Dr. Grossberg (00:04): Hello, I'm Dr. George Grossberg. I'm professor and director in the division of Geriatric Psychiatry in the Department of Psychiatry and Behavioral Neuroscience at St. Louis University School of Medicine in St. Louis, Missouri. Welcome to our fifth and final episode in this podcast series on Best Practices in Parkinson's Disease Psychosis. This episode is titled Getting Current on Data for the Management of Parkinson's Disease Psychosis and Parkinson's Disease Dementia. [0:38] I'm very happy to announce that joining me today is Dr. Stuart Isaacson, who's director of the Parkinson's Disease and Movement Disorder Center of Boca Raton. [0:49] Welcome Stuart. Dr. Isaacson (00:51): Great to be joining you on this podcast today. Dr. Grossberg (00:55): Great. Today, we're going to review the latest data on agents that are used or under investigation for Parkinson's disease psychosis as well as Parkinson's disease dementia. But first, just a reminder of the definition of Parkinson's disease psychosis. What we're looking at are the NINDS/NIMH diagnostic criteria. Initially, we need a primary diagnosis of Parkinson's disease, which is based on the UK Parkinson's Disease Brain Bank criteria. So, we start with that diagnosis. (01:33): In order to qualify for psychosis in Parkinson's disease, we need to have at least one of the following symptoms: hallucinations, [1:41] which I think you know are abnormal perceptions without a physical stimulus, most often visual, or delusions, [1:51] which are firm, false beliefs that a patient has despite evidence to the contrary, most often paranoid or accusatory. Illusions [2:02] might be another option where the individual has misperceptions of real stimuli in the environment. And then a false sense of presence, [2:12] another psychotic symptom quite common in Parkinson's patients where they feel that there's someone maybe looking over their shoulder, there's someone actually there who in fact is not. The symptoms occur after a Parkinson's disease diagnosis and need to be recurrent almost day-to-day, or continuous, for at least one month. [2:35] This is what we need to have to make a diagnosis of Parkinson's disease psychosis. (02:42): We know that about 50% or more of Parkinson's patients will experience psychotic symptoms during the course of their disease. We also know that Parkinson's disease psychosis is associated with decreased activities of daily living, with a higher risk of depression, lower quality of life, increased caregiver burden, and most importantly increased nursing home placement, morbidity, as well as even increased mortality. So obviously, it's something that we want to recognize and treat promptly and effectively. (03:21): Now, as far as recommendations for treatment, the International Parkinson and Movement Disorder Society recommends, like all of us recommend, non-pharmacologic approaches should always be tried first. But if the non-pharmacologic approaches are not working and the patient's quality of life is dwindling, or if they present potentially a danger to self or to others, then there may be pharmacotherapies, some of which are more effective than others. The society looked at four drugs and the data that we have relative to these four drugs. (03:59): The first drug was clozapine. Clozapine is indeed effective for Parkinson's disease psychosis. It seems to be a relatively safe and well tolerated drug, but only, only with rigorous monitoring for agranulocytosis, which involves weekly blood work for the first six months and then monthly thereafter. So, it's not an easy drug to use, but can be clinically useful in Parkinson's disease psychosis. (04:30): [4:30] A couple of other atypical antipsychotics, olanzapine and quetiapine, were also looked at. Olanzapine was found to be not effective and really could make the parkinsonian symptoms worse, so it was felt that the risk was not acceptable. Quetiapine may be a useful drug. The problem is we don't have enough data. [4:53] We have insufficient evidence. So, the risk may be acceptable. There's no need for specialized monitoring. It may be possibly useful, but we really need more data. (05:05): [5:05] The newest drug, which is pimavanserin, was found to be efficacious and indeed it is approved for this indication with acceptable risk without need for specialized monitoring. So, this is the scenario for Parkinson's disease psychosis. Now Stuart, tell us a little bit about the mechanisms of action of these drugs and some of the latest data that we have. Dr. Isaacson (05:30): Well, thanks. That's a really a wonderful overview of some of our knowledge and some of the recommendations from this evidence-based review. It really reflects our understanding of the mechanism of action of antipsychotics [5:43] in general, and pimavanserin in particular. All these antipsychotics, the first generation, the second generation, typical, atypicals, all block dopamine D2 postsynaptic receptors. This is thought to be integral to the antipsychotic effect for a very long time. Of course, in Parkinson's disease and that dopamine depleted brain where patients have already lost about 50% of striatal dopamine, they're probably very, very sensitive to any type of antagonism of these dopamine D2 receptors and readily develop this problem with increased parkinsonism and motor symptoms. (06:25): When we think about the first generation of conventional antipsychotics, these primarily block dopamine D2 receptors, but they also block a number of off-target receptors. These can lead to off-target side effects, which are particularly troublesome in people with Parkinson's who already suffer non-motor symptoms reflecting more widespread degeneration in the Parkinson's disease neurodegenerative process. [6:50] Many of our patients without medications have daytime somnolence, have autonomic failure with orthostatic hypotension, developed cognitive impairment, and some of dementia. All these problems already occur. So, they may be even more sensitive to these off-target receptor side effects. (07:08): When the atypicals emerged, we were really hopeful that these would be helpful in our patients with Parkinson's disease psychosis, because they seem to have less affinity or less receptor occupancy, at least, at the postsynaptic D2 receptors. They also blocked serotonin 2A receptors. [7:26] Whether it was a balance or was a rapid binding and off of the dopamine D2 receptor was lower affinity, many of these theories, clinically, we saw less parkinsonism emerge in patients without Parkinson's disease. (07:42): But unfortunately, in people with Parkinson's disease, all the atypical antipsychotics, [7:47] they all block dopamine D2 receptors, and they all worsened motor Parkinson's in Parkinson's disease patients, except for [7:54] two, and that was clozapine and quetiapine, and now pimavanserin. Now clozapine and quetiapine block dopamine D2 receptors, but they seem to have less possibility of causing motor impairment. [8:12] It's still possible, and some of the studies have demonstrated this in some patients, pimavanserin does not have any affinity to the dopamine D2 receptor, and hence we wouldn't expect it to have motor impairment. But probably equally importantly is that the off-target side effects are not thought to be present with pimavanserin, because it doesn't have affinity to other receptors off target or on target, seems to predominantly have inverse agonism activity and antagonist activity at the serotonin 2A receptor, perhaps a bit at the 2C, [8:51] but not to these other receptors, so we wouldn't expect it to cause some of this off-target side effects, like somnolence or orthostatic hypotension. (09:01): These drugs have been evaluated. Clozapine was evaluated in two randomized placebo-controlled trials. Both of these were four weeks in duration, [9:09] one by the French Parkinson's study group in France and the other in the US by the Parkinson study group here. Both of these are using validated scale, demonstrated improvement in psychosis symptoms and without worsening motor symptoms in these four week trials. Somnolence was seen in over 50% of the patients though, and orthostatic hypotension in almost 20% of patients. [9:34] Clozapine, based on these two studies, the evidence-based review, as Dr. Grossberg pointed out, leads to this being felt to be clinically useful by the Movement Disorder Society. (09:48): Quetiapine, on the other hand, has not been subjected to large randomized placebo-controlled trials. Anecdotally, it seems to have an antipsychotic effect, as one would expect from an antipsychotic. But in several smaller trials, [10:01] some very small or some a little bit larger, clear efficacy versus placebo was not consistently demonstrated. Contrary, perhaps, to some of our anecdotal experience. One trial actually demonstrated and suggested motor worsening in some of the patients. Orthostatic hypotension and somnolence can be limiting to titrating to effective doses. Many of these patients who do respond to quetiapine use a low dose, [10:31] sometimes call it low dose quetiapine. But because we don't have the evidence base for efficacy and the tolerability can be problematic in some patients, this was felt to, in the evidence-based review, to be possibly useful. (10:45): So, we were encouraged when pimavanserin in a six-week pivotal trial demonstrated, on a validated scale, significantly improved symptoms of psychosis, hallucinations, and delusions. [10:56] In looking very clearly at the motor scale that we use in the UPDRS, there was no worsening of motor symptoms [11:05] and no difference from placebo. Remarkably, somnolence was not significantly increased over placebo. And orthostatic hypotension, [11:14] measured as an adverse event, was actually less common in the pimavanserin treated patients. [11:20] I think most remarkably in this trial and somewhat very unusual in trials of any psychosis in that this study of pimavanserin in Parkinson's psychosis, 14% had a complete resolution of all psychotic symptoms compared to about 1% in placebo. [11:37] So, this was very encouraging. (11:40): Since that time and the commercial approval, a lot of work has gone on to understand what does this mean clinically to our patients [11:49] that we see every day who live with Parkinson's disease and experiencing psychosis symptoms, which is more than half of our patients in most assessments. We've tried to see, because as an atypical antipsychotic, it shares the boxed warning for increased mortality in elderly patients with dementia-related psychosis, do we have evidence concerning safety and mortality? There've been a number of studies looking at Medicare databases and some meta-analysis, [12:19] so we can go through some of this newer information, because I think it's reassuring to see this in total. (12:26): For instance, one study evaluated the efficacy and safety of pimavanserin compared with placebo in a meta-analysis of four trials. They demonstrated in this meta-analysis a significant reduction in hallucinations and delusions using those subscales of the validated SAP scale. [12:42] Now this was a significant of the P-value of 0.009, and no change in the motor symptoms using that UPDRS scale. And it seemed to be protective against hypotension [12:52] with the P-value of 0.001. So, that was an encouraging trial. (12:57): Another study was an observational study that looked at a Medicare database and compared [13:03] Medicare beneficiated who initiated on pimavanserin, over 3000 of them, versus over 18,000 who were initiated on a different atypical antipsychotic, quetiapine being the most common of that group, about 78% of patients. They looked at the first 180 days after initiating [13:22] either pimavanserin or one of these other atypical antipsychotics, most commonly quetiapine, and found that the mortality was actually approximately 30% lower with pimavanserin, [13:35] the hazard ratio of 0.65, than with the other antipsychotics. This translates to about one excess death per 30 antipsychotic treated patients. Beyond 180 days, or among nursing home residents, this difference did not persist, [13:53] but at least not first 180 days, they demonstrated this lower mortality. (13:57): Another retrospective cohort study that was based on a large commercial insurance database of new users of pimavanserin, about 775, versus [14:08] one of the antipsychotics quetiapine and clozapine, about 4,500 patients, and other antipsychotics, about almost 1300 patients. And then, looking at these three groups, there was no difference in mortality risk between pimavanserin and other antipsychotics, [14:26] so they did not demonstrate an increased mortality with pimavanserin. (14:32): More recently, there's been [14:33] a comparative review, a systemic review and meta-analysis that aimed to compare the efficacy, the safety, and the acceptability of pimavanserin, quetiapine, clozapine, [14:46] and then some other atypicals that we know block dopamine receptors and worsen motor function, olanzapine, trazodone, and risperidone. This looked at [14:54] meta-analysis of 19 studies that had been performed in Parkinson's disease psychosis. Using Clinical Global Impression scales for severity, CGIS, pimavanserin and clozapine were noted to significantly improve Parkinson's psychosis symptoms compared with placebo. Pimavanserin also improved psychotic symptoms based on the SAPS PD score, but again that was the only study that used that. (15:21): Clozapine, quetiapine, pimavanserin all did not worsen motor function in this meta-analysis, although quetiapine significantly impaired cognition based on Mini Mental scores. [15:34] This meta-analysis concluded that both Pimavanserin and clozapine improved Parkinson's psychosis symptoms without impairing motor function, and suggested quetiapine should be avoided in patients with Parkinson's psychosis who also had impaired cognition. (15:49): Another systemic review and a Bayesian network meta-analysis analyzed 17 studies assessing efficacy [15:56/15:57] based on the scale used in the trial, CGIS, or a Brief Psychosis Rating Scale, or the SAPS PD scale in the pimavanserin trials, as well as safety based on UPDRS and AE dropouts, and found that clozapine and pimavanserin were both efficacious. They had low impact on worsening motor function [16:21] that wasn't seen, and that dropouts because of AEs had an odds ratio of 2.9 for clozapine and 2.2 for pimavanserin. (16:32): Another study recently looked not at mortality but at other important factors for people with Parkinson's disease who have a progressive worsening of motor function and hence are at risk of falls. And they also looked at hospitalizations, [16:47] because many of our patients, as they age, are in hospitals for a variety of reasons, either a fall or other medical conditions. And this cohort study looked at 112 patients who initiated pimavanserin, and almost 1,000 [17:02] comparator initiators who initiated on clozapine, quetiapine, risperidone, olanzapine, aripiprazole and brexpiprazole. And what they found was that the incident rate for falls and fractures was much lower for pimavanserin, 17.8 versus 40.8 for the comparators. That was unknown if this reflected orthostatic hypotension or worsening parkinsonism leading to falls. But it was a significant difference in falls and fractures. We looked at the match incidence rate ratio and they found an incident rate ratio of 0.71 for pimavanserin versus the comparative initiators. [17:41] So, this was a interesting study. (17:44): Another one looked at health care utilization and mortality in a retrospective cohort study. They found that 12.1% of patients with Parkinson's psychosis used long-term custodial care within one year of being diagnosed with Parkinson's psychosis. This is a relative risk of 3.38 for long-term custodial care and 1.34 for death. The health care utilization and costs were higher for patients with Parkinson's psychosis than for those without, suggesting that people with Parkinson's disease, once they develop psychosis, are very different than other Parkinson's disease patients [18:20] who do not have psychosis, have increased mortality, increased health care utilization, as well as some of these other problems we've discussed. (18:30): This study [18:30] I found interesting as well. It was a study that looked at 844 new pimavanserin users and compared them to 2,500 new quetiapine users to assess hospitalization and mortality, [18:45] comparing pimavanserin and quetiapine. They found that the adjusted risk for hospitalization [18:52] for pimavanserin versus quetiapine was much lower for pimavanserin, 0.59 at 30 days, 0.56 at 90 days, 0.63 at 180 days. And in a year period, the risk was 0.70 for pimavanserin versus quetiapine. The most common reasons for hospitalizations was traumatic injury, probably resulting from a fall, and sepsis from an infection. Hospitalizations for heart-related issues seemed higher with pimavanserin though. And the adjusted hazard ratio for all cause mortality [19:25] for pimavanserin versus quetiapine was favored pimavanserin, 0.773 at 90 days, 0.80 at 180 days, and at one year, 0.94. (19:35): So, take it all together, I think we now have a lot of long-term information from looking at commercial use databases, and Medicare, and commercial insurance. And this augments the recent publications of the long-term extension trials that were carried out after the pivotal trials, looking out over seven years in some patients, that really demonstrated no new safety or tolerability concerns. It augments the information that the FDA has published, that they did not find new things in the commercial realm. [20:06] And I think it gives us some sense that we have a drug that has demonstrated efficacy, safety, and tolerability, both in the short term and now also looking out in the longer term. So, I think this is all very interesting. George, [20:21] we're good [inaudible 00:20:23] this data here for people with Parkinson's disease psychosis. Do you have any data for patients who also have Parkinson's disease dementia? Dr. Grossberg (20:31): Thank you, Stuart, for that comprehensive overview. Just to briefly look at Parkinson's disease dementia patients who are also having psychotic symptomatology, the International Parkinson's and Movement Disorder Society recommendations looked at basically the three cholinesterase inhibitors in wide use, donepezil, rivastigmine, galantamine as well as the NMDA receptor antagonist memantine. What they found was, [21:00] was that only rivastigmine was in fact efficacious, had acceptable risk without need for special monitoring, and indeed was clinically useful. And this resulted in FDA approval for rivastigmine as a treatment for patients with Parkinson's disease dementia. [21:21] We're going to look at more specifically psychosis as well. But rivastigmine in patients with Parkinson's disease dementia, with or without psychotic symptoms or hallucinations, was looked at in a large study of 536 patients, and it was found that rivastigmine significantly improved Neuropsychiatric Inventory, or behavior scores, whether psychotic symptomatology as well as agitation and digression. In fact, there was greater therapeutic benefit seen among patients with hallucinations. (21:58): We also have a sub-analysis of the [22:00] now famous HARMONY study with pimavanserin, which focused on Parkinson's disease psychosis, where we looked at a little over 15% of those patients in that study also met criteria for Parkinson's disease dementia. And these patients, in fact, also responded quite well versus placebo. Their likelihood of relapse was 13% in the of pimavanserin group, [22:28] these were responders to pimavanserin, versus 28%, more than double, risk for relapse in the placebo group. Let's keep in mind that there were some adverse events with pimavanserin, headache, constipation, UTI, and asymptomatic QTc prolongation. (22:51): As we finish up, Stuart, I just wanted to get your input on a case I had very recently in the clinic. This is a gentleman in his late seventies with moderately advanced Parkinson's disease who's been having really, really traumatic psychotic symptomatology that are really impairing his quality of life and the quality of life of his care partner, [23:13] in this case, his wife. I actually started him on pimavanserin. He's been on the optimal dose of 34 milligrams for three to four weeks now. We're seeing a very marginal effect, a minimal effect. So, what are the treatment options for this individual? If this was your patient in your clinic, what would you be thinking about? Dr. Isaacson (23:37): It can be a very difficult situation, because we're trying to balance both maintaining good control of motor symptoms, especially reducing the risk of falls, and also treating psychosis symptoms [23:47] can be devastating at home and allowing them to stay at home. So, we know, with pimavanserin, that we see the full effect by six weeks, and we don't see a greater effect by 10 weeks or later. [23:59] So, I think six weeks in patients had more improvement from four to six weeks. So, I'll be hopeful primarily that there'll be an increasing effect over until six weeks. I think that we have to look at his Parkinson's medicines [24:13] and see if there's any opportunity to reduce some of the drive. I think, in Parkinson's psychosis, we think perhaps in many patients, there's a serotonergic two-way overactivity, the fire burning, and then the dopamine drugs adds fuel to the fire. So, maybe temporarily we can take away some of the fuel of the dopaminergic medicines if they tolerate it motor-wise. So, that would be a second thing. (24:36): It can be difficult in the short term. Perhaps we have to add or change the medication. We know that clozapine has a demonstrated efficacy but requires blood monitoring [24:47] once a week for six months, every other week for six months, and monthly thereafter. They have to get into a registry before they can get it. So, it may take you several weeks. This is a patient I might begin to think about the registry for clozapine and get that underway so that if, after six weeks, there's not a good effect from pimavanserin, we can think about switching it and they'll be ready to get it. (25:07): And then, sometimes we add quetiapine, [25:09] which was found to be possibly useful in the evidence-based review, to the pimavanserin. Because pimavanserin only blocks 2A, it might allow us to use a lower dose of a different antipsychotic. [25:21] So, I wouldn't stop the pimavanserin. I would probably add on a low dose of quetiapine. Sometimes, we give pimavanserin [25:28] at night, in the morning, and quetiapine at night. Unfortunately, we don't have any data to suggest the safety or interactions, but we wouldn't think there would be any. We know both can prolonge QTc, quetiapine by six to eight milliseconds on average, pimavanserin five to eight milliseconds. This might be additive. So, sometimes we'll check an EKG and make sure that their range is below or closer to 400, so they don't approach that 500 millisecond threshold. But often, I would add a low dose of quetiapine at night and hope that that would be temporary, and that the pimavanserin will continue to improve. (26:03): As time goes on, we have some empiric medications. You mentioned rivastigmine. We often add that. There's been some work with ondansetron and other medicines. And we look forward to other non-dopaminergic medications. The TAAR1 agonist that's under development. We've looked at that recently in Parkinson's psychosis and found some hope. Because, when you look at the 25% of patients in the pimavanserin pivotal trial that did not have improvement, because it's such a high affinity serotonin 2A receptor inverse agonist, we might assume that those patients are not driven by serotonin 2A. They must be driven by a different network of causing psychosis. So, I think about that as well. Dr. Grossberg (26:43): That's a great discussion. I appreciate that. The only thing I would add is that we'd be also thinking about electroconvulsive therapy, if all else fails. But thank you again, Stuart. It's great to have you with us. I want to thank our audience for participating in this activity, and please continue to answer the questions that follow and complete the evaluation.