THE BEST OF TIMES
It was the best of times (for antivaccinationists). It was the worst of times (for antivaccinationists). It was the age of wisdom (definitely not for antivaccinationists). It was the age of foolishness (definitely for antivaccinationists). It was the epoch of belief (for antivaccinationists).
Such is the time we live in, my apologies to , even though he is long dead.
Let’s face it. If we ignore the science, it is, alas, indeed the best of times for antivaccinationists. I’ve lamented the rise of non-science-based fearmongering among the antivaccine brigade before many times. Indeed, I’ve lamented how the influence of ignorant, unscientific dolts like Jenny McCarthy spouting nonsense about vaccines has already resulted in the resurgence of vaccine-preventable diseases such as measles in areas of the U.S. to the point where I’m not along in fearing that the bad old days will soon return, just as Andrew Wakefield’s “studies” suggesting that the MMR was somehow responsible for autism and GI problems linked with autism resulted in the measles going from being conquered in the U.K. 14 years ago to , all because of the fear stoked in parents by bad science, paranoia, and anti-vaccine fearmongering.
Truly, it is the best of times for antivaccinationists, or so it seems from a superficial view.
THE WORST OF TIMES
For all the success antivaccinationists have had on the propaganda and public relations front, however, all is not well. If one concentrates solely on the science and nothing but the science, this golden age of antivaccine activism is also without a doubt the worst of times for antivaccinationists. Study after study over the last two years have provided strong evidence that vaccines do not cause autism or any of the other dreaded consequences attributed to them by antivaccinationists. For example, numerous studies have failed to find even a hint of a correlation between the most evil (in the minds of antivaccinationists) preservative used in some vaccines, thimerosal, and autism. Indeed, have failed to find a link between vaccines in general and autism or . In fact, the drumbeat of studies coming out that fail to find a link between either thimerosal-containing vaccines and autism or vaccines in general and autism (or, for that matter, vaccines and the dire consequences attributed to them by anti-vaccine activists) is becoming so routine as to be monotonous.
It’s a monotony that I like–and that you should like too.
You’d think that the pseudoscientists who are so utterly convinced that it absolutely, positively has to be the vaccines causing autism (and that if it isn’t vaccines causing autism they must be causing some horrible chronic health problem or other) would have gotten the message by now. You would, of course, be wrong, because to them science is besides the point. It doesn’t matter that the pace of these studies appears to be accelerating. Somehow, some way, no matter how many studies fail to find a link between vaccines and autism or other complications, antivaccine activists would have us believe that nonetheless vaccines must be the Root of All Evil, or at least all autism.
YET ANOTHER NAIL IN THE COFFIN
Last Thursday, yet another study failing to find a link between MMR and autism, this time out of Columbia University, was released. Even better, the first author on the study is Mady Hornig, who became a hero of the mercury militia a three years ago with her , in which she performed experiments of dubious ethics on some very unlucky lab mice and proclaimed the results “evidence” that thimerosal caused autism. She was even a common speaker at mercury militia–excuse me, I mean National Autism Association (NAA) –events, along with hard core antivaccinationists Boyd Haley, Richard Deth, Mark Blaxill, Jeff Bradstreet, and others, as well a , along with Mark and David Geier, Thomas Burbacher, Boyd Haley (again), Andrew Wakefield himself, and many others. The paper reporting her study is even , as well as . Unfortunately, , just as these endorsements imply.
This latest study makes me think that perhaps Dr. Hornig, contemplating her descent into antivaccine pseudoscience, may have had a “come to Jesus” moment and is trying to regain her scientific mojo (although I’ll only really believe that when I see her give up trying to show that thimerosal causes autism using the aforementioned “rain mouse” model). Either that, or she figured out that flirting with the band of pseudoscientists and ideologues making up the antivaccine movement is the surest way to kill her scientific career deader than dead and thereby end up working in the basement of her house like Geier, père et fils. In other words, perhaps she decided that tenure and a continued scientific career meant more to her than the momentary notoriety and adulation that came her way from her brief stint as one of the darlings of the mercury militia.
Let’s hope so, anyway.
Whatever the case, Dr. Hornig has done a study with W. Ian Lipkin as the senior author published in PLoS ONE entitled, . Before I get to the study itself, let’s look at the :
CHICAGO (Reuters) – Scientists who tried to replicate a study that once tied a measles vaccine with autism said Wednesday they could not find any link and hope their study will encourage parents to vaccinate their children to combat a rash of measles outbreaks.
Parents’ refusals to have their children vaccinated against measles have contributed to the highest numbers of cases seen in in the United States and parts of Europe in many years.
Measles kills about 250,000 people a year globally, mostly children in poor nations.
I love it when a mainstream news article lays out the price of antivaccination pseudoscience right from the beginning. The beauty of this study is that it tried to replicate the original study done by Andrew Wakefield. As you may recall, ten years ago, Wakefield published a describing a series of children with regressive autism and gastrointestinal complaints. He described intestinal abnormalities, including reactive lymphoid hyperplasia in the ileum, and the parents of several of these children reported the onset of symptoms temporally related to the MMR. This led to the hypothesis that the MMR was associated with autism and GI complaints, and what later followed were that purported to find measles RNA in the gut of autistic children. These results were spun by antivaccine advocates as “evidence” that the MMR vaccine was associated with gut abnormalities and autism.
Thus was born the MMR scare in the U.K., a scare that, a decade later, has resulted in the resurgence of the measles in there to the point that it has a mere 14 years after having been declared under control. Sadly, he had lots of help from the press, which published credulous inanities about this link, regardless of how shoddy the science was (and ).
Basically, the design of the study was pretty simple, and, better yet, it was an intentional attempt to replicate the research of Andrew Wakefield using rigorous methodology. The design was a case control study in which children with autism and GI problems were compared to children with GI problems and no autism for a number of parameters, including MMR vaccination and the presence of measles virus RNA in the gut of children who underwent endoscopy. The overall approach was to determine whether children with autism and GI symptoms were more likely to have detectable measles virus RNA in the gut. The result?
Zero correlation. Nada. Zip.
One out of the 25 autistic children in the study had detectable measles virus RNA present in their biopsies, while one out of the 13 children without autism also had detectable measles virus RNA present. There was not even a hint of a statistically significant difference between the groups, nor was there a correlation between the timing of the MMR vaccination and the onset of autism or GI symptoms. Basically, this study was negative, and represents some pretty strong evidence against the contention that MMR causes autism or chronic GI problems. So strong was the evidence that even committed antivaccine zealot Rick Rollens grudgingly admitted it in a :
“No longer can mainstream medicine ignore parents’ claims of clinically significant GI distress,” said Rick Rollens, a parent and autism research advocate.
He commended the researchers for their work but said, “This study by itself does not exonerate the role of all vaccines.”
Sound familiar? It’s just like the thimerosal shuffle. As studies come out exonerating thimerosal as a cause of autism, antivaccinationists invoke other “toxins,” sometimes going to ridiculous contortions to do it. Because, you know, it’s absolutely, postively, got to be something in the vaccines. It’s the same thing here. The MMR has been exonerated as a cause of autism, but Rollens knows that it’s just absolutely, positively got to be some vaccine or other that causes autism, even if it’s not the MMR.
One observation that came out of this study is that there does appear to be a correlation between GI complaints and regressive autism. Neither are associated with the MMR, but they are associated with each other. This is certainly an observation that should be followed up with further study; what this study contributes is that it frees researchers to stop wasting precious research dollars studying the hypothesis that MMR causes autism. It doesn’t. Between a previous study by and the current one, among others, the science is quite clear that there is no link between MMR and autism.
THE “EXONERATION” OF ANDREW WAKEFIELD? NOT!
The reaction to this study by the antivaccine contingent is instructive. There are two main reactions, the ridiculous and the even more ridiculous. To understand just the ridiculous, you first need to know that this study had one very interesting aspect to its design. All the samples were sent to three different laboratories for analysis, completely blinded so that the labs didn’t know which experimental group each patient was in, and tests of concordance were done, to verify that the three labs agreed. Prior to the beginning of the study, the three labs were tested with positive and negative controls. So what? You might reasonably ask. Let’s go back to a discussion of the Autism Omnibus last year, where polymerase chain reaction (PCR) expert Stephen Bustin examined John O’Leary’s laboratory, where the the reverse transcriptase PCR (RT-PCR) was performed to detect measles RNA. Bustin demolished the methodology of that laboratory demonstrating conclusively that the “postives” detected were all false positives made possible by , as .
The O’Leary laboratory was one of the three labs used in this study, and Dr. O’Leary was one of the authors.
Let’s put it this way. The day the study came out, a little birdie told me that himself was one of the reporters who called into a teleconference about this study held the day before its release and that he asked a lot of questions about the O’Leary lab. (Note added after publication: The press conference can be found .) I’ve also heard rumblings from the antivaccination underground that the talking point will be that this study is a “two-edged sword” in that it supposedly the fact that the O’Leary lab performed competently now somehow exonerates its horrible performance several years ago. It doesn’t. of the methodology in use at the O’Leary lab at the time was devastating, demonstrating a level of incompetence beyond belief. What is far more likely is that O’Leary was forced to clean up his act and tighten up lab discipline in order to be included as one of the labs in this study. The same little birdie also told me that Kirby asked questions about the case, which is in essence a non sequitur, because part of the claim in her case is that thimerosal was contributory to autim, and there has never been any thimerosal in the MMR. However, apparently Cedillo’s biopsy samples were also processed by Dr. O’Leary’s lab, and Kirby seems to think that maybe this study could therefore be used to support her test case in the Autism Omnibus. Again, don’t fall for that line; it’s virtually certain that Dr. O’Leary must have cleaned up his act. Remember, his lab was tested before it was allowed to process real patient samples from this study.
While David Kirby has not yet posted one of his typical obfuscation-laden feats of verbal prestidigitaion to criticize this study, that doesn’t mean that other anti-vaccine advocates haven’t jumped into the ring. Perhaps Kirby realizes just how silly such criticisms sound from a scientific standpoint and has decided for once to stay silent. Unfortunately, others are more than happy to leap into the fray to embarrass themselves, including Andrew Wakefield himself.
THE FASTEST WITH THE LEASTEST
Predictably, the Wendy Fournier of the National Autism Society was first off the mark with her , and a truly laughable bunch of nonsense it is. Amusingly, she starts out by calling this a “CDC study.” Actually, it’s a study out of Columbia University, Massachusetts General Hospital and Harvard University, Trinity College in Dublin, and the Measles, Mumps, Rubella, and Herpesvirus Laboratory and the Branch National Center for Immunization and Respiratory Diseases at the Centers for Disease Control, and its funding came from CDC grant U50 CCU522351 to the AAP and by National Institutes of Health awards AI57158 (Northeast Biodefense Center-Lipkin), HL083850, and NS47537. Predictably, Fournier wants to make this sound like a conspiracy of some sort or imply conflict of interest. Of course, she neglects the fact that investigators Mady Hornig and John O’Leary, among others, were until this study heroes of the antivaccine movement. Some “conspiracy”!
Fournier’s “complaints” are predictably lame:
The CDC study was designed to detect persistent measles virus in autistic children with GI problems. The assumption being if there is no measles virus at the long delayed time of biopsy, there is no link between autism and MMR. But NAA says this underlying assumption is wrong. The questions should have been: Do normally developing children meeting all milestones have an MMR shot, develop GI problems and then regress into autism? Do they have evidence of measles and disease in their colons compared to non-vaccinated age and sex matched controls
Point one: The study didn’t just look at the presence of measles virus in the gut. It also looked at the timing of MMR vaccination compared to a number of health parameters. Indeed, the authors analyzed not just whether MMR preceded symptoms but did statistical modeling of how long the interval between vaccination and symptoms was. The answer was that there was no correlation. Of course, Fournier’s whine is nothing more than an example of the old crank adage: “If you don’t like the answer, change the question.” This is, in essence, nothing more than a form of moving the goalposts.
Point two: Does Fournier know how impractical her demand for checking non-vaccinated age- and sex-matched controls is? She clearly knows nothing about clinical research or medical ethics. Colonoscopy is an invasive procedure. Not only does it require general anaesthesia in small children, but there is a small but real risk of serious complications. It’s not a procedure to be done without a firm indication, which is what the investigators did. They selected children for whom colonoscopy was indicated based on their clinical presentation and symptoms. Doctors can’t just go around doing colonoscopies and gut biopsies on “control” children. In addition, even if she meant that the controls with GI problems should have been free of exposure to the MMR, that is a methodological obstacle that would be very difficult to overcome, given how few children of that age have not received the MMR, and if ths hypothesis being studied is that MMR is causative for autism, autistic enterocolitis, or enterocolitis itself, looking for correlation between receiving MMR before the onset of symptoms is a valid methodology to pursue to test that hypothesis. Of course, an understanding of medical ethics was never the strong suit of antivaccinationists.
Neither are statistics, as Fournier demonstrates:
In the current CDC study, only a small subgroup of children was the correct phenotype to study. From page 7, “Only 5 of 25 subjects (20%) had received MMR before the onset of GI complaints and had also had onset of GI episodes before the onset of AUT (P=0.03).” The other 20 autistic children in the study had GI problems but the pathology developed before the MMR vaccine. Additionally, the controls all received the MMR vaccine and had gastrointestinal symptoms. The controls should have been free of exposure to vaccine measles in order to make a comparison relevant for purposes of causation.
This analysis was the one of the major points of the study, the point being that there was no difference, statistically speaking, between the number of children in the autism GI complaints and the GI complaints alone group, indicating that there was no association between having the MMR vaccination before the onset of symptoms and the development of symptoms. As for her complaint that all the controls received MMR and had GI symptoms, see my previous response. It applies here, too. The only grain of truth is that this study wasn’t all that large. It did, however, include four times the number of patients that Wakefield’s original study did, and it was far, far more rigorous in design and execution, including excellent blinding. Also, for reasons mentioned before, no study of this type is ever likely to achieve high numbers because it requires an invasive procedure. In any case, the study had more than adequate power for the questions answered and accrued enough patients to have highly statistically significant results.
The weakest of all, however, is this:
Inflammatory bowel disease in the absence of MMR RNA does not mean that MMR shot didn’t precipitate the GI disease and didn’t precipitate autism. A similar example would be rheumatic fever where the infection is cleared quickly but damage to the heart and/or brain last a lifetime.
Talk about moving the aforementioned goalposts! Remember, the hypothesis being tested was Andrew Wakefield’s hypothesis, which was that somehow the persistence of the measles virus in the guts of children vaccinated with the MMR live virus vaccine contributed somehow to autism and “autistic enterocolitis,” not that the MMR contributed to autism or autistic enterocolitis via some sort of rheumatic fever-like mechanism. The concept was that persistent measles virus in the gut somehow resulted in inflammation and increased gut permeability. Now that this study has provided strong evidence falsifying that hypothesis, antivaccinationists like Fournier are moving the goalposts yet again. If it’s not the persistence of the measles virus from vaccines in the gut that causes autism, then it must be some other mechanism, because to the antivaccinationist it absolutely, positively has to be the vaccines. They just know it, and no amount of evidence will ever sway them to think otherwise.
BUT WHAT DOES ANDREW WAKEFIELD THINK OF ALL THIS?
Of course, given that the Hornig et al study is the proverbial shot across the bow of Andrew Wakefield’s S.S. MMR-Autism, it would be shocking if the grandfather of MMR fearmongering himself didn’t take offense and decide to fire off a shot back himself, because, once again, to the antivaccinationist, it absolutely, positively, has to be the vaccines that somehow, some way cause or contribute to autism. It just has to be, don’t you know? is exactly as predicted in that he tries to spin this study as an exoneration of his work. (Brian Deer has done a nice description of Thoughtful House .) But first Wakefield pulls out the false equivalence gambit:
A study published yesterday in the Public Library of Science One (PLOS1), an on-line journal, failed to find evidence of measles virus in the intestinal tissue of 24 children with autistic regression and gastrointestinal symptoms. The findings contrast with those published in 2002 in which researchers from Ireland and the UK found measles in 75 of 91 biopsies from autistic children with GI inflammation, and in only 5 of 70 samples from non-autistic children(1). The children with autism in the 2002 study developed gastrointestinal symptoms and autistic regression after the MMR vaccine.
Unfortunately, for Wakefield, the reason Hornig et al‘s results contrast so markedly with his in his 2002 followup to his infamous 1998 Lancet study is because Dr. O’Leary’s Unigenetics Laboratory was so incompetent, its facilities so contaminated with plasmid containing measles virus sequences, that virtually all of its positives were false positives. In this, the during is quite . Back in 2002, no results derived from the polymerase chain reaction in Dr. O’Leary’s laboratory could be trusted. None. Just because sometime in the six year interim since then Dr. O’Leary has apparently cleaned up his lab’s act and his lab can do the PCR assay for measles now does not exonerate it for its sloppiness and incompetence in 2002. Not that Dr. Wakefield doesn’t try to spin it exactly that way in his first of three points:
In the study published yesterday, conducted by three independent laboratories, only 5 of the 25 children developed these symptoms after the MMR vaccine and therefore, only these five are comparable to the 2002 study. This new study confirmed that results from the laboratory of Professor John O’Leary (one of the collaborators on the new study, and senior author of the 2002 study) were correct, and identical to the results obtained by the laboratories of the Centers for Disease Control and Prevention (CDC) and Dr. Ian Lipkin of Columbia University.
No, the results confirmed nothing, and Wakefield is doing nothing more than disingenuously trying to make the study’s numbers sound lower by zeroing in on one subgroup of subjects. Never mind that the whole point was that the same percentages of childen in each group developed symptoms after the MMR vaccine, which was a major finding of the study. Nonetheless, Dr. Wakefield forges on:
In that this new study affirms the reliability of Professor O’Leary’s laboratory and therefore of his previous findings, a major impact upon the current hearings in vaccine court is likely, wherein the government’s defense relies largely on the claim that Professor O’Leary’s finding of measles in the intestinal biopsy of Michelle Cedillo (a child with severe autism and epilepsy) was unreliable. The historical reliability of the measles assay used in Professor O’Leary’s laboratory is now confirmed.
I’ll give Wakefield credit. He’s slick. He also reveals a clever, albeit obviously desperate gambit to try to make the best of a very bad situation for him. It won’t work, though. As points out, Michelle Cedillo’s samples were run in 2002 and Stephen Bustin didn’t inspect Dr. O’Leary’s lab until 2004. As I’ll repeat again: Just because Dr. O’Leary apparently can get it right now does not exonerate him for his lab’s ineptitude prior to 2004. According to the methodology of Hornig et al, before any patient samples from the study were tested for the presence of vaccine strain measles sequences, all three laboratories correctly identified the true negative and true positive samples sent to them for analysis as a test of their methodology. If Dr. O’Leary hadn’t cleaned up his lab’s act, it is virtually certain that his lab would have produced some false positives. Wakefield seems incapable of realizing this, or at least he hopes his readers won’t. Predictably, some of them, like longtime antivaccine activist and the crew at that repository of antivaccine propaganda and autism quackery, , lapped it up anyway.
Wakefield’s second point is an argument that the linkage between MMR and autism is an open question. It isn’t, at least not anymore. Hornig et al is only the latest of a long line of studies over the last several years that have failed to find a link. It is also the latest of attempts to replicate his results. All have failed, other than Wakefield’s group or groups affiliated with Wakefield. Add all of that to the evidence for incompetence of the laboratory where the biopsy specimens from his studies were analyzed by PCR, and in serious scientific circles the hypothesis that MMR causes or triggers autism or autistic enterocolitis is about as dead as a hypothesis gets. The entire continued kerfluffle over the MMR vaccine is nothing more than a .
Finally, Wakefield concludes with one of the most blatant examples of rewriting history and shifting the goalposts I’ve seen in a long time:
Dr. Wakefield comments, “The search for the ‘footprints’ of measles virus in the intestine is merited, based upon the previous findings and the intestinal disease that is commonly found in these children. This new study rules out only one possibility – that the measles virus must remain for the long term in the intestine. We need to consider that the MMR vaccine can cause autism as a hit-and-run injury, but not necessarily leave the measles virus behind.”
Time travel back about six years to the conclusion of the infamous , which was the followup to Wakefield’s original in which PCR for measles virus was first described:
We describe an association between persistent MV infection and ileocolonic lymphonodular hyperplasia and ileocolitis in children with developmental disorder.1 The molecular data indicate the presence of MV genomes in 75 of 91 affected children with the disorder compared with five of 70 control children.
Wakefield also then went on to make an analogy between persistent measles virus infection as an immunological trigger for enterocolitis with autism and how persistent HIV infection causes AIDS. As recently as February, in response to the last study failing to find a link between MMR and autism, Wakefield was still ; i.e., that a persistent measles infection was the trigger. Even if Wakefield’s original idea were that MMR was a “hit-and-run” injury, the Hornig et al study is still strong evidence to refute such a concept, as pointed out in its conclusion:
ASDs comprise a wide range of endophenotypes that may represent different routes to pathogenesis. The work reported here eliminates the remaining support for the hypothesis that ASD with GI complaints is related to MMR exposure. We found no relationship between the timing of MMR and the onset of either GI complaints or autism. We also could not confirm previous work linking the presence of MV RNA in GI tract to ASD with GI complaints.
In other words, this study was about more than whether measles virus could be detected in the guts of autistic children with GI complaints. It also doesn’t absolve Andrew Wakefield for his , , shoddy research.
THE BEST OR WORST OF TIMES?
I realize that I wrote a lot about this particular study. It could be argued that I engaged in massive overkill, but it’s overkill with a point. (Besides, overkill is what I do best, almost as well as my co-blogger Kimball Atwood.) This study was even more important than typical studies over the last few years that have failed to find a correlation between the MMR vaccine and autism for one simple reason: It was explicitly designed to attempt to replicate Andrew Wakefield’s results. It was also highly rigorous in both looking at any potential correlation between MMR and gastrointestinal symptoms or MMR and autism in children with gastrointestional complaints. Not only was no correlation found between epidemiological measures, but Hornig et al found no differences between case and control groups in the presence of measles virus RNA in the colon or terminal ileum. Although the study is relatively small, it is sound. When added to our previous body of knowledge and previous studies, it’s one more nail in the coffin of the MMR-autism hypothesis.
None of that, of course, matters to anti-vaccine activists, because science is irrelevant to them. It exists only to be either abused or cherry-picked to serve their preexisting fervent belief that vaccines do far more harm than good or, in extreme cases, represent a conspiracy to control the human population. (.) They dismiss all studies that fail to find a link between vaccines and autism or other conditions, or, failing that, move the goalposts, so utterly convinced are they that vaccines are a plot to poison our children. And they’ve gotten good at it. However, that wasn’t enough. It took a far more sustained propaganda effort with a catchy slogan, the good fortune to have sucked a not-too-bright D-list celebrity into their orbit and get her to organize and front marches on Washington. From a propaganda standpoint they’re flying high. However, from a scientific standpoint they haven’t been any lower. As study after study comes out failing to find a link between vaccines and autism or all the horrible other consequences blamed on them, it’s getting harder and harder for them to make a case that isn’t obviously based on blatant pseudoscience.
As vaccination rates are threatened and vaccine-preventable diseases come roaring back, let us all fervently hope that times get as bad for antivaccinationists from a propaganda standpoint as they are from a scientific standpoint. If they don’t, our children will be in for a world of hurt, and so will we.
1. Mady Hornig, Thomas Briese, Timothy Buie, Margaret L. Bauman, Gregory Lauwers, Ulrike Siemetzki, Kimberly Hummel, Paul A. Rota, William J. Bellini, John J. O’Leary, Orla Sheils, Errol Alden, Larry Pickering, W. Ian Lipkin, Mark R. Cookson (2008). Lack of Association between Measles Virus Vaccine and Autism with Enteropathy: A Case-Control Study PLoS ONE, 3 (9), 1-8 DOI: