By Randall Osborne

Editor

Having pieced together what it's calling a "major ramp-up" by way of a strategic technology arrangement that involves six collaborations with biotechnology firms, Pfizer Inc. is ready to elaborate on its plan to beat the attrition problem in drug design.

Even by Pfizer standards, the effort is big.

The pharmaceutical giant typically lets biotechnology companies do the talking about agreements, but Pfizer's effort has become "an integrated story," said Alan Proctor, vice president of the company's Discovery Technology Center in Cambridge, Mass., which opened in January.

"We haven't really made a lot of fuss about it, but this is a big deal at Pfizer, the biggest thing we've ever done in research," Proctor said, declining to specify how much the firm spent on its center and may spend on the partnerships in total.

"What we really tried to do is put together a suite of technologies that speak to various stages in the drug-discovery process," Proctor told BioWorld Financial Watch. "We realized we really need to ramp up our productivity. The process is beset with one major problem, and that's attrition. We lose things along the way."

About one in 100 would-be drugs makes its way out of early testing, he noted. As the race to sequence the human genome proceeds, the fight to devise better drugs first will become hotter as well.

Important to the Pfizer plan are deals with longtime partner Neurogen Inc., of Branford, Conn., and with ArQule Inc., of Medford, Mass., Proctor said.

The Neurogen collaboration, worth $27 million, non-exclusively licenses that firm's Accelerated Intelligent Drug Discovery (AIDD) to Pfizer.

AIDD has served as the basis of a couple of drug development collaborations with Pfizer, in the indications of anxiety, depression, insomnia, Alzheimer's disease and obesity in humans, and dementia and anxiety in companion animals.

Neurogen's program integrates combinatorial compound synthesis, high-throughput screening and advanced informatics to generate drug discovery data. The result is a process that can otherwise take several months can be condensed into two weeks.

"You start with a virtual library of chemical compounds, and then you sort of design [the prospective drug] in the chemist's mind," Proctor explained. "We try to make the compounds very drug-like, so they don't fail later on. We can design molecules that satisfy that quality, as well as design the compound in iterative cycles."

The ArQule deal, worth up to $117 million, calls for ArQule to devote the designated technologies, scientists and its AMAP Chemistry operating system to generating compound libraries for Pfizer.

More recently came the agreements with Celera Genomics, of Rockville, Md., and Incyte Pharmaceuticals Inc., of Palo Alto, Calif.

Celera will combine its expertise with Pfizer's to identify novel drug target genes derived from Celera's human genome sequencing. The collaboration will focus on selecting drug targets from the thousands of genes discovered by Celera's high-throughput gene discovery pipeline. One analyst valued the deal at more than $50 million.

Pfizer will have access to five databases developed by Celera: the Celera Human Gene Index, which provides a set of human genes derived from its expressed sequence tag (EST) sequencing program; the Drosophila Genome Database, which provides the complete Drosophila melanogaster genome sequence; the Human Genome Database, which is being developed to provide the complete sequence of the human genome and the entire collection of human genes with links to associated biological and disease information; the Celera SNP Database, which includes single nucleotide polymorphism data from the sequencing of several individuals; and the Mouse Genome database, which allows for comparative analysis with the human genome.

In the Incyte agreement, Pfizer will be analyzing sequenced data in developing full-length clones for drug targets. Incyte expects to file patent applications on the drug targets and then license them out.

"Basically, we have to feed this incredibly voracious discovery machine we're building with lots of targets," Proctor said. The process is "driven by rapid disclosure of genomics sequences, accelerated by Celera."

During the next year, the majority of genes will be the subject of patent applications, Proctor said.

"They're filed in pockets, here and there, and we're seeking to influence our access to a large chunk of intellectual property," he said.

Pfizer is "not trying to tie this up on an exclusive basis," Proctor emphasized. "Competitors can get in. We think that's to our benefit, and to the benefit in general of the human population. Attrition is a killer; you have to have a lot of shots before you get one hit. So this serves the public much better, and it serves us. It will allow us to springboard off others' peoples discoveries."

To round out its methodology, Pfizer has invested in high-throughput screening technologies by Aurora Biosciences Inc., of San Diego, and Evotec Biosystems AG, of Hamburg, Germany.

In the $50 million deal with Aurora, the firm will provide scientists as well as automated systems, including those for compound storage, high-throughput screening and multiple ion channel lead discovery platforms and development stations, Pfizer's research facilities worldwide.

Evotec, in a $17 million deal that includes a $1 million equity investment, will provide access to its EVOscreen technology and a non-exclusive, worldwide license to its hardware and software.

From these two firms, Pfizer gets "a 25-fold improvement in our ability to run drug-screening processes," with fluorescence-based technology, Proctor said.

"There again, it's a qualitative change as well as a major quantitative change," he said. "It increases the number of assays, and scales back to the nanoscale the volume of assays."

With the center in Cambridge, Pfizer wanted "a place where we could experiment with new paradigms, and new robotic and informatic technologies. That's our product here, not drugs per se."

Drug discovery firms, Proctor said, tend to "pay so much attention to getting drugs out the door that it's difficult to get your head up from the assembly line."

Pfizer's approach is becoming "a parallel approach to drug design, using gene families, clusters of genes that have the same characteristics. You can screen simultaneously many if not all members of a gene family against a library, but you have to have the robotics and informatics to massage and manipulate that data. Chemists make a lot of compounds, some of which don't react well with the target you're looking at, but may react with a related one, but you don't know that."

While refusing to describe spending in precise amounts, Proctor said the investment in the new technology center and in its "ramping up" is immense.

"It's really substantial, a significant portion of our annual research budget. This is a new game we're playing." *