Select Committee on Science and Technology Minutes of Evidence

Examination of Witnesses (Questions 199 - 217)


Dr Klaus Stohr

  Q199  Chairman: Dr Stohr, thank you very much for coming to talk with us. I think you have been in the room for a while already so I do not need to alert you to the fact that we are being web cast, et cetera, but thank you very much for coming to join us. Let me start the discussion by asking you if you can give us the latest information that you have on human cases of H5N1?

  Dr Stohr: You are aware that the first wave of spread of human cases linked with the H5N1 outbreak in birds in Asia started in December 2003. To date there are 121 individual recorded cases: 91 in Vietnam, 19 in Thailand, seven in Indonesia and four in Cambodia. The first wave subsided in around March/April 2004, the small second wave started in August/September last year, and the third wave started in November last year and has since not subsided, so we are in the third wave which has now lasted for a relatively long period of time. This comes as something of a surprise as we had thought that the activity of the disease, the virus, in animals would be linked to the season because the cooler the weather the more the virus can survive in the environment and the more people are going to be exposed. We have noticed a reduction in the number of cases since around November last year. There is another increase now occurring with the beginning of the winter period in Asia already. All these cases have occurred in healthy, mainly young adults. The case fatality rates, the number of those who die of those infected, is in the area of 60 per cent. Well, 63 people have died of the 121, so it is a bit more than 50 per cent. The surveillance in Asia is based on hospitals. Hospital-based surveillance implies that only the severe cases are being detected and considering that normally every infectious disease has a relatively wide spectrum of the disease from mild to very severe, it is very likely that a considerable number of cases with mild disease have gone unnoticed. There are a large number of respiratory disease symptoms which resemble H5N1 in humans so it is not difficult to overlook mild cases. People will not go to the hospital if they think they have normal influenza or any other viral respiratory disease. Human-to-human transmission is something we know an avian influenza virus can cause. We are not surprised to see a limited, we call it dead-end street, unsustained human-to-human transmission of the avian influenza virus. This has happened presumably several times. It is part of the epidemiology of the virus and nothing comes as a surprise. However, it would be a profound change if this virus mutates and acquires the missing capacity to cause a pandemic with easy and fast human-to-human transmissibility. The disease is still an animal virus and it is an animal virus which is very difficult to be transmitted to humans. There may have been tens of thousands exposures of humans to the virus. You have all seen the pictures of the culling of more than 150 million birds in Asia and the way this was done - by people unprotected and people who have not been able to have access to drugs - so there has been huge exposure to this and a relatively small number in comparison of severe cases. The disease is not easy to transmit and the disease will certainly also depend on the infectious dose, and presumably infectious people do not excrete enough to infect another person or there are some other factors which play a role. The human-to-human transmission is part of the epidemiology of avian virus but it is non-sustained dead-end street and there are normally not more than one or two links in the transmission chain.

  Chairman: Thank you. Lord Taverne?

  Q200  Lord Taverne: I understand that the recent research shows that the 1918 influenza virus was a pure avian virus rather than a combination of human and avian viruses. What are the implications of that? Does it make the development of transmission from person to person more likely or less likely or are there other implications?

  Dr Stohr: The research on the source of the 1918 virus has been going on for several years. There has been the expectation that when it is finished we will have evidence that the 1918 virus is causal or has come directly from animals. Thus the WHO's recommendations for control have been based on the assumption from two or three years ago that the 1918 virus was a pure avian influenza virus. Therefore, our efforts have always been focusing on trying to eliminate the animal reservoir because without the disease circulating in animals there would be no transmission to humans and there would be no chance for this virus to either - and this is one of the possibilities - to readily mutate and change its transmission characteristics. That is one of the ways an influenza pandemic virus emerges or a reassortment takes place. A reassortment would mean that the genetic information of the avian influenza virus and the human influenza virus would be swapped and the progeny virus would have the transmissibility of the normal influenza virus but, of course, the pathogenicity of these killer strains of the avian virus. The 1918 virus differs. A relatively small number of amino acids (and the influenza virus has more than 4,000 amino acids so in less than 40 of these amino acids) are very peculiar and very specific in the 1918 virus and differ from all of the other H1N1 viruses which were isolated after this. What this tells us is that presumably the small change in the H1N1 virus, which is a pure avian virus, not readily transmissible between humans, could suffice to make it an animal independent virus, jump to humans and be maintained in humans. This is a very important piece of knowledge which corroborates the WHO strategy to focus on the control of the disease in animals in order to reduce profoundly the risk of a reoccurrence of a pandemic. That is certainly one of the key conclusions. Is it going to be more likely that this virus will now change and cause human-to-human transmission? The answer is very difficult to give. The last three pandemics for which we have microbiological evidence of the source of the virus were caused by influenza viruses which were either partially or fully composed of genetic material from pig or avian influenza virus. In other words, all the past pandemic viruses in one way or another were linked to animal influenza virus. Two of the last pandemics were caused by reassortment, the mixing of avian, pig and human influenza viruses. One was caused by a pure avian virus. This is the only knowledge which helps us currently to deduce the likelihood for the next pandemic to be caused by pure avian influenza virus or by reassortment. There is a two to one probability in the future for a pandemic to be caused by pure avian virus (one) as opposed to a pandemic being caused by a reassortment of virus (two). The broad implication of these findings is certainly that we need intensified research and an understanding of how animal influenza viruses can change and which changes in an animal influenza virus will make them more pathogenic and transmissible to humans.

  Q201  Lord Taverne: Does what we have learned so far suggest that it is likely to be more deadly and more virulent by being a pure avian virus?

  Dr Stohr: Yes, all the evidence suggests currently that pure, non-adapted animal influenza viruses cause more severe disease in humans. The pathogenicity, the way the virus caused disease in humans in 1918 resembles very much what we are seeing H5N1 doing currently in humans. H5N1 elicits an over-exuberant immune response which is then causing damage to several tissues in the body rather than what we are seeing the normal influenza virus doing which replicates in the cells in the lung and through that mechanism causes the disease. Principally, this primary viral pneumonia issue, as we have seen in 1918, is very similar to what we have seen H5N1 doing currently in Asia in humans.

  Q202  Lord Soulsby of Swaffham Prior: You mentioned surveillance systems and laboratory facilities being largely based on hospital cases, but I was also taken by your comment that this is an animal virus. In terms of animal viruses there is a report from the World Health Organisation of avian flu H5N1 from cats and we know it goes into pigs and human viruses go into pigs. What surveillance is being done in the Far East of animals other than chickens of the presence of viruses in a sub-clinical way and possibly as carriers of the H5N1 virus?

  Dr Stohr: I might not be the best person to answer this question because I am working with the World Health Organisation and I am Head of the Global Influenza Programme in WHO. What I can say certainly is that the surveillance of the disease in animals in most of the countries in Asia tends to be sub-optimal. Just giving you one example, in some countries human disease was first reported before there was knowledge of the presence of the disease in animals in a situation where we know the virus does come from animals. Secondly, what is certainly also ill-understood is what role might pigs play as a possible carrier for H5N1 virus, or as we call it a mixing vessel, where the human influenza virus and the animal influenza virus mix and this progeny virus can cause a human influenza pandemic. There were isolated events of avian influenza H5N1 found in pigs in China. However, there has been no large-scale serological investigation studies in Asia which would give us a better understanding of whether or not this virus circulates in pigs. If that were the case, the chances of the outbreak of a pandemic would certainly increase profoundly. The number of pigs exceeds by far the number of humans in Asia and there is the proximity of pigs to poultry where the virus is circulated. As far as the human influenza surveillance is concerned, there are profound differences between countries. The sensitivity of the human influenza surveillance depends on very many factors. The first one is primary health care centres, being aware and being alert. The para-medical staff, the nurses, the doctors in the field need to be aware of the clinical science and need to be attentive. Then the patient has to have access to a hospital within proximity. Normally in Asia it takes four to five days before people will come to a hospital. That means they are very sick and then they have to be transported over a very long distance. In the hospital there should also be a good triaging/screening system in the hospitals. Samples should be taken and containers should be available to ship the samples. The country needs to have a laboratory which is capable of doing a good diagnosis. Very often laboratories are not able to propagate virus because this requires higher laboratory standards. At the end of the diagnosis being made the country must be willing to share the information rapidly and completely. All this has to be in place and we see here a difference between countries. For instance, the last cases in Indonesia have been very rapidly investigated, very well detected, and the virus has been shared very quickly with the WHO. There are good examples in Cambodia where the WHO has been invited to conduct field investigations very quickly. Without field investigations surveillance is severely hampered because knowing that there is a sporadic outbreak is one thing, but the key thing to understand is whether this is the only case or are there more cases in the family who have not made it to the hospital, and whether there is somebody else sick in the village or in the vicinity because that would indicate and signal the beginning of the pandemic, possibly by a virus that has increased human-to-human transmissibility.

  Q203  Lord Soulsby of Swaffham Prior: Do you think in coming westwards into Europe and in this country that there should be more testing of animals other than the avian host to see what the situation is?

  Dr Stohr: We have a relatively poor understanding, not only in Europe, about the pathways of transmission of avian influenza virus. If somebody were to screen birds one would expect between 10 to 20 per cent of them would have antibodies to a large variety of different influenza viruses. How these viruses are being spread between animals and within regions is relatively ill-understood. The focus for Europe perhaps now should be to have an alert general public, and to have a very well-educated group of ornithologists and hunters, as those who would be the first to detect death in migratory birds. We also need a very alert system in the agricultural sector so that the first cases of H5N1 in domestic poultry would be detected rapidly. Those are certainly the key component of surveillance in Europe.

  Lord May of Oxford: Just a very quick observation rather than anything else, and it is a supplementary to what Lord Soulsby was saying. I have a manuscript that is currently under review from somebody who works in conservation biology that has tried to pull together the serological data on a whole set of different animals, mainly in South East Asia and mostly mammals, that show evidence of H5N1 at fairly low levels, but even so it is an interesting thing because the medical aspects of it have intersections with pure conservation biology aspects. You can take a larger view and say there are other reasons why we should not be shuffling these parrots around the world like this.

  Q204  Baroness Finlay of Llandaff: You were talking about having been invited into Cambodia as the WHO. I wonder whether you were able to do serological testing on contacts in Cambodia and whether you have got any data to give a better indication on the case mortality rates, because you are citing 60 per cent from the cases that are known and that is a long way away from the 3 per cent that is being spoken about.

  Dr Stohr: Perhaps I can first try to answer the question about Cambodia. The WHO has country officers and we have one-third of our staff working in countries. The country officers in Cambodia have been working very closely with the ministry of health and immediately after the first suspicion that a H5N1 case had occurred our epidemiologists were invited to go to the field and support the ministry of health, and that was found to be, I hope, very useful on the part of the ministry of health and on the part of WHO, which provided the information necessary for risk assessment not only for this country but also global risk assessment to better understand outside this country what impact these cases might have. So far as case mortality is concerned, as I said, hospital infection control is the foundation, so we will be expecting a much higher case fatality rate, and perhaps a biased case fatality rate, because the mild cases are not being seen. The models, which are being based on 1957 and 1968, would anticipate that in a mild pandemic between 25 to 35 per cent of the population would be infected. Of those 35 per cent, 46 per cent would be ill but would not need to see a doctor necessarily. A bit more than that, 52 or 53 per cent (we are close to 100 already) would require some type of medical care according to the standards and activities in Europe, North America or Japan. Around about one per cent would be so ill that they would have to be hospitalised and 0.2 or 0.3 per cent would die. These are the figures that are being based on the pandemics in 1957 and 1968, which were very mild pandemics. The total number of deaths would then be two million to 7.8 million deaths and there would be up to 28 million hospitalisations. Those who would be sick and require medical care or health treatment would be in the billions. That needs to be put into perspective. In 1918 the situation was more severe but it was not even near to what you are currently seeing so far as the numbers are concerned, which I believe is because they are biased numbers and they do not consider that there are mild cases and it would not be, from a biological and evolutionary perspective, advantageous for it to kill all its host because there the selective pressure on these viruses would be to find a good balance between killing the host and then finding time to be transmitted to the next one. That is what the evidence currently suggests, that the case fatality rate, as we have seen in 1918, might be at the very upper range of what should be expected.

  Q205  Baroness Finlay of Llandaff: I was wondering whether serological testing in the field has been able to support those theories?

  Dr Stohr: There have been samples taken in Cambodia from 1,200 people and China has taken samples in 16 provinces from four occupational groups. However, these samples have not been linked with very good epidemiological data to understand what the risk factors are. Fortunately, recently there have been good studies conducted in Southern Vietnam with contacts. These arrived last week at one of our H5N1 reference laboratories in Hong Kong. These samples can give a short answer but, unfortunately, there has been not been enough research possible in these countries for various reasons. Ideally, one could look, for instance, into blood donors and one would also have a biased population but that would give a very good indication of what is the background noise between these latest instances.

  Q206  Baroness Perry of Southwark: Just a supplementary to your very helpful answer to Lord Soulsby's question. In the countries most at risk of a pandemic, do you feel that there are political factors which are preventing adequate surveillance? Also, in your own work in the WHO what sort of co-operation are you getting at national level from these countries?

  Dr Stohr: Principally, on the part of WHO of course we assist countries and countries assist us also in order to support the larger public health community. The assistance which we would hope to receive affects three areas. One is the very rapid detection and notification of cases. This will be very important to ensure that the rapid response stockpile of antivirals can be used in a timely fashion. It will also be important that the virus is shared with the WHO. That is the assistance we also hope we would receive. That would have implications certainly for the announcement of the pandemic because of course epidemiologically one cannot confirm that a pandemic virus has emerged; one has to do this in a laboratory. Secondly, it is important for WHO virologists to provide up-to-date diagnostic test kits to our member states. Thirdly, of course the virus would be important to turn it into a vaccine prototype 3-D that can be given to the governments. The third area is that currently we would also hope that countries assist in co-ordinating the research effort. Research efforts are particularly important in the area of the epidemiology of the disease. We still have only a limited understanding of the excretion pattern of the virus and the antibody genetics and other factors which could inform policy on hospital infection control and how people are infected. All these areas would require co-ordinated research for which we need more assistance from our member states. The serological studies were mentioned and that is something we would be keenly interested in conducting, but it is very costly and it requires support from countries. Thirdly, what is also important is the prevalence studies of the virus are being initiated in animals. There is a different level of support required from the currently infected countries. I have given already a good example from Indonesia of the rapid sharing of viruses, participation in field investigation, prompt announcement of cases. Cambodia is in a very similar situation. Thailand had a web site from the very beginning on which every day the number of suspect cases was being reported as well as those taken off the list which turned out not to be positive or put on the list those which turned out to be positive. So these are very, very good examples of good collaboration. There are various surveillance systems, particularly in animals. I think the presence of an economically important animal disease may or may not necessarily be in the interests of a given country. If a country announces that it had H5N1 in animals it would have to suffer consequences as far as trade is concerned. Countries like Thailand, for instance, which is where poultry production is the fourth largest contributor to the gross national product, will be heavily affected from the consequences of this announcement. Sometimes also the control of the disease in animals may not necessarily support the reduction of the risk for the disease in humans. So there are certainly areas. What we are seeing in the field generally is there is an increased understanding of the importance of fast investigation, reporting, and that comes certainly at a time where several countries have to realise that the disease is endemic in their territories. The FAO and other agricultural organisations would conclude that this will not disappear on its own and even with co-ordinated efforts it will take several years before H5N1 is going to be eliminated in Asia, so we have to live with the situation currently where we have to be planning long term and as long as the disease does exist in animals we have also to expect that sporadic human cases will occur and there is the risk of a pandemic.

  Q207  Baroness Perry of Southwark: Are you confident you will have that level of co-operation in China?

  Dr Stohr: I cannot speak for the agricultural sector because that is not my main area of knowledge and influence.

  Q208  Baroness Perry of Southwark: I understand that.

  Dr Stohr: But what we are seeing in Thailand is that the government has made enormous efforts to control the disease. There have also been announcements that Thailand is seen as an example on how the disease can be controlled in animals. The unfortunate reality now after one and a half years is that certainly five of 72 provinces in Thailand still report the presence of disease in animals. We have to come to grips with the reality which is if we do not devise a strategy which is long term and which takes into consideration the need for international co-ordination that we may have to live with the disease for much longer than anybody anticipated before.

  Q209  Baroness Perry of Southwark: I asked about China, do you have confidence that China will co-operate?

  Dr Stohr: China has shown after SARS, as you know very well, that these are not only being reported and recorded but also announced internationally. I can give you an example today, for instance, there is information on one suspected case of H5N1 in humans in China. This has been reported and very transparent. We have received in the last couple of months viruses very, very regularly from the agricultural sector. The cases in Central China in Inner Mongolia near the south western border have been reported immediately and teams were allowed in. On the other hand, of course we see that China has been using billions of doses of H5N1 influenza vaccine for many years. That may have been the reason. However, as I said, in the disease currently if there are outbreaks or if there are events the WHO is notified very quickly and we are satisfied with the co-operation.

  Q210  Lord May of Oxford: A quick supplementary. The best way, if the virus does hop into the human population human-to-human, would be to try and control it at the place where it first appeared. I am sure you are more familiar than I am with the two papers, one from the NIH and the other from here from the Ferguson group, that suggest if you acted very promptly with appropriately geographically targeted antiviral prophylaxis you could, provided this flu is roughly as transmissible as earlier ones, control it. I wonder whether your feeling is that that is a realistic hope, and if it is a realistic hope whether the machinery is being put in place in South East Asia, which is probably the right place, to try and achieve it or is it just that the realities on the ground preclude the possibility of doing that?

  Dr Stohr: The reality on the ground would not preclude it. The chances are not huge that it is going to succeed. However, it could not be forgiven if we did not try to do it even though the chances may be relatively small. The models say it can work; the reality would say we have areas in Asia where 80 per cent of the country can only be reached by four-wheeled drive vehicles; where the challenge is to treat in 10 to 15 days 80 per cent of a population with a drug which has to be taken over a certain period of time; where you have to have a very high compliance rate; where you have to practically seal off the territory and make sure nobody gets in or out. It is a huge challenge which has to be accepted and taken into account. We will have by the end of this year, from information from one of the major antiviral, one million treatments and by March next year three million treatments to try it out. Our current challenge is to put into place a system whereby these drugs can be rapidly deployed in the field. We are not there yet, we have to say that. It will also require co-operative companies to try it out to pilot test the system. We know it can be successful. Polio has shown that with the vaccine given to a large amount of people in a very short period of time. The infrastructure is there and the knowledge is there but we have to try it out.

  Chairman: Lord Patel?

  Q211  Lord Patel: I think Lord May asked the question that I was going to ask. You mentioned just now the one of the industry companies, Roche, had offered three million doses of Tamiflu. I think what I hear you correctly to say is that you have plans to use this to contain the disease if there is an outbreak and you have the means of proper storage and distribution?

  Dr Stohr: The storage, the warehousing, the rotation is taken care of by the company. The WHO is not involved. It is reassured that this has been taken care of by the company. The company will also deliver the product to the destination which the WHO will identify within a very short period of time - 24 to 48 hours. This will all be taken care of by the company. Our task will start with a risk assessment triggering the rapid deployment and then, of course, the putting into place the infrastructure so that the drug can be given to those who need it. That is the real challenge now. That is what I am saying where we are not where we would like to be.

  Chairman: Let us move on. Lord May had a question.

  Q212  Lord May of Oxford: I would like to ask you not to be too polite about this and tell us exactly what you really think. What is your assessment of the UK's pandemic influenza contingency plan? How do you feel it compares to other countries, both in the OECD and others?

  Dr Stohr: The UK has published its contingency plan in, I think, the autumn of last year. If you compare this internationally there are less than 50 countries which have published their national pandemic plans. Sometimes if you count the pages the quality is one page. Others have very comprehensive plans - Canada, UK, US, Australia and also Japan. Very few countries have translated their plans into national law. The quality of the plans, in our view, will not only depend on how comprehensive it is but also how much the plan is translated into policy activity and how much is it underpinned with activity in the field. The UK plan has all the components - communication, co-ordination, surveillance, information-gathering, the public health response, the medical response, the Civil Service response to reduce social disruption, but the WHO has not conducted a very in-depth analysis of this plan. That is not our duty but I must say we would be very happy if we were to be inundated with requests to assess pandemic preparedness plans from developing countries. Unfortunately, this is not the case. We are focusing on trying to find resources and support also from developed countries in helping us to increase the number of countries in the developing world to which have thought through the huge challenges of a pandemic in the absence of antivirals, which is probably the case for most of them.

  Q213  Earl of Selborne: Very quickly, following on from that answer, given the risks to the international community and the fact that, as you say, not every country is able to produce the contingency plans you would wish, have you got just one headline message as to what the international community needs to do in the longer term to reduce the risk of influenza pandemics?

  Dr Stohr: Influenza pandemics - and I would like to focus on those - have been part of our normal life in the past and they will be coming in the future. However, there is a panacea and there may be a solution not only to the current unsatisfactory use of the influenza vaccine but also to the situation where there will be (and there is no doubt about this) a lack of vaccines and antivirals during the next pandemic. There are no possibilities currently to stock vaccine. We believe that the international research community - and here governments will have a very important role to play - should try to find solutions to develop a vaccine compound. There are indications that it is possible with $500 to $800 million and in 10 years' time that a vaccine can be developed which can be given to protect not only against influenza but also in the long term to prevent the impact of a pandemic. That is one response. The other response certainly that is required is long term further strengthening of disease intelligence, disease detection, and alert and response mechanisms by countries as well as by the global response and alert capacities to detect the emergence of infectious diseases and control them. I would like to expand a little bit more on this if time would permit because that is very important for the WHO. Our capacity to respond to emergences like a influenza pandemic will depend on the capacities of the member states to detect and respond and also upon the WHO's and other international players' capacity to regionally and locally co-ordinate the response. If the WHO is going to be able to live up to the expectations, the resources which are currently available to do this will not suffice. There is no doubt about this. We would hope there will be in the future a match between the demand to transform the WHO into an operational organisation as well as the support which we receive from the international community.

  Q214  Chairman: Could I infer from what you are saying there that your feeling is that overall, let's say, the leading western nations are not doing all that they could do to help contain this situation?

  Dr Stohr: I believe the world as well as each individual country is certainly not entirely prepared for an influenza pandemic. A pandemic is always about damage control. The best preparedness will not avoid people dying or getting sick. 25 per cent of antivirals for a country will mean a reduction in hospitalisation by about 75 per cent but there will still be people who will require hospital care and there will still be people who are going to die. An ideal pandemic preparedness status does not exist. We can only get as close as possible. There are currently ten countries which have antiviral stockpiles for more than 20 per cent of their population. There are 37 countries which have put orders in for antivirals. There are nine countries which will have access to vaccine, and globally there are 250 countries that may. This shows what the gap is. If all the antivirals which exist currently were to be homogeneously globally distributed, it would suffice for two per cent of the world's population. Seeing this in context I think the argument as to whether or not we are properly prepared will come on its own.

  Q215  Chairman: I think in this country we are placing an order for something like 14 million Tamiflu doses. How does that compare with other countries? How does that compare with other European countries like France for example?

  Dr Stohr: It compares very well. That will suffice, if I understand it correctly, for 25 per cent of the population in the UK -

  Q216  Chairman: - It does or it does not compare very well?

  Dr Stohr: It compares very well. France is the only country which has a larger stockpile per percentage of the population than the UK. Australia has about the same stockpile. Canada, if one takes together the central and state resources, will also have less than the UK has. However, taking into account France, which has the largest stockpile per capita and it also has the largest vaccine production capacity in the world, that shows the trust of some governments in their capability and capacity to deliver vaccine in a timely fashion.

  Q217  Chairman: So France is the best equipped, is it?

  Dr Stohr: At least as far as drugs are concerned that could be one of the conclusions, right.

  Chairman: Dr Stohr, thank you very much. Do members of the Committee have any more questions? We could go on a long time but we have run out of time really. Let us thank you very much for coming and giving evidence. Please let me ask you if other things occur to you that you think it might be useful for us to know, would you let us have that information? Thank you very much.

previous page contents next page

House of Lords home page Parliament home page House of Commons home page search page enquiries index

© Parliamentary copyright 2005