National Editor

SAN FRANCISCO - How and why to nail down deals in tough economic times was the topic of talk from lecterns and in the hallways of the Westin St. Francis Hotel here, during Allicense 2003, a two-day conference sponsored by the Walnut Creek, Calif.-based consulting firm Recombinant Capital.

"In a perfect world, we like to think that a deal emerges from a very well-thought-out strategy that senior management spent a lot of time developing," said Jack Anthony, senior vice president of business and commercial development for Tularik Inc., of South San Francisco.

For better or worse, however, that's not always the case, he told attendees.

"Sometimes strategy' is your boss coming in on Monday morning with BioWorld [Today] in his hand, saying, Why can't we do a Pfizer-Neurocrine deal?'" Anthony said.

San Diego-based Neurocrine Biosciences Inc. and Pfizer Inc., of New York, signed a $400 million agreement in December, focused on the former's late-stage insomnia drug. (See BioWorld Today, Dec. 20, 2002.)

Anthony called for companies to find precise and worthy reasons - on both ends - for any deals that get done.

"With very few exceptions, there are no protective franchises in this business," he warned. "You have to get outside yourself and ask, Why would somebody want to do a deal with us?'"

In the post-genomics era, those reasons must be persuasive - and Perlegen Sciences Inc., of Mountain View, Calif., has managed (during the same month as the Neurocrine-Pfizer deal, coincidentally) to seal agreements with giants such as Eli Lilly and Co., of Indianapolis, to identify genetic markers for Lilly; with Pfizer Inc., of New York, to discover the human genetic contributions to cardiovascular disease; and with Bristol-Myers Squibb Co., also of New York, to identify markers of patient response to medications in Bristol-Myers' portfolio.

"The literature is full of people who've associated a gene here and a gene there with a particular trait, only to find out that no one's been able to reproduce it," said Brad Margus, CEO of Perlegen.

Perlegen uses high-density arrays of DNA probes on glass wafers. Each wafer holds about 60 million DNA probes that can be used to recognize longer sample DNA sequences. The method finds which probes the sample DNA hybridizes more strongly to in order to determine if a known sequence of DNA is present in the sample.

The company is able to "interrogate" 1.7 million single nucleotide polymorphisms, Margus said, using 500 cases and 500 controls in each instance.

A genotype - one SNP in one person - carries varying price tags, depending on which company you ask, he noted.

"You'll hear numbers all over the place," he said. Accepting the figure of 20 cents per genotype, "it doesn't take a rocket scientist to figure out that it's a pretty expensive experiment to look at 500 cases and 500 controls in 1,000 people [across] 1.7 million SNPs. If you do it once, it costs about $340 million. And you don't want to do it once, you want to do a replicate."

Perlegen's price tag, about $2.2 million, Margus said. He compared the company's capability to a car that could drive around the world at the equator on about seven gallons of gas.

"The bottom line is that it's really hard to find genes," he said. "All of us have heard for years now about a gene being discovered once a week. But what they are usually finding genes for is one-gene disorders."

While that approach works for many rare diseases, scientists believe that 30 to 50 genes are involved in more prevalent illnesses, Margus said.

Perlegen's technology has proved helpful in various stages, he said. Some companies might have "Phase III clinical trials up the wazoo on a drug they spent a lot of money on, and they've got a small group of people who have an adverse reaction," which needs to be explained. Others in preclinical are "eager for a real target that they can say is positive in humans," Margus said.

Two of Perlegen's deals involve drugs already marketed.

"Do those people want to find the genetic reasons why people don't respond to their drug so they can make a test and reduce their market size? I don't think so," he said. "But they are interested in making a better drug. Competition is on their heels."

The work needs to be done fast, too, Margus said.

"You'll meet geneticists who have spent five or 10 years looking at one area of the genome to see if it's associated with a disease," he said. "In four to six months, [Perlegen] can look across the entire genome and find all the genes that are involved in causing a common disease like schizophrenia, diabetes, cardiovascular disease."

What's more, companies with speedy technology need to exploit it early and as fully as possible to draw collaborators and make profits independently.

"If you come up in five years with a way to sequence any [entire] genome for a thousand bucks or for a dollar even, but the day before, someone else has already found the genes that play a role in cardiovascular disease, you're too late," he said.

Allicense 2003 concluded Wednesday.