Scientist Stories: Ed Engleman, Co-Founder of Bolt Biotherapeutics

Good evening, everybody. Thank you so much for being here for this really exciting talk that Dr. Engelman will be presenting. I have the privilege of introducing Dr. Engelman. I could actually spend the whole hour talking about his incredible accomplishments. I won't do that, but I will highlight 4 specific, really groundbreaking and leader— excuse me, industry-leading accomplishments. One is the foundation of the founding of Stanford Blood Center. So Ed was, has been with the university for, for many years, and it was under Ed's direction and innovation that Stanford Blood Center was first founded back in 1978. The next 3 that I'd like to highlight have again really led the worldwide industry into making and adopting changes in blood banking. Ed was the world leader in understanding what we now know as HIV, but at the time, some type of a viral infection. And he's really the one that made the connection with, could this viral infection be transmitted through blood? And as a result, he put some screening tests into place here at Stanford. And the data shows that we probably mitigated approximately 400 to 500 HIV reactions at the time because of Ed's innovation. The second I'd like to highlight is cytomegalovirus, or CMV, a relatively benign virus for those of us who are relatively healthy. But for those patients that are immunocompromised and for newborns, it really can be fatal in blood transfusion. And again, under Ed's directorship, leadership, and innovation, the connection was made that CMV was not benign at all in transfusion to the wrong recipients. Stanford Blood Center was the world's leader in adopting that viral testing, and as a result, the mortality in transfused newborns was decreased. The third that I'd like to highlight is the development of the world's first— excuse me— the cancer vaccine for prostate cancer. And Ed is really the individual who's developed that methodology, and that vaccine is available today on the market. So it is my privilege to not only introduce Ed, but to be able to work with him every day, learn from him, and be inspired by him to do the great things that we can do here at Stanford Blood Center. So without any further ado, I'd like to welcome Ed.

Thank you, Harpreet, for that over-the-top introduction. Actually, Harpreet is the boss. Well, for those of you that have heard me before, and I know there are some of you here that have, there will be some repetition. I'm sorry about that. But there will also be some new things. I thought I would start out with sort of a historical perspective. I don't think I've done that here before. The way— how did we get to this point? In terms of the change and growth of how we do science, especially science when it comes to immunology and the treatment of cancer. Actually, if you, if you go back to the 1900s, say 1900 to about 1970, from my perspective, it was a pretty primitive time. When I went to medical school— and certainly when my dad was practicing medicine, the focus was mainly on listening very carefully to patients, trying to make patients feel better, developing what we used to call bedside manner. Some of you have heard of bedside manner. It's sort of a lost art. But it was bedside manner. And I think the reason for that was because we had such a poor understanding of what actually causes disease. And we had an even poorer understanding of how to treat diseases since we didn't understand how they were caused, including cancer. And so the drugs that were available were generally pretty primitive. The first real sort of intelligent approach to drug discovery involved antibiotics. I think the first discovered antibiotic was penicillin, I mean, sort of deliberately discovered antibiotic in the 1940s. After World War II, there was growing interest in the idea of trying to discover new medicines. And beginning in around 1970, there developed a keen interest on the part of our government, your government, in actually funding biomedical research. And that really started at the National Institutes of Health in Bethesda, Maryland. And in the 1970s, and into the '80s, there was an explosive growth of government funding at the NIH. And it was partly helped to catalyze biomedical research. And during that period of time, which was the Vietnam War, many physicians who would otherwise have been drafted and taken to the front, or at least near the front, had the opportunity to go to the NIH and do biomedical research. And in fact, that's what I did. During that period. And then as the years went on, not only was biomedical research supported by the government at the NIH, but they began to support it in a significant way at medical schools and other nonprofit medical research organizations. And that led over the years— with great congressional support, I might add— that actually led to a modernization of of science, an explosion of interest in biomedical science, and greatly improved methodology. In addition to providing funds that were based on competitive grant awards, the government made a decision to grant to the, to the schools, including Stanford, the rights, the ownership of the intellectual property based on the discoveries that were made at those schools that were really based on government-funded grants. So you had, beginning again in the 1970s and into the 1980s, a collection of events that made for an explosive growth in the science and the biotech. One was the ability to patent discoveries, actually own them. And again, my discoveries are owned by Stanford because I work for Stanford. Secondly, the ability to get funding for biomedical research, which led to the discoveries that led to the patents and to the breakthroughs, some of which I'll talk about tonight. And third, which is more controversial, the fact that the US remains the only large developed country on the planet that doesn't have regulated drug pricing. So the carrot was there for the discoverers, the scientists, industry. And right here in Silicon Valley was born the venture capital industry, which is today still a uniquely American phenomenon. And again, it started really here in the Bay Area with the creation of funds designed to, I think in essence, translate— in many ways, translate the kind of biological discoveries into— of course, there was the internet. And that whole field, but in the area that I'm interested in, really catalyzed the translation of basic discoveries into useful products. And a number of important companies like Genentech and others, mostly in California in the 1980s and into the 1990s, capitalized on the new discoveries that were made mainly in academia. And they were called biotech as opposed to pharmaceuticals because the pharmaceutical companies, which were mostly older and mainly in the Northeast, in New Jersey and Pennsylvania, those companies were making what we call small molecule drugs, chemicals that could be very effective, but they were not biologicals. They were not based on naturally occurring biological molecules. While the biotech companies began to use newer technology, particularly gene cloning, to start making the same molecules that we have in our bodies that actually control everything that goes on in our bodies. And one of those molecules, those biological molecules or proteins, turns out to be antibodies. So one of our most important defensive weapons are are antibodies. We have literally millions of those antibodies, each antibody having a different specificity. And now we're going to transition into a discussion of the immune system as a whole because our immune systems, which include antibodies and the cells that make them and other types of immune cells, is remarkably— our immune systems are remarkably powerful. And they're powerful because they're exquisitely specific. So we have natural products, natural cells and antibodies, proteins, that are able to recognize other proteins and other cells based on their molecular features. And that is unique to the immune system. And what it translates to is if we can harness the immune system, we can take advantage of its its specificity. And now gradually, that has occurred in the field of cancer. And I will talk more about that as obviously we go through in terms of immunotherapy for cancer. Now before I get there, I want to go back to the fact that Harpreet commented that the Blood Center began in 1978. And I was certainly there at the time. I can't hide from that. It's actually with great pride. And when we started the blood center, at that time we knew very little about the human immune system. Amazingly little. We knew that we could introduce foreign organisms if they were killed or otherwise modified, and that somehow the immune system would react to those organisms. And that it would be a specific kind of reaction. There would be antibodies made, for example, such that if the same organism alive and well would infect the body, that that vaccine reaction could protect us from the infection. So we started developing vaccines a long time ago to prevent infections. What we were having a much harder time doing was trying to understand how we could use the immune system to treat infections and, of course, cancer. And one of the great problems was we had no idea, other than this interesting phenomenon about vaccination, we had absolutely no idea how the immune system worked. We knew we had one. We didn't know what it consisted of. We knew that there were white blood cells, speaking of our wonderful blood center. You have in your circulation red blood cells and white blood cells, and of course the plasma. And we only knew that the white blood cells weren't red. And we knew that if you had an acute bacterial infection and there was pus, that there were a lot of white blood cells in there. We knew that one type of white blood cell was the source of antibodies. And that's about all we knew when I started here. And one of the accidental advantages of the blood bank was as we were making our blood transfusable products, we were not using the white blood cells. In fact, we removed them on purpose. And Stanford was one of the first and still one of the few blood centers that doesn't just remove those white blood cells, but instead of throwing them into the garbage, which most blood banks do, we actually save it and make it available to researchers. So when I started, when I started way back when with the blood center and those white blood cells that were being collected for disposal, the first thing I did was to take advantage of a brand new technology that had been discovered only a year or so earlier in Europe to try to make what we call monoclonal antibodies. And the way we would do that was to inject our human white blood cells into mice. And then with the hope that the mouse immune system would recognize those white blood cells as foreign, which they did, and make a whole bunch of antibodies, which they did. But with this new technology, we could isolate each antibody, or I should say the cells that produce an individual antibody, and screen those antibodies to see if they could distinguish that distinguish one type of white blood cell from another. Not too difficult conceptually, right? And with that technology, my group was able to generate about a dozen of these monoclonal antibodies that could distinguish white blood cells from one another, which otherwise look the same under a microscope. Quite amazing. And I had no idea at the time that, A, those antibodies would prove to be commercially valuable. Not on my mind, but I thought they were going to be great research tools. And indeed, they turned out to be great research tools. And now fast forward about 35 or 40 years, and we have about 500 of those antibodies that can recognize different aspects of white blood cells. So what we learned with the availability of those monoclonal antibodies that were directed against human white blood cells is that in fact, we don't just have white blood cells. A myriad of different types of white blood cells. Each one of those cells has a different function. And yet they can't work alone. They work as a coordinated army, if you will. They communicate with one another via soluble molecules. Some of the cells, which we call B cells, go on to make antibodies, to be sure. But the other cells, some of which we call T cells, are very powerful fighters. We know that there are other cells that sound the alarm. We call those dendritic cells. I'll show you some pictures of those. But those are just a few of the kinds of weaponry, the different types of white blood cells that we have in our bodies. And that work that I did, my group did over the first 5 or 10 years I was on the faculty, really helped catalyze the learning about how our immune systems work. And over the years, we've taken advantage of that knowledge. And of course, many labs all over the world have done that. And with that background, we began to think about what can we do to manipulate the immune system for the treatment of cancer. And I'll talk about that. And I'll start talking about that shortly. But I just want to mention that this interesting background where we came from a time when we had little or no understanding of how the immune system works. We had little or no understanding how disease works. And then beginning, as I said, in the 1970s or so, as more government money started flowing into biomedical research, the discoveries started accelerating. And we are where we are today. So with that background, we'll talk a little bit about immunology and cancer. First, I always like to let everyone know when I give talks that I do have interests in industry. I believe that the only— in fact, I've learned the hard way over the years that the only way for discoveries to be translated into really useful and helpful products is to transfer our technology out of academia into industry, in this case, the biotech industry. And as was mentioned earlier, our group was the discoverer of the technology that ultimately became the first cancer therapy therapy vaccine, the only one, a treatment vaccine. That was done after this company Dendrion was formed and licensed our technology from Stanford. And I was the founder of that company. More recently, new technology that I'll describe for you tonight has led to a licensing of our more recent discoveries to a brand new baby biotech company called Bolt, which hopefully will be able to develop this technology and hopefully demonstrate its utility in humans. So what you see on this interesting photo— stay on the left side there. See if I can figure out how to— I know that Kevin showed me how to do this. There we go. OK. So the cells of the many white blood cells that I described for you, I mentioned the B cells go on to make antibodies. The dendritic cells sound the alarm. And then the T cells, we believe, are the most important cells around when it comes to killing other cells, particularly tumor cells. And what you see here in green is a T cell, shown here in green or yellow, attacking a tumor cell, which is hard to see because the lights are on. But this is what these killer T cells can do. When they kill their targets, they can move on and kill another one and another one and another one. So they can be very, very efficient. And the goal of the type of tumor immunotherapy that is currently being developed and used and is the newest approach is designed to activate these killer T cells, to make them go out and hunt and kill tumor cells. And as it turns out, That was not easy to do. However, we learned over the last 15 or 20 years that if you look inside tumors, when tumors are first diagnosed and you remove them or you make a biopsy, and inside tumors you find not only the tumor cells, but you find a wide range of immune cells. Immune cells that we only now, because of our monoclonal antibodies, can tell what they do. And when this is done for all of these different cancers listed here on the left, you find that depending on the type of immune cells that dominate in the tumor, the prognosis is either really good or really bad. And so you can see here that killer T cells, which we call CD8, are associated with a very positive, good prognosis in almost all of these tumors. So if you have a predominance of killer T cells, that's a good thing. If on the other hand you have a predominance of what we call Tregs, which are T cells designed to inhibit the immune response, maybe prevent autoimmune disease. But if you have a predominance of Tregs in the tumors, that's a bad sign. That doesn't bode well. And there are other cell types. Only here, these are all different types of CD4 T cells. These are subclasses of a subpopulation of T cells, to give you some idea of how complicated things are. Now, what's not shown here is that there are a lot of reasons why the presence of those good T cells in the tumor may have a helpful prognosis. But most tumors don't have all those good cells in them. Or if they do, they're counteracted by other factors that are produced or guided by the tumor. So there are multiple barriers, multiple barriers to effective immunotherapy for cancer. By definition, by definition, if the tumor is not removed completely surgically at the time of diagnosis, and even if you have those immune cells there, it's a battle. And without therapy, without therapy, it's a battle that is usually lost. So the spontaneous immune response in our bodies to tumors that develop in our bodies usually fails. They put up a good fight and then they fail. And the reasons for the failure are because— and this is a very short list, believe me. When I give my lectures to the medical students and the graduate students, I have 5 slides of all the reasons that make it hard for the immune system to combat cancer. But the tumor environment itself, even with those immune cells in it, is even more full of factors and cell types that work the other way around, that prevent an immune response from taking place. And of course, tumors consist mainly of normal tissue components. And we have a wide range of mechanisms in our bodies that are constantly telling our immune system, don't attack, don't attack. And so because we don't want the immune system to attack our normal tissues, because what happens if that happened? We have autoimmune disease. So, we have these great checks and balances. We have a system in our thymic gland, our thymus gland, which is here. And it's there when you're a baby, and then it gradually gets smaller and smaller. And the only purpose of that thymus is to get rid of T cells that can recognize self. That's what it does. So, by the time you're born, the majority, but not all of them, the majority of the T cells that can recognize your own normal tissues have been killed off. That's probably a good thing. However, only recently, only recently did we learn that in fact we have quite a large number of T cells that can recognize self, but they're prevented from doing so by molecules we call checkpoint molecules. And that's a fairly recent discovery. I wish I'd made that discovery. That was a good one. Because it turns out, and I'll talk about this later, that if you can block those checkpoint antibodies, those checkpoints with antibodies, you can actually release the immune system to react very strongly against cancer. Well, cancer is at least as smart as viruses. At least as smart as viruses. And some viruses like HIV or hepatitis C, if you try to treat those viruses with a single drug that attacks a single part of the virus, it almost always fails. In fact, it always fails. It always fails. Why is that? Because these viruses are clever, and they have ways that they can escape from a single type of bullet. Now we know that in the case of hepatitis C, if we provide 5 different drugs in the same tablet, each drug targeting a different part of the virus, we can cure. With HIV, we can keep it down forever, even if we can't cure it. But it requires multiple attack points. Well, the idea that cancer is going to be curable by just targeting a single molecule is not realistic. It's going to take multiple attack points. So tumors can escape most therapies that target a singular, a single molecular component, and I'll show you that. So the goal of immunotherapy is either to bypass the barriers or overcome the barriers. Overcome the barriers. So my main interest is not in bypassing the barriers, but actually overcoming them and convincing the patient's own immune system to be able to attack and destroy tumors, as opposed to giving the patient a magic bullet that seeks out and destroys the tumors. But there are two ways of going about it. You can try the magic bullet approach, or you can try coming up with a way of hypnotizing or convincing the immune system and educating the immune system and pushing them toward attacking a patient's own tumor. That's called active immunotherapy. And that's what we're going to talk about today. If we had about 3 hours, I could go through the whole thing, but we don't. So an early approach, perhaps the earliest approach that is still a standard today, had to do with immunotherapy with tumor-binding monoclonal antibodies. Now, I told you earlier that in the late 1970s, methods were developed that allowed one to make single purified homogeneous preparations of single antibodies that would recognize a single molecular target. Well, eventually that technology matured and allowed for the creation of monoclonal antibodies that were of human origin so that they themselves, the antibodies, wouldn't be rejected and recognized as foreign by our own immune system when they're injected. So if I inject a mouse antibody into any one of you, it will get rejected because it will be recognized as a foreign protein. But if I inject I inject one of my antibodies into you, you'll be perfectly happy. And that antibody will do its job. So, the work began to develop these antibodies. In fact, I think Genentech down the road really was the dominant player, still the dominant player in terms of the ability to produce these antibodies at scale and use them for treating cancer. Now, each antibody Each tumor-recognizing antibody recognizes only a single molecular target. I told you 5 minutes ago that that's probably not going to cure cancer. And they work by either convincing other cells to come along and kill the tumor cells or by somehow slowing it down. But these antibodies, beginning in the 1990s or so, became standard therapy for a number of common tumors. Lymphomas, breast, certain breast cancers, colorectal and head and neck cancers, and so forth. And I'll show you just a couple of examples. And I'm showing you these examples for two purposes. One is this is the incremental benefit that was reported for Herceptin. Herceptin, an antibody perhaps known to many of you because it's a standard therapy for certain forms of breast cancer. This antibody called Herceptin recognizes a molecule that is expressed on the surface of some breast cancers. And when it binds to it, it can slow the tumor down. So you see here, this is a big study that was published in the New England Journal of Medicine in 2001. This was the study, probably the major study, that led to the FDA approval of Herceptin. And what you can see here is a comparison where patients were treated with advanced breast cancer, were either treated with chemotherapy alone, which is combination chemotherapy, or chemotherapy plus Herceptin. And this is basically survival. This is survival. So the fact that this line is higher than this line means that you had greater survival when you used Herceptin in combination with chemotherapy. That's good news. That's good news. The bad news is the difference isn't all that great, is it? Amazing. And Herceptin is a multibillion-dollar blockbuster drug. Now, over the years, clinicians and companies have figured out better ways to use Herceptin so it's more efficient and more effective. But nonetheless, Herceptin— almost never cures anybody with cancer. Here's another very important blockbuster antibody called Erbitux. Erbitux, in this case, for the treatment of a very common form of lung cancer. And you can see here survival time. Does this curve remind you a little bit of Herceptin? Yeah. Yeah, that's right. So there's a benefit. This is chemotherapy plus the antibody, and here's chemotherapy by itself. You won't, you won't see a curve with the antibody by itself, by the way, and that's because the efficacy is almost none. Almost none. Now, I'm not here to tell you that monoclonal antibodies for cancer are bad. In fact, they're good. They're good, but they're not very powerful on their own. Let's talk about the advantages. Advantages of mouse— or of monoclonal tumor-binding antibodies. They're cost-effective because you can manufacture huge amounts of the same antibody and you have a benefit of one size fits all. Anybody who expresses the target of that antibody can be treated with it and hopefully benefit. They're generally safe and well tolerated. Why are they so safe? Well, they're a whole lot better than most chemotherapies because Chemotherapies that were developed by the large pharmas, not the biotechs, but the large pharmas in the 1970s and '80s, those were primarily chemicals that were designed to kill cells that were rapidly dividing, growing rapidly. But that's a nonspecific effect because we have a lot of rapidly dividing normal cells, especially in our intestines. So chemotherapy Traditional, conventional chemotherapy is generally very toxic. One advantage of a monoclonal antibody is just that. It only recognizes a single type of molecule. And therefore, in general, immunological approaches, including antibody-based approaches, are generally really well tolerated. So you can continue to use these antibodies over and over again. And generally, and generally, there is not a what we call cumulative toxicity, in contrast to chemotherapy where the more you use, the cumulative toxicity builds up. However, there's limited efficacy as monotherapy for all the reasons we've discussed. Now today, in the last couple of years, with recognition of the fact that the monoclonal antibodies are by themselves not very effective, industry has turned to interesting ideas. They're now conjugating or chemically linking various types of toxic molecules or poisons to the antibodies, and the antibodies are serving as delivery vehicles to deliver those molecules to the tumor and hopefully kill the tumor. And then we have other things that I won't talk about today. We call them bispecifics. But these kinds of approaches with antibodies are much more potent And my hope is that they'll gradually replace the conventional monoclonals that are currently available. Now, I want to jump ahead. And there are reasons why I'm telling you all of this historically, to what is now becoming standard of care. This is a leap, a huge advance. It is transformative. And that is the discovery and development of antibodies that inhibit the checkpoint molecules. Now, I told you a few moments ago that about 10 years ago or so, an important discovery was made, which was that there exist these molecules and their sole purpose is to slow down an immune response. We call those checkpoint molecules. We did not know of their existence more than 10 years ago. Amazing. We thought, in fact, that all T cells and all B cells were a limit that could recognize self or subtle differences from self were eliminated during development. And it's true that that process takes place, but it is not efficient. Now we know that it is these molecules, these checkpoint molecules, that actually hold the immune system back. Now, they're important. They're important because if our immune systems are allowed to react to anything they want, anything that gets in their way, if you will, the assumption is we'll have a whole lot of autoimmune disease. And I think that's a pretty valid assumption, as I'll show you in a moment. But the first antibody to recognize a checkpoint molecule, which is called CTLA-4, or ipilimumab, developed by Bristol-Myers, turned out to be remarkably effective, but also remarkably nasty in terms of inducing autoimmune disease. Nonetheless, ipilimumab And I know, I knew some of the patients that were getting the first doses of ipilimumab, particularly patients with malignant melanoma that had metastasized all over their bodies. And this was the first therapy that was able to slow and in some cases stop that kind of cancer. The problem with ipilimumab or CTLA-4 antibody is the high incidence of autoimmunity, which can sometimes be completely intolerable. But luck would have it that a few years later, another checkpoint molecule was discovered, which is called PD-1. It has a mirror image co-receptor called PD-L1. And it turns out that unlike CTLA-4, when you block PD-1 or the interaction between PD-1 and PD-L1, you do in fact block the checkpoints. And you do in fact release the immune system, but you get far far less autoimmune disease for reasons that we don't fully understand. And I'm not going to go through the molecular basis for this here tonight. Not to worry. There's no test at the end of today. But I will summarize for you how these antibodies are now being used in the clinic. And as I said before, this is revolutionizing not only immunotherapy, but the treatment of cancer in general. So anti-PD-1 and anti-PD-L1 antibodies demonstrate broad Oopsie, sorry. Broad activity and greater tolerability than CTLA-4. There are certain cancers that appear to be generally sensitive to this treatment, notably melanoma, renal cancer, bladder cancer, many lung cancers, stomach cancer, head and neck cancers, and ovarian cancers. But interestingly, only about 20% to 40% of patients are responsive. It's a remarkable thing because the number used to be 0. 0. This is with patients with advanced disease. But the majority of patients with these susceptible cancers, these cancers that are susceptible to this treatment, are resistant. And we don't know why. We can speculate. And I'll show you some recent data that suggests we're now beginning to understand that. But we don't know fully why. But I can tell you— because of the data— and I'll show you some of the data— the availability of these antibodies is transforming the way we treat cancer. And every major pharmaceutical company around the world— I think that might be a slight exaggeration, but I would say the vast majority of major pharmaceutical companies— are developing and/or commercializing their versions of these antibodies as we speak. All right. Let me show you some data. So a few moments ago, I showed you some pictures from, uh, the classical studies, uh, for the development of Herceptin for breast cancer, uh, for Erbitux for the treatment of lung cancer. Now this is a paper that was also published in the New England Journal at the end of 2016, uh, and this represents the treatment of patients with lung cancer non-small cell lung cancer, which is the most common variety, with an antibody that's targeting the checkpoint molecule PD-1. This is not an antibody that's targeting the tumor itself. Let me make that very clear. This is an antibody that's targeting the immune system. The immune system. And this is the response rate. This is survival. Survival. So here— and this is, by the way, no chemo. Just the antibody, just the antibody. So here you are manipulating only the immune system, and without the antibody, with standard of care, this chemotherapy, this is the lifespan over time. Here's the response in patients who receive that antibody. Now that's a huge, huge response. It's about a 50% response rate. Now I will tell you there's a little caveat. And that is that the patients who were eligible for this trial, their tumors had to express the PD-L1 molecule. Otherwise, they weren't eligible. So, and only about 30% to 35%, as far as I know, of lung cancers express that target molecule. But the bottom line is that this is extremely rational therapy. The drug is well tolerated. And you can imagine how important study like this. This is more details that I won't go into. Now, so knowing that some patients respond dramatically well and can live a long time with that therapy, the other thing that the companies are doing— and now there are about 4 or 5 of these antibodies out there that are FDA approved for one indication or another— now what they're doing is they're saying, well, what if we took one checkpoint antibody and treated patients with two different ones. One that blocked PD-L1, one that blocked CTLA-4. So that you double up. And what do you think happens? Well, this is malignant melanoma. So on the right here is what happens if you just use ipilimumab, which is the first anti-CTLA-4 antibody. And this is just tumor size. This is actually not survival. This is size of the tumor.. And you can see that in some cases, the tumor size goes down. Some patients, it goes up. It's kind of borderline benefit. Here's when you use both an anti-PD-1 and anti-CTLA-4 antibody. And you see that about 80% of the patients have a pretty dramatic effect on their tumor. Remarkable. Only problem, of course, is a lot of patients developed autoimmune disease. But it was a trade-off. That is today still debatable, but potentially a reasonable one. So let me summarize, because tonight's talk is really not around checkpoint antibodies other than to tell you how exciting that field is. What are the advantages of checkpoint antibodies? Cost-effective manufacturing. One size fits all. This is not personalized. The PD-1 and PD-L1 mabs are generally well tolerated. Generally. The therapeutic activity appears to be durable, well tolerated, and there are a number of different tumor types in which this works. And there's synergism, or additive activity, with numerous other agents. So now, what I didn't tell you is that these— I gave a lecture at Merck about 10 months ago, Merck is one of the companies that has an approved checkpoint antibody. And they were telling me that at that time, which is March of 2016, that they had started 200 clinical trials with their single anti-checkpoint antibody in combination with other agents. Think about that. 200 trials, one company. One company. And they were so proud. They said, we have 200 and Bristol-Myers only has 140. You may remember, or you may not, when Jimmy Carter— it was only about a year ago when I saw this on television. It was remarkable. Jimmy Carter was being interviewed. And he'd had malignant melanoma. And it had spread to his brain. Now normally, up until a few years ago, when that happened, it was get ready and say goodbye to your friends. And Jimmy Carter went on television and said, the tumor's gone. I'm cured. Truly remarkable. Truly remarkable. Now, I don't know if he's really cured. All I know is the tumor is no longer detectable. So there can be dramatic benefits from these antibodies. Now, what about the challenges? What about the challenges? Well, many patients and tumor types are resistant. In fact, the majority of patients with cancer are still resistant. The reasons for this resistance, I'm not going to go into. It gets very technical. But one reason, of course, is these therapies can only work if your tumor has T cells in it. No T cells, nothing to excite. And then the other problem, of course, is this autoimmunity burden. But let's keep going. So I'm going to show you— what we call a hot tumor and a cold tumor today. I mentioned to you that checkpoint antibodies can only work if they can unleash the power of T cells in the tumor. And here's a tumor, a breast cancer. This is an experimental breast cancer from our lab. Provided this photograph from one of my graduate students, Michelle Ataya. And what you can see, the little red dots are T cells. And these are not very common in this tumor. And breast cancer in general does not respond well to checkpoint antibodies. On the other hand, here's a melanoma. And look at all the T cells in this melanoma. So melanoma, as it turns out, often responds well. So just to give you some idea of the kind of research that can be done to begin to explore and explain why does a certain treatment work and another treatment doesn't work in a given tumor. I'd like to turn now and push the clock back. I do have a little time left, I hope, on another early approach, the one that was described by Harpreet. And that was our early work with dendritic cells. Dendritic cells, another type of white blood cell. This is the type of white blood cell that sounds the alarm. It's not the cell that kills. It's the cell that says, wait a minute. We better get the killer cells organized. And without these dendritic cells, the T cells don't get activated. This is what they look like in our tissues normally. You can see these long dendrites or finger-like projections. We have these cells everywhere, but they're relatively rare because they— each cell can cover a lot of territory. And they're there as scouts. And I'll show you how this works. So they come out of the bone marrow as progenitors. They migrate through our bloodstream. They land. In various tissues. I'm showing you here the skin, the top layer of the skin or the epidermis, the middle layer, the dermis, a little deeper. These cells then constantly sample the environment. They're looking for substances that shouldn't be there, what we call foreign substances. And if they come into contact with a foreign substance, they have a variety of antennae or receptors that recognize those substances. And they get very excited. They take them up. They gobble them up. And what do they do? They leave their post. This is the only time in the lifespan of these cells or of the host that they leave their post. They go via the lymphatics into the draining lymph nodes where they then come into direct contact with several other types of immune cells, including CD4 T cells and CD8 T cells and natural killer cells and B cells that go on to make antibodies. And what do they do? They induce an immune response. That's what they do. These are the cells that sound the alarm and turn on the immune reaction. And for that reason, those of us that were really interested in this area going back about 25 years or 30 years consider them to be central to the immune response or our universe of immune responses. So back in 1992, At the blood center, we developed ways to purify these cells from the blood, not only of healthy individuals, but also of patients with cancer. And the idea— I had this extremely simple-minded idea. It was, in retrospect, an embarrassingly simple idea. But the idea was, well, if we could isolate these cells, take them out of the body, away from that immunosuppressive environment that the cancer caused, and then load them with an antigen, that was expressed mainly by the tumor and then give them back to the patient, maybe they would go into the patient's immune system and induce a powerful immune response against the tumor. That was the idea. The likelihood that it would work was probably 1 in 1,000 because it shouldn't have worked for reasons that we can talk about later. But that was the idea. And this is the way it was done. It was done right here— well, not right here. Not right here, but right there at the blood center. Patients would come to our blood center and they would undergo a procedure called a leukapheresis. And some of you have maybe heard about apheresis, where we collect platelets. You can collect a particular type of blood cell without having to collect all the blood cells. So in plateletpheresis, we take platelets, but we give back the red cells and the plasma and so forth. Well, leukaphoresis— leuka stands for white. We take out the white blood cells and we don't take out the other cells. And we can get a lot of white blood cells. And from there, we could isolate the dendritic cells, which were very rare. And then we would manipulate them in test tube, load them with a tumor-associated antigen, wash them so that none of the free antigen was around, and we would inject them right back into the patient. So whereas we maybe took out 1 billion white cells, we maybe isolate 5 million dendritic cells, give them back to the patient with the hope that they would then induce immune response. And wouldn't you know it, they did. They did. So back in the early '90s, some 25 years ago, we published the first report that said you could actually you immunize a patient with cancer against their own cancer, and they would mount an immune response against it. A complete surprise. A complete surprise. And this is an old-fashioned CT scan. This is like a specialized X-ray right in the middle of the heart here, but non-invasive. And these are the two main chambers of the heart. This is your vertebral body. This is the aorta, actually, believe it or not. You're cutting through here. And I don't think you can see it very well here with the lights, but believe it or not, there's tumor here and here, big tumors. And at the end of 6 months, the tumor was gone. And we were able to demonstrate that in about 3 consecutive patients. And after treating 4 patients, we published a paper. 4 patients, and it looked really promising, no serious toxicity. We then went on over the next few years to treat several different types of tumors. This is not an easy therapy. It took a long time to isolate the cells and so forth. But we had very promising data in small numbers of patients. And that led eventually to the university, my employer, filing invention— I filed invention disclosures. They then filed patents. At the time, no drug company was interested in this because This was not a molecule in a bottle, one size fits all. This was much harder. You had to get dendritic cells from each patient, manipulate them separately, and then give them back to the patient. So none of the existing companies had any interest in this at all. So I went out and met with venture capitalists. And some of them were interested enough to want to finance a company that would try to develop this for patients. So the company licensed the technology from Stanford. And some scientists from my lab, after they finished their training, went to the company and transferred the technology. And eventually, the company then proceeded to design a vaccine with dendritic cells. They decided to proceed to prostate cancer, mainly because they were so worried that if we tried to treat lung cancer or cancer of a vital organ, we would induce autoimmune disease and bad things would happen. But with prostates, we men have them, but we don't really need them. They're not exactly vital. So if we induced an autoimmune response to the prostate, so what? So they decided to pursue prostate cancer. And they developed a manufacturing process. They filed with the FDA. And then over the next many years performed a number of clinical trials. And eventually, they showed— in a graph, a little bit reminiscent of the graphs I showed you earlier, that patients who received the vaccine lived longer than patients who didn't receive the vaccine. But like those early antibody studies, this looks nice, but we're not curing a lot of people, are we? No. And this took so many years to do and cost so much money. Nonetheless, By the time of 2009, about 700 patients had been treated. The demonstration that it was safe and effective was important. The vaccine was well tolerated. And in 2010, this vaccine, which technical name was Cipollucel-T, is called Provenge, was approved by the FDA. Happiest day of my life. I think my wife was pretty happy then too. Well, that vaccine is used today. It's available today. It's certainly well tolerated in general. But it's extremely costly to produce because each vaccine is based on cells from an individual patient. And because it's a monotherapy, the efficacy is modest. So over the last 20 years or so, my lab and the students that are in it and almost everybody in my lab here at Stanford is a student. And as a result, they graduate and they leave, which is very sad for me. I'll show you a picture of some of the key people from the lab toward the end of the talk. But these are my second family, my students. And they know that one of the things that I wanted to do and have sort of dreamt about for 20 or 25 years, ever since we came up with the dendritic cell approach was to come up with a way of manipulating those darn dendritic cells in the tumor without having to remove the dendritic cells from the patient and go through that expensive manipulation and giving them back. So the question was, is there a way to do this? And beginning around 5 or 6 years ago, a new postdoc in the lab and several grad students worked very hard on a project that I'll describe for you the results. And it turns out that by combining monoclonal antibodies to tumors, like Herceptin or other tumor-binding antibodies that don't work so well on their own, and specific molecules that can stimulate dendritic cells to become activated, it turns out you get remarkable synergy and remarkable potency. So I'm not going to go through the details. It's fairly technical. This is all published about a year and a half ago in a journal called Nature that describes this remarkable discovery. And I'm going to skip over all of these things and just show you a little example of data. Now, one frustration is that the discovery was made and tested in experimental animals. It was not made and tested in humans. To humans. That takes a long time and a lot of money that we don't have, but hopefully will happen over time. But this is an example of where we treated a type of cancer that is resistant to checkpoint blockade. This is breast cancer in a mouse, in a mouse. And what we're doing is treating a tumor, the breast tumor, at a time when the tumor has already metastasized to the lungs. And you can see here that when we treat the primary tumor in red, this is the combination therapy that I'm talking about, we can cure that primary tumor. The individual components of our therapy, the anti-tumor antibody, the dendritic cell stimuli by themselves do not work. But only in combination do they work. But remarkably, not only do we cure the primary tumor, we also cure the nodules or the metastases in the lungs. So here are the nodules in the lungs. Of the control animals, and here we are in the treated animals. So very exciting, and this could be done not only with breast cancer, but with a wide range of cancers as described in the paper. So I'm not going to go into great detail other than to say that we have a novel approach. It appears to be extremely potent against diverse tumors in mice. It's active against tumors that are resistant to checkpoint blockade. The challenge, of course, not yet tested in human beings. That's not a small step, I admit it. And so we have a company that's been formed called Bolt Bio. And they've licensed our technology from Stanford. And my hope is that they'll be able to bring this technology through and demonstrate that it's going to work in human cancer. Well, so that's what— our lab does here at the Blood Center. But obviously, as I've gone through this talk, I pointed out the pluses and minuses of each of these therapeutic strategies. And by the way, there are other therapeutic strategies. I didn't have time to go through them all. But what we do know is that while cancer immunotherapy can be effective, we really don't fully understand how they work, when they work. And we're only— we know that we don't really know, for example, whether does it work based solely on the immune response and immune cell content in the tumor? Or is it reliant on some more systemic type of immune response? What are the causes of resistance? So these are big questions. And I don't want to suggest to you that we're the only ones on the planet that are investigating these questions. Can you imagine? How important these questions are when these types of approaches are now becoming the standard of care for cancer. So we took it upon ourselves, the students in the lab, to investigate these questions. We developed a method that enables a detailed analysis throughout the body of the immune system. Think about that. A detailed way of analyzing the immune response throughout the body in different tissues all at the same time. Now, we can't do that yet in people, but we can do it in mice. And I'm not going to go through the technology in detail except to say— in fact, I don't think I should even mention it. It gets a bit too technical. Except we're using an approach called mass cytometry, which actually discovered or developed here at Stanford. By Gary Nolan's lab. And then we combined that approach with powerful informatic tools. And essentially, this is the way the experiments were done. So we would have our little mice and various types of tumors in these animals. And we would treat the animals either with an ineffective therapy— and in this case, the ineffective therapy is checkpoint antibody. We're using breast cancer. And we would also, in other mice, we would treat with our effective therapy, the new therapy that we published a year and a half ago in Nature. And after treatment, immediately after treatment, we would start analyzing the nature of the immune cells in various tissues, including the tumor, the lymph nodes, the spleen, the circulating blood, and even the bone marrow. All of those tissues. And every few days, we would sample them and analyze them. And we'd analyze literally 40 different components in the immune cells, 90 different immune cell types, if you can imagine, in all these tissues. And then we would subject the data to our bioinformatics analysis. And what we found— and it was just published about a month ago in a journal called Cell, if you really want to read the details— is that effective immunotherapy— effective immunotherapy, therapy that actually cures the tumor— generates a coordinated system-wide anti-tumor immune response that involves sites all over the body, near and far from the tumor, not just in the tumor. In point of fact, this systemic immune response persists long after immune cell activation in the tumor has ceased. And it's required for efficacy. It's required for it to work. We further discovered a novel population of immune cells, a population of CD4 effector cells— that's a technical detail— that seems to selectively expand in those mice that are cured and not in the animals that were not treated well. And we discovered that when we looked for those same kind of cells in patients, who had received immunotherapy. It was the patients who had responded that also developed those same cell types. So a very helpful biomarker of efficacy. So all of this is published in detail, uh, in, uh, about a month ago in, uh, in this journal. And I think it's going to really change the way we think about and design new therapies for cancer. In fact, not only therapies for cancer, but think about this. So we now have a way of analyzing the immune system system-wide. And we can subject any— well, eventually humans, but any mouse with any disease to an analysis of the immune system in those mice. And we can further analyze what happens if you treat those animals with that disease. What are the desirable changes? And what are the changes that don't count? So I think that this is going to turn turn out to be profoundly important over the next 5 to 10 years as we try to understand how best to utilize the immune system to treat not only cancer, but other diseases as well. So I'm going to finish up now, a few minutes late, and summarize where we are today in terms of immunotherapy. We know that immunotherapy can lead to durable tumor regression in a growing range of tumors. And what does that mean? That fancy language means we can control the tumor for at least years, at least years. And we can cure some patients of their tumors using immunotherapy. The most effective immunotherapies rely on tumor-attacking T cells, T cells that can be generated as part of a coordinated system-wide immune response that relies on many cell types, not just those killer T cells. Every cell and every molecule capable of influencing the immune response represents a potential therapeutic target. And I am not the first person to suggest that. I said that Merck was, as of a year ago almost, doing 200 clinical trials with combinations with their antibody, their anti-PD-1 antibody. Bristol-Myers was doing 140. And Genentech, Roche, probably doing 100— 25, for all I know. So every conceivable combination that you and I and others could think of is currently being tested. And over the next few years, we're going to find out which combinations work the best. In the future, immunotherapy will be comprised of combinations of products that engage different components of the immune response, just like we're doing with certain viral infections. I mentioned hepatitis C and HIV. And we're going to do the same thing with immunotherapy. So that's pretty much fact. Now the question is predictions, predictions. And these are hypotheticals. But I believe that most patients with inoperable cancer, that is, most patients with tumors that cannot be surgically removed, are likely to receive immunotherapy. That's most patients with cancer, not just lung cancer and renal cancer, bladder cancer. Combination immunotherapy will likely control and possibly cure many cancers. Many cancers. How far away are we from this? Maybe, maybe 8 to 10 years max. Many patients, however, will develop autoimmune disorders as a result of immunotherapy. Is that an acceptable trade-off? We now have at Stanford a clinic, a weekly clinic devoted entirely to the autoimmune disorders that result from immunotherapy. Think about that. We're creating a whole new industry. The good news, the good news is that with rare exception, these autoimmune phenomena are not life-threatening. And they are at least manageable. And often they're mild. Often they're mild. But I think it does represent a slight warning light that you can't fool Mother Nature completely. That if you want to turn on the immune system, nonspecifically, as you do with checkpoint inhibitors, you're likely to run into some problems with attacks on normal tissue components. Now, the last question is one that I really don't have an answer for. And that is, we have in the past developed treatments for cancer that I would call incremental. Useful, incredibly expensive, but only moderately useful. I went through— and described for you the monoclonal antibodies that are standard of care for so many tumors and how their utility is limited because their efficacy is modest. Now we're on the verge of having therapies— we already have them— that are not providing marginal benefit. They're providing dramatic benefit, really dramatic benefit. But the question is now, we have a lot of patients with cancer, millions of patients with cancer. Who's going to pay? Who's going to pay? I don't have the answer, but I leave you with that thought. Not only for these therapies that I've described for you tonight, but all the other breakthrough therapies that are right on the heels of these therapies. So on the one hand, I am incredibly optimistic. And I hope I've communicated that optimism to you in terms of what the future holds for immunotherapy. In cancer. On the other hand, it's a little worrisome. We're getting awfully good at this, and the question is, how will we pay for it? Okay, well, with that, I want to show you a photograph. Um, this is not the current lab. This is the lab as it was a year ago. And the reason I'm showing you these photographs is because these are the individuals who do this extraordinary work. I am the luckiest man around because These are brilliant, hardworking, really lovable students. This fellow here, Yaron Carmi, came to the lab from Israel where he started as a postdoctoral fellow in the lab. He's back in Israel. He's now a professor. But when he came here, he didn't know what the hell he was going to do with his life. Fortunately, he came to the right place and he worked with some really bright graduate students. This guy, Ian. This guy, Matt, this guy in particular, Michael Alonso, who was not only my graduate student but my postdoc. And then he was a lab manager. And lo and behold, he left for Bolt Biotech.

What can I do?

Will you back up 5 slides? Uh-oh. Oh, wait a minute. Maybe 6. Actually, it only goes forward, I think. You were able to go back earlier. Yeah. You have to tell me when. Keep going. That one. No, yes. You seem to gloss over the last entry there, and I thought that was probably the most provocative. That one?

Yes.

Okay. You can take the cells out of a mouse and transfer it to another mouse and and provide him immunity? Yes. Well, the idea that using T cells to transfer into a person to attack cancer is not new. And in fact, I didn't go through this as part of today's discussion because today's discussion was focused on what I call active immunotherapy, ways of convincing the patient or the tumor host to mount an immune response to the tumor. Another approach would be to give a patient what I would call a silver bullet. You could take T cells out of that patient. Maybe you can educate them in the test tube and give them back. In this case, yes, what we showed is that if you effectively treat these animals with our immunotherapeutic strategy, Eventually, you can identify the cells that are responsible for doing the work. In this case, it turned out that these CD4 T cells played a critical role. And we found the same types of cells in patients who responded positively to immunotherapy. So the point here was that by doing this detailed system-wide analysis, you can, in fact, ultimately work things out and figure out what's good and what's bad. But I hope I've answered your question. But we don't want— You didn't transfer the cells back to the original mouse. You gave it to another mouse that did well. That's correct. Right. That other mouse, however, is genetically identical to the first mouse. Well, genetically identical because my cells, if injected into you— remember, white cells and these lymphocytes they are going to always react against foreign things. So your cells don't like my cells, and my cells don't like your cells. Now, there's a whole other science that is being worked on to try to make us friendlier. But cell therapy, which was the dendritic cell approach that we took 25 years ago, is still a viable concept. My own personal view is that if it involves the removal and manipulation of a patient's own cells before you give them back, that the cost of doing that is so high that I question whether or not that can be a long-term viable solution. But anyway, that's another discussion. But it is absolutely another potential approach. Yes?

Thank you. Thank you. Back here. I have two questions, but first of all, I'm right here. Back here.

Yeah, thanks.

I'm one of your preemie donors, so thanks for making me feel special. And can you back those slides up? Yikes. More.

Just keep going.

Nope. Keep going faster. More, more, more. You could go faster if you want.

How far back do you want me to go?

You had so many, just keep going. I know. The ones where they had the red dots in the bottom. Ah, okay. And while you're getting there, I have two questions. So my first one is— No, the red dots in the bottom of the chart. On the bottom of the chart, keep going. I'm sorry, I should have stopped you.

Right there.

Whoa.

Go back. OK. So but let me ask my first question first. So since the body systems are so complicated, why is it that it's required that a person go through chemo before getting into a clinical trial? It seems that you would want the immune system to be as strong as possible.

Yeah. So let me rephrase. The question. So under our general ethical principles— when I say our, I mean what has evolved over time and is generally accepted and expected— we don't generally use experimental drugs in patients unless they have and are receiving standard of care. So it is viewed as unfair, inappropriate, unethical, unethical to try an experimental therapy, one that has not been shown to be effective by definition because it's an experimental therapy, on patients who have not yet been given the opportunity to have standard of care therapy. And I think that's the best explanation. Now there's another— there's an additional explanation. That turns out, and this is more complicated and I wish we had another hour, but it turns out that certain types of chemotherapy, just like certain types of radiation therapy, turn out to be incredibly synergistic with immunotherapy. It's very surprising. We used to think that chemotherapy and radiation therapy worked solely by killing tumor cells. Wrong, wrong, wrong. It turns out, for example, that certain types of radiation therapy— by the way, the majority of patients with cancer, with advanced cancer, receive radiation therapy. And it turns out that if you use radiation therapy in a mouse, it can work just like it does in a human. But in mice, we can take out their immune systems. And if you remove the immune system from a mouse, I mean, the T cells, specifically the T cells, and radiate that same tumor, the radiation doesn't work. So in other words, effective radiation appears to require, in many cases, appears to require an immune system to actually make it work. So there are many cases and many examples where the concomitant use of chemo, even though it's still crude, with immunotherapy is likely to make the immunotherapy more effective. That's a short, very abbreviated summary. So there's two issues going through there. One is the ethical issue about making sure the patient gets standard of care. And the other is the potential additive benefit of putting these things together.

It seems the first time I saw that slide, I could read what the red dot meant. Can you read that for me now?

The red dots? Say patient with confirmed response.

What—

why wouldn't all of them have a confirmed? Because, um, you have to be able to achieve a minimum of a 25% reduction in tumor size to be defined as a responder. That simple. I have no influence over who gets chosen for the next question.

My question is, you didn't mention anything about sarcomas.

And I was wondering if you could talk about any research or progress that's been made on sarcomas. Well, let me just say that I wish that every type of cancer and every type of tumor would behave the same. The opposite is true, that every tumor has its own unique characteristics. And in fact, even within a given tumor type, as you've seen today, some tumors behave differently than others. That's why only 20 to 40% of kidney cancers are responsive to single checkpoint antibody blockade.

Sarcoma—

I didn't get into this, but the tumors that are responsive to to checkpoint antibodies tend to be those that, A, have a lot of T cells in the tumors— talked about that— and B, have a large number of mutations. We call that mutational burden, mutational burden. So lung cancer and melanomas have a large number of mutations. All tumors have mutations, but some tumors have a lot more than others. And mutations are changes in genes that encode proteins. Not all of them, but many of them. So if you have a mutated gene that encodes a protein, that mutation will cause the protein to have a different amino acid sequence. And as a result, it'll look similar to the natural protein, but actually to the immune system, it looks foreign. So the more mutations in a tumor, the more likely you're going to have foreign proteins expressed by that tumor that are going to be attacked by the immune system, by the T cells. Does that make sense? It turns out in sarcoma, or at least— now, there's many different subtypes of sarcoma, as you may be aware. But in most sarcomas, sarcomas have a low number of mutations in general. So for tumors that have low numbers of mutations, they tend to be resistant to checkpoint blockade. That doesn't mean that immunotherapy won't eventually work for sarcoma, but today we haven't yet seen good responses in most types of sarcoma.

Yeah, I was curious, the anti-CDLA-4, you mentioned that that had a certain amount of autoimmune disease. That's a murine antibody, isn't it?

No. No. Because the CTLA-4 antibodies that are available now, ipi and the other ones that are now approved, are all humanized.

They're all humanized. Because the one you had listed up there, I'm pretty sure was a murine based on its generic name. Yeah.

No, it's humanized.

Now humanized.

Thank you.

I had two questions. First was, you talked about the side effects of autoimmune disorders that tend to develop with combination immunotherapy. Could you tell us a little bit more about that? Are these specific disorders like rheumatoid arthritis or Graves' or something? Or is it sort of an undefined autoimmune response?

Actually, that's a really good question. For the most part, the autoimmunity that we see does not correspond to the known autoimmune syndromes, the common ones like lupus and rheumatoid arthritis. However, there are certain manifestations, autoimmune manifestations, that are much more common than others. Skin rashes that are autoimmune are particularly common. Colitis, or inflammation of the colon, is a big problem with the CTLA-4 ipilimumab antibody. And you name it, if you think of it, you can get it. But they are truly autoimmune. And in fact, if you remove the drug, if you remove the antibody, generally speaking, the autoimmunity will get better. You can manage the autoimmunity with immunosuppressive drugs, typically corticosteroids. And another question that I'm sure would occur to you is, well, if you develop the autoimmune complication and you now you use the immunosuppressive drugs to stop it, what happens to the benefit? And I can't answer the question, but I can— definitively, I can say that with IPPI, with the CTLA-4, after several, several years of follow-up, it appears that the benefit is retained, which is surprising. We don't have enough years of follow-up with the PD-1 and PD-L1 antibodies to be more certain about this. But it appears to me, at least at first blush, that despite the fact that we're seeing more and more of these autoimmune disorders, they appear to be manageable and not nearly the problem of the underlying cancer.

Great, thank you. So my second question then was about your monitoring or evaluation of the entire immune system. Are you doing that through some sort of— could you describe that a little more, tell us a little more? I'm curious about that. Is that through some real-time monitoring? Is it through some genetic analysis of microwaves? You're drawing blood.

What are the procedures and— That's a good question. I'm happy to brag about our research. Sure. The reason why it can only be done in mice today is because this is an invasive procedure. So what we do with these animals, with these experimental mice that have cancer, is we treat them, treat them, and then we have groups of mice. Remember, they're genetically homogeneous, so they're all likely to respond in a similar way, which is a huge advantage over us. And nonetheless, they develop cancers. They die just like we do. And if they're not treated effectively, they'll die. And so the way this is done is we create the cancer, or we transfer the cancer. And then we wait a certain period of time until the cancer is of a large enough size. Then we treat. And then a certain percentage of the animals will be euthanized. And the various tissues in the animal will be isolated. And then the immune cells will be prepared from each of those tissues and subjected to our mass cytometry analysis. That analysis enables us to look at each of millions of cells for 50 different proteins. Think about that. That's how powerful the— modern technology has become. And by doing that repeatedly over a period of time, say 3 different times, or in this case 2 different times, day 3 and day 8, we did it at a time when the immune response was first getting into the tumor and then after the tumor was well on its way to being cured or not. And then we analyzed all of that data, which required incredible amount of computational computational power and what we call algorithms for providing the analysis. In fact, it would be about a 30-minute talk just to explain all this. But that allows you to literally visualize the changes that take place after effective therapy in each one of those tissues. And what we found is that the changes that correlated with effectiveness took place everywhere, everywhere, in the blood, in the bone marrow, in the lymph nodes. And we even went beyond that and we found a drug that prevents the immune cells from leaving the lymphocyte, the lymphoid tissues. In other words, we showed that there was this massive coordinated systemic immune response and then we treated the animals with a molecule that prevents those lymphocytes from getting out. And what do you think happened to the tumor, the anti-tumor effect? Gone, gone. So we were able to show the systemic immune response that was absolutely required regardless of what is going on inside that one tumor. So it was a surprising result, but a very powerful and important one. And we think that this approach allows us to generate what we call an immunological signature, an immunological signature that ultimately we will be able to find what is a good signature. You know, how will we know when a patient is is going to respond or is responding. And that's the initial use of this technology. But eventually, we're going to use it to design better therapies. So that's why I'm so proud of this approach and how powerful I think it's going to be. And I'm working hard to convince the new students in the lab to use it to answer a number of important questions.

Thanks for the talk. The slide that showed the breast tissue and the melanoma tissue with the different densities of the T cells in those, I was getting the impression that the T cells don't migrate at all. And I was trying to reconcile that against the other slide where you talked about the transferability from one mouse to another by sharing the T cells. So how does that all reconcile?

OK, if there's one message I want to get over, it's that these cells move all over the place. They're incredibly mobile. And in fact, in therapy at the very least, the importance is getting them from point A to point B. Once inside the tumor, unless you can induce a powerful immune response to the tumor, those cells are just going to sit there. Now, those cells that I showed you in the tumor, including the T cells, are pretty much sitting on their hands. The tumor, which I didn't go into in detail, produces a wide range of substances. And those substances can hypnotize or paralyze all of these immune cells. In addition to T cells, there are monocytes, there are macrophages, there are dendritic cells. They're all sitting there. In fact, they're not just sitting there, they're promoting the tumor. They've been completely subverted. If we can change that environment and bring new cells in to kill the tumor, that's what works. So I don't mean to suggest to you that they're not mobile.

And was the density important also?

The density is important. We don't know the exact— right now it's relative frequency. Right. Thanks.

Any other questions? I know one of the protocols of that combined therapy is to do a certain number of combined infusions and then continue with an anti-PD-1.

Why? Well, we know that when you inject an antibody, a human antibody, into a person, that on average, on average, it's going to last around a month. And then after about a month, you have about half as much left, and it keeps going down. And it varies between different monoclonal antibodies. So the reason is very simple. It's that you want to— if you think you need to continue to treat, and you think that the antibody is gone, you treat again. We actually don't know yet what the optimal therapeutic frequency is. We don't know. These are very new therapies. Very new therapies. The FDA approved checkpoint antibodies for the tumors that I mentioned. I mean, they're approved for non-small cell lung cancer, for kidney cancer, I think for bladder cancer, and for melanoma. And over the next couple of years, you'll see one approval after another after another. They're generally given at intervals maybe once a month. They're very expensive, or I should say they're very costly to the patient and to the payers. But until we know that you can stop the treatment, I think if I were the patient, I'd want to keep getting it, unless I'm getting terrible side effects.

Hi. First of all, Dr. Elgeman, fascinating talk. Thank you very much. I didn't want to lose the chance to ask you about—

Can you speak louder?

Ah, sorry. First of all, fascinating talk, Dr. Engelmann. And my question, I didn't want to lose the chance to ask you about another approach that is an antibodies strategy, but different from that that you just mentioned about monoclonal antibodies. My question is, what's your opinion? What do you think about that? Approach with natural antibodies, those that are generated by B cells by the type 1 in the marginal zone in the spleen. I am curious to know because I was working for 1 year in a therapy in mice based on infusing the mice carbohydrate antigen in a way to develop these natural antibodies. In these mice. And by doing that, try to decrease the growth of certain kind of adenocarcinomas which have in their surface carbohydrates. And when those cells transform into more cells, those carbohydrates lose the sugar chains in the backbone of those antigens in the surface.

I would say that when we did the dendritic cell— I'm going to go backwards. When we did this dendritic cell work 25 years ago, it was an exciting time. But I would say the community or the scientists were saying, eh, that's interesting, but OK, maybe. Now that lit a fire. And it led to many labs and many biotech companies and many academics doing more work on immunotherapy. That was the first demonstration that you could actually use what I call an active immunotherapeutic approach in a patient and get their immune system to attack the tumor. So that led— and now we're in this geometric growth phase where every conceivable way that can manipulate the immune system is being tried. We're trying our approach. There are papers or publications and announcements. You all hear the news almost every day with a new advance or a new excitement about an approach that seems to work. The reason that the checkpoint blockade approach is becoming so important is because it works in more than one type of cancer. It's one size fits all. You can use the same antibody to treat to treat lung cancer as you can to treat kidney cancer. And they're generally well tolerated aside from this issue of autoimmunity. So as a result, that— the checkpoint approach is viewed as one component of an ultimate combination. My view, as I said in the last slide, is that eventually we will have combination therapy. Maybe it'll be 3, maybe it'll be 5, maybe it'll be 6 different components. The critical it's critical that each component be well tolerated so that if you add them all together, you're going to get more toxicity, but presumably, hopefully, I would say a tolerable toxicity. I mean, the Merck people— I hope I'm not saying anything out of school— but I know that from Merck's perspective, and I'm not a Merck spokesman and I don't get paid by Merck, they would say that with their anti-PD-1 antibody, They're looking for things to combine with it that are nontoxic. They would say, well, Bristol-Myers is trying combinations of anti-CTLA-4 and anti-PD-1. Too much toxicity. Too much toxicity. But my point here is that many approaches are being tried, including the one that you described. I can list 30 of them. I teach a course in tumor immunology and tumor immunotherapy. And it's a 12-week course. We meet for at least 5 or 6 hours every week, and we still don't cover all the new approaches. So these are just examples, but they're the most prominent examples that we have today. So I don't want to be negative about any approach, but it has to be tested thoroughly. OK, we're past 8:30. Thank you very much, Dr. Ankerberg.