Note: The I AM GPH podcast is produced by NYU GPH’s Office of Communications and Promotion. It is designed to be heard. If you are able, we encourage you to listen to the audio, which includes emphasis that may not be captured in text on the page. Transcripts are generated using a combination of software and human transcribers, and may contain errors. Please check the corresponding audio before quoting in print. Subscribe now on Apple Podcasts, Spotify, Google Play or Stitcher Radio.
EP05 The Future of Gene Editing with Siddhartha Mukherjee and S. Matthew Liao
Deborah Onakomaiya: Hi guys, and welcome to this special episode of I AM GPH. I am your host Deborah Onakomaiya, and on today's podcast we're going to be featuring excerpts from a recent lecture given by Pulitzer Prize winner, Dr. Siddhartha Mukherjee. Dr. Mukherjee is best known for his 2010 book called "The Emperor of All Maladies: A Biography of Cancer" as well as his 2016 book called "The Gene: An Intimate History." On October 10, 2017 Dr. Mukerjee visited New York University to give the inaugural William C. Stubing Memorial lecture called "Are we ready to edit our genomes?" Here he presented a snapshot of the past, present, and possible future of gene editing. He also raised many scientific, humanistic and ethical questions about this rapidly evolving field. Joining us also on this podcast will be Professor Matthew Liao. He is the director of the Center for Bioethics here at NYU GPH and also the Arthur Zitron Professor of Bioethics. Please stay tuned after listening to Dr. Mukherjee's lecture as Professor Liao will be joining us in the second part of the podcast with some thoughts and reflections about this complex and exciting field of work. So let's get started, guys. I hope you enjoy this special episode.
Siddhartha Mukherjee: In the last five odd years. So if you think about again, reading and writing, reading genes and writing genes, reading being the capacity to decipher information from the genome, writing, referring to the capacity to intervene and make deliberate changes in the genome. Both of these technologies are changing and that's really where we are today. So I'll give you one example of something that happened last week. This is a paper published last week in which the authors use genetic information alone to predict human height. So on the X axis is the predicted height based on genetic information. On the Y axis is actual height based on genetic information. And the trick to this paper was to unleash deep learning on the human genome. So this is about several hundred thousand, so 2000 individuals. This is now being generalized to 500,000 individuals on a biobank. And human beings, human eyes, our intelligence, our standard ways of thinking about the world are not sufficient to predict height. It's too complicated. That was very safe because we could say we understand that height has a strong genetic and inheritable quality. Identical twins tend to be the same height, particularly in the West where their nutritional status is equalized if they don't have diseases and so forth. If there is no great environmental influence, if there's no accident or change in nature. So we lived in a relatively safe zone and that was a zone of ignorance. We said, we cannot predict this, it's too complicated. We just opened about 10 or 20 days ago. We opened a new box there by saying, okay, maybe human beings can't predict it, but what if deep learning computational techniques could predict it. And in fact, this is deep learning on 2000 individuals. This R value, people who are statisticians will know this is 0.64 it's not great, but it's not bad either. And we think based, I spoke to the author of the paper, we think the accuracy is within 0.6 centimeters. So in other words, forget about asking anything about family history. Forget about asking anything about the individual. This just takes the sequence of a genome and using deep learning techniques spits out your predicted height and you get to about 0.6 centimeters of real height. This is a morally complex universe because if we were to be able to predict this for unborn children in societies where height carries a premium, if we were to generalize this across other poly-genic diseases, and remember we were in the safe zone. We said it's too complicated. We can leave it all alone because it lies in some kind of weird black box. We just put the black, we just opened the black box and inserted it into the human genome. There are skeptics who will say, this is an unusual situation. You know, height is very heritable, absolutely correct. Others will say, okay, but what if. You know what if you get too close enough with other things, what if you get to the risk of cancer? What if you get to the risk of Alzheimer's disease? What if you get the risk of neurological diseases? On one hand, diabetes? On one hand you can imagine a new kind of medicine that will be born out of that. Powerful ways to interrogate human beings, predict people's futures. On the other hand, what if this enters territory like height? What if it enters the territory of IQ? Whatever that measures. What if it enters sexual orientation? We have just stepped using new arenas into territory that we thought was so complicated that we would leave it on its side. Now this is of course just the reading part, right? So this is just what's happening in reading. So just to finish and summarize that arena, obviously these points should be, should be evident to everyone. This has consequences for virtually every human disease. It has consequences for health maintenance, surveillance and risk as consequences for insurance and risks, risk sharing. People often ask me, well, what about insurance? And I said, well, insurance is a mechanism for risk sharing. If you have, if you're ignorant about risk. You can't have insurance when you are not ignorant, ignorant about risk. If all of a sudden you're all to be non-ignorant about risk, we would not be able to share risk anymore. So the standard mechanisms that we have insurance would be challenged. On a side note, this are, these are my, I'm revealing my personal political proclivities. I can't think of a more strong argument for universal basic health insurance that in fact one of the consequences of the deeper interrogation of human genetics has to be that we have to radically equalize ourselves when we realize that in fact we're radically unequal. There's no other consequence. So that is the so-called reading part of the human genome. And then I'll move on to the writing part and I'll stop. I'll read a section of the book where all of this is addressed. So while we're reading, we're also doing something called writing. And by writing I mean changing the actual code itself. So if reading involves going to the encyclopedia and deciphering the nature of information, writing involves going to page 347 in volume 16 and erasing a particular set of ACTG and replacing it with a different code. This is an article that I wrote for a magazine, which I would encourage you to read about BRCA 1, the gene that increases risk for breast cancer. Here's a picture of, there are several people involved in this. There are patents and cross patents and lawsuits going on about who owns this technology. I'm not going to enter that, but anyway, this is Jennifer Doudna. I'm doing a talk with Jennifer in January and her, a woman who is her lab manager. But Jennifer, of course, as many of you very well know, is one of the leaders in inventing technologies to allow us to change, to make deliberate changes in human genomes. This is the cartoon. I have to remind ourselves that this is not one technology, this is a suite of technologies. People somehow have become obsessed with one enzyme, one gene Cas9, CRISPR, et cetera. But CRISPR and Cas9 are really a suite of technologies that allow us to begin to think about making deliberate changes in the human genome. And here's one provocative kind of deliberate change that you might want to make on the human genome. This is obviously a cartoon, but do delete from a, embryonic cell or an embryonic STEM cell or a sperm cell, the BRCA one gene mutation that increases your risk for breast cancer. The part of the point that I'm trying to make is, is this idea of a suite of technologies. CRISPR and Cas9 are one element of a kind of array of technologies that are allowing us to circulate closer and closer and closer to making deliberate human, deliberate changes in the genomes of humans and in the genomes of other organisms. And the fact is that these technologies will only enrich themselves and deepen themselves with the other two things that I talked about, the writing, the reading and deep learning techniques that we understand. So the more we read, and again, reminders that it's like a child, the more the child reads, the more the child understands, the more the child learns to write himself or herself. And all of a sudden, if you see this in your own children, they go through a transitional phase. When all of a sudden language, the language that we communicate with becomes fundamentally amenable to their own manipulation. And I really like to press this analogy. It's not as if we're just reading or just writing or just learning. It's the fact that these three things are converging together, allowing us to learn the language of genetics and genomics and becoming more and more facile with using them. A couple of more slides. One is, that again to emphasize these kind of suite of technologies. Here is a cross section of a culture, cell culture in which cells that have been genetically modified are being persuaded using techniques to become other cells. And if you look closely, you might identify what this cell is. This is a sperm cell. So this, these are primordial germline cells, germline STEM cells which are being persuaded to make sperm cells. And it again does not take much to understand that you can make changes in these cells, in the cells that produce sperm and eggs. And ultimately if you make changes in those cells, the genetic changes that you're making will ultimately become changes in sperm, obviously. And once they become changes in sperm, they can therefore be transmitted not across one generation but across multiple generations. So the contrast is with the example that I gave you at the beginning of the talk, which is in my laboratory, we work on blood STEM cells. Blood STEM cells, if you make genetic changes in blood STEM cells, those cells remain in the person's body often for the lifetime of that person. But when that person generates another human being, they don't get transmitted. So they're restricted to that human being’s body. On the other hand, if you were to make changes in these cells, these primordial germ cells, then those changes would become permanently imprinted in every single cell of the person that's born. Because the sperm would give rise to every single cell of the person that's born, bar a few exceptions. But also then become the sperm, become part of the sperm and the eggs of that human being by definition. And therefore the changes that you would make would be permanently imprinted on the human genome. And the final possibility of course, is that we're getting to a place as part of this suite of technologies where we're saying, "God, you know, why make all these changes?" Like what I just said? God. But why make all of these changes in a gene on that already exists. Why don't make the whole thing from scratch? It's a chemical. You can string them together from scratch. Why don't we make larger and larger pieces of DNA? Why don't we get to the place of time we're making full chromosomes and make it all from scratch? And so the questions really have become, can we read the genome before the embryo is implanted, can be introduced, genetic changes before making sperm and eggs? We've talked about that. And the answer is 'yes', we can. And therefore, can we read and write information into the embryo by making important diagnostic decisions in the embryo before. So what we, what have we decided, and this is the landscape that we're inhabiting right now. So as you can imagine the National Academy of Sciences and the National Academy of Medicine met much like this meeting to make some decisions about what to do about all of this. And I found this report extraordinarily important. I would encourage everyone to read it. It's online. You can find it online, but this is of course the bioethical report that came out of all of this. I'm not going to ask you to read all of this but it's important to reflect because we've already taken this journey in this talk that three aspects were recognized. One is that the possibility, and I love this idea, that the possibility of diminishing the dignity of human beings is at risk as we move forward with these technologies. The possibility of restricting respect for our variety is important. And finally, the possibility of a lack of humility we might be engaging in things that have a lack of humility in moving forward. People who are in the field of bioethics know these principles very well. But it was probably one of the first times when I think the consequences for human future have been so stark. The limitations of the technologies we're also recognized and have become recognized, even in the last few weeks. There've been three or four attempts in altering genes in humans. Whether these are safe, whether they only hit one gene, whether they can by mistake make other genetic alterations that we don't know of remain a question. But I think it's fair to say that nothing in principle stops the field from moving forward. I think it's also fair to say that within the next 10 odd years that we will know the answer to whether we can definitively make changes in the human genome or certainly in the genomes of other organisms. I suspect the next 5 years we will be able to quite definitively talk about making genetically modified changes, deliberate changes in virtually most organisms, particularly the ones that are involved in crop and food production for humans. So the National Academy proposed, and I'm going to stop here. I'm going to stop here and this is going to be the basis of I think the next few years of work. The National Academy made 3 recommendations for the possibility of human genetic changes. Many of us, I think, including me, had thought that the Academy would say we should stop. We should put a code, we should go back to a coder moment, a moment of reflection in which we say that we are not allowed to make changes in the human genome or particularly in human embryos. But I was personally surprised that the Academy said in fact it would be permissible to make changes in the human genome, permanent changes in the human genome. And these were the three principles. It's interesting that they, perhaps they're obvious, but in fact they're remarkably can coordinate with the triangle that I proposed in the book, "In the Gene." Number one is that we make changes only if the gene of interest or the gene that we're changing produces what they called extraordinary suffering. In fact, it's the same word that I used a phrase I used in the book. That in fact extraordinary suffering is a criteria. Number two that there's confidence that the gene in fact causes the disease, the causal confidence between the link between the gene and the disease. And that no other justifiable alternative exists and there's no coercion. So the thought that I will leave you with as we enter the Q and A session is the thought which occurs, must've occurred to all of you by now, which is who defines extraordinary suffering? What is the limits of that definition? Is extreme shortness extraordinary suffering in societies where there's a primacy placed on height? What is the confidence line that we draw between the link between a gene and a disease? So let's take the example of BRCA1. I mean, I was giving a talk like this and a woman came up to me after the talk, she had a family history of BRCA1 of breast cancer and ovarian cancer, and she said she would love to eliminate the BRCA1 gene from her lineage. Seems like there's extraordinary suffering if you have breast or ovarian cancer, but perhaps there's also extraordinary suffering if you don't, and you live under the shadow of having breast and ovarian cancer in your future. Should we, where do we draw that line? What if we knew that there was only a 10 or 20% chance of having such a disease? I've often asked this question, you don't have to answer it. I often ask the question to this audience. What if I were to tell you that you're, you've sequenced the genome of your unborn child and that child has a 5% risk of developing a lethal neurodegenerative disease. How many of you would choose not to have that child? And what if I switched that number from 5% to 15%, would it make a change? And obviously individuals draw that number in different ways. And finally, this idea of no justifiable alternative and no coercion sounds quite reasonable, sounds like a reasonable idea. But to remind ourselves that we live at the end of a history where the nature of coercion and the nature of what's justifiable are so linked in with our aspirations that what is coercive and what's not coercive often becomes fuzzy when it becomes real in the world. The culture around eugenics in the 1920s and 1930s was a course of culture. It's the culture that took people who were otherwise supposedly liberal minded, interested in the rights of individuals and interested in understanding the understanding and sympathizing empathizing with the plights of individuals. And yet the culture was coercive enough that a state mandate was passed to move them into penitentiaries. So someone asked me in it at the beginning of the talk, 'Are we ready to edit? What's your answer to the question? Are we ready to edit our genomes?" The answer is technologically we are getting closer and closer to doing that. We have devised criteria to try to restrict our capacity to move forward. Whether you agree or disagree with that criteria is a question that you need to take up individually and just to remind ourselves that this cannot be a conversation that other people have. This is your genome. By definition, it is central to you. It is one of the most, it is one of the things that is most central to who you are. It's stewardship is a part of what we have to have stewardship over. And the fact that this is a conversation that will resonate very widely across the next 10 years I think is very evident to me and hopefully evident to everyone. So it's a moment I think to pause and ask the question, "Should there be deeper criteria? Should we stop? Should there be more than this that we devise in moving forward as technologically we get more and more equipped to change our own genetic information." Thank you.
S. Matthew Liao: Yeah, so I thought Dr. Mukherjee, his basic point in the lecture was that there's a lot of new developments in terms of how we can read DNA codes and but also how we can write to them now. And also our understanding of them is becoming more sophisticated and that together, the three elements coming together, there'll be kind of like language acquisition. We're going to get better and better at it and then we're going to be able to do more and more sophisticated things. And I think he's basically right on that. And so then my own view is I'm an ethicist, so my perspective is, well how do we think about the ethics of being able to do this? Cause this is leading us to designer babies, right? And so should we be engaging in these type of activities? Already I think in September there was, in the UK they did a gene editing of an embryo and they for the first time, so usually they try to correct diseases in an embryo. For example, a sort of like Tay-Sachs or cystic fibrosis. But this particular study they knocked out a gene called OCT4. It's a very important gene for pluripotency for development of the embryo. It sort of develops the placenta and a bunch of other things and they knocked it out and surely enough the embryos did not develop. And that's very significant because now they can do basic science using CRISPR with using gene editing technologies and they're going to be able to learn a lot about sort of why are there spontaneous abortions, et cetera, et cetera. So we're going to get a lot of knowledge in the way that Dr. Mukherjee has suggested. So in terms of the ethics, I think now is the time to start thinking about having some sort of ethical framework for thinking about whether we should be doing these things. And in my own work I've suggested that we at least should have some sort of lower bounds. I worked on, I've done some work on human rights where I basically argue that human beings have human rights to certain fundamental conditions. Those are sort of, and that includes some fundamental capacities, capacities such as being able to see, being able to walk, being able to think and so on. And so the suggestion here is very simple. It seems that whatever we do, whatever kind of designer babies we engage in, we shouldn't deliberately create embryos or offspring that are not going to have all the fundamental capacities. So for example, we shouldn't deliberately create someone without arms, right? Or someone who cannot think and more controversially, I also think that we shouldn't create offspring who cannot hear or who are going to be blind, right? But there are some controversies about that. And so that raises issues about what we can and cannot do. So they are also correspondingly, I think that if we can correct, if we get go can go into an embryo and suppose we find a defect in one of the fundamental capacities, say it lacks the gene to be able to see or to be able to hear and we can use CRISPR to edit that and to sort of allow the embryo to recover the capacity, I think we should do it. Or to acquire the capacity, I think we should do it. So those are some of the things that we can begin to, that's, I call this the human rights approach. And this is a way by which we can sort of think about when it's permissible, when it’s not permissible to engage in gene editing. And this view actually differs from a lot of different views in the literature. So some people think that, some people take a more libertarian view, so they think we should be able to do anything we want, right? But on that view, it would be permissible to create embryos without arms, offspring without arms and legs or sight, et cetera, et cetera. There are other views that are more what's called a perfectionist views, right? So they say that we should have the embryo with the best chance of having the best life possible. I think that view proves too much, right? So on that view, so let's imagine a case where somebody, you can have the gene for being able to drink very fine wine, right? And then there's, you can create an offspring that's going to be able to taste very fine wine and another offspring that has the super taste, there'll be able to taste even better wine, right? So that view implies that we have an obligation to create offspring that we'll be able to taste these very, very fine wine. But it doesn't seem like we have an obligation to do that, right? And so the human rights view shows why we don't have those obligations because it's focused on the fundamental capacities, the things that we, human beings all need in order to pursue a good life. My view is going to be that we need to be pluralist about these technologies. So maybe sometimes we do, maybe we want to be smarter, right? Maybe we want to have acquired additional capacities depending on what we want to do. And so one question is, what are we going to use the gene editing for? Right? Why do we need humans with all these super capacities? And then one answer is that, well, it's going to be the case that, so this is what we know. We know that eventually the sun will die out. In order to make sure that the human species survive, we have to leave this planet at some point, right? And so gene editing might enable us to create offspring that'll be able to survive better in space. So maybe a more radiation resistance, maybe offspring that'll require less water, that can absorb oxygen, you know, be more efficient at absorbing oxygen and so on and so forth. Those are the opportunities that we have now and that we should really look into because they could mean the survival of humanity. So I think there are a lot of opportunities. And one of the things I would encourage our graduates is to think in interdisciplinary terms, right? So technologies like this, like CRISPR, they're very complex and so, and they connect up with all different aspects of life. Not only do you need to know the science of CRISPR, you also need to know the implications. How is it going to affect us, our society, right? And also future generations. So they really need to think about all the different dimensions. And I think public health is particularly equipped to look at all these different issues because it's inherently interdisciplinary. And so they can, it's sort of working in a team, working together to sort of tackle like, so here's a technology, how do we use it to best further public health, to best further humanity? And so one of the things that gene editing, some people are using it for example, to fight off Zika viruses. So they want to edit mosquitoes using gene drives to drive out certain mosquitoes that are responsible for the Zika virus. And there are other uses like this. But again, it requires the students to be able to appreciate the science, to know the opportunities, but also to know the limits, right? So they need to be able to think also about the ethical implications, you know. What unintended consequences? When you create these mosquitoes and you put up out in the wild, what happens if they mutate? See that's a scientific question, right? And so other animals eat them and then they mutate, et cetera. So it's sort of being able to think through the implications of some of these technologies. I think that will be very important and that'll be a very important skill to have for our students in the future.