| Jesse Goodman is a man with a great deal of expertise and experience, in vaccines and the science behind vaccinations (see here and here). When someone with such solid credentials speaks, we'd do well IMHO to listen. His words are also important since the FDA has bucked the European and Canadian trends and has, to date, not yet licensed any vaccine containing non-alum adjuvants. It may be instructive to get a glimpse of why.
Goodman's comments center round the implications of increased reactogenicity, ie immediate short-term reactions to vaccines, and whether such reactions are indicative of problems in the future. Clinical trials as currently designed are not very effective for detecting delayed adverse effects, let alone establishing causal relationship between them and vaccination. (See examples here, here, and here.) Plus we are breaking new grounds with these new, non-alum adjuvants: there is no data anywhere, on the longer-term effects of such vaccines especially in non-elderly populations (and limited data even in the elderly), to guide us on such basic questions as what adverse effects to look out for, or whether they should or should not be considered 'possibly related to vaccination' (see below).
We cannot assume that such powerful vaccines will only cause the same kinds of side effects as you would see in conventional vaccines. (Here's why) The experience of Vioxx should tell us that such assumptions can be dangerous - that powerful new drugs or vaccines can have unexpected but serious, even life-threatening, effects. In the case of Vioxx, outside of those directly involved in that area of research, who would have thought that an arthritis drug would cause heart attacks? And yet it did. (There were in fact plenty of 'signals' - the issues were hotly debated within the FDA - but they were ignored by those whose fingers were on the button, at a cost of 28,000 extra heart attacks and sudden cardiac deaths.)
Similarly, in the GSK trials for the AS03 vaccine, there were 2 cases of stroke within 6 months of vaccination among 500 clinical trial subjects aged 18-60. (The expected risk of stroke for those aged 45-54 should be 0.6 in 500 persons per 6 months, less for younger people, but a bigger group is needed to test for statistical significance). Were those related or unrelated to vaccination? GSK says not, but do we know that to be the case? We don't have any data to guide us. Can we be comfortable assuming no causal relationship, when scientists are now telling us that inflammation plays a bigger role in strokes than we used to think? (Stoll 2006, Welsh 2008)
Another example is Cervarix, the HPV vaccine by GSK adjuvanted with AS04, which is a combination of alum and MPL, a derivative of bacterial endotoxin. There are no hard figures (ie statistically significant ones) but the FDA meeting transcripts do show a fair amount of concern over possible excess of autoimmune and neurological conditions, and spontaneous abortions. Some of these are not the kind of adverse events we would expect, (or look out for) in conventional vaccines, so it really is uncharted territory, what these new adjuvants may or may not cause months or years down the road. With regards to AS03, the FDA has asked GSK to do a 2-year follow-up study of their clinical trial subjects, which I think is a good idea. But it also gives you an idea of the kind of timescale we are talking about, before we have some vague idea of the range of risks (or lack thereof).
So, what Goodman is saying, is that we don't really know whether increased reactogenicity as observed in clinical trials, implies increased risk of longer-term serious consequences including, as he said, neurologic complications. It may. It may not. But we don't know. "There just aren't those data."
When the Chief Scientist of the FDA says that, we would be mad to not pay attention.
We do know that adjuvanted vaccines in general are more reactogenic. For example, here's a chart (adapted from their EMEA file) showing the local adverse reactions following the GSK vaccine in adults, compared to controls receiving the unadjuvanted vaccine.
Note that >50mm of swelling and/or induration (=hardness) is about the size of a tennis ball, which IMO is pretty darn big, for an injection site reaction. Such strong local inflammation invariably results in disorderly cell death and release of intracellular contents, which, as we have seen, are processed by immune cells in the same way and at the same time as they are processing the vaccine antigen. In the presence of a strong stimulatory environment generated by the adjuvant, is there a risk of activating autoimmune processes* against such self-antigens? There's no hard data to help us quantify that. Immunologists studying the effect of infections and other triggers on autoimmune diseases, find that typically it takes a while (ie months or years) to progress from the first subclinical autoimmune reaction induced by a trigger, to eventual disease, so that by the time the patient has symptoms, the original trigger is long gone. They call these 'hit and run' events. (Christen 2004, Ma 2005, Krishnan 2006) The science would suggest that adjuvanted vaccines causing strong local reactions may act as such triggers for susceptible individuals, but again there's no data to tell us for sure and to what degree, one way or the other.
(*For more on cell death and autoimmunity, see Sauter 2000, Savill 2002, Kremer 2009)
All that is for local reactions. Let's look at systemic adverse events, ie those beyond the injection site. Again, adjuvanted vaccines are more reactogenic. For example, for the GSK vaccine, for adults aged 18-60 yrs, > 60% of those in the adjuvanted group complained of myalgia (ie generalized muscle pain unrelated to injection site), as compared to 20-30% in the unadjuvanted group. Fever, fatigue, headache, and lymphadenopathy (swollen, painful, or inflamed lymph nodes) were also more frequent in the adjuvanted group.
Let's see what happens in kids. There was a 3-staged clinical trial with the H5N1 vaccine on children aged 3-9, as follows. Control groups were given the unadjuvanted seasonal flu shot.
I'm going to focus on the two groups (highlighted in red) most relevant to the current pandemic: those aged 3-5 yrs receiving half the adult dose in Phase A, because that is the dose that will now be given to kids 9 or younger. And those aged 6-9 yrs receiving the full adult dose in Phase C, because that is the dose that will be given to anyone aged 10 or over. (No trials were done for anyone from aged 10-18yrs.)
This first table shows AEs in kids aged 3-5 given 1/2 adult dose. Grade 3 symptoms are defined either by size of local reaction, or as those causing some restriction in activity, or >39C in the case of fever.
And this for those aged 6-9 given a full adult dose:
The very first thing to note, is how the current unadjuvanted flu vaccine is quite well tolerated by kids (0% fever), in general, and how differently they react to the adjuvanted ones. Let's look at some more comparisons of systemic symptoms:
 
As you can see, there is a huge excess of systemic symptoms, in both age groups. It's remarkable that the younger kids (aged 3-5 yrs) had very little reaction to the unadjuvanted vaccine, but had some really strong reactions to the adjuvanted one. In the older age group, the biggest differences are in fever and headache.
But what do these symptoms mean? Do they have any significance for the longer term, as Goodman seemed to be worried about? I don't know the answer to that, but again science comes to the rescue. The mechanisms behind these reactions, may give us some clue.
The following is a diagram (Perry 2003) of the acute phase response, which is your body's early/immediate response to injury of any kind, with or without concomitant infections. Soon after such injury happens, the innate immune system swings into action, to contain and limit the damage, prevent further spread eg of an infection, and promote tissue repair. Note that the acute phase response works the same way whether the trigger is an infection in the lung, as shown in the diagram, or a vaccine that causes your arm to swell up, or someone got hit by a car, or after surgery, or a non-infectious illness like acute pancreatitis. (Gabay 1999, Gruys 2005) Such sharing of pathways has implications, but first let's look at how it works.
Here are the basic steps, following the numbered arrows in the diagram.
- Cells of the innate immune system, like neutrophils and macrophages, respond to signals of injury. They move into the area and start the 'clean-up', eg by phagocytosis. In the process, they become activated and secrete a number of cytokines, IL6 being one of the earliest ones strongly triggered. Over a certain threshold of inflammation, cytokines and cell debris spill over into the blood, and travel to other parts of the body.
- In the liver, IL6 triggers the production of a large number of proteins (eg C-reactive protein, fibrinogen) into the blood, and causes a reduction in other proteins. These are collectively called acute phase proteins. Here's a list which also gives you some idea of what these proteins do.
- IL6 and other cytokines also have effects on the brain, which results in the symptoms that we recognize as being 'sick', eg fatigue, headache, loss of appetite, or myalgia, plus fever.
- The bone marrow is also stimulated to produce more white blood cells, to fight the infection.
The acute phase response is fundamental to all systemic immune responses, including when someone gets infected by flu. Cytokines play important roles in severe or fatal influenza eg ARDS or influenza-related encephalopathy, but we'll save that for another day. For right now, let's take a closer look at fever as a marker of systemic reaction to adjuvanted vaccines. The following diagram (Eccles 2005) gives some more clarity on how cytokines induce fever.
You can see that cytokines, including especially IL-6 (Chai 1996), are an essential component in the pathway for induction of fever. Whether these cytokines work by directly passing into the brain, or by stimulation of the vagal nerve (probably both), they have to be present in the circulation, even if transiently, for fever to occur. (Leon 2002, Rummel 2006) In other words, people don't get fevers unless they've had a systemic cytokine response, and IL-6 is an essential part of that.
Now, does this constitute 'proof' that AS03-adjuvanted vaccines induce a systemic cytokine response? No, not unless and until it's measured and the data published. However, for the purpose of non-specialists who are more concerned about whether to take the vaccine than the finer points of science, there IS a strategy called making an educated guess.
So what is an educated guess? Let me use an analogy. You know that your car battery needs to be working in order for you to start your car, right? When you get up in the morning, climb into your car, and turn on the ignition, and the car starts nicely and easily, what does that tell you? It tells you that you've got a battery that is working (well, at least not totally dead!!) Notice you didn't have to go look under the hood and test the battery; you can 'test' it (and 'prove' it) indirectly by turning on the engine. In the same way, there are certain biological phenomena that logically follow one upon the other, where the mechanism is well established. Just like a functioning battery is a requirement for the car to start, a systemic cytokine response, however transient, is a requirement for fever and other systemic symptoms after vaccination such as headache, myalgia, fatigue, etc.
Note however that such 'educated guessing' is useful only on the conceptual level, ie to answer the question of whether, in principle, an AS03-adjuvanted flu vaccine can or cannot induce serum cytokines. Whether that does happen in a particular person, is a lot harder to tell, because individual cytokine responses vary over a wide range, partly due to genetic factors. (Bennermo 2004, Hildebrand 2005) Which means it's much harder to make an educated guess as to how you specifically might be affected by such vaccines, even though we might think, statistically, (eg from the relative frequency of 'reactogenicity'), that if you vaccinate a large population with an adjuvanted vaccine, you're probably more likely to have some people who have stronger and systemic cytokine responses, than if you use an unadjuvanted vaccine.
In addition to the chronic illnesses that might be induced by a high or dysregulated IL-6 response, as mentioned in the previous diary, would a strong surge of systemic cytokines affect concurrent illness, eg influenza infection? I haven't found any data that specifically addresses this issue, but there are some interesting models that I hope to explore at another time, if only to improve our understanding of influenza as an illness. For the next diary, let's turn out attention to the implications for pregnancy.
In the meantime, I'll leave you with some food for thought - a slide from Dr Goodman's presentation. Note especially where he underlined the text:
 |
