Sequences from Tamiflu Resistant Immunocompromised Patients Released

[AlaskaDenise]

Re: Sequences from Tamiflu Resistant Immunocompromised Patients Released

Discussions about tamiflu resistance and immunocompromised patients can get very difficult for the average FT reader to understand. We'd all appreciate it if those posters with scientific backgrounds would do more "translation" into everyday terms.

What's being presented to the public about this topic is often mistranslated by media, so your understandable explanations are badly needed.

After 4 years of trying, I understand a little, but still get quite lost when the jargon gets too deep. I get totally lost with gsgs's statistical data.

While we can appreciate your desire to have scientific discussion, please remember what role that FT has played in your scientific pursuits, specifically, in post #54: "The story began when I was collaborating with NAMRU-3 on H5N1 outbreaks in Egypt." We'd be grateful if you could express your appreciation by phrasing your scientific discussions in terms we could all understand and other appreciation might also be in order.

.

 

[HenryN]

Re: Sequences from Tamiflu Resistant Immunocompromised Patients Released

After 4 years of trying, I understand a little, but still get quite lost when the jargon gets too deep. I get totally lost with gsgs's statistical data.

I think only one person understands gsgs's statistics. The numbers he posted on this thread really had no application. He took all isolates and all changes, added them up and posted as evidence that the non-random changes were random. For this thread, we were talking about the NA changes, which are relatively rare (generally 0-3 changes in the sequence of over 1400 positions). I had posted a travel log with 22 isolates with the same change (A1230G). These would be created by one change in one isolate, that led to multiple isolates with that change because of expansion (the same virus was isolated in multiple locations). He was counting those 22 examples as 22 changes (independent events). He also added up all changes in all genes (but changes in the other 7 gene segments have nothing to do with changes in NA) and then tried to use this composte number to say that matches were coincidences because there were so many changes in the database.
In the past he has looked at all possible changes at all positions to say that associations that were not random were random.

It is like someone saying that getting heads 15 times in a row was rare and GSGS saying that if you flip a coin 100 million times, you will get 15 heads in a row 10 times and arguing that 15 heads in row is common.

 

[HenryN]

Re: Sequences from Tamiflu Resistant Immunocompromised Patients Released

<a rel="nofollow" href="http://www.recombinomics.com/News/08240901/H274Y_WA_Match.html">Commentary</a>

 

[mixin]

Re: Sequences from Tamiflu Resistant Immunocompromised Patients Released

I think only one person understands gsgs's statistics.

That may be close to the truth; but that's not his fault that most of us aren't as mathematically gifted as he is. Sometimes it's hard for really smart people to "dummy down" for the rest of us.

It is like someone saying that getting heads 15 times in a row was rare and GSGS saying that if you flip a coin 100 million times, you will get 15 heads in a row 10 times and arguing that 15 heads in row is common.

This is not true. He would run all the stats and present what he found.

 

[Hogweed]

Re: Sequences from Tamiflu Resistant Immunocompromised Patients Released

That may be close to the truth; but that's not his fault that most of us aren't as mathematically gifted as he is. .

From reading gsgs posts I don't get the impression they are especially mathematically gifted and I doubt they would claim that. Certainly good at crunching numbers but that's something else.

 

[mixin]

Re: Sequences from Tamiflu Resistant Immunocompromised Patients Released

Thanks for the clarification. I guess I didn't know the difference.

 

[HenryN]

Re: Sequences from Tamiflu Resistant Immunocompromised Patients Released

That may be close to the truth; but that's not his fault that most of us aren't as mathematically gifted as he is. Sometimes it's hard for really smart people to "dummy down" for the rest of us.



This is not true. He would run all the stats and present what he found.

I think you have missed the point. On the sequences the number of changes is VERY small so the number of different combinations is not based on some random number involving 100's of millions of tries. Similarly,c oming up with a number and inappropraitely applying it may reflect a "gift" but the irrelevant number doesn't address the issue at hand.

 

[Hogweed]

Re: Sequences from Tamiflu Resistant Immunocompromised Patients Released

Thanks for the clarification. I guess I didn't know the difference.

I remember a maths professor who started a course with the words "This is pure mathematics. There is nothing in this course you will ever likely use in real life - except maybe for some of the polynomial stuff which unfortunately they've found a common use for in computing"

An exaggeration of course but still with more than a hint of truth.

 

[Hogweed]

Re: Sequences from Tamiflu Resistant Immunocompromised Patients Released

A few years ago the big banks started employing brilliant "applied mathematicians" to invent new financial instruments. They produced amazing imaginary money systems and "proved" with statistical models that they could only break down catastrophically once every few billion years - ie they were safe. So they then tied this imaginary money into the real financial world. Unfortunately many of their initial assumptions were wrong (they had missed "The Black Swan") as they had never been exposed to the "big picture" and the world's financial systems were almost destroyed and might still be as their creations sucked the real system dry.

Number crunching can be misleading if not understood within a wider context is all I am saying.

 

[mixin]

Re: Sequences from Tamiflu Resistant Immunocompromised Patients Released

Hogweed,

Yes, I guess everything relates back to if someone has an agenda to promote (or not).

 

[mixin]

Re: Sequences from Tamiflu Resistant Immunocompromised Patients Released

Here's a summary I did on the symptoms, testing, treatment, etc for the 2 cases. There is a big difference between them; maybe this is reflected by the differences in the sequences.

1. 18 year old male A/Washington/28/2009/07/14,m18
On April 29, 2009, he was hospitalized for a hematopoietic stem cell transplant, which he received on May 7. He received immunosuppressive treatment prior to his transplantation and remained hospitalized in a single-patient room after the transplantation.
May 31: he developed symptoms and tested positive
June 1: started a 10 day treatment of Tami
June 4: virus was again detected in a viral culture
June 4: Tami was extended for 20 days ~ June 20
June 7: patient was discharged
June 11: virus again detected in a viral culture
July 7: nasal wash specimen tested positive
July 8: he was started Tami again
July 14: became ill with a fever & staph infection
July 14 tested positive for panflu, Tami was increased to a high dose, 150 mg orally, twice a day This is the collection date of the released sequence
July 16: he had increased rhinorrhea and mild cough
July 18: he was discharged still taking oseltamivir
Aug 6: CDC testing on a June 4 specimen showed susceptibility but July 30 specimen showed resistance.
Aug 6: Tami was stopped, no further treatment, patient was assmptomatic
Aug 10/11: testing on a May 31 specimen showed susceptibility. June 11 & July 14 specimens showed resistance.

2. 47 yo female A/Washington/29/2009/07/28,f47
She had a previous hematopoietic stem cell transplant for leukemia and had a recurrence of leukemia in December 2008. She underwent two cycles of immunosuppressive chemotherapy during March--April 2009. On June 21, she was admitted to the hospital for further chemotherapy; she also had developed a fever and symptoms of an upper respiratory infection. The patient's hospital course was complicated by prolonged neutropenia and protracted bone marrow recovery, neutropenic fever, coagulase-negative Staphylococcus bacteremia, and Pneumocystis jirovecii pneumonia.
June 21: she developed symptoms.
June 21: test result was indeterminate because of an inadequate cellular specimen
June 26: tested positive for influenza A
June 26-July 1: Antiviral treatment with high-dose oseltamivir (150 mg orally, twice a day) and rimantadine (100 mg orally, twice a day)
July 3: the viral isolate was identified as novel influenza A (H1N1), and high-dose oseltamivir and rimantadine were restarted.
July 3: The patient's respiratory status worsened, and she required supplemental oxygen for hypoxia
August 4: treatment with inhaled zanamivir was attempted, but was poorly tolerated, and oseltamivir was continued.
Aug 4: testing showed positives for novel flu on samples from July 6, 14,16 & July 28 This is the collection date of the released sequence
August 6: the June 21 sequence did not have the H275Y mutation but was detected in the July 28 specimen.
Aug 6: Treatment of the patient with oseltamivir was discontinued when results became available and treatment with inhaled zanamivir after identification of oseltamivir resistance again was attempted but poorly tolerated.
Aug 7: intravenous zanamivir and aerosolized ribavirin therapy were initiated.
August 10: CDC received other previously collected virus isolates from this patient for testing, and pyrosequencing of a virus isolated from a specimen collected on July 14 had the H275Y mutation,
August 13: the patient remained symptomatic and hospitalized on intravenous zanamivir and had been switched to oral ribavirin because of intolerance of aerosolized ribavirin.

 

[Flufreak]

Re: Sequences from Tamiflu Resistant Immunocompromised Patients Released

http://www.recombinomics.com/News/08240901/H274Y_WA_Match.html

"Matching Sequences fromTamiflu Resistance Patients in Seattle
Recombinomics Commentary 11:11
August 24, 2009

Of the four oseltamivir resistant 2009 influenza A (H1N1) viruses, two were original clinical samples detected by the neuraminidase inhibition assay, and two were viral isolates detected by sequence analysis of a neuraminidase gene.

The above comments from the latest CDC update describe the four reported examples of pandemic H1N1 oseltamivir resistance in the United States. The four isolates are from two clusters. One cluster involves two immuno-suppressed patients in Seattle who were not epidemiologically linked, as well as two summer campers in North Carolina who were taking prophylactic Tamiflu because of a pandemic H1N1 outbreak in the camp. These two sets of clusters suggest the resistance is not due to independent spontaneous events.

The CDC has released the HA and NA sequences from the two immune-suppressed patients. Both NA sequences have a defining polymorphism (A1230G), in addition to H274Y (encoded by C823T), which also supports the presence of a minor species circulating with this sub-clade, and raises concerns that the resistance associated with this sub-clone is widespread.

The public sequences are from the 18M patient with a recurrence (A/Washington/28/2009) and 47F (A/Washington/29/2009), who remains hospitalized after treatment with Tamiflu, Relenza, and Ribaviron. Although initial samples collected from both patients were Tamiflu sensitive, the isolates collected significantly after the start of Tamiflu treatment had H274Y. Although both had clean signals for C823T (which encodes for H274Y), both sequences had mixed signals at other positions. Both sequences also had addition changes that were not in common, indicating those changes were acquired after the acquisition of A1230G. However, the linkage of A1230G and C823T may have been established in a minor population at an earlier date, raising concerns that H274Y is widespread. A1230G is present in multiple isolates from Washington state (see list here), but is also present in multiple isolates from Mexico and Nova Scotia. as well as sequences from Oregon, Florida, China (Jiangsu and Guangdong), New Zealand, and Singapore.

The concept silent circulation of minor populations with H274Y is also supported by the two cases from the summer camp in North Carolina. Although sequences from the two campers haven't been released, the clustering of two cases in campers at the same location at the same time further supports a minor species. Two independent events are unlikely since both patients were attendees of the same camp. Moreover, the two cases in July were followed by rapidly spreading outbreak in schools in August in adjacent regions (see map), raising concerns that H274Y is also widespread in the area and contributing to the rapid spread among students in the area.

Sequence data from the campers, as well as students in the area would be useful."

This may be off topic, and already answered, but does H274Y increase transmission among those NOT treated/ prevented with Tamiflu?

 

[HenryN]

Re: Sequences from Tamiflu Resistant Immunocompromised Patients Released

Here's a summary I did on the symptoms, testing, treatment, etc for the 2 cases. There is a big difference between them; maybe this is reflected by the differences in the sequences.

1. 18 year old male A/Washington/28/2009/07/14,m18
On April 29, 2009, he was hospitalized for a hematopoietic stem cell transplant, which he received on May 7. He received immunosuppressive treatment prior to his transplantation and remained hospitalized in a single-patient room after the transplantation.
May 31: he developed symptoms and tested positive
June 1: started a 10 day treatment of Tami
June 4: virus was again detected in a viral culture
June 4: Tami was extended for 20 days ~ June 20
June 7: patient was discharged
June 11: virus again detected in a viral culture
July 7: nasal wash specimen tested positive
July 8: he was started Tami again
July 14: became ill with a fever & staph infection
July 14 tested positive for panflu, Tami was increased to a high dose, 150 mg orally, twice a day This is the collection date of the released sequence
July 16: he had increased rhinorrhea and mild cough
July 18: he was discharged still taking oseltamivir
Aug 6: CDC testing on a June 4 specimen showed susceptibility but July 30 specimen showed resistance.
Aug 6: Tami was stopped, no further treatment, patient was assmptomatic
Aug 10/11: testing on a May 31 specimen showed susceptibility. June 11 & July 14 specimens showed resistance.

2. 47 yo female A/Washington/29/2009/07/28,f47
She had a previous hematopoietic stem cell transplant for leukemia and had a recurrence of leukemia in December 2008. She underwent two cycles of immunosuppressive chemotherapy during March--April 2009. On June 21, she was admitted to the hospital for further chemotherapy; she also had developed a fever and symptoms of an upper respiratory infection. The patient's hospital course was complicated by prolonged neutropenia and protracted bone marrow recovery, neutropenic fever, coagulase-negative Staphylococcus bacteremia, and Pneumocystis jirovecii pneumonia.
June 21: she developed symptoms.
June 21: test result was indeterminate because of an inadequate cellular specimen
June 26: tested positive for influenza A
June 26-July 1: Antiviral treatment with high-dose oseltamivir (150 mg orally, twice a day) and rimantadine (100 mg orally, twice a day)
July 3: the viral isolate was identified as novel influenza A (H1N1), and high-dose oseltamivir and rimantadine were restarted.
July 3: The patient's respiratory status worsened, and she required supplemental oxygen for hypoxia
August 4: treatment with inhaled zanamivir was attempted, but was poorly tolerated, and oseltamivir was continued.
Aug 4: testing showed positives for novel flu on samples from July 6, 14,16 & July 28 This is the collection date of the released sequence
August 6: the June 21 sequence did not have the H275Y mutation but was detected in the July 28 specimen.
Aug 6: Treatment of the patient with oseltamivir was discontinued when results became available and treatment with inhaled zanamivir after identification of oseltamivir resistance again was attempted but poorly tolerated.
Aug 7: intravenous zanamivir and aerosolized ribavirin therapy were initiated.
August 10: CDC received other previously collected virus isolates from this patient for testing, and pyrosequencing of a virus isolated from a specimen collected on July 14 had the H275Y mutation,
August 13: the patient remained symptomatic and hospitalized on intravenous zanamivir and had been switched to oral ribavirin because of intolerance of aerosolized ribavirin.

So was H274Y detectable in the July 3 and July 6 collections from 47F?

 

[mixin]

Re: Sequences from Tamiflu Resistant Immunocompromised Patients Released

So was H274Y detectable in the July 3 and July 6 collections from 47F?

You asked this before. I don't have any inside info so I'll venture a guess that the answer is "no".

Since they were testing backwards and the July 28 sequence was the one announced to be resistant and the one to be released, a person might assume the others tested sensitive.

Unless, of course, there is some reason for the CDC to keep the earlier results secret (if those also showed resistance)?

 

[HenryN]

Re: Sequences from Tamiflu Resistant Immunocompromised Patients Released

You asked this before. I don't have any inside info so I'll venture a guess that the answer is "no".

Since they were testing backwards and the July 28 sequence was the one announced to be resistant and the one to be released, a person might assume the others tested sensitive.

Unless, of course, there is some reason for the CDC to keep the earlier results secret (if those also showed resistance)?

July 14 had H274Y.

 

[mixin]

Re: Sequences from Tamiflu Resistant Immunocompromised Patients Released

Yes, I see that. I'll retract that part of my reply.

So maybe you would like to answer your own question?

 

[AlaskaDenise]

Re: Sequences from Tamiflu Resistant Immunocompromised Patients Released

.......... He took all isolates and all changes, added them up and posted as evidence that the non-random changes were random. For this thread, we were talking about the NA changes, which are relatively rare (generally 0-3 changes in the sequence of over 1400 positions). I had posted a travel log with 22 isolates with the same change (A1230G). These would be created by one change in one isolate, that led to multiple isolates with that change because of expansion (the same virus was isolated in multiple locations). He was counting those 22 examples as 22 changes (independent events). He also added up all changes in all genes (but changes in the other 7 gene segments have nothing to do with changes in NA) and then tried to use this composte number to say that matches were coincidences because there were so many changes in the database.
In the past he has looked at all possible changes at all positions to say that associations that were not random were random..........

Thanks for the explanation.

While statistical information can sometimes be useful, from what I understand about influenza, there are reasons for changes. So I'm less inclined to think all these changes are random.

In looking at the 356 NA sequences from the US (the 1st 470 positions that show on the NCBI database), they are remarkably the same with the most common differences being I106V and D248N (not sure of numbering if there are offsets), and even those mostly disappear after the 12 May sequences. But then there's that pesky WA/28 & WA/29. The WA/29 has double the differences as compared to the few others with differences. So is that the different leg you reference - with the 274Y change? Do you suspect that leg will maintain most of those 6 differences?

.

 

[HenryN]

Re: Sequences from Tamiflu Resistant Immunocompromised Patients Released

Thanks for the explanation.

While statistical information can sometimes be useful, from what I understand about influenza, there are reasons for changes. So I'm less inclined to think all these changes are random.

In looking at the 356 NA sequences from the US (the 1st 470 positions that show on the NCBI database), they are remarkably the same with the most common differences being I106V and D248N (not sure of numbering if there are offsets), and even those mostly disappear after the 12 May sequences. But then there's that pesky WA/28 & WA/29. The WA/29 has double the differences as compared to the few others with differences. So is that the different leg you reference - with the 274Y change? Do you suspect that leg will maintain most of those 6 differences?

.

There are free software programs to develop phylogenetic trees, but they are built off shared polymorphisms. I did a quick tree that included the sequences with A1230G. They all form a branch, but the two isolates from mexcio form a separate branch because they share polymorphisms with sequences from california and canada.
The other sequences form the main branch with identical sequences from Nova Scotia, Washington State, Guangzhoo, and Christchurch. WA28 comes off that branch, and WA29 comes off WA 29. WA 29 has more polymorphisms, but it was collected later. I think it is noteworthy that the CDC made public the two acknowledged sequences that were separated the most in time. I suspect the greater the differences between the two patients, the happier the CDC is, but they will have a problem with NC, because they will be virtually identical (so I suspect the CDC will hold those sequences as long as possible (but the cat really is out of the bag). 4 sequences in 2 clusters is a MAJOR problem for "spontaneous mutations", and widespread H274Y is cause for concern (and supported by the DATA).

 

[mixin]

Re: Sequences from Tamiflu Resistant Immunocompromised Patients Released

Unless, of course, there is some reason for the CDC to keep the earlier results secret (if those also showed resistance)?

Comparing the date of the released sequence and the test dates, the obvious escaped me Maybe they found resistance in the earlier specimens and are withholding the sequences because someone is writing a paper or any of the other reasons they withhold sequences?

---------------------------------------------------

When I did the summary, I realized just how much these poor people have been through with their original health problems and now the flu. The hospital confinements, home confinements, all the precautionary measures, the side effects from the meds, waiting for the many test results...

If their families ever find their way to this thread, I want them to know their loved ones aren't just some case numbers being discussed on a flu board.

 

[wotan]

Re: Sequences from Tamiflu Resistant Immunocompromised Patients Released

There are free software programs to develop phylogenetic trees, but they are built off shared polymorphisms. I did a quick tree that included the sequences with A1230G. They all form a branch, but the two isolates from mexcio form a separate branch because they share polymorphisms with sequence from california and canada.
The other sequences form the main branch with identical sequences from Nova Scotia, Washington State, Guangzhoo, and Christchurch. WA28 comes off that branch, and WA29 comes off WA 29. WA 29 has more polymorphisms, but it was collected later. I think it is noteworthy that the CDC made public the two acknowledged sequences that were separated the most in time. I suspect the greater the differences between the two patients, the happier the CDC is, but they will have a problem with NC, becasue they will be virtually identical (so I suspect the CDC will hold those sequences as long as possible (but the cat really is out of the bag). 4 sequences in 2 clusters is a MAJOR proble for "spontaneous mutations", and widespread H274Y is cause for concern (and supported by the DATA).

Wouldn't the presence of only a single polymorphism that leads to Tamiflu resistance itself tend to discredit random mutations. From following your various discussions in different threads here and elsewhere I gather that there is at least one other known polymorphism that leads to Tamiflu resistance and that it is at least reasonably fit. In the case of purely random generation of Tamiflu resistance, shouldn't we see roughly equal amounts of both of those?