Herantis Pharma Oyj (HRTIS) Earnings Call Transcript & Summary

Good afternoon or good morning, depending where you are in the world, and welcome to the Herantis Pharma webinar to discuss the CDNF Phase I/II data readout. My name is Julie Silber, and I will be your moderator for today's event. Presenting today will be CEO, Craig Cook, who unfortunately is experiencing some allergy or viral-like symptoms and is safely social distancing; Henri Huttunen, our CSO; Antti Vuolanto, our COO; and Sigrid Booms, our Director of Clinical Development. And I apologize profusely for my Finnish pronunciation. We're still working on that. The next slide shows our disclosure statement. As a quick reminder to our listeners, during today's webinar, management may make forward-looking statements involving known and unknown risks, uncertainties and other important factors beyond the company's control that could cause the company's actual results, performance or achievements to be materially different from the expected results, performance or achievements expressed or implied by such forward-looking statements. Please note that the forward-looking statements made during this webinar speak only as of today, and the company undertakes no obligation to update them to reflect subsequent events or circumstances other than to the extent required by law. Our next slide shows the agenda that we will follow today. We will discuss our drug candidate, CDNF; the unmet need, treatment, details on the study and the data; the conclusions from the readout and what lies ahead. We will conclude with a Q&A session followed by closing remarks by our CEO. For the Q&A session, please feel free to write your questions via the question section on the webinar dashboard. As a reminder, this call is being recorded and will be available for replay soon after the event. And with that, I would like to turn the webinar over to Herantis CEO, Craig Cook.

Thank you, Julie. So to set the theme, I'm going to start off with a brief description of Parkinson's disease and the pathology that we're treating. So Parkinson's is an as yet incurable, despairing disease that affects up to 10 million patients worldwide. And it has a cost to society of up to $50 billion, $25 billion of which is in the U.S. only and is caused by degeneration of specialized neurons called dopaminergic neurons that lie deep in the brain. And it's nicely illustrated in the graphic on the right-hand lower quadrant of the slide, where the illuminated areas are dopamine. And as we move from left to right over time, we see those illuminated areas diminishing. And that's typical for a Parkinson's disease patient. Now we've known the cause of Parkinson's for decades, but there has been very little progress since the mid-20th century when levodopa was discovered, and that still remains the standard of care for Parkinson's currently. Moving on to the next slide. CDNF being developed by Herantis is a natural protein whose natural role is to protect and preserve neurons and help them survive. In other words, it has the potential to reduce, stop, slow or even reverse progression of neurodegenerative diseases. So we aim to treat not only the symptoms but also, more importantly, the cause of the disease. On the next slide, we see how it's administered. We administer it directly to the key dopaminergic areas of the brain called the putamen and the substantia nigra. And we do this via a sophisticated surgical device that is implanted directly into these key dopamine areas. And once administered, CDNF promotes the functional recovery of these dopaminergic neurons, both within the putamen as well as in the substantia nigra. And how does it do that on the next slide? Well, we have a powerful multipronged, multitiered or multipillared mechanism, which comprises 5 key aspects. First of all, it strengthens the endoplasmic reticulum. Now the endoplasmic reticulum is the area of the cell just outside the nucleus, where all the proteins are produced and packaged or folded for distribution to other parts of the cell or other parts of the body. And in times of stress, the endoplasmic reticulum can become compromised. So this folding and packaging of the proteins becomes compromised such that you end up with faulty proteins that then can't perform their function and, in fact, become a spanner in the works, so to speak. And CDNF reduces this or prevents this such that the endoplasmic reticulum can continue to function even in times of stress. And one of these key proteins that are produced -- one of these key faulty proteins that are produced as a result of the endoplasmic reticulum under stress is called alpha-synuclein, and this is a key pathology for Parkinson's disease. And we know that CDNF acts to reduce the accumulation and the generation of alpha-synuclein. So it strikes right at the core, CDNF strikes right at the core, right at the heart of the cause of Parkinson's disease. In addition, it promotes cell survival and protein synthesis through its action on a signaling protein called protein kinase B or Akt, and this again functions to generate proteins essential in neuronal survival. Furthermore, it has an anti-inflammatory effect that reduces the damage done by cytokines produced or released by glial cells and that does long-term damage typically to the neuronal structures. CDNF acts to reduce that harmful inflammation that's caused by the cytokines. And finally, it also has an action on gene transcription, which is important as the protein synthesis moves from the nucleus into the endoplasmic reticulum. And that's a key transitional or transductional area where CDNF again acts. So those 5 tiered mechanisms underpin the potential of CDNF in not only Parkinson's but also beyond. And the next slide shows proof of those CDNF mechanisms in a preclinical experiment that we did in monkeys. And what we see here is these monkeys were part of a model where a substance called MPTP is administered. And that basically destroys and wipes out all the dopaminergic neurons in these -- in the brain of these animals. On the left, we see the damage that MPTP does. There's no dopaminergic neurons left. Well, 80% of them are lost as part of this model. And on the right-hand side, we see the brain of animals that were subsequently treated 6 weeks later with CDNF. And we can see the marked contrast in the regeneration of dopaminergic neurons. So this is a very powerful contrast between animals that didn't receive CDNF and animals that did receive CDNF. So based on this compelling foundation of both the science as well as the preclinical data that we generated, we moved the program into our first-in-human study. And I'm going to hand over to Henri at this point to take us through that.

Thank you, Craig. So because we're using here intracranial administration of the protein directly into the putamen, we actually couldn't perform the Phase I study in healthy volunteers, which would be a normal first step, the first-in-human study for any drug development. So in our study, we actually -- due to this invasive nature of the delivery, we directly started from -- started with patients. So in our study, we recruited 17 subjects with moderately advanced Parkinson's disease. And this was done in 3 different leading Nordic hospitals. We had 3 sites; 1 in Stockholm, Karolinska University Hospital; 1 in Skåne, Lund University Hospital; and 1 in Finland, Helsinki University Hospital. And the principal investigator of this study was Professor Per Svenningsson from Karolinska. And at this point, I would like to thank all the study teams, all the members at each study center, the neurologists, the neurosurgeons, the study nurses as well as the hospital pharmacy staff and also our collaborating CROs and the provider of the device, Renishaw Neuro Solution. And first and foremost, we would like to thank the patients and their families. This has been a very laborious, very extensive period in everybody's life, many, many visits, tough time. And we've had excellent patients who have really been committed to this study, and we would really like to warmly thank everybody's participation in this study. So we have presented the study design before. So I will just briefly remind you that we had a 3-part study. We basically had 3 different protocols here. So first, we had a randomized placebo-controlled 6-month main study, as shown here. And in this part of the study, we had a vehicle group where some of the patients, in a randomized manner, received placebo infusions. And after this main study, 6-month treatment ended, all patients were given the opportunity to enroll into an active treatment extension study. So the second 6-month period, everyone in the study was receiving active treatment, again, in a blinded fashion. So all the subjects and all the investigators were still blinded to the dose level and which treatment the patients belong in the main study. And at the end of the extension study, then the patients had -- could decide whether they could have the device explanted, full device explantation or parts of the device explanted. And then they enrolled in a follow-up study where there are no further infusions but rather a safety follow-up. And before the study even started, there was, of course, this -- a neurosurgical procedure, where the drug delivery device was implanted. And this was followed by a 6-week healing period and then baseline assessments before the first infusions. The end points in this study are quite numerous and complex, and please bear in mind that this is primarily a safety study. That means that the primary end point, primary objective in this study was to test the safety and tolerability of CDNF. We also collected data on safety and functionality of the drug delivery device. This data will be separately analyzed and presented by Renishaw Neuro Solutions. And in this presentation, we will focus only in drug-related outcomes. In addition to the primary end point, we had several secondary and exploratory end points that related to the features of the disease. The secondary end points were mostly clinical rating scales and tests, including home diaries for patients. And perhaps the most interesting is the so-called UPDRS, which stands for Unified Parkinson's Disease Rating Scale. This is a golden standard clinical rating scale in Parkinson's trials, and we also, of course, included this as a secondary endpoint. For the exploratory end points, we had various types of assessments. Perhaps the most interesting is the dopamine transporter PET imaging. We will present some of this data today. This is a very sophisticated imaging technology, which allows us to look into the -- deep inside of the brains of the subjects, and we can specifically look at the integrity of their dopamine neuron system during the study. Also, we collected cerebrospinal fluid samples, and we have been analyzing them for various markers, for example, alpha-synuclein levels, different species of alpha-synuclein, which is the main aggregated protein species found in Lewy bodies, the neuropathological hallmark of Parkinson's. We also analyzed CDNF levels in CSF and serum to understand what happens to the protein after the infusions of kinetic insights. And we also are performing proteomic biomarker screens from the cerebrospinal fluid in order to understand a little bit more what may be happening in humans which have been administered with CDNF. And finally, we used the Parkinson's KinetiGraph, a wrist-worn device that collates actigraphic data trying to complement the secondary end points with a more objective measure of the motor features of the disease. So 17 patients were enrolled. Here is the demographic and Parkinson's disease characteristic data at screening. And as you can see, we have a fairly well-balanced cohort divided in 3 groups. The average age being somewhere around a little over 60 years. We had a slightly male dominant study. And the Hoehn and Yahr Parkinson's staging scale, according to the Hoehn and Yahr, we had the patients from Stage 2 to 2.5 and 3. And very importantly, the disease duration since the first motor symptoms was a little over 10 years. And I will come back to this in the next slide, giving you a little bit of a further insight. The baseline UPDRS, the Part III, the motor score of the UPDRS, we had a little over 30 in off-state and around 15 or slightly below in on-state. The off-state is of most interest in this type of studies. Levodopa response, about 60%. And most -- and in this cohort, we had off time per day for patients varying from 4.7 to 6.1 hours each day. This tells you that these subjects were already moderately advancing their disease. Their motor symptoms weren't terribly well in control with their oral dopamine medications. So they are -- for example, could be candidates for deep brain stimulation surgeries and were considered to be good candidates for a safety study for intracranial CDNF. However, and now that we start looking into the data, I would like to remind you that Parkinson's disease is a chronic progressive disorder. And this data is actually not from our study. This comes from Jeffrey Kordower's paper in 2013, where they performed a thorough analysis of postmortem samples from human Parkinson's patients. And here, what is shown is dopamine transporter immunohistochemistry. So all the dark signals seen in the panel A here shows dopamine transporter positive fibers in the putamen and nerve terminals. So you can see that this area is really filled with dopamine transporter, which is a presynaptic protein mostly. So very much filled with dopamine neuron terminals. However, when we look at the Parkinson's patients, you can see that 1 year, 3 years after diagnosis, there's still clearly some dopamine neurons and terminals left. But if you go further up to 5 to 10 years, all this putamen region gets weaker and weaker in staining intensity and one could almost conclude that this area is nearly completely empty of dopamine terminals at the time of when patients reach 10 years post diagnosis. Now please remember that our patients in this cohort are 10 years since first motor symptoms. So that should tell everyone in the audience that this is by no means an optimal cohort for a therapy that aims to restore these remaining degenerating functionally compromised nerve terminals in the putamen. We really don't have many targets left to rescue. And this is very important when looking at any of the efficacy readouts from this data. And of course, it's important to understand as well that because this is a first-in-human study, we use an invasive route of administration. It's very much justified to use this type of patients for a safety study. And this is -- this was the viewpoint of the regulators and the ethics committees. In the first-in-human study, this type of patient population is best suited for this study. However, moving forward, in potentially next clinical studies, our goal would be to reach earlier-stage patients with more dopamine terminals left to risk. Then I will now give the speech to Sigrid, who will present the safety data.

Hello, everybody. Yes, I will present the safety data for you. So altogether here, I show you a picture of the treatment emergent adverse events that were recorded in the study. So in general, the majority was transient and mild. So in the first 6 months, we had altogether 108 reported adverse events after treatment initiation and in the extension study, 98. So more or less similar amount, and also the distribution over the different body systems were quite the same. So in the nervous system disorders, we recorded headache but also dyskinesias, and there were also imaging findings which were nonsymptomatic. In general, disorders fluctuated from chills and fever to fatigue. And then when you look at a little bit pink, purple, the infection and infestations, these were mostly related to device and infusion procedures and cleaning procedures of the skin around the port. But altogether, this looks quite a good profile from a safety perspective. So keep in mind that the reported AEs that are presented here are both related and not related to the treatment. Next slide. There were also altogether 9 serious adverse event reported in the main study and in the extension study. There were -- half of these were related to infectious events. So there were brain abscesses reported in the first 6 months period and also port skin necrosis in the first 6-month period soon after surgery. So our partner, Renishaw, has then done extensive risk mitigation improvements in the study, so both through the surgical procedures, but also to the infusion procedures, the device cleaning procedures and additional training to the investigators, both neurosurgeons and also the neurologist who were handling the infusion procedures. So after these risk mitigation steps were implemented, we were about halfway in all infusions in this study. So the last almost 100 infusions were conducted without any procedure-related infections or other AEs. Next slide. So the primary end point of the study was met. As I mentioned, the safety profile was very mild for CDNF. Majority of the drug-related treatment emerging events, they were -- they have also recovered, similar profile in the main study and in the extension study. And not shown specifically in this presentation, but also between those groups, we didn't see any difference in safety profile. The SAEs were partly related to device or procedures involved in the study. But with the improvement and risk mitigation steps implemented, this also helps us for preparation to the next steps. And no dose-limiting toxicities were related to CDNF. That's what I wanted to say about safety, and then I give the word again to Henri.

Thank you, Sigrid. So then let's look into the secondary end point data. And today, we will focus on the UPDRS, which is clearly the most important clinical rating scale in Parkinson's studies. And here, just a few remarks on UPDRS. So this is a clinician rating scale. So a movement disorder specialist at the neurology clinic will assess the patient at each visit where UPDRS data is collected, but the UPDRS contains 4 different panels assessing slightly different aspects of the disease. And most interesting, of course, for Parkinson's disease study is the Part III of medication or typically. And of course, I'm not saying that the other parts are not important, but typically Part III is primarily addressed when we're looking at the dopaminergic system and changes in its function. So the part -- UPDRS Part III scores typically worsen by 2 to 2.5 points per year. Of course, there's a lot of individual variation here, but this is a rough average from literature reported cohorts. The score in a healthy individual would be 0 and the maximum possible score is 199. And if you remember, our patients in off-state at baseline were roughly 30 -- between 30 and 32 points. So this is a rating scale, long time used in clinical studies. Nowadays, real-life data is also becoming more and more important. And Parkinson's home diary is one such assessment. So this is a patient-reported outcome. So basically, patients record in a 3-day period, every half an hour their states, whether it's an on-state or off state or whether they are troublesome dyskinesias or non-troublesome dyskinesias, and as well, sleep is reported. And then from these home diaries, a certain analysis can be performed. Both outcome measures are very noisy. There's a lot of fluctuation. They are best suited for large studies. Typically, like, for example, a small molecule Parkinson drug Phase II study would have 300, 400 patients. And in these [ double ] studies, the UPDRS and home diaries work rather nicely. In a study with 17 patients, of course, there are challenges with statistical assessment of results. I would like to start with an example of an impressive individual case. So here, we have a patient who was in a placebo group for the first 6 months, so receiving 6 monthly infusions of placebo, then crossing over to mid-dose CDNF here from 6 months to 12 months. And as you can see, the UPDRS in off-state was at baseline 24 or so. There was a little bit of a placebo effect from 0 to 6 months. And then the orange hatch line here is the expected decline in an average number. When the patient crossed from -- crossed over from placebo to mid-dose CDNF, there was, however, a very clear improvement. The patient improved by 43% in 6 months in the UPDRS Part III. So these type of examples are, of course, very exciting. But again, we don't see these in all patients. But it tells you that there may be something happening in this type of patients. This slide shows the averages from all the groups. So here, the red line is the expected annual decline. So we expect a little worsening over the 12-month period. And here, you can see the different dose groups. The gray one over here is the placebo from 0 to 6 months, and then we have the mid- and high-dose groups not really changing significantly in either direction. Again, improvement would be expected to be a lowering of the score. There has been a literature reported minimal clinically important difference in UPDRS around 2.5 points, but we would like to take a conservative position here. And because of the small study size and the noisy nature of UPDRS III, we have concluded that there is no clear signal in the UPDRS data, but we can clearly see that the patients are not getting worse. These 2 additional lines here show the placebo crossovers, the mid-dose in the sea blue green here and then the purple showing the higher dose CDNF across all the patients. Then a slightly different view into the UPDRS is comparing first 6-month respective treatment periods. Here, we have the placebo group from 0 to 6 months, the expected annual decline. And then the 3 different CDNF dosing groups here, the mid-dose, the high dose and the crossover mid-dose group. And you can see that there's a -- it seems like there could be a little trend of a decrease. But again, a small study, noisy metric. Then moving on to the exploratory end points. And the dopamine transporter PET is, at this point, has been analyzed, the whole data set, and we will be presenting this. The rest of the exploratory end points, the alpha-synuclein analysis, the proteomics and the actigraphy data is still being analyzed, and this data will be disclosed later. So positron emission tomography is the imaging technique, the PET imaging technique. It is based on radioactively labeled tracers that bind to specific proteins. And in this case, the target protein in our PET imaging is dopamine transporter. This is an enzyme, a transporter that's localized in presynaptic terminals in dopamine neurons. And there's a very high density of these dopamine transporters and the fibers in the putamen in healthy humans. And as you can see here, as shown in these images, the signal is significantly lower than what we would be expected in the putamen caudate nucleus area. So we used -- for this PET imaging, we used a tracer called FE-PE2I. That was 18 fluorine labeled. This is a fairly novel tracer developed at the Karolinska PET Center published a few years ago. It's a very sensitive tracer for dopamine transporter. And we used this tracer because of its sensitivity. And the patients were actually imaged at the baseline, at 6 months and 12 months. And there will be one additional 19-month time point in the follow-up study. We don't have that data yet. That will be collected over the coming winter. And regarding the dopamine transporter PET signal, healthy age match controls would be expected to decline in their signal by 0 to 2.5% each year. And Parkinson's patients, on average, have showed a reduction in the DAT PET signal by 6% to 13% annually. All Swedish patients were imaged at Karolinska PET Center and all Finnish patients at the Turku PET Center. And again, I would like to start by an individual example. So here, we have a patient who was receiving mid-dose for the whole study. And you can see that there's actually quite high UPDRS, again, maybe a little bit of a placebo, and then there's no significant improvement in the UPDRS scores over the 12-month period. However, we see a very important and interesting change in the PET. So this is the baseline PET signal level. And at the 6 months, this patient shows more than 50% increase in the PET signal. And this increase is actually maintained from 16 (sic) [ 6 ] to 12 months. So that's very interesting. And this was the most -- the patient who improved the most in the PET signal in this study. However, notably, we don't really see a clear correlation with the clinical outcome measure and the improved PET signal. Here's another example, and this is the patient that I showed earlier, the placebo crossover patients who improved in the UPDRS quite significantly from 6 to 12 months. And quite interestingly, this patient shows a decline in the dopamine transporter binding potential in PET from 0 to 6 months. But after crossing over from placebo to CDNF, there is a modest increase in the signal, about 5% increase from 6 to 12 months. And this seems to correlate now better with the clinical UPDRS score. Now if we look at the averages, and these are percent change from baseline values. So everybody is normalized to start from the same level. The expected annual decline line is shown here, and here are the different dosing groups. For the placebo high-dose crossover patients, we were unable to perform the 12-month time point PET imaging due to COVID-related reasons. So the patients were not -- were unable to visit the PET centers because of the COVID outbreak in the spring. However, for the rest of the patients, all 3 time points data have been collected. So here, the gray, again, is the placebo. There's a clear decline from 0 to 6 months. We also see a decline in the high dose group from 0 to 6 months, which, for this, we don't currently fully understand why this happened. We do see, however, a little increase coming back from 6 to 12 months in these high-dose group patients. The mid-dose group, mid-dose CDNF group, shows a very interesting average change, about 50% increase on average in the signals here, and the increase is fairly well maintained from 6 to 12 months. We again look at the same data from a slightly different perspective comparing the respective 6-month -- first 6-month treatment period to placebo. Here's the placebo, the expected annual decline. And here, we have the dosing groups, the first 6-month mid-dose group, high-dose group and then the placebo crossover patients for their first 6-month treatment period. So it looks like there may be some increase in the dopamine transporter PET in some of the patients. And in the mid-dose group, this was the strongest effect observed. Now we should remember that the starting levels in these patients, the signals were fairly low. And it's, of course, very important, as I mentioned in the beginning, that if we would -- if we could access earlier patients, we would have more dopamine neurons and neuronal terminals left in the putamen that could be rescued. And our current view is that this would show us a stronger response both in terms of PET signal and also possibly could result in improvement that shows in clinical rating scales. Finally, I would like to show these 2 images which are from gadolinium test infusion. So before any treatment infusions were performed, the functionality of the drug delivery device was tested. So this is a few weeks after the surgery, the implantation of the delivery device. So the patients are infused with gadolinium contrast agent, and then this contrast agent is visualized by MRI scans. And these analyses have been performed by our colleagues at Renishaw Neuro Solutions. And basically, it shows the striped area here in -- the purple here is putamen. The striped area is infused area of the putamen. And the green and yellow parts are a signal that is found outside of the putamen. So for example, in this right-hand side graph, you can see that there is very clear backflow along the catheter line with the contrast agent. However, the quantitative analysis of this data suggested that 60% to 75% of the putamen volume had been covered by the infusion. And this is a fairly reliable estimate on the functionality of the device and how well we can cover the putamen in these patients. This probably can be improved by adjusting the infusion parameters in the future. But the key take-home message from here is that we have definitely reached the target area of the brain with our infusions in this study, as shown by these test infusions. So at this point, I will hand over to Craig for conclusions and next steps.

Thank you, Henri. So next slide. So in summary then, from a safety perspective, clearly, we have established the safety of CDNF. And in addition to that, we've learned and improved about the administration procedure for both this study but also for future studies. From an efficacy perspective, as Henri mentioned and explained the state of the dopaminergic neurons in this group of patients, it's very difficult to gauge efficacy because of that disease severity and lack of target for CDNF and also the small unpowered, from a statistical perspective, study numbers. But despite this difficult patient group and despite the noise in the study, we've had some very encouraging observations where there hasn't been worsening of disease, which is the -- what would be expected in this patient group. And in addition to that, there's individual suggestions of both improved biological signals in some patients as well as some really interesting individual cases. And furthermore, the study has certainly contributed significantly to our understanding of CDNF and the treatment of Parkinson's disease and how to shape the program moving forward, which brings me to the next slide in terms of moving forward from here. We will now shape the program based on the data and what we've learned from this study to maximize our chances of success and, ultimately, likely partnerability. We still have some data coming through on alpha-synuclein, proteomics and some other CSF analytics that we expect shortly. Clearly, the key areas of focus moving forward will be earlier disease patients where there are dopaminergic neurons to act on. We will look to adequately power the study with sufficient patient numbers and optimize the dosing. And of course, as we've been clear in the past, and as you will understand from the mechanism of action, we have indications in Parkinson's disease clearly, but also potentially beyond Parkinson's disease such as stroke. And then our next-generation CDNF or xCDNF, as we call it, that continues, and that will be a key part of moving forward. And on that note, I'll just give you a brief update on that program. So the common challenge in CNS diseases is getting drugs across the blood-brain barrier, which is a membrane that functions to protect the brain and it keeps harmful substances out of the brain, including medications that could potentially be helpful. Now xCDNF is a -- sorry, CDNF is a large protein that cannot squeeze through or cannot cross the blood-brain barrier. That's why we have to use a surgical device to get the CDNF to the point where it can exert its effect. xCDNF, on the other hand, we've taken the smallest, most potent parts of CDNF. And these fragments are able to squeeze through the blood-brain barrier without the need for surgery and then exert their effect in the brain. And what we know about these small fragments is they retain the biological activity of CDNF. They can squeeze through the blood-brain barrier. They are highly potent molecules. So even though we've taken just a fragment of it, the potency is in the picomolar, which is really, really potent, and we are able to engineer them for improved pharmacokinetics. So we're at the point now where we're close to selecting the final compound. And moving forward, clearly, the target product profile for next-generational xCDNF would be to administer it without a surgical device, potentially with a simple subcutaneous injection, which, from a Parkinson's therapeutic perspective, would be a major advancement and contribution to the treatment of Parkinson's. And again, there's indications. Parkinson's is a key indication, but there are several other indications that this could have potential, too. So that wraps up the formal part of the presentation. I'm going to hand back to Julie. Hopefully, we've given you a good framework and a good understanding of the data, CDNF as a treatment and Parkinson's per se. We'll take questions now and concentrate on specific details that you might want clarified.

Great. Thank you all for such a wonderful presentation. I just wanted to do a quick reminder that you can ask questions by typing them into the Questions section on the dashboard of the webinar. Our first question is, has the response seen in the 2 patients with particularly good DAT PET binding been sustained? Or how has it changed?

Henri, do you want to take that?

Yes. As far as we know from the 12-month data, and as you actually saw in the graph, one of the patients who showed the very, very high response maintained this improvement in DAT PET. It will be very interesting to see the 19-month data, of course, because this is 6 months after the infusions were stopped. That would be, of course, very interesting addition to this data set. We don't currently have a date when this data would be available. The COVID situation may impact a little bit on how these subjects are able to visit the PET centers, for example. But at least some of the patients who showed improvements did also maintain it until the end of the infusions, whether it remains -- and the signal remains high after that remains to be seen.

Great. Thank you. The next question is, when did the infections occur that precipitated the termination of the trial by the 2 participants.

Should I take that?

Sigrid, do you want to take it?

Yes. So we had 2 cases of brain abscess. So the first one happened about 3 months after surgery, and it was found out that it was related to one of the infusion occasions. As it was found out, there was a hair entrained into the infusion port. So there, we found a clear cause. The other time was already later in the study. I think it was about 5 months after the surgery. So -- and that was also considered related to one of the infusions. So the infusion procedures seem to be the critical point, and those have been clearly improved by risk mitigation steps. So the site staff was retrained on cleaning procedures prior to an infusion, and also sterile drapes rates were introduced. So overall, what we have seen now also in the end of the study that the procedures went much faster, and the staff was also very happy with the new risk mitigation and cleaning procedures for the infusion. They are not directly related to the surgical procedures.

The next question is, what are the time lines for your next clinical study.

Well, I think the program is now at the point we're still waiting for some data, as I mentioned, coming through. We continue to perform our analysis. And in the final quarter of this year, we'll firm up the timings once we have a clear view of the protocol or the potential protocol moving forward for follow-up studies. And in principle, during the course of next year, a potential follow-on clinical study is conceivable.

Great. The next question is, did you get any specific cluster of characteristics among patients to relate the results.

Sorry, can you just repeat the question, Julie?

Yes, I will. Did you get any specific or -- yes, did you get any specific cluster of characteristics among patients to relate the results?

Well, what -- I mean, I think Henri could probably provide more detail on that. But from my perspective, I think it's -- given the patient cohorts that we've -- the study was conducted in and the state of the dopaminergic neurons in those patients, as Henri explained in the micrographs that we saw earlier in the presentation, it was difficult to identify any specific clusters. We probably wouldn't expect any clusters with that state of dopaminergic loss. Analysis continues. But to date, I don't know that anything specifically have been identified. Henri?

This is a little bit too early to say since we still have the proteomics analyses ongoing. We're also genotyping the patients for typical Parkinson's risk genes. It is a small study. And importantly, it's not powered to address efficacy. But it's possible, of course, that we can still, in the remaining analysis, could -- if we're lucky, we could see some interesting subgroups, for example, similar changes in some biomarker candidates or perhaps sharing a similar genetic risk for the disease. And these type of things possibly could be seen. But of course, we don't know until we have thoroughly analyzed the data. It's a very important and a very good question. I very much agree with this type of thinking.