The APA says that genetic testing isn’t ready for prime time, but is there anything worth salvaging in these panels? We get some answers from John Nurnberger, MD, PhD, co-founder of the International Society of Psychiatric Genetics, the founding editor of the international journal Psychiatric Genetics, and a Distinguished Professor Emeritus at the Indiana University School of Medicine.
Published On: 8/2/2021
Duration: 11 minutes, 22 seconds
Related Article: "Genetic Testing: What You Need to Know in 2021," The Carlat Psychiatry Report, June/July 2021
Your patient shows up for their first appointment with a full panel of genetic testing in hand. They’d like you to prescribe what the test recommends, but the APA has this position paper out that says genetic testing isn’t ready for prime time. Today, we interview John Nurnberger to find some middle ground.
Welcome to The Carlat Psychiatry Podcast, keeping psychiatry honest since 2003. I’m Chris Aiken, the editor in chief of The Carlat Psychiatry Report. And I’m Kellie Newsome, a psychiatric NP and a dedicated reader of every issue.
Kellie Newsome: The human genome was mapped in 2003, and with it came the hope that we could one day tailor treatment to a patient’s genetic fingerprint. The wait was not long. In 2005, the first pharmacogenetic tests were released for the metabolic enzymes CYP2D6 and CYP2C19. Those tests were approved by the FDA, but most of the commercial genetic panels that have cropped up since then are not approved.
Dr. Aiken: That difference is important to understand. When we talk about genetic testing, we often talk in black-and-white terms like “is genetic testing useful or not?” That’s a bit like asking “Is laboratory testing in psychiatry useful or not?” Clearly, it depends on the patient, and the test involved. So while individual tests like CYP 2D6 and 2C19 are FDA approved, most commercial genetic panels – the ones that drop promotional literature at your office – are not approved. One of the most prominent of these is Genesight, which was released in 2009.
Kellie Newsome: And there’s a paradox with genetic testing. While individual tests might be sound, when you pile a bunch of those together in a pharmacogenetic panel… not so much. In the Carlat Report we’ve covered half a dozen studies that tested these genetic panels out in real world populations – usually in patients with depression who failed to respond to an antidepressant. Now, you would think that was an enriched design – that if you’re going to see a strong signal with pharmacogenetic testing it’s in people who didn’t respond well to antidepressants that these tests ought to make a difference. But time and again they fall short – we covered one of those failures this summer in a large trial of the Genecept test.
Dr. Aiken: Yes it’s confusing – if some of the individual tests are valid – why does the genetic panel fail to work in practice? Well, they don’t exactly fail to work. When you stack all those failures up, you do see a small signal pointing toward successful outcomes with these genetic tests. This month with interviewed John Nurnberger about which of the dozen or so tests in these pharmacogenetic panels are useful, and which ones to take with a grain of salt.
Dr. Nurnberger: I think that the evidence is growing that there’s a place for this in clinical practice. The meta-analyses of pharmacogenetics tests are actually looking pretty favorable recently – not the ones on the S and L allele of the serotonin transporter, which have been more or less negative since the meta-analyses have started coming out as you are aware - but the ones on the CYP enzymes, particularly 2D6 and 2C19 are looking pretty good at this point. And I participated in one of them a few years ago and I was impressed with the care that they took in designing the protocol and I think it was you know a good study, and there have been a number of others coming out.
Kellie Newsome: Dr. Nurnberger recommended CYP 2D6 and 2C19 – and those are the same tests that were the first to gain FDA approval in 2006. Those are the top two – desert island tests – and in our online interview he delves into other genetic tests that deserve a place at the table. But whether the test is useful or not depends a lot on the patient. Here’s where Dr. Nurnberger thinks genetic testing has the most utility:
Dr. Nurnberger: I think it is worthwhile considering genetic testing for individuals that have failed a couple of antidepressant trials or in individuals where there are other reasons to think that they may be having very unusual responses to medication. And sometimes it’s very useful for the patient to get that kind of information and feel as though there’s a reason for them to understand their experiences with medications.
I don’t follow the directions rigorously, but I think the evidence suggests that if you have a test that gives you like a green light/yellow light and a red light for the medications it’s probably wise to avoid those that are in the red light category and you can use the others with some adjustment of dose and just being careful with the dosing.
So I do it personally in some cases, but I will say it’s not at all routine in my practice, and I don’t think it should be in anybody’s practice.
Kellie Newsome: In the interview, Dr. Nurnberger discusses the relative merits of ordering individual genetic tests – sticking only to the most high yield ones – vs. ordering a whole genetic panel. These days, many patients are coming to their first appointment with the results of one of those genetic panels already in hand. So let’s unpack what is inside these tests.
These tests fall into two broad categories:
- Pharmacokinetic tests. These tell us something about the blood levels of a medication – whether they are likely to go too high or too low. Around 20%-30% of people have genetic variations in their metabolic enzymes and these can lead to abnormally high (poor metabolizers) or low (rapid metabolizers) levels of psychiatric medications. Most pharmacokinetic tests focus on the genes that code for these enzymes.
Dr. Aiken: But there’s a whole nother level of pharmacokinetics at the blood brain barrier. Just as the liver determines the serum level, there’s a pump at the blood brain barrier called the p-glycoprotein transporter that determines the level the medication will reach in the brain. P-glucoprotein regulates the transfer of medications across the blood brain barrier, and some medications are more susceptible to variations at this transporter than others. Only a few commercial tests look at these gene, but it is one of the more practical ones in terms of the clinical research that supports it.
Kellie Newsome: After pharmacokinetic genes are the pharmacodynamic ones. These are the genes that supposedly shape the brain’s response to medications, such as genes for neurotransmitter receptors. We covered one of those – the serotonin transporter gene – last month in an interview with Vladimir Maletic. It’s these pharmacodynamic genes that hold the most promise – I mean after all what could be more useful than predicting how the brain will respond to an antidepressant – but it’s also these that are the most disappointing in the clinical research.
Dr. Aiken: So far the most useful genetic tests have been the ones that inform us about the potential blood levels – the pharmacokinetic genes. Which got me wondering – why not just check the serum level directly if you’re worried about it? You can order a serum blood level for most medications – not just lithium and Depakote – and the cost is pretty low, around $20-60.
Dr. Nurnberger: Nothing wrong with that. Yeah, the only disadvantage of that is you actually have to have the patient on the drug at the time and know what they’re taking. Sometimes, the point at which you’re using the pharmacogenetic test is to make choices going forward. A lot of time that’s the kind of situation you’ll be encountering, so the drug level is not very useful. But if you’re on the drug and you’re wondering well can I raise the dose? Sure, you can look at the level.
Kellie Newsome: So the bottom line is this. The APA and the FDA may not endorse pharmacogenetic panels, and the Carlat Report has certainly taken a skeptical eye toward them, but that doesn’t mean you should rip up the results and throw them in the trash when a patient shows up with pharmacogenetic results in hand. The test may not pinpoint the best medication with a laser-like focus, but they are not meaningless. You just need to know how to interpret them, and which ones are most likely to make a difference.
John Nurnberger is a Distinguished Professor Emeritus at the Indiana University School of Medicine. He was a co-founder of the International Society of Psychiatric Genetics and the founding editor of the International Journal Psychiatric Genetics. From 1989 to 2006 he led the NIMH Genetics Initiative Bipolar Group. He remains active in multicenter investigations into the genetic of bipolar disorder, alcohol dependence, and autism.
Dr. Aiken: And now for the word of the day….Serotonin 5-HT3
“Once upon a time, a princess came upon a tiny cottage. Seeing that no one was at home, she ate some of their tiny meals, drinks some of their wine, and then tested all their tiny beds.” So begins the story of Snow White and the Seven Dwarfs, but we’re not here to tell you that story. This is the story of Serotonin and the Seven Receptors. Today’s receptor is #3 – but we know him as the Chunder.
Dr. Aiken: Kellie what are you talking about?
Kellie Newsome: Chunder. You know. It’s the Australian word for vomiting.
Dr. Aiken: Is that where beer does flow and men chunder?
Kellie Newsome: Yes – but only in a land down under. Serotonin 5HT-3 regulates nausea and vomiting, or emesis if you want to get all technical about it. And that’s why SSRIs make patients nauseas when they first start them, and it’s also why 5HT-3 antagonists like ondansetron are used to treat nausea. Join us next Monday where we’ll talk about how to manage nausea on psychiatric medications.