NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Institute of Medicine (US) Board on Population Health and Public Health Practice. John R. La Montagne Memorial Symposium on Pandemic Influenza Research: Meeting Proceedings. Washington (DC): National Academies Press (US); 2005.

Cover of John R. La Montagne Memorial Symposium on Pandemic Influenza Research

John R. La Montagne Memorial Symposium on Pandemic Influenza Research: Meeting Proceedings.

Show details

5Afternoon Discussion: Reaction to Rapporteurs, Day 1

DR. FINEBERG: I will introduce one question to get us started, and it relates to the opportunities for studies of pathogenicity, and the degree to which those studies can be conducted in a sense, on viruses free of hosts, or whether those studies have to be conducted always from the beginning, thinking about a particular virus and particular host.

When I was hearing the first group's discussion about pathogenicity and the studies of viral genome, it was not clear to me what the strategy would be vis-à-vis variety of hosts. And since it is evident in nature that different hosts, for the same virus, have different pathogenicity, how is it possible to study pathogenicity without automatically thinking of it in terms of the combination of a virus and a host?

So, that is the question. And if the presumption is correct, what does that imply about the complexity of a research strategy to investigate pathogenicity? So, let me put that on the floor and ask anyone who would like to comment.

PARTICIPANT: If I can just summarize very quickly, in Group 1 I think this was discussed. I think other people from Group 1 can chime in; it was discussed quite a bit. And there was certainly a feeling among a few physicians in the audience that it was hard to understand pathogenicity for human disease without humans somewhere in the process other than doing the experiment. We debated what model was the best model. After some debate, it was felt that the ferret model was everybody's bet as the best model—given no model is perfect

DR. FINEBERG: Could you say just a word about why immunologically or otherwise, the ferret seems to be a preferred model as a small animal for human study?

PARTICIPANT: I am not a ferret expert. I think what was said in the meeting was that by and large, virulence factors co-segregated between humans and ferrets, so that viruses that tended to have a phenotype in humans, also tended to have that in ferrets more than in other models. I think there was the acknowledgement that some models, for example pigs, hadn't been tested to the degree to know whether they might be better than ferrets. But by and large, it was felt that ferrets were more reflective than say mouse. And there was also a strong feeling in the meeting that mice, which lacked the Mx system, that part of their immune response, and specifically their innate immune response, was quite distinct from what goes on in humans. And that might really throw things out of kilter.

PARTICIPANT: I was also in Group 1, another point that I would make is that I think we all agreed on is that any sequencing data and attempts to correlate sequence with virulence need to be accompanied by very careful and detailed clinical histories of the individuals from whom these samples were obtained. So, there was an epidemiologic link to the genomics was very important to all of us.

PARTICIPANT: I wanted to make a comment to kind of drive us away from the pathogenicity discussion. In regards to the third and fourth presentations, which I think were the fourth and third groups, both of them touched on something that I thought was incredibly important, and in some ways I wish they had gone further with it.

The third group talked about operational infrastructure research. And the fourth group talked about animals, the need for surveys of knowledge, attitude and practice among people who raise poultry, the development of economic incentives, and the development of educational programs all of which are subject to research. I think that as part of our long-term research strategy, we should be putting much more effort than we have into the non-biologic aspects of a pandemic, because there are factors that strongly influence the spread of a virus, and the ability to control the spread of a virus that really have very little to do with pathogenicity and with hosts and molecules and all those areas that we, as scientists, drawn to. I also think that if a pandemic emerges, we will not lose so many people to Tamiflu resistance as we will to the fact that they did not understand the importance of basic hygiene or they did not have access to personal protective equipment. I think those are factors that determine how a pandemic flows.

I also think that the type of research that we need to do will have to draw much more heavily on some people who we do not usually partner with. We need to talk to the people who know about immune modulators, but also the people who know about health education, and the people who know about international trade law, the people who know about organizational dynamics, some of the factors in getting these drugs out to the field, getting the supply chain, which we heard about earlier today.

Those are issues that we have not spent the time researching. It is a much more basic set of questions that we have. In contrast, we have gone very, very far with research on some of the biologic questions.

PARTICIPANT: I would like to return to the previous discussion, if we could, and maybe ask the group if there is any information that we can glean from historical perspectives of pandemics. Over the past few months I have been trying to understand the future of the pandemics, and have gone back and pulled out a lot of the original work or writing that have been done. But I think there has been a lot of what I would call also folklore around previous pandemics. And when you really start to peel the onion and look at what's there, it is clear that 1918 was not unique. It looks like 1830-1832 had a very similar picture of deaths primarily between 20 and 35 year olds, and a very classic W shaped curve again, just like we saw in 1918. Whereas, if you get into the 1880, you get into the other ones, there are at least 10 in the last 300 years; they really fall into two camps, those that had the classic accelerated or exaggerated Y shaped curve, and those that had the W shaped curve. Which would suggest to me, that there are several mechanisms for pandemics to occur?

The underlying necessary cause of a pandemic is a new virus for which there is a lack of overall population-based immunity. Pathogenicity is by definition, the virus' ability to cause disease. Virulence defines the virus' ability to cause severe disease. The virulence factors may determine which type of pandemic it is.

In other words, is it one of the W shaped curve, or the accelerated or exaggerated Y shaped curve? The answer has tremendous implications for how we deal with it, because one is probably more of a secondary bacterial pneumonia type picture and the other not. I think that this area needs a lot more research because we have artificially lumped all pandemics under one category, when in fact they may be of a common origin, but not of common outcome or cause. Perhaps an historian in this area could comment on it.

PARTICIPANT: One of the problems with deciding whether there will be another 1918 or not is that there will never be another 1918, because 1918 was a year replete with secondary bacterial pneumonia in many of these victims. We talk about the high mortality in that year. It was also true that 98 percent of the population had ordinary three day fever and flu. In addition, the summer of 1918 was an unusually cold one, in which case there would be more people indoors and the possibility for more abundant transmission and so forth. Another point about another 1918 is again, it is not 1918. We have learned a great deal. We know how to make vaccines. We know what kind of virus is causing this; we have antivirals.

PARTICIPANT: I think these points are really good. The point that I would make to that is that I think we really have to consider in a sense each pandemic as a unique event. And if you have a unique circumstance—when a unique virus emerges that can circulate in humans—that is going to be different from one pandemic to another in terms of what has happened in the past, what kind of other strains have circulated in the past, and what the immunity in the population by different ages would be. So, I think that we should not generalize. It is clear going back to the molecular biology, looking at the genetics of the 1918 virus versus the 1957 or 1968 virus is clearly where we are seeing differences by which pandemics may emerge, and clearly that is a useful model.

Another difficulty, however, is that even just looking at 1918, with all the work that has been done for 80 years on this virus; we are still missing so much of the primary data that we really need to understand what happened. We have no pre-1918 human samples to study. We have no sera collected before 1918; we are never going to have those data. And if we are not going to have data before 1918, it is going to be very, very difficult to go back to what happened in the very distant past. So, we have the epidemiologic evidence that exists, and we can use that data to sort of tease out hypotheses. But the problem is that going back in the past is going to be extremely difficult to develop experimental models.

DR. FINEBERG: It seems that the fundamental problem in one sense is that the time horizon for an observation—a pandemic—goes back centuries to obtain a handful of type of observations. And in our lifetimes, where we are accustomed to looking at organisms that turn over in the hours, rather than units of observation that take decades or centuries to become apparent, we are not accustomed to thinking as cautiously as you are now advising us about generalizing from such a paucity of real observation. And I think that is a pretty important caution. At the same time, the premise of all of our work on thinking of a research strategy has to be that we believe what we learned from the past can help us in the future. And so, being modest about it, and being properly qualified I hope at the same time will not deter us from making the effort that we are all engaged in today. And I know you certainly wouldn't want that to be the case.

PARTICIPANT: I think a very important point was made, and was made very quickly, and that is about the usefulness of having challenge strains available, contemporary challenge strains of virus. Economically, this can be very cost effective to do this, because instead of having to go to complicated and very costly and dangerous clinical trials, it means that you could develop at least attenuated strains of contemporary viruses to have those available. I say this with feeling, because I have recently been through an experience with a clinical trial where an up-to-date strain was not available. So, we came away with the conclusion that we could not affect enough of a control population to come up with any kind of answer. So, I think cost effectively, this would be very important.

DR. FINEBERG: Let me ask a question of the group who were reporting on the development of antivirals. The timeframes of that set of studies struck me that one would really have to look quite concretely and specifically at the state-of-science for each line of activity that one is thinking about. I wonder if anyone in that group could comment to help us understand where, if the group did discuss this, the most promising direction of development might rest today, and where things were still at a much more speculative stage across that list that we saw arrayed.

PARTICIPANT: As everybody is aware, there is a relative — and I just say relative paucity of drugs in the pipeline that are unique and different for the treatment of influenza infection. There are several that are available, but they have not been studied appropriately in high risk patient populations. We have tried to emphasize that, and quite honestly, it can be done immediately. That includes evaluating drugs like oseltamivir in infants less than 1 year of age.

There is a fundamental problem that we have, and that is the pre-clinical animal toxicity data would imply that oseltamivir is toxic for the newborn brain in a rat model. We have to understand how we can circumvent that problem. And if we can not, we have to find a back-up drug such as peramivir. Peramivir has gone through phase 3 studies. It is not orally bioavailable. It can be given parenterally. It is available for a group to develop as quickly as possible. After that, then it becomes a question of the long acting neuraminidase inhibitor applied topically, or alternative strategies. But there are clearly two or three clinical protocols that could be developed and implemented within a short period of time.

PARTICIPANT: These drugs are available and could be studied in next influenza season. There are other possible approaches that presumably in the next year or two could be undertaken with regard to looking at resistance emergence strategies to reduce the combination therapies with antivirals. Another point of discussion within our group was the need to start studies of the available drugs to determine their impact, and their dose-related impact on some of the important outcomes in H5 disease in South Asia.

DR. FINEBERG: Yes, I thought that point was really quite telling, as well as some of the others that emphasized the importance of proactive planning for clinical protocols at the time outbreak would occur. We learned the lesson from SARS on how valuable it would have been to have had those protocols in place, accepted and ready to go across a whole spectrum of geographic locations. I think that is something that could definitely be done in a short timeframe, and in preparation for the next flu season.

There is relatively little, as one looks at the list that we have arrayed so far, that actually would have a bearing on something that happens in the next 12 months. And I think the reality of all of us in thinking about the importance of a research strategy is ultimately to work it back to what we can do in what timeframe to provide the kind of preparation, preventive and ameliorative strategies that we all know are going to be needed.

The discussion that we have had is of course premised on the great worry of the grand pandemics. But a lot of what we are talking about truly is going to be relevant year after year to endemic influenza that is still underappreciated as a significant source of mortality and morbidity in countries all around the world. So, I think there will be a true value coming out of this discussion, even if we are fortunate to not experience for some time, any of the greatest threats that worry us.

Copyright © 2005, National Academy of Sciences.
Bookshelf ID: NBK25573


  • PubReader
  • Print View
  • Cite this Page
  • PDF version of this title (6.5M)

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...