Don’t underestimate those brainy types riveted to their laptops at coffee shops in Harvard Square, because they could be managing the development of a potential breakthrough drug. That’s what I was thinking last week during a dinner hosted by Xconomy at a Harvard Square venue, where we discussed the virtues and limitations of virtual business models in biotech.
Bob told our guests—consisting of biotech executives, venture capitalists, attorneys, and consultants—that their comments would be kept off the record (with one notable exception within the group). So I’ll present some of the key ideas from the dinner without identifying the sources. However, I will say that our out-of-town guest, Duane Roth, shared an interesting critique of the problems with traditional product development in life sciences and how a type of virtual business model is at least part of the solution.
Roth, who is the CEO of the nonprofit industry group CONNECT in La Jolla, CA, ignited debate and thoughtful comments with his prescription for the ills of current R&D models. Many of the thoughts he shared are explained in greater detail in a paper he co-wrote for the Kauffman Foundation, which made the paper available online on January 25. He said that fully integrated companies with R&D, manufacturing, and marketing capabilities have fallen short on delivering enough new products to justify their large budgets. For example, Roth’s paper notes that FDA drug approvals have declined as R&D budgets skyrocketed. To back this, the paper cites data from the merchant bank Burrill & Company that show that in 2004 the industry spent $47.8 billion in research while the agency approved 36 drugs, while in 2008 research spending had swelled to $65.2 billion, but only 24 drugs were approved.
Roth argued that one solution is the creation of “product definition companies,” which are largely virtual firms that license technology from research institutions funded by the government. Such companies would have only a small group of full-time experts to manage the business, and they would rely on a network of professional services firms such as contract research outfits to handle the R&D and clinical work required. The companies would essentially operate in “the cloud,” he said. And rather than trying to take their drugs into expensive late-stage clinical development, the product definition company would sell a product to, say, a pharmaceutical company, which would then advance the product to commercialization.
“The new model is about moving products, not companies, up the innovation ladder,” Roth said.
His concept quickly stirred debate. A venture capitalist at the dinner said that the fatal flaw in “product definition company” concept is that it wouldn’t work for certain fields such as RNA-interference, a method of gene silencing, which is too complex to be fully understood and developed by a network of contractors and consultants. In such cases, the VC said, you need to build a company with a team of scientists who can perform the research and develop an expertise in a bold new field. “RNAi and recombinant technology couldn’t be figured out in the cloud,” the VC said. However, he acknowledged that virtual business models such as the product definition company model are effective when the underlying science of an experimental product is well understood, and contractors can be hired to perform specific experiments to prove or disprove whether the product has a shot at gaining FDA approval.
We had several biotech CEOs at the dinner who are actively following virtual business models, and two of them were focused on developing specific molecules for an illness rather than trying to advance a platform technology with many applications in medicine. All these virtual firms have the potential to generate high returns for their investors, without requiring nearly as many employees or as much capital as some platform companies.
In the interest of protecting their identities, I’ll pull out a solid example from our recent beat coverage of a biotech with what could be called a virtual business model: AesRx. Stephen Seiler, a veteran biotech executive, formed AesRx in 2008 with a molecule he had licensed for treating sickle cell disease. When I spoke to Seiler late last year, he and his vice president of drug development were the only full-time employees, and the rest of their labor consisted of a group of advisors and consultants. Though Seiler didn’t describe his firm to me as virtual, the company could be called a virtual biotech based on the definition that evolved during Xconomy’s dinner.
At the other end of the spectrum is Cambridge, MA-based Alnylam Pharmaceuticals (NASDAQ:ALNY), which has built a substantial R&D organization to develop RNA-interference drugs. Alnylam, which had 175 employees as of September 2009, is doing research to advance the field, or platform, of small interfering RNA treatments that can turn off disease-causing genes. While the company works with outside researchers to help develop aspects of RNAi therapies, the firm likely needs lots of its own scientists to build internal expertise to help drive development of its drugs.
Still, it’s clear from the dinner that there are many degrees of virtual business models in biotech. And that’s probably the way it should be for an industry in which there is a high degree of variability in the products and science under development at different firms.