Cytokine storm

A helpful reader (hat tip to easy hiker) sent along a story from New Scientist concerning a new report in The New England Journal of Medicine. The NEJM paper is a case series of six subjects who almost died as a result of a clinical trial of a monoclonal antibody being tested as a drug for rheumatoid arthritis. Within an hour of receiving the drug, TGN1412, all six suddenly developed a cytokine storm syndrome, with severe pain rapidly developing to multiorgan failure and respiratory distress. This result was unexpected as nothing like it had occurred in the animal trials and the dose given was a mere fraction used in the preclinical experiments.

According to the New Scientist, within hours the subjects' lymphocytes and monocytes (their white blood cells) had vanished:

The injected antibody was unusual, because it was capable on its own of provoking lymphocytes, called T-cells, into becoming as active as they would be if they had to fight an infection. It normally takes two signals, not just the one provided by the antibody, to awaken T-cells. The antibody - known as a superagonist - was designed to be able to activate any type of T-cell without requiring the usual secondary signal. It works by binding to a receptor called CD28 on the T-cell surface.

In earlier trials on animals, the antibody initially triggered multiplication of T-cells, but a specific subset, called regulatory T-cells, ended up multiplying fastest and taking control. The regulatory T-cells calmed the immune system. TeGenero [the company developing the drug] hoped that this immune "calming" process offered potential therapeutic benefit, perhaps easing symptoms of diseases like rheumatoid arthritis, where normal T-cells attack the body's own tissues.

Unfortunately, the reverse happened. Around 60 to 90 minutes after the men received their injections, their bodies were flooded by a surge of inflammatory chemicals called cytokines, which combat severe infections like those seen in patients with blood poisoning. The cytokines caused severe inflammation.

1 hour - simultaneously, the six men begin suffering excruciating headaches, shivering, back pain, gut pain, diarrhoea, swelling and nausea

4 hrs - all have fevers, are flushed, their blood pressure drops dangerously low and their hearts start to race. Blood tests show their lymphocytes and monocytes are fast vanishing.

5 hrs - one patient begins fighting for his breath. All suffer lung pain. They are all given steroids and other medications to ease inflammation.

12 hrs - the patient fighting for his breath is so bad that he has to be taken into intensive care and put on a ventilator to keep him alive.

Suntharalingam [physician supervising the trial] decides to take all the volunteers into intensive care as a precaution.

24 hrs - two people are on ventilators, and the four others need support with breathing.

48 hrs - the four least affected men start to recover, but all six begin to suffer multi-organ failure, and have to be attached to kidney machines, one of them for a fortnight.

Thereafter, the men began to recover and their lymphocyte and monocyte counts began to creep up again. (New Scientist)

This paper is interesting because it illustrates how little we know about the genesis of the set of conditions we call collectively a "cytokine storm." Cytokine storms are suspected of being one of the deadly consequences of human infection with H5N1. As we emphasized in our brief discussion on cytokine storms at The Flu Wiki, cytokine "storm" is a metaphor, not a technical designation. It refers to a disastrous dysregulation of the usually localized chemical signaling system used by the many different kind of cells in the immune system that must coordinate their actions in precise ways. The cytokine system is a complex and interlocking network of positive and negative feedback signals. Sometimes very small changes can cause large consequences. For example, if your thermostat gets stuck in the position calling for more heat, your furnace can fail, or at least your wallet will when you get your heating bill. The drug, in this case, seemed to put the immune system into overdrive, although exactly how isn't clear.

It is a mistake to believe that if H5N1 kills because of a cytokine storm, interpreted as an "over active" immune system, then anything that stimulates the immune system is a risk factor or anything that damps it down is beneficial. The cytokine storm phenomenon, whatever it is (and it is almost certainly more than one thing lumped under a common name), is not amenable to simplistic formulations. Treating cytokine storms in various clinical settings, such as gram negative sepsis, has shown how difficult it is to manage. So far there is no magic bullet or agreed upon therapy. The apparent success of steroids in this case (and the role of steroids here is still conjectural), if true, may be only a reflection of the extremely specific effect of this drug. Steroids in the context of an infection with a virus that does many other things would work differently.

This paper tells us that detailed information from some animal models may not be applicable to humans. We still have very imperfect understanding of how cytokine storms develop and why. It may also be a clue of where to look for how H5N1 provokes this reaction, but it is likely there is more than one way to cause cytokine storm like reactions, although one candidate would be evoking a protein that acts like a super-antigen, as in the case of this drug.

As always, we know too little.

More like this

Revere: within an hour or two dozens of genes are upregulated to produce a bewildering assortment of cytokines once H5N1 gets into the nucleus, so this report also demonstrates how frighteningly fast this can happen. Thanks for the report, because I am actively researching the use of steroids in such cases, and my first review was turned down at one journal because the reviewers specifically said the evidence against steroids being useful agents in animals was strong. (Only up to a point, in my opinion.) This is also illustrates that no matter how good your animal models are in H5N1, they are far from good analogies to humans.

Ah yes. We are legends in our own minds.

The fact is that while we advertise our vast knowledge and technologies to the general public pronouncing that we know more than in 1918 and technologies are so much more efficient at controlling pandemic influenza...we actually only think we understand the various complexities of the body.

"the evidence against steroids being useful agents in animals was strong."

Marissa. Anyone making this statement has never been out of the lab or has never treated cattle with acute respiratory viral infections.

Are steroids useful in all animals for all infections...probably not.

However, with respiratory infections in cattle that mimic the clinical signs of SARS almost exactly, they were the difference between 5% mortality and 25% mortality.

We may soon see many individuals in the US die of a cytokine storm, as a result of an H5N1 infection unless something is done to stop the madness of insisting sick chickens are the source of human H5N1 infections in Asia. If you think it is a pleasant way to die, just read a few books on the 1918 Spanish Flu Pandemic, in which heroic doctors and nurses sat by the beds of patients dying of Spanish Flu victims, and recorded the horrific consequences in their bodies. The patients bodies often turned black, with blood coming from the nose, mouth, or ears. If you do not want to die that way, then insist immediately that WHO and public health officials in Asia start testing the blood of animals in asia to find a blood match to human H5N1 victims. Of course I have no idea how you would communicate with WHO or Asian governments, but this is real insanity. There needs to be testing done of dogs, cats, and pigs in Asia in the areas where there are clusters immediately.
I do not know how I can make this understandable except to say the scientific evidence shows the vector for H5N1 infections is probably a mammal. The only match we have is from a cat. They can kill every chicken in Asia, but it will not stop humans from becoming sick with H5N1.
Radio Australia yesterday reported China's health ministry says a 62 year old farmer from the northwest region of Xinjiang died of bird flu in July.
And it says: "He is not believed to have had contact with other infected people or outbreaks in poultry." Just for a moment please assume the above statement is correct. The question then becomes: Then by what vector did in become infected with H5N1 and die? And the answer may be a cat, or a dog, or a pig, or some other mammal infected him. And it is now time to stop looking in the wrong place for the vector, unless you want to see your friends and family die of bird flu if there is a pandemic.
The behavior of public health officials in Asia is very similar to the behavior of public health officials in the US during the Spanish Flu Pandemic. Many doctors insisted the deadly flu was caused by a bacteria, and they kept looking for that bacteria in the blood of those victims that died. Of course they never found the bacteria.
And now we humans are desperately trying to kill chickens, and insist the source is in chickens, while people continue to die at a rapid rate. Do you see any difference between what the doctors in 1918 did, and what the doctors in Asia are doing? I do not.
There is still a vast amount we do not know about H5N1 infections, and even if we know the vector, we may not be able to stop a pandemic; and there may be multiple vectors.
Also it is good to cull chickens because it may stop the virus from mutating to a form in chickens that can infect humans. But right now the source is probably not chickens.

In another post today, Marissa, in referring to the large fatality rate among pneumonia victims in Thailand, stated 6 to 7% may have been infected by H5N1.
If we assume 6 to 7% have H5N1, and if that assumption is factual, then would such a large number indcate an H2H pandemic may have begun?

My friend Dr. Hideki is getting info in from Thailand daily and he said that its becoming really difficult to get a confirmation of the H5N1 in people that they fully believe that have it. He said IHO that the testing is either not being done correctly or the virus has changed again and the test is inadequate. Also he said that IHO that to have that many cases of "pneumonia" its an epidemic in itself in the summertime. His other big concern is that these people are reporting to HCF's sick as dogs and certainly some of them likely have H5. They are being placed in to ward situations and dual rooms with no isolation protection because there is nothing that says they are H5 positive.

Revere, is the testing going inadequate to the task?

By M. Randolph Kruger (not verified) on 16 Aug 2006 #permalink

M.Randolph Kruger,
"He said IHO that the testing is either not being done correctly or the virus has changed again and the test is inadequate. Also he said that IHO that to have that many cases of 'pneumonia its an epidemic in itself in the summertime."
Thank you for this very important information. I am honestly unable to even guess what all this means, but the information is disturbing. This is like a bad dream where you would like to wake up, but cannot. I feel like I am in that dream, watching helplessly as a pandemic emerges, and knowing the authorities in Asia could be taking constructive action, but due to ignorance, they continue looking for the source in sick birds, while massive numbers of people are being infected by a mammal, such as a cat. Of course nothing is being done to find the real vector, and so the epidemic spreads. The mammal might be people, and so we may have H2H.

M.Randolph Kruger,
In a report from the Sourtheastern Cooperative Wildlife Disease Study at the College of Veterinary Medicine at the University of Georgia, dated April 2006, in a section entitled Avian Flu in Mammals, it says:
"The highly pathogenic avian influenza (HPAI) H5N1 virus of current concern not only affects birds and humans, but has also caused illness and death in a variety of other mammals."
..."Domestic animals can also be affected by HPAI H5N1 virus. The virus has been isolated from dead domestic cats in Germany and Thailand. An unpublished survey in Thailand identified dogs with antibodies to H5N1 virus, but whether the virus causes disease in dogs is unknown. Of greater concern with domestic mammals is the finding of H5N1 virus-infected pigs in Southeast China and Indonesia. Pigs carry swine influenza viruses and are susceptible to infection by both human and avian influenza viruses and can become 'mixing vessels' to produce a reassorted virus that could cause a pandemic in the human population."
It is time to start testing the above mentioned mammals for H5N1 in the areas in Asia where there are clusters of Human H5N1, to find out if a match can be found between the human H5N1 and a mammal H5N1.


Within an hour of receiving the drug, TGN1412, all six suddenly developed a cytokine storm syndrome, with severe pain rapidly developing to multiorgan failure and respiratory distress.

My professional training is as an engineer, not as a physician.

But I find this procedurally startling. First ever administration of an experimental drug in humans, and it was given to six individuals simultaneously?

If I had been designing the test protocols, I would have planned to start with one test subject, and would have scheduled to wait a week before moving to the next. For eff's sake, I would have done that if testing on a *machine*, much less on a human volunteer.

That procedural model would still have left one desperately sick individual fighting for his life, to be sure. But one beats six.

If stupid crap like this continues to happen, the pool of volunteers for medical trials will dry up, to everyone's collective detriment.

Eh..talk about confusion.My other half has a donor heart and takes immunosuppressants.Is he in the best or worst position re.human BF?Who knows.

Revere: I am just curious as to whether mice or rats could possibly become infected with H5N1 and if so, has anyone been testing them in outbreak areas?

Today, Thursday, a report on the cluster of H5N1 infection in West Java's Cikelet village it says: "Contact with sick fowl is the usual mode of transmission of the H5N1 bird flu virus, that is endemic in poultry in nearly all of Indonesia's provinces."
In another report on the same village it says: "Almost all human H5N1 cases have been linked to close contact with sick or dead birds, such as children playing with them or adults butchering them, according to WHO."
Thes 2 statements are part of the mythology of human H5N1 infections, and this myth is lethal and dangerous. Those public health officials that base their actions regarding human H5N1 infections on this myth are idiots.
As Niman said today, it is true that most human H5N1 infections occur in and around areas where there are infected birds. But the direct link has never been proven.
As the virologist Andrew Jereminko, who worked in Indonesia for many years, the only blood match they have for human H5N1 infection is from a cat. Scientists are ignoring this evidence as if it does not exist.
When the germ theory of infection became known in past centuries, many doctors laughed, and thought the idea was stupid. Why should a doctor wash his hands after visiting a patient. The idea is total nonsense, and the theory was ignored. Of course many patients died as a result.
And now humanity laughs at the idea that sick birds are not directly infecting humans with H5N1. That the transmission my be indirect from birds to a mammal, such as a cat or dog, and then to humans. And so, in their stupidity and blindness, doctors state over and over again, as if they reciting a prayer, that sick birds infected their patients.
And now, as the number of H5N1 infected patients increases in large numbers daily, the the clusters grow larger and larger; the evidence that the source, or reservoir, or vector for human H5N1 infections is probably a mammal, such as a rat, or a cat, or a dog; is ignored, while
doctors look on as their patients continue to become infected and die.
As I reported above in another post, an unpublished survey in Thailand identified dogs with antibodies to H5N1 virus, but whether the disease causes disease in dogs is unknown. What happens if asymptomatic dogs are the vector for human H5N1 infections? In other words the dog is infected, but is healthy, and infects humans.
My only wish now is that doctors continue to ignore the scientific evidence, and as a result the now epidemic of pneumonia cases in Thailand, many of which are very possbily H5N1 cases, explodes into a pandemic, and millions die.
The stupidity of humanity knows no limits. All we can do now is watch global climate change, caused directly by the capitalist system, destroy more and more animal and plant life, and soon destroy humanity as well.
Let the fools, meaning the public health establishment, keep the myth alive, and just set back calmly and watch,as a wave of certain death and sickness decends on humanity, just as it did in 1918. And if it kills you, be grateful doctors are so stupid.

William: If in fact some small percentage of patients in Indo have H5N1, it does not signify by itself that there is a higher level of h-2-h transmission; for example, there could be a rampant source of the virus that we don't know about. What concerns me the most is the opportunity for h-2-h transmission in hospitals where there is no or limited isolation. Further, it increases the chances for recombination of H5N1 with other strains.

Mara: I'm sorry to say that your better half will be vulnerable, based upon the research I came across this and last year.

Judy: I think the possibility exists for sure. What I find astonishing is that the current clades of H5N1 have managed to adapt to a fairly widespread cross-section of mammals.

Marissa,
"If in fact some small percentage of patients in Indo have H5N1, it does not signify by itself that there is a higher level of H2H transmission; for example, there could be a rampant source of the virus that we don't know about."
Your posts on the blog site are for me important, because I learn a great deal by reading them.
Your statement above is totally correct regarding H2H infections in humans. But what facinates me in that statement is the last phrase--"there could be a rampant source of the virus that we don't know about."
With that phrase you have summarized the problem much better than I ever could. I wish I could post that phrase all over this blog site in giant neon lights for all the world to see. Perhaps that would wake up the walking dead
experts at WHO, and the public health officials everywhere.
Today Henry Niman said in a post on his site that it is time to test mammals in Asia and elsewhere for H5N1. And of course I learned from Niman that the sequences released showed birds were probably not that source.
Therefore it is time to start testing mammals, to find the source; and WHO should be at the lead in this effort.
To ignore what Henry Niman is saying in this regard is insanity. I realize what a difficult field situation Asia represents, and I know there are not enough people to do the testing. But at least an effort could be made to do it.
I believe you used the adjective "rampant;" because; to observe those pneumonia patients in Thailand, many of which may be infected with H5N1; and to notice all the growing clusters of human H5N1 infection in Asia does indicate rampant is the correct adjective to use.
Please ignore my statement in my previous post where I said I wished the world would suffer a pandemic. That statement was made our of frustration and despair. Of course I do not want to see a pandemic. And many doctors and other medical staff were heroic during the Spanish Flu Pandemic of 1918, and should be admired for their heroism.
But you, who obviously know much more about H5N1 than I, give me hope that perhaps testing of mammals in Asia will be done before it is too late, if it is not already too late.
If the field work to determine the mammalian source or sources is not done soon, it will probably will not make any difference once H5N1 goes H2H. And even if we knew the source in mammals, it does not mean we could stop H5N1 human infections. But if we don't know the source, then many more people may die.

Marissa: If rats, then why not fleas? They infest dogs, cats, chickens, humans, houses, sand, and on and on. This might possibly explain cases where there is no contact with any obvious source, and if everyone is looking at chickens this could easily be overlooked. I admit I know next to nothing about this so if this is wrong thinking, please explain why with much thanks from me to you.

Judy, while this is a possibility, fleas would be a totally new vector for H5N1, and the fleas would have to have contact with both humans and rats, like the plague. Don't forget there that are many species of flea. I'm not an entomologist, so I don't know how many species there would be in Indonesia.

William, given the history of SARS, such an undertaking would not likely happen, though, like you, I call for it. It would require testing on a scale not seen before, since we are trawling for species that are unknown reservoirs. We must also consider the possibility that there is more than one species acting as a reservoir. I'm going to do some more research into SARS to see what models we might come up with. The situation in Indonesia and Thailand is clearly deteriorating, and unless the authorities there get a handle on it and become more transparent about reporting, I think we are going to run out of time in the next couple of months. All we need now is more efficient h-2-h transmission and we'll go from Phase 3 to Phase 6 in about a week. Many experts have hung their professional reputations on the theory that this is very difficult for the virus. (A question of probabilities, really.) I, for one, think that all the experiments done with H5N1 to date with glycan microarrays (and comparison with H1N1 1918 strains) are misleading; in truth, I'm not sure we know what markers are important. And that's another change for me in the last few months. With the new cleavage site and the PB2 polymorph E627K, I no longer think that the mortality rate would drop concurrently with HA subunit changes necessary to support higher h-2-h transmission. On the contrary, this is far less likely than retaining them. this, too, is a change of opinion for me in the last 6 months.

Marissa: regarding your statement that you no longer think the mortality rate would drop with the HA subunit changes necessary to support higher H2H transmission. That should have been bold faced type, caps. The "experts" have been basing all their pandemic preparedness plans on a certain relatively low rate of infectivity (I forget what) but I do recall they believed the virus mortality rate would drop to about 2% of those infected. Now we are seeing, however, with every new case and cluster an increase in the mortality rather than a decrease. Can you give an informed prediction? What is the expected % of the population that could reasonably be expected to come down with H5N1 if it goes to stage 6, and of those, how many might be expected to die?

By mary in hawaii (not verified) on 18 Aug 2006 #permalink

Mary wrote: What is the expected % of the population that could reasonably be expected to come down with H5N1 if it goes to stage 6, and of those, how many might be expected to die?

These are the $64,000 questions. As to the first question, I really have no idea. To qualify as a pandemic, H5N1 must increase its h-2-h dramatically. That means the CAR (clinical attack rate) must rise to about 20-40%. It is possible that it might be higher than this--say 60%, though this might not happen until the second wave. I would expect 2-4 waves, with the second wave by far the worst, though I base this on the 1918 pandemic, and H5N1 might not follow this course. Let's be conservative and suggest an MR of 50%. It could be a lot higher, though we pray it would be a lot lower. In the USA this scenario would lead to about 180 million infected, and 90 million deaths. Worldwide, about 2 billion deaths.

If the MR does not come down at all, those numbers will be higher. The reason I think that the MR will remain high is that the high MR is due in part to the E627K polymorphism on PB2, which arose from the Qinghai strain. Changes in the HA subunit do not materially affect the PB2 gene. The new cleavage site also contributes to the high MR, but I don't think that the changes required to the HA gene that will increase h-2-h transmissability will affect it. I could be wrong about that, and I hope so, but I doubt it. I base that on looking at the HA modeling results to date.

However, if H5N1 acquires increased transmissibility in the HA gene via polymorphisms acquired by recombination, it's possible that the cleavage site might also change, which might bring the MR down some. This is just one epidemiologist's opinion, though I know a few others have gone on record as essentially agreeing with most of these points--Robert Webster for one.

What's key right now is watching for the development of larger clusters in Indonesia--30 to 50, or larger. If we see that, we're dead trouble.

Marissa: I don't think we can be sure about the role of E627K in virulence. The question of genetic correlates of infectivity and virulence are still unclear and we just don't know. The uncertainties in attack rate and virulence are so wide I don't think it is informative to make predictions except for upper and lower bounds. It is quite clear that we need to prepare for a serious degree of community illness and mortality.

Marissa, revere et al.:
And what would you estimate the ciollateral damage be? Due to the collapse of societies, secondary and tertiary diseases and the crime rate going up?

Revere, I think you're being conservative, when you say "we just don't know," regarding infectivity and virulence. Both NS1 and PB2 genes code for virulence in varying degrees and operate separately from the HA gene. I agree with you, that we don't know all the polymorphisms that can change infectivity, and how these would affect virulence. In my opinion, our government ought to be funding several dozen research groups on this subject right now.

I was asked my opinion and I gave it; I know yours is different. What I would be ashamed of is if a pandemic occurs and we're body bagging by the truckload and we're still saying we don't know. We need bright unknown scientists to step forward with theories that can be examined in the public view. Journals tend not to publish these theories--"too unsubstantiated" they usually say. Well so what? Alfred Wegener died long after his theories of plate tectonics became accepted. And look at the fuss the medical community kicked up when 2 Australian scientists found the true cause of stomach ulcers--i.e. H. pylori.

As always, we know too little? Somebody needs to tell that to the Govt. cronies in Washington, who let their corporate masters feed us anything that'll fatten their wallets. We've become lab-rats for the elite.