Freeing the business of science
In this story
With most of his research conducted online, Associate Professor Matthew Todd, senior lecturer in organic chemistry at Sydney University, is used to not having met some of his closest research collaborators. As founder of the Open Source Malaria project, he is a pioneer of a growing international movement known as “open science”. A loose coalition of low-budget non-governmental organisations, ambitious start-up companies, frustrated academics and curious amateurs, open science is bringing together many parties with little in common other than a dissatisfaction with the closed-shop way the life sciences are currently practised.
It’s a movement that takes advantage of communication technologies to rally support, make connections and cut costs. Most important, it seeks to liberate knowledge and, in turn, transform industries and disciplines that many within the field consider stifled and bureaucratic, even corrupt, and to make them nimble, entrepreneurial and better able to deliver on the promises made possible by the past two decades of biological innovation.
“It was a sense of growing frustration with the way research was being done,” says Todd of his own move from conventional bench research to open science. “You see endless examples of people doing research in a closed environment in the lab and publishing academic papers, many of which are never cited, and doing work that is often very similar to work other people are doing, sometimes a duplicate.”
Todd, who won an Accelerating Science Award in Washington, DC, late last year, launched his first open science experiment in 2010. That was a project to find a cheaper way to synthesise praziquantel (PZQ), a drug used in the treatment of bilharzia, a parasitic disease that affects approximately 200 million people. Unlike a traditional study, all notebooks and results were published on the project’s website and anyone who felt they could was invited to participate – a similar model to open source software programs. “When you run things open source you prioritise freedom – anyone can do anything, from anywhere,” Todd says.
The project was a success and generated enough international attention for the Switzerland-based Medicines for Malaria Venture to fund Todd to start Open Source Malaria (OSM). Launched in mid-2013, OSM draws upon a huge amount of chemical information released online by a philanthropic wing of pharmaceutical giant GlaxoSmithKline – a major reversal of the norm in which intellectual property is jealously guarded and the release of a single compound might require months of paperwork.
Yet it’s just one example of a growing trend among drug companies. In February, Johnson & Johnson agreed to share data from its US pharmaceutical operations through the Yale School of Medicine Open Data Access (YODA) Project, a move the project’s director, Harlan M. Krumholz, described in an editorial in The New York Times as “an extraordinary donation to society, and a reversal of the industry’s traditional tendency to treat data as an asset that would lose value if exposed to public scrutiny”.
So why would a drug company release valuable intellectual property to researchers for nothing? Undoubtedly there is an element of good PR in it, but Todd notes there are also economic benefits with the cost of archiving vast quantities of research material not inconsiderable. Further, many pharmaceutical employees are just as keen as researchers on creating cheap cures.
Indeed, contrary to the caricature of the evil, hand-rubbing drug companies making millions by monopolising life-saving data, so-called Big Pharma was infinitely more helpful to Todd’s bilharzia project than his academic peers. With the potential for massively accelerated research, you’d be forgiven for assuming all scientists would want to shift to an open model. Yet there remain significant cultural barriers.
“Academia is structured in such a way that we’re not rewarded for this kind of [open] behaviour,” Todd explains. “Because it’s competitive, you keep secrets.” The competition is for university grants. The more academic papers they are credited with authoring, the more funding a senior researcher might get. Todd speculates that some academics simply do not realise that a significant contribution to a published paper would result in an author’s credit, even in a piece of collaborative research. There is also pressure on academics to “own” a particular field of research. Like being credited as assistant director of a film rather than director, not receiving top billing in a project isn’t quite as attractive to senior researchers as it may be to, say, a PhD student. The collaborative and meritocratic nature of open science subverts academia’s traditional hierarchies.
While projects such as Open Source Malaria offer a vision of medical research reinvigorated – like Linux for drugs – the science is only half the story. Drug development is an expensive business and even if, for instance, OSM were to find a promising compound, there is no guarantee any drug will be developed as a consequence. In fact, the odds are massively against it.
In 2014 it might be fair to talk about a crisis in drug development. In recent years, drug approval has plummeted. Meanwhile, the research and development cost of taking new drugs to market has spiralled out of control – multinational AstraZeneca, for instance, spends a mind-blowing average of $US11.7 billion per drug. This cost is, of course, passed on to the consumer.
One reason for this cost explosion has been the “de-risking” of drug discovery and emphasis on profitable drugs for the so-called “lifestyle diseases” more common in developed countries, such as diabetes and obesity, as well as repeated trials designed only to broaden the applications of a pre-existing drug, no matter how minor the benefits. There are, however, those in the open source movement who believe that a more transparent model can change not only the science of drug development but also the business.
Bernard Munos is a former executive for American drug giant Eli Lilly turned champion of open science. In 2006, at the urging of a professor at the Harvard Medical School, he published an influential paper in Nature. Originally intended only for internal company use, it argued that the open science model had the potential to reinvigorate drug R&D and, possibly, completely undermine the current system. It caused quite a stir within the pharmaceutical industry.
Munos blames de-risking culture on a new breed of “scientifically illiterate” drug company managers, motivated by profit rather than innovation. He points out that since 2004, two-thirds of US Food and Drug Administration-approved drugs have been created by minor companies, often at a fraction of the spending of the majors.
“Small companies, because they have fewer resources, must collaborate with each other. They’re more open,” Munos says. “By making it open, you make it easier to cross-pollinate your ideas. And we know now that this process is essential to innovation.”
Beyond the institutional and cultural challenges to open science, arguably the model’s biggest impediment is the international patent system. Impossibly complicated and tremendously expensive to access, intellectual property law seems to have few friends in the business of innovation. In Australia, the white knight of patent reform is undoubtedly Richard Jefferson.
Jefferson is a big name within the scientific community: his research in the 1980s created a “reporter gene” that became standard in a wide range of genetically modified crops and made him one of the world’s most influential biologists. Today he is the CEO of Cambia, a non-profit institute based at Queensland University of Technology.
Funded by the Bill & Melinda Gates Foundation, Cambia’s aim is to democratise innovation, in part by making the patent system more transparent. Jefferson says that unless the system of accessing patent information is made less intransigent, experiments in open science are doomed to remain “self-indulgent”.
“The patent system is so phenomenally complex it might as well be in illuminated Latin script in monasteries,” Jefferson says. “The reason we now have expensive drugs for erectile dysfunction and baldness instead of bilharzia or malaria is because there’s not a lot of money in bilharzia or malaria. So companies can’t justify the phenomenal amount of money it costs to get their guys inside the monasteries to figure out what the fuck is out there.”
In a move he hopes will change that, Cambia last year launched the Lens, a navigational tool for the international patent system. The Lens offers users the ability to search 80 million worldwide patents, to view critical information such as which territories the patent is registered in, who owns it and whether it is still in force. It’s the kind of information that would normally cost hundreds of millions and require teams of lawyers. It could mean the difference between a red or green light on a research project, or perhaps ensure that a drug company avoids wasting time and money by unwittingly infringing on someone else’s patent.
With tools such as the Lens cracking open the safe on patents, and signs that at least some academics and companies are learning to share, does open science have the potential to disrupt a moribund research industry? Matthew Todd certainly thinks so. Indeed, without changing the current model he fears complete stagnation of innovation.
“I can share raw data with people as quickly as I can ‘like’ something on Facebook,” he says. “Yet we’re not doing it as a society. Technology is a solution, but we first need to share what we’re doing. Remember Oscar Wilde’s Selfish Giant – when he put a wall up round his garden, it fell under a perpetual winter.”
This article was first published in the print edition of The Saturday Paper on Mar 8, 2014 as "Freeing the business of science".
A free press is one you pay for. In the short term, the economic fallout from coronavirus has taken about a third of our revenue. We will survive this crisis, but we need the support of readers. Now is the time to subscribe.