The Rise And Rise Of Research Consortia

“The name is selected to refer to energy amplification,” said Jan M. Lundberg, Eli Lilly & Co.’s head of R&D, referring to the acronym of a new drug development consortium in which his company participates. “AMP, for Accelerating Medicines Partnership.”

Lundberg was boosting the new group, which comprises 10 major drug and biotech companies and as many patient advocacy groups, while making his round of meetings with journalists at last month’s BIO International convention in San Diego. With the support of the Food & Drug Administration, the National Institutes of Health, and partners in academic research, AMP aims to accelerate the development of diagnostic tools and treatments for unmet medical needs through nothing short of a research transformation.

Members will combine forces to validate drug targets in Alzheimer’s disease, type 2 diabetes, lupus, and rheumatoid arthritis. They will pool data and make them accessible to the biomedical research community at large.

Although drug company partnerships and data pooling have become well-established practices in recent years, AMP is being hailed as a breakthrough, given the number of major drug companies that are partners and the involvement of government agencies. NIH is funding more than half of the $230 million project and running it as well. The consortium is being put forward as a new model for research, one that recognizes that important breakthroughs will require broad, trans-sector collaboration.

AMP is one of several consortia that have debuted over the past year. These groups, most of which are public-private partnerships, have purviews as varied as clinical trial design, antibiotics discovery, and lung cancer. But all share two key features: collaboration between competing drug companies and the compilation and sharing of data. And in coming together, they have the chance to learn from the shortcomings of past consortia and from each other’s failures.

Ernst & Young cites AMP and other new groups in its latest annual report on the pharmaceutical industry. “There have been partnerships in the past, but we would argue that things have changed,” says Gautam Jaggi, an Ernst & Young consultant. “The new consortia are different kinds of partnerships responding to different pressures.”

A key pressure, Jaggi says, is resource constraint. “The bigger companies are over the patent cliff,” he says. Blockbuster drugs are no longer cash cows, “and that means that companies are in an environment where they just don’t have the sources of cash flow they’re used to.”

Members of research consortia acknowledge that money drives them to collaborate. But they also point to chronic failure in the drug-testing clinic stemming from an incomplete knowledge of the biology of disease and an inability to ascertain drug targets. The realization that no one research entity, no matter how well resourced, can identify all the dots that need to be connected has created an environment for collaboration and data sharing.

Anna Barker, president of the newly convened National Biomarker Development Alliance at Arizona State University, says the emergence of the new groups marks a distinct trend in research. But a lot of groundwork on data-sharing has already been completed, she notes. Barker, who was deputy director of the National Cancer Institute in the mid-2000s, says NCI’s Cancer Genome Atlas, a project launched in 2005 to catalog the genetic mutations responsible for cancer, helped set the stage for industry collaboration on data gathering.

She also cites Sage Bionetworks, a database company started by two Merck & Co. researchers, as a pioneer in collaborative data initiatives. Sage, which was launched in 2009 using computers and data that Merck got with its 2001 acquisition of Rosetta Inpharmatics, hosts research data contributed by major drug companies as an open resource. The firm also will manage the data compiled in the new AMP consortium.

AMP’s path forward won’t be smooth, Barker warns. At NCI she learned that once researchers agree to combine data, new problems emerge. “Data quality is extremely important, and not too much thought has been given to it until the last couple of years,” she says. “As we move into the era of big data, it is going to become more and more important that you know the derivation of the data, that you have metadata that gives you the provenance.”

Industry’s concern about protecting intellectual property presents another obstacle in managing research collaborations, Barker says.

The biomarker alliance hopes to accelerate research by establishing two databases. One will collect published data on anything that has been put forward as a biomarker, documenting the provenance of the data. The second will track work on biomarkers from early discovery to clinical validation. The data, which will be open to researchers, will improve understanding of how to successfully advance biomarkers through each stage of development, Barker says.

Michel Goldman, executive director of the Innovative Medicines Initiative, an effort launched in 2008 by drug companies and the European Commission to foster research collaborations in Europe, points to the IMI-sponsored European Gram-Negative Antibacterial Engine, or ENABLE, as yet another new consortium. Goldman sees the flurry of collaborative groups this year as a validation of the concept of collaborative research.

“It is clear that industry realizes that it not only needs to partner with academia and other outside organizations,” he says, “but that companies also need to work together because challenges are so big.” Alz­heimer’s disease has been a major area of frustration, he says, noting that the $10 billion spent researching it has not produced a single significant treatment.

It is also true that the industry cannot point to a significant new drug developed in a consortium, Goldman concedes. But the success of collaborations so far can be measured in other ways. In schizophrenia, for example, data from 10,000 patients compiled by five drug firms working in an IMI-supported program showed that the number of patients enrolled in efficacy trials can be reduced by 40%. The duration of observation in trials can be reduced as well.

Goldman agrees with Barker on lessons learned. Managing intellectual property concerns takes time, “but with the help of honest brokers, it is possible to reach agreement where the interest of each party is maintained,” he says. As for pooling data, “it’s not just a question of clarity; it’s a question of interoperability. It is a real problem, and consortia are investing a lot of effort into this.”

Several major drug companies have a track record in collaborative research, in groups such as the National Center for Advancing Translational Sciences, an NIH-sponsored effort to reduce the time and cost of drug development, and TransCelerate, a nonprofit industry organization working to improve clinical trial efficiency. But according to Mikael Dolsten, Pfizer’s head of R&D, the AMP consortium is bringing collaboration to a higher level.

“For me, it has been the most comprehensive effort to take on major scientific and medical challenges and to get a better understanding of diseases.”

Dolsten argues that a “life sciences ecosystem” has emerged during the past five years as a result of interaction between researchers from disparate organizations. “Science has moved at an extreme pace to generate very large data sets to explore,” he says. “Companies need to work together and categorize them.”

Describing AMP, Dolsten uses the metaphor of digital maps leading to GPS technology. “We see that revolution in science moving medicine in the same direction, creating the same thing. But you can only do that with a high level of detail and sophistication,” he says.

For most consortia, map work is a matter of finding the most direct route in a drug development process replete with duplicated efforts, repeated mistakes, and wrong turns.

Sage Bionetworks has been on this road for five years, steadily expanding as a data analysis commons. For example, in 2011, with funding from Roche and Takeda Pharmaceutical, it launched a neuroscience effort called the CommonMind Consortium.

Lara Mangravite, director of Sage’s systems biology research group, says data generation and analysis are well-established practices within pharmaceutical research consortia. But opening access to pooled data is still a new frontier.

“What we don’t know is what will happen when we make the data public,” she says. “Will people download the data and work on their own, or will they come together and collaborate?”

But most of the momentum in collaborative research will be forward, Mangravite says, as the drug research enterprise comes to grips with having run into a wall on a spate of intractable diseases. There is little question that the future will be marked by increased sharing of data, a theme that held sway at the BIO convention.

NIH Director Francis S. Collins told attendees in San Diego that industry, academia, and patient advocacy groups need to come together in public-private partnerships to parse drug target data and pool research results in hopes of stemming the rising tide of clinical failures. “We have moved beyond the idea of the competitive nature of targets in drug discovery,” he said.