GRIMALDI: There have been just a handful of innovative collaborations. For instance, Genentech and Roche and Theravance had an interesting transaction structure. But most deals follow a preset formula, with a fairly standard list of deal terms. If you are looking for a specific product, you'll do a collaboration. But if you want to have that "R" keynote [as in R&D] to it, it makes sense to start with a product collaboration, and perhaps later acquire the whole company.

Now, there have been some bold R-oriented acquisitions that have gone counter to that point. Think about Sirna. Or that a lot of the antibody acquisitions were quite early stage and certainly focused on technology.

PERKINS: Across the board, younger biotechs are willing to share the risk for a larger piece of the return. With our relationship with AstraZeneca [on MedImmune], we both put in 50-50 and we have a 50-50 worldwide profit share. There are a lot of deal structures like that, where a young company is no longer willing to take a milestone, but because of access to capital, is willing to invest. I see that continuing.

RYAN: Biotechs want to take products to the beginning of Phase III to maximize value, and pharma wants those products at that point in time. But what happened—and Pfizer is one of the more prolific in terms of these kinds of deals—is that almost every one of those programs went backward two or three years because what biotech called Phase III was not Phase III in pharma's mind.

SCOTT: The decision to partner and the interaction between our company and the external company is straightforward. The challenges of getting a deal done and projects successfully integrated are often internal. Companies that get better at this will really succeed over the next five to 10 years. Companies should be able to shift from an internal project to internalizing an external project as seamlessly as a hybrid car shifts from gas to electric without the driver noticing.

GARTIN: The problem is that there are unequal standards, with pride of authorship for internal projects. At the end of the day, an acquisition is simply a different investment in R&D. Companies look at external projects with a sharper eye as to whether they will be more or less successful than the projects they have invested in for years.

CLINTON: Looking ahead, what changes do you predict to pharma's model?

DE ROSEN: We will see a more diversified world, where the most successful companies won't do everything. Look for the model where companies can access discovery and gain development input and support for its marketing efforts in partnership with different institutions.

FARINO: The industry has used the same R&D model for 40 years. This model must change to one where we begin with a much deeper understanding of the underlying pathiophysiology of disease to gain earlier confidence in mechanism of action and safety. We believe such developments will allow for much earlier "live licensing" of compounds into the marketplace with much more focused claims. That can impact how you value products and companies, and how deals get done.

FUNTLEYDER: Those who can balance risk and reward, or at least risk, will succeed. It's about innovation per unit of risk—or however you measure it. And it doesn't matter about cap size. Either you'll succeed, or you'll disappear.

CLINTON: So should Big Pharma get out of research?

FRANK: It's very interesting to look at the relevant strengths of biotech and pharma. There could be a new model, where pharmas don't do "R"—just "D," marketing, and manufacturing—and research dollars are used to acquire products.

HOFSTAETTER: It doesn't work like that.

FRANK: The argument used is if we don't have "R," we can't have "D."