Professor Stephen Bustin schools Andrew Wakefield, and I enjoy

I suppose that while I’m on another roll writing about the antivaccine movement I should just embrace it. I was going to start this post out again with one of my periodic laments about how blogging about the antivaccine movement has taken over and crowded out other topics that I like to write about. I realize it’s become one of my go-to cliche beginnings, to the point where I sometimes feel lazy when I use it. It is, however, an honest sentiment, and I hide nothing with respect to my opinion of the antivaccine movement and how it endangers public health through the promotion of pseudoscience and quackery. However, I can’t resist the topic of this post because it combines a discussion of a key figure in the antivaccine movement with a topic that I’m deeply familiar with. It also involves something that I first discussed way back in 2007, and, really, how often does a blogger get to “close the loop,” so to speak, on a topic he first wrote about six years ago?

I’m talking about the hero of the antivaccine movement, particularly in the U.K., namely Andrew Wakefield. Of course, I didn’t first write about Wakefield in 2007. My interest in his pseudoscience dates back to before that. However, it was in 2007 that I first read the masterful deconstruction of some of Andrew Wakefield’s work by a world expert whose work I referenced frequently and impressed me greatly. It was done in the context of the Autism Omnibus, a huge court proceeding in which the Vaccine Court tried to deal with the claims of so many parents who claimed their children’s autism was caused by vaccines by examining a few test cases, meant to be the best cases the plaintiffs could come up with, to see if they would fly. They didn’t, and in one of the cases, Stephen Bustin was the reason why. Professor Bustin is a world expert in the technique known as the polymerase chain reaction (PCR). PCR is a technique whereby small sequences of DNA can be specifically amplified, so that very tiny amounts of DNA can be detected. Done properly, PCR is very, very sensitive and can be very, very specific. As you might expect, Andrew Wakefield did not do it correctly. Indeed, PCR seems to be a crank magnet. Indeed, just in January, Sin Han Lee was at it again abusing PCR to claim that the dreaded HPV DNA from Gardasil is killing young girls.

But back to Andrew Wakefield’s use and abuse of PCR.

As Matt Carey points out, although Wakefield’s infamous 1998 case series in The Lancet that was ultimately retracted due to his scientific misconduct is his most famous publication, it was actually a later publication that seemed to produce the most convincing evidence that there was a link between the MMR vaccine and autism, finding vaccine-derived measles in the intestinal mucosa of autistic children. The PCR tests to detect the measles virus DNA sequences were performed by John O’Leary in his laboratory in Dublin. As part of litigation over MMR in the U.K., Stephen Bustin was hired by vaccine manufacturers to review the methods in the O’Leary lab. Although these results were not initially revealed, Dr. Bustin testified about them in the Autism Omnibus proceedings, which I covered extensively in 2007, using Professor Bustin’s testimony to limn the difference between real scientists and crank scientists. Now, he’s followed up that testimony with a publication in which he describes exactly why there is no link between measles virus and autism. The entire paper is worth reading in its entirety, but I will, as is my wont, discuss key sections, but first let’s summarize what Bustin found wrong with Wekefield’s paper:

Absence of transparency: the key publication shows no data; hence an expert reader cannot evaluate the reliability of its conclusions

Unreliable techniques and protocols: analysis of the qPCR data was incorrect

Disregard for controls: obvious evidence of extensive contamination was disregarded

Lack of reproducibility: the data could not be duplicated by several independent investigators

His conclusion was:

The only conclusion possible is that the assays were detecting contaminating DNA. Since MeV is an RNA-only virus and never exists in DNA form, these data must be ignored and it it is my opinion that the authors should withdraw this publication from the peer-reviewed literature.

Let’s take a look at Professor Bustin’s analysis of the “fruit” of Wakefield and O’Leary’s labor, a 2002 paper published in Molecular Pathology entitled Potential viral pathogenic mechanism for new variant inflammatory bowel disease. This is a paper that looked at biopsies of the terminal ileum of children with gut pathology and controls without such pathology, reporting that 75/91 children with histologically confirmed diagnosis of ileal lymphonodular hyperplasia and enterocolitis were positive for measles virus in their intestinal tissue compared with five of 70 control patients. They even estimated the number of copies of measles virus to be between 1 and 300,000 copies per ng, concluding that “the data confirm an association between the presence of measles virus and gut pathology in children with developmental disorder.”

There were, Bustin reports, many problems. First of all, are problems with the samples. For instance:

RNA was extracted from fresh frozen samples as well as formalin fixed, paraffin-embedded tissue (FFPE). However, there is no information on how the fresh samples were frozen, how long they had been stored,what percentage of patient and control samples were fresh frozen or FFPE and whether the same percentage was in each category. This is essential, since it is well established that FFPE treatment modifies and destroys RNA, or in Prof O’Leary’s own words “wax and fixation by itself breaks down RNA” [5]. Hence it was well known at that time that RNA-derived data obtained from FFPE samples must be analysed and interpreted with caution [12-14].

This is an obvious issue. If you are trying to use quantitative reverse transcriptase PCR (QRT-PCR) to quantify RNA transcripts, it can make a huge difference where the tissue came from and how it was handled. Fresh, flash-frozen tissue is best, and that’s what I’ve always tried to use in my laboratory whenever possible. It’s well known that RNA in tissue is pretty unstable and starts to degrade almost immediately as soon as it’s removed from its blood supply. When harvesting tumors from animals, for instance, I routinely drop them in liquid nitrogen as fast as possible in order to minimize variability. For human tissue, it’s usually not possible to do get the tissue frozen that fast, but it’s still doable. It is possible to detect RNA transcripts from FFPE tissue. It’s even possible to do cDNA microarray experiments using such specimens (“gene chip” experiments that measure thousands of genes at once) or even next generation sequencing (techniques I’ve mentioned before that allow the sequencing of every RNA, coding and noncoding) in the but the RNA will inevitably be degraded, and special techniques are needed to control for that. There’s no evidence that O’Leary did any of that, and, of course, next generation sequencing techniques didn’t exist in 2001, and cDNA microarray technology was in its infancy. So, while it’s true that these days we have found that RNA can be a bit more robust that doesn’t exonerate O’Leary in any way, particularly given the RNA technology that existed at the time.

Perhaps the most glaring issue I see, however, is this:

b. No information is provided with respect to quantification or quality assessment of the
extracted RNA; indeed there is no mention of RNA quality. This is vital information
needed to assess the validity of any quantitative or negative result [15].

“Vital” is putting it mildly. In any paper involving QRT-PCR, it’s absolutely essential to verify the quality of the RNA. It makes me wonder how this paper was ever accepted (much like nearly all Wakefield MMR papers, actually). Whenever I review a paper involving PCR, I always look carefully at measures of the quality of the RNA. Heck, if you want to go old school, you could even insist on an old-fashioned formaldehyde-agarose gel showing that the 28S and 18S ribosomal bands are intact! (Yes, to the scientists out there, I know that this is incredibly old-fashioned.) Not to do anything, be it old-fashioned or using newer technologies, to assess the quality of the RNA is completely unacceptable. This is particularly true, given that Bustin points out how QRT-PCR results are summarized without any actual data as either “positive” or “negative.”

Also damning is the observation by Bustin about the control transcripts examined. In most QRT-PCR, the level of the gene of interest, the gene that is expected to vary, is normalized to the level of a “housekeeping gene” (or, increasingly, multiple housekeeping genes) whose level does not vary much (preferably not at all) under the experimenta conditions being studied. Often a gene known as GAPDH is used. Sometimes β-actin is used. In this case, Bustin noted that the levels of GAPDH were much lower as detected in FFPE than in fresh frozen tissue, which as expected. However, levels of the measles F gene were identical.

Bustin writes:

Since any RNA present during formalin fixation would have been affected in an identical manner, the obvious implication of these results is that whilst the control RNA was indeed present prior to formalin fixation and so was degraded, the MeV target was not degraded and entered the sample after formalin fixation. Consequently, no MeV RNA can have been present in the tissue and the positive result must have been caused by a contaminant.

This also comes into play if there were large differences between the proportions of FFPE samples and fresh frozen specimens used in the control group versus the experimental group. of course, none of that matters if what is being detected is a contaminant. Supporting that what O’Leary detected was nothing more than contaminants, is another observation. In QRT-PCR, as I explained before, the RNA first has to be turned into DNA before amplification. This is the reverse transcriptase (RT) step, named after the enzyme used to do the step. In fact, we frequently use what we call “no RT controls” in which we do a PCR reaction on the sample without its having been subjected to reverse transcription because amplification of a signal under such conditions means that DNA is being amplified, not RNA.

Guess what happened when Professor Bustin looked at just that question:

Two tests accidentally omitted including the RT step before the PCR test. In the case of the control, the results are as expected: the assay works significantly less well (Figure 5A). This is because Taq polymerase is very inefficient at making DNA from RNA. In contrast, the four MeV samples tested give the same result, regardless, indicating that the test is detecting DNA (Figure 5B). Since MeV does not exist as DNA, the test is not detecting MeV but a DNA contaminant.

Here’s a general hint about PCR: If you get the same results no matter what, chances are, you’re dealing with a contamination problem. There are also many other reasons to suspect that what O’Leary was measuring was contamination, reasons that Bustin enumerated and I discussed in that post six years ago. For these reasons, Bustin correctly concludes that the results of Wakefield’s 2002 paper should be completely discounted. It’s true that Bustin’s new analysis doesn’t change anything in that Wakefield’s results have been completely discredited for several years now.

Amusingly, Wakefield is in a pugnacious mood and has shown up on the antivaccine crank blog Age of Autism. It turns out that Bustin’s criticism isn’t the only flak he’s been taking. A recent episode BBC’s Newsnight looked at the current outbreak of measles in the U.K. due to low vaccine uptake rates. Wakefield was, of course, given considerable blame. Now Bustin piles on. And how does Wakefield respond? He blames the government for the measles outbreaks because it didn’t believe his claims that vaccines can cause autism and wouldn’t allow parents to use the single dose measles vaccine. The chutzpah boggles the mind!

But Wakefield’s just getting warmed up. After working himself into a fine lather, he builds up to a climax worthy of his level of crankitude. Basically, like any good crank, he challenges his enemies to…a live public debate:

The more light that shone on this subject by way of informed, balanced debate, the better. I am offering to debate any serious challenger on MMR vaccine safety and the role of MMR in autism, live, in public, and televised.

What is it with cranks and live public debates? Of course, I know. Live public debates allow cranks to do the Gish gallop to their heart’s content, flitting hither and thither and yon among cherry picked and/or bogus studies to their heart’s contant, like a demented bee pollinating a field of pseudoscience.

Somehow, I bet that if Professor Bustin were to take Wakefield up on his offer, Wakefield would find a way to politely decline.