Functional Genomics Targets Cancer
Signals Magazine, Jennifer Van Brunt Editor.
signalsmag.com
Cancer researchers have made tremendous strides in understanding how this
disease wreaks its havoc at the molecular level. They've teased apart the basic
mechanisms of angiogenesis (which forms the blood vessels that supply a tumor
with oxygen and nutrients) and uncovered many of the mysteries behind
programmed cell death (or apoptosis, which tumors manage to avoid). They've
learned about growth signals, cell surface receptors and oncogenes. All these
insights have provided the keys to open a veritable treasure chest of potential new
targets for drug discovery.
Add to that the massive amount of data being cranked out by the human genome
sequencing effort, and you've got a rough idea of the untapped opportunities that
await researchers keen to find new drugs for treating cancer in its various
manifestations.
Now, several biotech firms have taken the further step of joining advances in
genomics research -- and technology -- with a heightened knowledge of the cellular
mechanisms underlying neoplasia to search for yet more potential drug discovery
targets.
Going Global
For instance, Chiron Corp.'s got a new drug discovery venture in Singapore, called
S*BIO, which not only gives the California company a conduit to transfer some of its
drug discovery technology but also a means to eventually reap the rewards of the
research collaborations through produce revenues or development rights. S*BIO
gets access to Chiron's gene expression and combinatorial chemistry technologies,
as well as its cancer-focused genomics libraries. Apparently, Chiron's cancer
genomics research has already generated validated targets for potential use in
vaccines, therapeutics and diagnostics -- but it's also produced far more targets
than the company can pursue. With the creation of S*BIO, however, there's now an
outlet. Chiron will invest about $8 million in S*BIO, giving it a 20 percent stake.
And, in its latest academic alliance, Incyte Genomics Inc. has partnered with the
U.K.'s Roy Castle International Centre for Lung Cancer Research (RCIC) to study
the role of genes in the prevention, diagnosis and treatment of lung cancer. In
particular, the partners intend to study the genetic damage that causes this sort of
cancer. It will even be possible to compare gene function in cancerous lung tissue
and tissue taken from a healthy lung in the same individual. Incyte and RCIC will
jointly own any intellectual property that results, which they will then make available
for licensing.
NCI's Treasure Trove
The NIH's National Cancer Institute (NCI) never stops its search for new cancer
drugs. Over the years, it's amassed an astounding number of compounds -- many
from natural sources -- that could prove to be useful in fighting cancer. Some of
those have turned out to be very powerful -- and lucrative -- drugs (Taxol springs to
mind). Others have demonstrated anti-cancer activity in preclinical and (in some
cases) clinical settings, but they are difficult to work with: Either they're not
bioavailable, or they're difficult to synthesize or they're not amenable to scale-up.
It's just these sorts of compounds (both natural and synthetic) that are the subject of
Exelixis Inc.'s recent CRADA (Cooperative Research and Development
Agreement) with the NCI. The plan is for Exelixis researchers to use the company's
functional genomics models to identify the targets for these compounds, as well as
the signaling pathways involved. Then, the company will develop alternate
compounds with the same target but with more optimal pharmacologic and
therapeutic properties.
"NCI did a thorough job of documenting the efficacy of these compounds [in
preclinical and clinical studies]," explained Geoffrey Duyk, Exelixis' chief scientific
officer. "This gives us a certain level of confidence in the target."
Since the NCI's database contains thousands of compounds, however, the task is
to winnow them down to a few thousand, and then to 100 or so, he continued.
Exelixis has already done that, Duyk said, through a process involving
bioinformatics and other selective criteria. And now, the first of those compounds
are arriving at the South San Francisco company's labs.
The job currently facing the researchers is to "make the experimental cut [on these
compounds]," Duyk explained. "We want to find out what happens when we put
them into our model systems." Generally, he said, Exelixis starts this process with
its simplest (and fastest) models -- Caenorhabditis elegans or yeast. This is
essentially a feasibility study, to ascertain which model system will be useful for
determining the mechanism of action for the compounds. Once researchers identify
the compounds that produce the desired result, they go after the target -- by looking
for mutant organisms that are either resistant or hypersensitive to the compounds.
This is followed by positional cloning to identify the genes that underlie the
resistance or sensitivity, biochemical assays to confirm the targets and
identification of the target in the corresponding human gene.
Exelixis retained the commercial rights to develop any of these compounds, Duyk
added. "The original compound could be the starting point for the use of chemistry
to create analogues or variants." Or, it may be possible to look for other compounds
that are structurally similar or distinct. But the real goal of the research is to find new
mechanisms for attacking cancer cells, he explained. "If we can identify these, they
could lead to entirely new classes of therapeutics."
Profiling Vaccines
A new class of cancer vaccines could also arise out of genomics-based drug
discovery. In this instance, the vaccines will be used not to prevent the disease from
occurring in the first place, but rather to prevent its recurrence following surgery
chemotherapy or radiation treatment. Theoretically, such vaccines will be able to
boost the cancer patient's immune system to the extent that it will be able to attack
any errant tumor cells that may have escaped the surgeon's knife.
In a major collaboration announced earlier this month, privately held Eos
Biotechnology Inc. will help its new partner Aventis Pasteur to identify
genomics-derived targets for the development of cancer vaccines in two specific
areas (which were not identified). Eos will identify, select and validate the targets
while Aventis Pasteur will take them into the clinic and commercial development.
Eos gets an undisclosed upfront fee, R&D funding, and milestones as well as
royalties from Aventis Pasteur, a vaccine heavyweight that produces over one billion
doses annually.
"This relationship leverages the wealth of genomics targets that we have identified
in our cancer research programs that are not amenable to antibody therapeutics,
our own area of strategic focus," explained David Martin, president and CEO of
South San Francisco-based Eos. That's because antibody therapeutic targets have
to be accessible; i.e., they must occur on the surface of the cell. But targets that
reside inside the cell could also serve as the basis for small molecule therapeutics
-- or to generate vaccines, Martin explained. Here, there's an opportunity to develop
DNA vaccines, which have a much greater ability to elicit a CTL (cytotoxic T
lymphocyte) response (to attack tumor cells) than do whole-cell or even
protein-based vaccines, he said.
Eos has developed "two custom Affymetrix GeneChips, containing DNA probes for
a total of 80 percent of human expressed genes," Martin explained. These are
based on sequences that Eos scientists picked from available from public sources,
he continued. "Affymetrix designed the chips for us." As well, Eos is working on one
last chip which will contain probes for about 15,000 genes, Martin said. The
advantage of this system is that "We know what these are already. Some are known
genes, some are known ESTs [expressed sequence tags], some are theoretical
exons." Thus, when Eos researchers interrogate various tissues or (cellular)
processes, "we're using almost a full deck" of the expressed human genome.
Eos uses this system for profiling gene transcription; it's able to identify both
abundant and rare transcripts of genes -- whether they are already known or totally
unique. In fact, the company has already profiled the transcript expression levels of
43,000 clustered genes and ESTs in normal adult and fetal human tissue.
In cancer, "We look for a gene that's expressed at high levels in tumor cells and
minimal levels in normal adult cells," Martin said. "Using bioinformatics, we can pull
out genes that fit our criteria." Eos uses tumor tissue microarrays to validate the
expression and the cell-type specificity of the candidate genes at both the mRNA
and the protein level. The tumor tissue microarrays contain archived tissue samples,
whose clinical outcome is already known. To top it off, Eos also has developed
various primary human cell culture assays, which can be used to test the ability of a
target candidate to induce a pathogenic or cancerous state in the cells.
More Lethal Weapons
PPD Discovery, a subsidiary of PPD Inc., has devised yet another scheme to
identify new targets for wiping out cancer cells -- and it's the subject of the new
alliance it recently signed with Agouron Pharmaceuticals Inc. (a Warner-Lambert
company).
In this functional genomics-based approach, PPD Discovery will employ its GSX
System to screen the genome for genetic suppressor elements (GSEs) capable of
inducing apoptosis in cancer cells. These GSEs modulate the expression levels of
target proteins, or limit their synthesis, mimicking the way in which drugs work. By
inhibiting a gene's function, it's possible to identify that gene within a given disease
pathway.
"The technology is based on a simple assumption, that any gene can be inhibited
by its own fragments," explained Tanya Holzmayer, co-inventor of the GSX System
and vice president of genomics at PPD Discovery. Thus, researchers produce
small, random fragments of DNA from the gene (or even genome) of interest and
introduce them into cells, which are then subjected to selection for the phenotype of
interest. These fragments can inhibit the expression of the corresponding full-length
genes by producing antisense RNA or an interfering RNA or peptide. Any cloned
gene fragment that specifies a functional inhibitor is a GSE.
For the cancer program, "we select for GSEs that can kill tumor cells of different
types," Holzmayer explained. "We screen for cell death." And, she added, the
company has already identified new genes (for inducing apoptosis) or genes that
are known but have never been associated with cancer cell death.
In its deal with Agouron (the second between the two companies), PPD Discovery
will search for drug targets that selectively kill cancer cells by inducing programmed
cell death. There's "some exclusivity as to the type of cancer" that's the focus of the
collaboration, Holtzmayer said, but it hasn't been disclosed. Agouron will support
the R&D and make an upfront payment to PPD Discovery; the latter also gets
milestone payments and royalties on Agouron's future sales of any products that
result. PPD Discovery retained certain rights to pharmacogenomic and diagnostic
applications, on which it will pay Agouron royalties.
The Cancer Imperative
Biotechnology companies have already spent considerable amounts of time and
effort -- not to mention cold hard cash -- in their efforts to come up with new ways to
fight one of man's most devastating diseases. In fact, nearly half (175/369) of all
biotech drugs in development at the current time are aimed at cancer, according to
the Pharmaceutical Research and Manufacturers of America's (PhRMA) 2000
survey of new biotechnology medicines in development.
There's already a few biological cancer therapies on the market -- and sales have
soared.
In its first full year on the market, Genentech Inc.'s Herceptin (a monoclonal antibody
that targets a growth factor receptor that's overexpressed on about 30 percent of
breast cancer cells) raked in $188M in product sales. The FDA approved Herceptin
for treating metastatic breast cancer in the fall of 1998. And Rituxan, Genentech and
Idec Pharmaceuticals Corp.'s monoclonal antibody that targets a cell surface
receptor on B cells, garnered $163M in sales its first full year on the market and
$279M in 1999. The FDA approved the product for treating non-Hodgkin's
lymphoma in late 1997.
It's already quite clear how powerful this new type of cancer therapy can be. The
clinical results are impressive, and skeptical physicians will be hard put to ignore
the mounting evidence that biological therapy not only works to eradicate tumors
and metastases, but also that it works at least as well as conventional treatments --
without the debilitating side effects. |
