The essence of
the crime: LYMErix was bogus because the testing for Lyme is bogus and
you can't use a test that does not detect Lyme to assess whether a vaccine
prevented it. This is
what I told the FDA--on the FDA's website.
My name is Kathleen Dickson. I am an analytical chemist from
Southeastern Connecticut. I would like to discuss the validity
of the results of the LYMErix adult vaccine trial, specifically--
the validity of serological standard used, and how that standard
affected the vaccine trial results.
THE PROBLEM IS THE DEARBORN/DRESSLER IgG STANDARD.
One of the testing procedures used in the trial, the Western blot,
looks for antibodies to specific antigens expressed by B. burgorferi.
The limitation of the Western blot, is that it qualifies the body's
reaction to the infection but does not actually identify the
In Lyme disease, patients produce variable antibodies over time,
most likely a result of antigenic variation - the organism changes
its outer membrane components, and even most of those identified
antigens are variable antigens. Current diagnostic methods now
target the invariable region of the variable antigens, for this reason.
According to Allen Steere, Chief of Rheumatology, Tufts: (2 reports)
1) 1986, Journal of Clinical Investigation, (Title: "Antigens of
Borrelia burgdorferi recognized during Lyme disease. Appearance of a
new immunoglobulin M response and expansion of the immunoglobulin G
response late in the illness.")
"...The IgG response in these patients appeared in a characteristic
sequential pattern over months to years to as many as 11 spirochetal
2) 1993, Dressler/Steere, (Title: "The Serodiagnosis of Lyme Disease",
which came to be the CDC/Dearborn IgG criteria), Journal of Clinical
Infectious Diseases, 1993 Feb;167(2):392-400.
"...The specific immune response in Lyme disease develops gradually
over a period of months to years to greater than or equal to 10
10 or 11 antibodies characteristically show up in Lyme. Some are
more specific than others. These 10 or 11 bands don't all show up
at once, however. They show up one or two or a few at a time.
Persistent infection is evidence by changing bands over time.
CDC decided to establish another serodiagnostic standard, based on
these specific antigens and called for a Second Serodiagnostic
to be held in Dearborn Michigan, late October, 1994.
However, in May, 1994, immediately prior to the start of the LYMErix/
ImmuLyme Lyme vaccine clinical trials, members of the CDC and others,
privately met in Fort Collins and decided that the Dressler/Steere
standard for IgG of 5 of 10 bands be the CDC standard, according to
transcripts of the June 1994
vaccine meeting, presumedly to
facilitate the vaccine trials.
[Slide 3, Table 1 of Dressler]
The problem with the Dressler IgG standard of 5 of 10 bands is that it
-calculated to be 99% specific, and was not empirically derivedů
-It was generated from strain G 39/40, a strain Barbara Johnson of
the CDC, later, at the Dearborn meeting, recommended NOT using,
-And represents an artificially compressed summary of what only the
arthritis-presenting patients showed over time.
And does not represent what's going on in neuroborreliosis, a much
more serious and disabling disorder.
Table 1 reports the frequency of certain antigens, polypeptide and
>From the arthritis data set, were derived the bands for this case
Dressler/Steere report that individual *specific* bands, such as OspA,
B, C, 18-, 93-, and 28-kD, generated from Bb strain G39/40, are specific
markers of infection.
Dresser/Steere report that 18, 28, 93 are the most specific, because
they never showed in the controls. That they never showed in controls
and are specific, would mean, in the presence of symptoms, that one
of these bands indicate that Lyme is the source of the illness.
P93 and 23kD (OspC) seem to be the consensus on highest specificity,
as seen in the literature. That Steere came up with 28, instead of 23,
could be a reflection of the potential of this odd strain, G39/40 to
generate sufficient antigen of diagnostic value.
Confoundingly, OspA and B were left out of Dressler/Dearborn IgG case
criteria. We surmised that this was because it was intended that these
be vaccine immunogens.
Therefore, the Dearborn case standard criteria for IgG excluded, to
quote Steere, "major", "immunogenic, outer surface proteins" from the
case criteria, the Osps A and Osp B.
The exclusion of Osp A and B has resulted in, is, for example,
unvaccinated people who have 3 IgG bands plus Osp A and Osp B, aren't
diagnosed as positive, according to the CDC case definition, even
though they have 5 bands.
So we really don't know what Dearborn IgG means.
[Slide 4- Imugen Report]
Further decreasing the potential for getting early and adequate
antibiotic therapy is the practical misinterpretation of what
the CDC criteria for IgG of 5 of 10 bands means.
For example, Imugen, uses reporting forms which state: "Normal Range: <
[Slide 5- Zoom of Imugen Report, Bottom Right]
Normal is not "less than 5 bands" --If the patient has clinical signs of
Lyme disease plus 2 specific antibody bands for B. burgdorferi, no
honest diagnostician would assert that the patient does not have Lyme
disease. This kind of misinterpretation of CDC criteria further
"Normal" is no bands and no clinical symptoms of Lyme.
[Slide 6 - Zoom of Imugen Blots, Show Strain ID]
Note that this lab uses G39/40 and FRG, a strain from West Germany. We
question how many people in the US will have been exposed to this bug,
such that they will have antibodies to it.
Clearly, the Dearborn Conference also did not resolve the another
problem of standardization, as demonstrated by this labs' use of
odd strains and reporting concepts.
To miss patients by using this Dearborn case definition serodiagnosis
standard, instead of weighting the specificity of an individual band,
such as Osp C or P93, both highly specific alone, will result in the
patient's lost opportunity for early and successful treatment.
THE PRE-DEARBORN DIAGNOSTIC STANDARD
[Slide 7 - page 29 Dearborn Conference Summary]
[Slide 8- Zoom]
Changing bands over time was formerly the criteria for determining
later stage Lyme disease, in place before the Dearborn conference, as
reported by David Dennis of the CDC:
"1) Isolation of Bb from Clinical specimens
2) Demonstration of diagnostic levels of IgM or IgG antibodies to the
spirochete in the serum or the CSF, or
3) Significant change in IgM or IgG antibody response to Bb in paired
acute-phase and convalescent sera phase
Although potentially useful in confirming active Lyme disease, neither
cultural isolation nor paired serum specimen testing has been much used
for validating cases in routine Lyme testing, since the procedures are
not often performed in the general medical setting."
The majority of the other recommendations made by the invited
researchers to the Dearborn conference on IgG serology, were
based on the frequency and identity of these known-to-be specific
bands, but these 8-9 other recommendations were ignored.
The overall accuracy of this Dressler IgG standard never exceeded 28% in
actual practice and these results were reported by the other invited
researchers at Dearborn. In other words, most people with Lyme disease
DON'T have a 5 of 10 band profile.
HOW DOES DEARBORN APPLY TO THE VACCINE TRIAL?
If few people have Lyme disease - and this Dressler/Dearborn criteria
will exclude most Lyme patients - the vaccine will not be shown to
be a failure or cause adverse events.
We believe this is exactly what happened in the trial.
[Slide 9 Table 2 of NEJM SKB Vaccine Results]
Only 22 people got Lyme disease the first year in the vaccine group,
while there were 515 unconfirmed cases - compared to in the placebo
group of 468.
There 10% more unconfirmed cases than in the placebo group in the first
year of the trial.
There were ~1750 Unconfirmed Lyme disease cases reported during the SKB
trial of ~11,000 over two years.
The Western Blot serology from these unconfirmed Lyme cases will need to
be reviewed for evidence of other Bb specific bands and compared to the
placebo group by an independent group of analysts. If there are any
other specific bands besides OspA, the case must be counted as a Lyme
disease case, in the presence of symptoms.
Note that there were only 2 asymptomatic cases the first year in the
group vs 13 in the placebo group. In the second year, there were 0
(zero) in the vaccine group and 15 in the placebo group.
We believe these results do not show that the vaccine is effective at
preventing asymptomatic Lyme, which SKB reports, but rather, that it
is turning asymptomatic Lyme cases into symptomatic ones.
As a support group leader in Southeastern CT, I have met ~10 people, who
found my name on the internet, who had adverse events and were ill,
looking for help. After learning more about these patients, I found
that all but one of these cases had previous Lyme, and that one got
the Erythema Migrans rash during the series of vaccination. NOT ONE
SINGLE PERSON DID NOT HAVE OTHER BANDS ON FOLLOW UP WESTERN BLOT.
It is because I have gotten so many calls from patients looking for help
because of their illness, that I am here today.
Continued follow up on these Unconfirmed patients should have been with
further Western blotting from one of the CDC recommended strains (B31,
297, 2591) and the original case definition, to look for changing
bands, and/or one of the newer antigen-decomplexing methods, like that
of Len Sigal's of RWJ or Steven Schutzer's, for IgM or IgG.
In the re-tabulated results, which we insist be performed, cases where
active infection is not found by these follow up methods, should be
resummarized as the "Uncomfirmed Lyme/Possible Seronegative Lyme".
VACCINE FAILURE AND ADVERSE EVENT
[Slide 10 Persing's Patent]
Dr. David Persing, formerly of Mayo, now with CORIXA recorded in his US
"Additional uncertainty may arise if the vaccines are not completely
protective; vaccinated patients with multisystem complaints
characteristic of later presentations of Lyme disease may be difficult
to distinguish from patients with vaccine failure. Vaccine failures
have been occasionally noted in animal models (E. Fikrig et al.,
Science, 250, 553-6 (1990)),..."
Vaccine failure and vaccine adverse event cannot be distinguished from
each other. An asymptomatic Bb infected adverse LYMErix event case may
never be detected until the patient is vaccinated and symptoms occur,
which we think explains the majority of the adverse events reported to
LYMErix. Many previously infected Lyme cases report systemic
symptoms after vaccination. Many find out they had Lyme after being
vaccinated, becoming ill, being tested for Lyme and finding other
therefore not be looking for only arthritis as a potential
adverse event, to the exclusion of systemic illness.
FREQUENCY OF ASYMPTOMATIC INFECTION
[Slide - 10]
According to Allen Steere's 1986 report, it is possible that, for
every one Bb-infected person with symptoms, there is one walking around
Vaccine failure and exacerbation of asymptomatic infection are
according to the patient data collected, and on the online VAERS
Dearborn/Dressler is not a valid criteria for assessing Lyme, the former
CDC criteria of changing bands is valid.
Until there is an independent review of the WB data from the trial, we
have no idea how safe this OspA vaccine is.
[Slide- SBK Results Table]
By what mechanism vaccination of the asymptomatic Bb infected patients
is causing the Lyme like illness, we do not know exactly.
Previous infection could be "priming" the immune system, as Denise Huber
of Tufts has suggested, in "Identification of LFA-1 as a Candidate
Autoantigen in Treatment-Resistent Lyme Arthritis" July 31, 1998,
Science, Vol 281, p 703.
or the vaccine is activating a dormant infection by the immune
dysregulation it causes, as demonstrated by the effect of Bb
infection and Osp A alone, on NK cells population, T cells,
neutrophils, and the effects on the various inflammatory
regulating biomoleclues, such as IL-10.
We simply don't know all the variables, at present, that effect systemic
illness from immune dysregulation caused by Bb infection, and especially
the effect of a sucha a large dose of a known immune irritant, Osp A
upon this system, the asymptomatic Lyme patient.
The vaccine should be taken off the market immediately, until the true
data, the acknowledgement of the presence of other bands besides Osp A
in all 4 groups of uncomfirmed Lyme is published and re-presented to the
Certainly this vaccine should not be approved for use in children, until
we know the true results of the adult vaccine trial.