CommentaryDrug Discovery

A Call for Sharing: Adapting Pharmaceutical Research to New Realities

See allHide authors and affiliations

Science Translational Medicine  02 Dec 2009:
Vol. 1, Issue 9, pp. 9cm8
DOI: 10.1126/scitranslmed.3000155

Abstract

From the dawn of time, the sharing of knowledge has been one of the main forces driving science and innovation. Yet in recent decades, a proprietary culture, which wrongly posits that all intellectual property must be restricted, has spread across the pharmaceutical industry and threatens to stall the engine that has given us so many valuable treatments. This paper argues that pharmaceutical companies, together with universities and government agencies, stand to gain much from reversing that trend and engaging in widespread collaboration early in the research process to expand foundational knowledge and create a shared infrastructure to tap it.

The first scientific journals in the late 17th century transformed the practice of science, which until then had often been a secretive occupation shrouded in mystery, and ushered in a culture of sharing that made it easier for scientists to build on each other’s contributions (Fig. 1). By speeding the compounding of knowledge, this, in turn, hastened the pace of scientific discovery and spearheaded the technological progress that spawned our advanced industrial society. Yet in recent times, the reward and recognition lavished on successful individual scientists have somehow become an apparent deterrent to sharing. This occurs at a time when, paradoxically, the available tools, with the Internet being most notable, make such sharing more productive and sensible than ever. In the pharmaceutical industry, which has struggled to keep up the pace of its innovation, sharing could be the key that allows companies to access the vast creative, intellectual, and technological resources required to tackle the formidable challenge of turning the riches of the genome into a treasure trove of new treatments.

Fig. 1.

Egypt’s “Fourth Great Pyramid.” The ancient Library of Alexandria was humanity’s first attempt at sharing knowledge on a massive scale. Founded in the third century B.C.E., it grew to house nearly 700,000 scrolls, attracting scholars from all over the Mediterranean. For seven centuries, it proudly stood as a symbol of knowledge towering over humanity’s secular pursuits. Sadly, it succumbed to civil war and neglect at about 400 C.E.

CREDIT: O. VON CORVEN

A SMARTER WAY TO DO DRUG R&D

Sir Isaac Newton credited his accomplishments to standing “on the shoulders of giants.” Modern-day scientists do the same with each other. Thus, the movement that was born in the 1660s with publication of the Philosophical Transactions of the Royal Society has blossomed into thousands of journals that vie for scientists’ manuscripts. Sharing is a concept that resonates with researchers because it has been an engine of scientific progress through much of the modern age.

Sharing also fosters cross-pollination, an essential driver of creativity. Over a thousand new scientific papers enrich the life science literature every day. Turning that knowledge into novel insights, hypotheses, or ideas for treatments is an endeavor that is broader than any single drug company or research laboratory. Pharmaceutical firms seldom have all the resources needed to approach a disease by the many avenues that can yield viable therapies. Take type 1 juvenile diabetes, for instance. It can potentially be treated with insulin, immunosuppressive drugs, vaccines, monoclonal antibodies, stem cells, tissue implants, gene therapy, allo- or xenotransplants, or a combination thereof. This combinatorial approach requires vast competencies across many fields that are unlikely to be found under one roof. Sharing addresses this challenge by bringing together people with complementary skills. Hollingsworth (1), who has studied breakthrough innovation across hundreds of biomedical research organizations, has observed that the most productive ones have numerous linkages to networks of scientists in diverse fields where the exchange of ideas takes place. When marshaled toward a common goal, these interacting innovation networks have been especially good at generating breakthrough solutions. They are also more efficient. Investing in a single or narrow set of options can lead to overfunding projects that are not yet ready for translation. Scherer (2) has shown that increasing the number of competing projects undertaken in parallel is more likely to optimize the tradeoff between the speed and cost of R&D than pouring all resources into a single or small set of projects.

Sharing can happen on at least two levels. One is sharing knowledge and resources to make the discovery process more efficient. If a company has developed biochemical and in vitro assays and an animal model for a particular disease, for instance, why screen only the compounds that its chemists can make? Why not leverage that investment by inviting the chemists of the world, many of whom do not have the resources to develop their own assays, to submit their compounds for testing? This is the rationale behind Eli Lilly’s recently launched Phenotypic Drug Discovery or PD2 initiative (3). If an interesting compound is identified, Lilly has the opportunity to negotiate a collaboration with the compound’s owner. Innocentive is another venture that seeks to harness the benefits of sharing by connecting problem owners (or “seekers”) to a worldwide online community of about 170,000 registered “solvers” (4). Results show that, in about 40% of the cases, some solvers have just the expertise needed to crack the challenges and are rewarded by cash prizes offered by the seekers.

A second level of sharing revolves around goals that are difficult to achieve but whose rewards are potentially quite valuable. Pioneering science has always been about dissecting and fusing existing and new knowledge, a process that often reveals new vistas. And sharing can take that process to a level that begets genuine collective creation. DeMonaco (5) has shown that many therapeutic innovations come from physicians who try to help patients for whom standard treatments have failed, and in doing so identify new uses for existing drugs. Why not facilitate this process by opening the companies’ massive databases to the scrutiny of the world’s scientists? Some companies are already doing it. Lilly has long championed transparency in clinical trials and was among the first to endorse their mandatory registration in a public database and to publish audited results. More recently, GlaxoSmithKline shared genomic profiling data on more than 300 cancer cell lines, and Novartis released the results of a large genomic analysis of type II diabetes. Ultimately, sharing should bring about an efficient division of labor in which companies collaborate and share the cost of advancing foundational knowledge, such as identifying single-nucleotide polymorphisms (SNPs) and other biomarkers; developing disease-state models; elucidating pathways in complex diseases; and bringing to maturity sciences and technologies such as translational medicine, epigenomics, and stem cell biology. Companies can then focus the bulk of their resources on areas that are likely to yield new treatments, such as designing small molecules that effectively modulate specific targets and pathways, with distinctive clinical benefits over the current standard of care.

Sharing can also foster innovation in developing countries. If scientists in geographically isolated locations can access online tools and data that until recently were available only to researchers from wealthy countries, they might no longer need to emigrate in order to innovate. Instead, these scientists can leverage the online resources to conduct research that matters to them. Hohman et al. (6) recently used this paradigm to identify several promising compounds, some of which are existing drugs that can restore efficacy to chloroquine, an inexpensive and often ineffective malaria treatment. If scientists in developing countries can create valuable intellectual property that then generates the resources to fund future discovery cycles, they might be more inclined to support intellectual property protection.

THE DAWN OF SCIENCE 2.0?

There are many ways in which scientists in industry and academia can share information. We have mentioned a few. Another form of collaboration is to help strengthen the research infrastructure. For instance, the Critical Path Institute, an independent organization founded at the behest of the U.S. Food and Drug Administration (FDA), has formed several industry consortia to develop new tools that can accelerate the development of medicines. One of them is creating disease progression models for Alzheimer’s and Parkinson’s diseases. Others work on validating biomarkers for cancer and cardiovascular diseases. A similar effort underpins the Innovative Medicines Initiative in Europe. And there are still more consortia searching for new SNPs or elucidating cell-signaling pathways, as well as cooperatives to create superior animal models of disease and groups who share toxicology data. Drug companies are learning to collaborate on a range of issues, both large and small. There are industry consortia to address privacy and data security issues, or the questions that arise between trial sponsors and independent review boards, or to advance the science of inhaled drugs by conducting joint R&D projects. These consortia are extremely cost-effective. Yet keeping them going has not always been easy because of industry’s strong proprietary culture, which tends to value internal work over collaboration. This total-ownership mindset, however, is becoming unaffordable, making sharing an imperative. Entrepreneurs are stepping in to help. Edwards (7), for instance, has banded with the U.S. National Institutes of Health (NIH), GlaxoSmithKline, and several universities to make a large number of chemical and clinical probes freely available to the world’s scientists, who must in turn share their findings. Other ideas, once unthinkable, may no longer be far off. For example, a consortium whose mission would be to administer a compound collection comprising most or all of big pharma’s chemical collections would offer obvious benefits. If a company has undiscovered treatments in its vaults, why not allow others to identify them? How long can mistrust stand in the way of the common good?

These examples target low-hanging fruit and tend to involve institutions rather than individual scientists. Yet it is possible to imagine other sharing schemes [for instance, based on the open-source concept (8)] that can empower true collaborative creation of novel therapies. In 2008, India’s Council of Scientific and Industrial Research launched such an Open-Source Drug Discovery platform aimed at tuberculosis (9). This platform breaks down the steps of drug discovery into 10 work packets such as target identification, target expression, drug screen development, in silico docking of molecules, etc. Any scientist interested in contributing his or her expertise can log on, join the appropriate work packet, and pitch in. More than 1250 scientists from 25 countries have signed up to do so.

In a similar vein, Scott (10) at Indiana University–Purdue University Indianapolis (IUPUI) has created a Distributed Drug Discovery system, or D3, which breaks down drug discovery into three stages that rely on the expertise of volunteers in low-cost centers around the world. The first step uses computational chemists and the idle power of numerous personal computers to scan the molecular space and identify a small number of promising drug candidates. The second step uses students at universities in the United States, Spain, Poland, and Russia to synthesize the molecules as part of their training. The last step, currently performed with the help of scientists at NIH, will eventually use students and inexpensive tests to perform primary biological screening.

Some pioneers from academia have gone further and are pushing to use the interactive tools of Web 2.0, which include blogs, wikis, YouTube, social Web sites, and RSS feeds, to create a robust framework for open science that would make drug R&D a broad-based participatory exercise. Web 2.0 is based on the idea of user-generated content: Someone sets up a Web site, and users create the content. The idea seems to be getting some traction. Scienceblogs.com hosts 75 blogs, some of them attracting 1.5 million visitors a month. Others such as Neurodudes (11), Brain Waves (12), and In the Pipeline (13) appeal to audiences of neuroscientists and pharmaceutical researchers. Social bookmarking Web sites such as Connotea or CiteULike help scientists locate material that their peers have found useful. Since 2007, the Journal of Visualized Experiments, a peer-reviewed, PubMed-indexed platform, has published biological research in video format (14). Some scientists, however, choose to use the more informal YouTube (15) to disclose material, such as negative or inconclusive results, which are difficult to publish in a peer-reviewed medium. Jean-Claude Bradley at Drexel University has pioneered the use of open notebooks to share his chemistry experiments in real time with the rest of the world (16). The OpenWetWare Web site (17) uses wikis to share experimental protocols and lab books that display everything, successful or not. Nature’s Network (18) and biomedexperts (19) are social Web sites designed to help scientists locate and connect with relevant expertise.

Some researchers have voiced concerns about the lack of quality standards that pervades this rising tide of user-created content. They fear that, because it is free, easy to find, and frequently offered in a colloquial, more engaging format than peer-reviewed literature, it may acquire a prominence unjustified by its scientific value. In fact, one should not pit Science 1.0—peer-reviewed literature—against Science 2.0. Far from competing with each other, these channels meet different but complementary needs. Shared user-generated content rules the forum where novel ideas and insights take shape. It meets the demand for informal and rapid-fire communications of participating scientists. Most of the work of researchers never makes it into a peer-reviewed journal and remains, therefore, inaccessible to most people. Yet there is much to be learned from the setbacks and meanderings that accompany the making of a scholarly paper. Open science helps lift the veil of secrecy so that researchers can learn from each other’s stumbling.

IF YOU BUILD IT, WILL THEY COME?

Despite the benefits that come with it and the availability of tools that allow it, many scientists remain ambivalent about sharing. In 2006, Nature launched an experiment that gave scientists the opportunity to comment on papers undergoing peer review. It was not a success. There were few comments, and most of them were not substantive. Despite vocal advocates, few scientists have adopted the tools of open science, and industry often restricts the ability of its researchers to do so. Part of the problem is that the system used to assess and reward scientific output favors “closed” science. By and large, scientists are still judged by their number of experiments, publications, patents, and citations. There is no credit, and thus no reward, for writing a blog or posting comments.

Employers are also wary. Hanging out on the Web takes precious time away from building molecules and other “real work.” Pouring your mind into a blog could result in inadvertent disclosures that might compromise intellectual property. It can also tip off a competitor, who could take your half-baked idea, improve it, and scoop you.

The infrastructure for sharing may be in place, but the cultural barriers are proving more daunting than the technical ones. Many of these fears are rooted in the proprietary mindset that has long dominated the industry. Yet sharing is helping to overcome them. For instance, many drug companies have developed robust methods to test the safety of their drugs. However, mutual distrust has kept these companies from working toward a common approach, leaving FDA scientists confused about which methods are better and should be the preferred ones. In 2006, the Predictive Safety Testing Consortium, organized by the Critical Path Institute, invited pharmaceutical companies to meet with scientists from the FDA and the European Medicines Agency to test and compare their respective methods, so that regulators could issue better guidance. They now convene monthly and involve over 200 scientists in the discussions.

REKINDLING THE CULTURE OF SHARING

The fear of jeopardizing intellectual property or giving away valuable ideas that might be appropriated by colleagues or competitors without due credit looms large in the enduring resistance to sharing. This concern, although legitimate, fails to appreciate that the process of innovation has been profoundly transformed by the growing complexity of science. Innovation once resulted mostly from large quantum advances. Today, however, innovation increasingly stems from the aggregation of numerous small contributions. This trend, which became noticeable a couple of decades ago, has reached problematic proportions. There is hardly an invention today that cannot be traced to numerous prior patented discoveries. Securing the rights to this intellectual property has significantly slowed innovation and increased its cost. It is not uncommon for such negotiations to take years and for the sum of the royalties on the prior art to threaten the viability of the innovation being pursued, a problem known to patent lawyers as “royalty stacking.” Any holdout can freeze the process, causing further delays and costs. The Internet has the potential to significantly increase the flow of microcontributions and leverage downstream innovation accordingly. Yet the fear of having one’s ideas stolen is hindering sharing and spoiling this opportunity (Fig. 2). It would help if the existing legal framework were to be supplemented in a way that would protect microcontributions as they arise, in a process somewhat akin to what already happens with copyrights. This could perhaps be achieved with an online registry that allows scientists to record and time-stamp their contributions as an optional but cheaper and faster alternative to patenting them. If such a contribution should subsequently find its way into a commercial product, its inventor(s) would be entitled to part of a statutory share of the product’s revenues that would be split among such claimants.

Fig. 2.

Seize the day. In 1959, John F. Kennedy made a speech in the authors’ hometown of Indianapolis, Indiana, and remarked that the Chinese word for crisis was made up of two characters, one that represents danger and another that represents opportunity. Fifty years later, his message still resonates.

CREDIT: C. C. CHEN

The drug industry today produces about 20 new drugs annually, roughly the same number as it did 50 years ago. Each company produces new drugs at a constant rate, and all attempts to speed drug innovation have failed to increase this mean drug output (20). However, evidence suggests that enhanced sharing of knowledge can change these dynamics by making everyone better informed and increasing success, a process economists call spillover. The Internet, as an example, offers unprecedented opportunities for sharing, which hold the hope of energizing therapeutic innovation and ushering in a new golden age of drug discovery. However, secretive behaviors inherited from the past and reinforcement by an intellectual property–protection framework that predates the Internet threaten to derail these opportunities. Changing this situation will take bold initiatives from senior industry and academic leaders as well as policy-makers to build in the incentives that are now lacking. One expected result should be greater collaboration upstream in the drug R&D process to expand foundational knowledge and create a shared infrastructure as well as tools to mine it. Such collaboration will eliminate much duplicative work, allowing pharmaceutical companies to refocus the resulting savings on downstream research through which new drugs and competitive advantage are created. Stated differently, companies should compete in areas that offer a viable return on investment, and share where pre-competitive collaboration helps all of us discover new therapies more efficiently and effectively, as patients and society demand. In sum, we issue a call to action to the pharmaceutical industry, universities, and government agencies to join hands and intensify sharing in order to help repower pharmaceutical innovation.

Footnotes

  • Citation: B. H. Munos and W. W. Chin, A call for sharing: Adapting pharmaceutical research to new realities. Sci. Transl. Med. 1, 9cm8 (2009).

References

View Abstract

Navigate This Article