|I always believe in giving people the benefit of doubt, and, at least in the first instance, always taking them at their word. In examining the safety concerns about MF59, I believe the best approach is to look at the work of the most experienced scientists on such adjuvants, and see how the case for safety is made.
For that reason, I'm going to explore a couple of articles written by top experts and published in peer-reviewed journals on this subject. Since these experts are also employed by manufacturer of MF59, Chiron previously and now Novartis, I believe we will not be in danger of ignoring any data that is favorable to their case.
I'm going to restrict my comments on issues of safety only, and will not address the efficacy of the vaccine discussed or potency of the adjuvant. With one exception only, I am going to pick out all comments pertaining to safety of MF59 and examine the evidence presented. The one exception is the use of MF59 in HIV vaccines, as I'm not clear whether host immune responses in these instances can be extrapolated to healthy subjects. In addition, this attempt of mine is for public discussion and understanding only, and is limited by my inability to access all primary sources.
The following paper is written by one of the top experts on the subject (a cursory search on PubMed returns 89 publications over the last 20 years mostly on this and related subjects), appears to cover the topic very comprehensively, and makes as clear a statement as you can find in its title. MF59 is a safe and potent vaccine adjuvant that enhances protection against influenza virus infection. O'Hagan, DT, Expert Rev Vaccines. 2007 Oct;6(5):699-710.
First of all, Finalcial & competing interests disclosure
Derek O'Hagan is Head of Vaccine Delivery Research, Novartis Vaccines and Diagnostics, Incl, MA, USA.
In preclinical studies, MF59 adjuvant offered improved protection against influenza virus challenge and significantly reduced the viral load in the lungs of challenged mice. In humans, MF59 is a safe and potent vaccine adjuvant that has been licensed in more than 20 countries (Fluad [Novartis Vaccines and Diagnostics Inc., MA, USA]). The safety profile of an MF59-adjuvanted vaccine is well established through a large safety database. MF59 adjuvant has had a significant impact on the immunogenicity of influenza vaccines in the elderly and in adults who are chronically ill. MF59 has also been shown to have a significant impact on the immunogenicity of pandemic influenza vaccines. MF59 allows for broader cross-reactivity against viral strains not included in the vaccine. MF59 has been shown to be more potent for both antibody and T-cell responses than aluminum-based adjuvants. MF59 has broad potential to be used as a safe and effective vaccine adjuvant for a wide range of vaccine types.
- The first comment is not on MF59, but an older adjuvant, Freund's incomplete adjuvant or FIA.
Although w/o emulsions containing mineral oils, such as FIA, have also been used as vaccine adjuvants in humans, including in combination with influenza vaccines , they are generally considered as too reactogenic for routine human use . Nevertheless, long-term follow-up in subjects who received immunization with FIA has established that there are no significant long-term adverse effects in humans with this adjuvant, although local reactogenicity was very common .Reference #6 refers to a paper that I cannot find but I posted the abstract and my comments here. There is a different published review of this study which is discussed at some length here. As you can see, both these reviews only compared mortality data, and there are some interesting findings eg cerebrovascular lesions and higher incidence of having a diagnosis of 'therapeutic misadventure' or 'late complications of therapeutic procedures' for the adjuvant group, that are not obvious from the gross data alone.
- There is quite an extensive paragraph on preclinical studies and toxicological studies, but with the glaring absence of any references for these remarks. I would really have wanted to look at the original published findings of these preclinical studies showing the safety of MF59.
In addition to immunogenicity studies, extensive preclinical toxicology studies have been undertaken with MF59, in combination with a range of different antigens in a number of species. Prior to evaluation in clinical trials, pivotal toxicological studies were performed with MF59, including repeat-dose toxicity (including local tolerability), genotoxicity, embryo-fetal and developmental toxicity and sensitization evaluations. No treatment-related safety issues were identified in these studies and findings were generally limited to inflammatory responses at the injection site, consistent with the use of an immunological adjuvant. MF59 is neither genotoxic nor teratogenic, and does not cause sensitization in highly sensitive preclinical models.BTW In case you are wondering, this paper has a total of 73 references cited. But nothing on preclinical safety or toxicological studies or any of the stuff that is just quoted!
In comparison, in the SAME section, on preclinical experience with MF59 the author gives 6 references (#25-30) for the efficacy/potency of MF59, but nothing on safety. ;-(
- Next are the comments on Fluad, the vaccine licensed for the elderly.
The registration of Fluad was based on the results of a large clinical development plan that demonstrated, in more than 20,000 subjects in company-sponsored clinical trials, that the adjuvanted vaccine is well tolerated and more immunogenic than conventional nonadjuvanted influenza vaccines.....Moreover, in the outpatient population, the use of MF59 was not associated with any significant increase in reactogenicity compared with conventional vaccines .
Well, we've already come across one example where such large numbers as 10,000 aren't always what they seem. Again, I wish they had included the references for that statement of 20,000 cos it would be great to go and check that data! Anyway, it can't be helped...
Now, reference #34 is this paper, which as you can see, covered 204 subjects in the Fluad group and 104 in the regular vaccine group.
- Next is one example that I've already mentioned, but is so important it bears repeating.
Importantly, the addition of MF59 to the influenza vaccine did not affect the safety profile of the vaccine, which was very well tolerated in all clinical trials performed in elderly and nonelderly subjects  #32 is this study which as you can see is THE review of 10,000 elderly subjects. Not nonelderly at all. It could just be an oversight, cos we know that there is at least one study on nonelderly subjects, although we are not clear how 'nonelderly' those subject are, since the paper did not elaborate.
With more than 30 million doses distributed, the safety of Fluad is supported by extensive pharmacovigilance data that shows that vaccination with Fluad is associated with a very low frequency of adverse reactions .Unfortunately, reference #43 is something presented at a conference:
D'Agosto V, Berardi S, Burroni D, Hennig R. Tolerability and safety of an MF59-adjuvanted subunit influenza vaccine (FLUAD). IVW 2006 - the Second International Conference on Influenza Vaccines for the World, 18-20 October (2006).
Which sort of makes it privileged information, you know. Well, I KNEW I should be going to even more conferences....
The second paper that I would like to go through is chosen partly because it is cited by O'Hagan in the context of 'extensive preclinical experience of MF59' having been reviewed previously. MF59-adjuvanted vaccines: increased immunogenicity with an optimal safety profile. Podda A, Del Giudice G. Expert Rev Vaccines. 2003 Apr;2(2):197-203.
Audino Podda, Chiron Vaccines Clinical Research & Medical Affairs, Chiron Srl, via Fiorentina, 1 Siena, Italy. Giuseppe Del Giudice, IRIS Research Center; Chiron Srl, Via Fiorentina, 1, 53100 Siena, Italy.
This paper has 47 references.
The need to enhance the immunogenicity of purified subunit antigens has prompted the development of several new adjuvants. However, many of these new molecules have demonstrated a reactogenicity profile that is not suitable for their inclusion in vaccines for human use. In this context, the adjuvant emulsion MF59 has been developed, tested in combination with different antigens in several animal models and subsequently evaluated in humans. Clinical trials with several MF59-adjuvanted vaccines have been performed in different age groups (from newborns to the elderly) and have shown an increased immunogenicity of coadministered antigens, associated with a high level of safety and tolerability. MF59 has been the first adjuvant to be licensed for human use after alum and, as part of an enhanced influenza vaccine for the elderly, is now available in the marketplace of several countries worldwide.
- Again, the first place that we land in, is in preclinical studies.
The adjuvant emulsion alone and/or in combination with different antigens (eg influenza, HSV, HIV and others) has been tested in several toxicological studies aimed at evaluating the safety profile of this compound before commencing clinical trials. Toxicological studies, performed in rats, rabbits, dogs and other animal species, included acute toxicity studies, repeated-dose toxicity studies, mutagenicity studies and teratology studies. The overall conclusions from this extensive toxicological evaluation was that MF59, either alone or in combination with different antigens, has an acceptable safety profile and can be safely administered to humans.
Again, I wish they had put down the references. I mean, by the sound of it, they probably spent a fortune and years on this, what is there to not share with as many people as possible?
- This next is on the adjuvanted flu vaccine, this time on 12,000 elderly subjects.
although the presence of the adjuvant is associated with an increased rate of local adverse reactions, particularly local pain, most of these reactions were mild in nature, of short duration and qualitatively similar to those induced by control vaccines [23-28]. Phew! FINALLY we are getting some references!
Here are the links to these references, for y'all to browse, cos these ARE the adjuvanted flu vaccine studies!
#23 is here but is also the same study discussed here
#24 is here
#25 is here
#26 is here
#27 is here
#28 is here
Good. Now we know there's lots of data on the safety profile of Fluad in the elderly, notwithstanding all the caveats that we already discussed.
- Next we have clinical trial on a pandemic vaccine with H5N3, first there is this Phase 1 trial, subjects mean age around 28 years. 32 given MF59-adjuvanted vaccine, non-adjuvanted 33. The second is when they rounded up the same subjects (found 26 of them) to study the effect of a booster shot. 15 of these individuals got the adjuvanted vaccine in this trial
Anyway, for these 2 trials, here's the author's comment:
From a safety point of view, the addition of MF59 was very well-tolerated and did not provide any evidence of clinically important reactogenicity.
- Other vaccines: Hepatitis B vaccine, link
As expected, due to the presence of MF59, the adjuvanted vaccine induced a higher rate of mild local reactions (pain at the injection site was the most frequent reaction). However, the incidence of systemic reactions did not significantly differ between MF59-adjuvanted and alum-adjuvanted vaccines.
- This is the interesting one, 2 studies on HSV vaccine - with 2 recombinant glycoproteins.
Several things to note about these 2 studies:
- They both used MF59 as placebo.
- Note the high drop-out rate 23% for the first study, 11% in the second.
- Of these, 10 patients (7.3%) in the first study, and 8 (4%) dropped out specifically because of adverse reactions
- The drop-out rates were the same for the 'vaccine' group and the 'placebo' group.
Here I'm indebted to the work of Matsumoto, who provided the following information in his book. Remember this chart from the first Tulane study?
Apparently, the 2 'NIH vaccine participants' on the chart came from these studies, I believe it is the second one.
The story as written up by Matsumoto, is that one of the patients in the trial, patient X, a doctor who had received 'a placebo' went to the Senate Veterans Affairs Committee to report that he had developed symptoms that were very similar to what the Gulf War veterans were having - muscle and joint pains, twitching of muscles, rashes, fatigue, tingling in limbs, "possibly auto-immune peripheral neuropathy", had to stop working, etc. Except of course he was not in the military, didn't get the anthrax vaccine, didn't have any of the exposures, etc. The Senate Committee who knew of the work of Pam Asa at Tulane, sent him to see her. He and another patient from the trials who got 3 doses of the vaccine and also got ill, both tested positive for anti-squalene antibody (ASA).
The significance of this is, here you have 2 people KNOWN to have been given MF59, got sick shortly after being given those vaccines with symptoms of auto-immune disorders, and then tested positive for ASA.
Leaving aside the issue of whether this validates the ASA test, what else does this story tell us? For me, the main purpose here is to arrive at some judgment as to how to interpret these papers published by Chiron and Novartis, like what does it take for them to tell you something went wrong?
When I look at how the adverse reactions are described in the studies, even though they did admit to the patient drop-outs, and there's this whole list of adverse reactions, and the lengthy description in the paper itself, the authors still asserted that Vaccination side effects were limited in severity and duration.
AND, despite all that, in this review of MF59-adjuvanted vaccines by Podda, you find this remark.
Three immunizations with these proteins together with MF59 were very well-tolerated.
- One final trial I want to write about, Immunogenicity of a recombinant human cytomegalovirus gB vaccine in seronegative toddlers.
As far as I can tell, this may be the first MF59 adjuvanted vaccine given to healthy toddlers except for those born of HIV-positive mothers. The important thing to note is that the purpose of vaccinating kids against CMV is not to protect the kids, not the ones receiving the vaccines anyhow.
CMV is frequently transmitted by kids to caregivers, including pregnant women who may then give birth to babies with congenital CMV infection.
I believe it is important to find ways of preventing congenital CMV, including vaccinating kids to protect the adults. But does that justify giving an adjuvant for which we have scant data on large nonelderly adult populations, let alone kids and babies?
Here's the chart showing the reactions of these kids.
And the remarks from Podda's paper
In this toddler trial, the CMV bG/MF59 vaccine was well-tolerated by all children and no significant systemic reactions occurred after administration of any of the three doses although transient local reactions were more frequent after the third dose.As with all the other studies, they only followed the kids for 7 days.
The rest of this series can be found via these links here part I, part II, part III, and part V.