By Jennifer Van Brunt


Banking on a diagnostic marker for Alzheimer's disease that's already widely used in drug development, Variagenics Inc. last month struck another deal in its ongoing quest to attain a prominent position in the burgeoning field of pharmacogenomics.

Through its alliance with Nova Molecular Inc., based in Montreal, Variagenics has acquired the exclusive rights to intellectual property covering the pharmacogenomic application of apolipoprotein E (ApoE) and other genetic markers for the treatment of Alzheimer's disease and other disorders of the central nervous system. Variagenics, a privately held company based in Cambridge, Mass., paid $12 million for these rights and also has the option to acquire Nova Molecular outright.

Nova Molecular has already worked with a number of large pharmaceutical companies in developing therapies for treating central nervous system disorders. Its technology is based on discoveries by Judes Poirer, an associate professor at McGill University, that genetic variants of the ApoE gene are associated with different clinical responses to drugs for Alzheimer's disease and other central nervous system disorders. Briefly, Poirer and his collaborators determined that ApoE plays an important role in the brain's response to injury. Together with its receptors (the low density lipoprotein receptors), ApoE apparently regulates the transport of cholesterol and phospholipids into the brain during repair. Individuals with certain variants of the ApoE gene (especially ApoE4) are less able to repair damage than are individuals with other variants (ApoE2 and ApoE3). Importantly, about half of Alzheimer's patients carry this allele and half don't, according to Poirer's research. Thus, drugs such as Cognex (tacrine) that work in ApoE4-negative patients will not work in ApoE4-positive patients, and vice versa. According to Variagenics, tests for ApoE are now commonly used in the development of new drugs for Alzheimer's disease and represent the most advanced clinical application of pharmacogenomics to date. "This marker has really gotten the attention of the pharmaceutical industry," said Fred Ledley, president and CEO of Variagenics. For it can also be used to predict the efficacy of experimental drugs for treating other central nervous system disorders, including dementia, depression, Parkinson's disease, stroke, multiple sclerosis and amyotrophic lateral sclerosis. As well, Ledley said that Variagenics has also found variations in a number of other genes in the pathways for central nervous system drugs, thus reinforcing its strength in this area of pharmacogenomics.

It's been common knowledge for decades that not all drugs work in all patients. As far back as the 1950s, physicians and clinicians realized that a patient's genetic makeup could affect how that individual responded to prescribed drugs. But turning that basic concept into a full-fledged field of scientific inquiry and then into a useful drug development tool has occurred only recently. With the realization that as much as 90 percent of humans' variability in drug response is due to genetic factors, every major pharmaceutical house has vowed to incorporate pharmacogenomic studies into its new drug development efforts. And many biotechnology companies are harnessing the newest gene-detection technologies to turn pharmacogenomics into a business opportunity. There are compelling reasons for this burst of activity. For one, clinical trials for experimental drugs can now be designed to include only that set of patients most likely to respond to the drug in the first place. That increases the success rate of a clinical trial and should make it easier to win marketing approval from the FDA. For another, it means that physicians can prescribe for their patients only those drugs that are most likely to be safe and effective.

It should come as no surprise, then, that there's been an explosive increase in the number of biotech collaborations and joint ventures set up to exploit the technology's considerable power. There were two ground-breaking ventures announced in 1997 that pointed the way to the future. In one, London-based SmithKline Beecham plc teamed with Incyte Pharmaceuticals Inc. (NASDAQ:INCY), of Palo Alto, Calif., to create a joint venture company in molecular diagnostics (diagnomics), diaDexus. In the other, Millennium Pharmaceuticals Inc. (NASDAQ:MLNM) launched its subsidiary, Millennium Predictive Medicine Inc. (MPMx), housed at the parent firm's facilities in Cambridge, Mass.

It took about a year, but MPMx recently signed on its first big pharma partner. In February 1999, it struck a $70 million deal with Becton Dickinson and Co., of Franklin Lakes, N.J., to develop pharmacogenomic and diagnomic tests for cancer patients. The idea is to be able to describe a patient's current medical condition and to provide prognostic and therapeutic information that will suggest the best course of treatment for that particular patient. MPMx will provide its partner with clinically validated markers for cervical, breast, ovarian, uterine, prostate and colon cancer, as well as for melanoma. In fact, the first clinical marker is melastatin, a protein encoded by a human gene that Millennium identified five years ago. It's already being tested in clinical trials for melanoma.

Several biotech companies have aligned with contract research organizations (CRO) as a means to tap into the clinical services marketplace. Most recently, Axys Pharmaceuticals Inc., of South San Francisco, joined forces with PPD Inc., a CRO based in Wilmington, N.C., to form a new company, PPGx Inc., which will be dedicated to marketing pharmacogenomics products and services. In February 1999, Axys (NASDAQ:AXPH) licensed to PPGx its pharmacogenomics intellectual property, bioinformatics software and Allele Frequency database of gene polymorphisms. PPD (NASDAQ:PPDI) will provide staff and bioinformatics services and contributed its two subsidiaries, Intek Labs Inc. and Intek Labs Ltd., to the new entity. The Intek operations specialize in genotyping, phenotyping and large-scale DNA purification and archiving.

Variagenics did it last year; in December 1998 it signed a far-reaching agreement with CRO Quintiles Transnational Corp. (NASDAQ:QTRN) to commercialize Variagenics' pharmacogenomics technologies in the clinic. Quintiles, based in Research Triangle Park, N.C., will market Variagenics' genetic variance discovery and analysis capabilities to its customer base of biotech and pharmaceutical companies. In fact, Variagenics will be a preferred provider of contract pharmacogenomic services to Quintiles' customers.

"This is a decision point for big pharma," Ledley said. "Do they build pharmacogenomics studies into their pivotal clinical trial or not?" Ledley believes that in the short term, "companies will still design their clinical trials the old-fashioned way, i.e., to get the drug approved for use in everyone." But, he added, those same companies may build in a prospectively defined subgroup (those who would benefit most from the drug, based on genetic variance) as a secondary endpoint. Being a basically conservative group of companies, not all the large pharmas could be expected to sign up for pharmacogenomics services at the outset. But, whether they have or not, "they're all interested enough to bank their [patient sample] DNA [for the future]," Ledley explained.

The business opportunity is hard to resist. For one thing, analysts agree that the CRO industry will continue to grow at an annual rate of at least 20 percent for the near future. There are already 200 to 250 new drugs that enter development each year, and with the rapid discovery technologies now available, that number should increase by about 10 percent every year, according to Ledley. Add into that the fact that big pharma spent more than $40 billion in research and development in 1997, of which it outsourced $2.8 billion to CROs, and it's easy to see why the opportunity is appealing. Analysts also calculate that pharmacogenomics may represent about 5 percent of total R&D costs. "We don't need to have uniform market acceptance," Ledley said. But he wouldn't balk at 3 to 4 percent of the total market or 5 percent of Quintiles' business. "Quintiles probably touches 30 percent of all drugs in development," Ledley said.