Consortium Sandbox: Building and Sharing Resources

See allHide authors and affiliations

Science Translational Medicine  25 Jun 2014:
Vol. 6, Issue 242, pp. 242cm6
DOI: 10.1126/scitranslmed.3009024


Some common challenges of biomedical product translation—scientific, regulatory, adoption, and reimbursement—can best be addressed by the broad sharing of resources or tools. But, such aids remain undeveloped because the undertaking requires expertise from multiple research sectors as well as validation across organizations. Biomedical resource development can benefit from directed consortia—a partnership framework that provides neutral and temporary collaborative environments for several, oftentimes competing, organizations and leverages the aggregated intellect and resources of stakeholders so as to create versatile solutions. By analyzing 369 biomedical research consortia, we tracked consortia growth around the world and gained insight into how this partnership model advances biomedical research. Our analyses suggest that research-by-consortium provides benefit to biomedical science, but the model needs further optimization before it can be fully integrated into the biomedical research pipeline.

Entrepreneurial teams focused on turning fundamental biomedical research into a commercial product fuel the translational research engine. Inventions aligned with a clinical need face scientific, regulatory, reimbursement, and adoption hurdles, many of which are shared across research and development sectors. However, the creation of tools that address these common challenges requires knowledge, research materials, technical expertise, infrastructure, and financial resources that extend beyond the reach of research and business plans of any single organization. In addition, tools that address shared needs require validation and adoption across multiple organizations if they are to be widely used as intended. A consortium is one collaboration framework that has been increasingly leveraged to meet shared challenges in biomedical R&D.

Here, we report our global findings on the motivations for consortia formation, output and trends in research-by-consortium, and some of the common challenges that these partnerships are beginning to experience. By highlighting this collaboration model, we hope to drive improved efficiencies into the biomedical research ecosystem and encourage the scientific community to explore ways of leveraging the diverse output emerging from consortia.

Consortia are neutral environments that use collective expertise and distributed management to create research tools that are designed and validated by scientific consensus. By temporarily putting aside their institutional differences, these collaborations aim to accelerate individual research efforts by building broadly accessible standardized resources (1, 2). Thus, researchers who do not directly participate in a consortium’s scientific efforts can still benefit from the partnership’s output.

For example, researchers lack robust methods that universally predict the safety of new vaccines. Such a platform can drive vaccine development forward, but its creation requires the coordinated efforts of multiple stakeholders, including biopharmaceutical and regulatory scientists who represent the end-users, experts in assay development, and academic scientists who conduct basic research on biological pathways, biomarkers, and prediction methodologies. This type of platform technology must be validated across multiple diseases and existing vaccines before it could be used in clinical trials to predict the safety of a de novo vaccine. This unmet need is the strategic focus of “Biomarkers for Enhanced Vaccine Immunosafety” (table S1; URLs are provided for all consortia mentioned), a 5-year Innovative Medicines Initiative (IMI) consortium engaged in developing panels of immune-response biomarkers that can predict the safety and efficacy of any vaccine. The consortium coordinates the scientific research needed to design, develop, and validate the technology, leveraging the resources and expertise from three large pharmaceutical companies, 13 nonprofit research institutions, and three small- to-medium-sized businesses. The preeminent outcome of this consortium would be improved public health. But even if the project does not meet all of its goals, the data generated by the consortium research would still be available to the broader scientific community and may inform alternative approaches.

This multistakeholder consortium is just one data point that marks the rapid emergence of biomedical research consortia within the past decade—evidence that research-by-consortium is fast becoming a permanent component of the R&D process (35). Our analysis of 369 consortia—with distinct objectives, disease focuses, research agendas, participants, and work streams—revealed that there is no single model for convening a productive consortium. However, one can glean common features that provide a design framework for emerging consortia.


Because the term consortium has multiple meanings, we applied generalized inclusion and exclusion criteria to simplify our analysis and broadly classified the scientific objectives of consortia into four categories: fundamental or basic research, tools, biomarkers (6), and specific products (supplementary methods). Then, in order to discern consortia motivations and identify global and domestic trends, we further categorized consortia by the sector that initiated the partnership. These groups were responsible for developing and advancing the strategic research agenda and promoting the scientific objectives to secure additional thought-leadership, participants, and resources. Broad characteristics of these groups and sectors along with a generalized list of their expectations when initiating consortia are shown in table S2.

To select our consortia study pool, we evaluated information from publicly available Web sites, press releases, publications, and presentations. We also conducted phone-based interviews of a select group of consortia in order to decipher their goals and motivations, operational frameworks, accomplishments, and challenges. In total, 369 consortia met our criteria and provided enough detailed information to be included in our analysis (supplementary methods).

Rise in research-by-consortium. There was a steady rise in the number of new biomedical research consortia, peaking in 2012 with 63 (Fig. 1A). However, only three continents—North America, Europe, and Asia—were found to have biomedical research consortia (Fig. 1B). All regions experienced increases over time, but none was more apparent than the rise in Europe between 2007 and 2012. This spike is mostly a result of the support for new consortia from the European Seventh Framework Programme (FP7), a multination funding program that aims to stimulate and support networked research within this continent. Of the 183 total biomedical research consortia (61%) launched in the European Union between 2007 and 2013, 111 were supported with FP7 funding, which includes those associated with IMI. Using broad categories—academia, government, health care organizations, industry, nonprofit foundations, and third-party organizations (table S2)—we assessed the types of organizations responsible for initiating consortia and found that government agencies were responsible for initiating the majority of the 369 consortia cataloged, followed by third-party organizations and industry (Fig. 2A). Data on trends by year (Fig. 2B) suggested that industry has increased initiation of its own consortia, with two-thirds of industry-initiated consortia launched between 2007 and 2013 supported by IMI.

Fig. 1 Number of consortia launched.

By year, 1995 to 2013. (A) All consortia. (B) By continent.

Fig. 2

Consortia: Who and why? A total of 369 consortia were studied. (A) Sectors that initiated a consortium. (B) Growth of consortia by top three initiating sectors (government, third-party, and industry). (C) Sectors intended to benefit by consortium activities (n = 369), by initiating sector. (D) Growth of consortia focused on advancing device research. The color key applies to (A), (B), and (C). For consortia launched in Europe and North America, n = 208 and 144, respectively.


Of all biomedical research consortia launched in Europe, ~58% operated under strategic research plans that were defined by a government agency. This level of government involvement is considerable compared with that of North America (24%); in Asia, government agencies initiated six of nine consortia. Scientific needs were not the only driver of collaboration; 36% of all government-initiated consortia had secondary objectives of economic development, and ~20% of the 369 consortia sought to advance regulatory science through the participation of a regulatory official in the consortium’s research or on an oversight committee (7).

Mechanisms of collaboration. The models used to structure a consortium reflect the requirements of their sponsors, complexity of the science being pursued, resources available, type of output expected, and methods for sharing output with the public. Despite these differences, we found some common features among the collaboration models used by the 369 consortia studied. The majority of consortia had clear mission statements outlining the targeted deliverables and timeframe; milestone-driven work streams that describe how each sponsor and participant contributes to the collaboration; formalized governance structures that define decision-making roles and opportunities for providing input into the strategic direction; mechanisms of transparency and accountability to sustain an environment of trust; and mechanisms for broader access to the consortium output.

The majority of consortia depend on their sponsors to provide not only financial support but also their expertise and resources in advancing the goals of the collaboration. The latter is often provided as part of an in-kind contribution. One popular model pools finances to support infrastructure and overarching project management but depends on a sponsor’s research team to execute the scientific strategy in collaboration with synergistic teams from other sponsoring organizations. For example, TransCelerate BioPharma has a small, dedicated staff and depends on in-kind contributions from its member companies to provide technical project management and laboratory resources to advance its research objectives. Another model spends its pooled finances to attract outside expertise into the collaboration—often by funding researchers from the academic and small- to medium-sized business communities. The IMI and the Québec Consortium for Drug Discovery (CQDM) use this mechanism, and CQDM’s model is unusual in that financing provided by its industry and government partners supports research projects that are performed entirely by the academic and small-business communities; industry scientists are assigned as mentors to each of the consortium projects as part of the sponsor’s in-kind contribution.

The majority of products created by a consortium are often deployed to the broader community, but access depends on the type of output, intellectual property, and licensing agreements. Consortia used a broad spectrum of mechanisms to govern access to output, ranging from immediate release to the public to exclusive access limited only to consortium members. Intangible products such as biomarkers, methods, and data standards were often described in publications, presentations, white papers, and guides. Access to other tools such as technology platforms, data sets, and biospecimens often was determined by intellectual property agreements. Some consortia, such as the Biomarkers Consortium, required that the inventions created from their projects become available to the community via a nonexclusive licensing opportunity. As an incentive for its funding sponsors, CQDM required that its funded inventors first grant industry sponsors an opportunity for a nonexclusive license to their technology.

Consortia outcomes. The majority of consortia (45%) intended to improve drug development, and of these, most were initiated by government (31%) and then industry (27%) (Fig. 2C). IMI managed ~59% of all industry-initiated consortia within the drug development category. About 21% of consortia focused on output that would broadly advance multiple research sectors, and once again, government was responsible for initiating a majority of these consortia (54%), followed by third-party organizations (15%). One example of a cross-sector resource is the International Mouse Phenotyping Consortium’s effort to produce standardized knockout mouse models that are annotated with genotypic and phenotypic data. These tools are essential for advancing basic research as well as preclinical development projects, and this consortium intends to have both the mice and associated data widely available to the research community.

About 15% of all consortia focused on the advancement of basic research, with government (49%) and academia (19%) initiating the majority of these types of collaborations. Consortia focused on accelerating the development of devices (11%) were primarily initiated by government agencies (68%), followed by industry and third-party organizations (each at 10%). However, the number of consortia focused on accelerating device research has begun to rise (Fig. 2D), with 75% of these partnerships operating in Europe.

Improvement of operational efficiencies of the health care sector was the goal of ~5% of consortia. One example is the Clinical Decision Support Consortium, in which a diverse group of health care providers and technology developers work together to build tools that can integrate various types of knowledge, clinical data, and medical best practices within a learning electronic health system. The remaining 2% of consortia aim to improve the development of both drugs and devices. One example is the Innovation in Medical Evidence Development and Surveillance (IMEDS) Program, which aims to create resources and regulatory tools to monitor new drugs and devices after obtaining regulatory approval.

From the horizontal perspective of the initiating sector, government (n = 160) was most active in developing consortia to accelerate drug development (32%), followed by consortia whose outcome would benefit a broad set of stakeholders (27%). This trend in desired outcomes was similar across most sectors. The majority of consortia initiated by industry (n = 63), third-party organizations (n = 70), academia (n = 36), and patient foundations and nonprofit organizations (n = 33) focused on accelerating drug development (71, 54, 39, and 58%, respectively), followed by consortia designed to broadly advance the entire research community (13, 17, 31, and 27%, respectively).

Consortia output. Regardless of initiating sector, a majority of consortia had as their mission to create tools that can be used by the entire research community, such as procedures for biospecimen handling, methods for clinical trials, predictive methods for designing safer drugs, and collective research resources (combinatorial molecular libraries, stem cells repositories) (Fig. 3A). Of the 176 consortia focused on tool development, ~29% concentrated on enabling the sharing of data among organizations via data standards and technology frameworks. Disease-based foundations initiated ~9% of all consortia, and most of these (41%) placed strong emphasis on direct product development (such as a specific diagnostic test), with 34% having a secondary objective to develop tools with broader research uses.

Fig. 3 Initiators and outputs.

(A) Intended products of consortia, by initiating sector. (B) Sectors that initiate consortia, by intended product.


Biomarker development was the second popular area of emphasis for most consortia, except for those initiated by disease-based foundations. Fitting with the broad-use requirements for the majority of consortium products, 72% of all biomarker-focused consortia aimed to develop diagnostic indications, with 59% focused on genetics-based biomarkers (8). For example, one effort from the Accelerating Medicines Partnership of the Foundation for the National Institutes of Health (FNIH) is focused on identifying genomic and genetic risk factors for type 2 diabetes. Aggregated and annotated data on 100,000 to 150,000 individuals will be shared with the broader research community.

Government agencies played a substantial role in initiating consortia across all output categories (Fig. 3B). A comparison of output from worldwide government-initiated consortia showed (i) primary interests in developing broadly used tools (83, 56, and 46% of all government-initiated consortia within Asia, North America, and Europe, respectively); (ii) secondary interests in developing biomarkers (31 and 23% of all government-initiated consortia within North America and Europe, respectively); (iii) secondary and tertiary interests in advancing fundamental science [for government agencies in Asia (one of six), North America (13%), and Europe (20%), respectively]. With the exception of Europe (12%), no government agencies initiated consortia focused on developing a specific product.

Disease focus. More than 50% of consortia were focused on creating a capability that could be used in the treatment of a specific medical condition, with oncology and rare diseases serving as the top two interests (n = 46 and 38, respectively) (9, 10), followed by Alzheimer’s disease (n = 22) and diabetes (n = 20) (Table1 and Fig. 4). For cancer, 39% of consortia focused on the development of biomarkers, such as those that can stratify patient populations into subgroups defined by their therapeutic response (for example, the Prostate Cancer Molecular Medicine project, a Netherlands-based consortium) (11). A second focus (37%) was on developing broadly used tools, such as those that enhance data sharing or evaluate new clinical trial methods (such as I SPY2, which, guided by imaging and molecular biomarkers, is testing a phase 2 adaptive clinical trial methodology for a targeted cancer drug plus standard chemotherapy) (12). The majority of oncology-focused consortia were initiated in Europe (n = 27), with government agencies responsible for starting 52% of oncology consortia in three continents studied (Fig. 4A). This group includes the MAMMOTH Consortium, which is managed by the Center for Translational Molecular Medicine, a Dutch public-private partnership. Third-party organizations initiated ~20% of oncology-focused consortia. One example is Worldwide Innovative Networking (WIN), a personalized-medicine consortium that uses collective resources from government, industry, and academia to advance research on cancer biomarkers and early-stage clinical trial methods.

Table 1. Disease-focused consortia.

Products and locations for the top four diseases. Categorization of rare diseases was based on information in publicly available databases (24, 25). One caveat with this broad categorization is that therapeutic areas such as oncology, Alzheimer’s disease, and diabetes represent an aggregate of multiple subtypes, many of which may be rare diseases in themselves.

[CREDIT (background): Science collaborations map/computed by O. H. Beauchesne, using date from Scopus and Science-Metrix]
Fig. 4 Focus by disease.

For the top four diseases (AD, oncology, diabetes, and rare diseases). (A) Disease-focused consortia by initiating sector. (B) Sector interests in initiating disease-focused consortia.


About 35% of rare disease–focused consortia aimed to create tools for their respective disease-research communities. For example, the International Consortium for Human Phenotype Terminologies is an effort to standardize the vocabulary used to describe the phenotypes of patients with rare genetic diseases so that harmonized data sets can be combined or cross-analyzed (13). Other rare-disease consortia are focused on direct product development (mostly diagnostic assays with some on therapeutics), such as the Accelerated Research Collaboration, a consortium of academic researchers who are guided by the Myelin Repair Foundation and industry experts to develop research tools and therapeutics for multiple sclerosis. In fact, foundations and other nonprofit entities initiated ~33% of all rare disease–focused consortia, with governments following at 25% (Fig. 4A). One example of a multigovernment group is the International Rare Disease Research Consortium, which aims to coordinate the development of diagnostics and therapeutics by sharing data and biological samples with the research community and harmonizing ethics and regulatory processes for product testing.

Nearly half of Alzheimer’s disease (AD) consortia were initiated in North America, and 37 and 36% of these aim to accelerate biomarker research and develop broadly usable tools, respectively. One tool-developing consortium, the Coalition for Accelerating Standards and Therapies (CFAST), joins several third-party organizations with the NIH and the U.S. Food and Drug Administration to develop disease-specific guidance documents that standardize data collection and reporting methods for products that require regulatory evaluation; AD is one of their 11 therapeutic areas of interest. Biomarker research was focused primarily on discovering those that can diagnose AD stages and track the disease’s progress, such as the Alzheimer’s Disease Neuroimaging Initiative (ADNI). The landscape of AD consortia represented an almost even distribution of initiating sectors, with government (27%) spurring a slightly larger share than most others (Fig. 4A).

Half of all diabetes consortia sought to develop biomarkers, and most of these focused on diagnostic indications for type 2 diabetes. For example, IMI’s DIRECT consortium aims to discover biomarkers that can identify subtypes by risk of progression. Nearly 20% of diabetes consortia were concentrated on advancing fundamental diabetes research, such as the Beta Cell Biology Consortium, which aims to provide shared resources for studies on the development and function of pancreatic islet cells. Half of the diabetes-focused consortia were initiated by a government agency, followed by industry (20%) (Fig. 4A).

We also analyzed the disease-driven interests from the perspective of initiators (Fig. 4B). Government groups, which initiated 57 consortia for the top 4 disease areas, placed a high emphasis on oncology (42%) followed by rare diseases (30%). Third-party organizations (32 consortia) also focused primarily on rare diseases (44%) and oncology (34%). Foundations and nonprofits initiated 29 consortia with a primary focus on rare diseases (69%), followed evenly by oncology and AD (14%). With only 20 disease-focused consortia initiated by industry, their interest was spread nearly evenly across the four top diseases (Fig. 4B).


The scientific community has much to gain even if it’s not directly involved in a consortium because the output has a value proposition that differs from output emerging from other closed collaborations (1, 5). If all of a consortium’s milestones are met, the public is delivered a rigorous tool that was designed for widespread use by a “superstar” team of researchers and validated across several use-cases (table S3). However, the true value and impact that a tool creates depends on an expanded user base: the larger research community. This dispersal provides opportunities for additional validation and adoption into practice.

Of the 369 consortia that we researched, the majority were focused on creating resources that advance research for a specific medical condition, and most research being performed by AD, diabetes, and cancer consortia was focused on biomarker development. This goal is aligned with current trends in the pharmaceutical industry, which is focused increasingly on molecularly targeted therapies. However, balancing proprietary strategy with risk reduction and shared expertise becomes a challenge for consortia (14). Most have decided that specific druggable targets are proprietary, but diagnostic biomarkers that can stratify patients into subgroups are considered to be precompetitive and broadly usable information (1517). For example, biomarkers validated through a consortium such as I SPY2 also must be recognized by a regulatory body for their indication, becoming potential gold-standard resources that can be leveraged by drug developers as well as the basic oncology, diagnostic, and imaging communities. A related challenge facing a consortium is the need to maintain a collaborative environment of trust, assuring participants that their resources are not being used to provide a special advantage to one of their competitors. A consortium does this by creating a “precompetitive” collaborative space, in which the shared need is the focal point of the partnership and is surrounded by borders that provide an assurance that the tool or resource is useful to the broader research community, including those not directly involved in the partnership. This motivation for collaborating to benefit the wider community also helps to address antitrust concerns.

The consortium provides a safe harbor with guarantees of transparency, using legal agreements that make sure each participant receives equitable benefits and access to the final product, with opportunities to monitor and adjust the projects through a formal governance process. Neutral organizations and government agencies that receive no direct benefit from the output have taken the majority of responsibility in identifying and validating shared needs, advancing consortia established by others, and creating the precompetitive space for collaboration (2). Many independent nongovernment organizations, such as the Critical Path Institute (C-Path), FNIH, and Reagan-Udall Foundation, were created to assist a U.S. government agency and thus have similar interests in secondary outcomes relating to societal benefits (table S2)—which renders these organizations accountable to the broader scientific and patient communities.

Trends. Consortia focused on advancing drug development most likely will continue to dominate the landscape, with government and neutral third-party organizations playing an important role in identifying and managing the precompetitive space among all researchers. The process of drug discovery is becoming increasingly reliant on collaborations (1, 4, 7, 18, 19), and the consortium framework offers a way to maximize individual resources, leverage those acquired and developed by others, and create scientific tools designed and validated for wide use, often with the guidance from a regulatory body. Most consortia are in the early stages of demonstrating the proof of principle in their collaboration structure. However, the steady increase in drug-development consortia initiated by industry, such as TransCelerate BioPharma, is one indicator that competing research organizations are not waiting for the proof of principal and are adopting this partnership model in order to solve unmet needs identified by their colleagues.

Nearly 18% of the 369 consortia analyzed focused on sharing and aggregating specific data sets or on developing tools that can enhance data-sharing capabilities. As research and clinical data become increasingly recorded as bytes and bits, and the complexities of biomedical research continue to drive collaborations, a consortium makes an ideal platform to create solutions that researchers need to keep pace with the growing volume and need to share data.

Consortia focused on accelerating device research remain in the minority. This is most likely because the industry is composed of a smaller number of large legacy companies as compared with that of the pharmaceutical industry (20). Devices also differ from drug development because historically, their development life cycle and investment periods are shorter, and the sector faces a shorter regulatory time frame. However, the medical device sector is increasingly faced with technically complex applications, including assisting the delivery of drugs and cell-based therapeutic products, and regulatory bodies are beginning to explore their oversight roles (21). In concert, there has been a slow increase in consortia focused on medical devices, which includes diagnostic instruments as well as those that assist in medical interventions (such as surgical tools and robotic technologies) (Fig. 2D) (22).

Even if a product is destined for the public good, a consortium relies on participants who are incentivized to contribute because they have strategies to use the output to advance their own “competitive” research. Many of the consortia that we interviewed were beginning to develop assessment frameworks that can be used to objectively evaluate their model of collaboration in the context of a participant’s anticipated return on investment. However, return on investment is based on the adoption of a consortium’s output and its impact on an organization. The utility of these evaluations would not be limited to the consortium and its participants; it could also provide increased transparency as well as assist a consortium’s management in the assessment and optimization of its own operations.

A clear need remains for rigorous evaluative criteria that provide metrics and associated data, as evidence in the development of best practices. However, the consortium framework is still in development; thus, it is too early to develop a best-practices approach. Yet even without a core set of best operational practices, the true value of the consortium framework can be demonstrated by the level of adoption of its created tools and resources and by metrics that track improved research performance and efficiencies. There are not only opportunities for individual researchers to leverage consortium-created tools; consortia must continue to find ways to collaborate with each other. A few have done so by leveraging their individual expertise, such as CFAST and partnerships between IMI and C-Path (to coordinate tuberculosis drug-discovery efforts) (23) and Parkinson’s Progression Markers Initiative and ADNI (to find the similarities and differences between their diseases of interest) (24), as well as the Predictive Safety Testing Consortium’s sharing of data with the Biomarker Consortium’s effort to detect drug-induced kidney injury in the clinical trial setting (25).

Opportunities for advancing biomedical research through consortia are endless. But the number of researchers, constricting resources, and perception that there is scientific overlap have contributed to the hesitations felt by many sponsors in their commitment to and participation in some consortia. These variables are just some of the root causes leading to an emerging feeling of “consortium fatigue” (3). Although our research uncovered few consortia with overlapping missions, there were opportunities for collaborations between consortia. A map of biomedical research consortia could drive efficiencies into the ecosystem by helping to reduce overlap and identify synergies for cross-consortium collaboration.

The translational biomedical research community is complex. Stakeholders come from industry, academics, patient and caretaker groups, clinical medicine, payers, and government. Compared with most R&D sectors, the life cycle for biomedical research is considerably longer, faces continuous unpredictable scientific challenges, and carries a higher risk of failure. Consortia represent valuable frameworks that allow stakeholders to mitigate risk, coordinate expertise, leverage resources, and use safeguards to address reproducibility (26). A consortium is not the only partnership framework; but unlike those that benefit one or a few partners (27), consortia are driven by shared needs across multiple organizations and a coordinated strategy to create high-value, widely accessible tools in a streamlined and open manner.

The data presented here aim to provide a semiquantitative snapshot of the consortium landscape in an attempt to direct the attention of the research, clinical, patient, and policy communities. By highlighting how these temporary collaborations advance sets of research outcomes, we hope to provide clarity on their value proposition to the research ecosystem. Collaborations are knotty endeavors, and integrating the right partners is not easy. Nonetheless, collaboration is becoming increasingly essential for translating medical innovations to patients, and the consortium is one way of raising the tide for all boats in the biomedical research waters.


Supplementary Methods

Table S1. Consortium Web sites

Table S2. Sectors or groups that initiate consortia through the development of strategic research agendas

Table S3. Examples of consortium output

References (2831)

References and Citations

  1. Competing interests: The author is a nonpaid member of the scientific advisory board for the CQDM.
View Abstract

Stay Connected to Science Translational Medicine

Navigate This Article