Good afternoon, everybody. Why don't we get started? My name is Ravi Ranjan, and on behalf of other panels of the Young Investigator Award Committee, Faisal Saeed, Eugenio Cingolani, and Mark Levine, and Marina Cerrone, I welcome you all to the basic science part of the Young Investigator Award Ceremony, or the award competition. The award ceremony happens tomorrow. First of all, let me start with congratulating all the finalists. You guys have done excellent work. I should be proud of it. There were lots of excellent presentations, so the fact that you got here is a mark of achievement in itself. It is a competition, so we have to follow some rules. Each presenter will have 10 minutes to present. There's a clock that comes up on the screen. It turns yellow when you have two minutes and red when you have one minute left. Please don't go over. As we have 10 minutes of questions and answers to follow, the questions will all be coming from the judges on the table here. Traditionally, we just haven't had time for the audience to ask questions beyond those 10 minutes, so we'll just stick with that format. With that, why don't we get started? Our first presenter is Lillian Gutierrez from NYU, and the title of her presentation is Exercise-Induced Dysregulation of the Adrenergic Response in a Mouse Model of PKP2 Arrhythmogenic Cardiomyopathy. Thank you for having me as the judges present me. I'm Lilian Gutierrez and I'm presenting the study, Exercise-Induced Dysregulation of the Adrenergic Response in a Mouse Model of PKP2 Arrhythmogenic Chiromyopathy. Before starting, I would like to acknowledge the work of my lab. I am part of Mario Del Mar Lab at NYU and also I would like to thank the contributors of this study, Dr. Giorgia Bertoli, Dr. Chantal Bano from Amsterdam UNC and the Microscopy Laboratory at NYU. So I want to start saying that glycophyllin-2 or PKP2 is an essential component of the desmosome. The desmosome is this specialized protein complex that provides functional and structural integrity to adjacent cells, in this case, cardiac myocytes. Pathogenic variants can lead to arrhythmogenic chiromyopathy or PKP2-ACM. And PKP2-ACM is characterized by the loss of cell-to-cell contact, structural damage and also the progressive fibrophatic infiltration in the tissue. But PKP2-ACM is not only characterized by structural damage. Another highlight of this disease is that it presents with arrhythmias and electrical alterations. These arrhythmias can be worsened or triggered by exercise and catecholamine surges. So in this study, we are very interested about the link between exercise and PKP2-ACM. So for that exercise, it has been described that is related to adrenergic response and metadrenergic receptors. And previously, it has been also described that there are two pools of metadrenergic receptors in cardiac myocytes. One in the sarcolemma, as we can see here in the diagram, and other intracellular pool associated with the diet. So the aim of this study is to dissect the relative contribution of these two pathways in the specific case of PKP2 deficiency and exercise. So for that, we designed and implemented a running protocol using cardiac-specific tamoxifen-activated mice. These are young adult mice that after tamoxifen injection, I don't know if I can see my pointer, but PKP2 protein is removed, and these mice run through six weeks, 75 minutes daily. We have in this study a control group and a PKP2 knockout group. We have sedentary and trained conditions for each group. As background of this study, Bano Pergen et al. previously described in trained PKP2 cardiac myocytes that the calcium transient amplitude of these myocytes is significantly increased in PKP2-trained myocytes during baseline compared to the sedentary condition. However, for time purposes, I'm only showing the background of the trained PKP2 myocytes. So these myocytes, after isoprotein administration that is a beta agonist that target the sarcolemma and intracellular receptors, we observed a disproportional increase of this calcium transient amplitude, as we can see here in the red bar. Later in this study, we wanted to test the response to different beta blockers. And first, we test propanolol. Propanolol is a membrane permeable beta blocker that will target and reach the sarcolemma receptors and also the intracellular beta receptors, as we can see here in the diagram of the left. And it will successfully block the isoprotein response. So after quantification, indeed, we block the isoprotein response using propanolol. And the values are similar as the baseline, as we can see here in the third column. Finally, we wanted to test another beta blocker commonly used in PKP2ACM, that is sotalol. But in this case, sotalol is a non-selective blocker that will only reach the sarcolemma receptors, as we can see here in the diagram. So the intracellular receptors are still active and sensitive to the isoproterol response, as I'm showing in the diagram. And after quantification, the local calcium transient amplitude remains as the isoproterol administration response. So with this data, we hypothesize that there is a decreased abundance of the sarcolemma beta adrenergic receptors while the intracellular beta adrenergic receptors are preserved in the trained PKP2 knockout condition. So to test this hypothesis, we wanted to use an innovative technique that is super-resolution expansion microscopy. So for this, I want to credit my colleague, Georgia, for implementing the technique in the laboratory. And this is a nice technique because while we expand the cardiomyocytes, we are preserving the structural integrity of the cells. I'm showing you here a not-expanded cardiomyocytes acquired with a 63X lens. We can see how the length of these cardiomyocytes is 100 microns. And after expansion, we acquired these cardiomyocytes with a 10X lens now. And now we can see that 100 microns is a fifth portion of the complete cardiomyocyte. And as I mentioned, we can preserve the structural integrity and if we see this desmin staining, we observe how the desmin is maintained and nicely distributed along the C-disk. So we use this technique to characterize the abundance of beta adrenergic receptors in cardiomyocytes. I'm showing you here in green the distribution of beta adrenergic receptors in a control myocyte and how nicely beta adrenergic receptors distribute along the lateral membrane and the cell end. Also in intracellular regions. However, for the purpose of this study, we are only focusing in the regions of the lateral membrane and the cell end. So for the results of this section, I'm showing you the control group and sedentary entering conditions and we image the lateral membrane and the cell end. Then we extract by ROIs the distribution of the beta adrenergic receptors in these regions, lateral membrane and cell end and quantify the density of beta adrenergic receptors. So for this group, we observe that there is an increase of beta adrenergic receptor density in the trained condition, both in lateral membrane and cell end. If we observe the opposite condition, the PKP2 knockout group, and we follow the same strategy after quantification of the beta adrenergic receptor density, now we observe that in the PKP2 knockout group in the trained condition, there is a significant decrease of the beta adrenergic density in the lateral membrane and cell end of this myocyte. So with this data, we conclude that indeed there is a reduction of the sarcolemma abundance, beta receptors abundance, sorry, in trained PKP2 knockout condition. But we wanted to test these results in a more functional way. So for that, we take advantage that catecholamines bind to beta adrenergic receptors in cardiac myocytes, and they reach the sarcolemma receptors and also the intracellular receptors. But to get inside, they need an specialized transporter named OCTRI. And this is really important because it has been described that the intracellular beta adrenergic receptors are needed to be activated to further or subsequently activate and phosphorylate sarcoplasmic reticulum proteins, such as phospholamban. And in a setting of trained PKP2 knockout condition, we have also demonstrated previously that there is a hyperphosphorylation of the phospholamban protein and also the ryanodic receptor that leads to an abundance of calcium, of leak of calcium from the sarcoplasmic reticulum that hence increase the risk of adrenogenesis in this condition. So with this, we take the strategy of removing the contribution of the OCTRI transported from the cardiomyocyte, and we are only going to see the scenario of abundance, sorry, of the contribution of the beta adrenergic receptors in the sarcolemma and in the intracellular pulse. I'm only going to show the results of the trained condition, but we have analyzed the different conditions that I have talked about in the presentation. I am showing you here that we administered different agonist, isoproteranol and the catecholamine norepinephrine. I am reporting the response to isoproteranol and norepinephrine relative to the baseline. As we can observe here, norepinephrine has a less response because we have lesser sarcolemma receptors and also the OCTRI pathway is silenced. However, isoproteranol, that all have no problems to reach the beta adrenergic receptors in the sarcolemma and in the sarcoplasmic reticulum, the response is still relatively high. So with this data, we conclude that indeed we have less abundance of sarcolemma receptors and a dependence of the OCTRI pathway. Finally, we wanted to test in a more microscopic level, the distribution of the sympathetic inhibition in our groups. So for that, we use the tyrosine hydroxylase staining to see the innervation across the tissue. Tyrosine hydroxylase will look like these brownish spots. And when we quantify these spots in the control condition, we observe a significant increase in the train control condition. However, if we look at the knockout condition, we observe a significant decrease of the train condition of the tyrosine hydroxylase distribution. So for that, I would like with all these data conclude that exercise leads to an increase in the abundance of sarcolemma receptors in control cells. There is a decrease in the abundance of sarcolemma beta adrenergic receptors in train PKP2 knockout cardiac myocytes and yet an availability of intracellular receptors. And all this heterogeneity can contribute to the adrenergic response and the arrhythmogenesis of PKP2 ACM. Thank you so much. Thank you. Thank you, great presentation. The question that I have, it's prior to dissecting the meganescent at the cellular level, how did the animals look, and do you see any changes in the ejection fraction in those animals if you checked? And, yeah, so that would be my first question. Thank you. So, in this study, we didn't address the more, let's say, clinical features of these mice. However, in a previous study of our laboratory in which they indeed analyzed these parameters in a trained condition of knockouts, we observed, indeed, a reduction of ejection fraction. So, yes, they have an ejection fraction reduction. The second question was? And then it was related. You know as you know Of course. That's a limitation in our study since we are using a treadmill to test and train these mice. The telemetry system haven't been implemented yet. And also we have to take into account that in our system of the treadmill, putting a telemetry system will be also, will have really noisy images. So we don't have that available to test if we are having spontaneous arrhythmias. And also if we even try to, let's say, these mice do not, are not exercising and we are taking AKGs during anesthesia, maybe that can influence also our studies. So no, we don't have that data. If I have to hypothesize or suggest something, I believe that they have. It's true that one observation with our colleagues was that indeed after treadmill running, our mice look tired, they knock out, I mean, tired and very compensated compared to the control. So if that is a syncope or pre-syncope, I think I wouldn't say for sure, but they look Thank you so much. Well, it's true that this is only an extra step of the previously thing we demonstrated about the beta-blocker mechanism or response in this mouse. Indeed, now we are demonstrating that there is a reduction of the sarcolemma receptors. And of course, one question that rises our minds is what happened to the intracellular receptors that are over there. If indeed there is a desensitization or something else, what is happening to those receptors? And a second thing we are following, too, is to go deeper into that innervation in case of the knockouts, the innervation of our mice, and see in a three-dimensional way we can get to see this innervation and what is the implication of those nice beta-receptors that we see in the cell N in relation with PKP2 more precisely. Yes, actually, I'm very proud, I'm very excited that I have seen some presentations across this Congress that, indeed, we are seeing heterogeneity of sales, not only caromicides, how they contribute one with another. So let's see in this model what is the contribution and the feedback they have between each other. So yes. Great. Thank you. I have a question. So you showed very nicely the changes of the heterogeneous distribution of the receptors. That data is quite convincing. Now I want you to speculate a bit. Thank you. That's a really nice question. Of course, a mechanism behind arrhythmias, or an easy thing to say, is that heterogeneity always underlies or relates to arrhythmias. We want to say that what we have seen is not only that the adrenergic reduction is prarythmic, we have to mention that we have previously seen that in a BKB2ACM model there are other changes such as calcium mishandling and also metabolic processes that are being affected that all together can contribute to the arrhythmogenesis of our model. So to see that everything is targeting into a complete puzzle is really nice. So I think it's the complete whole of processes related to the arrhythmogenesis, we are seeing that this is another thing that we are adding to the complete idea. That's the new thing that we are putting here. Great, thank you very much. Our next presenter is Christian Egli from Vanderbilt University and the title of the presentation is High-throughput screens identify genotype-specific therapeutics for channelopathies. Thank you. It's an honor to be here today to present my work. I'd like to start off with a brief background on Long QT syndrome and ion channel trafficking. So when I refer to trafficking today, what I mean is the intercellular transport of proteins. Specifically, what I study is the KV11.1 potassium ion channel, and variants of this channel can cause trafficking defects. So I developed a high-throughput drug screen to identify potential drug candidates to improve that trafficking in these variants to treat Long QT syndrome, and then followed that up with multiplexed assays of variant effect to basically test which variants are likely to respond to that drug. So casein H2 is the gene that encodes KV11.1. It's also known as the human etherogogo-related gene, or HERG. And as you can see here on the right, is a single cardiac myocyte action potential, where in the red here is a prolongation in this action potential duration. And so that corresponds with the underlying current, where at the end here, during phase three of the action potential, in normal current in blue, there is a decent amount of IKR current. But in Long QT syndrome, that current is drastically reduced, causing that delay in repolarization. So globally, we identify that as a prolonged QT interval in electrocardiograms. Approximately 25% to 40% of Long QT syndrome is caused by casein H2 variants, where the dominating mechanism is that 80% to 90% of missense variants cause this trafficking defect. And so here's another closer look at trafficking. The channel forms a tetramer in the endoplasmic reticulum before being transported to the Golgi apparatus, and then finally, the plasma membrane. And so during this process, you can observe this by a Western blot, where there's two separate bands, this 135 kilodalton core glycosylated protein and the 155 kilodalton form, which is the trafficked form. But in these genetic variants on a Western blot, you only see the core glycosylated form, suggesting that it's not being transported properly to the cellular surface. But numerous reports have shown that reducing the temperature or treating with inhibitors like E4031 improve that trafficking to the cellular surface. And then when you wash off E4031 or a drug inhibitor, you get an increase in the amount of current on the surface. So we wanted to use this information to develop a drug screen to try to identify drugs that improve trafficking without blocking the channel. And so for that, we use thallium flux. It's a technology that we use at the high-throughput screening facility at Vanderbilt. Essentially, you overexpress your channel of interest in hex cells and plate them in 384-well plates. And after a 24-hour treatment with the drug of your choice across all these different wells, you basically incubate your cells with a fluorescence indicator that responds to thallium. And during the assay, you add in thallium, and it fluxes across the channels. And so it becomes a fluorescent marker of how many channels are on your cellular surface. And so you can imagine that after a treatment with a therapeutic for 24 hours, it increases this fluorescence. And so that's our readout. So with that, we screened 1,680 drugs from an FDA clinical or NIH clinical collection in two separate trafficking-deficient variants in four replicates each. And so here's our data. And so these are the two separate variants where zero is basically everything is control. And you can see some drugs were fairly toxic, especially in the oncology area. And most of our hits, which are above this red-dotted line here, is three standard deviations from the median. Most of our hits here are actually these inhibitors. So we can't use a lot of these in clinical practice. But we did identify AvaciaPib. And so what is AvaciaPib? It's a cholesterol ester transferase protein inhibitor. This was part of a drug class that failed phase 3 clinical trials because none of them reduced cholesterol levels. But AvaciaPib was one of those drugs that was fairly well-tolerated. I did actually test some of the other CETP inhibitors. None of the other ones actually increased the trafficking of KV11.1. You can see here on the right, this is a concentration response curve of AvaciaPib in four separate trafficking-deficient lines. But what was interesting about this drug is when we started to look at it in acute effects, in wild-type channel. And so this is wild-type thallium flux, where one is your control conditions of thallium flux fluorescence. As you're increasing concentration with E4031, you're getting a steady block of the channel, as you would expect. But it actually appeared that AvaciaPib was activating the channel. So we know drug in the bath solution during thallium flux can have autofluorescence or cause some other false positives. So we went to patch clamp electrophysiology, where, again, we saw an increase in the amount of current through the channel without drastically changing the current-voltage relationship, as shown here in blue, compared to our standard vehicle control. So what's the underlying mechanism of how AvaciaPib is actually activating the current? And I tested inactivation, which some potent activators affect inactivation gating, and there was really no effect. It's affecting deactivation. And so this is our voltage protocol. We're holding different steps for 10 seconds periods of time. And so I want you to look at the negative 120 millivolt step here. So the dotted black line is essentially zero current. It's opening very rapidly and then closing. And so you can fit a biexponential function to that. With AvaciaPib, that is drastically slowed. So it's affecting the activation gating as opposed to inactivation gating. And you can see the different tau parameters here. It's drastically slowing the deactivation of KV11.1. So we had a drug in hand. It's increasing trafficking. It's activating the channel. But how do we potentially go into patients when there's more than 500 different variants and we don't know which patients might respond? So for that, we collaborated with Brett Kronke, who does saturation mutagenesis. And so essentially, he has these pooled libraries where he's taking one amino acid residue of the channel at a time and mutating in one of the possible 20 combinations of amino acids across the entire channel. And so when we have these pooled libraries, then we can do separate treatments to basically phenotype which variants are severe and which variants are likely to respond to drug treatment. For this, they're extracellularly HA-tagged. And so then when you add an antibody into solution that's fluorescently marked, we can do FACS and basically sort into different bins these variants based on if they have zero surface expression or if they have high surface expression and then use next generation sequencing to identify which variants are in which bin and their phenotype. So here's a result from an 100 amino acid residue where in the y-axis, these are the residues that we're mutating into. And you can tell in S5 helix and area around the poorer region, there's several variants that are severe trafficking deficient variants, like less than 10% of wild type channel. So that's our phenotyping. And then after drug treatment, we can identify which of those variants are likely to respond. And so here next to the poor region is a large amount of variants that are responding to drug therapy, whereas in S5 helix here, there's some that are not responding to therapies. So in conclusion, I've presented a combination of high throughput screens. So a high throughput drug screen to identify potential drug candidates to treat patients, and then high throughput variant effect mapping to identify those specific patients that might respond to therapy. So this is a new avenue for precision medicine. We currently have work in IPS cardiomyocytes ongoing. Future efforts could lead to clinical trials in patients with Long QT syndrome. And I'd like to acknowledge my mentor, Bjorn Nolman, for giving me the opportunity to develop my research in an independent direction, as well as collaborators in the Brett Kronke lab, Alex Shin, and also Tree Doe in our Nolman lab. And with that, I'll take any questions. Thank you. That was really nice work, and you very sort of elegantly showed how your screening and then sort of targeting the channel was increasing the functional response. The question I have is two, one sort of a more basic technical one. The two variants that we use, how did you end up picking those? And were they sort of different enough to sort of give you signal that would apply more broadly? And the second question is, can you speculate more on the mechanism of how this is changing the channel kinetics? So for the first question, how I selected the variants, both of them responded and were fairly well characterized in the literature to respond to E4031, which is kind of the gold standard of increasing trafficking. So that's why I chose those two. In my original paper, one of the variants actually had a double variant. It was G601S, and it had a truncation. And that had better characteristics for the screening to give us a better separation of our positive control and our controls. So that's what I used for one. And then N470D was one that had a full channel. And then can you repeat the second part? Speculation on mechanism of how it's altering the kinetics of the channel. I think because it's increasing trafficking and because it seems to be activating the channel, I do think it's acting somewhere directly on the channel. I think it's likely different from the poor domain. We still don't know exactly where it's binding, and that's going to require follow-up experiments. The reason I brought that up is when you should have did it on the different mutations, be it the poor or the helix part of it, you could see that the response is quite different, even in those regions, right? There are clearly mutations where it doesn't respond as well. That could potentially tell you a little bit more about the mechanism of how this is acting. Right, right. And I think once we get more data on that too, potentially we could identify a binding site. If we're mutating some of those residues, and E4031 is still working, but AvaciaPib is not, that could potentially identify where it's binding. But that's, again, going to take more works. Great work. Thank you. I was wondering if you checked other ionic currents or you focus specifically in CAV111, and if you could speculate if it is possible there would be an effect on additional currents. It's possible. This was already tried in phase three clinical trials, and they didn't have, I mean, patients were, it was fairly well-tolerated. That isn't to say that at some of these higher concentrations it doesn't have effects on other ion channels, so we're currently exploring that as well. One of the issues we've had, in vitro at least, is that at higher concentrations it seems to have some cellular toxicity. And so we've run into some issues there, even though it was well-tolerated clinically. We think that's because of how lipophilic it is. The other question, you brought the clinical studies. I know they were not powered for that, but are you, I couldn't find anything, but are you aware if there were any changes in repolarizations in those patients, like QT, QTC? So in one paper I saw of super therapeutic doses, it decreased QT about seven milliseconds. And in control conditions it was three milliseconds, so there was no difference from control. Again, those are wild-type patients, so we don't know, some of these long QT variants, how they're gonna respond to the drug. And it might actually be beneficial that it's not overly shortening QT interval. Thank you. Is there anything else? Is there time, or? Yeah, we have time. Thanks very much. I really enjoyed your talk. Thanks. I had, I guess, a couple questions, but I'll try and narrow it. Were you expecting differences in kinetics when you were doing these experiments? The reason I ask is, I wouldn't have necessarily predicted that there would be differences in kinetics if the underlying problem is a trafficking problem. I would have thought maybe you would just expect to see a higher amplitude. I was not expecting that. I was not expecting these kinetic differences on the channel. I mean, most of these drugs will inhibit the channel if they're increasing the trafficking. And I don't believe most of those actually affect kinetics of the channel. And I do have one Western blot from about two years ago that all of the other activators do not seem to improve trafficking. So this is something that seems very unique to AvaciaPib. And the other is another unrelated question. Would you think of thallium as a fluorescent indicator for potassium? Was that, had that worked for you? So thallium actually conducts through potassium channels more than potassium does. And the Nernst potential, the reversal, the Nernst potential is about negative 40, which is kind of in that sweet spot for KVLM 0.1. Not only that, but I really had to go through a lot of optimization for this. And so I didn't get into this and get to show it, but across all of those wells, I have an activator in there. So if those channels are on the surface, they should be activated. And so that's how I had the best difference between my positive control and other controls. So it's worked very well for us. Great, thank you. Thank you. Great, thank you for an excellent presentation. All right, our last presenter is Saranda Nemanja. From University of Bern. And the presentation is titled, KCNH2 Suppression Replacement Gene Therapy for Type 1 Short QT Syndrome. A Proof of Concept Efficacy Trial in Rabbits with Short QT1. Thank you very much for the kind introduction. Good afternoon, everyone. It's an absolute pleasure to be here today and have the chance to share some of our latest work we've done on this KCNH2 specific suppression and replacement gene therapy approach that we're currently investigating in our Short QT Syndrome Type 1 rabbits. Now, why don't we start with what this Short QT Syndrome Type 1, or Short QT1, actually is? Well, Short QT1 is a genetic channelopathy that is characterized by gain of function variant in the KCNH2 encoded potassium channel underlying the IKR current. Or in simpler words, we have an enhanced IKR. And with an enhanced IKR, we expect nothing less but a shortening of the cardiac repolarization and therefore a shortening of the corresponding QT interval on 12-lead ECG, which may in turn give rise to atrial and ventricular arrhythmias that can ultimately culminate in sudden cardiac death. And important to note, the most common initial manifestation of the disease is a survived sudden cardiac arrest or sudden cardiac death in that in half of the cases before the age of 40. As for current therapeutic approaches, these consist either of the pharmacological approach with quinidine being one of the drugs of choice or device therapy with an implantable cardioverter defibrillator for patients in whom breakthrough arrhythmic events still occur despite optimal pharmacological therapy or those with a survived sudden cardiac arrest. And while these luckily do their job in many patients, they not rarely come along with intolerable side effects leading to patient noncompliance and a reduced quality of life. And most importantly, these are merely symptomatic treatment options and they do not target the pathogenic substrate of the disease. But what if we do? What if we target the pathogenic substrate of the disease? Because after all, we're living in the era of gene therapy. Could gene therapy rescue the pathogenic phenotype in transgenic SRKT1 rabbits? So this is more or less the question that we've been trying to answer during the last years. Now, Dr. Mike Ackerman's lab from the Mayo Clinic have designed this KCNH2-specific suppression and replacement gene therapy approach, shortly KCNH2Suprep, that consists on one hand of a KCNH2 short hairpin RNA, or SHRNA, as a suppressive arm that targets a region of the gene that's devoid of common polymorphisms or disease-causing variants, thereby knocking down not only the variant allele, but also the wild-type allele in a variant. shim, thereby allowing increased expression of wild-type KCNH2 shim channels. Now after obtaining initial positive results in iPSC-derived cardiomyocytes, they decided it was time to test it in a suitable animal model that effectively recapitulates the human disease phenotype, which is where our transgenic rabbit model came into play. So to test the therapeutic efficacy of KCNH2 suprap in vivo in our rabbit, we've utilized an AV9 backbone, placed the construct under a cardiac-specific promoter, and also chose a cardiac-specific route of delivery. So what we did here is we accessed the carotid artery of the rabbit with a swan gaunce catheter, went down to the aortic root, and occluded it for 25 to 30 seconds, as can be appreciated in the X-ray image shown here. We then injected the construct downstream of the occlusion, leaving it with practically nowhere else to go except for the coronaries, aiming at maximal cardiac distribution and minimal off-target effects. This was repeated three times for rabbit, and the rabbit... Now, if we take a look at the heart rate corrected QT interval or the QDC, which let me remind you is originally short. And importantly, we did not only Importantly, we did not see such changes on the QDC or QDI in rabbits that received the sham treatment. Next, we wanted to assess what the KCNH2 suprap does at the whole heart level ex vivo. So to do so, we utilized the combined Langendorff and optical mapping setup that allowed us to acquire panoramic images of different views of the heart, as can be appreciated in the schematic here. And we selected regions of interest at the base, mid, and apex of the left ventricle across the different views of the heart, as we were not only interested in assessing the effect of the KCNH2 suprap on the optical APD90, but also on potential apicobasal heterogeneities. Now, if we take a closer look at the apex, you'll see that the optical APD90 was indeed significantly shorter in untreated trichity one as compared to wild-type hearts, but was completely normalized. and the wild-type hearts were less pronounced to start with. However, SupRep-treated TRKD1 hearts consistently exhibited APD90 values in between the two. And importantly, if we take a look at the apicobasal APD90 heterogeneity, which we calculated as the difference of the APD90 at base and the APD90 at apex, you'll notice that it was indeed significantly increased in the untreated TRKD1 hearts, but completely resolved in the SupRep-treated TRKD1 hearts. Finally, we were curious to see what the KCNH2 SupRep does at the level of individual cardiomyocytes, so we isolated these from rabbit hearts three weeks after treatment and performed PATGLAN measurements to start with. Now, if we take a look at the action potential duration at 90% However, this time, it was not fully normalized to the wild-type levels. Importantly, we did not see such differences on the APD90 and sham-treated TRK1 cardiomyocytes. Next we wanted to see what's happening with the IQR as it is sort of the direct target of the gene therapy and is pathologically enhanced. for KD1 cardiomyocytes. but not in all of them. levels, which is why we thought, okay, why don't we look at the individual cardiomyocytes and try to quantify the percentage of cardiomyocytes in each group that shows wild-type-like IQR? And we did that, and what we observed is that roughly 13 percent of untreated and sham-treated SRKT1 cardiomyocytes exhibit wild-type-like IQR levels, while roughly 50 percent of the suppressed-treated SRKT1 cardiomyocytes exhibit wild-type-like IQR levels. So these partial therapeutic effects at the IQR levels might explain the subtotal normalization of the APD90 at the individual cardiomyocytes. However, this was still sufficient to fully restore the pathologic phenotype at the whole heart level ex vivo, as well as at the in vivo level. And importantly, this roughly 50 percent correction of the IQR was complete. rabbit, making it the first study to demonstrate clinical perspective. Well, for one, KCNH2 suprap could be a Two, as mentioned earlier. The study encourages further development of this gene therapy towards a first in-human clinical trial for patients with short KT1 or long KT2. However, it's very important to note that long-term efficacy by distribution That's really an impressive I was curious, was there one area that you were focused on that you did most of the work on? Because it seems like a lot for one person to do. Yeah. To be honest, from the experiments that I've shown here, I actually did all the experimental work. That's impressive. It definitely took a lot of support from colleagues that had the experience and were willing to share that with me. But yeah, I was involved in all the experimental work. With mentioning this delivery of the viral construct directly into the aortic roots is something that we have to do at the experimental surgical facility with the help of interventional cardiologists. So that's something that I wasn't involved with. But then the experiments three weeks after treatment for its therapeutic efficacy assessment were all done in our lab. And I had the opportunity to actually carry these out. That's really impressive. Yeah. Congratulations. Thank you. Very nice work. So I had a couple of questions. Definitely, arrhythmia studies are super important because the whole reason we're doing this is to hopefully reduce the arrhythmia incidence. Now, the interesting thing is that in our transgenic 381 rabbits, during ECG measurements that we perform randomly, we don't have telemetric ECGs implanted, so these ECG measurements are really performed at specific time points rather than during the whole three weeks during which we keep these rabbits. So during these measurements that we've performed before and after treatment, we did not observe any baseline arrhythmias, so it was, in a sense, a bit difficult to actually derive any conclusions about whether the QT normalization or the APD normalization actually has antiarrhythmic effects. However, we've been working on several different studies, one of which is a Carnitin study, where we basically observed that supplementation with Carnitin actually prolongs the QT and prolongs the APD90 in transgenic 381 rabbits, and this then also led to a reduced arrhythmia inducibility as observed in a 2D in silico model. However, we were trying to actually see what's happening with arrhythmias in our rabbit model because, yeah, 2D in silico modeling is cool and all of that, but we would also love to have that sort of information in vivo or ex-vivo. So what we tried to do is during the ex-vivo whole-heart experiments with the Langendorff and optical mapping setup, we actually tried to inject some parasympathomimetic drugs, because as you know, arrhythmias in short QT1 typically occur during sleep or at rest, so when the parasympathetic tone is high. However, we weren't able to actually consistently induce any arrhythmias, which is why we weren't then able to actually derive any conclusions on whether the arrhythmia inducibility is indeed reduced, but as mentioned, there are other studies that we worked on where we actually saw a reduction in the arrhythmia inducibility, and on top of that, we also, for this project, since we weren't able to induce any arrhythmias ex-vivo, we teamed up with Jordi Heiman at the University of Graz, and he's actually working on some 2D silico modeling for this study as well. Fair enough. But that's definitely something that would have to be further elaborated and investigated. I'd like to ask you another question. Yeah. Unrelated. I was fascinated by... about that. Yeah. So it is actually well known that in diseases like short QT syndrome and long QT syndrome, it's not just the shortening of the QT and APD or the prolongation of the QT and APD respectively per se, but it's also its regional dispersion. So it is actually well known that in our transgenic 3D1 rabbits, the dispersion, the repolarization heterogeneity is increased. And this is thought to be one of the main contributors to arrhythmia due to the substrate for reentry arrhythmias. So this is why we were really interested to see whether we can, whether with our therapy we actually are able to reduce that and hopefully also reduce the risk for arrhythmias. But on the other hand, we were also interested to see whether by chance with this therapy, I mean, you never know, we wanted to rule out that we're actually increasing more heterogeneity and in turn causing more problems than the one that we're trying to treat. But as you saw, it seemed that the heterogeneity actually normalized. And also at the in vivo level, we took a look at the QT dispersion, which is known to be a marker for regional heterogeneity of repolarization, and also there we did not see any difference between the sub-prep treated 3D1 rabbits and the wild-type rabbits. Yeah. I wonder... Yeah. Yeah. I mean, definitely. So, if I'm not mistaken, at the, like, apical cells, they have more IKR, which is why the – okay, I'm getting some hints that I'm saying something that's wrong, but I'm not sure. They have more ITO, for sure. Yeah. Yeah. But I'm not used to it. Yeah. To be honest, we were discussing our results internally, and yeah, we were actually leaning towards the direction that apical cardiomyocytes indeed have some more IKR, which is why the APD90 tends to be shorter in our rabbit cardiomyocytes. But yeah, definitely regarding the question about where the cardiomyocytes that we patched arrived from, we were always speaking in the sense that we wanted to take cardiomyocytes from the left ventricle only, but then we didn't really discriminate between base, mid, and apex, or between endo, mid, and epi, and we're fully aware that there tend to be differences in the APD90 duration due to different channel expression patterns. So we can definitely not rule out that we had certain populations that we were patching at that specific moment. It was sort of blindly patching LV cardiomyocytes. Thank you. Yeah. Thank you. Great work. Any technical questions? titers, and when you reach the packaging capacity, which is about 5 KBs, the titers start coming down. So can you comment what's the size of your final construct that you're having? Yeah. It was basically, if I'm not wrong, 4.8 kilobases. So we were really at the limit of the packaging capacity, which is why we also had one of the limitations of our study. So due to the fact that the construct was actually going towards the limits of the packaging capacity, we weren't able to fit another tag or flag so that we could actually track down where the construct is going, which is why we then followed up with some biodistribution studies to circumvent this problem. But yeah, basically, we were actually towards the packaging limit. And regarding the doses that we administered, I mean, we're aware that for AV9-based gene therapies that are administered intravenously, the dose usually tends to be between 13 over 10 to 14 over 10 VG per kilogram. However, since we actually decided to go for a cardiac-specific route of delivery, we wanted to start with a much lower dose and see whether we can actually see therapeutic effects. So the dose that we used was 1,000 times lower than that. So we went for a dose of 10 over 10 VG per kilogram. And we actually saw very promising therapeutic effects with that dose. However, if one considers the possibility of actually switching to an intravenous administration, which is another thing that we thought about, because we always thought about the fact that it would be really cool if we could actually do the things in parallel and actually see what are the effects on therapeutic efficacy. Is one approach better than the other in terms of efficacy, in terms of safety, in terms of bad distribution, and stuff like that? So this is something that we have in mind and we would like to do for our upcoming studies that would involve long-kidney tool rabbits. But this is something that we haven't tried yet. However, I guess for those studies, we are opting for the conventional doses of 13 to 14 over 10 VG per kilogram. But it would be also interesting to try different doses and actually see what the dose response would be. Thank you. Thank you for an excellent presentation. Thank you. That ends the session. The winner will be announced tomorrow at the award ceremony. And once again, congratulations to all the finalists for making it here and for excellent presentations. Let's give a round of applause for them and enjoy the rest of HRS.

Young Investigator Award

basic science

cardiac studies

PKP2 arrhythmogenic cardiomyopathy

channelopathies

AvaciaPib

long QT syndrome

gene therapy

therapeutic approaches