Dr.  Stephen Cunnane – Brain Glucose and Ketone Metabolism

Dr. Stephen Cunnane – Brain Glucose and Ketone Metabolism

November 2, 2019 21 By William Morgan

– Thanks Ken and good morning everyone.
It’s a great pleasure to be here and
I want to thank the
organizers for this invitation
and for the energy in the room
and the progress that we’re making.
This is such an important area
and I hope that we can
contribute a little this morning,
Eric and myself, to the
neurological aspects.
I’ve extended the title of my talk
from what’s in the program to
focus on Alzheimer’s disease
and the implications for treatment.
We’ve developed ketone PET imaging
and I’m gonna tell you a little bit about
how we use that to try to
establish potential, the
therapeutic potential of ketones
in this important disease.
I have some disclosures to mention.
On the left side of my
public and nonprofit funding
and you’ll notice at the top,
we were talking about sources of funding
some of it’s gone from Canada to the US
for some of the cancer work,
well the American Alzheimer’s Association
has supported the trial that I’m gonna
talk to you about today.
So it does go both ways,
and there’s some private-sector support
and following Tim Nolks’
experience that he
so eloquently described yesterday,
I wanna mention in the bottom right there
that I’m not a physician and I’m not
providing any medical advice to anyone
from this presentation.
So we’re talking a couple
of people have referred to
Jeff Volek’s Ford and
I just wanted to show
a picture of it here.
(audience laughter)
It’s got a bigger
reputation than it really is
and little do you know that
it’s actually a Toyota Prius
which is dressed up as a performance car.
No, I’m making allusions to hybrid cars
because I think it’s a good
analogy for the human brain
which is what I want you to think about
in terms of its fuel requirements.
There is an interchange normally
between the ketones used by the brain
and the glucose used by the brain,
and this becomes faulty as we get older
and I wanna tell you a little bit about
how we explore that.
And I wanna start with the beginning.
And the beginning is infant development.
We’ve made the allusions
over the past couple of days
to the importance of ketones in starvation
which is a secondary
benefit of having ketones.
But in fact,
the infant brain cannot develop
normally without ketone.
So it’s not an alternative
fuel, it’s an essential fuel
to supply 20 to 25% of the
infants brain energy requirements
day in day out physiologically.
It’s not a question of caloric restriction
or intermittent feeding
or whatever you like,
it’s a breast-fed healthy term infant.
And part of that ketone
supply to the brain
is fueling the energy,
but part second point is supplying carbon
to make brain lipids,
which is in the image there
which if any of you do brain imaging
you’ll realize that’s actually a human,
an adult brain rather,
that’s superimposed on a CT of an infant,
or probably an ultra scan actually.
And the white part of the
image is the white matter
and the white matter is
made of primarily of lipids
that are made from ketones, cholesterol,
long chain fatty acids.
The myelin, the sheaths around
the axons and the nerves.
So it’s an important anabolic substrate
for the developing brain,
and it’s important catabolic
or fuel substrate as well.
The main source of the
carbon supply the ketones
that are constantly coming into
the infant during lactation
is the medium chain
triglycerides that are in milk.
Human milk, all mammalian
milk that we know of
contain medium chain fatty acids
and in principle, I mean that the infants
of most or probably all mammals
are in fact in mild ketosis
during the lactation period.
The advantage in humans
is that some of those
medium chain triglycerides
end up in storage fat,
in the body fat of the infant.
We are unique as a mammal
in having body fat at birth
and so this in a way
extends the ketogenic state
beyond weaning because
those MCT can then be used
from body fat,
and I’ve discussed this in
the implications of this
for human evolution both in
the book that Ken referred to
and in this article, the
general human evolution
a of couple years ago.
So I think the important point
I wanna make here is that
what we’re trying to do is
recapitulate in the aging brain
what the newborn brain
is used to and thrives on
day in day out.
The second I think important point
that I wanna make is that the fuels
that are used by the brain,
there’s two different strategies.
I notice on my screen
that this partly cut off
but I’m glad to see this not
out off at the top for you.
So the two strategies are that the glucose
gets into the brain
because it’s pulled into the brain.
This little stick person is
pulling a glucose molecule
into the brain which is on the right,
and that processes occurs
through the glucose transporter
and it’s driven by a falling
glucose in the brain cell
that’s doing some job of
communicating with its neighbor.
So this is a process that’s
driven by brain activity,
that’s strategy number one.
It’s still supplying most of the fuel
that the brain is gonna use.
But the alternative strategy
is driven by ketones
and it’s a push strategy.
So when the ketones go up
in the blood on the left,
they are pushed into the brain.
And this is the way the
brain works all the time
if glucose goes down
because you’ve been fasting for 24 hours
for argument’s sake, insulin is down
and ketones are being produced
and they will go into the brain
in relation to the concentration
as shown on this slide.
So this is a compilation
of a number of studies
which I’ve compiled and in
the publication that’s noted.
You’ll see on the x-axis
the plasma ketones
are on a log scale
and the fasting indeed
goes from the postprandial
to 24 hours fasting, 40 day
fasting which is referred to
George K Hill’s classic work,
but there is also a paper
published in the Journal
of Clinical Investigation
on a 60 day fast in
obese individuals as well
where you got over 10 millimolar
of beta hydroxybutyrate in the blood
and you’re supplying, as
you can see on the y-axis
on the right-hand side,
about 85 to 90% of the
brain’s energy requirements.
Obviously it’s an extreme situation,
a non-physiological situation,
probably ethically impossible to do today.
All I’m using it for is to
demonstrate the principle
that as ketones become
available in the blood
under these circumstances,
they will be used by the brain.
So there’s a linear relationship,
over a 600 fold range at least
which is an extraordinarily
efficient way of providing
a backup fuel for the brain.
The third point I wanna make is that
glucose is the main fuel of the brain
under most circumstances
except under extreme fasting
but it’s not the preferred fuel.
So there’s a difference
between the main fuel
and the preferred fuel.
And this is an example of a study
that we did with a PET scan.
I’ll show you how we
did the pet in a minute
but I wanted to set the stage
for the Alzheimer studies
with these couple of preliminary points.
So on the x-axis, you can see a change
in brain ketone uptake
caused by the ketogenic diet.
So these are Delta values,
they are increasing amounts
of brain ketone uptake
compared to the pre-ketogenic diet stage
and on the y-axis is a glucose uptake
which goes down in
proportion to the increase
in ketone supply.
So glucose uptake is going down
when there’s sufficient ketones around
to go into the brain.
So this is an inverse relationship
and it’s why I refer to ketones
as being the preferred fuel.
We’re not the first
people to have shown this,
and there’s a very useful
glucose sparing effect
that in fact occurs on the ketogenic diet.
So the context for this
in relation to aging
and Alzheimer’s disease is
shown by these two little arrows
and if you look at the two scans,
you can see that the
one where the arrows are
has basically missing
the orange to red color
over the years, the parietal lobes.
This is a classic image of what one sees
in Alzheimer’s disease and it’s been known
since the dawn of PET
scans in the late 70s.
Interpretation is almost always been,
well the brain cells are
dying in Alzheimer’s disease
so they don’t need glucose,
so the glucose uptake goes down.
And that’s a logical interpretation.
Except for two points that I wanna make.
The first point, and it’s in this concept
that we’ve developed over
the past couple of years
along with some other
people around the world,
is that in fact the glucose
problem is specific to glucose
and it precedes the cognitive deficit
in Alzheimer’s disease.
So how do we know that?
What we’re talking about
is still the definition
of the glucose deficit is
defined by the PET scans
where the color, the
reddish color is declining
and that’s showing lower
glucose uptake in the brain.
In the yellow box, there are
five categories of people
with presymptomatic bring
glucose hypometabolism
and they’re all categories of individuals
at risk of Alzheimer’s disease.
Who are they?
Older people, people with
insulin resistance independent
of age, people with a family
history of Alzheimer’s disease,
people that are APOE four carriers
and the presenilin mutation.
So those are very
different groups of people.
They’ve all been identified as
having a risk of Alzheimer’s,
an elevated risk of Alzheimer’s disease
and they all have presymptomatic
brain glucose hypometabolism.
So this cannot just be a
consequence of the disease,
they don’t yet have the disease.
They’re all cognitively normal.
They’re at risk, but they
are still cognitively normal
when the PEt scans have been done.
I haven’t given you the citation.
I’ll to, if you look up our work,
all this work has been referenced
and has been done by other groups.
So the way we look at this now
is that this latent glucose hypometabolism
is in fact helping precipitate
the neural pathology associated
with Alzheimer’s disease,
deterioration of synapses
and deterioration in brain function
which is associated with cognitive decline
and which is gonna push down
the glucose metabolism further
and is gonna create a vicious cycle
of brain energy exhaustion and
progression of the disease.
So, we could argue about
whether the neuropathology
comes before the latent
glucose hypometabolism,
but we have a way of
assessing this and saying
if there are two fuels
that the brain can use
what about brain ketone uptake
in people at risk with Alzheimer’s disease
or at risk of Alzheimer’s disease?
If it’s the same pattern
as it is with glucose
and we know the neuron’s
completely screwed up
or the astrocyte,
because it’s a different
transporter mechanism,
it’s a different access
point to the Krebs cycle,
it’s a different pathway
until you get to the
common pathway.
So let’s explore
brain ketone uptake in these conditions.
So how do you do that?
You do what we’ve called
dual tracer quantitative
PET imaging, and the
protocol is shown here.
We do the PET image with a ketone tracer
which is acetoacetate, carbon
11 labeled, short half-life
so you have time to collect the image
before the radioactivity
is decayed away completely.
And then you have a washout period,
and then you do the classical PET FDG scan
on the same individual,
on the same afternoon
they’re in the same physiological state,
so that images are quite comparable.
You can do this qualitatively with,
well I think Parker referred to SUVs
standardized uptake values.
It’s a relative measure.
But it’s not good enough if
you’re trying to figure out
what the problem with glucose is,
the extent of the problem,
what your therapeutic
target with ketones is.
You need to quantify these scans.
And in order to do that,
you need blood samples
and you also need an MR image
to localize the PET image
and get the exact information
about what regions are affected
and what the extent of
the glucose problem is.
So this is what the scan is.
You can see the references,
we’ve published a few papers now.
We’ve studied over 300 people
using this methodology.
And the units are classically
known as CMR units,
cerebral metabolic rate,
is a quantitative value with
micromoles per hundred grams
per minute and they’re basically two units
to this calculation.
One is an uptake rate constant.
K which is taken from the PET scan,
and the other is the plasma ketone value.
And as many of you know now,
plasma ketones can vary a lot
if you’re on a ketogenic diet,
or if you’re on a ketogenic supplement.
So the results become fairly noisy
but the uptake rate constant
is actually a fairly
tight value depending on the conditions.
So if we look at the people
that are of most interest
during brain aging and Alzheimer’s disease
is CTL or cognitively
normal healthy older people
age matched to the MCI, which
is mild cognitive impairment
and AD, Alzheimer’s, and
we’ve got 20 people per group
in this project.
And these are the K values.
This is the capacity of the brain
to take up the ketones.
It’s kinda like how many doors are open
if someone calls fire here,
if there’s only two doors that open,
we’re gonna backup
before we can get through
everyone through the doorways.
If four doors open or six doors open
that’s a larger K value
in the analogy of a fire
and trying to escape.
And as you can see here with glucose,
the capacity of the brain to use glucose
is higher in the areas where
the red and orange colors
are shown on the brain, and
as you go towards Alzheimer,
you can see there’s a
virtual disappearance
of the red orange and the
green is more predominant.
So the capacity of the
brain to use glucose
decreases in Alzheimer’s disease
and this has been established
for many many years.
If we look at the ketone uptake
in exactly the same individuals,
you can see that the green color,
in fact the capacity’s
lower than it is for glucose
under the normal circumstances.
But it if anything, it actually increases
as you get towards Alzheimer’s disease.
There is no loss of the capacity
to transport ketones into the brain.
It’s the same cells.
So they can’t be dead
because this is an active
transport mediated process.
So this encouraging us to believe
that we could use the normal
brain ketone uptake capacity
to in fact bypass the glucose deficit
and potentially have an
impact on cognitive function.
So the issue then is well
that’s basal uptake rates,
what about uptake rates when you provided
ketogenic substrate?
Simple way of doing this is with an MCT,
a medium chain triglyceride.
We arbitrarily hit on a
dose of 30 grams a day.
We did this over one month
and this rather busy looking slide
is basically showing you
that the plasma values
are on the x-axis and
the star and triangle
are before they start the treatment.
So the values are lower
around .35, .4 total ketones.
When they go on either of
these two MCT supplements,
we basically double the plasma levels
and the brain values
are shown on the y-axis.
And you can see that we’ve
doubled the brain uptake
on the y-axis.
The key values are in the background.
The dotted line is what
the reference value,
the slope of the line that
we would ideally like to see
that has been shown in our studies
with young healthy adults.
And basically they’re identical,
statistically identical slopes
for the Alzheimer’s patients.
If we were to extend these plasma values
to much much higher values,
we might see that the slope of this line
connecting the blue
triangles, the green triangles
might drift off but it’s very
close to what it should be.
So this suggests that the brain
can use additional ketones
in Alzheimer’s disease
when you provide them.
So the important thing then is,
well what happens to memory and cognition?
And we’ve done a study in
mild cognitive impairment,
we call it the BENEFIC trial
which is for brain energy,
functional imaging and cognition.
And the cognitive outcomes
are summarized on this slide.
This study was designed and powered
for the metabolic outcomes.
Is the capacity to use ketones normal
in mild cognitive impairment?
And the answer is yes,
it’s the same results
as I just showed you
for Alzheimer’s disease,
but we still have 19
or 20 people per group.
And we have a placebo control on this
so we have cognitive results.
And you can see that
there are several domains
of cognitive function that improve,
particularly the episodic memory
but also processing speed and language
that do not improve on the placebo.
So this is encouraging,
but it’s not definitive
because a clinical effect
with a cognitive outcome
would require probably closer
to 100 people per group.
Importantly though, with the
brain ketone PET we can show
that there is a statistically significant
positive relationship
between the plasma levels
or the brain ketone uptake
and the outcomes on cognitive functions.
This is one example with
the quite commonly used
trail making test,
which there was a positive effect
in mild cognitive impairment.
And you can see that there’s
a linear positive relationship
in relation to the brain ketone uptake.
But it’s also clear that the placebo group
and the active group are not
well separated here at all
so the ketone effect of the MCT
does not really pull and separate them
from the placebo group at 30 grams a day,
which I don’t have on the slide I see,
but that’s the dose we use for six months.
So that’s something that we now realize
is helping us set a therapeutic target.
We arbitrarily started with 30 grams a day
which was close to the
upper tolerable limit
we thought with the
formulation that we had.
And we were trying to
demonstrate a metabolic effect
and we’ve seen that metabolic effect.
But now if we expect to
see a cognitive effect
we’re going to have to increase
the brain ketone supply
a little more.
Still the trend is there
and as the yellow box says,
several other of these
correlations were also positive
but I don’t have time to show them to you.
So again proof of principle suggests that
you can achieve a metabolic
rescue of the brain
and that the rescue is gonna
have a functional effect
that is gonna correlate with
the level of ketones achieved
both in the blood and in the brain.
So we look at this now in
terms of the green band
in the middle here,
what is the energy rescue or the gap
in energy supply to the brain
that we’re trying to fill.
So let me take you through this.
The bar on the left here is
the young healthy controls
and their energy supply,
they’re in the normal group.
So they’re at 100%.
And their brain energy is coming
as you can see in the table at the bottom,
about 95% from glucose and
about five percent from ketones
using the dual tracer ketone PEt imaging.
So that’s the young healthy controls.
Older healthy controls,
we’ve got about 30 people in that group
and you can see that the total
height of the bar is lower,
the ketone contribution is a bit lower.
It’s not that the K value for the ability
to take up the ketones goes down,
but in fact plasma ketones drift
downwards as you get older.
So their contribution to brain
energy metabolism goes down.
But it’s glucose that is going
down in principle here mostly
from 95% contribution to about 89%.
But they’re all cognitively normal still.
Then you go to MCI which is the middle bar
and the values gone down a little further.
So there’s still a
significant drop in glucose
although the ketone
uptake is still the same
as it was in the healthy older people.
Then the second bar from the right
is the trial I just
mentioned, the BENEFIC trial
in which the MCI patients
were on 30 grams a day
of The MCT,
and now we’ve helped restore them
closer to the healthy
elderly with the ketones
which are now contributed
twice as much as they were
in a healthy elderly.
But the glucose contribution
has not been changed
by the MCT supplement.
So this is suggesting to us
that if we wanna get a more complete
reversal of the energy deficit,
we’re gonna get up, need
to get into 45 grams a day
roughly speaking.
That was gonna pit us somewhere between
the healthy elderly and the healthy young
which seems like a reasonable place to be.
We might be able to get
right up to this 100% value
with perhaps with a ketone Esther
or some other supplement,
we’re still exploring that.
So this is telling us what
our therapeutic target is,
it’s giving us a dose that
we should be looking for
in order to accomplish
this particular goal.
This is the same slide but
it’s a little busier now
because I’ve added the
Alzheimer’s group here.
So from the MCI we get a further decline
to a value of a deficit of about 19%
in early Alzheimer’s disease.
So person with early diagnosis
of Alzheimer’s disease
day in day out is facing a 20%
energy deficit in their brain
MCT is one way to help correct
that, as we show you here
but you can see that the group
next to the Alzheimer group
here is an exercise group.
And in fact exercise
helps get glucose in more
into the brain.
It also helps get ketones into the brain
without a ketogenic
supplement at the same time.
It’s helping to increase the K value,
the number of doors into the brain.
So exercise is potentially
quite complementary with this
and we’re excited about
the potential of that.
The problem is that it took seven years
to get five patients
through this protocol.
So it’s not easy to get Alzheimer patients
to participate in these sorts of projects.
But the proof of a
beneficial effect is there.
A second Alzheimer group
which are basically identical
to this one on 30 grams a day
is also improving the brain energy status
but it’s not changing glucose.
So medium chain triglycerides
are not affecting glucose,
exercise is affecting both,
and I think together we could probably
catapult them just over
this magical line here.
So it’s kind of exciting
to know that we can
do some things about the brain energy
status in Alzheimer’s disease
and we need to power studies
now for the cognitive outcomes.
So coming back to that busy slide
with the direct line relationship
between in normal young adults
who have been starved or not,
that’s the black line that’s shown here,
and what I’m showing next to it
is the uptake not of beta hydroxybutyrate,
which is the black line,
but the dotted line which is acetoacetate
and I put the Alzheimer patients on there.
I put our healthy older controls on there
and I put the ketogenic diet on there
and they’re all in the
same linear relationship.
So it’s very achievable.
People are on a ketogenic
diet are probably at least
at one millimolar ketones in the blood
and this should be really helping to
stave off if you’re in a preventive mode
the glucose deficit in the brain.
But this is what we imagine is probably
the target that we’re gonna
need on a daylong basis
24 hours.
It’s very hard to achieve
with a medium chain
triglyceride supplement
but much easier to achieve
with a ketogenic diet
although compliance is
tougher in older people.
So just to show you what
a ketogenic diet does
to the brain ketone uptake,
these are the cerebral
metabolic rates for acetoacetate
and the blue color is in
people pre-ketogenic diet,
very low ketone uptake,
around five percent
and that’s what you would see.
The blue represents quite a low uptake
and then everything basically lights up
when you’re on a ketogenic diet.
So those of you that
are on a ketogenic diet,
you can see how your brain,
it loves the ketones.
And in fact the ketogenic diet is
actually increasing the total
supply to energy of the brain.
This is in a normal adult
where there’s five percent
ketones, 95% glucose
around 100% there are two
different control groups here
for reasons that I don’t
need to explain it,
both about 100.
But in fact you can get to
113% or so on a ketogenic diet.
There’s more energy
getting into your brain
on a ketogenic diet than there
is on a typical Western diet.
So there’s not just a therapeutic effect
but potentially a cognitive benefit
in those that are not
facing aging associated
cognitive decline.
So to summarize,
I just want to come back to this image
and say that those arrows are definitely
a consequence of the disease,
but they’re also
contributing to the disease,
the arrows are present
before the disease starts
and it’s a glucose specific problem
that brain energy rescue by ketones
is definitely feasible in
mild cognitive impairment,
it’s feasible in Alzheimer’s disease.
What we’re to do is
basically let the brain
have the luxury of thriving
in a fuel environment
that you were born into.
And the fourth point is
that the cognitive benefits
still need to be better defined
as well as a mechanism of action.
We don’t know if it’s as
alluded to earlier today.
We don’t know if it’s just a fuel effect
or whether there’s some signaling affects,
whether there’s some
anti-inflammatory effects,
whether even affecting
the pathology in the brain
in this situation.
So I wanna finish by thanking
a terrific group of collaborators
and members of my lab.
And thank you for your attention.
(audience applause)