 All right. Great. So we're in the final call of the Insure Staff Education webinar series, and Dr. Vicky Pratt from Indiana University is going to talk to us about pharmacogenetics. All right. Great. Thank you, Bob, for having me talk for the final, final, final one on pharmacogenetics today. So I thought I'd give you a couple of objectives of what we're going to go over today. We're going to talk about the star allele nomenclature, which is specific to pharmacogenetics, and then differentiate between various metabolizer phenotypes. And this is all part of the pharmacogenomic and pharmacogenetic words that are often used when we talk about pharmacogenetics. And then recognize the common drugs that can have adverse drug reactions. So I'll use pharmacogenomics and pharmacogenetics somewhat interchangeably, but really we're talking about precision medicine and getting the right dose to the right patient. So if you look really what is pharmacogenetics, you can Google on the web and look at different sort of definitions, but it's really about genetic changes that can give rise to different responses to different drugs and looking at those genetic changes and how they affect drug metabolism. So this is from the personalized medicine coalition from a report a few years back. And if you look at various different kinds or types of drugs, people respond to them differently. If you look at hypertension, specifically ACE inhibitors, about a quarter of folks cannot who take this drug, it doesn't work for them. And beta blockers, it's about 15 to 25% of those drugs don't work well for them. Anti-depressants up to 50% of anti-depressants are ineffective. Statins, it can be up to 70%. And asthma or the beta 2 agonist, it can be also up to a 70%. So there's a lot of trial and error to see if a medication will work for people. And looking at pharmacogenetics is a way to help reduce that trial and error. So this is a partial list that I put together a few years back of some various kinds of changes in genes and some of the drugs that are involved with the sort of a drug gene pair. That's when you're thinking about pharmacogenetics. And I'm a geneticist, so that's my caveat is I tried to make this so that if I can understand it as a geneticist that hopefully everybody else can understand it, is that pharmacokinetics is what the body does to a drug. So if we look at the first in activates over time, then in the second profile, this is an inactive drug that the body activates and then slowly inactivates over time. Pharmacodynamics is what the drug does to the body. Where is the most effective dose of the drug? This is when you take a medication, you take it twice a day or three times a day with meals. You will see peaks or I mean troughs and peaks of drug concentration in the body. And the idea is to eliminate the effects of the peaks and troughs and keep the drug concentration in the body at the most effective concentration. So a lot of what we're going to talk about today is not necessarily the pharmacodynamics, but we're going to talk about pharmacokinetics, especially related to some of the drugs and genes that we're talking about. So if we look at adverse drug reactions, about six to seven percent of all hospitalizations have an adverse or result of an adverse drug reaction and it accounts for about 100,000 deaths each year or about the fourth leading cause of death. One in five or 20 percent of drug candidates that are under development are actually terminated due to an adverse drug reaction that is seen in patients. So I like to think, and here I'm talking about pharmacogenomics, that there are major kinds. We have our inherited or constitutional changes in our DNA, in a patient's DNA or our DNA that affect how drugs are metabolized in our body. And that's where we're going to spend a lot of this talk today. The other one is the somatic or tumor-specific changes, and this is the K-RAS and the BCR-ABLs and C-KIDS and B-RAS, where there is a lot of work of really targeting that cancer to a specific drug to kill that cancer or control that cancer. And then what we're not going to talk about today that I like to think of in pharmacogenomics is also for infections such as HIV and HCV, there's changes in those viral genomes that drugs are specifically targeted against. So based on the variants seen in HIV or HCV, there are certain drugs that are more effective in treating those infections. But we're not going to talk about those. I want to introduce you to a specific term called a companion diagnostic, and that's where most of the oncology type stuff is, area is right now. And this is where the drug and the pharmacogenetic tests are approved by the FDA at about the same time. So a drug manufacturer they're working on looking at lung cancer and they see that this drug works in ALC-positive breast cancers, and so that test and that drug went through the FDA at the same or similar time and were approved together by the FDA as a companion diagnostic. So let's talk about the inherited or constitutional variants. They're primarily close of a gene family called the cytochrome P450. Most of these proteins are located predominantly in the liver and they're involved in the toxin and drug metabolism of the body. So I want to introduce you to the nomenclature around the cytochrome P450 because it can be a little bit confusing. So the star is called star allele nomenclature and a star one is considered the normal allele where no variant or mutation is detected. All the star alleles are numbered in order of description and there are sub alleles that are alphabetized in the order of description. And there's a website that was started out of the Karolinska Institute in Sweden that summarizes all these star alleles. And I will disclose that this is actually, the folks that started it in Sweden are actually retiring. So this is moving to being a U.S. and multinational consortium and I am on that consortium where we're taking over the star allele nomenclature. So if I show you an example, so if we look at cytochrome 2D6, so cytochrome CYP is for cytochrome. This is the protein family, the 2D6, so it's the family 2, subfamily D, isoenzyme 6, if you're looking at the protein, this CYP 2D6 itself is just, is this also the gene? So since I mentioned this, we're mostly going to talk about the gene. Then we have a nice star here and after the star is the, so if we have a star 4, it's the third, remember star 1 is no variant detected, so it's the third variant described and then if we have an F, it's the fifth subvariant described in that gene. So this is an example from the CYP allele website looking at 2D6 star 4 and this is just a partial list. So to be a star 4, you have to have an 1846 G to A and this is in HGVS nomenclature, so you have to have an 1846 and that's what's in bold on this, in this slide. So that's the defining variant. This is old school haplotype analysis, so also on that same allele, you usually see a 100-C to T variant and a 4180 G to C variant and you can see in all these sort of sub alleles, there's different combinations of other variants that were observed in that haplotype. Also point out specifically down here, remember our 1846 is our defining a variant in star 4, but in this instance here in star 4M, 100-C to T and 4180 G to C were not observed, so there can be star 4 alleles that don't have the 100-C to T or the 4180 G to C in it. So if we look at each allele, one that we inherit from mom and one we inherit from our father, that taking all those variants, we're going to assign a function to that allele. So CPIC, which is the Clinical Pharmacogenetics Implementation Consortium CPIC, there's a website here, put together a project of trying to standardize that nomenclature of allele function designation. So a normal allele, or even if most of the time it's like our star 1, that is a normal functional allele. An allele can be a decreased functional allele, you can have a no-function or non-functional allele, an increased functional allele, and then if you're uncertain, you can have an uncertain or unknown functional, especially a newer allele as different methodologies such as next generation sequencing, and there's a unique combination that is observed and you really don't know whether the function of the allele in that individual. When a lab looks at the various combination of alleles and assign a function, they use that to assign a metabolizer status. And a metabolizer status is just the phenotype or the predictive phenotype based on the genetic variants observed in a person. So if you look at a normal metabolizer, and I have in parentheses that it's called an EM, it used to be called an extensive metabolizer, but again back to CPIC, that group standardized nomenclature in pharmacogenomics. So if you have two functional alleles, you're called a normal metabolizer, and in the old school it was previously known as extensive metabolizer, which you may see in the literature. Somebody who has a decreased functional allele and a non-functional allele or no-functional allele, that person is called a reduced metabolizer, excuse me. I have this one in sort of a different normally. It's called an intermediate metabolizer, and that's my bad, but it's sort of a reduced functional or reduced metabolizer. So the correct nomenclature here is intermediate metabolizer, which is this IM, but think of this as a reduced metabolizer. There's a poor metabolizer that has two no-functional alleles or a PM, and then there's a rapid metabolizer, which has one increased functional allele, or an ultra rapid metabolizer, which has more than one increased functional allele. So I want to sort of give you how I think of this, is that if you look at a normal distribution of people, and in general when you're talking about drug metabolism, most of the time you're worried about the people at the far end, the ultra rapid metabolizers and the poor metabolizers. In the middle of the normal metabolizers, oops, I clicked quick, are the normal metabolizers, and then you have this sort of reduced intermediate metabolizer between the normal and the poor metabolizers. If you're looking at different, even though your genotype doesn't change, you can look at various different drugs, and in some people that window is really a narrow window between for the intermediate reduced metabolizer, and in others, that window can be much bigger, and if that is drug specific, not necessarily genetic specific. I want to introduce you to also, it's not just your genetic changes, but there are other things that affect metabolizer status. There are certain non-drugs that can affect metabolizer status, so if you've ever taken a medication that says do not take with grapefruit juice, and the reason is, I should say grapefruit and grapefruit juice, is that it inhibits cytochrome 3A4, and this is one of the major pathways that many drugs go through. Actually, if you eat your broccoli, it actually induces cytochrome 1A2 to work a little bit better. If you take St. John's wort, and some people, this is an herbal supplement that many people take over the counter when they are feeling a little blue or for depression, this actually induces the 3A4. Now if you look, grapefruit inhibits and St. John induces, I'm not sure that they cancel each other out, and I wouldn't try this at home, but there are things that can affect how our cytochrome P450s work. So I'm going to go over one of our first examples, it's cytochrome 2D6. There are more than 100 different unique alleles that have been described, and it's estimated to metabolize about one fourth of all drugs. Remember, I told you that we need to worry about our ultra-rapid and our poor metabolizers. About 5 to 10 percent of Caucasians are actually poor metabolizers for cytochrome 2D6. Some of the medications that 2D6 metabolizes are antidepressants such as fluoxetine, and I will apologize in advance that I screw up the pronunciation of many medications. It's involved in pain management such as coding, antipsychotics, and then there are drugs that such as fluoxetine, if you're taking it, it actually can inhibit 2D6. So that's important for the people that are intermediate or reduce metabolizers where fluoxetine may not work very well because while it goes through the 2D6 pathway, it also inhibits the 2D6 pathway. I'll talk about a specific sample with 2D6 related to coding. Coding can be used in the treatment for pain. It also can be used to treat coughs and diarrhea. You may have seen on TV there's commercials that say, what is it, opioid-induced constipation. So coding is one of those that causes opioid-induced constipation. So it can be used to treat diarrhea. It is currently accepted for medical use, but it has severe restrictions. So essentially you need a prescription to get coding. In Canada and other countries, you can actually get it from behind the counter in the pharmacy without prescription. There is abuse of coding and that it can lead to severe psychological or physical dependence on coding. So coding is largely an inactive drug. It does have some, can alleviate some pain, but it's largely an inactive drug that is metabolized through 2D6 to its more active form morphine. And then there's clinical pharmacogenetic implementation consortium guidelines related to your genetics and how to dose for that. In the U.S., there's actually a black box label or black box warning on coding, especially for nursing mothers. Coding and morphine actually, I should say morphine, can get in the breast milk. And there's been babies that were nursing that actually died because they've got too much morphine at one time and these babies have died. So this has led to a black box warning about coding that people that are ultra-rapid metabolizers get too much coding too fast, especially children. And this is why coding is also contraindicated in kids getting tonsillectomies, that ultra-rapid metabolizers get too much morphine and they can stop breathing. So about 1 to 10 percent of Caucasians are ultra-rapid metabolizers, 3 percent of African Americans are ultra-rapid metabolizers, about 1 percent of Asian, especially Chinese, Japanese, and Hispanics are ultra-rapid. And in the population that's really high ultra-rapid metabolizer status is North Africans, Ethiopians and Saudi Arabians with a little over a quarter of people are ultra-rapid metabolizers. FarmGKB website, and if you look at coding, and remember all our cytochrome P450s are mostly in the liver, so coding gets into the liver and then goes through the 2D6 pathway here to become it's more active for morphine. So remember, if we looked at the Pharmaco, see now I get confused again. If we look at the profile of this drug, this is the inactive drug that gets activated and then inactivated through these pathways and eliminated from the liver and the body. So if we look at dosing for coding in intermediate metabolizers, it is recommended to increase the dose of coding because it may have insufficient pain relief and either that or consider alternate out amalgésia. For poor metabolizers, coding doesn't really work at all for these folks and that you should avoid giving coding to them. Another drug that's also used for pain relief, which is traminal, that goes through the same pathway and ultra-rapid metabolizers. So this one you're using for avoiding because of an efficacy of the drug. The drug just doesn't work very well. And ultra-rapid metabolizers is also recommended to avoid coding and traminal because in these folks the coding gets transferred or changed into morphine way too fast and it can cause issues with people to stop breathing. So in adults, especially in children and newborns, though in adults most people tolerate it fairly well. Because I've known a few ultra-rapid metabolizers and they really like coding because it really ticks the pain real fast. So if you're looking at CPIC guidelines, and this is from the CPIC guidelines, that ultra-rapid metabolizers and this is sort of what I just put in that summary slide is that avoid coding due to potential toxicity and poor metabolizers avoid coding due to lack of efficacy. There's I'm not going to go into this slide way too much but there are DNA testing platforms that are available for lab, this is more for the lab folks who would be interested. Another example we're going to talk about today is cytochrome 2C19. There's more than 25 alleles described with this gene and about 3 to 5% of Caucasians are poor metabolizers and about 15 to 20% of Asians are poor metabolizers for 2C19. Some of the medications that go through 2C19 are the proton pumps inhibitors such as Prilosec or Meprazole and antidepressants like amitriptyline, cetylopram, acetylopram, chomiphramine. And then what an example we're going to talk about today is clopidogrel or plavix. This is just looking at the star alleles and this is how the labs and people predict the metabolizer phenotype is looking at the various combinations. So as you can see where no variant is detected or our star one metabolizer or our star one star one are normal metabolizers and about 35 to 50% of people sort of fall in this range. The more common alleles are star two in Caucasians and star three in Asians and the rest of these changes in star and 2C19 are more rare. So if we look at our example of clopidogrel or plavix, it is used to treat coronary artery disease, peripheral artery disease and cerebrovascular disease. It is metabolized by cytochrome P450, again this is an inactive drug that gets activated in the body and it has this nice big long name that I'm not going to talk to say. A few years back the FDA announced that if you're taking clopidogrel it can't be taken with a Meprazole or SOMeprazole because they actually inhibit 2C19. It's also clopidogrel is used to end stemmy and stemmy and along with aspirin to prevent thrombosis after step placement. In 2010 the FDA changed the black box warning on the label to say that poor metabolizers do not effectively convert plavix to its active form and then there was a big media campaign in magazines and on TV to let people know that poor metabolizers should not be treated with plavix. At the time this black box warning went into effect there were no other medications that were available as alternative to plavix or clopidogrel. The FDA recently updated the label and I put this in red just this past fall in 2016 that the drug label suggests that there are different platelet P2Y12 inhibitors that can be used in patients identified as poor metabolizers. So if we look at clopidogrel here remember it's an inactive drug that goes through and this is a little always sort of different these pathways are there's probably a lot of things that are not here but it goes through and then 2C19 here as well as various other genes to become the active metabolite and then this active metabolite with the big long name then binds to the receptor on the platelet to prevent it from aggregating or developing a clot. So if this is black then clopidogrel is just eliminated through the body and cannot and then cannot bind to the platelet. Guidelines recommend that if you're an intermediate metabolizer that you take pressigrel or tachagrelor or some other alternative therapy if there's no contraindication as well as the poor metabolizers also you take pressigrel or tachagrelor or some other therapy if there's no contraindication. In the ultra rapid metabolizers it's fine to take clopidogrel you get a little bit more through the system a little bit faster it does not it doesn't cause any issues though it it has been reported with some association for a risk for bleeding. Again this is a CPIC guidelines that I just summarized on the previous slide so I'm not going to go through that again and for the lab folks or people that are interested there are there's lab develop test or lab develop procedures as well as FDA and research only platforms for performing testing. There is one test out here I'm just going to remark on this one this is more of a point of care test is verified now it's a platelet function test or platelet function assay that does not test for the genetic variants but only looks at the platelet function to see how well the clopidogrel is binding to the receptor. We're not going to go too much over this slide only to say that there is some variability in the platforms of what alleles that they test. We're going to go into our third example which is warfarin or kumadin and when it was named I thought a little interesting bit of trivia that warfarin got its name from the Wisconsin alumni research foundation or warf that originally patented so just to be a little silly so if Indiana patented it would have been IRF I think warf sounds a little bit better so good thing it was named in Wisconsin instead of Indiana. So it was originally marketed as rat poison so basically the rats would go it was mixed with corn the rats would go eat the corn with the warfarin on it and then they'd go off and bleed to death and die. In the early 1950s there was a military recruit that really didn't want to go into the army and so tried to become a rat like and eat the warfarin, eat the rat poison and then really started research into the anti-coagulation properties of warfarin. Right now it's prescribed more than 30 million times in the U.S. each year and accounts for about 40 little over 40,000 ER visits each year as when people can get unstable in it. There's a very narrow therapeutic window for warfarin so you get a little too much, you bleed, you not get enough, you clot so it's part of that whole clotting cascade that keeping that in control can be difficult at times. There's two major genes associated with about 40% of the variability in the drug response for 2C9 and those are 2C9 and B-core C1. I will tell you, I think I got the slide in here, there's a website called warfarin-dosing.org. There's some other websites but this is probably one of the ones that many people use that can take somebody's genotype and look at their genotype, their weight, their height, their age, other medications that the patient's taking and then help predict what's the most optimum starting dose for and how to change the dose of warfarin for a patient. Normally patients are monitored with an INR or an international normalized ratio so somebody who's had a blood clot and are taking warfarin, you really want their INR between two to three. If it goes significantly over four, that patient is at risk for a bleed. My grandmother who has since passed away, when she had a blood clot in her leg and was prescribed warfarin and the first time she took her first dose of warfarin, her INR went to, and she wasn't in the hospital at this time, her INR went to 12 and so just to say that that one dose, that's how fast the INR can go up and change and cause a bleed and somebody can die from that. So if you do a genetic test or have the genetic information that can help with, that can help keep that from going up so high. So if you look at the package insert on warfarin, based on the genotype, it really says what, how to get to a therapeutic INR, so if you have, remember star one, star one is no variant detected, so if you're star one, star one and you have the GG or non-variant for v-core C1, you really start this dose at five to seven milligrams per day. So in somebody that's a poor metabolizer for 2C9, somewhere down here, you want to start them with a much lower dose because then avoid their getting INR from getting too high. So if we look at 2C9, there's more than 35 alleles described and what I haven't completely mentioned, I touched on a little bit, is that certain alleles are more common in certain ethnic populations. So in for 2C9, the more common alleles that you see in the Caucasian frequency were star two and star three and when originally warfarin dosing and on the package insert, you really only see the star two and star three and doesn't really account for other ethnicities, people of other ethnic backgrounds as well. So in African Americans, you'll see the star four, I mean the star five and the star six and some other ones. So I do know C-pick is updating that and there are guidelines around that for different ethnicities and I do have a slide on that. Coming up, so v-core C1, vitamin K, 2, 3, epoxide reductase subunit one, really is a haplotype. Most people test for this minus 1639. There's some people that may also test for this 1173 C2T, but there's other variants in the haplotype and I believe in warfarin.4, it really only accounts for the 1639 because this is, you only really need to test this when these are additional variants that aren't necessarily needed to be tested. So this is the updated algorithm that C-pick just published earlier this year in 2017, that if you look at warfarin dosing, so is there a genotype available, yes or no, then it really talks about the different ethnicities that if it's non-African ancestry, it's really fine to do the v-core C1 and the 2C9, star 2, star 3, and then do a dose calculation on that. If you look at the African-American ancestry, you really need to test for different variants, specifically star 5, star 6, star 8 and star 11, and then that dosing may change, and then there's another recent variant that's been published, and I'm not sure which gene this is in, that looks at this and you change the dosing because warfarin dosing.org originally didn't have good algorithms around African-American, but I believe they have updated that, but I'm not sure. So there are different platforms to test for warfarin that include 2C9 and v-core C1, but the variants are different among those different platforms. Our fourth example that we're going to talk about today is UGT1A1. So mostly we've talked about phase one metabolism, taking a drug, sort of at the beginning of the pathway. Here is, this is UGT1A1, it's about phase two metabolism, and this is about eliminating the drug from the body. So UGT1A1 is a complex, and I'll show you a slide in a second that I'll show you that slide in a second. It's used for metastatic colon cancer, drug arenotican, and also arenotican is in a drug combination such as fullpox and fullfury, and then a tanzibir for HIV therapy. This also is a gene that's involved in Gilbert syndrome, a genetic disorder of hyperbili rubenemia. It doesn't cause any, that just causes transient hyperbili rubenemia or jaundice. So if we look at, we look at, this is the UGT1A1 gene complex. So there's these common exons, two, three, four, and five down here, but there's different alternative first exons. So depending on which first exon is spliced onto it is which complex it is. So we're going to talk about UGT1A1 right here, which is spliced to the common exons. They're in the promoter of UGT1A1 is a TA repeat or a top top box that turns the gene on, that turns the gene on, and this is the most of the variants that we're going to talk about for. For the drugs that we're going to talk about today. So if we look at UGT1A1 and areno t-can, the TA repeat in the promoter or the top top box can affect how well the metabolism of areno t-can. So it's, areno t-can is an inactive drug that gets activated to SN38. It's inactivated through UGT1A1, and I'll show you that slide in a second, and this is called phase two metabolism. So in the TA repeat, normally you have seven or six TA repeats, but there's some people that have seven TA repeats or eight TA repeats. So that causes areno t-can, so there's a promoter problem with the protein, and it causes, activates areno t-can more slowly, which causes an increased risk for toxicity, including high-grade neutropedia and or diarrhea. So areno t-can inactive goes through this pathway to become SN38. It's active form. And then to the UGT1A1 pathway here and eliminated from the body, but if there's a blockage in this pathway or a backup, you know, a partial blockage in this pathway, this builds up too much, causing neutropenia and the severe diarrhea. This is like death by diarrhea here, bad, bad diarrhea. So if we look at the UGT1A1 frequency, in Caucasians about a third carry this TA repeat, about half of African-Americans and about 10% to 15% of Asians. There is, you can have five TA repeats, and this is called a star 36, or you can have eight TA repeats, which is a star 37, and both of these are a little bit more common in the African-American population. Areno t-can is used for metastatic colon cancer. There's not really very many other drug alternatives by the time a patient gets put on areno t-can. So it is recommended to reduce the dose. So this testing is not used so much anymore because most people give the patients a loading dose because there's not really any alternative, and then just see how well they respond to it, and then slowly increase the dose for tolerance. So in the FDA label, it does say that UGT, there is a UGT1A1 test that's available, and that can detect the, and remember, six TA repeats is our star 1, star 1, star 1, star 28, and our star 28, star 28 genotypes. So here's the decision tree. Remember, in full fox and full fernox or full fury, this is a areno t-can in this sort of drug combination. And it's more about if you're doing high doses, need UGT1A1 genotyping, or actually, this is your low dose, sorry. Low dose, you don't really need it. If you're doing high dose, sorry, I forgot that backwards for a second. If you're doing high dose, it's really recommended to do the UGT1A1 genotyping and then looking at that there's a toxicity around it. But mostly in the U.S., what they do is they give you the low dose and then titrate it for just to tell you sort of don't get to start spanking your sick. There was an FDA-CLEAR platform that had been discontinued since it's not really used so much in the U.S. I'm going to introduce in our sort of end of our time here and talk a little bit about somatic. We spent the most time in our constitutional inherited ones where a lot of pharmacogenetic testing is done, but I wanted to mention briefly about somatic variants. So this is, again, a lot of our companion diagnostics where you have BRAF or ALC or various other ones, and then there are drugs that go with those changes in the tumor. Our last and sort of final example that I'm going to talk about is chronic myelogenous leukemia. It was in the early 60s described as a Philadelphia chromosome, which is a translocation, a 922 translocation. And then in the 80s, it was determined that there was two genes that were fused together and there's a major breakpoint, a minor breakpoint, and a other breakpoint or a very minor breakpoint. But in the late 90s, a matinee mesolate or GLEVAC was discovered that it can be used to treat CML or chronic myelogenous leukemia. So GLEVAC is a small molecule inhibitor of tyrosine kinases or TKI, tyrosine kinase inhibitor. I think that's not really discussed much, is that GLEVAC is also largely an inactive drug that is metabolized through the 3A4 and the 3A5 pathway. Remember, this is one of those medications that you can't take with grapefruit or grapefruit juice because it inhibits the 3A4 pathway. And in caucasians, actually, most all caucasians are poor metabolizers for 3A5. So if you're caucasian and you're a poor metabolizer for 3A5, this 3A4-5 complex, and you take grapefruit juice and you block this, you're really blocking GLEVAC to prevent it from becoming its inactive form and won't be able to work on the chronic myelogenous leukemia. You also take GLEVAC for gastrointestinal stromatumers or GIST at harbor kit mutations. So if we look at GLEVAC or matineb, and it gets in the body, again, it goes through this three, this is its major pathway through 3A4 to become its active form, which then can be binding to the BCR-Able fusion gene or fusion protein to prevent or entreat CML. There are people that, over time, become resistant to GLEVAC or matinemeasalate, and that many of them develop a second mutation in the tyrosine kinase binding domain, and the most common one is called T335151i. And if they develop this, there are additional drugs available that can be used to treat that, especially when you become resistant to GLEVAC. I will remind you that much of this information is available on FarmGKB as well as on the CPIC website. Here's a probably, I don't remember when I last updated this, but a fairly recent list of many of the drugs and the genes that have drug dosing guidelines. I just wanted to mention largely the CPIC guidelines. CPIC has walked a fine line in the sand. They neither promote nor do they not promote genetic testing for this. They actually look at is there evidence for the gene and the drug that if you have the genetic information, you can guide the dosage or recommend alternative drugs based on that. And they do systematic evidence reviews and they update this probably about every two years. They update the guidelines and the literature around each of the genes or the drug gene pairs. So our sort of last slide here in conclusion is that this is a rapidly, pharmacogenetics is a rapidly growing field. It's just not your genetics that affect drug metabolism, but also the environment affects drug metabolism. And genotype, phenotype, correlations, why it can help guide drug information and prescription information are still somewhat imprecise but can serve as guidelines to personalize medicines. And that's my end of my talk today. Thank you. Thank you, Vicki, very much. I appreciate that's a fantastic talk and I have a couple of questions. One is what do you think are the drug gene pairs or let's just say the drugs to start with which are getting the most attention in real practice right now? In other words, what are the ones that insurance companies may be most likely to see come across as a test with requests to pay for? So a few years back, one of the ones that probably would be seeing quite a bit, definitely the companion diagnostics. So those are sort of a given. The other ones people may see is for stents, you would see with clopidogrel and 2C19. That seems to be well covered by insurance and insurance companies. Some of the other ones, which are probably a little bit more rare, may be TPMT with the azothioprene, especially in children. It's one of the drugs that I didn't mention that's used in some childhood cancers. Let me go back to my list and see. The other ones are the HLA ones, such as I'm looking in my list real quick and I know I'm overlooking them. Carbamazepine, a little bit less so. Abacavir is a companion diagnostic and if you have HLAB5701 and you take Abacavir, your increased risk for Steven Johnson syndrome, which is a severe allergic reaction to the drug where your skin sort of comes off. That's one that is being seen. And the other one, interestingly enough, is I have a cath tour for cystic fibrosis. Many, before somebody's put on either cath tour, they already should know what their CF mutation status is. So I'm not sure that one is really combined because it's used to make the diagnosis of CF and then once you know what the mutations are, then you can take either cath tour. Some of the other ones, less so. And so just to get back to the companion diagnostic, so basically those drugs are approved for use in partnership with the test. So in some ways you're not supposed to use the drug without doing the test, is that correct? Yes, that's correct. Okay, according to the FDA, yeah. Yes. So if there's a good indication for that medication and poor indications for alternatives, then it sounds justified to go ahead and do those tests. Yeah, that's correct. So my next question is about what the insurers will see when a test is ordered or when it's billed. Do they see the CYP label? Do they see the list of star alleles that are tested for? What do they end up seeing? Or do they just see it like a Tier 2 CPT code? So most of the main pharmacogenetic genes, the HLA, the 2C9, 2D6, 2C19, there is actually gene specific or the Tier 1 CPT codes, as well as CS has a Tier 1 specific CPT code. Some of the other genes such as DPYD, TPMT, SLOCO 1B1, CYP384, 3A5, those ones are currently in Tier 2, currently in Tier 2 as of 2017. I think it's even harder for the insurance companies to sort of track what they're actually paying for and seeing and what the requests are. Yes. So they may have to do a little extra work to figure that out based on clinical reports and so forth. Right, but normally, and I don't know how each insurance company would do it, it is recommended in the AMACPT coding book that if you're at the Tier 2 gene that you're supposed to provide the gene, the Hugo gene name. And I don't know if all systems are set up to transmit the Hugo gene name or not. Interesting. So you're supposed to include it, but the systems may not have a field to transmit it, so they may not on the other end see it. All right. And then it also doesn't transmit what the star alleles that we're being tested for are. No. Right, okay. I have another sort of final just information question. So there's a high incentive of G6PD deficiency in like Mediterranean populations. Yep. And there are medications that those people shouldn't take or which the doses should be adjusted for. Is that considered a pharmacogenomic? Yes. And if you look, there's Raspere case. Oh, okay. That is associated with G6PD or G6PD deficiency. And interestingly, so there is CPIC guidelines around some G6PD deficiencies. Most G6PD especially because it's X-link in males is very easy to diagnose by a biochemical assay. The problem is in female carriers who have one G6PD deficiency allele and have a normal one. It's hard to that biochemical test doesn't work very well for them. And unless they're sick or they're taking a medication and then it makes them sick or sick or the biochemical test isn't very good and then the DNA test would be the preferred method for women who are carriers for G6PD deficiency. Great. Interesting. Super. Well, thank you, Mickey, very much for coming and giving this talk to round out our Ensure Education webinar series sponsored by NHGRI and the ISCC Ensure Education Working Group. And I look forward to hearing more from you in the future. Great. Thank you. Thank you so much. All right. Bye.