(Click on image for a larger version)
Have a look at the two pictures on the slide. We have chosen them quite deliberately. Those among you in the biosecurity laboratory field would know that the Australian Animal Health Laboratory (AAHL) in Geelong is a national facility managed by CSIRO and is a high-technology, state-of-the-art biocontainment laboratory. As its core responsibility AAHL is committed to protecting the Australian community from the impacts of foreign animal diseases, but it is also very much committed to deploying that technology to assist other countries in the region with their animal health problems.
From that perspective I would like to talk about some of the activities with which AAHL has been involved.
(Click on image for a larger version)
With respect to Indonesia, which will be the focus of this talk, AAHL is helping the veterinary laboratories managed by the national government to deploy state-of-the-art diagnostic technologies against the threat of avian influenza. The arrows indicate some of the areas with which we are involved , and where their laboratories are situated, right across the country.
You can see from the green shading that H5N1 is spread throughout the length and breadth of Indonesia. There are only two provinces that are not affected, to the present time.
(Click on image for a larger version)
Let’s pause to reflect on the term avian influenza (AI). Although avian influenza is perceived as a public health threat, it is also very much a part of nature. AI viruses are commensal infectious agents of wild birds and can occur in quite a variety of haemagglutinin and neuraminidase combinations.
(Click on image for a larger version)
Where AI infections occur as a disease, in poultry it is viruses of the H7 and the H5 haemagglutinin that are causing the problems. Consequently the World Organisation for Animal Health designates any H5 or H7 AI virus as a notifiable avian influenza. Some of these are low pathogenic and non-virulent, but where they are causing disease they are called high pathogenic (high path) avian influenza. Some of us would be aware that there is also concern for some strains of H9 becoming highly pathogenic, but basically avian influenza disease is H5 and H7 driven.
(Click on image for a larger version)
Against that background, how should the current AI epidemic be viewed? This diagram, which is taken from a publication by Li and colleagues from 2004, is very helpful, because it allows a better appreciation of the AI viruses involved .
Avian influenza viruses have eight genome segments, change can occur within any of those segments, and there can be reassortants. We see from this diagram that, over a period of time, an H5 haemagglutinin that was first detected in Guangdong in 1996, in a goose, has persisted through a whole series of reassortant events to end up in a virus that we first see described in 2002, labelled as the Z strain, which persisted through 2003, spreading throughout South-East Asia, and then in later years globally.
So we are specifically dealing with an H5N1 virus that in many respects is different from the other H5s and H7s that we call high path avian influenza. It is different in that it is killing waterfowl, which avian influenza viruses do not usually do, and it is also infecting quite a number of people. Far more people are infected with this H5N1 than with any other avian influenza viruses previously. And there is the issue of global spread of the disease in poultry, which hasn’t been seen before.
So, from a veterinary point of view, bird flu is not just any old avian influenza virus. It is this particular strain, and that is what the international community is trying to come to grips with.
(Click on image for a larger version)
This dendogram illustrates one of the important points about H5N1: where it has infected a country it is tending to be evolving within that country. That is quite different from human seasonal influenza, where public health authorities tend to deal with global populations of influenza viruses. To focus on Indonesia, for example: after the initial introduction in 2003 the sequence analyses are indicating that there have been no new introductions to that country. In Indonesia there is natural evolution of strains of H5N1 from perhaps an original single-point introduction. This interpretation is important in terms of managing the infection country by country.
(Click on image for a larger version)
The multilateral organisations combating bird flu internationally, the FAO and the World Organisation for Animal Health (the OIE), issued a situation assessment at the end of last year, and nothing much has changed. The risk of disease in poultry and its transboundary spread will persist as long as this H5N1 strain is ‘entrenched’. The term refers to the situation in countries like Indonesia, Egypt, Nigeria and so forth where, in spite of the best efforts of the veterinary authorities in those countries over a number of years, little has been achieved in successfully addressing the AI situation. So in some countries there are entrenched infections, and that is a major challenge.
(Click on image for a larger version)
Nonetheless, the FAO and the OIE believe that a reduction in levels of infection in these countries is achievable and, more importantly, is essential because the risk of pandemic influenza arising from these H5N1 infections has not ceased. The fact that it hasn’t happened to date has not changed the possibility that it might. It could be argued that as these viruses continue to evolve and change, perhaps the risk increases. Therefore veterinary authorities globally and even in ‘entrenched countries’ such as Indonesia accept that controlling this infection is a high priority.
(Click on image for a larger version)
What is necessary for the control of an infection such as this in poultry? There are four main issues and the first, from an animal health perspective, is absolutely crucial. Countries that cannot control the transmission of infection from one group of infected birds to another simply haven’t been successful in controlling the H5N1 AI epidemic. Most important are movement controls, or the new term that is being used, ‘movement management’, because it is a management issue. Culling can be seen as a definitive movement control. Movement management is a strategy that animal health authorities should be able to adopt, and where it is done in a timely manner it can be very effective.
But where infection is entrenched adoption of a whole parcel of other measures is needed. Susceptible birds have to be protected from infection, and issues such as quarantine are important, and also the biosecurity of production units. We have to be able to farm poultry and keep them free of the risk of infection from the surrounding countryside. In the complex range of issues around protection of susceptible birds, vaccination is one of the major tools being used at the present time.
Reduction of the viral load in the environment is another important concept. Again vaccination is a strategy that is supported by the FAO and the OIE as a tool towards that particular objective.
And the strategy will not work without involvement and commitment from the community, so public communication is absolutely essential.
(Click on image for a larger version)
The approaches listed cover the essence of how to manage an animal disease. Let’s present the issue diagrammatically. The black line down the middle of the slide represents the absolutely essential criterion, movement controls. Countries that control bird flu successfully have been successful in movement management controls, but entrenched countries have not been successful in that area.
Subsidiary measures such as biosecurity, on the right-hand side of this slide, or decontamination of infected premises, on the left-hand side, are important. And vaccination underpins the whole strategy and gives extra flexibility to it. But we do emphasise, whenever we meet with colleagues from affected countries, that we really have to be more astute in the way that we look at movement controls.
(Click on image for a larger version)
Dealing specifically with Indonesia is a very complex situation. Under the new arrangements for governance in that country, we have a national government but with real power devolved right down to the district level. There are 440 districts, which are responsible for animal health and the actual practical measures to deal with control of animal health issues. Coordination of 440 local governments to have an effective national campaign is a challenge.
We can see further down in this diagram that there are 60,000 villages and 240 million people, and that translates to a poultry industry with 1.5 billion birds per annum moving through the system. Translate that into doses of vaccine and, well, you’ve got a market, put it that way.
(Click on image for a larger version)
The poultry industry has two main components: the commercial sector, which has variable levels of expertise, and the village sector, which has little experience in the necessary animal management skills such as biosecurity.
Against this mix, Indonesia was in a situation in the middle of last year where it had 12 or 14 vaccines that were provisionally registered but were unable to be fully registered under the national government regulations until an efficacy test was conducted. Laboratories such as AAHL did not want to do the efficacy tests free of charge, because under our governance we are unable to provide a free service to commercial companies. So there was a point of indecision. International development assistance organisations such as AusAID, USAID and the World Bank were willing to provide millions of dollars in assistance to Indonesia to help with vaccination, but these agencies were unwilling to spend money on products that were not registered.
(Click on image for a larger version)
To simplify the issue the US Department of Agriculture volunteered to undertake the efficacy testing, or challenge testing, of these products. Three challenge viruses were selected, from all the isolates that had been available from the Indonesian agricultural sector, and these were sent from AAHL in Geelong to Athens, Georgia, for this work.
In the standardised challenge trials birds were vaccinated with a single dose of each product under test, according to the manufacturer’s instructions, and then challenged three weeks later.
(Click on image for a larger version)
In the first experiment they were challenged with the first challenge virus, A/chicken/West Java/PTW-WIJ/06, and then in the second experiment those vaccines that weren’t included in the first experiment round were included and the two other challenge viruses were also included.
(Click on image for a larger version)
In this graphic presentation of the data the antibody response to these products was variable. Some gave good to excellent antibody responses to the vaccine antigen; while with some others the responses weren’t so good. But basically all products gave an adequate minimal level of antibody response to the vaccine.
(Click on image for a larger version)
Including the data for survival after challenge into the graph, we see a situation that caused a lot of concern with the first of the challenge viruses. There is only one immunogen in that whole panel that is providing a satisfactory protection against that particular challenge, and that is an antigenic preparation made from the challenge virus itself, A/chicken/West Java/PTW-WIJ/06. Homologous immunisation gave 100 per cent protection, but none of these other provisionally registered products gave satisfactory protection against this particular strain of virus from West Java.
(Click on image for a larger version)
When the other two challenge viruses were used there were no problems. Virtually all the products gave good protection.
(Click on image for a larger version)
(Click on image for a larger version)
The results indicated the possibility that some strains of H5N1 AI circulating in Indonesia may not be effectively contained by available vaccines. Also, the serological responses to the vaccine weren’t necessarily predictive of protection, some birds with quite high antibody titres to the vaccine antigen succumbed to challenge with A/chicken/West Java/PTW-WIJ/06.
(Click on image for a larger version)
Sequence data for the HA gene of H5N1 isolates from Indonesia were examined. The first challenge strain came from a small cluster of viruses somewhat divergent from most of the sequences available. There were only four viruses clustering in that group, which may give some cause for hope that perhaps the situation that was paralysing the international response to vaccination in Indonesia might not be widely distributed.
Those particular viruses came from the commercial poultry industry while most of the other isolates in the analysis came from the village sector, raising the issue of whether the commercial industry has to be managed differently from the village sector with respect to the vaccines that are used?
(Click on image for a larger version)
The challenge virus A/chicken/West Java/PTW-WIJ/06, on the basis of the sequence analysis, would appear to potentially have 19 amino acid differences in the HA1 protein as compared with viruses from early in the outbreak, quite a significant change over that period.
(Click on image for a larger version)
AAHL receives viruses from countries in South-East Asia, including Indonesia, and works through a structured program of tests and analyses, as shown diagrammatically. Sequence analysis allows grouping of the viruses, on the basis of which decisions can be made regarding the need for further analysis antigenically. A system is being developed, in collaboration with Erasmus University and other groups internationally, to utilise antigenic cartography as used in the selection of antigens for human seasonal influenza vaccines, to identify H5N1 strains against which available vaccines may not be protective.
(Click on image for a larger version)
As part of that approach some viruses were selected, the 12 on the list shown here, against which antisera were raised and cross HI work initiated. The three that are indicated in the accompanying list with red arrows are those that group with A/chicken/West Java/PTW-WIJ/06 in the cluster of four that we believe are possibly of concern.
(Click on image for a larger version)
In the cross HI tests with the panel of antisera and each of the homologous and heterologous antigens, the reactions of the sera raised against those three viruses against a panel of other antigens from Indonesia gave very low HI titres. So in the HI serology the preliminary indications are consistent with the findings of the challenge study in the US, that there may be a serious issue of mismatch of antigens between vaccine and field strains.
(Click on image for a larger version)
The international response to this newly identified situation is being coordinated by the OIE and FAO through the OFFLU consortium. More information regarding OFFLU is available on the web. In this work there is a partnership of the Indonesian laboratories, AAHL in Geelong; the USDA in Athens, Georgia; Erasmus and Cambridge universities with their antigenic mapping; and the Veterinary Laboratories Agency in the UK. The concept is to develop for avian influenza a system analogous to the WHO-mediated system that works for choice of seasonal influenza candidate antigens for vaccines for people.
Indonesia is submitting more H5N1 isolates from poultry to AAHL under material transfer agreements, and the Indonesian agricultural sector is keen to be part of this international collaboration to address these specific issues in that country. It is not known yet how widespread these strains of concern are, or whether they are restricted to the commercial industry. It might be that special vaccines may be required for certain parts of the poultry industry in certain countries. The vaccine that is suitable to deal with this particular problem in Indonesia may not be suitable for use elsewhere, in other countries. This raises the issue of manufacture of products that may not have international markets, and whether such manufacture will be seen as commercially viable by vaccine companies.
There is quite a lot of work to be done; we are in the early stages. Hopefully this presentation has given an indication of the challenges and the sorts of responses that are being initiated at the present time.
Discussion
Question: Peter, if the aim is to avoid the infection becoming entrenched, in a country like Indonesia, what constitutes success?
Peter Daniels: We are in a situation where the infection is entrenched, and so what are the choices? One choice, for instance, would be for the world to say, ‘Oh well, it doesn’t matter. It’s just a problem for the poultry farmers in Indonesia. They’ve got to learn to live with this and the world can withdraw.’ But the best scientific advice from the WHO, the FAO and the OIE is that this is a problem of global concern, and so we have to try to work out what to do to deal with a very difficult situation in terms of animal disease.
Question (cont.): Does that mean that what you are really thinking about is eradication?
Peter Daniels: People aren’t using words like ‘eradication’. The policies being developed by the international organisations are to reduce the risk of pandemic influenza by reduction of the level of H5N1 AI virus transmission. That is seen as a more practical and achievable goal than to aim immediately for eradication. Of course people would like to eradicate it, but…
Question: I saw in your slides that you also tested a goose Guangdong-based vaccine, but it was not clear to me. Does that vaccine protect against the Indonesian virus?
Peter Daniels: My recollection of that data is that no, that particular product did not have a high level of efficacy. In general, the data from the study showed that H5 viruses from Indonesia, when incorporated in antigens, were giving better levels of protection than antigens that were coming from other international sources.
Question: I am just wondering if in the animal vaccine world they are thinking about providing some sort of immunogen that is not as variable as the H antigen. I don’t know if this is a fair question to ask you, but at least in the human H5N1 world people are starting to think about trying to control infection or reduce infection with the T cell response. I am wondering if that is an area of research.
The second question I have, which might be easier for you to answer, is: are people thinking about making small batches locally? That has been done in the animal vaccine world. That used to be done with flock-related viruses anyway.
Peter Daniels: In terms of management of disease, people are having to deal with the more conventional type of product, which is based on the haemagglutinin antigen. Our laboratory isn’t researching any other approach to vaccination, although there is other work going on.
With respect to production units for vaccine, Indonesia has two quite good commercial vaccine production companies, but they simply could not produce sufficient product to effectively conduct vaccination programs in Indonesia. Traditionally, also, the commercial industry has preferred to use imported products and the local production has tended to go to the village sector on government contract. But if the local companies were the only ones interested in producing the products that the commercial sector needed, on the basis of the international antigenic analyses, then those historical preferences would change, I’m sure.
Question: I was just wondering about the extent to which immunisation programs in Indonesia would simply shift susceptibility towards a different strain. Are there examples of that occurring with animal vaccines, where you create an ecological niche and something else comes in behind it?
Peter Daniels: It is a concern. I’m not sure what we are going to hear from China, but the data that I have presented is the first time that we have encountered or have had to deal with this particular problem in veterinary medicine with respect to avian influenza. We do see it already to a certain extent with equine influenza. I think the international community is concerned that we may get on top of this particular antigenic mismatch, but, as you say, the situation is going to continue to evolve, and we are going to be in a situation where this has to be ongoing activity – if we want to keep using this type of vaccine.
Question: Peter, how do you get the vaccine into the villages, given that there are many fewer vets in Indonesia than human health care workers, and there are many more chickens than people?
Peter Daniels: There is estimated to be a standing chicken population of 300 million in villages, so that too is a challenge. But people aren’t thinking that every dose of vaccine has to be administered by a veterinarian. Part of the AusAID funding in the year or two ahead, I am sure, will be for working with groups in Indonesia to have communities looking after their own vaccination needs themselves.
Question: Can you just give us a hint or feeling for the magnitude of Australia’s involvement, and your own in particular? Is this a big thing in CSIRO, assisting in Indonesia in this area, or are we ‘there’ but not to any major extent? Could you give us a feeling for how involved we are?
Peter Daniels: For us it is a significant involvement. I would say that 25 per cent of the budget that I have responsibility for goes into this. The money that is being expended at the present time is part of the $100 million that the Australian government committed to avian influenza issues in the previous round, and $18.6 million of that was designated to Indonesia. So at the present time, from say a year ago through to the next year or two, Australia is putting in at least that amount of money for assistance to Indonesia. And USAID, the World Bank and the Japan Trust Fund are putting in similar amounts of money. So those four funding sources are giving that sort of level of assistance.
Visit these websites
The Sir Mark Oliphant Conference Series is supported by the Australian Government under the International Science Linkages Program.
© 2008 Sir Mark Oliphant Conferences.
Website design acidgreen
