First of all, thank you very much all of you for attending. This is a great turnout on a Sunday night, at least I think it's Sunday. I'm not sure what day it is. I've lost track. We appreciate you all coming. We have, I think, a very interesting agenda for you tonight. As you know, we presented, I think, a really beautiful paper this morning presented by Dr. Luke, who's joining us this evening. Dr. Luke, of course, did this work while at the University of Chicago, but he's recently moved to the University of Pittsburgh. If you're a Steeler fan, you'll be happy. The paper today that Dr. Luke presented makes me extremely proud of the work of our collaborators and the work of people at our company.
For those of you who were there, and I've been doing this for 30 years, it was a great demonstration of a really nicely conducted phase I study that had beautiful safety, early assessments of efficacy, and beautiful correlative biology and immunology. It doesn't get any better than what you saw today, in my opinion. I think others in the room who are also our collaborators, who I'll introduce in a bit, would agree. That really doesn't get any better than this. What I want to do tonight is, I've asked Dr. Luke to run through the presentation again because I know a lot of people, ASCO's very busy, and a lot of people probably didn't have time to attend this morning. We'll run through the presentation quickly.
The slides are up on our website, you don't need to take pictures or take notes because it's all there for you to look at and download and do what you want. I've asked Dr. Luke to run through the slides. I'd like to hold questions until he's done, and then we can open it up for questions. I also have some slides that I want to run through that sort of put the data in context. As you know, anti-CD73 antibodies are sort of a hot area now. Targeting CD73, there's probably 10 or so companies working on this. What you saw today in our presentation was really very novel, unique behavior of this antibody.
I'll go through what those things are and why we think they're advantageous and why we think that they're going to play a great role in therapy, both in oncology, perhaps even other diseases as well. With that, I know it's already getting late. I want to turn it over to Dr. Luke. Before I do, I want to just make some quick introductions. We're also joined by Dr. Mehrdad Mobasher. He's been with us a few months. He's our Chief Medical Officer. Janet Ko, our Director of Product Development. Dan Hunt, VP of Business Development. Also joining us tonight is the most distinguished person of all, Lauren Harshman, who's from Harvard University. No, Lauren is from Dana-Farber Cancer Institute, and she's a genitourinary cancer expert, renal cell cancer and prostate cancer.
You'll see why we're interested in those diseases. I'll have to be careful what I say because Lauren's here. Lauren also trained at Stanford. We know she's a very well-trained oncologist, right? What's that?
I'll keep you honest.
Jason was not trained at Stanford, nonetheless, he's still a pretty good oncologist. With that, let me turn it over to Jason. He did?
I was at Harvard before she was.
Oh, is that right?
Yeah.
Awesome.
Appreciate all of you taking the time to come out. As was mentioned, we're going to go through the slides for the presentation. A bit less formal than the actual talk, if you want to interrupt, you want to ask questions, it's fine. We'll try to just get through the slides. If it's pressing, ask. Otherwise, we can wait. Sort of just to hit the highlights of the slides as we presented them. CPI-006 obviously is the molecule anti-CD73. Really talking about immunomodulatory properties of that and then the phase I results as they've been described so far. To set the stage on this, you can see on the right-hand side this cartoon that's busy, but important in identifying several aspects of what we think is important in this biology.
Obviously, everyone is aware of the importance of CD73 in this conversion of AMP to adenosine in the tumor microenvironment, adenosine being known to be immunosuppressive. One aspect of the activity here is to block CD73 to interrupt this catabolism of AMP to adenosine. More broadly than that, CD73 is expressed on a number of different cell types, including T and B cells, and has a number of other functions related to lymphocyte adhesion and migration and activation. We highlight that here, where CD73 is known to interact with other ligands and potentially can act as a co-stimulatory molecule for T cell activation. That's another aspect of CPI-006 that's important. The antibody itself is a humanized IgG1 Fc gamma receptor deficient anti-CD73. An important point there is that there's no ADCC function.
When you see the later data, it's not that the cells are being depleted. Obviously block this catalytic activity, we can agonize immunomodulatory activity here. You're all also aware that ciforadenant, or CPI-444, is the A2AR inhibitor that's previously been investigated by Corvus, with activity as monotherapy and in combination, and importantly with this adenosine gene signature that correlates with treatment response. It really will be an integral biomarker moving into the future. All of this to say there are multiple roles where adenosine could matter, CD73 could matter. The interaction with antigen-presenting cells is going to be important, particularly B cells, throughout the rest of the data that I show.
CPI-006 can interact in a number of functions here. So just quickly, preclinically, looking at CPI-006 in terms of CD73 catalytic activity, you can see that at increasing dose, you get a decrease in free adenosine interrupting the catabolism as you would expect. Interestingly, the molecule has monotherapy activity. This is MDA-MB-231, is a triple negative breast cancer orthotopic model. From those tumors, importantly, when you take them out, if you expose them to a non-competitive CD73, you can see that the antibody diffuses through the tumor and binds to the antigen. In the context of this competitive blocking assay, the epitope is blocked, and the enzymatic activity is interrupted, as is shown on the right-hand side. The immunomodulatory activities, as I mentioned, are really highlighted on this slide, which I really think is an interesting one. These are healthy donor PBMCs treated overnight with the antibody.
What you can see is an induction of a number of activation markers, particularly on B cells, CD69. I'll also note that besides being an activation marker, this molecule is important in migration and trafficking of B cells, and again, that will become important in a second. Broadly, their activation marker is 69, 83, 25, as you can see here, upregulated, as well as antigen presentation markers, 86 and class II MHC. Interestingly, these effects upregulating these activation markers are independent of the adenosine axis, as shown on the bottom panels here, where 69 and 83 are moved to the right here despite co-treatment with this high-potency adenosine agonist. We're not moving that back to the left, which means this is a different effect coming here.
These lymphocyte markers are consistent with activation of B cells, which a number of these markers are associated with B cells and other antigen-presenting cells. That led to the clinical trial design, which I won't belabor. It's a three plus three dose escalation. In orange, you can see the dose levels that have already been accomplished as monotherapy as well as combination. There'll be a third combination with pembrolizumab that will be initiated. All these are current and ongoing. Dose expansion group's in the middle. Just in terms of dosing, the molecule is a one-hour infusion every three weeks with ciforadenant, as previously described. The eligibility is pretty standard for a phase I population. We will note the CD73 expression, which is not required in dose escalation but is a potential marker to be integrated as a biomarker into the expansion cohorts.
Objective's fairly standard, biomarker assessments are noted on the bottom. We'll talk about some of them throughout the rest here. Talking about patients treated on the study, you can see monotherapy in the middle column and the combination on the right-hand side for CPI-006 or with ciforadenant. Again, it's a general phase I population, but we'll note that four prior therapies on median for each one of these groups. These are a heavily pretreated population. If you look at the tumor types, I won't read them out, but these are tumor types that generally speaking, are quite aggressive in the refractory setting. This is a pretty heavily pretreated population. In terms of adverse events that we've observed, you can just look at the numbers and see that it's been very well tolerated to date with mostly low-grade events.
All grades here in the middle, not so many even in combination, Grade 3 events really only just here with anemia. This has been very well tolerated to date, and I think that's important as we start to think about adding in other therapies like anti-PD-1 or the combination with ciforadenant. These are PK and receptor occupancy data. What you can see is that with increasing dose, we have increasing exposure to the drug, as you'd expect. There's an interesting phenomenon here where clearance appears to decrease. It's not so clinically important, but it's an interesting PK profile. We see that the CPI-006 is detectable at the 6 milligram per kilogram dose level out over 21 days. In terms of the receptor occupancy, this dotted line shows 100% receptor saturation.
You can see that doses we've already gotten to, we get 100% receptor saturation at 6 milligrams per kilogram or higher. Again, similar to what we previously showed from the peripheral blood, you can see that upon binding of the antibody here, the total cell surface CD73 doesn't change. This is kind of important. Whereas the free cell surface goes basically to zero, which means that the epitope for binding is totally blocked, but the receptor is still there. That's important when I think Richard's going to talk about some other antibodies in class. This is different than what has been described to date for other antibodies with just a different mechanism. From tumor, we see very similar. This is a patient treated at 12 mgs per kg, colorectal cancer. This is from a tumor biopsy.
Again, we see the same phenomenon where the antibody's able to penetrate and bind to antigen. The epitope's blocked by this competitive antibody binding. In the context of competitive antibody binding and CD73 enzymatic activity being eliminated. Really, we think about we've got preclinical tumor, we've got peripheral blood, and we've got the tumor from the patient. They're all showing the same thing throughout, which I think is what Richard was alluding to with really the beauty of this study, which really goes from preclinical translation all the way through the patients. These are the disease assessments for patients treated on the study so far. You can see the dose for monotherapy and combo on the right and dose increasing here, here. 1 through 12 milligrams per kilogram. Again, 6 milligrams per kilogram is 100% receptor occupancy on the dotted line here.
Again, would highlight this is early in the study, right? At the 6 milligram per kilogram is the first dose where we're hitting full saturation. You can see as we're going up, there's a suggestion here. It appears that we're getting longer-term benefit in these patients. Some of these patients have been on this therapy for a long period of time. What I remember mentioning before, these were highly refractory patients, more than 4 lines of therapy. It's very unlikely that this is just stable disease just because these are people that progressed through a lot of stuff before they went on the study. Similarly, this combination, even at earlier dose levels with ciforadenant, we're starting to see patients going out longer than you would really expect for just stable disease on a phase I trial.
These patients are out to 6 cycles on therapy. Would point out there are a number of patients, renal cell, that will be discussed a little bit later that are ongoing on the study, and there's high hopes that this could be an interesting combination approach there. From the peripheral blood in patients, we see some quite interesting changes in peripheral blood profiles. You can see in the CD73 positive B cells, which was sort of mentioned a few times, there's this rapid reduction in the peripheral blood compartment of these cells. There are also changes in T cells and other sorts. You can see CD73 positive and negative CD4s, which appear to go up. CD73 positive CD8s are basically stable, and CD73 negatives are slightly lower. Monocytes are also lower, consistent with the B cells.
We see that this is trafficking of these CD73 positive B cells out of the peripheral blood, we're really going to talk about that later, and redistribution of both those B cells and monocytes into different compartments. This is going to become important. As well as an increase in this ratio of 4 to 8 cells, we think that that could be important in terms of mediating antitumor activity as well. Profiling these changes in B cells, we can see that after treatment, we see this initial drop and a return to a new steady state thereafter, and that's true at each one of the dose levels. This appears to be an on-target effect independent of dose. Very interestingly, however, as I just mentioned, the B cells do come back to the peripheral circulation, and when they do so, they have a different phenotype.
Highlighted here from these three patients treated on the 6 mg per kilogram dose cohort, where upon return to the peripheral circulation, you have an increase in HLA-DR class II MHC expression. These are activated B cells that are ready to help prime an antitumor response when they return to the circulation. We're going to talk about what that could mean, we think maybe this is consistent with redistribution of these B cells into lymphoid tissues, where they may be able to prime immune response, and when they return, they have this profile. This is a patient with prostate cancer, a 72-year-old man treated with a 6 mg per kilogram dose cohort, again, refractory to standard therapies that are outlined here. You can see in this patient, we get this drop in the CD73 positive B cells.
It returns, as I discussed, to a new baseline and continues then to stable out over time. In this patient, we see a reduction in the patient's overall tumor burden with these paraaortic lymph nodes decreasing. Upon initiation on this study, it's interesting, we saw that the PSA velocity started to change, and then there were some fluctuations over, but the patient clearly benefited. The treating investigator's report is this guy had a fairly significant pain syndrome that essentially has resolved now, and this guy's going on now more than 11 cycles. Very unlikely this patient just has stable disease on a fatal trial for more than 11 cycles with a improvement in overall pain syndrome, like I mentioned. The treatment in the peripheral blood has also been associated with this cytokine profile that really suggests an immune activation phenotype.
You can see early on treatment at a half hour and two hours, the induction of a number of cytokines consistent with an early inflammatory response. These then sort of expand into a broader inflammatory response out at 24 hours and 8 days. You can see that sort of here with this initial burst and then sort of a broader expansion of this phenomenon out over time. Again, consistent with an inflammatory response to treatment associated with those changes in the blood cell types that I mentioned. To try to put this together then, what do we think is going on here? Excuse me. CPI-006 is obviously anti-CD73, it binds and it can block this conversion of AMP to adenosine. That's clear, and that's what many people in the field have been focused on so far.
In addition, I think it's becoming very clear, and we try to show you here, binding to CD73, however, has the potential to lead to activation of multiple immune cells. Here we show B cells, where you can have multiple pathways that can induce this activation with upregulation of CD69. If you remember at the beginning, I talked about CD69 being an activation marker, but also a migration and trafficking marker, and I think that's important. CD69 is a known regulator of S1P1, which is a molecule that helps to mediate egress of B cells from lymph nodes. This biology is all starting to link up such that we think that after we treat, these B cells are going down. They're actually migrating into and being retained within the lymph nodes, where they start to pick up this increased antigen presentation ability.
Again, returning to that cartoon then, that's why I mentioned before that we think antigen-presenting cells, and specifically B cells then, are now into the lymph node where they can interact with T cells, help facilitate showing the antigen to T cells. There's this additional co-stimulatory function of CPI-006 on CD73 in T cells. It's an addition to helping the B cells become activated. It's also directly helping to activate T cells. You put that in the context of ciforadenant eliminating immunosuppressive adenosine in the tumor microenvironment, we really think that this could be an attractive therapy, with the early data, we're quite excited about the potential here. In conclusion then, CPI-006, novel immunomodulatory activity with dual mechanism.
We talked about B cell trafficking quite a bit with increased expression of CD69 and other markers, an increased antigen presentation, and obviously inhibition of CD73 enzymatic activity without internalization of the receptor in that being different than other antibodies. Molecule safe is throughout the study to date, 12 milligrams per kilogram as monotherapy and 6 milligrams per kilogram with ciforadenant with no DLTs or MTD. At doses of 12 mgs per kg, we get sustained occupancy in PDL with target saturation and inhibition of enzyme activity from tumor biopsies in patients. The treatment associated with peripheral blood serum cytokines consistent with an inflammatory response. The preliminary data suggests that there are patients already who are benefiting with extended stable disease that appears to be clinical benefit associated with reduction in pain, as I mentioned.
The study continues to enroll both monotherapy combo and there are further stages as I outlined previously. With that, I'll pass it back to Richard.
Well, stay up there. I'm sure. Any questions for Jason, or? I have a question. These B cells are in the blood, they leave. What do they do when they get in the lymph nodes? What's the normal immunology?
Well, they're going to do a lot of things, right?
Well, teach me. I don't know what they do, tell me what they do.
B cells are one component of the antigen presentation compartment. They will go and, on some level, educate T-cells. We commonly think of that as being from dendritic cells, but they'll also become and be educated to differentiate into other cell types, obviously to differentiate into plasma cells and make antibodies. I think there's probably multiple things that I'm not sure 100% where you're going with this. There's lots of things that they could be doing.
Okay.
I think sort of the spectrum is multifold.
Since I give some of the lectures on immunology to freshman medical students at Stanford, the lymph nodes are not just these little things that are kind of hanging out there. They have a very characteristic architecture. B cells come out of the bone marrow, and they float around in your bloodstream. When there's a site of inflammation or a bad guy who's entered your body, they go to the lymph node, and they get stuck in there. When they're in there, they get exposed to the antigen. Then interesting things happen. They start to differentiate. They do a couple of different things. One is some of the B cells differentiate into what are called plasma cells and make antibodies. One of the questions today from I think one of your former colleagues was, have we found any of those antibodies yet?
Well, that's a really good question, we're going to look like that really hard because this may be the greatest way I've ever seen to make antibodies to your tumor, and to make even monoclonal antibodies, is you could harvest those B cells. Something really to do. One thing is that the B cells will differentiate towards making antibodies. Some of those B cells are so-called professional antigen presentation cells. Dendritic cells, B cells, macrophages. The other thing they do is that they have the antigen on their surface, they present it to T-cells. Remember, T-cells don't do anything unless they have antigen presented to it in the proper context, which is more complicated than we need to get into now, activation of T-cells. This is all a concerted process that results in how an immune response occurs.
Immune response is both cellular and humoral, usually. The action really occurs in those lymph nodes. Although not proven, although there's a lot of preclinical evidence, in fact, one of our collaborators, Jonathan Powell at Johns Hopkins, did this with our drug ciforadenant and showed that in the regional lymph nodes in a mouse tumor, and this is published, you got activation of lymphocytes and reduction of what are called suppressor T cells with the treatment with our drug in the regional lymph nodes of a tumor. Regional lymph nodes mean where the antigens from the tumor would be spreading because that's going to be the first site that gets exposed. What we're seeing here, and the reason for our proposed mechanism, is all really consistent with what you would expect in normal immunology. We obviously have a lot of work to do.
It'd be nice to look and see what's happening with those T cells. It'd be nice to see, are we finding antibodies that are tumor-specific? We certainly want to look at that. That's not that easy to look for. That's a challenging exercise. Of course, what we're primarily interested, are people's tumors going to go away? That's, of course, what's driving all of this. The thing I want to emphasize is that this antibody, if we can go to the next slide, this is a great way to segue into this slide. I mentioned there's a lot of CD73 antibodies out there. This antibody is unique. Why is it unique? Are we just lucky? Did we just stumble upon that? We're not. I'm definitely not lucky. I can tell you that. I've never been that lucky.
You think about most people who've been making CD73 antibodies have been saying, "I want to block the conversion of AMP to adenosine. Let me make an antibody that blocks that." Because people were thinking adenosine, and that's one of the functions of that protein, so it's not a bad idea. As I was telling some other people in the room earlier, CD73 was first described, had nothing to do with adenosine. It was first described, it was first called lymphocyte vascular adhesion molecule. It was known to be involved in lymphocyte adhesion, and it was called lymphocyte vascular adhesion molecule. Later did people figure out that it was also an enzyme that convert AMP to adenosine. This is an antibody that we licensed from a person named Linda Thompson, and I want to give a shout-out to her.
Linda was at Scripps at the time. She's now retired. Linda made this antibody not so much for looking for blocking adenosine, but more about its function. The antibodies you get are a result of what you screen for, what you look for. She was looking for effects on vascular adhesion and things like that. Of course, we licensed that antibody from Scripps, and we, of course, did a lot more work on it. We humanized it. We affinity maturated it. We changed the frameworks. It's an entirely different-looking protein now. It's different. It was made to interfere with the function of that adhesion molecule, and it also blocked the conversion of AMP to adenosine. I was telling somebody before, small molecules to block things have the limitation. Who was I talking to about that? You have a geography problem.
If you want to block these functional things, you need to cover a bigger surface area. It's hard to block protein-protein interactions. Easy to block AMP binding. ATP binding is easy. Usually, small molecules can do that. Not these other functional things are much harder. Anyway, how does this result? What's different about our antibody? In this slide, I've looked at Corvus' antibody, AstraZeneca. That's probably AstraZeneca right now they don't talk about. Their lawyers are very good. AstraZeneca, BMS, Surface Oncology, Innate, a company called Arcus. And I've highlighted in yellow there the key differences. CPI-006 inhibits enzymatic activity. No question, very potent at that. Kills it dead. You saw the slide that Jason presented at very low concentrations. It does it without internalizing the antigen.
A lot of the other antibodies, like the BMS antibody that I'm familiar with, the AZ antibody, some of the inhibition of enzyme is due to the fact that when the antibody complexes with the antigen, it gets internalized. That happens to a lot of antigens. When you cross-link them, they become internalized. Obviously, they become internalized, there's no enzymatic activity, right, because the antigen's not there. Of course, it comes right back because the cell produces the antigen again. This does not internalize. We knew that. We can study that in vitro. That's one difference. We know it doesn't cause internalization. We've also made some of those other antibodies from these companies, and in vitro we show that they don't cross-block.
If you react with the same epitope, if I treat with one antibody and then try to come in with the other, I won't be able to get in there because it's blocked and vice versa. Very straightforward experiments that immunologists do all the time. We know we don't cross-block. We know we don't internalize. Then there's a third thing. Many of these other antibodies react with human and mouse. We don't react with the mouse because the mouse doesn't have that epitope. There's no question that this CPI-006 is a unique antibody that has not only the property of inhibiting the catalytic activity, but also has this other property of these immunomodulatory functions, which we've demonstrated in vitro using human cells.
We've actually also demonstrated in subhuman primates, we haven't shown you that, monkeys who also have this antigen, and the antibody does cross-react with them. Now we've shown it in our human clinical trial. You see this happen even at the lowest doses. We were very pleasantly surprised. At one milligram per kilogram, you see these phenomenal changes in blood lymphocytes. Blood is like shooting ducks in a barrel, right? Because you give the antibody, of course, it's in your bloodstream, it's in a pretty good concentration. Of course, you want the antibody to get into the tissues and the tumor and all those other places, and that's why you need to go up on the dose. That's what we're doing. Inhibits the enzyme, doesn't cause internalization. The other antibodies there, they cause internalization. Some of them induce what are called allosteric changes.
If something binds to a protein and causes it to bend a different way, you can lose enzymatic activity. This company has a small molecule. Why you would make a small molecule to that defies my logic. I guess if you just wanted to block AMP, that would be fine. Again, you're trying to block proteins that interact with other proteins. I don't see any reason to do that. By the way, it'd be very easy to do that. If I wanted to block AMP binding to CD73, my chemist at Corvus could do that by next week. Pretty simple. Small molecule. It's easy to mimic it. There's already scaffolds out there and such, et cetera. We don't want to do that. The most interesting thing, though, is this activation of APCs and lymphocytes. I don't know, maybe somebody can get me a paper or something.
Jason, do you know? I don't know of any, Lauren, I don't know of any other CD73 antibody where there has been a description of this effect on immune cells. I just don't know of it. Okay? I think this antibody is very, very unique. I think, therefore, the antibody's unique, our combination with it is unique. This is why we're really excited about this agent because unlike the 12 other guys who have CD73s and A2A antagonists, we can now have something that's really pretty unique and special proprietary to Corvus and file patents on that and all that other sort of stuff, which of course we're doing very aggressively. Very different antibody for the reasons I've mentioned. And so far appears pretty safe even at doses of 12 milligrams per kilogram.
I just heard last week we dosed patients at 18 milligrams per kilogram. We're going to up on that. Enrollment in this trial is going very, very quickly. Obviously, there's a lot of interest in prostate cancer, not only because of what we're seeing, but AstraZeneca, as you know, reported a couple of weeks ago or a month ago at AACR with their A2A antagonist, some responses in prostate. We've also seen activity in prostate with our ciforadenant, which we didn't chase because we saw what we thought were better signals in renal and lung. Let me go to the next slide. I guess I can do the next slide. I'm used to having people working for me. I don't like doing it myself. Let me take a moment to put it together. I don't know if I need this, I'll use it anyway.
Ciforadenant, which Jason described, is in the trials that you heard about in combination with 006. Obviously we've also been using it as a monotherapy and more recently as a combination with atezolizumab in ongoing renal cell cancer studies that we're doing, and also in lung cancer studies that are being conducted by Genentech. Those studies are enrolling nicely. I think the recent really exciting thing with ciforadenant, and we've now presented this at a couple of meetings, is that we've identified a gene signature called the adenosine gene signature, which we believe predicts response to this treatment. Everybody says they have a biomarker. Biomarkers are hard. It's hard to prove this. There's a lot of variables and all that stuff. What I like about this biomarker is it actually makes sense. Our scientists in our laboratory did some very simple experiments.
I can't believe these experiments hadn't been done before. I told my guys, "The dumbest experiment I've ever seen," but it was the dumbest experiment that had to be done. He just took white blood cells, stimulated them, throws adenosine in, suppresses them. What genes are associated with that suppression? It turns out, shockingly, it's a bunch of myeloid-associated genes, which we've presented at meetings. Adenosine, the presence of adenosine is associated with these myeloid genes. Then, of course, we did the obvious thing. Let's look in the tumors, and we find that from these waterfall plots, again, this is old data. We've presented this. Patients who respond have this adenosine gene signature. Makes sense. Things got really interesting and even more exciting right around this time.
A paper by McDermott et al. was published in November, Nature Medicine in 2018. 400 and something patient frontline renal. Okay. Genentech, of course, does very nice biomarker work and genetic analysis, and they described what they called a myeloid gene signature, which was associated with lack of response to atezolizumab, a PD-L1. Let me repeat that. Myeloid gene signature they found. They're not looking at adenosine. Adenosine wasn't in the picture. There was no adenosine in the treatment, no CD73, A2A, nothing like that. It was just a study looking at SUTENT, atezolizumab alone versus atezolizumab plus Avastin. A myeloid gene signature was found to predict no response to or lack of response to atezo. When we look at their genes, they're virtually identical to our genes. I think of 12 or 13, 90% of them are the same. 95% are the same.
Putting this together, what does this mean? It makes very good sense. This adenosine signature, which is indicative of the presence of adenosine, appears to be a resistance mechanism to anti-PD-1 or PD-L1 therapy. That makes perfect sense then to combine a PD-1 or anti-PD-L1 with an adenosine antagonist. That would make perfect sense. That's what we're going to test in the clinic. I won't go through this again. This has all been presented. By the way, we looked at, of course the obvious things. CD73 expression, does that correlate? No, it does not. Okay. We've looked at hundreds of people. There's no CD73 in the tumor, CD73 in the serum doesn't seem to predict anything. Okay. I know a lot of people are talking about that. We don't find it, and we've got the most data.
I hope they keep working on it. This adenosine gene signature appears to make biological sense. By the way, the biomarker game is never that easy. It's good to find this connection. You might ask, why don't we just measure adenosine in the tumor? That's hard to do. Adenosine is a very short-lived molecule metabolite. Adenosine is actually a drug. It's used to treat supraventricular tachycardia. You give it IV. If you've got a rapid heart rate, it'll slow your heart rate, and it has a half-life of 10 seconds because there's a lot of enzymes that metabolize it. In a tumor, what some people have reported, and they know better than this, they do know better than this.
They take a tumor, they crush it, they say, "Oh, it's got a lot of." They put enzyme inhibitors in there, they say, "Oh, there's a lot of adenosine in the tumor." Yeah, because you've crushed the tumor, and all the adenosine from the inside of the cells leaks out and all that other stuff. You really care about is the adenosine in the extracellular space. By the way, that was measured 20 years ago. We've measured it. Charles River Labs will measure it for you if you can buy the test from them. It employs a cool technique called microdialysis. The little tiny microdialysis tube hooked up to a mass spec that's made for measuring these wrapped short-lived metabolites.
When you do that, you do it right, the level of adenosine, at least mouse tumors, several that we've measured and was found, again, 20 years ago, is around 100 nanomolar, 100 to 200 nanomolar, not micromolar. The levels of ciforadenant that we achieve easily outcompete the adenosine in the tumor, easily. I'm only bringing that up because some of our competitors make a big deal about that. Some people might ask about A2B. What about A2B? The role of A2B receptor is unknown. We actually have an A2B selective drug that we made at Corvus. It's really selective for A2B, and we're having a hard time figuring out what it does. The role of A2B is not yet clear. I'm being a little flippant. It actually is believed to be important in fibrosis and diseases like that.
Most people who have an A2A antagonist are also blocking A2B, because A2B is also known as a low affinity, look it up, low affinity adenosine receptor. The affinity of adenosine for A2B is like 25 micromolar. I would say pretty much anybody's A2A receptor antagonist is also substantially blocking A2B. If people are telling you that they've got A2A, A2B is important and they can differentiate it's hard to believe that. Chemistry doesn't work that way. Okay. Getting back to the adenosine signature, we've looked at TCGA, and we asked the question, are there other tumors besides renal cancer? Because what I showed you was elucidated on renal cancer biopsies. What about other tumors?
We go to TCGA, where we have sequences on lots of tumors, and we find that it's possible to use this adenosine gene signature for a lot of tumors. There's some really interesting things here. Some tumors are low, and some are really high. One that I like, I don't like it, but it's interesting as a target, colorectal cancer is really high. Lung cancer is high. Renal cancer, half the patients are high. You can go through the list. Pancreatic cancer is high. There may be some targets there where we can use our adenosine gene signature and our drugs for what are now IO recalcitrant tumors. These tumors that have been IO resistant, colon, prostate, pancreas, gastric, the list is pretty serious, may be recalcitrant because they have another mechanism of resistance. There's multiple checkpoints, and you have to hit the right ones.
This is precision medicine at its best. That's our plan, to use this adenosine signature both for A2A and for CD73. Will the adenosine signature predict CD73 response? That's a question. I don't think that we know the answer to that. It may not. It's a different agent working by a different mechanism. Just to go back to this slide. You heard about ciforadenant, you've heard about CD73. I want to mention our ITK inhibitor because we announced, I think, a week ago or so, that our ITK inhibitor. This is a drug that has nothing to do with adenosine. As you may know, our team at Corvus are the developers of a drug called ibrutinib, which is a very successful drug for lymphoma, approved for many lymphomas now. More than that, it was a paradigm shift in how you make a drug.
It was a covalent drug, which most people say, "You should never make a covalent drug." It was covalent. We ignored those people, and we made it anyway. It's a great drug. ibrutinib hits a target called BTK, which is in lymphomas. There's a homologous target called ITK that's in T-cells. We've now made a covalent inhibitor of ITK that's extraordinarily selective. Our clinical trial has recently been initiated. Initially, we'll be treating patients with a variety of T-cell lymphomas. That drug is in the clinic now, and we're doing a lot of cool biomarker work associated with that. What we're really interested in the T-cell lymphoma patients, of course, we're interested to see if their T-cell lymphomas go away.
That's what we're primarily interested in, T-cell lymphoma patients have normal T-cells, so we can also interrogate and evaluate what's happening to the normal T-cells. It turns out that when you hit ITK selectively, you affect lymphocyte differentiation. That's what we're going to be looking at. Did Jason get bored? He left me? It's all right.
Yeah.
Okay. Just to finish off, what we have to look forward to. Again, I mentioned this on our conference call last week. Corvus presents data at almost every major meeting, AACR, SITC, ASCO. We get oral presentations and stuff like that. Usually it hurts us. People say, "Oh, it's too early, it's this and that and the other thing." I've learned from my experience at IDEC and Pharmacyclics that ultimately you win that way. We're going to continue to do that. Of course, it represents good science and good drug development and allows us to work with the great people like Dr. Harshman and Dr. Luke and many others. I mean, we are well-wired in the medical community. Here's what you can expect. I think we'll see more data on ciforadenant at SITC.
I expect to see more clinical data on CD73 at SITC, which is in November. We should have more follow-up and more patients treated by then. I'm hoping that we can have some clinical data, although it's going to be very early, for ASH Meeting in December of this year. More likely that would be in some meetings early in 2020. Three products in the clinic, generating clinical data. We communicate that to the street. We're pioneers in each of those fields. Everybody's talking about they have better A2As and A2Bs and CD73s, but we're still the pioneers, and we can continue to do that. I'm still waiting for people to come up with better rituximab and ibrutinibs. It's 20 years later now, and they haven't. It's good to be a leader. All right.
I guess I started by answering a question that I asked. Oh, hi. Great. Nice to meet you. All right. Michael, you have a question? Yeah, you can use this.
That's great because Dr. Luke left. I had a question. The prostate cancer patient. I was wondering if you could just elaborate a little bit more about the type of tumor volume reduction that you've seen in that patient because the last scan on the swimmer plot shows progressive disease, not stable disease. I was just wondering if you could elaborate on that.
Yeah. That's a good question. I can answer that since he's not here. This patient has some lymph nodes regressing, about 20% regression of some of these periodic lymph nodes. He has one lymph node in the pelvis that's actually a little larger. By RECIST criteria, not iRECIST, but strict RECIST criteria, any progression is a PD. However, remember what I told you about regional lymph nodes. Right? This is a lymph node in the pelvis near the prostate, I'm not convinced. I don't know what that is. He has scored a PD now, but the guy's feeling great. His pain is better. Bone pain. He's having a regression of lymph nodes, and he's got maybe some slight enlargement, tiny enlargement in a regional lymph node. We'll have to continue to treat and see what that looks like.
Tony.
He's been I don't remember exactly when that scan was.
Five years ago.
Yeah, I think it's a Yeah. Yeah. Yeah. RECIST criteria is pretty tough. One of the things is interesting and a lot of people are having. Imaging techniques are so much better now. It's getting really harder to call PRs and CRs because the imaging techniques are so good. They pick up millimeter differences which we never really saw before. Anyway, he scored a PD now, but clinically, he's not a PD. I don't know what that lymph node is. Now, what I would like to do is biopsy that lymph node, that's going to be not trivial. Maybe in a dark alley some night I could. No. In New York, we would do that. Right, Mario? All right. Any other questions? There's got to be some. Tony.
Richard, there's a couple of papers I want
I think it's webcast, you need to
Thank you. Two other papers that I would love for you to help rectify for me the notion of what's going on in the adenosine signature. One is [Mary Antonson] in Nature in January last year in 2018. Let me just explain. IMvigor, just like McDermott. atezo is used. They get a 50% response rate. What happens in the patients that don't respond? They have elevated TGF beta, by the way. I'm curious how that relates here, because likewise in cord blood, in Blood Advances in May, this was again in a blood tumor. Looking back at NHL, they demonstrated that patients that responded also had that myeloid signature as well.
That seems to be a common theme as you alluded to, I'm trying to also just justify what happened in IMvigor with the elevated TGF and if in fact that's what you see with that adenosine signature as well.
First of all, you're right. The myeloid signature is being a repeated theme now, and myeloid cell biology is a hot thing in immunology. We only recently started studying that because it's not been easy to study myeloid cells. They're difficult to work with in the lab, and we don't have good markers for them. That's really the hot area now. We've looked at other genes, TGF beta. I can't remember specifically if we've looked at it, but we've looked at a lot of things. For example, we've looked at the T effector signature, which by the way is up on the slide. The T effector signature, which is a good biomarker for some tumors, and tumors may be different, Tony, also.
Basically, this up here where we saw, this is your T effector signature, and it looks like with our myeloid signature, about half of them express the T effector signature and about half don't.
Okay.
I mixed it up. Half of them express and half don't.
I only say this because IMvigor was in bladder cancer. If you go to your slide which had the
Oh, that.
It actually showed, if you look over here.
Bladder
at the far left. Way over here, far left. Yeah.
Oh, here.
It looks like.
Yeah
You get some.
Yeah. Different tumors might be different and the cutoff. The other thing to think about is the cutoff. What's your cut point for calling high or low? That's a struggle. That may not be the same in all tumors. Just like for PD-L1. PD-L1 is a great biomarker, but it's not true in every cancer. It's probably a great biomarker in melanoma for sure, lung cancer probably. Other tumors, kidney cancer, it's not a great biomarker in kidney cancer. The biomarker game is tough. Again, what's nice about PD-L1 is it makes sense, the target of your antibody. What I like about the adenosine gene signature is the target of your drug. That makes good sense. All right. Any other questions? Michael?
Webcast should've done this before.
No, it's okay.
Michael Morabito, Credit Suisse. If you're getting target saturation at 12 mg/kg, and it's 100% or near 100%, do you anticipate any efficacy benefit from raising the dose beyond that?
Great question. First of all, the occupancy data that was shown was in the blood. That 100% occupancy is on blood cells, not in the tumor. I don't know what the percent occupancy is in the tumor, and we don't have that much data on that. That was one biopsy specimen, and it's a core biopsy. I think we're pretty close to full occupancy, but I think we should go higher and get some more data and figure it out. Blood is going to be different because blood is, as I said, really easy. In terms of the immune function, CD73 is expressed on a lot of tissues, and I'm not sure exactly what the right occupancy is going to be. We're going to have to be a little empiric and do it the old-fashioned way.
Go until we think we're pretty high, or go until we see an MTD or maybe not. We may never see an MTD. If you take a drug like Rituxan, we went up and up and up and up and never had an MTD. Right? Then finally we said, "Wait a second, there's only so much antibody you can give." Right? In our monotherapy now, we're at 18 mg/kg. We just started that. That's a pretty high dose. Now, in monkeys, we've gone to 100 mg/kg, and we didn't see any significant toxicity. I think we went over 100. We're not going to do that in people. That's an enormous amount of antibody. We would need swimming pools full of antibodies to do that. I think it's going to be somewhere between 12 and 24. Maybe 12. We don't know yet.
I was just wondering when you're going to start dosing the cohort with Pembro, and then wanted to know, are you going to have to start with one milligram as well, or can you start higher when you start?
Great question. We can start higher. The protocol's written that we start two dose levels below where we are with the monotherapy. We wanted to fill up our monotherapy and our ciforadenant data first because we're most interested in that. I'm not that interested in making Pembro work better. Having said that, mechanistically, there's a beautiful story emerging here, and I'm hoping that you guys came to that conclusion yourself. The thing that makes the most sense here is ciforadenant, CD73, and PD-1. A triplet. When you think about it, you're removing immunosuppressive adenosine, you're unleashing your T cells with PD-1, and you're stimulating your APCs with CPI-006. That's a triple threat. What do they call that in soccer? The hat trick or something like that?
Yeah.
Yeah. That's really what makes the most sense. We didn't want to go start in our protocol with a triplet. We thought a lot about that. We thought that was a little too rigorous to go to the FDA with a protocol that had singlet, doublet, triplet in the first protocol, first-in-man study. It was a little aggressive. Now we have safety data. Now I think we can go to that. We want to do the Pembro, but the protocol's written so that we can fill up the others first, and I wanted to get more data on that because those are our products. All right. Any other questions? Yes, Mara?
Properties of CD73 and whether looking at that in a combination with an angiogenesis strategy makes any sense?
Yeah. Mara, you're too smart for me, Mara. CD73 is on some vascular endothelium, but it's very complicated. It's on some vascular endothelium but not others, and what all that means. It is involved in angiogenesis. Does it make sense? I don't know. I think we have enough things to do right now. I'm not sure that would be the top of my list, but it is an interesting thought.
I like it.
You like it?
I like it.
You like-
I've actually submitted investigator-initiated clinical trials.
Oh, that's right. I forgot about.
to these guys that we're talking about with cabozantinib. Any of the TKIs, but right now in kidney cancer, you have to think about what's going to be used up front, and then what's going to be used 2nd, 3rd line. We're kind of in an embarrassment of riches in kidney cancer with a lot of different therapies. However, we're still not getting enough cures, I do think we got to go to things like triple threats and thinking about how to take advantage of potential combinations and synergies with anti-angiogenesis. I do think what you actually said my trial really well.
Yeah. I haven't agreed to do your study.
I know. Not yet, Richard. We're talking.
Speaking of new studies, I should mention that we're very close to opening a study at Johns Hopkins in myeloma. Why myeloma? That's because CD38 is. I still think the best way to block adenosine is to hit the A2A receptor, because adenosine comes from a lot of places, and I'm not sure you can get them all. One of the places it comes from is CD38. CD73, CD39, but CD38 is another source through what's called NAD cofactor. There was a very good paper several months ago in Cancer Discovery in mouse models showing that anti-CD38 and A2A blockade work spectacularly in cancer. Of course, what is it? daratumumab, I always forget the name of that antibody. J&J's daratumumab is an anti-CD38 that's approved for multiple myeloma. Patients fail it. It's a very good drug.
It's a very good antibody, and it has a response rate, and of course, people ultimately fail it. We have a study we're going to do with the Hopkins guys adding our A2A antagonist to daratumumab failures. We're going to give our drug alone, and we've looked at the adenosine signature in myeloma. It's expressed there. We see it in the bone marrow cells and so forth. That's going to be interesting study. That's just going to be open at their center, maybe one others. I don't know. We are interested in these other things, I forgot about your study. Whoops. All right. Any other questions? I think it's getting late. Again I want to thank everyone for coming. Great turnout. Appreciate your time on this Sunday evening, we'll be hanging around for answer any other questions you have.
Thanks again.