Activity Transcript

Christoph U. Correll, MD: Hello everyone. My name is Christoph Correll. I'm professor of psychiatry and molecular medicine at the Donald and Barbara Zucker School of Medicine at Hofstra Northwell, New York, and I'm also professor and chair of child and adolescent psychiatry at the Charité University Medicine in Berlin, Germany.

Welcome to this program titled, "Looking Outside of the Dopamine Box: Novel Receptor Pharmacology for Schizophrenia." You may have noticed that our goals in schizophrenia treatment have evolved over time. In the 1950s when antipsychotics were initially called major tranquilizers, we thought they were dampening the agitation and aggression of our patients. Aggression and self-harm were the primary targets. It took 2, 3 years for observant psychiatrists to actually notice that voices decreased, delusions improved, and that was the birth of the term antipsychotic. And also, at that time, the iteration of postsynaptic dopamine blockade as the identified mechanism of action of antipsychotic activity. However, as we will also review briefly in the side effect side of things, blocking dopamine postsynaptically had some price, and that was too much dampening of the reward system, secondary negative and cognitive symptoms, secondary depressive symptoms, but also, extrapyramidal side effects, motor side effects, Parkinsonism, dystonia, akathisia, as well as tardive dyskinesia.

The birth of the second-generation antipsychotics came from clozapine, the first truly atypical antipsychotic because it helped psychosis, but it didn't lead to neuroleptization. It didn't block dopamine receptors so much that animals or humans would become stiff. Here we thought that serotonin blockade, in addition to dopamine blockade, was relevant. And that gave birth to the iteration of multiple dopamine serotonin antagonists, as well as dopamine partial agonists, treating positive, maybe secondary negative symptoms, but also improving some of the side effects, motor side effects of the first generation agents, but also improving some of the weight gain issues that olanzapine and clozapine had, also prolactin elevation that comes with dopamine blockade. And with that, we entered into the realm of, not just focusing on symptoms, functionality, quality of life, remission and recovery became important. But over the last 7 decades, what has eluded our grasp has been to engage novel mechanisms of action that truly treats psychosis. So far, we had postsynaptic dopamine modulation, decreasing dopamine transmission via dopamine receptor blockade. And this talk will be about an alternative way to handle psychosis by a muscarinic agonism.

But let's first review what's the state of the affairs with the current dopamine antagonist medications. On this slide, you see, in black, the outcomes with chronic patients and in red, the outcomes in first episode schizophrenia patients. And when patients come to us, they're exacerbated, they're sick. We want to response, a reduction in symptoms and we achieve that, currently, in 1 in 5 to 2 out of 3 patients. But responses in the eye of the beholder, mostly, it's like 20% in chronic patients. In first episode patients who respond better, up to 90%, maybe responders, some other studies say maybe 80% can respond. We can even achieve a 50% reduction in the total psychopathology.

But response is a relative outcome. What we want is stability, and that's an absolute outcome called remission. And here we already run into trouble with our current treatment paradigm. Less than half of our chronic patients achieve 6 months of no more than mild for positive symptoms, and mild for negative symptoms. Yes, first episode patients do better, but the lower range here of 17% makes us already alert that even in first-episode patients, symptomatic remission is not always possible. But we don't really want to treat symptoms, do we? We want to treat people, and people are defined by what they do, how they interact. So recovery is the principle of symptom stability for a year or 2, but also functionality, adequate self-care, psychosocial interaction, and doing something meaningful with one's leisure time and, ideally, something in terms of educational work. And that's where we still fail most of our patients.

The median recovery rate for chronic patients was 13.5%. In first-episode patients, in the study we took the numbers from, it was 17%. Recently, in another meta-analysis upped that just to 21%, meaning 8 out of 10 patients, even after the first episode, will not achieve recovery. That's partially due to relapse that fuels treatment resistance and is partly due to non-response. 20% of first episode patients are treatment-resistant from the beginning. That doubles to about 40% in chronic patients. So the hope is, with additional mechanism of action, we can, A, change the early trajectory of the illness, not dampening dopamine too much, blocking open receptors that may up-regulate, but also maybe treating some of the patients that are currently only partial responders or even treatment-resistant.

On this slide, you can also see that the responder rates that we can achieve in first episode patients really drops, from 80%, with about a 20% at a minimum reduction from total psychopathology in the beginning that's minimally improved on the Clinical Global Impression Scale (CGI), or more. That drops to 51% when people have multiple episodes. Wanting, really, much or very much improvement on the CGI scale, that's a 50% reduction on the total Positive and Negative Syndrome Scale (PANSS) or Brief Psychiatric Rating Scale (BPRS). We have 52% of patients in the first episode that achieve that. And instead of 1 in 2, it drops to 1 in 4 to 1 in 5 patients when people have had multiple episodes. So what we need are treatments that are effective and safe and also give patients the idea of medication interest, not just adherence. I want to stay on this because I feel more back to normal.

I don't have some of the dampening in my dopamine system, less juice in the brain, less engagement with myself and life. And the hope is that different mechanisms of action, either by themselves or in combination with the current agents, could achieve better outcomes. And better outcomes are not just symptoms. When we ask patients, when we ask families, they have different agendas. And here, you see something that we've called life engagement. So what about the goals that your patients might have in the social and family realm? What about physical health, living arrangements, or occupational and financial life? It's important to identify these goals and link them to treatment success in our patients whenever possible.

But the problem is that efficacy symptom reduction is just the spearhead of the treatment. In order to be, not just efficacious, but effective, broadening the term to global outcomes, we need tolerability, we need adherence or medication interest in order to then get to the higher order outcomes of subjective wellbeing, quality of life, and ideally, an improvement in functional capacity, which will translate into better recovery rates. Medications alone will not be able to do that. With a biopsychosocial illness concept, and we should have a biopsychosocial treatment concept. So that means medications that are safe and efficacious need to be flanked by psycho-education as well as psychotherapy. But without medication, we cannot treat schizophrenia adequately. So what is the side effect profile of the current so-called antipsychotics that are all postsynaptic dopamine blockers? Even the partial agonist block postsynaptic dopamine receptors. On this slide, you see the 3 side effects that are related to postsynaptic dopamine blockade, extrapyramidal symptoms (EPS) or Parkinsonian side effects that are treated with anti-Parkinsonian medication on the left, akathisia in the middle, and prolactin elevation on the right.

And predictably so, first-generation agents are the worst. They're more at the bottom, farther away from the vertical line, which is placebo. And the "-DONEs", risperidone, paliperidone, ziprasidone, lurasidone are somewhere in the middle. And the "-PINEs", olanzapine, clozapine, chlorpromazine, quetiapine. And the partial agonists, the "-PIPs-," aripiprazole, brexpiprazole, and cariprazine are more on top because they don't block dopamine as much. They have partial agonism to counter that, or they have also inbuilt anticholinergic, anti-histaminergic blockade in terms of the pines. When we look at another side effect cluster, on the left-hand side, weight gain, in the middle, sedation. Now this is turned upside down. Now the "-PINEs" are at the bottom, clozapine, chlorpromazine, quetiapine because they have histaminergic blockade, which translates into weight gain and also cardiometabolic problems.

The "-DONEs" are in the middle and the "-PIPs-," the partial agonists, are more on top. And on the right-hand side we see QTc, re-polarization, problems in the heart when there's a prolongation, and we have currently 5 approved safe agents. The 3 "-PIPs," aripiprazole, brexpiprazole, cariprazine, lurasidone, as well as paliperidone. In terms of weight gain, we have a pretty clear gradation, but weight gain is not necessarily the life-shortening event. It's the metabolic consequences of weight gain. Cholesterol elevation, triglyceride elevation, HDL decrease and glucose and insulin resistance. And here you see, again, a similar gradation in terms of risk for first-generation, high potency drugs and partial agonist, antipsychotics as well as lurasidone and now, also, lumateperone being on the safer side, with other "-DONEs" and then the "-PINEs" having the biggest risk factors. So, wouldn't it be great to have effective antipsychotics that don't have these traditional side effects of dopamine receptor blockers?

That's the hope, that's the goal because these side effects are not just physical events, they're not just nuisance variables, they actually impair patients' functionality and quality of life. Here's a study of 435 patients who are asked to rank activating and sedating side effects, as well as weight gain and sexual side effects. How much, on a scale from 0 to 100, they impair their functioning, with 100 being the highest impairment. And you can see here, that between the high fifties to high sixties, patients feel impaired. That obviously makes them question, "Why should I take this medication? Maybe I'm better, maybe I don't need it anymore but I have these lingering effects on my function." And we know that side effects are associated with non-adherence. And here you can see that EPS and agitation have the highest association, followed by weight gain and prolactin elevation or sexual side effects and then sedation.

Reducing these side effects has the hope of decreasing non-adherence and decreasing non-adherence-related relapses and hospitalizations. Moreover, metabolic side effects and metabolic syndrome or diabetes, hypertension, have also been associated with decrease in cognition, where we already know that cognitive impairment is part of the illness of schizophrenia. We don't want to worsen that any further. So having agents that don't interfere with cardiometabolic health or even improve it would be very welcome. So, let's talk about model agents in development for schizophrenia, both what they are and how they work. There are currently 4 front runners that are targeting either total symptoms, Ulotaront which is a TAAR-1 agonist, which we will not focus on today. Two muscarinic agonists, xanomeline plus trospium, which is an M-1, M-4 muscarinic agonist with no D-2 binding and is paired with xanomeline, with the peripheral muscarinic antagonist, trospium.

And we have also emraclidine, which is an M-4 muscarinic for positive allosteric modulator. Now what is that? It means that we have an orthosteric site where our neurotransmitters sit, but an allosteric site is to the side of that. And when you go there, you can either increase the functioning of that receptor or dampen it. It's a fine-tuning effect. And a positive allosteric modulator increases the muscarinic agonism of M-4. And then we have pimavanserin as a fourth agent that is currently approved for Parkinson psychosis, but is currently under development for negative symptoms in schizophrenia. And again, that's an augmentation for predominant negative symptoms. We will not talk about that. So, let's focus on the muscarinic agonist couple or family here. First, let's go a little bit into the cholinergic system, as it is involved in schizophrenia.

Acetylcholine can, at low doses, link to muscarinic receptors and at higher doses, it goes to nicotinic receptors. Muscarinic receptors are G-protein coupled receptors. First, nicotinic receptors are ionotropic receptors. We'll not talk today about nicotinic receptors. So, let's talk about muscarinic receptors. There are 5 of them. There are 2 families, the odd-numbered and the even-numbered. The odd numbered, M-1, M-3 and M-5 stimulate. They excite. When you stimulate these receptors, downstream effect is excitation. The even receptors, even numbered M-2 and M-4, when you stimulate those, they lead to inhibition. Why? Because you're stimulating an auto receptor, a presynaptic auto receptor, which is the break in the system. So, M-1, which is part of xanomeline, when you stimulate it, you excite M-2 or M-4 in this case, which is part of xanomeline and also emraclidine. When you stimulate it, you downward inhibit. Where are these muscarinic receptors located? M-1, in the brain, particularly in the cortex and frontal cortex and the hippocampus.

Stimulation of the M-1 receptor has been shown in animal models to improve cognition. So, the hope is that M-1 stimulation in those areas could help patients' cognition. At the same time, there is some contribution to the caudate and the putamen. So we have here also, the basal ganglia, where the striatum is and where we think psychosis resides. M-4 has much less distribution in the hippocampus, frontal cortex, and cortex, and it's much more distributed, relatively so, in the basal ganglia, nucleus accumbens, and caudate, where we believe that its stimulation can decrease psychosis and we'll talk about that mechanism soon.

But these receptors are not just in the brain, particularly M-1 has contribution in the periphery. For example, in the salivary glands, meaning here, you can have hypersalivation, but also in the stomach, which results in nausea and vomiting, and also in the intestines. M-4 distribution is not in the stomach and also not in the salivary glands. So here, the contribution to peripheral side effects is minimal. What is the effect of muscarinic agonism on the M-4 receptor on psychosis? I would conceptualize that as the bottom up approach to improving psychotic symptoms. We've said that the odd-numbered muscarinic receptors inhibit, so we're stimulating the auto receptor in 2 places. One, in the mid-brain, which is in the ventral tegmental area. And by doing that, we're decreasing acetylcholine, which then decreases the inference dopamine output.

In addition, in the striatum itself, where we believe psychosis resides, and that is basically in the associated striatum. Here, muscarinic agonism on M-4 receptors stimulates the auto receptor, decreases acetylcholine in the inter neuronal contribution to psychosis. So the cholinergic inter-neuron releases less acetylcholine, which then decreases dopamine output from the afferent ventral tegmental area (VTA).

We have now 2 mechanisms that, in synergy, reduce dopamine tone without occupying the dopamine receptor at all. What does M-1 do? M-1, as we said, has excitatory effects when you stimulate it. Now M-1, however, sits on the GABA-ergic interneuron. So, when you stimulate the brake, it's the same as the auto receptor, you're stimulating the brake but here in the frontal cortex, you're stimulating gamma-aminobutyric acid (GABA) and we believe that in schizophrenia, there is an excitation inhibition disbalance and M-1 stimulation may actually rebalance that imbalance. How does it do that? It stimulates GABA, which sits on the excitatory neuron for glutamate. Remember, GABA is the brake in the system, glutamate is the gas pedal. But glutamate input increases dopamine output by stimulating GABA, reducing glutamatergic input. We are, again, in the midbrain (the ventral tegmental area), decreasing dopamine tone, thereby reducing psychosis.

That's theory? Well, let's look at the clinical data. The first study, published in the New England Journal of Medicine, tells us reviewers and editors felt this is big news, not just for psychiatry but for all of medicine, and a non-dopaminergic blocking agent treats psychosis. This was a phase 2 study, xanomeline plus trospium with titration over 1 week starting at 50 mg xanomeline plus 20 mg trospium twice a day, and then going up over a week to 125 mg of xanomeline plus 30 mg trospium twice a day. And if there were side effects, one could go down to 100 mg xanomeline plus 20 mg trospium in a combination pill. A 5-week study, 1-week titration, 4-week outcome assessment. And in this study with a respectable 17-point difference from baseline and only 6 points, roughly, on placebo, that was an effect size for the PANSS of 0.75.

Remember effect sizes? That's the difference in the means divided by the standard deviation, and the 0.75 means that you have 75% of a standard deviation difference. 0.2, 20% difference is a small effect size, 0.5, half a standard deviation is a medium effect size, and all of our antipsychotics in meta-analysis are at an effect size of about 0.35 to 0.55. So a respectable effect that was seen. But not only total symptoms improved, also positive symptoms, negative symptoms and also the CGI scale improved. What was the downside in terms of side effects? Was the trospium, the non-central active anticholinergic able to counter some of the pro cholinergic side effects of xanomeline? It was. But now we have some anticholinergic side effects and some pro-cholinergic side effects. 17% of patients had constipation. That's anticholinergic. 17% had nausea, compared with only 3% and 4% in placebo.

Dyspepsia 9%, vomiting 9%, but what's crucial here is, although there were side effects, in terms of dropout, only 2.1% of patients dropped out in the drug arm and only 2.1% dropped out in the placebo arm, for side effects. So it was, overall, well-tolerated. There was no significantly greater weight gain or metabolic abnormality with xanomeline trospium vs placebo. There was no change in Simpson Angus scale of extrapyramidal side effects or in the Barnes akathisia rating scale. There was also no relevant sedation or somnolence. In post-hoc analysis, responder rates were also looked at and looking at the 20% reduction from baseline, 60% in the xanomeline trospium arm reached that vs only 23% on placebo, which was a number needed to treat (NNT) of 3, until 1 more person reaches the response rate on drug vs placebo. That's a very good effect. Anything less than 10 is considered, not only statistically, but also clinically relevant.

If you look at 30% or 40% decrease from baseline, the NNT was 4 and 7, and even a 50% reduction, 1 out of 7 patients reached that with an NNT of 11. The effect sizes for positive symptoms, negative disorganized symptoms, hostility, and even anxiety depression ranged from 0.48 to 0.66. So very respectable.

This was the phase 2B study. Now, we're looking at the phase 3 study data and here the effect size for total symptoms was still a respectable 0.61, and that was, although the placebo response doubled from around 6 to under 12, but the drug effect also improved from 17 points to almost 22 points. There was also significant improvement in positive symptoms, negative symptoms, and the PANSS Marder negative symptom factor score. What about the side effects here? Slightly higher dropout rates, but basically 1.5% difference between xanomeline trospium, 7.1% vs placebo, 5.6% dropout for side effects. There was a higher rate of constipation and nausea, 21 and 19% vs before 17%. Dyspepsia also increased from 9% to 19%, and vomiting went from 9% to 14%. But again, the dropout rate, one and a half percent difference between xanomeline trospium and placebo. Again, no difference in weight, no difference in people gaining at least 7% of weight. And in this study also, prolactin was measured, which did not increase at all from baseline.

The other approach to the muscarinic agonism is not an M-1, M-4 agonist like xanomeline plus trospium, but is the M-4 positive allosteric modulator (PAM), emraclidine. And here we're looking at outcome data from a phase 1B study. Meaning, this study was designed, really, to have the pharmacokinetic aspect that was phase 1 of the study and to estimate an effect size, not hoping to see any relevant, significant effect against placebo because only 27 patients were enrolled in the placebo arm, in the 30 mg per day arm and then the 20 mg twice a day arm. Since there is no M-1 contribution here, no peripheral anticholinergic is added. This is not a combination pill. It is just emraclidine itself. And to the surprise of many, actually, this study was positive in both active arms. The 30 mg once a day, which is now carried forward into 2 phase 2 studies, as well as the 20 mg twice a day. Significant effects for total symptoms, positive symptoms, negative symptoms.

Again, validating that muscarinic agonism is a very viable, effective, and also safe treatment for the treatment of schizophrenia. Whether you stimulate the muscarinic receptors directly, leading to inhibition downstream, less acetylcholine output, less dopamine output, or whether you are enhancing the physiologic muscarinic tone by having a positive allosteric modulator. Here also, significant in effects in terms of improvement as a responder rate of 20% reduction, 30% or 50% reduction.

What were the side effects with the positive allosteric modulator, emraclidine? Again, very few patients dropped out for side effects, 7% on the emraclidine, 30 mg arm, once a day, 4% on 20 twice a day. There was nobody who dropped out in placebo, but except for headache, which was equally high, roughly, in all 3 arms. The other side effects were all single digit. Nausea, only 7%, back pain, 7%. There was also some blood work, dizziness, 4% in the 30 mg arm that was carried forward. Dry mouth was the only one double digit, but again, it's a small study. We'll have to await the results of the phase 3 study. Again, no signal for summons or sedation, no signal for weight gain or metabolic abnormalities, no prolactin elevation, so a very safe, non-postsynaptic dopamine blocking antipsychotic agent in the muscarinic agonist group.

Let me summarize. Dopamine blockade has revolutionized the treatment of schizophrenia. This is clear. We've been able to deinstitutionalize many patients. We've been able to treat psychosis that has really scarred many people's lives. However, approximately a third of patients, or even up to 40%, is currently treatment resistant with the postsynaptic dopamine blockade treatments that we have. Another third or so has only partial response and maybe only 17% achieve recovery.

Furthermore, anti-dopaminergic adverse effects, cardio metabolic burden and sedation limit the utility of current antipsychotics to varying degrees. Therefore, novel mechanisms of action, agents that treat psychosis without blocking postsynaptic dopamine receptors are very much in need. Currently, we have 2 agents under development and a third one likely going into trials. And the 2 here, xanomeline plus trospium, as well as emraclidine, have already yielded promising results. Xanomeline plus trospium, 2 positive trials with a third study coming in soon, and emraclidine with 1 phase 1B study that's positive, with 2 phase 2 studies underway.

Therefore, we hope that we will, in the future, have novel mechanism of action antipsychotics that either monotherapy or even in combination with current postsynaptic dopamine blockers, can treat patients better with less side effects to achieve their goals.

Thank you very much for your participation. Please continue on with the evaluation.

This transcript has not been copyedited.