoping that the pill, with its relatively mild
side effects, would be an overnight sensation," said Dr.
Keith Flaherty, an instructor of medicine at the University
But BAY 43-9006 has not worked well against melanoma when
used by itself. Dr. Flaherty, however, combined it with
chemotherapy and found tumors shrank in 7 of the first 14
patients he treated, including Mr. Smith.
But does the drug work better in the patient with the RAF
mutation? That is not yet clear, suggesting full
understanding of this target is not yet in hand.
Mr. Smith is not troubled by such questions. "I left the
science to them," he said, "and I sit back and enjoy the