Newsmakers: Scinovo

GSK’s new unit may change the R&D outsourcing game

By Gil Y. Roth

The Scinovo group is one of GlaxoSmithKline’s new strategies to boost its pipeline. Focused on non-clinical development or, in GSK’s terminology, preclinical development (PCD), Scinovo is dedicated to working with external partners to accelerate PCD through a combination of GSK’s internal resources and the company’s network of external suppliers.

Photo courtesy of GlaxoSmithKline. Photo by www.tomwhipps.com

I first learned about Scinovo at the annual AAPS meeting when Terry Walsh, one of its members, stopped by our booth to talk about the new organization and its outsourcing perspective. I was intrigued by the notion of a major pharma company developing leverage with contract service providers by bringing emerging companies under its umbrella.

GSK’s chief executive officer Andrew Witty, in an internal note about the new program, remarked, “Scinovo is a unique model of collaboration, established by Preclinical Development, that supports GSK’s growing number of external and internal assets. The Scinovo team is able to couple the excellent preclinical development science that we have in GSK with our global network of preferred research suppliers to forge customized drug development solutions for each of our portfolio partners. Ultimately this will help ensure faster and more effective delivery of important medicines to patients.”

That AAPS conversation led to an interview with Scinovo head Paul Trennery, Ph.D. about the group’s origin, its strategy, how it fits into GSK’s CEDD and DPU structure, and how large pharma is adapting to pipeline pressures and R&D outsourcing. —GYR

---

Contract Pharma: Tell me about the origin and the mission of Scinovo. How does it differ from the company’s past partnering practices and what prompted this new approach?

Paul Trennery: There were three primary drivers behind Scinovo and the preclinical development (PCD) network.

First, there’s the need to share risks in high stakes drug discovery and to capitalize on global knowledge. That’s the realization that we cannot rely on our internal discovery engine alone. Throughout the industry for years, we’ve seen an emphasis on direct in-licensing of mid- to late-stage assets. We also work with multiple external innovator companies, where earlier-stage option-based deals yield risk-sharing opportunities while also increasing GSK’s chance to have broader involvement in new scientific discoveries. So, there are many more external partnerships.

Second there’s the need to ensure that our internal resources focus on added value activity. What are the real capabilities our scientists must focus on to tackle the drug development issues of tomorrow?

Third, we saw that, if everything goes according to plan, our late-stage portfolio would likely double over the next few years, so we had to evaluate the capacity of our internal infrastructure. We recognized that we cannot just build new infrastructure all the time; we must build stronger and long-term strategic partnerships with suppliers.

Scinovo consists of three main groups: the Senior Science Team, Scientist Account Executives, and CRO Management and Operations. The first group has our experts in non-clinical safety, ADME, chemistry and pharmaceutics, the second is a small team handling relationship management responsibilities, and the third covers delivery support.

From our internal statement, here’s a description of what Scinovo does:


CP: What’s the role of the “scientific account executives”? What were the precursors to that role at GSK?

PT: This is a new type of role for preclinical development at GSK and anywhere else in big pharma world, as far as I am aware. I think CROs use a similar model to enhance their relationships with key clients. Their role is to be the interface between GSKs preclinical departments and the partner drug discovery company. They not only build the relationship but bring personal expertise in drug development to the discussions too. Essentially they “represent the preclinical needs of the partner company within GSK,” and can offer options on how to get work done which may save money or time without compromising quality.

Biographical Note Dr. Paul Trennery is currently senior vice president and head of Scinovo at GlaxoSmithKline. In this role he has responsibility for external non-clinical development for GSK’s internal portfolio activities that are delivered through global suppliers (CROs/CMOs) as well as leading a unique model of non-clinical development support to the many external innovator companies that currently partner with GSK’s drug discovery organization. From 2000-2008 Dr. Trennery had responsibility for the non-clinical safety of all GSK R&D drugs in development as SVP Safety Assessment. Prior to this he was Worldwide Head of Medicines Safety Evaluation for GlaxoWellcome and Director of Drug Safety at the Glaxo Research Institute in the U.S., also serving as an adjunct member of the Health Sciences faculty of the University of North Carolina. Dr. Trennery has a Ph.D. in Biochemical Toxicology from the University of Birmingham (UK), and is a Fellow of the Royal College of Pathologists, the Society of Biology and the British Toxicology Society.

The account executive also can bring whoever is the most knowledgeable scientist across our PCD organization to help advise the partner scientists in those areas. Thus, they facilitate our consultancy service, be it in terms of problem solving, design/interpretation of work packages or how best to get work done and where in the world to do it. We have found a number of smaller companies are doing more than they need to get their asset to the next decision point, possibly because they are listening to external consultants who are being more conservative or are perhaps basing their recommendations on less than up to date regulatory positions. Of course, the converse is also true, because we are liaising with regulatory authorities all the time with our internal portfolio, we often will be aware of changing regulatory expectations that the smaller companies are not.

CP: To that end, is there an effort to “teach” early-stage partners about outsourcing practices in general?

PT: Not at all. Many do it already. They may be reticent to use a supplier on a different continent, and we can connect them and if desired help monitor work for them, to address those concerns. In particular, we can sometimes offer them benefits in time, quality or money if we get the work done via our preferred provider network. Suppliers we have approached on this case by case have been happy to agree where the biotech company is in a partnership with GSK probably because:

1. It is work they might not get otherwise,
2. It is likely they would get repeat work from that company if all goes well, and
3. It helps to cement the good relationships between the CRO and GSK where they can expect large volumes of business assuming the partnership is appropriately nurtured on both sides.

One strength is that we have all the PCD CRO/CMO management teams within the Scinovo group now, so non-clinical safety studies, ADME work, chemistry syntheses and drug formulation work are being awarded through a single team. This gives us stronger procurement options when negotiating costs with multi-service suppliers but also, critically, allows us to evolve working partnerships that allow more than one part of the development work stream to be performed with a single provider, where it makes sense to do so and quality is maintained to the level we require.

CP: How does Scinovo fit into GSK’s new structure? In mid-2008, the company announced a review of its therapeutic areas and launched those Discovery Performance Units (DPUs) within its Centers of Excellence for Drug Discovery (CEDDs). What role will you play?

PT: That review was a two-stage process. It involved the recognition that we could go down another level of simplicity. The average CEDD had around 300 scientists. The average DPU has around 60. So that DPU could focus on a particular aspect of a therapeutic area or disease, a particular type of delivery or a particular technology platform that has broad application across our portfolio.

That strategy was combined with the cross R&D evaluation of the areas of research we believe will have the biggest impact on the health of patients around the world. This activity meant we got out of certain types of research and we recognized that there were some others that we should’ve been involved in or that we needed to accelerate. That led to the question of what we should develop in house and where we should partner.

There’s an expectation that each CEDD should work toward having a proportion of their discovery activity via external partnerships. Some are already doing much, while others are working toward it. In addition, our CEEDD (Center of Excellence for External Drug Discovery, established in 2005) is 100% focused on building drug discovery activity via deals with external companies across therapy areas. Over our whole portfolio, we think that, by 2015, about half of our pipeline will come from external innovator companies.

CP: Is that a tacit — or implicit — admission that existing large-scale R&D models just aren’t yielding enough molecules?

PT: Plenty of molecules are being developed, but there’s just too much attrition! These rejections are either safety related or because the molecules profile is not sufficiently differentiated to be worthy of reimbursement in today’s challenging drug pricing world. The need to demonstrate real value to the patient means that many molecules that in the past would have made some small benefit to patients, are now no longer viable to develop.

CP: Scinovo is focused on GSK’s option-based, very early-stage partners. Are you seeing more partnerships coming from that phase, or is the overall volume of partnerships increasing to the point at which it just seems like there’s a surfeit of early-stage partnerships versus late stage?

Photo courtesy of GlaxoSmithKline. Photo by www.tomwhipps.com

PT: It’s a little of both. Obviously, everyone wants to buy into a late-stage Phase III asset which will make a real difference to patients and the company. But as far as the partnerships with our option-based deals, they follow the whole gamut, and are routinely pre-Proof of Concept (PoC). Some are merely platform based so we might be four or five years or more away from demonstrating clinical value with some of these deals, so GSK and the partner share the risks associated with developing those potential medicines.

CP: Were these sorts of deals on the radar 10 years ago?

PT: Definitely not. Our interests then were twofold: there was in-licensing of bigger assets that had substantial value attached to them, or they were part of larger partnerships. We had a deal with a company a number of years ago, in which both sides generated a few candidate leads and the best among them were chosen to develop. So those kinds of deals were going on, but they still involved the presumption that our internal expertise was at least as good as that of the partner.

I think now the emphasis is more the recognition that there’s a lot of great science out there and we’ve got a real chance to capitalize on global expertise by allowing external partner companies to do their thing their way.

The Scinovo option is that we don’t want to leave those partners entirely adrift. We can provide a support system in the area of preclinical development, in which these companies generally have less experience. They may have people who’ve come from large pharma and know what needs to be done, but it’s not generally their area of expertise. They’re probably working at these companies because they’re good at drug discovery, or with a particular technology. Why let them struggle when we can help?

We want to be a magnet for GSK. We want potential partners to say, “If we partner with GSK, we’ll also get this support system and I can see some value in that.” For us, we want our partner to succeed because we have a vested interest — if they succeed, then GSK succeeds because that compound becomes a medicine that patients want. The CEEDD in particular have been great champions of the Scinovo concept. Their proposition in terms of being the ‘partner of choice’ for risk-sharing alliances (www.ceedd.com) includes access to the services provided by Scinovo.

CP: Even with the lower buy-in for option-based deals, the large pharma companies still seem to have a lot of due diligence to engage in with these projects.

PT: Certainly. Due diligence has become a core competency for large pharma, not a hobby.

CP: What do you think has the biggest need for improvement, in terms of preclinical due diligence, both internally and externally?

PT: It’s multifactorial. Attrition on grounds of safety is still a huge problem. Selecting out high risk molecules in terms of safety earlier based on better knowledge of the target biology is a definite opportunity for the industry. We need to couple that to more useful screening tools that are based on known hazards where we can link the assay to a real end-point in lead optimization and candidate selection studies. There are no shortage of assays out there professing to help make selection of safer molecules, but knowing which make most sense and ensuring good decision-making processes are in place between toxicologists and the drug discovery scientists is still extremely important to make ground in this important area. We want to be able to integrate across all our preclinical science disciplines to manage this and other development risks so they are better understood by regulators.

Also, new effective medicines are getting harder to find and develop. There’s great promise with oligonucleotides and other biological molecules, but delivery to the site of action is much more complicated than for small molecules overall. This means that traditional formulation-based drug delivery groups need to be more innovative in preclinical research and build stronger links with DMPK and safety colleagues in progressing these new approaches.

And then regulatory expectations — in terms of clinical efficacy and long-term safety — are changing, resulting in higher costs and longer times to get to NDA. Clinical trials are getting huge, sourcing and paying for comparator drugs — placebo-only trials now being less common — is also an important consideration to factor into trial costs.

CP: That being the case, how soon do Scinovo’s “science account executives” get involved in the process when GSK is talking to a potential partner?

PT: The earlier the better for me, but we recognize that we don’t want to jeopardize a delicate partnering conversation. We leave it to our Business Development colleagues to steer it. But we do often get involved before there’s any signed deal, as part of due diligence or to discuss some facet of the science.

We’ve done a number of deals in 2009. From July onwards, we’ve given advice and input prior to some of those deal-signings. In 2010, we hope we can be helpful to some degree in every deal that’s being contemplated. It is a judgment call though, and the BD people orchestrate the process, so they determine when it’s best for us to come in to the discussions.

We’ve been trying to balance building early credibility with active involvement with partners at the opening stage in the relationship. We are finding that where we helped resolve a problem or were able to show how we could build value into a partnership, our internal business development teams, as well as our portfolio leaders will look to us to engage with the next partner they’re talking to. As I mentioned, the CEEDD in particular have taken this to heart and we are actively involved with all their alliance partners both pre- and post-deal.

Now, for direct in-licensing deals of compounds in later stage development, we know about them from the outset because we handle the due diligence from a PCD perspective. We’re called in to see the dataset for a particular compound, so preclinical expertise often can help identify the necessary work stream we need to tackle if that in-licensing proceeds. Those partners rarely retain any PCD-type responsibility for the asset, so they don’t often require post-deal support from Scinovo.

CP: Let’s talk more about outsourcing. Besides sheer scale, how do you think outsourcing relationships with preclinical service providers differ between large pharma and emerging pharma/biopharma?

PT: Large pharma, while still doing tactical outsourcing based on ‘overspill,’ are also looking to build suppliers into their capacity planning and prepared to undertake longer-term procurement deals. Biotechs and small pharma are still doing tactical outsourcing where cost is the primary driver. Cost matters to big pharma of course, but in the great scheme of things, clinical costs are so large, preclinical work must be efficient and effective, but it is more of a balance with time and quality, both of which are hugely important to us.

CP: Which criteria are most important in building Scinovo’s list of preferred suppliers? I’m sure that providers would like to know what they should be doing more of.

PT: Are they flexible in potential deal structuring for long-term partnership opportunities?

Are they prepared to influence their clients (us) proactively? Most big pharma companies will listen to ways to save money now. CROs and CMOs in general are more efficient than any big pharma operation at the routine work, so what can they teach us?

And, obviously, do they do the basics right? Some suppliers get caught between trying to be the development partner of choice vs. actually doing solid work fast with good quality and on-time delivery.

CP: What changes have you observed in the preclinical outsourcing market in recent years?

PT: We’ve seen interest in building stronger, long-term business relationships. Preclinical providers are recognizing that security in business comes with flexibility in pricing and operating approaches. There’ve been various deals with that in mind. Also, we’ve seen their desire to grow their offerings either in depth — e.g. toxicology suppliers offering more animal disease models, or problem-solving skill sets — or breadth — e.g. chemistry based CROs expanding into the ADME and toxicology arena. However, in the last year the financial situation — coupled with reorganizations within large pharma — has led to leaner times for suppliers in the nonclinical development arena. There’s more competition now and I suspect that could drive consolidation in the area. And lastly, everyone will be aware of the growth in recent years of low cost providers.

CP: Earlier, you mentioned clinical formulation as one of PCD’s area of expertise. How far along the formulation chain can you go?

Photo courtesy of GlaxoSmithKline. Photo by www.tomwhipps.com

PT: Most of our formulation development work is done in house at this time although more recently we have begun to work with external companies to access their particular expertise in formulations where we need to bio-enhance our molecule to ensure adequate human or animal exposure.

We tend to apply our extensive in-house expertise to assess the biopharmaceutics of molecules quite early on in development during pre-candidate selection, so we can achieve the desired preclinical and clinical exposure without having to go around the houses developing multiple formulations and wasting time and money before we get this right.

This expertise includes the application of our ADAPT (Accelerating Drug and Product Technologies) process, which couples various predictive technologies and decision making to select the optimum clinical formulation, while minimizing API requirements. The formulations we select for early development are chosen such that the PK/PD targets have the greatest chance of being met and are also chosen to be as simple as possible but with the commercial needs in mind. This overcomes some of the reasons why in the past formulators have had to produce relatively complicated formulations to overcome API variability. The other advantage of taking a simpler formulation through to commercialization is that Quality by Design (QbD) can much more easily be demonstrated with savings of time and costs in late stage development.

We are routinely applying ADAPT to internal GSK assets and it is a process which we can apply to external projects as required to achieve similar benefits.

CP: Speaking of which, you mentioned that Scinovo also helps GSK make in-house/outsource decisions on internal development projects. What criteria guide your recommendations on those?

PT: It depends on the discipline. With safety assessment, for example (which is my background), we historically have kept everything internal up to first-in-man, and then we outsource everything else in terms of routine studies, but retain problem-solving work. There’s a strategic line in the sand there. We know that once the asset is in the clinic, we’re going to outsource much of the rest of that compound’s development safety package. So it is easier to schedule that work — three six-month tox studies, etc. as it is not dependent on internal workflows.

In other areas, outsourcing is more contingent on capacity overspill. That’s a bit more difficult to manage, given the vagaries of portfolios and drug development timelines.

Large pharma is on shifting sands. There’s an evolving recognition that we’ve got to only use our infrastructure for the added value activities. There are some things for which we may have internal capacity but we don’t ultimately want to be doing that work stream internally because we want our scientists applying their knowledge to something else. Where it makes sense, let’s free up our internal scientists to work on things that are not necessarily available externally, but are still critical to delivering the next generation of successful medicines in a more cost effective way than the past.

Like many other companies, we’re working in the oligo-nucleotide arena. Everyone recognizes the difficulty in delivery to the site of action. That’s a big challenge to both our formulation scientists and our safety and drug metabolism scientists. Those groups need to work together to get these complex bio-molecules to the site of action. That’s something we are developing internally right now because much is not easily available on the outside. To facilitate that, we may need to outsource more of our conventional formulation work. But it is all work in progress.

CP: Paralleling the large pharma practice of outsourcing the manufacture of later-stage, commoditized products, rather than newer, more valuable drugs?

PT: Yes, that sounds right. Our position on outsourcing has four interlocking factors.

1. The expectations and capacity of our portfolio at any point in time.
2. There’s the infrastructure shift and the links to our Global Manufacturing and Supply organization. At some point in the development continuum, we will be transferring drugs to our manufacturing organization. Can we judge, based on their capacity and our early portfolio needs, how many pilot plants we need year on year? If less, how are we going to phase them out? If we need more, how are we going to source them?
3. There’s the evolution of our suppliers. A number of them have broadened their service offerings. Take WuXi AppTec, for example: they’re building a large capability in toxicology and yet their original expertise was in chemistry. They’re looking for companies to give them a molecule at the beginning, and they’ll pop it out at the end (almost; they don’t have a clinical organization yet). Are we going to place entire development packages with suppliers?
4. And there’s adding value internally. How do we want to use our own people? It’s not all about outsourcing so you can let valued internal staff go. Of course, that’s a big worry for people. If they don’t have adaptable skills, it might be difficult to change. We try to encourage our people to look to the future and ensure that they’re getting the training and learning in those areas where they’ll continue to add value to what GSK needs.

CP: Scinovo’s been around for nine months. What success have you had, and what adjustments have you had to make?

PT: As far as successes go, we’ve consulted with a partner to optimize a targeted development strategy which included PK/PD aspects, formulation approaches (including modeling) and assessment of API synthetic route. We have also saved some partners substantial sums of money in their toxicology programs at external suppliers by facilitating a preferential arrangement in our supplier network. In addition, we’ve enabled INDs to be more smoothly submitted by addressing CMC issues that would otherwise be a delaying factor with the agencies. Importantly, we’ve been partly instrumental in a partner company signing with GSK versus another pharma because of the strong preclinical capabilities we could bring to the development of their assets and the Scinovo operating model. We’ve been able to do some work for a company in-house (at a cost) using an assay not generally available at CROs. This gave them data to allow a decision to be made between two possible candidates. So, there are quite a few areas where we believe we have made a positive contribution to the development of a partnership and/or a drug candidate.

In terms of adjustments, well, the main thing so far is that we’ve had to be selective in what we decide to do! At present we adhere to three self imposed “rules”:

1. We do not make a profit. Thus, if we do work for a partner in-house we will charge at cost; we will not lose money but we are not crafting a CRO business model. In addition, we are not trying to do a lot of routine work in-house either; for that we want to recommend our preferred supplier network. Our value to GSK is when we can help a compound solve a preclinical problem that would otherwise cause a promising medicine to stop progression, or can accelerate the timeline to a key decision point. If we do that, our costs are easily covered.
2. We will only provide consultancy and delivery of work support to an asset that is part of a deal with GSK. Thus, there is the possibility that the compound will become a GSK drug at some time in the future. We’ve been asked by some companies to help them with issues on assets they may have in their portfolio but which GSK is not interested in, for whatever reason. Thus far, we have declined getting involved there but will continue helping them on the assets which are within the potential deal with GSK.
3. The third rule is that the above rules will probably change! We’re ensuring our operating model remains flexible enough to adapt if it makes good sense for GSK in the future. And I think the one thing we can be sure of is that the Scinovo operating model will evolve. Scinovo is part of the bigger PCD organization, which in turn is a key part of R&D, so we want to ensure the success of preclinical development and R&D as a whole; Scinovo is an instrument we can flex and apply to help drive toward that greater mission.

CP: In that case, where do you think Scinovo will be by 2015?

PT: For me, the excitement is that we can go in a number of directions with Scinovo. We did wonder if we should create Scinovo “pods” elsewhere. At the moment, we have centers in Philadelphia and the UK. That may be an evolutionary model for us. It may make sense for us to be more closely associated geographically with our partner companies, understanding their environment and challenges. Conversely or in parallel, perhaps the model will be taken up by other departments within GSK. We can tell a partner, “Scinovo is a non-clinical support model, but we can get you a person here who knows a lot about, imaging, genetics, pharmacovigilance, the design of clinical trials etc.” Partners have asked us to handle Phase I trials for them.

CP: And?

PT: We could go that way in time. The scientists currently in my team don’t undertake any clinical studies, but we have an excellent clinical development organization and it is an option to evolve the model to embrace larger capabilities and expertise, either earlier or later in the development process. Where is it going to go? We’re having a lot of conversations within R&D about the directions we can take. But we know we need to walk before we can run.

Gil Y. Roth has been the editor of Contract Pharma since its debut in 1999.