Loading summary
Transcript15 lines
- [00:02]
A
Welcome to the American Society of Hematology Conversations with blood authors. This blood podcast episode is hosted by Dr. James Griffin from the Dana Farber Cancer Institute in Boston. He discusses alternative AAV gene therapy for hemophilia A using expression of BI ATE, a novel single chain factor VIII memetic antibody with Dr. Vincent Mucinsky and CD371 targeted car T cells secreting interleu 18 exhibit robust expansion and clear refractory acute myeloid leukemia with Dr. Mark Guyer and Dr. Susan DeWolf.
- [00:42]
B
Welcome everyone to the Blood podcast. We have two really interesting articles today to discuss. I'm Jim Griffin from the Dana Farber Cancer Institute and I'm introducing first doctor Vincent Wuchinski. And the title of this presentation is alternative AAV gene therapy for hemophilia A using expression of Bi8, a novel single chain factor VIII mimetic antibody. Dr. Mochinski.
- [01:09]
C
Thank you. The purpose of that research project was to develop a new gene therapy strategy for hemophilia a based on AVs. One of the limitations of AV based gene therapy is the size of the capsids which limits the size of the expression cassettes that we can actually fit in the capsids. It's limited to around 4.4.7 to 5 kilobase. So for very large transgene, which is the case of factor VIII in hemophilia A, this is a real challenge. We can find some options to reduce the size of the transgene that we put in there. But what it does is that it also limits a lot the space that we need to put regulatory elements. And if we want to manipulate those regulatory elements to improve expressions drive specificity, that becomes a bit of a challenge. We've always been interested in the potential of antibodies and the combinations of antibodies with gene therap. For hemophilia A treatment, there is a class of bispecific antibodies called Factor 8 mimetic antibodies that's been pioneered with imicizumab that has an outstanding potential at mimicking the functions of factor VIII by binding to Coagulation Factor 9 and Factor X. What we tried to do is to harness the potential of those antibody in gene therapy approach. So the first steps of that project was actually to transform and generate a bispecific antibody of small size. So going from a traditional IgG structure to a small single chain bispecific antibody that retains the full functions of the factor 8 mimedic capacity. So what we did is instead of having traditional structure, we add single chain viable fragments so SCFV start getting factor 9 and factor 10 that were fused together and that gave rise to a molecule that we call viate, which is a small single chain tandem over CFES that retains the full factor vitamin functions of factor 8 mimetic antibodies. The size of that molecule is about 54 kilo D. And we evaluated the biological functions of that molecule across a series of biological assays in vitro and ex vivo. So chromogenic assays that measure directly the conversion of factor X into tana, but also crossing assays in human plasma and mouse plasma to show that human molecule was retaining the activity. And what was really striking is that despite the complete change of structure of that antibody, it was indeed retaining the absolutely equivalent function to the original mestizoba antibodies. So that molecule was fully functional. And that led us to the second steps of that research project which was to transform that newly designed single chain factor VIII mimetic antibody into gene therapy approach. So the interest is that the coding sequence of biate is only 1.5 kilobase, which leaves plenty of space for the design of expression cassettes for IVs. We design a liver targeted gene therapy approach to cassettes. It's about 4.4 kilobase. It's using a strong HDR hat parameter that drives the liver specificity for expression. And we express when we produce viral particles using the AV8 capsids, which again has a high liver tropism. And we had a first series of experiments where we tested in vitro the transdictions and the expression level of the new expression cassettes in different models, but mainly in the HQH7 cells. And to show that the molecule that was the single chain factor epimetic antibody was generated by gene transfer at the exact same biological functions that the recombinant molecule that we could generate, which was the case, it means that everything was preserved the way it was supposed to be. We then tested that new gene therapy approach in an animal model. So for that we used the Hemophilia a mouse. We infected the mouse with different dose of AVs from 4 times 10 to the 11, all the way up to 1.2 times to the 13 vector genomes per kilos. And we looked at stability of the transgene. So over the course of eight weeks, measuring the steady state levels of the single chain factorate mimetic antibody in the mouse, we had a very nice dose dependent response in production, but that remained stable all the way until eight weeks. We verified that there was no anti drug antibodies developing over the course of that eight Weeks, which wasn't the case, confident that we could move to a functional evaluation of that molecule. So for that we use a breeding model called the tvt. So the tail vein transactions, where basically we sections the lateral vein of the tails of the mouse and we record the blood that is lost during the assay. So if you do that on anemophilia a mouse, you have anything in between 800 to a mil of blood loss. And what we could see was the three different dose of AV that we tested in the mouse. We had a successful dose dependent reductions in the bleeding rate of the mouse all the way to the highest dose of AV that shows almost no bleeding in those animals. The actual bleeding rates that we could measure was similar to what we could observe in wild type mice in a mouse that was treated with regular injections of melissizumab by IV injections, or a mouse that was actually treated with normal recombinant factor VIII therapies. It really works. It really gives us the possibility of developing a non factor VIII based approach for the treatments of hemophilia A. It has a lot of interest for us and a lot of potential for further developments. Because of the small size, we have quite some flexibility in the design of the expression cassettes. We have some rooms to manipulate the regulation elements, to manipulate the promoter that we want to use and insert in an expression cassette. So it gives us really a lot more flexibility. The fact that it is also a non factor VIII based gene therapy product means that this is the kind of treatment that becomes accessible to hemophilia A patients that have developed anti factor 8 inhibitors and that cannot be treated with regular factor VIII, but are also not eligible with traditional factor VIII based gene therapy approach. So that makes it really interesting. In the next stage of developments we are looking at improving that molecule because right now it's based on the time of a cfe, which is a small molecule, which is critical for the packaging size, but it also a molecule that has a relatively short half life. Tandem of scavs have a half life in between one to three hours in human, which and the consequence of that is that it's eliminated rapidly. So you need to have a high level of expressions to maintain a high level of steady states. What we're looking into is developing the next generation of that molecule where we fuse that molecule to an albumin binding peptide or an albumin binding moiety, for example, in order to extend the half life of that molecule. The consequences is that we're expecting to be able to achieve the same level of that molecule with a lower amount of viral particle, which is again a critical element to provide a safety aspect of the gene therapy and sustainability over the long term.
- [08:09]
B
Thank you very much. Dr. Meczynski, just a couple of questions for you. Emicizumab has been associated with some incidents of clotting. Would you expect the same potential issue from this approach and did you see any evidence of that in your mouse models?
- [08:27]
C
That's a very good question. So in short, the answer is yes, there is a chance that you will provide the same phenotype. So the molecule is based on the exact same biological functions of emicizumab. It is just the formats that we've changed, but the principle of model of actions remains the same. So it will be subjected or is likely to be subjected to the same kind of risk as what has been observed with emicizumab in the clinic. We've not seen any of those in our mouse model. But our mouse models are not thrombotic challenging models. So it is not unexpected that we haven't seen anything. But realistically if we were challenging the model on the thrombotic aspect, we were likely going to see the same kind of outputs.
- [09:09]
B
Thank you. One other question. What can you tell us about the potential immunogenicity of this single chain antibody?
- [09:18]
C
In our hands we haven't seen any immunogenicity, we haven't seen any ADAs developing over the course of the eight weeks. With the steady state level it's difficult to estimate without evidence. What is existing in the literature is that emicizumab itself has a very low level of anti drug antibody incidence even with the repeated injections. That is the normal course of administrations of emicizumab. And the tone of the CFS that we're using is a design that is inspired by the oncology space. So that series are heavily used in the oncology. As far as I know, there's not a huge amount of NTADAs being reported for Toniomas CFS. So the risk seems to be relatively restricted. But this is something that will have to be demonstrated experimentally.
- [10:06]
B
Thank you very much for that. Really very nice presentation and good luck with your future studies. We're going to go on to our second article which is entitled CD371 Targeted Car T Cells Secreting Interleukin 18 Exhibit Robust Expansion and Clear Refractory AML. We'll have two presenters, Mark Guyer and Susan DeWolf and Dr. Geyer is going to start us off on this topic.
- [10:33]
D
Thank you so much for inviting me to be on the podcast, Dr. Griffin. As this audience knows very well, CAR T cells have really transformed the way that we treat patients with relapse and refractory disease. B cell non Hodgkin lymphomas, B cell all and multiple myeloma. And we haven't yet seen those same results in patients with acute myeloid leukemia, despite of course having a tremendous unmet need in our patients with relapse and refractory acute myeloid leukemia. There's a few reasons why it's been harder to develop effective CAR T therapy for aml. First of all, we don't have a good antigen like CD19 that's just restricted to the leukemia. You can see blood count suppression because of on target off tumor effects, including effects on hematopoietic progenitor cells. And there's a more hostile tumor microenvironment in the context of aml. Now, there have been a number of series of patients with AML receiving CAR T cells across a variety of different targets and cell doses with inconsistent evidence of T cell expansion and activity. And this work here that we're presenting today arose from preclinical work led by my colleague Dr. Tony Danian, who developed in the laboratory a novel CAR T cell product that has three significant features. It has a fully human SCFV targeting CD371. CD371 is also known as Cleck 12A and it's CLL1 and it's present on mature myeloid cells as well as bulk AML cells and leukemia initiating cells that but appears to be absent or at very low level on CD 34 positive hematopoietic progenitor cells. Also, we used a co stimulatory domain which provides immunologic signal 2 on target engagement that helps to protect against T cell exhaustion because of the way that the domain is engineered. And then most significantly, the cells are engineered to secrete interleukin 18 which helps to stimulate interferon gamma production and which we think based on the preclinical work is going to enhance CAR T cell cytotoxicity, including activity against cells that might have very low levels of the antigen. This work basically represented the pilot phase of a phase one trial investigating these CD371 targeted IL18 secreting car T cells in patients with relapse and refractory aml. There's a couple things that are remarkable here for the five patients that we enrolled in the pilot study, many of Them had circulating blasts at the time that the T cells were collected and had relatively low lymphocyte counts. And so we were able from peripheral blood mononuclear cells, even in the context of a very low lymphocyte count, to be able to generate a CAR T cell product reliably in all patients relatively rapidly. And we were able to use doses of CAR T cells that are 10 to 100 fold less than in some other series that have been used for AML. Specifically, we had a starting dose of 3 times 10 to the 5th car T cells per kg and then had a step down dose of 3 times 10 to the 4 car T cells per kg. The 5 patients that we treated on this study all had very highly relapsed refractory aml, all had received venetoclax, all had received one or more high dose cytarabine containing regimens. And in all patients we were able not only to generate the CAR T cells, but to see them expand in vivo, which is something that has not been consistently seen in a number of other series. We saw peak expansion between day seven and 14 and we saw production of interleukin 18 peaking in that same time period. In three of the patients who actually were. The three of the five patients who had had a prior allogeneic stem cell transplant, we were able to see a morphologic leukemia free state at just 14 days into treatment without evidence of minimal residual disease, which is quite remarkable in these extremely treatment refractory patients. And those responses were able to lead two patients to be able to receive further cell therapy in the form of a stem cell boost for one patient and in the form of a second transplant for another patient. We did see prolonged cytopenias and that remains a challenge with the product. But the fact that we were able to clear these highly refractory BLAST populations with a very small dose of cells was quite encouraging in terms of safety. You know, for the first two patients that we were treated at the 3 times 10 to the 5th car T cell per kg dose level, we did see dose limiting toxicity in the first patient because of just a hypocellular marrow with low platelets and low neutrophil that persisted for six weeks post treatment that was able to be rescued with a stem cell boost. In the second patient we saw transient, albeit manageable, grade 4 CRS that was actually in a pediatric patient. That's why we went to the step down dose of 3 times 10 to the 4th per kg at which level we were still seeing T cell expansion. The CRS was seen in all five of the patients, was grade three in one patient, grade four in the one patient that I mentioned, and as I said, was manageable. Only one patient had ICANS in the study. The prolonged cytopenias are a challenge that we're continuing to navigate in terms of how best to incorporate stem cell boost or stem cell transplant into the algorithm here. And we think that that may be mediated both by possible effects of the CD371 targeting, as well as just effects of the interleukin 18 in affecting Myelo coiesis. Now, I'm also aware that some of our colleagues have presented other data suggesting that cytokines secreted by CAR T cells could actually fuel resistance in myeloid blasts. And, you know, that certainly has been an area of interest in the field. It's possible that that effect may be context dependent and may be sort of specific to that particular CAR T cell model. In our case, despite the very high levels of Interleukin 18 that we were able to generate from these cells, we were able to see blast clearance in three of the five patients, which is encouraging. And we've continued that dose level to completion at 3 times 10 to the 4th per kg. And we are excited to reopen the study soon with a new cohort. I'm going to hand things over to my colleague, Dr. DeWolf, who led our translational science efforts.
- [16:41]
E
Thank you so much. In the context of these research, really exciting clinical outcomes, we wanted to delve deeply into the underlying biology. And what is unique about this therapy is it's an armored therapy. We have car ts that are also secreting IL18, which is interferon inducing factor. And so this gives us a very interesting opportunity to understand what IL18 is doing not only to the cars as they proliferate in the in vivo, but also to the tumor microenvironment itself in these patients and to the surrounding immune compartment. In parallel to the clinical work, we did multimodal analysis from longitudinal primary patient samples over the course of therapy that we were collecting in real time. And this included flow cytometry, single cell RNA sequencing, and T cell receptor sequencing, as well as extensive cytokine profiling, as Dr. Geyer mentioned. And just to sort of walk through some of the highlights of what we found, not only did we see in vivo proliferation of cars in many of our patients, but these cars had a highly cytotoxic profile, and that profile was different than what the T cells looked like compared to the end of production material. So there was a real change in the nature and the sort of tumor killing capacity of these cells in vivo. Furthermore, we looked at what types of pathways were being upregulated by these cells and the number one that stood out was interferon pathways. Not surprisingly, in the context of IL18, Dr. Geyer mentioned that three of our patients were actually patients who were treated in the relapse post allo setting. And one of the things we wondered is whether the cars that we found in the patients were actually of donor or Horst origin. And using our single cell approach, we were able to disentangle essentially all of the CAR T cells that proliferated in vivo were in fact of donor origin. This is really not only fascinating biology of this state of mixed chimerism in these patients, but also really clinically relevant because none of the patients developed graft versus host disease. So in spite of the fact that these were donor T cells, we did not see any clear secondary organ toxicity in the context of their proliferation. The other thing I'd like to mention is that in addition to seeing these highly cytotoxic things, CD8 T cells, that we think is part of the reason these cells were perhaps effective in during this highly refractory disease, is that we also saw a change in the NK cells. NK cells are a cell type notorious for their traditional anti leukemic activity, particularly against acute myeloid leukemia. And what we found is that the NK cells, when we profiled them one to two weeks and after infusion of CAR T is that they too were highly cytotoxic, very activated cells. Again suggesting that perhaps some of what is going on is remodeling of the immune environment in the context of this therapy. We think that we've learned a lot about perhaps why these cells are working. And I also think that it's a really nice proof of concept that there's so much basic biology that can be learned in these highly unique clinical trial samples. We really were able to study in very quick turnaround primary patient material.
- [20:38]
B
Very nice. Really an elegant set of studies taken to patients already and very encouraging results. Can you tell us a little bit more about the antigen? There's only a handful of antigens that are really considered to be specific for the hematopoietic system. The second part of that question is given the neutropenia and thrombocytopenia that occurred, are we sure that there isn't much expression or any expression on hematopoietic STEM cells. What's CD371?
- [21:08]
E
CD371, also known as CCL1, also known as CLEC12A, is an antigen that we see on many mature myeloid cells, but is actually known to have some expression on, you know, stem and progenitor compartments. So it's certainly not something that we think of as entirely sparing of stem cells and therefore may be part of of what we're seeing here in terms of what its function is. It's actually, I think, something that literature is rather limited on. I think there are some data that it may be important in immune response to actually uric acid crystals.
- [21:49]
D
It's involved in sensing monosodium urate crystals. That appears to be one physiologic function. How relevant that is to leukaomogenesis, I don't know, but basically came up in all of our screens as among the sort of most ideal targets that you could come up with. In a sense, that appeared to be HPC sparing and appears to be present on most bulk leukemia cells. You know, and frankly, even in the mirroring model, even in a model where there's not uniform CD371 expression, even partial CD371 expression is efficient to generate expansion of the cells. And you have some off target effects as well, even in the antigen negative or antigen low cells, just with the IL18 secretion probably improving the endogenous immune responses against those cancer cells. I think that it's probably multifactorial. Why these patients have cytopenias. These patients generally have poor marrow reserve from pretreatment. And then we know that interferon gamma can interfere with myelopoiesis. In fact, there's been some elegant work showing how that's relevant in the pathophysiology of graft failure after hematopoietic cell transplantation. And so I suspect that that is probably part of the story as well. In addition to any direct effects of the CD371. That's part of the reason why we're working in close coordination with our transplant colleagues regarding giving these patients a stem cell boost or potentially a second or first transplant shortly after they are demonstrating response to this product. Just so that we're able to basically capture these patients in a window where they have disease control, but to help spare them from some of the consequences of cytopenias.
- [23:19]
B
Thanks very much. Terrific presentation, both of you. And we'll conclude the podcast here.
- [23:26]
A
Thank you for listening to this blood podcast of conversations with blood authors. To read the articles, please visit bloodjournal.org this episode is copyrighted by the American Society of Hematology.