Dr. Angelo De Marzo Video (Text Version)
Title of Talk: Introduction
Speaker
Dr. De Marzo will speak for about 10 minutes with some introductory remarks. He has expertise in Pathology and Molecular Patho-Biology of Prostate Cancer, Tissue Banking, Tissue Microarray Technology, Software and Database Design, and Image Analysis. He is a former member of the Integration Panel; Angelo.
Angelo De Marzo, M.D., Ph.D.; Professor of Pathology, Urology and Oncology, Director, Tissue Microarray Core Facility, Johns Hopkins
Thank you, Alvin, for the kind introduction and I certainly want to thank the organizers for inviting me. It’s been a wonderful meeting as it always is. Also thank the DoD of course; I’ve been supported by the DoD for many years in my laboratory off and on, including myself, postdoctoral fellows, and other students in the laboratory, and it’s been incredibly gratifying as all the other speakers have said about high-risk type of approaches that would not be funded for instance by the NIH.
The other thing I wanted to mention was I did serve on the Integration Panel for 3 years, and I have to tell you even though I got money from the DoD for my research I felt that—that experience was far surpassing anything I did in terms of getting grants and doing because of working with these wonderful people that you’ve all heard about. I don’t want to announce them all by name; it would sound like I was getting some sort of Academy Award or something but working with them has felt like working with family, and it’s been just outstanding and I think this program is fantastic and I certainly will try to continue to support it to the best of my ability.
So I wanted to talk a little bit; I'm not going to talk specifically about what the speakers are talking about. I’ll let them do that but I wanted to introduce the topic. Really it’s an issue of advanced disease and how difficult of a problem it really is to literally cure an advanced cancer. And I’m going to try to go through that step-by-step. And so I think most people remember that cancers develop through a series of multiple steps that happen, and you know it’s a disease of the genome. So—so somatic changes happen to the DNA in precursor cells or cells that are sitting otherwise minding their business and—and mutations or epigenetic changes, these are both called somatic, meaning they’re different than what you inherited. So yes, you inherit clearly risk factors that increase your risk for cancer but what happens during the development of cancer is changes happen to that DNA that are called somatic. And then what happens is an individual cell might get a change or two that gives it a selective advantage. That cell becomes selected for simply as in evolution and it goes on—obviously I’m simplifying this to acquire some new properties and then other mutations or changes happen. And again—and you have these waves of clonal selection and expansion and the cells obviously have to interact with their micro-environment bringing other cell types like angiogenesis. Inflammatory cells come in until it takes—and it takes many years, other than obviously childhood cancers that can happen very quickly. These—this process takes years—20 years of cigarette smoking or—or gastritis for 20 or 30 years or colitis. This doesn’t happen overnight so many, many changes happen to the genome.
And then you—you end up with you know a malignant neoplasm that in most cases kills you by metastasizing and so other genetic changes happen for that process as well.
And so what I wanted to talk a little bit about was this concept of heterogeneity. I think you’ve heard quite about this and—and certainly anyone in cancer biology hears about this all the time because the way I wrote it here, it’s a major problem—is it the major problem—and very well might be for advanced cancer treatment and I think I’ll try to walk you through why I think that might be the case. And so there’s lot of different types of heterogeneity. I think we tend to forget that when we’re talking about that especially to audiences. There’s heterogeneity of morphologic types, obviously there’s different types of cancer, so in prostate cancer most people who have prostate cancer have adenocarcinoma. It’s over 90–95% of cases are adeno—but even with an adenocarcinoma there’s different Gleason patterns and we don’t know too much about what drives that.
And then there are other cancer types in the prostate. There’s small cell neuro-endocrine, very rare, but very aggressive. There’s ductal adenocarcinoma and other variances as well—squamous cell that are very rare. And certainly it’s not known how this happens that heterogeneity at the morphologic level—could it be driven by different cell types of origin, yes, although we don’t have really good data for that in the prostate. There is good data for that in the breast I would say. Or, are there different pathways altered and—and there’s obviously some data for that too but we’re not—we don’t have anything definitive at this time.
But there’s other types of heterogeneity so heterogeneity within a given morphological type, and what do I mean by that? So within a given type of cancer for instance, you can have similar pathways that are driving the cancer, but each cancer from each patient has a different constellation of patient-specific changes. And so we—we knew this for many years, and now with the efforts to sequence the entire genome either all the exomes or all the—all the entire genome we find that yes; cancer patients of a certain class have very similar types of pathways that are altered by mutations, so P10 often in prostate cancer, TMPRSS2-ERG obviously you’ve heard a lot about these but each patient also has a unique malignant neoplasm with unique genetic changes in that case even when they look the same to we pathologists under the microscope.
And you know does this happen in prostate cancer and—and absolutely it does and then very recently published what I would consider landmark paper on whole genome sequencing of seven cases of prostate cancer really illustrates this. Levi Garraway, Mark Rubin, and others, many others contributed to this project; I think it was partly funded by DoD and certainly the Prostate Cancer Foundation. But I don’t want to go through all the complexities here but these three—each one of these as a person and their—their genomic DNA from a tumor was sequenced, and the top three represent lesions that have TMPRSS2-ERG rearrangement and the bottom three do not have TMPRSS2-ERG rearrangement. These were all reasonably high-grade cases and what you can see is that intra-chromosomal and inter-chromosomal changes are quite dramatic, and there can be a lot of them and they’re different in every patient. So even though you have some very similar changes that get you to the cancer, there’s also heterogeneity within a given type of cancer and in this case adenocarcinoma, in this case all three TMPRSS2-ERG adenocarcinoma(s) on top and they still have a lot of heterogeneity.
So that’s across patients; what about an individual tumor. And I don’t think this gets talked about too much. I know Bob Vessella talks about it but when you carry a type of tumor and we often can't do that for prostate cancer where you put the cells in culture and grow them and try to—and look literally at their chromosomes with relatively old-fashioned technology although we have better technology now—when you actually do this for many cancers no cell is the same. So yes; if three cells have the same chain, say P10 loss or TMPRSS2-ERG rearrangement, you call that a clone. But they usually look at 20 cells and I can tell you in—in advanced cases with a lot of changes, every single cell is different. So it’s sort of this amazing constellation, so if you have billions of tumor cells you actually have billions of separate genomes in that tumor.
And so Steven Elledge has this beautiful paper that gets into this a little bit that I would highly recommend from a few years ago where he basically goes to this concept and he said you start out with multiple types of tumor cells, so again clonal changes happened that got them all to this point but then there’s sub-clones and genetically that’s shown by these different sizes and so for instance you give therapy one and you might target all the cells for instance that have cube shapes so they—they get taken out. And then but there’s inherent resistance because of the genetic instability or—or genetic heterogeneity at the population level. And then therapy two might take out all the circles and then you’re left with triangles. And the problem is there’s not just three types. So obviously this becomes a major problem in cancer therapy because you already have billions of cells sitting there that are potentially already resistant to the compounds that you’re going to put on them. So how do we obviously get out of this and—and obviously we don’t have the—all the answers but there’s a couple concepts that are emerging that seem to be very, very promising obviously and we have examples in humans now where this happens.
And some of these are shown here; one is called oncogene addiction, first talked about by Bernie Weinstein meaning when you activate an oncogene, for instance BCR able and CML even though there can be many other clonal genetic changes as that cancer evolves it’s amazing that the tumor cells seem to be addicted to signaling through this pathway such that if you inhibit it later you often get a very dramatic response and prolonged and increased survival. Obviously not everyone gets cured but it’s—it’s really amazing that even in advanced lesion it’s still dependent. And the same thing happens obviously in breast cancer with HER-2 Neu signaling where you can treat patients with metastatic breast disease who have HER-2 Neu amplification and often get these amazing responses.
Now c-Myc is an important oncogene in many cancers and we think in prostate cancer as well and at least in animal models there’s some very good examples where if you reverse Myc expression even in an advanced tumor often the tumors will have a dramatic response.
Now the other concept that’s similar to this and I think it is related to the AR is lineage addiction and this was also—I didn't put the reference but Bill Sellers and Levi Garraway and wrote about this a few times and I thought it was an amazingly nice synthesis because what happens is obviously cancer arise individual cell types so tumor cells, brain cells are very different than colon cells, than prostate cells and you have these lineage survival factors. And I would argue one of those of course is AR in prostate cancer that operates during the development of that organ and somehow the cells that develop there seem to become totally dependent or very largely dependent on the signaling through that lineage factor. This happens also in melanoma for example and so targeting that factor as a survival factor is obviously something we’ve all—everyone has been doing for 40 or 50 years in prostate and we’re going to hear about that even some more.
And then finally you know how can you deal with this and—and Bill [Kaylen] has talked about this beautifully, this concept of synthetic lethal and that just basically means if you have a mutation in one gene what other gene throughout the genome if you mutate will that cell have to die—it cannot tolerate. And so maybe for instance in this building if you took out one column nothing would happen; what other column in this room can you find to make the whole building collapse? I wouldn’t recommend trying it but that’s really what the synthetic lethal concept is and—and the example here that looks very promising is in breast cancer. And so, for instance, in BRCA-1-dependent breast cancers or BRCA-1-mutated breast cancers, this discovery that inhibiting the p-Akt pathway has dramatic effects on these tumors is really showing that synthetic lethality probably holds.
And so what are we doing in this session that relates to this? Well, Tim Thompson is going to talk about really a new mechanism by which the Myc-oncogene in prostate cancer might be targeted and controlled—very exciting new work from there and then Marianne Sadar and I think I’m saying that right—she just told me, so if you—if you meet her you have to say it right—targeting the oncogene lineage addiction. Of course we’re talking about the AR in a totally new way, and I don’t know if anyone saw her talk yesterday but she’ll get to expand on it today. It’s very exciting. And then Bob Vessella for many years has been characterizing circulating tumor cells, tumor cells that have gone to the bone, disseminating tumor cells, and getting them out a few at a time. And the ability to understand the heterogeneity here I think is going to be unbelievable you know based on his work of doing this, and he’s been at this for many years. So I’m excited to see all these speakers. And I’m going to leave you with that and hopefully I think we will have time for questions at the end and maybe all this can come together. Thanks very much.