Flu News: D225G Follow-up

In case you missed my latest update on the flu, I’ve added to it here. – Ilene

Flu News: D225G Follow-up

By Ilene with guest expert Dr. Henry Niman

Dr. Henry Niman heads the research company Recombinomics Inc. Recombinomics has a small group of researchers who analyze the sequence data from viral samples isolated from patients diagnosed with swine flu.

I spoke with Dr. Niman last week about the evolution of the H1N1 virus as it circulates through the world’s population. These changes are natural—and Dr. Niman’s research on the subject allowed him to predict how the virus would change as infection rates increase and time goes on. The outbreak in Ukraine was initially described in many media reports as a new lung-blackening "mystery disease," leading to many false and misleading Internet stories. According to Dr. Niman, it was clear from the start that H1N1 was killing an unusually high number of previously healthy young adults. 

H1N1 infection seemed to cause more severe illness in Ukraine, and Ukraine officials asked the WHO for assistance. The WHO interfaced with Ukraine labs in Kiev and sent groups into western Ukraine to survey the problem and gather information. As part of the investigation, the WHO sent samples to Mill Hill in London, one of several regional centers that performs genetic analysis for the WHO.

Subsequently, the WHO issued several non-informative reports but held the sequences at Mill Hill for analysis. Dr. Niman wrote a number of commentaries on the rising death toll and the need to make the sequences public. He predicted the deaths would be associated with a receptor binding domain change in the wild-type H1N1 virus (the predominant virus) to a variant form, characterized by the D225G genetic marker. Wild-type H1N1 has a D at position 225 of the viral protein Hemagglutinin (HA), and is referred to as D225. The variant protein, D225G, has a genetic change causing a change in position 225 of the protein. "D" is the symbol for the aspartic acid which is present at position 225 of the wild-type protein. Glycine, symbol G, replaces aspartic acid in D225G variants. Hemagglutinin is one of two surface proteins projecting out from the surface of the virus. The function of the HA protein is to bind viral particles to susceptible cells in the host animal.

The receptor binding domain change D225G causes the virus to bind to receptors in the lungs rather than in the upper airways where D225 would bind. Dr. Niman’s prediction was based on his analysis of the public database for H1N1 sequences, descriptions of fatal cases, and what the WHO did and did not say in their reports.

Before releasing the sequence information, the WHO issued an update claiming that the data showed no significant changes and suggested that the swine flu vaccine would remain effective in cases with the D225G variant. They also indicated that a second regional center, the CDC in Atlanta, had been brought in.

The sequences were released on November 18 at GISAID, a public database that requires registration and is password protected. Reporters are not registered for access, and even if they were, most are not able to read the sequences. Because Dr. Niman has access to and expertise in analyzing genetic sequences, he became an important, if somewhat controversial, source of information.

Dr. Niman found that the released sequences confirmed that the predicted genetic change, D225G, was in four of ten samples. The sequence data was followed by demographic data which indicated that the four patients with D225G were the four patients who had died, suggesting that this change was indeed significant. The WHO hadn’t reached this conclusion itself, and Dr. Niman feels the WHO’s failure to recognize the importance of the D225G change is irresponsible.

Further, Dr. Niman observed that D225G has been identified in lung tissue of patients who died from the effects of cytokine storms. A cytokine storm, or hypercytokinemia, "is a potentially fatal immune reaction consisting of a positive feedback loop between cytokines and immune cells, with highly elevated levels of various cytokines. The primary symptoms of a cytokine storm are high fever, swelling and redness, extreme fatigue and nausea. In some cases the immune reaction may be fatal." Wikipedia. "Cytokines are any of a number of substances that are secreted by specific cells of the immune system which carry signals locally between cells, and thus have an effect on other cells." Wikipedia.

It would be interesting to learn whether the D225G variant was also present in lungs of patients who died without signs of cytokine storms, and those who did not die at all, though lung tissue is not typically sampled from recovered patients. Statistical information regarding the increased ability of D225G to provoke cytokine storms is, in some cases (e.g. live patients) challenging to gather.

Recently, antigen characterizations have been released. A throat sample from of one of the four fatal cases was tested and the isolated virus was designated a "low reactor" by Mill Hill and/or the CDC. The antibody titer (amount of antibody against the virus) stimulated by the test sample was at least four times lower than expected. “Low reactors give a reduced titer of AT LEAST 4 fold, which would lower activity below the traditional cut-off of 40, which is a very low bar. Not all patients with a titer of 40 will avoid infections.” 

The low reactor designation contradicts the WHO’s statement on vaccine activity, because it suggests that the available flu vaccines are considerably less effective against the D225G variant than the wild-type D225 viruses.

According to Dr. Niman, the "low reactor" designation could seriously impact those who developed a borderline response to the vaccine. Nevertheless, Dr. Niman still believes it is a good idea to be vaccinated: “I think the vaccine still protects against more isolates that do not have changes at position 225 and should be taken.”

Dr. Niman expects viral samples from the other three fatal Ukraine cases would be low reactors, if tested. Conversely, he expects samples from patients without the D225G marker to stimulate the normal antibody response. His predictions, which have so far proved prescient, should be confirmed by further research and appropriate testing.

In summary, evidence presented by Dr. Niman supports his conclusion that the D225G marker is associated with greater invasive activity in the lungs, and the occurrence of cytokine storm reactions.  Further studies are warranted to draw more definitive conclusions.  For instance,

1. The D225G genetic marker is poorly recognized by the immune system according to data collected from one sample—is this result confirmed in other cases?
2. Is the D225G variant statistically more likely to be recovered from fatal cases?
3. Is the D225G variant statistically more likely to be present in cases showing cytokine storm activity in the lungs?

How well the D225G variant transmits is another unanswered question. According to Dr. Niman, D225G is likely transmitting as a mixture with the wild type virus, so replicating natural conditions would be difficult, but data on transmission via multiple ratios would be interesting. An alternative view suggests that the transmission of the D225G variant may be less efficient compared to the wild-type virus.  

An important consideration is that the D225G amino acid change has a negative impact on transmission. The change from D to G at amino acid 225 of the 1918 HA significantly impairs transmission among ferrets. When both D225G and D190E are present, transmission is abolished. These changes do not impair viral replication or virulence in the respiratory tract of inoculated animals. The D225G change in 2009 H1N1 influenza virus is not a concern.

Even so, the variant D225G is being seen in numerous locations worldwide, and appears to be spreading in a mix with the wild-type virus. Dr. Niman’s theory that this is due to recombination explains this phenomenon and suggests concern is warranted. See Swine Flu News: What is the significance of D225G?.

Today, November 28, the WHO acknowledged that the D225G change in one patient would have likely resulted in failure of the flu vaccine. Dr. Niman writes about this development in WHO Confirms D225G Vaccine Failure.

One isolate from Ukraine with the mutation had changed so that swine flu vaccine probably would not protect against it well, Britain’s national medical laboratory reported Friday.

Flues mutate so fast, Dr. Fukuda cautioned, that announcing each change is "like reporting changes in the weather."

The above quote from tomorrow’s NY Times piece by Donald McNeil, acknowledges the vaccine failure for viruses with D225G.  However, although WHO has publicly confirmed the failure, they don’t think an announcement is required.  Thus, they continue to offer altering opinions on the significance of D225G, which directs H1N1 to the lung and was present in four of four fatalities in Ukraine.

The associate of D225G with the Ukraine fatalities led to a survey of samples in Norway, where D225G was found in three patients (two who died and 1 who was in serious condition).  Similarly, France found D225G in two fatal infections, including one who was Tamiflu resistant.

However, even though this change is drawing additional attention daily, WHO has taken a position that the vaccine failure against H1N1 with this D225G is not worthy of an announcement.

This mindset is significant cause for concern and is hazardous to the world’s health. 

On a final note, the death rate in children continues to rise and this trend will continue, at least until the counting starts over. Dr. Niman writes: "35 fatal pediatric deaths in week 46 in the US will be published on Monday. The 35 ties that weekly record set two weeks ago, and the new entries raise the 2009/2010 season total to 178, which is a new record.  The 138 last week represented the highest tally since [reporting] adolescent fatalities became mandatory, when 142 died in 2003/2004.  178 is a new all time high, but it will be broken each week until there is a break, which may not happen in 2010." Week 46 Sets Record H1N1 US Pediatric Fatalities

Background Reading

For more detailed information, please visit at Dr. Niman at Recombinomics. He has posted a number of articles within the last few days–here are some excerpts with my comments. 

Worldwide Transmission of D225G
Recombinomics Commentary
November 27, 2009

The recently released sequences from patients in Ukraine provided valuable insight into the pathogenicity of H1N1 and the genetic change associated with the total destruction of both lungs in fatal cases. The description of the patients and the spread of receptor binding domain change, D225G, to multiple genetic backgrounds via recombination led to the prediction that D225G would be found in the lung samples from fatal cases.  The release of the sequences by Mill Hill confirmed the prediction.  Sequences from 10 isolates were released and all four fatal cases had D225G.  Moreover, all 9 cases from western Ukraine, which were from three Oblasts (Ternipol, Lviv, and Khmelnitsy) were from the same sub-clade as the fatal cases, but the samples from the upper respiratory tract did not have D225G. The absence of D225G from the upper respiratory tract is not a surprise because the specificity of D225G included…receptors which are present in the lungs. Thus, the sub-clade with D225G can expand and cause a cytokine storm which destroys the lungs. Moreover, sequences with D225G have been designated as low reactors by Mill Hill, raising concern that immune responses and vaccine will select for D225G.

My comment: "a clade is a group of organisms including a common ancestor and all the next generations of that ancestor."  Simple English Wikipedia.  "A sub-clade is a group of sequences within a clade that are closely related to each other." Dr Niman.

Although D225G transmits from patient to patient, only the samples from the fatal cases, which were from lung and throat samples were positive for D225G. The sequences from Ukraine led other countries to more fully investigate samples. Norway, which had seen an increase in fatalities announced the detection of D225G in two fatal and one severe case.

The finding of D225G in four of four fatal cases in Ukraine leaves little doubt that the polymorphism is transmitting and the recent classification of Ukraine sequences carrying D225G suggests the spread will accelerate….

More surveillance of low respiratory tract infections would be useful.

WHO Silence on D225G Immune Escape Raises Concerns
Recombinomics Commentary
November 27, 2009

The [D225G] change affects receptor binding specificity and allows the virus to bind [targets] on the lung, and also affects the antigenic site. 

The failure of the WHO or CDC to comment on the low reactor status of the Ukraine sequences from fatal patients is also cause for concern…

My comment: it appears that the D225G variant is not as susceptible to the immune response stimulated by available vaccines. The WHO previously said the D225G change was not significant.  At least in one test, the D225G sample proved to be a "low reactor." Because D225G appears to be circulating as a mixture with the wild-type virus, immune responses that fail to target D225G might shift the ratio in favor of D225G, leading to more severe cases.  More studies should be done to confirm this finding. 

D225G and H274Y in Fatal Infection in France
Recombinomics Commentary
November 27, 2009

[There were] two fatal cases in France with D225G.  Moreover, one of the two cases also had Tamiflu resistance, presumably H274Y. 

…The same strain in two patients who had no link to swine or each other signaled efficient transmission.  The same is true for D225G.  It is in multiple patients in multiple countries and appearing at increasing frequencies at the same time.

WHO Mis-statements in D225G and H274Y Raise Concerns
Recombinomics Commentary
November 27, 2009

Preliminary tests reveal no significant changes in the pandemic (H1N1) 2009 virus based on investigations of samples taken from patients in Ukraine. Analyses are being performed by two WHO influenza collaborating centres as part of the global influenza surveillance network.

Preliminary genetic sequencing shows that the virus is similar to the virus used for production of the pandemic influenza vaccine, reconfirming the vaccine’s efficacy at this time.

The above WHO comments in their latest Ukraine outbreak update are unfortunate.  The comments were made after sequences had been generated which showed that four of four fatal cases in Ukraine had the receptor binding domain change D225G. This change had been predicted because it is the type of change expected for the large number of deaths which were linked to the rapid destruction of both lungs.  D225G had been identified in 1918 and 1919 lung samples from fatal infections and analysis of the change identified a change in receptor binding specificity…  The failure of WHO to consider such a change significant raises serious concerns about the agency’s credibility and scientific underpinnings.

Moreover position 225 is in one of the known antigenic sites, so to declare a confirmation of vaccine efficiency was false…

The failure of WHO to correctly report on the receptor binding domain after the sequences had been generated destroys confidence in the agency at a most crucial time.  In addition to targeting of the lungs and a reduction in vaccine efficiency,Tamiflu resistance is on the rise and one of the fatal D225G cases in France also has H274Y, raising concerns at a time when WHO is posting situation updates which are not credible.

They then compound this lack or credibility by claiming that the D225G, which is four unrelated cases in Western Ukraine are "spontaneous" which is also the characterization of oseltamivir resistance, H274Y.  These claims have no scientific basis and are simply absurd…

WHO Silence on D225G Vaccine Mismatch Confuses Media
Recombinomics Commentary
November 28, 2009

The vaccine can still prevent the virus from entering body cells to reproduce. These new changes should instead encourage all French to go to be vaccinated with adjuvant.

The above translation is from a French story on the discovery of D225G in two fatal French cases, one of which was also Tamiflu resistant.  Because of the failure of WHO to disclose the "low reactor" status of Ukraine H1N1 with D225G, media reports are distributing the false information, stating that the activity of the vaccine is unaffected by D225G.

This misinformation is fueled by the WHO update on Ukraine, which stated that the vaccine appeared to be unaffected based on the sequence.  For the "low reactor" the HA only had one amino acid difference, which was D225G.

As has been noted daily in new media reports, D225G is widespread,…

The designation of a "low reactor" means that the titer of a reference anti-sera is reduced by at least 4 four.  A four fold reduction in titer is typically called a mis-match and mismatched vaccines are a concern because a vaccination will not eliminate the new changes that reduced the titer, but will eliminate the wild type that competes with the variant.  Therefore the use of a poorly match vaccine leads to increased vaccine resistance and in this case would select for D225G.

The failure of WHO to address this issue is the height of irresponsibility.  Although the NY Times noted the Ukraine low reactivity due to D225G, it is not clear that the WHO comments were in response to the vaccine failure, since similar statements were made at the WHO virtual press conference and the reactivity of the vaccine was not addressed…

Therefore a statement by WHO is long overdue to end that false information distributed through media reports…

WHO Weather Report on Ukraine H1N1 D225G Required
Recombinomics Commentary
November 28, 2009

The question is whether this mutation again suggests that there is a fundamental change going on in viruses out there or whether there is a turn for the worse in terms of the severity.  I think that the answer right now is that we are not sure.  I want to answer why we are not sure in a way which explains why more investigations are needed. As you know these influenza viruses change frequently.  Their gene properties change because these are viruses which frequently undergo mutations and so mutations in and of themselves are not necessarily important and in fact, if every mutation was reported out there, it would be like reporting changes in the weather –

The above comments from Keiji Fukuda at Thursday’s virtual press conference are in context, which highlights the need for additional data on D225G.  However there is a large body of evidence on this change and the Ukraine data is the most current and quite compelling.  Ten samples were collected from ten patients and nine represented the same sub-clade. However, only four were from tissues from fatal cases, but all four had D225G. In contrast, nasopharyngeal washes were collected from the five surviving patients in western Ukraine and all sequences were the same sub-clade, but did not have D225G. The finding of D225G in 100% of the fatal cases raises concerns, but the number of such cases is small…

An aggressive campaign on this change is warranted because the only reported tested sample for antigenicity was found to be a low reactor, and the only amino acid change in the HA sequence was D225G.  Since the same sub-clade was found in earlier collections in Norway, and the cases positive forD225G were fatal or severe, more interest in the change has been generated, but media reports state that the vaccine is effective against D225G, when the data presented on the Mill Hill characterization sheet at GISAID cites the antigenicity as being "low reactor", signaling a need for a new vaccine… 

Finding a receptor binding domain change which alters specificity in 100% of fatal cases is cause for concern, and release of additional sequence data is overdue, since some still need a weatherman to know which way the wind is blowing.

H1N1 RBD Changes at 225 Create Vaccine Mismatch Concerns
Recombinomics Commentary 14:22
November 29, 2009

Antigenic characterisation:

A/California/7/2009 like. Low reactor

The above characterization of A/Lviv/N6/2009, which was placed on deposit at GISAID by Mill Hill, raises concerns about the evasion of pandemic H1N1 sequences which change position 225. The above isolate has only one amino acid change in HA, D225G, which strongly implicates D225G in the low reactor results. A low reactor reduces the titer by four fold or more, which signals a mismatch. Mismatched vaccine create the potential for the section of the variant, which could create problems since D225G was found in four of four fatal cases in Ukraine, and several countries (Brazil, Ukraine, Norway, France, China) found D225G in fatal and or severe cases.

Moreover, since D225G changes receptor specificity, it may transmit undetected because of an emphasis on nasal pharyngeal samples where levels would be expected to be lower.  This was seen in five additional nasopharyngeal washes from Ukraine survivors, who were infected with a sub-clade that matched the fatal cases, but lacked the D225G.

My comment: A vaccine that is effective only against the wild type D225 HA protein would select for the D225G variant, so the ratio of D225G to the wild-type virus would increase. 

The designation of one of the Ukraine D225G isolates as a "low reactor" raises concerns that the H1N1 evolution is outpacing the vaccine as well as immune responses from unvaccinated hosts.  This concern was present ealry when changes began to appear at position 225, a known antigenic site.

My comment:  A "low reactor" means that a test subject’s immune system does not generate enough antibody to the changed virus to be normally protective.  It indicates a mis-match between the virus and the vaccine used to protect against the virus. A mis-matched vaccine will not be effective, allowing new changes in the virus to avoid detection by the immune system. 

The above data reinforced concerns that the changes at position 225 were becoming more prevalent, especially in samples collected from affected organs of fatal cases.

The concerns ratcheted up another notch when one of the four cases was characterized antigenically, and was reported to be a "low reactor"…

However, WHO has yet to address this result.  The "low reactor" status was quietly added to the characterization sheet at GISAID and there has been little comment in the mainstream media, other than a sentence in the New York Times on Friday.

Comment form WHO and other agencies, including Mill Hill and the CDC, who were conducting testing of the Ukrainian samples, is long overdue.