Issue Date: February 23, 2009
Exelixis After GSK
EXELIXIS, the South San Francisco biotechnology company, has traditionally done things in a big way. And it likes big numbers. Its 300-some researchers in drug discovery have at their disposal a library of 4.5 million compounds. From screens of these compounds, they have identified 15 drug candidates worthy of clinical trials, several of them first-in-class oncology treatments. The company has partnerships with six large drug firms.
Last October, however, Exelixis found itself looking at the number one. That is the number of compounds GlaxoSmithKline took at the end of a six-year collaboration under which it could have selected any two that had reached proof of concept. It signed on for XL880, a dual VEGF and Met protein inhibitor for treatment of various tumors, but declined to take another of the five compounds that Exelixis had developed through their partnership.
Exelixis announced GSK's decision on the same day that it presented data on several of its compounds at a major cancer symposium in Geneva. The decision drew a lot of attention from investors and other industry watchers weary of the high-spending, zero-profit nature of the biotech industry. Already on the decline for several months, the firm's stock fell more than 25% that day.
But two months later, Exelixis' shares jumped that much and more when Bristol-Myers Squibb signed on to codevelop two of the molecules identified in the GSK partnership, including Exelixis' lead compound, XL184. Bristol-Myers agreed to pay $195 million up front, compared with the $55 million GSK would have paid had it taken a second compound under the original agreement.
Although this financial rollercoaster ride ended smoothly enough, it raised a big question: Does Exelixis' high-stakes approach to drug discovery still makes sense in an era when many biotech firms are scaling back and focusing their efforts on just one or two products?
Not everyone thinks so. "Looking back, the collaboration didn't work out all that well for Exelixis or for GSK," says Eric Schmidt, a stock analyst at Cowen & Co. "Neither really got the productivity out of the relationship that it might have hoped for." Exelixis, he adds, is still indebted to GSK for about $100 million that it borrowed under the arrangement.
Viren Mehta, managing partner of the pharmaceutical investment research firm Mehta Partners, says Exelixis' stock drop last fall was a reaction to more than just the end of the GSK partnership. "It was a convergence of many negative forces," he contends. Investors, he says, were losing patience in a company that spent $225 million on R&D in 2007 but took in only $114 million in revenues, none of which were from actual product sales. "Exelixis is a poster child of how innovation in this industry has continued to lag amid receding timelines for achieving profitability," Mehta asserts.
He blames Exelixis' high R&D spending on its "multiple shots on goal" strategy of drug development. What appeared to be a prudent approach of moving multiple compounds into single therapeutic areas has proved less effective than "prioritizing early and going all out on the goal," he says. "Hopefully Exelixis has learned its lesson."
George A. Scangos, Exelixis' chief executive officer, sees things differently. He argues that the GSK partnership was, overall, successful. "We had a long-standing collaboration, from 2002 to 2008," he says. "GSK paid a substantial amount and got in return a compound, XL880, which is moving forward well in their hands."
Michael Morrissey, president of research and development, agrees. "GSK's expected outcome, as we understood it, was that if it got one good compound, that would make it a win for them," he says. For Exelixis' part, he says the deal provided cash to build its R&D operation and develop a pipeline of first-in-class compounds.
BOTH EXECUTIVES point to the quality of the lead compound that Bristol-Myers signed on for, XL184, an inhibitor of the Met and VEGFR2 proteins, currently in Phase III clinical trials for thyroid cancer. Morrissey speculates that similarities between XL880 and XL184 may have influenced GSK's decision not to take the second one. "The question is whether they would double down or not—whether they would invest again in the same target class," he says.
Scangos suggests another spin. "One way to look at it is that GSK's failure to select XL184 reflects their enthusiasm for XL880," he says.
Both executives say the success of the GSK deal must be judged in the context of its six-year span—a period during which the biotech industry matured significantly. GSK had entered a drug discovery collaboration with an open-ended commitment to take or leave what came of it. Bristol-Myers, on the other hand, is signing on to a well-developed compound. "There is a reasonable chance we can file our first New Drug Application in 2011," Scangos says of XL184. "We're within shouting distance of that."
And although some analysts question Exelixis' strategy, they can't deny that the company is in better shape than many biotechs in a difficult economic time. With a wave of downsizing that has seen some firms cut much of their staff and eliminate research on all but one or two advanced prospects, Exelixis made a relatively minor reduction in its workforce and has not dropped any projects. The company maintains the "critical mass" that Scangos and Morrissey say has supported a steady march toward becoming a pharmaceutical company with drugs on the market.
Exelixis started that march with a very different purpose at an early stage of the biotechnology industry. Formed as a genetic research firm in 1994 in Cambridge, Mass., Exelixis moved to South San Francisco in 1997. It signed its first significant partnership—an agricultural chemistry deal with Bayer—the following year. It went on to become one of the first companies to evolve from offering genetic research services into a full-fledged drug discovery firm.
"The idea when we started, prior to the human genome," Scangos says, "was that there would be sufficient conservation in biochemical pathways and their control points between humans and simple organisms like fruit flies and C. elegans worms that you could use those organisms to do powerful genetic experiments to elucidate the pathways and find good points of intervention." The company started analyzing organisms such as fruit flies, worms, and zebrafish.
THE BUSINESS PLAN was to perform sophisticated genetic experiments and market the information to drug companies that needed help with biology. "Back then, people felt that genomics had a virtually unlimited, infinite potential," Scangos recalls, "and basically what we were selling was a small piece of infinity."
This turned out to be a bad way of making money. "You would get an up-front payment and headcount support that would cover your costs and then a royalty rate on compounds when they hit the market," Scangos says. But the royalty payment wouldn't show up until 15 years later. "It was a high-risk, cost-plus business model. And cost-plus, high risk is not viable," Scangos observes. "It became clear that we had to build the capability to make drugs based on the biological insights that we were generating."
It was a risky shift—made all the more so by a conviction to do it right. "With the backing of the board," Scangos says, "we made a commitment that we would build up a drug discovery group of at least a hundred researchers, that we would do this big-time, that we would build critical mass, and that we would build excellence."
Morrissey liked the idea. The self-described Wisconsin boy had worked at Ciba-Geigy and Berlex Laboratories before coming to Exelixis in 2000. "Exelixis was of the class that started doing postgenomics discovery," he says. "We had a view of investing in a critical mass of people and technology, a soup-to-nuts discovery that would allow us to be able to compete not so much with our biotech brethren but with any big pharma."
During the past eight years, Morrissey says, Exelixis has amassed the skill and technology to screen rapidly, leveraging "the density and diversity" of a sizable compound library and applying leads to multiple scaffolds for analysis. "We try to complement these scaffolds with structural biology early in the process," he says.
Indeed, biology is key to the discovery process. Morrissey says Exelixis has "fantastic people on the biotech side" with skills in protein analysis, high-throughput screening, and protein crystallography. "Over the past five years, we have solved over 900 enzyme inhibitor enzyme complexes," he points out.
That biotech expertise has attracted partners like GSK, Bristol-Myers, Wyeth, Genentech, and Daiichi Sankyo, Morrissey says. He says GSK approached Exelixis in early 2002 during a search for a biotech partner that could carry out drug discovery the way a well-established company would. Moreover, GSK wanted a partner whose processes would integrate well with its Center of Excellence for Drug Discovery system. Exelixis came out on top in a field of about 25 biotechs, Morrissey says.
Between an up-front payment, research funding, and milestone payments, the six-year partnership funneled $155 million to Exelixis and validated its discovery platform. Morrissey concedes, however, that the deal was initially perceived as ending in a letdown for Exelixis—and one that raised questions.
"We were penalized by GSK saying no," he acknowledges. "It caused some concern among investors in regard to whether the compounds were as good as we said they were." The Bristol-Myers deal alleviated these concerns, he says, and the analysts Schmidt and Mehta endorse this view.
Scangos says GSK's decision can be interpreted several ways, but he prefers to look forward to new partnerships, confident that Exelixis' pipeline will attract significant deals. "GSK didn't choose XL184 or any of the other compounds," he says, "but we have six that had been part of the collaboration. We have had a lot of interest from a variety of companies in a number of those compounds and now have a collaboration with [Bristol-Myers]."
The economic downturn is now prompting Exelixis to seek partners for compounds it might normally have developed on its own. Frank L. Karbe, chief financial officer, says the firm launched an aggressive partnership program last year partly in response to the credit crisis but also to minimize future development costs and alleviate investor concerns over R&D spending. "There is a huge misperception about R&D burn rate," he says. "We have significant operating expenses, but much of it is covered by partnership commitments."
And more of those expenses will be shared by partners soon, Scangos says. "We have a number of earlier assets that are in either preclinical or clinical development both in oncology and metabolic disease," he says. "And we have a number of additional collaborations on which we are in discussion. We expect to complete those over the course of this year."
- Chemical & Engineering News
- ISSN 0009-2347
- Copyright © American Chemical Society