Rebuttal to Wakefield Called 'Weak' with 'Distortions': Shattock
January 26, 2001
This Wednesday we posted a report from Mike Watson in the UK which
purports to correct items in the recently released Andrew Wakefield study.
The study challenges the safety of the MMR vaccine. "A Rebuttal to
Wakefield's "Measles, Mumps, Rubella Vaccine. . ." was the newsletter item.
Paul Shattock, the Director of the Autism Research Unit, University of
Sunderland, writes in to report his reactions and counter-rebuttals to what
he asserts are specious and even deceitful arguments by rebuttal author Mick
Watson. Watson is the Medical Director of Aventis Pasteur MSD.
"For us it is always a pleasure when the opposition shows what it has
in terms of arguments, remarks Shattock as he removes his gloves, "it is
even better when their case is so palpably weak that they must resort to
distortions of fact that are easily detectable to even a casual reader."
Below are quotations from the Wakefield study followed by Watson's
rebuttals. Following that are the counter-rebuttal comments from Shattock.
Wakefield: The first thing to note is that these were short-term
safety studies with periods of observation lasting at most 28 days, and
often considerably less.
Watson: The observation period in the study that the authors
concentrate on (Stokes et al, 1971) is clearly stated in the first page of
the paper as being "six to nine weeks". That is up to 63 days, rather than
No parent whose child has been involved in a clinical study will
forget this. Indeed, a sticker is often placed on the front of the child's
notes to ensure that everyone is aware of this. If the parent or a
healthcare professional perceive there have been any unusual problems in a
clinical trial subject, they are extremely likely to report it.
Shattock: No idea what this sticker business is about but let
us quote the paper:
The abstract states, "Joint involvement was notably absent during
the six to nine week follow-up".
On the final page comes the main reference to this element of the
"The present studies with queries at six to nine weeks following
vaccination did not reveal any occurrence of arthritis or arthralgia
beyond the 28 day period for close observation."
It is so abundantly clear that the trial was for 28 days. No data
are provided for anything over that save this particular feature
which (quite correctly) was sought by the researchers.
Wakefield: Prior to its licensing in 1975, trials of combined MMR
vaccine safety were the subject of two relatively small-scale controlled
Watson: MMRŽ (1st generation, Merck) was licensed in the US in 1971
and in the UK in 1972.
Safety of MMR vaccine (from one manufacturer alone - Merck) was
investigated in four studies prior to licensure. In addition, seven studies
(two of which are published) investigated the safety of MMRŽ II prior to its
licensure - these are not even mentioned in the paper.
The safety of another MMR vaccine, ImmravaxŽ, was studied in seven
studies prior to its licensure in the UK.
Shattock: I am positive beyond any doubt whatsoever that Dr Wakefield
was absolutely assiduous in seeking data. There would have been no
point in any paper that did otherwise. It is unclear whether these
studies were published or retained in secret files which are not
available. We would greatly appreciate the published citations so that
any errors can be drawn to the attention of the authors.
Wakefield: The potential for delayed intestinal pathology is borne out
by Fournier et al's demonstration of persistent measles virus infection of
the diseased appendix in 1968.
Watson: Virologists generally accept that wild measles virus only
causes persistent disease in the central nervous system - subacute
sclerosing panencephalitis (SSPE) and measles inclusion body encephalitis
(MIBE). MMR has been demonstrated to protect against these diseases.
No group has ever shown reproducibly by accepted methods that measles
virus, either wild or vaccine-derived, causes persistent GI infection in
children with autism. This includes Wakefield's own group (Chadwick N, Bruce
IJ, Schepelmann S, Pounder RE, Wakefield AJ. Measles virus RNA is not
detected in inflammatory bowel disease using hybrid capture and reverse
transcription followed by the polymerase chain reaction. J Med Virol 1998;
Shattock: Hmm. Strange that there appears to be an SSPE death related
to the MMR a couple of weeks back. I concede that the Wakefield team,
in the paper quoted above, failed to isolate the measles virus RNA.
The method used was not sufficiently sensitive and would require a
huge infection to show up. Fortunately, it now appears that adequate
techniques such as those employed by the Japanese and by O'Leary have
remedied the situation. The Japanese have published.
The principal of publishing results, even though they fail to
support your case, must be unusual in a commercial or non
investigative laboratory. Dr Wakefield and his colleagues have
been much derided by opponents for doing this. The principle is
called "integrity". It is an alien concept in Departments of
Health and certain areas of commerce. This is the area where
science is lost to political or economic requirements.
Watson: However, it is thought that mutant measles virus genetic
material can persist in the tissues of apparently healthy people without
causing disease [Katayama Y, Kohso K, Nishimura A, Tatsuno Y, Homma M,
Hotta H. Detection of measles virus mRNA from autopsied human tissues.
J Clin Microbiol 1998; 36(1): 299-301].
Shattock: So "mutant" measles can persist but Moraten and Schwarz
strains cannot? Those would be interesting references (please).
Watson: Background rates of gastroenteritis in Philadelphia and
Costa Rica-San Salvador.
These data are presented as 5.6% and 44% respectively. However, it is
not appropriate just to sum the percentages in the gastroenteritis rows of
Tables 6 and 7 of the paper (Stokes et al, 1971) since they may result in
double counting (subjects may have reported gastroenteritis in more than one
of the time intervals). This is especially true for countries with higher
rates of enteric infections.
A similar argument pertains to the "unrelated illnesses" referred to
later in the paper.
Shattock: Exactly the same, the same caveats apply to both groups
equally. It is the overall comparison of the two groups that is
important. One group (MMR) had way more GI problems than the controls.
Please examine this for yourselves and thank you for drawing the
casual reader's attention to this, one of the key points of the whole.
Wakefield: Combination of the data sets [gastroenteritis in
Philadelphia and Costa Rica-San Salvador], as presented, obscured these
Watson: These data are not presented combined, but rather separately,
in Tables 6 and 7 of the paper (Stokes et al, 1971).
Shattock: The data are indeed presented separately. There is no
evidence, in this particular paper, that the figures from Philadelphia
and Costa-Rica were combined to obscure the differences seen in the
developed and undeveloped world.
In the next paper (Schwarz AJF etc 1975) there is no attempt to
separate results from the 282 children from Daytona (Ohio) - a
developed country and the 1192 from the undeveloped Santo Domingo and
Panama. Given the significant differences referred to above, this
would seem a very unscientific and unsatifactory approach. It was also
difficult to understand why Gastro-enteritis was completely omitted
from the list of "side effects" considered when the differences were
Watson: High rates of gastroenteritis post-MMR vaccine. The gold
standard in safety studies for common events following MMR vaccine was a
placebo-controlled crossover study of 1162 twins in Finland in 1982 - this
is not even mentioned by the authors.
More detailed data from this study has recently been published
[Virtanen M, Peltola H, Paunio M, Heinonen OP. Day-to-day reactogenicity and
the healthy vaccinee effect of measles-mumps-rubella vaccination.
2000 ; 106(5): E62]. In this study, diarrhoea, nausea and vomiting after MMR
were reported with the same frequency as placebo.
Shattock: Was this 1982 study published? If so where? It would have
been difficult for Dr Wakefield to quote the 2000 paper which appeared
after his own paper had been submitted. If the results are as you
suggest they would appear to be at variance with the previously
published report referred to above. Checking is clearly required.
Wakefield: Follow up for detection of adverse events was reduced from
4 weeks in the initial controlled trial, to 3 weeks in subsequent studies.
Watson: As mentioned earlier, the observation period in the initial
studies was not 28 days but up to 63 days.
Shattock: Again I must disagree. Observations were for 28 days only.
At the time of "the second bleeding" (up to 63 days) parents were
asked about "any significant illness". Given the nature of the
questions and the fact that there was none for the whole 63 days, the
side-effects listed in the paper were clearly excluded. There is no
doubt that the 28 days is accurate.
Watson: The later development studies of MMRŽII had an observation
period of 42 days. PriorixT studies had observation periods of 42-60 days
Shattock: Publication details please! You could be right and we
need to know.
Watson: Numerous post-marketing surveillance studies of MMR vaccines
have been conducted and published - these are not even mentioned by the
Shattock: All that has been requested from day one is the science
that demonstrates safety. If these numerous articles exist why have
they not been quoted by the Department of Health and why has no-one
been able to locate them? References please.
Summary - Watson: This review has been published in an obscure
journal aimed at toxicologists and clinical pharmacologists. It has a
worldwide circulation of 400 copies (www3.oup.co.uk/drugsj/adrates/). It
would not have been accepted by a mainstream journal. The paper has been
widely leaked to the media but its first author would not provide Aventis
Pasteur MSD with an embargoed pre-publication copy.
Shattock: The paper was about "Adverse Drug Reactions" so a journal
about "adverse drug reactions" would seem to be appropriate. There
are few journals designed for such studies.
An advance copy was sent to the Department of Health. Leakage did
nothing to help the authors' case and was unfortunate all round. Not
only did it obscure the authors' presentation but interfered with
the Department's own pre-emptive campaign of the previous 3 weeks.
For Aventis Pasteur to expect advance copies of such material
when their own 30 year old research is still unavailable for
assessment by interested parties is really a bit much.
Watson: The paper is a selective, biased and flawed attempt to
undermine confidence in MMR.
The authors themselves concede that the contents of the paper have no
relevance to autism.
The paper is littered with inaccuracies. The worldwide literature on
the positive aspects of MMR safety has been totally disregarded.
No published epidemiological study has ever shown a link between MMR
vaccine, autism and inflammatory bowel disease.
Shattock: I would invite and encourage readers to compare the
quality of the evidence and the tone of the Wakefield paper with that
presented by the company representative before making any final