AnaptysBio, Inc. (ANAB) Earnings Call Transcript & Summary

Good day, and thank you for standing by. Welcome to the rosnilimab Updated Phase IIb rheumatoid arthritis data update. [Operator Instructions] Please be advised that today's call is being recorded. I would now like to hand the conference over to your speaker, Dan Faga, President and CEO. Please go ahead.

Hello. Thanks for joining us today. We're thrilled to share exceptional results from our Phase IIb trial in rheumatoid arthritis or RA rosnilimab, our potent depleter and agonist that targets the PD-1 co-inhibitory receptors, which are selectively expressed on the surface of activated T cells. We believe these updated data reinforce the promise of our best-in-class antibody and the potential for a best-in-disease profile in RA. Today's press release and this presentation contain statements that are forward-looking. Anaptys' CMO, Dr. Paul Lizzul, and I are joined by 2 prominent rheumatologists, Professor Paul Emery, Professor of Rheumatology at the University of Leeds; and Professor Jonathan Graf, Professor of Medicine at UC San Francisco, who is also an investigator in our trial. Both physicians will be offering their views at key points during our presentation and will also join us for Q&A after our prepared remarks. In February, we reported differentiated initial safety and efficacy data for rosnilimab from our 424-patient placebo-controlled study. The primary statistical analysis was conducted at week 12, where all doses of rosnilimab hit statistical significance versus placebo on the primary endpoint, DAS28 CRP as well as on ACR20. Furthermore, rosnilimab demonstrated impressive achievement of CDAI low disease activity or LDA at 3 months, which as of that time, initially showed a signal of stable outcomes through 6 months. Today, we're excited to share updated clinical and translational data, and we have 3 key points to make. First, rosnilimab's efficacy results are JAK-like in both speed and in magnitude on ACR20 from just 3 monthly doses. Then for 6 months, we observed JAK-like efficacy on CDAI-LDA, CDAI remission and ACR70, all high order responses representing disease control. We also observed corresponding clinically meaningful reductions in pain and HAQ scores. These are key factors in a patient's own decision to stay on or switch drugs. Importantly, the safety profile is notably unremarkable, and this compares favorably to standard of care. Second, we have not observed rosnilimab's full potential in our reported response rates as continuation criteria prevented certain patients from receiving treatment after 3 months. Many of these patients that were ineligible to advance did have significant benefit on rosnilimab, including a dozen patients who achieved LDA in their first follow-up visit off of drug at around 4 months. And third, after 6 months of treatment, patients saw durable responses off drug for at least 2 months. Commercially, this allows for monthly treatment from baseline, followed with potential extended dosing intervals and maintenance phase. The RA market after patients have progressed on first-line biologics like anti-TNFs generates greater than $10 billion of revenue yearly in the U.S. alone. We're excited about the significant difference rosnilimab can make for these patients with RA, a disease where there hasn't been a new mechanism approved since 2012. In a mature market like RA, first-line therapy is usually mandated by payers. That said, in a clinical trial, treating patients naive to b/tsMARDs provides a great comparison on relative potency. Rosnilimab is comparable and even surpasses 6-month results on CI-LDA, ACR50 and ACR70 versus results from Phase III head-to-head active comparator trials of today's most utilized therapies on the market. On this slide, in the blue bars is pooled data across rosnilimab's 3 doses. It is shown relative to comparators in 3 gray shaded columns. First in dark gray is Rinvoq's data from its head-to-head Phase III study versus HUMIRA in naive patients. Second is HUMIRA's results from that same trial, where HUMIRA outperformed itself relative to other trials, including when it was a comparator to Kevzara in a Phase III head-to-head [indiscernible] trial. And the third gray column shows how Kevzara, which is the best-in-class IL-6 antagonist, performed in that same Phase III head-to-head study. Note, we are not using a responder analysis here. Given the trial is complete through week 28, we are able to use a nonresponder imputation or NRI analysis to calculate rosnilimab's response rates. Paul will later explain this in more detail, but I'll just say for now, this is the most stringent and conservative way of assessing response rates in our Phase IIb trial. To be clear, the denominator we are using on this slide to calculate the response rates represents 188 patients, which is the intent to treat with a total number of naive patients randomized to rosnilimab in the trial, regardless of the eligibility to advance in the all active treatment period. Now let's look at the data for the experienced population. Here, we also use a conservative NRI analysis. And for clarity on this slide, the denominator is 130 experienced patients randomized to rosnilimab in the trial and allocated across the 3 dose arms, which are graphed separately. The rosnilimab low dose monthly is in the same color, the mid-dose monthly in blue and the high dose twice monthly is in green. In this experienced population, of whom 29% have utilized a prior JAK inhibitor, rosnilimab's data in both the mid and high doses are compelling and comparable on CI-LDA and ACR70 to Rinvoq, which is the first dark gray bars for each of the endpoints. This Rinvoq data is from its head-to-head all active comparator Phase III study versus Orencia. Orencia's data is in the second gray bars. These comparator studies, patients know that they're on an active drug from baseline, and you typically see a significant boost to the response rates for the same drugs versus their results in placebo-controlled studies. As you can clearly see that benefit on Orencia's response rates in the second gray bars, which, again, are from Phase III all active head-to-head trial and relative to Orencia's results in the third gray bars, which are from Orencia's own Phase III controlled trial. Importantly, the rosnilimab results also beat our 6-month target product profile indicated in the orange horizontal lines, where we have stated the need to be clearly bio better and approaching JAK-like in this experienced population. This can be seen very clearly when rosnilimab data is compared to that third gray bar, which again was Orencia's data in its own Phase III trial. I'll now hand it over to Paul to review the details.

Thanks, Dan. Let me start with a quick reminder of the study design, which includes 6 months of active treatment and 3 months of off-drug follow-up. We enrolled 424 moderate to severe RA patients and approximately 40% were b/tsDMARD experienced, meaning they were previously treated with a biologic or JAK inhibitor. All of these patients regardless of study arm were required to be on a stable background conventional DMARD such as methotrexate. Patients were randomized to either 100-milligram subcu rosnilimab monthly, 400 milligrams monthly, 600 milligrams twice monthly or placebo. At week 14, rosnilimab patients who achieved LDA or low disease activity continued with their assigned treatment through week 28 or 6 months in a blinded all active treatment period, during which they remain blinded to their dose of rosnilimab. If a patient did not achieve LDA by 3 months or they were on placebo, they were required to discontinue from study treatment and then entered an off-drug follow-up period with continued clinical assessment. Let me explain this important point a bit further. The requirement to achieve CDAI-LDA within 3 months in order to continue on rosnilimab was not only mandated by regulatory authorities, but was also a very high bar for patients. This is earlier than rheumatologists typically expect to see this sort of results for patients who are starting with high disease activity. The CDAI scale on the left side of the slide shows the breakdown of disease activity starting at high disease activity and down to low disease activity with a cutoff below 10 and then remission with a cutoff below 2.8. 95% of patients entered our trial with high disease activity, defined as a CDAI of greater than 22 and patients had a median CDAI score of 36. Some high disease activity patients who saw impressive and meaningful reductions in their CDAI score but didn't yet hit LDA by week 14 were ineligible to continue receiving treatment in the study. For example, one patient with a CDAI of 54 at baseline improved to a CDAI of 12 by week 14. Despite achieving an ACR70 response, this patient was unable to move into the blinded all active portion of the trial, even though they subsequently achieved LDA at their following visit. More generally, there were patients who didn't achieve LDA by 3 months and were ineligible to enter the all-active treatment period. However, many of them did achieve ACR20, a proxy for adequate improvement. These patients missed their opportunity to deepen their responses out to 6 months. Excluding patients who hadn't achieved LDA by 3 months, but did achieve ACR20 is far from a reality in clinical practice. As these examples illustrate, preventing patients from advancing in the all-active treatment period created a ceiling as of week 14 for the max response rate achievable in our Phase IIb trial. On this slide, let's look at the patient population and their disposition. In the red outline box, we highlight the ITT population of 318 rosnilimab patients randomized in the trial, which includes 188 naive patients and 130 experienced patients. Unless otherwise specifically noted, these are the numbers used in the overall denominator for the nonresponder imputation or NRI analysis of the ITT in this presentation to calculate rosnilimab response rates, regardless of whether we are focused on the first 14-week blinded placebo-controlled treatment period or through the full 28 weeks, including the all-active treatment period. This is the most stringent and conservative way of assessing the response rates regardless of the trial design. At week 14, 220 or 69% of the 318 rosnilimab patients across all arms achieved CDAI LDA response and therefore, continued treatment in the blinded all-active treatment portion. But for clarity, the ITT population for week 28 includes all rosnilimab patients randomized in the trial. As shown in the box on the right, 77 rosnilimab patients made it to week 14, but had not yet achieved LDA and were ineligible to enter the all-active treatment period, but instead entered the off-drug follow-up period. In an NRI analysis, these patients are all considered nonresponders for all endpoints and all time points beyond week 14. Okay. So now with this accounting out of the way, I also want to highlight a few key points captured in the green box on the top of the slide about the disposition of patients through the all-active treatment period from month 3 to 6. 95% of patients completed the all-active period or said differently, only 8 of the 220 patients who entered the all-active treatment period discontinued, demonstrating remarkably high patient acceptance of study treatment. 7 of those 8 patients were in CDAI LDA response at the time they discontinued and dropped because of things like missed visits, patient relocation or even 2 patients who were required to discontinue at or near their 24th week visit only because their particular study site location was shut down due to a severe system failure. Only 1 patient in this time frame continued due to an adverse event, which was a moderate headache treated with Tylenol at week 20. Altogether, this is a strong indication of the tolerability of this drug, and this is important because we're talking about a lifelong chronic disease. Lastly, while all patients are through week 28, the study is still ongoing with the off-drug follow-ups. As of March 11, 2025 data cutoff, approximately 75% of patients completed week 34 and 50% completed the full 38 end of week study visit. Now it's time to dive into the trial results. As Dan said, our 3-month data were JAK-like on ACR20. Here, we have graphed the 3-month ACR20 data for the 174 experienced patients, 130 of whom received rosnilimab across the 3 doses as well as the 44 placebo patients. Four notes when looking at this graph. First, as will be consistent on all slides moving forward, the rosnilimab low dose is in the salmon color, the mid-dose is in blue and the high doses in green. Second, absolute and placebo-adjusted response rates for both experienced and naive patients are in the chart on the right. The data for naive patients, as expected, are even better than the experienced patients. Third, all patients are on background conventional BMARDs, so think of placebo as placebo plus methotrexate. And fourth, for the experienced population, 29% were previously on a JAK inhibitor. This analysis shows separation from placebo starting as early as week 4 with rosnilimab patients showing a consistent deepening of response through week 12 as well as through week 14. What we're going to do next is layer on top of this graph, the ACR20 responses observed in the Phase IIb placebo-controlled trial of Rinvoq in experienced patients. Now in orange are Phase IIb results of Rinvoq in dark orange and placebo in light orange. Rosnilimab absolute and placebo-adjusted efficacy numbers are comparable to those observed in Rinvoq's as early as week 4 through the time where Rinvoq starts to plateau as early as week 8. Additionally, by only 4 weeks, rosnilimab's rapid onset of response and separation from placebo is comparable to Rinvoq's, which is notable because JAKs are often lauded for their quick speed of onset. Now let's go a step further and show data from Rinvoq's Phase III all active head-to-head trial, which you can see added in the burgundy line. The apparent improved performance of about 15 percentage points for Rinvoq in this Phase III study is likely attributable to the bias from patients knowing they were all on active drug starting from week 0 or baseline. As you can see highlighted in the green circle, even with the inherent bias, rosnilimab patient data are still comparable at week 14. These data demonstrate that rosnilimab's efficacy generally looks as good as Rinvoq's Phase III data by 3 months. The same trend is echoed in CRP data for rosnilimab and Rinvoq. CRP is a biomarker that represents an objective measure of inflammation. There are consistent reductions in CRP in rosnilimab patients, which separated within the first month from placebo. These robust reductions in inflammatory disease biomarkers further substantiate the clinical results in RA and provide support for other diseases where systemic inflammation contributes to disease pathology such as ulcerative colitis. Here, you see the mean change in CRP for experienced patients in all 3 rosnilimab doses through the placebo-controlled part of the trial. Now let's layer in data from Rinvoq's Phase IIb placebo-controlled trial in experienced patients in orange and Rinvoq Phase III head-to-head trial in [indiscernible]. Despite substantially higher ACR20 rates in Rinvoq's Phase III head-to-head study relative to its placebo-controlled Phase IIb, the CRP reductions are very similar. This emphasizes that an objective lab measure like CRP is not subject in the same way to trial bias and can serve as a useful way to compare the biologic impact across drugs. As you can see, rosnilimab shows comparable or even deeper CRP reductions in experienced patients than Rinvoq in both of their trials. Okay. So before we continue, I'd like to bring in Professor Emery and Professor Graf to comment on the data we've shared so far. So Professor Emery and Graf, could you please comment on our clinical data in the first 3-month induction phase and the comparisons to Rinvoq's clinical data?

Thank you very much, Paul. Well, there's a number of points I think we can pick up on here. And I think we could probably start with the issue of comparing head-to-head versus comparing to placebo when there's the uncertainty of not knowing whether you're getting active drug. And this is now very well researched and the data that you showed for Orencia when it is head-to-head versus against placebo indicates that the impact can be in excess of 15%, 20% and does explain and validates what you say. I think you're very brave going up and doing a CRP because JAK inhibitors directly inhibit IL-6, which acts on the liver. I'll pass over, I think, to Jonathan to allow to say a little more about that.

Yes. I mean I think we've known for a long time that drugs that block interleukin-6 JAK inhibitors certainly block interleukin -6 signaling have a disproportionate effect on CRP, sometimes out of proportion to their ability to make clinical improvements in people. So I think going head-to-head against just the CRP alone is kind of a very brave and bold thing to do. And I was frankly surprised to see such dramatic drops in the CRP this early in the trial for patients who received rosnilimab. And I think you're absolutely right, Paul, about the head-to-head data. I mean when you're -- you want to take placebo out of it, patients who are in head-to-head trials know they're getting 1 or 2 active treatments. And I think you are -- if you're going to compare rosnilimab's results to other drugs study in active comparator trials, you also remove any argument about whether placebo affects your results or not because everybody in the active comparator trial is taking active drug. So I think looking at both the naive population where I think rosnilimab performed exceedingly well and even in the biologic DMARD experienced population to see rosnilimab match it point counterpoint, I think, was again very, very surprising and very impressive to me.

And I think it's very valuable that you actually tested it in these 2 populations. One is the experienced population, which almost certainly is where it will be used initially where there's a big unmet need and a lot of patients don't respond and they're difficult to treat. They have longer duration disease. They have a ceiling on response. But the naive population gives some idea of what it's capable of, some idea of the ceiling of response. And it is, without doubt, the best that's been seen in that respect. JAK inhibitors are the only ones that really beat methotrexate and as first line, the only ones that have beaten TNF are methotrexate partial responders. You are partial responders and you're getting responses numerically higher than JAK inhibitors, which I think is quite remarkable, as is the speed of onset for a monoclonal that's acting mainly through a cellular effect and yet is producing responses, which are as quick as the market leaders.

Great. I think we could spend a lot more time talking about this. It's very interesting, but we should probably move on for the benefit of folks on the line. So let's move on to further objective biomarker support for rosnilimab. So specifically on the top left of this slide, you see the impact of the mid-dose on percent change from the baseline on T cells in the periphery. While there was no reduction in total T cells, it drove a rapid, deep and sustained targeted reduction of about 90% in PD-1 high T cells in red and about 50% in PD-1 positive T cells in orange. You can also see higher levels of Tregs in blue, particularly when off of therapy, which is a finding that is often associated with a decrease in flares. On the lower left-hand side of the slide, you can see the pooled impact of rosnilimab on Tph cells, which are key upstream drivers of inflammation in both RA and ulcerative. You again see a deep reduction regardless of baseline values, which is sustained through the entire treatment period. In contrast, no reduction in these cells were observed in the placebo arm. We see a similar impact with the high dose with a slightly less robust effect for the low dose on the right-hand part of the slide. So in summary, rosnilimab's rapid and deep reduction in PD-1 positive T cells supports durability of response and favorable T cell composition that is reflective of immune homeostasis. Now we're also very excited to show some of the most robust ever reported synovial biopsy immunohistochemistry results that provide dramatic visual depictions of the magnitude of PD-1 positive T cell reductions in the target tissue. Synovial biopsies of the most impacted joint were taken at baseline and at week 6 in a subset of patients, which was after only 2 or 3 doses of rosnilimab. When we look at the tissue biopsies of the high mid-dose rosnilimab patients, we see a deep reduction of 90% in PD-1 positive T cells, showing a dose response relative to the low dose. The baseline images on the left and the treated tissue on the right illustrate a striking and profound reduction in these pathogenic cells. In contrast, PD-1 positive T cell counts increased on placebo. This is exactly what we hope to see and confirms that rosnilimab is working as we would expect at the site of inflammation and disease pathology and consistent with the reductions observed in the blood. Gene expression analyses conducted on synovium demonstrated highly significant decreases in T cell activation and B-cell activation pathways. Which were expected based on our mechanism of action, which reduces T cell activation and Tph-mediated B-cell maturation. The impact of rosnilimab is visually dramatic and captures the difference between the baseline on the left and at week 6 on the right. This includes key mediators of inflammation and damage in RA, such as IL-2 receptor alpha, CD28, CD40 ligand and rank ligand. There is a 5.5-fold decrease in expression of genes associated with T cell activation and a sixfold decrease in B cell activation signatures in the tissue of rosnilimab CDAI responders at week 6. These differences were significant between rosnilimab responders, nonresponders and placebo, demonstrating that we not only confirmed expected pharmacology, but also that these changes have clinical relevance. Additionally, we observed reduction in pathways relevant to both RA and ulcerative colitis disease activity, including those related to TNF and IL-6 production. We see significant downregulation of target genes, which confirms rosnilimab's broad impact across T cell, B cell and myeloid biology as well as genes that mediate RA structural damage. In addition, we see significant down regulation of genes that would also be relevant to inflammation associated with ulcerative colitis, including NOD2, which has a strong genetic correlation with the disease; TREM1, which is known to be upregulated in TNF failure patients and IL-12 receptor beta, which is the shared subunit of the IL-12 and 23 receptors, which are validated clinical targets in DC. All right. So before we continue, let's take another pause and bring Professor Emery and Graff back again to comment on the significance of this translational data. So 2 points I wanted to ask about. One are to comment on the reductions on PD-1 positive T cells in both the periphery and synovial biopsies impacting both depth and durability of clinical outcomes. And then secondly, thoughts on impacting the reduction on the T and B cell pathways.

Well, I'll start if I may. The results are -- leave me no one specialist because they are so good. At 6 weeks, you're seeing this really striking change in the synovium. And as someone has told the synovium for many years, it's rare to see this sort of consistency. And then to find the gene expression in the peripheral blood is again a marker of response and correlates with response. I think this is a very profound validation of the principle of the mode of action and the speed with which it occurred at week 6 the sum because when you think of the turnover of lymphocytes, which are probably acting, it's a much longer process. But of course, you are actually killing the high cell high PD expressing cells, which is very important. With regard to the [ BD ] cells, I'll let Jonathan, but clearly, it's targeting the population that are pathogenic here.

Yes. I mean there are several things that sort of stand out for me as I sort of digesting all this. I think the first is trying to -- someone who's participated in the trial and see the durability of effect that this has on patients, even patients who stopped the drug at the end of the trial, looking at the ability of this drug to deplete what we are now increasingly believing are some of the instigators of RA in the first place and depleting them, really, if you think about it, you're not just blocking a cytokine, you're affecting a cell and cellular pathways. And that takes a lot longer to reconstitute again, if ever. And I think for a lot of understanding the biology, this just makes sense for why this drug appears to have the duration of effect that it actually has. Also, if you just sort of look at -- again, and we've become increasingly suspicious again, that these peripheral helper cells or at least these PD-1 high T cells that are in the periphery or in the lymphoid tissues are driving disease activity in a whole bunch of different ways. And you can see that because when you remove them from the periphery and the synovium, you have so many pleiotropic effects throughout. I mean, we know that they secrete cytokines that attract B cells and cause them to mature and act very, very badly in the synovium making autoantibodies and immune complexes. And those B cells and these are B cell activation markers appear to be completely aggregated here in the presence of this drug. And so looking at TNF production, IL-6 production, all these multiple pathways appear to intersect upstream at the level where this drug is acting. And I think the translational data really support that.

I'd just like to come back to a couple of little points. Tregs, we love because they predict a lot of things in advance of happening, in particular, the most powerful predictor of being able to take the drug therapy in patients on biologics. What you're seeing amazingly is an increase in Treg, which continues off therapy, which I think will be relevant to the discussion we have later about what happens to these patients and why they don't flare. The other thing is the gene expression strongly shows that structural damage genes are diminished, which means I'm sure this is going to have a profound effect on structural damage, which one would expect by the reduction of activity within the synovium.

All right. Well, thanks, both. I think we should move on again as time is moving on. So on this next slide here, moving past the translational data. Now we've broken out the rosnilimab patients who achieved CDAI LDA by week 14 into 2 groups, 137 patients that achieved LDA week 12 as shown in the green line and an additional 86 patients who achieved LDA week 14 and are represented in the dark blue line. Together, these are the 220 responders or the 69% of patients that were eligible to continue on the all-active treatment period. They represent the max number of potential responders at week 28 in our NRI analysis. While the denominator for the purposes of calculating percentages remains the most conservative 318 patients randomized to rosnilimab at the beginning of the study. It's exciting to note that not only do these responding patients start with high disease activity, but when you look at the data beyond week 14, these patients sustained stable CDAI scores through the end of treatment at week 28 and even off of drug. Now I want to direct your attention to the light blue line, which represents a subset of patients who did not achieve CDAI LDA by week 14 and were ineligible to enter the all-active treatment period. So they received their last dose at week 14. These patients achieved LDA when assessed at their first off-drug follow-up visit 1 month later at week 18. On average, this is not an unexpected observation as patients with higher disease activity at baseline may take longer to achieve low disease activity. But regardless, it is pretty clear, this group of patients in the light blue line started at a relatively higher disease activity and improved dramatically through week 14. In the real world, they likely would not have come off of drug as required in this trial. Instead, for our 6-month NRI analysis of the ITT population, they were imputed as nonresponders. Now let's look at the remaining patients captured here in the purple line who did not achieve CDAI LDA by week 18. Approximately 50% of these patients were at ACR20, which would be a more typical bar for staying on therapy post 3 months. The purple plots reflect the 25th percentile, median and 75th percentile values for this population. Minimally, at least 25% of patients in that box were doing quite well with a median CDAI 15 and trending towards low disease activity. For example, 5 patients in this group had already achieved ACR 50, but did not enter the all-active treatment period and were imputed as nonresponders in our 6-month NRI analysis of the ITT population. As these examples illustrate, preventing these patients from advancing in the all-active treatment period created a ceiling as of week 14 for the max response rate achievable in this study. Dan showed this slide earlier in experienced patients, showcasing rosnilimab's 6-month data using an NRI analysis as compared to Rinvoq and Orencia's all active head-to-head Phase III trial. To recap, rosnilimab's mid and high doses at week 28 for CDAI LA and ACR70 are comparable to Rinvoq and ACR50 is within range of the comparators. Now for illustration, we layer in incremental experienced patients on top of the NRI analysis from the light blue line on the prior slide that achieved these endpoints at week 18, which was 4 to 6 weeks after their last dose. It shows how across CDAI , ACR50 and ACR70, response rates would have continued to deepen and is a simple example to begin to show how the trial design may have capped the month 6 response rates. We could go even further by adding back the 7 of 8 patients who discontinued the all active treatment period while still in LDA as you would do in an LOCF or last observation carryforward analysis. But I think you get the point. Needless to say, we are really excited about these trial results and the positive impact that rosnilimab could have on patients living with RA. Now let's look at CDAI remission, which is a high bar requiring a score of less than 2.8 on a scale of 76. This is considered a very stringent endpoint, indicating near complete absence of clinical signs and symptoms. Again, this slide shows data for the experienced population in the stacked bar chart for both the NRI and the NRI plus week 18 analyses. The naive patient data are in the table on the bottom right, which are also very compelling. Rosnilimab CDAI remission rates are JAK-like and show how LDA response rates at 3 months deepened with improved responses by 6 months into remission. Additionally, in this responder analysis, rosnilimab demonstrated significant improvements by week 12 that further improved to week 28 across multiple validated patient-reported outcomes, including the HAQ-DI, a self-reporting tool to measure function and disability as well as the pain VAS. HAQ-DI scores of 1 to 2 indicate moderate to severe disability. As you can see on the left, rosnilimab patients entered the all-active treatment period with a score of about 1.6 and achieved a reduction of 0.9, which is greater than 4x the clinically meaningful difference. On the upper right-hand side of the slide, rosnilimab patients approved on the pain VAS from a mean baseline of approximately 65 down to 15, also a highly clinically meaningful change. Not only did patients report these very positive outcomes on symptoms and quality of life, you can also see on the lower right panel that CRP reductions that were sustained through 28 weeks objectively reinforce these improvements. Finally, we want to share preliminary data from patients off of drug for 8 to 10 weeks at week 34. This is an example of a completer analysis of the approximately 75% of patients in the trial who had reached this time point as of the cutoff date. We'll be able to update this in the future using an NRI analysis at the conclusion of the study. 83% maintained responses and were still in ADA at week 34 across all doses and consistent whether the patient was either naive or experienced. In patients with entrenched chronic disease, you would expect most patients to flare within a few months of withdrawing drug. For the other 17% of patients who weren't in LDA at week 34, the median CDAI was equal to 13, indicating these patients were still generally well controlled. We are very excited about these off-drug durability data as they highlight the uniqueness and differentiation of rosnilimab's novel mechanism and provide us with options on convenient extended dosing as we think about induction and maintenance dosing regimens. We also wanted to spend a minute on how rosnilimab performed relative to Lilly's peresolimab in their Phase IIa study. The CDAI LDA efficacy data are clearly differentiated, both in magnitude and durability at all time points. On the top left of the slide, you can see that on an apples-to-apples MRI analysis of the ITT population with similar week 14 continuation criteria, rosnilimab's LDA scores are higher at 3 months and stay higher at 6 months compared to peresolimab, shown in the purple lines on the top right, which markedly degrades through 6 months while on treatment. It is also impressive that rosnilimab at week 28 is higher than peresolimab's peak at week 14. Also as shown on the bottom right, Lilly only presented their PD-1 positive T cell data reductions for the first 12 weeks. On the bottom left slide, you clearly see rosnilimab's rapid and robust 90% reduction in target PD-1 high T cells that further deepens and is sustained through week 28 and is strongly differentiated compared to peresolimab's 57% in the purple line. It is notable that our preliminary data of PD-1 positive T cell reductions at 3 months off of drug continue to show a greater reduction versus what peresolimab saw at any reported time point. We have consistently stated for years that rosnilimab is a more potent depleter and agonist than peresolimab. Our data continue to recapitulate this and rosnilimab has one might say is a clearly differentiated benefit risk profile for peresolimab based on clinical data, translational data and our tolerability and safety data. Okay. Well, let's bring Professor Emery and Graf back again to comment on the 6-month efficacy response rates as well as the off-drug durability.

Thank you very much, Paul. Can we go back to Slide 19. I think it's worth going over a little bit what you said because people might think there's a bit of special pleading here. I'd like people to look at the 3 lines the dark blue, which is the middle line, the light blue, which is the group that were excluded from the study and the green line, which is the ones that reached the target as planned. These 3 lines are parallel. They're improving exactly the same rate. The only difference between them is where they started. Ones with the light blue line started with much more aggressive disease and we have the problem of having high disease activity requiring going into low disease activity state. If you want to do that, the best thing is to just be a small amount above that low disease activity state. If you have a very high disease activity such as this blue line, they've improved by an ACR70 and yet they don't get into the study. And it's a design fault. If you wanted to use this design, you should actually not include very active patients. You actually did include very active patients. and you suffered by having these patients who, as a clinician, there is no way in the world we would stop the drug when the line of improvement, if anything, is steeper than the other 2 lines. So I think you're entirely valid in including these patients. And I think it does cause pause for thought for the FDA to think is this the best way of describing the activity of a drug if you want to include very active patients. I don't know what you think, John.

I couldn't agree more. I mean, would you ever turn away a patient from a drug that's responding like the typical patient in that light blue line? I wouldn't getting that kind of response. I would just bring them back in a month and I presume they're going to continue to get better as they are. I'm also blown away a little bit by the remarkable reliability in the response and the duration of response throughout the duration of the trial. There have been a lot of hemming and hauling over. If you look both at the peresolimab and the rosnilimab trials at the change in responses between week 12 and week 14. I mean, just look at week 14 and carry it all the way out to week 34, that's internal validation of the actual response that was seen during that time even with a little blip in there. I'll remind you that after week 28, patients all knew they were not on drug. They came back to see us for 3 subsequent monthly visits, knowing very well they were off drug. They had every reason under the sun if they were having any symptoms to tell us and to get rescue therapy if they needed to or they may have joint pain or joint swelling or whatever. And what's remarkable is you see that they are maintaining a near -- they are in remission, near completely disease-free state throughout the duration of the active arm of the study and in the follow-up period off of active drug. I think that's just remarkable. And I'll just stress one more time. I think CDAI remission is an extremely high bar to achieve. If you think about how a CDAI is done, if there is -- if there aren't any tender or swollen joints, but a patient who initially felt that they were 90 out of 100 on the scale of pain and disease activity drops all the way down to 15. And if you, the investigator assess them to be at a level of 15 out of 100, right, that's 30, and you are now no longer in remission even without tender or swollen joints. It is a very high bar to reach and to see that consistently maintained throughout the study on and off drug, I think, just speaks to the power of this therapy.

I suppose, unusually, there's consistency between the translational science and the clinical response. You see reduction of PD-1 high T cells in the sum, which would require longer to repopulate and longer to change. have the high Tregs in the peripheral blood. And what we see is a persistent good response, which is consistent with both of those. So you have a therapy which is working on cells and allows it to continue working when you take the drug away, very attractive in terms of -- for a physician to use.

I'd say my patients in the study were persistently happy as well. I mean they were resuming their activities of life, taking care of their grandchildren going to Disney World. I mean -- and this was continuing after they were off study drug. That is, I think, a very, very good point.

Well, thanks, both. We should move on. And now let's move on to the important safety summary. So this is a slide we've shown before. And as a reminder, the typical RA patient when compared to the general population lives with a two to threefold increased risk of significant comorbidities, such as infections, MACE and malignancy. To compound things even further, these patients are typically on background immunosuppressive therapies such as methotrexate or steroids that can also drive higher infection rates. Additionally, the approved classes of therapies such as the TMS or the JAKs or CD20s all have black box warnings, indicating they can further exacerbate the risk of these comorbidities. So I think this is some of the real-world context in which we should reflect on when reviewing the notably unremarkable 6-month safety data for rosnilimab. So on the left, you see the baseline to week 12 placebo-controlled safety data provided as patient incidents with adverse events, and we shared this back in February. On the right side of the page, we provide baseline week 28 safety data in the form of exposure-adjusted incidence rates per 100 patient years. This allows for a more reasonable apples-to-apples comparison to placebo given that the duration of time patients were on placebo was only 3 months, whereas rosnilimab patients were exposed for 6 months. The bottom line is rosnilimab continues to be well tolerated with no safety signals or trends. We see no treatment-related SAEs, no malignancies, no anaphylaxis or systemic hypersensitivity and a low incidence of injection site reactions. Most adverse events were mild to moderate in severity. They look similar to placebo across arms and the rosnilimab dose groups importantly also did not show any dose response, even though our highest dose was indeed 12-fold higher than our lowest dose. The overall SAE rate was very low. Since our last update, there were only 3 additional SAEs reported between the third and 6 months. In the low dose, there was an isolated event of embolic ischemic stroke in a 6-year-old patient from Poland who had an identified embolic stroke from a 60% stenotic right common carotid artery. Importantly, none of the SAEs observed in rosnilimab patients in the entire study were deemed to be treatment related. All of the SAEs plus the severe Grade 3 events were singular and unrelated, and therefore, no safety trends or signals have been observed with rosnilimab. For a chronic severe RA population that is also on background conventional DMARDs such as methotrexate, it's not unexpected, particularly given a study of this large size scope and duration to observe single one-off events or SAEs. Of note, rosnilimab was highly tolerable. Less than 2% of patients in the entire study discontinued due to an adverse event. This positive safety and tolerability profile also particularly compares favorably to the standard of care. All right. Let's bring back Professor Emery and Graf again to comment on safety. So could you please comment on the profile of an RA patient and what you'd expect to see with other agents on tolerability and safety in trials for the real world over 6 months? And then also maybe let's put rosnilimab's emerging safety and tolerability profile into perspective, including the SAEs in our trial.

Just to start, the first thing you look at is the completion rate of the study. If you've got a large number of dropouts from the study, then clearly, in a lifelong disease, this is going to be a problem. You expect to see quite a large number of dropouts in a study of experienced patients, of which a large proportion of this population were. So to have so few dropouts is really very impressive. The second thing in this population, again, infection is a real problem. and yet the actual numerical infection rate was lower than placebo and no opportunistic infections, 2 big factors in choosing therapy. Very well-tolerated drug. And obviously, we will be comment on 1 or 2 things. Johnson, is there anything you want to add?

Yes. I mean if I had to compare it to in terms of safety to what -- I mean, we rheumatologists know that we're treating our patients with very complicated biological and anti-metabolic therapies. But if I had to look to compare it, it looks almost like [indiscernible] like safety to it. There's just very, very little that sort of jumped out at us, particularly the lack of serious or opportunistic infections, as you said, Paul. And this is a lifelong disease. Patients typically get this early in their adulthood and they -- and it lasts throughout the entirety of their life and being able to stay on a given therapy that maintains its efficacy and tolerability is very, very important because many look at this market as a market with many therapies on it. I think if you are living with the disease for 50 or 60 or 70 years, you better hope that there are other drugs that you can take that in case your drug stops working. So I think the tolerability data here are extremely encouraging.

I think just to tie a bow around this, I'm sure some people may be wondering about the one case of stroke and perhaps you could comment on that.

It could at the lowest dose of the drug. We would -- in 400 patients studied for 6 months, it wouldn't be unusual in this population of very long-standing exposure to inflammation, which patients to have an event. And it's always -- it's actually impossible to comment on the significance of the signal that except was there a reason -- a preexisting reason? Yes, they had carotid disease on the correct side. They have 60% stenosis. It was an embolic stroke and in fact, with good recovery as it turned out. But I think you cannot -- there is an explanation for it. And you can't really say any further than that.

Yes, I would agree. And also the investigator in the trial didn't feel it was related as well on the front lines. I think having done a lot of these trials, these things tend to happen. And a one-off is you really can't make a judgment one way or the other. It's impossible, alone in this population of patients that are extremely high cardiovascular risk.

You just need to compare this to other comparable populations. That's right. And the safety here is exceptional. That's right.

Great. And now to cover on next steps. So beyond ramping up this RA trial, we are focused on completing the rosnilimab Phase II trial in ulcerative colitis. While that study is still enrolling, it is well on track to report initial data in the fourth quarter of this year, where all patients will be through at least the week 12 primary endpoint. Importantly, to date in the UC study, based on our surveillance data, we have had no serious safety events reported of concern, including no malignancies, no MACE and no serious infections. This further supports rosnilimab's overall favorable safety profile. We are assessing 2 alternative strategic paths forward for rosnilimab. One is securing a global partnership to help advance rosnilimab across all indications, including dual Phase III programs for both RA and UC. Or two, assuming the UC Phase II trial is both positive and hit the PPP, we have the option to independently advance rosnilimab into Phase III specifically for UC. In parallel, we'll enable activities that will be executed on in 2026, including Phase III drug supply scale-up and the end of Phase II meetings with regulators. We believe rosnilimab's mechanism has the applicability to other inflammatory disease. So also for 2026, based on our clinical and translational data in RA and UC, we are assessing additional indications to advance into Phase II trials. Beyond rosnilimab, we have 2 additional clinical stage programs that we're excited about, AMB033, CD122 antagonist and AMB-101, BDCA2 modulator. We'll host an R&D IR event later in the second half of 2025 to walk through AMB033 in more detail. Our cash runway is through year-end 2027. Included in this runway is the Phase III enablement activities and costs. However, most capital commitments to facilitate Phase III development would not come until 2026 after seeing the results from the UC trial. Of note, our cash runway excludes any potential inbound dollars from BD or the substantial future royalties and milestone payments due [indiscernible] from our immuno-oncology financial collaboration with GSK. In closing, I want to thank Professors Emery and Graf for joining us and to thank all of the patients and clinicians who supported our trial. Also thank you to our team at Anaptys, who worked tirelessly to achieve our vision to transform patient health by delivering innovative immunology therapeutics. For our open Q&A portion of the call, besides Paul and myself, Professors Emery and Graf will participate alongside of us. Now we will turn it over to the operator for Q&A.

Our first question will come from the line of Yatin Suneja from Guggenheim.

And putting all the data in context. So 2 quick ones for me. First one is around the dosing. Could you just talk about the dosing between the 600 and the 400 and the different regimen and how you are thinking about a Phase III program? And then related to the dosing is, if we look at the off-treatment effect, they continue to persist. So could you talk about introducing a drug holiday? Is there a possibility that patient could be off treatment for a certain time period? So that sort of is on the dosing front. And then just one question for the physicians. If you can talk about the perceived safety for this mechanism, given that this mechanism is used in cancer on the other side. So just love to hear your thoughts on that.

Yes. Thanks, Yatin. Maybe I'll let the visiting professors comment on your second question first, and then I can come back to the dose.

Well, I think PD-1, of course, antagonists are very well known. This, of course, is quite different. This is a homeostatic mechanism, which is actually normalizing PD-1 expression rather than actually making it abnormal. And of course, there have been similar molecules CTA-4-Ig works in a similar way of normalizing interaction between activated cells. And I think the concern, it's not going to stop cancer, that's certain. And rheumatoid occurs more frequently in smokers. So there will be cancers, but there's no need to attribute this mechanism of action as the cause of that.

Yes. And I think sort of tying all your questions together, which I think we're very, very, very smartly put together. This is an exciser of the cells that are driving the disease. It doesn't necessarily mean that as it's given or as it will eventually be dosed, that patients will be continuously have their PD-1 receptors agonized, PD-1 cells agonized. And so you can imagine that the cellular effects normalizing the immune response in patients and the way that the drug is dosed and its current safety profile, even with continuous dosing, you put all that together. And yes, I think that the safety profile, I have a growing confidence. You'll never know until you'll know when this is entered into the marketplace, hopefully. But from what I see now and from the way this mechanism may potentially work, I don't think it's a simple -- well, one drug fights cancer, therefore, the opposite mechanism will cause cancer. I don't think that's the case that we're seeing here.

All right. And just getting back to the first question about dose. So I mean, when you look at the data in aggregate, whether it's at 3 months, 6 months or off drug as well as the translational data also, there's certainly a dose response between the low dose that we're picking up on in the mid and the high dose. And the translational data, both in the blood and in the synovium support the superiority of the mid and high dose relative to the low dose. I think it's also important not just to comment on efficacy, but it's also important to note that we don't see a dose response on safety as well, really important. And I think all of this together, this gives us confidence that at least some monthly dosing regimen could be used as we go on into Phase III. And obviously, with the off-drug period as well, that provides an opportunity to think about even more extended dosing regimens as...

The off-dose is very attractive because frequently, patients have to stop therapy for independent reasons, operations, travel, all sorts of things to not have patients just about to flare before the next dose, which is what happens with many cytokine inhibitors is a real advantage. It gives stability to the patient. It gives much more flexibility to the patient.

Next question will come from the line of Joseph Thome from TD Cowen.

Definitely a lot of data here. Maybe one for the physicians. When clinicians are deciding what therapies to use in practice, I guess, what scales and time points, I guess, are most important when deciding what drug to use next? Obviously, a lot of data here, and it looks like rosnilimab is obviously performing well across all these measures. But maybe what's most important? And then maybe for the company, the finding that the high disease activity patients maybe did eventually get to low disease activity over time. When you look to a Phase III, is there a way to kind of best capitulate that in a Phase III trial design? I'd assume kind of limiting high disease activity patients wouldn't be ideal. But how do you kind of best capture the benefit that you're seeing with this drug in Phase III?

Regarding your first point, we're looking -- if we're taking patients who have experienced biologics, there's a reasonable amount of data, some of it not terrific, but what is there, that if you switch mode of action, you're more likely to get a response. This is a different mode of action to anything we've got and therefore, would be very attractive to the increasing number of patients we find which fall into the category of difficult to treat. And that is what people are finding is almost the biggest population now in the clinic as they've worked through all the advanced therapies and have reached a stage where they become resistant to them. This has the advantage. If you notice, it actually kills the activated cells. So abnormal signaling in those cells would not be relevant. and it has a sustained response, which, again, and most importantly, is tolerated. These patients are very intolerant because they have abnormal liver. They've been exposed to inflammation for up to 25 years. And to have something that they tolerate and works in a different way would be very attractive to a physician.

Yes. I'll take the -- that's the biological experience difficult-to-treat population, which obviously is the obvious first target market for a product like this. But I'll take -- let's look at the naive patients and look at how surprisingly well they responded. It will take, I think, convincing of practicing physicians like ourselves. But I think if you were to offer physicians 1 of 2 modes, one is to put someone on a medicine that temporizes symptoms versus we put someone on a medicine that could potentially relieve the symptoms, right? If you have a flat tire, right, you can pump up your tire enough to get you to drive to the next gas station and pump it up again and get to the next gas station and pump it up again to get to where you need to go, right? But if you can change the tire in the first place, you solve the problem. And there's a very real possibility that we can -- that it may be possible to show that certain mechanisms of action like this have the potential to really reset the disease itself. And from that standpoint, seeing somebody with brand spanking new RA in your clinic, if you have the chance to offer a patient a drug that basically will remit their disease and potentially reset their immune system in a safe and effective way, that may be a very attractive mechanism for practicing rheumatologists as well. That's another hurdle to climb. I think the obvious first targets are the patients who are failing current therapies. But if you think about the potential of a mechanism like this, let your imagination run wild, you can begin to see where this could potentially go.

Joe, it's Dan. I'll pick the other question here. I think I really appreciate you asking because it tees up the exact point we're trying to make is even in this Phase II trial, regardless of any patient that entered the all active treatment period, month 3, month 6, so many of these patients were seeing a very dramatic response. We exactly want to treat patients whether they're various levels of high disease activity. It was the handicap of this trial design pushed on us by the regulators that didn't allow those patients with very, very high disease activity not to move forward and ultimately impacts the response rates on NRI analysis. So the question you're asking is the point we're making is that in a Phase III study, you wouldn't have that constraint at 3 months. all these patients to Dr. Emery and Graf's point, if you're seeing that type of response in these patients had dramatic improvement, they would not only not be taken off therapy, they wouldn't let you take them off of therapy. They were responders. And so that's the point. And we're breaking this out just to show the impact regardless of where you start, ultimately, you're getting to dramatic response rates, whether that's low disease activity or remissions.

Our next question will come from the line of Yasmeen Rahimi from Piper Sandler.

Congrats to the amazing for the great data and for sharing it as great details. Two questions. One is for KOLs. Dr. Graf, wonderful to hear your voice again. I guess if you were up to the task to design a Phase III study with this product profile, what are some of the key considerations that you would want to implement into the study? I know the team from EA provided their color, but if you could comment on what you would recommend to them. Also, if you could remind us what the treatment effects look like when we go from a Phase III to a Phase II, that could also be helpful when we think about RA studies. And then the question for the management team is, obviously, you guys have a great partner, Sanofi. As you're thinking about strategic, I don't know if you want to remind us have you had a chance to share those data with them to the extent you can comment on it, that would be helpful. And I'll jump back into the queue.

It's good to hear your voice, Yasmeen again. I you asked a great question. If I were to design a Phase III trial for you, well, the good news is when you come to meet with the FDA, the FDA has a known quantity now. They know that this drug met its required efficacy levels in the Phase II, right? So the -- you can make the argument that you could allow your placebos potentially or anybody theoretically who's not -- your placebos who are not responding at whatever time point, week 12 or wherever you want to make your first assessment to cross over into active drug in a blinded way, right? So adding a crossover element, I think, would be very, very effective and take care of the placebo issues that you see with the sort of artificially mandated structure of the Phase II trial. Also, I think to just sort of remember, I mean, to get to Paul's point, I think I would love to see active disease patients continue to be enrolled in the Phase III study, just like they were in the Phase II study. That's where you see some of the most dramatic responses as well. And I think, again, if you allow one of the primary endpoints to be a 6-month follow-up instead of a 3-month follow-up and allow that to be the case because you have crossovers allowed, you'll get a full sampling of what the -- assuming that 6 months is even -- there isn't further improvement beyond 6 months. I mean, we don't know, but it seems to be getting pretty close to where you're down to no disease activity. If you can get out to 6 months, you'll get a full understanding of what the magnitude of effect of this drug might actually be.

Well, thanks, Yas, for the question. I can't answer directly as to where we are. But what I'll say is that we have a novel MOA in immunology. And to me, that's somewhat unique right now when we look at it in the landscape of Phase IIb, Phase III ready or drugs in Phase III. We're doing something different here. And I'll applaud Sanofi, who as an investor in us, did provide capital last summer in the equity raise that we did. But they're an investor versus a strategic partner with us today. But what I'll say is a year ago from where we were relative to the results at this point, where were we last year, we were running a 424 robust Phase III trial. Fast forward to today, it's now shown a best-in-disease profile, which is the aggregate of efficacy, right? That's JAK-like on a stringent basis, forget what we were talking about earlier of patients that responded and just weren't counted there. But the safety and tolerability as well as the once-monthly dosing that we come, the whole profile here is differentiated. And it's exciting what we're trying to do here that's different in these diseases where there's been nothing new for a long time. So I think Sanofi is one of many companies that are focusing in allergy in the autoimmune area that potentially are looking for something that's different as opposed to me better drugs or tweaks or combination of things that have been around for a long time, where you get an approach like we have with a PD-1 depleter where you're targeted at the top of the immune cascade, having a broad impact that's safely driving very high rates of remission for so many patients. So I think we have a choice over time as to where we go with the drug. But for now, we're excited that Sanofi is an investor in the company like many other folks on the call.

Our next question will come from the line of Alex Thompson from Stifel.

Congrats on the data. I guess one on safety. On the liver enzyme elevations that you highlighted in the safety table, I wonder if you could talk about the time course of the course of those elevations, what was observed and when and your confidence in the broader safety here as well. And then on UC, based on the biomarker data that you generated in this trial, how confident are you in the doses that you have in UC? I know you haven't quite disclosed what those are relative to RA. So any comments there would be helpful.

It's Alex. The liver function test, if you know, nobody stopped the drug because of them. If you study rheumatoid patients on methotrexate with long-standing disease, you find that there's frequently mild liver function test abnormality. And as you probably know, there are studies done where they've intervened or not intervened and stopping the drug makes no difference. And in fact, the drug wasn't stopped because it wasn't severe enough. There was only one slightly higher grade liver function test abnormality, and that resolved spontaneously with continuation of drug. So I would say, as a clinician, there is no liver signal here whatsoever. It was given for 6 months continuously without a single patient stopping. That for me is a very clear positive safety aspect.

Yes. And I'll just add on that, that each one of those that are listed in the table were a single spurious event at one time point that normalized immediately afterwards. So these are just variations that happen over time.

You also asked around the dosing in the UC study. So we have 2 active doses from baseline. One is once monthly and the other is twice monthly. And just for some context on that, the high dose in the UC study is higher than the high dose, which was 600 QQW in the RA study. The second point there is after the first 6-month period, we do have a treatment extension period where patients who have responded can continue on through 1 year. So not only do we have the benefit there seeing sustained activity over the long term as well as safety data, we're also taking the opportunity in that treatment extension period to look at a Q8-week dose from month 6 to month 12. So we're exploring 2 different doses as an induction and a longer-term extended dosing in that 6- to 12-month period afterwards.

Our next question will come from the line of Anupam Rama from JPMorgan.

Congrats on the update. Maybe just a quick one for the company. Have you thought about which dose you're moving forward in the pivotal? Sorry if I missed this. And then for the KOLs, we've heard a couple of times now like, hey, have we seen the full impact on the efficacy side of rosnilimab? And if we followed it longer, could there be more, right? So just remind us in registrational studies in this space, like how long is the randomized portion? How long are the OEs? And is there optionality in the OEs, not just in terms of learning the durability of effect here, but also would you recommend trying different treatment intervals too in the OLE portion?

Okay. So a couple of questions there on following. But I think one was on the dose. And obviously, data looks really good. We are seeing somewhat of a differentiation between the low dose, the 100-milligram monthly and the 400 monthly and the 600 dose. So it's probably a bit premature to make a final determination at this point in time for Phase III. But I think what we can say is it looks like we have a monthly dose, and that's very reassuring.

Doctor, you want to speak about the mass response and we potentially see moving forward?

We made a big point about the fact that even with this study, we would have seen a much higher percentage response level of ACL-5070 as we've been if normal rules had applied. In terms of the depth of response, we're seeing a very sustained response. We haven't looked in this current data very closely, but that's clearly a very interesting concept. And if you gave it long enough, would you just wipe out all activated cells and the drive. And as Jonathan has referred to really, the sort of initiation of disease. I don't think we can say much more at the moment. But with future studies, I would certainly hope that would be one aspect that will be explored because clearly, what we'd love to do is to give this a set period of time and then have a period perhaps in remission or perhaps with some mode of action that maintains the response. many possibilities.

And just one more point there. Noting that we had that at 3 months to enter the third to 6 months, if you just look at things like where were we on CDAI remissions in the first 3 months through the next 3 months, you see a clear deepening of those CDAI LDA responders moving to remission. And then another way to look at that on a relative basis you look at the ACR50, you can see a clear deepening for ACR70 responses over time between week 12 and through the 6-month endpoint.

Yes, that's what I was going to say. you give it more time, in particular, the patients that were the slightly slower responders upfront or if you look at the translational data, right, you had a very nice sort of dose response between -- or at least gene response between the nonresponse -- those who didn't reach low disease activity, those who were on placebo and those who didn't reach low disease activity. I would love to see a trial that gives more time for those in the intermediate range to achieve the full effect of the drug.

Our next question will come from the line of David Risinger from Leerink Partners.

So thank you, Dan and to the KOLs for the presentation and discussion. It was also interesting to see efficacy comparison to Lilly' peresolimab. And actually, my RA data questions have been answered. So maybe, Dan, could you just look forward to the forthcoming UC results, just provide a framework for that data set that will read out. Any key points you want to make, including considerations relative to J&J's PD1?

Thanks for the question, David. So in UC, we have in Q4, we'll have all patients through 12 weeks, and we'll have a partial responder set through the 6-month period, similar to where we were in February for RA. I think that really tees up what's important here in UC as you think about a target product profile. So the market dynamic is pretty different than in RA. In RA, we have a mature market. There's biogenerics that are frontline therapy, and we needed to deliver data that was clearly bio better approaching JAK like. Well, in UC, it's a different dynamic. It's a growth phase. There's the initiation of class cycling. You have branded assets that are still penetrating in the front line, where those -- generally speaking, biologics. And what we're still searching for there is ensuring that the -- the 1/3 to 1/2 of patients that lose response at 1 year can we improve on that, right? And so this is where now a similar has come back with RA is we're looking with our primary endpoint, which is the change in MMS. We're looking to see symptomatic improvement and distinction from placebo at 3 months. But what's really going to matter here and that's to be differentiating, we hope to see from the other biologics is not just a Skyrizi-like endoscopic remission rate. But then when we look out past a year, do we continue to see durable responses like exactly what we're seeing in RA. That is going to be significantly differentiated in the UC market compared to not just the IL-23p19 but also the TL1As, which once you got to a year, look no different than anything else. So we're really excited about the potential here. And I think the RA data potential read through to that profile in the mid- to longer term, what we hope to see in [indiscernible].

Our next question will come from the line of Derek Archila from Wells Fargo.

So just one from us. Just on Slide 14, I guess I wanted to understand the increase in PD-1 high T cells between week 6 and 12 for the every 4-week dosing regimens. I guess how do you think this plays a role as you kind of pursue potentially 8-week dosing?

Yes. So you're looking at, I'd say, interesting quirk as it relates to trial design and timing of dose right out of the gate, right? And so until you get the stable activity over the longer term, what you're seeing in the difference between the monthly doses and the every other week dose in the earlier phases is some slight up and down, but still within a very tight range. That's what attributed that is timing of dose. When you get out over time, I think very importantly, this is exactly why we're showing that off-drug period. The data looks stable and actually start to differentiate between the 600 and 400-milligram dose versus the lowest 100-milligram monthly dose.

Got it. Understood. And then maybe just a follow-up to David's prior question on the J&J PD-1. And I guess maybe how you differ from that molecule. It seems like they had some trouble with their UC cohort. So just kind of curious your comments on that.

Yes. So we'll end up seeing them present at UR in 1.5 weeks. We're looking forward to more disclosure there. But a few things. One, we've said on the record that we understand J&J has had issues scaling the drug on the manufacturing side. They only enrolled 5 patients as they disclosed, 3 of them dropped out pretty early. In the totality of all of J&J's data, these were less than 3-month time courses of administration. So they have effectively no data in our eyes from a clinical perspective for us to read through at all. As I mentioned earlier in our prepared remarks, not only is our study targeting 132 patients to enroll in UC, we are getting pretty close nearing the end of enrollment, and we look forward to reporting the data in Q4. But I wouldn't look at the J&J less than a handful of patient data in UC and be able to tell you one way or the other at this point. But let's see what they report in 1.5 weeks are.

Our next question will come from the line of Andy Chen from Wolfe Research.

Back to the topic of durable response with these patients 2 months or 3 months off treatment. I know Dr. Emery or Dr. Graf, you mentioned that the CDAI remission bar is very high and it's hard to achieve. But I'm just wondering like what is the real-world application of this? Is this just a thought experiment that it's nice, there's off-treatment sustained response? Or is it does actually influence therapy selection here? And also, if you can remind us, are there other drugs in rheumatology that have off-treatment durability and it's really translated to real-world commercial use?

I can take that. There's very few drugs that especially in a population like this, you can stop in which there is a fairly immediate relapse. We've done tapering of every type of rheumatoid disease and every type of drug. time -- the only time you take the successfully really is if you get patients right at the beginning of disease. And these are at the beginning of disease. So what it's suggesting is it's having a more profound effect. And you'll say, is this just thought experiment? No, it's not because patients love having stable disease. We -- I can tell you about a study where we ran in patients who were relatively early, watch them for a year in stable remission. And as soon as we have their drug, 50% of them relapsed. Rheumatoid patients are rarely in stable remission. If they are, they're extremely lucky. If you're in late disease and you've got stable remission, you really are very thankful and you will not change your drug rapidly.

I was going to cite your study actually. to the price. I completely agree. Patients want to live a normal life. And there are different flavors of RA patients. Some patients have chronically active disease, some flare in and out, have bad days and good days and things like that. Smoothing out the edges and having a consistently extremely low level of disease activity throughout their lifespan, I think, is the holy grail for patients who live with this disease for decades of their life. So it's a good question, and I think it does matter clinically in a big way.

And with that, this concludes the question-and-answer session. Thank you for your participation in today's conference. This does conclude the program. You may now disconnect. Everyone, have a great day.