Chordoma - Drug Repurposing on Cancer Cells

Uploaded by ChordomaFoundation on 21.12.2010

Diagnosed and had his initial surgery and I said "that is very interesting" and there
was this guy standing next to me during this conversation and it turned out to be Josh.
And I said "Ah, this is the patient" What that started was really an Odyssey which we
are still on. I think that the fundamental kind of theme behind what we are doing and
what we have been doing for the last two years is to try to figure out how to as rapidly
as possible go from some of the model systems that you heard talked about, the most reductionist
model systems, i.e. are cells to patients as rapidly as possible circumventing as much
of the traditional drug discovery process as possible. Fundamentally, what this comes
down to being able to test as many hyposthesis in parallel as possible. The way one does
this is the same way that one would do it in any other field which is that instead of
doing things in series, i.e. you test things one at a time, you test everything simultaneously
in a single experiment. Simone asked me to bring these because they are sort of illustrative.
There are some of these that are hanging around. I think that David Alcorta actually washed
this out last night so as far as I know there is nothing toxic or living in these things
but I still would not lick them or use them as eating utensils or anything. What they
are is what we do our experiments in and it is a plate that has one thousand five hundred
thirty six wells in it. They hold about a thousand cells each. What we would ideally
like to do is to take each patient with a Chordoma and stick them into one of those
wells and pour a drug on top and see how they respond. But, a tad impractical, so what we
did is take the cells out of body, put them into those wells, and put the drugs on top
and see how they respond. There are a number of issues associated with that. If you take
the cells out of the body as some of the previous speakers have said, they stop responding in
the way that might potentially in the body. And you also have to, as David was saying,
the cells are what they are said to be. The point is that one can test and were able to
test and have been testing every drug that has been approved not only by FDA, but every
regulatory agency worldwide in a day using this kind of technology. And we do it using
robots like this. I will take heart in hand here and hope it doesn't crash the computer
again. What you are seeing here is, think of these robot arms as essentially researchers
or post-docs or graduate students or technicians. What they are doing is handing plates back
and forth to each other and in the back we can hold four or five million wells of compounds.
In the middle, there are tissue culture incubators. This sort of Jenn Air looking things in the
These are tissue culture incubators that keep the cells warm and happy and then at this
end there are so called readers which are essentially glorified microscopes that allow
you to assess whether the drugs are doing you want them to do on the cells. In this
case, what we want the compounds to do, the drugs to do is very simple. We want them to
kill them. So, it is a very simple assay. We do a lot of more complicated ones, but
in this assay, it is pretty simple. I am now going to try to hopefully get to. We are going
to have some fun with this because this is going to do a PC to Mac transition. If any
of you know, or are in the computer world know them are fighting words. So, we will
se how this goes. This may...who knows maybe there will be new data here that we will discover.
So the concept here is that rather than discovering a new drug in a way that I will show you in
a second, what we are actually working on is to take drugs that are currently approved
for other uses and find which ones might be useful in Chordoma. The reason we are taking
this approach with Chordoma and I can be straightforward with this crowd is that the conventional drug
discovery process which I think is on this slide. This process takes 10-15 years. When
I first talked to Simone and Josh about this, their reaction was that doesn't apply to what
we need because the life expectancy in our disease is four years, five years, less than
fifteen. So what else can we do. And so we had been thinking about taking this approach
for other reasons, but I must say that the conversation that I had with these two, three
years ago really gave us a boost working in this direction. So let me just tell you have
drug discovery normally works and then I am going to try to distinguish that from what
we are doing. So, normally what one does is you study the basic biology of a system for
a long time. And the first talk about the notochord is a perfect example. I grew up
in the same world of developmental biology. You are studying a fundamental mechanism usually
and then frequently, you stumble across an application of what you are doing which you
really didn't anticipate. But you might also identify a target. And what we mean by a target
is usually a gene or a protein of which acts as a pressure point in a regulatory system,
in a system that regulates whether a cell is growing or not or whether it remembers
what it is supposed to be. And you can think about this as a lever in a systems of other
levers and you push this one to bring the system back in balance again. That is really
what a target is. So what you doing, you are trying to identify a drug or small molecule
compound which can act on that target to either increase its activity or decrease its activity
and bring the system back to where the balance is supposed to have. Normally what we do in
probably eighty or ninety percent of what we do at our place and how typical drug development
is done is that you create an assay system and we have been doing this in Chordoma too,
that is a testing system that allows you to test four to five million tests in a week.
That is a typical drug development project. That is not what we are doing in Chordoma,
but this is a typical project. So you are testing greater than 100,000 chemicals on
an individual target for its activity and then you do a lot of additional chemistry
to try to make modifications of these molecules to make them suitable for initial animal use
and then human use. And then eventually you get into humans. This whole process, you have
probably heard for those of you that know about this, this whole process around ten
to fifteen years including cost of failures this is about a billion dollars to get from
here to here. For a lot of diseases, you really have to go this route, and I would argue that
we want to go this route eventually in Chordoma too, but the problem is we really don't know
what the targets are in Chordoma and so in that case, one simply asks a more practical
question, well "What do we know about Chordoma?" What we know about Chordoma is - they have
a certain cell type and they grow to much. So, the issue is, Can you get that cell out
and make them stop growing or kill them. It is pretty simple in principle. So, what is
the center that I run and Mahong has really done all the work here, so as usually, I get
to stand up and present. I think there is probably one person that does the work that
has actually given a presentation so far. So, anyway it is about sixty five people.
We have a hundred collaborations, a bunch of them with foundations. Fundamentally what
we do is bring up the things, the kinds of technologies that go on in biotechnology and
pharmaceutical organizations to rare and neglected diseases in academic or government drug discovery.
So this is another way to put that last chart I gave you here. If you sort of break this
down into this sort of vernacular of the field. You start with a target at one end and you
want to get to FDA at the other, the conventional way that this is you start out with 300,000
compounds and it takes about ten to fifteen years to go from here to here. What we are
trying to do here is to take all the drugs that are approved by FDA, take them back to
a screen and then try to get back to a clinical trial within a year or two. That is the idea.
So I am not going to do this in detail, but just to show you what the numbers are. Two
informatics people, two compound acquisition chemistry people at our place have spent the
last three years putting together the largest collection in the world of approved drugs.
We have everything that has been approved by FDA, UK, EU, Canada,Japan. These are a
number of compounds that are in clinical trials and haven't been approved yet. And that number
is about three thousand and it is growing all the time. This was an enormous, both informatics
challenge and a procurement challenge. These compounds are often very expensive. Some of
them we had to synthesize ourselves, or have them synthesized in China. Or I still have
my medical license, so I was able to write prescriptions to obtain some of these and
then we purified them out of the tablets. The group that does this is very unusual for
an academic group. This is what a pharmaceutical or biotech company looks like. It is very
unusual for an academic organization, that is, it has biologists, chemists, informatics
people, computer scientists people and robotics people who run that beast like I showed you
at the beginning. And this is the other thing that is unusual as these are some of the places
that these folks come from. About eighty percent of the people come from pharma or biotech,
so we know how to do this. Mahong and I were both at Merck before but there are many other
companies that these folks come from. It brings a knowledge of how to very practically move
the kinds of discoveries that you heard about before into the drug development world very
efficiently. This is actually a picture of the robot that you saw in the video. It is
about the size of this room. It is about the size of a small school bus. It can do about
four million tests in a week. We work on a lot of rare and neglected diseases. Some of
them are rare like Chordoma and Gaucher and Huntington's. Some of them are neglected like
diseases of the developing world. I am not going to show you a lot of data. I just want
to give you a sense of the approach. This is the headway we have made so far. This was
initiated officially in two thousand six. What was required to do this and again it
sounds like a broken record from other speakers, but it is really important to say that you
can't underestimate the role of the Chordoma Foundation in doing this. When Josh, Simone
and I first sat down to do this, what I told them was that I personally know a lot about
Chordoma because I used to treat patients as I am a Neurologist. Mahong is not, most
of my scientists are not. We are not Chordoma experts. If we are going to do this, you have
to put together a brain trust of Chordoma researchers who can help us pick the right
cells, pick the right outcomes, be able to followup the results. And that drug discovery
team was assembled by the foundation. We screened a bunch of cell lines, some of which you heard
about before. This was the line that a number of folks have talked about before. We are
looking at cell killing. We've just starting to initiate screening primary cells directly
from patients as opposed to cell lines. We can get into what that means if you are interested.
There is twenty eight hundred and sixteen approved drugs that have been screened at
multiple concentrations over about a ten thousand fold range of dose. As you know, dose is important
as to the effect of the drug. We are now in the followup stagen, in addition to doing
an ongoing screening here. And what we are trying to do is to find what is the shortest
path from a cell based screen which is necessarily artificial to a person. And the goal here
is to patients by the end of the year. That is quite ambitious, but I actually think we
can do it. This is the kind of really straight forward data you get out of this. What we
are looking at here is a bunch of different cell lines and a bunch of different compounds
and these are the Chordoma lines here. I just show you this to give you a sense of how complicated
that these data are. And the way you deal with these data is you do the experiment and
a robust a way as you can and then you work with experts to go to other model systems
whether they be in cells or animals or what have you as Xenograft models in order to validate
the results that we get. In something that is intermediate in complexity between a cell
and a dish and a person. I would be glad to take any questions, but I just want to stress
at the end that this is team really is a team. We have a team at the NCGC who works on this.
Michael, David, Neil Spector have been critical in this, particularly Michael and David and
Josh when he was in Michael's lab. Derek Park who is at the University of Pittsburgh who
has talked earlier and will talk later today. Larry Baker of Michigan, Andrew Wagner and
the folks from the Chordoma Foundation. We get together on the phone very regularly to
discuss these results and really hash through them. We met again yesterday afternoon and
one of the real pleasures for me in this has been to be able to work with these folks who
are just incredibly committed to doing something for this disease and I am very competent that
we are going to be able to make some headway and I hope in the next six months we will
be able to give you more definitive information.