Although some pharmas are cutting back on internal neuroscience R&D programs, the number of precompetitive consortia backed by industry has continued to rise, with the latest two launching within the last two months. Most initiatives are focused on developing better disease models and discovering new biomarkers-the challenge going forward will be to push the boundaries of the precompetitive space to advance the discovery of new disease-modifying targets and mechanisms.

At least three pharmaceutical companies have publicly disclosed major cuts or reorganizations of their neuroscience R&D programs in the last two years. In February 2010, GlaxoSmithKline plc announced plans to eliminate discovery research in pain and depression. In March 2010, AstraZeneca plc ceased research in schizophrenia, bipolar disorder, depression and anxiety.1

Last December, Novartis AG said it would close its Basel CNS R&D facility, disband its Neuroscience Disease Area and shift to new directions based on the genetics of neurological diseases. This February, AstraZeneca said it is shrinking its neuroscience R&D group to 40-50 people focused exclusively on outside opportunities.2

In contrast to reductions within individual companies, there are at least eight new or ongoing precompetitive neuroscience collaborations with substantial industry participation (see "Selected ongoing precompetitive neuroscience industry collaborations").

According to Paul Chapman, general manager of Takeda Pharmaceutical Co. Ltd.'s Pharmaceutical Research Division, "Competitive advantage in industry comes mostly from making better judgments based on all available biological data and using it to create better molecules. Everything else can and should be shared."

Last month, Takeda joined two precompetitive initiatives-the Structural Genomics Consortium and the newly formed CommonMind Consortium.

Precompetitive initiatives have aims that vary widely-from developing better mouse models of disease to obtaining regulators' approval of enrollment criteria for clinical trials. Pharmaceutical partners and consortium leaders told SciBX that, ultimately, these programs all share the common goal of reducing the high failure rate and cost of drug development in neuroscience.

They also said the biggest obstacles to expanding the scope of precompetitive research include concerns over IP and demonstrating that collaborations can provide a worthwhile return on investment.

"It is a zero-sum game. Everyone is short on resources, and you have to feel confident that precompetitive investments are worth it," said Husseini Manji, global therapeutic area head of neuroscience at Janssen R&D.

Janssen is a unit of Johnson & Johnson.

Last December, Janssen committed $3 million to One Mind for Research, a program sponsored by the International Mental Health Research Organization that fosters precompetitive neuroscience research, education and awareness-building programs.

Diagnosing the problem

Companies involved in precompetitive neuroscience initiatives told SciBX the space is an area particularly in need of precompetitive collaboration.

Peter Andersen, SVP of external scientific relations and patents at
H. Lundbeck A/S, said some companies have shied away from neuroscience R&D in part because fundamental aspects of the biology of neurological diseases are not understood.

"The underlying problem in brain research is that the biology is not sufficiently mature. CNS diseases are defined based on symptoms relative to normal behavior, whereas you normally would develop drugs based on a specific molecular target you want to interact with. This leads to a large risk of failure when you ultimately test compounds in patients," he said.

Chapman agreed. "Neuroscience is particularly fruitful because the system is so complicated; the diseases are so poorly understood. In spite of some great science and some genuine progress in the field, compared to diabetes, cardiovascular disease, even oncology, neuroscience is, frankly, way behind."

Manji emphasized it is not only the lack of understanding but also the need for new drugs that is driving precompetitive collaboration in the area. "Disorders like Alzheimer's disease are going to bankrupt the healthcare system, and many companies are pulling away because it's just too complex," he said.

"Neuroscience is still an area with a very large unmet need, and there is a lot of opportunity for industry," said Elisabetta Vaudano, principal scientific manager at the EU's Innovative Medicines Initiative (IMI). "However, there has been a lot of failure-the success rate in neuroscience is among the lowest of therapeutic areas. There are bottlenecks which industry has not been able to tackle, and thus it is an area very ripe for public-private partnerships like the IMI."

Indeed, IMI has been a leader in organizing precompetitive collaborations in the neuroscience space. Since 2009, IMI has put together 4 consortia focused on the topic, each with more than 20 members. Combined, the consortia have committed funding of €90 million ($117 million).

Observing the patient

Whereas newer initiatives led by IMI are beginning to broaden the definition of precompetitive collaboration to include preclinical discovery and tool building, the most tangible results over the last decade came from pooled efforts by companies to share data obtained from observational clinical trials, particularly in neurodegenerative diseases.

In 2004, after about three years of discussion with the pharmaceutical industry, the Alzheimer's Disease Neuroimaging Initiative (ADNI) was established to monitor disease progression in 800 normal older individuals, people with mild cognitive impairment and patients with mild Alzheimer's disease (AD) over 5 years. It has since expanded to enroll more than 1,000 patients.

ADNI has $130 million committed thus far through a combination of support from the National Institute on Aging (NIA) and 28 current partners coordinated by the Foundation for the National Institute of Health (FNIH), including 25 from industry.

Neil Buckholtz, chief of the dementias of aging branch of the Division of Neuroscience at the NIA, told SciBX, "Even more critical than the funding has been the interaction of industry scientists with academic scientists, through the private partner scientific board as well as through participation in individual cores. ADNI is managed through a series of cores-biomarker discovery, MRI, PET, etc. People from industry sit in on the discussion from the cores, and they don't manage the study, but the interaction is critical for success."

Data from ADNI are made publicly available through a database accessible to researchers who register, and Buckholtz said that from the beginning it was clear this would be an IP-free initiative. "Some of the companies we spoke to in 2002-2003 were thinking of mounting this kind of study, but none could do it on their own, to this scale," he said.

Among the first direct outcomes from the ADNI is the optimization of the use of cerebrospinal fluid b-amyloid (Ab) levels as a selection criterion for enrolling patients at risk for developing AD.

The Biomarkers Consortium, another precompetitive collaboration sponsored by the FNIH, also is using samples obtained from ADNI to perform proteomics studies to identify biomarkers linked to AD progression.

Following the ADNI model, in 2010 The Michael J. Fox Foundation for Parkinson's Research launched its Parkinson's Progression Markers Initiative, a multicenter observational trial of 400 newly diagnosed patients with Parkinson's disease (PD) and 200 healthy controls. The initiative is funded by the foundation and at least 10 industry partners.

Separate from these initiatives, the Critical Path Institute's Coalition Against Major Diseases (CAMD) is analyzing clinical trial observations with an eye toward regulatory validation of biomarkers and disease models.

Diane Stephenson, associate director of the CAMD, told SciBX that CAMD aligns with other public-private partnerships such as ADNI and other discovery initiatives yet is distinct because its aim is to work closely with the FDA and the EMA to achieve qualification of new regulatory tools that can be used to speed drug development.

CAMD was launched in 2008 and currently has partnerships with about 20 not-for-profit and industry partners to accelerate the development of therapies for AD and PD.

One of CAMD's signature achievements thus far is the creation of a unified clinical trial database. The database contains results from 6,000 AD patients, comprising placebo arms from 22 clinical trials contributed by 9 companies.

"The database is available to qualified researchers around the world and was originally set up to build a quantitative disease drug trial model for AD, a project aimed at informing clinical trial designs and optimizing the chances of success in the future clinical trials by learning from the past," said Stephenson.

The Biomarkers Consortium has a similar ongoing project to examine industry clinical trial placebo data focused on pre-AD patients with mild cognitive impairment.

Stephenson said CAMD is now trying to get companies to share data about clinical biomarkers.

"I would say in all honesty the biggest challenge is that some companies define the precompetitive landscape differently than others," she said. "Many companies will fund studies on biomarker performance, but because these studies may pose risks in a competitive landscape, they may be reluctant to share the data. We are hopeful that, in the future, companies will provide more than just placebo data, but it's a fine line we are continuing to expand."

She added: "Attitudes are changing quickly. Working together across diverse stakeholders to advance our understanding of AD is significantly easier than it has been in the past."

Group therapy

The next step in precompetitive neuroscience research will be to share in the creation of tools and knowledge that will enable the discovery of new therapies.

Janssen's Manji and Lundbeck's Andersen said industry's involvement in precompetitive collaboration is now expanding to include the development of biomarkers, disease models and even discovery research.

However, they said that IP is one of the biggest concerns as collaborations move into the grey area between the competitive and precompetitive space.

"There is room for both IP and IP-free precompetitive collaboration; both are important," said Manji.

An example of a completely IP-free effort is the CommonMind Consortium, announced last month by Takeda and Sage Bionetworks. The partners and collaborating academic institutions plan to obtain genomic information from neuropsychiatric patients and share it in an open-access database.

"The academic partners who founded this consortium saw the need to generate and analyze large-scale genomic data from human subjects with neuropsychiatric disease, which is exactly the type of effort that is best served by precompetitive collaboration especially as the data must, and will, be made broadly available to the public," said Takeda's Chapman.

Andersen cautioned that when IP is brought into the picture, it must be clearly negotiated up front. He told SciBX that Lundbeck had previously encountered problems with IP when participating in a collaboration with a biotech and two academic teams that was organized by Top Institute Pharma, a not-for-profit that manages academic-industry partnerships.

"We very quickly realized the drawback to this setup, which ultimately necessitated we buy out the other company involved in the collaboration after a year and a half because the rules around how to share IP were too complicated," he said.

In addition to his role at Lundbeck, Andersen is chair of the research directors group of the European Federation of Pharmaceutical Industries and Associations, which coordinates the solicitation of IMI research proposals. He favors the IMI approach to IP. "Everything you bring into the collaboration you own; everything you invent together you share; everything you do afterwards is your own," he said.

He said any problems with IP were eliminated after IMI made a dedicated IP lawyer available for consultation to adapt these rules to the needs of individual participants.

IMI is putting this flexible approach to the test with its four consortia focused on neuroscience, in which a broad range of potential projects are captured within each initiative.

For example, NEWMEDS, led by Lundbeck, has nine separately organized "work packages" that each have a distinct focus. While one work package is pooling clinical trial data to identify genetic factors underlying schizophrenia, another is developing animal models of cognitive dysfunction that relate to clinical endpoints.

PharmaCog, led by GSK, is seeking to develop better preclinical and clinical models of cognitive function. Jill Richardson, who leads the effort and is director of external alliance and development at GSK, told SciBX, "GSK is involved in several precompetitive neuroscience initiatives, but, to date, the one we have invested in most is the IMI PharmaCog project."

She added: "Joining up with external companies and organizations is as much about developing robust, harmonized methods, models and tools, networking, and sharing ideas as it is about intellectual property or access to drugs. So we don't view IP as a barrier to participation."

The other two IMI projects are EUROPAIN, which is focused on understanding new mechanisms involved in chronic pain, and EU-AIMS, announced in March, which is building new models for autism spectrum disorder (ASD).3

In addition to these government-backed efforts, disease foundations are beginning to consider ways to encourage precompetitive collaborations. For example, the Myelin Repair Foundation has been funding academic collaborations over the last 7-8 years and is now soliciting interest from industry to form a precompetitive consortium modeled on that effort.

Jay Tung, VP of drug discovery at the Myelin Repair Foundation, told SciBX the foundation has had discussions with at least four industry organizations about setting up a precompetitive consortium to identify biomarkers and imaging tools for myelin repair. They hope to launch an initiative within a year.

"We are an organization that represents patients and can serve as a neutral convener, and our interest is in identifying the tools necessary to run a clinical trial for myelin repair," said Tung. "We are in the process of surveying pharma and biotech to ask what they would want addressed in this type of consortium, and at this point we have gone back to them two or three times and are continuing to hold discussions. We don't want partners who just want to attend and listen; we want actively participating partners."

Manji told SciBX he would eventually like to see precompetitive collaborations extend to target validation. "Many companies are investigating the same targets, and if we're doing it anyway, we should do it together. We'd have to share the spoils, but we could do more things together, and a quarter of the pie is better than no pie at all."

But he added, "We are trying to be pragmatic, and that type of collaboration is going to be much more complicated." 

Cain, C. SciBX 5(19); doi:10.1038/scibx.2012.483
Published online May 10, 2012


1.   Ward, M. BioCentury 18(12), A6-A7; March 15, 2010

2.   Hansen, S. BioCentury 20(60), A1-A4; Feb. 6, 2012

3.   Osherovich, L. SciBX 5(16); doi:10.1038/scibx.2012.406


      AstraZeneca plc (LSE:AZN; NYSE:AZN), London, U.K.

      The Biomarkers Consortium, Bethesda, Md.

      CommonMind Consortium, Seattle, Wash.

      Critical Path Institute, Tucson, Ariz.

      European Federation of Pharmaceutical Industries and Associations, Brussels, Belgium

      European Medicines Agency, London, U.K.

      Food and Drug Administration, Silver Spring, Md.

      Foundation for the National Institutes of Health, Bethesda, Md.

      GlaxoSmithKline plc (LSE:GSK; NYSE:GSK), London, U.K.

      H. Lundbeck A/S (CSE:LUN), Copenhagen, Denmark

      Innovative Medicines Initiative, Brussels, Belgium

      International Mental Health Research Organization, Rutherford, Calif.

      Johnson & Johnson (NYSE:JNJ), New Brunswick, N.J.

      The Michael J. Fox Foundation for Parkinson's Research, New York, N.Y.

      Myelin Repair Foundation, Saratoga, Calif.

      National Institute on Aging, Bethesda, Md.

      Novartis AG (NYSE:NVS; SIX:NOVN), Basel, Switzerland

      One Mind for Research, Rutherford, Calif.

      Sage Bionetworks, Seattle, Wash.

      Structural Genomics Consortium, Oxford, U.K.

      Takeda Pharmaceutical Co. Ltd. (Tokyo:4502), Osaka, Japan

      Top Institute Pharma, Leiden, the Netherlands