|  |
Sleuthing For Cancer Pathways & Markers
It’s ASCO time again – and there’s as much buzz as ever around the annual meeting of the American Society of Clinical Oncology. Investors, physicians, patient advocates and biotech and pharmaceutical companies alike are keen to learn the details of clinical trials on exciting new cancer therapies. All eyes will be on Genentech Inc., for one, since the company has already reported that its colorectal cancer therapy Avastin is able to delay the progression of both breast and lung cancer. And Genentech and partner OSI Pharmaceuticals Inc. will report more trial results on Tarceva in lung cancer and other tumors. Celgene Corp.’s results for thalidomide analog Revlimid will be scrutinized, as well – and the list goes on.
It’s an exciting time for cancer research: The advent of specifically targeted therapies heralds a new era in treatment, which does not rely totally on non-specific chemotherapy or radiation treatment to attack malignant cells. The ultimate goal, of course, is to have an arsenal of medicines that all attack wayward cells at the molecular level, while leaving healthy cells untouched.
We’re not there yet, but relentless research continues to yield valuable information on the underlying mechanisms of various cancers. In one approach, which is slowly gaining adherents, scientists create predictive models to design experiments that test hypotheses regarding systems -- metabolic pathways, cells or organs, for instance. Systems biology integrates disparate data sets (genomics, proteomics, metabolomics) with sophisticated computer algorithms and high-speed analytical instrumentation to test and fine-tune complex biological models.
Gene Network Sciences Inc. (GNS) is one of a growing number of firms that have developed tools to assist in systems biology-driven drug discovery. The small company, located in Ithaca, NY, uses pre-clinical and clinical data to create computer models of cells and tissues. These data-driven models and simulations are used in drug development to determine the mechanism of action of new drugs, as well as associated biomarkers of drug efficacy and toxicity.
Thus, GNS’ approach to drug discovery and development infers unknown pathway relationships from the experimental data and also creates mechanistic dynamic simulations of known pathways. It’s already attracted the interest of some heavyweight pharmaceutical companies, too: In March, Swiss giant Novartis revealed an ongoing relationship with GNS when researchers published a paper in FEBS Letters (Aksenov et al.; Vol. 579; p. 1878; 21 March 2005) describing GNS’ integrated approach and detailing a case study on a compound of clinical interest.
Also in March, GNS announced a drug development contract with Johnson & Johnson Pharmaceutical Research & Development (J&JPRD) under which the latter will use GNS’ systems approach to determine the pathways associated with mechanism of action, biomarkers and tumor specificity of an unnamed pre-clinical oncology drug.
Recent Biomarker Discoveries
|
Company
|
Partner
|
Marker (s)
|
Disease
|
Details (Date)
|
|
Genzyme
|
Massachusetts General Hospital; Dana-Farber Cancer Institute
|
Mutations in gene for epidermal growth factor receptor (EGFR)
|
Non-small cell lung cancer (NSCL)
|
Presence of markers correlates with clinical response to Tarceva and
Iressa
(5/05)
|
|
Neurogenetics
|
Massachusetts General Hospital
|
Specific changes in gene for ubiquitin-1
|
Alzheimer’s disease
|
Gene variant increases risk for late-onset Alzheimer’s
(3/05)
|
|
Predictive Diagnostics (subsidiary of Large Scale Biology)
|
Perkin-Elmer
|
Blood biomarkers
|
Alzheimer’s disease
|
ND
(4/05)
|
|
Sequenom
|
Boston University School of Medicine; University of Texas Southwestern
Medical Center
|
Point mutation in gene for complement factor H
|
Age-related macular degeneration (AMD)
|
Genetic variation is strongest known risk factor for AMD; may account
for 50% of cases
(3/05)
|
|
ViroLogic
|
AstraZeneca
|
Protein biomarkers (identified by ViroLogic) that indicate activated
cell signaling pathways
|
NSCL
|
Iressa biomarker study on tumor samples from treated patients
(3/05)
|
According to Colin Hill, GNS’ CEO, over the past five years the company has developed a computational platform that allows the incorporation of data of all sorts – molecular profiles, cell proliferation assays, tumor shrinkage, and so forth. “We take heterogeneous data and [from them] extract pathways. We feed the data into the computational engines. We probe all possibilities of network configurations that work best with the data.”
And, when it comes to modeling cancer, this platform is “agnostic” as to type (breast vs. lung, e.g.), Hill explained. “In molecular pathways, we can pull out biomarkers for efficacy and tumor specificity,” he added.
GNS’ task is aided greatly by the fact that so much genome-wide molecular data is publicly available. “The core of our method involves the extraction of pathway relationships from data,” Hill said. When it’s available, “we use existing pathway knowledge, especially in cancer where a lot is known about cell cycles and apoptosis. But most of the circuitry is unknown,” he continued. GNS’ technology brings together information on known mechanisms and discovers the missing circuitry that is “most important and specific to drug action.”
The search for relevant biomarkers has escalated in recent years, and the results of promising research efforts appear on a relatively regular basis. For instance, the table above lists five new discoveries that have been reported in just the last three months.
As well, scientists from BioCurex Inc. recently announced that they had identified and characterized a clinically significant biomarker that can detect lung and breast cancer. In fact, the marker’s been shown to successfully detect over 90 percent of all cancers tested through blood or tissue samples. In the latest findings, BioCurex’ blood test detected 92 percent of cervical cancers with a specificity of 95.7 percent.
And the small Canadian firm licensed the technology to Abbott Laboratories in late March for use in certain types of tests to diagnosis and monitor cancer. Under this deal, Abbott gets worldwide semi-exclusive rights to the technology and BioCurex gets the usual upfront fees, development milestone payments and royalties on sales. Importantly, BioCurex retained the rights to develop radioimmunoassay-based tests; the rights to detect cancer cells in tissue samples; and any applications using the marker for tumor imaging, therapy or vaccination.
This marker, dubbed RECAF, is actually the receptor for alpha-fetoprotein (AFP), which is apparently found on malignant calls in various types of cancer but is absent on most normal cells. Alpha-fetoprotein was the first oncofetal antigen to be discovered, about 30 years ago. It is normally made by liver cells in the fetus, but levels drop off drastically after birth – except in some malignant diseases such as hepatocarcinoma.
However, while the protein itself disappears after birth, its receptor apparently does not. “When cancer cells develop, they have a problem with differentiation,” explained Ricardo Moro, BioCurex’ CEO. In that way, they resemble fetal cells. “The receptor re-appears in practically every tissue where it was originally synthesized,” he added. “When we looked at different cancers, we found RECAF in every single kind we studied,” and that included about 75 percent of all known cancers.
No wonder that BioCurex believes that the marker can be used for routine screening for most common cancers – and that, eventually, it may serve as the basis of new targeted therapies for specific cancers.
originally published 05/13/2005 |