Hey everyone, we're back. Happy to have Daniel Faga, your CEO of AnaptysBio, for a fireside chat today. Maybe I'll kick it over to Daniel just for a quick intro to AnaptysBio and then we'll get into the Q&A and cover some of the recent questions around ECR as well. Dan, over to you.
Sounds good. Good timing for this this morning. Nice to see you, Alex. I appreciate the opportunity. You know, I think most folks know AnaptysBio is focused on the development of antibodies that modulate disease-driving and overexpressed immune cells. The key here is the cells that we're targeting across our development portfolio do not exist in significant numbers in healthy tissue or blood. Our goal is to target those cells and restore immune homeostasis to patients who have autoimmune and inflammatory disease. We have three programs in clinical development. Our lead program is rosnilimab, which is a selective and potent depletor of pathogenic T cells. We completed a phase IIb trial this August with data that's extended out six months on drug, then three additional months off of drug out to nine months, showed high impact and durable responses, as well as a safe and tolerable profile.
We'll come back to that later. We also completed the enrollment of a second trial in ulcerative colitis. The initial data will read out in Q4 at three months. Subsequently, we'll follow that out to six months and 12 months efficacy data and safety data. Our second program, ANB033, is a CD122 receptor antagonist. We are nearing the initiation of our initial phase IB indication, which will be in celiac disease. We are hosting an IR event dedicated to this program, also upcoming in Q4. Last, our third program, ANB101, is a BDCA2 modulator. We initiated the phase IA in healthy volunteers earlier this year. We're excited about all three of our programs. We're well capitalized. We had almost $300 million coming into the summer. We are also backstopped by a substantial royalty interest in a drug called dostarlimab (Jemperli), which is sold by GSK.
That is going to do over $1 billion of revenue this year, which will trigger a $75 million milestone to the company. It's in a significant and very early part of the growth at Jemperli. We have big royalties around that program. I'll kick it back over to you, Alex.
Yeah, great. By way of background, you know, the ECR abstracts came out yesterday. Lilly's drug parasolumab, which was a PD-1 agonist like rosnilimab. We have the abstract from the phase IIb study that they discontinued last year. I think the top line here was that by week 12, they saw four malignancies, and then they saw four more out through week 60. You know, just given the mechanism here, and I think when folks think about PD-1 agonism, you think about this is the opposite of Keytruda. How is this not like the worst-case scenario for the class?
Yeah, yeah, and we've been waiting for this, for some insight into this trial from Lilly for quite some time. I will address that spot on, Alex, but I want to give some background on how rosnilimab actually works, what we designed around, and what we've seen so far with our drug. Then I'll respond to what our competitor's abstract says. I think, as everyone knows, we engineered our program to be a potent depletor, but it's also a selective depletor of pathogenic T cells. We're targeting pathogenic T cells. Generally speaking, we're talking about TPH cells, peripheral helper T cells. As I mentioned in my opening remarks, these cells, they don't exist in large quantities. They're actually quite rare in healthy tissue and periphery, 2-3% in the periphery. We see elevations of 2-3-fold higher in diseases like rheumatoid arthritis, like ulcerative colitis.
Also then in the synovium in this disease in rheumatoid arthritis and lamina propria in ulcerative colitis, you see almost all your T cells are these overexaggerated. What we ended up seeing in the synovium was almost all of them were TPH cells, not even T effector cells. We designed our drug to hit this very, very hard. What we ended up showing translationally was greater than 90% depletion, both in the periphery at all doses and then in the synovium at our middle and high dose of greater than 90%, close to 100%. Almost a full depletion. Again, this is targeted. You don't want these cells in a healthy environment, and that's what we talk about when we say restoring immuno homeostasis. T cells are constant. We do not prevent the activation of naive T cells to PD-1 expressing T cells.
It's like drugs like Abatacept, which are much broader hitting. It's been on the commercial market for decades, and there's been no linkage to any safety concerns there. Our drug does exactly what we want it to do. Now in hundreds of patients out over time, in phase I and II trials, and I think importantly in our phase IIb trial, this is a 424-patient study. We dose patients for six months, and we observed safety out through nine months. Our colitis trial is over, up to a year long in terms of active dosing. We've seen nothing of concern in any of our trials from safety. I think importantly, when you look at tolerability, less than 2% of our patients in the rheumatoid arthritis study came off due to AE. Only one patient after three months came off this trial.
And the drug was highly effective, and patients are staying on this drug because they're feeling better. Our placebo rates look like the active rates in the first three months. Right down the line, they include infections. We had very few severe infections. It looked like placebo. No deaths, no malignancy. We had one case of MACE in our RA trial, which was explained as at the lower dose. It was an embolic situation. They had severe blockage, probably shouldn't have even been in the study, but has nothing to do with the drug and was ruled that way. We're not seeing anything of concern on the safety or tolerability side. On efficacy, we had very high ACR20 rates, best in disease. Those responses deepened between three and six months. We saw conversions to ACR50s and 70s. CDI low disease activity patients, we saw move into remission.
Off drug, we saw stable clinical outcomes. Overall, the profile we're quite excited about in RA, and we'll talk more about UC later. Eli Lilly, on the other hand, this is a very different drug. We have been talking about this for years before clinical data was even available in arthritis by either party. When you look at Lilly's phase IIa study to start with, which was 100 plus patients, what they showed in translational data was less than half of those TPH cells were depleted, meaning more than half of those TPH cells, and this is just in the periphery, were still activated and driving disease.
While I think everyone was excited that there was a new class in RA at the time, this is going back two or three years, what Lilly's phase IIa data showed was a decline in efficacy over time from week 12 out through week 24 and then off drug. The exact opposite of what you saw with us, which was a deepening of effect and then a stable outcome. Very different profiles. They did not see any safety events in that phase IIa study linked to this mechanism. There was one malignancy in their phase IIa trial that happened to be in placebo, which wasn't discussed, but it was there. I think that's a key point. In arthritis, the comorbidities here are real. You have a two to three-fold increase in MACE, two plus fold increase in malignancies just by having disease.
All patients are on background methotrexate in these trials and in the commercial setting. You're already immunosuppressed just from the companion drug that's on board with these mechanisms. I think there's a lot of context that needs to be set. With that background, Alex, what came out yesterday was an abstract from Eli Lilly's phase IIb study, which was something more in range with the size of our phase IIb study. It was a 500-patient trial. They were placebo controlled for 12 weeks, and then they went out through 60 weeks. They did complete the trial, as according to clinicaltrials.gov. The trial ended in January of this year, after the last patient hit week 60, from what it looks like based on the publicly available information. Efficacy, not surprising. We'd heard all of this. The benefit wasn't there with their drug.
It looked very similar to what they showed in the phase IIa study. The ACR20 rates weren't interesting. They barely hit stat siga only their high dose. They had a placebo-adjusted delta of 12% on ACR20. We had a placebo-adjusted delta, at our mid-dose, of over 19% ACR20. Very different efficacy. I think conveniently here, Eli Lilly chose not to describe the efficacy after week 12. Now, we know what happened in the phase IIa study, and I think we've talked about this publicly for quite some time. We do believe that they lost very significant efficacy again after week 12. This is everything to do with their drug, not the mechanism. You have a drug that's not potent or effective, doesn't sustain efficacy. What does that mean? They ran a 60-week trial, and sometime between week 12 and something beyond that, they lost efficacy.
Patients now on an active drug, not controlled against placebo, with a drug that's not working that great, and they lost a response. You effectively have patients who have been on one or two plus prior biologics, including now parasolumab, Eli Lilly's PD-1 depletor, and methotrexate, and there's no control. We knew there were going to be events in this trial. We've been saying that for a long time. To your point, Alex, they did announce there were some malignancies in the trial, this time on the active arm, and they cited some possible MACE events sometime after week 12. They gave no context or detail around any of this. It was a very high-level abstract. Hopefully, they're responsible and they put out more specific information at ACR in the poster that they received. We've been talking with KOLs.
There's a lot of KOLs even in this abstract that I recommend folks reach out to. It is almost impossible for a drug to drive cancer in the first 12 weeks, right? They have four malignancies. We've heard there's been a lot of discussion even within Lilly, going back in the last year plus of how to position and think about this. At the end of the day, if you don't have any efficacy, you don't have any benefit, and your benefit risk profile is not going to be attractive. It's really hard to explain what this is until we see more details. We do not believe it's linked to the mechanism specifically. We think it's linked to a disease that has comorbidities with a drug over at Lilly. It just doesn't work good.
Okay. Lots of follow-ups. I guess the first question though really is just, you know, do you think you seem to be pretty confident this is not on target? I guess, you know, how can you rule out that this is not on target? Like, how can we get comfortable with this profile? If I were to push back and just say they've still dosed more patients than you, is this just a matter of time for rosnilimab?
I mean, they didn't dose grossly more patients than us in the study. We had a very large study in arthritis as well. We are also running an ulcerative colitis study in that trial.
What can you say about ongoing blinded safety data, if anything?
Exactly where I was going to go, Alex, is, you know, I'll get into the trial design, I think, in a little bit, but at the end of the day, it's a one-year-long study. The highest dose is higher than what we observed in RA. Just for context around that, at some point, regardless of the dose, if you're depleting very specifically the target and there's nothing else to interact with, you're not doing anything else with the drug. There's nothing else to engage with. I think that's an important note. The reason we're dosed higher is because we need enough antibody to get into the gut. That said, we do have higher doses in that trial. The trial's been ongoing for about 20 months so far in terms of FPI to present. We've seen nothing in blinded data that suggests any issues with this drug.
No severe infections, no malignancies, you know, et cetera, again, on a blinded basis. Who knows what arms are showing up in the data, but we're not seeing anything of concern. When we report that data out at three months, six months, and a year, we are generating our own data set that goes out to a year. It's not a 500-patient study. It's 136 patients. At some point, it all adds up across the totality of what we're seeing with their drug, and we're not seeing anything here of concern.
Is FDA or other regulators worried about this?
Yeah, thanks. We've heard nothing from the FDA regarding the class. If there was a class issue perceived by the FDA, I'm sure we would have been notified of them, you know, going back 6, 9, 12 months from now.
Yeah, nothing as it relates to the trial designs that you think has been related to concerns around this, like not crossing over your placebo patients in the RA study, etc.?
The trial design in our phase IIb and not crossing over placebo patients had everything to do with conservatism from the room division because there's treatments, there's plenty of treatment options out there post 12 weeks, and we hadn't shown proof of concept. It has nothing to do with safety. Again, the GI division let us do a crossover at three months and let us go out a year. It has nothing to do with safety. It had to do with efficacy, lack of proof of concept. Everyone's running the same phase IIa trials in RA, pre-proof of concept. Like I said, we've heard nothing from the agency of concern.
Where do Johnson & Johnson and Gilead fit into this whole story here?
Johnson and Gilead have different programs. Johnson & Johnson has a depletor. Gilead has a non-depleting antibody, so they are slightly different. Johnson & Johnson, admittedly, it's been less, less raw number of patients and shorter trials over three months, but their safety was also clean. Drug was also less potent, less effective. We haven't heard anything out of Gilead in terms of their data to date.
Okay.
There is the mechanism specifically, and then there is modulating T cells. We are far away, not the first company or the first mechanism to target T cells in autoimmune disease. Across multiple classes targeting a broader array of cells, not as specific as what we're doing. We haven't seen this, but you asked a question, and I got to be balanced on this. Like any drug, it's going to take thousands of patients and a lot longer to clear it, right? I think that's an obvious comment, but it needs to be said. We just haven't seen anything to date, and we've been testing this drug now in hundreds and hundreds of patients.
Okay. We'll see at ACR. I think we're good for this for now. I want to pivot to UC, this will be your next readout in the fourth quarter, as you said. Maybe at a high level, can you talk about this trial design and what you think good looks like for this drug in UC?
Yeah, so counter to what we were able to do in RA, where we did have to take patients off a drug if they hit a pretty high bar in efficacy in three months, the dynamics here in UC are very, very different, in the commercial markets, as well as what the TPP ultimately is. Going back two and a half years when we designed this trial, what we wanted to do based on the hypothesis here is that patients are going to want to get deeper response as homeostasis is restored over time. We wanted to run a trial that maximized our ability to drive clinical remission out through six months plus. That's exactly what we did. We ended up enrolling 136 patients. Half the trial is going to be in advanced therapy with advanced therapies, half naive to advanced therapies. We have three arms, two active doses.
We have a higher, every other week dose, Q2W, and a kind of a mid-low dose in this study, Q4W versus placebo. It's a treat-through design, meaning patients on the active arms will take drug for six months. At six months, they'll get assessed formally for symptomatic benefit only, so relatively low bar. They're eligible to stay on this drug for a year. At six months, whether you're in the Q4 or Q2 week, you are moving to a Q8 week dose. What we ended up seeing in RA is we did drive to that full depletion, stable activity, and off drug for 12 to 14 weeks. We saw stable results. We went with a Q8 week dose in this trial once you get out beyond six weeks.
At some point, like I said earlier, the PD here is such that if you're driving the depletion, you are sustaining at a concentration high enough where you should need less drug over time to sustain the translational effect, which we are seeing, at least in RA, result in longer-term clinical benefit.
Yeah.
The placebo patients, that arm at three months, they do crossover. If you're not in symptomatic benefit, they will cross over to the high dose. At six months, you end up with four arms. You have a crossover group. You technically still do have placebo control. It's all blinded. It happens to the background, the crossover. I'm not sure how many placebo patients will make it six months. That's a bit of its own experiment. The key thing that's going to end up reading out in Q4 is the initial three-month data. Patients will still be materially in progress out through six months plus at that period of time. What we're looking for here is benefit versus placebo on the change in MMS. We will not have max remissions at three months. There's no way. That's not how we think the mechanism is going to play out.
We're looking for something that's competitive to other biologics in the space, the Skyrizi's T01A. When you look at things like clinical remission, the ultimate bar here that matters. There will be a subsequent read at six months to see if we're driving to effective remission rates that are differentiated. Just for some context, if you will let me, what are we looking for here in terms of results? There's a lot of numbers that got thrown out in the context of UC, clinical remissions, in naive patients, in bio-naive patients with Skyrizi and the T01As. We're talking best case 30% on an absolute basis.
Yeah.
Now, that's very different when you look at experienced patients on other drugs. It's now in the high teens. On a blended basis, when you look at our trial, 50-50 advanced therapy naive, something like 25% remission, that would be clinical remission at three months. That would be on par. You don't even know if we have to be there. We have to be directionally in that ballpark. Why is that? Because with every other drug that exists, that 25% average clinical remission rate, when you get out to six months and a year, a third to half of those patients lose response.
Yeah.
If you do the math, multiply those together, we are talking in the mid-teens. Patients are in clinical remission out a year.
Yeah.
Said another way, 85% of patients sometime between three months and a year, not even just at a year, are looking for a different treatment option. That is a massive market allowing class cycling, and you need different mechanisms of action here, not just the me-better approach of other stuff that patients have been treated on.
Yeah.
We think the market's huge for a biologic that just looks like what else is out there. We have the opportunity to drive deeper remission over time and have stable outcomes to a year. If anything, that's the proxy off of RA with what we've already shown.
Yeah. As it relates to specific three-month readout, totally hear you about the comparability. One of the questions heading into the RA top line was, should we expect to see a dose response? Here you're adding a higher dose due to sort of driving greater gut exposure. What do the expectations look like there?
In the RA study, we had three active arms.
Yeah.
In the periphery, they were all potent. There was no difference translationally, right? It was very quick. We saw in the first read, full depletion of the cells at every dose. However, in the synovium, the lower dose wasn't as effective. We did see a real dose response when you looked at bio-experienced patients, like the harder to treat ones. Particularly once you get between the second and then into the third line patients, there was a dose response. The sustainability of those results were different over time. That looked at, you could see translation in the synovium. The low dose just didn't get enough drug there. We think if we got enough drug to the synovium, it would have been as potent. The corollary in UC is it's harder to get drug into the lamina propria than it is the synovium. We don't have the low dose here.
The question mark is going to be, does that monthly dose that we're using, is it potent enough? Not in terms of how the drug works, but is there just enough antibody to drive in the ALP? I think it's reasonable that we could have a dose response for that reason alone. If drug gets there, we think it will be potent. It's really a question of, does the antibody get there or not?
Is it?
At that middle dose, we don't think we're going to miss this at the high dose in terms of getting enough drug in.
Do you think this data set will be sufficient for you to make strategic decisions around the path forward for rosnilimab, or do you think you need six months plus data?
The TPP is defined at six months. We will not, as a company, make a decision on how to move forward until we see six-month data. We need to see how deep the remissions are and that they're stable and the drug is ultimately competitive in the mid to long term. That's the development. That's what makes this a developmental drug. Now, again, we saw it in RA. I think in 2026, we'll have the collective of information around UC to make a judgment call. Do we hit the TPP and does it make sense to move this forward into phase three?
There is no shot that we get initial six-month data in this readout like you kind of had for RA at the top line?
Yeah, and so the difference here between UC and RA is RA was a huge study. This is a much smaller study, and the degree of error on partial data sets at six months wouldn't be as reliable. When we announced the RA data, we had 60% of a 424-patient trial out through six months. If you run the math through, that's still bigger than the trial that we're running in totality in UC.
Yeah, yeah.
This is going to be a three-month, does the drug work? Does it beat placebo? Is it safe, tolerable? I think the other key thing here, we'll show clinical response and clinical remission. These endpoints, these secondaries are important. Tolerability and safety, we will give a safety update for the totality of the trial. The important thing here is the tolerability. Are patients staying on the drug? Are they feeling better? In the commercial setting, you don't get scoped at three months. It's a discussion. Are you feeling better? You're getting scoped at six months or maybe even a year, right? We're running our trial like it is in the real world. I think that's what's important here. The TPP is reflective of real-world commercial, and that's what three months is, a kind of a signal that our patients are feeling better, staying on the drug.
We'll see the full remissions at six months in 2026.
For the rest of the pipeline, can you remind us sort of what the catalyst path is for moving towards the clinic across both programs?
Yeah, so ANB033, like I mentioned earlier, is a CD122 antagonist. We initiated the phase IA trial in healthy volunteers 11 months ago. We're effectively through that. We are going to have an IR event coming up specifically on this drug, a little bit of a teach-in on why targeting CD122. We think it's a lot more interesting than just IL-15. CD122 is a dynamic receptor that blocks both IL-15 and IL-2 on subsets of T cells, cytotoxic CD8s and CD4s. We'll do a lot of education on this. We'll show differentiation data of our drug versus competition. We'll walk through the phase IA healthy volunteer data sets, as well as lay out the rationale for why celiac disease makes a lot of sense for initial indication, as well as show, I think, pretty compelling data to support our drug in that disease and walk through the trial design.
We'll probably give some perspective of other areas we're going to go or potentially could go in additional phase IBs next year.
Great. Dan, appreciate you taking the time and going through everything. Have a great rest of your day.
All right. Nice to see you, Alex. Thank you for your time.