Scientist Stories: Siddhartha Mukherjee, Genes as the Mastercode

My name is Siddhartha Mukherjee. I'm an MD, and I have a PhD, a DPhil, in immunology. I'm currently an associate professor of Medicine in the Oncology Division at Columbia University.

Who were some important mentors early in your career?

So, I had 2, 3 incredibly close early scientific mentors, and then I had mentors later. The first and the most important was Paul Berg. This was— I was an undergraduate at Stanford University, and in my second year, I joined Paul Berg's lab as an undergraduate. Paul was and is an incredibly influential geneticist, one of the pioneers of modern genetics. His discoveries are legion. But when I had joined Paul's lab, this was in 1993, '94, we were coming off the end of a very important series of discoveries which Paul had played a very crucial role in. And that, of course, was the capacity to create recombinant DNA, to stitch together two pieces of DNA, a piece from a virus, a piece from a bacteria, a piece from a bacteria, a piece from a mammal, into one piece of DNA using bacterial enzymes. And had begun to create the recombinant DNA revolution, allowing us to basically change genetic code by stitching together two separate pieces of DNA which came from different animals or from different parts of the genome. This had led, as you very well know, to a series of important ethical questions. Because for the first time we were managing, we had managed, we had opened up the capacity to make synthetic pieces of DNA, synthetic piece of genetic code, which were unlike anything found in nature. As I said, you could stitch together genetic code from a bacterium and a genetic code from a virus. And the possibilities of being able to manipulate genes was enormous. This led to an enormous public interest, but also public fears around what had happened. And Paul was central in creating the Asilomar Conference. And a conference which addressed Asilomar 2, which addressed concerns, public concerns, and questions about what to do now that this technology was— had become available. So, Paul's lab, when I joined Paul's lab, I was working extensively with recombinant DNA and with new technologies around recombinant DNA, and in my case, putting them into yeast cells and to mammalian cells to figure out how to mix and match pieces of DNA together to make them work in yeast and mammalian cells. So, Paul was a very early mentor. And what's interesting about Paul is not only was he a mentor as a scientist, but he also opened my eyes to the potential ethical and clinical concerns, safety concerns, clinical concerns around genetics and the modification of genes. Concerns which will, of course, echo over time as we go along. So, this is now 1993, '94. My second major early mentor was at Oxford where I went to pursue my PhD, my DPhil, and that was Alan Townsend. Alan had solved an incredibly important problem in immunology. He had helped solve an incredibly important problem in immunology with many colleagues. But the problem was, it's very simply stated, which is if you have a a cell that's infected by a virus, and the virus is sitting within the cell, its DNA or its RNA is inside the cell, how can a T cell possibly recognize the infected cell? Because the contents of the virus, the DNA or the RNA, is sitting inside. How can a T cell recognize something that it can't see, because the cell, the infected cell, is visible to the T cell, but only its outside is visible to the T cell. So, it was a kind of, you know, how do you scan the inside of a cell problem. And Alan had solved it partially by making the discovery that viruses— sorry, that cells have an innate molecular apparatus by which they pick up the contents made by these viral genes. They pick up these contents, broken down pieces of these contents, peptides, and display them like molecular tags on their cell surface. It's a very specialized system to do this, a series of molecular steps that allows essentially a cell to take its innards and display them on its surface as if saying, you know, this is what I contain within me. And if that happens to be from a foreign invader like a virus or a microbe, then the T cell can sense it and thereby kill the infected cell. So, Paul and Alan were very influential to me for very different reasons. Paul, because he introduced me to the ideas around genetic engineering, recombinant DNA cloning, and its potential applications and the ethical concerns. Alan was— worked on a problem that was much more close to a human problem. Why do infections get cleared? How does one obtain immunity? And Alan really pushed me towards thinking more about medicine. Which is where I, of course, mainly have my, the center of my work. There were two people who came later. One was Connie Sepko, and I should say Connie worked very closely with a geneticist, Cliff Tabin. Connie and Cliff were both at Harvard. I went to medical school there. And they were interested in develop— problems of development. In Connie's case, how does the nervous system develop, how does the retina develop in particular. And she was a cell biologist. And so, from Connie, I learned how to do a technique that I still use today called lineage tracing. But the questions there were, you know, how does a single cell become you? So, it was, again, you think about this as a kaleidoscope. Paul thinking about recombinant DNA and DNA, Alan thinking about some of the clinical problems of DNA, of human diseases, Connie and Cliff thinking about how DNA becomes actualized. So, you have a long genome, but it's inert without being actualized into making diverse kinds of cells.. And those diverse kinds of cells have to make a retina, they have to make a brain, they have to make your fingers. How does all of this unfold in space and time? And so, you know, Connie came— Connie, I was in Connie's lab and again, took on projects that had to do with solving this kind of, you know, how does the genome become a person, an animal, an organ? And finally, I should mention David Scadden. I started working with David. David was a hematologist, and he was, of all these people, the closest to the clinic. He was creating trials. David was involved in working very closely with stem cells, blood-forming stem cells, hematopoietic stem cells. And that is what— and he had made some fundamental discoveries about how hematopoietic stem cells live inside bone marrow. And so, he brought me to problems which were now directly adjacent to patients. You know, can we do gene therapy with blood stem cells? How could we? What were the ethical and safety concerns with those kinds of therapies? How does one actually bring this to human practice, which is where, you know, the area that I sit in. Today.

When did you first hear of the Human Genome Project? What were your initial thoughts?

I heard very much about the Human Genome Project because it was very richly discussed in Paul's lab. Paul had been very closely involved along with many other people, obviously Jim Watson, Wally Gilbert, John Sulston, virtually, you know, Francis Collins, virtually this idea had started coalescing of a kind of consortium, as it were, of trying to not just sequence single genes or 2 genes or 5 genes, but the entire human genome. So, around '93 is when I first even heard the word Human Genome Project. I was then unaware of the private effort and the public effort and how these would intersect and coalesce and compete. So, I was very unaware. I would say by about the late '90s was when it was very clear what the project looked like or was looking like. There was widespread newspaper coverage by then. And by the time I was in Alan's laboratory, it was very much something that at Oxford, very much something that we spoke about. You know, just to give you a sense of how important it had already become to us, these— this T-cell problem has to do with specialized molecules called, in humans, labeled with the word HLA, human leukocyte antigen, or MHC, major histocompatibility complexes. Complicated words, but really is a set of genes carried in cells which allows these peptides from viruses to be presented on the surface of cells. This is an incredibly important part of the genome because it determines whether your tissue can be transplanted into mine, whether your bone marrow can be transplanted into mine. But it was also a very tough part of the genome to sequence. For various structural reasons. There's lots of repeats, there's lots of elements in that part of the genome. So, its importance to us as a solution to be able to scan a genome and to figure out where these genes lived, where they came from in evolution, what they did, what functions they had, etc., and how those functions to what extent was encoded or determined by where they were located in the genome. All of these questions were very, very relevant to the work we're doing, we were doing, and continue to do. As I thought more and more about bone marrow transplantation, a field that I would eventually really become the center of what I worked on, these, the genes that allow us to do transplantation across one, from one human to another for conditions such as leukemia and other diseases became of crucial importance. So, the Human Genome Project was not an abstraction anymore by the mid-'90s. It was a reality. It was a necessity. We needed that information to try to solve some of the thorniest problems that we were trying to solve. To solve in Alan's lab. Tell us how you came to write "The Emperor of All Maladies." Yeah, so, I came to it, I would say, a little bit backwards. So, I'll tell you the story of the journey that unfolded. You know, when I was at Stanford, I was in Paul's lab, and one could say that there were these echoes of what it meant, what the public discourse around science meant. And it was really inspired by Paul talking extensively about Asilomar and what it had done to him as a scientist, as a thinker, as an intellectual. We would have lab meetings. Sometimes we would have lab retreats. Sometimes we would— so, in fact, the first lab retreat that I went to was at Asilomar. It turned out that, you you know, the lab still used Stanford, still used Asilomar as its site. So, I had this strange and wonderful opportunity when I went in 1994, I think, or 1995. I had a poster as an undergraduate in the Asilomar retreat. But then, Marian Dieckmann, who had been Paul's assistant, research assistant, and really a phenomenal researcher in her own right, And Paul, you know, we— walked me through Asilomar saying, oh, you know, this is the building where the worst argument broke out around between the journalists and the scientists. Here's where we signed, you know, this document. This was where, you know, we stayed up all night eating Chinese food, you know, trying to write what would become one of the central documents of Asilomar. There was, you know, the graduate students were there. Here was, you know, Jim Watson. This is where Jim Watson stood up and said, you know, let's do this, et cetera. So, it was a kind of visual tour, as it were, of history. And echoing in the back of my mind was, you know, this story has got to be told somehow. This story is so important, so moving, so crazy, so powerful. And has such public impact that it has to be told. But I wasn't in a position to tell it or write it. So, I then went on to Oxford, but this kind of echoed public discourse around science, echoed, kept echoing in the back of my head. At Oxford, I was a nose-to-the-ground PhD student, you know, solving problems, trying to write my thesis. But at the end of when I finished my thesis, again, this this kind of old idea that this story needs to be told, these stories need to be told more publicly, came back to me. So when I came to medical school, this was in 1997, I first began to start writing. And I started small. I wrote for The New Republic, The New York Times, usually smaller things. I was learning medicine, learning clinical medicine, and most of these first ideas were around sort of the life, the life of a medical student, the life of medicine. I remember writing a piece about resident hours. I remember reviewing books that had to do with medicine for the Times and for the Times Magazine. But the idea of writing a 600-page book hadn't really occurred to me. I then decided to become an oncologist and started undergoing training as an oncologist. And that's where I sort of began to think that you know, we've thought about this public discourse. I've been thinking about how little and how little some things are known and how important they are. And so, I began to think there should be— the story should be told. And that was the inception as I was— went through my fellowship in oncology, that was the inception of The Emperor of All Maladies, the first book on cancer. And that book really started out as a journal. I was keeping a journal. I would write every day at the end of rounds, at the end of my day, and slowly began to grow and grew out of a patient's question. The patient asked me, you know, I was giving a woman a new trial for chemotherapy and she said, "Where am I going and what are we doing?" And it was sort of a kind of a cry from the heart, you know, desperation, but also a philosophical question. You know, cancer is, was, and is one of the most talked about words in the English language. You can't open a newspaper without encountering the word cancer somewhere. And yet, its history was unknown. Its story was unknown. There was no biography, as it were, of cancer. And that was the inception of the Emperor of Maladies.

For you, what makes cancer such a fascinating subject?

Well, I think it's not only for me. I think very broadly speaking, the public perceives cancer as an exceptional disease. It is partly because we have pitted the best of our minds, the best minds of our generation against it, and we have had sobering success against cancer. It was, for a while, it was the second most feared word in the English language after shark. And I don't work on sharks. So, the public fear of the disease, the fact that we had pitted our— the most brilliant minds of our generation against it, the fact that it had yielded important insights and certainly treatments and cures, but, and yet, it sort of was unreachable, had remained outside conquest. It was, you know, is the essential stories like, you know, the man battling the bear story. It is the man, it is humans. Primal. There is a very primal quality about the, the story. It is the, you know, humans battling the unknown in a cave. So, I think all of these elements make it one of the most fascinating stories of the 20th and 21st century.

Why do you think our progress combating cancer has been so sobering?

Well, there are many insights. About why the progress has been sobering. First, let's identify the fact that there has been progress. There's no doubt about it. Cancer mortality has fallen about 1% every year for the last 20-odd years. That translates into a 20-25% decrease in cancer-related mortality. First of all, there's— it's not one disease, but many diseases is almost a cliché. But the idea that, you know, you don't have one magic bullet solution to cancer has become very clear. So, some cancers have become extraordinarily tractable to prevention and treatment, prevention, early detection, and treatment. Breast cancer is a good example where treatment has made an enormous difference. There is certainly early detection with modest success. And— but not a lot of, you know, there's some prevention, especially these days now that we know people with genetic susceptibilities for breast cancer, we've been able to do— launch successful prevention campaigns, particularly focused on people with genes such as BRCA1 and BRCA2. On the other hand, pancreatic cancer, particularly in its advanced form, has remained doggedly resistant to treatment. So, the diversity of cancers is one reason that has been— it's been very sobering. The word magic bullet has never turned out to be either a bullet or magic. And so, that's been a very sobering realization. And second, I think it was very clear early on that the campaign against cancer would be built like a pyramid. The bottom of the pyramid would be prevention, then there would be early detection, and the top of the pyramid would be treatment. In the last decade or so, we made at first an enormous success in prevention, basically by eliminating cigarette smoking as a major carcinogen. And then, you know, finding things— finding viruses that cause cancer such as the human papillomavirus and creating vaccines against them. So, these were very successful efforts. But then in the last 10-odd years, I think to some extent it's been sobering to realize that the number of chemical carcinogens with large human impact that we've been able to remove, as much like cigarette smoking was— we were able to diminish, we haven't found so many. And so, the we have had to really rethink to some extent how to prevent cancer in a way that, you know, had seemed maybe 20, 30, 40 years ago perhaps simpler. 20, 30, 40 years ago, one might have made the argument that there were a dozen or so common carcinogens, and those— and by systematically removing those dozen or so common carcinogens, cigarette smoking, being one of them, viruses being some of them, that we would make enormous strides against cancer. And that there would be biomarkers, markers of risk that we could find in people and we could then use whatever tools that we had to diminish those risks. Both of those have not turned out— well, have turned out to be the case, but they have turned out to be much more sobering than we had imagined. In contrast, just to give you a contrast, in coronary disease, in heart disease, both turned out to be true. So, it turned out that, number one, that it was for the most part one disease. I should say all three turned out to be true. For the most part, it turned out to be one disease, essentially one disease, although people could have arguments about what the pathophysiology of the the disease, essentially one disease. Number 2 is that prevention, primary prevention became possible by behavioral modification, dietary modification, and potentially the use of a new series of drugs to lower cholesterol. So, primary prevention became possible. And finally, the— it became very clear that there were clear biomarkers. So, having a, you know, having a high level of a particular form of cholesterol was an indicated risk, and therefore you could be placed on medicines such as aspirin and, for instance, Lipitor, to reduce the risk. Hypertension was a risk. Smoking was a risk. And there were other risks that were familiar— risks. Family histories were risks. Age is a risk, and so forth. So, you had all sorts of markers by which you could identify patients before they had the frank disease, and if— and you could intervene and change the biology of people before they had the frank disease. So, just to summarize, all 3 things, the fact that it was largely one disease, the fact that it was largely— that there were behavioral and other mechanisms of preventing it, and number 3, that you could identify the highest risk patients in for heart disease made that an enormous clinical success, so much so that the mortality from heart disease, as you know, has declined up to 70% if you have the appropriate access. In contrast, as I said, the mortality from cancer has drip, drip, drip, drip, dripped slowly, frustratingly, because it is just the opposite. It is in many diseases there was, aside from some obvious candidates, It was not easy to find mechanisms of primary prevention. And number 3, biomarkers of risk, aside from some genetics and some other exposures, the biomarkers of risk were few and far between. So, exactly what happened in heart disease in some ways did not happen in cancer because they're two fundamentally different illnesses. And that's been some of the frustration. To some extent, the challenge of working in this field.

How have efforts like the Cancer Genome Atlas, TCGA, changed the fight against cancer?

So, in many ways. So, again, one could— we could divide this up, this question up systematically. First of all, one should say that the TCGA was built on top of the Human Genome Project. So, the Human Genome Project provided the template to build the Cancer Genome Project, allowing us to figure out what are the genetic changes in cancer cells. I should preface this by saying that this must be known to your audience, but at its base, cancer is a genetic disease. So, and by genetic disease, we mean that cancer occurs or is caused by mutations in genes that alter cellular growth, cellular metabolism, and allow cells to grow when they're not supposed to grow, and occasionally don't allow cells to die when they're supposed to die. These genes further co-opt metabolism, they co-opt growth pathways such that the cancer cell grows without control. It has no longer the control that normal cells have. When you have a cut in your hand, a wound,. It heals, but you don't grow a new hand. That's because the cells in your hand have a capacity to grow to fill the wound and repair the wound, but then have also a capacity to stop growing so that you still don't have a new hand growing out of it. In cancer, these pathways, because of mutations in genes, these pathways are dysregulated so that now you have a cell that keeps growing, or cells that keep growing. But that brings us to the question, or brought us to the question, what are these mutations? What are the genes? How many are there? Is it 1 in a cancer cell? Are there 10? Are there 20? What is their diversity within a cancer? 2 breast cancers, do they have the same set of genes or do they have different sets of genes? What is the diversity across cancers? If you compare breast cancer to pancreatic cancer, what is the diversity across cancer? So, the first thing that the human— that the Cancer Genome Project provided, building on the Human Genome Project, was answers to these questions. How many genes are mutated in cancers? What are the genes? What is the identity of the genes? What is the relationship between, let's say, two breast cancers? What is the relationship between breast cancer and pancreatic cancer or lung cancer? What is the compendium, what is the encyclopedia of mutated genes that allow cancers to behave like cancers? That was a major and very important series of of discoveries. The second was to ask the question, now that you know these mutated genes, can you use them to understand how to derive new therapies? And that has been an ongoing project. That might— those might include molecules like Gleevec or imatinib, which take a particular product of a cancer gene called BCR-ABL, the protein BCR-ABL, and basically jams the function of that protein, enabling it to not function anymore. Now, BCR-ABL was discovered long before the Human Genome Project, but the Human Genome Project put mutations such as BCR-ABL and others in the context of the entire human genome. So, we began to imagine new forms of therapy based on our understanding of the complexity of cancers, cancer genomes. And the third was cancer prevention. So, and we are, this is also an ongoing effort. Again, based on early efforts to map genes, which eventually would coalesce into the Cancer Genome Project, we began to find genes that increase susceptibilities for cancer. These are, in fact, biomarkers, the few and far between that we have. The gene BRCA1 is one of them, the gene BRCA2 is one of them. The Cancer Genome Project, again, these were found before the full compendium was reached, but it put these genes in the context of the human genome and put the— gave us a sense of really, as I said, what the complexity and what the compendium was. The next step, the phase that we're in, is to take all those, all that material, these insights, and translate them into prevention, detection, and cure. And that's the process that we're undergoing right now. How do we interpret the alterations in the cancer genes, the mutations? You know, a breast cancer could have millions of cells, but, you know, could have 130 mutations, whereas other cancers maybe have only 1 or 2 mutations. How can 2 mutations drive the physiology of a cancer cell? What happens when you have 100 mutations? How do— so, So, really, the Cancer Genome Project, building on the Human Genome Project, closed a certain set of questions. How many genes are there? What are their— what is their identity? What do the mutations look like? But opened a new set of questions, which is how do these genes make cancer's abnormal behavior happen? What is the physiology of cancer? And now we've gone on to yet another set of questions. Now that we've understood the physiology of cancer, we're asking how is the physiology of cancer related to the physiology of the host. So, immunotherapy is that third set of questions. So, you really think about all of these as sequential steps, eventually always asking the same question, which is how do we take the genome and use that to understand better ways of preventing cancer, better ways of detecting cancer early. For instance, you know, you might find a way to detect cancer by genes that are shed in the blood. It's a very common theme these days. And thereby detect cancer in its early phases. And finally, treatment. You know, how can we understand the— how the genes make cancer behave abnormally? How can we understand how the cancer interacts with the host? Cells, again, the host genome and the cancer genome juxtaposed against each other or interacting with each other through cells, and how can we thereby drive new kinds of treatment.

Do you think of the HGP and TCGA as question-generating enterprises? How are they distinct from other types of scientific inquiry?

Oh, I absolutely think of the Human Genome Project and the Cancer Genome Atlas as a question-generating enterprise. You know, and I think we would not have embarked on it if it was not a question-generating enterprise. In this sense, it's very different, fundamentally different from things that it's been compared to. So, it's been often compared to the Manhattan Project. It's been compared to the Apollo projects. But here is, and I've written about this extensively, it's a fundamentally different— the Manhattan Project was not a question-generating enterprise. It was an answer-generating enterprise. Enterprise. Can you split an atom and cause a bomb, detonate a bomb? That is not a question, it is an answer. It raises certain questions, but it doesn't raise fundamental questions. It answers questions. It is an engineering project. The moon landing also is not a question-generating enterprise, it is an answer-generating enterprise. It is driven by when you land on the moon, you've landed on the moon, you've done something concrete. It shows that, you know, if you ask the question, can human beings land on the moon, you can say, yes, they can. We did. And, of course, it generates a small number of questions. The Human Genome Project and the Cancer Genome Atlas should not be viewed really. Because if they're viewed as the Manhattan Project and the Apollo Project, then the public says, well, what do we get out of it? What are the answers? But really should be seen as template projects. They should be seen as question-generating projects. And what we are seeing out of it is the answers to those questions. The answers to those questions are, can we prevent cancer? The question is, you know, what does the Human Genome Project tell us? Or the Cancer Genome Project tell us about how to prevent or treat a cancer? So, I've tried very hard in this kind of long discourse to make a very simple distinction between question-generating projects and answer-generating projects, hypothesis-generating projects and hypothesis-proving projects. The Human Genome Project and the Cancer Genome Atlas lie very much in the hypothesis-generating arm or the hypothesis-generating arena of the scientific enterprise. The Manhattan Project and the Apollo Project lie very much closer to the hypothesis-closing aspects of the enterprise. Of course, that's not absolutely true, but for the most part. And I think it's very important for the public discourse around science to be able to distinguish these two things.? Because if they cannot, then you mix up these two ideas, and then you ask the wrong kinds of questions. You ask the question, what did I get out of it, right? Well, you know, out of the Manhattan Project or out of the, you know, you got a bomb that some people might say brought the war to an end. So, what do I get out of it? That's a very concrete answer. What do I get out of the Human Genome Project? I would say you get a whole bunch of interesting questions.. And if you didn't realize that that was the purpose of the Human Genome Project, or that was one of the purposes of the Human Genome Project, you as a taxpayer, as the public, as a journalist, as a writer, as a thinker, might feel disappointed. You could say, "What did I get out of it?" Well, I got a whole bunch of questions out of it. And you would say, "Well, did I pay all this money to get this effort, put these brilliant minds to work to get questions?" And you— and some people might say that's That's really disappointing. But to me as a scientist, it's fundamentally important to explain that these are two different ideas. They have different aims. And once you know, if you know that they have different aims, all of a sudden you say, that's very interesting. I got a whole series of questions out of it. And those questions, when we answer those questions, as we answer those questions, we will find new treatments for not just diseases like cancer, but of course illnesses in general, and understand the human body in a fundamentally different way. Given that, how do you contextualize personalized medicine? So, that has been, again, that has— the question was posed that now that we have the capacity to sequence human genomes, how much do we have now the capacity to personalize medicine for an individual disease or an individual cancer? The answer turned out to be Somewhat, but not very much for certain diseases. And the answers are still emerging. So, what have we learned? Well, we've learned some very important lessons. One, we've learned that most common human diseases, hypertension, heart disease, breast cancer, and others, are— the risks for them are not contained in a single gene. But really in tens of genes, thousands of genes, even tens of thousands of genes for most common human diseases. There are certainly rarer human diseases where a single mutation in a gene will cause the disease or greatly increase your chance of having a disease. Cystic fibrosis is one of them. There are many others. Sickle cell anemia is one of them. Huntington's disease is yet another one. But for common human diseases, some of which I named, we've discovered that in fact the— for most humans, the genetics or the genomics is quite complicated. And therefore, delivering precision medicine isn't going to be a first-order problem. It isn't going to be a kind of match, as it were, a jigsaw puzzle. In which the genome will match a medicine and you'll put these two things together and all of a sudden, boom, you'll get the treatment for a particular disease. It has turned out to be more complex. So, the question that it asked was, is it possible to use genetic information to go from genes to medicine in a simple manner? And the answer is for genes where the genetics is simple, the prediction of risk is simpler, the capacity to diagnose is simpler, the potential to intervene has— we've now begun to see attempts to intervene that have been successful. But for most human diseases, it has turned out that the genetics or the genomics is complex. The match between the genes and the medicines is not one-to-one.. And we are still trying to figure out the extent to which we can match these complex genetic susceptibilities or vulnerabilities to the appropriate medicine. And that has been—

that's the answer to the question. How do you define the public and effectively communicate scientific inquiry to that public?

Well, I mean, I have some fundamental— first of all, I think you never underestimate a the public. These are people who have, you know, it is their generosity to some extent, our generosity, their generosity that has sustained this institute, that has sustained the, you know, decades-long project to investigate human diseases and to emancipate us from illness and suffering. So, and the public, as you know, however you define it, however broadly you define it, has fundamentally always supported the mission of the National Institutes of Health, and that there should be a national health program. So, research program. So, what does the public mean? Well, the public is very broad. It has very broad sense, you know, you range from people with deep interest in health literacy to people who have no interest in health literacy. Access is complicated. Educational levels are complicated. Socioeconomic levels are very wide. There are racial disparities. In health. So, the public is not one public. There are many publics. How does one speak to that public? Well, first of all, as I said, never underestimate. And I think the fundamental rule is that one should be able to communicate the central goals of a project, any project, scientific project, in a way that's understandable to a, what I sort of, I would say, to a 5th grader. Not because the public has a 5th grader's intellect, but your capacity as a communicator of science should be so simple that someone who is generally able to speak and understand language in, with some degree of complexity, should be able to understand why you're doing the project. This was, epitomized by my, you know, by Gary Gilliland, another one of my mentors, a leukemia biologist, who basically said that if you're a graduate student and you cannot explain your project to your grandfather or your grandmother, and in two sentences or less, and have them figure out why it's important, then you're doing something wrong. It's not that you're communicating wrong, there's something wrong with your project. It is too complicated. It's not worth it. It's not worthwhile. It's maybe you require simplification. Your brain requires some simplification. It's not that they, it's not that the recipients of the information have the problem. It's actually the giver of the information has the problem. And it's not a communication problem per se. It's not as if you, it is the problem that things are muddled in your head. And if they're muddled in your head, they'll be muddled as you go through what you're doing in the lab. So, and so, you know, this, you know, this could be, this basically, it turns out to be a mechanism by which, you know, your father, your mother, your fifth grader, your family member around Thanksgiving table, should be able to understand what you're doing and why you're doing it in 2 sentences or 3 sentences. They may not get the micro details, they might not get the technical details, but they should be able to understand it. And that's the kind of public you want to be able to reach. Never underestimate, never patronize, never soften, but you should be able to describe the essential goals of what you're trying to do in a way that's intelligible to someone who understands the basic instruments of language.

What role do you think scientists should fill within the public discourse?

Well, they have many roles. I mean, you know, they are to some extent the experts, so they have roles to ensure that the knowledge is, that the correct knowledge is is dispensed and dispersed. That false knowledge, which is false news, which abounds today is not dispersed and is contained and is refuted. And it can only be refuted, as I said, without using technicalities. Technicality is actually the friend of false knowledge. Technical speak is the friend of false knowledge or false news because you can hide behind words. You can hide behind technical words even more than anything else. So, I think that, I think, you know, we have a responsibility. As I said, as we move along in various things that we're doing within the realm of research and health, one of the responsibilities is to be able to communicate that effectively, to refute false news. And sometimes the false news might be a disagreement with another scientist. To be able to refute that and do it in a manner that's civil and based on evidence. Science communications around science, like science itself, has to do with evidence. Bring evidence and create a civil argument around evidence. I mean, this seems to have been lost in this last 5 years or so, but that is the basis. And it's been going on in the discourse in science for centuries.

How did you decide to write The Gene: An Intimate History?

Well, at the end of the book on cancer, the question that's raised at the end of the book on cancer is cancer cells are abnormal. How does normalcy happen? What is normalcy? And by normalcy, I don't mean a kind of, you know, I don't mean a kind of moral definition of normalcy. How do cells develop? How do we not have cancer? What makes our bodies develop? I mean, to some extent I was going through my scientific mentors. The book on cancer really came out of my fellowship. The book on genetics and genomics went backwards to people like Paul and to people like Connie and Sepko and Cliff Tabin, who were asking fundamental questions, not about cancer, but about, you know, how does— what do genes look like? How can we manipulate them? How do genes make organisms develop, come out, become real? So, it was a kind of going backwards to some extent. And it was also motivated by a very personal story, the history of mental illness in my family where there was and still is a suspected genetic link, schizophrenia and bipolar disease that has crossed several generations in my family where there is a genetic link. There's certainly environmental links, but there is a genetic link. So, that's why it's called an intimate history. And so, that's how the second book came to be realized. Why is genetic information an important topic for you? Well, it's the idea of how essential it is and how private it is and how personal it is. It is the most intimate information we have. It belongs to us in a way that no other, really, information belongs to us because it's writ into ourselves. It is writ into our, you know, we use the word written written to DNA in popular— as a popular phrase, but in this case it is literally written to DNA. I think there is a, you know, a kind of essentializing quality about it. We find it's essential to you. It is private to you. It belongs to you. Your exposures, your environment, your experiences somehow are choices that you make. But the genome is what you're handed, what you're given. These ideas go back in history to people like Rousseau and other philosophers, you know, what is the tabula rasa, what is the table that you begin with, what is the blank slate that you begin with. And the genome, your genome, is the slate that you begin with. You begin with it before you have your first experience, before you make your first choice, before you do your first activity, it is inside you. And therefore the level of privacy, the level of personhood, the level of intimacy that you have associated with it metaphorically, culturally, biologically, socially, remains exceptional. And I think there are people who've argued against this very successfully saying, you know, your genome doesn't determine you. And most people who are thoughtful about genetics would say yes, there is a component in which genes determine who you are, but there are large interplays between genes and your environment. There's a large interplay between genes and chance. You know, if you have an accident, and break a leg, that is not a genomic event, but it will— it has enormous consequences for— may have enormous consequences for your life and your fate and your choices and your decision. So, to think of human fate and human destiny entirely in gene and genetic and genomic terms is clearly false, but also clearly a fallacy. But on the other hand, to not think of human destiny and choice and future in terms of genomes is also false. And so, part of the education around the idea of what genes do, how they work, how they make us, how they don't make us. Is to draw that line and to understand that it's genes plus environment plus chance plus triggers that ultimately results in human phenotype, human fate, human destiny, human choices, et cetera.

How does the HGP figure into the debates over genetic determinism?

Well, it was not the Human Genome— I think— I don't think it was the Human Genome Project's task to address either the intimacy or the— or to resolve genetic determinism. I think the Human Genome Project's task was to sequence the human genome. As we sequenced the human genome, we found out what we somewhat knew, but really found out in real terms, that human beings are very similar to each other genetically, but we also contain differences from each other. And these differences are responsible for differences between you and me and have consequences for our bodies, have consequences for illness, and have consequences for our destiny. As I said, they're not the only things that have consequences— environment, chance, fate, and so forth. So, I don't think it was the Human Genome Project's project to solve or unsolve genetic determinism. Genetic determinism existed before the Human Genome Project and will exist after the Human Genome Project. It is a part of how humans think of genes. The Human Genome Project illustrated that we are very similar to each other and that there, in some ways, we have critical differences between each other. And it basically allowed us to deepen the discourse about these similarities and differences and raise the question. As I said, it's a question-raising enterprise. Raise the question, you know, to what extent are we determined by genes? And the answer has, had been clear for a long time. The answer had been that, you know, it is a flawed and foolish thing to say that human destiny or fate or human beings are entirely genetically determined or, you know, strong genetic determinism is a flawed idea. Weak genetic determinism is a flawed idea.

Somewhere in between lies the truth. How should scientists approach ethical concerns over determinism arising from social genomics?

Well, the first word is cautiously. And thoughtfully. And the second idea is, again, to try to hold one's sense of balance against this powerful new wave of genetic determinism. As we associate SNPs and other things with complex, socially complex ideas, educational achievement, sexual orientation. We are witnessing a kind of revivalist phase of genetic determinism. If you have these genes, you will be that. So, the first thing that needs to be addressed is that that is generally not true. As I said, there are exceptions to this case. If you have a mutation in the Huntington's disease gene, you will likely experience that disease in your lifetime. That is a powerful form of, that is a powerful genetic determinant. In contrast, most complex human traits have genetic determinants behind them, but are highly modified, refracted as it were, through the lenses of environment, triggers, chance, and choice. Such that when we communicate these ideas to the public, We have to stand against this revivalist buffeting wind of the new genetic determinism. And the only way to do that is to do that by tackling the problem head on. Is to say that there is going on right now, and in fact the book The Gene ends here. It makes very clear there is this this wind that is this new buffet that is trying to knock off really a sober knowledge about how complex human traits and destinies and futures are attained, which is an intersectional interaction between genes and environment. This was known, you know, since Dobzhansky's days. But it is important, I think, now especially, to stand steadfast against that buffeted wind and to reinforce this idea that no one is saying that your genes are you and you are your genes and there's a one-to-one map between one and the other. In rare circumstances, there is, one could produce such a map, but for most complex human traits, it's not the case. [Speaker:JOSHUA_ELKINGTON] How can science communicators help us avoid deterministic frameworks? Well, it's to emphasize what we've learned over, you know, the last several decades. That, as I said before, most complex human traits, some exceptions aside, are determined by multiplicities of genes and multiplicities of genes interacting with environments, chances, triggers, choices, exposures, et cetera. So, in other words, you can't create an epidemiology of human destiny just by reading the human genome. In rare cases, you can create some sense of what a person's future would look like. Because those genes, the mutations in those genes, are so autonomous, so powerful, that they will swerve your fate. But those are rare cases. The more—

most common cases are not that. What most excites you about the future of genomic medicine?

Well, so, we are working on several aspects of this. I'll tell you the one thing that we are working on that is very interesting to me. I treat leukemia, and it's a disease of— it's a cancer of blood cells. And really, for the kind of leukemia that I treat, acute myelogenous leukemia, AML, we know a lot about the genome. We know a lot about the genetic mutations. But unfortunately, we have not been able really thus far to be able to target any of these genetic mutations with drugs or match the mutation with the appropriate kind of drug. There's been a couple of exceptions to that. There's a mutation called FLICK3, which there is a drug for, but again, the effectiveness has been pretty limited, and we're using for this disease drugs that were made in the 1970s. One of the approaches that we've taken, which is a completely novel approach, is to say, Well, what if we, instead of changing— so, the— an amazing capacity in this disease, in AML, would be to be able to distinguish between the cancer cells and the normal cells and kill just the cancer cell and spare the normal cell. But whenever we've tried to do that, we've never found things on the cancer cell, tags, markers, that were not present in the normal cell. So, we've been stuck in this disease because every time we try to kill the cancer cell, we kill the normal cells as well. And obviously, you can't do that because you will very soon reach a threshold and you can't cross that barrier. And we've struggled and struggled and struggled against this, and then all of a sudden we had this idea that what if, rather than concentrating on the cancer, we could use genetic technologies to change the normal cells, thereby making the cancer unique. In other words, we would bioengineer a tag in a cancer cell, which would be eliminated using genetic techniques such as CRISPR from normal cells, thereby creating a cancer-specific or cancer-unique target. We developed this in the laboratory and we are launching a clinical trial. This would not happen if we didn't have tools like CRISPR at our disposal. So, gene therapy and CRISPR and various other things at our disposal. So, I'm excited about the use of these tools to enable novel forms of medicine which could not be imagined before these new genetic and genomic tools were available. That's one. The second idea that I'm very interested in is preimplantation genetic diagnosis. The idea that you can biopsy potentially an embryo and make decisions. First of all, figure out whether there are— whether it carries mutations that are that cause serious genetic diseases or serious diseases and choose or not choose to implant those embryos. It's a very ethically complicated arena, as you can imagine. Lots of debate around it and a lot of questions that have been raised. You know, what should parents be allowed to choose for? How much information should parents be given or can anyone be given? To make that choice. So, I'm very interested in that technology and all its ethical— the debates that follow. And the third, of course, is using tools such as CRISPR and others, synthetic DNA, et cetera, to make alterations in human cells, particularly in sperm and egg cells, to alter human genomes potentially for the future. So, if you're talking about sperm and egg cells, to make alterations that will be carried generation after generation. And this is, again, an ethical minefield. What alterations should we be seeking? How safe are they? Can it be done technically? So, those are the three arenas that I think are most exciting.

How can scientists effectively navigate and communicate these complex ethical issues?

Well, first of all, one, The first thing one does is to learn the history. You know, this idea that we are coming close or approaching the language of positive eugenics, first of all, you have to explain what positive eugenics is and what it did and what it was, you know, in the United States, not far away from here. It's not, we're not talking about the atrocities of killings in Nazi Germany. We're talking about sterilization, the sterilization of Carrie Buck. We're talking about forced sterilizations of women, usually, in order to prevent genetic diseases based on very poor science usually, and very poor public understanding. So, that's one thing that, to begin the migration or the navigation across these ethical minefields, one has to talk about history and why that history is, what happened, why that happened. How did we let it happen, and what its consequences were. The second thing I think we need to focus on is defining rigid criteria, strong criteria, when we— about the kind of information that we are allowed to manipulate and allowed to have. So, for instance, information about diseases where the causal link between the disease and the gene is very strong, I gave you a couple of examples, Huntington's disease being one of them, is information that might be valuable to a parent for preimplantation genetic diagnosis. That information seems to be, seems to lie in the realm where an action could be justifiably taken. In other words, you could choose not to implant an embryo that carried a mutation which might cause a lethal neurodegenerative disease in the future. You may choose to, but you may not choose to. That choice might be left to you, because no one speaks for the embryo. You, as a parent, must make that decision. Whereas information about complex traits where there's— chances are you don't understand the traits, there's only a propensity, it is an accepted part of human diversity, it is a socially accepted part of human diversity, I think that information is not very useful. Providing that information only causes greater confusion among people. And this is not to take a paternalistic view of it, but just, and say, you know, you can't have that information. But it's to remind us that having that information and making a choice about it, making an action on it, is not justified because that action is based on incomplete information. Information is incomplete. It is never going to be complete. It is the nature of that information that it is not deterministic. I think that information should be limited. It should be limited by several mechanisms, but limited primarily because as a scientist, giving you that information could not and should not cause you to make an actionable choice because I cannot tell you in any manner, way, or scheme the link between that information and what you hope to achieve with it, which is a kind of a certainty about a future destiny. So, that's another one. The third one is the simplest of all, and that is that we really limit the information and the choices to diseases of extraordinary suffering. That the centerpiece of all of this, and this I'm speaking as a physician, should be suffering. The definition of suffering, our understanding of suffering, who suffers, because as I said, no one speaks for the embryo. But we do have an understanding of suffering, what suffering means. It's a malleable understanding that changes over time. And you might decide that a particular form of suffering isn't meaningful to you. You know, for a community of people with different stature, short statured people might decide that it is not a form of suffering. That, and that in fact there is a kind of, there is a, a malignant patronism to defining that as a kind of suffering. So, I think the way to navigate these fields are these, the things that I told you about. Number one is to have a very public conversation about the definition of what extraordinary suffering is. Number two, to understand the history of positive eugenics and the closeness that we've had to it historically in this country. And number 3, to be particularly cautious about information where genetic determinism is weak but is likely to be interpreted as strong. And so, I think these are early steps. I talk about them in the book. I call this a triangle. I think, you know, maintaining yourself within the triangle, extraordinary suffering, strong causal link between gene and disease, and no other justifiable alternative is a triangle that is kind of a safe zone. And of course, this is superimposed on the history of positive eugenics. You can, if you depart that zone, if you begin to intervene on something that we don't consider extraordinary suffering, whether, where the questions about extraordinary suffering of suffering are not resolved, are very much still in the public realm and in debate, we begin to depart from this triangle.