Scattered Clouds, With a Chance Of Developing Miracle Drugs - Industry Trend or Event

Will the study of proteins transform the pharmaceutical industry? investors are betting big on the latest boom in biotech.

IN A CLUTTERED LAB ACROSS THE BAY from San Francisco, chemist Bob Chapman grasps between his thumb and forefingers a glass plate containing a titanium chip. Though the device looks like something thrown together in a garage, it's actually a prototype for a technology that Chapman and his 54 colleagues at Signature BioScience hope will radically accelerate the discovery of new drugs.

In biotech as on the Net, speed is money, and Signature's all-about-speed technology has attracted $21 million in venture funding since the company launched in October 1998. Signature is one of the first players in the burgeoning field of "proteomics" The term is a mouthful, and the science of applying advanced technologies to protein research in hopes of making drug discovery smarter and faster is even harder to articulate. That hasn't stopped investors -- many of them from the tech and Internet world -- from making proteomics the next hot thing in biotech.

The promise of proteomics is breathtaking: a flood of scientific data that will help the pharmaceutical industry pinpoint new drugs that are more effective and less toxic than what's currently on the market.

Throughout history, scientists have discovered cures on a mostly hit-and miss basis -- penicillin, for example, was pretty much a lucky guess. Genetic researchers take a different approach. By studying the genome -- the blueprint for human life -- they hope to better understand the root causes of disease and build cures from the bottom up.

The field of proteomics has taken this research to the next level. Rather than study the blueprints, proteomics specialists study the actual building materials: proteins, the living matter that genes produce. Proteins are essential to all organisms; they are responsible for everything from how your body grows to how it digests food. But when they get out of whack, they can cause a body to break down. By figuring out which proteins do what, proteomics firms promise to significantly narrow the search for drugs to fight cancer, heart disease and other illnesses.

"Understanding how proteins work, how they interact, is incredibly important," says Terry McGuire, a co-founder of Polaris Venture Partners, which has investments in several genomics companies that are flirting with proteomics. "Each step along the way, we're going to open a new door."

The surge of interest in proteomics can be traced to the exhilarating 10-year effort to map the human genome. As the completion of the first draft neared last year, investors stampeded back to biotech. The influx helped the industry recover from a dramatic slump -- known in the industry as biotech's "nuclear winter" -- that stretched through most of the 1990s.

In 2000, VCs poured about $2.8 billion into young biotech firms. That was double the 1999 amount and triple that for 1998. While countless Net businesses collapsed last year, 67 biotech firms were confident enough to float initial public offerings -- a single-year record for the industry. Biotech's market capitalization rose 38 percent in one year, to $429 billion. That figure has fallen as the Nasdaq biotech index has dropped about 25 percent this year, keeping pace with the overall market decline.

The first beneficiaries of the human genome project were genomics companies like Incyte and, most famously, Celera, which shares a large part of the credit for sequencing the genome. Both saw their stocks soar last year. But once the hype cooled, investors realized something that scientists knew all along. Though genomics firms can help isolate disease-related genes, that information by itself won't significantly improve the time-consuming and expensive process of developing drugs. Proteomics, on the other hand, promises to speed up things considerably.

The field has grown quickly. There are now about 50 companies primarily focused on proteomics, according to Frost & Sullivan analyst Justin Saeks, and another 150 or so that spend at least some of their time on protein research. These companies are trying to squeeze revenue out of proteins several ways: research services, analytical software and scientific instruments.

As the firms strive to make drug development less of a crapshoot, virtually every biotech VC company has added proteomics to its portfolio. Even big tech players like IBM and Compaq are investing in proteomics ventures [see sidebar, at right]. The potential payoff is high, but as Signature's story suggests, so are the risks.

Beginning with $1 million of venture money in 1998, founder John Hefti grew Signature BioScience the hard way. After germinating in a rented lab in Silicon Valley, the company moved to a dinky starter office, then a shared space, and only recently into its own 30,000-square-foot building in Hayward, 27 miles southeast of San Francisco. A clone amid acres of generic office boxes, the facility's bland countenance befits an industry in which substance is everything, style nothing.

Perhaps that's a sign of maturity. Signature's founders -- old hands with impressive credentials -- are experienced enough to have learned from past mistakes in both science and business. More than half of Signature's employees have M.D.s or Ph.D.s; Hefti and a few others have both. Signature President and COO Andrew Sandham, a compact Englishman at his fourth startup, has a resume full of executive positions in the pharmaceutical industry, as does CEO Mark McDade.

But the success of Signature hinges on its technology. In the demonstration lab at the back of Signature's half-finished new headquarters, Sandham gestures with pride at a jury-rigged contraption. It's a work in progress. The. central device, a clear acrylic box roughly the size of a toaster oven, encases a motorized metal cylinder. The box is attached to a microwave transmitter with front-panel readouts that jump like those of a heart monitor. A narrow tube carrying electric-green liquid runs over the top; Sandham calls it the "Bob tube" (after Chapman, an inveterate experimenter who joined Signature fresh from his postdoctoral work at Harvard University).

Signature dubbed this unique technology Multipole Coupling Spectroscopy. Lab scientists shoot proteins and drug compounds through the Bob tube and examine the interactions between the two. The compound and protein may not interact at all, which means the compound has no potential as a drug. The compound might stick to the protein but not change it -- more interesting, but also a dud. Or the compound could stick to the protein and actually change it. That protein would be, in industry parlance, a "target." As a target, this protein becomes a focal point for further experimentation [see illustration, page 87].

Targets are gold to pharmaceutical companies, but finding them has traditionally been an expensive and somewhat scattershot process involving the extended pursuit of numerous dead ends. If Signature perfects its technology -- challenges include turning it into an assembly-line process and miniaturizing the machinery -- drug companies could send it a set of proteins that may play a role in, say, breast cancer, along with a library of potential drug compounds. Signature's technology would identify the most-promising candidates, removing much of the early guesswork and saving pharmaceutical companies time and money.

This is an infant business, however, and after nearly two and a half years the company, like many young firms, has zero revenue (it expects to break even in 2004). And, as with so many startups, the corporate vision can sound a bit fuzzy. "It depends on who you talk to," Sandham says. "John Hefti's vision is that he's basically got a new paradigm for drug discovery, for looking at how proteins are structured and how they interact. He's very excited by that vision. That's more the academic viewpoint." Getting down to business, Sandham adds, "The real vision of the company is to have products accelerated into preclinical and clinical development by our partners and eventually by ourselves."

The key word is "eventually." The lure of manufacturing its own drugs is clear; successful products typically enjoy 80 percent to 90 percent gross margins. Still, investors are becoming increasingly wary of product-only startups, many of which crashed in biotech's early days as their promising drugs died in clinical trials or were shot down by Food and Drug Administration examiners. That's why many proteomics companies, like Signature, are focusing first on technology and services that can bring in money relatively soon, while talking about someday getting into the drug-development game. Even then, most won't end up manufacturing their own drugs, which would take a major infusion of capital.

For now, Signature is pursuing a variety of revenue streams: licensing its targeting technology to manufacturers that will make Multipole Coupling Spectroscopy instruments; developing databases of protein information and selling access to them to pharmaceutical companies; and providing protein analysis services to drug companies. A balanced portfolio betters the odds of survival.

The future of proteomics companies hinges on how successful they are at winning over the $350 billion pharmaceutical industry -- the ultimate customer for the information, tools and data they develop. With a significant number of drug patents approaching the 20-year expiration date, the industry has been under constant pressure to maintain its historically healthy growth rates with new rounds of breakthrough drugs. It typically takes 10 to 12 years, however, to bring a drug from development through approval and to market. Worse, there's a 9-in-10 failure rate for medications that reach the clinical-trial stage.

Drugmakers are eager for innovations that can improve those numbers, but they have yet to carve out a big budget for unproven proteomics technologies. A few are starting to, however, including Pfizer and Novartis. And for good reason. Revenues for proteomics-related instruments and supplies, laboratory analysis and bioinformatics (a category encompassing databases, analytical applications and computer modeling) amounted to an estimated $1 billion last year, according to Frost & Sullivan, and are predicted to hit $5.8 billion by 2005.

As pharmaceuticals companies warm to proteomics, VCs are expecting a shakeout among the dozens of startups. "There's a lot of talk about proteomics, but not as many technical breakthroughs as there are companies being funded," says Robert Nelson of Arch Venture Partners. "Everybody's getting excited because it's the next big thing, but the problem with biotech in general is that innovation occurs when it occurs; you can't force it."

Maybe so, but the entrepreneurs at Signature are moving as fast as they can to force a breakthrough or two. So much so that Andy Sandham hasn't even had time to settle in at company headquarters. As he searches for a quiet and furnished place to talk, he walks into a room chosen at random and then realizes, "Oh, this is my office."

Sandham is refreshingly candid about Signature's challenges, and he's careful not to overstate his case. "We have a mission that is pretty complex at the moment. This is a very new technology, and there are a lot of problems to solve."

Unlike tanking dot-coms, proteomics companies are building on more than two decades of innovation in biotech. There are hundreds of biotech pharmaceuticals in clinical trials and on the market. Even if hopefuls like Signature fail, other companies and technologies will continue to bring drug development closer to blackjack than roulette -- more computation, less speculation. The odds are improving.

                   Teaching the World to Say Proteomics
                 In recent years, dozens of companies have
                 begun focusing on the study of proteins.
COMPANY              FINANCIAL POSITION [*]
STARTUPS
Structural Genomix   $85 million in funding from Atlas Ventures,
San Diego            Prospect Venture Partners and others.
Syrrx                $70 million in funding from Bay City Capital,
San Diego            MPM Asset Management and others.
PUBLIC
COMPANIES
Large Scale Biology  Stock price down to $5.19 after August 2000
Rockville, Md.       IPO at $26.
Oxford Glycosciences Stock price up to [pound]11 ($15.95) on LSE
Abingdon, England    after April 1998 IPO at [pound]2.80
                     ($4.70 in 1998).
GENOMICS
CROSSOVERS
Celera Genomics      Stock price up to $36.50 after April 1999 spin-
Rockville, Md.       off from Perkins-Elmer (now Applera) at $18.
Myriad Genetics      Stock price up to $40.80 after October 1995
Salt Lake City       IPO at $18.
COMPANY              TECHNOLOGY
STARTUPS
Structural Genomix   Turning X-ray crystallography -- a standard
San Diego            but slow method of discovering protein
Syrrx                structures -- into a high-speed process.
San Diego
PUBLIC
COMPANIES
Large Scale Biology  Developing protein analysis software
Rockville, Md.       platform; selling data from Human Protein
                     Index.
Oxford Glycosciences Developing tools for industrial-scale
Abingdon, England    separation and identification of
                     proteins.
GENOMICS
CROSSOVERS
Celera Genomics      Hoping to catalog up to a million pieces
Rockville, Md.       of proteins a day at expanded computing
                     center.
Myriad Genetics      Working on industrial-scale system for
Salt Lake City       discovering protein-protein interactions.
(*.)AS OF MARCH 9, 2001.
SOURCE: COMPANIES AND MARKET REPORTS

Putting the Tech in Biotech

Some familiar names from the computer-technology and Internet worlds are making the move into genomics and proteomics. Former Microsoft CTO Nathan Myhrvold has been traveling the biotech conference circuit, delivering speeches proclaiming genomics -- and biotech in general -- the next "exponential industry." Serial entrepreneur Jim Clark sits on the board of gene research firm DNA Sciences. Softbank, Yahoo's sugar daddy, plans to launch a life-sciences venture capital fund this spring. This is less a matter of carpetbagging than it might seem, since biotech is increasingly becoming a data-driven business. Indeed, the biggest biotech converts so far are the info-tech powerhouses. Here's what they're up to and why.

COMPAQ

Biotech ventures: Supplied systems for most major gene-sequencing centers; formed an R&D alliance with Cetera Genomics and Sandia National Laboratories; invested $100 million in early-stage genomics, bioinformatics and related companies, directly and via venture capital funds.

The draw: "There are two interesting aspects," says Ty Rabe, director of high-performance technical computing solutions. "First, pharmaceutical and biotech companies developing products want to take advantage of genomics information, and those companies have been making major investments in their research capabilities. The other is the more dynamic, brand-new industry of specialized companies operating to supply the information to biotech and pharmaceutical companies. We believe this will be a major new industry."

IBM

Biotech ventures: Committed $100 million to Blue Gene, a supercomputer designed to simulate protein structures, and another $100 million in initial funding to build a life-sciences business unit. Also invested in biotech research partners and contributed to biotech venture funds.

The draw: "Early in 2000, we started to see the growth in new biotech companies and took the view that with the human genome there would be a real growth spurt," says Caroline Kovac, VP of IBM's life sciences unit. "We believe this will be a very hot opportunity for IT over the next decade."

SILICON GRAPHICS

Biotech ventures: Developing specialized servers, workstations, 3-D visualization tools and database applications for biotech research.

The draw: Science research accounts for about one-fifth of SGI's business, and the company expects continued growth, says David Zirl, director of life- and chemical-sciences marketing. "The genome is just the start of things. Protein analysis will produce much more data than we have today, and it's very complex."

SUN MICROSYSTEMS

Biotech ventures: Formed a consortium to support development of an open platform for the life-sciences industry based on Java and XML.

The draw: "Biotech growth will more than replace any loss of dot-coin revenue," says Sia Zadeh, manager of Sun's life-sciences group. "The growth of computational resources and storage needs is very significant. Biotech requirements are real and they're addressing a real market, not a hypothetical market. How many Internet pet companies do you need? Not too many. But there's a real need for human health care."

COPYRIGHT 2001 Standard Media International
COPYRIGHT 2001 Gale Group