Ina Liko:
Hello everyone. Thank you for joining me today. My name is Ina Liko and I’m the Director of Clinical Pharmacogenomics Operations at RxGenomix. Thank you again for joining us in May. I’m finally back after attending some conferences. So I’m excited to be back for our first today’s session and continuing in the following months, as well. May is Mental Health Awareness Month, and this month, for the last two sessions today, and next week, I’ll be actually be focusing on some mental health topics. As always, if you have any questions, please put them in the Q&A portion or the chat portion, and then I’ll address them at the end.
Today, like I said, we’ll be focusing on mental health medications, mainly on SSRIs. The two main genes that affect these medications are CYP2C19 and CYP2D6. I’ll be reminding us about these two genes, how they work, how they affect these medications. We’ll see if there are any civic guidelines. And then, also, if there is any FDA language. And then we’ll go back to our patient case and see how we can help her.
Before we dive into some of the details on CYP2C19 and CYP2D6, again, I’d like to focus our attention on this patient. And then keep this patient in mind throughout the presentation, especially when we’re going through different recommendations for different drug gene pairs, as well. So, this is Mindy. She is a 45 year old, Asian female who works as an executive director of client engagement for a big financial firm. Now, due to the pandemic, she actually has been working from home, like a lot of us have, where she lives there with her husband and three children. The pandemic has really disrupted her lifestyle and she was diagnosed with depression last summer. And she actually has been trying to control it with a few medications that haven’t necessarily worked for her.
So she trialed Citalopram for three months and didn’t see any improvements, didn’t see any response or improvements on that. Then she tried Paroxetine and she experienced severe headache, dizziness, difficulty concentrating, forgetfulness, and confusion within the first month. She continued for a few months because she did not see… I think it was actually causing or helping with her depression, but also did not recognize that Paroxetine was the one that could have been causing some of the side effects after a few months. So she goes her doctor back again and [was special 00:02:52] Sertraline, where she also trialed that for about three or four months. And even with that medication, she didn’t see any success, so she’s not responding. So, at this point, it’s been quite a while for her. So it’s been a few months, at least six or seven months. She’s tried three different medications and nothing has really worked.
So she comes back to the doctor’s office and her psychiatrist says, “Oh, well actually, there is this pharmacogenetic testing that’s available. We’re going to do that for you. See how that can help.” So, he orders her the pharmacogenetic test. And then these are the results. The psychiatrist now comes back to you and asks, “Well, what medications should we start her on, given her history?” And then, we’ll talk through some of those options as we go to the end of the presentation. But for now, we’ll keep her condition in mind, and then as we move through some of these other slides where we’ll gain a better idea of what are some of these drugs and interactions, why those medications did not work for her, why she had some of the side effects, and then what the best option would be for her.
Before we go into some of the details with the drug gene pairs, I also wanted to remind us about depression and especially depression through the pandemic. We’ve seen that those numbers have actually increased a lot more than before the pandemic. In the US, the lifetime prevalence of major depression is about 17% of the population. Now during the pandemic, the Center for Disease Control and Prevention reported that the percentage of adults with recent symptoms of depression, it increased from 24.5% to 30.2%. And we see that the prevalence is increasing. However, depression is very difficult to treat due to the delay of antidepressant effects, as well as there’s this interpatient variability and response, as well as the experience of side effects, because a lot of patients will experience side effects and then they will not want to go back with their doctor to try different medication, because they think that that other medication would also have more side effects for them, as well.
That’s also the challenge with treating depression, just mental health in general. Now there are associations of genetic variants that are associated with antidepressant drug response have generated great clinical interest from clinicians, but as well as patients. And that’s why we have also a lot of other development of pharmacogenetic tests. So today, we’re mainly focusing on SSRIs and the genes that affect them, but again, the rest of this month and actually next week, I’ll be talking about some other medications that affect mental health.
Now that we have an idea about our patient and just journal idea on how prevalent depression is, I’d like to dive in a little bit more on some of these details. And one of the main genes that affects SSRIs is CYD2C19 and, oh, sorry, there’s a typo here. CYP2C19, not CYP2C9. But CYP2C19 metabolize about 10% of common medications. There are over 30 known allelic variants of CYP2C19. And they can range from single nucleotide polymorphisms, or SNPs, to where can have mostly commonly reported alleles categorized as increased function, normal function, or no function. Now, these genetic variants can alter the pharmacokinetics of the enzyme and can lead to increased, decreased, or no function enzyme activity. And here is where we will actually derive a lot of the phenotypes for CYP2C19, depending on whether the patient has an increased, decreased or no function enzyme activity. And then depending on these combinations, we’ll get different phenotypes.
These are the four genetically-mediated CYP2C19 phenotypes. An ultrarapid metabolizer would have two increased function alleles or one normal function and one increased function allele. For a normal metabolizer, would have two normal function alleles, so this will be the default tier. *1/*1. For an intermediate metabolizer, we have one normal function allele or one increased function and one no function. And then, for a poor metabolizer, it’d be two no function alleles. So, generally speaking, unless a medication is the pro drug, if a patient is an ultrarapid metabolizer, we would expect that they would either need a different drug, or they would need a higher dose because the medication is not in the system for as long as it should be in order to have an effect on them. That’s why either switch them to a different drug, or they may need a much higher dose.
And then on the flip side of that, we have the intermediate or the poor metabolizers, where there’s just not enough of that enzyme to metabolize the medication, all of the medication at the rate that it needs to be. In that case, the patient may actually have some side effects because we have that leftover of the medication in the system. And then of course, with the poor metabolizers, it would be at a greater extent. Now on the flip side of that is actually, if a medication is a pro drug, so it needs… In pro drugs, I’m sure you remember from class, but we do need that enzyme to activate the medication. In the case of an ultra rapid metabolizer, if we’re activating that medication to just a really fast, so then you get more of that active metabolite in the system. It may cause more side effects.
And then in the case of an intermediate or poor metabolizers, for a pro drug, you’re just not activating that medication fast enough. So you don’t have enough of the active metabolite in the system and the patient may actually not have an effect, or they may need a different medication.
And then here, I would just like to point out the pathway for just one of the SSRIs here that the patient was taking. So this is Citalopram, mainly metabolized by CYP2C19 and CYP1A4. We first need to find out what this patient’s phenotype is, and then go from there when it comes to the recommendation. And then, just as a small alert here, but the main enzyme that a lot of literature is out there when it comes to pharmacogenetics, is with Citalopram and CYP2C19. I just wanted to point out that there are these other two genes or the enzymes that do help in the metabolism of Citalopram. So we have to take into account all of them, but at different extents.
Now, you may wonder if this is just a rare variant for our patient or not. And this chart shows the CYP2C19 phenotype frequencies. I want to point out that the ultrarapid metabolizer is not necessarily as common. However, it is found in 4.3% of African Americans and then for Central or South Asians, about 2.9%.
So now that we talked about CYP2C19, I’d like to overview CYP2D6. And for those of you that have attended my talks before, this will be a review. But briefly, CYP2D6 is a highly polymorphic gene, and there are over 130 known allelic variants of CYP2D6. These are structural variations. Some of these allelic variations, they’re single nuclide polymorphism, so, similar to what we saw to CYP2C19. Some of them will be insertions or deletions, duplications, as well as multiplications.
Now, again, these genetic variants, they can alter the pharmacokinetics of the enzyme and they may lead to increased function in those that are carrying additional functional gene copies, as well as lead to decreased or no function. For example, gene deletion, and some of these variations, they have no functional consequences that we know of, as of yet.
Similarly here, to CYP2C19, we have four genetically mediated CYP2D6 phenotypes. However, what’s different about CYP2D6 is the way that we calculate these phenotypes. Each of these alleles for CYP2D6 is assigned an activity value that ranges from zero to one, and it’s the combination of these activity values or alleles that will give us an activity score. As humans, we’re diplotype organisms. We got one strain from mom and then one strain from dad. And that’s why we’ll always have this diplotype. Now for CYP2D6 here, this, for example, if we look at this particular genotype, *1/*2xN. So, this times N, it means that there is a multiplication. There is at least, this number could be 1, 2, 3. It’ll depend on the particular genotype of the patient.
In this case, *1 times, let’s say this is one times one. It will be two. So one, and then the activity score for a *1 is going to be one. And then we just add these two together. We get a score of three. Activity score of three, puts us in the ultrarapid metabolizer. Now, some of these other alleles, for example, *4 and *5, they’re a poor metabolizer. They have an activity score of zero. If we add these two together, they’ll have an activity score of zero, and that makes a poor metabolizer.
An intermediate metabolizer would be someone that has an activity score between zero and 1.25. And a normal metabolizer is someone that has an activity score between 1.25 and 2.25. This is how we are mainly calculating CYP2D6 phenotypes and then from this phenotype calculation, we then derive different recommendations for different drug gene pairs.
Similarly here, I just wanted to point out was the frequency of these phenotypes. For intermediate and poor metabolizers, we see that in Americans, about 25% of the population has a variation. And then this number is a lot higher in East Asians, particularly, it’s about 40% of the population that will have a variation in either being an intermediate metabolizer or a poor metabolizer. And you can see that the intermediate metabolizer status is more common than the poor metabolizer status.
All right, so now it’s an exciting part of finding out what’s the phenotype for our patient. So for CYP2C19, as a reminder here, she was *17/*17. Now, *17 has an increased function allele, so she’s an ultrarapid metabolizer for CYP2C19. For CYP2D6, her genotype is *4/*4. Now, the activity value of *4 is zero. We just saw that and few slides ago. Now, the total activity score will be zero, and that puts her in a poor metabolizer phenotype for CYP2D6. She’s kind of opposite when it comes to these two genes. CYP2C19, she’s metabolizing medications very, very fast. And then CYP2D6, she’s just a poor metabolizer, not metabolizing medications as fast.
So, and again, we’re not talking about pro drugs here. In this case, medications that are metabolized or broken down by CYP2C19, we would expect that she would maybe need a higher dose, or maybe she wouldn’t respond to that particular dose that was given to her. And then for CYP2D6, maybe she’ll have more side effects because again, she’s a poor metabolizer, so more of that medication is in her system for longer than it’s supposed to be there.
Now that we know the patient’s phenotype, let’s look at some medication, examples here, and I’d like to focus first on the CPIC guidelines. We do see, these are for SSRIs, mainly. Here, I’ve included Citalopram, Escitalopram, and Sertraline. So if you remember, our patient did not respond to Citalopram and Sertraline, and mainly CPIC guidelines say if you’re an ultrarapid metabolizer, for Citalopram, Escitalopram, consider an alternative drug that is not metabolized by CYP2C19. For Sertraline, they say to initiate with recommended starting dose, however, if the patient does not respond, consider an alternative drug.
And then for a poor metabolizer, you would consider a 50% dose reduction or to select an alternative drug. If you’re an intermediate metabolizer or a normal metabolizer, these three medications, CPIC recommends to initiate therapy with recommended starting dose.
And then now, let’s look at CYP2D6, because our patient also tried one of the medications that is affected by CYP2D6. In this case it’s Paroxetine. CPIC recommends for an ultrarapid metabolizer, for Paroxetine, select an alternative drug, that it’s not metabolized by CYP2D6. And for Fluvoxamine, there are no recommendations because of lack of evidence. When it comes to a poor metabolizer, so our patient did have some side effects with Paroxetine. CPIC recommends to consider a 50% dose reduction, if Paroxetine use is warranted here or to select an alternative. Fortunately, there are other alternatives, we can use for her, as well. And then, just for completeness here, I have also included Fluvoxamine. Consider a 25 to 50% dose reduction or select alternative drug.
The FDA also has some language when it comes to some of these medications. And this is from the FDA Table of Pharmacogenetic Associations. For a poor metabolizer and the patient is taking Citalopram, they say that will result in higher systemic concentrations, as well as adverse risk reaction QT prolongation, and the maximum recommended dose is 20 milligrams. If a patient is taking Escitalopram, and this includes pretty much a lot of the phenotypes, so ultra metabolizer, intermediate, or poor metabolizer, the FDA says that this may alter systemic concentrations, doesn’t necessarily have any particular dose recommendations. And then for CYP2D6 poor metabolizers, particularly for Fluvoxamine, it will result in higher systemic concentrations, the FDA says to use with caution. And for CYP2D6 ultrarapid, intermediate, or poor metabolizers with Paroxetine, again, it has a general statement saying that this may alter systemic concentrations, however, doesn’t necessarily have any particular dose recommendations.
All right. Now, let’s go back to our patient and just a reminder here, again for her assessment. Citalopram and Sertraline did not work for her because she’s a CYP2C19 ultrarapid metabolizer and for CPIC guidelines, we would want an alternative drug. Paroxetine, she experienced a lot of different side effects. Again, she’s a CYP2D6 poor metabolizer and per guidelines as well, they would expect that she would have actually more side effects with it, since she’s a poor metabolizer. And CPIC would recommend a 50% dose reduction. In terms of her plan, I’ve just included some alternatives here for her. These are just some of the alternatives that I chose. There’s definitely other alternatives out there as well. But a good option for her could be Duloxetine. It’s not affected by CYP2D6 and the Dutch Working Group does not necessarily have any recommendations when it comes to those adjustments. So, that would be a good choice for her.
Mirtazapine is also a good option. Again, not metabolized by CYP2D6, not necessarily affected by CYP2C19, either. And then, Paroxetine could also be a good choice for her, but again, we would start with that 50% dose reduction. This will depend on the patient, how she feels about it. She did take Paroxetine in the past and did not have a good experience with this. So she may be a little bit more reluctant to try it again. However, if she’s open to it, we could start Paroxetine at a lower dose, 50% dose reduction. These are my references. Thank you so much for listening to this presentation. Please let me know if you have any questions. And any questions come up later, that’s my email address. Feel free to email me any time.
Interested in registering for PGx Clinical Pearls?
Every Thursday at 3:30pm Central Time
https://us02web.zoom.us/webinar/register/WN_Xmg9jxiASRmA2wAeUo3YeQ
For any questions, you can email us at [email protected]
