Biomarkers bring improvement to medicine and will
increase overall survival
October 29, 2008
George S. Mack
BioDecade
When molecular biologists pursue drug discovery, the process normally
involves finding targets linked to a disease. Sometimes disease-associated
molecules are just that—associated with or the result of the disease but not the
proximate cause. Of course, a valid biomarker could be in the disease pathway
as a contributor to the problem, but it could turn out to be so far upstream that it
would be totally impractical as a drugable target. Breaking a pathway circuit far
in advance of the disease phenotype could cascade into sweeping off-target
and undesirable side effects; however, instead of treating these remote
markers as dead ends, a researcher might just see these discoveries as low-
hanging fruit that could be developed as diagnostic, prognostic or treatment-
related progress indicators.
From the investor’s point of view, the advantage of biomarker development is
that it costs significantly less than drug development. A startup biotech could
develop a marketable biomarker in two or three years, instead of the 12 to 15
years and a billion dollars it might take to get an investigational drug through
clinical development. That’s achievable because a minimally or mildly invasive
diagnostic test presents little or no danger to the patient per se. The test must
only prove-out that it can give the clinician more helpful information about a sick
patient, such as saving time in the diagnosis or predicting which drugs might
be superior treatments or telling the physician when a disease has begun to
advance and become more aggressive. Moreover, a diagnostic test might be
approved by the health department of a single state thus allowing clinicians and
pathologists to send blood, urine or biopsy samples immediately to the
reference lab for assay. Clearly this does not present nearly the burden of drug
development.
Biomarker detection is ushering in the new field of personalized medicine.
Nowhere is this better illustrated than with the monoclonal antibody Herceptin
(Genentech/Roche/Chugai’s trastuzumab) which is very effective but only in
breast cancer patients whose tumors overexpress HER2/neu, an epidermal
growth factor receptor (EGFR) tyrosine-kinase that only occurs in 25%-30% of
patients. Today patients are routinely assayed for the HER2 biomarker/target,
but during development of Herceptin, the project was nearly scrapped when it
was being tested in a large population of random patients who were not being
screened for the marker. The drug just did not appear to be statistically
significantly effective in that larger group. However, when the decision was
made to screen patients, the HER2 cohort turned out to be the sweet spot for
the drug.
Because of Genentech’s experience, drug developers are working hard to find
specific populations of patients who will be responsive to a drug. For instance,
Durham, NC-based Adherex Technologies is conducting a Phase II trial with its
cyclic pentapeptide ADH-1, for metastatic melanoma. The drug is designed to
attack more malignant cancers that are in the process of advancing and
metastasizing, which can only occur when tumor cells break away from each
other and assume a migratory phenotype. In its study, Adherex is including only
those patients that express higher levels of the molecule N-cadherin that is
known to be present in many advancing and malignant tumors. In melanoma
and hepatic cell cancers N-cadherin is expressed with great frequency, but in
other tumors such as breast cancers it is expressed only during later advanced
stages of the disease. Following the lead of Herceptin, management at
Adherex believes it has found a regulatory pathway to approval by specifying
which patients are candidates for therapy. If and when ADH-1 is approved for its
lead indication, the company will seek to expand the product’s label through
further clinical trials in other cancers. This is a tried and proven method of
regulatory management.
Sometimes a specific disease phenotype is difficult to discern or may not be
definitively diagnosed until it’s too late. New technologies could be the answer.
Israel-based Rosetta Genomics announced on July 22 that its first cancer
diagnostic test based on the company’s microRNA technology platform was
approved for use by surgical oncologists and pathology labs in the State of New
York. The test was developed, tested and authenticated by Columbia University
Medical center in New York, and it can differentiate squamous cell versus non-
squamous non-small cell lung cancer (NSCLC) with high degrees of sensitivity
and specificity. This is significant because the monoclonal antibody Avastin
(Genentech/Roche’s bevacizumab) has been shown to be less useful and less
effective in squamous-cell NSCLC. In addition, the FDA is requiring a black-box
warning on the Avastin label to alert clinicians that there are significantly higher
rates of severe or fatal hemorrhage in that same group of patients. There is
high probability that the differences between squamous and non-squamous
NSCLCs will lead to additional discoveries relative to drug sensitivity and
potential risky side effects.
Rosetta is also developing ways to diagnose disease from microRNA levels in
blood and other fluids including urine. Others are also working on the problem.
In a study performed by investigators at the Fred Hutchinson Cancer Center in
Seattle and published in the Proceedings of the National Academy of Sciences
at the end of July, blood from mice and humans with metastatic prostate cancer
was tested and compared to controls. Researchers were able to specify which
mice and patients had certain cancers. The study of microRNA biomarkers
could lead to highly useful and lifesaving diagnostics for very early cancer
detection—long before clinical manifestation.
Biomarkers and the algorithms that give them meaning will also be used to
monitor disease progression and efficacy of treatment. Predictive Biosciences
of Lexington, Mass. is creating original diagnostic technology around the field of
cancer. The idea is to inform physicians and patients about the absence,
presence or progression of disease so that therapies and other diagnostics
can be used more effectively and economically. The company will focus its first
efforts on bladder, breast and colorectal cancers with the goal of spotting
biomarkers and using the information much the way the prostate specific
antigen test is used to monitor disease progression in prostate cancer patients.
Assay of urinary Matrix Metalloproteinases (MMPs) and A Disintegrin And
Metalloproteinases (ADAMs) will be the first biomarkers developed for use in
bladder cancer survivors. MMPs and ADAMs are key regulators of tumor growth,
progression and metastasis because they mediate cancer advancement
through breakdown of the extracellular matrix allowing tumor cells to detach and
migrate away.
Biomarkers will also be useful in determining if drugs are safe for particular
patients with genomic variations. In July the FDA recommended to infectious
disease specialists that patients who might be candidates for the AIDS drug
Ziagen (GlaxoSmithKline’s abacavir) should be tested for the human leukocyte
allele HLA-B*5701 prior to therapy to guard against a sometimes fatal
hypersensitivity response that includes fever and rash that affects multiple
organs. FDA has also ordered GSK to modify the drug’s label to warn
physicians and patients that the genetic test should be performed and that if the
allele is present, avoiding the drug would “significantly decrease” the risk of the
dangerous reaction.
More sophisticated biomarkers will come along and increase survival of
patients with cancers with each advancement producing incremental
improvement in overall treatment success. The day will come when drugs, dose
and dose intensity will be selected according to not only the disease, but also
the patient’s own genotype and phenotype. Although that time is somewhere in
the distant future, researchers continue to whittle away around the edges as
they gradually improve the art and science of personalized medicine.
| © 2008 BioDecade and George S. Mack. All Rights Reserved |
 |
| |
 |
| |
 |
| |
 |
| |
 |
| |
 |
The use of biomarkers
|