Biotech Grows Up

(3 of 3)

Proteomics is all the rage in Europe as well. The British biotech Oxford GlycoSciences announced this month that it had filed for patents on 4,000 proteins. It has $280 million in cash reserves and is awaiting U.S. and European approval of a drug for Gaucher disease, a rare inherited disorder.

To speed the discovery of drugs that attack their targets without poisoning those who take them, many biotech firms rely on design rather than serendipity. Vertex Pharmaceuticals of Cambridge, Mass., has halved, to 18 months, the time it takes to discover candidate drugs and is today among the most prolific generators of leads.

Yet the nuts and bolts of its discovery strategy are not radically new. Vertex essentially finds the structures of its target proteins and designs molecules that slip selectively into the grooves found along the proteins' surfaces. It did this expertly in designing the anti-HIV drug Agenerase, which it promotes with GlaxoSmithKline. A handful of other firms have developed their own drugs in similar ways.

Because a protein's structure is essential to its function, those proteins that perform similar functions often have stretches where they are similarly shaped. Instead of focusing on the structure of just one target, Vertex homes in on entire families of proteins as it seeks out its leads. This is done by crystallizing a protein of interest and exposing it to intense X-ray beams. The way those beams are scattered reveals how atoms along the molecule's length are arranged--information that is converted into a protein structure by computers. Chemists use this structure to digitally model molecules that should fit--like the pieces of a jigsaw puzzle--into grooves on the protein's surface. When the company has found about 100 that might work, it produces them and tests each out against the target.

Here's the clever part. While all this testing is going on, Vertex's chemists are already at their computers using the protein's structure to model what its cellular kin might look like and thus develop drugs against each one of them. Rather than starting from scratch, however, the chemists dip into the compounds they derived from their digital screening of the first target and apply them to each member of its family. The power of this approach is that these steps are taken simultaneously--with all the tedious testing needed to move a drug into clinical trials.

There's more than bragging value here. Vertex's approach has won it 25 partnerships with drug firms. Last year Vertex signed an $800 million alliance with Novartis to develop eight drugs that target a family of proteins called kinases to treat diseases such as cancer and diabetes.

Actelion, based in Basel, Switzerland, applied parallel drug discovery to produce a drug for pulmonary hypertension that received FDA approval last month. It has other drugs in late-stage development against heart failure. The French company Cerep, meanwhile, also boasts a computer-savvy parallel drug-design strategy but plans to license out its drug candidates after the first phase of clinical trials. And there is no shortage of competition in the U.S. "The future of medicine is in structure-based drug design," says Celera CEO Craig Venter, explaining his company's acquisition of Axys.

The strongest biotechs and pharmaceutical firms are working to integrate what Vertex CEO Joshua Boger calls "all the technologies that are advancing drug discovery." Eager to boost its target- and drug-screening expertise, Vertex scooped up Aurora Biosciences for $600 million this summer. Millennium has bought four companies in the past four years to master each step of the drug-discovery process. "To think of any of these technologies in isolation," says chief technology officer Michael Pavia, "doesn't really help you much. You have to put them all to work together."

Managing the information these technologies generate and distributing it efficiently across the company are just as important. "The people who do this in an outstanding way will have the competitive edge," says Daniel Vasella, CEO of Novartis. What's harder to say is when--and for whom--that edge will pay off.

Most of the drugs that will emerge from the current revolution in rational drug design won't reach the market for at least another seven years. And some will fail spectacularly. So will some of the companies. On the other hand, because most biotechs are relatively small, their potential for growth is enormous. Modest successes can give a powerful boost to share values. Then there's the science. "This is not e-commerce," says Jan Buck, managing director of Arthur D. Little's Global Health Sciences Fund. "It may have been like e-commerce back in the '80s, but this is stuff that's coming to fruition. This is real product."