Here is that article from Red Herring:
November 5, 2000
Trend number ten: Biotech
Functional genomics makes drug discovery cheaper and faster, reinvigorating all of biotech.
By Stephan Herrera
From the December 04, 2000 issue
Consequences:
Now that the human genome is mapped, biotech startups are racing to profit from it. The field of functional genomics will change the way scientists discover and develop drugs, making the process faster, cheaper, and smarter.
Winners:
Functional genomics companies that offer innovative approaches and new analytical tools to improve our understanding of the relationships between genes, proteins, diseases, and drugs: Rosetta Inpharmatics (Nasdaq: RSTA), Illumina (Nasdaq: ILMN), Athersys (proposed Nasdaq: ATHX), Millennium (Nasdaq: MLMN), Celera (NYSE: CRA), Affymetrix.
Losers:
Companies that fail to make the transition to the post-genomic era: Orchid Bio-sciences (Nasdaq: ORCH), Lyon Biosciences, Aurora Bioscience (Nasdaq: ABSC), Human Genome Sciences (Nasdaq: HGSI).
See also:
Data gold mining
Microclimate
Talking Points
Think our trends are off base?
LIVE Top ten trends Q & A on November 20 at 4pm Eastern
Go to our discussion forum to debate biotech issues
Until recently, functional genomics -- the business of using genomic data and analytical tools and algorithms to build drugs better, cheaper, and faster -- was little more than Holy Grail coffee talk among scientists. But 2001 will see the field become a major preoccupation of investors, entrepreneurs, and "big pharma." Functional genomics will move to the fore as biotech shifts into the post-genomic phase that began this summer, when scientists finished mapping the human genome. This is a change we've been eagerly awaiting (see "Beyond the Genome", April 1999). While mapping the genome was a mighty milestone, by itself the map tells us little about the relationships between genes, proteins, and disease -- and nothing about the impact of drugs on those relationships.
There have been previous proclamations about functional genomics and its key enabling technology, proteomics. The resulting technologies were supercomputers that processed bigger bundles of data more quickly; brilliant maps of gene sequences were placed on the Web for all to tap. But without a better understanding of proteomics -- the precise role of proteins in disease management -- genomics had little function. Companies like Rosetta Inpharmatics (Nasdaq: RSTA), Illumina (Nasdaq: ILMN), CuraGen (Nasdaq: CRGN), Millennium Pharmaceuticals (Nasdaq: MLMN), Celera Genomics (NYSE: CRA), Oxford GlycoSciences (London: OGS), and Athersys (proposed Nasdaq: ATHX) have spent the past two years developing the tools that will finally make genomics functional and profitable.
The $350 billion pharmaceutical industry revolves around only 400 drug targets known to wreck human health. Functional genomics should help explode that universe to 10,000 or more targets. "Functional genomics will become the singular focus of any company that has any hope of turning all that genomic data into something truly useful for human health," says Paul Gilman, director of policy and planning at Celera Genomics.
Many biotech businesses that went public in 2000 described themselves as functional genomics companies (see chart "Microclimate"). Underwriters are telling firms that are only marginally in the field to play up their functional genomics tag line. The classification pays off: Rosetta Inpharmatics and Illumina, for instance, both achieved $1 billion-plus market valuations in their first month on Nasdaq, at a time when biotech as a whole is still struggling to regain market momentum. Corporate investors have taken notice of the sector (see "Data Gold Mining"). And one of the most eagerly awaited IPOs in years is a functional genomics outfit, Athersys, which is expected to file by year's end (see "Protein Powerhouse," September).
For the past several years, the discovery of a single gene -- like the one thought to cause cystic fibrosis, or obesity, or Huntington's disease -- was cause for celebration. Then bioinformatic advances in high-throughput analysis, which helped Celera make a lasting name for itself in the genome-sequencing business, became the big deal.
But pharmacologists and pathologists reminded everybody that a single gene or protein won't change the world. As Mark Boguski wrote last year in Science, "A single gene or protein may have multiple forms and functions that are context-dependent and that may never be fully understood by sequence analysis alone." Dr. Boguski knows of what he speaks. Formerly with GenBank, the repository of gene and protein sequence data, he joined Rosetta last August because he sees functional genomics as the solution for studying all genes or proteins at once, rather than simply one at a time.
Rosetta has attracted investors like Paul Allen, plus top management personnel and scientists from fields beyond medicine, like physics and mathematics. The company also has inked collaborations with Agilent Technologies (NYSE: A), Dupont (NYSE: DD), Merck (NYSE: MRK), and Amgen (Nasdaq: AMGN). Rosetta offers its partners the chance to get in on the ground floor of a new, multidisciplinary approach to drug discovery and development. Rosetta is a cross between a software company and a pathology lab, working to enable researchers to better predict drug toxicity based on a person's genomic profile. The company has a nifty way of interpreting genomic data using pattern matching. Previously, researchers looking for genetic culprits of a disease searched door-to-door, one gene at a time. In contrast, Rosetta looks at the entire human genome at once. This fall, University of Chicago researchers used this technique to find a previously unknown gene that triples the risk of adult-onset diabetes, which affects 15 million Americans. Divining natural protein function is the province of proteomics, and Rosetta's technology delves even deeper, analyzing how protein function is altered by drugs.
Another company leading the march toward functional genomics is Illumina, which makes its arrays from fiber-optic probes. Like Rosetta, Illumina's technology draws cells from the blood samples of many patients, with a variety of diseases. The genomic data is already out there for any company to download. The trick is mining the data to get at gene and protein function.
And Athersys, a much-hyped startup, can make a cell produce, or "express," all 100,000 or so of its genes and trigger them to express their proteins. Such a broad technology will be enormously useful, enabling researchers to study genes and proteins by the millions.
Regardless of the technological promise of functional genomics, biotech's biggest hurdle remains the U.S. Food and Drug Administration approval process, which nine out of ten drugs fail. Even the one that makes it to market occasionally kills or causes harm. Until a drug has been tested in a large and diverse population of patients, its true tendency to harm or heal is hard to predict.
The biggest reasons drugs fail in animal and clinical trials are inept drug metabolism (effectiveness) and organ toxicity (side effects). The biggest reason "big pharma" has been consolidating and striking billions of dollars in biotech deals over the past decade is that the drugs in their pipelines are falling short of expectations. How many lives -- and tens of billions of dollars in health care costs -- could be saved if these riddles were solved? How much would big pharma pay to improve the odds of FDA approval by even 10 percent? And how much would investors pay to get a piece of these new companies? Functional genomics has the answers.
Merck and a few other drug companies spent years and close to $500 million developing a new immunosuppressant called FK506, only to scrap it in clinical trials because of toxicity troubles. "In three weeks' time, we could have told Merck that FK506 would fail," says Rosetta's president, Stephen Friend. Easy for him to say. Biotech has promised big things before, only to fall short: just look at gene therapy and model organisms.
Allen Oliff, who directed cancer research at Merck for 20 years prior to joining DuPont, says the functional genomics trend is different, and that's why DuPont is paying Rosetta $200 million to co-develop new drug compounds. Mr. Oliff says he's more cynical than most big pharma executives: "Not every functional genomics concept will pan out. Functional genomics won't miraculously change the 90 percent failure rate in the drug business overnight. But incrementally," he says, "companies like Rosetta will improve that rate to 70 to 80 percent. That's certainly the biggest change I've seen in my career." |
