Making the Cut | Session 1: Science and Society || Radcliffe Institute

– Good morning, everyone. I’m Tomiko Brown-Nagin, the
Dean of the Radcliffe Institute for Advanced Study, and I’m
delighted to see all of you here today. I’m so pleased to welcome you to
the Radcliffe Symposium focused on the critical and urgent
topic of human gene editing. Now, I say urgent
because we find ourselves in a new scientific
era, an era that demands thoughtful analysis
of our capabilities and of their potential
implications, both positive and negative. The technology to make very
precise genetic changes is far more accurate and
more accessible today than ever before, and technical
acronyms like CRISPR now pop up regularly
in the news and even in conversation, not
just in science journals and on syllabi. Yet, it’s fair to say
that the underlying concepts and the practical
realities of gene editing aren’t common knowledge. I’m delighted that the
Radcliffe Institute is hosting this forum, one in which we’ll
explore crucial nuances as we consider the scientific,
ethical, social, and legal implications
surrounding human genome editing. In true Radcliffe
form, we’ll do so by engaging with scholars
and practitioners across disciplinary and
professional boundaries. It’s only by drawing
on diverse perspectives that we can understand
the scientific, medical, and ethical implications
of the new capabilities and begin to answer
questions like, what are the important
distinctions between gene editing techniques,
some of which produce heritable changes,
while others do not? Where are the gray
areas, the unknowns in human genome editing
as an area of research and applied science? What are the benefits and trade
offs in clinical applications? And as we move forward
in this new era, what does consensus on standards for
research and clinical care look like? Now, I’m not a scientist. I’m really not a scientist,
as many of you know. I’m a law professor
and historian. Yet as the Dean of Harvard’s
Radcliffe Institute for Advanced Study, I
embrace the responsibility of institutions of
higher education to foster informed
dialogue around issues with important
societal implications. Gene editing and
its applications in medicine and
scientific research is clearly one of those issues. It is a scholarly and a
decidedly public matter, and today’s program will
address both of these facets. Before I turn things over to
Professor Immaculata De Vivo, I’d like to offer some things. First, I’m grateful to
our Provost, Alan Garber, who encouraged at Radcliffe
to take on this critically important and timely topic. I’m also grateful to
our planning committee, led by Professor De Vivo. Thank you. And to all of the
committee members, including Associate Provost
for Science, Kathleen Buckley, Professor Janet Rich-Edwards,
and Professor Alyssa Goodman. Thanks as well, as always,
to the Institute’s staff, especially the
academic ventures team, led by Becky Wasserman,
and the events team, led by Jessica Viklund,
for all of their hard work behind the scenes. And of course,
we’re honored to be joined by a distinguished group
of panelists and moderators. Thank you all for being here. Finally, I want to acknowledge
the members of the Radcliffe Institute Leadership Society
and all of our annual donors. You enable the Institute’s work. Thank you for your support. With that, I’m pleased to
introduce Professor Immaculata De Vivo. Mac, as we call her, is
the life sciences advisor here at Radcliffe, a professor
of medicine at Harvard Medical School, and a professor
of epidemiology at the Harvard T.H. Chan
School of Public Health. She also directs the
genotyping and genetics for population sciences core
facility at the Dana Farber Harvard Cancer Center. Please join me in giving
Mac a warm welcome. [APPLAUSE] – Thank you. Thank you, Dean Brown-Nagin. Colleagues, participants,
attendees, everyone watching live stream,
I am pleased to have the honor and responsibility of
shepherding this timely topic. So how, you may ask, did this
symposium come into being? It started when,
truly, the provost contacted the dean wanting
to engage Radcliffe to host a symposium of this topic. Given that the topic
of gene editing broadly has captured
the imagination of the scientific community,
as well as the public, Radcliffe is well positioned
to showcase this event, because the magic I like to
call of Radcliffe Institute is their ability
to bring together a multidisciplinary
approach to topics. And as such, today’s program
design is within that spirit. Scientists, clinicians,
ethicists, journalists, and all have come together here
to explore this topic. Gene editing, broadly
and specifically CRISPR, we have constructed
a timeline which captures the milestone of
the development at CRISPR, and right now I’d like to take
you through a few of them. So they’re just a few– the larger, more comprehensive
timeline is on the website. I urge you to go to the website
at the end to click through. But today, just for
the framing remarks, I’m going to give
you the timeline. A few milestones. So the timeline was written
for an expert audience, and I will do my
best to translate it to the non-expert audience. And basically, the timeline
illustrates the stride over the past few
years that have been enormous,
from the discovery of these little repeats
of DNA in bacteria to the implementation of
CRISPR into clinical trials. Truly remarkable. So with that, I’d like to take
you through this timeline. So how did we get to today? These repeats, and
these are bits of DNA that are repetitive, that were
found originally in bacteria. It’s a primitive immune system
in bacteria, to be to be exact. It is a simple,
yet powerful tool to change your DNA sequence
to modify gene function. It has enormous implications
from clinical to basic science to even improving crops. So these selected
milestones, again, I will translate in a non-expert
way in the best of my ability. And this is a very
characteristic display of CRISPR, what it
means, and I will try to take you through
some of these cartoons within the little
time that I have. So in 1987, these
repeated sequences were recognized by Ishino in
e. coli, which is the bacteria. And Mojico in Spain actually
coined the word CRISPR. Bolotin discovered
the Cas9 protein, which is actually the
scissor that everyone equates with CRISPR. This scissor is
critical, as you’ll see, with further discoveries. And it was Horvath
in his discoveries that showed that, in actuality,
the bacteria, when it’s invaded by a virus, which
is called a phage, it’s the bacteria’s way
of sort of digesting, for lack of a better
word, the virus, because it’s trying to attack. This, ergo, is
the immune system. And of course, it
was Sylvain, who’s one of our speakers
today, who showed that the scissor cuts with
precision, great precision. He will talk about
that today, hopefully. And one of the more profound
discoveries within CRISPR was by Chorpentier and
Doudna and colleagues where they show that with the
use of these guiding RNAs, they can direct the scissor to
cut exactly where they wanted. So they were able to take
this system out of nature and into the lab, so
to speak, to engineer. It rendered it a
programmable system. Huge implications. And of course, Zhang Feng
showed this in humans. And then, of course, today,
where we are actually using this technology in
clinical trials for something like sickle cell anemia. Truly remarkable. In 30 years, we’ve gone from
this discovery of these weird looking repeats in bacteria
to the implementation in clinical trials. And as a researcher, I can’t
contain my own enthusiasm how it’s quite remarkable. And it’s partly, or mostly,
because the entire world came together to work on this
technology from all aspects, whether it was
the acknowledgment or the recognition in bacteria
to the discovery of the scissor to how the scissor
cuts and where it cuts and how do you manipulate it. So it’s truly a
multidisciplinary discovery and implementation in
the clinical setting. And so with that, I
would like to introduce the first session, which
is on science and society. And for the sessions today,
the way we have designed them, is that they’re case
studies, and then we will have a famous– Sharon Begley, the
journalist, at the end to do sort of round table
discussion with the moderators. I will introduce
each moderator and do a brief bio and the
speakers for the session, and the moderator will
then introduce each speaker with their bios to
give the session talk. So the first moderator for
today is Charmaine Royal, and the speakers
within that session are Sylvain Moineau
and Jonathan Kimmelman. Can you please come up? [APPLAUSE] [INAUDIBLE] And
the guys can sit. So Charmaine Royal is
an associate professor of African American studies,
biology, global health, and family medicine,
and community health at Duke University. She also has appointments
at the Duke Initiative for Science and Society and at
the Kenan Institute of Ethics. If I were to read
you the bios of all these distinguished speakers,
we’d have three days of bios. So I will show deference by
just giving you a few lines. So with that, Charmaine,
I give you the podium. – Good morning, everyone. It’s great to be here. I’d like to thank Mac and Becky
and Rebecca and everybody, Tomiko, and everyone
for inviting me as well as the other speakers
to this timely event. This first session,
Science and Society, I think sets the tone
for the entire day, because we’re going to talk– our speakers are going to talk
about the scientific aspects of CRISPR and the ethical and
social implications and policy implications of CRISPR. And then we’ll have
the case studies later on that allow us to incorporate
what we get from this session into those discussions. So Mac presented a wonderful
timeline in terms of CRISPR and how we’re coming to think
about CRISPR and the evolution of the technology. But I’d take us back
even further in time. I’d take us back to 1953,
when Watson, Crick, Franklin, and Wilkins told us about
DNA and introduced us to the structure of DNA. And it’s soon after that that
people started thinking, OK, if we can figure out the
exact changes in the genome– now we call it the genome. Then they didn’t
call it the genome. The exact changes
that cause disease, then we should be able
to target those changes and fix them and cure disease. And the idea of
gene therapy, let’s start with gene
therapy, gene transfer. Because CRISPR is just a
variation on the same thing. It’s fixing the problem in
the genome to cure disease. And so gene therapy, the
idea of gene therapy, which we eventually started
calling gene transfer, came about in about the
1970s, 1972 or thereabout. And later on, we had the
first successful studies in the 1980s, 1990. We had the first successful
gene transfer trial for SCID, Severe
Combined Immunodeficiency with DeSilva, who
was cured then. So we had gene therapy 1990. And then in 1999, everybody
knows what happened in 1999. We had Jesse Gelsinger,
the first reported death from a gene therapy trial. Raised a lot of issues,
a lot of questions about this procedure
and the way we do it and the ethical
and social implications. And that put a halt on
gene therapy for a while. But the work was still going on. And fast forward to
2012 and CRISPR, and now our conversations
about this technology. But the ideas, the
fundamental ideas, have been there for a long time. And the issues have
been there too. Germline versus somatic. Disease versus traits. Issues of equity, the
rich versus the poor. Who has access? All of those issues. Some are old, and some
are new, and we’re going to talk about
some new ones today. So I applaud Mac
and her colleagues for bringing us together. Science and Society,
recognizing the need for us to think about the
scientific agenda in the context of society. And I know our
audience here comprises people who are scientists
who know a lot about this but people who are from
the broader community who are trying to understand. I’ll tell a short story. Last week I was at the American
Society of Human Genetics. The American Society
of Human Genetics is the largest
professional organization of human geneticists worldwide. And some of my colleagues,
Vince and Kiran, were there. And the final day,
the final session of the American Society of
Human Geneticists, I mean, there’s 8,000 members
in this society. And we were talking about what’s
on the horizon for genetics, for human genetics. And Kiran moderated
that session. And one of the panelists,
Cynthia Morton, who’s here at Harvard,
told of a ride she was having in a taxi cab. And the cab driver
asked her what she did. She’s a geneticist, a
human geneticist here. Asked her what she did, and she
said, I’m a human geneticist. And he said, oh, are you
involved in that gene sniper thing? [LAUGHTER] And I thought that was precious. That just goes to show
that it’s out there. It’s out there, and people
under– and he’s right. It’s a sniper thing, right? People understand this
in all kinds of ways. And so today, I look forward
to our conversations today as we think about the scientific
and the ethical and social implications of gene editing. So I’m going to
introduce our speakers. And I’m just going to read
their short bios in the program. And as Mac indicated, you
have the longer bio there. So our first
speaker, who’s going to talk a bit about the
science behind all of this, is Sylvain Moineau,
Canada research chair in bacteriophages in the
department of biochemistry, microbiology, and
bioinformatics in the faculty of science and engineering at
the University Laval in Quebec. Sylvain. – Thank you. [APPLAUSE] Thank you, Professor
Royal, Charmaine. I certainly want to thank
the organizing committee. I want to thank Dean
Brown-Nagin for the invitation. I want to thank Mac, Professor
Vivo, for the invitation as well. Always a pleasure to
come back to Boston. So yes, I am at
University Laval. A lot of people actually last
night asked me where this is. Well, this is in Quebec City. This is a French
speaking university, so of course, I’m going
to give my talk in French. [LAUGHTER] No, I won’t. So it’s about 40,000
students, and I hope you’re going
to take the time to visit our lovely city
of Quebec City in Canada. So today I’m going to talk to
you about a bit of the history. So my talk will be
separated in two parts. The first part will be a
little journey in microbiology, and the second part will
be, well, why you’re here. All right? We’ll talk about the
gene editing technology. As Mac mentioned, what a ride. CRISPR has been crazy. There’s a lot of
front newspapers but also in scientific journals. Scientific journals with one
word are usually pretty good. [LAUGHTER] So it made the front
page of those as well. And of course, you are all here. This is usually the
first slide I start when I give a talk on CRISPR. When I talk to my students,
I show this slide, and I told them, this is– if you type CRISPR in the
research article databases, this is the number
of publications that are related to CRISPR. So as Mac mentioned, the
acronym started in 2002, and she showed you the timeline. And the technology
appeared in 2012, and today we’re almost
over 6,000 articles on CRISPR per year. So that’s about
16 papers per day. You can’t keep up
with the literature. It’s insane. So this is a career opportunity. Some people in the
audience could say it’s a business opportunity. So there’s a lot of people
involved and interested in this technology. So today, I guess my take
home message is right there. There’s two things. There’s a CRISPR-Cas
system that you’re going to find in
bacteria, and there’s a CRISPR-Cas9 technology, which
is the tool that we’re trying to use for genome editing. And today I’m going to try
to show the differences between the two. And of course, by
explain a little bit where CRISPR-Cas
came from in bacteria is going to lead to
the technology itself. This OK? Well, don’t have
a choice anyway. So this is a single tiny drop of
water that you see on the right here. That drop of water was mixed
with a DNA binding agent. So what do you see here,
the little dots here, the little ones, are viruses. The big dots are bacteria. In a single tiny drop of ocean. So if you go to the
harbor in Boston, you take a little
water, you’ll find this. What it means is that
we are surrounded not only by bacteria, but
we’re surrounded by viruses. Viruses are the most
abundant biological entities on the planet. It’s not bacteria. We’re full of
bacteria, by the way. 40 trillions of
bacteria in our body. 40 trillions. There’s more
viruses in our body. All right? So we’re surrounded by viruses. These bacteria, because
they’re surrounded by viruses, they need to defend themselves. Otherwise, they
are going to die. So they need a
defense mechanism. They need an immune like
system, otherwise they’ll die and the viruses will win. So what these viruses
do in our environment, they actually control the
population of bacteria. And these viruses are
called bacteriophages, meaning eating bacteria. All right? So we’re surrounded by those. And bacteria are
surrounded by those. So they have a really key
role in our environment, because that’s one
of the reasons we don’t have piles of bacteria. They’re killing those bacteria. So this is a very small
schematic representation of how these viruses replicate. Sorry. There’s a big difference
between a virus and a bacteria. Bacteria is a cell. So it replicates by itself. It’s got all the
machinery to replicate. If it’s got a good
condition, nutrients, the cells will become two,
four, eight, 16, and so forth. A virus cannot do that. A virus need a
cell to replicate. A virus by itself
cannot do anything. So it’s inert, except
when it’s inside the cell. That’s different. So the phage and viruses
need the cell to replicate. So that’s just a little
schematic here of how viruses or phage replicate,
in that case. So they will absorb
to the cell surface. There will introduce
their DNA inside the cell. Then at that time,
they take over the cell to produce new
variants, new viruses. In this cycle or lytic cycle,
we call, it takes about half an hour to do. It’s very fast. So they can kill
bacteria very rapidly. They’re like the Usain
Bolt of virology. They’re really fast. They can kill your cells. So of course, bacteria,
they need to defend, like I just said. And they’ve got multiple
defense mechanism. I mean, really, many of
those, and many more need to be discovered, by the way. So they can block
this replication of viruses in multiple ways. And CRISPR-Cas is
actually one of them. There’s many more, but today
we’re focusing, obviously, on CRISPR-Cas. And this function was
discovered in 2007, as Mac was mentioning
earlier, in publication with Philippe Horvath. And my group is also
involved in this finding, because we’re studying
phages and we have interest in how bacteria are defending. So we already
presented the acronym, which was published in 2002. Actually, in the
literature, there was a bunch of
acronyms that were proposed for this structure. But CRISPR was the
one that stuck, so it’s Clustered Regularly
Interspaced Short Palindromatic Repeats. All together now. [LAUGHTER] And there’s also Cas for CRISPR
Associated Genes or proteins. And we’ll talk about this
in just a little bit. So this is your
typical DNA sequences. Four letters all mixed
up and gives us genes and give us what we are today. So if you go into some bacterial
genomes, you’ll find this. You’ll find this. You’ll find these
repeat sequences. It’s the same letters
next to each other, and they are separated
by our variable regions, like Mac was mentioning. These are spacers
or variable regions. So this is what CRISPR is. It’s found inside of about half
of the bacteria at the CRISPR. The other half does not. But that’s what it is. It’s repeats and variable
regions and so forth. And in the literature,
in the press, it’s often presented that
way with color coded. So you’ve got the black diamonds
here, which are the repeats. They’re repeats, so
they’re repeated. In between them, you’ve
got variable sequence, and here they’re shown by
rectangle different colors, because the spacers
are actually different. So this is, again, inside
the genome of the bacteria. If you go next to that
region, you’ll find genes. Again, in the bacteria. And what a mess it is. There’s a bunch of
different genes. Cas 1, 2, 3, 4, and so forth. There’s a lot of
different Cas proteins that have been associated with
these CRISPR structures I’m talking to you about. And really today we’re going
to talk about this one. CRISPR-Cas is only– Cas9 is
found in one type of CRISPR-Cas system. And that’s why sometimes in the
journals or in the literature or in the press, you’ll
find another story about another CRISPR. Because scientists
are studying this one. They’re studying this one. They’re studying this one. Because we don’t know
how they all work. So there’s a bunch of– bacteria
has a highly diverse CRISPR-Cas system. So a lot of scientists are
actually studying these things. It’s kind of nice
there’s such a diversity. So you avoid to be
scooped by other groups. But that’s part of science. So CRISPR-Cas in bacteria, and
I’m going really briefly here, works in three ways. So the first step is called an
adaptation or a vaccine step. So the phage will
infect the cell, but the CRISPR
machinery will pick part of the phage
genome, the virus genome, and will introduce it
into its CRISPR array, if you can believe that. It will just introduce a
piece of the viral genome into the CRISPR array. At that time, there
will be some RNA. So our GPS that
will be produced, this RNA molecule
will be produced that will be matching
what has been introduced into the CRISPR array. And then the CRISPR-Cas protein
will bind to this GPS molecule and be inside this cell. So if another virus
comes in, it has a perfect matching sequences
from this and this. It will bind to it, and it
will cut the DNA of the virus. So this is how it
works in bacteria. And I’ve got this
educational video just to illustrate what
I’m trying to say here. So this is a bacteria
having fun, multiplying. This is actually in
a cheese environment, or in milk fermentation process. So you’ve got bacteria that
we’re using to make cheese. It is infected by a
virus, by a phage. Its DNA gets inside the cell,
as I was trying to tell you. And then at that time,
that’s the end of the cell. The bacteria is done. This virus wins, because
it’s going to produce itself. Proteins self assemble. And after half an hour,
the cell will blow up, and the cell will
be– and new variants will be released
into the environment, ready to infect other cells. And that’s nature. That’s happening in water,
in yourself, and so forth. Sometimes when a defective
virus, so a virus that’s not very efficient,
gets inside the cell, the CRISPR will pick
part of its genome, introduce it in its array,
in the CRISPR array, and then at that time started
mounting its immune system, producing this RNA
molecule, this GPS, producing that Cas9 protein,
bind together, and waiting for another cell to be
infected by the phage will be recognized by the
machinery and the phage– whoops. Oh God. It’s a virus. There you go. I wish. So this is– this big blob
here, this is your scissor, your Cas9 protein, your sniper. And it is actually armed with
this little RNA molecule. And this is what we
call, really, a GPS. Because you can modify
that RNA molecule to go and bind to another
part of the genome. And that’s why it’s
became a tool, because you can modify that GPS. Unfortunately, this is a GPS. It doesn’t go always
where we want it to go, but that’s the point. And at that time when the
DNA of the virus is cut, the virus cannot repair itself. It’s done. The cell win. So it’s a defense mechanism. Have you ate yogurt today? – Not yet. – I hope you did, because
it’s good for you. Well, if you did, you
actually ate Cas9, because this is found
in yogurt bacteria. This is natural. This is a way that bacteria
we’re using to make cheese, to make yogurt. When you look at
the ingredients, you’ll see bacterial cultures. They carry a CRISPR-Cas system. So this is a natural process. The technology now. CRISPR-Cas9. I will finish with that. So this is a very, very
powerful research tool, and never forget that. Because today I want to talk
about medical application. But CRISPR-Cas9 is a fantastic,
very successful research tool. A lot of people in
various life science labs are using it with a
massive amount of success. We’re making discoveries,
new discoveries, because we’re using
CRISPR-Cas9 as a research tool. Of course, because it’s a
research tool so successful, people are trying to
see now, all right, can we move this from the lab
and try maybe to cure people? But if we succeed or not,
at the end of the day, it’s still a very, very
powerful research tool. So what’s the technology itself? We took part or scientists
took part of the natural system to develop this tool. So they took Cas9. So you remember there
was a bunch of protein, bunch of RNA molecule there. But we took just part of it. The Cas9 and this little GPS
to form this Cas9 and the GPS together. Again, remember, we can
change this at will. We can determine
where we’re going to bind to a piece of DNA. And this technology,
as Mac was mentioning, was actually published for
the first time in 2012. It was a collaboration
between Emmanuelle Charpentier and Jennifer Doudna. This was published
in August 2012. You see the number of citation. And then Feng Zheng
and George Church’s lab here in Boston publish
two papers showing, a few months later, showing that
this can work in cell lines, in human cell line,
animal cell lines. Again, August, February. This is the number. This is a very short
list of organisms that have been modified using
the CRISPR-Cas9 technology. Look at the dates. I mean, this technology
was rapidly adapted, because it’s so easy to use. Pretty much everything has
been modified using CRISPR. There’s difficulties in
some cases, of course. But it’s been really useful. This is another video,
educational video, to show how it works. This is from MIT that
with their permission, we have modified a little
bit for educational purposes. You go inside the nucleus of
the human cells or a cell. This is DNA. You can introduce Cas9 armed
with the GPS that you want. It will bind to one
strand of the DNA. There’s a little motif
that is necessary. And then Cas9 will cut
the DNA, the DNA you want. Then at that point, you
will rely on the cell to repair itself, because
the cell can repair itself when you see a break like that. Then if you introduce at
the same time another piece of DNA with little
tricks here, you can actually introduce it where
you want that piece of DNA. You can also introduce a
mutation, correct a mutation, and voila. This is as simple as– well, not really. But it’s close. It’s close. So again, this is a very
powerful research tool, because if you’ll sequence
the genome of, let’s say, a plant, a human, you
analyze the genome. You’ll quickly realize there’s
a number of proteins and genes we do not know what they do. Now we have a tool where we
can go and modify that gene, modify that protein, and look
at what’s happening to the cell when you do so. So we’re understanding the
function of these proteins now, which is fantastic. But of course, like I was
trying to say earlier, there’s medical applications. And you can use
CRISPR-Cas9 technology in at least three
different ways. We might heard a few more today. Ex vivo, in vivo, or germline. So ex vivo is very simple. You need the consent
of the patient. You will fake maybe some cell,
immune cells, most of the time, from the patient. You’re going to introduce
your editing system. And then you’re going to
check if it cut or replaced at the right spot and
then reintroduce that into your patient. So this is called ex vivo
somatic editing therapy. There’s certain
clinical trials ongoing. A lot of people are– a lot of people. Scientists are looking
into that if you could modify some immune
cell to attack cancer cells. So we take cells
from the patient, modify it, return
it into the patient. So that’s called ex
vivo somatic editing. So like Mac mentioned, there’s
a couple of clinical trials are underway in
the United States. In vivo now. In vivo is that we
want to modify you. You’ve got a genetic disease. You’ve got something
you want to correct. You might be blind
for something. You might be blind. So you can take– there are some vehicles
where you can introduce Cas9. You can introduce the GPS
into these viral particle. And these viral particle
will be sent to your body targeting the area where we want
to modify and try to mutate. So some cells, not
all of your body, but some cells will
actually be modified and hopefully adding
a positive impact. And again, there’s some
clinical trials ongoing. There’s one, at least,
that started in July 2019. In this particular
case, they want to treat some form of blindness. So we’re going to inject
the virus containing Cas9. A viral vehicle, should
I say, carrying Cas9 GPS behind your retina and try
to modify some photoreceptor cells and maybe
correct the blindness. And finally, yeah, in vitro
germline editing medicine. It’s very obviously, we’re going
to talk about this today, where we’re deliberately
changing the genes and pass onto our children
in the future generation. Should we do that or not? So essentially, it’s
in vitro fertilization. You get your fertilized egg. You introduce your
CRISPR-Cas system, check it has the right
modification or not, [INAUDIBLE] cells, and return
the embryos to the woman. We’re probably going
to discuss that today. [LAUGHTER] And there are many
applications where I talk to you about CRISPR-Cas9. But there’s also
Cas12, Cas13, Cas14. Last week there’s
prime editing now. There’s many more
new technologies. I mean, it’s insane. Again, 16 papers a day. Really crazy. So in conclusion, I
hope what I tried to say is that viruses are abundant. You’re full of viruses. CRISPR-Cas is a natural defense
system that bacteria are using. You eat this stuff, if you
eat yogurt, you eat cheese. It’s highly diverse. It works in three
step in bacteria with the parts of
this natural system to make it a tool, a
genome editing tool. And maybe it has great potential
in medicine and definitely in agriculture. And you’re going to
see that in your foods way faster than you’ll
see medical application. And on that, I certainly
want to thank or CRISPR team in my group. So you see the list
of all the people that have been involved in
the CRISPR-Cas9 technology. And maybe highlight
Josiane Garneau was on the paper that showed
that Cas9 and CRISPR-Cas system was cutting the DNA
at a very specific. So our collaborators
at DuPont Danisco. I do not have any stock
on DuPont and Danisco. But our collaborators
for a long time. And I want to thank Yannick
Doyon, who provided me some slides for my talk today. Yes, we are working
on CRISPR-Cas, but we are a phage lab. So I certainly want to
thank all the people on my team right now working
on phage and on CRISPR. And I certainly want to
highlight here [INAUDIBLE],, which is in charge of our
phage collection, because we do have it in University
Laval in Quebec City a very large phage
collections that allow us to play around
with these bacteria and these viruses. And of course, all
our funding agencies and, of course, I’ll
be here all day, so if you have any questions,
I’ll be happy to answer them. Thank you very much. [APPLAUSE] – Thank you, Sylvain. Our next speaker is Jonathan
Kimmelman, James McGill professor and director
of the biomedical ethics unit in the department of
social studies of medicine at McGill University. [APPLAUSE] – Thanks very much. So nice to have the other aspect
of Quebec represented here. McGill’s an anglophone
institution. I’m going to be speaking
in English today. I always joke that if
you live in Quebec, the orchestra concerts
you go to are twice as long, because after they
do the French horn solos, they need to hear the
English horn solos. Anyway. OK, good. So let’s get started. So probably a lot of
people here in the audience are familiar with this story. Mila Makovec, who was
a six-year-old kid who was diagnosed with
Batten disease, a very rare, aggressive
neurodegenerative disorder, invariably fatal. And shortly after
her diagnosis, it was discovered that she had
a particular mutation in one of the genes called
CLN7 that interrupted the proper splicing
of this sequence, leading to the development
of her Batten disease. Within a very short
period of time, the scientists that had
diagnosed her condition as well as a group of scientists here
at Harvard, Tim Yu’s lab, were able to identify the
exact genetic sequence and develop an
oligonucleotide sequence that could disrupt that
glitch and enable proper splicing
of that CLN7 gene in order to produce a properly
functioning CLN7 gene product. So within a span of about
a year and a year or so, they went all the
way from diagnosis through to testing a
small genetic sequence, an oligonucleotide in rats. And then within
two months of that, they were already putting
this oligonucleotide sequence into the spinal
cord of this patient in order to control or
manage her Batten disease. So this was reported
widely in the press. Just recently, about two
weeks ago, the actual results were published in the New
England Journal of Medicine. And although in a
lot of measures, you don’t hear about this as
much in the press reports, on a lot of measures,
there wasn’t really too much of a change
in Mila’s symptoms. There was maybe
some interruption of the decline of some
function, but there was still some decline in function. There was, in fact,
a dramatic decrease, objectively measured, in
the frequency and duration of her seizures. And that decreased in
lockstep with the increase in dosage of all
oligonucleotide sequence. So this is a bit of
a window, in a way, into the future of
using gene editing. Now, obviously this is not a
gene editing technique exactly, but it’s very similar, has
a lot of properties that are really similar to this. And in fact,
oligonucleotide sequences are analogous to the
guide RNAs that would be used in a CRISPR-Cas9 system. So I want to use
this case, in a way, to motivate some of what I think
are some important limitations and policy concerns associated
with the application of gene editing in medical contexts. My talk’s going to divide
into four different sections. First, I want to talk about
some of the expected impact, medical impact, of
these techniques. I would argue,
actually, that far from having completely
revolutionizing medicine, I think that their impact is
going to be relatively modest. There will be, to
be sure, an impact, but it will be an
impact in niche areas. Secondly, I want to
suggest that there will be important challenges
in terms of definitively establishing the
safety and efficacy of these different techniques. And thirdly, and related
to that, that will also create major challenges in terms
of ensuring equitable access to these various approaches. In the last part, I want to
talk a little bit about sort of why I think these concerns– why they matter morally. OK, so let’s start
with expected impact. So we’ve already
heard a bit here about some of the
projected impact of these various techniques. It’s not hard to
find headlines that proclaim a completely
transformed medicine by using gene editing approaches. I want to sort of cloud
this picture a little bit and provide some context
for some of these claims. I think there are
three reasons to why the impact of these
techniques in medicine may be smaller than we
might think, at this point. The first has to do with the
incredibly long timelines. Developing products in medicine
takes a really, really long time, partly because the
science is just so complex. And as anyone here knows
from basic economics, there’s this phenomenon
called discounting. $1 today is worth less than
$1 20 years down the line. And so that means
that if we’re going to think about the value
of these interventions, we have to think about this
in the context of it taking many, many years to go
from a basic concept, from exciting science, into an
actual clinical application. So we can look at gene
therapy, or gene transfer, as a window into how long
those timelines can be. There’s been a sort
of recurrent theme so far that things go really,
really quickly in CRISPR-Cas9, but we haven’t yet really
talked about how slowly things go in the context of
clinical development. So for gene therapy
or gene transfer, the very first
trial was in 1989. It wasn’t really until
2017, about 28 years, that you had the very first
licensed FDA licensed product for gene transfer. If you look at a particular
disorder, hemophilia, this was always considered
the low hanging fruit for developing a
validated gene transfer technique in human beings
for various reasons. First clinical trial, 1993. But actually, the first trial
really showing some efficacy doesn’t happen until 2017, 2018. I’m told that we’re
on the cusp of an FDA approval of a hemophilia
gene transfer product. So again, we’re
talking about 27 years of a period of germination
for this technique. Lots of time. Now, that sort of interfaces
with my second point, which is if we want to think
about the value and impact, we have to think about
this in an incremental way, the way economists
think about increment. So as I said, we have about 25
years from the first hemophilia gene transfer study
into an intervention that looks like it’s
effective for hemophilia. And so it’s tempting, if we
want to think about the value, to look at how much
value we’re getting today compared with the way we
managed hemophilia back in the early 1990s. But of course, during that
period of germination, there were a lot of
other research lines that were progressing. For example, there’s a long
half life factor replacement therapy that made tremendous
advances in this period. There are monoclonal
antibodies that basically mimic the effect of factors and
some very promising findings in New England Journal
of Medicine in 2016. And then 2017, an RNA
inhibitory approach for managing
hemophilia A as well. So again, we have these various
parallel research trajectories that are non gene transfer that
are happening at the same time. And so if we really want
to think about the value and impact of gene
transfer for hemophilia, we need to actually
measure that against where we are with these
other rival techniques. And it’s actually a
bit smaller than it would be if we were
comparing it back to the benchmark of where
we were with hemophilia back in the early 1990s. Now, the third
phenomenon I want to talk about here that limits the
value of these approaches has to do with the
narrowing eligibility. When you develop
an intervention, at first you have
it in your mind that it’s going be
used for everyone with this particular condition. But through the process
of clinical development, gradually various populations
fall by the wayside, because they’re ineligible
for that approach. So let’s just go with
hemophilia again. Venn diagram all the
hemophilia patients in the world that will be
candidates for gene therapy. Right off the bat, we
can eliminate children. Given the current techniques
that we use for gene transfer, they can’t really
be used in children because of the particular
liver physiology of children. At least right
now they can’t be. Now on top of that, we
have to eliminate patients from low income countries,
middle income countries who aren’t going to be able
to afford hemophilia gene transfer, at least in
the foreseeable future. And then we have to
eliminate patients who have neutralizing
antibodies to the viral vectors that are used to deliver
the gene transfer approach. And then also have
to eliminate people who have inhibitors
to the factor that’s produced by the vector. And so at the end
of the day, you’re left with a smaller
population than you might have envisioned
that’s going to be eligible for this technique. And that’s a general theme
one sees in innovation. So bottom line, in
this section, I just want to say that for sure there
will be an impact, a really important impact. There will be transformation
in niche areas, but at the end of the day,
a fairly modest impact against all the other
innovation that we have elsewhere in medicine. My second point has
to do with evidence. I think there are
two reasons why collecting really good evidence
about safety and efficacy will be particularly challenging
in an era of precision medicine of use of gene editing. The first is a
fairly obvious point, which is that when we’re talking
about population precision medicine, we’re talking
about very small populations. Now, most of us
here probably know that the gold standard for
establishing the relationship Between getting a drug and
its safety and efficacy is the randomized
controlled trial. Now, randomized
control trials are premised on there being
a large enough population to be able to
accrue into a trial so that you can
eliminate or control the effects of random
variation in a population and really get a good
measure of the treatment, the cause and
effect relationship between a treatment
and a disease response. When you start moving into niche
areas like precision medicine, you’re taking that
large population and dividing it into really,
really fine grained strata. Now you have a problem of
actually populating trials in order to run those kinds of
randomized controlled trials. That’s a challenge not
only for measuring efficacy of these interventions. It’s also a challenge
for measuring safety. This is a really
humbling slide here. It’s a lot of information. All you need to know– imagine you have a new
medical intervention, and it increases your risk of
cardiac mortality by about one out of 1,000. And cardiac mortality
is already happening in the population at
about six out of 1,000, so your increase is going
from six of 1,000 to seven by taking this drug. You would need a
clinical trial that would enroll about
160,000 patients in order to detect that level
of increased elevation. So the bottom line, even
for licensed products, non-precision
medicine drugs that are used in large populations,
it’s incredibly difficult. You need large populations
to able to really detect subtle safety signals. That’s going to be a
major challenge in an area where you’re talking about
tiny populations that are going to be taking highly
personalized interventions. And on top of that,
in order to be able to up your statistical
power in that setting, typically you have to
use surrogate endpoints. These are indirect measures
of disease response. In cancer, for example,
tumor shrinkage is a surrogate endpoint
for what we really care about in medicine,
which is survival. Well, those are
imperfect measures. Oftentimes surrogate
endpoints don’t really give an accurate read on whether
there’s actual meaningful clinical benefit. And again, it’s a device that we
use to up our statistical power at the cost of really
being able to make certain and confident inferences
that a treatment is effective. My third point has
to do with access. Now, of course,
everyone here probably is aware that many of
these interventions are likely to be expensive. We worry about
whether people are going to be able to access
these interventions. But I want you to think about
this in some broader context. If we have a health care
system that has a finite size, it means that if
we’re going to afford these expensive
interventions, it means we have to cut back
on other interventions or the access to
other interventions. So it’s not merely about who can
get access to these techniques. It’s also a question
about at what cost, at what we’re going
to have to restrict in a finite system of
health care resources in order to enable those
patients to have access. And I think there are
three reasons why we can expect major challenges here. The first has to do with
the fact that we’re, again, we’re talking about
precision medicine products. These are highly customized. Number one, if a
company is developing a product for a really, really
niche market, a small market, it needs to charge a
very high price in order to recoup its
development expenses. So that’s number one. Number two, if you
have a bespoke product, it means you can’t
necessarily mass produce it and gain
those economies of scale of mass production. Now on top of that because
you have interventions that are developed
for niche areas, you also have less competition
in that economic space. And in principle, at
least, competition at least is supposed to be good
for bringing prices down. Doesn’t always work
that way, but that’s how it’s supposed to work. On top of that, there are major
challenges for health care systems that relate to the
fact that these interventions that we’re talking about are one
time expenses that occur at one point in a disease trajectory. So normally when
health care systems are reimbursing for
a treatment, it’s over the lifecycle
of a treatment. You take a drug chronically,
blood pressure medication chronically, and the
cost is spread out across that lifecycle. But also whatever cost
gains for that insurance system or the payer are
gained across that lifecycle. So if you have a one
time intervention, even if in principle that
one time intervention is cheaper across
the entire lifecycle than these other
interventions that are a little bit
less effective, it creates really big
challenges for payers, because they have to front that
money, a substantial amount of money, up front, and they
never know whether or not that person is going
to remain in the system long enough for that
system to accrue whatever cost savings are going
to occur from that treatment. So think about this. I think you know we
both live in Quebec. We’ve got a wonderful
health care system. Don’t let anyone
tell you it’s not. And so if someone gets
gene therapy in Quebec, I can tell you we also
pay high taxes there. I help to pay for that. That person goes off to
live in the United States, it means that our
Quebec health care system is not accruing the cost
savings of having delivered that gene transfer approach. Now, on top of that, in
order for us to even estimate the cost savings of
these interventions, we need to have really good
evidence of the long term benefit. How long after you introduce
gene therapy into the liver does the liver continue
to produce enough factor replacement to actually
correct the hemophilia? It may take many years
for us to actually have a sense of what
the actual trajectory is in terms– or the durability
is of those interventions. Oftentimes the
cost estimates are based on assumptions
that are really not very well evidence based
about how long these are going to last. Because of course, we’d
have to wait five or 10 years or 20 years to actually
measure those things. So this is just to
close this section out, just to mention the cost of
some of the products, at least in old school gene
therapy, not gene editing. If we look, for example,
at this product developed for a condition
called LPL deficiency. It was licensed in Europe. The cost of this product
was about 1.1 million euros. I’ll let you do the
conversion to US dollars. This is a product here that
probably people have heard of. It’s a CAR T therapy for
acute lymphoblastic leukemia. Here the cost is about $475,000
for a complete treatment that doesn’t include the ancillary
costs of the medical costs of managing a patient who
is taking this intervention. Substantial costs there. And then here’s another. Here’s the second licensed
gene therapy product. It’s a treatment for a rare
hereditary form of blindness called LCA deficiency. And this treatment
costs about 470– or is it $435,000 per eye. So again, a substantial cost. Now, let me sort of
close off this section by talking about a couple of
reasons why I sort of think it’s important to consider
all these issues about access, impact, evidence, et cetera,
in the context of gene editing. The first is that there
were tremendous pressures on our regulatory
systems around the world, but especially here
in the United States, to lower some of the standards
for licensing products, because there was so
much hope and expectation about these interventions
being effective. I think it’s really
important that we have some context for
assessing the impact of these interventions before
we acquiesce to these demands to lower regulatory standards. In fact, if we are going to
actually accrue those cost savings and accrue value
for health care systems, it’s crucial that we have
really good evidence of efficacy and safety before they actually
go out onto the market. Number two, I think
it’s important for us to remember that as
exciting as CRISPR-Cas9 is, amazing stuff
to be done, it’s important to keep other
research initiatives going along in parallel. Again, think about
that incremental value. There’s lots of other
value that we’re accruing through other
kinds of techniques as well. Thirdly, a little bit
counter-intuitive here. I just want to underscore
a lot of times people when they talk about
CRISPR-Cas9 and gene editing, they have in mind all these
sort of scary scenarios of customized babies, designer
babies, and people enhancing their intelligence, et cetera. I think, again, if you
think about how long the development timelines
are and how difficult it is to accrue really,
really good evidence, I think some of those
kinds of scenarios are things I worry
a lot less about. I’m much more worried
about issues of access and the quality of evidence. Let me just close really
quickly with a coda going back to the case of Mila
Makovec, which is what I started with originally. So in this case, first let’s
think about the expected impact. Clearly this seems
to suggest that this had an impact on her
illness trajectory. But again, there are
other development programs for Batten disease, numerous
interesting development programs. In fact, New England
Journal of Medicine 2018 published a really
important clinical trial about enzyme
replacement therapy, a kind of an old school
technique for managing genetic disorders. So not CRISPR, or at
least in this case, oligonucleotide therapy,
not the only dog in the race here in terms of managing
these kinds of disorders. Number two, let’s talk about
the evidence challenges here. We have one case
study amidst who knows how large the forest
is of other case studies. Now, because negative
case studies are never published in the
medical literature, we don’t really have a sense
of what the success rate is and even how
reliable, necessarily, this particular success is. Is it a fluke or is
it actually real? We don’t really know in part
because, again as I said, we don’t really know how large
that ocean is of other case efforts to try to
bring about remission of similar kinds of conditions. With respect to the
access, my understanding is that the oligo-sense
anti-nucleotide research effort for this
patient costs somewhere on the order of $3 million that
was raised through a GoFundMe campaign. So again, this is
not an intervention that most people are going
to be able to access. And with respect to issues
about why it matters, let me just note some
of the different rules that we normally have for drug
development that were suspended or that were under
considerable pressure in this particular case. And I’m not knocking the case. I just think it’s
important for us to consider some of the
context of this case. Number one, we had a
very abbreviated period of toxicology testing,
basically a month and a half of testing
a product that is supposed to be used
chronically in this patient. Number two, normally
with clinical trials, we are obligated to
prospectively register them before publication. To my knowledge, this
is the second time New England Journal
of Medicine has ever published the results
of a clinical trial without any prospective
registration. It’s a little bit
ambiguous whether this should have been
registered or not, but I’m just saying,
here’s a case where the rules about registration
are kind of a little bit under pressure. And number three, think
about the normal mechanism by which we prioritize
research, scientific research. Scientists are scarce resources. We really want to
make sure that we allocate that scarce
resource towards the most productive social ends. In this case, you don’t really
have a proper peer review system that is allocating
research effort. What you have is a
GoFundMe campaign and a scientist willing
to work with that funding. Now, again, I don’t want to– I’m not completely
knocking this case. I just think these are issues
that we ought to consider. And so I’ll just close
by saying that I think there are tremendous promise
of using CRISPR-Cas9 gene editing in many
areas of medicine, but I do think that
it’s important for us to be thinking about
some of the cost issues. And I think it’s probably
much more realistic that we think about these having
impacts in particular niche areas. They will have big impacts
in those niche areas. But I’m not sure
how generalizable this will be across
other areas of medicine. Thanks. [APPLAUSE] – So thank you both. Now, we’re going to have
a conversation here, and you’re just going to
listen in on our conversation, and then we’ll open the
floor for questions. OK? So thank you both,
Sylvain and Jonathan. As I listened to
both of you, I kept thinking the hope
and the hype and how we balance those things. So Jonathan, you talked a
bit about the long timelines for impact, and I wanted to
hear your take on that, Sylvain, in terms of we’ve
seen what’s happened with gene therapy,
gene transfer, and how long it has
taken for us to get to FDA approved products. What is your sense on what will
happen with CRISPR or with gene editing? Let me be broad. Gene editing in
terms of how long it will take for us to get
to a point where we can think of this as being ready to go
and even talking about safety, as Jonathan mentioned. – Want to go first? – Oh yeah, you go. – I don’t have the
answer to that. Something I try to say is
that CRISPR-Cas9 is already a success. It’s already been
used in research. And there are new drugs
that are being developed because we use
CRISPR-Cas9 to understand how a human cells are working. So from a research perspective,
using the technology, for me it’s already a
clear 100% success, because we know more about the
human body or human cells today that we knew five years ago. So for me, it’s
already, it’s a given. Now, how can we translate
that research knowledge and using the
CRISPR-Cas9 technology to treat people and have
safety issues and all that. That’s why the clinical
trials are ongoing. There’s only a few
of them are ongoing, but they’re ongoing here. And when I don’t
know the answer, usually I stop here,
because I really don’t know. Hopefully it’s going to work
in some cases, but for sure. And that’s why sometimes
when people are calling for to stop any research or any use
of the CRISPR-Cas9 technology, I really hope it’s
never going to stop being used in the research. Because again, it’s a
tremendous, tremendous research tool. – I’ll just add something
really quickly to that. So I think it’s
helpful to distinguish between this as a research
tool versus as an intervention. So as a research
tool, again, I talked about these different parallel
development trajectories. This is a research
tool that will help those other parallel
research trajectories. So it may very well
speed any number of different applications,
not all of which are direct applications
of this in human beings. When it comes to direct
applications in human beings, I think about this just– who knows what the percentage
is, but 90% of the time when you have really,
really novel platforms, it takes decades to go from
basic concept into application. We look at angiogenesis
inhibition, 25 years, monoclonal antibodies, a similar
amount of time, gene transfer, any number of area. So it generally takes a
really, really long time to go. It’s not like with computers
where we have Moore’s law. Biological systems are
just so much more complex. Humans are hard to control. There’s so much
variation, et cetera. But having said that, I
said that 90% of the time it takes a long time. I don’t know. 10% of the time, you
get really quick wins, or maybe 5% of the time. I don’t know what
the percent is. So I wouldn’t rule it out. We have to think in
terms of probabilities. – Yeah. Because I guess my
question should not have been how long do you
think it’s going to take. I think it’s more a
question of, do you think this will have a shorter
timeline than gene transfer, for example, given that we
know more and the technology is quite different? – Yeah, well, I’ll say something
about that really quickly. So I study a little bit of this
in more sort of small molecule context, not so much– but you have this sort of
funny thing going on in science and therapeutic development. So we’re learning more and more. We’re becoming more
and more clever. We have different kinds of tools
to get at different problems. So on the one hand, things
should be getting quicker, but that’s always
offset by the fact that we’ve solved the low
hanging fruit problems and that we’re getting to more
and more difficult problems. And so we’re constantly
sort of at this tension between getting smarter
and dealing with more and more difficult problems. Now, where exactly the
equilibrium ends up, I don’t know exactly. But I think that we should
not lose sight of the fact that our cleverness
is always competing against our solving
easier problems or moving on to harder problems. – Sylvain, what do you
think is the bottleneck or could be the bottlenecks
in the technology and the advancement
of the technology? – You mean for
human applications? – Yes. – There’s a lot of
discussion about is it going to cut or modify
at the right place. We call that off target issues. And it’s much better
than it used to be. So essentially, when the RNA– I mean, the GPS, you hope
it’s going to go and cut the right piece of DNA, and
it’s going to be replacing, or you’re going to
modify what you hope for. And we have seen that when
you do this in viruses or you do that in
bacteria, most of the time you get what you want. But when you do this experiment
in a much bigger genome like our genomes,
it doesn’t always go and cut the right place. So that’s a huge issue. As a research tool,
of course, if you do this experiment
in your lab and we’ve got the tools to see if it
cuts at the right place right. And if it doesn’t cut the right
place, you throw the plant away or you throw the cell line away. You can’t do that with
a human, obviously. So that’s really a big issue. And it’s getting better,
but it’s not 100%. And where it’s going
to be a cut off, that’s going to be–
that’s definitely a bottleneck, for
sure, for sure. And also it’s interesting,
because there’s also all this area– it’s
not human medicines, but there’s a lot of research
going on in agriculture. Can we use this technology,
for example, to improve– I mean, we’re 9 billion people. We need to feed this planet. And can we improve on all
the agricultural crops? And so CRISPR has
obvious been used a lot. But then you get
into countries that see this GMO use
differently, like in Europe or in Canada or here in
the United States, which is a bit different. I was trying to [INAUDIBLE] in
my talk is that for sure CRISPR has already been used
in crops and it’s going to be much better than
what we used to be doing. Will people embrace that or not? It’ll be interesting. So it’s like GMO debate
again with more information, faster technology. So that’s also–
yeah, we could discuss later about this as well. – Any response there, Jonathan? – No, I think Sylvain captured. Yeah, thanks. – All right. Let’s talk a little bit
about precision medicine. So Jonathan, you brought
that up quite a bit. And there are
varying perspectives on what precision medicine
is and what it’s not. And the way you presented
it, it was presented as it’s all about the genome. The way the NIH
has presented it is that it’s looking at genes
and environment and behavior and looking at those
variables that come together. And for us to be able to
treat and cure and prevent disease based on our knowledge
of all of these factors and how they interact,
not just the genome, many people interpret it
to be just the genome. And that’s part of the hype is
that it’s all about the genes. And so you talked about the
fact that precision medicine as conceptualized
around the genome will– primarily around the genome– will cause us– the impact to
be limited to small populations. But the way the All of
Us program here in the US and the Precision Medicine
Initiative in the US is framed, it’s about all of us. It’s about all of us. So we’re talking
small populations, but we’re also talking
the population as a whole. So what do you think of that
critique of gene editing technology as being limited
to just a select few or small populations versus the way
I think NIH and others are thinking about it? Whether they’re doing the
work that’s going to get us there is a whole other question. But the way they’re
thinking about it as being applicable to our
broader society and larger populations, not
just a select few. – I’m trying to think
of how to respond. I mean, to be honest,
I don’t really even know what the term
precision medicine means, although I’ve tried
to study this. I study precision medicine. But it just seems like
people use this term in so many different ways. I’ve heard people say that
having a Y chromosome is a– Y chromosomes are biomarkers. I suppose, that’s
true, in some sense. The version of precision
medicine that I have in mind is the version that you
see implemented in cancer. And of course, the
aspiration is to be able to take a patient’s
tumor and type it and be able to match a drug or
match a treatment regimen based on whatever markers. Now, it turns out that there
are, at this point at least, a limited pool of markers that
we can match people to drugs, but the aspiration
eventually is to be able to take 100% of cancer
patients, type their tumor. And so there is an– and determine the treatment
regiment based on that. So there’s an
aspiration, obviously, to reach large populations. But again, at the
end of the day, if you are talking about taking
a population of lung cancer patients and breaking them
up into splinters of patients with BRAF mutations
and KRAS mutations, EGFR amplification, et
cetera, et cetera, again, you’re talking about
having smaller populations in which to validate your
particular management strategy. And I just don’t see– I mean, if you think about
the very concept of precision, I just don’t see any
way around the fact that you’re talking about
having fewer patients in which to understand how an
intervention works. – OK. Let me hear if Sylvain
has a response to that. – Now you’re talking about NIH. And I think what
personalized medicine, I think what you want is to have
therapies that will, yes, will be personalized, but also
applicable to other people. And I think you definitely
need, I’m sorry to say that, but you definitely need more
funding for basic research. And I think everyone, I mean,
of course, I’m biased here. But we do need that funding
to increase our knowledge and try to make it, yes,
personalized medicine, but to be available to many
diseases and to many people. And that’s why there’s a hype. There’s a hype,
because as a scientist, when you find something
interesting, you talk about it, you’re passionate about it,
you want to tell people, this is cool, this may work. And then you need to tell our
elected officials, hey, look at what we did with
the funding you got, but we need to go further. So that’s why there’s a hype. There’s a hype, because
we’re really passionate, and then we know
we can go further if the funding is there. I know it’s a cliche,
but it’s so true. It is so true. – Yes. And of course, the
hope piece is that we want to give people hope. There are lots of families
suffering with individuals suffering from devastating
diseases, genetic diseases, monogenic diseases that could
benefit from the technology. So the balance is,
I think, what we’re going to be talking about
all day today really. It’s how we balance
those things. We have just a couple of
minutes for our discussion, and I wanted to wrap up
our conversation by talking a little bit about access. Jonathan, you mentioned
access, and that is a very important part
of the conversation that often gets left out. But recently, it’s
interesting that we’re having this now, this
conversation here now, this symposium here today,
because just recently, the NIH and the Bill Gates Foundation
announced that they’re going to be working
together to bring some of these technologies to
Africa looking at sickle cell disease, which we’re going to
hear about in a little bit, and HIV. So they’re addressing
the access issues. What’s the problem? I mean, we’ve got Bill
Gates on board, right? [LAUGHTER] So we don’t have to
worry about that. Do we, Jonathan? – I know that’s a
rhetorical question. I think you’ve
answered it yourself. I mean, obviously
the Gates Foundation and many other initiatives
have transformed access to tuberculosis medications,
vaccines, HIV management regimens. So there has been– back in the ’90s when AZT
was first being developed, there was a lot of
discussion about whether or not we should run
clinical trials in low income countries. And many people
objected, because they thought there’s no way
these countries are going to be able to afford AZT. But actually, at
the end of the day, partly because of the positive
results on those trials, NGOs were able to
bring enough pressure to bear on pharmaceutical
companies and governments to actually lower the cost
to make them affordable. So even though your
question is rhetorical, I do think there
is some promise. There are certainly
some areas where we have been able to increase
access, whether that’s universalizable outside of
neglected and highly prevalent diseases to me is a really,
really big question. I’m not sure I
see that happening in the context of gene transfer,
bone marrow transplant, et cetera. – Sylvain, you’ve
got the last word. – Well, access is
obviously key there, but we’re still at the
beginning of all this. The first clinical
trial started this year. They were approved last year,
but they started this year. There’s only a few patients that
they’re starting to be treated. So we’re a long way to go. So yes, there’s hype. It’s a very powerful
technology in the lab. Is it going to be safe? Probably, maybe, I hope. So we’re still– I don’t want to make it– I don’t want to add
to the hype already. And hopefully down the road,
everybody will have access. That’s what we hope for. Or come to Canada. [LAUGHTER] – We’ll probably come anyway. Thank you both so much for that. We’re going to open up
now to the audience, to you, for your questions. There’s a mic in the
middle, and we’re going to ask you to come to
the mic to ask your question. We’re also going to ask you to
make sure that your question is a question and not a statement. So let’s go. Want to hear. Oh, lots of questions here. – Can we have more coffee? [LAUGHTER] – We’re going to ask you
to state your name as well before you ask your question. – My name is Alyssa Goodman. I’m another science
director at Radcliffe, but this is not a
planted question. My question is you said,
Jonathan, that this is not like computers. And one of the things that Mac
pointed out in her introduction and in my conversations
with her is that the big deal
about CRISPR-Cas9 is that it is programmable. And I’m just curious, if I was
going to play devil’s advocate, if what you were saying
about maybe this isn’t really for everyone and it’s too
expensive per dose, et cetera, it sounds a lot like
what people said about when Bill
Gates said things about personal computers. And now we have this
incredible, open source, very modular software
environment where people make solutions
very, very cheaply and very individualized
solutions. And so I’m just curious whether
you and Sylvain and Charmaine see a future where
personalized medicine is sort of programmable, iterative,
very personalized, and somewhat safe, hopefully, medicine. Is that possible? – So I’ll start. So you raise important points. I mean, for example,
the Mila case. I mean, in principle,
academic hospitals can produce oligonucleotides
that are customized for particular mutations. So in principle, there
are different kind of production paradigms
that we can think about. And I think it is an
important consideration. I do think that,
nevertheless, we’re still stuck with
major challenges in terms of delivery
of these techniques, in terms of being able
to achieve the expression levels that we need,
if we’re correcting a particular gene that we
can up the expression level. In the case of neuro
degenerative disorders, some of the cases that have
been successful stories are cases where,
again, the physiology is kind of on our side. You don’t need to have full
diffusion into all the tissues of the brain. The spinal cord is
sufficient, et cetera. So I think, again, we’re sort of
stuck working against biology. And the other thing
also is we’re still stuck with these issues
about really demonstrating efficacy and safety
in populations and really having good
grounds, not only for believing the risks are worth it,
but also that it’s worth reimbursing these. So I take your point,
and I think it is a– I think it’s conceivable that
there are niche areas, niche kinds of diseases,
for which we can have academic laboratories
that are producing these oligonucleotides. Harder for me to imagine
that’s going to be scaled up to apply for the kinds
of prevalent disorders that we’re stuck
with and that we’re going to have good evidence
to suggest that’s worth doing. – Thanks. – I agree. Delivery will always
going to be an issue. And so I think if you want
to do personalized medicine to mass people
like today, I think you have more chance
with microbiome or foods to change than CRISPR-Cas9. That’s for sure. – I want to ask about
sort of international road uses of CRISPR. Dr. He in China, of course. The expectation that a
[INAUDIBLE] scientist has claimed that
he’s going to do the same kind of way
preliminary intervention in the human germline. So question mark. Is there anything
that can be done other than sort of
international pressure from governments and scientists? How does one control
kind of access road preliminary dangerous uses? – That’s a great question. It’s hard to– I mean, mad scientists
are everywhere. There are people that
you can’t control. But I like to think that
there’s not that many. And what happened in China,
I mean, his career is over. So I think if people start
doing that, their career will be over right now. So that’s one issue. Yeah, that’s the
main issue, I think. I don’t think we can stop
people doing anything. But I think– and also sometimes
if your parents and emotion takes over and it’s hard. And if parents wants to
do that and then you’ve got a scientist that is willing,
it’s hard to block that. But you would hope that
the scientist would or a doctor will actually
tell them that it’s very risky and there’s a lot
of ethics concerns that we need to
think about that. So yeah, it’s not
an easy question. It’s really not. I’m sure we’ll talk about
this at lunch or at break. Yeah. – Yes. – My name’s Elizabeth
Guggenheimer and I want to thank
the panel and also Radcliffe for all the
science conferences. My question is actually a follow
up from a previous conference. You didn’t touch on
gene drives to eliminate vectors of diseases. So for example, mosquitoes
and all the mosquito borne diseases and
ticks and Lyme disease. I wonder what your
thoughts are on that. – That’s me, right? No. – I’ll just say that
it’s not an area that I consider myself
sufficiently expert on to say anything
intelligent, so I’m going to let Sylvain take this one. – Oh, lovely. [LAUGHTER] No, also I’m really not, but
it’s a very good question. I mean, should we
modify mosquitoes? And if we start removing
insects or we start removing– what’s going to replace it? And what’s going
to happen there? It’s a very complex question. And we actually see
that in our field. What’s happening is we’re
studying a phage interaction with phage and bacteria. So we’re having
bacteria now that are resistant to phage
thanks to CRISPR. And then we’re putting this out
there in the field in foods, for example, because
that’s what we– so just step back
a little bit here. So when you make
cheese, you make yogurt, you’ve got bacteria
inside of your– because those are
the good bacteria. Phage are found in milk. They’re naturally found there. So sometimes when you
taste your cheese, your cheese doesn’t taste
great, because phage have killed the good bacteria. You don’t have the
flavor development. So that’s a process problem. So to solve that,
you use CRISPR. It’s a natural system. And then we can make the
bacteria resistant to phage. So we’re doing that. It’s great. So what happened is
that when we start putting those strains
out there, we’re killing the phage that
are causing issue, but new phage are coming up. Something we have
never seen before. They’re fine, they’re phage,
but they’re different. So in the gene
[INAUDIBLE] scenario, what’s going to happen if
we start removing insects and what’s going to take over? That’s the question, really. So I don’t know that. Yeah. – All right, thank you. Thanks. – Hello. Sorry. My name’s [INAUDIBLE]. I’m just a high school student
who’s interested in CRISPR. – Oh, great. – My question is I know that
CRISPR-Cas9 has been successful in editing different organisms
such as mice and even in the agriculture field. But in terms of human gene
editing, I want to know– I know that there isn’t
exactly a blueprint, and it’s not 100% accurate
to fix genes in humans such as diseases and such. But how will scientists
ensure a better oversight of gene editing in the future? – I mean, I can start. This is a little bit– your
question is a question that in some ways is similar
to a question that was raised by two people ago. In terms of oversight
of human studies, we do have ethics
committees that review that, regulatory
bodies, et cetera. So I don’t necessarily see
it being too much different, the oversight of that, in terms
of, again, the human studies. But again, we have
this phenomenon of potentially rogue actors. And I guess I would
echo Sylvain’s point. It’s hard in a
large world that’s driven by all sorts
of different forces to ensure that you have
absolute sort of uniformity in terms of adherence to
whatever international policies there are. What I can say is you can
make it really, really hard for people to do
these kinds of things by making it very difficult for
them to advance their careers. So just as a quick
example, many times people want to use these
techniques because they want to be the
first one to do them and they want to get the
scientific notoriety. They want to get a
paper into Nature or one of those single named journals. I love that line. And those journals
now will typically require ethical
review of manuscripts before they accept them. And if a study has
been done inconsistent with international
consensus or guidelines, they’ll reject them. And so it gets much harder
to advance your career by stepping out of line with
what our global consensus on policy is. So I see that as kind of
one of the main vehicles by which you can make
it really, really hard to do that kind of thing. – I’d also add to that quickly
by saying that scientists also have the responsibility
to let others know about other scientists
they know who are going rogue, right? When Dr. He was doing his work,
there were some scientists who knew or who had some
inkling of what was going on. Some didn’t even know
who to go to talk with. So scientists having
responsibility for more than just
their own science but for the field as
a whole and for being bold in speaking out
when they know things are going on underground. Some scientists are more
likely to do that than others, but I think all scientists need
to take on that responsibility. – And by the way, that’s
great that a high school student is here. – Yes, absolutely. [APPLAUSE] – That’s another reason why it’s
so great to have CRISPR-Cas9 discussions, because it raised
the interest in high school students, and that we
do really need that, have more high school students
interested in science. And actually, it’s funny,
because there’s a lot of– it’s not funny, it’s great. There’s a lot of
schools that are introducing CRISPR-Cas9
or CRISPR-Cas into their curriculum. So students, undergrad
students, are doing CRISPR-Cas experiments,
which I think is also great. – Great comment there, Sylvain. Yes. – I’m [INAUDIBLE]. I’m a neurologist in
[INAUDIBLE] Canada, working on rare diseases
and precision medicine at [INAUDIBLE]. – We have a lot of
Canadians around here. [LAUGHTER] – My question is beyond therapy
for genetically determined diseases. I’m talking about what is your
take on disease prevention. In particular,
concerning, for instance, factors like gut flora,
environmental factors for conditions that are
killing our people more, like obesity or maybe autoimmune
diseases, multiple sclerosis, for instance. We know there is a link and
we can modify flora positively to prevent those conditions
from developing the symptoms, basically. – I can go. So that’s a bit outside
CRISPR-Cas9 in a way, isn’t it? – Not necessarily. – Well, I think the gut flora
is clearly part of our– I mean, we need to take care
of that, that’s for sure. And there’s so many
microorganisms there. The viral population
in our gut microflora is we need to know
what’s in there. And it’s clearly,
again, I was trying to say that viruses
are controlling bacterial populations. So every time you find
bacteria, you’ll find viruses. How can we domesticate
that to understand how phage or
viruses and bacteria are interacting together? That’s clearly part of
the personalized medicine thing that’s going to happen
sometimes down the road. CRISPR-Cas9 in that
area, I don’t know. Yeah. Maybe I missed that part. – CRISPR is for, for instance,
in your area of work, can modify the bacteria. – Oh, I see. Yeah, yeah, OK. – And those bacteria,
for instance, the ones we have seen
in multiple sclerosis, like [INAUDIBLE] bacteria,
they can actually be linked to causality in some studies. And if we modify them, we can
probably prevent the disease. – OK. We’re going to have to move
on, because we’re running short on time, and we want to
make sure we get everybody. So no follow up questions. Just the question
and the answer. – All right. So first of all, hello. My name is [INAUDIBLE],, and I
am also a high school student. And specifically, I’m
interested in the intersection between machine
learning and how it can be used to
optimize gene editing technologies like CRISPR-Cas9. So my question is for
Professor Kimmelman. And in your
discussion, you talked about the
developmental timeline. So I’m curious to
see your thoughts on using machine learning
and neural networks in order to better predict
the kinds of gene technologies and their specific
applications and what effect you think that would have on
the development of timeline you mentioned. – OK, thanks. [LAUGHTER] Maybe that came out wrong. – Thanks but no thanks. – It probably won’t
surprise people in this audience to
know that I’m also a bit of a skeptic about
the whole machine learning thing in medicine. [LAUGHTER] – Machine learning is great
if you have really, really reliable data that you can
put into your algorithm, and I think it’s
really challenging collecting really,
really good reliable data to feed into those algorithms. The other challenge
with machine learning, obviously, is you
can’t often– it’s really hard to
peer under the hood and to understand what
those algorithms are doing and whether they
really make sense and what kind of assumptions
are built into them. So I actually
think that I’m just a bit of a skeptic
about how much machine– I’m sure, again, there’ll
be niche applications, to be sure, particularly
in areas like diagnostics. I’m just not sure
that machine learning is going to solve the
big problems of taking a basic principle and developing
a useful intervention for it that’s going to work in
populations of human beings. – Thank you. – Thank you. – Joel Heller. I’m not a high school student. [LAUGHTER] I have a question about the
[INAUDIBLE] model for CRISPR. It seems like it’s a
technique and not like a drug that you’re developing. And so the access could be
just a lab person and a doctor. And so how does that work in
a global sense and as far as patents go? – I could take a stab at that. So again, it depends
on what applications we’re talking about. But if you have to deliver
something with a vector, I assume that the
vector is going to have intellectual
property around it. So if you have to package it
in a sort of [INAUDIBLE] virus or something. So there’s going to
be IP around that. If you’re talking oligo-sense,
oligonucleotides, there may be particular chemistry
around the nucleotide sequence that has intellectual property. And I think one thing
that we have to remember, again, if we look
at the history, look at the innovation
in medicine, you don’t just invent
something and you just use it for 20 years exactly as it is. You look at medical devices. They always have these
kind of incremental changes that bring a little
bit more value added. And by the way, they also give
you new intellectual property patents that you can extend. So I think it’s a
little bit enticing to think that maybe
we have just– we’ve put it into a program and
it spits the oligonucleotide and we use that in patients. But I think the reality is
that the way that innovation proceeds, there’s always going
to be some space for protecting the intellectual property around
whatever the sort of cutting edge techniques
that we’re using. But maybe I’m wrong. That’s my sense. – Absolutely. – Thank you. Next question. – Hello. My name is Mikey. Unfortunately, I am another
high school student. – Not unfortunate at all. – Thank you. So I’m specifically
interested in how genetic technologies
like CRISPR-Cas9 can be used to solve global
problems like climate change. So a question for
you, Sylvain, you’re studying the co-evolution
between these phages and the bacteria, which
is incredibly complex. I mean, you have CRISPR and
then you have anti CRISPR. So I guess another question
about how computing innovation is going to come into play. With this, are you guys
finding yourself working more with computer scientists to
aid in this process or do you think like quantum
supremacy [INAUDIBLE].. Do you think that’s
an over hype? – Yeah, so great questions. Great high schools around here. [LAUGHTER] So yeah, indeed. So climate change is one thing. Obviously, crops is one
thing that’s really clear. I mean, we need to improve
crops because of global warming. So CRISPR-Cas9 is
definitely a technology we can use to improve crops. I was just reading
recently that I think there’s so many
patents in agriculture related to CRISPR-Cas
and there’s even more patent in the
agriculture side than there is a human health. So there’s a lot. So that’s definitely going
to be an issue in the sense that people are
going to study this. In plant science, you gotta
also think about this. So when you do, for example,
you want to change a crop. So there was
different people were using different technologies. You don’t want to know
what they were using. But it was really,
really complicated, and it was an error trial base. And now we’ve got
this technology where you can do
your modification, and then you can almost be sure
that you will get your plant after a couple of months. Before that, you
were not even sure. You would do your
modification randomly and then you’d, again, try
to see if this would work. And now you’ve got
this technology where you’re getting get your plant. For sure you’re going to get it. And then you can study it. Then you can ask
research questions that are totally
different today than five years ago before
the technology was there. So in the plant
science, CRISPR-Cas9 has changed totally how you
do research, because you can do this much faster. So in terms of climate
change, definitely it’s going to have an impact. Now, in terms of
bioinformatician, yeah, we work with
them all the time. I mean, yeah,
because [INAUDIBLE].. Seriously, they’re really–
this is definitely– and I think we’re
seeing more and more people that are of
interest in computer have also degrees
in biology and then they’re understanding
both sides is fantastic. There’s more and more
bioinformatician. And I come from a department
that teach biochemistry. Actually, when I arrive
at my department, we were a biochem department. A couple of years later,
we became a biochem and microbiology department,
and now we are biochem, micro, and bioinformatics. So clearly it’s really
a good field to go into. – Thank you. – Thanks very much. – Hi, my name is [INAUDIBLE]. I am also a high school student. And I’m interested in
the sort of intersection between computer aided
genome analysis and gene editing there. My question is about the sort
of economics regarding this and how giving the sort
of inherent unscalability, as you mentioned, of
personalized medicine and personalized gene editing. Do you think that it will
ever get to the point where the prices can
come down where it’s open to the average consumer? Because you can only
apply the specific therapy to a small subset. Will the prices ever come down? – The pricing is not
related to the technology, because doing CRISPR-Cas9
is not expensive. What is expensive is the
clinical trial it goes with it and finding the right
delivery vehicle, like John was mentioning. Because treating blindness
or treating liver disease or treating blood disease
is not the same delivery. So in the clinical trial to
make sure that this is safe, that’s where it’s costing. I don’t think it’s going
to go down, though. I’d be very surprised. But yeah, I don’t know. But like a research tool, that’s
why so many people are using it, it’s not very expensive. You just need to– you buy the Cas9, the vector. You just need to change the
guide RNA, which is very cheap. So to do this in a lab
is not very expensive. But to expand that
to treat somebody, you’ve got safety issues. So you need clinical trials. – Well, I guess I was wondering
about like a different process, because right now you have to
go through a similar process to clinical trials for
each individual operation. But now given that you think
about doing a different for every person, is
there another process that could be implemented now? – I don’t know. – So we’ll– follow
up with him later. We have to– thank you so
much for your question. We’re going to try to have
everyone ask their questions, so we need to move along. Next question, please? – Hi, my name is [INAUDIBLE],,
and I’m another high schooler passionate about personalized
medicine and structure based drug design. So in current target
based drug design, you have to screen
billions of molecules, then do lead optimization. I know that process
takes a super long time and costs a lot of money. So if we’re thinking
about the uses of CRISPR in the long term,
wouldn’t it benefit us and save a lot of money
from this process? Because instead of going
through an integrated proteomics transtomics genomics
approach, we’re going directly to the
genes and editing them. So we wouldn’t have to go
through all the ligand receptor analysis and save a
lot of time there. – So there are CRISPR
screens going on right now. So Cas9, you can modify. You can make it to go bind
something and not cut. So you can play with expression
of genes and production of proteins. So you can design
thousands of guide RNA and do screens to
look for targets. So what you’re trying
to say is there’s some of that going on already. So it’s not– yeah. – All right. Thank you. – But it’s a great point. Yeah. – All right, last question. Oh, she’s– oh, OK. – [INAUDIBLE] That is the key. – Thank you all so much. – I’m sorry,
Charmaine, I’m sorry. – No, you go ahead. – OK. I want to thank you,
Sylvain, and Jonathan. They truly set the
tone for the day, which is to really strike the
right balance, if we can, between the innovations, the
discoveries, and the ethics, and the economics. So with that, I’d like
to suggest a break. Actually, take a break. Not even a suggestion. And there’s coffee in
the back, and there are bathrooms on the first
floor and downstairs. – Thank you. – Please be back at 11:00. 11:00, please. [APPLAUSE]

Leave a Reply

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