>> I think maybe we should start.
Dr. (indiscernible) is away this week so I will be filling in for him.
and the talks today for the demystifying medicine will continue on the new and emerging
diseases and this week's talks on west Nile disease we have two outstanding speakers
today, Dr. Amy Agrawal who is trained mostly at the NIH and is part of clinical center
and who is the key person and chair of a clinical study looking at west Nile disease and
the epidemics that's spread through this country very quickly.
we know that west Nile disease is considered to be a very low pathogen FLAVO virus
type and by mutation, whatever occurs with this virus and we'll learn about it today I
hope has become a highly infectious, very devastating disease.
so this is one of the areas that is of concern and it was a very hot topic as I learned several
the response of the NIH was to institute this multi-center clinical trial and we will hear
today about the results and some of the potential therapies that are coming our way.
then Dr. Murphy will continue with more a basic science aspect of west Nile disease and
Dr. Murphy is Director of the laboratory molecular immunology at the national institute
of allergy and immunology.
he'll tell us how he bridged his research on cytokines to host defenses and genetic risk
factors for west Nile disease.
without more further to do, please welcome Dr. Agrawal.
>> Thank you very much.
can you hear me?
Thank you for that nice introduction and it's a great pleasure and honor to be able to
speak to this group.
as I said earlier to be able to see what building 50 looks like.
I have never been in building 50.
I know this is a group of mostly Ph.D.es and I am first and foremost a clinician soy I'm
going to probably slip into medical jargon F. there's anything unclear please stop me and
raise your hand and I'll be happy to explain things.
I'm going to go lieu the clinical and therapeutic aspects of West Nile Virus.
I'm going start with a clinical case description so you can get a flavor of what patient with
severe West Nile Virus might look like.
then go on to a description and epidemiology of West Nile Virus, I'll go through the
clinical spectrum, very heterogenous of the disease and diagnosis and then investigational
therapy and give you some of our results of our recently completed multi-center clinical
so to start with a case history, this is a patient I saw who was a 52-year-old white female,
stage 1A non-hodgkins lymphoma, a very limited disease.
she had finished cycle 4 of chemotherapy, a week later developed fever, headache nausea
after three days went to the oncologist and was found to be a temperature of 105
Fahrenheit, severe headache and was admitted to the hospital in Rhode Island.
she there had a temperature of 103 her vital signs were otherwise fairly unremarkable
except an elevated respiratory rate of 29.
her physical exam was notable for power, her PIC line central catheter was clean and
didn't look infected and there were no focal findings.
Her chest x-ray was unremarkable and her white count was on its way down from the
her A and C is neutrophil count was 1800 and going down and 1,000 is considered the
threshold where someone is called neutropenic, if they're on fever go on antibiotics in the
treatment of cancer.
Hematocrit was low and platelets at 84 from chronic disease and from chemotherapy.
she had a head CT scan unremarkable and lumbar puncture which was remarkable for
elevated white cell count of 40.
otherwise normal red cell protein glucose and gram stain were unremarkable.
she was started in anticipation of neutropenic fever since the white count was on its way
down on a broad spectrum anti-buyiottics.
to treat the possibility of herpes simplex virus.
on hospital day three she had further evaluation for the on going high fevers CAT scan,
chest pelvis was unremarkable.
MRI of the brain was unremarkable and she developed a rash all over her body.
this was thought possibly a drug -- to be a drug reaction and her antibiotics were switched
she did not improve on hospital day four she developed the inability to speak.
her ANC now neitherred at 200.
and on that day she had repeat lumbar puncture which was even worse with 79 white
the gram stain was unrevealing.
at this point she met the definition for encephalitis.
she had altered mental status more than 24 hours.
it can be any kind of depressed level of consciousness or lethargy that doesn't clear with
for example treatment of fever or waiting a few hours.
It has to last more than 24 hours confusion or personality change.
plus one or more of the following, fever, a focal neurologic finding, she had see insures
which she didn't have.
CSF which she did have, and abnormal electro enreceive la gram or imaging consistent
on day 6 she had repeat MRI.
she did have medial lesions shown with arrows here.
and a third lumbar puncture was performed 8 days after symptom onset which show it is
white count declining to 11.
possibly due to nutrapenia but possibly due to virus burning out and leaving her with
sequella but finishing its acute phase.
her approaching was now elevated at 90.
and several days later she had a West Nile Virus PCR from her initial lumbar puncture
the IGM remained negative and all of her studies that were sent out came back negative.
so she was transferred actually air lifted down here to participate in a clinical trial, that's
I'm going to get late tore our clinical trial but that's the extent to which we were going to
try to get patients into this clinical trial.
she was air ambulance and enrolled as a patient under our study.
here at the NIH clinical center from Rhode Island.
and now I'm going to move to a description and epidemiology of the West Nile Virus.
West Nile Virus is an ARBO virus, a functional rather than biological grouping that
includes other mosquito, sand fly and tick born viruses which cause encephalitis.
it's a single stranded RNA virus in the family FLAVI VERDI.
part of the Japanese encephalitis Cihra complex which includes a group of closely related
JE, St. Louis encephalitis and Murray valley encephalitis.
including DENGI virus and tick borne encephalitis.
This is an electron micrograph of West Nile Virus from the 1999 initial arrival of the
virus in North America.
this is from a crow's brain, an envelope virus 40 to of 0 nanometers in diameter and that
1999 New York strain was found to have a 99.8 sequence hemology to a strain from
Israel supporting that as a source circulating the North America.
this virus mutated and that longer is circulating in North America but it's still causing a
similar type of disease.
this slide is a picture of the global distribution of the Japanese encephalitis Cihra group of
As you can see in blue West Nile Virus, let me see if I can get this pointer from my angle.
before the arrival in North America was the most widely distributed was the most widely
distributed of the Japanese encephalitis Cihra group.
In pink you can see Japanese encephalitis which causes severe outbreaks in Southeast
Asia and in orange St. Louis encephalitis virus which occurs in the Americas and
interestingly not yet explained has almost been replace bid West Nile Virus in the last
CDC can't find it on surveillance, mosquito surveillance, sent knoll chickens and no
human disease reported in the last few years almost as if West Nile Virus is unfeeding it
and that's not reasons for that are unknown.
then Murray valley encephalitis is closely related to West Nile Virus in Australia.
this map is a compilation of the years from the CDC compilation of years 99 to 2001 and
the hash mark states are areas that had human disease and the solid colors are areas that
just had mosquito other animal disease reported.
The pink is '99 through 2001, started in New York and spread very minimally less than a
hundred cases each of those years.
and then spread only in animal force the most part except for some human disease
starting to emerge along the Gulf Coast in 2001.
in 2002 however the disease really exploded and went across almost every state in the
United States, human diseases are in brown, a few animal cases were on the border on the
verge of the epidenimmic.
and there are many more case.
I'll show the case count in another slide.
The first three years less than a hundred cases not many fatalities.
This year several thousand cases.
subsequent years show that the disease remained across almost every state in the
continental United States and several in the southern provinces of Canada.
and I'm not sure how well you can see the numbers on the East Coast but it did almost
disappear from the East Coast and it's been predominantly in the upper Midwest,
California and Mississippi.
and epidemiologists are slowly trying to get at why this is, it's not fully understand why
seems to have big outbreaks after unusually hot summers, it's transmitted very effectively
by mosquitoes that live in irrigated agriculture sok occurs a lot in rural areas far away
from major medical centers often.
and in 2007 I'm showing you here a county by county map.
this is what the last four years has pretty consistently shown, north and South Dakota,
Montana Idaho, California and then there's a hot bed in Mississippi.
the rest of the country has been relatively spared, that's not to say that bird populations
are not coming back with another unusually hot summer with birds flying around several
hundred mile radius that we wouldn't have an outbreak in any area of the country, it's
been extraordinarily difficult to predict during the years of our where the outbreak would
be but this is a pattern lately been emerging.
so this is a summary of the cases and deaths.
You can see the first three years wasn't much to worry about, it was kind of interesting
and scary but in 2002 the year that brown map I showed you that really covered the
whole country and everybody that's when you started seeing it over the news all the time.
in 2003 was the peak outbreak so it's been '04 through '07 we have seen that pattern ton
county by county map starts to be relatively consistent.
the modes of transmission primary mode is mosquitoes and there's an ensowOT cycle
where mosquitoes bite birds, not necessarily the crows and blue Jays that die from the
disease but birds that develop a high level of viremia and live long enough to pass to it
the next mosquito.
That's the ensowOTIC cycle.
when it bite as humaner horse or mammal, reptiles, that's a dead end host.
they are not developing high enough viremia to pass it on to another mosquito but they
get sick and can die from that disease.
other modes of transmission have been documented since the North American outbreak.
blood transfusion, we have had cases, mull multiple cases occur through blood
transfusion and now nucleic acid testing is done on on donated blood during West Nile
six cases could recollected despite testening 2003 and one in 2004 and the blood banking
people greatly intensified the way in which they test and there haven't been transmitted
cases since then.
they were doing mini pools and some got now they're testing individual units.
several case where is a donor wasn't found to be West Nile Virus positive but the
recipients came up with a positive test after the organs are donated where you're
immediately immunosuppressed with drugs to tolerate the translated organ and several
deadly weaponed encephalitis and died from West Nile Virus.
needle stick accident, one documented case in the 1950s, that was known before this
outbreak that's transmitd through laboratory accident.
one via breast feeding the child was atim tomatic and four transpregnancy registry and
there have been no other cases identified.
so on to the clinical spectrum of disease and the diagnosis.
so this disease has become more virulent as mentioned in the introduction, first isolated
in 1937 from a woman in the west Nile district of UGANDA.
and it caused outbreaks in subsequent decades in Africa and Middle East of west Nile
fever which was a self-limited viral illness that never caused very severe disease.
There was a well documented outbreak in the mid 70s in south africa that had 18,000
cases with no records of encephalitis and no death.
there's been a major change in virulence since 1996, an outbreak in Romania, then Russia
in '99, Israel in 2000 and the outbreaks we have had yearly in the western hemisphere
have all had been much more virulent with unprecedented rates of encephalitis an
mortality indicating evolution of a new strain of virus.
so this is the iceberg pyramid of west Nile virus infection so the vast majority, this is very
rough estimates the vast majority about 80%, possibly more are asymptomatic after being
20% develop west Nile fever, less than 1% the initial estimate was 1 in about 150 cases
get neuroinvasive disease so encephalitis, meningitis, polio like or paralysis disease.
this comes from several studies where they go on the heels of a hot outbreak season in a
geographically area and go house to house and take a survey if people had any
recollection of symptoms.
Since the blood banking group versus started doing testing of donated blood the number
at the tip of the pyramid is probably even smaller so there's unreported asymptomatic
probably neuroinvasive cases are more on the order of 1 in 300 of people who cera
going through the different types of phenotypes of disease that West Nile Virus produces
west Nile fever is themidest of the diseases has sudden onset high fever, headache,
sometimes ocular pain, malaise, nausea, vomiting, diarrhea and rash in many patients
much like the patient I presented earlier.
in other outbreaks it was described as a self-limited illness usually lasting less than a
even in those severe outbreaks in Romania, Russia, et cetera, west Nile fever didn't get
that much attention and was not that severe.
new data has shown in the American outbreaks that it's probably a lot more severe now
than it was ever thought to be and I have just highlighted what I thought were some of the
more interesting symptoms.
This is a retrospective interview, 98 west Nile fever patients, they excluded anybody with
96% reported fatigue.
what they report, I know this from talking to the patients and investigators that are multi-
sites is really disabling fatigue.
it's like marathon or previous marathon runners are unable to walk up the stairs of their
house for six weeks. The medIan duration was six weeks in this retrospective series.
muscle weakness by 61% lasting four weeks and difficulty walking reported by 47%.
the median time to feeling like you have got back to normal was reported to be 60 days in
as you can see with west Nile fever people have variable kind of symptoms, it's not like
they all have a little bit of each, some only have trouble concentrating, some only have
fever, some only have a physical symptom.
sometimes they cluster together and overlap a bit.
in hospitalized patients with more severe illness than fever this initial report of the 1999
New York outbreak has stood up consistently every case series that's published about the
more severe spectrum of the disease in hospital supports what they found here, a two
third, one-third split in the patient whose end up in the hospital between encephalitis and
patients with just fear and headache are not classified at about 8%.
and there's overlap of people with fever and headache and don't quite meet someone's
idea of needing a lumbar puncture.
many of them probably do have it and weren't tapped if they're sick enough to get into the
hospital from fever and headache.
PORESIS, some kind of weakness or hypoflexia in 66% of patients.
So some kind of weakness, hard to classify in more than half the patients.
the Frank polio or paralysis is in a minority of patients, 10%, that statistic is held up in
other case series.
in this series it is mortality, this has been consistent going forward, overall the mortality
was about 12% in the hospitalized patients if they had encephalitis it was 19.
encephalitis plus motor weakness up to 30%.
most significant risk factor and significant illness and death was in patients over 75 years
The odds ratio for death in those people was 8.8.
neuroinvasive disease due to with, that's the tip of the myrrh -- due to West Nile Virus,
tip of pyramid, a third encephalitis and maybe ten percent or less polio my lie at this and
other things reported in the North American outbreaks are Parkinson features, and
This is a slide I have had in talk force clinics but included for you too, there's nothing
pathopneumonic for West Nile Virus, nothing that's a hook to the diagnosis.
meningitis is a typical.
very non-specific and includes in the diagnosis if they are present are a viral that includes
GI symptoms that's not always there for aseptic MENINGITIS.
headache or rash, the rash is significant during the West Nile Virus season.
MRI find they're there, they're helpful but a normal MRI is more common.
asymmetric motor weakness is suggestive of West Nile Virus in the West Nile Virus
not a lot of other things that do that.
routine labs are not helpful.
there is a neuroepidemiology gist named Jim SAV aR at CDC who has been tracking this
outbreak and he travels to outbreak areas and sort of camps himself for several months
and finds a cohort of patients to follow.
he's done this now with a group of patients in Louisiana and -- who had all manner of
disease and then a group of paralysis only patients in the whole state of Colorado.
he captured every patient in one year's outbreak and he's following all these people
so he's been the generator of much of the prospect active data about how people do with
his case series albeit small.
this is a Louisiana series.
-- Louisiana series.
16 patient he is followed 8 encephalitis, 5 with acute paralysis.
he's the first to report the movement disorders.
Seen in 16 out of 16 patients.
tremor, and Parkinson illness, the basal ganglia will light up on MRI in this disease if it's
abnormal and that's also the area affected by Parkinson's disease the question remains
what happens to the people as they age, whether they will in their old age develop
Parkinson's disease from the initial hit to their basal gain glia is unanswered.
8 months follow-up in these patients there was one death the acute flaccid paralysis
patients only three in the series had no improvement.
encephalitis 5 out of 7 improved to baseline.
MENINGITIS patients had a complete recovery.
five patients had persistent tremor at 8 months and 5 ad par kin sewnian features.
this is an example of the typical MRI findings.
I put in here the herpes simplex findings to compare to West Nile Virus.
that's the most common cause of sporadic encephalitis in the United States and it has
pretty reliable temporal lobe enhancement commonly seen when somebody presents with
herpes simplex encephalitis.
This outbreak afforded the opportunity to show that West Nile Virus Japanese
encephalitis and Murray valley encephalitis have similar findings on MRI when present
which are enhancement on flair imaging of the foulness and basal ganglia.
they're prevalent among the Japanese cera group.
Our parent had two MRIs, first was normal and in a non-research institute where money
is no object that's tall patient is going to get is one MRI.
she happened to have a second one that's when it popped up through enhancement.
but these are two case series one had 39 patients, all 39 had IGM in their CSF, all
confirmed west Nile cases.
37 out of 37 had normal CT scan, only one of 14 ad MRIs that show encephalitis.
and they had sub cortical and hyperintensity in flair.
I use the word enhancement that's wrong.
's hyperintensity on radio flair.
the second case series was 16 patients, prospect TVly followed 16 out of 16 CAT scans
two out of ten MRIs were abnormal.
when abnormal they had enhanced of basal gain glia thalamus and ponds on imaging.
it doesn't exclude West Nile Virus but characteristic findings maybe suggestive of the
the other concerning thing is the portfolio owe like illness, similar to polio virus type
illness and many people call it West Nile Virus polio.
it is an acute flaccid paralysis that's asymmetric and caused by anterior horn cell loss in
the spinal cord t-cells infected and knocked out by polio virus.
I'll show you a case series how those patient patients do in a moment. There's the
spectrum of this muscle weakness, a less severe presis.
56% of patients had some kind of weakness that was document on physical exam.
then there's peripheral neuropathies and in the Chicago department of health study there
was a fatigue weakness with no objective abnormality, that's very poorly understood.
This is the Colorado cohort.
Jim went to Colorado and Mced to capture virtually everyone in the state that year in
2003 who had paralysis.
and so he followed these 32 patients and continues to follow them.
27 of them had asymmetric polio type paralysis.
Four had a (indiscernible) type ascending symmetrical weakness.
one had an isolated peripheral neuropathy and at four months three were dead, only two
at baseline strength.
most remained completely paralyzed.
17 had respiratory muscle weakness and 12 or 40% were intubated.
and he was there actually for some of these patients, affected only with paralysis, not
with encephalitis or cognitive difficulties and several participated in their own decision to
withdraw support and die.
it's kind of a scary illness, another scary thing is it affects younger healthier patients not
surely has to do something with the immune response to illness.
but these are the types that the median age is younger for patients who get paralysis than
those who get encephalitis.
the diagnostic criteria for West Nile Virus infection, this is from CDC diagnostic criteria,
you can isolate the virus.
it's a P-3 biohazardous pathogen, I think.
they were talking about shirting it to P-2 but guess not yet.
Or positive PCR.
four fold rise in IGG.
so acute and convalescent titers.
West Nile Virus IGM in CSF is considered diagnostic.
although IGM does persist in CSF for possibly up to a year in some cases beyond a year
so it could be an infection from the previous season but right now it's considered
diagnostic to find that in CSF with compatible clinical symptoms or IGM and IGG
together in the same serum sample.
so a key point for the diagnosis are peek viremia occurs one to two days prior to
symptom onset by the time you get the patient there's a limited role for PCR.
and the IGM is more often what's going to be positive.
the notable exception is patients immunosuppressed.
She was post chemotherapy, becoming neutropenic and negative IGM so they need to be
checked with PCR patient whose are otherwise healthy.
many hospitals and insurance companies won't pay for testing with a PCR because it's so
unlikely to be positive.
most patients should be tested sere logically, very sensitive.
IGM and IGG should be in the blood and CSF.
if negative titers should be checked.
it can take up to three to four weeks.
so on tovational therapy.
after that big spread across the country with 4,000 some cases in 2002 in the fall of that
year NIH convene ad workshop on West Nile Virus therapeutics.
just inviting concerned parties from around the country to discuss what should be done
about this, were there any potential clinical treatments that ought to be studied.
at that time we didn't know.
Epidemiologists best guess was we didn't know.
is it sporadic or going to cause yearly outbreaks like japanese encephalitis, and we
thought if it is not going to cause outbreaks this maybe our last chance to study the
disease so there was urgency to try to test a o promission compound in humans using the
collaborative anti-viral study groups infrastructure.
intravenous globulin pooled antibodies from donors that are used for other diseases.
These are safe and proven efficacy for other animals.
Ribovirus was shown effective in very high doses.
at that time there was no data.
the IV juiced the achieve the high serum levels and it was available only in limited
quantities from the producer.
and riboviral caused anemia in high doses and interferon alpha it was effective against
West Nile Virus in high concentrations in one in vitro study.
no animal data.
at that meeting Tom Solomon from the UK presented his data from a trial in Vietnam
where it was completely ineffective.
it was a blind placebo controlled trial.
the decision was made to test the immune globulin with high titers of anti-west Nile
two case reports from Israel in 2000 suggesting a clinical response.
was was treated with IVIG and made a dramatic response.
they tested the IVIG for antibodies and found high titers.
they tested some U.S. source plasma IVIG and found it has no deckable titers against
West Nile Virus so suspected perhaps there was a cause and effect relationship.
there was a second case report late their year where they used it and somebody else made
the dramatic clinical response.
so with our outbreaks we started -- people around the country we included had a patient
here at the clinical center who we used emergency IND obtained that product on the
market in Israel and had it shipped an treated patients.
All were uncontrolled and no conclusions could be drawn.
our patients cleared his PCR but died and never recovered from encephalitis how far he
was treated late in the course of disease.
another thing, supporting trying this compound was that it's very early 1970's mouse
model had responded to immune serum so the virus was detectable in the mouse at day 6
and immune serum given at that time produced 80% survival versus 3% in controls after
innoculation with West Nile Virus.
mike diamond from Washington university presented this data which was as yet
unpublished, the only thing out there about this product from Israel called AMERGAM
seen in green in the legend.
he inoculated mice with 100 TFU of West Nile Virus at day zero and then he compared
saline AMERGAM the brand name of this IVIG product and non-immune globulin which
was American source IVIG.
and as you can see here out to day 5 there's no difference between the non-specific IVIG
and saline and the percent of mice surviving.
but there is a rescue of about 50% of mice when the virus consistently appears in brain at
after day 5 this effect -- this drops off.
So we knew we had a very narrow treatment window to shoot for because patients don't
present until they have symptoms an often they don't have symptoms until the virus is in
to try to run a clinical trial was risky but we felt we should go forward and hoped for
some difference between mouse biology and human biology.
so we developed in collaboration with the CASG the phase 1, 2 randomized placebo
controlled trial to assess safety and efficacy of IVIG, AMERGAM containing high anti-
west Nile Virus antibody titers with or high progression to West Nile Virus encephalitis
so only the sickest patients.
There were two inclusion categories.
Because of the narrow treatment window we enrolled patients not yet confirmed with a
test and just had presumptive disease and the West Nile Virus season.
so we would take anybody who had certain criteria that looked like encephalitis, nothing
obvious pointing to some other diagnosis.
and a little bit of white cells in their CSF and enroll them and then wait for the diagnosis
to be made and go ahead and treat them anyway with one dose only of this AMERGAM
or we would take confirmed patient whose had acute West Nile Virus, had a positive test
but whose disease was not yet neuroinvasive.
we thought if they were positive and has been in a coma two weeks it was too late.
If they were positive in racek group like older immunosuppressed and not yet had
encephalitis they would possibly benefit from treatment.
we randomized these two groups in a three to one to one ratio to the punitively activate
American source I IVIG and saline.
and this is just -- I'm going to flip through quickly to show you the effort put in to trying
to enroll this trial.
our peak we had 70 sites.
This is I think 56 with us throughout most of it.
we had a lot of sites in the U.S. and Canada trying to be there on the ground when the
We didn't know where it was going to hit we tried to do best guess, constant
communication with CDC.
we put a site, get it through the IRB V an investigator there waiting and ready and try to
just wait and see if the disease would occur in that area.
the end points for the study phase 1, 2, the primary end point was safety so we wanted to
compare serious adverse events compared to treatment groups.
For efficacy we did power calculations.
I did them for five minutes with the statistician.
If you want to use a hard end point like mortality to look at efficacy to get a statistically
and clinically significant difference shown in thousand patients treatment arm that's
impossible when the outbreak is 4,000 patient as year, you hope it will be at least that
much following year.
the other problem is the disease is so heterogenous, if you look at morbidity what are
your enpoints going to be?
We used a compilation of index activities of daily living scoring system, ability to bathe
yourself, feed, clothing, whether you work or not, whether you can walk.
modified rank and scale and outcome scale are smaller scales from 0 to 5 and 0 to 6 that
are one end of the spectrum is dead, the other is completely normal and there are things in
between like dependent for feeding and bathe bug not depending for this or cognitively
that kind of thing.
and then to assess neuropsychological effects of disease we used a modified mini mental
This is an expanded version of that with more breadth.
so we had a lot of difficulty occurring subjects, primarily chasing the outbreak at peak, I
mentioned we had 70 sites.
Patients were almost always referred to late.
People would despite our best efforts wait until they had a positive test and then call
investigator at that time site and try to get a patient transferred to that site.
We had inability to transfer even if everything was right and in the treatment window, the
receiving hospital had no beds in the ICU or insurance company wouldn't pay for the
subsequent hospital stay, they won't transfer somebody just to participate in a clinical
so they'd be ten miles down the road at another hospital unable to get into the trial
because it wasn't through hospital's IRB and there was no investigator there.
so the first year we started late due to IRB delays at that time sites.
We caught the tail end of that explosive outbreak in 2003 and enrolled 7 patients.
in subsequent years we enrolled 1914 and then 22.
baseline demographics into the results.
I don't show P values they're all non-significant.
no difference among human groups, there was a number of Caucasians.
what have I done?
There was a male predominance and this was the same in all treatment groups, just
remember if you look at raw numbers here there's a three to one to one randomization
27, 12 and 13 enrolled in the treatment arm.
the male to female predominance is probably consistent with other vector born diseases,
probably due to males being more exposed to mosquitoes being outside doing outside
activities more but not for sure, maybe there's some immunodeficiency that's as yet
and the median age was the same in all groups, mid 50s of patients with severe illness
that we enrolled in the trial.
this is the primary end point, a summary of serious adverse events and I don't show P
values here because there was no difference, no significant safety differences or SAE
differences among treatment groups.
even no matter how you look at it number of death, number of subjects within SAE
absolute numbers of SAEs which there were more than subjects and number of SAE per
subject, any way you slice it there was no difference among treatment groups and the
numbers are small.
so no suggestion of a safety issue.
the primary investigator at each site was responsible for calling an SAE definitely
probably possibly not likely or definitely not related to the treatment.
it was a one time infusion.
so most were not associated and a handful were thought to be possibly associated with
and just to show you what those were, the ones people thought might be associated with
treatment were all things that are known to be side effects of IVIG so low white count is a
known side effect of IVIG.
chest pain respiratory distress can sometimes be caused.
As you can see here two patients with low white count received saline.
AMARGAM caused chest pain and worsening leukopenia maybe but maybe not.
could have been in effect the underlying disease.
all of these were resolved without sequella.
these are the efficacy results.
and I have just compressed these to one slide, not showing error Baars because these
were all non-significant so this is essentially no difference between treatment groups in
this study, so all this can be thought of as natural history data essentially the bar tell index
is activities of daily living score, low score bad, high score good.
so from pre-infusion from the time at the trial they were given a score to day 90 follow-
up all patients improved.
same with the modified mini model, all improved.
No difference among treatment groups.
The modified rank in all patients improved the scale is reverse.
the high numbers are bad T low numbers are good.
and this NCI that scoring system so everybody got better.
We're going to dig into this data a little bit more and look for more trends about natural
history but this is just the overall secondary end point efficacy results.
we were looking for -- we knee we were underpowered but looking for some preliminary
evidence of efficacy, there is none.
and B, an effort, we always knew this was a very underpowered study. we were having
trouble getting enough patients but hoped to be able to validate these as end points since
nothing was validated for this disease to use as end point.
we were able to show that there's a nice spread, patients do improve and if there are
subsequent clinical trials for future compounds and there's one in the pipeline that's a
monoclonal antibody that maybe starting phase 2 next summer could use these and have
something to refer back to as efficacy end points.
so in summary the primary end point was safety.
no significant safety concerns among any treatment arm.
secondary end point was a preliminary indication of efficacy, no difference among
One thing I mention here in the third bullet point is we had planned initially I already
altered the slide we had planned to do two dosing cohorts when it became apparent we
were not going to be able to accrue the trial we wanted to accrue 100 patients 50 at 0.5-
grams perky low, 50 ant 1 gram perky low.
we wanted to go hire and FDA didn't allow it.
When our study agent was expiring and our accrual was slow we made the decision to
stick with one to get a large pool of patients to evaluate similarly so we stuck with a
so the -- what would have happened with a higher dose is simply not known.
additional human treatment studies that started at the same time as ours were industry-
One was alpha interferon plus placebo. It was hemisphere biopharmaceuticals and
queens hospital, that failed to accrue with three subjects.
and an anti-nucleotide by AVI biofar Marks theirs failed to accrue, trying to help each
other because it was so hard and they were not able to accrue their trial for similar
so thanks to the following colleagues.
people here at the clinical center and my colleagues at collaborative anti-viral study
group at the university of Alabama Birmingham.
and now I can take any questions.
>> We have time for a couple of questions and then we'll probably have another
discussion at the end of the session.
I wonder what you think about when you showing the MRI is what we're looking at, are
we looking at inflammatory response, we're looking at white cells, what exactly is the
>> I think -- well, the radiologist -- I think would be able to give you technical answer of
fluid infusion so it isn't inflammatory response but the virus does cause neuron necrosis.
>> And same thing -- that causes then an inflammatory --
>> Peripheral neuropathy as well in -- is that similar?
>> Yes it causes anterior horn cell loss identical to polio virus when it's in the final cord.
-- in the spinal cord.
>> You mentioned about the (inaudible) neuropathy with the IDD.
>> That's a very good question.
we looked at -- we looked at -- that was another sort of thing that would make us less
likely to work is whether it was going to get lieu the blood brain barrier, we looked at
radio label studies of IGG and gets through with miniscule amounts of intact blood brain
the hope is with encephalitis your blood brain barrier is broken down enough to let some
IGG which is a big molecule but not as big as IGM which is a pentamer, has to be -- the
first response in CSF and CNS is its own IGM.
so if you can get IGG in there before you develop your own IGM we hope that that would
have some effect.
>> That causes some of the myalgia and other complications?
>> I don't know, maybe Phil can answer that.
people are studying various cytokine effects and I'm sure probable things like TNF alpha
was known to cause but I don't know that's directly causing what symptom.
How it occurs?
>> I don't think so.
but, yeah, no.
still shaking his head no.
>> I don't think so.
I just think they have different -- they're a different viruses that end up doing the same
and there are several other viruses that rarely -- there's several types of antivirus that
causes -- a flat process like polio.
and there's that here and there but nothing on this scale.
this is the first thing since polio there's done this in a major way.
and it does occur in Japanese encephalitis but small proportion of patients.
>> Could you tell us a little bit about the animal models, the animal model, I know you
mentioned some of those are causing death as higher titer virus.
does the animal have the same mouse or other models have a similar symptoms as the
>> That's something that we have been -- we have struggled with.
I have been talking a lot as have the members of the CFC and the company is maybing a
monoclonal where we're talking to the same two animal people, John has hamsters at the
University of Utah and mike diamond has got mice.
and we clinicians pushed for them to extend it out longer and see what happens.
Can you keep the animals alive until they get symptoms and then treat them, try to
minimummic the human situation where you don't go to doctor until you're sick for a
couple of days, most people don't.
and they have tried and they -- a lot of animals simply die.
a lot of time what happens is they just get so weak they don't take food and water so you
need to van IV in them to keep them alive which you would codo with humans.
and we've looked for John has tried really hard with his hamsters, a larger animal, to look
for morbidity end points that you could use to correlate with human and I just -- he's
looked for that and sort of scraped for things that he can use, he has big charts with all
symptoms and inability to walk up a ramp.
I just -- there are a little bit of correlation but not enough.
what this company wanted to do because it's so hard to accrue a clinical trial is ask the
FDA to go for the two animal roles, which is possibly something that can -- it's -- I think
without precedent so far but if you can't do a clinical trial like a bioterrorism agent and
you have two separate animals that you can prove it's safe and efficacious.
I don't think they're going to go for it.
but we try to get John Moyer to show as much as he could to show they were like humans
and I don't think it's ever going to be enough.
not close enough.
he's trying to do now some test analogous to nerve conduction studies and he only has --
he presented this must have been four or five months ago at a meeting where he very
preliminary data he's only done a few animals but showed nerve conduction studies
maybe predictive of death of the animal.
so that's -- that was raised an eyebrow with us who are clinics in the room because in
humans nerve conduction studying abnormalities don't show up until well after the
clinical symptom is there so it doesn't make sense that it would show beforehand in an
that was very preliminary, so we'll see.
>> Are there examples in primates for West Nile Virus type --
>> They don't get sick.
>> You can immunocompromise them.
and that would be a route to go.
>> Seems to me this is going to be an on going research area an certainly lots of
opportunities in that to expand on that disease.
thank you very much and hang around.
we may have some more questions.
We'll continue with Dr. Murphy.
please exchange the microphone.
I want to remind everybody those of you that have come last week we did not have the
books to sign up, there are two books circulating, on this side green books please sign
your name and email address, if there is anyone who has not signed up we need those for
follow-up so please do so.
there's another book circulating on the other side so please pass it around.
well, thank you all for coming and thank you Amy forgiving a beautiful clinical
presentation for West Nile Virus.
it's a little bit like chasing a tornado I guess in Oklahoma, trying to track all these patients
so we got interested in West Nile Virus.
when we started to see crows dropping dead in our backyard and hearing about people
dying there was little known about the pathogenesis.
though I'm an M.D., I really do mostly basic and translational research.
we started out asking some very basic questions about the immunopathogenesis.
Here is the virus as you have seen and Amy has told you about the interaction between
mosquitoes an birds in nature.
and dead end hosts.
so this is where we really focused our attention.
that is as Amy has shown you the outcome of West Nile Virus infection in humans is
highly heterogenous with 80% of infecteds remaining asymptomatic and 20% developing
of those some 40% or so developing just west Nile fever and the rest developing disease
and a small percentage of those dying.
the outcome of most infectious diseases is heterogenous and we wanted to know whether
there were any particular West Nile Virus restriction genes and whether there was some
immunopathogenesis that could account for this.
in order to understand pathogenesis we have taken three complimentary approaches.
one is to use mike diamond's mouse model of West Nile Virus infection here at the NIH
in which we could explore issues of pathogenesis, genetics and therapeutics and couple
that to the epidemiologic studies potentially where we could look at the genetic
components of West Nile Virus.
and possibly even looking at human tissue with respect to immunopathogenesis, genetics
I'm going to show you data where we have actually done all three of these things in a
complimentary fashion to try and get at restriction factors for this disease.
so here is another way of looking at the virus and here is the mosquito inoculating the
skin with the virus.
skin dendritic cells pick up the virus and are thought to migrate to draining lymph nodes.
at this point there's very little known as to what happens.
is this really a check point for viral replication or is it a sort of a launching pad to the
Maybe it's one thing or the other, depending on the host.
so there could be differential outcomes at the level of the lymph node that determine
whether the host stays atomatic or goes on to develop central nervous system signs and
so this is the mouse model that we have been talking about.
and we found as mike diamond has found and others have found that if you inoculate a
mouse, C-57 black six mouse in this case with west Nile virus we can kill about 35 to
40% of mice, never any more in our hands.
it's not clear why that's the case.
but it's actually an advantage because one could imagine that you could explore genes
that might improve this curve or actually make mortality worse.
so this is actually a good position from the point of view of scientifically understanding
what's going on in this disease.
you can look at viral load in the brain and this is days post infection and you see there's --
you see the virus at day 5 there's a peek on this log scale at about day 7.
then in animals that survive you start seeing resolution of virus in the brain by day 12.
18 it's below the limit of detection.
this is actually terrific model of West Nile Virus infection, the animals developed central
nervous system disease as patients do.
and one can try to imagine that there might be similar pathogenetic factors underlying
so in this regard there have been a series of studies in this model looking at cellular
elements of the immune system that are involved in pathogenesis.
so it's known for instance an animal's lacking CD-8 positive t-cells you can increase
mortality up to about 85%, animals lacking gamma and delta t-cells are uniformly
susceptible to this virus.
animals lacking B cells and NK cells also develop 100% mortality.
if you can deplete macrophages chemically you can also show an increase in mortality.
so it's -- and of course cera can be protective as Amy suggested.
all of the elements of the immune system at the cellular level and at the antibody level
seem to be important in protection in this model.
so we wanted to go on and start looking at actual molecules that maybe important in
so will glass in the lab at the time set up the model and started to look at
immunoregulatory factors, genes turned on in the brains of mice infected with West Nile
so these are mock-infected mice.
you can see these are just some of the factors that we looked at on this RNA protection
you can see that in the brains of these mock-infected mice you can see very little if any
expression of these chemokines and cytokines in the brain.
however, in mice infected with the virus soon after infection you see the appearance of
these bands indicating the induction of RNA and coding these various factors.
and I'm not showing you all the ones we have looked at, we looked at many, these are all
chemokines that we have looked at.
and I just wanted to point this out to say that of all the ones that we looked at, perhaps the
most highly induced genes were these three here.
CCL-5, CCL-3 and 4, these are particularly interesting because they all bind the same
receptor, receptor known as CCR-5.
this was very interesting to us because we had actually cloned CCR-5 back in the mid
'90s and done a lot of work hon its role in infectious disease pathogenesis.
isle show you some that just in passing.
so before I show you the rest of our analysis of chemokines in West Nile Virus infection I
want to take a moment to tell you what chemokines are.
there are a type of chemo attractant that binds to these seven transmembrane domain G
protein coupled receptors.
Here are the G proteins.
triggering of the receptor acts the G protein, there's a series of signal transduction events
that occur that all get integrated to allow leukocytes to traffic in a directed fashion to sites
of infection and inflammation.
so the chemokine system at a glance shown here, this is the largest known collection or
destruction of cytokines.
45 in human depending on how you count.
they have complex relationships with 19 known chemokine receptors.
so there's a lot of promy skewsty in the system, individual receptors able to bind multiple
chemokines and the same chemokine able to bind multiple receptors.
there's also evidence of monogamous lye Ligand receptor relationships shown here where
there's a dedicated chemokine to a specific receptor.
these are interesting patterns.
because where there's promiscuity in the system these seem to be inflammatory factors,
turns on at the time of infection or injury.
but off in the absence of injureious systems.
in the monogamous system, these are the more homostatic basal regulators allowing cells
to go to their destination in the bone marrow, lymph node, et cetera, throughout the day.
this is just a summary of the system, as I side there are 45 chemokines and 19 receptors,
also three chemokine binding proteins.
There are many virally encoded chemokines, chemokine receptors and binding proteins,
made by all cell types in a differential manner and they target all leukocyte subsets but in
a differential manner.
we think they're associated with many diseases and they maybe important in many
diseases but the evidence for importance in specific human diseases is actually fairly
the one where there's excellent information about the role of chemokine and a receptor in
human disease is in the case of CCR-5, the receptor we showed is expressed in the brain
of mice infected with West Nile Virus and HIV.
so this is a summary of a huge amount of work started in 1996 in which we first
identified CCR-5 and showed in collaboration with Ed burger here and others showed
contemporaneously and independently that CCR-5 is used in conjunction with CD4 to
allow specific types of HIV strains, the ones that transmit disease, that is the important
ones in -- we think in pathogenesis, to enter cells.
So this is a critical HIV co-receptor you could say.
we also found as others did, that there's a mutant version of CCR-5 in populations that we
call CCR-5 delta 32 which is a truncated form of CCR-5 unable to make to it the plasma
membrane and therefore, unable to interact with HIV.
so this turns out that this individuals who are homozygous for this mutant form of CCR-5
are highly resistant to infection with HIV even if heavily exposed to the virus.
individuals who are hetero zygous for this mutant form can be infected with HIV but
once infected they progress very slowly.
to AIDS compared to individuals who lack this mutant allele.
so I should also tell you that individuals who lack CCR-5 because of homozygous delta
32 appear to be healthy.
they -- when we first found this they didn't -- we couldn't find anything wrong with these
it appeared to be sort of a VESTIGULE gene at the time.
this data provided proof of principal for targeting CCR-5 in HIV/AIDS and since 1996,
'97 that's been done by several drug companies most prominently phizer has developed a
compound known as MOROVARAC, a small molecule that has gone through clinical
trials and is now FDA approved as of 2007 for treatments of patients with HIV/AIDS.
I should say someone shamelessly that I think this is -- shamelessly this is the first drug
that's come out of the intramural research program of NIAID.
since I have been here since 1985.
This is kind of a big deal for the chemokine field and I think it's a big deal for HIV.
but it raise as lot of questions about whether this is going to work on a broad scale, when
given to millions of people and what kinds of toxicities one is going to see.
that's where west Nile virus comes in.
This is how this all works at the biochemical level, TP-120 envelope glycoprotein of HIV
interacts with both CD4 and chemokine receptor.
this causes a confirmational change in the protein allowing a cryptic fusegenic peptide of
the the GP-41 component of the envelope glycoprotein to be revealed.
this induces a fusion core between the envelope of HIV and the plasma membrane of the
target cell allowing the HIV genome to pass the virus into the target cell.
so here is a little bit more about CCR-5 delta 32.
it's a truncation of an area of the second extra cellular loop of CCR-5.
32 not being an even multiple of 3.
this causes a frame shift and a premature stop codon.
so this is actually bogus, incorrect sequence and then here is the premature truncation of
it's found predominantly in Caucasians.
and I've -- hetero zygous individuals have intermediate level on CCR-5 on cell surface
and appear healthy as I said.
so here are a few take-home points about CCR-5 and HIV.
MORAVAROC is the first therapeutic targeting a host determinant infectious disease I'm
the first chemokine therapeutic and CCR-532 homozygosity is one of the stronger
resistance factors known in human inFERC.
this is all an example of the candidate gene approach in determining human disease
resistance and susceptibility.
this is just by way of asking this question, that is, what are the beneficial roles of CCR-5?
Since it's going to be -- the drug is going to be given to lots of patients with HIV who are
already immunosuppressed, what's going to happen when you knock out CCR-5?
Well, some information to answer this question came from our analysis of the mouse
model of West Nile Virus.
we found that in this mouse model as I said CCR-5 Ligands were induced, CCR-5 RNA
We could find CCR-5 positive leukocytes from the CD4, CD8 NK and macrophage
so it was reasonable to ask what happened to the survival curve in animals lacking CCR-
we had access to the animals through the Jackson labs and when we tested them we found
that all of the animals died within 13 days.
so this was a dramatic phenotype and it's suggested that CCR-5 was extremely important
in controlling infection in this model of disease.
how does it work?
We think by conventional mechanism for chemokine receptor and that is by allowing
leukocytes to find their way into the brain to control virus.
here is a slice of a brain from a mouse infected with West Nile Virus.
encircled here are aggregates of leukocytes the virus induce as multi-focal encephalitis
and though it primarily infects neurons there are many leukocytes that enter the brain and
collect in these areas.
Here in this one slice there are 33 lesions.
if you do this in a brain from a knock out mouse there's few of these leagues aggregates
cells present in the brain.
one can also do an adoptive transfer study of splenosites from a wild type donor.
infected with west Nile virus.
and transfer those into a knock out recipient mouse infected with west Nile virus and ask
whether there's change in mortality of susceptibility of these mice relative to expectation.
When you do that, there is almost complete transfer of the total mortality phenotype in
the seen in the CCR-5 knockout mice back to what is observed in the wild type mice.
These are knock-out mice receiving wild type splenocytes and this curve is basically the
same as what you see in the wild type infected animals.
some of these splenocytes that you can knock out fatal for West Nile Virus infection.
another is to take cells out of brains of mice.
that's what this shows.
when you take this leukosites out of the brains you can show that mock infected mice
have few CD4 CD8s or NK cells shown here.
whereas West Nile Virus infected mouse brain shows an accumulation of these cells all
three of these cell types, you can see from 7 to 12 day as lot of these cells are appearing
in the brain.
what happens in the absence of CCR-5, well, all of these levels of accumulation drop off
so you can see there's a large increase in the accumulation of these different cell types.
same happens with macrophage subsets.
so these are resident macrophages, macro glial cells in the brain of mice.
Nothing happens there.
but in mice when you look at the infiltrating macrophages in the brains of these mice you
see a drop off of the accumulation of these cells in CR-5 knockout mice.
It seems to be important in getting all of these different subsets into the mouse brain
where they seem to be controlling infection.
I can't tell you yet exactly which one or maybe all, maybe it's all of these subsets that are
important in controlling the infection but that's something we need to work on.
another thing we need to work on is whether this means anything in terms of human
fortunately as I said we had the ultimate genetic probe to test this -- answer this question
and that is CCR-5 delta 32.
it's a complete loss of function allele so we can do a loss of function genetic test.
and one can look at either the homozygous genotype or the hetero zygous genotype and
correlate with different phenotypes such as symptomatic West Nile Virus infection or FA
fatal West Nile Virus infection.
we have done this in various study populations.
so West Nile Virus sero positive individuals symptomatic initially we tested two, one
from Arizona, another from Colorado.
here are the number of individuals in those studies.
And control populations, one was an NIH random blood donor control population from
here and another was a symptomatic West Nile Virus sero negative population that was
geographically controlled for these individuals from Arizona.
when you do this you find freak so I have CCR-5 deficiency in the random blood donors
from NIH, 1% and shown in many studies in the United States over the years.
when you do that in the Arizona West Nile Virus sere row negative population you get
about the same result but when you look at these disease cohorts West Nile Virus sere
row positive individuals who are symptomatic from either Arizona or Colorado you get
about the same answer which is that there's a big increase in the frequency of CCR-5
deficiency in these individuals.
now, you might say how big is big?
This odds ratio of about 4 to 5 is very large and in fact if you look at it compared to what
I already told you is the very strong association of CCR-5 deficiency with the HIV
exposed uninfected phenotype you see you get an odds ratio of six here which is
comparable to what we found in west Nile virus.
another way of looking at this is if to ask a question how strong an association is this, is
that we found in Arizona and Colorado about 400 symptomatic cases of west Nile virus.
if you back calculate how many infecteds you would have to have to yield that many
cases, you get a number of about 2000.
and if you then convert that into the number of expected CCR-5 delta 32 homozygoats
1% of that would be 20.
in fact we found 17 homozygoats so we found almost all of the ones that we would have
expected to find.
so this is a very, very strong one could say screening tool for identifying CCR-5 delta 32
Find a cohort of individual whose have west Nile virus disease and you find a large
number of CCR-5 delta 32 homozygoats.
We expanned the study as it's important to do in these kind of epidemiologic studies
because you can get fooled often.
we have expanded these studies, gene in the lab tested two other cohorts one from
Arizona, one from California look agent the odds ratios -- odds ratios plotted for these
various cohorts and the combine Meta-analysis.
you see a strong association odds ratio of about 4 in each of these cohorts with an overall
odds ratio of 4.3 per 620 individuals.
so this is a -- results consistent with what we found for the mouse study.
and it has some therapeutic implications.
so here we have HIV and West Nile Virus.
CCR-5 is facilitating HIV but restricting West Nile Virus.
if you come in with a drug to block CCR-5 it would have a positive effect on HIV, that is
promote decreased viral load and restrict disease but if you do that you might facilitate
the susceptibility to West Nile Virus disease in individual whose become infected.
will that happen?
We don't know right now but it's certainly something to consider and going forward this
is something that will have to be monitored as the large number of patient whose are
treated with MARAVAROC are evaluated.
so with one exception, that is West Nile Virus and CCR-5 all known human diseases
involving the chemokine system are linked to normal or increased function, not loss of
function which is kind of an interesting point.
so this is really the first beneficial role of a chemokine or chemokine receptor in
infectious disease pathogenesis in humans.
this is the another way of looking at it, CCR-5 has a YEN yang relationship with HIV and
West Nile Virus.
bad for you in the kiss of HIV, good in the case of West Nile Virus.
and it raises questions about whether CCR-5 might control over pathogens including
other related viruses.
So this is something that I think is very important and in the mouse system we know
CCR-5 can have some effect in the control of listeria infection, as perGILLAS infection,
others, in human there's only one other infectious disease where it's been suggested to be
playing a role in control of infection.
that is tick borne encephalitis virus, which happens to be a flavivirus and as Amy told
you at the beginning West Nile Virus is a flavivirus.
so we now have two flaviviruses that appear to be -- for which CCR-5 delta 32
homozygosity appears to be a disease susceptibility factor.
so what's the epilogue of this?
I showed you the human data and what does this say in it says that the glass is 95%
empty and 5% full.
because only about 4 or 5% of the West Nile Virus cases we studied in man are CCR-5
95% are not.
so you might say how does this -- why is this so different from what you observed in the
Of course you know, most individuals who become very sick with West Nile Virus go to
intensive care units and get supported and don't die so we didn't see a strong association
of CCR-5 deficiency with the end point of death and it suggests there might be other
factors maybe at the genetic level that are also restricting West Nile Virus disease.
so gene LIM in the lab has looked and will glass when he was here looked for those
He and she have tested other chemokine receptors that are expressed in the brain of West
Nile Virus infected mice.
As you can see that we have found some effect for CCR-1 and CX 3 CR-1 on mortality in
the mouse model of disease but nothing like with CCR-5.
gene has gone on to test CCR-2 where there seems to be a more profound effect but
interestingly the CCR-2 Ligand, CCL-2 when mice are deficient in that Ligand we don't
see any effect on mortality.
so it maybe one of the other Ligands for CCR-2 that appears to play a role in the CCR-2
dependent effect on mortality in this model.
and there are other known West Nile Virus restrictions genes identified by others
including chemokine named CXCL-10 by Robin Kline.
TLR-3 by Richard FLAVELL's group has been shown to affect outcome in West Nile
Virus in the mouse model.
I told you about gene's work on CCR-2, it's been known for a long time that there is a
general tuck restriction or genetic susceptibility gene, strong susceptibility gene in mouse
strains, for not just West Nile Virus but many flaviviruses.
so this genetic locust was originally named the flavivirus locust and recently FLV or the
flavivirus locust has been found to actually be a polymorphism in the OAS-1D gene in
the mouse where OAS-1B is a subtype of an oligoaden late synthetase.
I'll tell you more about that in a second.
I'll show you some of gene's recent work on OAS-1 which is the human homolog of
so here's more information about FLV.
in inbred mice are sensitive to west Nile virus, wild mice are resistant and that defines
FLV has been mapped to mouse OAS 1B T homolog of human OAS-1.
a nonsense mutation in exon 4 of OAS-1.
so the question was could we translate this information from the mouse in OAS-1B to
So this is some information about OAS.
again, I mentioned to you it's oligoaden late synthetase.
what is this?
One of the mechanisms by which type 1 interferon or the interferon alpha that Amy
mentioned executes a anti-viral effect on cells.
these are some of the other ones.
But oligoaden late synthetase when activated is able to convert ATP into oligoaden late.
and these become factors that can activate RNAL to conform which is capable of
degrading RNA including the RNA of flaviviruses which are RNA viruses.
so in humans this would be OAS gene cluster, there are several related genes in the
in mouse there are even more.
it's a very complex cluster of genes.
OAS-1 appears to be the related gene in the human to mouse OAS-1B and gene limb in
the lab has looked at this locust and found a number of polymorphisms there that some of
which are very interesting including this one which is boxed in red.
the reason why I'm highlighting that is because this is a single nucleotide polymorphism
which affects a splicing site in the gene for OAS-1.
and we're going to call this one where there's a G in this position the G allele which
allows splicing here.
and this encodes an enzyme of 46 kill la Daltons with high enzymeATIC activity.
over here the G converts to A, so it's a the A allele which doesn't slice here but later on
this encodes a protein with 52 kill la Daltons with -- kilo Daltons with low enzymatic
since this is a common polymorphism one can look at its distribution in the same cohorts
of West Nile Virus infected individuals that I showed you before.
so when you do that, here are individuals who are homozygous for the A allele in blue
and these are random blood donors.
so in this study you see that there's about a 35 to 65% distribution of these genotypes AA
versus individuals who have a G allele either homozygous or hetero zygous.
and a cohort of individual whose are West Nile Virus sere row negative you find the
same distribution of these genotypes.
In contrast to individuals who are West Nile Virus sere row positive symptomatic West
Nile Virus seropositive individuals where there's a skewing of the expected genotype to a
so one can convert that to odds ratio which is what you're looking at here and gene has
done this to all of the cohorts that she's collected.
and you see that there's an odds ratio of about 1.7 to 1.8 across these four cohorts.
So there's a consistent association of this polymorphism, this functional polymorphism
with symptomatic West Nile Virus disease.
This is the combined analysis over 312 individuals with a strong P value.
we want to go on and test whether this polymorphism affects whether we can show it
affects some other aspect of West Nile Virus pathogenesis.
so gene looked at its ability to influence West Nile Virus replication.
in human lymphoid tissue.
again remember the virus comes into the skin, goes through the draining lymph node and
NCI collaboration with MARGOLES group was able to obtain tissue as surgical excess
from otherwise healthy humans and look at the ability of West Nile Virus to infect these
to do this, it seems to me has to cut the tissue into small pieces and put on the collagen
beds and then can actually directly inoculate the tissue with virus.
and ask whether virus replicates or not.
so this is immunohistochemoindustry showing an antibody for West Nile Virus.
you can see in this section of human lymphoid tissue the virus is detectable in this tissue
suggesting that it's replicating.
gene went on to look at 20 tissue from 21 individuals and looked at the ability of the virus
to replicate over time and there's actually a lot of information in this slide.
the first is that tissues from all 21 individuals supported replication of virus.
this is the first pathogen for which that's actually been the cases in system.
the second point is that the virus replicates the very high levels so you can see three to
four log enhancement of viral replication in the system without exogenous activators or
and the third -- from the point of view of the rest of this talk, most important point is this
tremendous variability in the ability of these tissues to replicate virus so there appears to
be some heterogeneity in tissues from different individuals to support viral replication.
so this is actually a very nice property of this system.
and fortunately gene had saved DNA from the tonsil tissue from these individuals and
was able to go back and genotype all 21 of these individuals for the OAS-1
polymorphism that I told you about from the epidemiologic level.
and look at the distribution of viral replication as a function of OAS-1 genotypes.
That's -- here are the data, she did this for three different days and found that there was a
consistent association of the OAS-1 single nucleotide polymorphism association with
West Nile Virus refullycation in human lymphoid tissue in this ex vivo system.
so so these are the AA homozygoats.
These are individuals who have low levels of OAS-1 activity or one would say type 1
interferon activity or anti-viral activity.
and consistent with that idea you see high levels of viral replication in these tissue
relative to other tissues in individual whose have the G allele, the form of OAS-1 that has
higher levels of OAS-1 enzymatic activity.
consistent throughout this nine daytime course and statistically significant throughout.
so the conclusion from this is that there's a common functional polymorphism in OAS-1
at the slice accepter site that's a genetic risk factor for West Nile Virus consistent in
humans with odds ratio of 1.7 over four cohorts.
West Nile Virus conducts lymphoid tissue in an OAS-1 manner.
the data strongly suggests that endogenous type 1 interferon regulates West Nile Virus in
I want to emphasize this is endogenous, this is an endogenous conclusion because there's
really very little data in humans for an effect of endogenous type 1 interferons on viral
most of the data come from mouse models of infection or from the use of exogenous type
1 interferon to affect viral replication as in the case of hepatitis C virus for instance.
and based on this we would say that the data support the investigation, further
investigation of type 1 interferon as a therapeutic in symptomatic West Nile Virus
infection in the future although as we have heard from Amy this is a very difficult
with that I will stop and just credit again gene limb and will glass who have done most of
this work along with Dave Mcthermon who helped a lot with the epidemiology.
and we got the virus originally from Alex PUTINOV in this building and did the lymph
node work with Andre LUSKO and lonIA MARGOLES NICHD and got help from the
comparison medicine branch, Jerry ward in particular and from the department of health
services from California, Illinois, Arizona and Colorado, without whom the epidemiology
work could not have been done.
Kara Glaser, Bernard Johnson, John pate and Ron CHESSIRE respectively so I'll stop
there and answer any questions you might have.
>> Well, if you have any questions from the audience, please go ahead.
>> If I understood you correctly, for the 22 homozygotes to show a high level of West
Nile Virus disease symptom, if you had some kind of agent that stimulates (inaudible)?
>> You're saying if you could heighten CCR-5 activity?
>> I don't think so.
I think in fact it would make them worse.
and the reason is in order for CCR -- in order for CCR-5 to work the way we think it's
work you need to have a gradiant of chemokines from the brain to the blood.
this is our vision of how it's working.
so the Ligands are produced in the brain, there's a gradiant across the blood brain barrier
to the blood.
and then that is able to attract cells up the gradiant to the brain.
if you somehow enjected chemokines into the blood you would destroy that gradiant.
it would be re-- gradiant.
it would be reversed and you would have a harder time getting CCR-5 positive cells into
if it's working the way we think it's working.
>> I wonder if you could tell us if there is a connection between the OSI polymorphism
and other types of virus and which would be a potential target?
Potential viruses that could be -- have a different or at least similar RNA degradation
it's -- well, the OAS-1B polymorphism, the mouse, is associate -- the flavivirus resistance
so any flavivirus that's been tested infects susceptible mice in an OAS-1B dependent
manner so it's not just West Nile Virus, it's other flaviviruses as well.
in human we don't know.
but that would be I think that -- certainly it will be worth looking at other fly Sri viruses
>> The medical doctor before talked -- told us different people get different symptoms.
when infected with this virus.
can you ascribe them to the different genotypes of the patients or of the virus?
>> We try to do that but when we first started doing this work we started looking at the
most extreme phenotype which is death.
we did have an association of CCR-5 deficiency and death in that initial study.
But we haven't been able to consistently show that in other cohorts.
so I would say that -- I would say that it's not really clearly associated with that.
that's a very difficult thing to do because as I pointed out and as Amy pointed out, the
percentage of subjects who die is actually relatively small.
so the number of individuals in these groups who have fatal infection is low.
and the lower the number the harder it is to do meaningful epidemiologic studies.
so I think we need to -- it needs to be addressed further.
there maybe some association with death.
we haven't been able to show a significant association with other outcomes as well.
it's just the one outcome writ's clearly and consistently associated as symptomatic West
Nile Virus infection to take anyone who falls into that category.
>> Could you speak to other chemokine receptors?
They're -- I looked up and found that the CCR-5 as co-receptor in the HIV process, have
you looked that the or do you know if that also plays a role in the West Nile Virus?
Does CXCR-4 to CCR code receptor for entry?
>> You're asking about CCR-5 or other --
>> Other co-receptors -- other chemokine receptors.
I mean since you've done all this work and you show that CCR-5 is playing a large role,
could other chemokine receptors similar to CCR-5 also be playing a large role and have
you looked at any of those?
>> We have looked at those.
there were two slides that actually I showed on other chemokine receptors in the mouse
model so we look at CC R-1, CX 3 CCR-1 and also CXCR-3.
You can show a rank order of -- in the mouse model where CRR-5 is clearly the
CCR-1 would be next, CX 3 CCR-1 would be about the same.
I'm sorry, CCR-2 is next and then 1 and CX 301 would be trailing down the end.
we haven't been able to do that in humans because of a lack of appropriate genetic
polymorphisms that you can use to do epidemiologic studies like the ones I showed you.
there is one actually for CX 3 CR-1.
that's worth looking at but at least in the mouse model there wasn't much of a signal
there, could be different in humans.
>> I had a question to Dr. Agrawal.
perhaps you can fill us in if there was -- have been any cases of reinfection with virus in
human model or human disease?
>> There was case of an immunosuppressed child who appeared to recover from an
indetection and acquired it again and was positive but that's exceedingly rare and nobody
else has reported that.
>> Is there an association between age and a person that develops ParkinsonIAN signs?
>> -- and a lot of this has been case series and he's the one who looked and followed it
if you remember that was 16 patients.
so -- I have talked to him about that data and that's nothing we have talked about.
I don't remember if we did a published gradiant by age.
so I don't think so.
but I wouldn't want to say that there for sure isn't.
>> There's sort of a feeling that from both of you is that the system participate in all
levels from t-cells to NK cells to probably antibody level.
and so where is the clue?
Where is the key to this virus?
What do you think?
>> Well, I would certainly at least in the I would say CCR-5 is a major factor.
at the cellular level we don't -- we don't know.
we don't know what the key CCR-5 positive cell type is in the brain that appears to be
able to protect mice from lethal infection with the virus.
and the cellular studies that are published already don't help us because just about any of
the cell types could be the relevant CCR-cell types.
and that if you eliminate CD4, CD8 or NK or macrophages they all play a role when
they're totally eliminated from the mouse.
so we need to look, just don't know.
>> -- to the lymph node?
Is it dendritic cells?
>> Yeah, I -- that would be the notion, it's probably longer cells in the skin but it could be
free virus too.
>> And the drug, how does it work, the (indiscernible) you mentioned, is it anti-viral
drug or what is the mechanism of action?
>> I mean, no, the neat thing about MARAVARAC, maybe the most noteworthy thing
about it is it doesn't target any element of the virus.
it targets a host factor, CCR-5.
so it's a classical big Pharma antagonist to a G protein coupled receptor.
it blocks CCR-5 exploited by HIV to enter cells and in blocking CCR-5 it blocks access
of HIV to CRR-5 which it needs in the cell.
>> Must be --
>> Well, you could think of it as maybe being useful in preventing infection if you have
been -- you have had acute needle stick injury or something something like that.
it's going to be generally used in patients who are already infected with the virus.
and they are the ideal -- there the idea is to try to reduce viral load as you would with any
other anti-retroviral drug protease inhibitor or reverse transcriptase inhibitor.
>> Any other questions?
>> Two questions.
You said just to reiterate there's no phenotypic difference between humans that are
deficient or have the delta form of the CR-5, and also -- I for get my question.
>> To answer your first question while you try to remember your second.
what I said is that there isn't phenotype in individual whose are missing CCR-5.
that is much higher susceptibility to West Nile Virus -- to symptomatic West Nile Virus
disease if they get infected.
>> Other than that, there's --
>> Other than that, no one has found a -- well, now tick borne encephalitis as well.
Other than that, no.
they have one head, two arms, they -- anatomically and physiologically the otherwise
>> And have you knocked out CCR 5 Ligand in mice and see if that affects their
susceptibility to West Nile Virus?
>> We haven't done it but others have done it for us and gene has checked some of the
CRR-5 Ligand knock-out mice and has not found the same phenotype for the Ligand.
you have to remember that the receptor has -- this is one of these inflammatory receptors
that has a promiscuous Ligand profile with many different Ligands so it maybe that you
need to knock out all of them or we just haven't knocked out the right one yet.
>> In reference to your survival analysis, have you looked into other strains, is there a
particular reason why you chose the C-57 background?
Would you expect to get similar results in a Bob C or AJ?
>> I don't know.
we have not looked at any other strains.
Reason for that is the -- all of our knock-outs are on C-57 black six.
so -- black six.
so most knock-out mice are on C-57 black 6.
so from a technical point of view that's the strain that's most feasible to analyze.
to do others you would have to go and do a lot of breeding.
I think that could be done, not sure it would be worth doing because the information that
we've got right now is that there's other strains that are also susceptible to West Nile
Virus and we've already translated it to humans.
so I think we have the strain that we want to do the mechanistic studies.
>> Can you tell us anything common among individuals who are CRR-5 sort of human
knock-out, is there a certain advantage evolutionary that you can assign to that group of
Are they all from certain part of the world or certain genetic extraction or anything
common among them?
>> That's been widely debated.
It's been on TV in fact.
and so the idea is that the allele frequency is not homogenous.
as I mentioned, it's -- it was found -- the allele was founded in Caucasians.
so it's not found in Africa, it's not found in Asia.
accept in very, very rare numbers of individuals.
and it's not found in Native American Indians.
so in the United States you find it predominantly in Caucasians.
You do find it in some African Americans and some Asians in the United States but not
in Africa or Asia.
so the question is why was it -- once it was founded why was it retained?
And that's a very interesting question.
if you look at Caucasians and go across Europe and look at the distribution of CRR-5
delta 32 you find that it's most common in northern Europe, Finland is I think number
there's a decline A gradiant of CCR-5 delta 32 frequency as you go south.
so it's -- there aren't that many -- it's low in Greece and other Mediterranean countries.
so people tried to correlate that with different plagues that are known from history.
and there was an idea that maybe it has to do with the plague but that hasn't held up based
on for one CCR-5 knock-out mice are more susceptible than wild type mice
(indiscernible) and number two there was an excusemation of bodies in Italy from
individuals who I think had died from the plague back 1600s and the frequency of CCR-5
delta 32 was the same in the -- in those individuals relative to expectations from modern
so it's probably not plague but it could be any number of other things.
Like smallpox, there was an argument for smallpox that came and sort of went.
one could imagine ancestral plagues that burned out that are no longer -- that have no
>> Maybe there was another virus that disappear and we never know about it.
there was even been a paper within the last year suggesting that it had nothing to do with
positive selective pressure.
that the drip of CCR-5 delta 32 throughout evolution is no faster than any other random
so -- there was actually some good data in that paper.
so it's still being widely debated.
of course the intuitively the most comfortable answer is that it was positively selected for
but we just don't know what that may have been.
>> Any other questions?
Well, thank you, both of you for very very stimulating talks and wonderful discussion to
the audience and we'll see you next week again.