WHY HAVEN'T WE CURED CANCER?
 
       Newspaper science pages practically scream with good news about
cancer; new genes are being discovered, new mechanisms uncovered every week.
So why is there no cure? It's been 20 years since Richard Nixon declared War
on Cancer, and yet cancer killed more people in 1980 than in 1960. Does this
mean cancer research money is wasted? Why haven't we cured it yet?
 
THE CELLULAR MERRY-GO-ROUND
       The short answer is that cancer is not a single disease with a
single cure. It is a collection of hundreds or even thousands of diseases
that share a common set of cellular symptoms. There are as many kinds of
cancer as there are different kinds of cells--liver, blood, skin, brain, and
so on. Most cells go through a limited number of rounds of the cell cycle,
duplicating their DNA, resting, dividing, resting, and then leave the cell
cycle to differentiate into their mature, useful form. Cancer cells, though,
become "stuck" in the cell cycle, like being on a merry-go-round that's
going too fast for you to jump off
       Now think of the ways the merry-go-round could get stuck. The
gearshift could get bumped, and if it were rusty, or if a gear tooth were
broken, the operator wouldn't be able to slow it back down again. Or the
throttle could get stuck, or the choke. Or suppose the operator's boss liked
the calliope music and called "Turn it up!", but the operator thought he
meant the speed of the ride. The problem, in other words, could be one of
mechanics, communication, or both.
       Cancer results, then, when a genetic sabotage disrupts the normal
regulation of a cell's growth. Depending on the type of tumor, the cell may
grow rapidly, it may become "immortal" and thus out-divide its neighbors by
sheer longevity, or it may never fully mature into its final form. Whichever
way it happens, the cancer cell becomes dominant over its neighbors and
begins to overgrow them. Early on, the cancer cells stick together, usually
forming a benign tumor that often can be removed with minimal injury. If the
tumor metastasizes, however, the tumor cells break out and spread to other
parts of the body, where they can lodge in bone marrow, liver, lung, Iymph
nodes, and other tissues. Cancer therefore can kill you in many ways,
ranging from physical obstruction (of the brain or bowel), to organ
dysfunction, to general body wasting. An extremely diverse range of symptoms
results from this complex disease.
       The gears and pistons of a cell are made of protein. A major,
complicating difference between a machine and a cell is that when the cell
needs, say, to shift into second gear, it makes a clutch, uses it to change
gears, and then destroys it. The blueprints for these parts are the genes,
and reading the right genes at the right time is crucial for normal cellular
function.
       But just as an engine will not start unless you turn the ignition,
a cell won't divide on its own. It requires elaborate communication, from
outside the nucleus and from outside the cell. The outside of the cell is
covered with receptors to receive these signals; the information is passed
along to the DNA by a series of molecular levers and switches more complex
than any Rube Goldberg cartoon. The process of getting chemical messages
from outside the cell into the nucleus to alter gene expression is called
signal transduction, and is one of the most promising areas of research.
       Each step in this chain reaction has modulators and regulators,
giving the cell exquisite subtlety of response. Like a Maserati compared to
a 1966 Chevy, more parts mean higher performance, but also more things to go
wrong
 
GENETIC GAS AND BRAKES
       In the 1980s, two basic classes of cellular disasters were discovered
that can both lead to cancer, but in opposite ways. Both involve
mutations--alterations, either inherited or incurred--of genes that code for
fundamental cellular proteins.
       The first class involves genesthat promote cell growth. When mutated,
these genes are called oncogenes, or "cancer genes." Oncogenes make proteins
that put a brick on the cell's gas pedal. One example is the ras oncogene,
of which Cold Spring Harbor Laboratory scientist Mike Wigler was one of the
discoverers. The protein that the ras gene encodes, notated "Ras," is a
vital link in communicating to the nucleus that it's time to get ready to
divide. Normally, Ras is switched on and off by other proteins. But when
altered by mutation, Ras becomes stuck in the "on" position, playing,
"Divide! Divide! Divide!" like a broken record. The Ras protein appears to
be a vital step in cell division; mutated ras is found in half of all
colorectal tumors, and in 25 percent of all tumors.
       The second class, anti-oncogenes or tumor suppressor genes, normally
act like brakes to cell growth; when mutated their proteins are stuck in the
"off" position. P53 is a tumor suppressor gene. It's not known exactly what
P53 does, but when inactivated by mutation, cell proliferation accelerates.
Like a colon cancer cell. "Tumors usually carry over many traits of the
normal tissue," says Mike Wigler. "Differences in signal transduction are
almost total between tissue types." The engines are more or less the same,
in other words, for each type of merry-go-round, but the operators all speak
different languages. Each type of cancer will have some properties in common
with others, but also some that will be unique.
 
FOUR HITS
       A rule of thumb is that it takes four or five mutations, or "hits,"
to cause cancer. Often it's three or six, sometimes two or seven. The best
model to date of the way cancer develops uses colon cancer, because it
usually shows a clear, linear progression from polyps through adenomas
(benign tumors) to metastasis (malignancy). Eric Fearon and Bert Vogelstein
laid out a schematic diagram of how cancer might develop as mutations accrue
in a group of colon cells. They argue that the number of mutations is more
important than the order in which they occur, as though the mutations chip
out keystones from a many-arched system, eventually reducing the whole
structure to rubbly chaos.
       In Fearon and Vogelstein's model, the first mutation causes epithelial
cells to proliferate more than usual. In patients with Familial Adenomatous
Polyposis, a high-risk condition for colon cancer, this mutation is
inherited. A second mutation leads to early and intermediate stage adenoma.
A third causes progression to late-stage adenoma. A fourth converts the
benign tumor into a carcinoma, and other mutations may affect how the now
spreading cancer binds to and lodges in other tissue.
       "It may be that ten percent of all human genes are involved in cell
proliferation," says Mike Wigler. If even a small fraction of those can
actually contribute to cancer, take every possible combination of four
mutations and you end up with a lot of causes for cancer. Even two cases of
breast cancer, for example, are likely not to be identical.
       This is why there cannot be a single cure. Bruce Stillman is a senior
Laboratory scientist with a broad view of basic cancer research. He
explains, "There's not going to be any magic bullet for cancer. There's no
equivalent of antibiotics." Each cancer has to be understood individually,
and then placed in the context of cellular function. In a set of diseases
this complex, there are no short cuts; you have to understand the complexity
in order to draw generalizations.
 
SHOTGUN THERAPY vs THE SURGICAL STRIKE
       Current strategies for destroying a runaway cellular merry-go-round
are roughly equivalent to blasting the entire amusement park with a shotgun.
Given enough birdshot, the malfunctioning engine probably will be struck and
eventually stop, but it isn't a particularly elegant method, and a lot of
innocent bystanders are likely to get killed. Basic research on how the cell
works seeks to discover how good cells go bad, so we can begin to reject the
shotgun blast for a surgical strike.
       Cold Spring Harbor cancer researchers study a wide variety of
cellular events and molecules. Mike Wigler continues his studies of ras.
Mike Gilman researches the molecular pathway by which an oncogene called
c-fos is activated. David Beach investigates a class of proteins called
cyclins, which guide the cell through the phases of the cell cycle. Other
scientists, including Bruce Stillman, study specific aspects of the cell
cycle, such as the transcription of DNA into proteins and replication of the
DNA Itself. Each of these processes is crucial to cell division, and each
one, if activated too much, too little, or thrown out of sync, could lead to
cancer.
       Stillman says that basic research will have two main impacts on cancer
treatment. First, it will improve the diagnostics of cancer. The better we
understand how tumors are formed, the earlier and more accurately we can
detect them. A tumor is tiny throughout . most of its life first two cells,
then four, then eight. It is often only in the last few doublings that it
becomes detectable, and by then it may be metastasizing. Since cells grow
exponentially, early detection is critical. By learning what makes a cancer
cell different from a normal cell we can learn to identify an early-stage
tumor. Early detection means a better chance at successful therapy. Second,
basic research may provide the basis for better therapies. "Basic research
will give us in. sight into how to keep cancer in check, which will lead to
a better understanding of chemotherapy," Stillman says. "The more we
understand oncogenes, tumor suppressor genes, and so on, the better we will
be able to design chemotherapies to attack them." Better chemotherapies mean
greater remission rates, lower dosages, and fewer side effects all likely
outcomes of better specifying the drug's target.
       According to Wigler, though, a different tack may bring us closer to
the mark. "The best hope may be . immunological," he says. "If a tumor
carries with it certain vulnerabilities," he says, "you can shape your
treatment to those vulnerabilities. The trick here is "training" the immune
system to recognize a cancer cell as something to be destroyed. Since tumor
cells divide extremely rapidly, they have a greater chance of "evolving"
slightly different proteins just by random mutation and recombination. These
proteins need not have anything to do with cell division, but if the unique
region of the protein pops out through the cell membrane, a roving killer T
cell can recognize it and kill the cell. Antibodies to proteins unique to a
tumor would bind to the tumor cells, acting like flags for killer T cells,
which would seek and destroy them.
       Again, however, much basic research is required before this becomes
practical. An important point to note about these studies is that they are
not directed specifically at cancer. Basic research into the cell cycle,
signal transduction, and even immunology is our best hope of understanding,
treating, and ultimately curing cancer.
       After 20 years of the War on Cancer, how much closer are we to curing
this diverse collection of diseases? "About twenty years closer," quips
Wigler. Laboratory director James Watson recently estimated it would be
another 20 years before the fruits of basic research ripen into real cures.
That's realistic, says Bruce Stillman, and it really isn't that long, given
the complexity of the disease. "If you look at infectious diseases,"
Stillman says, "you see that many of them were known in the nineteenth
century, although penicillin wasn't developed until the 1940s. It was
seventy years from recognition to cure." Compared with polio or other
diseases caused by a single kind of microbe, understanding cancer is like
dissecting a computer as opposed to an abacus.
       "There's going to have to be a revolution in our understanding of
how cells work" before we can cure cancer, Stillman says. "It's the old
'forest and trees' problem. We've been studying the trees and we know where
a lot of the paths are, but we still don't have a map." Each scientist is
inching along with a magnifying glass, mapping the details of his or her own
path, but two paths could abut or even cross and the scientists wouldn't
know it. The work being done now is essential, but a synthesis will have to
take place before we truly understand its significance. "We need to identify
the molecules that connect the pathways, to tie together all the facts that
we have--that's really understanding how cells work. And that's beginning to
happen."
 
The Harbor Transcript
A Newsletter of Cold Spring Harbor Laboratory, Fall 1992