Well, it is my pleasure to welcome you to San Diego and Harlem 2025. Please note that the visual reproduction of Harvard in 2025, either by video or still photography is strictly prohibited. So I'm very glad to see all of you this morning. And next to me is Dr. Veronika Ducey from Turin, Italy, who will co-chair this session. The first presentation is titled, Biomarkers of Autonomic Nervous System Activity, will be presented by Dr. Neil Herring. He has, as well as other speakers, 12 minutes followed by three minutes of discussion. So Dr. Herring, please. Excellent. Well, thank you very much for the invitation to talk about biomarkers of the autonomic nervous system. Now, when I was at medical school, the autonomic nervous system, as far as the heart was concerned, was pretty straightforward, really. The sympathetic nervous system through the stellate ganglia could be thought of mainly as like an accelerator, positive inotropy, chronotropy, leucotropy, dromatropy, whereas the parasympathetic nervous system, mainly via the vague eye, acted mainly like the brake, if you like. And whilst this is sort of broadly true, it's a gross oversimplification, to put it simply. So the way we think about the autonomic nervous system from the heart's perspective these days is as a sort of series of feedback loops, both down in the ganglionic plexi and the intrathoracic ganglia, also coming via the brainstem and higher centers. And I want to pay tribute at this point to the late Jeffrey Ardell, who sadly passed away earlier this year, who was a great friend, also a mentor, and is a co-author, senior co-author on the recent white paper that we've published in Journal of Physiology. But complex as perhaps this sort of diagram looks, even this is an oversimplification. If you look at some of the beautiful anatomical work coming out of UCLA and the innovation of the heart, it is incredibly complicated, particularly like some of the tissue clearing work, for example, that they've done, looking at the innovation of the heart. As Drew Armour put it, like an entire brain embedded in and around the heart, interacting with all the different cell types down there. And I would just like to flag the latest issue of Journal of Physiology, which is focused on cardiac neurobiology, and contains several white papers, reviews, and articles in this area. So given all this complexity and our sort of perhaps oversimplified interpretation of it, how have we approached biomarkers of this complex system? Well, you can do it through directly recording nerves. You can do it by measuring variability in end-organ behavior, I guess. The other approach, I suppose, is to look at circulating neurotransmitters in the blood and those levels. And to cut a long story short, I think it's safe to say that in myocardial infarction, hypertension, heart failure, then we really are slamming on the accelerator and taking our foot off the brake. And both increased sympathetic drive and reduced vagal tone in these conditions is linked to high levels of mortality and indeed arrhythmic death. So in 12 minutes, I've not got time to talk about all of this in detail. But just to tilt briefly about one or two of these things and some of the limitations. Direct nerve recording is generally via micro-neurography, such as done here with this sort of acupuncture needle to measure muscle sympathetic nerve activity. Or indeed, more recently, ultrasound guided at the survival of vagus. For any of you who have ever attempted this, it's extraordinarily technically challenging. And whilst it's been incredibly useful as a research tool, it's certainly not made its way into the realms of clinical medicine as yet. End-organ behavior, there are lots of measures here. Heart rate variability, even resting heart rate, heart rate recovery after exercise. And perhaps the best validated within the realm of heart rate variability is the standard deviation of the RR interval or NN interval, particularly through the UK Heart Study and Atrami studies that were done in the late 1990s. Again, linking SDNN with mortality, both in the heart failure population and indeed even in the first month after a myocardial infarction in the case of Atrami. Although it's still worth remembering that low ejection fraction in SDNN still didn't identify patients who would benefit from primary prevention ICDs and Dynamit, for example. So where are we with heart rate variability? Well, it's still useful. We still use it as a monitoring tool in patients with CRTs and ICDs. Crikey, we even have it in our smart watches and things like this, giving us information back. In terms of neurotransmitters, well, we've known since the late 70s and early 80s that plasma catecholamines, for example, are linked with mortality in patients presenting with sort of crashing heart failure, albeit being in the 1970s and 80s where all you got was diuretic oxygen and crossed fingers. The problem is that plasma catecholamines have a very short half-life. Even the act of taking a blood test can raise plasma catecholamines. So unless you're in crashing heart failure or have got a fear chromocytoma, the clinical utility again is somewhat limited. And of course, we have beta blockers. So trials like CIBIS-2 and others have shown that these improve mortality in the context of myocardial infarction and heart failure. So more recently, we've been looking at alternative biomarkers. So here's the accelerator pressed down and the foot off the brake, high levels of norepinephrine which were targeting beta blockers. But what about sympathetic co-transmitters? Neuropeptide Y, for example, is generally only released at very high levels of sympathetic stimulation. It has a longer half-life and is more stable, perhaps, than catecholamines. And in the last 20 years, we've been working on trying to understand what the cellular pathways are and the crosstalk between the different cell types are in this situation. We believe we've identified some of the key receptors that may be involved in triggering ventricular arrhythmia, influencing infarct size through the microvasculature, and even modulating remodeling after a myocardial infarction. And also, we've identified neuropeptide Y as perhaps a useful biomarker in these circumstances. I'm going to give you a little bit of our basic science data that's unpublished and then move on to the biomarker headline data that we've published recently. So we've tried to address this and approach it through a number of ways. We've used iPSC cardiomyocytes and patient data. We've also used rat models, for example, as well. So here's just a biopsy from a papillary muscle at the time of mitral valve surgery showing the expression of the main neuropeptide Y receptors in the heart, which thankfully are also expressed in iPSC cardiomyocytes. And through examining the signal pathways, we've used things like EPAC-based FRET sensors transfected into the cells to look at cyclic AMP signaling. If you look at the red line here, then these receptors are GI coupled. They reduce cyclic AMP. But the receptors have, you know, antagonists have different efficacies. We think perhaps the Y5 receptor in blue is the one that's most well coupled to that GI pathway. But it's not just GI. So they also appear to be phospholipase C coupled and increased levels of IP3 generation that are measured here by the breakdown product IP1. So that gives us a clue that NPY may be affecting electrophysiology, may also be affecting calcium handling potentially in those myocytes. So one way of looking at this is to culture a monolayer on a multi-electrode array of the iPSC cardiomyocytes and look at how it affects activation recovery and interval. And interestingly, NPY in these myocytes seems to actually prolong the iPSC recovery in ARI. And it seems that the Y5 antagonist seems to be the best target for addressing this. In fact, in the dish, it increases automaticity that we see as well. But is this just a quirk of iPSC cardiac myocytes? So to validate this, we go back to human data. And this is patients presenting with ST elevation myocardial infarction as part of the Oksami study who've been treated with primary PCI, where we measure the peripheral venous level of neuropeptide Y. And sure, we do see a prolongation of QTC interval in those patients with high peripheral venous NPY levels. We see similar and interesting patterns happening with calcium handling as well in these iPSC myocytes. But of course, we don't have the structural integrity of these of a whole heart. So to look at what's happening in terms of calcium in the whole heart, we've done this in the RET with optical mapping, for example, tying off the LAD and imaging the scar border zone in different areas to see how the relative amplitude of the calcium transient changes. And the interesting thing is that in remote, normal myocardium NPY increases the calcium transient, but then produces a big drop in the level of the calcium transient between the infarct border zone going to the normal area. And this heterogeneity in calcium through an infarcted heart can be highly proarrhythmic. Of course, in the RAT heart in the Langendorff mode, we can also do ischemia perfusion and look at how NPY induces ventricular arrhythmias. And indeed it does. And we can prevent this in particular via a Y1 antagonist. Again, is this RATology or is it actually relevant? Well, we go back to our ST elevation myocardial infarction patients, and indeed those with high levels of peripheral venous NPY are those that are experiencing generally VT and VF in the first 48 hours after their PCI on the coronary care unit. This work was published in European Heart Journal a number of years ago. But is it just arrhythmias? We think it's a bit more than that. So in a separate larger study where we recruited 164 STEMI patients and followed them up for five to six years, we actually found a significant increase in the level of heart failure and even all-cause mortality in this group. And this is adjusted for everything that we know that contributes towards those things. And it's not just those patients perhaps developing heart failure after their first myocardial infarction. We've also collaborated with the heart failure group, John McMurray in Glasgow, to look at a large heart failure cohort, 833 patients with a new diagnosis of heart failure with a range of ejection fractions. So about a third of these people, for example, also have HEF-PEF. And indeed, neuropeptide Y levels as well correlate even when adjusted for all-cause mortality. Not just all-cause mortality, but cardiovascular death. And the interesting thing is the levels of heart failure hospitalization are no different, making us think that it's MPY that's specifically driving arrhythmic death in this population. Indeed, you're better off having an above-the-median level of BMP than you are having an above-the-median level of MPY. And crikey, if you have both, then your event rates are incredibly high. It was really nice to see published just this year the Chinese group looking at MPY levels in new presentations of VT-storm, seen as this very much linked with our hypothesis. And again, adjusted hazard ratio of around four for survival in these patients with a MPY level measured immediately at the front door. So we think MPY is both potentially a helpful biomarker, but also there's some very interesting physiology and pathophysiology here that may bring in the possibility of new medications that could be adjuncts to beta blockers in this situation. Thank you very much for listening. Thank you for the presentation. We do have a minute or two for questioning. Anybody in the audience? So are there antagonists of MPY that is ready to be tested clinically? So there was one Y1 antagonist that was patented through a small pharma company in the 1990s. Now that pharma company no longer exists, so the answer is there perhaps was. The ones that are certainly readily available at the moment are used purely for, you know, research purposes. But I hope there will be, I think, is the answer, so I think it would be a good idea. The trick is getting the right receptor at the right time in the right condition. And that's a challenge. Oh, you have a question. Okay, go ahead. Yeah, a great talk as always, Neil. My question relates to the somewhat paradoxical response to MPY in the iPSC where, you know, you think maybe through Y5 there's actually a prolongation. And the question I had was whether if you applied MPY and norepinephrine at the same time or even MPY with epinephrine, whether you'd actually see a reduction in ARI. Yeah, that's a good question. Quite possibly. Again, our feeling with that observation is that it's linked to the GI coupling of the Y5 receptor. And the obvious candidate for that in terms of, you know, ARI, action potential duration, might be, you know, cyclic MPPK and modulation of IKS, for example. So that's patch clamp experiments that we've got ongoing at the moment. But you're right, what it does on its own versus what it does when you've treated a patient with a beta blocker already and taken away some of that sympathetic drive, you know, it might be paradoxical. That's why I say you need the right receptor at the right time and the right person. Okay. Because of time, we have to cut it here and thank you very much. Excellent, interesting presentation. Dr. Tan, Alex? Thank you. Good morning. So, do I hit? Hits hit start. That's the wrong talk, so let's end slideshow, and I hit this, nope, I hit my name, oh I see, okay, thank you. Good morning, I want to thank the organizers for this invitation. My topic today is to talk about the Intrinsic Cardiac Autonomic Nervous System as well as its modulation. So I think first of all we have to talk a little bit about basic anatomy and physiology of the Intrinsic Cardiac Autonomic Nervous System so that we can understand what are the ways we can modulate it and what are its effects in terms of arrhythmogenesis. And in general there are two ways to modulate it, we can either suppress it or we can stimulate it and in terms of suppression we can do it by destructive ways, meaning either surgical resection or ablation or non-destructive targeted chemical suppression. So the Cardiac Autonomic Nervous System is organized into extrinsic and intrinsic, meaning extrinsic everything outside of the heart and intrinsic everything within the heart in very simplistic terms. And these are connected to each other to and fro with efferent and afferent nerves. Efferent nerves connect the brain towards the heart and afferent nerves relay signal in the opposite direction. The Intrinsic Cardiac Autonomic Nervous System on the other hand are subepicardial clusters of neurons consisting of hundreds to thousands of cell bodies as well as the exons and dendrites that interconnect them. And together they form a network allowing the Intrinsic Cardiac Autonomic Nervous System, otherwise known as ganglionated plexi, to operate as a network to connect the brain and the heart. These ganglionated plexi are located most densely in the atria around the roots of the pulmonary veins and in the ventricle around the roots of the great vessels as well as in the atrial ventricular and interventricular groove. From a physiologic and neurochemical standpoint, these ganglionated plexi consist predominantly of efferent parasympathetic cholinergic cell bodies and efferent sympathetic adrenergic postganglionic fibers. However, they also contain other neuropeptides such as calcitonin gene-related peptide that are a marker for afferent or sensory nerves. And at the same time, they also contain other neuromodulators such as nitric oxide derivatives. Together, this myriad neurochemistry allows the GP, or ganglionated plexi, to function as trafficking centers for afferent and efferent nerves that interconnect the brain and the heart. And in this way, they can modulate cardiac electrophysiology as well as contractile and vascular function. So going to our first model, I think this is well known that sympathovagal nerve discharge is a trigger for AFib. This is data that comes from direct nerve recording of the stelic ganglion and the cardiac vagal nerve, showing in a canine model, showing that sympathovagal firing precipitates paroxysmal atrial fibrillation, and that the cryoablation of these cardiac nerves eliminated all episodes of PAF consistent with a causal relationship. Now these are extrinsic sympathetic and vagus nerves. What about the intrinsic cardiac nerves? This is elegant work done by Dr. Chen's group, first author Choi et al. And here they recorded not only from extrinsic sympathetic and vagus nerves, but also from the ligamental martial ganglionated plexi as well as the superior left ganglionated plexi adjacent to the left superior pulmonary vein. And you can see that in this case, all nerves, the intrinsic GP as well as the extrinsic sympathetic and vagus fires immediately preceding the onset of AFib. But at the same time, they also saw episodes where the GP fire alone and the extrinsic nerves are quiescent. So this kind of goes to the idea that these nerves behave as a network, but these GP can also function independently of central control. This is work from the Oklahoma group by Dr. Patterson that demonstrates the cellular mechanism in that sympathetic activation increases calcium transient, vagal activation shortens action potential duration. Together this combination leads to activation of the sodium calcium exchanger and promotion of EADs and triggered activity. That brings me to the first model of GP modulation, which is something that happens inadvertently during AFib ablation. And here from the same study in canines, they found that firing of the ligament marshal nerve activity contaminates the HO electrogram in the same area and gives the appearance of a complex fractionated HO electrogram. So investigators have targeted these complex fractionated HO electrograms with high frequency subthreshold scan to produce a heart rate response. And what they found was they also compared this with an anatomic approach. Now these ganglionate plexi have very conserved locations across species. They are generally located in the inferior aspects of the inferior veins and the superior or superior anterior aspects of the superior veins. And in comparing an anatomic versus an electrical approach, what they found was that the anatomic approach had better outcomes in terms of AFib prevention, in part because under anesthesia the ability to trigger a bradycardic response may be impeded. So again, work from a multi-centers including the Oklahoma group, this was a randomized controlled trial of over 200 patients that compared a GP ablation alone versus PVI alone versus PVI and GP ablation utilizing an anatomic approach as we mentioned. So you can see here in your typical wide antero circumferential ablation, the ablation lines intersect with the GP. But in this study, additional ablations were applied where the GP clustered. And what they found was that PVI plus GP had a better outcome than either PVI alone or GP alone. And even though there was more ablation lesions, there was no significant increase in post ablation left atrial flutters with this approach. There is corollary evidence that changes in heart rate and heart rate variability as a sign of autonomic modulation provide contributory benefit to AFib prevention. This is a study performed by Andrade from Canada, and they examined via a Holter monitor heart rate variability as well as average heart rate. And what they find is that in patients who had successful ablation, they had a greater elevation in heart rate, daytime and nighttime, and a more marked depression of the standard deviation of the NN intervals, indicating that autonomic modulation provides additional benefit to AFib ablation alone. I just wanted to talk a little bit about PFA. There's limited evidence about the effects of PFA on autonomics. Now the idea of the PFA is that it has tissue selectivity for the myocardium, and it's a non-thermal form of ablation. And in this case, this is a study done by Gerstenfeld et al, showing that compared to thermal ablation, PFA causes less change in the heart rate and less change in the standard deviation of the NN interval. Another study from New York showed that in 76% of PFA-ablated GP had preserved electrical response to high-frequency stimulation. So this suggests that PFA, by unknown mechanisms, whether tissue selective or a matter of tissue depth, seems to have less effects on autonomics than thermal ablation. I wanted to bring up neurocardiogenic syncope because the issue with AFib ablation is that AFib is a substrate disease in a large majority of patients by the time they present. And so the effects of autonomic modulation may be tempered by the presence of structural modeling. On the other hand, neurocardiogenic syncope are in younger patients who have normal hearts and represents a more pure autonomic condition. And in this case, patients with a very prominent cardio-inhibitory response, in this case AV have been shown to benefit from cardio-neuroablation by an anatomic approach or guided by high-frequency stim. And the most important panel that I wanted to show the audience here is this Kaplan-Meier curve because it shows the durability of autonomic ablation, of GP ablation. And you can see here the curve starts to only dip after about 60 months, suggesting that ablation of the GP has pretty durable effects. One of the issues with cardio-neuroablation is exactly that. It is not just neuroablation, but it is also cardiac ablation. So the second method for us to modulate the GP is by a more targeted chemical suppression. In this case, botulinum toxin, which is a reversible muscarinic antagonist, is injected into GP during coronary artery bypass surgery in order to prevent post-op AFib. So post-op AFib just means AFib manifesting for the first time in patients without prior clinical AFib who are undergoing open-heart surgery. And here you can see that the suppression of this GP by Botox significantly reduced the incidence of post-op AFib, which is typically defined as AFib within the first 30 days. But the results are mixed when applied to a larger population of CABG as well as valve. So with the shortage of time, I'm not able to talk about other methods of modulation, chemical modulation, and the other methods include calcium and glutamate. The idea of calcium and glutamate is that glutamate opens neuronal-only NMDA channels that promote calcium entry into cells. And this calcium entry causes calcium overload in neurons that leads to neuronal apoptosis. So our study here demonstrated a significant reduction in AFib vulnerability, not just electrically induced AFib, but also paroxysmal AFib. In closing, I wanted to add that the effects of GP modulation on ventricular arrhythmias is an area of active investigation, and at this point in time, it is unclear whether or not it is proarrhythmic or antiarrhythmic. The evidence in normal hearts is that GP modulation has no effects, but in the setting of acute myocardial ischemia, it may have some effects, whether pro or antiarrhythmic remains unclear. So in sum, I wanted to finish off with several conclusions. Number one, intrinsic cardiac autonomic nervous system is the final common pathway of neurocardiac control, representing the convergence of extrinsic and intrinsic neurons, afferent and efferent neurons. And GPs can be stimulated or suppressed by destructive or non-destructive methods with benefits for AFib, neurocardiogenic syncope. The antiarrhythmic mechanism is suppression of both sympathetic and parasympathetic components, and the effects of ventricular arrhythmias remain unclear, and the extent and optimal method for GP modulation also remains unclear. Some of the potential downsides include proarrhythmic nerve sprouting and neurodecentralization. So I want to thank the organizers for this opportunity. There's a question, please. Do I have time? All right. Alex, very nice, refreshing talk. Clinically, I'm doing PFA a lot, and since you mentioned about PFA, I used to do a lot of cryo-balloon ablation for AFib and destruct a lot of GP along the way. Now I do PFA exclusively. What I found is, although you have vagal response acutely during PFA, but for the next few months, you don't see this 20 beats per minute increment of heart rate after a cryo-balloon. We often see, especially in younger patients. But if PFA is more tissue selective, you don't destruct GP. There's a recent New England Journal of Medicine paper comparing PFA and cryo-balloon. You see PFA has almost a superiority of efficacy. How do you explain, besides you can argue that PFA has more tissue myocardial ablation, but without a GP modification, why PFA has a superiority in efficacy compared to cryo-balloon? As we all know, cryo-balloon has definitely a destruction of GP along the way. Thank you very much. I think that is a difficult question to answer. I think that patient selection is the key. I think that the effect of autonomics is greater in some patients than in others. So when applied generally to a general population with varying degrees of structural atrial myopathy, I think the effects are unclear in general. But I think in select patients, I think that it is beneficial. I think the trial that was done by the Sunny Post group, where they looked at RF and they looked at additional GP ablation. I think that probably is the strongest paper that I've come across that shows benefit of GP ablation. But those were also paroxysmal AFib as well. Yeah, please, I think just one, probably, okay. Thank you, thank you so much for your presentation. I think my question is similar. I see some of the mixed results on the GP ablation and I think about some of the GP that are hiding in the fat and some of the, most of the ablation category, I believe, is not designed specifically for fat ablation. I guess my question is, do you think if there's any value to for a new device or maybe new recipe for ablation that can be more effective for GP or would you think RF is sufficient for effective GP ablation? Well, I think if you take the data from neurocardiogenic syncope where RF is used, you will see that there are benefits that are durable, at least for about four years. As far as PFA is concerned, I think we don't have the ability to alter the pulse waveform and the characteristics. So with the current settings that we have, it may be less penetrative, I wanna say, or I don't know that it is necessarily tissue-specific as such. So I think that story is not clear at this point. All right, thank you so much. Thank you. I think now we have time to move to the next presentation. Thank you. Okay, thank you, Alex. It's a great presentation. Next presentation is titled Models and Effects. Oh, but Olu from UCLA. screen. Okay, good morning. Thank you all for coming to this very early session. I want to thank the organizers and moderators and all of you for being here. Here are my disclosures and my funding sources. I always like to start my talks by showing this image captured by my dear colleague Peter Hanna to illustrate just how densely innervated the heart is. What you're looking at is a piece of a human heart taken from the operating room and rendered optically clear as shown on the bottom part here. Let's see. Oh, you can't. Yeah, there you go. As shown here, and an image for a long period of time, and you can see just how densely innervated the heart is. I mean, it's just really full of nerves. And Dr. Herring showed this version earlier of how the system's organized and so did Dr. Tan. Essentially, information is taken from the heart on a beat-to- beat basis to this intrinsic cardiac nervous system that Dr. Tan just talked about, but also to various aspects of the neuraxis, all the way to the forebrain, and even to the frontal cortex. And it ensures that the heart's working on a beat-to-beat basis. It ensures that the heart's working very well. But in the setting of chronic pathology, which is why many of us are here, our group and many others, including Dr. Chan and Herring and Patterson, have shown that there are a lot of alterations that happen to the neural control of the heart. So that includes increased afferent activation, so that's information traveling from the heart to the nervous system, that drives a variety of processes within the stellar ganglion and other neural structures that ultimately causes dysfunctional neuronal activity. It leads primarily to two things, which again Dr. Herring talked about, increased sympathetic drive to the heart and decreased parasympathetic drive to the heart, ultimately causing arrhythmias and heart failure, etc. And so this is why we do sympathectomy, is to target these adverse remodeling that happens within a stellar ganglion. And for those who may not know, cardiac sympathetic denervation includes resection of the lower half of the stellar ganglion down assisted thoracopic surgery. And in this case report that we published, but there are many publications and examples of this, you have dramatic reductions in arrhythmias. Now, why does this help the heart? Dr. Chan and his colleagues actually quite a while ago showed that the propensity or the density of these nerve sprouts in the scar border zone of an injured heart actually relates to the likelihood of a patient dying from cardiac death, cardiac arrhythmic death, or non-cardiac death. And so we thought about trying to look at what the function of these nerves in the border zone are. And this is an EP audience, so I want to make sure I show some EP before I move on to the other aspect of my talk. And what happens is that in the scar border zone, when you turn on sympathetic activation, in this case right stellate ganglion stimulation, you can see just how dramatically altered the same region is during sympathetic excitation. And this is what's happening in our patients when they're excited, when they're running to catch a bus, is that the action potential duration, or in this case activation recovery interval, becomes dramatically altered. You see these very steep repolarization gradients, which we know are markers of very arrhythmogenic substrates. And that occurs. And by sympathetic denervation, we're able to reduce this change. The same thing happens with depolarization. And I won't belabor that. You can almost even see the emergence of what looks like a monomorphic VT circuit here during sympatho-excitation, again, in the scar border zone in an injured heart. So to put it a different way, our understanding of how neuromodulation works is that in a chronically injured heart, where there is neural remodeling that leads to chronic sympatho-excitation or withdrawal of parasympathetic tone, sympathetic blockade reduces your drive for arrhythmias. And things like vagal nerve stimulation or tragus stimulation will enhance that protective parasympathetic effect on the heart. But many of you may ask, well, why not just go upstream of this whole cascade and actually target whatever drives this neural remodeling process? And I mentioned that that's those increased afferents that essentially become enhanced after myocardial infarction or any form of cardiac injury. And we have ways of doing that. So these afferents here, these are nociceptive afferents that express a channel called TRPV1. And there's a toxin that you could actually inject. It actually grows where I'm from right here, one of the few places in the world where this grows. And you could actually target the TRPV1 channel with this compound. And it depletes the afferent nerve endings. And so a former postdoc in the lab, Kiyoshi Masuyama, did a study that he just recently completed where he took pigs with injured hearts, randomized them two weeks after an infarct. So this isn't at the time of infarct or before the infarct. He randomized these pigs two weeks after an infarct to saline or to this RTX injection. And then he studied them four weeks later. And many of us in here are used to looking at voltage maps. I'll just start by showing what a voltage map looked like. This is a follow-up study done by Abdallah Sarkar and Yumei Shen. And what they found is that when you look at a heart that has where very cool was injected, you see very heterogeneous scar border zone, a lot of scar. But once we treat with RTX, again, this is two weeks after the injury, you can see how spared the myocardium is, which suggests, again, that going upstream is important. Here's another sort of representation of the amount of scar and how much smaller it is. And here, looking at arrhythmia burden. So black here is the number of PVCs or non-sustained VT episodes that you get after giving cesium chloride in MI saline animals. But the blue trace shows what happens with RTX. You can see much less arrhythmias. And that's quantified here. And also, with program stimulation that we're mostly used to looking at, you can see that, again, in the RTX-treated animal, you're able to substantially reduce arrhythmogenesis. So it's no surprise, then, that in our clinical guidelines today, autonomic modulation now has a class 2B indication. I think it should actually come up higher in our clinical guidelines for managing patients with ventricular arrhythmias. And here's a list of approaches that are being used at the bedside today that decrease sympathetic drive or enhance parasympathetic drive to the heart. So in the remaining five minutes that I have, I wanted to switch gears a little bit and talk about a new direction that we're taking in terms of trying to understand the effects of cardiac sympathectomy. Now, baked into the English language are adages, sayings, proverbs that talk about the connection between the brain and the heart. So here's a word cloud that was generated to show that. And you hear things like young at heart, brokenhearted, heartfelt, heartsick, heartstrings. That suggests that what's happening at the level of the heart influences your brain, influences your emotion. And for example, some of our speakers today may or may not have taken a beta blocker prior to giving this talk as a way of really calming your nerves beforehand. And so we went back to some of our patients who had had sympathectomies. And cardiologists are often accused of not listening to their patients. So it actually took a psychiatrist, my colleague Saeed Khalsa, who went to a patient who had sympathetic denervation and was just talking to him. And before the sympathetic denervation, this patient was getting a lot of shocks. And he would say things like, I'm trying to calm down, trying to get my heart rate down, but it's almost impossible. I feel my heart beating faster. The patient will complain of these things before getting a shock. And after sympathetic denervation, look what happens. I can feel my heart racing. I'm starting to feel better. There's an absence of a rapid pulse. So I think that tells us that the sympathetic denervation we're doing is not only having an impact on the heart, but also actually on our patients' emotions, our patients' thinking, and their clinical state. In that patient that I showed just a second ago, here is a quantification of his anxiety over the period after the sympathectomy. You can see here that there was a complete decrease in his somatic, cognitive, and interoceptive, which is basically your sensation of the body scores afterwards. And it wasn't just in that patient. We took the next 10 patients who had sympathetic denervation, and we quantified anxiety, depression, and PTSD. And interestingly, we only saw a signature. As you can see here, green is before and gray is after. You can see here a reduction across all the patients in their anxiety scores, all 10 patients. But we didn't see that for depression, at least not as much, and definitely not for PTSD, suggesting that this spinal pathway that we're interrupting with sympathectomy is having some enceolytic effect. So there are parts of the brain that have been implicated in this work, I'm sorry, in this paradigm that I'm describing here. And we started to ask, well, how is this mediated? There are two pathways that allow you to get from the heart to the brain. That's up the vagus, nodus jugular ganglion, or through the spinal afferents that we are interrupting during sympathectomy. So two of my trainees decided that they were going to study this in a rodent model. And I can't emphasize enough how hard this is to do. In the last couple of minutes here, I'll just show you that they were able to do a right sympathectomy and a left sympathectomy, or bilateral sympathectomy. And you can see here very nicely that when they transect the right stellate ganglion, there's an immediate drop in heart rate. And that persists, actually, over time, over six weeks. So we are looking at this as a model of sympathectomy. And when you look at the heart in these patients, this is work done with my colleague, Jack Cheng. You can see here, control hearts have nice nerves. And here, the sympathectomy hearts, you can see that there's a reduction in the burden of sympathetic nerves. So we then developed a protocol to actually study this anxiety behavior in these mice to actually prove that they're, or at least to investigate this enceolytic effect. Here's the paradigm here. I won't belabor it too much, but we did a sympathectomy. And then we did behavior testing. To prove to you that we're getting sympathectomy, here is a quantification of the heart rate across animals. One of the things that the sympathetic chain does is innervate the upper eyelid. And even these mice showed a Horner syndrome, showing that we're clearly impacting the sympathetic chain. These nerves also control brown adipose tissue. And same thing, we're able to see a reduction in body temperature in the suprascapular region, but not, obviously, rectal temperature. So what happens to behavior? Here's a behavior test for mice called an open field test, where there's a bright light in the middle, a nervous mouse would stay in the corners, and a less anxious mouse would go and venture towards the middle. And you can see here that, I'll just go through this really quickly, that if you look at sham sympathectomy animals, you see that they avert the center region to some extent, and left sympathectomy animals also do. But when you look at right sympathectomy, you see, definitely, a tendency towards more exploration of the center zone, suggesting that they're actually less anxious. And this is very preliminary, by the way. The last data for this came just a few days ago. But you can see the trend with right sympathectomy, that these animals explore the center zone more than the sham and left sympathectomy animals, which suggest this might be due to heart rate. We did another test called the novelty suppressed feeding. Starve the mouse, then put it in also like an open field with a bright light with food. Anxious mice are going to not go to the food right away, but less anxious mice will go towards the food. And you can see here that the right and sympathectomy animals clearly showed a reduced feeding latency, meaning that they went towards the food pretty quickly, and they tended to eat more. So I'll wrap up here and conclude that cardiac function and innervation are very tightly linked, and neural remodeling is a critical aspect of what happens to our patients after the hearts are injured. And this important effect of anxiolysis following sympathectomy is an important effect of the procedure and something that I think we need to explore more in our patients. And with that, I'll stop and thank my lab, and I appreciate your attention. Thank you. Hello, this is really a very wonderful presentation. Years ago, Dr. Wei Zhongcai was my fellow, and he did a renal degeneration study in dogs. And we took out the stelic ganglion. We see the neural remodeling goes all the way back to the brain. So in your mouse, your rats, after stelic ganglion resection, are there any brainstem changes? Absolutely, great question. So ongoing studies, I'll tell you what we're doing is actually taking an approach called a whole brain or whole spinal cord, CFOS. Basically allows us to look in an unbiased way at lots of brain structures and brain areas. I didn't show it. We also have some animals where we've actually implanted electrodes in the brain and actually have an electrical signature for how various brain regions talk to each other. That signature has been shown to be active in depression. And interestingly, the mice that have myocardial infarction, they also show that same. OK, thank you very much. Thank you. It's a pleasure to introduce the last speaker, Dr. Stavraskis. He's going to talk about the models and the effect of vagal neuromodulation. But, you know, all the levels of the neuraxis can be targeted for autonomic modulation from the epicardial ganglia, the stellar ganglia, spinal cord. In the interest of time, I will focus on vagal stimulation and tragus stimulation. Over the last 20 years, we learned that the brain controls the heart and the immune system through the vagus nerve. The vagus nerve provides a two-way communication between the brain and the immune system. The vagal afferents travel through the celiac ganglia and up in the spleen, activate a subset of T-cells that express choline acetyltransferase, and they produce acetylcholine, which binds to the alpha-7 nicotinic acetylcholine receptor and suppresses inflammation. Now, inflammation plays a central role in HFPAF. Risk factors such as obesity, metabolic syndrome, and diabetes drive low-grade inflammation, also known as meta-inflammation, which leads to eventually fibrosis, diastolic dysfunction, and the clinical syndrome of heart failure. So, a few years ago, we hypothesized that chronic intermediate transcutaneous vagal stimulation may suppress inflammation and reverse diastolic dysfunction in HFPAF. In this proof-of-concept study, we took a rat model of HFPAF. These are called salt-sensitive rats. You feed them with high-salt diet, and they develop HFPAF. And we performed active and sham VNS 30 minutes every day. And we found that diastolic dysfunction was indeed suppressed or reversed in this model. And in the tissue level, we saw that fibrosis and inflammatory infiltration of the myocardium were also reversed by active stimulation compared to sham stimulation. We then went on to study the mechanism. We added two additional groups. One had received MLA, which is a selective inhibitor of the alpha-7-nicotinin colon receptor. In another group, we gave olmesartan, which is an antihypertensive. And as shown here, MLA reversed the effects of TVNS, whereas a decrease in blood pressure by olmesartan at the same levels as TVNS failed to reverse the effect. And we also saw changes in inflammatory genes in the myocardium. We then focused on cardiac resident macrophages. These are a subset of macrophages that are in the heart. And we hypothesized that the protective effect of TVNS in HFPAF is dependent on acetylcholine-mediated signaling, which reduces cardiac resident macrophages in their pro-inflammatory and fibrotic cytokine production. And then we asked the question, what is the role of cardiac resident macrophages subtypes in HFPAF? Which are the specific proteins produced by cardiac resident macrophages that promote HFPAF and could serve as potential therapeutic targets? And what is the effect of TVNS on modulating CRM and their secreted proteins? So we took a clinically relevant mouse model of HFPAF. These are mice that are fed with high-fat diet and are given L-NAME, a specific inhibitor of the nitric oxide synthase. We confirmed the effect of TVNS by slight slowing of the heart rate, increase in cycle length, as you see in this schematic here. And TVNS rescued the phenotype of HFPAF compared to the SHAM group. And there was, again, effect on fibrosis. We then performed single-cell RNA-seq and, again, focused on cardiac resident macrophages. We identified four groups, as previously described, TLF, MHC2, CCR2, and interferogamma responsive. And importantly, we saw that TVNS decreased the number of CCR2 macrophages in the heart. To prove the causative role of CCR2 macrophages, we did the same model in CCR2 knockout mice. And we found out that the effect CCR2 knockout had a better phenotype compared to wild type. And this was seen also with pro-inflammatory and pro-fibrotic proteins in the cardiac tissue. To identify potential targets, we looked at our data a little bit more closely. We found that this protein, SPP1, which encodes for osteopontin, and it's been shown to have a causative role in atrial fibrillation models, was increased in the HFPAF group and decreased by TVNS. So we also confirmed the specificity of this protein in CCR2 macrophages. And we performed the same experiment in a global SPP1 model. And again, the knockout of SPP1 resulted in an improved phenotype. We also looked at another protein, IGF1. And in order to prove the causative role of this, we performed the experiment in macrophage-specific IGF1 knockout, which attenuated the effect of TVNS. We also examined the cholinergic signaling. We used MLA, which inhibits the alpha-7 nicotinic acetylcholine receptor. And MLA attenuated the effect of TVNS. We confirmed the specificity of alpha-7 acetylcholine receptor on CCR2 macrophages by flow cytometry. And of the cardioaggressive macrophages, it was more prevalent in TLF macrophages. And in this last experiment, we examined the effect of loss of T cell-specific CHAT, which again attenuated the effect of TVNS. So how do we put this data together? There is a neurocardiac immune axis in HFPEF, where HFPEF is associated with increased SPP1. TVNS acts through CHAT T cells, alpha-7 acetylcholine receptors, and decreases SPP1. And it also increases the pro-reparative IGF1 to improve the phenotype. So in conclusion, the neuroimmune axis plays a central role in HFPEF. TVNS improves HFPEF by reducing pro-inflammatory SPP1-expressing CCR2 macrophages and inducing the expression of pro-reparative IGF1 and TLF MHC2 macrophages via acetylcholine alpha-7 nicotinic acetylcholine receptor signaling. These results were corroborated in a pilot randomized clinical trial, which was published a couple of years ago, in humans. And I will submit to you that this data help us identify new targets for therapy and may allow us to identify subgroups of patients that are more likely to respond to vagal neuromodulation. I would like to acknowledge our funding source, members of my lab, collaborators, and thank you all for your attention. Sorry, it was a little bit quick. Yeah, go for question. Thank you. Yeah, great talk. Thank you. STP1 plays a lot of different functions in normal physiology. Are you thinking about any side effects of suppression of STP1? We did not see any side effects. But, you know, mice are not humans, so we'll have to see. I'm not aware of any drug development for STP1, at least for cardiac uses. Okay, thank you. David Patterson, University of Oxford. Nice, really nice data. What's the clinical data look like on patients that have been vagotomized? You know, I'm thinking of transplant patients now. What's their cardiac phenotype like for heart and lung transplant patients? Do we know? That's a good question. I'm not aware of this data. Yeah, because it might be interesting to retrospectively go back and have a look at that, because clearly, you know, you're doing low-level stimulations through the tracheas, and there is a patient group that doesn't have the innovation. Just be curious to see what their phenotypes are like, cardiac-wise. Thank you. That's a great comment. How do you see the future, like in the next years, with discontinuity in humans? Because after VNS, there is a big deal. So if you were to say what's going to happen in the next, like, five, ten years in the clinical field? Yeah. I think, so two things. First, we need to improve the technology, and I think it's like a wireless device that can monitor physiological signals, apply AI, and then provide vagal neuromodulation in a closed-loop system. We need to, I don't think we have the exact biomarker that would provide this closed loop, but I think that's where we are headed. We can close. Okay, I think it's been a great session really. Thank you very much to all the speakers, it was a really great talk, thank you.

Harlem 2025

autonomic nervous system

cardiac neuromodulation

biomarkers

heart rate variability

neuropeptide Y

ganglionated plexi

cardiac sympathetic denervation

vagal nerve stimulation

neuro-immune axis