Virus Sharing Key Against Next Flu Pandemic: Global Database Hosts Genetic Data Of Flu Viruses

When a deadly influenza virus appears and threatens to become a pandemic, time is of the essence. The 1918 flu epidemic infected a large portion of the global population and killed millions of people. The next pandemic is inevitable, and surveillance of flu viruses is essential through the timely sharing of flu virus genetic data with the scientific and research communities. A collaborative database has become prominent in recent years.

Some countries have been reluctant to share flu viruses in the past, over concerns about transparency and lack of return in terms of access to resulting vaccines and benefit-sharing. However, an initiative created in 2008 appeared to have won international trust, in part because it addresses basic protection of intellectual property rights. It now hosts nearly 600,000 flu virus sequences, from all over the world, freely accessible on the conditions to adhere to its sharing mechanism, according to the initiative database.

The Global Initiative on the Sharing All Influenza Data (GISAID) was launched during the 61st World Health Assembly in May 2008. GISAID is an open collaboration comprised of experts worldwide. GISAID’s EpiFlu database, hosted by the government of Germany, is, according to the WHO Collaborating Centre for Reference and Research on Influenza (VIDRL) the most comprehensive database of flu viruses genetic sequence data.

GISAID plays a primary role in the sharing of data among the WHO Collaborating Centers and National Influenza Centers for the bi-annual influenza vaccine virus recommendations by the WHO Global Influenza Surveillance and Response System (GISRS). GISRS “monitors the evolution of influenza viruses and provides recommendations in areas including laboratory diagnostics, vaccines, antiviral susceptibility and risk assessment,” according to the WHO website. It also “serves as a global alert mechanism for the emergence of influenza viruses with pandemic potential.” [Note: GISRS is expected to be examined in a future story]

Genetic sequence data contain the genetic information that determines the biological characteristics of an organism or a virus.

The sharing of influenza virus genetic data although indispensable for the world to be prepared for the next pandemic, and also for the creation of seasonal flu vaccines, has been and still is an economically and politically-charged issue. Recently developed procedures of synthetic biology render the virus biological material and the virus genetic sequence interchangeable.

Countries have different reasons for not wanting to share data. For example, scientists may be concerned that their scientific contribution is not properly acknowledged. This was confirmed by WHO Director General Margaret Chan in March (IPW, Public Health, 8 March 2016) during a press briefing after a meeting of the International Health Regulations Emergency Committee on Zika.

Countries also may wish to retain ownership over any potential intellectual property associated with the data. This may be the case for low and middle income countries because they want to make sure that their populations will be able to access and share the benefits of vaccines developed from the data they shared.

The same issue about a lack of sharing virus information was also reported with the Ebola and most recently in the Zika virus epidemic.

A Bit of History

In 1952, the WHO had established the WHO Global Influenza Surveillance and Response System (GISRS), which did not address the issue of benefit sharing for developing countries.

When Indonesia declared in 2007 that it would not share its H5N1 bird flu samples with the World Health Organization, unless the country was guaranteed affordable access to vaccines, it signalled a dysfunction in the international flu research system for influenza surveillance (IPW, Public Health, 8 March 2007).

It was not, however, the first issue with the sharing of flu virus information.

The Los Alamos National Laboratory’s Influenza Sequence Database got its start in the late 1990s as an international repository accessible to researchers around the world, according to Science Daily. However, the lack of confidentiality was an obstacle for countries to share their data, and according to news articles, a secured section for bird flu virus was created 2004, to which only 15 laboratories had access.

In 2006, scientists cried foul about the Los Alamos database, concerned that it was only accessible to a small number of labs.

Dozens of scientists worldwide called for the release of all sequence data for the H5N1 avian flu strain into the public domain, according to articles in Science and the Wall Street Journal. The New York Times called the news of the WHO operating a secret database that holds the virus’s genetic information ‘distressing’ while health authorities were racing to head off a possible avian flu pandemic. According to Science Daily

A new paradigm for access by scientists was introduced at the 2008 World Health Assembly. GISAID offered a database hosted by the German government. A letter from the German government, obtained by IP-Watch, showed support for GISAID in response to a letter expressing concern over the process sent by a group of collaborating centres.

Three years after GISAID was launched and after the swine flu pandemic in 2009, the WHO developed and adopted in 2011 its Pandemic Influenza Preparedness (PIP) Framework. This was an answer to developing countries’ concerns over the sharing of flu viruses with pandemic potential, by providing an access and benefit-sharing mechanism.

The framework was mainly meant for virus biological material, however, the framework recognises that “greater transparency and access concerning influenza virus genetic sequence data is important to public health and there is a movement towards the use of public-domain or public-access databases such as Genbank and GISAID respectively” This was intended to give countries a choice as to how they share their sequence data with the public, emphasised in a speech given by WHO Assistant Director General Dr. Keiji Fukuda.

Meanwhile, according to sources, concern has been raised about GISAID not being sufficiently included in the discussion on how to handle genetic sequence data in the PIP Framework.

A WHO PIP Framework review team is meeting from 29 August to 2 September (IPW, WHO, 24 August 2016).

Ironically scheduled to overlap this PIP meeting is a “WHO side meeting during the IX options conference with the GISRS” being held in Chicago (US). Many of the collaborating center representatives and other interested parties are obliged to be at the Chicago meeting.

GISAID’s Database and Access Agreement

“Providers of data to GISAID retain all their inherent IP rights to the data they submit to GISAID, which is a result of the licensing terms whereby providers license their data only to registered users of GISAID that agreed to specific and clear licencing terms,” said Dr. Jörg Paura, IP lawyer and Board member of Freunde von GISAID e.V. Germany.

Experts in licensing media content developed and eventually negotiated with countries a legally binding licensing mechanism known as the EpiFlu Database Access Agreement that recognised the ownership of rights to the data and provide legal certainty on its use, aimed at regulating an unregulated exchange of flu data. The agreement states the licence terms as granting users with a “non-exclusive, worldwide, royalty-free, nontransferable and revocable license to access and use the GISAID EpiFlu™ Database.”

To have access to flu data in GISAID, every user must be positively identified and agree to the terms of use, which call on all users to recognise the contributions and interests of data providers and users, including the condition that originating laboratories providing the specimen and submitting laboratories who generate the sequence data must be acknowledged. The fair exploitation of results coming from the data should be ensured, and no restrictions shall be attached to data that have been submitted to GISAID, it says.

Data providers agree to allow “a non-exclusive, worldwide, royalty-free, and irrevocable license to collect, store, reproduce, access, modify, display, distribute, coordinate, arrange, and otherwise use the Data submitted…”

Data users agree that they may not access and use the EpiFlu database “or collect, store, reproduce, access, modify, display, distribute, coordinate, arrange, and otherwise use the Data in connection with any other database related to influenza gene sequences, including, without limitation, by enabling others to access or use the Data through a separate portal or across a network of institutions, except for operators duly authorized by GISAID.”

The data can be used to author, co-author or publish results obtained from analyses of the data, if such published results acknowledge, as the original source of the data, the laboratory where the clinical specimen or virus isolate were first obtained.

Users also have to agree not to distribute data to any third party other than GISAID authorised users to maintain a traceable chain of custody of the data. Users are encouraged to make “best efforts” to collaborate with originating laboratories and involve them in further research.

Data providers can restrict or prevent persons not bound by the GISAID EpiFlu Database Access Agreement from using the data they provided but they agree to inform such persons that said data or technology is available through EpiFlu (see section h of the agreement).

Breach of the access agreement would result in the termination of access rights, the agreement says.

While GISAID currently offers some 592,000 sequences of flu viruses, human and animal, 210,000 of those, sourced from over 800 institutions worldwide, are exclusively available in GISAID, as shown in the database. The database also collects associated metadata such as date of specimen, specimen source, date of virus harvest, antiviral susceptibility, and for human samples patient information such as age, gender, health status, treatment, and vaccination.

GISAID has some 6,500 active users, from National Influenza Centres, universities, research institutions and universities, public health authorities, both animal and human health experts, e.g., clinicians and epidemiologists, members of industry, and all of the six WHO Collaborating Centres, which are an integral part of GISAID’s governance structure.

GISAID: Not a Public Domain Database

GISAID offers a publicly accessible database, which cannot be considered a “public domain” database, mainly because of its database access agreement, which explicitly preserves the ownership rights to the data and regulates its use.

In particular, according to a GISAID background presentation, contrary to a public domain database, such as GenBank, GISAID provides unique accession numbers for virus isolates, and verify users’ identity. GISAID also protects submitters’ rights, and prevents the attachment of intellectual property restrictions on data, such as on the fraction of a sequence. GISAID also enforces acknowledgement of original specimen providers.

While the legal concept of public domain differs from jurisdiction to jurisdiction, public domain databases are always accessed anonymously. Public domain mainly refers to materials that are not protected by intellectual property rights and free to be used as pleased by anyone, without any restrictions or obligations to the submitter. Intellectual property rights can subsequently be taken on products invented from those materials, if they meet the required criteria.

The GISAID licence explicitly permits use of the data for development and distribution of medical interventions. It does not, however, permit making claims to the original virus through the underlying data through the modification.

The owner of virus given to GISAID does not allow modification of the underlying virus and taking ownership of it, sources said.

From Indonesia to China

In 2008, after the launch of GISAID, Indonesia immediately shared its bird flu virus genetic sequence data with GISAID.

According to a 2013 article published in Nature, on 31 March 2013, China reported the first human cases of infection with a new H7N9 avian flu virus and on the same day, uploaded the genetic sequences of viruses isolated from the first three human cases into GISAID.

On 5 April, the Chinese scientists submitted their first major H7N9 paper to The New England Journal of Medicine, and learned almost simultaneously that several other research groups were preparing to publish papers on the virus, or already had done so, including analyses of the sequences in GISAID.

Novartis and the J. Craig Venter Institute in Rockville, Maryland (US) planned to use the uploaded sequences from EpiFlu to develop H7N9 vaccines, without involving the Chinese researchers, who felt that this was contrary to the spirit of GISAID, the article said.

After some mitigation by GISAID, communication channels were opened between the various parties who agreed to collaborate, the article concluded. It was GISAID’s biggest test to see if it could enforce its sharing mechanism.

GISAID’s Public-Private Partnership and its Contributors

Starting in 2010, the GISAID initiative entered into a public-private partnership with the government of Germany to ensure its sustainability and to promote the global sharing of genetic sequences of all influenza viruses, related clinical and epidemiological data associated with human viruses, and geographic as well as species-specific data associated with avian and other animal viruses. Germany’s role as the official host and guarantor of GISAID’s database is to provide the public with free and open access as well as quality assurance of its data through manual data curation.

In addition to Germany, the *STAR’s Bioinformatics Institute of the Republic of Singapore and the United States Centers for Disease Control and Prevention, are prominent contributors, making substantial in-kind contributions to GISAID, according to the initiative.

William New contributed to this story.