Low Fliers Behind the Drugs
Wednesday, January 2, 2002
By Terence Chea,
Washington Post Staff Writer
Inside the laboratories of the world's major pharmaceutical companies and biotechnology start-ups, an emerging science is quietly transforming the drug industry. Bioinformatics — the use of computers to analyze the inner workings of biology — is helping researchers pinpoint the roots of diseases and design sophisticated medicines to treat them.
But even as it becomes a vital part of drug research, bioinformatics as a business is losing favor with investors. Shares of publicly traded firms that sell biological data and software tools are slumping, and venture capitalists are increasingly wary of investing in such companies.
"Generally speaking, they've lost faith in both tool and data companies," said Victor Li, who manages a biotech hedge fund for Arlington investment bank Friedman, Billings, Ramsey & Co.
The mapping of the human genome is producing vast quantities of information that will lead to greater insight into how disease unfolds at the level of individual genes and proteins. But this volume of data cannot be crunched in a researcher's traditional lab book. Bioinformatics provides the computer tools and databases to search, store, analyze and compare that data and use it to develop safer, more precise medicines. So experiments traditionally conducted in laboratories with petri dishes and microscopes can now be performed by clicking through reams of biological data at a computer workstation.
Bioinformatics companies face many challenges as employees struggle to generate profits and reignite enthusiasm among investors. A growing number of companies are competing for a limited pool of customers. Current business models aren't producing the kind of revenue that attracts investors. And many potential customers aren't willing to pay for bioinformatics products and services.
The fledgling industry's growing pains are especially important to Washington area entrepreneurs and economic development officials who harbor ambitions to make the region a bioinformatics hub alongside Maryland's biotechnology prowess and Virginia's information technology expertise. In Virginia, the Howard Hughes Medical Institute is building a new research campus devoted to bioinformatics in Loudoun County. Fairfax County is creating a business incubator for bioinformatics start-ups.
And the companies themselves, even those with healthy revenues, are evolving from data companies to the potentially more profitable drug-discovery enterprises. But the path to that evolution has had many twists and turns.
In the late 1990s, as technology stocks soared, investors poured hundreds of millions of dollars into new bioinformatics ventures. As scientists raced to decode the human genome, venture capitalists and Wall Street traders speculated that these companies would sell drug researchers the maps and tools needed for the coming medical gold rush.
Several types of companies sprang up. Some companies developed software and hardware to store, organize and analyze vast amounts of biological data pouring out of efforts to map human genes and proteins. Others sold access to huge databases of genetic information that could be mined for drug research.
"Bioinformatics was seen as the bridge between the data generated by the mapping of the human genome and the commercial pathway to better drugs," said Chris Ehrlich, a senior associate at InterWest Partners, a Menlo Park, Calif., venture capital firm that invests in early-stage technology start-ups.
The idea was that bioinformatics would reduce time and money needed to develop new medicines. One blockbuster drug can generate billions of dollars in annual sales, but a new drug typically takes 10 to 15 years and about $800 million to develop, according to a new study by researchers at Tufts University. About one in five experimental drugs fails in patient trials, which adds significantly to drug-development costs.
Bioinformatics could shave years and hundreds of millions of dollars off that process by lowering the rate of failure, proponents said. Detailed genetic information could lead to safer drugs that treat disease with more power and precision.
"Better drugs faster. That was the promise of bioinformatics," said Mark Edwards, managing director of Recombinant Capital, a Walnut Creek, Calif., biotech consulting firm.
But bioinformatics has yet to fulfill its promise, at least from an investor's perspective. Companies that went public last year are trading far below initial offering prices. Compugen Ltd. of Israel, which went public on the Nasdaq Stock Market in August 2000 at $10 per share, now trades for about $4. Lion Bioscience AG of Germany went public that same month for about $40; shares now trade at around $16.
InforMax Inc.'s recent difficulties are illustrative. The Bethesda company, founded in 1990, provides software and services to more than 18,000 research organizations, including about two dozen of the top drug companies. But after going public in October 2000 at $16 per share and climbing as high as $32, its stock now trades for about $3. In October, chief executive Alex Titomirov and President James E. Bernstein resigned after the company announced it would not meet its third-quarter revenue projections.
Another red flag for investors came in November when Genomica Corp. of Boulder, Colo., was purchased by Exelixis Inc. of South San Francisco for $110 million in stock, less than the company's $119 million in total assets at the end of its third quarter ended Sept. 30.
Venture capitalists, who recoup their investments when a company goes public or is acquired, say they're reluctant to invest in bioinformatics companies when they see such low valuations by Wall Street.
"Unless we find a bioinformatics company come to us with a business model that not only shows a clear path to profitability but an exit strategy, we're going to be more skeptical about putting money into that sector," InterWest's Ehrlich said.
One of the major problems, they say, is that the number of customers, never high, is shrinking. The target market is pharmaceutical companies, biotech firms and academic research institutions. And though bioinformatics is important, many researchers aren't willing or can't afford to shell out big bucks to pay for these tools.
Compounding the problem is the growing number of players. The industry generated $1.4 billion in revenue in 2000 and is expected to reach $6.7 billion in 2007, according to a recent study by market research firm Frost & Sullivan. But bioinformatics firms don't just face competition from other upstarts; major technology giants such International Business Machines Corp., Oracle Corp., Compaq Computer Corp., Hitachi Ltd. and Sun Microsystems Inc. have launched their own life science divisions. Meanwhile, many major pharmaceutical and biotech companies are developing in-house bioinformatics capabilities.
Bioinformatics companies also face the same problem as many dot-coms: They must persuade customers to buy their products when similar products are available free over the Internet. Human genome data is losing its commercial value as more becomes available free in online public databases. And free bioinformatics software developed by academic researchers can be downloaded from the Internet.
"How do you make money in a world in which a lot of information and tools are free?" said Pradip K. Banerjee, a senior partner at Accenture. "That sets up a huge challenge for the smaller bioinformatics companies."
With those obstacles, venture capitalists see little potential for big returns. A company developing an innovative cancer drug faces big risks, but it could have a huge market if it succeeds. By contrast, bioinformatics companies selling products and services could generate steady revenues, but will never reach the sales potential of a new drug.
"It may be a good business, but it's just not going to get that big," Recombinant Capital's Edwards said.
After the collapse of tech stocks, investors are most interested in companies developing new drugs and therapies. Therapeutics are riskier and take longer to develop, but they have huge upside potential. Linda Powers, managing director at Bethesda venture firm Toucan Capital, said the firm's second $120 million venture fund is focused on new drugs and therapeutics.
"Increasingly, investors are turning towards therapeutics," Powers said. "It's really the therapeutics that give you the huge revenue potential. They also carry larger risks and longer time frames."
Investors say bioinformatics companies may not survive as stand-alone companies. They want to see companies that incorporate bioinformatics into their drug-development efforts rather than just sell those services to other firms.
Not surprisingly, many bioinformatics companies are shifting strategies. Companies whose original plans revolved around selling genetic data to drug researchers are reinventing themselves as drug discovery and development companies.
One example of this evolution is Celera Genomics Corp., the Rockville company that raced the publicly funded Human Genome Project to sequence the human genetic code. The company's original business plan was to sell access to its gene maps of humans and other organisms to drug researchers. That business generated $89.4 million in revenue in its last fiscal year. But over the past year, the company has started to reinvent itself by using its gene-splicing expertise to discover new medicines. To that end, in November Celera paid $174 million to purchase Axys Pharmaceuticals Inc. of South San Francisco.
Like Celera, Incyte Genomics Inc. of Palo Alto, Calif., built its business around selling genetic information to drug companies. Last month, Incyte hired two former executives at DuPont Pharmaceuticals as its new chief executive and president, a move that bolsters its bid to develop its own drugs.
Drug giant Merck & Co.'s May acquisition of Rosetta Inpharmatics Inc. of Kirkland, Wash. underscored bioinformatics' value to the pharmaceutical industry and may signal future mergers and acquisitions in the sector. Merck paid about $615 million in stock to acquire Rosetta's technology, which uses sophisticated computers and gene analysis to search for new drugs.
"The key challenge the drug industry faces is how to translate this human genome knowledge into some kind of drug molecule or disease target," Accenture's Banerjee said. "Companies that can help make that happen will be successful." |
