Hello, everyone. My name is Laurie Doyle, and I would like to welcome you to the Inflarex conference call to discuss the results from the open label extension part of the SHINE study with IFRS-one and to provide a strategy update. On the call today are Inflarex's CEO, Doctor. Niels Riedeman and CFO, Arndt Christ. On this call, Inflorex will make forward looking statements, which are statements regarding the company's intentions, beliefs, projections, outlook, analyses and current expectations concerning its business.
For a discussion of risks and uncertainties that could cause actual results to differ from such forward looking statements, please refer to the factors described under the heading Risk Factors in Inflarex's Annual Report on Form 20 F that was filed with the U. S. Securities and Exchange Commission. Also, please refer to the notice and disclaimer associated with the slide deck for this presentation. These statements speak only as of today, and we assume no obligation to update these forward looking statements even if new information becomes available in the future, except as required by law.
With that, let me turn the call over to Niels. Niels?
Yes. Thank you, Laurie. Also on my behalf, welcome, everyone, to today's call. We are very excited to share with you today the SHINE study results of open label extension study as far as our snapshot analysis go. And then in the call, I will hand over to Arndt to go through some financials and also finish up with some strategy update.
So let's jump right together at Page 4, if you will. Let's look at the SHINE study patient disposition. Before we do that, I just want to remind you this is a snapshot analysis that was done after the end of treatment, which occurred on September 6. That was week 40 of the trial. The last patient last visit as planned was occurring on October 4, so just very recently.
So the final data will be available early 2020 after closure of the database. But as far as the end of treatment at week forty goes, we feel that these data are very definitive, and that's why we feel comfortable sharing them with you today. So 179 patients enrolled in the trial, and about 177 patients really started dosing on either placebo or IFRS-one treatment. Out of these patients, 156 entered the extensions part of the Schein trial and 122 completed the open label extension phase. Now this is roughly 70% of all patients that started the trial, which speaks for that a lot of patients finishing the 9 month treatment and even the 1 month observation.
Again, today, we're looking at the end of treatment at week 40, which was encompassing 116 visits. So you see that 6 patients skipped this visit or missed it or had a later last visit for EOT. We discontinued 34 patients in the open label extension part of the SHINE study. So let's jump to the next page, which is Page 5. So just to remind you of the SHINE study details.
The main period that we already reported on had 5 groups, a placebo group and 4 dosing groups, which we called minimal low, medium and high dose. And it was a 16 week double blind placebo controlled study. At week 16, we then looked at all patients regardless of their initial group and stratified them in either patients that had a high score response on week 16 or didn't have a high score response. Now the responders the high score responders went into the responder group and were continued with IFX low dose, IFX-one low dose treatment, which is 800 milligram every 4 weeks. The nonresponders transitioned to the non responder group receiving IFX-one medium dose, which is 800 milligram every other week.
Just to remind you, the open label extension phase key goals were for the high score responders to determine if a maintenance could be achieved with a low dose IFX-one therapy and for the high score non responders to determine whether transitioning them to a medium dose of IFRS 1 would make them become responders. Just to remind you, the medium dose was calculated by us as 800 milligram every other week to be efficacious. Now we have recently reported on a postdoc analysis that the 1200 milligram group, which was a high dose group in the initial period of the trial, was in fact clearly the most efficacious group. Important note here is we did not retrospectively unblind patients. That means patients transitioning into the open label extension period did not know in which group they had been before transitioning.
So please let's go to the next slide on Page 6 together. So this answers the key goals of the open label extension phase. So at week 16, all the responders, which is the blue dotted line here, were set to 100%. And you see that throughout the additional weeks of dosing, 70% or over 70%, in fact, remain high score responders. The non responders, 84 patients here, that were set on 800 milligram biweekly dosing gained response up to 42% at the end of the 40 week treatment period.
Now this makes the overall response rate at 9 months of being in the SHINE study 56.3%. But please take a close look to the confidence interval, the 95% confidence interval that we provided here. So with this being 116 patients, we felt that this is relatively meaningful. And we have reported on the high score, and we have also mentioned the limitations of the high score with a very high end count variability and an extremely high variability of this score. So we have reported recently that we had quite a substantial effect, especially on the high dose group, on the overall inflammatory lesion count.
So that includes draining fistula, which are not captured by the high school. So when we now progress to the next slide, let's first look at the question, how did the overall patient population that completed the study, remember, roughly 70% of those that even started it, how did this 70% of patients perform when we're looking at their lesions? So please, let's go to the next slide, Page 7. So I want to walk you through the slide, and please focus on the left side of this slide first with me. Now what we did here is we looked at all the patients that finished the 40 week treatment, and then we compared them to the baseline of the open label extension population, which was 156 patients that entered the OLE.
So when we say baseline, we mean day 1 of the SHINE study. So what you can see here is that the excess nodule count, the draining fistula but also the related score, IHS IV score and the ANF count, which just counts all lesions together, are remarkably reduced at the end of this volume week treatment period. These are the mean values. I didn't show the median values, but you can add approximately 10% to each of these parts when you look at the media. So it's a remarkable response, and we ask ourselves the question, can that happen by chance?
Or how does it happen? How does it compare the initially very well performing placebo group that showed a 47.2% response in the high score? So this very placebo group is depicted here on the right side. Look at this for a second and look at the different counts. And this is the same measure here as the mean.
And remember, that group had a 47.2% high score response. But when you look at the lesions, in mean the end count went down only 26.5%, and you see the other numbers. So this was a well performing placebo group, but this is not nearly comparable to the majority of the patients at the end of our treatment. So we feel this is remarkable, and we are very happy that we can share this with you today. So one question would be now how does this how do these scores or how do these lesions evolve over time?
How are they reduced over time? And instead of going through all these scores and all the lesions, we took the end count to AN count to give you one example on the next slide. So please move to Slide 8 with me. Slide 8 shows you those 156 patients that entered the open label extension study and retrospectively looked at them how did they do in the main period altogether irrespective of in which treatment group they were and how did these overall OLE patients behave in the long term extension study. This is the end count.
You see that there is remarkable reduction in the main part of the SHINE study. But then there's a continuous decline of the end count. And this is the total count here, and we're coming from around 12 counts down to below 4. Again, if I display the median, this becomes even much more impressive. So we have talked about the IHS IV score in our last press release because of 2 main reasons.
We have seen that our drug, especially at the high dose, significantly reduced the lesions, the inflammatory lesions that really matter. That includes the trimming system. And we looked at the IHS-four score because it's an established score not invented by us to capture the lesions, and it has certain advantages and was developed for a certain purpose. So let's take a brief look at this IHS IV score together on Page 9. So what this score really does, it looks at each lesion, the inflammatory nodules, the abscesses and the draining fistulas, and it gives them a score.
1 nodule gets 1 score, 1 abscess gets 2 scores and one draining fistula gets 4 scores. Now originally, this was developed by KOLs and physicians to have a measure of severity but also to be suitable for tracking treatment response. Now I mentioned already that it does capture draining fistulas unlike the high score. But what it really does as an advantage is that it gives less weight to the most fluctuating lesions, which are the inflammatory nodules. And it gets more weight on the abscesses and even more weight on the draining fistulas that don't fluctuate nearly as much.
So by looking at all lesions, this score has the advantage that it has overall less of a variability when you compare it to the very high variability of the high score, which unfortunately was previously not published, so that the estimates going into our study when it comes to the group size were not adjusted for this extreme variability. We already did internal validation work, and that's important because this work shows that this IHS IV score in the overall data set of the SHINE study correlates nicely with the DLQI and with the pain score in a statistically significant manner. So let's look at the performance here as one scoring system that looks at all lesions in the open label extension part. So please move to Page 10 with me. Again, this is the split between high score, non responders and responders.
This time, it's upside down because the responder you find here on the bottom. So these are the 72 patients that responded. Please keep in mind as we move through this and the next two slides here that when we say responders, we mean high score responders on week 16. We are looking here at their IHS IV scores. So the responders, interestingly, with a low dose treatment of IFRS 1, largely remained a good reduction.
I mean, we're coming from around 28, 30 points in the overall treatment group down below 10. So there's a slight decrease of improvement, I would call it, but overall, relatively good maintenance. Now the non responders that still displayed an overall score in this IHS IV score of close to 30, they had a continuous reduction of below 20. Now we come to the most typical question, how does this reflect the initial treatment response? Or in other words, how did the original dosing groups do when you are looking into their performance in the IHS IV score?
So as we move together to the next slide, what we have done in this slide, which is Page 11, is a little bit tricky, so I need to guide you through that. Again, this is IHS IV scores. And the slide you see here is all non responders. Remember, nonresponders means week 16 high score nonresponders. And we ask ourselves the question, how did they perform at week 16?
And you see the different bar graphs of the different groups. So you see, for example, with the first bar graph that the placebo group, actually the nonresponder on highscore, the placebo group, when looking at IHS IV, showed a worsening, which is reasonable. So it had a 20% worsening of the inflammatory lesion score, IHS IV score. And the minimal dose group had a slight worsening as well. Now interestingly, when you see what happened to these two groups after 800 milligram every other week dosing until week 40.
These patients that had never seen our drug, and again, they were not unblinded for their form of being placebo or minimal dose, had a remarkable response because from having a worsening, they went to having a 20% to 25% improvement on the score. You see that still effects exist for the next two dose groups because the 800 milligram 4 weekly got a little bit higher dose and the 800 milligram biweekly was basically continued. But the only patient population that didn't have an additional effect was the 1200 minuteimum group, the high dose group. So you might speculate whether that means that this group selected 2 non responders in a way that these patients clearly did not respond to the drug. And you see that because over time, they don't improve.
Now the next question obviously is these are the nonresponders. So let's look at the same slide by just displaying the responders, which is then Page 12, please. And same metrics, same idea, but these are all high score responders on week 16. And you see, of course, this select patients that do relatively well on the IHS IV score. These are very low patient numbers.
But of course, one measure here is relatively obvious that there is I wouldn't call it a full maintenance, but there's a decent maintenance of this improvement and maybe a slight decline depending on the group. So with this, I would like to move to the next slide, Page 13, and give you, last but not least, some learnings from the extensive work that we have done on TKPD modeling and also PBPK modeling, which is called physiological pharmacokinetic modeling. So what we have learned here is that IFX1 consumption in HS is much higher than in other diseases. And we learned that by looking at the trough levels over time. And we have compared it with similar dosing in other studies.
And the interesting finding is that at the same dosing, you can have a multiple lower trough level. We are talking here Factor IV lower trough level when compared to the other disease. And that means IFRS 1 is more consumed. Now how come? So we believe that the results clearly indicate to us that this consumption is largely driven by a very high C5a turnover rate.
And why do we think so? Because the modeling shows that we have a so called target mediated drug clearance or drug deposition. That means the more C5a target is produced, the more IFX-one binds to it and gets cleared as complex out of the system. The model suggests also that IFX-one achieves a good tissue penetration rate, especially for the higher doses. This is at least in one aspect very important because the one thing that we really see from all the data that the relatively strong effect on reduction of draining fistulas that we detected on the high dose and which we have seen before in the initial Phase IIa study in Athens and Greece was only seen at the very high dose.
So we believe there's a special link to that because we see still a n count reduction at lower levels. So this is the learnings. I think that's very well. We believe we can better explain what happened to the placebo group, and we have good information on what doses are required for maximum effect. With that, I would like to conclude here on the last slide for the data.
So the conclusions and key takeaways are on Page 14. The long term treatment with IFRS 1 leads to a market improvement of inflammatory lesions, and that is remarkable for HS patients for a long period of time, 9 months treatment time. Highscore responders maintained their response over time over 70%. Well, that was probably not the optimal dose for maintenance, but it still showed a good maintenance. And that's a good news for us.
Now placebo and minimal dose group patients that were non responders in the high score and transitioned then into what we believe is still a somewhat efficacious dose, 8 100 milligram every other week, showed a marked improvement in their patients. So I haven't shared you details on safety, but I wanted to at least conclude that it was Alexfron was well tolerated and we had not a single drug related SAE in the entire open label extension study. With that, you hear our enthusiasm about the data. I'm coming to the facts of the financials. And I would like to hand over to Arnd, then we skip to the next slide, Page 15.
Arnd, please.
Yes. Thank you very much, Niels. I'm also very pleased to speak with you today regarding the company financials. And for purpose of this call, I will provide the most important key financial figures for the 1st 3 quarters of 2019 summarized on Page 16. First, the company's total operating expenses during this period reached €43,000,000 after €25,200,000 in the previous year.
The expenses for research and development amounted to €33,600,000 versus €60,000,000 in the previous year. This increase is primarily attributable to an increase of CRO expenses associated with preclinical studies and clinical trials that we conducted with IFRS 1 as well as an increase in expenses for our manufacturing efforts, mainly due with or in connection with IFRS 1. G and A expenses for the 1st 9 months of 2019 slightly increased from €9,200,000 for the 1st 9 months to €9,300,000 in 2019. Our loss for the periods for Q1 until Q3 2019 amounted to €39,600,000 compared to a loss of €19,600,000 for the same period 2018. The earnings per share were minus €1.53 for the 9 months 2019 compared to minus 0 point €79 for the previous year.
Importantly, I would like to provide an overview of the company's capital situation. For the 1st 9 months of 2019, we spent €27,000,000 for operating activities compared to €15,200,000 for the same period in 2018. Together with our marketable securities valued at €108,500,000 we had €135,500,000 cash and cash equivalents and marketable securities as of end of September 2019 compared to €162,100,000 as of end of September 2018. Please let me conclude this overview of our financials with the statement that the company continues to run operations very efficiently with a lean organization. Combined with a strong balance sheet, the company has a 3 year runway, allowing us to reach multiple value inflection points within our cash reach.
With this, I conclude my presentation of the financials for the 1st 9 months, and I would like to hand over to Niels. Niels?
Yes. Thank you, Arnd. Thanks for this comprehensive overview. And I would like to conclude now with Slide 17, the strategy update. So as you could hear and see from today's talk, we are very much convinced that we have an interesting technology that holds value.
So we have clearly decided to continue development of our anti CVA technology. We are developing it in the current and also new indications. We have just recently discussed and decided to enlarge the RUNNING trial in pauderma gangrenozone, which is an interesting rare disease, inflammatory disease with a very high unmet medical need and no current approved treatment paradigm. One of the reasons behind this was clearly that we saw in our data and got conviction that we have a very good activity of our drug in the skin. And we also have built recently a very strong IP situation around usage of our drug and actually even the pathway in inflammatory skin disease or neutrophil driven skin diseases.
So we are very confident, and this shows that we are enlarging the study to 18 patients from POMENT12. We also initiate clinical proof of concept trial in oncology in 2020. We have spent now about 1.5 years of serious efforts into building up an oncology indication for our drug, which is based on a lot of literature on the role of C5a in cancer immunology, but also on some in house data. And we will announce this indication as we get close to it in 2020. With the today presented positive open label extension results, we, of course, want to evaluate options for further development in HS.
That means for now that we will discuss the data and next steps with the regulatory authorities, and this will also include serious discussions on potential alternative endpoints. So one thing that was anyway part of our discussions before in our strategy, which we are now forcing or focusing on, is broadening the R and D pipeline beyond the anti CDW technology. That's part of our diversification strategy. So the focus will remain on rare inflammatory diseases with high unmet medical needs and the defined oncology space. And we have just very recently hired a very experienced Head of Global Business and Development and Strategy with a strong pharmaceutical and biotech background in the U.
S. Who will help us foster this diversification strategy and who will be on board with us still within this month. So to conclude the overall overview of today, I would like to say that the company has sufficient financial resources to carry out this strategy and to reach multiple value inflection points within the next 3 years. With that, I'm very happy to conclude today's call, and I'm also happy to move over now to Q and A and give back to the operator.
We will take our first question from Yatin Sonijia from Guggenheim Partners.
Good morning.
Perfect. Yes, perfect. Good morning. Thanks for taking my question. Neil, again, a lot of details, very encouraging.
Can you maybe put these results in context to the published long term data that we have on Humira? I think it is our understanding that half of the patients that get treated with Humira start losing response over time. But in your case, these responses seem pretty durable. So just help us understand or comment or maybe put those results in perspective? And then I do have 2 more follow
ups. Yes. Thanks for this question, Jatin. Obviously, this is a very interesting question that we also already asked ourselves because when you're looking at the PIONEER data, which are the only published long term data to our knowledge, you're right that about 47 percent was the rate that they could keep it to, so slightly below 50%. But that was only 8 months, so that was a shorter treatment period.
And we hear probably with a suboptimal maintenance dose, we don't know that yet, but this is how we feel about it, we're able to keep 70% response. Now I think those trials are not fully comparable because PIONEER study had also placebo arm in the long term study, and it's always difficult to compare trial to trial. But I think what is fair to say, and this is what I would like the audience also to focus on, is that we have here a large amount of patients, close to 70% who started the trial, that have a really, really meaningful improvement in their overall lesion. And you mentioned it, Jatin, it's durable. Durability is one of the things that people look for.
They don't want to be told they're great at 3 months, but with a 50%, sixty percent chance, they are not great anymore 3 or 4 months later. So I think despite the fact that the comparability is always a little difficult, I think that these results are very meaningful. And I would like to also share that we have vetted these results with 2 HS Foundation presidents and with some of the most well known KOLs in the field. And they fully confirmed our view on these results, which made us feel strong, and this also led to the decision that we will further evaluate this indication.
Got it. And then in terms of the non responders, could you maybe perhaps tell us like how big of the pool was from the placebo group? And if there was any difference in how the placebo responded versus the patient who were actually on your drug and they were nonresponded and then they responded. Can you comment on that if there was any meaningful difference?
Absolutely. I mean, this is kind of the problem when you come to very small patient numbers. So you know that the placebo response rate at week 16 was 47%. So you could say roughly 50% were non responders, a little bit over 50% were nonresponders, and we had 36 patients. So we're really talking about 18, 19 patients, 16, 17, 18, 19 patients per group here that we showed.
So that is, to look at high scores, certainly an extremely low group, considering that at 18 patients, you have 95% confidence in the value ranging up to 50%. So this was also part of the problem we had that we didn't know the extreme variability in the end count. But we're talking roughly between 16, 18 patients here per group. So it's a very small number when you go to the former treatment groups and split them to responders and nonresponders and high score.
Got it. And then I remember in the 1st 16 week, I think there were some differences in how the patient responded between 10 14 weeks on drainage fistula. Did you see any sort of deviation or how was the drainage fistula over time? Was it more consistent over time from here?
Yes. Thanks for that excellent question. So the variability you see, I mean, the end count, we haven't even depicted that yet, but the draining fistula you saw is, especially for the end count, is enormous. For the draining fistula, there is still a high variability. Now here in this open label extension study, we're looking at 72 patients that are responders on Highscore and 84 that are nonresponders.
So this gives you already a much better feeling and a much lower smaller confidence intervals. Remember the first slide that I showed, there were the 95% confidence intervals on the high score. So the variability of the fistulas is largely gone by increasing the numbers. So we don't know how much of the variability we saw in the first period, which we published and that you just mentioned, is due to the fact that this could happen by chance or if it's really like a short weakening of the signals in these 1st weeks. We have commented on that before that this happens in other diseases as you think of psoriasis, where you also have a certain waxing and weaning while the patients get better until this slows down.
And we don't know if this is part of it, but I can certainly share with you that the variability the only way you can get a good handle on it is increasing the patient numbers, and that is very much reflected in the open label extension arm.
Got it. And I promise final question. Can you maybe comment on the regulatory interaction, what the next steps are in terms of meeting either with the FDA? Have you met it? What sort of feedback do you expect or you have received?
Thanks.
Yes. I mean, this is part of our strategy. Thanks for this, Jatin. This is part of our strategy to go to the regulatory authorities. We have not yet done that because, of course, we need this data yet in hand to do that.
And we also want to prepare this well. We have gotten excellent advice from various parties, including former FDA officials that were specified in dermatology. So we want to get well prepared. We can't tell you yet what the outcome is. I mean, obviously, you see we have an effect long term, a very interesting effect on highscore.
And probably, we would have seen that also earlier if we had a much higher patient number, but that's speculation. But our drug really does more than just a n count reduction, and that's what we're trying to argue. And now that we have evolved our understanding of the disease, I think there's also a lot of other arguments. So we're preparing that. It will take a few months to get this scheduled and done, and we certainly will share the outcome of this with the public.
Very good. Thank you so much.
Sure. Thank you, Anthony.
We will now take our next question from Anupam Rama from JPMorgan.
Hi, all. Good morning. This is Tessa calling in for Anupam. Thank you for taking our questions and for the updates here. A couple of million were already taken, but I guess I'll start with, can you remind us if you've talked with regulators about the IHS score before.
And just remind us of kind of what their feedback was or if those conversations have not been taken place yet, just kind of update us there. And then I guess that's my first question.
Okay. Sure, Ted. Thanks so much for tuning in, and thanks for your question. Yes, your question was, did we talk to regulators before on this quarter? No, we did not.
But I would like to mention that, first of all, this is not our only focus, right? I mean, another score that equally reflects it is just looking at ANF counts, just counting all lesions, and you see the same picture. So we are not necessarily fixed on one score. That's one thing. But the other thing I would also like to mention, this is not a score that we pulled out.
It's partially validated already, has been published by the Foundation Society and with the senior authorship of the Foundation President. There's a whole yes, I would say, a clear majority of the QOLs is behind the score like the IHS IV score. And again, they did a large undertaking to find a score that looks at all lesions. And the reason why they came out with the IHS IV score is that upon modeling this and looking at all the statistics, this kind of weighting that I was alluding to, 1, 2 and 4 points for nodules, abscesses and fistulas, it comes out as best score with lowest variability, with which is also best correlating with the improvement of the patient. So it's nothing we invented.
There's also a decent amount of publications already existing, and that's why we feel it's a natural FDA to talk about this score.
Okay. That's very helpful. That's very helpful. And then I guess, Neal, just to remind us on germinating fistulas, specifically, can you remind us of what improvement has been shown with HUMIRA? And just remind us of that data and perhaps like the inflammatory lesion count overall as well?
Sure. I mean, as far as my knowledge goes, in the original publications in the New England Journal of the PIONEER trials, there was not specific focus on the lesions per se, unfortunately. Otherwise, we could have calculated the end count variability and understood the implications for powering the trial. And the other thing is I don't think that there was any focus on draining fistula. Now I do believe that in the documents that were handed in to the U.
K. Authority for approval, there are some numbers or mean numbers, which in size were not comparable to what we've seen here at all. But I don't remember any concrete publication or anything that came out of these trials that really depicted a clear effect on draining fistulas.
Great. Nino, thanks so much for taking our questions.
Sure. Thank you. Thanks, Ted.
Our next question comes from Joseph Schwartz, SVB Leerink. Please go
This is Dae Gon dialing in for Joe and congrats on the progress and thanks for providing more granularity on this data. So I just have a couple. So maybe the first question I'll start with. You mentioned in your press release and today's prepared remarks that the IFX1 consumption rate was considerably higher in HS compared to other diseases. So actually, it's a 2 part question, but just wondering, can you expand on that a little bit?
What diseases were you specifically referring to as a reference point to HS? And then secondarily to that point is leading up to the SHINE top line data in May, the 16 week data, we had a lot of investor concerns regarding a paper that came out showing sort of a lack of correlation, if you will, on the C5a count and disease severity or Hurley Stage 2 and 3. So maybe now that you have more data, can you maybe help us reconcile that correlation? And I've got a couple of follow ups. Thanks.
Okay. So maybe I'll start with the latter first. So that paper that I think was an abstract, but I'm not also a small paper published on the C5a. These were very low numbers. And as you alluded to, the variability from patient to patient on C5a plasma levels is quite enormous.
Now this was a very early work when C5a levels were correlated in the patients to certain lesion count or to certain scores. So with the inter patient variability, with the numbers published and also with the retrospective way that this was conducted, I believe that the conclusions from this are taken had to be taken with care. Now what we have done is something very different. So while it's I believe it's very difficult to say when you have this amount of C5a, then I can guide your therapy and you will profit or benefit from me putting it to that amount. For that, in the complement system, that's way too complex, and it's way too many factors into that because the receptor density something you don't measure, the amount of neutrophils around us you don't measure, etcetera, etcetera.
So this correlation is going to be very, very difficult. But what we are doing here is quite different. We are looking at trough levels, and we are looking at, of course, also C5a levels across time. And what I have referred to in my and that was your first question in our publication when it comes to the question, how much consumption do you have? We compared it to non disclosed but other life threatening diseases where we believe we should see also a high turnover rate.
And I can while I would not disclose today which indications these are, I would like to let you know that the factor is considerable. We're talking about Factor 4. So with all of our knowledge, and I mentioned that during this call today, that 800 milligram every other week was firmly believed to be a fully efficacious dose. In other diseases, you see a super control with this dose. In this disease, you don't.
So there is something very special about HS, and I think it's a large amount of C5a turnover that drives this. And when you that's my last aspect here, Dae Gon, which I think is important. When you look at the other drugs being studied, but especially at Humira being approved, it's approved at double dose. And every KOL that we went into said like, take the double dose, take higher than you think you should. We know that anything that should help in HS has to be dosed as high as possible.
So while this is not scientific, clearly, I believe there is something to it, and our data show that.
Great. Well, thank you very much for that. And so just if we look into this a little more, perhaps with regards to the IHS IV score, which had been previously asked, just wanted to get your take on I know you've got meetings planned and we'll get an update on that. Your previous call when we were discussing the post doc analysis of the 16 week data, you mentioned IHS 4 score being partially validated. You spoke today about the foundation having done a substantial amount of work on this.
And I think in the previous call, you mentioned that in the, I guess, regulatory documents of HUMIRA, there were also some discussion about AN count and the FDA might want some draining fistula component included. So I guess if we think about going forward in the regulatory discussions, what do you think are the main sticking points if you do propose IHS-four as your primary endpoint going forward? And I guess what are some of the education that needs to be taken place in order to make sure physicians are also aware of comparability, if will, between HUMIRA and IFRS 1?
Yes. Excellent question. Thanks so much, Dione. I appreciate that. So let me first start by the question whether Highscore is the right score to capture it's a binary score to capture benefit for the patient.
That's when you just go with a statistical approach and you validate it by saying, is there a correlation, which we have done here also for IHS IV, is there a correlation between high score and benefit like DAQI, pain scores, and they all have like limitations. But yes, you will find a correlation, and we find that too also for the high score in our data set. So there's a correlation. But I give you a good example of a patient that was introduced by one of our investigators, which was actually one of his the sickest patients he has ever seen in his life, in his career as dermatologist with over 50 lesion count in this disease. He had a second disease, which we are also studying, very interesting.
And this patient had, I think, a lesion count like roughly over 50. Now I'm talking lesion count, not IHS IV score. And this was reduced by something like 27@week16, and she had the most remarkable impact on that disease of the last several years, an impact that no one and she had tried any type of therapy before and an impact that he didn't expect to be able to be happening to her anymore. But she was a nonresponder. And this shows you and I can give you ample other evidence, like what if you had a patient with 20 lesions, you reduce it to 0, but one draining fistula non draining fistula is now when you press on it giving you a drop off pass, which is now making it a draining fistula.
So that patient has a remarkable benefit clinically and experiences the benefit, but you don't really capture that in the high score. So I think this is something that most KOLs clearly support and that the FDA needs to hear and probably knows already about. But another good argument for us is besides the fact that a binary score has limitations and oftentimes leaves benefit uncaptured, besides that, I think a really important argument is also that all lesions matter to these patients. It's a very complex disease and a disease with great suffering. And it's very hard to say for an individual patient, you only have benefit if your AAN count is reduced by 50%.
Give you another example. You go to an imaginary late stage patient who has 15 draining fistulas and 1 abscess and 2 nodules, right? So or let's say, 1 abscess and 3 nodules. So now the patient has 1 abscess and 1 nodule. So 50% reduction of the end count, but he still has 15 draining fistulas.
So to say this is meaningful for him is difficult. And there are other patients that have a lot of excess nodules and maybe little or no draining fistulas that greatly benefit from an AM count reduction. So long story short, you want to make sure that you capture benefit at all lesion levels. And leaving out the drilling fistulas wasn't liked by the FDA in the 1st place, and you mentioned the documents. So they were okay with taking that one note in there that they can't increase draining fistulas.
But in fact, we should count all lesions and look at benefits at all inflammatory lesion levels. This is not only our view. This is really the view of most of the KOLs we talked to.
Great. Thank you very much. Last question for me is going forward, this OLE data as great as it is, I think people are still kind of thinking about your placebo controlled 16 week portion, specifically the 47% response rate. I know the IHS IV data on the placebo side looked dramatically lower, but any strategies or implementations that you're thinking of to make sure that the placebo response, the abnormal placebo response does not surface if you do go into your next study?
Yes. This is really right on this question, and I thank you for that. And we did a lot of work into looking into this. So the strategy we have is very clear. You cannot power a trial with 36 patients per group.
When you look at the AN count variability, at the high score variability and the confidence intervals overspending 40% with 36 patients, it's very clear that our high score response of 47 could have equally meant 30 or 60. So you can only control this problem by taking large amount patient cohorts. And our statisticians say you need to have at least 80 to 100, maybe even better, more than 100 patients to get this variability under control through the patient size. So powering a trial with 36 patients, knowing what we know today about the high score, you may get really unlucky. The other side of it is, and that is something we have reported before, that the trial had two points that may have artificially also helped placebo patients.
1 is the four-one randomization. So patients knew that in all likelihood, they would get drug. There was somewhat a positive drug buzz around our drug in some centers, obviously. So that may have interfered. And the other one that so we would abstain from that.
We would clearly say, if we go into another study, we would like to choose one dose. We have good information about the right dose, but certainly not more than 2 doses and a placebo arm. And we power it with enough patients. And the last point here is, obviously, our trial had a lot of touch points. So every 2 week IV infusion, getting wounds cleaned, getting taken care of may have an effect such that you're not actually testing against standard of care, but maybe, if you will, against intensified standard of care.
These are the 2 most obvious reasons we could find. But I would like to take this opportunity of your question to add one important point here. We have really spent a tremendous effort to understand whether we had any quality issues with the trial. And I would like to mention that these conclusions we take here are pretty definitive for us because we could not find any quality problems, Patient selection, training of the physicians, site selection was very similar to the PIONEER trials. The inclusion, exclusion criteria were practically the same.
And that is also evidenced by the fact that across all five dose groups, we have baseline characteristics which beautifully fit to the PIONEER studies. So I think we can exclude that there was a problem with this trial. And we have learned a lot, and I think we can now understand and now avoid a problem with the placebo high score response in the future.
Awesome. Thanks very much for taking the questions, and good luck.
Thanks so much. Much appreciate,
Yigal. We will now take our next question and this comes from Matthew Luchini from BMO Capital.
Hi, thanks. Good morning. Thanks for taking the questions. 2 from me. So just I guess coming on that the discussion of dose, I just would like to get a little bit more on how you feel the open label data impacts your thoughts around proper dosing of IFRS 1 if you were to redo another clinical trial?
Do you feel like at this point you've got a clear dosing strategy for moving forward? Would it be just 2 doses as you sort of just touched upon? Or do you feel like multi doses would still be necessary at this point? And then the second question is just, and if you said this and I missed it, my apologies. Could you just let us know if any patients needed any kind of antibiotics for flares during the open label extension portion of the study?
Thank you.
Yes. Thanks, Matt. Thanks so much for the question. So I think on the antibiotic question, take the last first here last question first. I think antibiotic is the standard of care.
I think trials that exclude patients from having access to antibiotics in general is becoming more and more problematic. The question that we have alluded to before, there's a specific question to certain antibiotics that are used for base treatment of the disease. These are certain tetracycline like drugs that are used like doxycycline and others. And some patients are stably treated on them. And there was always the debate.
And we allowed those patients to be on this treatment if they had been stable on the treatment for a period of time before. And the reason we did that is because the KOLs advised us if you take those patients off these kind of also anti inflammatory working drugs, which don't cure the disease, but they keep them a little bit more stable. So if you take those drugs off, those patients have a clear tendency to become more fluctuating vaccine weaning with their disease. And there was a large debate whether that could influence results. This was also allowed in the PIONEER 2 study.
But the number of patients that really qualified for that had really no impact on our results. So my take here is that this is not a crucial question for this disease in my eyes anymore. The next question was like how do we think about future dosing? I think we have really a very good understanding. Remember, we did a small study open label before with 800 milligram weekly dosing, which we also used for modeling.
And I think we have a very good understanding. While I cannot give you a definitive answer today whether we do 1 or 2 doses or whether what type of trial we will do in the future, obviously, a lot of this will depend on our discussions with the regulatory bodies. But while we can't give you a definitive answer, the one thing that I can mention and share with you clearly is that we have a very good understanding about an effective dose. The other question you had is how do the open label extension data feed into this? And what do we think?
That's a very interesting question because honestly, what we didn't think is that we could maintain a response largely with only 800 milligram every 4 weeks. So if you will, that part of the questions was relatively successful because I mentioned today and Jatin asked the question, how does this compare to other data published, especially the PALMUNE data. And I think it compares pretty well. So that lets us think that if we were to ever do another study here, which is an interesting thought, obviously, whether or not we would still do an open label extension or at least an extension phase where we test different maintenance therapies. This is all thinking out loud.
Obviously, none of this is yet decided, but I thought I'd give you as much color as I can.
Great. And one last one. Can you just talk what are your where are you in terms of publication planning for both SHINE and I guess ultimately as well the Phase IIa study, which unless I'm mistaken has not yet been published either?
That has that is true. That study is handed in for publication. It's pending. And we have decided that we will publish the results of this study here. Obviously, we need to have database lock and the whole study set complete.
But we want to publish the results as a whole, so as one part, not separating the main part from the other part. So that will happen then early next year. We'll have the complete data set after data cleaning and everything complete. And at that time, we can hand in a publication. So I think with our investigators now being very excited about this data here, we have a lot of support, and we will try to manage the publication here of the Phase IIb in a much smoother way than it did with the Phase IIa.
Great. Thank you for taking the questions.
Sure. Thank you. Thanks, Matt.
Ladies and gentlemen, we will now take our next question from Adhukumar, OR W. Baird. Please go ahead.
Yes. Hi, everyone. This is Jennifer Tubal on for Madhu. Thanks for taking my question. I just had two quick questions.
1, sort of an overview. You mentioned that you were planning on talking to regulators in a few months. And I'm just wondering that if they sort of give you the go ahead and you decide to do another trial, what the timing of that might look like? And then sort of more granular, of that non responder group, could you give us an idea of how many of those received a different dosing schedule in the open label trial as compared to the main period?
Yes. So thanks, Jennifer. With the first for your first question, as I mentioned today, we are planning to go to the regulators here with this data and discuss details. So it's very hard for us to speculate now what a future study may look like. But I do think the regulators will likely acknowledge that there was a long term effect on these patients, which is beyond doubt present.
And then I think we will take it from there. We will prepare this well. Again, it takes a few months. As you know, we have to ask for a meeting with this data set now being at hand and being analyzed. And then a few months down the road, we will have the actual meeting, and then we will share that with the public.
And then hopefully, at that point, we can get more clarity and give you more clarity how we envision the next phase of development. So I think that's as far as I can go on this one. And with respect to your second question, you asked how many if I could give you more granularity on how many patients per group. I mentioned that earlier on the call already, the placebo high score response rate was 47%. And you've seen the other groups that were between, I think, roughly 40% 50%.
So you can roughly say for each group, non responders and responders, there was a little bit more non responders, which is also reflected in the overall population, 84 versus 72. So each group has roughly 14, 15, 16 patients, sometimes 17 patients as responders and the same amount more a little bit more towards 14, 15 patients in the nonresponding group, just as a rough estimate. So that's very low numbers, and they're taken with care. I mentioned the variability of these scores. And that's why we feel it's very meaningful to look at the overall patients that the 72 and 84 patients and the 116 that finished.
Okay, great. Thanks.
Sure. We will take our next question from Steve Seedhouse from Raymond James.
Yes. Hi. This is Timur Ivankov on for Steve Seedhouse. And thanks for the update today. And Neel, I guess just to clarify your conclusion here.
I guess are you suggesting your trial could have been underpowered or the patients were under dosed or both? And just to make sure, are you open to running another Phase II study? Thank you.
Yes. So what I clearly suggest is if you want to show a difference here, you cannot do that reliably with 36 patients. So yes, to show a meaningful difference, you as I mentioned, on the high score, like that is a really rough estimate. You need a lot more patients, maybe 80, 100 per group at least. And that's one conclusion.
Were we underdosed? This is not the question here because we had 4 dose groups, right? And I shared with you that the high dose group clearly showed efficacy on the lesion reduction. We had mentioned the problem with the high score. I think we can explain it well now.
And if you just look at the individual lesion counts, like take the AN count, you have patients that go from 18 to 7, back to 15 to 9, from visit to visit each 2 weeks. So it becomes really clear with this variability, which by the way, we looked at looked into whether that is caused by a change in the examiner, and we can confirm that our analysis clearly shows that this is independent of whether there is an examiner change or it's always the same examiner. So it is also, according to our experts, true that these patients jumped at these lesions. Not all, but many. So yes, to show meaningful difference in the high score, this trial was underpowered.
To show meaningful lesion reduction, we showed it. And I think with the variability at hand, it showed a clear and statistically significant effect on the high dose group. And I discussed this a lot with our statisticians. And I want to mention one thing. There is a difference between post hoc and post hoc.
We have not included the IHS IV score in the secondary endpoints. That is true. But the trial, and I want to remind you of this, was designed and conducted in a high quality manner to capture inflammatory lesion reduction, which is what you have to do for the high score. So I think it's fully valid to look at lesions. And you can take the ANF count as well, and we showed you the results there, too.
So I think this is meaningful what we have. I think we know that with 36 patients per group, it's a huge risk to power trial with that on the high score. And I think we are very well informed about any potential future trial. Now I can't tell you whether there is another Phase II or Phase III trial. This is certainly nothing that we can speculate on.
And it's because it very much depends on our interactions with the regulators.
Okay. Thank you very much for that. And just a couple of quick questions on patient disposition. So I think there's 21 patients who did not enter the open label extension. And you mentioned patients were unblinded early in the open label extension?
What was the procedure for unblinding the patients?
No, I sorry, maybe there was a misunderstanding. We did not unblind the patients. The patients did not know in which group they were before. They were not unblinded, very importantly. So the patients transitioned to either the maintenance therapy low dose or the medium dose therapy for the nonresponders, and they were not unblinded to their initial treatment.
So until the end, they didn't know whether they started as placebo, lower, medium dose or high dose patient, which was very important for this trial.
Okay. And the reason why the 21 patients did not go into the extension is that what was was there a common theme there?
Look, we will share details on this when we have the final data set, but I can already share that there was not a common scheme. Of course, you have patients that move away, that discontinue, can't afford to travel anymore. And then you have patients that don't feel it's worth the time, then you have patients that don't feel an improvement. And then you have also some patients where you have a flare. But that was very, very rare.
In fact, we already published that at week 16, there was a clear signal for flare reduction in the high dose group. So I think it's a normal wide array of why patients transition out. But certainly, there's always a few that and a reasonable amount of patients that also transition out because they're not happy. That's normal.
Okay. And then I guess the final question is, so regarding the response rates in the open label extensions. So if a patient responded but didn't complete the study, how were the patients treated in your final in the data analysis you presented today?
Yes. So it's mentioned always on the bottom of the slides. For almost all slides, we did fast analysis. For the detailed analysis for the groups, we did last observation carried forward. But I can share with you that we have done numerous and different sensitivity analysis, including last observation carried forward, including analysis, which is a non responder imputation and the full data set, and it doesn't change the picture.
So I think that's what you're probably alluding to. But here, you have most of this is the full analysis set, And the ones where you see the bar graphs of the different groups is last observation carry forward. And that is a specific reason. If you go to low numbers and you don't do that, you see even broader effects, but they're really due to a couple of patients missing one day or the other. And that's what we try to eliminate in this picture.
Okay. Thanks for the update.
Ladies and gentlemen, our
and thank you for taking the question and squeezing it in at the end. I just have 2 really quick ones, make sure I didn't miss anything. The first is, have you done any kind of biomarker analysis or patient subsetting? Anything that would maybe give you an idea of who is and is not likely to respond? I guess I'm thinking in particular, say, SHINE non responders who also non responded in the open label extension that may be able to eliminate them from a future study to bias things in your favor?
Yes. Derek, thanks. Excellent question. We are working on biomarker. We have found a very interesting biomarker that looks like it tracks very well responders versus nonresponders when treated.
But your question, if we conclude this, we may even publish on that one as well. But your question alluded more towards the fact that do we have a biomarker that would tell us upfront, can we predict who will respond or not. This analysis has not been conducted yet. It's extremely difficult to find such biomarkers because in order to know whether that is true or not, you have to do a prospective trial. I can speak a little bit to biomarker development because I did that for years in the sepsis field.
But I wish we would have that. But we have a very active group with Ren Feng here, our cofounder and CSO, looking into this right now. So that analysis would take a little longer. If we find something there, for sure, we'll share it with you. But at this point in time, we don't have this data
Got it. That's very helpful. Thank you. And then just the last question. I'm looking at Slide 12 on the deck that you guys put out today.
And I'm just looking at the placebo transition and the guys who look like they were IHS-four responders under placebo and then afterwards. I just want to make sure I'm looking at this right. So is this indicating you had a decent number of guys who were IHS score responding under placebo as well as high score responding? I just want to be clear on what I'm looking at here.
Yes. Very good question. This is and Derek, this is mean, and it's driven by some very extreme responses. But if you if I can like briefly direct you back to Slide 7. And you go to Slide 7 and you look at the placebo response, this is a placebo group, right?
And you see the overall placebo group performance here on our platform. So this is driven by a few patients that had an extremely good response, right? And they were in this high score responder group. So when you look at the overall placebo picture, please go to Page 7, and you get a good feeling for what happened there. But it's an excellent question, and it points to the problem that you get food when you look at a few patients in this type of disease.
Yes. It's a skewed sort of presentation. That makes perfect sense. Okay, that's everything I had. I appreciate it.
Thanks and congrats again.
Thank you so much, Derek. Appreciate it.
It. Ladies and gentlemen, as we have no further questions, I would like to hand back to the speaker for any additional or closing remarks.
Well, thanks so much. I just want to close this by thanking everyone for tuning in today. I think these were very