Swine-origin influenza H1N1 as of now

1 December 2009

nytimesThe New York Times has published an editorial entitled “The Swine Flu, as of Now” which presents their views of the current influenza pandemic. The piece covers the number of infections, hospitalizations, and deaths; the distribution of vaccines, and lists those most at risk for severe disease. It begins with a positive view:

So far, the news about swine flu is better than expected. The pandemic may have reached its peak and is heading downward in all regions of the country; weekly deaths from the swine flu have started to decline; the virus remains relatively mild; there seem to be few claims of serious side effects from the vaccine; and despite widespread complaints about shortages, vaccine supplies are steadily building up.

The wording suggests that the pandemic is over, but I would not agree. Based on the patterns of previous pandemics, it’s likely that a third wave of infections will occur later this winter. To their credit they acknowledge this possibility several paragraphs later:

This is no time for Americans to let down their guard…if the swine flu follows the pattern of some previous pandemic strains, it could return in a new wave early next year.

When the editorial delves into viral genetics it moves to unsteady ground:

There have also been scattered reports of mutations in the swine flu virus that cause harm deep in the lungs of some patients or make the virus resistant to one of the standard drug treatments. However, neither mutation seems to be spreading widely.

As I’ve written before, there is no evidence that the G225D amino acid change in the viral HA protein is causing more severe disease in the lower respiratory tract. The fact that this change has been identified in viral isolates from a number of fatal cases does not prove a causative role. To do so it will be necessary to identify the change in a significant number of cases with severe lower tract disease, and not in individuals with less serious infections. But I do agree that neither the G225D change nor the mutation conferring resistance to Tamiflu is yet spreading widely.

But why does the Times still use the name ‘swine flu’? The virus is of swine-origin but is now clearly a human strain of influenza virus. The virus can infect pigs (and other animals) but has not caused outbreaks in these animals. If it does it will have changed sufficiently and will no longer be a human influenza virus. Perhaps like the ‘Hong Kong flu’ of 1968, or the ‘Russian flu’ of 1977, ‘swine flu’ is simply a moniker, not a biologically correct appellation. ‘Swine flu’ is clearly more catchy than ‘swine-origin influenza virus’.

  • gsgs

    > The wording suggests that the pandemic is over, but I would not agree

    we didn't see another wave in the Southern hemisphere.
    There was one in 1918/9 but not in 2003/4 in USA
    what was 1957,1968,1977 someone please add …

    > there is no evidence that the G225D amino acid change in the viral HA protein is causing
    > more severe disease in the lower respiratory tract

    I don't understand. We have several reports from fatal cases with this mutation
    saying it causes severe disease in the lungs. No evidence ?

    > The fact that this change has been identified in viral isolates from a number of fatal cases
    > does not prove a causative role.

    and you didn't disprove it, so we are back from proof to evidence gathering.

    > But why does the Times still use the name ’swine flu’? The virus is of swine-origin
    > but is now clearly a human strain of influenza virus

    yes, but “human flu” is ambiguous – it could be H3N2 or old H1N1.
    I call it mexflu.

  • ARR_3

    “…there is no evidence that the G225D amino acid change in the viral HA protein is causing more severe disease in the lower respiratory tract.”

    Your comment above is deeply flawed. Four deaths out of ten cases (from early in the recent Ukraine outbreak) all of which showed positive for the G225D mutation whilst the SIX SURVIVORS FROM THIS SAMPLE WERE NEGATIVE FOR THIS CHANGE is strong evidence of a putative causative relationship between increased virulence and viruses containing the G225D mutation. Your good at maths: what's the statistical probability that such an outcome would occur by chance alone?

    Clearly, whilst a larger sample size (and appropriate sampling) is now required to move from a putative to a demonstrated causative relationship – to say there is no evidence is simply ridiculous Vincent.

    “But I do agree that neither the G225D change nor the mutation conferring resistance to Tamiflu is yet spreading widely.”

    Agreement – based on what? Perhaps an inner voice told you?

    Plenty has now been written on the hopelessly flawed and biased clinical surveillence methods relied on by those ostriches at the WHO/CDC (cf mixed populations-consensus sequences) to repeat here.

  • http://www.virology.ws profvrr

    My argument is not deeply flawed. It is based on how science is done.
    Keep the discussion serious here please – leave out the inner voices.
    Four and six are not numbers on which to base a conclusion that the
    change is lethal. There are more data in ferrets which indicate that
    the change does not compromise lethality. But we'll see – when the
    data are in we can make a conclusion. As for Tamiflu resistance – I
    didn't tell the Times what to say. I'm sure the resistance will be
    global at some point but at the moment it hasn't spread everywhere.

  • http://www.virology.ws profvrr

    I do not call it human flu. It is a human strain; the strain
    designation tells you exactly what it is. There is a nomenclature for
    naming influenza viruses established by scientists, yet everyone else
    chooses to ignore it.

  • ARR_3

    With respect, it was you that made the conclusive claim that there was “no evidence” – I was merely stating the obvious that notwithstanding the small sample size of the data concerned – they are STRONGLY suggestive of a putative causal relationship between the two variables lethality (ie. mortality risk) and infection with an H1N1 virus containing the G225D mutation.

    If we were to compute the Chi-square statistic for the observed and expected frequencies we end up with a value of = 5.64 with 1d.f. (after the appropriate Yate's correction) yielding a p vlaue = 0.018.

    We can now conclude that although based on a limited sample size, the null hypothesis (that there is “no evidence” of lethality arising from the presence of the G225D in the 10 Ukrainians samples under discussion) is unequivocally REJECTED at this point. Notwithstanding the clear and strong putatuve causative relationship (G225D and increased risk of death) obviously more surveillenace is needed.

    With respect, in terms of Tamiflu resitance, you were in your own words in agreement with the Times. So, please forgive my sarcasm (“inner voice”) but the oft repeated logical flaw: asbence of evidence = evidence of absence, well, it's starting to 'get on my goat'.

  • ARR_3

    And that's how science is done.

  • niman

    Speaking of scientific names, the change of interest in HA is D225G (not G225D). It is bad enough with WHO calling it poistion 222, which reversing what is wild type and what is newly acquired. Individual isoaltes have a specific name that includes species for the isolates. The human isolates have no species listed, per convention. However, the identical virus has now been isolated from swine, turkeys, domestic cats, dogs, ferrets, and a cheetah. This list is only limited by teh limited testing.
    Scientists can it a swine virus (the lineage is swl for swine-like), because it has been circulating in swine for decades. Although it now efficiently transmits in humans (and MANY otehr species inclduing swine), is is a swine virus based on the origin of each flu gene (the human PB1 has been in swine since 1993 and the two avian flu genes have been swine since the 1990's.
    Only pig farmers and want to make up a new name for the swine virus.

  • niman

    Speaking of data (in humans) the sequences for the first fatal case in in Norway and a recovered case in Sweden were released today. Both have D225G, but have two tandem polymorphisms (one encoding the change found in most D225G isoaltes and the other in D225E isolates). This combination is rare, but matches two earlier isolates from Spain. All four represent the same sub-clade. Do you think this supports D225G transmission (and the isolate from Sweden was a nasopharygeal collection), or do you think these are “spontaneous” coincidences?

  • niman

    Speaking of science and data, I posted several comments on your “D225G doesn't transmit” story, but haven't seen any responses. Your argument that D225G doesn't transmit is not consistant with examples of D225G in humans.
    Any comments on reality (dead humans with D225G in their lungs)?

  • Pingback: Swine-origin influenza H1N1 as of now | H1N1DEATHS.US()

  • niman

    Speaking of science, Ukraine, and D225G, it is worth noting that there are sequences from 2522 swl isolates at GISAID, and the vast majority include HA. The number of HA sequences with D225G is less than 25, so about 1% of HA sequences have D225G, yet in fatal cases from Ukraine, the frequency is 100% (4 of 4).
    Sounds like another “spontaneous” mutation “coincidence”.

  • Pingback: Swine-origin influenza H1N1 as of now | H3N2FLU.US()

  • gsgs

    these are secret sequences fom GISAID, we need the full list of mutations.
    Evidence for spread would be mutations which occur together with
    225G(H3)s but not with D225s=D239s(H1)

  • gsgs

    you said “evidence” , then “prove” , now “conclusion”.
    Better give a number for your subjective probability
    estimate for -say- that within one year the number of
    225G vg. 225D samples can be linked with significance
    level <5% to fatal outcome.
    The problem with science is, that people don't give such
    numbers for estimates but rather use unclear phrases like “there is no proof”
    “more than expected”,….

  • Pingback: Tweets that mention Swine-origin influenza H1N1 as of now -- Topsy.com()

  • ARR_3

    Following-on from gsgs's comments above, at the very least, an updated piece entitled something along the lines: 'The inappropriateness of definitive conclusions on H1N1 evolution – as of then' might be in order, don't you think?

  • niman

    The pandemic swine flu in humans spreads efficiently in multiple species, including swine. In Alberta, Canada several attempts were made to cure the H1N1 infected herd, but the virus kept reappearing and the entire herd was culled. Swine flu is called swine flu by scientists becasue H1N1 has circulated in swine for decades and triple reassortants have circulated in swine since the 1990's. WHO made up a name to satisfy agriculture interest and had NOTHING to do with science.

  • niman

    The Ukraine and swine flu sequences from 2522 isoaltes can be viewed by anyone resgistered at GISAID. The four that have D225G and teh 6 that do not are not secret. Additional sequences from Ukraine are secret.
    There are polymorphisms in the two Lviv sequences (which both had D225G) that are not in any which are not linked and have D225G.other human H1N1 sequences, but that doesn't prove spreading any more than the 4 of 4 fatal cases
    Not many scientists think that for the four (unrelated) fatal cases spontaeously generatd D225G in western Ukraine within a day of each other (other than those unfamiliar with the actual data, as was written in the “does not spread story – note that the author has not addressed Ukraine other than to suggest that media ignore the actual data from dead patients).

  • gsgs

    inappropriateness comes in degrees …
    and even “conclusion” does (except in math)
    how about : “no conclusion yet”
    oops, I had to correct – I almost mistyped it as confusion ;-)

  • gsgs

    > niman wrote, in response to gsgs:
    > The Ukraine and swine flu sequences from 2522 isoaltes can be viewed by anyone
    > resgistered at GISAID.

    which requires a secrecy agreement, right ?

    > The four that have D225G and teh 6 that do not are not secret.

    so, where can I download them ?

    > Additional sequences from Ukraine are secret.
    > There are polymorphisms in the two Lviv sequences (which both had D225G) that are not in
    > any which are not linked and have D225G.other human H1N1 sequences, but that doesn't
    > prove spreading any more than the 4 of 4 fatal cases

    I don't understand. all which have D225G are “linked” ? define “linked”

    > Not many scientists think that for the four (unrelated) fatal cases spontaeously generatd
    > D225G in western Ukraine within a day of each other

    but that's what they say

    > (other than those unfamiliar with the actual data, as was written in the “does not spread story –
    > note that the author has not addressed Ukraine other than to suggest that media ignore the
    > actual data from dead patients).

    WHO presumably is familiar with that data

    58 comments 0 likes 2 points

  • niman

    I don't think you have a grasp of what is and isn't in the database. The original vaccine target, A/California/7/2009 and collected in April had D225G, and D225G is probably VERY widespread, yet is only in 1% of sequences at GenBank/GISAID, which is becasue most samples are from nasopharyngeal swabs which greatly under-represnt D225G, so your focus on numbers is rather misplaced (and when 4 of 4 fatal cases have D225G, you claim that frequency is due to a random event, which is nonsense).

  • niman

    The Ukraine sequences can be downloaded at GISAID, which is where Mill Hill deposited them. CDC and the WHO regional center in Australia have also deposited sequences with D225G there and anyone who seriously discusses flu sequences has registered.

  • Pingback: uberVU - social comments()

  • PP

    I'm very sorry Pr. Racaniello to interrupt with a (somewhat) poorly related and mundane question regarding h1n1 vaccines, but I just got my shot – as the carer of an ill child – and have a question in that regard to which I haven't been able to find a clear answer.

    I have a fairly basic understanding of virology, immunity and genetics.

    I hear the immunizing effect of the shot is relatively short-lived. Still I'm not sure as to what factors influence its duration, and how they exert influence – I would like to get the grasp of it. I'm guessing it has to do with the potency of the vaccine to induce an immune response. Maybe you would be kind enough as to shed some light? Or maybe point out some reference paper or material I could read?

    May I expect that my shot will have some degree of effectiveness, maybe just at damping the infection, a year from now, when next-year northern winter season starts?

    Thanks a lot for your efforts, I'm a frequent reader and am learning quite a bit.
    Best regards

  • gsgs

    I think it has longterm and shortterm effect. This is a new strain against which we have
    little immunity that's why it infects 20%-40% and not 5%-20% as seasonal flu.
    If everyone had it or was vaccinated, then it should behave like seasonal – even
    without the yearly re-vaccination and adaption to new strains.
    So, the first vax for a new strain is somehow different from seasonal repititions.
    The 1976-vax does still protect against mexflu, maybe half as good as the new
    vax, I guesstimate. But I also read that it probably increases protection when
    “boosted” one or two times later. So seasonal infection boosted the 1976 vax's
    protection against mexflu

  • Pingback: [Crof's H5N1] Racaniello on H1N1 so far | Influenza Virus Mashup()

  • gsgs

    > Your argument that D225G doesn't transmit is not consistant with examples
    > of D225G in humans.

    how does that show transmission ? We had many examples of H275Y developing
    in the individual and not spreading. Show transmission of other obviously acquired
    polymorphisms or family clusters of 225G or such. I haven't seen this.

    > Any comments on reality (dead humans with D225G in their lungs)?

    people die from lots of things without transmission, I don't see the argument.

  • niman

    D225G or D225N are present in about 1% of pandemic H1N1 samples. In Ukraine D225G was in four fatal cases and D225N was in two. That is 6/6 in fatal cases, demonstrating clear transmision.

    The data really don't get any clearer.

  • gsgs

    niman wrote, in response to gsgs:
    > D225G or D225N are present in about 1% of pandemic H1N1 samples.
    > In Ukraine D225G was in four fatal cases and D225N was in two.
    > That is 6/6 in fatal cases,
    > demonstrating clear transmision.

    what does the mere number of fatal cases or 225-mutations therein have to do
    with transmission ?

    > The data really don't get any clearer.

    that, if true, would be a clear argument against transmission.
    If it transmits then we should see clearer evidence earlier or later,
    clusters or transmission of acquired polymorphisms or such.

  • niman

    D225G and D225N are VERY rare. So far there is data on 11 isoaltes from western Ukraine. All 11 are the same subclade. The five cases that recovered are wild type at position 225. The six fatal cases have one of two changes at position 225 (D225G or D225N). The cases have no other non-synonymous changes in HA.

    None of the six cases are linked. Transmission is quite clear. If you don't understand why, you are beyond help.

  • Matt H

    In casual conversation with colleagues I typically refer to this strain as “2009 Pandemic Flu,” which seems to me the easiest way to distinguish it from other strains. As Prof. R. notes, it's not really a swine flu anymore, it's a human flu now. As the CDC's Ruben Donis said to an interviewer, “Can you tell I have an accent? I’m a U.S. citizen but I have the roots in Argentina. It’s like me. I’ve been in the U.S. since 1980. I’m a U.S. citizen but I have an accent.” The full strain designation is rather long for casual conversation, but simply saying “H1N1″ doesn't distinguish it from other H1N1 strains. When it matters which exact isolate is being discussed in a report or something then I will give the full designation (such as A/California/07/2009).

  • Matt H

    In casual conversation with colleagues I typically refer to this strain as “2009 Pandemic Flu,” which seems to me the easiest way to distinguish it from other strains. As Prof. R. notes, it's not really a swine flu anymore, it's a human flu now. As the CDC's Ruben Donis said to an interviewer, “Can you tell I have an accent? I’m a U.S. citizen but I have the roots in Argentina. It’s like me. I’ve been in the U.S. since 1980. I’m a U.S. citizen but I have an accent.” The full strain designation is rather long for casual conversation, but simply saying “H1N1″ doesn't distinguish it from other H1N1 strains. When it matters which exact isolate is being discussed in a report or something then I will give the full designation (such as A/California/07/2009).