Vicore Pharma Holding AB (publ) (VICO) Earnings Call Transcript & Summary

Welcome to today's webcast with Vicore Pharma. There will be a presentation followed by a Q&A session. [Operator Instructions] I'll now hand over to Carl-Johan to begin. Please go ahead.

Thank you very much, and welcome, everyone, to this webinar. We're thrilled to share this unprecedented data set today with you. Since last update, we have doubled the number of patients included and data from double the number of patients and the compelling results remain. Can I have the next slide, please, #6. Now as you all know, idiopathic pulmonary fibrosis is a deadly disease and with a rapid progression. And we all know that licensed medicine today slow this progression, but at a price of severe GI in tolerability. Now C21 show long-term stabilization and stabilization above the baseline. So we stimulate regeneration, repair the alveoli, stop the progression and stabilize the disease above baseline. Can I have the next slide #7. First of all, I would like to put our data as a [ bit ] in a context. To the left, you see the normal decline of IPF patients, whether they're moderate, severe or mild, they all declined with a rate of about 250 mLs per year. The other thing that's also important to remember that this, of course, also applies to both early and established IPF. So the typical and possible UIP cohorts, they show the same decline, both in placebo setting and as a response to antifibrotic treatment. So we have the same decline in both these cohort. And I think this is now important because here, we are different because we see an increase and a stabilization and even so more in the early type. And that is based on the science that we have for our drug, C21. Now can I have Slide #8, please. So the explanation for the strong data now is to be found in the underlying science. And in the normal alveoli, that's a little bubble where the gas exchange occur in the lung, -- there are 2 types of cells. The first call is the cell where the gas exchange occurred through. That's the type 1 cell. And there is a constant replacement of these cells and the type 2 cell is a progenitor cell that proliferates and differentiate and replace type 1 cells that needs to be replaced. So this is on a normal circumstances. Now in fibrosis and in IPF, this mechanism doesn't really work properly. So the type 2 cell is functional. It may be senescent or it may be infected or apoptotic for any reason. And then there is a wound generated on the alveolar surface. And this would then is healed, like any other wound by a scarring, scarring process that starts and perpetuate itself, which is the trigger point for fibrosis. Now C1 stimulate these 82 cells. And it's actually the only cell in the lung parenchyma where our receptor is expressed. The target is expressed on this very cell, and it is a regeneration repair mechanism. So we heal the [ alveolar ] integrity, which also leads to a functional repair of the lung capacity. So we're targeting -- we're hitting really the trigger point of IPF. We're not blocking enzymes or blocking receptors ubiquitously spread around in the body. We stimulate the repair in this very context. Can I have the next Slide, #9. So we have both preclinical and clinical support for this mechanism. And we have shown in human lung that we have a target engagement at the right concentration corresponding to the Cmax where we have the effect of our drug. We have shown that in slices from human IPF lung tissue, we can reduce the expression and protein release of TGF-beta 1, which is a marker of ongoing fibrosis. And this we also see actually in the clinical trial that you will see later on. So that kind of closes the loop on the biomarker side. We reduce the drive for fibrosis, and that could be -- then seen as a reduction in [ IPF ]. Finally, we restore lung function in the wild type COVID infection. The wild-type COVID infection did infect with type 2 cells with the same kind of pathology as in IPF. And here, patients were in a much higher degree and statistically significant less in need of supplemental oxygen. And if you don't need supplemental oxygen, you have functional lungs. So we can say that we restore lung function. So in a way, it's not a surprise that we also hit IPF. But in all honesty, these data are a little bit beyond our expectations. Now I hand over to our CMO, Rohit Batta, who will describe the study on the study population. Thank you.

Thanks, Carl-Johan. So I've heard IPF other patients describe IPF as an unstoppable disease, but that certainly hasn't stopped us at Vicore Pharma. If anything, you will shortly see from Toby's presentation that C21 is getting ahead of the disease. But before we dive into the data, I would like to share an industry physician's perspective on how and why we designed the AIR trial the way we did. Firstly, some learnings from other IPF trials are: a, it displays homogeneity, two, as a cohort, and as Carl-Johan, there's this constant consistent decline in lung function, whether you are mild, moderate, severe or early or established; C, you have a reliable FVC endpoint which regulators favor. And D, there is certainly room to make an impact. There's a large gap to fill. We have 2 licensed therapies that don't impact the mortality but negatively impact the quality of life. So we took these insights and applied them to the AIR trial. If you go to the next slide. So firstly, air is an ongoing multicenter global trial, and although the primary was safety, we designed it to allow accurate interpretation of efficacy. Our original goal was to enroll 60 treatment-naive IPF patients. And our thinking behind it was that with a treatment-naive population, we could fully characterize what C21 does in monotherapy and have a clear go/no-go signal. And with the 60 patients, we estimated that we could demonstrate better efficacy and better safety and you'll see from Toby side that we certainly overshot that target. Now AIR is an open-label trial, and we carefully applied a historic control is 240 mLs loss of FVC per year, which has been demonstrated in placebo arms of registrational trials like the nintedanib studies. From a diagnosis perspective, [ AF IPF ] requires correlation of the clinical setting with radiographic images and these patients were locally diagnosed with -- by published guidelines. However, we wanted to doubly confirm they had IPF. So the HRCT scans, these are the diagnostic scans were sent to Professor Devaraj, who's an ILD radiologist at the Royal Brompton Hospital, who's been participated and been the reader for many international trials like the [ EMBO ] and he confirmed that they do indeed have IPF. The primary efficacy endpoint was changed in FVC, and this is the FDA preferred [indiscernible] for mortality. And then from our FVC monitoring perspective and data collection, we had a militant approach, and we engaged the market leader, [ Clario ], which performed the majority of studies out there with FVC, and they provided the hardware and the software and the centralized over read. So the data goes up into the cloud and then the Clario physiologists interrogate it, making sure that it meets relevant guidelines. And then finally, we wanted to treat these patients long enough, not a short 3-month trial, which sometimes raises eyebrows, but a 9-month duration with frequent data points, removing any doubt that this was due to chance. Let me go to the next slide. So let's drill down into the numbers, so you get a better understanding of the population. So starting on the left-hand side, IPF belongs to a family of ILBs and with it, it accounts for about 15% of cases. And in such cases were screened by the local PI and 43% failed screening as the ILD radiologist, our reader wasn't convinced that it was indeed IPF. 51 subjects were enrolled in this interim and 57% of patients had probable -- sorry, 39% had probable early and typical established was 57%. And it's thought that the remodeling fibrotic process is a continuum from the early the bubble through to the established typical where you then get the honeycombing. Now given these unprecedented results, we decided to get a second opinion on those diagnostic scans. So we engaged with the group called [ Qurate ], which is a spinout from Cambridge University U.K. And they have a 3D imaging platform where they have fibrosis quantification algorithms that have been trained on thousands of IPF patients from the U.K. And in short, the human and the machine agree that these patients do indeed have IPF. So we've had extreme accuracy of the diagnosis for this population. If you go to the next slide. So there is a baseline overview of the interim results that we just shared at ATS on Sunday by Toby during his podium presentation. And the demographics and the clinical characteristics are very similar to other IPF international trials. And it's not surprising that the mean age was 68, and we've had predominantly male population. And we recruited patients predominantly from India, and there's no scientific or medical reason to suggest that IPF is different in different parts of the world. The HRCT pattern is -- the split there is very similar to the impulses pool trial, you can see there on the right-hand side. The FVC percent predicted 75.3%, which is very close to the pool impulses. If anything, we recruited a more FVC impaired population. And no subjects had been previously treated with antifibrotics. So in summary, the most recent recruitment has put the patient base very close to what's seen in other trials. And at that point, I'll hand over to Professor Toby Maher.

Great. Thanks, Rohit. So if we can have the next slide. So as Rohit says, this data was presented, I think it was earlier on this week at ATS. I've lost track as the meeting started on Sunday. So I'm not sure what day of the week we are anymore. But yes, we presented this data at the American Thoracic Society meeting in Washington, where it generated a lot of interest. So Rohit has talked you through the background of the trial and the design and the data here from the May interim analysis is just going to be presented to you in a few different ways to account for patient dropouts and missing values. So what you're seeing on the screen here is the actual data from the study as of the May data cut. In orange, we have the observed values for patients in the trial. And then to provide context, we have the standard of care in the untreated lines, which are the rates of FVC decline observed in the pivotal studies for current antifibrotic drugs, which is to say that historically in IPF trials using the same inclusion criteria, we would expect patients not on antifibrotic drugs to decline by about 240 mLs a year. And with the anti-fibrotic drugs, we've seen that approximately halved. So we would expect patients on standard of care, the pirfenidone or nintedanib declined by approximately 120 mLs a year. So those lines are added in for context. But as you can see, patients treated with C21 show initial stabilization. And once you go beyond week 12, even improvement in their FVC -- and Rohit has just talked you through this concept of patients with early and established IPF, which is to say the difference between those who have honeycombing, which is clear and irreversible destruction of the lung versus those who don't, but have otherwise the very typical pattern of IPF. And you can see in this data cut that the proportions of patients with the 2. And there is some missing data as the database is still being cleaned. If we move to the next slide, this provides the same data, but it's a 3-month rolling average of values for each subject. So it smooths out the line. It slightly pulls down the effect. But nonetheless, we continue to see a very positive impact of treatment, particularly once we get beyond week 12, week 16. And then if we have the next slide, this is the most conservative approach that has been taken. So this has imputed missing values, assuming that those patients who dropped out had a rate of decline equivalent to an untreated patient in previous pivotal trials, which is to say that they would have lost 240 mLs a year or 120 mLs every 24 weeks. So those values have then been imputed for those that were missing data from the point at which they left the study. And you can see here that, again, despite this conservative imputation, there is still ultimately an improvement in FVC again, once we go past the week 12 to week 16 time point. We can have the next slide. And then this is to try and provide some context. So this is actually taken from data that myself and colleagues generated when I was back in London now 7, 8 years ago. And in the study that we ran, we recruited patients with newly diagnosed idiopathic pulmonary fibrosis, and we gave them home [ spirometer ] and asked them to measure their forced vital capacity every day for 18 months. And the data presented on here in black is the disease trajectory of each one of those patients in that study. And so you can see that -- and these patients weren't on any background therapy. We actually completed the study before defend pirfenidone and nintedanib were available in the United Kingdom. So this is the natural trajectory of untreated patients with IPF. And you can see that untreated IPF patients, there are a small group, less than 10%, who do show stability over 12 months. There's a much bigger proportion, 20% who show rapidly progressive disease, which is to say they lose more than 10% -- more than 50% of FVC over 12 months. And then the average patient, which is the sort of cluster in the middle, lose somewhere between sort of 7% to 15% of FVC over that 12-month period. And overlaid on here, we have the disease trajectory of the complete set of patients who've got data going out to 36 weeks in the study. And you can see that the trajectory for these patients is distinctly different to the trajectory that we saw in our untreated prospective cohort. Next study, next slide, sorry. And then this is an analysis, just looking at the difference in trajectory for patients with early and established IPF. So as a reminder, the established ones have honeycomb disease. So this is really a very clear architectural destruction of the lungs. And you can see in orange here that although this group do better than our historic expectations. The patients who see the most clear improvement in FVC over the 36-week period of those with early IPF, which is to say less destruction of the lung who start off with lung architecture that is closer to normal with less irreversible scarring and this is quite an interesting observation, suggesting, of course, that the earlier we can treat the disease, perhaps the greater probability of reversing fibrosis. And once it's established, although we can see some treatment effect, the opportunity for improving disease is somewhat diminished. Next slide. And then this just tries to provide a subgroup analysis. And obviously, numbers are small. But if you look within this bar at 36 weeks, you've got here the trajectory of the observed average data -- and then within this box, you can see the 36-week outcomes for these different subgroups. So the patients with probable or early IPF were the ones that did best, female and male have very similar outcomes, albeit, as expected, with IPF, the proportion of female patients is a lot lower. You can see that the patients from India did the same as the average patients and those from the rest of the world had very similar outcomes. And then at the bottom, we have the patients who had typical or established or advanced IPF with honeycombing. So as a sort of a sense check of the data, we can see that these groups are behaving very consistently out towards the 36-week endpoint. Next -- and then this is just providing some context based on other recent data that has been published. So you can see here, based on the legend of the data as has been published for some of the other drugs in late-phase trials in IPF. So we have the client data with [ alpha v beta 6 ], inhibitor, showing the data from highest dose arm -- we have the data from the BI PDE4 inhibitor study, which obviously was a 12-week study, so the data mainly available to that point. And then the data that was published at ATS by Bristol-Myers Squibb with the LPA1 receptor antagonist. And again, all of these studies have shown slowing of FVC decline compared to placebo, the [ BI-1 ] has shown stabilization out to 12 weeks. But none of them have shown this improvement in FVC that's been observed with this study. So the next slide. And then this is the tolerability of the safety data from the AIR trial. And it put side by side with the data from the INPULSIS trial. Of course, the INPULSIS trial was 52 weeks. So there are some slight differences given that the AIR data is, first of all, an interim analysis and second of all, out to 36 weeks. But nonetheless, you can see that there are very few gastrointestinal side effects reported with C21 and the proportion of these events is similar or less than we've seen in the placebo arm of the INPULSIS studies. Also, and I always think this is good to look at in trials is the respiratory adverse events. So episodes a progression of IPF or cough. And these are low in this study. And again, that compares favorably with what we've seen in the INPULSIS data. And really, the one adverse event that has been seen, and this was known about from the Phase I data is hair loss. Again, we don't have granular detail on exactly what is meant by hair loss. But from what has been said by investigators, this is sort of excess loss of hair rather than total loss of hair and is an expected side effect based on the drug's known mechanism of effect. And then just finally, looking at the sort of fatal and severe AEs, the number of these events again compares favorably to what we've seen in pivotal trials. The next slide. Yes, this has been thrown in. There's a little bit of mechanistic data to test [ to me ]. But it's just speaking to the data that has been generated so far, showing that the sort of preclinical data using a combination of in vitro, in vivo and ex vivo assays is generally very consistent that the effects we see an important mechanistic pathways, including TGF-beta are borne out in the clinical trial, where when we look at biomarkers of -- that are important in disease. So in this case, plasma TGF-beta 1 levels, we see an appropriate reduction in the plasma of patients treated with C21 compared to their own baseline. So you can see here a greater than 50% reduction in TGF-beta 1 levels in plasma patients. And that is, as I said, in keeping with the preclinical data showing an effect of C21 on downstream expression of TGF1 beta. So providing further support for target engagement in this study. And on the next slide. So obviously, when one designs trials in retrospect, there are things that one would love to have done. And I think in this study, in retrospect, it would have been good to collect quality of life data to provide additional support for the FVC data that we've seen. But at the time the study was designed, the goal was to keep it lightweight and deliverable and to avoid unnecessary burden on patients. So there were no formal quality of life questionnaire administered to patients. Nonetheless, there were exits interviews performed with the investigators and their subjects. And on the left-hand side, at 12 and 24 weeks, subjects and their investigators were asked to make a decision about whether their experience in the study favored continuing with the drug. And you can see here that the vast majority, over 90%, both at 12 weeks and 24 weeks felt that their response to treatment favored them continuing over any concerns about side effects. And then on the right-hand side, you can see individual or statements from individual patients and their physicians relating to their experience on drug -- and there were many requests for patients to continue to have access to drugs such was their experience with treatment. You can see on there some comments about improvement of cough during the treatment period and cough is an important symptom that has a very negative impact on patients' quality of life. And then the final statement, again, just reiterating the general desire that people have for continued open-label access to drug, suggesting that the study participants themselves clearly perceived benefit from therapy. And if I can have the next slide. And then finally, just to address the challenge of dropouts. So as people will be aware, this study was conducted during COVID. It was a very difficult time to run clinical trials during COVID, particularly trials for patients with respiratory disease around the world are patients with respiratory disease were being told that they were at increased risk of death from COVID. They were also getting the message that health care facilities and hospitals were a good place to catch COVID. So we had enormous difficulties in continuing to run trials during this period with many patients just refusing to come up to health care facilities. And this wasn't a problem just with the AIR trial, I was involved in many trials that ran into COVID. And while some of them managed to complete successfully, there were many trials that actually ended up closing down because they just couldn't deliver. So it's a testimony to Vicore and their study sites that actually the trial was able to continue during COVID, but it's also unsurprising that there were an excess of patient dropouts during this period of time. And so we're just trying to illustrate on this slide the effect that COVID did have on study discontinuations. And so you can see that in the 18 months during COVID, so from March 2020 and 18 months onward as a sort of approximation of when COVID was at its worst. That is when the majority of study dropouts were seen particularly the very early dropouts before week 12. And you can see that as the world has normalized in the last 12 months, the rate of study dropouts has declined considerably. There have only been 3 subsequent dropouts before week 12 and only 2 later dropouts. And in fact, during this calendar year, during 2023, there haven't been any dropouts whatsoever. And it's worth saying that on the imaging reanalysis that was done by [ Qurate ] that didn't appear to be any imaging differences between the patients who dropped out of the study and those who discontinued. So we're not finding that this group of dropout somehow have some unique phenotype that might explain the finding seen in actual fact, the dropout seem to be very consistent with those patients who continued later in the study. And as I say, COVID itself is what has driven the majority of dropouts rather than any perception of treatment failure that might have skewed the results. So I think that brings us to the end of my section. So as we've seen, C21 thus far in the interim data in this open-label study has shown that patients improve their FVC over 36 weeks. Subgroup analysis shows a consistent effect across different subgroups of patients with that interesting difference between patients with definite or more established IPF and those with probable or more early IPF without the honeycombing. And then as I've shown you, the study has had very good recruitment rates, and the majority of the dropouts that we saw can be attributed to the COVID pandemic. So I will hand over now...

Thank you very much, Toby. Now you've seen the data. And with this data, we are very confident to move into the next Phase IIb trial. And if I can have the next slide, please. So the design of the IIb trial is fairly conventional IIb trial. It is a randomized double-blind placebo control with parallel groups. Multicenter, it will be a global trial. We will have a dose confirmation. So it will be the dose 100 milligram twice daily that we used in the AIR trial, but also a lower dose and then a placebo group. So all in all, we expect 270 patients, and they will be all comers, meaning that they will have background therapy that could be pirfenidone or nintedanib or no background therapy, and we expect about 1/3 of each to enroll in this trial. We're going for a 24-week treatment duration, and we will, of course, assess the efficacy. And this is a trial that we have vetted with the FDA and design is endorsed by the FDA. And also to the degree that we have a confirmation that if we have a 2-sided alpha, a strong result, we will have substantial evidence of efficacy, which means that we only need to do one extra Phase III trial. But I would say that if we can repeat the AIR trial, I'm convinced that we would have a good chance to also go for an accelerated approval, of course, this is nothing that an agency will tell you on beforehand. Now can I have the next slide, please? So not only did we report outstanding data last week. We also got a very important patent approved granted in the U.S., and that's the formulation principle of enteric coating. And it's not a conventional formulation pattern that it's easy to work around. This covers all enteric coatings. And I have a very hard time to see how a generic company could copy and do an enteric coating without doing enteric coating. So we're covering all enteric coatings. And I think that is something that is, in my mind, as strong as an NCE patent. And now we have that granted in the U.S. We are allowed in Europe. We have a positive IPRP. So we expect to have product protection to 2041 in U.S., Europe and many other territories within short. And of course, with this in mind, I think we have a much better prospect of both attracting partners, but also to be able to capture the breadth of the ILD field, not only to rely on orphan designation, orphan drug designation. So if I can have the final slide, please. In summary, then since we last met in this medium, we have updated the interim analysis with now twice as many patients as before. And I think what is new and what is strong is that we now have also confirmed the IPF diagnosis of our patients in the third way with the 3D reconstruction of the high resolution CTs. We continue to see spectacular efficacy data. The number is a little bit smaller than previous time, but that's because the patients that we have added have primarily been of the type that is typical UIP so more advanced disease. So we should expect the number to come down, and I think that's important. And even though that has been the case, the typical UIP patients, the established IPF patients still stabilized above baseline. We have continued a very good safety and tolerability profile and no GI signals. And finally, we have supported biomarker data and also a set of encouraging patient testimonies. Now we are convinced that C21 stops disease progression stabilizes and cease above baseline at 36 weeks, which I think is an extraordinary result. And to be honest, I think there is nothing like this. Thank you. I will now open up for questions.

[Operator Instructions] Our first question comes from the line of Gonzalo Artia from ABG Sundal Collier. Please go ahead.

First one from what you explained and the different responses seen with C21 seem to result from differences in severity of the patients. If we compare it with [indiscernible], I think if I hear correctly from one of your slides, these 2 drugs do not show this difference of response based on UIP pattern, right? I mean throughout your 3 interim analysis, we have seen that your range of responses with C21 seems to be quite wide based on UIP pattern. So I was wondering if these differences in response have any implication for your upcoming and [ trial ] design. And also, what would be the percentage split between UIP pattern probably typical at diagnosis in the real world?

Maybe I can start and if anyone wants to chip in, please do. I mean we will have an all-comers population. And typically, there is about 1/3 that is the probable UIP pattern and 2/3 that is typical. We had 39%, I think, in this stride and I think that's also a very common split. So we will not see any major differences in the population. Apart from that, these will be all-comers and they will -- some of them have background therapy as well.

Okay. And the other part is why -- well, am I wrong that of [indiscernible], they don't show this difference in between the UIP pattern and why is it?

Correct. And they all stay below on a core basis baseline, we are about with both typical and probable. And there is a large post-hoc analysis of the 2 registration trial for nintedanib and that was shown here in the slide deck, and they go just in parallel, both in the treatment group and in the placebo group.

Yes. Okay. And second question, in the second interim readout last November, I think you reported that you only had one acute acceleration of the reported patients at that time. I think that you did not provide this information from the updated -- for the updated patients now. Is it still number low?

I think that's a question for Rohit.

Yes. Thank you. So we have had 3 reported cases of acute exacerbations. One of those cases were actually the patient had COVID at the same time. So it was really difficult to sort of untangle what was due to COVID and what was actually a true exacerbation given sort of the lung injury there. The 3 cases. And actually, I think I believe one of them is actually a nonserious -- so point I'm making -- they're not adjudicated, but 3 in total, which relatively is a low number. If you think about what's in the literature, I think the annual incidence of acute exacerbations perhaps between 15% and 20% a year. So still a relatively low number. And obviously, we will be monitoring this closely and looking at this in the next Phase IIb trial.

The next question comes from the line of Viktor Sundberg from Nordea.

Yes. Congrats on the data update here. So I have one on the 24-week data here in the update. As far as I calculated, it looks like the 40 new patients that reached 24 weeks and just update had a much lower FVC on average versus the average update that you had in the 4 last years. My numbers, they must have had an average FVC of about minus 101.3 milliliters if the whole group has an average of 50 milliliter across all 27 patients. And I just wanted to understand if the 14 patients that reached 24 weeks in this update at any different baseline characteristics other than the probable and like the [ UIP ] pattern or how the dispersion look like, i.e., was it close to the mean? Or do you see any outliers that adversely impacted the results of this new group of patients that reached 24 weeks.

Well, maybe I can start again then and ask you to chip in, if you feel. First of all, I think there is a large intra-individual variability in these measurements as well as a large interindividual variability. And that's also the reason for us to use the 3 values, the mean of 3 values, 3 adjacent value or 3 visit rolling average, that kind of evens out this intra-individual variability since we have this high number of data points, we can allow ourselves to do that. And I think that gives you a much more even spread out. And I think it's -- or even shape of the curve. So if you look at the weak -- the measurements before and the measurements after 24, you will have the line, so to speak. And I think it's more by chance that you have a dip there. And then what is true and what we said in the beginning also is that the new added patients, the new numbers that have been added is primarily derived from patients with established with typical UIP. So they're expected to be lower than the first data set where we had more probable UIP.

And maybe on that same topic, if you compare it to 3 data points rolling average from the previous studies, do you see a different trend then? Or how does that look but that would also be interesting.

No. I mean the trend has been the same all the time from February -- from the first interim we did in February, but that was an interim where we had a large proportion of probable now that has been kind of leveled off with typical. But again, if you do the subgroup analysis, we'll find typical on the positive side of the x-axis and probably on a bit higher up on that end. So the pattern has been the same from the very beginning until now, the numbers are a little bit different, very much depending on how many probable and typical you have in your analysis.

Okay. And maybe one for professor Maher here. If you look at the baseline characteristics, patients have been diagnosed or have had the IPF for a shorter time. It looks like in the AIR study compared to other trials that have been done here. Is that even relevant for any of the outcome measures? Or how should we view this.

All right. So it's been an interesting observation actually in recent studies, not just the AIR trial that we are capturing people closer to diagnosis for entry into trials than we were perhaps sort of a decade ago when the pivotal studies for pirfenidone and nintedanib were done. So you're right, if you look at INPULSIS and ASCEND from memory, the average time from diagnosis was about 3 to 3.5 years. And at the time, we didn't have treatments available for patients and often those that were most desperate to go into trials were those who had disease for a while, whereas if you look at the recently presented ISABELA studies, which were the negative trials from Galapagos, average time from diagnosis in the patients who were untreated with antifibrotics was under 2 years. I think it was about 1.5 years from memory. So I think what we're seeing in the AIR study does reflect what has been happening in trials in general. As to whether that makes a difference, I don't think it does. Historically, we've tried to look for factors that might enrich clinical trials. And so work has gone into trying to understand whether disease trajectory changes after time from diagnosis, and it doesn't appear to. So one would expect disease behavior to be consistent whether patients a year, 1.5 years from diagnosis or whether they're 3, 3.5 years from diagnosis. So I don't think that has an impact, but it certainly reflects trends that we're seeing elsewhere.

The next question comes from the line of Chien-Hsun Lee from Pareto Securities.

Congratulations on the data and great progress with Vicore. So maybe a few questions for Dr. Maher. Since there's a high percentage of Indian patient in this trial, do you see any biological difference between the Indian patient and the rest of load? And if so, has such difference ever been observed in other trials such as INPULSIS 1 and 2.

No. So in straightforward clinical practice, we see lots of Indian patients with idiopathic pulmonary fibrosis. And we've never identified differences in disease behavior in prospective studies that we've done, for instance, the [ PROFILE ] study, about 10% to 15% of the patients recruited at [indiscernible] Center in London, where of Indian origin. And again, their disease behavior was entirely consistent with that seen in the Caucasian population in the study. And I think as Rohit went to lengths to explain at the start, there was central overreading of the CT scan. So an experienced ILD radiologist was making sure that the patients from India have CTs that were indistinguishable from those recruited in the rest of the world. So I don't -- I can't think of any reason why the Indian patients should have behaved in a way that was different to what we would expect for any other patients with idiopathic pulmonary fibrosis. And like I say, we've not ever recognized ethnic differences in disease behavior when it comes to FVC decline in patients with IPF.

Okay. And maybe a follow-up question. So there seems to be quite some variability on the FVC measurement. Is it common to see this kind of big variability in IPF trials? Or does this big difference really indicate that there's something positive happening in the [ lungs ].

That for me. And by big difference, you mean the improvement at week 36. I know that is sort of striking to use Carl-Johan's word unprecedented -- if you mean visit-by-visit variability, again, important steps were taken to exclude technical variability as a cause of that. So the spirometry was performed on standardized spirometer. So all sites have the same spirometry equipment. All sites were trained in the same way. And there was central overreading of the spirometry output to ensure high-level quality control. So technical variability was removed as a consideration. But we do recognize that there is some visit-by-visit difference in patients. Some of that can be driven by consequences of recent infection, mucus retention. So it is normal to see some up and down in individual patients readings from visit to visit, which explains some of that variability that one sees over time. But nonetheless, just to see that consistent on average improvement across the whole population is not something we've ever seen before with more than growth.

The next question comes from the line of Havin Kanda from Carnegie.

Congratulations on the positive data. I just wanted to follow up a little bit on the previous questions. So I was wondering if you could elaborate a little bit on the expectations you have on the effect size as you move towards randomized trials -- assuming that the lower FVC compared to the last readout is explained by the proportion of early IPF and established IPF in the added patients and that the Phase IIb study would have a proportion of these groups in line with what you would expect in a large randomized trial. Do you expect the FVC to erode further compared to what we see in this readout? Yes, I'll start there.

I think, there's a question for you Toby as well.

Sorry, I didn't catch the very last part of the question.

Yes. I guess based on the observed trends we're seeing here and assuming that the proportion of early IPF and established IPF would be in line with what you expect in the upcoming Phase IIb study? Should we then expect the average FVC to rode further in this study over 36 weeks?

I think, as has been said, we would expect the proportions in the trial to be about 35% of patients with definite UIP and the remainder to have probable UIP -- so not dissimilar to what we've seen in this study. I think the normal -- in the normal situation when powering an IPF trial, our expectation is that we're going to be slowing FVC decline. So the usual power assumptions would be to show a relative reduction in decline of about 50% and to power accordingly. Clearly, with the data that's been generated in this study, one could take a more optimistic approach to powering. However, I think it would be much better to take traditional assumptions and then to far exceed them because one would be able to see relatively early on in a double-blind trial, if there was a treatment effect assuming that this is replicated at a Phase IIb in a randomized controlled trial that actually the power assumptions based on this data would require about 40 patients per arm. But I think sensible to use traditional powering approaches and to vastly outperform that than to under power.

Okay. And I guess, secondly, how sure can we be that the proportion of these UIP pattern follow what you would expect in a large randomized trial since I understood that this has not been reported with nintedanib and pirfenidone. How convinced are you that this has not been a more selected population?

I want -- so again, when patients go into these later phase trials, there is central overreading of the CT scan before patients are enrolled. So for instance, in the pivotal study of nintedanib in patients with progressive pulmonary fibrosis, the study team there decided that they wanted to try and manipulate the inclusion to make sure that they had enough patients with UIP pattern. So they use the central reading as a way of providing real-time data for the pattern of CT that was going into the study. And in the same way, it would be possible to monitor this -- the Phase IIb study for C21 as it goes in real time to ensure that the proportions don't sort of massively change from that seen in the open-label study. So I think one can, if necessary, put caps on the particular patterns that are included. And if one wanted, one could take a decision to try and enrich the trial for patients with earlier IPF. And I think certainly, it's a point for discussion based on the data shown today and something that could be monitored in real time with the trial delivery.

Okay. Great. And the last question, if I may, to the Vicore team. On the ANDAS study, are you aiming to have any sort of adaptability into this study, making it possible to upsize it? Or any open-label extension that perhaps could help you if some accelerated pathway would be on the table?

Yes. I mean we -- whether or not it will be an interim analysis, we will have to discuss a little bit more. We will certainly be able to put in a futility analysis that doesn't cost anything from a statistical point of view. And of course, we would have a [ DSMB ] that will have also access to what's going on. But of course, I can highlight any big differences in -- from what we expect and especially of course, on the safety side. when it comes to an open-label extension, I think that's something that we consider and consider it in a positive way because that is also something that will increase recruitment because people do like to have the chance to have a drug that is working. So that could be something and that would be a way also to increase the safety population for an accelerated approval, for instance. So we are considering that, yes.

And we believe you have met all the requirements for breakthrough designation with the generated date? Or would that require you to show something else?

For a breakthrough designation? Was that the question?

Yes, that's correct.

I think we fulfill all the criteria for breakthrough designation. We have to file an IND to apply for it. So it will come in due course.

The next question comes from the line of Sushil Hernandez from BLK.