To create a potential vaccine for SARS in 2003, a group of Canadian researchers had to break the law.
Nearly 800 people died from this viral respiratory condition and some 8,000 infections were reported across the globe. By April 2003, when the SARS Accelerated Vaccine Initiative, or SAVI, was formed in British Columbia, Toronto had been hit with the first of two outbreaks it would experience. Using existing vaccine parts that had already been approved for human trials, the researchers developed three vaccine candidates in less than a year.
In the process, they circumvented university lawyers battling for a piece of whatever profits might come down the line from patents and infringed on the intellectual property rights of scientists who had come before them. Recognizing how devastating this virus could soon become, the researchers behind SAVI prioritized protecting people from it as quickly as possible – the law would sort itself out later, they hoped.
“No one wanted to grant permission to use their vaccine virus backbone as the SARS virus’s if they didn’t get anything out of it,” says microbiologist Brett Finlay, a professor at the University of British Columbia who led the project. “We just went ahead anyway. We figured if SARS really came back ferociously the next year – and we thought it would – then they [the legal teams representing the parties involved] could figure it out, or mandate it, or legally change it.”
A second wave of SARS never came, the members of the team returned to their regular research activities, and those legal quandaries seemingly disappeared.
When it became clear early this year that Canada would have to contend with a new coronavirus rapidly sweeping the globe, the federal government quickly decided that it wouldn’t put itself in such a legally precarious position again. The COVID-19 Emergency Response Act was passed in March, giving the government the power to do exactly what Dr. Finlay and his team had counted on back in 2003 – to appropriate patented inventions as needed to address a public health emergency.
It seems only logical that to protect lives from a deadly virus, researchers should be able to freely mobilize existing scientific knowledge and tools. And we are seeing that play out today with an unprecedented level of collaboration and knowledge-sharing. Processes that would have taken months or longer now take hours: the Canadian Institutes of Health Research administered $54.2 million for COVID-19 research in the organization’s fastest grant competition ever; researchers and clinicians are sharing lab notes and patient treatment protocols in real time, pushing concerns for academic credit aside; scientific publishers are expediting peer review processes and more than a thousand open-access articles on COVID-19 have already been published. The World Health Organization, UNESCO and national science advisors from around the world have called for open data sharing; and the WHO is considering a proposal to make patented diagnostics, drugs and vaccines available to all. Just a few months after the release of the COVID-19 viral genetic sequence by Chinese researchers in January, multiple vaccine candidates are already in clinical trials.
In many ways, the global response to COVID-19 has strengthened the case for open science, a movement that has been gathering momentum in the biomedical fields and beyond over the past few years. Underpinning the movement are principles like open access publishing and the free sharing of data, tools and biospecimens like cells, antibodies and animal models. Where appropriate, like in the case of a vaccine for a viral pandemic, the movement also advocates for an open approach to intellectual property and commercialization.
It’s an approach that Mona Nemer, Canada’s chief science advisor, has been promoting. “There is agreement between funders, publishers and researchers that the only right thing to do in these unprecedented times is to make science related to COVID-19 open as quickly as it becomes available,” she says. “Collaborations generally come easier to researchers now, as they are fighting a common enemy and time is of the essence. Also unprecedented is the speed at which discoveries are being translated into public health policy. I hope that this experiment will influence people’s views about open science.”
To the average person, it may come as a surprise that science, particularly academic science, which is largely publicly funded, is not always conducted in an open and collaborative way that makes it accessible for anyone to build upon. “After we did the SARS rapid vaccine development, people said, ‘Why can’t we do this for cancer and all the other major problems in the world?’” Dr. Finlay recalls. “Unfortunately, I had to say that science, as it stands now, doesn’t really work that way.”
In fact, science today involves paywalls for papers, restricted access to datasets, licensing delays and researchers who sometimes refuse to share their data, says Viviane Poupon, chief operating officer at the Tanenbaum Open Science Institute (TOSI) at The Montreal Neurological Institute-Hospital.
The institution, better known as the Neuro, claims to be the first academic institute to adopt such a model. Its foray into open science with the founding of TOSI nearly four years ago follows that of non-profits like the Structural Genomics Consortium and precedes government open science initiatives, like Canada’s Roadmap to Open Science – a set of guidelines released by Dr. Nemer in February which outlines steps to make federal science accessible to all – and the European Commission’s Plan S, which seeks to have all results from publicly funded research published in open access journals by 2021.
Open science aims to overcome what some researchers describe as a culture of competition, secrecy and premature commercialization in science, which slows down the pace of discovery and hampers our understanding of the molecular mechanisms behind the most challenging diseases of our time.
“We need to better define what we mean by ‘open science.’”
In the United States, the Bayh-Dole Act of 1980 gave universities the power to patent innovations funded by public dollars. In Canada, no such law exists although the 2002 Framework of Agreed Principles on Federally Funded University Research, developed by the Association of Universities and Colleges Canada (now Universities Canada, publisher of University Affairs) and the federal government, struck a bargain in a similar vein: it promised universities a doubling of federal funding if colleges and universities tripled their commercial performance, defined in part as income from intellectual property, by 2010.
According to Dylan Roskams-Edris, open science alliance officer at TOSI, the thinking was that if knowledge generated at universities might have commercial application, the best way of making sure that was realized was to apply for patent protection. Universities opened technology transfer offices and patent applications rose. So did the administrative costs of filing those applications.
As a result of this increased patenting, each time researcher materials like a biological sample are transferred from one institute to another, lawyers are tasked with negotiating material transfer agreements to determine who has ownership over any resulting discovery or profit. Negotiations can cause administrative delays of weeks or months when sending even simple reagents between institutions. Sometimes the negotiations fall apart, shutting down scientific projects altogether, says Mr. Roskams-Edris.
“When you multiply the delays by the number of scientific interactions between institutions, it amounts to a significant loss of time for IP that is unlikely to actually be of any value,” he says. “The majority of patents that get applied for and even granted don’t end up leading to agreements, let alone products.” Yet it costs university technology transfer offices anywhere from $10,000 to $50,000 of public money to file for a single patent, says Mr. Roskams-Edris.
As it turns out, it’s not uncommon for researchers to try to manage delays by sidestepping intellectual property policies and sharing resources via informal channels –even outside of crises. But doing so disproportionately benefits those with seniority and extensive networks of collaborators. When legal negotiations can’t be avoided, it is the least well-off institutions and researchers who suffer, says Mr. Roskams-Edris.
Not all researchers agree with Mr. Roskams-Edris and his colleagues at the Neuro that the current system is problematic. Despite the issues they faced researching a SARS vaccine, Dr. Finlay says he’s not entirely sure that science is actually hindered or slowed by these IP processes. He says that seasoned researchers know to avoid or go around institutions that tend to have onerous processes in place for patent negotiations.
But if laws and institutional policies are too costly or have to be circumvented to get results, should those policies be changed? What could be achieved if researchers worked more openly all the time, and not just in times of crisis?
The Neuro adopted an open science framework in 2017, after an 18-month consultation that saw 70 principal investigators and 600 other scientific faculty and staff members opt in to the experiment. The goal? To accelerate understanding of central nervous system diseases. “We barely understand the molecular pathogenesis of Alzheimer’s disease, of Parkinson’s disease, of Frontotemporal Lobar Dementia. We’re still trying to understand what’s happening at a molecular level,” says Jason Karamchandani, a neuropathologist at the institute.
The transition required the institution to expand its existing open-source data and project management software, build a biobank – a collection of biospecimens – and develop an open transfer agreement that eliminates the majority of intellectual property claims when institutions share materials. The Neuro is also working on a toolkit for quantitatively measuring the impact open science has on innovation.
Such infrastructure is essential to practicing open science over the long term, says Dr. Nemer, who cites the Neuro’s model as one to follow. “Having agreement ahead of time on standardized protocols, approvals and format of research output actually adds value to the resulting data, which can then be easily compared among researchers in different institutions, provinces and countries,” she says.
A key aspect of the Neuro’s experiment is the institute’s Clinical Biological Imaging and Genetic Repository, or C-BIG. The collection of biological samples, clinical information, imaging and genetic data includes pluripotent stem cells, a unique tool derived from the institute’s patients. These cells are self-renewing and can be reprogrammed and grown into neuron cells and organoids, or collections of cells, called mini-brains. They give researchers an unlimited source of brain tissue on which to conduct tests and develop new therapies. “This is brand new,” says Dr. Karamchandani. “We’ve had bits of cancer but we haven’t had the tissues involved in neurodegenerative disease. These cells are a new tool in allowing scientists to investigate these diseases and they’re sharable because we can create more of them.”
C-BIG features more than 24,000 specimens collected by researchers over the last three years. Samples are collected from the same patients over time, providing insight into diseases like multiple sclerosis, which is known to impact the body differently during active and dormant periods. Unlike traditional biobanks, which pool cells between established collaborators, Dr. Karamchandani says C-BIG will be open to any researcher. “It’s about anyone who has a good scientific question being empowered to conduct meaningful scientific interrogation,” says Dr. Karamchandani. And it’s an example of how open science can level the playing field and encourage diverse collaborations – the platform goes live to the public later this year and yet it’s already led to partnerships with the Canadian Open Parkinson’s Network and Capture ALS.
“Researchers who publish in the open are more widely read. … They also reach a more diverse audience, are cited more often and have a higher chance of making an impact.”
But it’s not a data free-for-all. The institute has implemented a three-tiered data classification structure for C-BIG: data that poses no risk to patients – like demographics – are available open access; more detailed genetic and phenotype data are accessible only to researchers who register with the institute; and access to biosamples that could re-identify patients when cross-referenced against other databases is determined by committee review.
C-BIG also feeds the Neuro’s Early Drug Discovery Unit (EDDU). The unit brings together researchers and industry partners to identify molecular targets that hold promise for the drug-development process. The institute has partnered with multinational pharmaceutical companies like Merck and Takeda, as well as open science biotech firms M4K and M4ND Pharma. Over the last three years, a third of the unit’s $25-million in funding has come from industry.
Researchers investigating a question like whether a specific molecular mechanism has implications for Parkinson’s, work together and with industry on procedures for analyzing the effects of a compound or drug on a diseased cell. That partnership gives pharmaceutical companies and biotech firms direct access to research expertise while providing investigators with funding and early access to new technologies developed by the companies. The open transfer agreement also requires companies to share with the Neuro the results from investigations that make use of the institute’s platforms and specimens, something that Dr. Karamchandani says doesn’t happen with most traditional collaborations.
Even with these rules, industry has been eager to collaborate because progress on drugs targeting central nervous diseases has been slow, says EDDU associate director Tom Durcan. “We haven’t really seen anything new pretty much in the last 10 years,” he says. “In a way, the pipeline is broken for both of us.”
The Neuro’s success in bringing biopharmaceutical partners on board is a testament to its efforts to collaborate with private sector and to better understand how open science can contribute to business and commercialization while also benefitting academic research. Reconciling the two remains one of the biggest barriers to wider implementation of the open model – the huge cost just to take a drug through clinical trials is one of the main justifications for pharmaceutical patents.
“What we need is a balance between public knowledge without IP and private knowledge with IP,” says Mr. Roskams-Edris. “Public institutions should be producing the best possible quality public knowledge that private interests can then use as the base for their own private development.”
Dr. Poupon sees the early-stage research that academic scientists do as a complement to the role pharmaceutical and biotech companies play in drug development. “You take high risks when you develop a molecule commercially, and it takes a lot of time and investment,” she says. “It’s a very specific business that is not what academia does and we totally respect that.”
However, some prospective private-sector partners remain skeptical due to what they see as a lack of clarity around a business model based on open science. “We need to better define what we mean by ‘open science,’” says Diane Gosselin, president and chief executive officer at the Consortium Québécois sur la Découverte du Médicament, a biopharmaceutical research consortium funded by public and private donors. CQDM aims to support early-stage, high-risk research that leads to tools for scientific discovery, and has partnered with the Neuro on a platform to identify new drugs for Parkinson’s Disease and ALS.
For Dr. Gosselin, open science is a collaborative way of working between academic and private institutions where both parties benefit. “It doesn’t mean that there’s no IP all the time,” she says. Instead, she believes whether and how IP might be applied down the line should be addressed in the early stages of such collaborations.
Ownership over discoveries also poses a challenge for the open science model in academia. For his part, Dr. Karamchandani thinks universities will be hesitant to give up their IP because it’s “been deemed a measure of success for universities in Canada, and even for individual investigators,” he says.
At the Neuro, initial concerns over whether the switch to open science would keep young researchers away have eased now that it’s attracted more than 30 new trainees. Nevertheless, the issue of how to acknowledge the contribution of individual researchers remains a barrier to implementing open science, especially in academic institutions. “Science is a competitive business and you don’t just go and tell everyone your very best data long before you publish it because then others might beat you to it,” explains UBC’s Dr. Finlay. “Your tenure and promotion are all based on your abilities to publish… and unless we come up with a better way of defining someone’s abilities as a scientist that’s not based on peer reviewed papers, then that competition is always going to be there.”
To address this issue, the Neuro is developing additional evaluation criteria, including whether the investigator has released open source datasets or published open source code. It’s also experimenting with researcher resource identification, a type of digital barcode for cell lines developed by the EDDU, as an alternative to using patents for documenting the evolution of a discovery.
Still, staff at the Neuro know that widespread buy-in on an open science model will take time and education. “Even when you’ve designed a better system, people who have traditionally operated in a different and more closed system are going to take some significant convincing,” Mr. Roskams-Edris says.
“Even when you’ve designed a better system, people who have traditionally operated in a different and more closed system are going to take some significant convincing.”
Time will tell if the volume and speed of collaboration inspired by the COVID-19 crisis will be a watershed moment for the open science movement. Dr. Nemer, for one, is optimistic. “The COVID-19 pandemic is demonstrating to the research community that working in the open is not only doable, but is also beneficial to the researchers and knowledge-users, she says. “Researchers who publish in the open are more widely read, both domestically and internationally. They also reach a more diverse audience, are cited more often and have a higher chance of making an impact.”
Just as the flu pandemic of 1918 drove the creation of global health agencies and helped make a case for socialized medicine, so too could the present pandemic inspire a change in the way we address the medical and scientific challenges of the next century.