Using Modern Genetic Tools to Study Breast Cancer with Dr. Charles Perou
By BCRF | October 20, 2016
By BCRF | October 20, 2016
When many of us hear the word technology or consider the life changing power of new technologies, we think about companies like Google or Apple or Amazon or Tesla. We think about products like internet search or I phones or drone deliveries or even self-driving cars.
Dr. Charles Perou, a BCRF investigator since 2003, thinks about breast cancer. You might say he sits at the intersection of technology and tumors.
Dr. Perou is the May Goldman Shaw Distinguished Professor of Molecular Oncology Research and professor of Genetics and Pathology and Laboratory Medicine at the University of North Carolina's Lineberger Comprehensive Cancer Center. He also has been a BCRF grantee since 2003. Among Dr. Perou's many areas of study, using new genetic and genomic technologies, technologies that look at single genes, that's genetics, as well as all parts of an organism's genes, that's genomics, in order to do what he calls personalized medicine, tailoring patient care to individuals based on their particular genetic makeup. The goal? To significantly improve patient outcomes.
Dr. Perou is the recipient of the 2016 Jill Rose Award presented at BCRF's Annual Symposium and Luncheon for outstanding research excellence - his seminal work in breast tumor profiling has revolutionized personalized medicine.
Chris Riback: Dr. Perou, thanks for joining me. I feel like I need to ask the most complicated question first. How do you describe what you do for a living?
Dr. Perou: I'll start by thanking you for the opportunity to discuss my work today and breast cancer in general. I would describe our work as using modern genetic tools, now often referred to also as genomic tools, to study human breast cancers. The great advantage here is, we can now look at the entire genome, meaning we can look at all the DNA sequences and all the genes at once as opposed to what we had to do ten or certainly 20 years ago where we would go one or two genes at a time. Now we can look at all 25,000 genes at once and that's really allowed us to see things that we couldn't see before, like breast cancer is not one disease, it's at least five or more diseases. With these tools, not only can we say it's five diseases, but we also know precisely which genes are now involved in which of these diseases and that is really providing the foundation for personalized care.
Chris Riback: That was really a landmark discovery, wasn't it? Realizing that there are, and you say now, as many as five different types of breast cancer. Take me through that. What happened? Historically there was the view that, well, breast cancer is breast cancer. Even today, many of us, when we think about the people who have it or about the disease and we say, it's breast cancer. I'm sure that you hear it and you're like, no, wait a minute. There are ... which type of breast cancer? Tell me about that discovery and tell me did you get immediately that this was going to massively change things?
Dr. Perou: I don't think I did get, at the time, potentially the implications of the work that we had done and how far reaching it has been. This takes us back to maybe the late 90s, probably 97, 98, when we first started to apply this tool, this genomic tool, which back then, the term was DNA microarrays. I was fortunate to be able to go to Stanford University and be a member of one of the laboratories that helped to develop this tool and one of the first laboratories in the world to apply it to the study of human cancer, so that was incredibly exciting to me and we did this, right? We got a set of 70 different breast tumors, applied this DNA microarray, which at the time allowed us to look at 8000 genes at once. We didn't have the full genome back then. Even with as few as a third of the genome, we could clearly see that breast cancer was four or, we've now learned, at least five groups based on patterns of gene expression.
Previous to these studies and contemporaneous with those studies were other studies in breast cancer where we already had a good idea that it was a heterogeneous disease, that there were for example, estrogen receptor positive disease and that's important because there's a whole series of drug targeting the estrogen receptor positive tumors. Then, at about the same time was really the advent of the importance of HER-2 gene amplification, because there's a drug called Trastuzumab, which targets HER2. At this time, people were going "Aha! We've got estrogen receptor positive tumors, we have HER2 positive tumors." Then we came along with the microarrays and said "Well, there is at least a third group of tumors here that we called the basal like type and that basal like type doesn't have the estrogen receptor and it doesn't have HER2 and it has a very different genomic pattern and we think this is a group."
That was, it took a while for people to appreciate this basal like group, which is also known by another name, triple negative breast cancer, which is really sort of its clinical name, because it is negative for the estrogen receptor, the progesterone receptor and HER2 and sort of this confluence of the genomics and the clinical, at least brought to light this third group and then further subdivisions have become appreciated since then. It was really this genomic technology allowed us to see this basal like subtype and then also beautifully incorporated the previous knowledge. There's a group that was defined by HER2. There's a group that's defined by the estrogen receptor and a whole set of genes that are regulated by the estrogen receptor. Not only did the genomics show us new things but it also showed us all the things we already knew and put it into one comprehensive framework and I think that was also an impetus to bring this into the limelight.
Chris Riback: Yeah. I saw a line where you said that it was like turning on the lights and being able to see the entire field. That's exactly what it sounds like you're describing. There were these things that were there and some of which you knew and some of which, and all of the sudden, the lights went on and through the technology and through the being able to see the entire genome, you were able to make discoveries and recognize and say wait a minute. There's more here than we thought there was before.
Dr. Perou: That's exactly right and we can now see the whole playing field. People shouldn't interpret that to mean we can interpret everything we see, right? There's still a lot of discovery, there's certainly a huge amount of understanding that we need to gain, but with these genome tools, we basically can now see the blueprints. We now have the parts in front of us. We're learning how the parts go together and we're learning how the parts come apart in certain types of cancer and so it's just an incredibly exciting time, still, to be doing this. The technology has now moved from these DNA microarrays to sequencing based approaches, sometimes referred to as next generation sequencing. I actually like to call it current generation sequencing because it is what we're doing now, and we're discovering more because, compared to even ten years ago, it's so much easier and faster to analyze the genome and now we can do it with even more precision that there is even more important subdivisions that are being discovered and evaluated for potential of new personalized medicine markers.
Chris Riback: Let me ask you about that deeper understanding and these discoveries and all of the information that's coming at you. I can almost see, I've got this vision in my mind of all this information coming into one side of your brain but then you have to do something with that to make it, I would think, and as we discussed earlier and I mentioned earlier, actionable and a part of personalized care. I'm struck by, you once said that you consider yourself a translational researcher because you live in both the basic science and the clinical research worlds. What did you mean by that and how important is your ability to translate the science that you see? How important a part of your work is that?
Dr. Perou: In many aspects, for me, that's the most important part. I love discovery and learning new things on the basic biologic side, but I also want that to improve patient care and improve health care and improve patient outcomes and so that is the translation, right? Literally, you discover something and now it needs to be translated to a new language, and that new language is clinical medicine, whereas the starting language was the basic research. There's a very large number of things that you can learn in the basic setting, which may have no effect on the clinical setting.
Part of my mission, right, is to wade through the genome of results that we get and try and identify those that I think could make a difference in the life of a breast cancer patient and to try and now put on a different hat, which is the hat of clinical research and clinical trials to try and bring these genomic findings into the clinical world. We're doing that, right? We've developed some clinical tests. There's more in the works and certainly, again, with the technologies continue to get better and better and actually, part of the mission is to even bring the technologies into the clinic, which can be a challenge in and of itself.
Chris Riback: You see this in so many different industries, I feel. Again, please tell me if I'm interpreting what you're saying correctly, but it's almost this, the tension and the ability to connect scale and personalization. On the one hand, you operate in massive scale, if you will, because you're trying to find these discoveries and shine the light and find what's going on across breast cancer in these different types of cancer in your lab, actually, of course, you look at, there are multiple types of cancers that can get looked at. In the end, that scale needs to get translated into something highly personal, down to the individual, that personalized care. Do you think about that tension, and maybe it's not tension. Maybe it's the benefits, or maybe that's part of the translation of going from scale to personalization and maybe back to scale and back to personalization. Do you think about those two areas at all?
Dr. Perou: I live this everyday, right? I wouldn't describe it as tension. I'd describe it more as, this is the process and it's actually a very helpful process because it forces you to look at this myriad of data and to think about it hard, to try and make a hypothesis about which few points within it might have a clinical implementation or might have an impact on this person, right? You don't need to have ... the genome of results from a given individual is not clinically actionable. You have to pick out the few that you think are the most important as we call cancer drivers and then target that. That is an extremely complex process but it's this important sort of sieve. They're filtering down the genome into a few of the most important components.
I just want to further comment that we have this exciting technology development, right? That allowed us to now sequence your genome in a day when it used to take years. Hand in hand with the development of the physical technology is the development of the computer analysis tools and that is equally as important as the development of the technologies, because it is the sieve that looks though the ten million data points that we get in some of these experiments to figure out the handful of them which we think will make a difference in the life of a breast cancer patient. We spend a great deal of time and money and effort on the development of these computational analysis tools and certainly on the usage of these tools. Half my lab now is computational people who just analyze the data and extract biological insights from that.
It's kind of fascinating, over the course of my career, certainly computers are important, but now we're literally just doing experiments on the computer. We can make discoveries from published data sets of genomic data coming from individuals at other universities, not even our own data and this is what's often referred to as big data mining or analysis, but it really is true. Again, it's incredibly exciting and I want to encourage every one to participate in this, both from potentially patients will, to encourage them to donate their samples to research, for researchers to make their data publicly available upon publication because in both contexts, this is benefiting the global community and it is this exact data that is the basis for the advances to be made in personalized medicine.
Chris Riback: You've been very upfront and out front about the importance for every breast cancer patient to try and get into clinical trials and I guess on other levels share data and share information. I was thinking about that. One the one hand, obviously it makes total sense and one can look at only a sub portion of the discoveries that you're making and recognize the importance of big data and access to data and that more data is needed and you can, then you or other people like you can make new discoveries.
At the other side of that, that's maybe the scale side. At the other side of that, that personalization side, is the human being who's going through the personal challenge and is trying to think about all the different things that he or she is needing to worry about and focus on and the thought of, okay and also can you get into this clinical trial? Can you make sure to get into clinical trials so that we can advance this, our research for others? I understand it, of course, and yet it feels like that can be, maybe something hard to hear at a time when people may be focused on their own personal health. Do you get ... What types of reactions do you get when you really advocated for that or do people get it right away and they really don't have a problem with it and they want to do whatever they can to help?
Dr. Perou: I think most of the audiences I speak to, scientific audiences or actually advocacy agencies and things like that are very aware of the benefits of clinical trials. We hope and are trying and maybe this discussion can also reinforce the point of, it is very important for the clinical trial organization to get patients into it. I think it actually also benefits those patients. Clinical trials, when you think about it, basically take what is the best standard of care treatment and then add to it what we think is now going to make it better. Now, we're not always right in picking the drug that is going to make it better, but at the very least, the patient is going to get, in essence, the best treatment possible and then what we also think might be next best.
It really does benefit the patient to be in a clinical trial and certainly it benefits the scientific and clinical community because it is ... clinical trial is where these advances are made. There's no way around that and so we can only advance personalized medicine as fast as these clinical trials can get designed and run and completed. It is all dependent upon the patients going into the trial. Then I also would want to, again, advocate for the patients to donate their materials for medical research, because that is the samples that we then apply these cutting edge tools to. Without those samples, we'd be nowhere, so it's really thanks to these patients and their donations that we are making these advances and we're forever indebted to them for that gift to society that they're giving.
Chris Riback: That is so well put. Yeah. It really does feel like a gift to society. To segue from that, again, tell me if I have this wrong, but an example, to make this really tangible, one area of your research, as I understand it, is focusing on interactions between tumor cells, at least the research in your lab, between tumor cells and immune cells. In fact, you and colleagues recently co-published research in the Journal of the National Cancer Institute that looks at what the number of immune cells inside tumors means for different cancer types and the answer turns out to be pretty complicated, doesn't it?
Dr. Perou: As always, nature is very complicated and you're exactly right. We're now gaining this amazing and tremendous appreciation for what we call the micro environment, so, the tumor is often thought of as just the malignant cells, which in the case of breast cancer are malignant epithelial cells. The tumor lives in a world that includes more than just the tumor and now we've known this for decades or more, right? Just getting an ever increasing appreciation for the immune system's interaction with the tumor. Many tumors are just filled with immune cells and the immune cells are attacking the tumor but actually what we're now learning is they're attacking the tumor but they're also being restrained by the immune system at the same time so there's an exciting new class of drugs called immune check point inhibitors, which are, in some patients able to remove that restraint, thus activating the immune system. In many patients, particularly in lung cancer and melanoma, these drugs are showing tangible effects, improvements in patient survival, which in melanoma is just amazing. I mean it's amazing in all cancer types when we see new drugs having this effect.
Now these drugs are being applied in breast cancer patients, particularly triple negative breast cancer patients and it's just scientifically so fascinating about, in essence, how the immune system is interacting with the tumor and the tumor is talking back to the immune system and so I'm now becoming an immunologist. A part of my laboratory is devoted to these immune therapy studies and the biology of immune tumor interactions. I think this is just, in some ways, another illustration of how science can be so exciting and new things come up and we're able to somewhat change directions or pick up a new project and go learn a whole new set of interesting biologic interactions.
Chris Riback: As I'm talking with folks like you, I'm really struck at the, what I might call the importance of flexibility and the benefits of surprise. It has not failed, in one of these conversations where I talk to someone in your position and they have been going down one road, and even possibly looking at one type of cancer or something going on within one type of cancer, and they extrapolate and have the kind of intellectual flexibility and say "Wait a minute, I can apply this elsewhere." Do you agree with that?
Dr. Perou: No, I totally agree with that. Again, this is, I think, a key to, I'll say a key to good science and one of the beauties of science, right? You can try and be smart and make hypotheses and pursue a determined path, but things pop up all the time to the left and to the right that you never anticipated and that's great, right? If I discovered something new that I didn't anticipate, I've still discovered it and now I'm going to pursue it. So serendipity in science or just seeing in a different way is just critical because nature is so complex. You can try and predict things that are going to happen, but you're often wrong and you're probably more wrong than you're right. When something new pops up and it's important and fascinating, I want to study it. This keeps the science fire lit in many respects.
Chris Riback: To close things out, let's talk about that fire. What's next?
Dr. Perou: What's next?
Chris Riback: We appreciate everything you've done to date, but what matters is what's next (laughing). You understand, doctor?
Dr. Perou: Absolutely. I think what's next is more experiments in the lab and more studies on clinical specimens. Particularly, I think the immune system is going to be very critical and is an important determinant of breast cancer tumor behavior and we need to understand how that works. Now we can potentially begin to manipulate it for a therapeutic benefit. I also think genomic analysis, gene expression analysis, DNA sequencing are continuing to identify important biomarkers and these biomarkers need to be rigorously tested and if they pass that test, implemented in the clinic and so I think, over the next handful of years, a few of these new biomarkers are going to get identified each year and so in this way, we're going to continue to make improvements in personalized medicine. We may not shatter the glass ceiling all at once and have suddenly a hundred new biomarkers, but if we get one or two or three a year, and we do that for the next ten or fifteen or twenty years, that is going to be really significant improvements for individual patients and it's going to make the outcomes for patients better. That is the path we're taking and I think it will include the immune system.
There is also important tumor aspects that we're also going to get at, like what are the tumor features that the immune system is seeing? Then, of course, we also, there's still many important advancements to be made at specifically targeting the signal transduction pathways that are mutant in tumors and I think much of that is also going to come from this computational analysis, right? Can we model these signaling pathways in the computer, such that they make predictions that you can then test. Then, if that model passes that test, it can be applied to all the patients. I also think in the next five to ten years that computer models of tumor behavior are the future and will really be really important for individualizing treatment for this patient and that patient based upon a conglomeration of all this genetic and proteomic data that we have and will continue to collect.
Chris Riback: Do you see why I asked how you describe yourself? Keeping a little track, I've got statistician, I've got oncologist, pathologist, computer scientist, immunologist.
Dr. Perou: I don't know what I am sometimes, right? I kind of like, I'm going to call myself a geneticist at the heart, but, as you've alluded to, you now have to be versed in statistics, in immunology, in cancer biology, in clinical medicine and so I encourage young scientists to try and have a diversity of interests and knowledge base. Also, critically important, I am not all those different hats, right? I have wonderful collaborators at UNC and other institutions who are experts in those field. This is a team effort, right? The many accomplishments I've had are applicable to all these individuals who I've worked with. Without them, we wouldn't have made these discoveries and implementation, so it really is team science and multidisciplinary science that is the current gold standard and will be into the future as well.
Chris Riback: No doubt. I've got to say, again, for somebody like me, outside of your field and what you do, understanding that and seeing the ways that the different disciplines have to come together and the role of somebody like you helping orchestrate those teams and bringing together the different disciplines, that's where it certainly seems new discoveries are made. That's where new impacts, positive impacts for individuals get generated. Dr. Charles Perou is the May Goldman Shaw Distinguished Professor of Molecular Oncology Research and Professor of Genetics and Pathology and Laboratory Medicine at the Lineberger Comprehensive ... It really is comprehensive, isn't it? Comprehensive Cancer Center at the University of North Carolina. Dr. Perou, thank you so much for your time. I'm Chris Riback. This is BCRF Conversations.
To learn more about breast cancer research or to subscribe to our podcast, go to BCRFCure.org/podcasts.
When you give to BCRF, you're funding critical hours in the lab. More time for research means longer, healthier lives for the ones we love.