better convergence at its own peril. Even under so-called “ideal” drug development conditions where all R&D is in-house and the potential for convergence and overlap is the greatest, only a minority of all drug developers have ever been involved in a commercial product launch; it’s still rare to see a deeply embedded, true commercial presence on drug development teams.
This is why the search and develop (S&D) methodology is such a challenge. We have taken what would have been ideal conditions for solving the applicability problem and traded them for conditions that solve a number of other important problems instead. This is a high level problem to have, but we have it and it’s not going away anytime soon.
Is this a solvable problem?
How do we incorporate metrics and endpoints for applicability that might help increase our odds of success? In essence, an applicable solution will be one in which data needed for eventual commercial success has been built into the earliest stages of clinical development. This tightly woven development/commercial fabric produces an asset which, in addition to having the strongest possible data to show regulators, is also ready-made for a firm’s strategy and underlying competitive advantages. This solution gets used in the right patients, is happily reimbursed by local, regional or national payers, and is eventually integrated into increasingly standardized clinical care pathways.
Let’s take, as a base case, the notion that the externalized, lean, capital efficient, asset-centric model (like Annovation Biopharma) is an optimal R&D structure for advancing early-stage science to clinical proof of concept. How would solving the applicability problem impact that model? Remember, what makes this an issue is that the owners of the asset are asking how one can get a drug to proof of concept quickly and efficiently, and what the optimal amount of capital invested is to maximize returns to their investors and limited partners; the partner company is asking how it can source, select, and acquire new solutions that fit its existing strategy.
• We need to add cost, and risk. Conducting extensive healthcare systems analysis, generating true comparative effectiveness data, and integrating large clinical outcomes data into development can add upwards of 50 percent to the cost of any clinical trial. As costs escalate, so does risk, particularly when venture firms are trying to hit that sweet spot of a 4x or greater return on invested capital. Despite the extra costs, and risks, we need to do this. The issue with the build to buy model is that both cost and risk are being born exclusively by the acquirer.
• We need to make the data matter, to both parties. Lets assume we augment our early-stage data package to provide support for a project’s commercial viability. What’s the point of these data? To matter to the acquirer, they need to be part of the metrics around proof of concept. For our VC partners, we will need to define phrases that trigger later milestones and earn-outs (or clawbacks) such as ‘reasonable commercial effort’ with more granularity and specificity. Adding to the development program to get the data is the easy part; making the data matter to each party is more of a challenge.
In my previous post, I said that some sort of equilibrium is needed between the Annovation development model and the much more capital intensive and commercial success-enabling one needed by big (and medium) pharma companies. Taking that a step further: would anyone be prepared to add a go/no-go decision to the buyout option in these deals based on commercial viability? I am not talking about early, pre-funding pass/fail decisions based on small, qualitative market research that VC firms already conduct during due diligence. I am talking about decisions to proceed with an acquisition, or forfeit the option, based on data that show that, although a drug is safe and effective, its intended use and commercial role is unclear, and therefore would not lead to any meaningful reimbursement. This is exactly the kind of decision that pharma makes every day and it’s the primary reason why good assets end up being jettisoned for “strategic reasons”—the entire basis (for what it’s worth) of The Medicines Co.’s original business plan in 1996. As I noted, this is not a new problem.
Where does the dust settle?
We have a high level problem. On one hand, we have managed to make real progress against the age-old biopharma R&D problem without having to resort to platitudes like “biotech is better because it’s more efficient.” The S&D methodology coupled with the financial discipline, governance, and hands-on operational excellence provided by the major early-stage life science VC’s has resulted in a veritable treasure chest of de-risked assets for biopharma companies to find. This is very good for every participant in the life sciences circle: VC’s return good money to their LP’s, biopharma gets the sweet, chocolaty center, and most wonderfully we are boosting the number of potentially innovative new drugs for patients more quickly and efficiently than ever before.
But—all is not good with this model. The professionals who regulate, reimburse, and prescribe new drugs are dissatisfied and have been for some time. They continue to ask the same, often simple questions about new drugs and all too often, they are still not getting the answers they need: How does this new drug compare to existing therapies in real world settings? Which patients should ideally get the drug? Are there patients in whom this new drug is not appropriate? How do we best to use the drug to get maximum value? Why should we ask our citizens to pay for this drug? Is this drug priced for the value it brings? How did your trials show that value?
We can answer these questions, but it requires drug developers to have a very good understanding of the jobs that users are trying to do and that they integrate this knowledge into their clinical development programs. This has long been an industry weakness, and build-to-buy deals are exposing it—and widening the divide between innovative new technology and much-needed solutions.
