GRTS B. Riley 2022 Transcript

Mayank Mamtani (B. Riley)  5:53  

Hi, good afternoon and good morning for folks on this side, including the presenter for next company presentation. I am pleased to have with me the CEO of Gritstone bio. Andrew

Allen, appreciate you being part of the conference. I believe you have a few slides to go through, and we can do a q&a at the end.

Dr. Andrew Allen (GRTS CEO)  6:21  

Very good. Thanks, Mike. Thanks for inviting us to participate. Good afternoon, everybody. My name is Andrew Allen. I’m the CEO and one of the cofounders of gritstone bio. And today I’ll be giving a presentation focused on our oncology portfolio and I shall be making some forward looking statements. gritstone has two core technologies. The first is our edge machine learning platform that enables us to identify T cell targets or epitopes from sequence data alone. As you know, T cells don’t recognize whole proteins they recognize fragments. And the identification of which fragments will be good targets for T cells is a challenging but rewarding proposition. 

That’s what our platform does. Secondly, we have proprietary vectors that deliver selected antigens to the human immune system to drive strong immune responses, including the challenging CD8T cell response that most vaccines struggle to elicit. Using these two platforms. We’ve developed products both in oncology and in infectious diseases. And we have several partnerships in the space listed here on this slide. We have multiple near term catalysts, both in the oncology and infectious disease arenas. Little more color on our vectors, we have developed a self-amplifying mRNA construct, which of course is becoming very popular given the interest on mRNA-based vaccines. This has some potential benefits over the first generation mRNA because it is a potentially dose sparing and be by virtue of double-stranded RNA intermediates formed within the cell. This may drive a stronger immune response, including more potent CD8 immunity. This does drive very good antibody responses as well as many of you will know. Secondly, we have a chimpanzee adenovirus vector. 

And this remains the king when it comes to driving CD8 responses in humans. And for cancer immunotherapy. It’s really about CD8.

 And that’s why the adenovirus remains a core tool in our therapeutics for cancer. Off of these two vectors, we’ve built a fairly extensive pipeline. And this is shown here I won’t be speaking to the infectious disease elements which are at the top of the slide, I will be speaking to our oncology products. Granite is our individualized new antigen based immunotherapy. Slate is our off the shelf product. So let’s turn to these. The premise for Cancer Immunotherapy is shown on this slide. We know that humans with cancer live on a spectrum. Some tumors have not successfully fully evaded the immune system. And Neo antigen-specific CD8T cells have been generated by the host. They’ve entered the tumor. And the only reason the tumor has survived is because at the last moment that tumor throws up PDL one expression as an escape mechanism that induces energy in the T cells which are adjacent to the tumor cells. So effectively they’re blocking at the last moment. And I think to our delight, and somewhat To our surprise, antibody therapy, just with an anti PD one or Pdl, one antibody is able to reactivate those T cells and drive tumor destruction that leads to durable clinical benefit in many subjects. This obviously underpins the transformation that checkpoint inhibitors have driven through solid tumor immunotherapy. However, it is worth remembering that most solid tumors do not respond to checkpoint inhibitor therapy, and most patients diagnosed with metastatic solid tumors will still succumb to their disease and treating such subjects with checkpoint inhibitors. Does precious little and the likely reason is that they do not have large numbers of new antigen reactive CD8T cells. There is no substrate for the checkpoint inhibitors. So if you treat them with a PD, one antibody, all you will do is induce autoimmunity. To close the gap, we need to drive new antigen specific CD8T cells. And our vaccine based approach seeks to accomplish that, of course, there are other ways to do it with cell therapy. And I think what we’re seeing now is that everybody recognizes the power of Neo antigens as targets for cancer immunotherapy. And obviously, our goal is to try and deliver strong CD8 responses just using a simple intramuscular vaccine similar to the one we’ve all had for COVID-19. If we can do that, obviously, that potentially is transformational for solid tumor immunotherapy.

Now, one of the key challenges is recognizing which aspects of the tumor we want to include in our vaccine. We’ve started with exome mutations, meaning mutations that change a single amino acid typically within proteins encoded by the tumor genome. And these are altered proteins which of course can function as antigens to the immune system. Because your immune system is not tolerant to these they were not present when the process of tolerance was taking place, as a fetus or neonate. And this, of course, makes them very attractive targets. The challenge is that most mutations do not create Neo antigens. And the question then becomes how do I find which are the good ones from the sea of many mutations? How do I identify the ones that are relevant targets for T cells, we’ve trained with abundant human data derived from human tumor cells, a machine learning model called Edge, which uses a deep neural network to establish a high quality predictive model for identifying just from sequence data alone, which mutations will create good antigens. It works extremely well, for class one presented new antigens, which of course, are the target for CD8T cells. And we’re getting ever better at class two prediction, which is another interesting area that we can talk about maybe in the q&a.

I will note that there are likely to be other forms of target within tumors that may, although they’re not mutations, they are somewhat tumor specific. And there’s growing interest in altered splicing in endogenous retroviruses in cancer, testis antigens and oncofetal antigens, there are other classes of epitope, that may be relevant. Of course, there may be some degree of immune tolerance to these because these are not mutated. And so the question really, that we’re all seeking to address now is whether they have as much utility in vaccines as these foreign Neoantigens, an interesting arena, perhaps for future discussion. 

The two products that I’ve mentioned are granite, which is our fully individualized products made unique for a patient based upon their own set of tumor mutations. This presents some manufacturing complexity, but obviously can be done. And of course, we’re getting better, faster and cheaper added all the time. And I think the success of kite is showing that even cell therapy, which is much more complex to do than what we make, even cell therapy can be done and scaled for individuals if the clinical data merited. We also have an off the shelf product, slate, very attractive concept. And I’ll show you some data and how we’re thinking about designing the slate programs that start with Granite. 

We completed a phase one to study in advanced subjects, patients with advanced disease who really reached the end of the road from a cancer therapeutics perspective. These are not the ideal subjects for vaccine based immunotherapy, because it takes about eight to 12 weeks to generate the necessary numbers of CD8T cells. And some of the patients that we’ve treated are unfortunately dead within that timeframe. Nonetheless, we obviously have to start here for phase one novel therapy studies. And we have some very exciting data which we shared at ESMO in September last year. In the Phase One two study, we did a flat dose of the chimpanzee adenovirus, that’s our priming dose containing the Neo antigens for that patient, we boost with a self amplifying mRNA construct and that was given at monthly intervals for six months and then bi monthly to the to a year. And this is all done on a background of systemic nivolumab full dose together with low dose subcutaneous ipilimumab i The CTLA four antibody from Bristol Myers and this was given sub q adjacent to the vaccination site. The goal being to maximize drug concentration in the vaccine draining lymph node, which is where you want the Paluma map to augment T cell expansion in response to the administered new antigens. The nivo you want in the tumor to prevent new antigen specific ca T cells from being shut down when they enter the tumor site. So that was the phase one doing this dose escalation. Phase two was cohort expansion, primarily actually in colorectal cancer and gastric cancer. As we showed in September, we treated and presented data on 26 subjects, almost half of them had microsatellite stable colorectal cancer. And given the huge unmet need, and the lack of activity of checkpoint inhibitors alone in this context, this is a focus for us going forward. So I’ll be speaking to our colorectal data in particular, gastric data, while promising are a little harder to interpret because there is some activity of checkpoints in some subjects with gastric cancer, so the backdrop is not quite as clean. From a toxicity perspective, we do see some injection reactions and transient fevers as you would expect with potent vaccines, but no other unexpected adverse events. We did of course, see checkpoint related toxicity as you would imagine.

encouragingly, we see strong induction of new antigen specific CD8T cells in all subjects treated. And these data are unique. No one else can show these forms of data. Because this is a hard thing to do. We’re using the low sensitivity ELISpot assay that just take cells out of the patient’s blood stimulates them ex vivo with peptides, and you measure the proliferating cells just the next day. There’s no in vitro stimulation. This is a simple assay, very standard. And mostly when you look at the history, vaccines that have been tested failed to generate CD, as you can see here, on the left is baseline on the right, we have strong induction of CD8T cells, essentially in all subjects. And gratifyingly, this led to a very prominent effect on the biomarker that is increasingly well recognized as having particularly utility for immunotherapy. And this is circulating tumor DNA. And so here using a waterfall plot, we’re characterizing the molecular response of these treated subjects with colorectal cancer, recognizing, of course, that the data are not in yet for this to be a registrational endpoint. But there’s accruing data suggesting that indeed, measuring CT DNA, and measuring its change in response to therapy is likely to be a superior biomarker, to assess benefit from immunotherapy. And there was an important step forward in this field yesterday when immuno core had the approval of their product, essentially advice specific, which really didn’t show any benefit when US judged using standard racist responses as a criterion that had a dramatic impact on circulating tumor DNA that correlated closely with survival.

 And I think you’re going to see more and more data suggesting that CtDNA response is a superior biomarker. So here we are showing that nearly half of our subjects had molecular responses, and it correlated very well with prolonged stable disease using resist for what that’s worth. 

But much more importantly, when we dichotomize the subjects into those who had a molecular response and those who didn’t.  And we then look at overall survival, you can see that there’s a profound separation of these curves. And we showed data last September on the left here. And as of January, all of the patients who were alive in September remained alive. And this is surprising for third line colorectal cancer, where most subjects unfortunately succumb to their disease within 12 months. The median is typically around six to seven months in trials of third line colorectal patients. 

And because of the way we have to select our patients early to make product, we have very limited opportunity to select for, quote, good actors. So of course, these are single arm data. This is not randomized. This is just dichotomizing around molecular response. Nonetheless, the data are extremely encouraging. Now, this form of immunotherapy is novel. And what we’ve observed is a pseudo response or sorry, a pseudo progression phenomenon. And you can see that here. This is one of our early cases, treated the dose level three, and you can see that multiple pulmonary nodules increased in size at week eight, and then start to diminish cavitate. And they’re essentially absent after a year. But this kinetic of response is obviously quite different from just checkpoint inhibitor therapy, and something we need to be very mindful of as we assess patients in the early stages of their treatment. Based on these impressive data, we went to speak to the regulators the FDA and have now launched a randomized Phase Two three trial in the maintenance treatment of newly diagnosed patients with metastatic colorectal cancer. The phase two component is up to 80 subjects, and the primary endpoint for that aspect of the study is molecular response. We anticipate having those data by the middle of 2023. So obviously, this year is a important year from an execution perspective. We’ve also launched an earlier stage trial in patients with high risk adjuvant disease.

Again, this is a small randomized study, looking at immunotherapy versus his observation which is the standard of care in subjects that very high risk of relapse from their disease based upon persistent CtDNA postoperatively. So those are the granite data, let me now pivot swiftly to sleep the off the shelf product. Slate captures a set of K RAS mutations in a vaccine that is pre made. And so we’re now just looking for subjects who have the appropriate K RAS mutation and the appropriate HLA molecule to present that mutant K wrasse peptide. The combinatorial pairs are what matter here. And we anticipate that between 10 to 15% of subjects with adenocarcinoma of the lung, and colorectal cancer will be eligible for this product. We show data last September using the version one product 26 subjects, half of them have lung cancer, and all of those have progressed on prior checkpoint inhibitor therapy. Again, no unexpected adverse events beyond injection site reactions, fevers and immune related adverse events associated with checkpoint inhibitors. encouragingly, we also see molecular responses in around half of the subjects. And here we’re looking at the lung cancer subjects. Not all of them had measurable KRAS mutant DNA baseline. But for those who did, you can see the responses shown in this waterfall plot. And we did see indeed, of recessed response in a lung cancer subject. And that’s shown here. And you can see quite impressive reductions in multiple lesions including the liver, which of course is traditionally one of the harder sites to impact. Sadly, this patient developed breakthrough at around five months and then succumbed to his disease. He had very aggressive disease, and actually blew through Pember chemo in three months, and was on our immunotherapy for five months before acquired resistance emerged. Now the first version product of slate had multiple epitopes, in addition to K RAS. And we learned that we were actually diminishing the magnitude of the K RAS specific effect. So we’ve re engineered into a version two product where we’ve taken out the non K RAS epitopes and focused in on K RAS alone. And that has a dramatic impact in the magnitude of the T cell response. And you can see that shown in an appropriate transgenic mouse model on the left, where very little response to g 12. V was observed using the version one product and clearly substantial immune responses to g 12. V in the with the version two program. And so that is now in phase two in humans. And we anticipate data off of that study, in the middle of this year, around mid 2022.

We manufacture our own products. And this is important, I think all of you are aware that bio manufacturing is now a valuable commodity. And if you don’t have it, it can be very constraining in terms of development timelines. We recognize this several years ago, and we have a fully insourced by manufacturing program, split across Cambridge and Pleasanton are two manufacturing sites. This is a hugely valuable strategic asset to us now. In terms of milestones, we’ve been very active over the last six months or so. And already this year, we’ve we’ve checked the box on several of our key milestones, we do anticipate some further important milestones of the infectious disease platform, which are listed here. And as I mentioned, we’re launching now these randomized studies with granted in newly diagnosed subjects with colorectal cancer. And we expect our data off of the version two product of slate in KRAS mutant subjects in mid 2022. With that, I’ll finish the prepared remarks and hand back to my anchor for q&a. Thank you.

Q&A

Mayank Mamtani (B. Riley)  23:45  

Yes, thank you, Andrew. That was That was great. So maybe, you know, touching on this topical development around immuno core and, you know, things around circulating tumor DNA has been talked about for a little while. And it’s great to see, you know, agency being progressive, although overall, they have sort of been a little, you know, stranger and broadly speaking, you know, relative to where maybe we were two years ago, but this was clearly a scientific, you know, progress that we need to have. And it’s great to see that that happens. So just maybe help us understand the implications to your program. And as you’ve thought about, you know, Colorado is, you know, it’s not the same disease, first line second line adjuvant, very hydrogenous. It’s not, you know, like lung cancer, it can be but so just can you sort of give us some insight into how you’ve thought about these this disease setting and and what sort of success looks like for you to you know, think of registration at the end of the first readout you will happen?

Dr. Andrew Allen (GRTS CEO)  24:58  

Yeah, so good. questions, obviously. So the colorectal cancer really actually does split for us fairly easily into two groups. There’s the huge group of common and garden, microsatellite, stable colorectal, that’s the group we’re focused on. They don’t respond to checkpoints. And actually even though of course, there are some biological nuances and some have KRAS mutation, some don’t. Some have winter, some have APC mutations, he has many, many different flavors. The depressing reality is they behave as a monolith in that they all progress and die from their disease. This is a tough to treat disease, and it is the second biggest killer of the lung cancer. So it’s an important disease for us to try and tackle. Now, the traditional endpoints, obviously, overall survival clearly, and then progression free survival was used earlier on for some of the chemotherapy agents. But there’s been remarkably little innovation in this space. For many years now. oxaliplatin and irinotecan remain the stalwarts of therapy. There are a couple of approved third line drugs, which have very minor benefits, they’re approved, but their benefits were pretty marginal, median survival was still six or seven months, even in the active arm, they’re barely greater than the control. So lots of room to improve. So as we come in here, we have a vaccine based immunotherapy treating end of life subjects, obviously, is not the right place to be. But Notwithstanding that, we’ve seen dramatic effects in particularly the subjects who are still alive. Some of them now over two years on study, very impressive and obviously very encouraging small data sets. But nonetheless, given the relentlessly poor outcomes for these subjects, this is encouraging. But now we want to move into a randomized study, to definitively address efficacy. So we need an endpoint, there are three endpoints to choose from overall survival, progression free survival, and molecular response. Those are the ones that make sense. First of all, recognize what we’re doing is novel’s never been done before. And what we’ve seen is that looking for short term research response may not be a good endpoint does not to say we won’t get there eventually, in some subjects. But in the short term, we’re driving T cells into lesions where they proliferate. And lesion expansion is actually what you expect, not immediate response. It’s completely different from checkpoints where the CDs are already there, they’re in the tumor, they’re just been shut down. And so when I give them a PD, one antibody, I turned them on again, and they go to work killing the tumor cells. And so I start here, and then I shrink, what we’re doing is starting here with something that has basically no T cells, it’s all tumor, and then we’re driving T cells in and seeing this is actually probably a good thing. So pseudoprogression, we’ve seen it, and it means that looking for short term research response is just a bad idea. I think, long term that may we may see someone you saw those molecular those pulmonary lesions disappearing. But that’s a slow phenomenon. And frankly, PFS and O S will capture that. So as we think about early stage, colorectal, OH S, of course, is a standard endpoint can be done, it is confounded by subsequent therapy. Now, subsequent therapy is not that effective in colorectal. So I don’t think that actually is a huge problem. And I suspect O S will be an important secondary for us. But generally, we’re looking for proximal endpoints that can isolate the benefits of your therapy, rather than waiting for a confounded delayed endpoint like Oh, s. PFS, obviously, is more standard. But for the reasons I’ve just described, pseudoprogression is a real concern. And that’s why we talked to the agency about AI resist as a potentially better tool. And they were open to the idea that they wanted to see phase two data before they commit to it as a registration endpoint. And so it’s something we’re capturing in the phase twos. It may be better, I suspect, it obviously will be better than PFS because we’ve certainly seen pseudoprogression That would be called progression completely inappropriately. You’re just mislabeling patients, flagrantly, so PFS is terrible IPFS is I suspect better. The good news is we have something that appears to be better still, which is molecular response. We just need more data. So we and others are generating those data. And at some point, I think, obviously, it will become a surrogate endpoint that will be used because we all want something good. And this looks to be very good. And it makes sense that it’s better than radiology. So I think we’re part of a trend. I don’t know which of those will become our endpoint, because we’ll have the conversation and

Mayank Mamtani (B. Riley)  29:25  

on that trend, like it looks like the argument was in uveal cancer, you know, that approved indication, kind of, you know, it was probably held by a small group of people. Like with colorectal, like, you know, we’ve done some checks, and we are doing this panel needed today. There’s definitely, I think conversations happening of early diagnosis and, you know, use of molecular markers to predict responses or predict, you know, treatment paradigms. Where are we with, you know, maybe publishing some information on circulating tumor DNA that we’d like outcomes in colorectal or in, you know, in tumors of your interest, because that looks like an important sort of package to put together.

Dr. Andrew Allen (GRTS CEO)  30:09  

Yeah, Natera has published actually quite a bit. They underpin a lot of this early work, and we’re working with them in our in our programs. And so there’s now a lot of data actually published showing that in the adjuvant setting, if you’re, if you have a surgical resection for colorectal cancer, and you remain CtDNA, positive after your resection, you’re in trouble. Because obviously, that tells you your tumor has not been removed, by definition. And so that group of subjects will do badly. And that’s the group that we’re doing our adjuvant study in. As you move forward into metastatic disease, the biology is basically the same, right. And this is not rocket science, right? It’s a marker of tumor burden, the more tumor you have, generally, the higher the CT DNA, if the if you treat somebody in their CT DNA goes down and stays down, that is good. If it keeps rising, that is bad. And so we showed these data in our small data set last September, where we showed that if you had a CT DNA of over 1000, maybe even over 500, and that’s human genome equivalents per mil, so there is now a more standardized assay unit that’s used, then, frankly, we weren’t able to help you because you were likely to die probably within two or three months. And you are too far gone for our product to be able to help, unfortunately, now hopefully other therapeutics can help those subjects. As we moved upstream. Of course, this is the population that we’re much more likely to be able to help. And that’s what our data suggested. And so the data are growing. And I don’t think the biology is going to fundamentally transform as I go from adjuvant to metastatic, we’re talking about essentially the same phenomenon.

Mayank Mamtani (B. Riley)  31:46  

I was at Dr. Andrew, maybe last last question, could you just remind us about your runway, your balance sheet situation, obviously, you have a different part of a pipeline, we didn’t talk about that. It is in the near term been helpful the balance sheet and which definitely is an important metric to follow in current market conditions. So just to kind of round out the story that you’re showing your profile on the capitalist profile.

Dr. Andrew Allen (GRTS CEO)  32:11  

Yeah, so we were active with an ATM in the fourth quarter of last year. And we finished the year with more cash than we had at the end of q3. So we have over $220 million in cash at the end of 2021. So that gives us runway into the 2H23, which obviously is a good place to be, as you say in current market conditions. We’re given the power of our platform that the novelty of the platform and the fact we’re generating exciting clinical data, both in oncology and in infectious disease. We’re obviously pretty active in business development and looking to bring in capital through non equity mechanisms. That’s obviously the prudent path given current equity prices. So we’re in good shape and I don’t think we’re going to be looking to raise through equity issuance anytime soon.

Mayank Mamtani (B. Riley)  32:59  

Thank you. I appreciate you being part of our conference and good luck with the rest of the meetings.

Dr. Andrew Allen (GRTS CEO)  33:04  

Okay, good. Thanks, Mayank.