The influenza vaccine innovation system and lessons for PDPs

Abstract

As Product Development Partnerships (PDPs) emerge and evolve in response to the need for vaccines, this paper re-examines the oldest and most successful PDP in the vaccine field; that which year after year, produces and reinvents influenza vaccines. This paper describes the influenza vaccine production and innovation system and reviews some of its most recent major innovations. Innovation in this system is a result of collaborative partnerships between various actors from both the public and private sector. It is argued that the influenza vaccine innovation system is a Product Development Partnership (PDP), be it an unconventional one, with a central coordination role allocated to the WHO rather than a private company or charitable/not for profit entity. The unusual structure of this PDP overcomes some of the organizational issues surrounding vaccine research and production faced by other documented PDPs. These are first, the need to coordinate knowledge flow via an effective knowledge broker. Second, the need to build in-house capacity and fund essential research and elements of production where private partners find involvement too risky or costly.

Keywords: :

• H1N1
• Influenza vaccine
• Product Development Partnership
• innovation
• vaccine production

The importance of vaccines to public health efforts is widely recognized, as is the lack of vaccine innovation for neglected diseases and diseases that are a burden for the world’s poorest nations. Neither the public sector nor the private sector alone have the resources or ability to coordinate knowledge sharing in order to successfully bring to the market, vaccines that may have low profit potential but nevertheless contribute to the public good. Acknowledgment of this problem has given rise to new organizational forms that attempt to coordinate and facilitate research and development of vaccines by including all relevant stakeholders within a partnership. Product Development Partnerships (PDPs) are a particular variation of Public-Private partnership. They mobilise around a particular disease, bringing together skills, knowledge, and resources from a variety of sectors including academia, non-governmental organizations, philanthropists, government and intergovernmental agencies, and pharmaceutical and biotech companies. Examples include the Medicines for Malaria Venture (MMV), the International AIDS Vaccine Initiative (IAVI) and the Global Alliance for Vaccines and Immunization (GAVI). As PDPs emerge and evolve, it is worth re-examining the oldest and most successful PDP in the vaccine field; that which year after year, produces and reinvents influenza vaccines.

This paper describes the influenza vaccine production and innovation system and reviews some of its most recent major innovations. Innovation in this system is a result of collaborative partnerships between various actors from both the public and private sector. It is argued that the influenza vaccine innovation system is a Product Development Partnership (PDP), be it an unconventional one, with a central coordination role allocated to the WHO rather than a private company or charitable/not for profit entity. The unusual structure of this PDP overcomes some of the organizational issues surrounding vaccine research and production faced by other documented PDPs.

The Upstream Influenza Innovation System and Public Coordination

Currently 122 National Influenza Centers exist in 94 countries for the primary isolation and identification of influenza strains.¹ Sentinel physicians across the world take nasopharyngeal swabs from patients with influenza like symptoms and send these to national centers. If a new virus strain is detected, detailed antigenic and molecular analyses are sent from these national centers to one or more of the four WHO collaboration centers for Influenza Reference and Research located in London, Atlanta, Melbourne and Tokyo.

The assay data used to determine the antigenic drift (mutational change of the virus) is sent from the collaboration centers to The Center for Pathogen Evolution at the University of Cambridge in the UK, where antigenic cartography is used to construct antigenic maps. Each antigenic map is added to a composite ‘master map’ which collates data from all laboratories.²

Twice a year, the data are reviewed to decide which variants should be included in the next season’s influenza vaccines. The review is undertaken by the WHO Consultation on the Composition of Influenza Vaccine and includes representatives from the WHO, the University of Cambridge and the four WHO Collaborating Centers; the National Institute for Biological Standards and Control (NIBSC is a UK government body), the Therapeutic Goods Administration (TGA is part of the Australian regulatory agency), the Center for Biologics Evaluation and Research [CBER is part of the Food and Drug Administration (FDA) in the US], and the National Institute of Infectious Diseases (NIID in Japan).²

The WHO collaborating centers then spend three weeks in preparing a high growth seed strain which is passed to manufacturers for testing and mass production. A further three weeks is required to verify that the hybrid virus produces the outer proteins of the identified strain, is safe and grows in eggs. The vaccine virus is then passed to manufacturers.

In response to the 2009 H1N1 pandemic, WHO laboratories produced 5 seed strains; two by reassortment, and three by reverse genetics. Rather than coinfecting eggs with the field strain and a safe laboratory strain and then testing the new viruses that form through traditional reassortment, reverse genetics can isolate the genome segments encoding haemagglutinin and neuraminidase which can then be directly engineered so that they are less pathogenic.³ Reverse genetics presents numerous advantages including faster access to all antigens rather than only those expressed in vitro.⁴ Importantly reverse genetics can reduce the time taken to produce a seed virus for influenza by weeks. The first work using reverse genetics was pioneered by a team led by Yoshihiro Kawaoka, at the University of Wisconsin-Madison in 1999 using influenza.³ The first experimental vaccine for influenza using reverse genetics was created in 2003 with the H5N1 virus, and was undertaken by researchers at St Jude Children's Research Hospital in the US. In 2009 as it happens, the best yield came from NYMC X-179A, a traditional reassorted seed strain produced by New York Medical College. However, the significant advances made with reverse genetics in preparation for the pandemic will go on to further research and development of influenza vaccines in the future.

While the manufacturer mass produces the vaccine virus, the WHO collaborating centers create a reagent. The reagent is a standardized substance used to measure potency of the final vaccine and will ensure that manufacturers are packaging the correct dose.⁵ The reagent is created by the NIBSC in the UK and is calibrated by CBER, TGA, and NIID.⁶

The Downstream Influenza Innovation System and Private Production

The WHO delivers the vaccine virus to manufacturers who test different growth conditions in eggs to find the best conditions. The vaccine virus is injected into thousands of fertilized hen's eggs which are then incubated for two to three days. The egg white is harvested, and the virus is extracted. The partially pure virus is killed, the outer proteins are purified, and the result is several hundred or thousand liters of antigen. The size of the batch depends on how many eggs a manufacturer can obtain, inoculate and incubate and the yield per egg.⁵ Following bulk manufacture, quality control begins as soon as the WHO makes available the reagent. Vaccine filling and release is the last stage of production. The manufacturing process up to this point takes approximately three months.

Each year 300 million doses are required for the seasonal flu vaccine. Estimated need for eggs range from 350 million⁷ to 900 million.⁸ The need to expand production in a short space of time was highlighted by many observers before 2009 who were aware of the impending threat of a global influenza pandemic. Cell based production methods have been used for a long time for other biological products, for example insulin and polio. Large bioreactors which cultivate cells in a liquid medium can produce biological products much quicker and in bigger volumes. Egg based production methods had been used reliably for seasonal flu since the 1950s and there was no impetus to change until the prospect of the pandemic began looming on the horizon with outbreaks of H5N1 in South East Asia. In June 2009, Novartis announced it had successfully completed the production of the first batch of influenza A(H1N1) vaccine weeks ahead of expectations by using cell based methods of production.⁹ Cell-based manufacturing technologies allow vaccine production to be initiated without the need to adapt the virus strain to grow in eggs. While eggs are perishable, cell lines can be safely kept frozen indefinitely, increasing the capability to rapidly produce vaccines if an influenza pandemic were to occur.⁸

For supply to Europe in 2009 Novartis used its cell culture production facility in Marburg Germany,⁹ and later the same year the company opened the first cell culture manufacturing facility for influenza vaccines in the US. The investment in this facility is a partnership between Novartis and the US Department of Health and Human Services (HHS). It is estimated that the plant will be running at full scale commercial production by 2013.¹⁰ In fact the HHS awarded five contracts in 2006 totalling more than $1 billion to accelerate development and production of new technologies for influenza vaccines within the US including the advanced development of cell-based production technologies for influenza vaccines.⁸

In the recent past there has been significant innovation in the delivery system for influenza vaccines. For the first time a live attenuated influenza virus was used in the US in MedImmune’s FluMist®. It was first introduced in 2003 for seasonal flu and was also used in 2009 to vaccinate against pandemic H1N1. The vaccine is delivered as a nasal spray, thus removing the need for syringes and needles.

Innovation was demonstrated not only in methods of production, but also in the use of novel adjuvants. During the production process in July 2009, it became apparent that the seed strain distributed by the WHO was producing only half the antigen yield compared with that gained from production processes used for annual seasonal flu vaccine production,¹¹ highlighting the need for adjuvants. Clinical studies demonstrated the effectiveness of adjuvants and a reduction in the dose required from 2 to 1,¹² showing the potential for adjuvants to expand supply. Novartis used its MF-59 squalene based adjuvant which had been licensed for use in its seasonal flu vaccine in Europe in 1997. Glaxosmithkline similarly developed AS03, a squalene based adjuvant that was formulated and licensed specifically for Pandemrix^TM, its H1N1 pandemic vaccine.¹³

Elements of a PDP

The vaccine innovation system comprises an ecology of actors including immunologists, virologists, bacteriologists, clinical research organizations, hospitals and public health workers, commercial research organizations (CRO’s) universities, small biotech firms etc. The vaccine industry is heavily reliant on progress in the basic sciences, which is usually publically funded. While knowledge collaboration and exchange do to an extent happen naturally, many authors argue that it needs to be enabled.¹⁴ The actors need to be coordinated by an integrator or network organizer, usually a private company, in order to bring a vaccine product through the research and development stage, to the market.

The aim of PDPs is to act as an integrator or knowledge broker¹⁵ and to create long-term partnerships and build trust between the actors, toward the achievement of a common technological goal, in this case, a vaccine. They are best described as virtual non-profit R&D organizations, outsourcing research activities to academic or private sector partners, while linking together expertise and providing public funding, technical oversight and portfolio management.¹⁶ There are many different types and forms of PDPs, but most have been explicitly modeled after partnerships in the private sector with contracts, business plans and corporate management structures.¹⁷ This is in contrast to the Influenza vaccine innovation network which is tightly controlled and directed by the WHO, an international publically funded body.

Orsenigo et al.¹⁵ document the case of IAVI and suggest that vaccine research and development performed by PDPs, is hampered by organizational issues. In the case of IAVI, the role of the PDP was initially questioned as to whether it was a facilitator or knowledge broker, simply connecting the actors, or was it an integrator with a stronger leadership role? Orsenigo concludes that the latter eventually appeared to make the case for a more effective PDP in product development. Indeed with the influenza network the WHO is a strong integrator. The WHO dictates the direction of the research, the timing of reviews and collates and issues the relevant information to the various actors and to the public. In general the public sector plays a strong role, governments for example granting contracts and supporting the development of manufacturing capacity.

The case of IAVI also illustrates the need for PDPs to develop in-house competencies that its partners do not possess both in the basic sciences and the social context. For example, the IAVI partnership recognized a need to promote and fund basic research, for instance on broadly neutralizing antibodies. This however required the development of a range of complementary tools, such as assays, which were difficult to contract out.¹⁵ The influenza innovation network coordinated by the WHO over the past 60 years, has gradually built capacities for surveillance and monitoring, data analysis, knowledge distribution, reagent production and virus seed strain production. In-house publically funded capacity is required because new influenza vaccines are linked to the shifting properties of the influenza virus. A single company cannot alone undertake the monitoring required to track the global spread and change of the virus. Likewise, the global demand for information dissemination about the virus, its properties, the vaccine, rates of infection and death, etc. are tremendous and it may be argued, is most efficiently done by the WHO.

Finally, a PDP led and managed by a public interest body such as the WHO, has a better chance of not being perceived as western led effort. This problem affected other PDPs and has led to skepticism of their activities, to some extent hampering progress.¹⁸

Conclusions

The example of vaccine innovation and production for influenza shows that innovation is not only possible in a publically funded innovation system, but that the state and international publically funded bodies are essential in bringing new vaccines to the market. These bodies set the priorities for innovation, coordinate knowledge flow and exchange and fund essential basic and applied research and elements of production in the face of uncertainty. For other PDPs this network demonstrates the importance of a central organizer in brokering knowledge and building in-house capacity to fill the gaps that private partners find too risky or costly to engage in.

Abbreviations:
CBER	=	Center for Biologics Evaluation and Research
FDA	=	Food and Drug Administration
GAVI	=	Global Alliance for Vaccines and Immunization
HHS	=	US Department of Health and Human Services
IAVI	=	International AIDS Vaccine Initiative
MMV	=	Medicines for Malaria Venture
NIBSC	=	National Institute for Biological Standards and Control
NIID	=	National Institute of Infectious Diseases
PDP	=	Product Development Partnership
TGA	=	Therapeutic Goods Administration
WHO	=	World Health Organisation

Acknowledgments

This work is supported by Innogen, The Open University and by the Canadian Institutes of Health Research (Catalyst Grant CVC-99978), Principal Investigator, Janice Graham.