Targeting The Genetic Basis of Kidney Cancer - A metabolic disease - W. marston linehan

Marston Linehan: Can you all hear me okay? Okay. Thank you, Dr. Passamani, appreciate the kind words and the kind invitation. As Dr. Passamani mentioned, I work across the street at the NIH. I'm in the National Cancer Institute, and it's nice to -- he asked me before this if I had -- if I'd been to Suburban before. I said, "Well, of course. I live in Bethesda. I've been here many times," I was thinking, when he asked me that question. We have two daughters, and when they were playing soccer, you know, they were always doing something. I'd be calling Dr. Goldstein about something, or somebody about something, somebody's ankle or, you know, God knows what. And I brought those kids so often to ER, I was worried they were going to start calling the police or something, you know, about some sort of abusive parent. I had all these ankle injuries and all that from soccer. So I have a lot of -- a lot of -- I used to think I owned a wing of this hospital after I brought my kids here so many times. So I have a lot of good friends over here, and it's great to see. So what I'm going to talk about -- Dr. Passamani mentioned, I'm a urologic surgeon. And so -- and I focus on urologic cancers. So when you think about urologic, you think about cancer in general, urologic cancers make up nearly 25 percent of urologic cancers. If you look at kidney, prostate, bladder, ***, ***, you got 24 percent -- 23 percent of human deaths in this country, in this hemisphere, and -- incidents -- and about 10 percent of deaths. So what I'm going to focus on primarily is kidney cancer. Now kidney cancer affects about nearly 300,000 patients worldwide, about 100,000 deaths. In the U.S., 65,000 will die of this disease this year -- 65,000 will be diagnosed this year, and 13,000 will die of this disease. And there's about 200,000 patients alive with kidney cancer in the U.S. Now, as Dr. Passamani mentioned, I'm a urologic surgeon. I'm not a medical oncologist. I'm not a molecular biologist. But when I came here, and I came to NIH, in the early '80s, I trained, as he mentioned, Duke Hospital in Durham, North Carolina. And so if a patient -- and this hasn't changed that much between then and now, a little bit, but -- so when we started, a patient comes to someone like me, a urologic surgeon, with a localized kidney cancer, we take that tumor out surgically, and we can give those patients a 95 percent five- or 10-year survival. We don't use the C word, the cure word, but, basically, and essentially, I mean, the situation with this disease does not return in their lifetime, if you want to say that. However, if patients come to us with advanced disease, 82 percent of those patients will die within 24 months. So it's a very lethal disease. Among all urologic cancers, kidney cancer is the one that we say costs the most -- we used to say "man years." We now say "person years," in other words, causes the most years loss of life because you get a lot of kidney cancers earlier in life, and it's such a lethal disease. So if you remember just one thing from this talk, it's really this slide, that kidney cancer is not kidney cancer. Kidney cancer is a number of different types of cancers. When we started working on kidney cancer, it was a single disease, in the early '80s, and we treated them all the same way surgically. We gave them all the same drugs, none of which worked. And we now know that kidney cancer is not kidney cancer. No way. It's a number of different types of cancer that just happen to occur in the kidney. They have different histologies, different clinical course, respond very differently to therapy. And as I'll show you -- and I think is the purpose of this lecture series -- these are caused by different genes. So how do you -- my God, you're a urologist. You're talking about genes. What, have you lost your mind? What -- I said, "Well, lot of people are dying of this disease. I can't think of a better way to go than try and understand the basic cause of these cancers." And so when we started, actually, no one had found a human cancer gene at that point. And people said to me, [laughs] "Son, you're over your head. I mean, what are you talking about?" People in my field said, "Look, you're losing it here. No one's found a cancer gene. You think you can find your" -- you know. I said, "Well, I -- you know, I've got some time here." It's early in my career, as we say. And I don't -- I can't think of a better way to treat these people. You know, at that point, at my first slides, you know, we would show, you know, 300 drugs that have been given for this disease, none of which worked. So we then established a program at the NCI to study -- we decided to use -- to study families in which kidney cancer runs. So we could hopefully use the power of genetics to identify cancer genes. So what I'm going to basically just walk through with you is our journey over the past 28, 29 years in identifying these cancer genes. And as I said, I'm a urologic surgeon. What I -- in our laboratory, we have, I don't know, 700 mice that have knockout genes. And we have all sorts of cell lines growing in a lab and all that. But our main model -- our model is the human model. We study the human model of cancer. That's it. Any progress we've made has been because of that. And so I'm going to kind of walk you through how -- where we have gone, and sort of where we are. So the first model of inherited kidney cancer we'd studied was one that had been previously discovered -- described. This was described in 1887. It's called Von Hippel-Lindau, or VHL. Von Hippel-Lindau is a hereditary cancer syndrome in which patients are at risk for the development of tumors in multiple organs, including, of course, the kidneys, which is why we got into this. So these patients are at risk to develop bilateral multifocal kidney cancers. They also get cysts in their kidneys. And the type kidney cancer they get -- in science, we don't say "always" often, but this is an always. These are always clear cell kidney cancer. We've looked at thousands of these, obviously, at this point. And they're always clear cells -- certain types. So that's the most common type of kidney cancer. Now we set out to try and identify the hereditary gene. The idea was -- because this was before there was any Human Genome Project. Francis Collins was up in Michigan at this time. There was no genome program at the time. And it was really a bad -- I mean, it was really tough. And so the best strategy we could think of was to study these families to use the power of genetics to try and trace a cancer gene through a family. So these patients with VHL -- so -- but we didn't know cores. If the hereditary gene -- what we were looking for was the gene that caused non-hereditary kidney cancer. Of course we're interested in hereditary, we're interested in any patient with [unintelligible] cancer, obviously. But far and -- you know, huge number of patients with non-inherited sporadic kidney cancer. And we were looking for the clear cell get the most common type of kidney cancer. That's 75 percent of kidney cancers, what you call clear cell. So we wanted to go after that gene. And we had spent six years working on sporadic kidney cancer, and then run up against a wall. So then we switched and we went to the -- study hereditary kidney cancers. But we didn't know if -- of course -- if the gene that caused the hereditary type of kidney cancer would be the same gene that causes sporadic. That's not always the case. If you look at BRCA-1, BRCA-2, the hereditary breast cancer gene, that's not necessarily the gene that causes breast cancer in a lady with breast cancer, who doesn't run in the family. So, you know, you just don't know. But we figured, well, we'll move and see what happens. At least it's good science if nothing else happens. So -- but this was a remarkable type of kidney cancer. So these people get up to 600 tumors per kidney. So managing these people surgically [laughs], you got to rethink things. We're not curing these people with surgery, obviously, because if they have 600 tumors, the only way to cure them is take out their kidney, take out both kidneys, put them on dialysis. And that is what I did with the first patient I had. Patient came with bilateral multi-focal kidney cancer. The conventional way to manage that was to do an nephrectomy. This was in the early '80s. I saw a patient from up north, a bilateral multi-focal clear cell kidney cancer VHL, talked to all sorts of people -- people in my field -- read everything, talked to everybody at NIH. The way you handled cancer is you take those kidneys out. You can't leave cancer in there. I did that. And I'll never forget it. I'll never forget when that guy left NIH, I took him downstairs personally to the front door and put him in a taxicab. And I sent that guy home to be on dialysis. And I said, "I'm never doing that again as long as I live." So we developed an approach -- the management of these people -- that involves partial nephrectomies. Now, people used to think clear cell kidney cancer and VHL wasn't real kidney cancer. You can look at my journal -- our journals in the early '80s, and you see journals with titles like "Clear Cell VHL Kidney Cancer's Not Real Kidney Cancer." My field didn't understand the term "lead time bias." They didn't understand that if you detect a cancer early, it's not necessarily going to be the spreadable moment. You have to give it time before it's going to spread, like prostate cancer. A patient to came to us with a large tumor, about an eight-centimeter tumor, and she already had a pulmonary metastasis. We've managed many of these patients over the years. We have the world's largest experience with this, and we have about 900 patients now from about 400 families. And we've developed an approach, which I lost a lot of sleep over for many years -- and that is not taking out their kidneys, doing partial nephrectomies. Today, it's rather obvious. But when we started this in the middle 1980s, I woke up many, many mornings, many mornings at 3:00 a.m., knowing that was the day. So what we would do is we'd do partial nephrectomies for tumors smaller than three centimeter. When we operate, we operate. I mean, we've taken out as many as 74 tumors from a single kidney. And I saw a lady last week where we recently did the ninth operation on her kidney. That's going all the way back to '82 -- '83. Anyway, so we do active surveillance. We watch these people. At the time, people in my field -- tomatoes, rocks -- I can't remember what else they threw at me. That's not the way you handle cancer. And I said, "God, I'm here by myself." At NIH, we now had more people, but I'm going, "Gee, it's going to be over," you know. I'd wake up at 3:00 in the morning. I'd come and see patients that day in clinic, and I would say -- I'd have nightmares that that would be the day I'd see -- I can still remember where I was in bed when I'd wake up, and I remembered saying, you know, "That's going to be the day I'm going to see people with pulmonary metastasis." And people are going to say, "You idiot." As of September 6, 2013, we have not had one, not one single patient develop metastasis when managed in this fashion. Now have we had patients with advanced disease, patients with VHL develop metastatic disease? Yeah, we sure have. We had a bunch. But when the tumors are between three and four centimeters, our metastasis rate is low. It's about 4 percent. Between four and five centimeters, they get to be that size, it's about 20 percent. Between -- now, when you start hitting five centimeters and above, it's about 50 percent. So I'll come back to what that might mean in a minute. But anyway, so that's how we manage these people. It's called NINCI three-centimeter rule. People refer to it as that. And we apply this same approach to managing VHL, managing the next disease -- next type of hereditary kidney cancer, I'd say [spelled phonetically] hereditary papillary kidney cancer. And the next one, one called Birt-Hogg-Dube, but we do not use this approach for the final two I'll show you, which are different types of kidney cancer. So the bottom line here is, is all the gene. We've managed these people based on the gene, and that's it. Now, most of our surgeries now, we do robotic. I wanted to show you this one just to give you a little idea. So here we're doing a robotic partial nephrectomy. This is a kidney cancer here, a patient with VHL. See the tumor here? So we're coming around, and you're saying, "Wait a minute. I can see that that's tumor." That's normal kidney. "You're not getting much of a margin there, son, are you?" No, we're not. We do what we call a nucleation on these. Now, in the old days -- the old days being the '80s -- I used to do wider margins on these. But after a while I realized, "My God, I do wide margin, I got no kidney left." Now -- so we now understand a lot about the biochemistry, and I'll show you a little bit of that in different things of these. But this approach, we've not had a problem with in a single patient. So we go -- we enucleate these tumors. And, as I mentioned, we've taken out up to 75, I think, in a single patient. Anything like that, we'd do open. Although robotically, we're getting so good -- our team's getting so good now -- one of the fellows did one, I think, two and a half weeks ago, we took out 35 tumors robotically. So as our skill set increases, we're getting better and better and better at this. And, of course, it's much better for the patients if you can go robotically than if you do a big open operation, which, of course, is what I did forever. So we're right -- as you know, we're right across the street. And this is, of course, the hospital. So we want to find this gene. So we brought patients in from mostly U.S., but really around the world. We'll bring anybody. And we looked at -- we saw families. So we would -- how do you do this? So we would bring them in, screen them. We have a multidisciplinary team that screens these people. These people get -- also get cerebellar and spinal hemangioblastomas, CNS tumors. We work with our neurosurgeons -- sensational. Our retinal -- they get retinal tumors. They get retinal angiomas. They get pancreatic endocrine tumors. They get pheochromocytomas -- of course, kidney cancers, Epididymal cystadenomas. And they get a tumor in the meso salpanks [spelled phonetically] of the ovary, it's a benign tumor. We work -- we manage them with Dr. Pam Stratton, whose actually here. So we brought these people in, and we determined who was affected and who wasn't in these families. So that, then, gave us the power of genetics to trace these genes through these families. Now this took us 10 years to do. We evaluated DNA from 4,317 patients. And we localized this gene, this gene was localized here, to the -- this is chromosome 3. This is what you call the long arm. This is the short arm, excuse me. This is the long arm. And this is where we localized it to. Then we mapped and mapped and mapped. And that's what this shows, our physical mapping. And we identified some candidate cDNAs, some candidate pieces of DNA. And then we evaluated them. And this one here was the seventh cDNA, piece of DNA, that we looked at and sequenced. And we're looking for a change in that sequence -- whatever this is -- that segregates with the disease, and that, of course, is what we found. This was -- this -- we called -- there was the human VHL gene. This was the sixth human cancer gene when we found this. And we're looking for changes in that gene, mutations. I don't actually use the word "mutation." What do you mean you don't use the word mutation? Well, I had a lady who said to me once, she said, "Look, don't call this a mutation." She said, "This is me. This is my family. We have VHL. My mom had this. My aunt Susie had it. My cousin Fred. My daughter Sarah, and my other daughter Emily, has this. Don't call us a mutation." I said, "Right. We're not." So I say "alteration." So we look for alteration of these genes that segregate with the disease, and that's what we found. We detected these alterations in 365, actually, and now we're up to 400 families. So we're basically 100 percent. Now we wanted to know was this the gene we looked for for so long, for 10 years? Was this the gene for the non-inherited form of sporadic kidney cancer? So we took tumors from patients with non-inherited, non-hereditary kidney cancer, and we look for alteration of this gene, and that's what we find. We find alteration of the VHL gene in 95 percent of tumors from patients with clear cell kidney cancer, of the VHL gene pathway. The VHL gene itself about 91 percent. So this is the clear cell kidney cancer gene. We find these alterations in clear cell kidney cancer, but not in the other types of kidney cancer. So the first evidence that there is a genetic -- there's genetic differences between these different types of kidney cancer. So what kind of cancer gene was this? When we found it, it was a completely novel gene. We had no idea. Was it an oncogene, which is, you all know, would be a one-hit gene that drives cancer. Or is this a loss-of-function tumor suppressor gene? So we each have two copies of each gene: one from our mother, one from our father. These patients inherit an alteration of one copy. And what we showed was, in the other copy, they lose that. They delete that. So this is what you call a classic two-hit tumor suppressor gene. This concept of a tumor suppressor gene was developed by the great Al Knutson, who's been our mentor and advisor almost 30 years. Al is now 90, and we went up and did a big symposium for him. But this is a two-hit loss-of-function tumor suppressor gene. Now the other type cancer gene you might think about -- and everyone would associate that with Al Knutson -- the other type of gene I'll show you next is a dominant gene. It's called an oncogene. That's where -- this is a loss-of-function. An oncogene is a gain-of-function. So you have one mutation there. This is a two-hit gene. This is a one-hit gene, the next one. I'll show you that in a second. So, well, that looks good, Marston. That's great. Yep. You got that. Runs in people's families. You lose the second copy of that gene. It gets deleted. But what other -- I mean, you say, "That sounds good," but what evidence do you have this is a loss-of-function gene? Okay, so if we take a cell line from a patient with kidney cancer that has a VHL gene alteration, we put that in the lab gorton [spelled phonetically] culture, then we put it in a mouse -- an immunocompromised [spelled phonetically] mouse -- it goes up and makes a tumor. So that's what that looks like. And if you leave it there, it'll -- which we don't, of course -- if you leave it there, it'll do just what it does in our patients. It'll spread and kill that mouse. Now, if you make one single change, though, in those cells, only one, you take the cells and you put in those cells that one gene, just one. And then you grow the cells, put them back in a mouse, you get little or no tumor. So that is a loss-of-function tumor suppressor gene. So we said, "Okay, we're on the right path with this gene." Now, when we found this gene -- 1990 -- gosh, it's been 20 years now. We found this gene in 1993. We had no idea what this was. It was a completely novel gene. People would say to me, "Marston, gee, you guys are really the world's experts now in chromosome 3. When the Human Genome Project came, you and your colleagues kind of became the go-to people for that chromosome. You going to work on lung cancer now or something?" No, what are you talking about? No way. We're going to work on kidney cancer. I said, "We're going to work on this gene." They said, "Well, you're not a biochemist." I said, "Well, we'll learn it." And so over the years, what we have learned is the following. We had no idea how this gene worked. So what we know now is that the VHL gene makes this protein, VHL protein. And this protein has two domains. One domain binds these partners here. We call this the VHL complex. And this is what you call an E3 ubiquitin ligase. This is a very conserved mechanism throughout biology. You can -- we learned about this because we just, of course, discovered this, and then we identified that this protein that I just showed you, this clear cell kidney cancer gene, binds these two proteins here -- we published that in Science -- but it didn't help us. We still couldn't figure out how this thing was working until we found this gene here cullin 2. So once we found that, then we had others -- we're then able to put that together, going back and actually studying yeast, if you can believe that, that this complex is what you call an E3 ubiquitin ligase. This is a normal gene, of course, becomes a cancer gene when it becomes altered. And that this complex targets this family of proteins called hypoxia inducible factor, or HIF, for degradation, for ubiquitin mediated degradation. It puts a ubiquitin grouping on here, and that then degrades. It's just like it takes these proteins to the trash can, basically. It's a normal process in itself, okay? Except this is -- I got to learn about this because this is critical to our cancer. Now the way this works is it's an oxygen sensor. What's that mean? So when there is a normal amount of oxygen in the cell, oxygen in the cell, this complex targets HIF and degrades it. Now, on the other hand, when the cell is short of breath, hypoxia, it doesn't have enough oxygen, this complex cannot target HIF and degrade it, and it accumulates. How's this work? What's that? Well, HIF is a transcription factor. So it transcribes things. It turns things on, you might argue. So what it turns on is things like vascular endothelial growth factor. That makes more blood vessels. It turns on erythropoietin that makes more red blood cells. It increases something called platelet-derived growth factor, which we'll come back to in a second, PDGF, which tells the cell next door to, you know, grow, and help us grow. So, basically, you could argue that if you had a 4-year-old child and you took out his right kidney, his left kidney hypertrophies. How does that happen? Well, you could argue a very good mechanism for that, is the cell becomes hypoxic -- says, "We got to grow. We need more blood supply. We need more blood vessels." And then when it gets a bunch of blood to bring oxygens come, and this, that, and the other, then it reaches equilibrium in oxygen and becomes normoxic, as we say, and then it kicks back into degrade HIF. That's a pretty -- it's just a pretty simple system once you understand it. However, in our people with kidney cancer, what happens is they get a change in the gene, here in the alpha domain, which binds this complex, or the beta domain, which is target specificity, which targets HIF and degrades it. So what happens is, in our patients with kidney cancer, you get this mutation clear cell. You get this mutation. And even in normal oxygen, basically the cell thinks it's short of breath. That's essentially it. Thinks it's short of breath. "I need to grow. I need more oxygen." So as a urologic surgeon, my God, I can understand this. My tumors are very vascular. This is saying, "We got to grow. Got to make more vascularity." My tumors are very vascular. My tumors make a lot of erythropoietin, believe it or not. My tumors stimulate the cells next door. So this, I can understand. So that simple principle, then, is the basis for therapy. So you could target this pathway. This is VHL. This is HIF. These are the downstream things I mentioned. This is VEGF, platelet drive growth factor. This is another one called transforming growth factor alpha. Whatever. Anyway, so this is -- so you could target with an antibody to target VEGF. That makes sense. Or people develop what are called tyrosine kinase inhibitors, TKIs, a tyrosine kinase inhibitor. I'll show you in a minute where the tyrosine kinase domain -- what that means. It's not so mysterious, but it's called a TKI. So these are drugs that target these pathways. So you could say, "Well, we understand the pathway. Let's target this pathway in kidney cancer." So we're going to fast-forward another 10 years to today. Well, anyway, another 10 year -- well, to today, anyway. So as of today, the FDA has approved seven targeted drugs against our first cancer gene pathway. Bevacizumab, an antibody -- Avastin, people call it. Temsirolimus, Everolimus. Sunitinib, Sorafenib. Pazopanib, Axitinib. Now you could argue -- well, Sunitinib is right now considered the best first-line drug. You get about a 25 percent partial response with that, and about a 20-month disease-free progression. The most recent drug goes -- you read New England Journal, article came out just recently, within the past month, comparing this drug called Pazopanib, and this drug, Sunitinib. And this drug looks equal to this. In medical oncology lingo, you'd say it's non-inferior, but it seems to have less toxicity, so this will probably move into first line. So what do I think about this? Well, it makes my knees weak. I mean, it's -- wouldn't say humbling. It's unbelievable that targeting this pathway, we're seeing tumors get smaller in people, and we're extending people's life expectancy. And what you can do, then, is you can sequence these drugs. You can start with this, and go to one of these other ones or two of these other ones, and maybe go back to them. So we're working the field -- I say "we" -- the field is working to make this chronic disease. Why aren't we doing better? So that's clear-cell kidney cancer. Why aren't we doing better, though? We understand this gene. Why can't we cure this disease? Well, isn't it true that most of these people eventually fail and die of this disease -- die of this -- yes, it is. Okay. So we just are -- so the Human Genome Project, the -- well, the NHGRI, National Human Genome Institute, and the National Cancer Institute collaborate on a project that's called the Cancer Genome Atlas, and we just published in Nature a few months ago The Cancer Genome Atlas of kidney cancer. Five hundred tumors. I had the good fortune to work with the 300 smartest people I've ever met in my entire life, and did whole genome -- you know, sequencing whole exome, whole genome, all sorts of stuff, and looking at RNA and all sorts of things, and looking at a whole bunch of kidney -- 500 kidney cancers. So the VHL gene has mutated, you know, here in about 70 percent. In many other studies we've done, other people have done, it's really higher than actually that, showed about 90 percent -- 95 percent, if you look at the VHL pathway. So this is the VHL gene, but these other genes called PBRM1, SETD2, BAP1, they're also mutated, and this gene here, BAP1, appears to be critical in progression. So this gives us new targets. In other words, why aren't we doing better? Now, another thing, just to -- for those of you who want to philosophize about cancer, this was -- this is a paper in New England Journal. I'm sure many of you saw it. It's like in May of '12, and it was by an incredibly gifted group in England at the Wellcome Trust. Andy Fatrayal [spelled phonetically] is basically head of this group, and we know these guys really well. And so they looked at a big kidney cancer. Now, how long does it take for a kidney cancer to grow, to develop? Well, if you'd asked me this about 10 years ago, I would've said, okay, so let's say you have a VHL -- a clear cell kidney cancer. VHL and non-hereditary are the same. You had a clear cell kidney cancer that's two centimeters in size. How long do you think it'd take to get there -- to get two centimeters? Well, I would have said, "I'm pretty experienced in this area. I'd say three to five years." Not even close. We've done 28,000 tumor measurements. We look at all these growth rates, we follow all these patients: 25.2 years to go from zero to two. Now, so if you look at this tumor that was nine centimeters, how long has that boy been there? About 30 years, okay? So it causes us to rethink everything we think, at least about my cancers. They're very -- it's like prostate cancer. It's just very slow-growing. But these people looked at multiple parts of this nine-centimeter tumor, and they looked at a couple metastases, and they sequenced all the genes. And what they found was not surprising to me, but it was to a lot of people. What they found was big time heterogeneity. In other words, the gene mutated here were different than the ones here and here and here and here and here and here and here. The genes mutated here were pretty similar to here, to the metastases, consistent with they came from one place. Makes sense. And if you compared the genomic pattern of mutations here and here, they were identical to here, supporting that this went there. My God. So how do you -- [laughs] -- how do you think about therapy, then, if you're going to target cancer genes? My God. Should we be sampling a whole bunch of -- this is called -- people call this now "precision medicine," where you target a specific gene. And the -- one of the leaders at our place said, "Well, Marston, you've been doing precision medicine for 30 years." I said, "Well, maybe." But, anyway, so this is now a big thrust of combining genomics, Human Genome Project, genetics, with therapeutics in cancer, and something we're making a huge push on. But conceptually, how would you -- how do you think about this? Should we take a tumor and do a whole bunch of sequencing on a bunch of parts, or should we take a metastasis and do a bunch of sequencing, and find, maybe ,driver -- you might call those driver genes. It's an -- it's a question no one knows the answer to. I'll tell you what I think. But, so, these people when they publish this -- I -- the guy who did this worked -- it's a friend of mine, and I -- he spoke at one of my -- our meeting I sort of help run. And I said to him after his talk, I said, "Andy, what does this mean for therapy?" Well, he said, "You are asking me?" He said, "I'm a genomics guy." I said, "Well, okay." I said, "I'll tell you what I think." But in that -- did you read that? If you read carefully that paper in New England Journal, what they said was they think what makes the most sense is targeting what's called the truncal gene, in other words, the VHL gene, which is what we have always thought as well. So, you know, so we will see. I guess what I'm trying to say is there are complexities in this, and, you know, we have miles to go before we sleep. Okay, so that's the VHL gene. Now the next genes I'm going to show you -- and that really came from our study of patients with non-inherited kidney cancers. I'm going to show you some other types of kidney cancer now, and I'm also going to show you one fundamental theme that runs through all cancers, which, to us, is really, we feel, the key to effective therapy. And I'm going to tell you what I'm going to tell you, and that is, I'm going to tell you that all these cancers are fundamentally metabolic. They're fundamentally metabolic. So, everything -- as I mentioned, we study patients. Everything we've learned, we've learned from patients. So this was a little girl -- young woman I saw in April of '87. She came from Ohio with her mom, her worried mom, and I took out that left kidney. Big tumor, T3A, 11 centimeters. I got it all out, I thought, but she went on to die in January of 1988, that little girl, that young woman. Cheerleader. And I took the pathology to Maria Marino, our wonderful pathologist, and I said, "Maria, what kind of kidney cancer is that?" She said, "Marston, it's papillary." Yeah, okay. It's not clear, so, no, it's papillary. All right. Another little -- another young woman I saw -- this one was 18, came up from Charlottesville in the spring of '89. I took out that left kidney -- came up with her mom, too. I took out that left kidney. She still went on to die in February of 1990. Her mom died 14 months after that of metastatic kidney cancer, and it took me -- took us 18 years to figure out what she had. I took the path to Maria, and I said, "Maria, what is this?" She said, "Marston, it's papillary kidney cancer." All right. Third patient, patient we saw in April -- March of '92. It's a family, another family. This guy comes up -- this guy's 71. He's got multiple tumors in his kidney. Okay. This is his sister, 68, she has multiple tumors in her kidney. This is his son, 42, there's multiple tumors in his kidney. I got the pathology, I took it to Maria, I said, "Maria, what is this?" She said, "Marston, it's papillary kidney cancer." All right. What I'm going to show you is, yes, it's papillary kidney cancer. Each one of these is a different disease. It's a different gene, big time different clinical course, very different approaches to therapy. All right, I'll start with the third one first. So, another form of inherited kidney cancer. It's -- this hadn't been described before. We called it hereditary papillary renal carcinoma. Each of these individuals in this family have papillary kidney cancer. This is what you see, bilateral multifocal. This is the first patient I saw who -- that was the son, the 42-year-old. All right. Remember I told you about the 21-year-old? I told you about the 18-year-old. My God, I was worried about this spreading and killing this patient. So I took out this patient's left kidney right there. That's the last time we've done that, certainly for small tumors in this, and I'll show you why. So this is multiple tumors. This is Type 1 papillary kidney cancer. These people get up to 1,100 tumors per kidney. We manage these the same way we manage VHL. For this type of papillary kidney cancer, we know the gene for it. We watch them. We do this all the time. We watch them, we do active surveillance, I guess you might say, until the largest tumor reaches three centimeters. Then we recommend surgical intervention. When we operate, we operate. We clean that kidney out. But until that time, we do active surveillance. Now, the first patient I saw in this was '92, so it's been 21 years. Have we had people develop metastatic disease? Yes we have. But people with larger tumors, we've never yet had one of these develop metastatic disease when managed in this fashion, i.e. active surveillance until the largest tumor is three. So we brought people in, again, across the street -- NIH. We did genetic linkage analysis, localized this gene to chromosome 7. This area here, we mapped here -- it's a really tough area to map -- and identified the hereditary papillary renal carcinoma gene as this, what we call Met, M-E-T. Met. Okay, what is that? Met is the cell surface receptor for growth factor, ligand you want to call it, growth factor -- called hepatocyte growth factor, HGA -- HGF. So HGF targets here, activates this, and causes cellular growth. When you don't need to grow, it stops, and everything's cool unless you have a mutation. This is a single alteration, okay? Single. So this drive cancers to cell -- to grow. So this is what you call an oncogene -- actually, a proto-oncogene. It becomes an oncogene with the alteration, but this is a proto-oncogene, okay? We talked about tumor suppressor genes, oncogene. All right. Well, this -- these are the real mutations we see in our families. We've detected -- this is a rare disease. We've seen 22 families -- detected mutations in all of them. Now, this is -- we see this in sometimes early onset -- 38-year-old, 27-year-old, and we've got a 19-year-old in here. So this, targeting a loss-of-function gene in cancer therapy is kind of a -- kind of tough. I mean, it's -- targeting a gain of function you could argue should just be tinker toys. I mean, this should just be screening. I mean, you know? This is just chemistry. I mean, you ought to be able to do this. So I'm not saying that it's true, but conceptually, it should be. So -- oh, yeah, I promised you I'd -- remember a minute ago I said a tyrosine kinase inhibitor. So this is the gene here. This is the receptor. This is the extracellular domain. This is the trans-membrane domain. And this area here -- this, you might argue, is the engine room. This is where phosphorylation, which is how you affect other proteins, kind of happens. This is called the tyrosine kinase domain. Tyrosine kinase domain. So, you know, if you want to sound smart with other people, you can see, "Oh, these are TKIs." You know, what's a TKI? Tyrosine kinase inhibitor. Okay? So many of the huge, you know, billion, billion, billion dollar drugs are called TKIs, right? That's tyrosine kinase inhibitor. So here I've shown you this gene causes this cancer, and so we want to inhibit this tyrosine kinase domain. So we'd like to use a tyrosine kinase inhibitor. So we did first trial with this drug called Foretinib, which was a dual kinase inhibitor, a VEGFR, and Met. It had some toxicity we didn't like -- the VEGFR side -- and if you all are medical oncologists, of course you know this stuff for breakfast and far better than I do, but you get things like hypertension, malaise, cutaneous, diarrhea, you know, a number of things can happen with these. A lot of that is due to the VEGF pathway inhibition. The Met pathway inhibition, we think, is a whole lot less toxic. We'll see. So, anyway, so where are we with this? So we know the gene, we have a drug that targets that pathway. So this patient here -- remember the first guy that I mentioned. I took out his left kidney. So that's this guy here. So I took out that left kidney in January of 1992. Then in May of 1992, we did a right partial nephrectomy and took out 12 tumors from that remaining solitary right kidney. Guy's doing fine, goes back and forth to work, little league baseball games, this, that, and the other. We're not carrying him with this partial nephrectomy. He's got this many tumors. So developed -- continues to develop tumors, there was a cord [spelled phonetically]. And in 2000, we took out an additional 59 tumors from that remaining right kidney. Okay, guy's doing fine. His renal function's a little off, his creatinine's about 14, EGFR's about 55, but he's doing fine. You know, you sit next to him on a bus, the guy looks great. You don't -- he's looking fine. However, continues to develop tumor, and tumor now -- largest tumor now was 3.4 centimeters. Well, we don't like that. I don't want him to die of metastatic disease like his father did. So we put him on drug, and after 49 cycles of this drug, all these other tumors became undetectable, and this one almost undetectable. We've had dramatic response in the lungs. People in medical oncology right now, people in my field will tell you there is no drug that works in papillary kidney cancer. This is papillary kidney cancer. Now, this is what you call a waterfall plot. So this shows the decrease of these tumors. Now a lot of [unintelligible] experimental thing, works first time, and all that sort of stuff. A lot of these people I had -- we had heavily operated on. They're heavily pretreated with surgery. Many had marginal renal function, and many had to come off drug for all sorts of nickel dime reasons. But every single tumor, every single tumor, got smaller during therapy. So this proof of principle that targeting this cancer gene could have an effect on these cancers. So are we home? No, we're not home, but we're encouraged about the progress of the work. Our next trials are going to involve specific TKIs, tyrosine kinase inhibitors, that just hit Met. And then we have a number of other things that in a lab look sensational against this pathway targeting Met, and we're going to be sequencing those trials and we will hope. But I'm hopeful. I'm optimistic. Now how about this little girl? This is the first one I showed you, this 21-year-old cheerleader from Ohio who had papillary kidney cancer. So we took that out, I put that -- we put that in culture, and we grew it in a lab. And we showed a funny chromosomal pattern. We showed that part of the first chromosome translocated to the X chromosome. I've got to move. And this translocation involved a genome chromosome one and the X chromosome gene called TFE3. So this is like many leukemias. This is a fusion cancer. So -- and it's very aggressive. We saw a 23-year-old the other day, law student, with this very small tumor, already had local nodes. This you don't do active surveillance on. These spread early. This is called TFE3 kidney cancer. It now makes up about 1.5 percent of all tumors, but 20 to 45 percent of tumors in young people. We now know this is a family. Subsequently, TFEB has been discovered, which is another type of very similar kidney cancer, often in kids, and another one is MITF -- another member of this family, and that gene is mutated, and this can run in families. So I'm going to scroll through this and talk about this one. This is called Birt-Hogg-Dube. This is a hereditary cancer syndrome where people get skin bumps. They get fibrofolliculoma, benign hair follicle tumors. Runs in families. We showed these people also get kidney cancers. They get different types of kidney cancer. They can get clear cell, they can get something called hybrid oncocitic, they can get chromophobe kidney cancer. They get up to 3,000 tumors per kidney. We manage these the same way: three centimeters, active surveillance. So we brought these families in, searched for this gene, used the skin marker, the fibrofolliculomas, the benign skin lesions, as a marker to trace the gene in the family, identified it on chromosome 17 in this region right here. This is the gene. We call it FLCN. This is the BHD gene. We've detected mutation in this now in 97 percent of these families. These people also -- this is pretty common. You'll see these, those of you who are urologic surgeons, if you are. So they get multiple cysts in the lungs. Thirty percent of these people get pneumothorax. We wanted to know what kind of gene this is. I'm going to screen through this because we're running short of time here, and basically to show you that when this gene is mutated, it activates two important cancer pathways called mTORC1 and mTORC2. Okay? How's that help me? All right, because there are drugs that target these pathways. So in this mouse model here, we knocked out this gene in mice, the same human kidney cancer gene in the mouse -- in the kidney of a mouse, and what we get there is we get a big kidney, a cyst. They're starting to form little tumors, but they die of kidney failure in 30 days before they get kidney cancer, they get full-blown kidney cancer. So I said, "All right. How about if we treat these guys?" So these guys died about 30 days of kidney failure. So I treated these guys with this drug -- actually, Rapamycin, which targets that pathway I showed you, and we saw a dramatic effect. We doubled their life expectancy. So that's targeting this part of the pathway, and we're now gearing up to do clinical trials in humans hitting both this part and this part. Now, in closing, I'll show you this last patient. This was this little 18-year-old, came up from Charlottesville with her mom. You know, we -- you never -- I know you are the same. You never forget a patient. You know, this is -- it took us 18 years to figure out what this was in her. In '95, we described another type of hereditary kidney cancer that was redescribed and renamed in '99, and now goes by the term hereditary leiomyomatosis renal cell cancer. This people get cutaneous leiomyomas. They get skin bumps. They're little muscle tumors, and they get uterine leiomyomas -- some people call these fibroids -- and they get kidney cancer. Runs in families, autosomal dominant, and these are traces in these families. These are these, quote, "skin bumps," and they can be very sort of -- these are benign, although many pathologists get them confused from leiomyosarcoma because the pathology's a little fussy. But -- and what these are, are the following. These are little muscle tumors, and so if this is a hair follicle, there's a muscle underneath it called a rector p-light [spelled phonetically]. So when you go out on a cold morning, and it's cold and you get goosebumps, it's that muscle. That muscle is an energy sensor. It senses that you're short of energy, as it were -- energy in the cell called ATP, and it contracts. All right? That's the same muscle a porcupine uses to fire his quills. Now, these can be, in these patients, really bad. They can be very painful. We've lost one. Or we had one -- actually, it was a family member of one of ours, committed suicide, and we've had another one who's -- we're concerned about. Anyway, they can be -- they can be very symptomatic. Sometimes they're not symptomatic at all. Sometimes you don't have these. But -- and it's particularly -- I've got one now. It's a college student in Boston, and got the disease here. Whenever it's cold, he won't go to class. He also gets embarrassed. I mean, it's -- we're working on it. But anyway, so that's the skin manifestation, and this is a remarkable one. This is -- they get cutaneous leiomyomas, and it's early onset. And we manage these patients with Dr. Pam Stratton, our sensational colleague who's really the world's expert on this -- managing these patients. So when we saw our first patients, and we reported them -- so we look at this. Ninety percent of the women have fibroids. This is, by the way, not that uncommon a disorder. I think you will see this at Suburban Hospital. I know for sure many hospitals -- I can think of many hospitals [unintelligible]. So you'll see these patients. Now, it's early fibroids. In our initial report, 90 percent of the women get these early onset fibroids and are initially is a catastrophic -- can be a catastrophic phenotype for women. Fifty percent of our women in our first report had hysterectomies in their 20s. So we hate this. So now, we and others -- I say "we" -- NIH and others, Dr. Stratton and her colleagues, are doing myomectomies, and I think now she has, I think, four patients who've had babies after that, and we're very proud of that. But this is a very dominant phenotype. Now, we got into it because of this, obviously, the kidney cancer. So this patient here, that 18-year-old I mentioned to you -- I lost her. Her mom died. God, I tried for years to find her. This was in Charlottesville. We found out 18 years later that what happened was her siblings moved in with the father who had a different last name. The parents were split. And between then and when we found them, the brother died, her uncle died, aunt died, uncle died. We found them when we saw this one, who then went on to die, came to us with very advanced disease. This is a very catastrophic phenotype if not diagnosed and treated. If you treat it, you're good. This is a very -- this is the second patient I saw, a 21-year-old came to me with this tumor here, came up from -- Cuban descent from Miami, and I took this out. I know, aggressive cancer, but I'm good. He died 17 months later of metastatic disease. This guy came to us -- 32-year-old whose father had died of metastatic cancer. You can see the skin bumps here. It's really -- we don't understand this yet. [laughs] It's really remarkable. They don't cross the mid-line. They stop right in the mid-line. Boom, look at that. You can see we biopsied this guy, and when we screened him -- oh, he's an asymptomatic guy, doing fine, 32. When we screened him, we found they can get cysts in their kidney, and they can get tumors inside the cysts, and they can get just plain tumors, as it were, and they can get just plain cysts. We go crazy over managing these people. I mean, we agonize over them. So this guy had cyst, but he had tumor inside it. I'll show you that. Doing fine. Very small tumor, half centimeter, one half centimeter. When we took this out, you can see it's a half centimeter, the rest of it was totally cystic. But he already had a big, positive node. These spread early. You do not do active surveillance, and you have to screen them every year. Every year, MRI. These can be -- it's an unusual type, of course, of papillary kidney cancer with these prominent -- our colleagues in -- if we have any in the audience who are pathologists -- would tell us these are prominent nucleoli, and they would describe this as perinucleolar halo. So it's a very sort of characteristic pathologic phenotype. Once you see it, you can make the diagnosis on this. Maria Marino, our wonderful pathologist, she looks at that and she says, "This is it." She's batting 100 percent. So -- but this can be early onset, too. We've got this in kids, 10-year-old came to us with this. Big tumors. We've seen it first time in a 77-year-old. I mean, his first tumor. These people are at risk for multifocal and bilateral. This is Sophie's Choice. What, do you take out the kidney? Horrible cancer. What are you going to do? So how are you going to manage them? So this patient here, 24, comes to us with this. So you get this CT. I don't know. I can't call anything in that. That's a cyst. There anything in there, I'm not -- I don't. Do an MRI. Well, I don't know. Is that just a cyst? Is that volume averaging? I don't know. Come back -- come back in nine months, we'll get a new CT and a new MRI. Got a CT. Nothing, that's nothing. I don't -- just a cyst, right? MRI, boom. We call that a double bump. See that? So we said, "We're operating." When we operate -- remember I showed you how when we did the VHLs, clear cell kidney, we know the gene, know the pathway, know they grow slow. We just did nucleation here. No way. So here, we go wide, big time. Why don't you just do a nephrectomy, Marston? Well, we don't [laughs] -- these people are at risk for multifocal. I don't want to be just whipping out these kidneys. So we do partial, but -- and we've had number of patients come to us, where people just did your usual partial, and it was a disaster. And this lady shows you why. So we went way wide. We went all the way up to the hilum of the kidney. We dissected out the hilum, did very, very wide operation; did really almost a heminephrectomy on that. So, Maria Marino looked at the path, I thought, she was, she said, I thought she was going to throw up. She said, "Marston, you've got to come over and look at this." Said, "You know, your little girl, is it 24-year-old?" I said, "Yeah." I said, "The one with the cyst, and the tumor in the cyst?" She said, "Come over here." I said, "All right." She said, "Look. See this path?" I said, "Yeah." She said, "Here's your tumor. This is the cyst; here's your tumor." Said, "But look at this. This is the normal kidney. It's infiltrated all up into the normal kidney." I said, "Don't tell me my margin's not clean." She said, "No, your margin is fine." She said, "You did a huge margin." [laughs] I said, "Yeah." She said, "You have two centimeters clean here." She said, "But this thing --" she said, "This thing was invading all the way up into the periculum." I said, "Maria, I knew that. I mean, I sensed that, but we couldn't see it. You can't see it on imaging; nobody can. She said, "Well, this is a nightmare." Now, she's fine. This is '08, and we're now five years out. She's disease-free, doing fine. But we've had a number of patients who came in after just partials, just regular partials. Nightmare. Positive margins, disease spread. If you spill these cells, it's bad news bears. So you got to be real careful how you manage these people. So, well, what are you going to do? Just wait until stuff happens? No. We're going to image these people every year. So, here's a lady we saw. Came up from Baltimore. In '03, we saw her. She's germ-line positive for this disease. We told her that imaged every -- we recommended imaging every year. Whatever reason, wasn't done. In '06, she had imaging which was called normal. Then, in -- right before Christmas, right before the holidays in '10, I got a call from guys at Maryland -- University of Maryland. They said, "Marston, we've got one of your over here." She didn't have screening in four years. This is what she had then. We took this out in January, you know, right after the holiday. Ten of 59 nodes positive. We now -- she's now in one of our therapeutic trials, actually doing very well with metastatic cancer. When we first saw her, she had no disease. So we do not do active surveillance. And when we do surgery, we go wide. We don't do any of these robotic. We go wide. I don't want to do -- I want to go open and go wide on these. They also can be multifocal, so we have to be real careful about them. This gene is on chromosome one. This is what the gene is. It's a Krebs cycle enzyme called fumarate hydratase, which takes you from fumarate to malate. I know we've got to -- we've got to stop, so I'm going to -- I'm going to just abbreviate this. To make a long story short, we had mutations in all of these cancers. This is a very aggressive cancer in which the metabolic -- these tumors undergo a metabolic shift. This is the TCA cycle to glycolysis. And make a long story short, by understanding this pathway, we are targeting this with bevacizumab and erlotinib, and we're seeing very -- I lost every one of these patients until just recently, until we started doing this. This we do with Rom Srinivasan, medical oncologist, works with us. And we've seen really dramatic response in patients. This woman here, both her sisters died, father died of metastatic kidney cancer. She came to us with advanced disease after a partial nephrectomy, in which disease was left behind. And it spread all throughout her retroperitoneum abdomen. We put her on this therapeutic approach. In three months, she was a complete response. It's now seven years; we cannot find disease. We don't use the "C" word, the cure word, but we're at seven years. It's actually seven now. So, what I'm -- went through a lot of things, but the fundamental point is kidney cancer is fundamentally not kidney cancer. It's a number of different types of cancers caused by -- with different histologies, different clinical course. I showed you a little bit about different approaches to therapy, how different they are, knowing the gene for those cancers, and that they're caused by different genes. So I want to acknowledge my colleague, Peter Pinto, Adam Metwalli, Piyush Agarwal, Rom Srinivasan, our medical oncologist, world's best urologic oncology fellows, and the entire Urologic Oncology Program, and also the other colleagues in gynecologic surgery, neurosurgery, ENT, ophthalmology, endocrinology, endocrine surgeon -- endocrine surgery, that we have the honor to work with. So, are we home yet? No, we're not. But I'm optimistic. And, again, I'm like you. I'm a physician. We take our patients like everyone does. And everything we've learned, we've learned by studying these patients. That's where everything we know has come from. Every single thing. Are we home? Are we there yet? No. But I'm optimistic, and I'm hoping to live to see it. Thank you very much. [applause] Male Speaker: We have time for a few comments and questions. I have one. Marston, I understand why you're real careful of not having a big margin in clear cells. What's with that tumor that lets you do that? Marston Linehan: Well, it's just that it's -- that -- well, that's a very good question. It's just that it's a slow-growing tumor. It hasn't developed the machinery to invade yet, until you get a certain point in time. Now, you -- it -- there is complexity in life. And the complexity in unraveling all these questions is kind of the following. Now, I hate it when I see a patient with metastatic cancer, you know? We all do. You hate it when a friend, or someone's spouse, or, you know, it's just -- you know, you hate it. But, you say to yourself, "That's an altered gene." And genes alter at a certain predictable rate. Ten times nine for cell divisions, on average, you get a gene alteration. Now, you could argue -- it's not, actually, it's not an argument. Well, unless you're into scientology, I guess, or something. But, if it weren't for that, we wouldn't be here? As Norman, I think it was Norman Thomas said, "If it weren't for that, there wouldn't be any music." In other words, we'd all be single-cell organisms. If it weren't for alteration in genes with cell division, there would be no humans. There would be no birds. So -- but it's that same mechanism that causes cancer. Now, there's a normal change -- rate -- change of genes in a cancerous cell. Now, that patient from the New England Journal article had been there 30 years at least. A normal cell only lives seven cell cycles, and then it turns over, right? But those cells became immortalized 30 years ago, so they're going to pick up all sorts of stuff along the way. Now, I'd like to say your tax dollars [laughs] are going toward figuring out that. In order words, there's a whole lot of changes when you do whole-genome sequencing, and, oh, look at all those genes. But knowing which ones are, you want to say drivers, and who's passengers, well, that's where it takes thoughtful science Male Speaker: It's really happenstance that you really wanted to preserve kidney function. That made you make -- do that experiment, I think. Marston Linehan: Yeah, well, you're -- well, you're right about that. Yeah, you're right about that. Male Speaker: Other unanswered questions? If not, I want to thank you for a wonderful presentation, and we will stop now. Marston Linehan: Apologize for going over. [applause]