The Wall Street Journal makes its text available via a news/retrieval service. The selection here is protected by their: Copyright (c) 1995 Dow Jones & Company, Inc.

New Technique for Discovering Medicines Takes Hold

Drug Makers Seize on "Combinatorial Chemistry" to Churn Out Compounds

A new drug-discovery technique is driving major pharmaceutical companies to shake up their research departments and begin a new round of gobbling up small biotechnology companies.

Dubbed "combinatorial chemistry," the new technique allows chemists to quickly make huge numbers of new substances, which can then be tested on automated machines for disease-fighting activity.

In the past two years, the new technique has moved from the province of a few pioneering biotechnology companies into the mainstream of medicinal chemistry. "It's a super-hot subject," says William DeGrado, a lab leader at DuPont-Merck Pharmaceutical Co. "Almost everyone is doing it."

Expertise in combinatorial chemistry largely prompted Glaxo Wellcome PLC to pay $533 million for Affymax NV, which has a mere 200 employees in its Palo Alto, Calif., labs; spurred Eli Lilly & Co. to buy Sphinx Pharmaceuticals Corp. of Durham, N.C., for almost $80 million; and led Marion Merrell Dow, now owned by Hoechst AG, to buy Selectide Inc. for $58 million.

Other pharmaceutical companies are rushing to add combinatorial- chemistry experts to their staffs. Merck & Co. is even erecting a new building to house its burgeoning combinatorial-chemistry program.

"Never in my life have I seen a [chemical] technology catch on so quickly," says the head researcher at a small California biotechnology company that does combinatorial chemistry. "It's like recombinant DNA for chemists."

Until now, the discovery of new medicinal chemicals has been an exercise in tedium. Working with solutions in test tubes, chemists can only conveniently assemble new molecules one at a time. If several chemicals are synthesized simultaneously in the same solution, it's almost impossible to separate them.

It has taken a century or more for drug and chemical companies to accumulate the hundreds of thousands of chemicals that now sit on their lab shelves. When scientists find a disease-causing microbe or gene that could be the target for a new drug, they start testing the shelf compounds one by one for activity against the new target.

Combinatorial chemistry is more like a chemists' version of ordering from a Chinese restaurant menu. It allows the drug chemist to create new molecules essentially by combining a molecular fragment from Column A with one from Column B and another from Column C, and so on. With this combinatorial technique, a chemist can assemble thousands of new molecules almost overnight.

The old rule of thumb was that a medicinal chemist might create one or two new compounds a week; with the new chemistry, researchers can generate new chemicals at 100 times that rate, once their labs are set up correctly. These can then be screened for promising medicines.

Greater efficiency is something that drug companies desperately need. Screening natural substances for medical potential is not only difficult and time-consuming but also tends to come up with products that are devilishly difficult to make in large quantities. Moreover, rational drug design-engineering new drug molecules from scratch with the help of computersonly works in limited cases.

"A lot of the easy drugs are already out there," explains Mario Geysen, director of biotechnology at Glaxo Inc., a unit of Glaxo Wellcome. For new drugs, he says, companies will increasingly focus on less common diseases, representing smaller potential markets. "If we don't become more efficient, costs will preclude the development of new drugs."

The key that opened the way to combinatorial chemistry was found in the mid-1960s, when chemists developed a method for synthesizing molecules on the surfaces of tiny particles of polystyrene. Making chemicals this way had several advantages over brewing them in liquid- filled test tubes, the traditional method. In particular, each particle carries a different molecule, making it practical to keep track of many different chemicals at once.

In 1984, Glaxo's Dr. Geysen used this "solid-phase-synthesis" technique to create a "library" containing millions of variants of simple protein-like molecules. That attracted the attention of many biotechnology companies, because proteins are the products of their genetic-engineering efforts.

Most big drug makers, which deemed the complex proteins to be the province of the biotech companies, initially paid little attention to the technique. That changed in 1992, when chemists at the University of California at Berkeley used the new technique to concoct variants on a class of small molecules similar to such profitable drugs as Valium.

Today, companies are using combinatorial chemistry to generate huge libraries of chemicals. Closely held Pharmacopeia Inc. in Princeton, N.J., for example, has contracts with several major drug companies. To date, tiny Pharmacopeia has generated about 100,000 new chemicals, and by year end, it will have one million compounds, according to John Chabala, Pharmacopeia's chief scientific officer.

"The drug industry sees [these libraries] as a way to match the enormous rate at which potential drug targets are becoming available," says Glaxo's Dr. Geysen. The Human Genome Project, for instance, is uncovering thousands of genes, enzymes, hormones and other biological products that are potential targets for new drugs created by combinatorial chemistry.

For companies that already have many new-drug "leads," combinatorial chemistry is speeding later stages of drug development. Typically, drug companies will produce 500 or so slight variants of a promising lead to see which one works best. That process takes a year or more with traditional chemistry. But at Eli Lilly, researchers working on a new nervous-system drug used combinatorial chemistry to slash the time to a few months.

Lilly's nervous-system drug, considered one of the most promising in the company's pipeline, should enter human testing by year end. That would make it one of the first products of combinatorial chemistry to reach such a late stage.

"Two years ago, most of the other chemists at Lilly were skeptical" of combinatorial chemistry, says Stephen Kaldor, head of Lilly's combinatorial-chemistry effort. "I don't hear any skepticism anymore," he adds.

As more drugs from other companies reach clinical trials in the next two years, the new technique is expected to grow in influence. Currently, roughly half of the medicinally important types of molecules can be made using the new techniques. But researchers in academia are scrambling to extend combinatorial chemistry to most classes of molecules.

"This is not a fad," says Lilly's Dr. Kaldor. "It is here to stay."

This article originally appeared in The Wall Street Journal on 9/11/95.