The Role of Antipsychotic Therapeutic Drug Monitoring in Clinical Decision Making: Example of Clozapine

Speaker 1 (00:00:12) Welcome everyone. I am very excited to welcome you to the role of antipsychotic therapeutic drug monitoring in clinical decision making, the example of clozapine. I have an excellent faculty with me today to present this very exciting topic. I'd like to go through the agenda, a few housekeeping items. First of all, I'd like to thank our partner Saladax for supporting this symposium today. If we could all, please give Saladax a big round of applause. They're all in the back. For having us today, I appreciate it. Thank you so much. So today we're going to be focusing on why, when, and how to use antipsychotic plasma levels. We're going to be talking about issues with laboratory reliability for our clozapine assays, and then we're going to be going through some clinical practical applications. And we hope to have a lot of engagement throughout the discussion. So, we're going to transition to some time for questions and answers and discussion with the panel later at the end of the talk.
(00:01:10) So housekeeping, handouts are available. Everyone should have received one. I was instructed to do this as a flight attendant, so everyone should have received one. Index cards are in the back. Please notice the last page has a nice rip out so you can rip it out. There are pens so you can submit any questions so we can get them to the right faculty at the end. Additionally, there's a nice survey. We'd like you to know this is the first, hopefully, in a series of education around therapeutic drug monitoring. So, if there are ideas, questions, anything that you would like to see moving forward, please let us know. We are so happy to be able to bring this to you. So, make sure to fill out that survey. So, without further ado we'd like to introduce the esteemed faculty.
(00:01:53) We have Dr. Jonathan Meyer. He's a clinical professor at the University of California, San Diego. We have Dr. Rob Cortez, he's the director of the clinical research program for psychosis at Grady Health System. He's also an associate professor for psychiatry and behavioral sciences. Also, the psychiatrist on the SMI advisor. And Deanna Kelly, the empower professor of psychiatry at the University of Maryland strategic partnership, empowering the state. She's the acting director for the treatment research program at the Maryland Psychiatric Research Center. And I'm Megan Ehret. I'm a professor at the University of Maryland School of Pharmacy and serve as the SMI pharmacy consultant, the SMI advisor. So, without further ado, we are going to turn it over. We're doing objectives? Excellent.
(00:02:40) We're going to describe the therapeutic drug monitoring, providing evidence-based and personalized psycho pharmacologic treatment. We're going to review some of the evidence-based derivatives of antipsychotic level response thresholds. We're going to talk about the point of futility, how to use levels for managing efficacy related to decisions in patients with clozapine. We're going to look at some of the limitations of the LCMS method for ascertaining clozapine plasma levels and the clinical value of a new immunoassay method. And then we're going to open it up and discuss some cases with the utilization of therapeutic drug monitoring in day-to-day clinical scenarios. So please join me in welcoming Dr. Jonathan Meyer.
Speaker 2 (00:03:23) Thank you all very much. And Manny in the back, if you could start the timer, that would be great. If you all can start the timer there, thank you all so much. Antipsychotic levels are like other drug levels in some ways, but in some ways, they are very different. I think most of us are familiar with the value of TDM, I know I'm preaching to the choir. But for antipsychotic levels, it's not something which has been commonly used. And we're really here to try to change that culture. Using clozapine is probably the best example, but we feel like this generalizes to other antipsychotics as well. I give you the economic argument right off the bat. So that analysis of 1580 was done now almost half a decade ago. You can say $1,580 per year per patient if you measure antipsychotic levels. And part of that relates to the enormous economic cost of treating patients—in particular with schizophrenia.
(00:04:18) Why else do we get drug monitoring? I think you know the answers, there's biological variations in all the things which affect absorption, distribution, metabolism, etcetera. Some people don't respond also and need to know when to cut it off and maybe go to something different. Surprisingly, haloperidol doesn't solve the world's problems with schizophrenia, yet there's some people you could literally give them as haloperidol as you are allowed to prescribe legally. They won't get staph, they won't get akathisia, and they won't get better. You need to know when to cut it off and move to something different. There was a consensus paper which was published two years ago with ASCP and a German group AGMP. The [inaudible 00:04:58] t working group has been publishing these papers now for over a decade and a half. And ASCP finally partnered with them because they recognized as we were getting this new technology. FDA was skeptical that people thought this was of value.
(00:05:13) These are the agents which we think are the most evidence based in terms of getting levels. But that doesn't mean it's not a value for other anti psychotics, it clearly is. But this is where we have the most data on these molecules. We're going to talk about clozapine, but there's reasons to get levels on all patients in many ways, especially those on oral therapy. So, here's all the reasons, and over the next one and a half hours, we're going to go through each one of them. I'm joking. You can read. You have the handout. There's lots of reasons to get these levels. Okay. I'm just going to say that, and I'm going to move on to more practical stuff. One of the issues, as many of you know, who here has ever ordered a clozapine or other antipsychotic level? A lot of people. Who here has ever gotten the level back in less than a week?
(00:05:59) Yeah, almost nobody. A few. And if you're lucky to be one of these institutions which have this new technology, you're very fortunate, but that's not true for most. And that is a source of frustration. Even if you think it's of value, the fact that you have to wait one or two weeks really, really interferes with their clinical decision making. And this is unique to antipsychotic level reporting. If I order a lithium level on almost any lab on the world, they'll probably give me an upper limit of 1.2, 1.3, 1.4, it kind of depends. Small, small variation. I can tell you, I wrote a paper a few years ago on what labs report for upper limits. I'll give you an extreme example for fluphenazine, first generation agent, well studied. One lab said the upper limit was 2.0, another lab said it was 10.0. Five-fold variation. The labs are the least reliable sources of information in terms of what is a useful therapeutic range.
(00:06:58) And that is a big part of the problem for us as a field. We got to get everybody on the same page and get to some agreement about what is a response threshold and what is the point which I call the point of utility where we're like, you're wasting your time. Even if they tolerate it going beyond that, you have less than five chances of response. Here's some info from various labs about clozapine. It just illustrates the point of the variability you get out there. You all know this, we do 12-hour troughs for a lot of drug levels, and it has nothing to do with the drug itself. It has everything to do with convenience, which easiest, I hate to say it, for us. If you're an inpatient unit, what's most convenient? 12-hour troughs. We do that by convention. That's fine because, in the end, we're all speaking the same language.
(00:07:44) So, at the state hospital I work, we actually did a DUE on people using lithium. Where all these people are giving folks lithium at noon, how can they possibly make sense of that level drawn at 6:00 AM when they're getting an 18 hour trough. And that's just something occasionally can be a problem and getting people in roughly 12 hours after their dose. We certainly know that as clozapine is titrated, there's value in getting levels partly because you want to see how close am I to my initial target, which will be the response threshold of 350 nanogram per ml. It just gives you some of that information. Also, you get a sense of people who are actually taking it. As I joke, occasionally there are people who don't take their oral medications, not my patients, of course, other people's patients, but you just want to know, am I making progress? Where are we going?
(00:08:33) If they're perhaps very slow now you know. Maybe you can slow up the titration and help manage tolerability concerns so they don't complain and say, especially if for clozapine, they don't want to take it. What are you going to give them? I don't know. You don't have a lot of options there. And it's really important particularly for this and a psychotic to get on top of some of this early, get that information, see what your dose response curve looks like for that person and then make dose adjustment decisions for the future titration. So how do we use this for managing schizophrenia and that's for efficacy decisions the most robust area for any psychotic levels. At some point we might have them for other disease states, maybe adjunctive use for unipolar depression, maybe even for acute, maybe if anyone's really interested, but this is where most of the literature is and where I think it's of most use right now in 2022, I put somebody on clozapine or any psychotic. When do I stop giving them more?
(00:09:34) Well, number one is they got a lot better. Okay. So, oh, you're a lot better. I'm not going to give you more. I've done my job. You're a lot better by whatever... I wrote a book on clozapine. But eventually people as kind of a field converge around the number. I would pick a number target. It just helps you in deciding what to do when people are non-responders. This is the best type of analysis it's called a ROC curve. This is for olanzapine. And what you do is you try to separate out the true responders from the non-responders mathematically. This analysis came up with a therapeutic response threshold of 23 nanogram per ml for olanzapine. Guess what? There's two other papers on this. They show the exact same number, which is really cool with slightly different populations and methods.
(00:10:21) How do you use this number though? Well, the way you use this number is in the context of knowing what the range is of what you're shooting for. Clozapine, we think the response threshold is at 350 and I'll give you example of how to implement that. Below that your chances of response are not zero. I've consulted for over a dozen years for the state hospital in California, nobody calls me up and said, I have a person who responded really well to a clozapine level of 275, my God, what should I do? Nobody calls me for that. They call me when the person's not responding. And then my first question is what is their level? I mean, they know [inaudible 00:10:58] to contact the Psycho Pharm service. Now, the state hospital without drug levels cause I know the first question we're going to ask them is, what is the level? Not only for clozapine but for many other antipsychotics.
(00:11:09) As you get beyond a thousand, it's not inherently unsafe, but your chances of getting people to respond are really, really remote. Occasionally, maybe in the state hospital we'll go a little bit beyond, but we know that we're really at the end of the curve there. And the chances are pretty small. And that also is consistent with the upper limit from the consensus paper, they call it the lab alert level. It doesn't mean you have to jump up and down if the level's 1100, it just means you have to actually have to go that thing, which nobody wants to do, which is talk to the patient. And you're thinking, God, I got to talk to the patient. You just have to assess the patient. How you doing? I'm fine. We get spuriously high levels for clozapine all the time and we'll hear some examples from Dr. Kelly about what that issue is. But the first rule of thumb is, go assess the patient. They're like, I'm fine. Your level is 1200. I'm fine. All right, well, let's see if it makes sense. And then we'll take some appropriate action.
(00:12:01) This again is just an initial target for people who are non-responders, it's just your initial goal. And I have this language down below. It is incorrect to state that the patient's plasma level is therapeutic because it's above the initial response threshold. We get these consults all the time. People have fixed in their mind that number of 350, which is good, at least they have a number, but they misunderstand how to use it. Essentially, they say, well, it's beyond 350 so we get a consult, Hey, this guy's level is 550 for clozapine and he's a non-responder. We're like, okay, give him more. I mean, is there a reason you can't? They have misunderstood it, but think if you get beyond 350, they've done their job and that's all they need to do. So I really encourage you, especially if you are the prescriber or you're an educator, you're working with clinicians to get them to not say the level is therapeutic, that's what burns me on.
(00:12:54) Oh, the level's therapeutic. Well, this guy just hit the staff. I don't think it's therapeutic for him anyhow. Get them to say the level is X and whether the person is a responder or not. Once they learn to think that way, then the light bulb goes on. It's like, oh, if the level is 450 is no guarantee response. No, it's not. Guess what? Half the people on clozapine may not respond. You have to accept it. For [inaudible 00:13:18] antipsychotics maybe only two thirds will respond. Don't think, well, it's above the [inaudible 00:13:22], it's going to work out. You have no idea where it's going to work out. Here's an example I give right here again, the levels below 350 and we would tell these people kind of get into the ball game. For clozapine when you order it from certain labs, you'll get the norclozapine as well.
(00:13:38) All the efficacy decisions though, are based on clozapine, not the ratio, not the sum, but just clozapine itself. That's very important. The norclozapine can give you helpful information about how people metabolize the drug. And there are certain contexts in which that can be very helpful. But really, all of our efficacy decisions are based on clozapine levels itself. And I gave you the population ratio there. What we call the metabolic ratio, the means, where some people who are poor metabolizer, some who are ultra rapid, you all know this stuff. But a lot of this has nothing to do with your genotype. If I'm smoking cigarettes, I'm feeding it typically a rapid metabolizer, right? I've induced my 1A2. There can be a lot of environmental exposures which affect this, which is why I say the genotyping thing, not so helpful. There's multiple inputs into this system which can affect what your level is. None of which may have to do with your P450 genotype.
(00:14:35) When do we say this is it, this is the end of the road? And this is a concept which was not well developed in the antipsychotic literature. But I felt it was important in my work with clinicians to give them a name. As you know, a lot of it has to do with marketing. I'm pretty good, pretty good marketer. How can I put you into an essay today? Now, But it's explaining the concept in a way that people get it. And once they understand it, then everything makes sense. We've had people at the state hospital providers, they think they could cure every with haloperidol if they can give them enough, right? They could give them enough. And clearly they didn't know the literature. It's pretty obvious that there's a certain point, a haloperidol level of 18 nanogram per ml, beyond which your chances of response are less than 5%.
(00:15:22) And we wanted to get people out of that mindset that more is better with whatever antipsychotic. And this is the contact we came up to help them use and define this upper limit of their titration. Number one is, there's some people who you can give them all the drug in the world and they'll never have tolerability issues. As I mentioned, the upper limit for haloperidol we think in terms of the point of futility is 18 nanograms. I would get consults all the time if people with levels in the 50s sometimes, 70, I'm like, what the hell? But they didn't really understand what to do. Some of these people really are impervious to education also. And they feel like they can cure everybody if they just give them enough haloperidol. And it's usually the next person who took over the patient say, well, what am I going to do with this? And then there's a whole different discussion.
(00:16:11) The idea is had they been checking levels at some point, their level would've been 20 or whatever. And if they consulted us, we'd say, look, you can give them more but your chances of response are less than 5%. It's an active futility. Don't waste a patient's time. This person is probably going to need clozapine. If you want to maybe try high dose of olanzapine, we can have a discussion about that. But even then the response rates are very small, maybe seven or 9%. But once people understood this concept, then they got it. We get this consult. Well, I have this guy started him on haloperidol, he's almost 22 to me. He's beyond the point of utility. Should I go right to clozapine? And I sit there and I weep at my computer, not literally, but I'm like, God, somebody got it. They got it.
(00:16:51) And once they get it, then they can use that for all the other levels as well. And this is how you put it into play. So here's somebody on clozapine 700, the level comes back a little over a thousand. It's fine. The patient's fine. They've been on that level now for two months, they're tolerate again, everything is managed. All the side effects are managed, just magically. And the question we got on this patient is, should we give them more time? Should we give them more drug? I mean, really these are the two questions. If you've been in this field for a while. And I think I'm older than almost everybody, except for a couple, there was this old literature about late response for clozapine. Let me tell you, it ain't true. And it ain't true for any antipsychotic. If we were talking about an oral meaning, if somebody does not have at least minimal response, I don't need them to be cured, but minimal response after two weeks on a given dose, they are unlikely to be a responder at week six. Okay?
(00:17:52) Once you give them more time, it's not going to solve the problem necessarily. And you can make that decision much earlier. And guess what? The same is true for clozapine. So at University of Maryland, Rob Cortez was there. He did this dose titrations [inaudible 00:18:04] clozapine. Everyone who responded, among those who did respond, responded on average. It says right there within two and a half weeks, when they got to the point where they responded with a standard deviation of 14 days.
(00:18:17) The one outlier took 56 days after his dose escalation to responded. But everyone on average, two and a half weeks. The point being, don't want people languish for months on end with a level which is not therapeutic for them without doing something about it, which usually is going to be further titration. So at this point, I'm going to pass the Baton to Dr. Kelly and she's going to talk to you about some of the issues with clozapine assays and some new technology which we think will really change how we view, how to use flow zine levels in real time to make our patients lives better. Dr. Kelly, take it away.
Speaker 3 (00:19:02) Thanks so much.
Speaker 1 (00:19:04) [inaudible 00:19:04] and questions.
Speaker 3 (00:19:05) Oh, reminder from Megan, the boss here. If you have questions, write them down on your paper, turn them in and we'll be able to answer at the end. So, thanks for the reminder, Megan. So, do you want to hear about the new technology? So, I'm going to tell you about the immune immunoassay technology. And then I'm going to show you some of the data that we've had working with this immunoassay. To set everybody on the same page, how do we currently analyze plasma clozapine levels? Most of you know that currently we use majority of places around the country use liquid chromatography tandem mass specs. So LC/MS MS. So back to the days of chem lab, the horror stories that we have with HPLC if you can remember the columns and the flowing and all that, but high-performance liquid chromatography, it actually separates the components in the plasma.
(00:19:54) And then the mass spec component provides the spectral information to identify what's in there. So, the tandem means there's generally two mass spectrometers and that's what we typically use when we send out blood levels to a national reference laboratory. The system synergistically allows us to separate and then to identify. The new technology that everyone's been talking about is a clozapine immunoassay. And down in the corner here is the My Care kit. And these kits can work on general chemistry analyzers found in most hospital laboratories. But what is the immunoassay? So, most of you are aware that immuno is antibodies binding to the drug or the drug and the conjugate they're covalently bonded to nanoparticles. And then we can look at the aggregate and we can identify using an antibody technology. This is different than the LC or the mass spec.
(00:20:52) So, the new technology has been cleared by the FDA and the information's here. What I want to point to really overall, as you can see here on this graphic, that this graphic shows you on the X-axis that LC MS values. So, your clozapine blood levels in your nanogram perineal range and then you have the MyCare psychiatry kit on the Y-axis. And this is looking at the correlation between these two values that we get back. And you can see it's highly correlated, very tight line here. So, the statistics are all listed here of the slope is near one. So, it's just off of one. The intercept is just below zero at negative 27. The R value, the tight correlation coefficient is 0.93. And the average bias is only about 2%. So, you can see that this is a very tight correlation, which is why obviously the FDA has cleared this device.
(00:21:47) The lower limit of detection for reliable detection is listed at 39 nanogram per ml. So, it can go down very low in our therapeutic range that Dr. Meyer was talking about. The typical range of detection that they do see is 68 up to 1500. We typically don't see many patients beyond 1500. I have using mass spec, I have seen a few, but not many over time so most people fall in that range. On the literary aggression models also, with the modeling we see very tight correlations and the deviation from linearity is less than 10%. So, what been the differences between these two? We have the current method and then we have the new technology using the immunoassay. So, on the left-hand side here, Dr. Meyer alluded to this, the turnaround time that we're all dealing with our clozapine plasma levels is generally two to five days or longer.
(00:22:42) So as he had everybody raise their hand to this many times, we have people that go longer than a week. Sometimes I have levels that don't come back, but I'm generally not going to get a level back for at least four or five days. Everybody sort of many days on the level, of course. With the immunoassay, you can get the results back very quickly. So, if it's running in your hospital, you can, less than 30 minutes, you can generate a result, you can order stat labs. You have this technology on the chemical analyzers, you can get that back in the same day. And Dr. Cortez will talk a little bit more about that. But if you have many levels in many days, you can look at this, you can do many things with the blood level data now that we couldn't do before.
(00:23:20) So the time, I think, is one of the biggest benefits of using immunoassay technology using this technology. We mentioned you have to generally send it off site, you can keep it in house. That's another benefit for the immunoassay. And Dr. Meyer alluded to this too. The reference ranges for the labs that we send these to are all over the place. You'll have ranges that are sometimes tenfold different. And so, with the immunoassay, we have the ability to look at one range and the ranges we don't have to get back from different laboratories that are different. The actual LC MS has no actual FDA clearance or approval or guidance around this while the new immunoassay is formally FDA cleared, which devices don't go through approval process. Here they go through clearances. So, this is a clear device.
(00:24:13) Cost also plays a factor in here too. So when you send your clozapine assays out, what do you pay? 50, 75, $100 for your clozapine assay. I see heads nodding around. Some people have better deals than others with laboratories. But on the immuno assay, the average cost varying by site might be around $20. So the kits that come, you can do many tests. I think it's around a hundred tests that you can run with the kit. The reagents come, you mix them in, there's two different reagents and that's how it's used. So the reason why I'm very interested in this is that about 10 years ago, I'm working with the University of Maryland College Park. And we are interested in developing something different than LC MS. We're looking at the electric chemical detection of clozapine in the blood. Clozapine can give and take electrons. We can pick up an electrical signal. We can look on the voltage of clozapine and find out what the voltage is.
(00:25:15) And we can see the plasma levels in the blood. We can estimate that by the voltage. So we get a patent, we work on this, we have high sensitivity. We can decide what the levels are, but it's not as not specific. It picks up uric acid at that same voltage level. So we're working on the specificity. We're working. We have an NIH grant where we're having a lot of fun with this. We're hoping this can move forward, but we're struggling because electrochemically, we're having difficulties. And along comes the immunoassay. And so we had collected several samples and I'll show you this data. We collected lots of blood samples and froze them for our development. But when this came along, I thought let's take a look at the bloods that we have saved. And I'll show you this data for this new technology, because in essence, in science, we want to move this technology forward. And for clinical care, we want to move this technology forward.
(00:26:08) So our project purpose was then and I'll show you. We were able to look at testing the evaluation between the standard reference laboratory mass spec, LC MS spec and the new immunoassay. And then we were looking at different variables in the blood demographic and clinical information. And then we examined the concordance. So we're looking at how well we can replicate those same results over time. So we were interested, Saladax was interested in us doing that. And so they allowed us to go ahead and move forward and helped us fund, donated the immunoassay results for us.
(00:26:43) So our study design was single site, cross sectional. This was a one-time blood draw. We had 117 participants. So it had 48 people come in with schizophrenia who were on clozapine. We had 24 with schizophrenia who were not on clozapine. We had 45 healthy controls come in that had no history of mental illness, no diagnosis and were not taking psychotropic medications. So, we sent an aliquot of serum to the national reference laboratory, for some reason, just not naming today. But a national reference laboratory, if you can imagine. And then also the aliquot to Saladax to use the MyCare psychiatry clozapine assay kit.
(00:27:24) The immunoassay at Saladax was then tested on three occasions across two other instruments. And then we also sent a separate aliquot back to LabCorp so that we would have two aliquot sent a LabCorp. So now if you can imagine, we're trying to determine if they can replicate the findings at LabCorp, if they can replicate these findings at the same blood level over and over again at Saladax. And that would be very informative. So we looked at demographic, we looked at clinical information, including smoking status, communications, their CBC withdrawn, et cetera.
(00:27:58) We used a linear regression and mixed effects model to look at the correlation between these. And then we did what's called a concordance correlation coefficient to look at this over time, to see how well, if you want run it one time and run it a second and a third time, if you get the same results or is that all over the place. And that's important as we try to look at what is a valid and accurate result. So, take a look at this data. We published this, Tiffany Buckley was one of our residents a few years ago and she's the first author on this. And we published it in therapeutic drug monitoring. The red lines are the national reference laboratory results. The yellow lines are the immuno assay results. So, the clozapine group is on the left. The not on clozapine group is in the middle.
(00:28:43) It's not on clozapine. It's not on clozapine, non-clozapine and then the healthy controls. And what we do, we find very similar results between the national reference laboratory and the immuno assay. We know that this is about a 10% difference. We found here just a little under 15%. So very similar overall on those taking clozapine for the mean blood level at the time of the draw. And I'll show you the correlation there in a minute. We'll come back to this but take a look. We have a red line showing up for the non-clozapine and the healthy controls when I sent this out for national reference laboratory. We did not see any values come back as false positives with the immuno assay. On average in the non-clozapine group, our average was a blood level of about 30 and in the healthy controls, we had an average blood level of 44.8 with a standard deviation of 332.
(00:29:36) I had a few people come back on the national reference laboratories with clozapine levels, not on clozapine, never have taken clozapine in their life, not on the other psychotropic medications with values of about 300. So, we have many false positives. In fact, 15 false positives using the national reference laboratory. Interesting sending this data out didn't experiment like this previously. So, this is our correlation between the immunoassay and the national reference laboratory. We do see a high and significant correlation between the two. This is including the national reference laboratory and then the three immunoassay samples all listed. So our correlation is very high as was the initial testing for Saladax. Oops. As I mentioned, I said 15, there's 18 false positives that we had for clozapine from the national reference laboratory. None from the immunoassay.
(00:30:31) What we do find is if we look at the concordance correlation coefficient between the two on many runs, we see again, as you continue to run it over time, you still see a high correlation. So you can replicate this over time. We didn't see any association between any clinical variables on clozapine levels except for a couple of things. And I'll point that out in a minute. The Concor correlation coefficient, this is looking at it over time. If you look at the national reference laboratory on the two samples we sent, it's a 0.86 correlation. So it's a high correlation. We see it, we send the blood level, we send it again. They can correlate that and have a very similar result. But if you look at the immunoassay results, it's a correlation CCC of 0.99. So they run it, they run it again, they run it a third time and they get the same results.
(00:31:21) So it's very highly reliable on the amounts that you get. Other variables that we did see. And this is interesting. We do publish this in the paper and this is probably for discussion and follow up. Somebody needs to do this, somebody's residency project next year. But for each unit increase in total protein in the blood that we saw, there was an increase of 145.6 units. If you did the mass spec and 225 units, if you did the immunoassay. So interestingly enough, you expect a higher plasma level for higher total protein in the blood. There's not a lot published on this. I'm not going to go into anymore detail on it. But I wanted to point that out cause we do point that in the paper, but no other variable was associated with the clozapine levels.
(00:32:07) So the summary of our results. So the immunoassay results were in good agreement, just like they had seen with the FDA approved and good agreement with the gold standard. We did have false positives with the gold standard. We did not see it with the immunoassay. I don't know if you'd send your blood bill if you've had that happen or not. It likely is cross-contamination at the lab. There could be other things picking up. Although these technologies, the mass spec with the tandem should be highly specific and highly sensitive. So I'm guessing there is some cross-contamination that happens at the lab level and we do get some differences. But also remember that the total protein, that's just an important side here, an extra thing to think about. But on repeat analysis that correlation that CCC is very high with immunoassay. You're going to get that same value again, it's a highly reliable result.
(00:33:00) So overall, just to kind of summarize, as we move into Rob's talk, the benefits to help transform care. I mentioned this higher CCC, that's important to us. The no false positive, that's going to be important to us. The results quickly, that's going to be really important to us, right? We're going to get these results back. If you have this technology, you're going to get these results back within a day, if not that same day from your laboratory. So this is really a great technological advance for us.
(00:33:28) And you can manage people's care in real time and we're going to give... This is what's going to be really exciting to see some examples that Rob has done. But the cost effectiveness is going to be important too. So these are sort of win, win, wins. But I also wanted to mention that in the future, this could be available potentially if it'll worked out as a point of care. So we could be thinking about something like this, like the atheist device, where we can do something in real time at the point of care. So this technology lends itself to that. So hopefully this is a transformative points to think about. I'm going to turn it over to Dr. Cortez so you can tell us about how this can be used in patient care.
Speaker 4 (00:34:12) Thank you so much, Dr. Kelly. Thanks Dr. Meyer. Hard to follow them after their great presentations, but I'm going to try. I'm going to talk a little bit about before the practical clinical scenarios. We're going to talk just a little bit about how you get to having the assay. That includes building a clozapine program. It includes the assay, it includes interdisciplinary collaboration, education, education, education. So I'm going to tell you a little bit about our practice at Grady Hospital in Atlanta, about how we set our clozapine clinic up. And then a little bit about how we introduced the immunoassay and then how I think that there were key things that helped us to set up the clinic that then helped us to then get the immunoassay out there, but it took a big, big effort.
(00:35:07) So when I was walking around the inpatient unit in 2011 at Grady, I worked as an inpatient psychiatrist for eight years. I was struck that no one was on clozapine. And I thought to myself why? Because the majority of people had schizophrenia and I wondered why wasn't 28% of these people on clozapine. And the response that I got was, well, we don't really do clozapine around here. It requires a lot and that just was unacceptable for many reasons. So eventually we got it together and we developed a clozapine program called PSTAR, which stands for persistent symptoms, treatment assessment and recovery.
(00:35:47) And I just have to say how important it is to frame everything within a recovery oriented framework, everything talking about clozapine, talking about the levels, thinking about person-centered care. So since PSTAR's inception in 2012, we've had 500 referrals and the vast majority are either for a second opinion, which basically is someone who has treatment resistance, schizophrenia or treatment resistance schizophrenia identified by the person who makes the referral. Interestingly, we get a lot less people for psychosis and suicidal ideation, one of clozapine's other FDA indications. So we're working on some educational efforts around that. And now we have about 110 or 120 people on clozapine.
(00:36:35) One of the first things we do when we get a referral is we differentiate TRS from pseudo resistance. And after the trip guidelines came out, it's very clear after two antipsychotic failures, we need to be thinking about clozapine. So even for people who are deemed as having TRS, there's a significant number of them who have subtherapeutic antipsychotic levels. And in one study that was done, it was 44% of people had subtherapeutic antipsychotic levels. We don't know if it was a lack of adherence issue. We don't know if there were kinetic issues, but that's pretty significant. So we learned a bit from this study and thought when people come to see us in our outpatient clinic, one of the first things we do is we get antipsychotic levels on what they're taking. Sometimes they come to us on clozapine, most of the time they don't. So we have had a lot of experience getting other antipsychotic levels.
(00:37:26) The other part of my job has been over time, we've realized that individuals that come into our service have an incredibly long duration of untreated psychosis. This is a problem in the Southeast, this is a problem everywhere for many reasons. And we've developed a first episode clinic called Project Arrow, where we try and work on reducing the duration of untreated psychosis for people. But really what this has turned into is more of an exercise in reducing the duration to treatment resistance. Because a lot of times people have had psychosis, untreated, no experience with mental health services for in the triple digit of weeks. So we really had to do something to try and get a first episode program going. So we have a coordinated specialty care program that serves people it's a SAMSA funded program. But I think what's interesting is that when you have this sort of continuum of care from the first episode to people who are on clozapine, you have the opportunity to make sure that people just get two antipsychotic trials, not eight.
(00:38:36) You really can start to reduce that number of antipsychotic trials. It's much easier said than done, because for those of you who work with early psychosis folks, there's a lot of buy-in that people need. You have to really do a lot of work with engagement. It's not that easy to just say that this person has had two antipsychotic failures, but in our first episode program, we have a large number of people who actually are on clozapine. But I think that the big thing that has been helpful for us is with clozapine levels, it helps us to determine, I guess, with other antipsychotic levels helps us determine, first of all, that someone might be sub therapeutic. We might need to do something, but we have a lot more confidence, I think in starting clozapine for some of these folks.
(00:39:24) Okay. Just to give you a little bit more about the evolution. So PSTAR began in 2012, it didn't take us very long to figure out that clozapine levels were incredibly important. It took us about a year. We just didn't really have it. But after a year we realized that this was really, really important. And we'd been systematically obtaining clozapine levels since. In 2020, we sent out 1047 clozapine levels. And we began to talk with the Saladax team about purchasing the immunoassay in January 2020. It was approved in April 2020. And then our assay went live in May 2021. And it took a long time, but part of that was due to the pandemic and there was a lot going on at the time. But our clinical chemistry team was really, really great in helping us to get the immunoassay going. So a big shot out to them.
(00:40:21) And it was a great collaboration that really hadn't happened before. I didn't know a lot of the people in the clinical chemistry team, but I found them really, really smart, helpful and we got this thing off the ground. So right now in our system, we are running the immunoassay for some people and then other people are still going to an external lab. So it's pretty easy to kind of see the difference because in the Southeast, as I was mentioning, it's kind of a clozapine desert. Clozapine is not widely prescribed. People are coming from all over the state of Georgia. We see a lot of people, virtually. Some people are getting labs at their lab, so they don't have to come in. But I'd say that maybe about 70% of people come in and get both their labs and their prescription in our clinic.
(00:41:13) So just a couple of notes about the advantages that we've seen with the Saladax assay. The core advantage is rather than having to wait five to seven days, you get it on the same day. And this has led to an improved workflow. And I'm going to talk a little bit more about that. So this is the scenario that we see for people who get their labs elsewhere. On day zero, we'll meet with the patient, we'll do a blood draw, we'll get a clozapine level on day seven or day 14 or never. The level will return. We call the patient on day seven, if something is unexpected. And there's some advantages and disadvantage here, at least the advantage is people are checking levels, but some disadvantages you have to respond quite a while after the blood drawn, this always happens on Friday night, every time. They have a system where it comes back Friday night, 7:00 PM, you get the lab, which is great. We've come to understand that and it works.
(00:42:16) But you can't respond in real time to what's going on with the level. Seven days is a long time. So then we kind of send up a little bit and said all well, we're going to get the blood draw seven days before the patient comes in, we're going to get the level. Then we meet with the patient on day zero. The level comes back and at least you can talk about a relatively recent level with the patient in front of you. But a lot can happen over seven days or 14 days or whatever it is. And it's impossible to respond real time to issues.
(00:42:50) So let me tell you a little bit about the current workflow that we have. So I run a clozapine clinic half day in the afternoon. So what we do is we have people come in if they can do it in the morning, we have them get their blood. They then go to a group. Then they come and see us in the afternoon. And the big advantage here is we can actually discuss that level that was obtained earlier in the morning. And it gives us much more confidence, I think in making a dose adjustment. And I think it comes up a lot, especially if somebody has a low level providing people instructions about how to reiterate the clozapine, which is a tricky issue.
(00:43:36) Now this isn't the case for everybody. Many people are working and people are just trying to fit it in however they can. But this is ideally how we like to structure it.
(00:43:49) The other thing, just to talk a little bit more about interdisciplinary teams, in our setting we don't use the point of care A and C device. We actually do phlebotomy for everybody. And we have a group of nurse practitioners who go out into the community and they can draw blood and they can get people their medication. So this has been incredibly helpful. And I think during the pandemic, it allowed people to continue clozapine that never would've been able to do it otherwise. So what we do with them is we have them go out to see the person, maybe the day before the visit with the prescriber or maybe the day of and this gives us a much better chance to get real time decision making on the levels.
(00:44:33) So might a better workflow equal a better work life balance? This is a bit of an aspirational kind of thing, but maybe. I know that being a clozapine prescriber over the last decade, it doesn't really work in nine to five hours. So in a survey that we did, which Dr. Meyer was co-author on actually, we surveyed 32 clozapine clinics and we looked at what kind of services they had, about half of them actually have on call services. And sometimes that's just people taking a pager, sort of all the time. From some of Jessica go's work at the VA, she found that many clozapine clinics were quite under-resourced with over alliance on too few staff members. I thought to myself, that's me. I'm one of those people.
(00:45:21) So I think that having the app and the level come back on the same day is just sort of one less thing you have to deal with. You have patients who are in crisis, all sort of things going on. The last thing you want to do is sort of deal with a clozapine level that comes back on 6:00 PM on a Friday night. And I think that families, at least my family... Somebody, good work, good work out there. So this happens all the time. My family's used to it. They say, well, instead of dad, it's Rob, Rob, you're always on the clozapine REMS website. Just put it down for a second. I'm like, this is the stuff that we have to do. But this has made it easier and it's obviated the need for the fax machines, which is I think a big breakthrough.
(00:46:12) All right. So some additional advantages of same day results. I think this quarter gets back to recovery oriented philosophy. So you have improved collaboration with prescribers and patients, give you a couple of examples of this. I mean, people really like hearing what their clozapine level is. I mean, it's kind of one thing that people have more control over. People don't have control over their absolute neutrophil count. Nobody cares if their ANCs a 2.6 or 2.4, maybe they try and exercise and they try to manipulate, but it doesn't really matter to them as much. Clozapine level is something that they did the last five days had something to do with what their clozapine level is. And it resonates with them. And with the right kind of education, people can really understand what this clozapine level means for them. And people take pride in it. And we say things like your clozapine level is perfect right in the middle, not too high, not too low. Great work.
(00:47:07) And also I think as Dr. Kelly was mentioning, this is helpful in the titration process. And Dr. Meyer was mentioning this too. So we have a sense, especially early on, we're shooting for a goal level of 350 nanograms per ml. And then with these early levels, it helps people to say, well, we can probably accomplish that in the next three or four weeks. So it's not this indefinite titration that sometimes can wear people down. And also I think that it gives our prescribers a greater confidence to increase or decrease the dose.
(00:47:42) So here's an example of someone, this is all de-identified with some of the details slightly changed, but this is a 44 year old person who's on 300 milligrams of clozapine. They come into clinic, they tell us they're still having symptoms. The clozapine level result comes back less than 68 on the immunoassay. So with this information, we ask them, which we would ask them all the time, have you been taking your medication consistently? And the person says they missed their last four days. Maybe they would've told us that that they missed their last four days. But when we presented with them, the less than 68 value, that's one thing that helped the last four days history come out. So rather than reflexively increasing the dose to 350, we decided to reiterate clozapine over the next three days. It just seems like it would've been pretty dangerous to go ahead and give that person 350 after they've missed four days of clozapine in a row. I don't know how else we would've figured that out without having the immunoassay.
(00:48:44) The other really big place that this can make a difference is in emergency room settings. So people who are coming in urgently and I think we do a great job in our clinic, but people have multiple challenges and multiple medical issues and people go into the emergency room for all sorts of reasons. And in the emergency room, the immunoassay can be very, very helpful. Let me give you this example here. So this is a 55 year old man with schizoaffective disorder. He's clozapine 450, lithium 900, Depakote 1500, his levels historically been running in the 500s. Seven days ago he went to the PSTAR clinic and he was doing well.
(00:49:31) So when he came into the emergency room, he had a fall and he had altered mental status. So when they got the clozapine level, it was about 1100 and everybody said, what gives? So this is another important point here, is the education piece, the collaboration piece, because I don't work in the psychiatric emergency room. It's really important to be able to sort of disseminate this information if you work in a healthcare setting across your healthcare setting. So there's been some educational outreach to the psychiatric emergency room team, to the emergency room team about ordering clozapine levels if somebody comes in on clozapine, do this for everybody. And now it makes a lot more sense than waiting seven to 14 days. That was less of a great argument that we had. We used to tell them before, just get it and we'll figure it out later.
(00:50:20) Nobody liked that. But now we get the level. Everybody knows there's been sort of an educational effort about what to do with that and it just makes more sense. So back to the case, this was a patient that was found to have COVID we suspect they had a fairly symptomatic infection. There's going to be more about clozapine and inflammation, I think at this very conference. So just to jump into it, basically the infection probably leads to increased. IL-6 TNF alpha interfering gamma and that inhibited CYP1A2, then we have hot clozapine levels. So with this knowledge and then also based on the consensus guidelines about what to do for patients on clozapine, who are in the midst of COVID infection, we reduced the dose to one third of its original dose. And then we had levels help us to guide the registration.
(00:51:23) Okay. I think another big benefit has been greater multidisciplinary awareness of pharmacokinetics and safety. So I knew that what we had done had made an impact when we had this patient who's 24 years old, had a diagnosis of schizophrenia. He developed right ankle osteomyelitis and he wasn't taking the Seltorexant through the PICC line. So I get a call on Friday afternoon at 5:00 PM, which is perfect. And the ID doc and the clinical pharmacist says, look, I know that you all are really interested in clozapine levels, what do you think about oral levo versus oral ciprofloxacin? And you know what I said, of course, no way do we get ciprofloxacin. And we got the patient on levo and ended up doing quite well. But I think that this entire outreach really has had just sort of overall increased the awareness, increased the collaboration and it's overall, I think really impacted the patient outcomes.
(00:52:27) Now the limitations, so with the immunoassay, you don't get a norclozapine level. But what we found is the benefit of having a rapidly available clozapine level has outweighed considerably the need to have a norclozapine level and to check the MR ratio. So what we tell our residents is, for everybody you do concentration dose forever and if you can, you get the metabolic ratio too for everybody. So we think about it in both ways, the CD and the MR for everybody. But in rare situations, we still might get the send out test, particularly if there's perhaps an absorption issue.
(00:53:06) So when do we get clozapine levels? And I'm coming down on a side where it's, when would we not want to get a clozapine level? And I have to be pretty much talked into not wanting to get a clozapine level because I find them so valuable. And one of the examples for not getting a clozapine level has been this patient, for example, they're taper and clozapine because they're doing well and maybe they don't need to be on an antipsychotic anymore and we're reducing the dose. And I'm trying to tell the residents, look, we don't need to get a clozapine level now this patient's on 25 milligrams of clozapine, they're going to be off of it. What's the point here? But other than that, I mean, I think there's a lot of really justifiable reasons for wanting to get a clozapine level. I know Dr. Meyer mentioned 200 milligrams during a titration, but we often get it at about 100 milligrams just to get a sense of kinetic what's going on. Any dose increase suspicion for medication inconsistency, any new adverse effects and then lack of response too. All right.
(00:54:12) So in summary, plasma antipsychotic levels are the best proxy we have for CNS action of clozapine. And I know that everybody is setting up measurement based care here. And it's so so important that not only do we think about obtaining levels, but we think about really how to get the best information about medication adherence. There's a big educational effort that's required about how to use a tolerability threshold and point of futility to make efficacy decisions. So new immunoassay methods for clozapine levels, often more rapid turnaround time, improved accuracy compared to the existing methods, which was kind of something that we had to struggle with a little bit. We kind of got some results that were different than we expected from the send out tests. And people thought it was the immunoassay, but when we showed them the data, they were like, oh actually, maybe it's the send out test. That's been the problem all along. So it took a little bit of frame shifting.
(00:55:13) I think that there's a lot of positive impacts from getting clozapine levels and that can include improved workflow, improve improved team collaboration, improved patient outcomes and I think that patients and their family members seem more invested and it's sort of easy to understand what's going on with the levels, from the patient perspective. All right. So with that, we are at the end of our prepare remarks and we would love for you all to put any questions you have down on the card, we'll come by and we'll pick them up.
Speaker 1 (00:55:50) Yep. We're collecting [inaudible 00:55:51].
Speaker 4 (00:55:51) Just hold up your hand and we have people...
Speaker 1 (00:55:54) Hold them up. The runners are coming.
Speaker 4 (00:55:55) People circulating, who will answer your questions. [inaudible 00:56:02].
Speaker 1 (00:56:07) Okay. This is excellent for you.
Speaker 4 (00:56:10) If I can read it.
Speaker 1 (00:56:11) We got more.
Speaker 4 (00:56:13) How does one determine minimal response on clozapine? So this is actually an important concept across all antipsychotics. What is minimal response? When we see people on antipsychotics, all we're looking for [inaudible 00:56:33] within that first couple of weeks after a dose increase is clinically perceptible change. If you notice it, if you can say in your mind, and there's a way you could quantify this, which is using the seven point clinical global impression of severity scale, CGI severity scale. You do this every time you see people, anyhow, you say, man, this lady's really sick. This is the sickest person I've seen, which is like a seven. That is people start to get better they drop down the CGI scale.
(00:57:02) But if you see this person, you say they're a bit better. That's minimal response. So if you can detect it in terms of an improvement of level of severity, that is sufficient. So the idea is after let's say a dose increase, so maybe their level was subtherapeutic, or maybe it was above the response threshold and they didn't respond at a level of 450. Now you get it up to 550 with a dose increase. After two weeks, the patient is notably a bit better. That's all you need.
(00:57:32) I'm going to let them roll with that. And we'll see what we get. But if after a couple of weeks you say, this person looks exactly the same. That is less than minimal response. And it's time to move along. Typically with the titration. Again, the titration guided by levels because occasionally once in a while, like Rob's patients don't take the, not my patients, of course, but other people's patients don't take the meds. You just got to figure it out. And the one statement I'd like to make more than anything about adherence is, adherence is dynamic. It is not static. You can have people take their meds every day for six years, who decide for a variety of reasons. They're not going to do it tomorrow. And that's why periodic levels over time can really help with that.
Speaker 3 (00:58:15) I have an excellent question that had asked, besides Dr. Cortez are any of the rest of us here on stage using the immuno assay at our clinical sites and have your sites implemented this. We are hopefully fingers crossed. Saladax maybe plug your ears a few weeks away at the University of Maryland Medical Center from using this. And so we are in the process of working to get this up and running. And I know there has been other interests throughout the room as well. And Rob, if you had any other sites that you know are up and running.
Speaker 2 (00:58:45) Hopkins?
Speaker 3 (00:58:46) Hopkins. Yeah.
Speaker 2 (00:58:47) Hopkins. Definitely. Yeah. It does take a bit of effort. I think maybe you want to speak to who you had to talk to in the lab to get them to buy to that? Well, this is valuable. Not everybody, but...
Speaker 4 (00:59:03) I think it depends on your health system, but the way that ours is set up, I mean the clinical chemistry team was the first group to get the buy in. And it really was pretty easy to get the buy in from clinical chemistry then that we have a whole another organization that deals with purchasing and trying to figure out if the cost can justify, can you bill for it. The purchasing thing was a major labyrinth. There was 18 different steps to go through. A lot of it I didn't have to do actually, but it just took a lot of time. And the system was in the midst of COVID and they had their hands full with a lot of stuff.
(00:59:43) So I think it took a little while, but basically it was the purchasing and it was the clinical chemistry that helped to get it over the finish line. And then also IT at the end. IT, There was a lot of issues, but ultimately when people ordered a stat clozapine level that resulted in the immunoassay, if they didn't choose stat, that resulted in the send out test. So there was a lot of education about that that had to happen. But now actually, if somebody in our system orders a clozapine level, whether it's stat or not, they just get the immuno assay and they now have to specify whether they need, they want the send out or not.
Speaker 3 (01:00:20) Yeah, I've been working to at Maryland. I think we had to get the buy-in from the lab that they would work with us.
Speaker 5 (01:00:26) Exactly. What I wanted to say is that you might not have a chemical analyzer at your place. And so where I work specifically, we don't have a general chemical analyzer, so you can be creative here, we're working with the university. And so the University of Maryland medical system does. And so the lab there is willing to work with us with a courier service that we could have pick up on a daily basis to take it down and back. Now that adds something to us. We don't have it done in our hospital, but we're still going to get it back in a much, much quicker time than we would if we would send it out. So we have things to work out. There's ways you can be creative in working with hospitals that might have a chemical analyzer if your site does not. But think about it.
Speaker 1 (01:01:03) Do you want to go ahead and take that question over here.
Speaker 5 (01:01:05) Sure.
Speaker 1 (01:01:07) So this question says since immunoassay has to 10 to 15% higher concentrations in the reference level for efficacy and the efficacy is based on concentrations from LC MS, do you shoot for greater than 385 nanograms per ml to be equivalent? And so what I would say here is that, you send it out for LC MS. You're going to get different results each time too. You're going to get this variation that occurs. So even if one time you get 350, the next time you might get 275, the next time you might get 385. So you're still dealing with this variation that you're going to get. Even if it's 10, 15%, sometimes you see a little bit of variation in those results.
(01:01:46) So I would base it still on your clinical care, on seeing what your patient's doing. I would still be aiming sort of in the range of 350, knowing that you're always going to be a little bit above that or below that. And that could be perfectly fine. It's when they're not responding that you would think about, but I wouldn't spend your time trying to nail it down to look at the difference because that coefficient of variance happens even with the results that you're getting. Normally, I don't know if you want to add to that.
Speaker 2 (01:02:12) I think the point that I made is that if you look at different studies, they show different response thresholds. Eventually you just pick a number as a target. It's just a guidepost. If you're well below that you're not in the ballgame, then once you get above it, then you're like, okay, we're in the ballgame. Now the clock is ticking every couple of weeks, three weeks. If they're not responding, we'll move up. So don't get fixated too much on the number because I can find papers, which show different numbers in 350. We've just kind of said, we're going to pick 350. Just remember a number wherever you are. You want to say 400, I don't care. Really, I wouldn't argue with you. Just pick something where if it's below that, you'll say in your mind, this is likely to be subtherapeutic for the majority of patients and let's kind of move on from there.
Speaker 1 (01:02:53) Dr. Cortez.
Speaker 4 (01:02:55) All right. So this is a great question. If you have an established clozapine patient, for example, is on 450 milligrams that stopped it for greater than a week. Can you titrate faster?
(01:03:07) Great question. And first of all, the package insert says that if somebody has missed two days of clozapine, you need to restart at 12 and a half to 25 milligrams. And I think that two days or more and I think if you've practiced clinically you know that if you do that, you might be messing around with a recipe to have somebody go into the hospital. I just don't know if that's really realistic for a lot of people. So we have developed some formulas to help guide clinicians in days 1, 2, 3, 4, 5, 6 and seven. And then if it's greater than seven days, we usually will kind of go back to 25 milligrams. I think a key thing here is people as Dr. Meyer was saying, adherence is dynamic and you may very well be working with somebody who has been off clozapine for a couple days, and then decided to restart clozapine at the full dose.
(01:04:04) You want to ask that person, if they've ever had gaps in their clozapine before and how they did or what they did to try to get back on there. Some people are very, very sensitive to a titration that's too fast. And if you have to figure that out, if your patient is one of those individuals. But typically what we do is we kind of do a pretty detailed history about other periods where patients have missed longer doses and then have an individualized plan. Also, we talk about what the package insert says and say, we're a little bit worried about this. We may need to go a little bit faster.
(01:04:40) So typically if somebody has been off of clozapine for greater than a week, I'll probably reiterate it back over a number of days. And there's a documentation issue that goes into that. And also you have to think about how likely is the patient going to execute the titration that you want them to happen. You can be very, I want them to be on 117.5 and then as you go up to 175, sometimes that's just not how it really works. So you have to figure out what is going to be doable. What can you agree on? You have people give back what you've said. But it's a great question. I wish that there was more guidance about how to do that.
Speaker 1 (01:05:27): Thank you.


© 2022 Beckman Coulter, Inc. All rights reserved. Beckman Coulter, the stylized logo and the Beckman Coulter product and service marks mentioned herein are trademarks or registered trademarks of Beckman Coulter, Inc. in the United States and other countries. All trademarks are property of their respective owners