Future of CAR T-Cell Therapy

Podcast Transcript

Dale Shepard, MD, PhD: Cancer Advances, a Cleveland Clinic podcast for medical professionals. Exploring the latest innovative research and clinical advances in the field of oncology. Thank you for joining us for another episode of Cancer Advances. I'm your host, Dr. Dale Shepard, a medical oncologist here at Cleveland Clinic overseeing our Taussig phase one and sarcoma programs. Today, I'm happy to be joined by Dr. Jos Melenhorst, Director of the Cell Therapy and Immuno-Engineering program and Vice Chair of the Center for Immunotherapy and Precision Immuno-Oncology. He's here today to talk to us about the future of CAR T therapy. So welcome Jos, happy to have you here.

Joseph Melenhorst, PhD: Thanks so much Dale, and appreciate being invited to this podcast. I'm a great fan of the show as you know, and this is absolutely a great way to get connected with folks at Cleveland Clinic.

Dale Shepard, MD, PhD: Absolutely. So maybe start out, tell us a little bit about what you do here.

Joseph Melenhorst, PhD: I was recruited to Cleveland Clinic and started in March, coming from the University of Pennsylvania, where I worked at the mechanistic translational aspect of chimeric antigen receptor T-cell therapies for cancer. I work with Carl June. We roll out the first FDA approved CAR T-cell therapy for B cell malignancies. And in the process of manufacturing and applying these therapies, we discovered a number of mechanisms of response and resistance, and that's really what I'm bringing with me, this expertise to Cleveland Clinic, as well as deep understanding of CAR T-cell therapies and mechanistic aspect. Part of it is developing these therapies, and part of it is really to do the correlative mechanistic studies to improve on the therapies that we administer to patients.

Dale Shepard, MD, PhD: All right. Excellent. So, we've actually had previous podcasts and we've talked about clinical use of CAR T therapy. So, we've kind of covered that and people can go back and listen to those things, but maybe just as a general overview. We have a lot of people that might be listening from different backgrounds. CAR T therapy. What is it? And how does it differ from some of the other things we do?

Joseph Melenhorst, PhD: So CAR actually are chimeric antigen receptor. They are synthetic molecules that have been engineered to include an antibody fragment, if you will, that binds to a service protein on target cells. In this case, for instance, in B cell malignancies that typically express CD19, we have an antibody against CD19 included. Then we have additional moieties that stabilize it on the surface of T-cells like CD8, and additional molecules that allow T-cells to respond, CD3, and to continue the memory function of the cells like CD28, or four on BB.

So the typical designs include all these elements with the antibody binding fragments and the signaling domains. So that is a CAR. Then the CAR therapy entails really the engineering of patient cells, typically stably introducing the CAR into the T-cells genome. And these cells are then infused and go after the tumor cells. What makes this therapy different from chemotherapies is that these cells will actually sustain immune selection if they are doing the job on the tumor cells. So tumor cells... Is different from chemotherapies, which are administered and washed out. CAR T-cells aren't washed out. They continue, they are referred to as living drugs and they have the ability to suppress even quiescent tumor cells, once they re-activate, if you will. So they're very different from the standard therapies.

Dale Shepard, MD, PhD: Another characteristic is that most of our therapies are treatments that are more patient specific. And I guess that comes with advantages and disadvantage. Is that true?

Joseph Melenhorst, PhD: True. Absolutely. Yeah. So for these CAR T-cell therapies, they're patient specific, in that we use the patient's own immune cells. There's also iterations where we don't, where we take normal donor cells, but most CAR T-cell trials have used the patient cells. Patient T-cells aren't always up to the task and sometimes feel, in mounting a meaningful de-tumor response, then we have to think about other ways of targeting the tumor.

Dale Shepard, MD, PhD: What are the biggest challenges as we move forward in terms of from an engineering, a CAR T? What are the biggest barriers to moving forward in the field right now?

Joseph Melenhorst, PhD: Yeah, that's a great question. There's many. If you still keep thinking about the B-cell malignancies, one of them is that patients cells are just not healthy, not viable enough. In those case, we found a solution in using normal donor T-cells, which are then engineered to not use the T-cell receptor, but rather use the CAR. So that's one. The second is identifying the best molecules that allow these cells to expand and persist in patients. And there's a number of developments in that domain. A third I would say is if the therapy works really well and patient has high disease burden, we oftentimes see a toxicity that we call Cytokine Release Syndrome.

It's a rapid release of inflammatory cytokines that make the patient sick. That's a side effect that is manageable clinically, but there's innovation happening there where patients are treated in MRD. Then the side effects aren't nearly as dramatic. And these others. So the first generation therapies. They are called second generation CARs. They were designed in a certain way. We find that these aren't always working as well as they should. So there's challenges in optimizing. And then of course, what do we do with solid tumors? That's always the elephant in the room. And I think there's some really exciting new developments also, like in pancreatic cancer.

Dale Shepard, MD, PhD: So as the guy who treats solid tumors, and kind of jealous of some of the responses that are seen in the he malignancies, what kind of progress is being made in solid tumor?

Joseph Melenhorst, PhD: I think in solid tumors, it's finding the right antigen, that's really. So in prostate cancer, we have found a good target antigen, and we do see clinical responses even to the point that we find Cytokine Release Syndrome, which is somewhat exciting for CAR therapists, because that's typically associated with liquid tumors. Haven't seen it much in solid tumors, but one more recent report from groups in California and from Carl June at U-PENN, there have been some meaningful responses seen. So I think that's a real exciting and development. And mesothelial cancers have also been... We also see progress in those domains using a CAR against a service protein that we call mesothelin. So those are two areas of development, and there's a lot of more research happening to better understand the tumor and find new molecules to go after with CAR therapies.

Dale Shepard, MD, PhD: And then is there... Maybe with mesothelin connection, perhaps ovarian cancer as well?

Joseph Melenhorst, PhD: Yeah. And pancreatic cancer also of expressed mesothelin.

Dale Shepard, MD, PhD: You mentioned before about looking at mechanisms of, and things related to response and resistance. And ultimately all therapies we use, it ends up being like, what patient is best for what therapy? And how do we match those up? What kind of things have we learned about predicting response to these treatments?

Joseph Melenhorst, PhD: Yeah, that's a great question. I think what we have learned is that the number of aspects in the therapy play a really important role. Being the T-cell health itself, how the tumor cells resist the therapy and the tumor mic environment, and home. Four elements seem to be important. And to start with latter, what we have seen on, for instance, glioblastoma, targeting with a CAR, and a recognized antigen only expressed on tumor cells, cells were infused peripherally, single infusion, and they did find a way to tumor cells. So seemed that the homing is working, but that's still something that we can improve. If you look at the T-cells before we engineered them with a CAR, we have found that immune features... Actually features related to cell health and exhaustion and whatnot are really important in predicting response to the therapy.

When we engineer the cells with the CAR, those features will be amplified. In addition, we found and others have found also, that a phenomenon called exhaustion has an impact on function of the cells. And also in the tumor end, we've seen tumor cells resisting by modulating pathways involved in cell death. So T-cells kill tumor cells. If tumor cells have modulations, then that won't happen and the side effect of that is that, chronically a tumor cell that stays alive longer, will exhaust T-cells because they can't do the job well. There's a number of challenges that we have seen, but we have learned to identify and predict T-cell health. We have learned more about the tumor cells, how it resist. We have learned to predict Cytokine Release Syndrome. And we're learning a lot more about homing off these T-cells. I mean, actually we can inject CAR T-cell local, regionally, like what we've been doing in glioblastoma.

Dale Shepard, MD, PhD: You think there'll be an opportunity, for instance, to do an immune test? To test T-cell function or something upfront before we ever pursue a CAR T therapy and make those predictions for instance?

Joseph Melenhorst, PhD: I think so, Dale. Yeah, I definitely think so. In work that I did before we did develop such a predictive assay for chronic lymphatic leukemia, and where we could predict such a response. And we have since gone on to apply this predictive algorithm to multiple myeloma, non-Hodgkin's lymphoma, with multiple myeloma treated with a very different CAR, and then non-Hodgkin's with the same CAR. And there we found that common principles also predict the response to CAR therapy. So you're right. It's thinkable that we can even predict patient response before any CAR T-cell is manufactured. And of course we can think about what else can we then give to this patient? Aren't they able to mount a meaningful response against a tumor?

Dale Shepard, MD, PhD: And I guess the flip side of response is resistance. What are the exciting things that we're learning about predicting resistance?

Joseph Melenhorst, PhD: The surprise finding. What I did with a graduate student McKenzie Collins at U-PENN, was that even in administering or using the KIMRIAH product that we developed testing against primary patient tumor cells and CLL. So the circulating tumor cells, they were quite resistant to second generation CARs, which surprised us greatly. And in CLL we have seen about 25% or so of patients achieving a durable remission. And we call that a cure, we probably said earlier this year, in nature. So it does work, but there's some level of resistance that we don't quite understand.

So we found that phenomenon, but we also found that in CLL, it's a disease with an overload of quiescent apoptosis resistant tumor cells into circulation. Whereas those in the tumor mic environment were actually more actively dividing and susceptible to CAR therapy. So we've learned about how our second generation CAR didn't quite do its sharp right against bulk tumor cells. And that's leading to novel designs in amplifying the therapeutic index. And you can of course think about optimizing CAR, but can also think about combination therapies that change something in the tumor itself.

Dale Shepard, MD, PhD: And so, an interesting point. You can change the therapy itself, or you can change either the tumor or the patient environment that the CAR is going to be introduced into. Of course, if there's a second generation CAR, what would the third generation CAR look like?

Joseph Melenhorst, PhD: That's a great question. The third generation CAR... I think maybe one step back. What I've also learned from these correlative mechanistic studies, that we don't really need persistence or CAR T-cells in every indication. And I think non-Hodgkin's lymphoma is really good example where two different, entirely different products, one made by Cyte Pharma, has a 28 Zeta CAR, or a 4MBB Zeta signaling CAR. Both of them have a complete response rate of about 40%, but not all patients, like most patients, will not sustain their CAR T-cells. So there's something else happening in the patient that we haven't fully characterized yet.

So I think it's, part of it is to really do detailed mechanistic and correlative studies to understand this phenomenon and actually exploit that. And that's a phenomenon that might actually, until the patients own immune cells are taking over, once the bulk of the tumor is eradicated, then eliminating tumor cells. And many tumors have mutated cell antigens. Neoantigens they're called. Chair of my department, Tim Chan, is a leading expert in neoantigens and he's developing a deeper understanding of what that entails. So you can think about combination of CARs with neoantigen targeting therapies.

Dale Shepard, MD, PhD: Along those lines, what other type of cellular therapies are you particularly excited about?

Joseph Melenhorst, PhD: I think the type of therapies that are being evaluated in other cancer that we haven't treated yet. The really big challenge is in acute myeloid leukemia. With two of my colleagues, we are developing CARs against AML, which is very difficult disease, very heterogeneous. So going back to basics, what's the tumor like? What does it have? How can we evaluate CAR T development and responses? And I think one I'm also particularly excited about is, with all this innovation taking place in cell engineering, there's ways in more selectively targeting a tumor and those developments I think are going to be really meaningful in both AML as well as T-cell malignancies and solid tumors.

Dale Shepard, MD, PhD: So actually, of course, you know Tim Chan's been on the podcast as well, talking about the Center for Immunotherapy and Immuno-oncology. Where are we at this point within that program? In terms of resources that we've developed to answer some of these questions.

Joseph Melenhorst, PhD: So, Tim has developed a really robust foundation, if you will, to understand the therapies. So we use checkpoint inhibition like PD1, pembrolizumab or others. And he has developed a deep understanding of what the antigens possibly are for CDH hydrotic T-cells what proportion of cells recognizes these. So that mechanistic understanding is important. Identifying potential T-cell clones that are most responsive to, and de-correlates as well. So that part is a deeper understanding, but also the pipelines developed, like the computational immunology.

And that particular aspect nowadays, that's really important because we implement actually very complicated analysis that cannot... Well, can be explored by post-docs in the lab, but that really needs specialized analyst, if you will, computational biologists, really, to make sense of and develop machine learning algorithms. So I think that's really what some of the things that I see that Tim's done, and exploiting that in renal sub carcinoma and auto diseases too. And together with Brian Gastman, also part of the center, evaluating this in skin cancers, and Natalie Silver in mucosal cancers.

Dale Shepard, MD, PhD: Yeah. Excellent. More of a clinical question, I guess, but how do we make these more accessible? I mean, these are therapies that are sometimes difficult for patients to get to. How do we improve that?

Joseph Melenhorst, PhD: Yeah. Great question. Part of the cost for CAR therapies, cell therapies is hospital costs as you know. And therapy may cost $475,000, like KYMRIAH, or 360,000 for axi-cell. And that sounds like a lot, but the hospital costs at least doubles that if not more, so the therapy can quickly become a million dollars. And if we can find ways to minimize the hospital stay, that will be a great advance. While I know there's work being done in my lab to generate a more precise CAR therapy.

So that's what I'm particularly excited about. We have learned about cells that actually are driving the therapeutic advocacy. So we're using that now to generate these therapies, but also efforts in shortening the manufacturing process. So it takes typically, when I was starting on these therapies at Penn, we cultured the cells for nine days, and we've done studies also with colleagues there to shorten this to three days.

And that actually generated much more potent product, but of course reduce the cost of the manufacturing too. So there's a number of elements that we are using in the manufacturing that are reducing and fine tuning, which will reduce the cost of manufacturing, but also will make the therapy more predictable and easily to manage. So that means that you don't need highly specialized staff, maybe in the future to administer these therapies, but can be more broadly expanded. And automation of course, is the next thing.

Dale Shepard, MD, PhD: Right. We sort of talked about the individualization, and in many ways it's probably very, very helpful, but you often hear terms like, "Off the shelf CAR Ts." Is that something that is realistic? To be beneficial in a meaningful way?

Joseph Melenhorst, PhD: I definitely think so. Yeah, off the shelf CAR T-cells such as what I mentioned before. And in lymphomas, when we see the persistence of CAR T-cell is not necessary, we may actually be able to achieve a much higher response rate overall, if we administer those. And patient with CLL 75% or more, doesn't respond to a autologous CAR T-cells, and off shelf has definitely a role there. I'm not saying that I think that these will induce durable remissions. I think there's still a need for some other therapies, but at least as a bridge to another therapy, that might be me very meaningful. And I think off the shelf CAR T-cells also have a potential big role in diseases like T-cell malignancies and myeloid malignancies and solid tumors too.

Dale Shepard, MD, PhD: See, I haven't had the advantage of using these therapies, as a solid tumor guy, but I guess the question, especially when we start thinking about durability of the CAR Ts and things like that. Is the current thought, and you talked about following with other therapies... The thought, CAR T therapy, and then perhaps additional CAR T therapy or other cellular therapies? Or initial CAR T, but then having to boost with another type of treatment?

Joseph Melenhorst, PhD: I think that off the shelf CAR T for inducing a remission, and like in chronic lymphocytic leukemia where we know that the T-cell compartment oftentimes is dysfunctional, some call it exhaustion, I would like to refer it as just general dysfunction. There's such a high tumor burden, T-cell numbers are sort of normal, even sometimes enhanced in patients, but dysfunctional. I think what off the shelfs CAR T-cells will allow us to do is induce this remission, allow patients’ immune system to recover. And then as a second line, we could then think about autologous CAR T-cells, but it's a study that still needs to be done.

Dale Shepard, MD, PhD: We've talked about a number of different things that are being worked on to improve these therapies. What do you think is the biggest gap that gets us to that next plateau? This is a really big advance to have this as a therapy, and really revolutionize a lot of treatment. What's that next gap that's going to get us to the next plateau?

Joseph Melenhorst, PhD: I think implementing the precision manufacturing process, that's a big one. At least for CLL and multi myeloma and non-Hodgkin's lymphoma. I think for ALL, because lymphoblastic leukemia, we have seen patient responding really well. Up to 80 or 90% of patients, pediatric leukemias, achieving CR at one and three months after infusion. But we have seen also a great number of relapses, which are related often to a loss of antigen. That's a big challenge. So that's one challenge that the field is dealing with. I think that's going in the right direction. I think AML is still something that we really need to tackle, and finding out what works best in solid tumors. So there's a number of developments happening across different institutions.

Dale Shepard, MD, PhD: That's great. Well, you're doing great work and-

Joseph Melenhorst, PhD: Thank you.

Dale Shepard, MD, PhD: ... I appreciate your insights in this field today.

Joseph Melenhorst, PhD: Thank you.

Dale Shepard, MD, PhD: Thanks for being with us.

Joseph Melenhorst, PhD: My pleasure. Thanks so much.

Dale Shepard, MD, PhD: To make a direct online referral to our Taussig Cancer Institute, complete our online cancer patient referral form by visiting clevelandclinic.org/cancerpatientreferrals. You will receive confirmation once the appointment is scheduled. This concludes this episode of Cancer Advances. You will find additional podcast episodes on our website, clevelandclinic.org/canceradvancespodcast. Subscribe to the podcast on iTunes, Google play, Spotify, SoundCloud, or wherever you listen to podcasts. And don't forget, you can access real-time updates from Cleveland Clinic's cancer center experts on our Consult QD website at consultqd.clevelandclinic.org/cancer. Thank you for listening. Please join us again soon