Observations on Vaccine Production
Technologies and Factors Potentially
Influencing Pandemic Influenza Vaccine
Choices in Developing Countries
A discussion paper
World Health
Organization
SoulMiitAtiaR* WestwnhcMc~
� SEA-TRH-006
Distribution: Limited
Observations on Vaccine Production
Technologies and Factors Potentially
Influencing Pandemic Influenza vaccine
Choices in Developing Countries
A discussion paper
World Health
Organization
Regional Office for South-East Asia
�@ World Health Organization 2009
This document is not issued to the general public, and all rights are reserved by the
World Health Organization (WHO). The document may not be reviewed,
abstracted, quoted, reproduced or translated, in part or in whole, without the prior
written permission of WHO. No part of this document may be stored in a retrieval
system or transmitted in any form or by any means - electronic, mechanical or
other - without the prior written permission of WHO.
The views expressed in documents by named authors are solely the responsibility
of those authors.
Printed in India
� Contents
Page
Acknowledgments .................................................................................................. v
Acronyms ..............................................................................................................vii
Introduction...........................................................................................................
1
Background ...................................................................................................
2
Overview of influenza vaccine production technologies ................................4
Classic influenza vaccine produced in eggs ..................................................................... 4
Live attenuated influenza vaccine ................................................................................... 5
Influenza vaccines from cell culture ................................................................................ 7
Second generation biotech vaccines ............................................................................... 9
Candidate seed strains and antigens.............................................................................. 10
Issues and challenges...................................................................................11
The question of adjuvants............................................................................................. 11
Conditions imposed on commercial use of reverse genetics ........................................... 14
Biotechnology and public perception ........................................................................... 15
Export controls ............................................................................................................. 17
Options .......................................................................................................
19
Timing and technology choices..................................................................................... 19
Fill/finish projects and importation of bulk antigen ........................................................
21
The option of animal vaccine plant conversion.............................................................. 22
Concluding discussion ...................................................................................24
Annexes
1. Overview table of influenza vaccine technologies ........................................
27
2. Relevant reports available online .................................................................
31
Page iii
� Acknowledgments
This paper has been written by Edward Hammond for the WHO Regional
Office for South-East Asia. It is intended as a contribution to the debate on
the sharing of influenza viruses and access to vaccines and other benefits
arising from their commercial exploitation, and to efforts to move forward
the issues raised by resolution WHA 60.28.
Page v
� Acronyms
CBW chemical and biological weapons
DNA deoxyribonucleic acid
GAP global action plan
GISN global influenza surveillance network
IGM intergovernmental meeting
IIV inactivated influenza vaccine
LAIV live attenuated influenza vaccine
MTA material transfer agreement
PIP pandemic influenza preparedness
RNA ribonucleic acid
TRIPS (Agreement on) Trade-Related Aspects of Intellectual
Property Rights
VLP virus-like particle
WHO World Health Organization
Page vIi
� Introduction
As a result of concerns raised over the sharing of influenza viruses and the
lack of affordable vaccines and medicines, the Pandemic Influenza
Preparedness (PIP) Intergovernmental Meeting (IGM) is discussing the
possible establishment of a new system for sharing of potentially pandemic
influenza viruses, a well as sharing of the benefits resulting from research
s
utilizing them.
Among the possible benefits being discussed is expanded transfer of
vaccine-related technology to developing countries, and a sustainable
financing mechanism for developing country pandemic preparedness.
WHO Member States hope that this financing and technology transfer
would help close the gap between pandemic vaccine supply and demand.
But what specific technological approaches are best suited for
developing countries? Influenza vaccine technologies need to be
categorized and assessed for their costlbenefit implications and respective
tradeoffs and risks. Not all technologies are freely available or equally easy
to use, so this codbenefit assessment needs to be made in the light of the
constraints imposed by intellectual property claims as well as "hard"
technology and know-how requirements.
Other important considerations, including export controls and
regulation of biotechnology, remain underexplored, but may influence
decisions by developing countries with respect to a possible PIP IGM
benefit sharing system.
This paper discusses these issues in five sections. Section I provides a
short background. Section II describes briefly the main technologies that are
currently available or that are under development, as well as their
comparative advantages and potential challenges1. Section Ill discusses a
number of cross-cutting issues of practical significance (adjuvants,
conditions related to seed strains, public perception of some of the
technologies, and export controls) that lie outside the production-related
questions, but that nevertheless need to be addressed. Section IV considers
various options. Finally, Section V contains some concluding remarks.
' A table summarizing key features of the various technologies is attached as Annex 1.
�A discussion paper
Background
Ensuring adequate availability of pandemic influenza vaccines is not an easy
task in any country of the world, and no single solution will be universally
appropriate. Limited global production capacity for human influenza
vaccines is the result of limited demand for seasonal influenza vaccines and
technical challenges to influenza vaccine production. Adding to the
difficulty is a recent sharp increase in patents and patent applications
related to influenza vaccines, which may impede access to vaccine
production technologies.
Pandemic preparedness efforts cannot be considered in isolation from
other public health concerns and must be weighed in the context of
programmes to address other priorities, complementing them when
possible. For example, the infrastructure to produce some types of
influenza vaccine is useful for making other kinds of vaccines, yet
paradoxically, the flu vaccine technologies that are most adaptable may be
the most expensive and technologically-challengingto utilize, as well as the
most impacted by intellectual property claims.
Some have proposed to expand seasonal influenza vaccination in
order to expand pandemic production capacity. This strategy is a key part
of the WHO Global Action Plan (GAP), whose overarching goal is to
increase pandemic influenza vaccine supply by stimulating demand for
seasonal influenza vaccines. Greater seasonal demand, it is reasoned, will
stimulate the private sector and others to construct additional influenza
vaccine production capacity that can then be used in a pandemic.
But in many countries, and especially developing countries, there is
low demand for seasonal flu vaccines and limited prospects of expanding it,
particularly among citizens in lower economic strata with competing health-
care priorities. The cost of implementing the GAP, even with optimistic
economic and antigen assumptions, is estimated to rise to US$ 3.5 billion
to US$ 5 billion annually by 2012, with an emphasis on spending in
developed countries to stimulate demand there, and the questionable
assumption that excess pandemic vaccine will quickly be used to vaccinate
those in other countries.'
* WHO IVR. The Global Action Plan (CAP) to Increase Supply of Pandemic Influenza Vaccines, First
Meeting of the Advisory Croup, WHO/IVB/08.10, 19 October 2007, Geneva.
Page 2
� Observat ions on Vaccine Production technologies and Factors Potentially Influencing
Pandemic Influenza Vaccine Choices in Develooine Countries
It is unwise not to squarely recognize the limitations on seasonal
influenza vaccine demand and the great challenges facing the GAP. Even in
developed countries where demand and income are higher, and despite
hefty economic stimuli, manufacturers currently are hesitant to expand
production capacity. This is in large part due to limited seasonal vaccine
demand. For instance, a large European manufacturer recently backed out
of an agreement to build an influenza vaccine facility in the United States
because it said that a US$ 298 million government subsidy was in~ufficient.~
Others have proposed emergency conversion of animal vaccine plants
if a pandemic strikes, particularly of poultry vaccine facilities with egg-based
production systems that probably can be adapted to produce human
influenza vaccine. With global human influenza vaccine production
capacity at least 70% short of providing vaccination for the global
population within six months of a pandemicI4 this suggestion makes
obvious sense. Where such capacity exists, this could expand pandemic
vaccine supply, but there are significant technical and safety hurdles.
Another strategy that has been proposed is to concentrate vaccine
antigen production in a small number of developed countries, on the theory
that making vaccine antigen is best done in a few expert facilities and that, if
these facilities are collectively made large enough, their surplus production
can be exported to developing countries in the event of a pandemic. Yet this
strategy, encouraged by the WHO GAP, leaves developing countries in a
state of dependency and at the end of the queue to receive vaccine.
All of the above factors, together with mounting pressure on health
budgets as a result of the global economic downturn, make ensuring
availability of influenza vaccines particularly difficult for most developing
countries.
Several studies have recently discussed options for expanding
prepandemic and pandemic influenza vaccine production capacity. A
number of these reports are listed in the annex to this report. While these
are valuable and discuss some technical aspects of influenza vaccines in
greater detail, there are key issues related to pandemic vaccination
strategies that remain under-contextualized for policy-makers. This paper
seeks to fill that gap.
McKenna M. Plant cancellation shows problems in flu vaccine business in CIDRAP News, 3 Ocl. 2008
WHO. Business Plan for the Global Pandemic Influenza Action Plan to Increase Vaccine Supply,
February 2008.
Page 3
�A discussion oaoer
This paper assumes that developing countries will largely not be
satisfied with reliance on pandemic vaccines and/or bulk antigen exported
from Europe, North America, or Japan, particularly because such supplies
currently cannot be made available in a timely fashion. Therefore is difficult
to argue that such reliance is an adequate pandemic vaccine supply plan.
Rather, here it is presumed that developing countries will continue to seek
the development of national or regional 'vaccine production capacity
through technology transfer and sharing of benefits of influenza research.
2. Overview of influenza vaccine production
technologies
There are a variety of technologies that are used or have been proposed for
production of human influenza vaccines. Often, significant parts of the
production process are similar. This is especially true in the later stages of
manufacture, such as packaging. The technologies may be categorized in
several ways. Below, they have been divided into four basic technological
approaches.
Classic influenza vaccine produced in eggs
With few exceptions, currently available seasonal and prepandemic
influenza vaccines are manufactured through egg-based production
methods. The system is cumbersome and inefficient in comparison to the
theoretical possibilities of newer cell-based production (see below), leading
some to characterize egg-based production as antiquated. Such
comparisons, however, are invariably made against technologies that have
yet to be fully commercially deployed and proven. Moreover, although it
may not be new, this decades-old technology is relatively cheap, very well
proven, and largely unencumbered by intellectual property claims.
Egg-based production is employed throughout the world for animal
and human vaccines. Apart from influenza, however, the only human
vaccines for which the egg-based system is utilized are yellow fever and
Japanese encephalitis vaccine. This means that apart from making flu
vaccine, egg-based production lines have limited broader utility for human
public health.'
Egg-based lines arc important for animal heakh, however, as discussed below.
Page 4
� Observations on Vaccine Production Technologies and Factors Potentially Influencing
Pandemic Influenza Vaccine Choices in Developing Countries
Egg-based production requires a supply of fertile chicken eggs
produced under relatively stringent conditions (in comparison to eggs
produced for food consumption). This is to ensure that they do not carry
pathogens that might taint the vaccine. The eggs are infected with a vaccine
strain and the fluid harvested from them yields vaccine after separation and
further production steps.
The reluctance of H5 viruses to grow to high titer in eggs (because the
virus strains are too efficient at killing chicken embryos) is a problem that
has bedeviled H5 vaccine development. While this remains a significant
technical challenge, the problems with growing H5 viruses in eggs are being
overcome, mainly by attenuating the hemagglutinin (HA) gene of the
vaccine strain, typically through reverse genetics (see below). It may be
noted that some of these techniques are proprietary however.
Major requirements of the egg-based production system include the
process of "candling" the eggs (inspection under bright light); equipment to
inoculate the eggs with virus; incubators in which to keep the eggs while
the virus is reproducing; and equipment to harvest, separate, and purify the
vaccine virus after incubation.
Some of the technology required to produce the vaccine strain is
specialized; however, none of it is reported to be particularly expensive,
complicated or difficult to operate. In the newest facilities the entire
process is automated, while in others some steps in production (for
example, candling and harvesting) are conducted by human technicians.
Later steps of egg-based vaccine production, including formulation
and packaging, may be similar or identical to the process used with other
technologies.
Live attenuated influenza vaccine
Live attenuated influenza vaccine, abbreviated "LAIV", is an influenza
vaccine production technology in limited commercial use in the Russian
Federation and in the United States. LAIV offers the possibility of producing
significantly more vaccine than classic egg-based production using same
production line; however there are significant additional scientific and
intellectual property hurdles that may reduce LAIVis attraction for
developing countries.
Page 5
�A discussion oaner
The production process for LAIV vaccines is similar to that of classic
egg-based vaccines, with some notable exceptions. LAIVs are administered
live. This means that when the vaccine strain-containing fluid is harvested
from eggs, it is not exposed to a detergent. Thus, if adventitious pathogens
are present in the eggs, these may survive the formulation process and
eventually infect human vaccine recipients. Therefore, eggs used in LAIV
production may require even higher production standards than those used
to produce classic killed vaccine. This increased danger of contamination
means biosafety practices in production need to be more stringent than
those used for classic killed vaccine.
While there are a number of well-characterized backbone strains6
available for classic vaccines, the "cold-adapted"' backbones used in LAIVs
are proprietary, such as the "Ann Arbor" strain used in the United States
and the "Leningrad" strain used in the Russian Federation. LAIVs thus
require a proprietary backbone strain6 and cannot be produced using the
vaccine seeds strains currently distributed by WHO global influenza
surveillance network (which do not have "cold-adapted" backbones).
Harvesting is simpler for LAIVs (no detergent wash is needed), but the
final product is more delicate because the live vaccine must be kept viable
"alive") until it is used. This means LAIVs require cold storage.
The fact that LAIVs are not killed potentially offers a major advantage
over classic vaccine, but at a cost. LAIVs reproduce in the body of
immunized persons; thus, they effectively act as their own adjuvants, which
means they should require a lower dose of antigen than killed vaccine. This
means the same production line may yield considerably more LAIV than
classic killed vaccine, although estimates of the increased yield vary widely.'
Influenza vaccines typically are comprised of a(n) HA gene(s) taken from a viral isolate that is inserted
into another, laboratory-adaptedstrain by reassortment or recombinant (reverse genetics) means.
While the immunogenic HA gene is the most important part of the vaccine, the labadapted strain into
which it is placed has typically been selected for useful characteristics for lab and industrial use (high
growth rate, tolerance for lab conditions and temperature ranges, etc). This labadapted strain is called
the "backbone" strain.
' "Cold adapted" influenza strains are laboratory-adapted types that are suitable for use in live vaccines,
which must be kept cold until use in order to maintain the vaccine's viability.
' Discussions to license the Russian "Leningrad" LAIV strains for H5 vaccine production .ire taking place,
however, no detailed information concerning the terms and restrictions of any possible license is
available, and no final agreement has been reported to have been reached.
' The WHO CAP estimate is 4.5 times, whereas others have estimated a yield as high as 10 limes that of
'
the c:lassic trivalent killed vaccine process.
Page 6
� Observations on Vaccine Production Technologies and Factors Potentially Influencing
Pandemic Influenza Vaccine Choices in iheloping Countries
One price of this antigen efficiency is that LAIVs are administered as
an intranasal aerosol (i.e. sprayed into the nose), rather than being injected.
They thus require a special closer instead of standard syringes. Sufficient
supplies of this doser are required in order to use LAlVs.
Another limitation of LAIVs is that they are unsuitable for
prepandemic vaccines because of the possibility that the live prepandemic
vaccine strain could mutate or recombine with circulating strains,
potentially causing or contributing to a new epidemic or even pandemic flu
strain. While this concern is not applicable to seasonal vaccines (because of
the antigens used), it does seriously limit the ability to test LAIV procedures
and formulations prior to an actual pandemic.
Of note, in the future it may become practical for LAIVs to be
produced in cell culture (see below), although at present they are produced
in eggs.
Influenza vaccines from cell culture
Influenza vaccines produced by cell culture are currently under
development in several places but so far are not produced commercially on
a large scale. In the cell culture process, animal or other cells are infected
with a vaccine virus, which is then harvested and formulated into vaccine.
The process takes place in vessels called bioreactors (or fermenters), in basic
design not dissimilar from those used in brewing.
Cell culture typically starts by growing cells in a nutrient-rich fluid in
small containers, scaling up to larger ones as the cells reproduce. When the
desired cell density and scale is reached (hundreds or thousands of litres for
commercial production), the cells are infected with vaccine strain virus.
After the virus reproduces, the cells are harvested and virus processed into
vaccine.
In some cell culture systems, gently agitated cells grow freely in a sort
of "soup" mixed with nutrients and (eventually) with vaccine virus. In other
cell culture systems, the cells grow affixed to a substrate such as tiny gold-
coated beads. They are then released by agitation.
--
Page 7
�A discussion oamr
For large-scale commercial production, the process requires large
bioreactors, from hundreds to thousands of litres in size. Production of cell
culture vaccines also requires equipment t o build and maintain a "cell
bank" to provide a new supply of fresh identical cells after batches of virus-
infected cells are harvested.
Cell culture systems are likely to be more flexible than egg-based
systems for production of other human vaccines, potentially increasing a
facility's utility. For H5 influenza viruses in particular, there are claims of cell
culture systems that grow the virus to a higher titer than is possible in eggs.
Although cell culture vaccines are a major focus of research and
development (R&D), as yet they remain in limited commercial use.
Scientific limitations for their use in flu vaccines include the inability to be
certain ahead of time that a particular cell line will be appropriate to grow
the pandemic strain, and the need for substantial bioreactor capacity, of
which there is little to no global surplus. Although investment may be
recouped through a multi-use facility, cell culture has considerably higher
facility construction costs at an industrial scale.
In addition to potential patent claims over the influenza genes (which
also impact egg-based vaccines) and backbones used in vaccine strains,
there are additional intellectual property issues related to cell culture
influenza vaccines. The cell lines that are used are themselves often
patented, and the information necessary for their use and for regulatory
approval is proprietary.
Table 1 Examples of proprietary cells lines used in cell culture vaccine production
:
Avian embryonic stem cells Vivalis (France) Licensed to Novartis,
GlaxoSmithKline, and others
Vero cells per se are not
cells (Vero) proprietary, but Baxter's
process of using them is.
To date, few cell culture-produced vaccines have been approved for
human use, and they are likely to prompt more intense regulatory scrutiny
than egg-produced vaccines. Cell culture vaccines require approval for the
vaccine as well as characterization and safety demonstration of the cells used.
Page 8
� Olwervations on Vaccine Production technologies and factors Potentially Influencing
Pandemic Influenza Vaccine Choices in Uwelooinc Count rics
Second generation biotech vaccines
If cell culture vaccines, because of their sophisticated biological
manufacturing process, may be considered the first generation of
biotechnological flu vaccines, then a basket of different technologies currently
under development could constitute the second. While supporters of these
technologies believe they may be useful for pandemic influenza vaccines,
several of them are at early stages of development and none are proven and
ready for commercial use. Therefore, although it is difficult to generalize
about these new technologies, they are unlikely to be selected in the short
term for pandemic vaccine production in developed or developing countries.
These technologies depend on the WHO global influenza surveillance
network (GISN) for antigens and WHO selections of the best antigens for use
in vaccines, but do not utilize WHO candidate seed strain.
Taking a longer a view, however, it is possible that some of these
technologies, for example virus-like particles (VLPs), may become viable for
large-scale use. The fact that they are new and mainly privately developed,
however, means that in general they are heavily covered by intellectual
property claims and may require very new kinds of know-how. For most
countries it is too early to tell, however, if national patents will be issued, so
the extent of intellectual property impediments for any particular
developing country remains unclear.
Briefly, second-generation biotech vaccines include, among other
approaches:
> production of recombinant HA protein in other, easily grown,
organisms (e.g. transgenic bacteria);
> "naked" and plasmid DNA vaccines in which "codon
~ p t i m i z e d ' " flu genes are used directly as vaccine, and
~
>- genetically engineered systems to co-express HA, NA, M2
genes from flu, manufacturing a "virus like particle" (VLP) that is
purified from culture and used as vaccine.
lo "Codon opiiniized" genes have nucleic acids that have been altered-typically changed from RNA to
DNA-so 1h.11ihe gcnr can be he~terexpressed in a biotechnological application (e.g. . v,iccinc').
I
Page 9
�A discussion paper
Summary of the basic technological approaches to influenza vaccine
production
1. Egg-based "classic" influenza vaccine: Vaccine virus is injected into fertilized
eggs. The eggs are placed in incubators and the virus reproduces in the eggs. Fluid
is then harvested from the eggs and washed with detergent The resulting killed
virus material is separated and used for vaccine formulation. This type of vaccine is
one kind of inactivated (i.e. killed) influenza vaccine, or "11V".
2. Live attenuated influenza vaccine ("LAIV"): Vaccine virus is grown in eggs (or
in the future, potentially in cell culture) in a process similar to classic flu vaccine.
The live virus uses a special type of genetic backbone (currently of limited
availability since they are proprietary). Harvesting and formulation is simpler than
with killed vaccines. The final product is more delicate and requires a cold chain,
but the process potentially is considerably more efficient, producing more flu shots
with the same number of eggs.
3. Cell culture influenza vaccines: Mammalian, avian, or other cells are cultured in
growth media. This culture is scaled up to the desired density of cells in large
bioreactors (fermented up to thousands of litres in capacity. The culture is infected
with vaccine strain, which multiplies in the cells, producing large quantities of vaccine
virus. Harvesting, purification and packagingare essentially the same as with egg-
based vaccines. This is another type of IIV, produced by a different method.
4. "Second generation" biotechnologicalvaccines: Many techniques are under
study, including: producing recombinant HA protein in other, easily grown,
organisms (e.g. transgenic bacteria); "naked* and plasmid DNAvaccines in which
"codon optimized" genes are used as vaccine; and genetically engineered systems
to co-expressflu genes, making a virus-like particle (VLP) that is used a vaccine.
s
Candidate seed strains and antigens
The W H O system develops and distributes candidate H5 vaccine seed
strains. he& seed strain; are suitable for producing vaccine in eggs and
incorporate antigens that have been selected by WHO. In the event of a
pandemic, the W H O system may develop and make available LAIV-
suitable seed strains; however, W H O does not presently have rights to the
proprietary LAIV backbones.
Although at a technical level, current W H O candidate seed strains can
be used to produce H5 vaccine, there are legal restrictions imposed on
them in a required Material Transfer Agreement (MTA).ll This is because
l1 For example, the Material Transfer Agreement for the WHO candidate seed strain NIBRG-23, made
from an HSN1 strain isolated in Turkey, can be viewed here:
h~p://www.11ibsc.ac.uk~flu_site/Docs/spotlighl/H5N1MTA-NIBRG-23.doc
Page 10
� Observations on Vaccine Production Technologies and Factors Potentially Influencing
Pandemic Influeni'a Vaccine Choices in Developing Countries
they are created using proprietary reverse genetics technology. (See also the
discussion on conditions imposed on commercial use of reverse genetics in
section Ill below.)
The advantage that the WHO seed strains theoretically offer is that the
strain is a known quantity that may be quickly used, reducing the amount
of work and time needed for new strains to go into production.
Vaccine makers and other companies, however, may choose not to
use WHO candidate seed strains for any of several reasons. These may
include a desire to avoid the intellectual property restrictions imposed by
the WHO MTA, or they may wish to use a technology type for which the
WHO strain is not suitable (e.g. LAIVs), or they may wish to make other
alterations particular to their production system (for example, to introduce a
mutation intended to make the virus grow to higher titer).
If a vaccine maker does not use the WHO candidate vaccine seed strain
in actual production, however, it is still highly likely to use the antigens selected
by the WHO system C
SImost immunogenic. In this case, the maker would
obtain the HA (andlor NA) gene(s) from the WHO system or synthesize them
from sequence data. The maker then incorporates the WHO-selected
antigenk) into its own vaccine strain. Thus, particularly in the future, of
arguably even greater importance than the WHO candidate vaccine seed
strain are the genes that the WHO system determines to be most suitable for
use in vaccines, because these will be used by manufacturers whether or not
the manufacturer utilizes the WHO candidate seed strain.
3. Issues and challenges
The question of adjuvants
Adjuvants are substances that are added to a vaccine in order to enhance
its immunological effect. Most adjuvants act on the human immune system
and are not linked to a particular vaccine strain or even a particular disease.
Thus a particular adjuvant may be used not only for influenza vaccine; but
also in vaccines against other diseases.
Adjuvants can both reduce the amount of antigen needed per vaccine
dose (potentially of great importance in a pandemic) and increase the
"take" of vaccines-lhat is, the rate of successful vaccination.
Page 1 7
�A discussion paper
Many adjuvants that may be used in influenza vaccines today are
inorganic chemicals. These are sometimes aluminum-related compounds,
such as aluminum hydroxide (or gibbsite, (Al(OH)J), which is more familiar
in medicine for its use as an oral antacid. One adjuvant that has long been
used, alum, is patent-free and easily obtained, but it is not generally
considered promising for H5 vaccines. ,
Major influenza vaccine manufacturers are increasingly using newer
adjuvants of a type called oil-in-water emulsions. Companies claim these
offer substantial improvements over other adjuvants. The proprietary oil-in-
water adjuvants used by Novartis and Gla~oSmithKline'~ based on
are
squalene, an organic compound produced in small quantities by many
animals and some plants, and are subject to patents and trade secrets.
A large number of biotechnological adjuvants, such as short pieces of
DNA that are active in the body and are designed to make vaccines more
immunogenic through specific gene or protein-level effects on the immune
system, are undergoing research. These, however, remain experimental.13
Not all vaccines contain an adjuvant. LAIVs do not need to be
adjuvanted because they are alive and reproduce in the upper respiratory
tract. One H5 vaccine, produced in cell culture by Baxter International, is
a killed virus vaccine that is unadjuvanted.14
One problem with assessing the potential use of adjuvants for
pandemic vaccine production in developing countries is that they are often
highly proprietary. For instance, detailed information on production of
vaccines with oil-in-water emulsion adjuvants is limited, as the adjuvants
are often patented and their use is covered by trade secrets.
Table 2 provides an overview of a number of adjuvants.
Sanofi's
'¥ proprietary formulation is reportedly similar, but its exact composition does not appear to
have been made public.
" Because these adjuvants are varied in nature and generally in earlier development stages, this paper
focuses on adjuvants in current use or advanced development
l4 The Baxter vaccine has an unusual composition and production method. It uses unaltered H5N1 virus
isolates that have not been placed on a labadapted backbone or had genetic alterations to reduce
pathogenicity. Because the live vaccine virus is virulent for birds and, potentially, humans and other
animals, it must be grown under very careful biosafety procedures in P-3 (BSL-3) containment. This
method of production requires a cell culture system, with the added challenge of stringent BSL-3
practices and facilities.
Page 12
� Observations on Vaccine Production Technologies and Factors Potentially Influencing
Pandemic Influenza Vaccine Choices in DevelopingCountries
Table 2: Adjuvants that may be used in pandemic influenza vaccines
What is it? An inorganic Chemicals An oil-in- An oil-in- An oil-in- Biological
chemical, related to water water water materials
potassium alum, emulsion, emulsion, emulsion, designed to
aluminum including consisting of consisting of whose boost
sulfa~e. aluminum squalene, squalene, formulation immune
hydroxide polysor'Jate plysorbate does not syst-
and 80 (Tween 80, and DL- appear to response.
aluminum 801, and a- have been
phosphate. sorbitane tocopherol. published.
trioleate
(Span 85).
Yes 1 yes 1 Yes. Includes
JVRS-100
Avenlis) fluventus),
(Intercell) etc.
Use Has long Clinical trials Used in Used in Does not Experimental:
been used in are underway vaccines vaccines currently some have
various of pandemic licensed in licensed in appear to be advanced to
vaccines. flu vaccines some some licensed. human trials.
utilizing countries. countries. Regulatory
these. Also hurdles likely
used in other to be quite
vaccines. substantial.
Efficacy issues Used in trials Potentially Deemed Deemed AF03 is Unproven.
and in one more effective and effective and thought to
US-licensed effective than licensed for licensed for be similar to
prepandemic alum, but less use in non- use in non- MF59 and
vaccine; but so than influenza influenza ,4503.
often proprietary vaccines.
regarded as adjuvants.
inadequate Mixed results
for use with in research to
' H5 vaccines. date.
Based on their reported composition, adjuvants such a MF59 do not
s
appear to utilize unusual or expensive ingredients; however, it cannot be
assumed that effectively incorporating them into vaccines is as
straightforward as their reported chemical composition because details of
their use are proprietary.
In some countries, vaccination has been associated with social
controversies due to perceived risks. Some vaccine critics have claimed that
certain adjuvants are unsafe, including aluminum hydroxide (alleged to be
Page 13
�A discussion caper
linked to Alzheimer disease) and MF59 (which has received scrutiny for its
use in a controversial US anthrax vaccine). While the scientific merit of
these criticisms is debated-the compounds have passed regulatory review
in many countries-where concern exists it would be inappropriate to ignore
the potential disruption to vaccination campaigns due to widespread worry
over adjuvant safety.
It is clear that in the event of a pandemic, the presently limited global
vaccine virus production capacity means that the supply of pandemic
vaccine antigen (in any form) will be far outstripped by demand, especially
in the early stages. With the exception of unadjuvanted LAIVs, in the
dominant planning scenario, widespread use of the most effective adjuvants
is highly desirable because it will enable more people to be vaccinated with
the limited amount of antigen available, especially at earlier stages of the
pandemic. Failure to use the most effective adjuvants would "waste"
antigen because each suboptimally adjuvanted dose would "rob" antigen
from the global supply.
Conditions imposed on commercial use of reverse genetics
Reverse genetics is a relatively new proprietary technology that is being
applied to the development of influenza vaccines as well as other products. At
present the technology is used in the creation of WHO GISN H5 vaccine seed
strains, although it is not strictly technically obligatory to use it when making
pandemic vaccine strains. Because of the advantages it offers, however, the
technology will likely be increasingly used in future vaccine strains.
Primarily developed by American and British universities, and covered
by a large number of patents, reverse genetics intellectual property has
been accumulated by Medimmune, a US-based subsidiary of the United
Kingdom's Astra Zeneca, a large flu vaccine maker. Meclimmune has thus
far allowed use of its reverse genetics intellectual property in pandemic
vaccine R&D, however, it has indicated that it will not permit commercial
use of the technology without a license.
Material transfer agreements for WHO candidate seed strains of H5
vaccines thus include protections for Medimmune's intellectual property and
thereby impose restrictions on those that receive seed strains (through contract
law), even in countries where Medimmune's patents have not been issued.
Reverse genetics technology involves creation of loops of DNA called
plasmids whose key parts encode for influenza genes. When the plasmids
Page 14
� Observations on Vaccine Production Technologies and Factors Potentially Influencing
Pandemic Influenza Vaccine Choices in Developing Countries
are introduced into cells, the DNA is transcribed into RNA and influenza
virus is produced. The technology enables scientists to "edit" the influenza
viral genes by making alterations to the DNA plasmid, for example, deleting
bases from the HA gene to make the virus avirulent.
In addition to allowing manipulation of individual genes, reverse
genetics allows scientists to relatively easily mix and match genes from
different influenza strains, particularly when inserting new genes onto
'backbone" strains for which plasmid systems are already constructed. This
is useful for research purposes and for creation of vaccine strains, because it
can be more straightforward and reliable than the traditional reassortment
method, whereby cells are coinfected with different strains and the resulting
hybrid viruses identified and selected by scientists.
Reverse genetics is potentially a very useful technology for egg-based,
cell culture, and other types of flu vaccines. It is, however, controlled by
Medimmune and because it is used in current WHO candidate seed
strains, recipients of those strains are already obligated to negotiate with
Medimmune should they choose to commercially produce vaccine from
those strains. This point has perhaps not received the attention it warrants.
Biotechnology and public perception
An important policy and health consideration underappreciated to date is
the potential for problems with social and regulatory acceptance of
recombinant pandemic influenza vaccines-that is, those that are the
product of biotechnology. Some countries may have additional regulatory
requirements for such vaccines. This may influence the decisions that
governments take in vaccine supplies. Decisions may be complicated by
the fact that influenza vaccines make use of biotechnologies that might or
might not be popularly and legally understood as "genetic engineering".
It is logical that in the event of a severe pandemic the vast majority of
people would opt for vaccination even if concerned about the safety of a
recombinant vaccine, for the simple reason that fear of severe illness or
death from the disease is greater than concern about the vaccine. It is also
true, however, that genetically engineered products used in humans remain
controversial in many parts of the world and some citizens may be reluctant
to be vaccinated, particularly in scenarios such as a slow-spreading
pandemic or widespread use of a recombinant (pre)pandemic vaccine.
Page 7 5
�A discussion paper
Although not strictly tied to biotechnology, recent cases of problems
in polio vaccination campaigns and the rejection of childhood vaccination
among some religious communities are evidence of the importance of
safety perceptions and belief. In the case of pandemic influenza vaccines,
the degree to which the vaccine could be termed "genetically engineered"
varies by the technology used. Perceptions may be further influenced by
other factors, such as use of animal products in cell culture, and whether
the vaccine is live or killed, with killed vaccines presumably engendering
less resistance.
A brief breakdown of some pertinent influenza vaccine technologies
and how they might be consideredis given in Table 3.
Table 3: Brief overview of key influenza vaccine technologies
lechmtogy - What is it?- Islt-geneticc~ng?
Reverse genetics Assenlbly of influenza viruses through Viruses produced by reverse genetics are
the creation of DNA plasmids bearing recombinant products and are, as it is
influenza genes that are transcribed into generally understood (and regulated),
virus in infected cells. Although not genetically engineered. If the virus genes
strictly necessary for most influenza have not been significantly changed,
vaccines, it may offer time savings and however, then the resulting vaccine virus
other R&D advantages. may not substantially differ from
reassortant viruses or natural virus isolates.
HA gene deletions To facilitate safe handling of H5N1 The manipulation of the HA gene creates
research viruses and vaccine production. a recombinant product. The modified HA
part of the HA gene is deleted to make gene is not transgenic, however, because
it nonpathogenic. This altered gene is it does not incorporate foreign genetic
then used in the vacane strain. material.
Virus-like particles (VLPs) Insertion of nucleic acids coding for The VLP vaccine itself is non-living;
influenza virus genes into other cells, however, i t is the product of an organism
triggering the production of non-living that is genetically engineered to express
particles that mimic key parts of non-native genes.
influenza viruses, and can trigger an
immune reaction.
Recombinant LAIV While it is possible to create LAIVs A live genetically engineered vaccine is
without use of recombinant DNA, lor the type most likely to encounter stricter
technical reasons i t is likely that a regulatory requirements and safety
pandemic LAIV would be produced questions.
with reverse genetics and possibly
incorporate additional genetic
modifications.
For many of the same reasons as LAIVs, These vaccines will contain a genetically
(pre)pandemickilled flu vaccines, engineered product Regulatory and social
produced in eggs or cell culture, may be concerns may be fewer, however,
recombinant products. because the vaccine virus is killed before
administration.
Page 16
� Observations on Vaccine Production Technologies and factors Pofentially Influencing
Pandemic Influenza Vaccine Choices in Revelopinc Countries
Export controls
Export controls are imposed by national laws. They are designed to regulate
and sometimes prevent the transfer of technologies (hat may be used to
create nuclear, chemical, or biological weapons as well as certain other
items, such as missile-related technology. They are discussed in particular
here because they have generally not been discussed with respect to
pandemic vaccine production to date.
Export controls are necessary to consider because research on highly
pathogenic influenza viruses and production of vaccines require facilities,
know-how and equipment that could be abused in biological weapons
programmes. As a result, some of the same technologies that can be used to
protect public health by producing vaccines can be difficult to acquire
because they may fall under export control laws.
Biological export control laws are controversial and have been a
matter of intense debate at the Biological and Toxin Weapons Convention.
The countries that impose the most rigorous export controls (mainly
developed countries) argue that they are necessary for national security and
anti-proliferation reasons. On the other hand, the countries that are most
often denied technology (mainly developing countries) counter that export
controls are arbitrary and unfair, and that they are often motivated by
political or economic considerations not related to weapons proliferation.
Export controls are not governed by any international agreement.
Some countries that have biological (and chemical) export control systems
attempt to coordinate them through the Australia Croup, a collection of
countries whose stated aim is "to minimise the risk of assisting chemical and
biological weapon (CBW) proliferation".
The majority of the members of the Australia Croup are OECD
Member States. The Croup calls itself an "informal arrangement" that
"meets annually to discuss ways of increasing the effectiveness of
participating countries' national export licensing measures to prevent
would-be proliferators from obtaining materials for CBW programme^".^^
'" See hltp://www.ai~straliagroup.net.
Page 7 7
�A discussion paper
For influenza vaccines, export control laws may limit the transfer of a
wide variety of research and vaccine production-related technology, and
even shipments of vaccines themselves.16
Export controls are applied to equipment, organisms, and ideas. The
different types of items that can fall under export controls,include:
> Physical items used in research and vaccine production such as
bioreactors (fermenters), lyophilizers (freeze dryers), separation
and packaging (filling) equipment.
> Know-how such a blueprints, design and engineering services
s
for high-containment laboratories and biological production
facilities, as well as certain kinds of scientific procedures and
knowledge.
> Biological materials-for example, highly virulent disease strains
or, in some cases, vaccines.
Export controls apply in different degrees to different countries and
technologies. Items considered by export-controlling countries to be of
highest risk1' may be more difficult to export than items that are considered
lower risk (for example, vaccines). Generally, when an export license for a
controlled item (or technology) is sought, the item is classified for its
intrinsic risk and then cross-referenced against a list of countries that
themselves have been categorized according to the degree of weapons
proliferation threat they are alleged to impose.
An additional pertinent consideration may be the entity in the
importing country that seeks access to the technology. For example, a well-
known international pharmaceutical company may be less likely to be
denied an export controlled item than a government research institute in
the same country, if the exporting county is suspicious of the aims of the
importing country's government research programme.
Finally, when export licenses are issued, typically they are contingent
upon the recipient of the controlled items agreeing to no further transfers of
"' A particularly severe export conlrol has recently been highlighted in news articles pointing out that
export controls in the United States would apply to H5N1 vaccine exports to several countries. See,
for example, URL: http://www.exportlawblog.com/archives/406(accessed 25 November 2008).
l7 For example, a large, high-quality fermentcr, which might be used to produce biological weapons
agents instead of vaccine.
Page 7 8
� Observations on Vaccine Production Technologies and Factors Potentially Influencing
Pandemic Influenza Vaccine Choices in Developing Countries
the technology. While as practical matter this type of re-export restriction is
difficult to enforce, entities that transfer export-controlled technologies
place at great risk their future ability to obtain export-controlled
techn~logies.'~
While the Non-Aligned Movement and others have been critical of
the Australia Group's biological export control system,'' there are no signs
that export controls are being relaxed even with the prospect of an
influenza pandemic. Countries must therefore take into consideration the
issue of export controls when making pandemic preparedness decisions.
Many developing countries are subject to Australia Group's export control
restrictions, which could impede their access to influenza vaccine
production technology.
The impact of export control regimes will vary by country and
technology. While export controls will not be a major issue for all countries,
particular technologies, such as cell culture systems, may be more prone to
export control problems than others. Countries that wish to develop a
domestic production capacity that utilizes imported technologies will need
to address these issues.
4. Options
Timing and technology choices
It is difficult to reconcile the severity of fears of an imminent pandemic with
the slow pace of expansion of global influenza vaccination and vaccine
production capacity. Years of meetings and rhetoric have passed since the
H5 pandemic scare began, yet most countries in the world-including many
wealthy countries-have thus far not ensured pandemic vaccine supplies for
their own populations.
18 Countries that impose export controls maintain lists of commercial, governmental and other entities
that have received (or sought to receive) export-controlled items for transfer to others without the
approval of the original exporting country.
l9 See, for example, the statement of Cuba (on behalf of NAM) and other statements at the 2007
Meeting of States Parties of the Biological and Toxin Weapons Convention, URL:
http://www.opbw.org/newprocess/msp2007/msp2007_s>a~emcn~.htm
�A discussion oamr
If the pandemic threat is so dire, why is the practical response so
muted? Limited resources are certainly a factor; but clearly, not everyone
shares the same views with respect to the imminence and likely severity of an
outbreak.
Those who warn that a pandemic may envelop the world within
months from its onset, and there are many experts that do, suggest a health
emergency that arguably would require strong government action such as
nationalization of pertinent production facilities and invoking of TRIPS
flexibilities to allow for greater availability of affordable treatments. A
pandemic could circle the globe so quickly that initiating such steps after the
appearance of a pandemic strain might be pointless.
Despite the dire predictions, steps like compulsory licensing of antivirals
have yet to be taken, suggesting that governments may be dubious of the
claims made by some scientists of the imminence of a severe H5 human
pandemic. Is this foolish, or an efficient use of overstretched resources?It will
only become clearer in retrospect
Nobody argues against improved pandemic preparedness now and in
the future, for everyone seems to accept that a new pandemic will occur,
sooner or later. Yet, at the same time, it is clearly not possible today to
abandon other public health efforts because the argument that a highly lethal
pandemic strain is nearly upon us may turn out to be correct
For those seeking to get ready for a pandemic now, proven technology-
mainly egg-based production of classic flu vaccine-offers degrees of certainty
that emerging biotechnologies cannot. Methods to grow H5 viruses in eggs
are improving, and egg-based production is already available and does not
require any potentially expensive and unreliable "bleeding edge" technology.
And in theory, the same production facility can also be used for production
of pandemic LAIVs.
Although egg-based production is sometimes maligned as "antique", it
is telling that major vaccine makers investing in biotechnology remain heavily
reliant on egg-based systems for their own flu vaccine production. The major
problem, of course, is what-if anything-to do with the production capacity
when it is not required for (pre)pandemic vaccines, in view of the fact that
there is limited other use for egg-based facilities and, for many developing
countries, seasonal flu vaccination is a losing economic proposition.
Maintaining an unused production base is expensive. WHO estimates that
maintaining an idle capacity to produce 200 million seasonal vaccine doses
would cost US $100 million per year.
Page 20
� Observations on Vaccine Production Technologies and factors Potentially Influencing
Pandemic Influenza Vaccine Choices in Developing Countries
Viewed in a longer timeframe, technology selection may become
more complicated. The flexibilities and potential efficiencies of cell culture
are attractive because they may offer a faster pandemic response and,
especially, a facility with potentially broader public health uses-if the
technology is available and markets exist for the other types of human
vaccines that may be produced in cell culture.
However, the relatively unproven status and considerably greater cost
of hardware for cell culture technologies (estimated at ten or more times
the cost of egg-based facilities), both in terms of equipment and intellectual
property, at present make them a daunting proposition for most developing
countries.
Fill/finish projects and importation of bulk antigen
Indonesian and Mexican vaccine manufacturers, with W H O support, are
developing filllfinish capacity for local vaccine sales. In the filllfinish
approach, developing country manufacturers import bulk vaccine antigen
produced by an overseas company and use it in a locally branded, finished
product. In the current WHO-supported projects, the antigen
manufacturers are Biken (to Indonesia) and Sanofi-Aventis (to Mexico).
The imported bulk antigen, suitable for a classic killed vaccine, is
processed in-country into a finished product. The national manufacturer
creates filling and packaging facilities, and some associated technology
transfer takes place.
Importation of bulk antigen and fillinglfinishing in developing
countries favours the argument, advanced by some, that it is rational for
global influenza antigen production to be concentrated in a few locations
with well-developed capacity and expertise.
Local manufacturers importing bulk antigen remain dependent,
however, on product supplied from abroad, which is unlikely to be
available in the event of a pandemic (particularly in its early stages), so long
as global production capacity remains well below that which is necessary.
Page 21
�A discussion oaoer
The option of animal vaccine plant conversion
Current global capacity for human influenza vaccine production falls well
short of that needed for pandemic response, even with optimistic
assumptions about demandlyield of pandemic antigen. Often unmentioned
is the substantial additional manufacturing base that uses egg-based
production systems to make animal vaccines. These facilities could lessen
the gap between pandemic vaccine supply and demand. They use a very
similar production process as that used for human influenza vaccines.
Estimates of the global size of the egg-based animal vaccine industry,
however, vary wildly.
On the high end, according to one source20the annual global egg-
based animal influenza vaccine capacity, as of 2006, was approximately 41
billion avian doses (at 100 doses per egg), or about 410 million eggs. In
terms of human vaccines, this implies a capacity of approximately 410
million doses of human trivalent seasonal vaccine. Using this capacity
estimate, output of a monovalent pandemic LAIV could be between 1.8
billion (WHO conversion factor) and up to 4 billion or more vaccinations
per year (other conversion fa~tor),~'
depending on antigen assumptions. In
either case, this would allow vaccination of a substantial proportion of the
world's population.
But WHO CAP consultants, also citing industry sources, come up with
very different numbers for the potential contribution of animal vaccine
facilities. They report that the animal vaccine industry can handle only
about 78 million eggs annually. This implies an annual pandemic LAIV
output of approximately 340 to 750 million human vaccine courses per
year, a much lower but still substantial figure.
It is thus difficult to be precise about (pre)pandemic capacity of
animal vaccine facilities because of conflicting and limited data and the
2fl Hcldens, J G M. Production capacity for human and veterinary influenza, June 2006,at URL: http://
www.dut&bio.org/'meetings/list/dutch_vaccines_group/files/influenza_dag/
DVC%2Ojacco%20Heldens,%2026.06.06.pdf (Heldens represented Akzo Nobel, which owned
Inletvet, a major animal vaccine maker, until it was sold to Schering Plough in 2007.)
21 See: Fedson DS, Dunnill P. New approaches to confronting an imminent influenza pandemic. Perm J
2007;11:639, URL: http://xnet.kp.orp/permancnteiournal/SUM07/influeriza-oandemic.h~rnl and
Fedson DS, Dunnill P.From Scarcity to Abundance: Pandemic Vaccines and Other Agents for "Have
Not"Countries in journal of Public Health Policy (2007) 28,322-340.
doi: 10.1057/palgrave.jphp.3200147
Page 22
� Observations on Vaccine Production Jechnohgies and Factors Potentially Influencing
Pandemic Influenza Vaccine Choices in Developing Countries
variety of assumptions that could be made about antigen production and
vaccine type. But even a low-end estimate would represent a large addition
to human production capacity. Notably, a large proportion of global animal
vaccine production capacity is located in Asia, and additional capacity
exists in Latin America.
Converting an animal influenza vaccine facility to human vaccine
production is not, however, as simple as switching vaccine seed strains.
There can be significant hurdles, the severity of which will vary with the
specific equipment and process used at each manufacturing plant.
Major issues to be addressed in such a conversion are regulatory
certification of the manufacturing process to human vaccine standards,
ensuring appropriate biosafety practices, adequate egg supply, improved
virus purification processes, and adoption of adjuvants approved for human
use.
Regulatory hurdles will be country-specific. In some places, animal
vaccine plants are already held to manufacturing standards near or equal to
those for human vaccines; however, this is not always the case. A related
issue is biosafety practices which, in some animal vaccine plants, would
need improvement-both in operating procedures and, potentially, to
equipment.
Conversion of animal facilities to human vaccine production may also
strain egg supplies, especially in countries or regions where H5 vaccination
of poultry currently occurs, because eggs laid by hens vaccinated against H5
cannot be used to produce vaccine.
Human flu vaccines produced in eggs go through an extensive
filtration process to remove egg proteins and other contaminants that can
cause an adverse reaction. Animal vaccines are generally not subjected to
the same level of filtration, and improvement of filtration in converted
animal vaccine plants would be necessary.
The adjuvants used in animal vaccines are not typically approved for
use in humans, and animal vaccine plants would have to switch to
appropriate adjuvants, unless they are producing a human pandemic LAIV
(which is unadjuvanted).
Page 23
�A discussion oaoer
Human vaccine producers and pharmaceutical companies own a
significant proportion of global animal vaccine production capacity. For
example, Merial, the world's largest animal vaccine maker, is a joint venture
of Sanofi Aventis and Merck. Ft. Dodge, another large animal vaccine
company, is a division of Wyeth. Intervet, a third large animal vaccine maker,
is owned by Schering Plough. Other drug companies, such as Pfizer and
Novartis, also have animcil vaccine businesses. Thus, human and animal
vaccine makers should not be thought of as wholly separate industries.
5. Concluding discussion
Which technologies should developing countries seek multilaterally to
improve pandemic preparedness? The answer, of course, depends on many
factors.
Reliance on a s m d number of developed country sources for
pandemic vaccine and/or antigen is unlikely to remain an acceptable
solution for most developing countries, particularly in view of the fact that
the developed country industry is not currently in a position to offer
sufficient quantities of antigen in a timely manner after the appearance of a
pandemic strain.
Practically, the present situation of dependency, which is effectively
unaltered by the WHO Pandemic Action Plan, means that the vast majority
of developing countries will only receive significant quantities of vaccine
after the needs of developed countries are met, which will likely be many
months after the onset of a pandemic-months during which pandemic
mortality may be severe.
As a result of the inequity, in the event of a pandemic, developing
countries will suffer a disproportionate burden of serious disease and death,
a problem that could be ameliorated by increased and equitably distributed
global vaccine supplies, particularly in the developing world. These vaccine
supply problems may be further exacerbated by non-health factors, in the
form of export controls that may inhibit the ability of some countries to
prepare for a pandemic because some kinds of technology transfer are
unavailable to them.
Developing country leaders are likely to face question from their
citizens if they remain vulnerable while the citizens of wealthy countries are
vaccinated; this situation could become especially tense if a pandemic is
severe enough to cause serious socioeconomic disruption.
Page 24
� Observations on Vaccine Production Technologies and factors Potentially Influencing
Pandemic Influenza Vaccine Choices in Developing Countries
Vaccination for the population at the earliest point possible following
the onset of a pandemic isn't the entirety of pandemic preparedness; but it
is a high priority. But at present, there is little consensus among experts
about how best to achieve that.
It is also clear that no single technological approach will be
appropriate for all countries or regions and that greater funding and
improved access to proprietary technologies will be necessary for
developing countries to improve protection of their citizens from pandemic
flu. Regional cooperation in production and technology to take advantage
of economies of scale will likely be far more fruitful than trying to go it
alone for most countries.
Several options for financing and technology transfer have been
mentioned in the context of the Pandemic Influenza Preparedness
Intergovernmental Meeting (IGM). These include increasing vaccine
production in developing countries, possibly supported by royalty-free
licensing of vaccine production technology. Contributions to a global fund,
and contributions of vaccines to a WHO stockpile by entities that use
pandemic preparedness biological materials in research and development
of vaccines and other biomedical items have also been proposed.
While a WHO stockpile may be useful to help stamp out or slow down
the emergence of a pandemic influenza strain, it is not designed-nor will it
serve-to ensure any country's vaccine supply. A WHO vaccine stockpile is
also mandated by WHO Member States outside the WHO PIP IGM
discussions, and is thus not a central objective of the benefit-sharing
discussion.
Because increasing national or regional vaccine production capacity in
developing countries requires flexibility in technological approaches, no
single technology transfer and cooperative arrangement is likely to be
effective. There is strong evidence that proprietary and emerging
technologies, such as reverse genetics, adjuvants, and in the future cell
culture, could serve to greatly increase the efficiency of preparedness
efforts. Specific technology selections, however, must be made in the
regional and national contexts.
In principle, developing countries may seek to formalize a system of
equitable reciprocity wherein those developed country companies and
other entities that utilize Global Influenza Surveillance Network (GISN)
Page 25
�A discussion oaoer
materials to develop vaccines commit to transfer their vaccine technologies
so that they may be used by developing countries.
Therefore, in the PIP ICM negotiations, developing countries have
explored the possibility of creating a mechanism for transfer of influenza
vaccine technology, through mandatory royalty-free licensing and other low
or no-cost means, including for both formal patents and related know-how
and trade secrets. The technologies prioritized by any such pandemic
preparedness technology transfer program should be those that are used by
industry to manufacture products that include WHO GISN materials (e.g.
H5 vaccines) or are developed utilizing WHO GISN materials.
Reducing proprietary barriers to the technology needed to produce
pandemic vaccines would represent a significant step forward; however,
making technology available does not guarantee that it will be effectively
used. Ways to optimize the use of technologies include, for example, the
creation of a financing mechanism by which the real-world transfer of these
technologies can be effected (for instance, to pay for the necessary
equipment and training to utilize them). A pandemic preparedness
cooperation fund could also be established, with contributions from
manufacturers that utilize WHO GISN materials in commercial products (to
be defined in a WHO material transfer agreement), and possibly
contributions from governments. A cooperation fund could also help enable
the use of nonproprietary technologies, such as egg-based production lines
and fill/finish capacity, which will be important elements of any national or
regional effort to increase vaccine production capacity.
Pandemic preparedness is a problem of daunting complexity, and
solutions will only come with time and contributions from many quarters.
The PIP IGM is an important process but not one that by itself can solve all
problems. Developing countries may wish to focus on important specific
benefits that will enhance their preparedness.
With the timing of a pandemic uncertain, but the time needed to
construct and validate vaccine facilities typically measured in years, it is
urgent that progress be made now to expand developing country vaccine
production capacity. Developing countries will have to work together to
identify the best technologies for their circumstances. The PIP ICM's
decisions may help to make key technologies affordably available to
developing countries, and through a cooperation fund, provide means
through which to effect their transfer and use.
Page 26
� Observations on Vaccine Produaion Technologies and Factors PotentiallyInfluencing
Pandemic Influenza vaccine Choices in evel lop& ~ountries
Annex 1
Overview table of influenza vaccine technologies
.- - -
Cell culture
produced vacdnes
-
-
-
. -
vaccines, =-.
-- - -
etc) F
-
Z --
Description Inject vaccine Animal, insect, or A vaccine seed Many techniques are
virus into other cells are strain using a under study, including:
fertilized eggs, cultured in growth special type of
allow virus to media, scaling up backbone (from a - Producing
grow. Harvest the quantity to the lab-adapted flu recombinant HA in
fluid from eggs, desired density of strain) is grown in other, more easily
wash with cells in industrial eggs (or potentially grown organisms
detergent to kill bioreactors cell culture). The (e.g. transgenic
cells, separate out [fermenters) of live virus is bacteria).
virus material for hundreds to harvested, purified - "Naked" and
vaccine thousands of litres and formulated for plasmid DNA
formulation. capacity. The use. Harvesting and vaccines in which
Generally, the culture is infected formulation is "codon optimized"
vaccine strain with vaccine strain, simpler than with flu genes are used
consists of the H A producing large killed vaccines, but directly as vaccine.
and NA genes of a quantities of the live final
"wild type" of vaccine virus. product is more - Genetically
influenza fused Harvesting, delicate. engineered systems
onto a lab- purification and to co-express HA,
adapted packaging is NA, and other flu
"backbone" strain essentially the genes, making a
with the other same as with egg- "virus-like particle"
viral genes. based methods. (VLP)which is used
as vaccine.
-
"Hard" Large BSL-2 Large BSL-2 +
Large BSL-2 . May Will vary with specific
technology space, incubators, bioreactors for cell be grown in eggs or technology; however,
requirements inoculation and culture, equipment bioreactors (see all are likely to require
harvesting to maintain cell respective BSL-2 CMP space and
equipment. bank and scale-up. requirements at microbial fermentation
Centrifuges In some cases left), but currently /coil culture capacity.
(separationand growth substrates. the process is done
purification) and After harvestingof in eggs. Purification
packaging virus, purification and formulation
equipment. and formulation differs from that for
requirementsare killed egg and cell
the same as for culture vaccines.
eggs.
�A discussion paper
- Egg-based W&sicn'
flu vaccine I
-
0 1 1 culture
producedvaccines -
Liveatenuated
("lAIVw)
-'
-
- - --
Major The technology is Theoretically higher Live vaccine is likely Dependent upon
advantages well known and has antigen output than to be effective in specific'
been successfully eggs, theoretically much lower doses technology. Nearly
utilized for decades. more scalable. May than killed vaccines. all purport to be
Essentially the same grow some U S Assuming able to provide
technology is used vaccines viruses to production capacity more vaa5ne
for some poultry higher liter. Cell can be harnessed, faster, but these
vaccines, making culture vaccine more LAIV vaccine claims are as yet
conversion of animal plants likely to prove may be made unproven.
vaccine plants and more flexible for available in a May offer more
personnel a producing other shorter time period
dependable
possibility in an kinds of human than with other
vaccine yield per
emergency. vaccines than egg '"T'es. production run.
based plants.
Major Not as efficient as Cell culture vaccines Not suitable for Dependent upon
limitations cell culture are a major focus of prepandemic specific
theoretically is. R&D; but as yet they vaccines due to technology.
Requires many eggs, are in limited recombination risks.
the availability of commercial use. Difficult to test with
which may be Inability to be certain prepandemic
limited in a that a particular cell strains. Requires
pandemic (eggs may line will be more stringent
be available from the appropriate to grow conditions on egg
broiler industry). Egg- 11iepandemic strain. supplies than killed
based plants are Requires substantial vaccine production.
unlikely to be used bioreactor capacity Higher biosafety
for other human of which there is little requirements for
vaccines (except for to no global surplus. production. More
Japaneseencephalitis Much higher cost difficult to store
vaccine). facility at industrial vaccine. Intranasal
scale. administration
requires special
delivery device.
Regulatory1 Fewer impediments Few cell-culture Seasonal LAIVs These products, if
safety as these vaccines will produced vaccines have limited use in successful, will be
approvals he produced in the hwe been approved the U ("FluMist")
S new to regulatory
same manner and for human use and and in Russia; systems and are
with the same they are likely to however, most highly likely to
facilities as seasonal prompt more intense regulatory require substantial
vaccines, although regulatory scrutiny. authorities would safety review.
converted animal Require approval for be encounteringa
vaccine plants likely the vaccine as well as live (and likely
will not already characteriition and genetically
possess approvals to safety demonstration engineered)
produce human of the cells used. influenza vaccine
vaccine. for the first time.
Page 28
� Observations on Vaccine Production Technologies and factors Potentially Influencing
Pandemic Influenza Vaccine Choices in Developing Countries
Administration Syringe Syringe Intranasal (requires Method of
appropriate delivery administration
device) (injected, oral,
intranasal, etc.)
will depend on the
specific product
Recombinant Probably. Seed strain Same as egg-based Probably, with the Vaccine will be
(genetically- may be produced production. Other notable difference genetically
engineered?) with reverse genetics genetic modifications that the vaccine is engineered or be
and, for example, of the vaccine strain administered live. the product of a
may contain a may occur to genetically-
modified HA gene optimize growth in engineered
(deletions) to make cell cullure. organism.
the virus less
pathogenic. Virus is
killed before use.
Adjuvanled Almost certainly, to Same as egg-based. Probably not. The Depends on
(pandemic make more efficient One exception is vaccine virus specific
vaccine?) use of bulk antigen unadjuvanted killed replicates in the technology.
and potentially to ''wild-type" virus upper respiratory
reduce the number (being tested by tract, stimulating
and size of required Baxter); however, immune response.
doses. producingsuch a Nevertheless, some
vaccine is a biosafety research has
challenge, requiring focused on
cell culture in large increasing immune
scale BSL-3 response to LAIVs
containment. with adjuvants.
Intellectual Few IPR problems Many IP R Only a small Impedimentswill
property for egg-based impediments. These number of depend on
process, except for include patents on backbone strains specific
adjuvants where IPR cell lines and are suitable for use. technology;
and supply problems production systems, Intellectual property however, it may
may exist. as well as trade impediments exist be anticipated that
Potential additional secrets on safety on the use of these technologies
problem if seed profile of cells. Cell strains; patents and will have robust
strain is produced characterizationis trade secrets cover IPR coverage as
using reverse only reportedly the formulations. they are mainly
genetics. publicly available for Seed may need being developed
Vero (monkey) cells. reverse genetics. by biotech
companies and/or
universities
seeking to sell this
technology.
-
Page 29
�A discussionpaper
Other issues Some (generally Requires supply of Safe production The vast majority
minor) side effects growth media and will require more of R&D in these .
from egg proteins other relatively stringent biosafety lines of research
and other possible exotic supplies. In procedures than appears to be
impurities. addition to technical killed vaccines to conducted by
challenges, cell prevent companies in a
culture production contamination of handful of
may be especially live final product. developed
prone to export Societal resistance countries.
control issues for a may be significant,
number of countries. particularly for
seasonal use.
Page 30
� Observations on Vaccine Production Technologies and factors Potentially Influencing
Pandemic Influenza Vaccine Choices in Developing Countries
Annex 2
Relevant reports available online
Friede M, Serdobova I, Palkonyay L, Kieny MP. Technology transfer hub for pandemic influenza
vaccine. Vaccine. 2008 Nov 18. (ht@://dx.doi.ore/lO. 1016/i.vaccine.2008.10.080 - accessed 9
January2009).
Lozano B. The veterinary hiological industry and the production of human pandemic influenza vaccines
in Mexico. Geneva: WHO FA0 OIE Consultation Seminar. 2006.
(http://www.who.int/entitv/csr/disease/influenza/Bernardo Lozano A v i m e x d f
- accessed 9 January2009).
National Academy of Engineering. V-36-3 engineering and vaccine production for an influenza
pandemic. The Bridge. Fall 2006; 36(3). h~~://w.nae.edu/nae/brideecom.nsf/wel~links/MKEZ-
-
6SZRM2?OpenDocument- accessed 9 January 2009).
Oliver Wyman Consultants. Influenza vaccine strategies for broad global access: key Findings and
project methodology. Scatlle: Path, 2007. (htl~://www.~ath.or~/files/VAC publ rpt 1 0 - 0 7 . d -
infl
accessed 9 January 2009)
World Health Organization. Business plan for the global pandemic influenza action plan to increase
vaccine supply. Geneva, WHO, 2008.
htt~://www.who.int/entitv/vaccine research/documents~Re~ort%2520McKinsev%2520Business%2520
Plan%2520Flu3.pdf - accessed 9 January2009).
World Health Organization. Mapping of intellectual property related to the production ofpandemic
Influenza Vaccines. Geneva: WHO, 2007.
(httD://www.who.int/vaccine research/diseases/influenza/Ma~~inp Intellectual Propem Pandemic I
nfluenza Vaccinemdf - accessed 9 January 2009).
World Health Organization. Meeting with internationalpartners on influenza vaccine technology
transfer to developing country vaccine manufacturers. Geneva: WHO, 2007. Document
WHO/IVB/08.09. (ht~://w.who.int/immunization/documents/WHO IVB 08.09/en/index.html -
s
accessed 9 January 2009).
World Health Organization. Tables on the clinical trials of pandemic influenza prototype vaccines.
Geneva: WHO.
1 httD://www.who.int/vacrine rcsearch/diseases/influenza/flu trials tables/en/index3.html - accessed 9
January 2009).
World Health Organization. The global action plan (CAP) to increase supply of pandemic influenza
vaccines, first meeting of the advisory group. Geneva: WHO, 2007. Document WHO/IVB/08.10.
(htt~://www.who.int/imm~~nization/documents/WHO 08.10/en/index.html - accessed 9 January
IVB
2009).
Page 3 1
� This paper presents an overview of technologies currently available for the
production of influenza vaccine, as well as others that are under
development. It draws attention to pertinent issues and challenges that
policy-makers in developing countries may need to consider when
reviewing their options for accessing influenza vaccine production
technologies. It is intended as a contribution to the debate on the sharing
of influenza viruses and access to vaccines and other benefits arising from
their commercial exploitation.
World Health
Organization
~ e ~ k nOffice for SouthEast Asia
al
World Health House
Indraprastha Estate,
Mahatma Gandhi Marg,
New Delhi-110002, India