Hello and welcome. I'm Dr. Rob Cotez, Director of the Clinical and Research Program for Psychosis at Grady Health System, an Associate Professor at Emory University School of Medicine, and I also serve as Physician Expert for SMI Advisor. I'm so pleased that you're here to join us for today's webinar, Pharmacogenomics in Psychiatry, Is There a Role? SMI Advisor, also known as the Clinical Support System for Serious Mental Illness, is an APA and SAMHSA initiative devoted to helping clinicians implement evidence-based care for those living with serious mental illness. Working with experts from across the SMI clinician community, our interdisciplinary effort has been designed to help you get the answers you need to care for your patients. Today's webinar has been designated for one AMAPRA Category 1 Credit for Physicians, one Continuing Pharmacy Education Hour, and one Nursing Continuing Professional Development Psychopharmacology Contact Hour. Credit for participating in today's webinar will be available until January 2nd, 2024. Slides from the presentation today are available to download in the webinar chat. Select the link to view, and that is there now. Captioning for today's presentation is also available. Click to show captions at the bottom of your screen to enable. Click the arrow and select Full View Transcript to open captions in a side window. Please feel free to submit your questions throughout the presentation by typing them into the question area found in the lower portion of your control panel. We'll reserve 10 to 15 minutes at the end of the presentation for Q&A. All right, now I'd like to introduce you to the faculty for today's webinar, Dr. Megan Aratt. Megan Aratt, PharmD MSBCPP, is a professor at University of Maryland School of Pharmacy, and serves as SMI Advisor's Pharmacy Expert. Most recently, Dr. Aratt has practiced at Fort Belvere Community Hospital as their Behavioral Health Clinical Pharmacy Specialist. She has experience in treating the spectrum of mental illnesses and substance use disorders. Additionally, she is a past president of the College of Psychiatric and Neurologic Pharmacists, CPNP, and serves as Senior Editor on the Psychiatric Psychopharmacology Review Course. Dr. Aratt, I'll turn it over to you. Thank you so much for leading today's webinar. Thank you, Dr. Kotez. I'm very excited to be here to talk about pharmacogenomics. I did my training as a fellow many years ago in pharmacogenomics, and excited to share some of the new data and information that we have on pharmacogenomics. So I have no disclosures or conflicts related to the subject matter being presented today. Our learning objectives for today, we're going to describe pharmacogenomics in the current literature as it relates to the treatment of serious mental illness. We're going to cover a broad overview of where things lie within guidelines and package inserts. If we delved into all of the individual clinical research trials, we could be here for several weeks. So we're going to try to provide some high-level recommendations today. We're going to look and think about available pharmacogenomic testing options. We'll describe current utilization of pharmacogenomics testing in clinical practice, and then discuss implementation of practice, of this testing into your practice, and what that might look like, and some of the considerations which are very important when thinking about pharmacogenomics. I wanted to start today just to make sure we're all on level playing field with pharmacogenomics, really thinking about what this means and what we're really looking at. So we're not going to talk about predicting disease today or genetic testing in the sense of predicting maybe a cancer type. Today, we're really focusing in on medication and how a patient responds to a particular medication. So in our practice, I'm sure we all have groups of patients that all have maybe the same clinical diagnosis and are issued the same prescription. In a select group, that drug may actually be very beneficial and work for the patient and not have any side effects. In other patients, the drug's going to work, but the patient's going to have side effects. In others, it might be that they just have side effects, but it doesn't work. Or perhaps it doesn't work and doesn't have side effects. But predicting which group or subgroup our patient's going to fall in can be very challenging. And many times within psychiatry, this can even be more difficult to try to determine where the patient will fall. And so we spend a lot of time educating and changing medications and switching strategies when we get into one of these subgroups. There are three larger buckets where we'll start today. There are pharmacokinetics, then there are pharmacodynamics, and then there's the immunologic mechanisms by which we're going to discuss some of the pharmacogenomics today. So we'll briefly go through a quick review of the different types of items, and then we'll dive into some literature. For pharmacokinetics, looking at how we metabolize medications. Many of us consider CIP enzymes. So this might be some of the most common ones that we see in psychiatry of 2D6, 2C19. But it also can include UGT, the cattle O-methyltransferase, and the P-glycoprotein. And so all of these are different things that can be measured to look at how we are metabolizing or moving medications. It's important to understand the standardized terms. And so these terms that are outlined here that we're going to go through are from the Clinical Pharmacogenomics Implementation Consortium. So this is an international consortium group of volunteers. They have a very small dedicated staff who help facilitate information into a distilled information for us to consider as clinicians when we're thinking about pharmacogenomic testing. And they have a set of guidelines that we're going to review today. The allele functional status. So all of the genes have allele functional status. So some of these would include CIP enzymes. So for the example here, we see the functional status as CIP 2C19 star 17. And so there are many different stars. And these indicate the phenotype of which the person has. So when we think about phenotype of our drug metabolizing enzymes, you might be ultra-rapid, you might be rapid, normal, intermediate, or poor. When we think about ultra-rapid for a phenotype, if you are a 2C19 ultra-rapid metabolizer, you're going to break down medication very quickly. You actually might remove the medication so quickly that it doesn't have time to provide symptom relief. Rapid, normal, and extensive all seem to get bunched together and really consider the normal rate of metabolism. This is the standard or status quo, where a normal amount of medication at a standard dose would be ideal. As you get into the intermediate and poor, this is where patients have a much slower rate of metabolism, and you may have too much medication at the standard doses. And poor is when you might start to really start experiencing adverse effects. When we look at some of our transporters, we may have increased, normal, decreased, or poor function for some of our transporters. And then for our phenotype, for our high-risk genotype status, maybe for our HLAs, you would either be positive or negative for those high-risk status. So it's important to sort of understand when you're seeing these particular information, which is usually what's reported in a pharmacogenomic test, that these different alleles indicate what the phenotype is. And these are always ever-changing. There are rare ones. There are new ones. So we'll talk about the importance of understanding some of those in the presentation today. For pharmacodynamics, this is the biochemical, cellular, or physiological effects of the medications and the mechanism of action. With pharmacogenomics, we have a ton of data on pharmacokinetics. We have some data on pharmacodynamics, but it's not nearly as telling. And we'll demonstrate some of that today. But we think about this like neurotransmitter receptors, reuptake transporters, signal transduction, genes transcriptions, protein folding and trafficking. So many of the dopamine receptors or the serotonin transporters, we can look at some of the changes in those particular receptors as well to determine response in adverse effects. And then our immunologic mechanisms. These would be our drug hypersensitivity reactions. The HLA genes are the ones that are most frequently noted in psychiatry. There are two sets of available guidelines. I briefly talked about the CPIC, which is the Clinical Pharmacogenetic Implementation Consortium Guidelines. And these are both freely available online so that you could review these and see where the guidance is. And these are nice collated reviews of the literature and package insert and data of that sort pulled together. So we have the CPIC. And then there's also the Dutch Pharmacogenomics Working Group that also has annotated guidelines as well. And so we'll review both of those guidelines today. Before we delve too deep into pharmacogenomics, I wanted to highlight one case because I think this is important as we think about implementation of pharmacogenomics. So several years back in Northern Virginia, Inova Health Systems was promoting their Medimap tests and suggesting patients could come get their Medimap tests done. And it would be a pharmacogenomic test that would provide a wealth of information regarding drug metabolism, pharmacodynamic effects, and all things of the sort. Well, the FDA was concerned about the clinical validity of this Medimap test and determined that it hadn't been established for the intended uses for which it had been marketed. Specifically, they were unaware of data that Inova might have had that established some of the relationships between the genotypes that were being assessed and then the responses that were being given. And the FDA felt that these tests gave significant public health concerns such that inaccurate test results could affect decision making of healthcare providers and patients such that it might be detrimental to patient health. They were also concerned that providers could be ordering this test and the results being sent to patients and patients may inadvertently or decide to change their medication without the advice of a provider. And so at this point, Inova then decided to end their genetic testing because they felt that it was concerning based on what the FDA was providing and some of the action the FDA has written to take. Most recently, the FDA has begun discussions about implementing a rule on regulating some of these lab-developed tests that previously could be done with little supervision. So, there is a lot of discussion now about if this rule is implemented, that laboratories may need to provide data on testing accuracy. So, as we think about some of the genomics today, what we see today in our own labs, if they are homegrown tests, it will be interesting to see how the FDA acts upon some of this in the near future to think about the availability of some of our homegrown testing versus using a commercial lab or using FDA-approved pharmacogenomic testing. So, we'll switch gears for a moment and now let's dive into the classes of medications and think about guideline and product labeling and where the evidence stands currently for pharmacogenomic testing, and then we'll take that data and think about utilization of it in clinical practice. So, we'll start with antipsychotics. Within the antipsychotics, there are 10 currently that have either guideline or product labeling information available for pharmacogenomic testing. Each of these classes of medications is set up very similar. I'll make sure we give the highlights and then I'll show you some of the specifics. So, within these 10 antipsychotics, the highlight is that 2d6 poor metabolizers, if we know that about a patient, starting at a lower dose or choosing an alternate drug that's not metabolized by 2d6 could be recommended. Looking at genetic variation in our dopamine 2 receptors has really demonstrated inconsistent results either for efficacy or adverse effect, so currently not recommended to be utilizing that particular pharmacodynamic test. When we look at the specific 10 antipsychotics, we have the antipsychotic names and then whether the CPIC or the Dutch working group has evidence guidelines and then which product labeling has particular guidance on these antipsychotics. You can see that they all surround the 2d6 and the need to potentially consider a lower starting dose if you do have a poor metabolizer of this. When we look at antidepressants, this tends to be where there is a host of literature with pharmacogenomics. The bulk of the antidepressants have really been derived from studies of major depressive disorder and have focused on pharmacokinetic mechanisms. Studies such as the STAR-D, the GEN-DEP, and the MARS, as well as the international SSRI pharmacogenomics consortium GWAS analysis, have not consistently supported any single pharmacodynamic gene variant as a significant predictor of antidepressant treatment response. We do have some low to moderate levels of evidence for things like serotonin 2a receptor, the glutamate ionic receptor kinase 4, and the FK506 binding protein 5, but this is low to moderate evidence and it hasn't made it to guideline or package insert. Currently, there are 17 antidepressants which do have guideline or product labeling information surrounding pharmacogenomics. For 2C19, for poor metabolizers, there is a recommendation of a 50% decrease in starting dose, particularly for those that are metabolized through this pathway, so citalopram, s-citalopram, sertraline, and our tertiary amine tricyclics. For rapid or ultra-rapid metabolizers, you might see an inadequate response for those particular ones through 2C19. For 2D6 poor metabolizers, again, a 50% decrease in tricyclics, fluvoxamine, peroxetine, and if you're an ultra-rapid 2D6 metabolizer, you may need to select an alternate antidepressant that's not metabolized through that pathway. So, I listed out the 17 different antidepressants that you see here and then where there are actionable guidelines in which particular metabolizing enzyme and then the product labeling as well for these products. When we think about mood stabilizers and anticonvulsants, here is where most of the immunologic data lies. So, we have the carbamazepine and oxcarbazepine. This is the HLA-B star 1502 allele, and this is in particular for genetically at-risk populations. Current evidence suggests maybe the Han Chinese, the Thai, Vietnamese, Indonesian, Malay, Filipino, or Indian descent. For carbamazepine, there's also the HLA-A star 3101. There are some pharmacokinetic for 2C9 poor metabolizers for phenytoin. There needs to be a decrease here. Additionally, valproic acid is contraindicated, or we can recommend particular genetic tests to individuals suspected of having certain rare metabolic disorders, and use in those that have some of these rare metabolic disorders can lead to liver toxicity, hyperammonemia, and encephalopathy. Considerations vary through some of the mood stabilizing medications. Here are the actionable guidelines and product labeling for your review to consider where these would be appropriate. For anxiolytics and hypnotic agents, there are two recommendations currently. For STIP 2C19 poor metabolizers, clobanzame, starting dose of 5 milligrams per day with titrations proceeding slowly according to body weight, and then 2C19 poor metabolizers, diazePAM could present with marked differences in drug clearance. So additional monitoring would be warranted if considering the use of that in a poor metabolizer. I think we're at our final class of medications, ADHD medications. We get quite a few consults for patients who are utilizing higher than expected doses or higher than the FDA max of some of the ADHD medications. And so we've done a lot of literature searching to see if there is evidence to suggest pharmacogenomic testing and the need for these increased doses. The only evidence currently available is for adamoxetine. So a 2D6 inhibitor or a poor metabolizer of 2D6. Adamoxetine, you start at the same dose as a normal metabolizer, but your approach to the dose escalation is different. You need to consider increases after four weeks if the medication is tolerated and the symptoms haven't improved, just to make sure that you're accommodating for that poor metabolizer or potentially someone who might be on a 2D6 inhibitor, which is inhibiting the metabolism of adamoxetine. So sort of having a lay of the, at least the very broad guidelines and package inserts, I think it's important that we delve into testing and thinking about some of the logistic considerations of pharmacogenomic testing. By no means are these all of the things that you may consider within your own practice site, but hopefully some of this information will give you some thoughts and ideas about where you might include this in your clinical care of patients. So we briefly talked about the Inova Health System as a homegrown MediMap sort of test. So there are lots of different ways to order pharmacogenomic testing. There are commercial tests. Some of these have gatekeepers where the provider orders the test, the provider gets the result and potentially has actionable items on that. We also know that there are direct consumer testings. Many of us may have already utilized some of the direct to consumer testing for pharmacogenomics. And those results then go directly to the patient. There's no clinician involved in those. And then there are non-commercial tests. And these are ones that are done within a healthcare organization or system. Recent estimates suggest that there are approximately 75 laboratories in the US that currently offer pharmacogenomic testing. So it's well beyond this particular presentation today to think about the differences in those 75 different laboratories that are offering this testing, but there are a lot of options available. When you think about the various testing that can be done, it's important to think about, are you testing for a single gene? So the first FDA approved genomic testing looking at the Roche 2D6, which was only looking at one gene versus are you looking at a panel of genes? So you're getting a multitude of information back. Often these panels of genes will include genes that lack sufficient evidence to guide prescribing. So it's important to look at what is actually being reported back within these panels of genes. Keep in mind that the genes that provide the most content and relevant to clinical practice are the 2D6, 2C19, 2C9, and the HLA A and B. So those are the ones that are most relevant. So a lot of the other information that you're getting may not have actionable items on those. The number of gene sequence variations or alleles that are currently being tested at each of those 75 labs, there is no regulatory standards currently mandating what needs to be tested. And so it's important to understand which alleles and how many of the various alleles are being tested by each of those panels of genes. The Association for Molecular Pathology and the College of American Pathologists have published recommendations for 2C9 and 2C19 allele selection. And they're working on their 2D6, they're underway. But those are good recommendations for what your pharmacogenomic test should be reporting on and what alleles they should be looking at to make sure that you're hitting all of the important things that are known in the current literature. There is a lot of discussion about test analytic validity, making sure that all of those various star alleles that are being tested are accurately being linked to a phenotype. There can be at times some discrepancy between what one test might say is a poor metabolizer, one might call that a normal metabolizer, and really the identification of these structural variations is important. These can lead to inaccurate phenotype assignments, which can in turn lead to inaccurate recommendations to the provider. Additionally, you wanna make sure that the presence of novel or rare allele variations are accounted for. These can account for up to 20 to 30% of the variants that we see in inter-individual response to medication. And the functional impairment or impact of these rare or novel alleles can be uncertain or unknown. So it can make interpretation of genotypes difficult, which can lead to inaccurate recommendations as well. So it's important to really understand what the test that you're using is testing for and how those alleles are being identified. Testing feasibility, the availability of the testing. Is this testing that a patient will need to go to a lab for? Is this something that you're going to routinely be doing in your office space? Patient and provider acceptability of testing. It's important that you always read the fine print when looking at what you're signing or what your patient is signing. Companies have data use policies on the data that they've collected. And so you wanna make sure that you're understanding all of those data use policies. Who's going to have access to this genomic data? Are they banking the data? Will it be used for future research? So it's important to understand all of those use policies. Testing turnaround time. Many of the labs now have pretty quick turnaround time results usually within a couple of days. Previously, it had been weeks. So it is a pretty quick turnaround time. But thinking about changing of meds and how your turnaround time might work is important. And then testing affordability, understanding whether or not these particular tests would be covered via insurance or is there gonna be out of pocket cost to the patient for these particular tests. When we think about clinical efficacy and cost effectiveness, I think these are important considerations for utilizing these. Mixed data on efficacy currently. So there are two meta-analytic evaluations of the clinical efficacy of commercial pharmacogenomic testing. These tests suggest that it improves the likelihood of achieving symptom remission compared to treatment as usual. But there are also some inconclusive and negative trial findings as well. And this efficacy, this evidence of this clinical efficacy has primarily been constrained to adults of European ancestry with MDD, who really, again, have the history of antidepressant non-response or adverse drug reactions. So we don't have a diverse group of individuals to suggest that this does improve efficacy. When we think about cost effectiveness, most evaluations conclude it's cost-effective or a cost-saving strategy relative to treatment as usual. But a limitation is that most of these have been completed by pharmacogenomic testing companies. So there isn't a wealth of information of independent data resources to suggest that they are cost-effective. Most of the cost-effectiveness is seen in a reduction in visits to healthcare providers and pharmacy costs related to medication switching, as well as ER and hospitalizations due to adverse drug reactions. So just a small bit of data currently available for efficacy and cost-effectiveness. When you think about test results, the interpretation and delivery of these, one, it's important to understand who is assigning a function to the alleles that are possessed by the patient, again, making sure that those are adequately recorded, and then combining the functions to drive the phenotype that recommendations are typically made from. The process currently is pretty inconsistent and there's no gold standard approach that exists. Most of the pharmacogenomic companies have proprietary algorithms, which can lead to discordant recommendations depending on the company you're using. So it's important to make sure that you're considering how those phenotypes are being decided and then how those are being relayed. And when we think about the test results, it's important to also think of the information you're providing the company, depending on the types of recommendations they're giving. So you need to think, is the results that you're getting based solely on the genetic information, or is it taking into consideration the clinical factors or characteristics of the patient as well? So there are a host of other factors to consider when thinking about pharmacogenomic testing. Age of the patient, sex of the patient, concomitant medications. And I think this is very important when you think about the potential for drug interactions. So today I was thinking through some examples in valproic acid and lamotrigine. We know valproic acid inhibits UDP, which can increase lamotrigine levels. And so understanding that the patient's UDP is increased and this is a cause of concomitant medications versus an actual pharmacogenomic concern is important. Thinking about renal and hepatic function, inflammation, lifestyle. Smoking is one that raises a big red flag. We know that smoking induces CYP1A2. So the patient may look like a rapid or ultra rapid metabolizer of that particular medication. And then if we stop those particular medications or we stop smoking, other medications can be affected. So it's important that these are weighed in as well when recommendations are made for a patient. And then ultimately therapeutic drug monitoring. This is probably essential in being able to understand where the patient actually is based on all of their characteristics and factors, but also their genetic makeup as well and can provide some accurate objective data and understanding how someone might be metabolizing through that particular medication. So it's important to think about all of these various things. Many times when I sit down to think about a patient, looking at past medication trials, past responses, adverse effects are very important because sometimes we can deduce what their genomic makeup might be based on a good thorough medication history. When we think about test results, interpretations and delivery, one size doesn't really fit all. Again, most of the testing and effectiveness has really been in a European ancestry model with major depressive disorder. So we don't have a host of data really delving into whether this is effective and cost-effective for a very diverse population of patients. So definitely need to continue to think through that as well. Let's think about now how we might capture all of this. So if you've had your patient come, you've determined that pharmacogenomic testing is appropriate and which tests you want it to do, how do we capture all of these results within our system? The electronic medical record, good, bad, or otherwise, sometimes can be helpful and other times can be a bit of a nuisance potentially in thinking about this. So when we think about pharmacogenomic testing, we're probably not the only providers that might be considering this. If the patient has a cardiologist or an oncologist, maybe an infectious disease provider, they may also have thought about ordering pharmacogenomic testing. So thinking about where in the chart you might find pharmacogenomic testing and the results is gonna be the first hurdle to cross. Some facilities may put it as a laboratory result, others might put it as a pharmacogenomic consult. Ideally, it would be put in so that there can be automated alerts. So if someone ordered pharmacogenomic testing and we know this person is a 2D6 poor metabolizer, it would be great if that was put in similar to what we might see for an allergy so that if the next provider came along and ordered a medication that's metabolized by 2D6, that then that potentially could fire an alert so that there would be an actionable decision made about the test and the new medication. Our state hospital facilities are still on paper charts. We are moving to an electronic medical record, but where our information gets stored, who has access to that information and how do we track changes is, it's almost virtually impossible unless the patient remembers that they've had this testing done and then we need to go in search of the results. So I know some progressive hospitals and facilities do have actionable pharmacogenomic alerts, which would be the ideal way to have that going just to make sure that everyone's aware of the testing that's been done. Thinking about capturing the impact, novel research study designs that really reflect the heterogeneity and complexity of our real world patients is essential at this point. We need medical comorbidities, patients with polypharmacies, patients with diversity in their genetic ancestry, age, gender and sex, environmental exposures, all of that needs to be included as we think about future study designs and really understanding where pharmacogenomics may play a role. Routine clinical practice, we are still in need of additional, clear, curated, peer-reviewed pharmacogenomic guidelines. I think we've got a good start. As you've noticed from some of the charts that were put together, many of the medications there for antidepressants might be some of the older tricyclic agents. So really thinking about some guidelines on when to order and who to order this in is gonna be really important as we continue to move forward with pharmacogenomic testing. Comparing cost of utilization to standard of care and really looking at that cost-effectiveness analysis of genotyping patients immediately at first diagnosis, maybe versus waiting till later when they might be in a more difficult to treat situation. So the cost of genomic testing is decreasing. It is being covered by various insurance plans and it allows for opportunity for increased utilization, including medication costs. The cost of switching medication, the potential hospitalization costs that occur as we do do this. The cost of potential pharmacist time for potential consultation. I know there is a lot of expansion currently within the VA sector, but also other healthcare entities and starting pharmacogenomic services. And so the cost of that implementation of that service and making sure that the recommendations that are being offered are actionable. And then quality of life costs and ability to get a patient back to a better quality of life faster by getting to the right medication. I will say I have seen things go well and things go potentially terribly wrong with pharmacogenomic testing. At times, if we have a patient who's struggling with adverse effects or concerns about medication efficacy, sometimes getting a pharmacogenomic test done helps alleviate why something isn't working. So understanding, okay, so we figured out that you are a 2D6 poor metabolizer, so we're gonna avoid those medications now and let's move over here instead of continuing to pick those medications. But I've also seen on the other hand where we've gotten testing done for a patient who potentially is doing very well but maybe has a side effect or something to a medication and the genomic testing comes back and says, well, they should not be on that medication, and it leaves us in a conundrum of what to do. You know, the genomic testing says, don't use it, but the patient's doing well on it. Could we manage the adverse effect without changing? And so it's important that when you're getting these results back that there is a plan in place to think about how you might utilize some of those results as well. When you're searching for laboratories, here is a great site. The National Library of Medicine has genetic testing registries, so this is a great place to be able to search where you might find labs, locations, and some of the information on what they are testing for. The final algorithm that we'll kind of go through here today was recently published in the last several years that really looks at thinking about, is this patient the most appropriate for pharmacogenomic testing? Is this something I should consider? So, I highly recommend this particular article if you're looking at this and thinking about whether your patient is a candidate for pharmacogenomic testing. So, thinking first, you know, does the drug have pharmacogenomic dosing guidelines or a drug label that has genomic advice? If they, you know, you can find those resources in CPIC or on the Dutch workgroup to see if there is guidance on that. If there is guidance, are the genes implicated in the guideline, the drug label that's tested by the lab? You know, who is doing the testing? Is this a CLIA or a CAP equivalent certified lab? Does the lab disclose the variants that are tested for each gene? Are there actionable variants tested appropriate for the patient's ethnic background? So, are they likely to carry the variant? Sometimes you might want to think through, you know, if the patient's very unlikely to carry the variant, then maybe this isn't the most appropriate test or that the variants could be missing. Is the lab transparent about how genotyping results are being translated into the phenotypes so that you can understand the recommendations that are being provided? Are the results accompanied by clinical decision support? And who is making that clinical decision support? Is there a phone number to call? Are you providing additional clinical characteristics of the patient as well? And is that being incorporated into the decision support? Is the test feasible? Is it accessible? Is it something that can easily be done? Is there an acceptable turnaround time? Is it affordable? What do the data use policies look like? And then, you know, if all of those things line up nicely, the test in lab, you know, could be done and then reevaluated periodically to assure sustained appropriateness. But making sure that you're utilizing your resources along the way and thinking about where things could go wrong is very important. So, thinking about decision support tools, walking through that algorithm to me has been very helpful in trying to determine if pharmacogenomic testing is appropriate for the patient that I'm seeing. Thinking about the demographic, clinical, and lifestyle information that also plays a key role is very important when we think through whether or not pharmacogenomic testing is applicable to the patient. The major genes that currently have evidence or guidance, again, 2C19, 2D6, 2C9, and the HLAs. So, it's pretty small in comparison to what you might see within particular companies. Many companies are going to give you a host of cytochrome P450 enzymes, a lot of different receptors and pharmacodynamic information. So, some of that has less evidence and there's not a great deal of guidance yet. For pharmacodynamic genes, again, we don't have any current evidence in package inserts or guidelines. There are studies that have been done, but not enough to curate some direct guidance. It's always important to think about education or consulting an expert when getting these results and making sure you're understanding the recommendations that are being given so that you can make the most appropriate choice for your patient. So, I know that's a lot of information on pharmacogenomics in a very quick fashion, but I want to thank you for your time today and turn it back over to Dr. Cotez for some additional information. Great. Thank you so much, Dr. Ehret. Before we go to the question and answer, I just want to take a moment, next slide, to let you know that SMI Advisor is available via your mobile app and you can use the SMI Advisor app to access resources, education, upcoming events, complete mental health rating scales, and even submit questions via our consultation system directly to our team of SMI experts. Download the app now at smiadvisor.org app. All right, so let's go into the question and answer. Again, Dr. Ehret, that was really, really interesting. I learned a lot. A couple questions came up. I guess one of the first questions is, can you kind of differentiate a little bit for us pharmacogenetic testing versus pharmacogenomic testing? Oh, this is a great question, Dr. Cotez. Many times the terms are used interchangeably. Originally, way back at the beginning, genetic testing, pharmacogenetic testing was one entity versus genomics was a panel of genes, but they very much get used interchangeably at this point. Either one of them probably means the same thing. Okay, sounds good. But that's a good question to always start with. I'm curious, you know, at University of Maryland, how do you use pharmacogenomic testing clinically? Like, do you have a service or kind of what are the main clinical applications that you see in your setting? Unfortunately, we don't have a pharmacogenomic service in psychiatry. We do have one in cardiology. So we have partnered with the pharmacogenomics and cardiology service, which is run by an advanced practice pharmacist or clinical pharmacist who sort of oversees those consoles. Our internal lab will run 2D6 and 2C19 for us, so we can order those on our patients. I'm very interested now with some of the FDA ruling to think about if we're going to need to show data or if those will just be grandfathered in as acceptable if this rule is implemented about lab-based test versus FDA approved or commercial testing will be very interesting. We very thoughtfully think about patients if we are considering pharmacogenomic testing. I think I'll make a plug for pharmacists. This is a great time if you have a pharmacist, student, or maybe another trainee to do a great medication history. Go try to find all the medications the patient's been on. Think of if you can find efficacy or adverse effect data on each of those different trials, because I think at times if we think about how all of those drugs are metabolized or which transporters they potentially hit, sometimes we can predict what it looks like. You know, if they've done poorly on drugs that are metabolized by 2D6, maybe they're a 2D6 poor metabolizer, so our next choice needs to be something that's away from there. It's not something that we're routinely doing on all of our patients. It would be patients potentially where we might be stuck on next steps but don't have an idea of medication trials in the past where we can't predict what to do next. But I know some places do have a more routine pharmacogenomic service where, you know, they would be doing more prospective testing in patients. So, patients would be going to the pharmacogenomic service to get testing done and then recommendations are made. I struggle a little bit with some of those services to think about sometimes the actionable recommendations. If the patient's doing well, you want to make sure that you're not discontinuing a medication just because the genomics test said, you know, this isn't a good medication for them, but they're doing well, they're not having side effects, so making sure it's well thought through. Okay, the next question I wanted to ask you a little bit about is, what would you say would be the strongest recommendation for using pharmacogenomic testing, like, that you would almost consider reflexively? Is there anything that's there that would rise to that level, maybe say like carbamazepine for people of certain Asian ancestry? That's about the only one that I, off the top of my head, that I would think of that would be the knee-jerk reaction. And it might, you may or may not even do it unless carbamazepine was your only option for the patient. You know, you might think about a different mood stabilizer instead of, you know, doing the test and doing that, working down that pathway, but none of the others I would say are like knee-jerk reactions. I'm going to use aripiprazole, let me get, let me know if you're a 2d6-poor metabolizer. There's not any of them that have to be done versus in other disease states, you know, for oncology or some of the HIV medications where it's, you know, you need to get this test before we issue this medication because we need to know it to dose it or whether or not it may be effective against that illness. Would you put oxcarbazepine up there with carbamazepine? Most likely I would just to avoid the potential for the adverse effect. Yeah. Okay. You know, one question I want to ask you is, it seems like, at least in psychiatry, and many psychiatrists, I think, have a fairly basic understanding of how drugs are metabolized and that, you know, I'm wondering, you know, are there resources you would recommend for prescribers or pharmacists to really kind of get a better handle on drug metabolism to really, I think, because it would really help, I think, put this talk into context a bit more. I prefer, and this is my own professional opinion, I like the Flockhart cytochrome P450 table. I find it to be very comprehensive in looking at how, one, which medications have substrates, right? So, all of the medications are processed through our body in some factions. So, understanding which enzymes metabolize the particular drug we're looking at, but then also looking to see if the drug we're using is an inhibitor or an inducer as well, because it's really three things you have to look at for medications. And the Flockhart table is very well set up and designed to look at those three areas and allows for some information. So, you can click on things to sort of see and read a little deeper into it. It's also important to think about, too, when you look at these enzymes and it can be overwhelming. Is it a major metabolite or is it a major inducer or is it a minor inducer? Because sometimes the minor ones may not play as large of a role as the major ones. But sometimes also our body can accommodate and think about if this pathway is shunted, it might move the medication to another pathway. And so, sometimes all of those things are happening and thinking about where these medications may interact is important. Thank you. Can you talk also just a little bit about phenoconversion? And you spoke sort of about it a little bit, but I'm curious if maybe we can just talk a little bit about phenoconversion and sort of the differences between the genotype and then sort of the phenotype and considerations there. So, phenoconversion is very important when we think about pharmacogenomics, but also thinking about medication in general. It's really the phenomenon that someone who might be a normal metabolizer gets converted to a poor metabolizer of a drug, and this can be due to all kinds of things. Maybe another medication, so there's an interaction happening that makes it, so you're a poor metabolizer, so you have a medication that's inhibiting a particular cytochrome. Or it could be a comorbid condition. I was thinking today of what comorbid conditions are important, COVID. Look at all of the struggles we had with COVID and clozapine, and that inflammation potentially changing how our clozapine was metabolized. And we don't want to draw conclusions on someone's genomic information based on potential other medications or other disease states or conditions that the patient are on. So, looking at drugs that are susceptible to this, things that are metabolized via these different pathways, and thinking about what dosage adjustments might need to be made because of these different other factors. And so, these sometimes aren't taken into consideration when you think about pharmacogenomic testing. So, if you just base your results just on the test alone without thinking about your clinical factors, you might miss a big opportunity for something to go wrong. I think that's really helpful. The next question I wanted to ask you a little bit was about, are there any differences in obtaining the sample via buccal swab versus the whole blood testing? Is there a preferred option, do you think? Or is it fairly similar? Results are probably fairly similar. I don't think there's a large difference. It's really probably patient, you know, acceptability, ease of collecting that sample. I spent two years in my research lab as a fellow collecting buccal swabs. And this was old school, like swabbing people's mouths versus, you know, some of them now are just spit into the little tube or the cup and shipping it off as long as it is done according to the standards that are set forth in the particular test and aren't subjected to, you know, degradation, then they should be both pretty equitable and similar. Okay, sounds good. And one more question. I'm going to ask you to speculate a little here. 10 years from now, where do you think we're going to be with pharmacogenomic testing? I don't know that we're going to be, I hate to say this, much further than we are now. I hope we are. I was thinking back, where were we 10 years ago versus where we are now? We have a lot more tests now than we did 10 years ago. But I don't know if our data and recommendations are any further along than where we might have been 10 years ago. We have a lot of research, I think, a lot of great information on it, but it just hasn't collated together to say it's actionable yet. Yeah. Okay, well, sounds good. Dr. Aratt, I want to just thank you again for this really wonderful talk. And now we'll just go into a couple of slides here. So, if you have more questions about this topic that you'd like to discuss with colleagues, post a question or comment in the SMI Advisors Discussion Board. This is an easy way to network and share ideas with other clinicians who also participated in this webinar. If you have questions about this webinar or any other topic related to evidence-based care for SMI, you can get an answer within one business day from one of our SMI Advisors National Experts. This service is available to all mental health clinicians, peer support specialists, administrators, and anyone else in the mental health field who works with folks who have SMI, completely free and confidential. All right, next slide. SMI Advisor offers more evidence-based guidance on psychopharmacology, such as the resource Myth vs. Fact for Serious Mental Illness on psychopharmacology. You can see on there, the first one is, you should not prescribe clozapine until all other medications have failed. Myth. And then we have some of the fact information there on the right side. And we also have other Myths vs. Facts on LAIs and weight gain and clozapine. Okay, next slide. So, to claim credit for participating in today's webinar, you'd need to have met the requisite attendance threshold for your profession. After the webinar ends, please click Next to complete the program evaluation. The system then verifies your attendance for the credit claim. This may take up to an hour and can vary based on local, regional, and national web traffic and the use of the Zoom platform. Pharmacists should claim a credit of participation credits after completing the evaluation. Your Continuing Pharmacy Education CPE certificate will be mailed to you by the end of the day tomorrow. If you're claiming CME or NCPD credit, you can also find the certificate in your SMI Advisor account after completing the evaluation. All right, next slide. And finally, please join us in two weeks on November 14th for Dr. Scott Zeller, who's going to be talking about Empath Units, Improving General Hospital Behavioral Emergency Care. And this webinar, again, will be on November 14th from 4 to 5 Eastern. And we hope to see you there. Thanks for joining us. And until next time, take care.

pharmacogenomics

psychiatry

pharmacogenomic testing

antipsychotics

antidepressants

mood stabilizers

anxiolytics

ADHD medications

drug metabolism

genes and alleles