All righty, let's get going. Thank you all for coming. I know you guys are the diehards of the meeting, and we appreciate you staying through the end. It's been my pleasure to organize this session on viral oncolytic therapy, which I think you'll find informative and educational. I think there's so many things about oncolytic viruses that are so promising from a standpoint of cancer therapy, in that there's the promise of selective tumor replication, natural tumor, anti-tumor immunity, and the potential to reduce the need for these very genomic-based personalized cell therapies. So many of these are in clinical trials. And I think we're all fortunate as a group and as a specialty to have leaders of the field within our own specialty. And so it's been a real honor to have the various people pictured here speaking on their work, and really their lifetime's work, in designing therapies and moving the field forward as it pertains to oncolytic viral therapy. So I don't want to take time related to their talks, so we'll get started with Dr. Martuza. Steve will introduce you briefly. Thank you, Rohan, and congratulations not only for this session, but for the entire meeting, which you were the scientific program chair for. And on behalf of the tumor section, welcome, everyone. It's my honor to introduce Dr. Robert Martuza from the Massachusetts General Hospital, who's going to talk to us about the history and potential future applications of oncolytic virus therapy. Thank you. Steve, Rohan, thank you very much for organizing this. So as noted on the slides up here, I'm going to talk about how this all started, where I think it's going, and to show you that this is really an iterative process that's still ongoing and is not completely done yet. So I'll talk about the conception of it, how it all started in a paper in Science 91, and then this concept of in situ vaccination with three papers that we did in the late 90s. And then the concept of engineering a virus to kill not just the bulk tumor cells, but the glioblastoma stem cells. And then the synergy with immune modulation in a recent cancer cell paper that we had, and others. And then finally, going to mainstream cancer therapy. So that'll be sort of the genre of what I'm doing in this talk. And then you'll hear from a number of my scientific colleagues and neurosurgeons who have done some creative stuff with both the same virus type, as well as other viral types as well. So let me go forward. So this is our current lab group that you see here. These are my disclosures over here. And the people, there are many people in the past that have been involved with this. In this short period of time, I'm going to talk about the ones that are listed in yellow here, or that are underlined up top. So this all started as I began to look at what the physicists call symmetries. If you have a up quark, you know there's going to be a down quark. If you have a positive particle, there'll be a similar negative particle. And the symmetries in cancer are such that things that cause cancer and the things that cure cancer are related to each other. So for example, in terms of chemicals, if you give nitrosourea to a rat or a mouse, you get a glioma. Back in the 80s, when we began this work, the main therapy for glioblastoma was BCNU, nitrosourea. If you give radiation to an animal or to a human, you get tumors. And the first thing we do is give radiation. And again, both of these things have effects on the DNA that are somewhat nonspecific. And they can cause a tumor, or they can cure the tumor. But at the time, no one was using viruses, despite the fact that viruses, in fact, can cause cancer. So the question was, could we somehow create viruses that were safe and that would, in fact, be able to treat cancer, and in the case of neurosurgery, glioblastoma? So looking back at what had been done in the even distant past, I call it take a walk on the wild side. In the late 19th century, there were viral infections that were noted to be associated with aggressions of cancer. In the early 20th century, various viruses had been noted to be associated with cancer, remissions, rabies, influenza, varicella, measles, hepatitis. In 1952, West Nile virus was used. In 1956, adenovirus. In 1974, mumps. And in all of these cases, there were regressions, but these were wild-type viruses and associated with the toxicities, and particularly neurotoxicities, that we associate with these viruses. So basically, what happened was surgical techniques chemotherapy, radiotherapy could be intentionally modulated. Viruses, at that point, could not. And the toxicities of these wild-type viral trials led to abandonment of viral therapy for cancer. And surgery, chemotherapy, radiotherapy basically became the mainstream therapies for cancer. So the first studies that we did were with non-replicating retroviruses. They were very inefficient. We never took them to human trial. Others did. They showed they were inefficient, but we knew that from the preclinical models. So the question then became how to amplify this effect. And the answer was really to have the virus replicate. We even published this one paper using a replicating retrovirus. Now, you've got to realize that at the time this was done, this was the beginning of the HIV-AIDS era, where it was skyrocketing. And we were concerned about lymphoid, germline, insertional mutagenesis, and FDA never approving something like this. So we really abandoned that. Bob Carter gave a great talk yesterday on using the TOCA5 virus, which is a replicating retrovirus. These issues of insertional mutagenesis are still not answered for that. So that still needs to be addressed. But we looked at other viruses, some of which I've already mentioned, and finally settled on using herpes simplex virus. These are some approximate sizes of various RNA and DNA viruses. And you can see that DNA viruses are, in general, much larger. Basic concept, as you all know, is that basically you engineer the virus so that when it goes in the tumor cell, it kills the cell, makes multiple copies, 1,000 copies come out, it spreads, et cetera. Whereas when it goes in the normal cell, if it's engineered correctly, it doesn't harm the cell, or at least doesn't replicate and spread. We chose herpes simplex because the DNA viruses are much more stable than RNA viruses, very large. There's about a 30 kilobase area that you can insert additional genetic material. It doesn't integrate, so you don't have the worry that you had with retroviruses. We knew that herpes could spread in the brain. So if we could at least control that, we might have the possibility of multiple flavors of viruses, replication, defective, latent. And you could use them not just for cancer, but for other neuromodulation uses in neurosurgery. And very importantly, antiviral drugs are available for herpes viruses, acyclovir, valacyclovir, a bunch of others. So we thought if we had a problem, we could turn it off. And that was the reason we used this. Now, you'll hear from a variety of the other speakers about other viruses, which I think are very interesting as well. So this is the first paper back in 91. And basically, this is mice that had a human glioma and were inoculated with a virus, with one inoculation. And you can see survival curves with no virus, 10 to the 3 virus, 10 to the 5 virus. And you got this dose effect curve where you finally had cures in these tumors. And we basically said the following things in that paper. A virus could be genetically engineered to kill cancer cells, in this case glioblastoma, without harming normal tissue. Mechanisms include direct cell killing by the virus and production of new antigens on the tumor to induce immunologic rejection. And that genetically engineered viruses are worthy of further exploration as a means of therapy of tumors that are resistant to currently available treatments. And what I'm gonna show you and what the other speakers will show you is that basically this has all come to pass. However, at that point we entered what I call the dark ages of oncolytic viruses. So I would apply to the National Cancer Institute and they would say, well, glioblastoma is not a cancer. It doesn't metastasize, so we don't want your grant. And I would send in things to gene therapy meetings and they'd say, well, this is not really gene therapy. You're not inserting any genes. And I would say, well, the herpes virus has a lot of foreign genes, but not gene therapy. We applied to pharmaceutical companies. They said it was too risky. Ultimately, if a company's foreign, we'll buy the company out and that'll take care of that. Nevertheless, we persisted, as Mitch McConnell once said of Elizabeth Warren. Anyway, it really took off when other groups could confirm this. And that was about four or five years later with herpes virus and with an adenovirus. And that really set the stage for a whole bunch of other studies, as opposed to people just in our group who were doing this. So the next virus that we made, other than that first one, which was a thymidine kinase deletion mutant, was this virus called G207. This was in 1995, it was published. And this is, I won't go through the map of herpes virus, but basically it has two mutations. They're very widely spaced. They affect different viral functions. So the chance of it mutating back in the patient to a wild type virus was essentially nil. And the mutations included gamma 34.5, which is the main thing that causes herpes encephalitis. So you take that out and you have a very safe virus. And then it also had an ICP6 mutation, which selects for replicating cells, as opposed to cells that are not replicating. So glioblastoma versus brain, if you will. The next thing that we did, which was a number of years later in 99, 98, 99, was introduce this concept of in-situ vaccination. And basically, I'll show you what this means. It means that if you inject one tumor, you get an immune response so that if there are tumors elsewhere, they will also get rejected. Not because the virus goes there, but because the immune response that's generated by the virus growing in cancer cells takes care of the other tumor as well. And we did this with the virus alone. We also did it with a cytokine called IL-12, which I'll get into later, and I think Jim Markert may talk about as well. And also, we did it with GM-CSF, which is the basis of the current Amgen vector. So here's an example. This is a bilateral colon cancer model. So there are colon cancer in these mice on both sides. And G207 not only decreases the growth in the inoculated tumor, but somewhat less so, but still substantially and very significant in the opposite tumor, and that's with a single injection. And I don't have the data up here, but if you want to go back to the paper and look, we showed that you had to have the virus growing in the tumor. If you inoculated in the skin, even just next to the tumor, it didn't work. We could show what the tumor antigen was that was expressed when the virus killed the cell, and we could show that lymphocytes were necessary. If you depleted them, you got rid of this effect. So that was this concept of in situ vaccination, which I think has become a hallmark of some of the work that we're doing and that others are doing now. And again, this is the work with IL-12. Again, you can see this dose effect as doses increase with GM-CSF as well. So you could potentiate this by putting a cytokine in that would stimulate the immune system. And then this finally led to a paper, Jim Markert was the primary author, using this virus G207 that we developed, inoculating into glioblastoma and in patients with recurrent pathology-proven glioblastoma and showing that there was no evidence of herpes encephalitis by MRI or clinical exam. There were apparent MRI responses in about a third of the patients. There was one histologic complete response in two long-term survivors, and there was no chronic toxicity that we could see. So this kind of set the stage for where we wanted to go, but obviously patients were still dying and the problem wasn't solved. So that then led, as I said, as an iterative process to the development of this virus called G47 Delta. Everything I'll tell you about in the next number of slides deals with this virus. It's basically that prior virus, G207, with one additional mutation called ICP47. And that mutation causes early expression of a gene called US11, which complements that gamma 34.5 genetic loss and therefore enhances growth and increases oncolytic activity. It also has an immune effect by inhibiting MAC class one presentation. And importantly, it allows replication in the cancer stem cells, okay? So this is critical. So G47 Delta replicates in glioblastoma stem cells. If you take, these are human glioblastoma stem cells raised as spheroids, MGG8, BT74, and others. But if you put that earlier virus, G207 in, you only get out the same amount of viruses you put in or sometimes even less. If you put in G47 Delta, you get an increase in the virus and so you get a spread. And you can see that here in the spheroids of these MGG8 spheres on 24 hours and 48 hours where GFP is expressed. And Cole Peters, who's currently a postdoc in the lab, has a paper that is submitted, not yet accepted, but it's in revision, showing that the gamma 34.5 deletion that we initially had in G207 will allow growth in the monolayer glioblastoma cells, the bulk cells of the tumor, but not in the glioblastoma stem cells. And it's due to a block during translation of the true late genes in the glioblastoma stem cells. So this is, I think, critical for taking care of recurrent tumors. And G47 Delta, the one I just mentioned, is currently in trial at the University of Tokyo. Tomoki Toto was formerly a postdoc with us. He's now a professor at the University of Tokyo. And basically, this is the scenario. I won't read it through. You can read it yourself. And some patients are showing remarkable results. Here's a patient with a recurrent glioblastoma. And what you see in the ones that have good results is a pseudoprogression initially, which is really the immune response. And then ultimately, you get a further regression, and ultimately, in this case, a complete regression. So this trial will be evaluated by the Japanese version of the FDA this summer, and hopefully will be approved and go on to other trials in various other tumors. It's also in a trial for prostate cancer, I might add. So since that time, here are a few examples of some viruses that we've made and studied. You can make it with increased replication, I've mentioned. You can make it with specific promoters, so it only grows in prostate cancer, but not breast cancer. You can make it with retargeting of the receptors. You can target various specific cancer cell pathways. Pick one and make a virus that'll target that pathway alone. But the problem is, I think, for glioblastoma, targeted therapy does not work. You know, you get rid of the EGFR variant three cells, and other cells just take over, okay? Targeted therapy is great for the various liquid tumors, leukemias, you know, some lymphomas, et cetera. And you can use CAR T cells, et cetera. But when you have a complex heterogeneous tumor, it's not ideal. So I'm gonna talk to you now about G47 Delta designed with vascular and immunogenic properties, as I mentioned down at the bottom there. So to do this, you need a immunocompetent GBM model. And this is the OO5 model that was developed by Indervirma and Maramoto and published in Nature 09. You take mouse OO5 glioblastoma stem cells, which you can see down here, and you can use them to populate tumors in the brain of syngeneic animals. And then you have an immune-competent animal model. IL-12, as I mentioned, is a pro-inflammatory cytokine, anti-angiogenic, causes vascular normalization, promotes proliferation of T and NK cells, stimulates Th1 differentiation, induces interferon gamma. But if you give it systemically, it's toxic. It's shown to be effective in cancer, but it's toxic systemically. So these viruses are perfect for things that work in cancer, but you can't give them systemically. So we made this virus containing, expressing IL-12 with a CMV promoter, Tubachemos. The postdoc published this in PNAS 2013. And basically what it does in these OO5 tumors is it normalizes the vasculature, decreases Treg, so that's good for the immune response, increases survival. So if you look here, this is the quantitation, this is the histology, and you can just see the reduction in the vasculature with the IL-12 vector compared to the vector without IL-12, E stands for empty. And you can see the reduction in Tregs as well, and you can see the increase in survival, okay? And we could show, I don't have this data up here, but the anti-tumor activity is dependent upon T cells. So this becomes sort of a key item in our thought of going forward. Now the next thing that happened in cancer therapy, as you probably all know, is checkpoint blockade. And again, this gets away from the idea of, you know, people for many decades trying to find a very specific antigen and getting rid of it. With checkpoint blockade, basically, you take the brakes off the immune system, and you say, let the body figure out what the antigen is, and don't worry about having a specific antigen involved. And the two that are most commonly used are anti-CTLA-4 and anti-PD-1, and they go by various trade names, which I'll try not to use, because I frankly find them more confusing. Every company has their own name. The problem is these checkpoint inhibitors are only effective in immunologically hot tumors with high mutational lows and preexisting immune response. They're not effective in cold tumors like glioblastoma. So if you look at this paper from Lawrence, for example, it's the high mutational low tumors like melanoma, some lung cancer, in which they work. And that's the problem, because glioblastoma is way down the line. So we found that if you use a checkpoint inhibitor like PD-1 or anti-CTLA-4 with this virus, you get a slight improvement in the overall outcome. But we hypothesized that if you could use two of these, plus the IL-12, you'd get an incredible result. And so here's one viral treatment, three checkpoint inhibitor doses of two different pathway checkpoint inhibitors, CTLA-4 and PD-1, and you get 90% survival in this otherwise really stringent model. And if you re-challenge them, you can't grow a tumor in the other hemisphere. And then this could be repeated with CT2A gliomas. And again, you get immune memory. So we think G47 delta is ideally suited. The base vector grows in the bulk tumor cells, takes care of heterogeneity. The ICP47 deletion allows the stem cells to be killed and express antigens. And the IL-12 expression takes care of angiogenesis, the immunosuppressive effects and the recruitment of Tregs. And so basically what you're doing is using this concept of in situ vaccination to turn a cold tumor hot, okay? Now, none of these things work unless companies get interested in it. And finally, a few years ago, a number of companies became interested, the biggest being Amgen that bought BioVax for about a billion dollars and then ultimately got FDA approval of TVEC, which is a herpes virus expressing GM-CSF. And I'll just mention two recent papers. This one was by Chesney. Basically, TVEC plus anti-CTLA-4 therapy doubles the objective response rate in metastatic melanoma. More patients had a complete or partial response in the combination arm. Significantly more patients had disease control in the TVEC plus anti-CTLA-4 arm. Another recent paper by Ribas and Cell, similar story. The TVEC plus anti-PD1 therapy in this case changes the tumor microenvironment. You get increased interferon gamma, CD4, CD8. Responses are independent of baseline CD8. So if you just treat people with a checkpoint inhibitor, then patients who have a low interferon score or don't express PD1 don't respond. In this case, if you add the virus to it, you get complete responses in some of those patients. So TVEC plus anti-PD1 and metastatic melanoma double the response rate relative to anti-PD1 therapy alone. So where does this go? Herpes is large. You have room for a lot of inserts. This is a partial list of other immune modulators and cytokines. I just mentioned, oh, I don't know what happened here. Anyway, I just mentioned the two that were there. Can we get to that prior, can we get to that prior page? I might've clicked something. Can we get to the prior slide page? I don't know if the person is, there we go. Great, perfect. So anyway, there are a lot of things that could be inserted, cytokines, et cetera, because again, as you start loading these up systemically, you get various complications as shown down here, and some of them are fatal like myocarditis. And we recently had the 11th International Oncology Virus Conference. There are now more than a dozen viruses in clinical trial. At least half of them are being used for glioblastoma plus other tumors. So basically where I see this going is we currently have approval for melanoma, skin cancer. We're looking at brain cancer, but I think all of these others are potential targets. There are multiple oncolytic viruses in human clinical trials worldwide, already showing clinical improvement as I showed you. It's a unique therapy, and we have to change our ideas about use of steroids. Pseudoprogression can be a good thing. Xenoantigens can be involved. Multiple targets can be attacked simultaneously. The microenvironment can be targeted. It's an iterative process. And importantly, there are multiple synergies with other therapies such as immunotherapy, chemotherapy, radiotherapy, anti-VEGF, et cetera. So you'll hear from my other colleagues about a number of ways they are looking at this in some exciting talks to follow this one. But remember, it started in neurosurgery with neurosurgeons. Thank you.

oncolytic virus therapy

selective tumor replication

anti-tumor immunity

in-situ vaccination

immune checkpoint inhibitors

clinical trials