– So as all of us who are studying ketogenic diets and cancer do well, I'll
show you in a second. Our study from a few
years ago is colloquially called the recharge trial. Because it was a very stupid acronym, but I won't tell you what it was. So this is, you know, from
recharge the novel biomarkers. And hope to just update you on some of the stuff we're doing now, collaborators are many and I'm indebted to all of them wonderful people. Richard Feynman has been my colleague and collaborator for many years this year. So we all aim to target the Warburg effect and I can call and described it very well. And this is the glucose and the glycolytic dependent phenotype. It's common to many aggressive cancers, and it's evidenced by positive PET scan with 18 fluorine labeled FDG, FDG is fluorodeoxyglucose, you can hear the glucose
is in the molecule, right? Do we have any pointers here? No, okay. Anyway, I think you can
see what's going on though, you can see the heart 'cause this is a rather fuzzy but
still valuable PET scan that we obtained a long
time ago, I still like it.
And when a person is on a standard, rather high carbohydrate diet, well, insulin is secreted and
drives glucose into the heart. So fluorodeoxyglucose is
clearly visible in the heart. You can see the tip of the brain there, the liver is less well seen. Now, the lowest portion of the liver, and on the other side, you can see the kidney collecting systems because this is radioactive
urine and thanks. All right, so there's
the collecting system of the urine and bladder. So the fluorodeoxyglucose
is excreted into the urine. So you get that, but what
you don't really want to see, but you do, all these multiple
sites within the liver, which are glucose added metastasis, and there's also another one in the lungs. So this is a patient with
the metastatic colon cancer. In 2012, somehow we're all stuck in 2012 due to funding problems. But this was this was our
study published at that time as again, as I say,
called the recharge trial.
And it was a pilot study
in 10 patients, oops! You give me two clickers
and I get confused. Alright, so the objectives
were safety and feasibility of a 28 day ketogenic diet in 10 patients with advanced PET positive cancer. We use the entry and exit PET scan as a surrogate marker of efficacy. We're looking at the change in the FDG uptake between the two. And the patients were
monitored for compliance and the systemic metabolic effect by getting a weekly fasting
beta-hydroxybutyrate level. I'm not gonna go over this in any detail, more information that I want. But you can see that well, first of all the patients were not selected as you can see by the
tissue type, all right? Because there are a
whole bunch of different, there's breast cancer, ovarian,
so forth, a lot of addition.
The patients were selected
by the metabolic phenotype by the fact that they
had a positive PET scan using fluorodeoxyglucose. That was the selection of the patients. The circled patient here
was excluded from analysis. We weren't thinking, but you see that she had a biologically
different tumor. She was surviving for 14 years, she had had no surgery
on her breast cancer, she had no chemo, or no radiation therapy and was still alive. These other patients had
multiple chemos failing. So she, we ended up having
to exclude her from analysis. She was a biologically different patient. In any event, actually all
the patients experienced a certain amount of calorie deficit. This was not by attention,
they did not on average, they ate about 35% less calories. Five patients you can see in the blue had a stable disease or partial remission and forehead progressive disease, but there was no relationship
between the calorie deficit that was the same in both groups. They also lost about 4% of weight, but again, it was about
the same in both groups. What did matter was the level
of ketosis between the groups, and you can see that the level of ketosis in the stable disease
partial remission group is about three times higher
than in progressive disease.
These are all pretty consistent. I mean, even the lowest one is
around 10 fold over baseline. These three of course, are below of, you know, four fold in lower baseline. This one, of course, looks like it, this patient should have responded but that's actually part
of the problem with cancer. Not everyone's going to respond, then we'll get to that in a little bit. Still, I thought was
pretty promising overall. And the summary is that this was a prospectively designed
and executed pilot study of ketogenic diet as a cancer therapy. 10 patients completed it
without unsafe adverse effects.
And ketosis was observed that correlated with the level of insulin inhibition. I didn't show you that
slide as you'd expect, and it also correlated
with disease stability, which is consistent with the hypotheses affects on glucose dependent cancers. So new work, which is what
I really came to talk about is we now are in the
process of just starting a human trial of 65 patients with just diagnosed
ER-positive breast cancer. Patients are randomized to two groups, one of ketogenic diet of 45 patients, and then versus a standard low fat diet of 20 patients total of 65. The ketogenic diet is
supplied by Natural Ketosis, which is a company in UK these are vacuum sealed meals, we get them in boxes of 28 day supply, so they have long shelf life, and we can ship it out to the patient right after their diagnosis to start.
The standard diet is high
in carbohydrate foods, which is supplied from a list we called from a Fresh Direct menu and patients selected so they can come up to a high carbohydrate diet on that arm. And our question is do
biomarkers of cell proliferation, this biomarker, KI 67 or Key 67, I guess, depending on where you're from, versus cell death, which is a tunnel acid whether these biomarkers change between biopsy and surgery, which is a period of
usually around three weeks between two and four depending on how long it takes to get the patient schedule.
Now, biomarker study isn't
exactly a therapy trial, but it's clearly related to one because it provides information about the likely behavior
of a therapy trial, right? And question you might reasonably ask is, is three weeks or two to
four weeks enough time to see a difference in a biomarker? And Goodwin and Dowling, I mentioned Goodwin before from Canada studied Metformin in a
very comparable study. They did it a few times until they got up to about 500 subjects and during the same time interval. And what they found was that
there was a significant drop in KI 67 in the Metformin
treated patients, but not in the control group.
And I think everyone in this
room will agree that Metformin, you know, while we don't
really know quite how it works, we know it drops blood glucose
and thereby drops insulin. But we also know that a ketogenic diet is quite a bit better at doing that. And so if this work for Metformin, we at least have some reason to believe that we may be successful
in showing a change in a three week trial using our diet. So, you know, I came to show
you a study of 65 patients and so far, we've managed
to complete one patient.
So, I don't have a lot of
data, but, it's illustrative. So this is KI 67. This is the biomarker of proliferation. And you can see these brown
stains in a lot of the cells, you can see a lot of other cells which don't have that stain,
they're sort of grayish. And Susan Feinberg, a pathologist has the laborious task of counting the numbers of cells and
determining the percentage, and you have to do this in
four high powered fields. So it's very laborious, but
and then you get an average. So she found one of
these high powered fields at 44%, another 72, 73 and 79%. That stain for KI 67, and
this is on the biopsy. Now, the tumor excision specimen, this is, you know, three weeks later, you can see, it looks like it's more, but I'm just, you know, I
deliberately picked that 'cause I wanted you to
see that it got worse, but it's still she did all the counting. And it was, you know,
a little bit more hard to say if this is
statistically significant from four high powered fields slides, but it was 51%, 79, 80 and 81 that stained in this follow
up, you know, biopsy, you know, from the from
the actual tumor excision, so it certainly was no better.
Now, when I actually try to press Susan for what's the normal value of KI 67 and breast et cetera she waffled you know, but I'll just say that by large, the value is probably less than 10%. So these are both very abnormal values. This is a patient who
didn't start off well and maybe got worse but
certainly didn't get better. This is the standard human toxin and a nuisance name the HNA and the central area is the tumor with normal tissue around it mostly, and this is three weeks later.
So higher power field, but
the important thing here is here you can see my mitotic figures, this is a nuclear materials
separating and a mitosis and it's only one slide,
but it looks worse. And basically, there's a lot of mitosis if you look at a lot of other fields. So in three weeks, the tumor
appears to have gotten worse. Displaying mitotis figures, certainly not a good possibly higher KI 67 and also apoptosis relating actually to tumor growth, not to an improvement. And yet patient one was
randomized to the ketogenic diet. So this isn't very good, right? This is not what we were expecting anyway, but you know, we have to
take the data as it is, however, it turns out, the
patient was not in ketosis. The baseline shows ketone body, sort of at the upper limits
of normal first week, if you want to call that marginal, you know, borderline
physiologic ketosis you can but you can see that
really nothing happened.
And the problem was our
fault it wasn't hers, she was supposed to be
on a ketogenic diet, but turns out I don't know what happened. But when we gave her the
list from Fresh Direct, we included some foods which
were not actually ketotic, and she jumped at them,
and then she ate them. And so entirely our fault. I forget, someone once told me that whenever you do a clinical trial, you have to expect that
you're gonna screw up the first two patients. Well, we've done it on the first one.
But anyway, we're still on the second we're about, so far,
we have 1.67 patients, the second patient is two thirds of the way through the diet. So we're pretty close. But in any event, the other thing, so I just wanted to give you a flavor of what we're doing now, in
terms of the human trial, we're excited to be doing it. We're hoping to get more patients and to start doing it
correctly, and that'll be good. And the other things we're doing is we're trying to develop
some novel biomarkers.
KI 67 is not a novel biomarker, but it's a cell proliferation. It's been around, but it's not used routinely in clinical practice, it's too cumbersome to use and the standardization is difficult. So it's not really truly standardized, as I say, not part of routine evaluation. And our team is working on developing a standardized method from
large statistical samples, basically, through
automatic cell recognition, or cell properties, counting these stains in a computerized fashion, and being able to get a large representative number. So we're very excited about possibly bringing
this to the clinic. Probably the most exciting
thing that we're also doing is another novel biomarker. And Stephanie Mattingly
is a colleague of ours who's working at University
of Alberta in Canada and she's after three
years of very hard work she managed to actually label
ketone body with 18 Fluorine. So this could be very exciting. We actually don't know what what the use of 18 label Fluorine ketone body will do. But it is the third major
radio label metabolic tracer after FDG and F-18 label glutamine and it could be for
potential widespread use.
Here is an example of an
image that she obtained and where it's going is
just where you'd expect. So you see some in the brain, okay, the eyes always light up,
the bone, and the marrow lights up very typically. But here's a tumor. This is actually an MCF-7 tumor, which is an ER positive tumor. One that we're studying
as well animal models, I'll tell you in a minute. This is a bone uptake, not as intense, but we're gonna be playing with that to determine how to manipulate the images and not the images, but
the actual physiology, just the black and white
version of the same. From the side this is not
optimally shown for imaging.
This is the sternum in front
but the heart is behind. So the heart also takes up. So this is just where you'd expect beta hydroxybutyrate to go. It goes to heart, goes to
brain and it goes to tumor. So we believe this is the tracer, we believe this is actually an accurate analog beta hydroxybutyrate. So to be explored with this tracer, our first, first thing you have to do is find out normative image findings under different dietary conditions. 'Cause you have to know
actually how it behaves. And we're in the process of doing that.
And because it goes to these areas, it goes to the brain, it goes to the heart and goes to cancers, its distribution is similar
to fluorodeoxyglucose but perhaps it will be complimentary in terms of the information
for specific applications. We don't know yet until we try it out and it may permit exploration
of individual mechanisms of ketone bodies and cancer
therapies and perhaps identification in advance
of susceptible patients. And the final thing I'd
like to discuss briefly is the use of diet and
drugs and mouse models. We're studying particularly two different mouse models and her collaborator Yuzu CD1 nude mice MCF again, ER positive breast cancer
xenograft in five mice per group and we also have a spontaneous
breast cancer tumor model just to see what the difference is, you know both all animal
models have limitations. So we're gonna try both. And the idea is to couple diet and again rapamycin. Rapamycin as (mumbles) points out, well, he didn't mention rapamycin, but you mentioned the pathways. Rapamycin is a drug as discovered on Easter Island actually, and it inhibits mTOR and I was described as some
more than the most of us who know nothing about alphabet soup, but when I was growing up, there was something called alphabet soup.
And that's what all these
signaling molecule sound like xmTOR and AKT and that sort of thing but anyway, mTOR is a growth molecule. It stimulates growth and
rapamycin inhibits mTOR. So it's been conceived
of as an anti cancer drug or to be used as one. But sadly, like the PI3 kinase inhibitors that Dr. Cantley developed, it also causes counter regulatory effects and
namely causes hypoglycemia. So we've been planning to do
this for a number of years, or crushingly disappointed
that Dr. Mukherjee and Dr. Cantley beat us to it, but we're in the process
of doing it now at least. And so we're looking at a
spontaneous tumor mouse model and CD1 nudes and
comparing a standard diet standard diet plus rapamycin
versus a ketogenic diet, and ketogenic diet plus rapamycin.
And we're looking at
changes in tumor growth, cancer death markers,
and overall survival. And we're obviously interested in testing whether a ketogenic diets synergizes with drug therapy to
increase overall efficacy at lower drug doses and lower toxicities. So that's the basic idea. Just wanted to tell you
very briefly about CD1. I stumbled on CD1 about 10 years ago, when I was doing a study for
someone entirely other purpose. I picked it because I figured one day I'll eventually be doing
animal research in cancer. And I looked at regular CD1s and put them on a ketogenic diet, and I was stunned to find
out that it is a mouse.
It develops about 2 millimo
ketosis on a ketogenic diet. That's very unusual for a mouse. So this was terrific. I tried two different diet variants, 10% protein and 16% protein for reasons too long to discuss on the time that I have left, but basically, the amount of
ketosis was the same at both. So we're still working on
which one we're about to use. Animals have now been ordered. We hope to start the animal study to evaluate the dietary, you know, plus drug
regimen I just described.
And so that'll be the fourth
and final thing I discuss. Thank you very much for your attention. (audience clapping).