A Conversation with Elizabeth Jaffee
, by NCI Staff
Editor's Note: Dr. Elizabeth Jaffee is Deputy Director of the Sidney Kimmel Comprehensive Cancer Center, Co-Director of the Cancer Immunology Program, and Professor of Oncology at John Hopkins Medicine. Recently RAS Central editors spoke with Dr. Jaffee about her career developing immunotherapies for cancer.
How did you first get interested in immunotherapy? When I was in college in 1977 to 1981, Milstein and Kohler's hybridoma technology had just come out, I think it was in 1976, and I was really taken aback by it. I was a biochem major but I was very interested in virology and the immune system. Then I heard a lecture by David Baltimore and I really got excited and I did research using hybridoma technology as an undergrad to try to understand, believe it or not, heavy chain switching, which of course didn't really get understood until molecular technologies were improved. But it was a good project in the context of the time, and I kept thinking and I read some of the history of cancer. I had read about patients with infections developing immune response to cancer, the whole Coley story. Then I took a year after completing my internal medicine residency to do a year of NIH funded research, basically a year in which you do a small research project and you take some didactic classes. I worked with a cancer immunologist at the University of Pittsburgh, John Kirkwood, and that's when T cells were becoming understood better, because IL2 had been cloned and you could now isolate and grow T cells. The field was starting to boom, and then I came to Hopkins with the concept that that's what I was going to do. I got involved in these genetically modified vaccines which basically looked at all different cytokines to see what would cause the cascade of events that could induce an antitumor response.
Can you define a neoantigen as opposed to a shared antigen?
Shared antigens are antigens like mesothelin that are overexpressed in the tumor, so it doesn't have a specific mutation that's patient specific, such as the common KRAS mutations, which are neoantigens.
Antibodies that target PD1 or PDL1 or CTLA4 rely on neoantigens already being present in the tumor, correct?
I think that's still an open question. Right now the current agents are releasing the brakes on T cells. What those T cells see that is most important is still not clear. A few years ago everyone said it was shared antigens, and now you hear people say it's obviously neoantigens based on newly formed mutations, but I think it's going to probably be a mixture of both and it may depend on cancer types. You have melanomas and some lung cancers where they have high mutation burdens and that's why they attract T cells, but you have other cancers like pancreatic cancer or breast cancer where you can induce a T cell and then release the brake and you get regressions too. I'm a little bit more tempered in how I look at the field having been in it for quite a while. We are still building databases of what's important, and I think in the end it's going to be a combination of neoantigens but also antigens that are shared that are directly linked to driver gene pathways. The newest mouse models, the genetically programmed, genetically engineered mouse models, have been adapted by the immunotherapy world over the past 10 to 15 years to really start to understand links between the driver gene genetic changes and inflammation. Again I think it's going to be a while before we really have the true answer, of what are the most important antigens that the immune system sees, and I think it's going to vary between cancers. We need more data.
What are GVAX-type vaccines, and how do you see them being used with other immunotherapy treatments?
The GVAX we use is an allogeneic pancreatic cancer cell line that overexpresses GM-CSF to attract T cells, and that has been irradiated to eradicate its potential to grow and divide. The idea is that the immune response to the GVAX cells will spill over onto the pancreatic cancer tumor cells and target and eliminate them. GVAX induces T cell responses to the cancer-associated antigens within the tumor, and so you get these T cells that traffic into the tumor, but they get down-regulated once they come in because the T cells are making gamma interferon and they upregulate the checkpoints such as PDL1 and PD1. What we believe is that in cancers where you don't naturally have T cells you need to induce them in some way, and vaccines is one way. There's data to suggest some chemotherapies can do that or radiation, we don't know yet what is the best way. You'd think a vaccine is the most specific way to induce a T cell response against antigens in cancer. Then you need to come in with immune modulators such as ones that block the CTLA4 or the PD1 pathway. We have clinical trials that are ongoing where we are seeing regressions of metastatic cancer resistant to chemotherapy, partial responses that are durable, similar to what you see in some of the other cancers. And we've published studies comparing vaccine plus the immune modulator compared to the immune modulator alone, and you don't see it with the immune modulator alone. So we are accumulating evidence, more and more with larger studies to support the concept that in cancers where you don't have this natural attraction of T cells due to lots of neoantigens you have to induce an immune response to antigens that are in the tumor microenvironment, and then come in with a checkpoint to release the brakes. There are many people now in the field who are testing this in different ways.
You're a cancer doc, so do you see patients and treat patients?
I used to see patients and run my own trials; now I've brought on a lot of clinical researchers who do this. I oversee the clinical trials as well as run a laboratory of new concepts that go into the clinics, and I also hold the INDs with the FDA because I have a fair bit of experience in doing that. Because I'm on the inventor side, I don't want to have a conflict of interest in signing patients up for clinical trials.
Do patients who respond to immunotherapy develop resistance?
We don't know because it's too early, but we suspect that they do develop resistance. There are patients who do recur after a year or two years. That's a whole area of research that we're just starting to begin, trying to understand patients who don't respond initially, and then patients who do respond but develop resistance. I suspect that we're going to find other checkpoints, and this is why it's so important to have serial biopsies as part of the clinical trial. I think one of the areas where clinical advocates have been extremely helpful is in educating patients that they need to be willing to undergo repetitive biopsies. In pancreatic cancer we're able to get paired pre- and post- biopsies in 60 to 70% of the cases, and when we can't it's not because patients refuse, it's because it's a site that's not safe enough.
Is RAS a reasonable target for immunotherapy?
Yes, I think RAS is a good target. We've seen T cell responses against RAS in our immunotherapy studies when we vaccinate with GVAX and particularly those which have mutated KRAS, G12D, and I think we really need to reevaluate it as a target. In the early years it was evaluated as a vaccine without checkpoints, so I think it's time to reconsider what happens when you target RAS and also unleash the brakes on the T cell.
Is President Obama's cancer moonshot a good idea?
I think the field is really excited that cancer has gotten on the agenda for the President and Vice President, and who better than Vice President Biden to really push the agenda. Now we need to see how this is going to materialize into really supporting the efforts of all of us who do cancer research, and how quickly can we move new developments into patients. I deal in a disease where we still haven't made big progress, and until recently it was really hard to get funding for pancreatic cancer. Now it's better with the Recalcitrant Act [the Recalcitrant Cancer Research Act], but there's not a lot that has been done for these bad cancers, and pancreatic cancer is just one example. I'm at the age where I have friends who are dying of breast cancer in their late 40s and early 50s, and that is just not acceptable. Same thing prostate cancer bad, lung cancer, so many bad cancers out there where we have drugs now but we've only made some progress. So I'm very hopeful that this awareness that has come about will turn into real funding and opportunities to really make progress.
How do you come down on the slow, steady, continuous, reliable funding for basic research, as opposed to the fits and starts that are stimulated by these spectacular successes that are often not very penetrant in cancers with large patient numbers?
I think you learn from all of that, and I think it would be wrong not to have a discovery pipeline. What makes our country great in research is that we have a lot of components to it. There is innovation at the level of basic science, there's innovation at the level of translation, there's innovation at the level of getting things through to patient care. And I think that you can't have one without the other. I think it's short sighted to think that you can ignore any of the components. And even things where we fund small areas, niche science if you want to call it that, where they have an impact on a small percentage of patients, we can learn from it. And you know what's great now? What I tell my graduate students and postdocs, don't look what's happening at the NCI and funding, look at what is happening at the revolution of technologies that are available to us now. Because these technologies are what are allowing us to really make progress now. I mean the whole immunotherapy field is only making progress because in the last ten to 15 years the genetic technologies have become available so we can actually look at things on a small piece of tumor. So from my point of view I think we need to be funding all areas of research. I do think that the old ways of doing research are less relevant now, and I think we do need to have new funding mechanisms as well. Don't get me wrong, it shouldn't all be all R01, it all shouldn't be basic science, but I am a person who believes that you need to have mechanisms for team science, not just at one university. You have to be able to bring people together with different expertise.
What do you tell your postdocs and grad students about a life in science, a life in medicine?
What I tell them is that it is the most privileged career, because you have the privilege to think about science, to continually learn, continually contribute to patients, whether it's indirectly or directly. A lot of people have to take jobs to put food on the table, and here we are, we have this luxury, we're getting paid to do something we're passionate about. So I think we're very lucky, you should only do it if you are passionate about it. And you're always learning something different. I started out developing vaccines, I've gone into trying to understand, based on what the immune system is seeing, how cells metastasize based on physical changes, because the immune system is seeing proteins that are involved in cell stiffness. Who would have thought 20 years ago I'd be talking to cell biologists and physical scientists about cell stiffness?
Why are tumors firm?
Right, exactly. Well, not firm, but cells themselves change their myosin component, regulation of myosin and change how they move within a tumor to metastasize. We've developed this process to look at what antibodies are seeing post-vaccine in patients who do well. We're finding that some of those targets have to do with myosin regulation, or interactions with neurons. So that's how they metastasize along neurons. It's really fascinating.
Do you have a philosophy for approaching cancer as an opponent?
The way I look at cancer, cancer wasn't a very common problem up until the mid 20th century. I think the reason is that you didn't get to live long enough to see cancer, because you died of infections. This is why I think vaccines are amazing, this is what always pushed me to do vaccines for cancer, because we eradicated all of these childhood illnesses so that people can live to their 60s and 70s. I think the body as it ages is kind of gasping for survival and you get mutations that occur and then on top of it they're helped by all of the environmental dangers that we are exposed to. And I think 80% of cancers at least are in part due to environmental changes as our cells are aging and undergo this need to stay alive, stay productive.
You're on the oversight committee for the RAS Initiative. What lessons should we learn from its organization? Should there be an Immunotherapy Initiative?
I've been suggesting this at a number of different think tank-type meetings that I've been involved in at the NCI, most recently on pancreatic cancer, and then one on immunotherapy. I think what has made the RAS program special is that there are a lot of resources that are available at the NCI Frederick, and you had the flexibility to bring in different experts based on the questions you were addressing. And it's very goal oriented, and it has flexibility to change based on the regular evaluations. There are not a lot of mechanisms for that kind of a program. One of the challenges with immunotherapy right now is trying to understand the different inflammatory pathways within different cancers that are likely linked to the genetic makeup of those cancers. We need to develop an immune atlas of data, so that eventually we can do a biopsy of a patient and we can say, “they need induction of T cells against this series of antigens and we need to hit these immune checkpoint pathways to best harness those T cells.”
That's patient-based heterogeneity. How about heterogeneity within the tumor?
With the new technologies we're able to multiplex and look at one, two, three, four slides from a biopsy at once, and look at the signals. I think the next phase, and I've been trying to do this on a local level here, but I think this would be something that should really be done on a big level, is to get a 3D picture. This would take pathologists, and software engineers, and groups of people who have these different technologies to come together and figure out how do you do this. How do you get multiple slices from a small tumor biopsy and be able to look at not only what the different signals are but how far is one cell is from another cell, and making all these correlations with clinical outcomes. And so you have a database and you are able to reconstruct, from one tumor biopsy, what it looks like, is that piece of it that you're seeing with these targets in this piece up here the same as what you are seeing in the middle of the tumor.
Can you talk a little about how you balance work and your personal life?
I'm very proud that I have wonderful relationships with my daughters and they're both doing very well, and I feel have the right goals in life. Even though I've had an academic salary I've paid for people to clean my house and to do shopping and errands so that I could make it home and have dinner with my children and my husband every night. Sometimes dinner didn't happen until 8 at night, but we tried to sit down six times a week, having dinner and talking about our work and our school and things that were stressful and things that made us happy. We always took family vacations twice a year, so summer and usually the winter holiday break, that was our time together. If I traveled out of the country I usually made sure that I could bring my family, so my kids have been to a lot of places and met a lot of interesting people. They both have a love for different cultures and travel and they both have now spent time in different countries. Certainly you get home and you spend some time with the kids and when they go to sleep when they were younger I would go back on my computer and do stuff. It's funny, when they were in high school they had a lot of work, they were taking AP classes so we all worked on Sunday together and then had our family dinner, and usually grandparents came over.
Do you come from a scientific background? Were your parents in science?
No, both my husband and I were the first, my husband's family never went to college, his parents had to work, and my mom never finished college, so I was the first woman to finish college in my family. My father was a business man, and everyone else in my family had been in business. So we come from pretty humble means, both my husband and I.
How do you get to work in the morning?
Currently I drive. Is that what you mean? But I actually, we were living out in the suburbs to use the public schools and now the kids are grown so I have purchased a town house downtown and I can walk if I want to work. I love physical activity and it's been frustrating to have to drive to work.