Exploring synergies between academia and vaccine manufacturers: a pilot study on how to rapidly produce vaccines to combat emerging pathogens

Size: px

Start display at page:

Download "Exploring synergies between academia and vaccine manufacturers: a pilot study on how to rapidly produce vaccines to combat emerging pathogens"

Shanon Robinson
5 years ago
Views:

1 Clin Chem Lab Med 2012;50(7): by Walter de Gruyter Berlin Boston. DOI /cclm Exploring synergies between academia and vaccine manufacturers: a pilot study on how to rapidly produce vaccines to combat emerging pathogens Thomas Strecker1, Jennifer Uhlendorff1, Sandra Diederich1, Claudia Lenz-Bauer2, Heidi Trusheim2, Bernhard Roth 2, Larissa Kolesnikova1, Christian Aepinus 1, Reiner Dornow3, Jens Gerlach4, Mikhail Matrosovich1, Ulrich Valley 2, Markus Eickmann1 and Stephan Becker1, * 1 Institut für Virologie der Philipps-Universität Marburg, Marburg, Germany 2 Novartis Vaccines and Diagnostics GmbH, Marburg, Germany 3 Kreisverwaltung Marburg-Biedenkopf, Fachbereich Gesundheit, Marburg, Germany 4 Regierungspräsidium Gießen, Gießen, Germany Abstract Background: In spring 2009, a new swine-origin influenza A (H1N1) virus emerged in Mexico. During the following weeks the virus spread worldwide, prompting the World Health Organization to declare the first influenza pandemic of the 21st century. Sustained human-to-human transmission and severe disease progression observed in some patients urged public health authorities to respond rapidly to the disease outbreak and vaccine manufacturers to develop pandemic influenza vaccines for mass distribution. With the onset of the pandemic we began to explore the potential of academic/ industrial collaboration to accelerate the production of vaccines during an outbreak of an emerging virus by combining the use of an academic BSL-4 laboratory with the expertise of a commercial vaccine manufacturer. Methods and results: To obtain virus seed stocks used for the production of a vaccine to combat the pandemic H1N influenza virus (), we followed various strategies: (i) optimization of cell culture conditions for growth of wild-type isolates; (ii) classical reassortment of and standard influenza vaccine donor strain PR8; and (iii) generation of corresponding reassortant viruses *Corresponding author: Prof. Stephan Becker, Institut f ü r Virologie, Hans-Meerwein-Str. 2, Marburg, Germany Phone: , Fax: , becker@staff.uni-marburg.de Received October 12, 2011; accepted April 22, 2012 ; previously published online May 23, 2012 using reverse genetics. To ensure a rapid transition to production, the entire potential seed stock development process was carried out in a certified canine kidney suspension cell line (MDCK PF) under Good Manufacturing Practice (GMP) conditions. Conclusions: The outcome of this study indicates that a combination of different experimental strategies is the best way to cope with the need to develop vaccines rapidly in the midst of an emerging pandemic. Keywords: influenza A virus; influenza vaccine development; pandemic; reassortment; reverse genetics. A novel swine-origin H1N1 influenza A virus emerged in humans in April Within weeks the virus spread globally via human-to-human transmission, forcing the World Health Organization (WHO) to declare the first influenza pandemic of the 21st century. During the last decade, concerns about the highly pathogenic avian influenza A (H5N1) virus led the WHO and national health authorities to develop influenza pandemic preparedness and response plans. These involve public health and pharmaceutical interventions that include pandemic influenza vaccine programs in which a clear strategy is suggested regarding how pandemic vaccines should be produced (1, 2). These plans were put in place with the declaration of the 2009 H1N1 pandemic, and the WHO recommended the production of a -specific vaccine to provide protection for the global human population. As it is absolutely essential to react as quickly as possible to emerging pandemics, we explored the possibility of accelerating the process of vaccine production. For this purpose, we established a strategic collaboration between an academic biosafety level 4 (BSL-4) laboratory and an industrial vaccine manufacturer, and evaluated the time required to produce seed virus vaccine candidates using three different experimental approaches. Traditionally, influenza vaccines are produced in embryonated chicken eggs. However, human influenza strains often do not grow well in eggs and thus, the propagation of human isolates in eggs requires the adaptation of the virus, which is, in turn, accompanied by mutations in the envelope glycoproteins hemagglutinin (HA) and neuraminidase (NA). If these mutations lead to changes in immunodominant epitopes, the resulting egg-based vaccine will elicit antibodies that recognise the circulating influenza virus less efficiently (3). In

2 1276 Strecker et al.: Development of vaccine candidates against the event of the emergence of a pandemic caused by avian influenza viruses, some of these strains are highly lethal to chickens and will therefore not grow sufficiently in eggs without prior genetic modification to decrease pathogenicity, thus limiting the rapid use of the egg-based vaccine strategy. Moreover, the potentially limited availability of embryonated chicken eggs may be a critical concern when it comes to meeting global vaccine demands in a pandemic situation. The number of eggs required to produce a certain number of vaccine doses must be calculated approximately five to six months before the start of vaccine production. Therefore, the inability of an egg-based strategy to produce enough doses of required vaccines in a timely manner underscores the urgent need to explore alternative approaches to influenza vaccine production. In recent years, cell culture-based technologies have been developed that have a number of theoretical and practical advantages over the traditional egg-based production of influenza vaccines. The reported benefits of this technology include the easy scalability of continuous cell lines, allowing adaptation to the actual number of vaccine doses required. The use of well-characterized cell lines reduces the risk of introduction of exogenous or endogenous adventitious agents. Additionally, cell-based vaccine manufacturing in closed bioreactors reduces the risk of exogenous microbial contamination. Mammalian cell lines enable the production of influenza vaccines using avian virus strains. Moreover, individuals who are allergic to eggs and egg components can benefit from influenza vaccines derived from cell cultures. As potential vaccine viruses do not require an egg adaptation process for optimal virus growth, cell-based influenza vaccine production not only accelerates the vaccine manufacturing process but, more importantly, also minimizes the selection of virus isolates with reduced immunogenicity and altered antigenicity. Following the identification of the pandemic strain, cell culture-based vaccine production eliminates the long manufacturing lead times and supply-chain vulnerabilities that eggbased production systems are subject to. The new technology could therefore significantly reduce the time from detection of an emerging virus to the delivery of the final vaccine a factor that is particularly important in a pandemic-driven need for influenza vaccines. At present, two approved mammalian cell lines are used for production of influenza virus vaccines: African green monkey kidney (Vero) cells and Madin-Darby canine kidney cells (MDCK). Both cell lines support efficient growth of influenza viruses and are used to produce both seasonal and pandemic influenza vaccines (4 6). In addition, several other continuous cell lines including PER.C6 (derived from human embryonic retina cells), NIH-3T3 (derived from mouse embryo fibroblasts), BHK (baby hamster kidney) and CHO (chinese hamster ovary) have been reported to support replication of influenza viruses, although only PER.C6 cells reached influenza virus titers that are considered commercially relevant. Despite its suitability for industrial-scale manufacturing and regulatory requirements, a cell line destined for the production of human influenza vaccines must be susceptible to a wide range of influenza strains of different origin and thus be capable of propagating these viruses in large quantities. Prior to manufacturing a vaccine designed to combat pandemic influenza viruses, the virus seed stock for large-scale production of the vaccine needs to be prepared on the basis of a virus isolated from clinical material. The development of virus seed stocks from an unknown, potentially highly pathogenic virus is carried out in laboratories equipped to bio-containment level 3 or 4. In our experience, the rapid decontamination procedures and specialized ventilation and air filtration systems present in the Marburg BSL-4 laboratory have proved to be ideally suited for the GMP-compliant production of virus seed stocks. Furthermore, the Marburg laboratory s operating license guarantees a rapid grant of permission to handle pathogens that embody unknown biological risks. The authorization to work with wild-type isolates was granted within 3 days and we received approval from the national authorities to work with the genetically engineered virus within a week. Faced with the global outbreak in April 2009, we explored three different experimental approaches to rapid production of influenza vaccines for combating (Figure 1 A C): (i) wild-type virus isolates were cultivated and passaged to optimize virus yield in cell culture; (ii) classical genetic reassortment between and the influenza A vaccine donor strain; and (iii) a recombinant reassortant virus comprising the HA and NA genes from and the remaining six genes from PR8 strain was generated using reverse genetics. We assessed the potential of multiple virus seed stocks and selected those with optimum virus growth characteristics for use in further downstream processes. Importantly, to limit the time between identification and isolation of the virus and availability of a virus seed stock for vaccine production, all steps from virus isolation to production of the seed lot were performed in compliance with GMP using the licensed MDCK suspension cell line PF grown in serum-free medium. Most commonly, influenza vaccines are prepared from inactivated virus material that undergoes various purification steps, resulting in the enrichment of the envelope surface proteins HA and NA, which are the major targets of neutralizing antibody response. High virus yields are required to maximize the yield of surface proteins. However, human wild-type influenza A viruses often have a limited ability to replicate in eggs or cell cultures, thus restricting the potential to produce the large amounts of vaccine doses required in a pandemic scenario. To overcome this problem, chimeric viruses are produced which combine the surface proteins HA and NA of the pandemic wild-type isolate with the characteristics of a well growing attenuated influenza donor vaccine strain [e.g., PR8, A/PuertoRico/8/34 (H1N1)]. Classical reassortment is an ideal approach to produce such chimeric viruses. During simultaneous infection and multiplication of the wild-type virus and the vaccine donor strain in a single cell, the genome segments of the two viruses can be randomly exchanged, leading to the emergence of reassortants that possess characteristics of both the wild-type and vaccine donor strains. Using specific antibodies against HA and NA of the donor strain, viruses are subsequently selected that contain genome segments of

3 Strecker et al.: Development of vaccine candidates against 1277 A Wild-type virus B Classical reassortment C Reverse genetics Vaccine donor strain Vaccine donor strain Co-infection and genetic reassortment Six plasmids encoding genes from donor strain M NP NS Two plasmids bearing NA and HA genes from PA PB1 PB2 HA NA Co-transfection of mammalian cells Selection process using neutralized antibodies against the envelope proteins of the vaccine donor strain Reassortant virus Rescued recombinant virus Figure 1 Schematic diagram of vaccine seed strain preparation. (A) Virus propagation of wild-type. (B) Generation of pandemic influenza vaccine seed viruses using classical reassortment. In cells co-infected with an influenza vaccine donor strain and, random combination of genome segments leads to the emergence of reassortants that possess characteristics of both viruses. Ideally, the reassortant virus contains the six genome segments from the donor strain that ensure optimal growth in the cell culture as well as the two segments from the pandemic virus coding for the hemagglutinin and neuraminidase. (C) Generation of recombinant viruses using reverse genetics. Reverse genetics technology allows the generation of infectious influenza viruses using the building block principle with plasmid-based cdna, thus enabling the generation of recombinant viruses with defined genome segments. the vaccine donor strain, thus ensuring optimum growth and genome segments that encode HA and NA from the desired wild-type virus (7, 8). An alternative approach to developing virus seed stocks for influenza vaccines is DNA plasmid-based reverse genetics. This is a straightforward technology that enables the generation of defined chimeric influenza vaccine seed viruses using molecular biological tools (9, 10). Because the process does not require scientists to work directly with potentially highly pathogenic pandemic strains, the safety aspect of reverse genetics makes it a favorable technology for generating recombinant influenza vaccines. This approach uses non-hazardous DNA plasmids containing all eight genome segments of influenza viruses that are transfected into mammalian cells (11, 12). The DNA plasmids serve as intracellular templates for both the synthesis of viral genome segments and the viral proteins that then assemble into infectious viruses that are capable of replication. This technique allows the recovery of infectious viruses directly from cells that are approved for the production of human vaccines (13, 14). Moreover, in the case of genetic engineering through classical cloning procedures, because proteins are denatured and removed during the initial cloning steps and the subsequent preparation of plasmids, plasmid-driven transfection represents an efficient purification step for eliminating potential adventitious agents that may be present in human influenza virus isolates. A full description of our adaptation of reverse genetics technology for generating a potential seed virus in response to the 2009 H1N1 pandemic is presented elsewhere (manuscript in preparation). Our timeline for the development of vaccine seed candidates is shown in Figure 2. We produced a potential seed virus stock based on the cultivation of wild-type isolates within 12 days of receiving the reference virus [influenza A/ California/04/2009 (H1N1)] from the CDC. A recombinant seed virus generated by classical reassortment comprising HA and NA from the pandemic strain was available 6 weeks after initiation. A reverse genetics-based vaccine

4 1278 Strecker et al.: Development of vaccine candidates against Sample arrival -X Virus isolation from clinical material Days Sequencing Virus inoculation and growth Wild-type Virus propagation Reverse genetics Cloning Rescue Virus growth Classical reassortment Reassortment procedures Expansion Figure 2 Timeline for vaccine seed virus production for combating the 2009 pandemic H1N1 influenza virus. strain was rescued and grown within 18 days. All viruses were grown or rescued in the certified MDCK cell line in our BSL-4 laboratory using GMP-compliant procedures that assured the quality of the reference virus for further downstream use in vaccine manufacturing processes. During the initial stages of a pandemic, the use of new, poorly characterized wild-type virus strains to develop a vaccine may pose considerable biosafety risks for scientists and technical personnel. Unfortunately, there are very few vaccine manufacturers in the world who are able to produce vaccines under BSL-3 enhanced biosafety conditions. This means that many vaccine manufacturers are unable to start production until the biocontainment restriction has been downgraded to enhanced BSL-2 level following safety testing in a ferret model and in accordance with national safety regulations. The BSL-4 laboratory in Marburg has the advantage of being able to handle pathogens that require the highest levels of safety precautions. The pathogenic potential of the new virus was not clear as the 2009 pandemic began, so the ability to work safely with the new influenza virus provided a head start in evaluating and testing the purification characteristics of the viral surface antigens HA and NA. Current pandemic vaccines are reassortants comprising the envelope surface genes of the pandemic virus and the remaining genes of an attenuated influenza vaccine donor strain. This method is well established and has been successfully applied in production of seasonal influenza vaccines over the course of several decades. Moreover, this approach allows the selection of well growing viruses that yield large amounts of envelope surface proteins. However, due to the complex screening, selection and genotyping procedures for the desired vaccine strain, this process requires a lead time of up to 2 months from strain identification to the start of vaccine production. In contrast, the development of seed viruses based on reverse genetics can help to significantly reduce the time required to generate influenza virus vaccine seed strains. During our study, we achieved a saving of 3 weeks compared to the classical reassortment method a critical advantage when preparing a pandemic vaccine. However, the use of reverse genetics in human influenza vaccines may have some limitations, including the need for approval of cell lines suitable for efficient virus rescue and vaccine production. Currently, the pharmaceutical company MedImmune holds the intellectual property rights for reverse genetics technology in human influenza vaccine manufacturing. Although MedImmune offers other influenza vaccine manufacturers non-exclusive licenses, the additional cost associated with the payment of royalties may reduce the attractiveness of this methodology to other vaccine manufacturers. However, it is expected that these issues would not restrict the development of reverse genetics-derived vaccines in the case of a severe public health threat. A critical factor in the production of pandemic influenza vaccines is the timely availability of suitable reference seed viruses. During our study, we successfully generated influenza vaccine seed viruses using both classical reassortment and reverse genetics techniques. Fortunately, the 2009 H1N1 influenza pandemic was less severe than anticipated. This pandemic provided a good opportunity to learn how to improve response to future outbreaks of viruses that may be more pathogenic. Our pilot study concluded that alternative strategies are necessary to rapidly produce a safe vaccine strain for mass vaccination purposes. Close collaboration between an academic laboratory and a vaccine manufacturer helped to significantly reduce the time required to produce seed vaccines. The availability of the BSL-4 laboratory during the development of vaccine seed stock enabled immediate response while adhering to the most stringent safety conditions. Moreover, BSL-4 containment ensures seed virus production under conditions that meet the quality assurance requirements necessary for human vaccines. No industrial BSL-4 vaccine manufacturing facilities currently exist. We believe that strategic alliances between academic or government-owned BSL-4 laboratories and vaccine manufacturers will become an essential component in the timely development of effective vaccines during pandemics caused by emerging pathogens.

5 Strecker et al.: Development of vaccine candidates against 1279 Acknowledgments The authors wish to thank all members of the Institute of Virology in Marburg and Novartis Vaccines and Diagnostics, who participated in the development of a vaccine seed virus candidate for combating the 2009 pandemic influenza virus. We are particularly grateful to P.R. Dormitzer for scientific advice, and G. Ludwig and M. Schmidt for expert technical support with regard to BSL-4 procedures. Conflict of interest statement Authors conflict of interest disclosure: The authors stated that there are no conflicts of interest regarding the publication of this article. The findings, interpretations and conclusions detailed in this paper are those of the authors. They do not necessarily represent the views of the corresponding affiliates. Research funding: We received financial support from the Philipps- University of Marburg and the Department of Medicine, and Novartis Vaccines and Diagnostics. Employment or leadership: CLB, HT, BR, and UV are employees of or shareholders in Novartis Vaccines and Diagnostics, which manufactures influenza vaccines using MDCK PF cells. Honorarium: None declared. References 1. WHO. World Health Organization guidelines on the use of vaccines and antivirals during influenza pandemics. Geneva: World Health Organization, WHO. Vaccines for pandemic influenza. Geneva: World Health Organization, Kodihalli S, Justewicz DM, Gubareva LV, Webster RG. Selection of a single amino acid substitution in the hemagglutinin molecule by chicken eggs can render influenza A virus (H3) candidate vaccine ineffective. J Virol 1995;69: Brands R, Visser J, Medema J, Palache AM, van Scharrenburg GJ. Influvac: a safe Madin Darby Canine Kidney (MDCK) cell culture-based influenza vaccine. Dev Biol Stand 1999;98:93 100; discussion Kistner O, Barrett PN, Mundt W, Reiter M, Schober-Bendixen S, Dorner F. Development of a mammalian cell (Vero) derived candidate influenza virus vaccine. Vaccine 1998;16: Genzel Y, Reichl U. Continuous cell lines as a production system for influenza vaccines. Expert Rev Vaccines 2009;8: Kilbourne ED. Future influenza vaccines and the use of genetic recombinants. Bull World Health Organ 1969;41: Fulvini AA, Ramanunninair M, Le J, Pokorny BA, Arroyo JM, Silverman J, et al. Gene constellation of influenza A virus reassortants with high growth phenotype prepared as seed candidates for vaccine production. PLoS One 2011;6:e Neumann G, Kawaoka Y. Reverse genetics systems for the generation of segmented negative-sense RNA viruses entirely from cloned cdna. Curr Top Microbiol Immunol 2004;283: Subbarao K, Katz JM. Influenza vaccines generated by reverse genetics. Curr Top Microbiol Immunol 2004;283: Neumann G, Watanabe T, Ito H, Watanabe S, Goto H, Gao P, et al. Generation of influenza A viruses entirely from cloned cdnas. Proc Natl Acad Sci USA 1999;96: Hoffmann E, Neumann G, Kawaoka Y, Hobom G, Webster RG. A DNA transfection system for generation of influenza A virus from eight plasmids. Proc Natl Acad Sci USA 2000;97: Suphaphiphat P, Keiner B, Trusheim H, Crotta S, Tuccino AB, Zhang P, et al. Human RNA polymerase I-driven reverse genetics for influenza a virus in canine cells. J Virol 2010;84: Murakami S, Horimoto T, Yamada S, Kakugawa S, Goto H, Kawaoka Y. Establishment of canine RNA polymerase I-driven reverse genetics for influenza A virus: its application for H5N1 vaccine production. J Virol 2008;82:

WHO biosafety risk assessment and guidelines for the production and quality control of human influenza pandemic vaccines: Update

WHO biosafety risk assessment and guidelines for the production and quality control of human influenza pandemic vaccines: Update 23 July 2009 Introduction This document updates guidance 1 from the World