Peter Mansell examines the impact of heavier in-licensing/out-licensing traffic on pharma forecasting.
More focused R&D strategies that prune out dead wood or high-risk therapy areas are leaving pipeline gaps that make management and investors nervous.
The patent cliff is swallowing dependable revenue streams that can no longer be replaced with a new generation of big-hitters, as emphasis shifts to targeted therapies and niche indications.
With access to capital thinning out and development programs becoming more complex, protracted and demanding, there is growing pressure, too, on biotechnology companies to find partners for viable compounds while they are still afloat.
And academic researchers are looking to translational research and relationships with an industry hungry to counteract squeezes on public funding.
These factors combined add up to heavier in-licensing/out-licensing traffic in the life sciences—and all of them inevitably have an impact on pharmaceutical forecasting, at both ends of the equation.
Not everyone is as bullish as the analysts at Morgan Stanley who last year suggested that pharmaceutical companies could triple their returns by exiting early-stage drug development and reinvesting in outsourcing. (For more on outsourcing, see Will big pharma become a collection of marketing and distribution firms?)
In fact, the industry appears to have become more circumspect about what it will buy into, albeit not averse to paying a premium for real value.
High prices for some early-stage biotech assets, expensive failures in the latter stages of clinical development and the effects of global recession have all encouraged closer scrutiny of pipeline opportunities, whether internal and external.
Writing in Innovations in Pharmaceutical Technology last summer, Tibor Papp, head of corporate advisory at PharmaVentures Ltd, highlighted a fall of over one third in collaborative deals for pharmaceutical assets between 2005 and 2009.
What was licensed in, though, could be handsomely rewarded.
In 2000, Papp noted, less than 10 percent of all licensing deals with published financials promised more than $100 million in upfront and milestone payments.
By 2009, the proportion of these high-vale agreements had jumped to over 70 percent.
Even with the number of deals contracting, around 40 percent of pharmaceutical pipeline candidates were still coming from external sources, while a comparable proportion of sales in the marketplace were driven by in-licensed or acquired technologies.
More recently, PharmaVentures has highlighted a trend towards recovery.
While collaborative R&D deals continued to decline in volume and value during 2010, licensing activity was on the rise compared with 2009, even if the average value (excluding royalties) of licensing deals with disclosed financial information fell markedly from the high seen in that year.
“The company that’s selling the asset obviously wants to get top dollar for it,” Talbot-Watt notes. For the company licensing in, though, the priority is to “pay as little as possible.”
As such, companies weighing up in-licensing opportunities will want to take a fresh look at the potential asset rather than extrapolating from any value attached to it by the vendor.
“In my experience, they start afresh,” comments Carla Peek, head of business modeling and analytics at Inpharmation.
“They commission their own models. We’ve never been presented with, ‘Here’s a model from the company we’re licensing from. How can we improve it?’”
For Talbot-Watt, this is probably a “bit of game play.”
Given the inevitable conflict of agendas, if a company can say it has used an external vendor, “it lends a little bit more credibility to their valuation.”
There are also practical reasons for sticking with what you know, Peek observes. Often different companies license different forecasting software.
“They go with what they know and they present their results in the way that they always have presented them, because they have their standard templates and standard charts, and that is what management is used to.”
What the external forecast really comes down to, Peek observes, is fleshing out and corroborating a business strategy that will already be well advanced in house.
“The main thing is that they want to see this is going to be possible, it’s going to be a good opportunity,” she notes.
“And hopefully, by the time they are commissioning forecasts and market research, they already have a good sense that this is a good therapy area to go into and to plan a strategy around. They’re looking for a forecast that can validate what they’ve said.”
The level of information a client will want on a licensing opportunity will likely depend on the client’s internal proficiency and resources as well as where the product sits in the clinical development process.
So it may range from epidemiology and disease characteristics to a full-blown business case.
If a product is in Phase II trials, for example, there will generally be details of the clinical program and the therapeutic target, Talbot-Watt explains.
Black Swan will then work primarily on patient sizing and the physician management flow, to pin down where the product would fit in the marketplace.
Alternatively, the brief may be for comprehensive modeling, with first and second to market, uptake, benchmark analogues, etc.
“Other companies will want a full-blown business case,” Talbot-Watt adds.
“And part of that will be the epidemiology and the forecast, and another part could be SWOT analysis and things like that.”
New products closest to market are more likely to be me-toos, Peek observes, so those forecasts “may be driven by order of entry, because that’s going to be the biggest determinant of your market share.”
The process also calls for quite a lot of qualitative market research.
“If it’s a therapy area they’re not used to, they need to define the market size, treatment paradigms, adoption rates, the number of competitors,” Peek notes.
“It might be a huge market but it might be saturated.”
With in-house products, she adds, companies will spend much more money on quantitative market research, such as a lot of conjoint studies to define the target product profile and determine clinical endpoints.
“But with in-licensing you hope that’s already been defined,” she says.
“It’s competitive and some of their shareholders might be getting impatient because there are gaps in the pipeline,” Peek comments.
“So if they’re looking for opportunities, it’s usually a very rapid forecast.”
It also has to be affordable since, when making those decisions, executives don’t want to spend a lot of money, either.
“They can’t justify spending hundreds of thousands on huge conjoint analyses,” Peek observes.
With an in-house asset, the company is likely to have planned out its timelines at the beginning of the year and set objectives.
As Talbot-Watt points out, forecasting for an in-licensing opportunity must be not only time-sensitive but sufficiently robust to form the basis of an investment decision.
With out-licensing, the process tends to be little more extended, as “it’s their asset and they know what they can do with it.”
Essentially, forecasters should be applying the same degree of rigor to licensing opportunities as they would to an in-house product.
They may have to be much more explicit with the sources and references they use, though, “because often if they’re publishing the deal, they want data they can actually stand on,” Talbot-Watt comments.
Forecasters may also need to do a little bit more of the legwork in terms of disease pathology for in-licensing, as they will not have access to medical staff in the originating company.
One marked difference, Talbot-Watt notes, is that an in- or out-licensed product will require input around deal structures and financing rather than just internal components such as R&D investment.
“They have to structure some sort of deal,” she says.
“And depending on the nature of the deal, they’ll probably want to put some sort of financial analysis into the models.”
The way these deals are structured—with their upfront payments, milestones for approval filings and/or product launches, and further payments or royalties tied to preset sales projections—is becoming increasingly complex.
This reflects a trend both toward product originators taking more of a stake in their discoveries and toward pharmaceutical companies wanting to spread more of the risks of drug development.
Whereas companies used to buy assets outright, there now seems to be more willingness to partner, some of it driven by biotechs or universities wanting more of a working relationship.
“It’s their intellectual property and they mostly know how to develop it,” says Talbot-Watt.
“They just need the cash and a little bit of commercial direction.”
“Previously forecasters didn’t need to incorporate NPVs [net present values] into their forecasts,” Peek comments. “Now it’s becoming the norm.”
These forecasters “were certainly doing scenario modeling but sensitivity analysis, risk analysis—that wasn’t the norm. Now most of our forecasts are going out with NPVs and with sensitivity analyses, risk analyses, feasibility.”
In the near future, Peek predicts, it could be that pricing and reimbursement strategies will enter the equation, which is at the moment a different department.
But she believes it is coming to the stage where these two are going to have to start contracting.
One further trend that may affect forecasting requirements for in- or out-licensed products in years to come is that these assets are being traded much later in the development cycle.
A lot of companies, Talbot-Watt says, are on the rebound from “investing in very early-stage assets that never got anywhere. We’re seeing things that are either in Phase III or they’ve already completed their Phase III and are going through registration.”
This could be “a bit of a poker game,” she speculates, one reflected in the speed of analysis now required in product valuations: “They leave it until a pivotal trial has reported favorably and then they jump in. But then, of course, it’s a dash to the finish line.”