Dr. Francis Collins: Vision for Medicine in the 21st Century

Uploaded by NIBIBTV on 24.07.2012

Our first speaker – outside guest speaker today is Dr. Francis Collins. All of you know
Dr. Francis Collins as the director of the National Institute (sic) of Health. And as
you also know, he had the vision to recognize the incredible importance of translational
research as part of the portfolio of NIH. Now, when the – as president of American
Institute of Medical and Biological Engineering, when we heard this wonderful vision, we recognized
that translation is a concept that’s fundamentally in the breadbasket of medical and biological
engineers. That’s what we do. We translate discovery to devices for impacting patient
care and to drugs that are impacting patient care.
So we asked to meet with Dr. Collins and his staff to talk about how AMB and NIH might
work together to catalyze the maximum success of this effort. We had a wonderful meeting,
and that was followed soon after with a talk at the annual event of the AMB to all of our
fellows, which was absolutely sensational in engaging us and making us understand how
medical and biological engineering could have such a positive and important impact for the
mission of NIH.
It’s my pleasure to introduce the director of NIH, Dr. Francis Collins.
(Applause.) Thank you very much. Good morning to all of you. It’s an absolute delight
to be part of this remarkable symposium celebrating the 10th anniversary of NIBIB and all the
great science that is coming forward as a consequence of this really innovative and
exciting aspect of what NIH is all about.
I want to particularly give my thanks to the director, Robert Pettigrew, who has been such
a visionary leader in this area of bioimaging and bioengineering and I think has, through
the remarks that he put forward already, given you just some snapshots of the really remarkable
scientific opportunities that this institute supports. It’s a very exciting time scientifically,
and it’s great to have such leadership in a component at NIH that really focuses on
this. And our interactions with AIMBE and other organizations only add to our excitement
about where this whole field is going.
I confess to being a bit of a techie myself, so it is really delightful to have a chance
to think a little bit about what I might say to you this morning and pick out a few examples
of areas in this space that seem to be particularly groundbreaking. And it’s obviously a broad
field of opportunity, so I will pick just a few examples. And please forgive me if I
left out some of the ones that you might have chosen yourself or that you might have thought
were even more appropriate. But in my brief remarks I wanted to both celebrate NIBIB and
to point out how this fits together in a broader sense with where NIH is going and where medicine
is going.
First of all, congratulations on this anniversary. Here’s a representation of the act that
created this particular part of NIH the 24th day of January 2000. And congratulations to
all in this room and outside this room who have made this into such an exciting part
of NIH. Congratulations to Dr. Pettigrew for having also made it possible for the Senate
of the United States to make a resolution celebrating this achievement. You can see
here that that happened just earlier this week, with support from Senators Burr and
Mikulski but wide recognition, therefore, about what’s happening and lots of positive
things said about NIH and NIBIB.
NIH has a noble mission. It is science in pursuit of fundamental knowledge about the
nature and behavior of living systems – that’s our basic science component – and the application
of that knowledge to extend healthy life and reduce the burdens of illness and disability.
That simple sentence sums up what we are all thinking about every day when we come to this
remarkable place and have the chance to work with some 19,000 individuals who have their
sleeves rolled up trying to make this happen. And of course, that’s a small proportion
of what NIH is all about, since most of what we support is research going on out there
in our nation’s finest institutions and some abroad, as well. And for me as the NIH
director, it is truly a privilege to be able to stand at the helm of such an exciting operation
and try to steer it a little bit here and there. But most of the really important scientific
leadership comes from the institute directors of our 27 institutes and centers, and hence
the very important role that they play. And again, Dr. Pettigrew plays that role extremely
I thought I would talk about four T’s here, NIH investments in innovation. Talking about
the future, innovation is what we need to focus on to make those things happen. So I’m
going to talk about technology, about translation, about talent and, well, yes, indeed, because
we are at a challenging financial time, how do we also justify this in terms of economics
– because, believe me, I get asked that question a lot, including at my House hearing
yesterday in front of Energy and Commerce. And I think we have a very good story to tell
there, as well, and we shouldn’t be shy to tell it.
So let’s talk first about technology. Obviously, in the sweet spot of NIBIB, with so many things
that are coming out of the bioengineering, bioimaging approaches that are leading us
towards new ways of diagnosis, prevention and treatment, I’ll just mention one that
is now actually beginning to take shape in the basic arena of technology development,
teaching us about how, exactly, the brain is really wired. This is coming out of the
NIH blueprint for neuroscience research, a collection of institutes that are all interested
in the brain and has led to the formation of this Human Connectome Project. And part
of that has been the support of a diffusion magnetic resonance imaging scanner which has
very high detailed resolution and is faster than conventional scanners.
And if you’ve not seen any of these pictures or movies, I would really encourage you to
look at some of the literature -- whoops, let me see if I can get that movie to run
– which has recently come forward in this effort, because it is really quite a remarkable
kind of image that one can achieve of this, two-dimensional sheets of parallel neuronal
fibers that cross paths at right angles -- there’s a paper in Science that will show you more
of this – and teaching us things about how the brain is working in the living individual
that we probably would not have imagined we could learn about quite this soon. Lots of
relevance for understanding normal and pathological development. And again, very much driven by
clever, innovative engineering approaches.
In translation, where we take the technology and try to move it forward into clinical applications,
that is obviously an area of intense interest across NIH and all of the institutes and centers.
One of the new kids on the block is the effort to try to focus this in a way that would enable
even more rapid development of therapeutics. And one of the motivations for that is what
you see on this graph, which shows you what we have learned over the course of just the
last couple of decades about the molecular basis of disease. This comes out of the compendium
Mendelian Inheritance in Man, which has been keeping track of what diseases do we actually
know the cause of at the molecular level. And you can see we’re up now to about 4,500
conditions, most of them relatively rare, and many of them caused by single gene mutations.
But look at the rate at which that has happened. And this is about to go up, I think, even
steeply again because of the availability of whole-genome or whole-exome sequencing
that enables you to find the cause of disease even in very small numbers of individuals.
The cost of DNA sequencing is a major driver of a lot of this technology. When we finished
that first human genome sequence in 2003, it cost us about $400 million to get that
reference copy. Now your genome or mine can be sequenced for about $8,000. So we’ve
dropped the cost by roughly 50,000-fold in the space of less than a decade, and there’s
no evidence that that particular drop in cost has reached any kind of limit. No laws of
physics are going to be violated here. We can keep on dropping this down, potentially
to the point where a human genome sequence is down to a hundred dollars or even less.
And that opens up all kinds of possibilities in terms of medical applications. Certainly
also DNA sequencing has found its way into basic science in lots of applications, not
only in terms of whole genomes or whole exomes, but just using it to count things, like RNA
expression, for instance, or epigenomics. It’s been quite a revolution. And the people
in my lab over in Building 50 actually can’t quite imagine how we did anything in human
biology before we had the genome sequence and the ability to collect this kind of data.
That’s all the good news. The bad news is, if you look at this same diagram and ask how
many of those 4,500 conditions where we know the molecular basis actually now have a treatment,
it’s about 250. So we have this huge gap between what we know and what we can do about
it. And that clearly is an opportunity as well as a responsibility and a challenge.
And that’s one of the motivations for trying to develop a better way to tackle the problem
of developing therapeutics.
Most of you know how that works, but let me just run you through a little cartoon here.
If you have a disease that you are seeking a small-molecule therapeutic for, you’re
faced with the challenge of sifting through the universe of shapes of small molecules
to try to find the one that actually has benefit for that disease, that hits the target in
just the right way to result in clinical improvement. That means you’re starting with a very large
library of structures, here cartooned, and you’re trying to pick out, using some kind
of assay, which of those actually has some potential of hitting the target in a beneficial
way. And then you have much work to do to sift through that in the preclinical space
to come up with something you might actually feel was safe enough to offer to a patient
in a phase one trial.
The problem is, this is a terribly inefficient process with huge losses along the way. You
start with maybe 10,000 of those compounds, you sift them down into maybe 250 in the preclinical,
and ultimately perhaps five of these make it into clinical trials. But experience being
what it is, right now only one of those will actually achieve approval by the FDA. Notice
the timetable at the bottom, 14 years, the average time that it has taken over the last
many decades to get to success. Because of all the failures, which are 99 percent plus,
you are also facing enormous costs because you have to pay for all those failures in
order to actually have some successes along the way. So the estimated cost of each success
is now in the neighborhood of $2 billion.
Well, that’s just not sustainable. With those thousands of diseases that are waiting
for answers, presumably we have to come up with a better way to approach this. And we
need to do so and do something fairly drastic in terms of a scientific approach, because
the experience here is truly daunting. There’s a paper from Nature Reviews Drug Discovery
just published a couple months ago. These authors basically tallied up what the cost
was in terms of drug development since 1950. And using inflationary corrections, they plotted
here the number of drugs per $1 billion of R&D spending, and they plotted this out on
a log scale. And you can see it looks really troubling here. It almost looks like some
sort of law is at work here, and it’s a law that’s going the wrong direction. You’re
now down to sort of less than one drug per billion dollars, where back in 1950, goodness,
there was more like 30 or 40, on inflationary-corrected dollars, of that same thing.
So what’s going on? The authors, somewhat tongue in cheek, decided maybe this needed
a name. It sort of looks like a law. So they called it Eroom’s Law, which, if you notice,
is Moore’s law written backwards – (laughter) – because this is going backwards. So what’s
the problem?
Well, the engineers in the room would probably say, that’s your pipeline? That’s the
best you can do? What’s going on here? Why all the failures? Why the slow progress? What
could you do to choose your targets better to begin with? How could you figure out how
to fail early instead of late? All of those are critical issues. And science has come
along in interesting ways in the last few years to enable us to tackle some of those
questions in truly innovative fashions instead of continuing to do things the way we have,
which clearly, from this diagram, is not working so well.
So this really was the motivation for us here at NIH. After much consultation through many
different venues, particularly the Scientific Management Review Board and consultations
with the private sector through my advisory committee to the director, we established
the National Center for Advancing Translational Sciences just last December 23rd. And it is
not intended in any sense to be the place where all translation happens at NIH; not
at all. It’s intended to be a hub, a catalyst to look at actual bottlenecks in this pipeline
as themselves scientific and engineering problems, not specifically attached to any disease or
any specific small molecule or diagnostic or device project, but looking systematically
at the way the whole pipeline works or in many cases does not work.
So the NCATS has already gotten itself deeply engaged in a variety of projects. One of which
fits rather nicely with one of the examples that Roderic showed you is to try to utilize
what we’re learning about tissue engineering to be able to do a better job of predicting
whether a drug is safe or not before you give it to a patient in a phase one trial. And
this has led to an unprecedented collaboration between NIH, DARPA – the Defense Advanced
Research Projects Agency -- who are pretty good at wild and crazy engineering ideas,
and FDA. And collectively here we’re putting $140 million into this over five years. And
the goal is to develop a chip loaded up with as many as 10 different cell types, human
cell types, in a fashion that is as much as possible representative of what happens in
So as you saw in the video a few minutes ago about the human liver that was being re-implanted
in mouse, in this case we’d want to have human liver in a three-dimensional organoid
on this biochip, as well as cells representing heart, kidney, brain and so on, the places
where you most would be interested in seeing a signature of toxicity if that was likely
to be a problem.
And of course we need to wire this chip up with a lot of different outputs in order to
find out everything we can about what happens to cells when they’re exposed to that new
test compound. So that probably means looking at gene expression, at metabolomics, at proteomics
and so on. And here is where we hope the science that NIH is pretty good at and the science
that DARPA is pretty good at can come together.
So there is between our two agencies now efforts to put together a consortium, and this is
an interesting cultural experience as well, bringing together staffs of NIH and DARPA
to work on this. Awards are anticipated imminently, next month, where we hope to put together
some of the best ideas about tissue engineering, biology, toxicology and so on to create a
pathway towards generating such a biochip. Obviously, the goal will be to assemble this
in a fashion where you can then test it with compounds where you know the answer, that
they are safe or they’re not safe, and then develop the idea about what a signature would
look like that would give you confidence that it was reasonable to go ahead with a phase
one trial or not reasonable.
If this works, presumably it should be much faster and cheaper than the slow, expensive
and not very reliable animal testing, which is the current way in which you have to generate
evidence that your compound is safe enough to get an IND from FDA. So FDA’s involvement
here is critical, because if this begins to work, we would not want to see this as an
add-on, as an additional thing that you have to do to test for safety, but rather as a
substitute, you know, for the methods that are currently used. And that is an example
of the kinds of things that NCATS aims to support.
And you’ve already seen one part of this on the chip. I won’t go through it again
because you already even saw part of this video, and saw it much nicer because it had
some – a narrative to go with it.
I will show you another video quickly. This is one I showed yesterday to the House Energy
and Commerce Committee, and it is work supported by NIBIB, the work of John Donoghue and others,
which is basically, then, developing this interface between the brain of individuals
who have quadriplegia and allowing them, through the process of connecting that interface with
their own brain function, to learn how to move a robotic arm simply by the thought process
that the subject is undergoing.
I must tell you the members of the Energy and Commerce Committee were quite riveted
by looking at this. Here is the example where the patient in this case is attempting by
her thoughts to pick up this canister which has coffee in it and see whether she can bring
it to her mouth and take a sip of her own coffee; all of this being accomplished, as
you can see, by the box mounted on the top of her head that allows her thoughts to be
translated into that robotic arm. And I think you can see by the smile on the face of the
patient and the investigator that this was a pretty good day – (laughter); that this
really was a remarkable moment.
Well, it’s great to have all of these capabilities, but we certainly need not only to have great
science and great ideas; we need to have the talent of the individuals to pursue that.
NIH is intensely interested in encouraging the most innovative ideas to come forward,
and worried that in the current climate, where our budgets are very tight, that peer review
sometimes can be a little conservative in terms of not taking as many risks, because
there’s such great solid science in front of them as well.
And one of the things we have done in order to try to encourage that kind of innovative
approach is a series of special programs, the three you see here, all supported by the
Common Fund, that aim at high-risk, high-reward research.
The Early Independence Awards, which we’ve just started, aim to take the most talented
graduate students and take them directly from the Ph.D. degree to independence, skipping
the postdoc and giving them the chance to show their stuff while they are still in that
oftentimes most creative phase and not in a circumstance where they are following other
people’s guidance but allowed to put their own ideas forward. We just made 10 of those
awards in this first year. The best day I think I had in the last six months was coming
to listen to those 10 grantees talk about what they were going to do, a truly remarkable
and inspiring group of fearless young investigators who are aiming to do things that are quite
bold indeed.
The New Innovator Award also aims to try to bring new investigators to NIH who have not
previously received grants from us, not requiring nearly as much in the way of preliminary data,
but demanding that what they put forward has to be a groundbreaking kind of approach; otherwise,
they can’t be evaluated as part of this program.
The Transformative RO1 – and now called the Transformative Research Award, because
it isn’t really quite so much like an RO1 – also has to be transformative, but can
support large teams, and has less in the way of budget limits than many of the other kinds
of formats that people are used to – and the Pioneer Award, which has now been around
for more than five years, which gives an investigator the chance to write a relatively brief proposal
with some bold ideas and then to be funded for a five-year period as an investigator
and basically allowing them to move in the direction that makes the most sense scientifically,
not particularly tied too tightly to original plan, very much like a Howard Hughes position.
And those also have turned out to be very much productive parts of our portfolio.
As an example, I’ll talk about an individual who’s actually received a Pioneer Award
and, more recently, a Transformative RO1, and who is a grantee of NIBIB. And this is
Sunney Xie. This is a investigator at MIT who’s a chemist who has actually applied
some very innovative approaches to do single-molecule visualization in single cells, an area of
intense interest. In fact, we’re about to start a new Common Fund effort in single-cell
Sunney Xie, using his skills as a chemist, has been able to come up with systems that
allowed you to look at expression of single-protein molecules. What you’re looking at here from
a publication in 2006, those yellow dots there are single proteins which are marked by a
version of GFP called Venus. And basically, this is a circumstance where they’ve created
a transgene in this bacterial system that is only rarely expressed, because it’s actually
repressed by the lack of repressor, but occasionally, one copy of the RNA gets read off anyway,
and that one copy then is used to make three or four proteins, which you appear – appear
as bright dots in that particular bacterium, and that is kind of fun to watch. Here you
see the solid static version. We can run a little movie here. Now, watch the yellow dots
as they begin to appear. They’re just sort of firing off there as an RNA was transcribed,
and then proteins are briefly made from that RNA in that same cell.
That was fun; let’s watch that again. (Laughter.) See, that yellow dot now will pop right up
there. You can see, oh, must have made an RNA there, now maybe four or five copies of
the protein being generated from that transcript. It’s teaching us a lot of interesting things.
It’s obviously nice eye candy, but it’s also teaching us a lot about biology.
He has gone on with a colleague at MIT, Jaiei Zuang (ph), who’s been also a very creative
individual in terms of figuring out how to see things at a resolution that you thought
you couldn’t do with light microscopy using this technique called STORM. And here is just
one image, here again, of a E. coli system, looking at a nucleoid associated protein called
H-NS. This is what this would look like in phase contrast; fluorescence of a conventional
sort would look like that. Look at the level of resolution showing where that protein is
in these individual bacterial cells.
So lots of exciting technology being supported through these grant mechanisms, the Pioneer
Awards and the Transformative RO1s.
I wanted also to mention something that NIBIB is doing which I think is kind of exciting,
bringing young scientists at undergraduate level into the realm of competing with each
other for this DEBUT challenge. And this is basically NIBIB’s opportunity for people
to compete for prizes. The deadline for receipt of the applications was June 2nd, so presumably
there will be some announcements before too long. Diagnostic devices is one approach,
therapeutic another, or technology to aid underserved populations in low-income circumstances
or individuals with disabilities – so very much an innovative way to try to drum up excitement
in undergraduates. They have to be submitted as teams. And as you can see, there is money
involved, which tends to get the attention of undergraduates.
Finally, there is this issue about how we need in the current climate to be sure that
we’re making our case for the value of what biomedical research is doing, not only for
the future of human health but for the economy. And again, I think as all of us are called
upon to defend the investment of taxpayer dollars in this activity, we should be prepared
to say why this is a good investment. We need to make that case because we are not at the
present time enjoying a particularly favorable environment for the support of biomedical
research through NIH.
You can see here in this series of bars, the purple bars tell you what the appropriations
have been in NIH over the course of the last 13 or 15 years. You can see the doubling that
happened between ’98 and 2003, which was a wonderful opportunity for growth and for
new investigators to come into the field, but you can also see that we flattened off
pretty badly after that. And if you look at the yellow bars, which take account of inflation,
we are actually losing ground and have been since 2003. Our purchasing power now is down
by about 20 percent over what we could do nine years ago.
And obviously, with the current focus and appropriately so on our difficult financial
circumstance and deficits, there is certainly no indication that this is going to get any
better, and presumably it could even get a lot worse, especially if those sequesters
that are being talked about happen to kick in on January 2nd of next year. If that were
to happen, we would in one fell swoop lose $2.4 billion of the NIH budget. We would have
to cut back our grant awards for FY ’13 to unprecedented low levels. About 2,300 grants
that we wanted to give would not be given.
So there is a cloud on the horizon. And we need to be defending this, as I tried to do
yesterday in the hearing, but as all of us, I think, are called upon to do from time to
time, why this is such an important investment for our economy, for American competitiveness,
but for human health as our number one approach.
If you’re looking for evidence to cite about that, there are a number of really remarkable
analyses that have been done by credible economists that make this case. Every dollar that NIH
gives out in a grant results in more than twofold return on investment in the first
year, in economic goods and services to the local community. Our grants support about
432,000 jobs, high-quality jobs. And the spinoffs from what NIH does probably results in multiplying
that number by a factor of 20 in terms of employment through pharmaceutical and biotech
companies. And we’re a major driver, therefore, of the American economy, and cutbacks at NIH
will have ripples that will be really quite severe. Again, there’s lots of information
about this. If you go to our home page, NIH, and look for the button that says impact,
you can see a long list of documents that go into this issue about economics and the
evidence that supports the value of what we do.
And I think this is a particularly moment, then, as we are facing this potential challenge
to the future of our enterprise, to emphasize that return and to do so in a fashion where
it’s clear we’re all speaking together. It’s one thing that have somebody stand
up with one voice and talk about the value of medical research; it’s even better if
that’s a part of a symphony with a chorus, and we all kind of have the same song sheets
in front of us basically arguing what we are about is a noble mission to try to improve
human health. We have a remarkable track record in terms of what’s happened in the last
few decades with – for instance, heart attacks and strokes have been reduced as a cause of
death by more than 70 percent in the last 40 years, HIV/AIDS no longer being a death
sentence but consistent with lifespan to age 70 and beyond, and a variety of other really
remarkable achievements. And we shouldn’t be shy to sing that song to people who are
interested in asking the question about whether this particular part of government investment
is worth the dollars. It is more than that.
As far as where you’re going in terms of this symposium and the future of NIBIB and
the rest of the NIH and all the disciplines that we represent, I have a hard time imagining
that very clearly. And I’m fond of this quotation from the guy who wrote “The Little
Prince” – “Le petit prince,” if you read it in French – Antoine de Saint-Exupéry,
who says, “as for the future, your task is not to foresee it,” because that’s
too hard, “but to enable it.” Of all the parts of NIH that I think is really focused
on that remarkable task and opportunity of enabling, it seems to me NIBIB is right in
the thick of that.
I congratulate all of you for the progress you’ve made, and I wish you another wonderful
decade to come. Thank you very much. (Applause.)
We have time for a question or two for Dr. Collins.
I will ask a quick question.
What do you see as the more important argument to be making with lawmakers, the economic
impact or the health impact at this current moment?
You know, I think at this point it’s still the health argument that gets people’s initial
interest, because everybody has concerns about themselves, their loved ones, their constituents
in terms of health issues that need more attention. But as soon as you get that part of it, oftentimes
there are then responses, well, yeah, but can’t the private sector do this better?
Or why are we giving you so much money? Couldn’t you do this for a lot less? Then I think you
have to be prepared to come right behind that with the economic argument, which is a very
compelling one as well.
You showed the depressing trend in the number of drugs per billion dollars invested, and
I see – excuse me, I understand the effort to try – (off mic) – early stage – (off
mic) – faster. But where is most of the cost? The cost is in the final testing of
those five targets and the total deduction in costs – (off mic).
No, you’re quite right. The major source of cost that doesn’t get us anywhere is
in failures in phase 2 and phase 3, because by that time, you’ve invested tens or hundreds
of millions of dollars in a compound. And the major cause of failure is efficacy: The
compound turns out to be safe, but it doesn’t work. So another effort – and I didn’t
have time to go into this – which we’re deeply engaged in right now with industry
as major partners is how to do a better job of target validation. Right at the front end
of this development, how do you pick the right targets so that you don’t go down this path
very far and ultimately discover you’ve got a drug that just doesn’t work?
There are some exciting potentials there. Again, because of the ability to use tools
like genomics, we have the chance to do target validation in humans instead of perhaps in
cell culture or in animal models in ways that we couldn’t have before. We have a human
knockout project that’s under way right now. Nature’s been at that project for all
of human history, and now we have a chance with things like exome sequencing to detect
the consequences of that.
Take an example like this gene called PCSK9, which Helen Hobbs and colleagues in Texas
discovered carries knockout mutations in about 2 percent of people. And those people have
very low levels of LDL cholesterol and very low risks of cardiovascular disease. She even
found a few homozygotes who have no functional PCSK9 at all. They’re entirely well. Their
cholesterol levels are, like, 12 and their risk of heart disease seems to be extremely
Now, talk about a great way to validate a target. That tells you that if you built a
drug that’s an antagonist to PSCK9 and you actually hit the target, you should provide
substantial clinical benefit with a reliable biomarker, and it should not be toxic because
the people who don’t have this for life seem to be just fine.
That was a great example, but could we do this more systematically by identifying such
loss of function, human knockouts, for lots of other genes and focus on the ones like
PCSK9 where the loss of function actually seems to provide some advantage? Those would
be perfect targets, then, to mount a campaign to build a small molecule against that target.
That’s kind of a thought process that we’re going through right now with industry as our
close partners, because they need this too. They’re aware that the way in which targets
have been chosen has resulted in a great deal of failure oftentimes very late in the process.
One quick question. Yeah, Dr. Collins, in terms of the dollar amounts or the time being
taken for drug development, shouldn’t we include a general statement that, apart for
trying to be more sophisticated, the problems remaining are much harder to solve? Therefore,
it is taking much larger effort, both in money and time, to achieve the goal.
That’s certainly part of it, and people would argue that the low-hanging fruit in
terms of developing drugs has already been plucked and that the targets that we now need
to go after are going to be tougher, less well-understood, perhaps less easy to hit
with our particular small molecule or biologic approach. That’s certainly in there, that’s
part of Eroom’s law, undoubtedly.
But it’s not the whole story. And if we have a chance to stop that downward curve
or even start it back up again with new science, facing the fact that we may have more difficult
science in terms of the targets themselves, we should put lots of energy into that. The
future really depends on that. I very much oversimplified all the factors that go into
this, but I do think we have an opportunity here if we work together to try to do something
fairly interesting.
One of the thing we just announced, which is a way to shortcut all of this, which I’m
pretty excited about, is to take those compounds that got all the way to phase 2 or phase 3
but failed to show efficacy. But where it’s shown not – shown at that point to be safe,
then figure out, OK, so it didn’t work for diabetes; might this drug have worked for
schizophrenia or cancer? We’ve seen lots of examples where that kind of repurposing
actually works out pretty well, whether it’s thalidomide for myeloma or whether it’s,
goodness, AZT for HIV/AIDS – AZT wasn’t developed for that; it was developed for cancer.
There are now eight companies that have agreed to open their freezers and make 58 compounds
available for investigators in academia or in small businesses to look for new uses.
And if that turns out to look promising, one could go almost straight to a phase 2 trial
with compounds where a lot of that investment’s already been made. That won’t work for everything,
but if it works for a few, it’s well worth the effort.
Thank you very much.
Thank you all. (Applause.)