Connect Biopharma Holdings Limited (CNTB) Earnings Call Transcript & Summary

Ladies and gentlemen, thank you for standing by. Welcome to the Connect Biopharma CBP-201 Phase IIb Atopic Dermatitis Data update announcement call. [Operator Instructions] Please be advised that today's conference is being recorded. I would now like to hand the conference over to your host today, David Carey of FINN Partners. Please go ahead.

Thank you, operator, and welcome to Connect Biopharma's CBP-201 Phase IIb Atopic Dermatitis Data announcement conference call. Joining me today are Dr. Zheng Wei, Co-Founder and CEO; Selwyn Ho, Chief Business Officer; and Dr. Jonathan Silverberg, Associate Professor at George Washington University School of Medicine and Health Sciences. Today's call is being webcast and will be posted on the company's website for playback. The slides reviewed during today's call will also be posted to the company's website after the call is concluded. During today's call, management will provide additional data from WW-001, the Phase IIb clinical trial to evaluate the safety and efficacy of CBP-201 for the treatment of moderate-to-severe atopic dermatitis or AD in adult patients. Following the prepared remarks, we will open the call to Q&A. Before we begin, let me briefly review our forward-looking statements. During today's call, we will make various forward-looking statements. Investors are cautioned that our forward-looking statements are based on current expectations and are subject to risks and uncertainties that could cause actual results or outcomes to differ materially from those indicated by our forward-looking statements. Please read the safe harbor statement contained in our press release, which was issued earlier today, as well as risk factors contained in Connect Biopharma's registration statement on Form F1 for a more complete discussion of these risks and uncertainties. Now I'd like to turn the call over to Wei.

Thank you, David, and thanks to everyone joining us on today's call. We are very excited to report the positive dataset from WW-001, the Phase IIb trial of CBP-201 in adult patients with moderate-to-severe AD. As many of you know, in November, we reported positive topline results from this trial, indicating that all 3 CBP-201 arms, which was 300 milligrams every 2 weeks or Q2W, 150 milligram Q2W or 300 milligram every 4 weeks or Q4W, met the primary endpoint of eczema area and severity index, or EASI, percent reduction from baseline at week 16 and was statistically superior to placebo. Multiple key secondary endpoints were also met with CBP-201. These initial results clearly demonstrates that CBP-201 provided significant benefit to patients in the study. Today, we are reporting the positive data set from the results of the primary analysis of this endpoint as well as results from several sub-analysis that were undertaken to provide insight into how CBP-201 may be positioned in the competitive landscape of other treatments in this indication, including the approved IL-4Ra antibodies. As Jonathan will discuss in a moment, and has not been in a press release we issued this morning, the patient population recruited in this study had a marketed lower baseline AD disease severity. And the trial had higher patient discontinuation rate relative to previous IL-4Ra Phase III study. We believe this is still largely to the impact of the COVID-19 pandemic, during which our trial was conducted. The sub-analysis that we will review today will undertaken to address the difference in patient disease characteristic between the population in this study and the populations in previous IL-4Ra antibody Phase III trials to determine the effect of these factors in the magnitude of treatment effect we called for CBP-201. There have been no head-to-head trial between CBP-201 and other antibody therapies and the ability to draw conclusion from non head-to-head trials is subject to inherent limitations. Despite these caveats, we are extremely encouraged by the data we have seen from the Phase IIb trial, and the additional data we are reporting today further increase our confidence that CBP-201 has the potential for a highly competitive safety and efficacy profile. Importantly, results from the Q4W dosing group support the potential for CBP-201 to have a more convenient and differentiated dosing profile than other AD biologic therapies. Combined with the potential for an improved safety profile compared to other AD therapies, we believe that CBP-201 has the potential to be an important treatment for patients with moderate-to-severe AD. Planning for a global Phase III clinical trial program evaluating CBP-201 in this indication is now well underway, and we expect to enroll the first patient in the second half of this year. Before [indiscernible] reduce the data, let me remind you that despite the availability of approved therapies for moderate-to-severe AD, many patients living with this condition continue to have unmet medical need. AD which has an estimated lifetime prevalence of up to 20% and is increasing globally. It's most commonly diagnosed chronic inflammatory skin disorder. It's estimated that more than 26 million people in the United States alone have AD, of which 6.6 million have moderate-to-severe disease. It's estimated that over 58% of adult with moderate-to-severe AD have disease with physicians considered to be inadequately controlled by approved therapeutic modalities, including topical anti-inflammatory agents and systemic agents. Dr. Silverberg can provide additional insight into its real-world experience, with currently approved therapies during the Q&A process of the call. Based on the data we are reporting today, we believe CBP-201 has substantial potential to address the needs of many AD patients who have inadequate responses to the current standard of care. With that, I will turn the call over to Selwyn, who will review the Phase IIb trial date asset and analysis. Selwyn?

Thank you, Wei. Today, I will be presenting the topline results of the CBP-201-WW001 trial, or WW001 for short. This was a global Phase IIb trial evaluating CBP-201. Our novel human monoclonal antibody targeting into the IL-4 receptor alpha in adult patients with moderate-to-severe atopic dermatitis. The data in this presentation is based on trial information that has been accepted and will be presented in 2 separate abstracts and posters at the Maui Derm 2022 conference, which commences on January 24 this year in Hawaii and that will be made available online to view at the start of the meeting. We are grateful to the meeting organizers for allowing us to share this data in advance. Before I would start, I would like to point out that I'll be making forward-looking statements and point to our disclosure here. As a reminder, in November 2021, we disclosed the headline data from the primary analysis of the WW001 trial, which showed that CBP-201 met both the primary endpoint and multiple key secondary endpoints, with significant improvements on skin clearance, disease severity and itch as well as favorable safety data with treatment-emergent adverse events, or TEAEs, similar across CBP-201 doses, and with low rates of conjunctivitis and injection site reactions. As such, from these primary analysis, we believe that 300 milligrams, either 2 weekly or 4 weekly appear comparable to the current marketed interleukin-4 receptor alpha biologic. However, making direct cross-trial comparisons with dupilumab based on the primary analysis within the WW001 trial versus the dupilumab Phase III registration trials, known as SOLO 1 and SOLO 2 are difficult due to 2 key differences being apparent between the trials, notably at WW001 recruited a less severe patient population in SOLO, and also the WW001 had higher dropout rates and discontinuations in SOLO 1 and 2. To enable a more like-for-like comparison, additional A priority and post-hoc analysis of the WW001 trial populations were performed. And these showed that as the baseline severity for the disease increase, CBP-201 efficacy response further improved beyond that seen in the primary analysis. Also, with the baseline severity more closely matching SOLO 1 and 2, side-by-side comparisons of CBP-201 300 milligrams 2 weekly and 4 weekly appeared at least comparable, with some endpoints numerically better than dupilumab 300 milligrams 2 weekly. Finally, we believe that CBP-201 300 milligrams has the potential for a differentiated efficacy and safety profile with the convenience of 4 weekly dosing. As a consequence of these positive data, Connect plans to advance the global development of CBP-201 into Phase III in atopic dermatitis, with an estimated first patient enrollment in the second half of 2022, subject to favorable discussions with the regulatory authorities in end of Phase II meetings. I will now present the trial design, patient disposition, baseline characteristics and primary analysis followed by the additional analysis performed. In terms of trial design, WW001 Randomized and Double-Blind, Placebo-Controlled Multicenter Trial of the Efficacy, Safety, Pharmacokinetics and Pharmacodynamics of CBP-201 in adult subjects with moderate-to-severe atopic dermatitis. This was a 4-arm trial with 3 active CBP-201 groups, 300 milligrams every 2 weeks, 300 milligrams every 4 weeks and a 150 milligrams every 2 weeks admin subcutaneously following a loading dose of 600 milligrams on day 1, with a matching [indiscernible]. Randomization is in a 1:1:1:1 manner. And following a screening period to 16 weeks, followed by an 8-week safety follow-up. The key inclusion criteria for EASI, IGA and BSA at baseline were identical to existing Phase III trials with [indiscernible]. There were some differences such as duration of atopic dermatitis or greater than or equal to 1 year in our trial compared to 3 years or more than that in the SOLO 1 and 2 trials. WW001 the monotherapy trial, therapies prohibited other than over-the-counter [ ammonias ]. And if topicals were used at the investigator's discretion, this was considered as rescue medication, leading to [ complication ] as a nonresponder for outcomes from the point of [ rescue ] therapy initiation. The primary endpoints was the percent change in EC from baseline to week 16 for each active group separately versus placebo. Key secondary endpoints include proportion of IGA 0,1 responders, EASI-50, EASI-75 and EASI-90 responders at week 16 and the change in Peak Pruritus-Numerical Rating Scale score from baseline to week 16. The trial was conducted globally with 59 centers from the United States, China, Australia and New Zealand, with the majority of centers in the U.S. Further details can be found on clinicaltrials.gov with this NTT reference. The patient disposition was as follows: 226 patients were randomly assigned to the 4 groups, with 56 or 57 patients per group, all of whom received at least 1 dose of treatment. WW001 was conducted during the ongoing COVID-19 pandemic, with movement restrictions likely impacting scheduled patient attendance at clinics, unlike the SOLO 1 and 2 trials. As such, there was a high discontinuation rate in the CBP-201 active arms, with up to 19% of the patients discontinuing, and that was approximately 3x higher than the 6.3% discontinuation rate in the [ dupilumab milligrams ] every 2 weeks in SOLO 1 and 2. Of note, the majority of discontinuations were due to withdrawal of consent and patients [ forced ] to follow up with very few patients discontinuing due to adverse events and no patients discontinuing due to COVID-19 infection. Moving on to the patient baseline characteristics in the trial. The groups were generally well balanced. However, unlike the SOLO 1 or 2 trials, WW001 recruited a much less severe trial population in SOLO 1 and 2, with a shortage duration of disease at 16 years, a lower baseline EASI with a median of 21.2, a lower proportion of patients with an IGA score of 4 at baseline of 31%, a lower baseline Peak Pruritus-Numerical Rating Scale Score of 6.9%, and the lower mean baseline BSA involvement of 39.5%. All of these were lower than seen in the SOLO 1 and 2 trials. A summary of the key differences between WW001 and SOLO 1 and 2 patient populations are provided later in the presentation. Turning to the results. The focus for this presentation is for the efficacy results at the week 16 time point, starting with the primary endpoint of EASI percent change from baseline, or CFB at week 16. Here, you can see that all doses of CBP-201 met the primary endpoint, with a statistically significant reduction in percent EASI from baseline to week 16. With both the 300-milligram arms delivering numerically greater improvements than the 150 milligram every 2 weeks of 63% and 65.4%, respectively, and that were numerically similar to that seen with dupilumab 300 every 2 weeks in the SOLO 1 and 2 trials. It is important to note that these results were despite a higher placebo response than seen in the SOLO 1 and 2 trials and occurred with a much lower baseline median EASI than seen in SOLO 1 and 2 of 21.2% for the whole population compared to 29.7% for dupilumab 300-milligram every 2 weeks within SOLO 1 and 2. For the key secondary endpoints, you can see for IGA 0,1% responders, both the 300-milligram 2 weekly and 4 weekly deliver numerically greater improvement than the 150 milligram every 2 weeks, but with only the 300-milligram 2 weekly being statistically significant. The magnitude of response seen is generally lower than seen with dupilumab 300 milligrams 2 weekly in SOLO 1 and 2. But again, it is important to note that the IGA severity at baseline across all groups in WW001 was lower than that for placebo and for dupilumab 300-milligram 2 weekly in SOLO 1 and 2. For EASI-50, 75 and 90% responders, again, both to 300-milligram 2 weekly and 4 weekly show numerically greater improvements than the 150 milligram 2 weekly with all those who's being statistically significantly different from placebo for EASI-50 and EASI-75. And the 300-milligram 4 weekly was statistically different from placebo for the EASI-90 response. CBP-201 300 milligrams 4 weekly for EASI-50 and 300 milligrams 2 weekly for EASI-75 showed numerically similar percentage of responders to that dupilumab 300-milligram 2 weekly in SOLO 1 and 2. And again, this is despite a lower baseline EASI score in WW001 versus SOLO 1 and 2. Finally, for the key secondary endpoint of itch, again, we saw a consistent result for Peak Pruritus-Numerical Rating Scale reduction from baseline as with the other endpoints. With both 300-milligram 2 weekly and 4 weekly showing numerically greater improvements in the 150-milligram 2 weekly and both 300-milligram doses being statistically significant versus placebo. Absolute PP-NRS improvement at week 16 for both the 300-milligram arms were numerically significant to that dupilumab 300-milligrams 2 weekly in SOLO 1 and 2. Consistent with the lower medium-base PP-NRS score of approximately 6.9% compared to 7.7% in SOLO 1 and 2, there was a greater placebo improvement than seen in SOLO 1 and 2. Moving to the topline safety outcomes in the primary analysis. For CBP-201, rates of Adverse Events, Serious Treatment-Emergent Adverse Events, discontinuations due to adverse events and Treatment-Related Adverse Events were generally similar between placebo and CBP-201, except for a low rate of conjunctivitis, which was not seen with placebo. In addition, rates of other adverse events of special interest, or AESIs, such as injection site reactions and herpes infections were also low. When considering the established safety profile of dupilumab, no new safety [ symptoms ] were observed in the WW001 trial. In conclusion, the efficacy and safety results from the primary analysis showed that 300 milligrams 2 weekly or 4 weekly appeared comparable to dupilumab. However, as described, in trying to understand how these results could be put into context, we recognize clear differences in the trial populations in WW001 that prevented our direct comparison to SOLO 1 and 2. There are 2 potential key differences between these 2 populations. Firstly, there is less severe disease in WW001 as a result of the COVID-19 pandemic potentially contributing to no opportunities for disease flaring during movement restrictions. For example, less exposure due to environmental allergens and less stimuli to disease flaring. There was also additional increased competition for a decreasing number of the most severe eligible patients in clinical trials that has occurred over time. Finally, there are fewer clinical trial sites selected in our trial from academic centers as well as a different geographical mix for trial site selection. All of these may have contributed to an increased placebo efficacy response and a decreased efficacy response for the active treatment groups. The second key difference was a higher treatment discontinuation rate in the WW001 trial, again, in part due to the COVID-19 pandemic movement restrictions, potentially affecting trial conduct as previously described. This has led to increased patient dropout rates and potentially decreased patient clinic attendance for scheduled visits, and again, they have contributed to a decreased [ efficacy ] response, especially for the active treatment groups. With these differences in mind, we reviewed and conducted 4 additional analyses, both A Priori and Post-Hoc, that would allow us to determine how increasing baseline disease severity as well as how a more closely matched population recruited to SOLO 1 and 2 would affect the magnitude of efficacy responses seen. The first A Priori analysis was the analysis for the China population recruited solely in China. This group represented disease severity higher than the global population and closer to SOLO 1 and 2, with higher baseline EASI and baseline TARC. [indiscernible] reduced discontinuations. A Priori Analysis 2 was an [ analyst ] results. And this accounts for the normal distribution of baseline , reflecting the very low baseline disease severity seen in WW001. Analysis 3 was our first Post-Hoc Analysis and looked at the effect of EASI baseline stratified into tertiles. Finally, our fourth analysis was looking at the same stratification, but using it TARC at baseline. All of these analyses were very consistent and showed that with increasing baseline disease severity, CBP-201 efficacy resulted in further improvements across all the doses with placebo responses trending lower. With this, Connect believes that 201 at 300 milligrams has the potential for a differentiated efficacy and safety profile, with the potential convenience of 4 weekly dosing. This also reinforces the impact of the clinical trial design and study conduct can have on efficacy outcomes and informs our Phase III atopic dermatitis program plan. Before providing you with the detailed findings from our additional analysis, this slide summarizes some of the key trial population differences and highlights the much lower population severity enrolled in WW001 compared to SOLO 1 and 2. When looking at the baseline disease characteristics, whether it be EASI, IGA score of 4, PP-NRS score or BSA, the WW001 trial enrolled a much less severe atopic dermatitis population than SOLO 1 and SOLO 2. Of interest, though, is that the China subgroup disease severity across many of the metrics is somewhat greater than the overall WW001 overall population and closer to, but still less severe than the SOLO 1 and 2 population. Now moving into the detailed analysis. The first A Priori Analysis examines the patients enrolled in this China subgroup, which was performed to meet the requirements of a future regulatory package to CBE and to support our ongoing pivotal AD trial. As you can see, the baseline EASI, PP-NRS score and the proportion of patients with IGA score of 4 are higher than in the WW001 overall population and closer to the SOLO 1 and 2 population. Correspondingly, as the baseline severity increases, absolute and placebo-adjusted responses generally increase compared to the overall WW001 population across the primary end point of percent EASI change from baseline and all the key secondary efficacy endpoints. For the second analysis, we reviewed an A Priori Analysis, looking at the median responses for the continuous endpoint variables of percent EASI change from baseline and absolute change in PP-NRS from baseline. This analysis reduces the impact of the non-normal distribution of baseline severity, which is due to a low EASI score in WW001. With the medium of 21.2, 50% of all the patients in the WW001 trial had a baseline severity of less than 21.2 compared to a median of 29.7 for dupilumab 300 milligrams every 2 weeks in SOLO 1 and 2. Again, as per the China subgroup, as the baseline severity increases for both continuous endpoints of EASI percent change from baseline and PP-NRS change from baseline, both absolute and placebo-adjusted responses generally increased compared to the overall WW001 population. For our First Post-Hoc Analysis, we examined the effect of baseline disease severity using EASI baseline, stretching out the population into tertiles equally. This is similar to analysis performed for dupilumab from the SOLO 1 and 2 trials, as previously reported in the literature. The cutoff for our population tertiles were less than 18.4, 18.4 to 26.4, and a high EASI group of greater than 26.4, with approximately 70 patients per tertile. Again, as per the prior analyses, as the baseline severity increases, both the absolute and placebo-adjusted responses generally increase compared to the overall WW001 population. However, only the higher Tertile group most closely matched the baseline severity seen in SOLO 1 and 2, and that is where the greatest responses are seen. Finally, for the TARC analysis or Analysis 4, we performed a similar analysis to EASI, but now using baseline TARC, a biomarker for atopic dermatitis and type 2 inflammatory disease activity, again, split into Tertiles. The cutoffs for our top population Tertiles were less than 116, 116 to 291, and for the high-TARC group, greater than 291 picograms per ml, again with approximately 70 patients per Tertile. Consistent with the EASI baseline Tertile analysis, as the baseline severity increases, again, absolute and placebo-adjusted responses generally increase compared to the overall WW001 population. Again, it is important to note that even in the high TARC Tertile, there were a few patients with a baseline TARC greater than 1,150 picograms per mil, a figure that represented the upper limits of the low TARC Tertile in analysis before we dupilimab from the SOLO 1 and 2 trials and reinforcing a very different patient population and less severe population enrolled in WW001. So with this additional context, we now present both the primary and additional analyses in a side-by-side comparison for WW001 and SOLO 1 and 2, recognizing that these are not head-to-head trials, showing both the primary and key secondary end points, starting with the primary end point, which is the data for EASI percent change from baseline to week 16. This shows the CBP-201 300 milligram, either 2 weekly or 4 weekly is potentially at least dupilumab 300 milligrams 2 weekly from SOLO 1 and SOLO 2 or from the China Phase III dupilumab trial, with some evidence of numerically greater responses. A similar pattern seen with IGA-01, again showing that the China subgroup with CBP-201 300 milligrams 2 weekly is potentially at least as good as dupilumab, with some evidence of numerically greater responses than dupilumab 300 mg 2 weekly. When looking at EASI-50, 75% and 90% responders, these are also generally consistent and shows again that in the China subgroup, 300 milligrams 4 weekly dupilumab 300 milligrams 2 weekly for both the EASI 75% or EASI 90% responders with some evidence again of numerically greater responses. CBP-201 300 milligrams 2 weekly is also at least as good as dupilumab 300 milligrams 2 weekly on the EASI 75% responders. Finally, results for PP-NRS are improved as baseline severity increases, with absolute and placebo-adjusted changes from baseline generally increasing [indiscernible] both 300 milligrams 2 weekly and 4 weeks, we've shown the greatest responses and a magnitude of effect at least as dupilumab 300 milligrams 2 weekly. So in conclusion from the WW001 trial results presented today. CBP-201 and the WW001 trial met its primary endpoint and key secondary endpoints and showed significant improvements in skin clearance, disease severity, and itch compared to placebo in adult patients in moderate-to-severe atopic dermatitis. Crosstrial comparisons to SOLO 1 and 2 are difficult due to a less severe population being recruited and higher patient discontinuations due to the impact of the COVID-19 pandemic on trial conduct in the WW001 trial. However, additional A Priori and Post-Hoc Analyses of WW001 trial population showed as the disease severity increases at baseline, CBP-201 efficacy response further improved. With baseline severity that more closely match SOLO 1 and 2, side-by-side comparisons of CBP-201 300 milligrams every 2 weeks or every 4 weeks, appeared at least comparable, with some endpoints numerically better than dupilumab 300 milligrams 2 weekly. Finally, we believe that 300 milligrams CBP-201 has the potential for a differentiated efficacy and safety profile with the potential convenience for the 4 weekly dosing. Planning for a global Phase III atopic dermatitis program is now underway, with first patient enrollment estimated in the second half of 2022. Thank you. And I'd like to hand back to Wei.

Thank you, Selwyn. We are excited about the results of this trial and are very encouraged by the finding from the additional analysis and remain confident on the potential better efficacy and safety profile of CBP-201 coupled with more convenient and differentiated Q4W dosing schedule compared to other biologics totally available. We look forward to leveraging insight from the additional analysis as we initiate a global Phase III clinical trial program in the second half of this year. Additionally, as announced in September 2021, we have already initiated a clinical trial of CBP-201 in adult patients with moderate-to-severe AD in China. This multicenter, randomized double-blind placebo-controlled trial is designed to assess the efficacy and safety of up to 2 doses of CBP-201 in patients with moderate-to-severe AD. We expect complete enrollment in this trial in the second quarter of 2022. Data from this Phase IIb trial, including the China subgroup data, will be part of the submission package for approval in China. Meanwhile, the China pivotal study data will add to the overall efficacy and safety profile of CBP-201 in future global submissions for approval. That concludes our prepared remarks. Operator, you may now open the call for questions.

[Operator Instructions] Our first question comes from Kelly Shi with Jefferies.

So this is Hao calling in for Kelly Shi from Jefferies. So maybe 2 questions from me. The first 1 is given the additional analysis that you have done, it seems that the result is correlating to patient baselines. Does it impact your Phase III trial design? Do you anticipate any design change based on this additional analysis compared to your Phase II trial studies? And my second question is, for the ongoing Phase II/III China trial that you have in China, do you anticipate enrolling more severe patients as you show in the subgroup analysis in this Phase IIb trial in China? And maybe if you can also talk a little bit about the status of the trial? And what's your discussion with the regulatory agents in China regarding the regulatory path for CBP-201 in China.

Thank you, Hao. Thank you for the question. So there are two questions here. Let me quickly comment on the second question, and I will ask Selwyn to comment on the first one. The question is whether or not we need to enroll more severe patients in China in our currently ongoing pivotal study in China? Well, the answer is that we have been obviously very closely monitoring and have been working hard to make sure that we enroll the appropriate population of patients. And as you can see from our global study that we presented today, the patient population in China does have higher disease severity and we'll continue to monitor that study. So that is question #1. And I'm going to pass over this to Selwyn to address both 1 and 2.

So thank you, thanks for the -- Hao, thank you for the question. With regards to our global Phase III design, there are a number of considerations. First off, we have to have negotiations, obviously, with the help of authorities as the final trial design. So I can't commit at this moment in time ahead of including discussions with FDA, EMA, et cetera, on what the final design will look like. However, I think it's a fair question to say and ask what is the target population that we're trying to recruit within the Phase III? And it's exactly as you described. We will be trying to look at a more severe patient population that represents an opportunity to demonstrate the greater efficacy that we've seen from our analysis that we presented today. Now in terms of the logistics of doing that, I think I pointed out a few things within the presentation, but just to reiterate. We recruit to patients only from the U.S., China, Australia and New Zealand. And we know that patients in some geographies, based on the severity of disease and their lack of access to therapies in best cases, particularly in Europe and Latin America, generally will have more severe baseline disease, and therefore, we'll have opportunities to recruit those in a global Phase III program. I think I also pointed out within our presentation that we recruited a very limited number of centers that has very high or very strong academic involvement. And so these academic centers, generally speaking, will have the more severe patients to enroll, and we'll have the greatest experience of conducting clinical trials, which I think is particularly important during a time of the COVID-19 pandemic, where you want absolutely the best possible opportunity to perform these studies. Finally, these centers as well will also have the most trained and experienced investigators who can be very persuasive to ensure that patients remain in studies, despite, for example, the drugs working even if they don't know whether it's active or placebo. I think if I could, I'd let Dr. Silverberg into this discussion because as an experienced investigator and who's obviously seen a lot of date as well, I'm sure you'll have additional points to raise around the Phase III program and how to maximize the disease severity, if at all possible. Jonathan?

Sure. Good morning, everyone. I think it's a great question. I think you heard it was a great presentation. You really got to hear several of the important considerations that drive -- likely were responsible for some of the differences in the trial population and just understanding that it's likely multifactorial. I think, going forward, there's definitely opportunities, just very technical little things that can be done that can make all the difference in the world. One, just making sure that investigators are really appropriately selected and trained for dealing with diverse patient populations, making sure that they're really able to do a great job when it comes to the investigator assessments and that you have reliable assessments wherever possible. And I think in terms of setting recruitment goals, and again this is something that, obviously, the FDA and the EMA have to opine on, but 1 can set caps to make sure that there's an appropriate proportion of patients who meet severity quite -- who are severe. So this way, you have the opportunity to really see that subset, and you don't get this dilution or excess placebo response that comes with a little bit of a milder patient population. These are very easy things to be done in the trial setting. And I think there's also opportunities for going beyond in terms of the outcome measures. Obviously, there's a fairly prescribed set of tools to be used for regulatory purposes, but they don't always tell us the whole story when it comes to where the clinical value is or where -- which subsets tend to do even better with a particular drug. And I think there's opportunities there to take a deeper dive and collect even more information in Phase III and something that is already discussed and underway. So I think those are big ones. Look, the reality is, we all want COVID to go away. And I think we're not there yet, but we're definitely heading in the right direction longer term. And I think that we now know that there's great -- we've always known there's great safety with this class. I think that's such a major strength. But I think we don't -- early on in the pandemic, there probably was a certain degree of unfounded fear around staying on anything with the word immune in it during the pandemic. Now we know we have safety across the class suggesting that we shouldn't really have any concerns with COVID. I think that's going to improve study retention, even if we have COVID lingering for on and off for the next few years, so I think your retention rates in the trial will inherently be better without having to do much other than time has passed. And so I think -- and then just getting back to life closer to normal, patients getting more into routine settings, I think will also change the landscape of things. We've seen across many trials, potentially different signals than we would have expected because of COVID. And I think as we get back to return to work and life is normal, I think things will go back to a baseline of what we're used to seeing in terms of placebo responses, et cetera, in trials. So some of it is already in a sense done. But I think with just attention to a few important details, I think there shouldn't be really any challenges navigating these issues in Phase III.

Our next question comes from Joe Catanzaro with Piper Sandler.

I wanted to follow up on something, I think, you elaborated a little bit on, Selwyn, and even something Dr. Silverberg, I think, just mentioned. But wondering if you could comment on the performance of the placebo arm and whether the change in easy you observed there is also reflective of a population with lower disease severity. As I look at your placebo response and compare it to dupilumab, some metrics are comparable. It looks like maybe you have a little bit of an exaggerated placebo response. So I'm wondering if you could just help me understand what's going on there.

Sure. I'll take that. And I think Jonathan has also got a very clear view of what I think is happening as well. So, yes, I think the lower baseline EASI does generally affect the placebo response and leads to that trending higher. It's not unusual to observe such changes, and we've seen these -- if you look at [indiscernible] over time for the AD studies that have occurred, probably even since the dupilumab Phase IIb, not only do you see the trend in your baseline EASI as a mean or median fall over time, but you also see the same trend temporarily for the placebo response increasing over time. And I think that's probably due to the effect of a higher baseline disease in [indiscernible] reduction before reaching a nominal threshold. I think we've discussed before, we've seen potentially some of these indications before, not just with AD, but also with psoriasis, talking about a nominal threshold of [indiscernible], for example. And the efficacious -- or the highly efficacious therapies that need this more headroom to show these reductions. As you get to higher baseline passing of EASI, you start seeing greater reductions with these higher more highly efficacious products for the more stringent thresholds, such as EASI-90 or [indiscernible]. And so I think this is a clear trend that we've seen. I think it's it's clearly documented, but I'll leave it to Jonathan to add some more information on that.

Sure. Yes, I agree completely. The -- some of it is just a statistical numbers game. When you have a continuous endpoint like looking at just percent change of EASI, the higher you are to start with the greater is to show differences and that's maybe more of an issue or more of a consideration when you look at the actual treatment effect rather than placebo responses. But it is an issue that comes up with a placebo responses as well to some extent. The big issue, I think, that comes up is, for the -- there are several quantitative considerations for EASI. When you get to the lower end of severity coming into the trials, the EASI becomes less responsive as you get down to the real extremes, down to the really low levels. I mean, it's not so hard to get from, let's say, an EASI of 8 down to an EASI of 2, but it's really hard to get from an EASI of 8 down to an EASI of 0. So when you can see a continuous change, you can see lots of these modest placebo responses come up. But where you won't really see it as much in general is when you start pushing for those deeper end points, that's really where you'll see better differentiation of active drugs from placebo. So I tend not to -- it's often in the early phase studies. It's my recommendation and others to use the percent change of EASI as the primary efficacy endpoint just because it's best powered. But it's one that we know notoriously gives these kinds of crazy high placebo response. We don't pay too much attention to it. It's the deeper end points so I think that matter. But there's no question also that -- I think there's 2 other factors: atopic dermatitis, psoriasis was just mentioned by Selwyn, and I think there are a lot of analogies, but there's some differences, too, because psoriasis tends to be a relatively stable disease. Of course, we see flares, and we see fluctuations. But whatever we see with psoriasis just exponentiated with atopic derm, very, very unstable. And that can be both a challenge when it comes to clinically to lots of flares and towards disease control, but it also is a scenario where, a lot of times, there can be seasonality or fluctuation with different triggers, et cetera. And that can also be that with change of seasons, with drift over time, there will be what looks like a placebo response, but it's really just the sort of waxing and waning of the disease. And with longer-term studies, you have a better opportunity to monitor so it becomes less of an issue. And then I think the last consideration is just the more severe patients get just -- as a general rule, it would be true with any therapy, the more refractory they're going to get to any therapy. So someone who's the worst of the worst severity wise, often nothing works. So it's -- and conversely, when you go to the more moderate and milder patient populations, you'll see effects of [indiscernible] improving or just sort of a natural -- a little bit more of that waxing and waning, whereas the severe ones tend to be more severe and persistent. So I think this is something that -- it's not surprising, but the nice thing is seeing this and just anticipating, and I think it should be fairly easy to address as in considering the Phase III population because it is inherently tied to just making sure you have consistent -- sorry, tongue [indiscernible], consistent severity of the target population coming into the study. And I think then you have less to worry about with this.

Okay. If I could just ask 1 quick follow-up, maybe along a similar line of questioning, but with a different endpoint. I'm just wondering if you could help us understand why having a population with lower AD severity would negatively impact the rate of IGA responders? It seems to me that getting to an IGA response when you're starting at 3 is a smaller hurdle than if you're starting at 4. And if I remember correctly, I think in the dupilumab review documents, the rate of IGA response was notably higher in patients with baseline IGA of 3 than they were with patients with baseline IGA of 4?

So I can take that. And then again, I think Jonathan can add to that. I think, pretty much, Joe, is exactly, as you said. Actually, as our severity increases you can see a key thing which is probably the placebo response also drops as well. We've got, I think, a population where the average percentage was 31% in the IGA 4, so the more severe group, versus it was closer to 49% within the dupilumab Phase III studies. And the interesting thing is, as you see -- as you get very, very severe where the proportion rises in the China Phase III population for dupilumab also gives you a view of that. The responses for IGA do seem to change a little bit again. If you look at the China Phase III study overall for dupilumab, every other endpoint seems to be relatively consistent. I think the final thing to add, just really from our population though is, generally speaking, again, what we would expect to see is the IGA-01 responders as well as the EASI-90 responders tracking to a similar kind of number. And we do see a little bit of discrepancy between our numbers. And without going into the super, super detailed of individual patients, this is where things like discontinuations have a big effect. One patient is 2% in our study. So it's trends, I think, that we're very clear about, but I think the overall position is pretty clear for us. Jonathan?

Yes, I'm not sure I have that much to add. I mean, I think, it is -- it is technically true. I mean the more moderate the patient is, in general, the easier it would be to get to that clear, almost clear. Although, there -- even there, there's caveats because when you're -- well, in this particular, it's for the moderate to severe patient population, in general, that would be the case. But you would also see changes that would happen where it would be easier to see more placebo responses there. I think the other issue is, as you start looking at some of the the binary endpoints, whether that's EASI-75, whether that's IGA clear, almost clear, I think we also have to always be careful with this because -- and this is where I agree with Selwyn. So you look at the overall picture and the totality of the data and you get a very clear picture here that there's efficacy and that there is indication to move forward. But I think we also -- as a statistician myself as someone who's trained, I'll always really say, you have to be so careful about making these kinds of cross-trial comparisons and you shouldn't do it. And then, of course, the investor community, right? This is everything you do is trying to make those comparisons. But sometimes, in a smaller Phase II study, this is not even a small study, but as opposed to just the largest size Phase III programs, 1 or 2 patients can shift so many things. And so when you look at some of the binary endpoints, I wouldn't make too much of it other than I think there are strong positive signals here and definitely more than enough to say, okay, we should move forward. Beyond that, I think that's where we have to then really test it out in a larger Phase III program and really see across an even larger pool of patients and even more diverse patient population how well the results hold up. But I look at this and I'm quite optimistic.