Autism Research Update: Implications for Practice

Autism Research Update: Implications for Practice

(electronic tones beeping) – Dr. Hansen is the Director
of Clinical Programs at UC Davis Mind Institute, and Professor in Chief of Developmental Behavioral Pediatrics, Department of Pediatrics, and Director of the Center for Excellence in Developmental Disabilities
at UC Davis Health Systems, and a good friend. (laughs) (audience applauding) – Where’s my clicker, here? – And this is your pointer. Oh, I forgot to say what
you’re gonna talk about. – That’s okay, my slide’s up. – Good.
– It’s right there. So it is nice to be back, and I realized as I was
thinking about my updates that I will be updating
you on some of the research that’s evolved since the last time I was here presenting on
autism research updates. So I have nothing to disclose,
financial disclosures. I do have to disclose that the day after I sent my slides in for the syllabus, we had a new study that came out online, so I’m gonna talk about it. So this is not in your syllabus. I think those of us who
treat patients with autism and certainly those of us doing research in autism spectrum disorders understand how complicated they are
and how heterogeneous they are in terms of presentation, certainly in terms of etiology, and increasing in prevalence. Our research has really
tried to understand what are some of these etiologies? Can we identify biomarkers? Can we find endophenotypes
or subtypes of autism that help us in understanding
potential targeted treatments so that we can do a better job in not only early diagnosis
but early treatment as well. So as I said, I’m gonna try and highlight
some translational research that’s been ongoing at the Mind Institute, I realize for about 20 years, and some of the most recent research that we have that primarily relates to looking at
gene-environment interactions as well as taking us
along this path to looking at immune dysregulation as
a risk factor in autism. So the learning objectives are one, to understand the current
prevalence estimates of autism in the United States, to understand or identify four maternal metabolic conditions
that we think increase the risk for autism as well as more generalized developmental delays. To understand or list the
clinical characteristics of a subtype of autism called maternal auto-antibody
related autism or MAR autism, and to list some of the
core and co-morbid symptoms associated with some different immune endophenotypes in the children themselves. So we are fortunate in the United States to have the CDC doing
ongoing prevalence studies across the country at various sites, and each time that they report out, which is about every two years, the prevalence of autism has increased and we don’t know how much longer it’s going to keep increasing, but right now it’s currently
one in 59 or 1.7% of children, and this work is done where they evaluate multiple records at eight years of age. So it’s up from the last
time they reported it, which was one in 68, and one thing I think is significant is that less than half of these children, looking back, were diagnosed by age four, although most of their parents
had concerns way before this. I think it really pushes us to try and figure out ways to
do earlier diagnosis so we can really get earlier treatment. This is a schematic
just to start the story of the research that we’ve been doing and I’m not quite sure
how to use my pointer but it all started with
a study that was funded by NIEHS and the EPA
called the CHARGE study, and this was really as we were beginning to try and figure out why
would a highly-heritable, at least genetically-based disorder be increasing so rapidly? There’s gotta be something
else that’s changing because genetic disorders
don’t change that fast. So can we try and figure
out by looking at multiple, multiple, in a population-based
case-control study, can we look at the shotgun approach? We really weren’t particularly
focused on looking at immune dysregulation but
it was one of the things that we figured we should be looking at, and things in the environment
that we knew were changing, many of which are
potentially very neurotoxic. Something that came out of that, so this was primarily
a retrospective study. I’ll show you another slide that goes into a little more detail, but an outshoot of that
then was looking at a study, we developed a study called MARBLES, Markers of Autism Risk in
Babies – Learning Early Signs, so this is a population of children who were mothers who were studied from the time that they
were pregnant or before they were pregnant who already
had a child with autism, so we knew that they were at greater risk for having another child with autism. So these are all co-collaborators and we all lived at the Mind Institute and at the Children’s Center
for Environmental Health. This is just another schematic
to show you that in CHARGE, these kids were two to five years of age. We had three groups. We had kids with autism,
with a diagnosis reconfirmed, kids with typical development, and kids with developmental
delays who did not have autism. We looked at a lot of biological factors. We looked retrospectively at
a lot of historical factors. We had newborn blood spots to look at, blood findings at the time of birth, and then this is MARBLES, so you can see this is
a prospective study, as I said, of children, of pregnancies, and then the children out to age three where we looked at a lot of
the same biological samples, we did a lot of the same
environmental questionnaires, except that it was prospective. All right, so that sets the stage. One of the things that
we found and reported on and I think I presented
at a previous study but we’ve got some
updates is the important of folic acid and reducing
the risk of autism. So this was in CHARGE, this was retrospective
data asking the mothers, did they remember, did they take prenatal
vitamins during this pregnancy, when did they start? And what we found was that
there was about a 40% reduction in the risk of having a child with autism if mothers took prenatal vitamins which have extra folate and preconception or at least in the first
month or trimester. This has been replicated several times. We didn’t have any studies that looked at, were there actual differences
in the maternal blood folate at the time of birth or
in the first trimester, and then the other thing
that was interesting about this that allowed
us to look at this impact was primarily on women
who had a genetic variant that made them inefficient
folate metabolizers, so in order to have what we
think of as normal folate, they needed to have extra folate on board because they were slow metabolizers. So it was this first clear
gene-environment interaction that we found in one of
our studies in CHARGE. This just shows you
that in this next study, we were able to actually
measure the difference in maternal folate between
mothers who had children with autism in red and then
mothers who had a child with typical development out here. So this was at the time of
the first month of pregnancy. So now this study is from MARBLES. So this is one of the first studies out of the prospective
study looking at children who were followed out to age three who ended up with a diagnosis of autism, non-typical development,
or typical development. So a lot of children who are at risk probably from shared
gene-environment interactions for autism may not end up
with a diagnosis of autism, but have some delays in terms
of their language development, have some behavioral
concerns around attention, emotion regulation, and sometimes we call that
the broader autism phenotype. It’s not quite autism
but it’s getting close, so that’s what we’ve
called, in this study, non-typical development. And you can see that the mothers who used pre-natal
vitamins in the first month were about half as likely to have a child on the autism spectrum, so there was a big, big difference here. No significant differences, it’s maybe a little bit different in terms of non-typical development, but clearly having extra folate on board in the early, early part of pregnancy was really important and
this is a group of mothers, oops, sorry, this is a group
of mothers that we know are at risk and a group
of fetuses are at risk because they already have an
older sibling with autism. In this study, about by
the time of mid-pregnancy, about half of the moms were
taking pre-natal vitamins. In preconception and in the first month, only about 1/3 of the mothers were taking pre-natal vitamins. By the end, almost everybody
was taking pre-natal vitamins, so it looks like this
is a critical window, when extra folate is really important. Okay, so I’m gonna stay with
the maternal environment, but I’m going to shift to another group of studies that we’ve done looking at metabolic conditions. These are all clearly disorders
that impact an increasing number of women in the United States. You can see how common diabetes, other related metabolic
conditions like obesity, chronic hypertension, metabolic syndrome, and what I’m adding here
is the pre-eclampsia, and what all of these metabolic
conditions have in common are that they have underlying
chronic inflammation which then causes insulin resistance, and just in the course
of normal pregnancy, insulin resistance increases anyway, just because of the metabolic changes that go along with pregnancy. This is a flow chart
looking at our schematic in terms of why we thought
this was important to look at in terms of some of the potential impact on fetal brain so we know that
maternal metabolic conditions put moms in a pro-inflammatory state, so they have elevated levels
of pro-inflammatory cytokines. These cross the placenta. They stimulate fetal
immune cells in the brain which are really important
for early brain development, so you have a pro-inflammatory state in terms of fetal brain, then the other mechanism in
terms of insulin resistance is when you’re resistant to insulin, maternal glucose levels go up. This stimulates fetal insulin to go up. Stimulates increased fetal growth, more oxygen consumption and
the potential for fetal hypoxia as well as iron deficiency, and this is particularly
important if we think about pre-eclampsia and placental insufficiency. So the questions that we were asking is can we show that these
metabolic conditions during pregnancy increase the likelihood of having a child with
ASD or other delays? So you could see, in terms of the initial
study that we looked at, that both children with
ASD and developmental delay but not autism had a higher likelihood of having mothers with
any or a combination of these different syndromes. And you can see how significant, actually, these risk factors are, both for delayed development but autism specifically as well. So when we looked at pre-eclampsia, we found that the same thing was true, that in women who had pre-eclampsia, they were twice as likely to have a child on the autism spectrum. When we looked at children who ended up with a diagnosis of developmental
delay, but not autism, it wasn’t really until there
was very severe pre-eclampsia and placental insufficiency
where we saw this spike in terms of impacting fetal outcomes. So again we’ve added
pre-eclampsia to one of these risk factors that we think is
very important in increasing or at least mitigating risk for autism. So now I’m going to shift
gears and talk about the other side of the immune system, so the cellular immune
system, and what we found, one of the observations that we have found initially in the CHARGE study early on was that some women develop antibodies that have a target on fetal brain, and unfortunately these are IgG proteins which means they can cross the placenta and their targets happen to be circuit fetal brain cells and their impact seems to reduce the development of dendritic spines and as we’ve looked at
children whose mother have these antibodies and again, this started out in the CHARGE study, about 1/4 of the moms who
had children with autism had these patterns of antibodies
with fetal brain targets and we didn’t see them in the mothers of children with developmental
delay and we didn’t see them in mothers of children
with typical development. And as we continue to look, we actually realize
that there’s particular some characteristics of these children. Not only do they have
autism but they seem to have distinct subphenotype where
they have more severe behaviors, they tend to have higher autism severity. Lots of stereotypical
behavior, and in boys at least, they tend to have larger
brains, so macrocephaly. And the story of how we developed or took this information
I think is a really interesting story in terms of translation and the importance of
having both clinicians and basic scientists on your team to really develop this
story and understand what it is you’re observing clinically, what those mechanisms might
be and then how can you use that information to get
back to clinical targets? So I’m just gonna walk you through, as I said we just, on
Western blots, early on, we saw that these
different patterns existed in some of the mothers who
had children with autism. We weren’t really sure what their clinical significance was but we
thought they probably were significant because
they weren’t in other moms. And it was really hard, this was when studies were first starting to really think about or
look at the immune system and it’s all over the place,
continues to this day. It’s hard to replicate
some of these studies. But one of the things that
I think really helped us along this path is we
took antibodies from moms who had kids with autism and
had these antibody patterns and in four different animal models, used these antibodies,
these maternal antibodies, and injected them in
two early mouse models and two primate models
and this intervention resulted in demonstrating that both mice and monkeys develop symptoms that
are similar to at least some of the behavioral symptoms
that we see in children. They had social abnormalities, they had lots and lots
of repetitive behaviors and a lot of self-harm. So we had some proof of concept here, and then some of the other
things that we started figuring out is that many of these women who had these antibodies
also had a genetic variant in the C-MET gene that
seemed to be associated with producing these auto-antibodies. We also found again by
observation, clinical observation, that a lot of the boys whose
mother has these antibodies and who had autism had larger brains, and then we also found
that mothers who had these auto-antibodies
were more likely to have metabolic conditions during pregnancy, so again, tying both of these
lines of research together. So the next step was
really trying to figure out if we could target what
are these antigens, and where are their specific targets? And this really took a lot longer than we thought it was gonna take. But we finally identified the proteins and they were quite common proteins. LDH, CRMP1, and STIP1
which are very important in terms of early neurodevelopment, and so Jackie Crawley
and her team developed a mouse model and injected
these specific proteins, not from maternal serum, but developed these proteins and developed this mouse model of autism. So again, another proof of concept. The step clinically then
is to try and figure out how can we screen for these antibodies, identify them in a clinical way that’s valid, that’s efficient, that you could do a lot of screening. And this has been
complicated because a lot of the recombinant proteins that are used in the ELISA assay have His tags on them, and it turns out that a lot of vaccines also have this tag, so you get a lot of non-specific binding in women who don’t have these
antibodies but have His tags. So it’s taken a very long time
to get this assay completed. Lots of stops and starts, but I think we’re getting very close to having a clinically-available screen for these auto-antibody patterns. Okay, so this is my last shift, I hope I have enough
time to do this, I do. Sorry for running through this so fast, but now I want to just look
at some immune dysregulation in the children themselves, so all of this has been related
to what we’ve understood in the maternal environment and
with maternal dysregulation, so this is actually from another
study at the Mind Institute called the Autism Phenome Project. These children were between
the ages of two and three. They had a diagnosis of autism that was confirmed and
we had a comparison group of typically-developing boys
as well of the same age. What we found is that
there were differences across these groups in their patterns of immune response to stimulation, and particularly when we
looked at within the group of boys that had a diagnosis of autism, and stimulated their mononuclear cells with lipopolysaccharide which usually just causes a upregulation or a
pro-inflammatory response. We found that some kids in this group were high responders in these dark bars, and kids that didn’t quite respond, they didn’t get such a great
response as this group did, so we call them high and low responders. And then we looked at autism severity, so this is from the ADOS, the Autism Diagnostic
Observation Schedule, and you can see that the
boys with the high response had much more impairments in their social affective relating. No real significant differences in terms of repetitive behaviors and not clinically statistically significantly different but some increase in their
overall autism severity. And when we looked at, oh they also actually had lower overall developmental quotients and particularly in the non-verbal developmental quotient. No significant differences in
terms of their verbal skills. Then we looked at correlated behaviors, sort of co-morbidities, and again, found significant increases
in the high responders in terms of sleep problems
and aggressive behaviors. Some suggestion that
there’s some increased attention problems but again
not statistically significant. So what this may help us
develop is trying to be able to segregate not just by
diagnosis but by subtype, particularly as we’re trying to do more targeted clinical trials. It may be that when you
look at a group of boys who we all know have autism
and we all can agree upon that, and some seem to respond to a particular targeted treatment and others don’t, and so it all kind of
washes out in the middle, so you don’t find statistical significance or clinical differences, it may be because these kids
have different etiologies and need different targeted treatments. So this is just one way of
thinking about how we might subdivide groups when we
look at clinical trials. So just in summary, I think
immune studies and autism really do reflect the
complex heterogeneity both in etiology as well as
trying to sort things out in terms of gene-environment interactions that are very complicated, and our line of research would suggest that it’s both maternal
and child immune systems that may be very important
in subtyping children as we try and develop
more targeted treatments. They may also be early
biomarkers that help us look for risk and hopefully, we can get towards prevention. So I think making sure that
women of child-bearing age understand the importance
of taking pre-natal vitamins when they’re planning a pregnancy, before they get pregnant, is a public health
message that is important and I think trying to address
some of the health factors that contribute to metabolic
conditions as best we can is also an important
aspect of autism treatment or perhaps prevention. So thank you to all the families that have participated in these studies, to our funders, and to
my co-collaborators. So thank you. (audience applauding)

One Reply to “Autism Research Update: Implications for Practice”

  1. huh. well, that is significantly closer to "vaccines cause autism" than i expected the research to go. still, more work to go. & ive never heard of this immune involvement before. & confounding by recombinant vaccines makes testing difficult. i suppose that would be true w/any immune related condition. Prospective folate needs a control, ideally, bc i dont know how to interpret that one prior sibling OR biz. Or why ORs at all. But pretty solid & interesting stuff. Better than most people, in a lot of field! Keep it up. ^^

Add a Comment

Your email address will not be published. Required fields are marked *