Hi, everyone. Thanks for joining us today at the last day of the Cowen Healthcare Conference in 2021. Thanks for joining us for the NGM Fireside Chat. I'm covering analyst Ritu Baral, and with us from NGM today, we have CEO David Woodhouse and CFO Siobhan Mangini. Thanks, guys, for joining us. Thanks for the invite. We're always happy to attend the Cowen conference, whether virtual or in person. Yes. Well, hopefully this is the last virtual. Hope to see everybody in the fall for the FAIR Conference for those clients on the line. Let's start with aldafermin, your lead program in phase II for NASH, before we go to your phase II geographic atrophy program. Let's start with FGF19. Where do you see the FGF19 mechanism fitting amongst the NASH landscape today? We have a bunch of new mechanisms that fit in sort of different parts of the therapeutic landscape. As you think of maybe complementary mechanisms, what might be a good fit for an FGF19? David, I'll let you take it from there. The way we think about the landscape and the disease generally is, these are very big populations of patients. Like many of these big metabolic diseases, it probably will require multiple mechanisms. Where we got interested in FGF19, in particular, is related to its mechanism in that it has both a powerful metabolic effect, which of course is one of the main drivers of NASH as a disease, but it also inhibits bile acid synthesis, and that makes it relatively unique when you look across the different mechanistic landscape. Why is that important? Well, we think later in the stages of disease, as measured by fibrosis stage, that bile acid inhibition actually becomes increasingly important as one of the drivers of the disease. The way that you can think about the landscape is many of the mechanisms are primarily metabolic focused, so be it a thyroid hormone receptor beta, the FGF21 class, PPAR. Those are like the anti-steatotic ones. [crosstalk] Exactly. We bring this extra dimension of bile acid reduction that pairs well with, mechanistically within, we have metabolic control as well, but we're focused on these later stage F3, early F4 patients. In which fibrosis reversal is really the key focus of the hepatologist treating them. Earlier stage, kind of F2, F1 NASH patients, where they have a lot of different things going on generally, whether it's type 2 diabetes, whether it's hypertension, hyperlipidemia. The idea that you could quiet the metabolic activity in the hopes of preventing those patients from progressing probably is suited towards one of these metabolic mechanisms. We think as you get more advanced, particularly with the speed we're able to do this is a really useful tool in the FGF19 mechanism to reverse disease quickly. That's what we've demonstrated in our clinical trials is, particularly in this F3 patient population, a 30% placebo-subtracted response rate. What we're interested in repeating in our upcoming trial, which we can talk about in a minute, is that observation of F3 patients, and we're even testing a higher dose too, that we've seen some really promising activity in the past. You may also ask why- This is the 3 mg dose, correct? 3 mg dose. You're right. ALPINE 2/3. You may ask why you don't see another FGF19 analog in development. We actually spent a lot of time in the early days where we got interested in this factor because of its changes around gastric bypass surgery. That's how we clued into the activity of FGF19. When you overexpress it in animal models, you actually see an HCC effect. We had to engineer the molecule in order to not only avoid it, but we actually can block that effect now in these rodent models. We've had this in over 200 NASH patients, so we're very confident that- and over 500 patients across multiple indications. We're very confident that's not a liability of the drug. Yet that created an IP landscape around what modifications you need to make that we think makes it very difficult for someone else to come in with an FGF19 analog. Got it. Can you review for us the trial design of ALPINE 2/3, which will read out in Q2? I guess it's ALPINE 4. How many of those early F4 patients are you studying within ALPINE 4? Okay. I'll start with ALPINE 2/3, which you're right, we're reading it out in the second quarter of this year. This is about 160 patient target enrollment. Four arms, placebo versus 0.3 mg, 1 mg, and 3 mg of aldafermin daily. The real objective of the trial It's designed prospectively as fibrosis reversal as measured by biopsy histology as the primary endpoint. One of the primary objectives is to have a very clear dose response, which we've seen in individual cohorts over time, but this is all in one trial, versus placebo. One of the interesting things about ALPINE 2/3 is that actually the 1 mg dose is a very close reproduction of what we just reported out last year in our Cohort 4 data. So same 24-week time frame. We biopsied those patients baseline and end of treatment. We saw from the combined F2, F3 population, 38% of the patients respond with a 1-stage improvement versus 18% in placebo. 20% placebo subtracted effect rate. When we looked at the F3 specifically, that was the 30% that I mentioned. The interesting thing about this trial is we've got that kind of link to the trial we read out a year ago, and then we have the 0.3 mg dose in there for the dose response curve aspect of it. The 3mg dose is the first time we'll be seeing results from 24 weeks of use at that level. When we tested it at 12 weeks, we saw 42% of the patients in an open label trial achieve a one-stage improvement. We'll be curious to see if we can even do a little better than that at 24 weeks. We'll be looking at safety and tolerability. The Cohort 4 data we reported out last year with 1 mg was very clean between the placebo and treatment arm, and we're hopeful to see that again in ALPINE 2/3. This will be the gating trial, so to speak, to moving to phase III. We'll package up our adaptive phase II that we had four different cohorts in this ALPINE 2/3 trial and have an end-of-phase II meeting with the FDA that will lock in the phase III design that we hope to start by the end of next year or early next. Sorry, early next year? Early 2022? Well, we have to figure out when we can get that at end of phase II meeting and then. Got it. How we go from there. It's around that timeframe. Around early 2022. Okay, got it. ALPINE 4. ALPINE 4. Thank you. Yep. ALPINE 4, we initiated last year. This is in an early F4 population, well-compensated cirrhotic population. The trial is designed as very similar to ALPINE 2/3 in that there's placebo versus those same three active doses, 0.3 mg, 1 mg and 3 mg. This time it's a 48-week treatment period. The feedback from our advisors was with these even more advanced fibrotic patients, that treating for a longer period of time could even be more beneficial. We were in the midst of enrollment on that one. The trial is being run. This is sort of our warm-up to phase III in that we're running it in North America, Europe, and we even have a couple of sites in Asia. Operationally it's getting a warm-up for us. One of the issues is in Europe, how they're handling the pandemic is a little different than the U.S. in that some of the sites have been a little more difficult to open up. We're making progress on enrollment for sure, but this is one of these instances where COVID's having a bit more of an impact than we've seen in some of our other trials. Got it. You won't have that data in hand by the time you start the phase III trial for the F2, F3 patients, correct? I think that's a safe assumption. Hard to tell right now. Your investigation of the F4 patients, again, will have to be like everybody else, just separate from investigation of the F2. That's right. Got it. One thing I wanted to round back on was that 3 mg dose, 24 weeks of treatment, 3 mg. The two major safety concerns around this drug have been the LDL, and initially you had a little bit of a GI signal. Can you maybe talk about how you are trying to manage that GI signal, especially with the 3 mg, and how that GI signal is different or the same as the GI signal seen with some of the FGF21 drugs? Sure. Yeah. What we know is one of the on-target activities of FGF19 is there's actually a prokinetic action in the gut. It speeds up transit time of food in the gut. That manifests as various symptoms. Most frequently, it's just a more frequent need to go, so to speak. One of the things that that taught us is if you actually teach physicians to coach their patients to take the drug around meals rather than right with the meal, that seems to help quite a bit in avoiding these symptoms. These are mild AEs. These are not severe AEs. We were really pleased with the 1 mg dose, which previously, when we hadn't been doing that coaching, actually, you could see a increased rate of these mild AEs in the treatment arm versus the placebo arm in our first cohort. What we were pleased to see in Cohort 4 is you actually couldn't distinguish GI AEs between placebo and treatment. The 3 mgs, we actually saw pretty similar rates, although they persisted for a little bit longer on the 3 mg in our earlier studies. We'll be interested to see in ALPINE 2/3 again, whether that comes down to some form. These, just to put it in perspective, when you look at GLP-1s, this is more mild and more transient than GLP-1s, which are obviously very successful drugs on the market. As it relates to the GI effect of FGF21s, those do seem to be more severe in that we, particularly in Cohort 4, didn't see a significant amount of vomiting per se. Probably not surprisingly, the placebo arm typically doesn't see a lot of vomiting either. You do see kind of a baseline rate of patients. Got it. Subjects having diarrhea. There is a difference of severity from a GI side effect. LDL, we have an algorithm that tells the physician if the patient goes up 10 mg per deciliter after two weeks. T Heir LDL level, whether on placebo or drug, which they're blinded to, then they're put on a statin. That statin use mitigates and actually brings below baseline all the patients. Many of the patients in the placebo arm, you're probably not surprised to hear, actually also go on a statin. Yeah Because these patients generally should be on a statin, and they're not. That really is quite well taken care of with the statin use. Got it. As we think about the Q2 data, what is good data for you? What's good data from the 1 mg arm? What's good data for the 3 mg arm? What would great data look like? Yeah. Good data relates to what I was saying earlier, which is reproducing what we saw with Cohort 4 in ALPINE 2/3 with that dose response curve, which we have a lot of confidence in because we've tested all 3 doses in previous cohorts in different trials. That's really positive in terms of moving forward. We have an outsized fibrosis reversal rate in a relatively short amount of time, clean safety tolerability profile. It's a really good place to be. Great data is, what I like to say is, if we want to get greedy on efficacy with a 3 mg dose and do even more. Already we're seeing a great effect from the drug, that would be really kind of a nice outcome to have even more efficacy from a patient's point of view, of course, as well. What numbers are we talking about? Obviously you have that 20% placebo-adjusted difference in the 1 mg. What's better? We've had so many conversations about what the delta in fibrosis improvement means, especially around the Intercept data. Say 1 mg, you showed 20% placebo-adjusted. If 3 mgs, if you showed, say, 25% or 26%, is that better? Would you really need a 30% to be considered better? Assuming it's the same. Numerically, it is better, obviously. I think it's always the balance of efficacy and tolerability. That's something we'll be keeping an eye on. One of the things that we've talked about before is we probably will bring both 1 mg and 3 mg forward in the phase III. We have enough patients to work with from a safety database requirement. Some of the feedback we've been getting from physicians is, "I actually wouldn't mind having two doses to work with depending on the patient I'm treating." There's this concept of fast progressors, maybe use a higher dose in the fast progressers. They would just, as typical, I think, want as much flexibility as working with a new agent as possible. What we look at is our 12-week study with 3 mg. We saw 42% of the patients achieve a 1-stage change. That's pretty close to 38%. That was an open label study, so we don't know what the placebo-subtracted rate was in that. That's pretty close to 38% at 24 weeks. Our 1 mg dose at 12 weeks showed 25% of the patients achieve a 1-stage response. Some people look at the difference between 42% and 25% and try and extrapolate out. Yeah. I'm just a little careful of reminding people about how pharmacology works sometimes, which is maybe the 3 mg dose is just reaching maximal effect in half the time, and we're sort of reaching a 40% response rate across the population, either quicker or over 24 weeks. I see. Yep. Yeah. Got it. Do you think that the 0.3 mg is subclinical, or do you think that it's somewhere on the dose response curve? I think it's somewhere on the dose response curve. When you look at the non-invasive, we haven't biopsied 0.3 mg patients before, so that will be also new news in this trial. Of course, we show a very consistent effect across a variety of non-invasive markers like liver fat, like ALT reduction, like C4 inhibition. Pro-C3. That looks like consistently, when you look across all those measures, that it's about an ED 50, which is actually why we selected it, because that's a really nice sort of central point to drive that dose response curve off of. Got it. Let's move now to some of the commercial aspects before we move to GA. Oops. Here we go. What do you think? We've always talked about aldafermin as a potential induction therapy. How do you see the economics of that playing out and the reception amongst at least hepatologists? Because if you are focusing on F3 and F4, these folks are going to squarely be. They're not in primary care anymore. That's more than they can handle. How How do you see the economics of an induction therapy, and how do you see the commercial effort there? Sure. Thanks, Ritu. This is Siobhan. We've done quite a bit of research, commercial research, with hepatologists and gastroenterologists to really understand what the primary drivers are of their decision-making when it comes to NASH patients. Probably not a surprise, but what was really interesting in terms of the quantitative analysis is fibrosis improvement is absolutely the number one driver of how they're going to select a therapy by far, compared to anything else. The other piece that we took away from that study is that it's not necessarily about induction therapy. When they find something that is working, hepatologists want to keep patients on that therapy and not necessarily switch them immediately if something's working. I think that's just something too that's important as we think about our commercial and go-to-market activities, and the hepatologist call point. I think the underlying point of your comment, which is the right patient for aldafermin, you're right. It's fast-progressing F2s, that F3 population that David was talking about, where we've seen really strong efficacy or potentially an F4 patient. These more advanced patients, the fibrosis impact of aldafermin is going to be a very powerful agent for them. That is what we're thinking. It's still very early, obviously, in terms of the economics, as we're thinking in terms of the go-to-market and call points with hepatologists. Got it. All right. Let's go to geographic atrophy. This is your anti-C3 antibody. Can you go through the purported mechanism in C3 for GA? How do you think it'll be differentiated from other players with the same mechanism, specifically the Apellis compound, which is going to read out soon-ish? Okay. Yeah. The mechanism, well, there's a big clue in this disease about complement involvement in the pathology, which is the human genetics association. In that there are genetic aberrations that occur throughout the complement cascade that generally are associated with overactivity of the cascade that is well correlated with the likelihood that someone will develop GA eventually. That was the big clue for us and others that, hmm, there's something here. The question is, "Where do you work in this extremely complicated cascade that has all sorts of feedback loops?" The thing our scientists focused in on was C3, because complement C3 is at the node. Well, there are three activating pathways. They all flow through C3, and there are also multiple effector pathways, and they all are downstream of C3. Each of these other complement targets that you hear about, they tend to be on one branch or the other, but they don't knock out the entire cascade. In the early days, that was viewed as maybe a good thing because complement's obviously active in your body for a reason. Right. Infection. Right. Particularly in the eye, which is relatively walled off from the body, that seems like a good place to work, and subsequently, we've learned you can modulate the complement system safely. The connection with photoreceptor loss and complement overactivation is less well understood, but there's an observation of this kind of accumulation of what's called drusen that is associated with these geographic atrophy lesions. That looks like a deposition of cellular debris associated with this complement activity. We started on this program related to the human genetics. Then we, and I think others developing this area, were pleased to see Apellis readout their FILLY study and demonstrate clinically that inhibiting C3 has a significant improvement in the rate of progression of the geographic atrophy lesions. We view that as clinically validating. Now, the difference that we bring to the table is we have an antibody that is very specific to intact complement C3. By virtue of it being an antibody, we don't have to pegylate the agent in order to achieve a half-life that would be suitable for intravitreal injection. We think that brings with it benefits from a half-life point of view, but that may stretch out our dosing to every other month instead of monthly. Also importantly, we think it can help avoid a safety finding that was observed in both the APL-2 study as well as in the Zimura study, which is an increased rate of occurrence of exudative disease in these geographic atrophy patients. [audio distortion] active disease, basically. Yeah. What looks like wet AMD, but they characterize it a little different, but it is treated with anti-VEGF, and so it effectively is like a wet AMD. What we were interested in seeing in our ongoing phase II is if we can see a similar effect as they've seen but avoid that safety effect and be able to stretch out dosing. We think that would be a really good positioning for the agent. Great. Tell us a little bit about your CATALINA study that's ongoing? As well as the safety data that you presented, the safety and PK/PD data that you presented at AAO this past year, and what that tells us about the potential clinical profile of NGM621. Well, like any phase I, you're really focused on safety. We were pleased to report at AAO that the highest dose we can fit in 100 microliters, which is the maximal dose for one intravitreal injection, was safe and tolerable in these geographic atrophy patients we tested it in. That's 15 mg, and that's what we've taken forward into CATALINA. That's one important check in the phase I. As you mentioned, the other thing we were able to learn from this trial was to model the PK and resident time of the agent in the eye. The way we were able to do that, ethically, you can't do regular intravitreal taps on these patients, but what you can do is measure systemic levels of the drug. Based on our non-human primate studies, we know the correlation between resident time in the eye and systemic, and we can model that into humans. Actually, our collaborator, Merck, helped us with that modeling. The result of that modeling showed that the amount of NGM621 that's in the eye is equivalent to an inhibition of 90% of the C3 activity for up to seven weeks. That supports this idea that we can stretch it out to every other month in terms of dosing frequency. Of course, we'll have to read out. CATALINA is designed as a year-long treatment with monthly and every other month sham and active drug arms. Four arms overall. Yeah. That will, of course, ultimately be the test. If we can see our monthly and every other month dose are similar in effect rate, then I think we're in great shape for taking it forward that way. Got it. Right now, what are you saying about your- w ell, I guess two things. One, have you had DSMB looks in CATALINA at this point? Two, what is your current messaging on when top line CATALINA might be out? We haven't seen any safety concerns as we've been running the trial so far. We haven't guided yet on timing for the trial either. We are just finishing opening up all the sites. That will give us the steady rate of enrollment we'll be able to project out and give some guidance on timing. I do remind you, it's a year-long treatment period. Once we give the timing for enrollment completion, you have to go out a little over a year. Yeah. All right. We do not have much time left, and we have three oncology programs to get through. They are sort of early stage, so maybe David, can you take us through the three programs, and very importantly, data catalysts for when they might start kicking off data? Sure. Why don't I start there? One of them actually isn't so early, which is NGM120. We just announced that we're moving it into a placebo-controlled study in first-line pancreatic cancer patients on a background of gemcitabine and nab-paclitaxel. We've been running this phase I-A/I-B in preparation for this transition into what really looks like a phase II, but it's an expansion of the phase I-B arm in this trial. That's really exciting because we'll be able to clearly see if our thesis here, which is both from a cancer-associated cachexia improvement in these patients that typically develop that syndrome, unfortunately, but also on this theory that we may be able to have a greater impact on the cancer itself, because either they can tolerate more therapy because they feel stronger, or there's preclinical data we have to support that there may be effect on cancer directly with this agent. That's exciting. I can't tell you when we'll have data from that part of the study, but I can tell you that in the second half of this year, we'll be sharing data at a medical meeting from the phase I-A/I-B that allowed us to move into this. NGM707, we expect to have in the clinic by the middle of the year. This is an antibody that is a dual antagonist of ILT2 and ILT4. We have some clues of clinical relevance and effect related to some Merck data that was put out last year that has an ILT4 inhibitor. What we're bringing to the table is an extra dimension to that inhibition by virtue of this extra activity being antibody. This program, in addition to NGM438, which is our third oncology program, stems out of research we started about six years ago around the concept of myeloid cell suppression in the solid tumor microenvironment. As you're probably aware, there's a large population of immune cells associated with a tumor. They're just inactive in many cases, and in fact, suppressive. When you put a T-cell checkpoint inhibitor on board, they can't sort of crack the activity around these tumors because of the suppressive environment. What we're trying to do with both NGM707 and NGM438 is come up with receptors that are associated with overexpression in these "cold tumors" as a way of taking a break and reprogramming that myeloid-derived suppressor cells into activating myeloid cells. That's what we'll be testing with both those agents. Got it. Great. Well, we are at time now. Thank you guys so much, David. Thank you, Siobhan. Look forward to catching up with you guys, especially after your Q2 data, and maybe having a broader conversation about your phase II GA program after competitive data towards the middle of the year. Great- Terrific. On the catalyst flow. Thanks so much, guys. Appreciate it. Thank you. Have a great conference. Thanks for having us. Bye