Aaron Besterman, MD
Child & adolescent psychiatrist at the Rady Children’s Hospital inpatient child and adolescent unit, San Diego, CA.
Dr. Besterman has disclosed that he has no relevant financial or other interests in any commercial companies pertaining to this educational activity.
CCPR: Welcome, Dr. Besterman. Please tell us who you are and where you work. Dr. Besterman: I’m an adult and child & adolescent psychiatrist at the Rady Children’s Hospital inpatient child and adolescent unit in San Diego, and I’m a clinical investigator at the Rady Children’s Institute for Genomic Medicine studying psychiatric genetics.
CCPR: Clinicians are inundated with ads for genetic testing. We’d like to help our colleagues understand how these tests work and whether they’re helpful, plus discuss a little of the underlying science. Can you start with the basics? Dr. Besterman: Essentially, there are two types of genetic testing: diagnostic testing, which attempts to identify a specific disorder; and pharmacogenetic testing, which looks at a cluster of genes to try to understand how well an individual might respond to a medication.
CCPR: Let’s talk about diagnostic testing first. Dr. Besterman: Diagnostic testing is only indicated for patients with neurodevelopmental disorders. This includes intellectual disability, a history of developmental delay, or autism spectrum disorders. The current guidelines state that we should be requesting chromosomal microarray (CMA) testing for all of these patients to detect genetic deletions (having one copy of a gene instead of the normal two copies) or duplications (having three copies instead of two). (Editor’s note: The International Society of Psychiatric Genetics [ISPG] guidelines for genetic testing and genetic education are available at www.ispg.net/genetic-testing-statement.) In 10%–20% of patients with neurodevelopmental disorders, CMA can detect a change associated with a specific genetic syndrome. It’s also recommended that both boys and girls get testing to look for Fragile X syndrome.
CCPR: Does insurance cover these tests? Dr. Besterman: Order the CMA through your usual lab. You may need to advocate with the insurance company. If you get the CMA and if there is something of interest, refer to your genetics or dysmorphology clinic and they may order more testing. Currently, the process is a cascade approach where you get authorization from insurance for one test; then, if that comes back negative, you try to get authorization for the next; and so on. It can take a lot of time and often is not approved by insurance companies. But technology is developing rapidly, and now whole genome sequencing can detect all the known genetic changes that confer risk for neurodevelopmental disorders with one test. I’m trying to provide it to kids with neurodevelopmental disorders who are on our inpatient service.
CCPR: How does the genetic testing you suggest affect clinical care? Dr. Besterman: In general, genetic testing doesn’t significantly alter psychiatric care, and that’s important for patients and families to know. But it can be useful in a number of other ways. For example, it can inform providers about the need to monitor for common comorbidities with some genetic conditions. It can also provide families with an answer for why their child has a certain condition, which can provide relief to those who’ve been on a diagnostic odyssey, going from specialist to specialist looking for an answer. It can also help these families connect with other families who have a child with a similar genetic condition for social and emotional support. Genetic testing also helps with reproductive counseling, if a family wants to know whether their next child could have the same condition. Lastly, genetic testing can plug families into clinical research opportunities where kids could be enrolled in research studies specifically for their genetic condition.
CCPR: What do you do when you discover a known genetic disorder such as 22q13 deletion syndrome (Phelan-McDermid syndrome), which is becoming better recognized as a cause of autism-like syndromes? Dr. Besterman: You need to provide proper counseling. If you feel comfortable doing all of it on your own, that’s fantastic. If not, there are alternatives. You can work alongside genetic specialists to make sure patients get what they need. If you order a test and it comes back positive, you could refer the patient to a medical geneticist for further counseling. Even if it comes back negative, you could refer to a geneticist if you feel the patient might benefit from more sophisticated testing. Medical geneticists are MDs. Non-MD genetic counselors are also a great resource.
CCPR: What are genetic counselors? Dr. Besterman: Genetic counselors are medical professionals who are experts in genetic medicine and often work alongside medical geneticists. They can facilitate genetic testing, consenting for genetic testing, and counseling to accompany test results. If you live in a rural area and there aren’t many genetic counselors nearby, you can find one through the National Society of Genetic Counselors (www.nsgc.org).
CCPR: There’s another topic in psychiatric genetics called polygenic risk. Can you talk about that? Dr. Besterman: Polygenic risk refers to the idea that complex medical disorders, including disorders like depression, schizophrenia, anxiety, or basically all common psychiatric diseases, are typically not caused by a single gene change. Instead, they’re caused by the cumulative effect of dozens, hundreds, or thousands of tiny genetic changes: ones that we inherit from our parents and ones that arise anew during development as part of the impact of the environment and experience on genetic expression and activity (epigenetics). Researchers compare large groups of individuals with a given disease to groups of individuals without that disease and look for genetic differences between the groups. Instead of looking at each genetic difference individually, you look at a cumulative polygenic risk score.
CCPR: How do we use that risk score clinically? Dr. Besterman: With our current research, polygenic risk scores are not good enough to predict with clinical certainty whether someone will have one of these psychiatric disorders. Still, the goal is not to use these risk scores in isolation, but to combine them with other risk factors, such as environmental risk factors, to improve our prediction of onset of disease.
CCPR: OK, we’ve spent some time unpacking diagnostic testing. What about pharmacogenetic testing? Dr. Besterman: Pharmacogenetic testing looks at genes involved in drug metabolism. There’s an international organization, the Clinical Pharmacogenetics Implementation Consortium (CPIC), that puts out guidelines for pharmacogenetic testing (www.cpicpgx.org). Right now, the recommendations in psychiatry include three genes. The first is CYP2D6, which metabolizes many antidepressants, antipsychotics, and atomoxetine; the second is CYP2C19, which metabolizes citalopram, escitalopram, and sertraline. Testing for these genes can predict how fast or slow a person will metabolize these medications. It doesn’t tell us how well they will respond to them. The third gene you might check is HLA-B; a particular variant of this gene is often present in people of Asian descent and predisposes a person to carbamazepine hypersensitivity. If a person has this gene variant, their risk of developing Stevens-Johnson syndrome is greatly elevated. So, this is definitely indicated for Asian patients who are being started on carbamazepine.
CCPR: The advertised tests cover a broader set of genes. Is there a scientific basis for their approach? Dr. Besterman: There are a lot of direct-to-consumer commercial companies, especially in pharmacogenetics, where the patient and the doctor receive this report that puts medications in green, yellow, and red boxes. But how the companies come to those conclusions is not well understood. If you compare the recommendations between companies, they’re often very inconsistent. This information can be misleading and potentially even cause harm.
CCPR: I saw two patients recently whose families insisted on getting a genetic test to see what medication their child should have. In both cases I told them I didn’t think the test would tell us anything, and it would be expensive and might even suggest things that aren’t helpful. One family hasn’t done it yet, maybe because of the cost. I had already started the other patient, a teen, on fluoxetine because that’s what the clinical data supported. The kid had a great response to the fluoxetine, but the test report said it was in the red box and therefore recommended against using it. Dr. Besterman: Yes, that’s one of the risks: not using medications that could potentially be helpful, or stopping medications that are already helpful, because the families don’t understand what it means for the medication to be in a red box.
CCPR: Patients say, “My relative got a test that showed them why their medication didn’t work. Now they’re on a different medicine and they’re doing great. I want the test, too.” What should we say to them? Dr. Besterman: I say, “Every person is unique, and we need to assess you individually.” Try to understand their goals, then decide whether pharmacogenetic testing is helpful for that patient. Emphasize that testing can only really help us decide whether an individual should be on higher or lower doses than average of certain medications—it doesn’t tell us anything about whether a patient is going to respond overall to a medication. Sometimes they’re still interested in getting tested, and I go along with that. It can be helpful for patients when they’re particularly anxious about medication side effects and want some objective data. But it’s really important to not make decisions just on the pharmacogenetic data; instead, use the data together with other information to make a decision based on your best clinical judgment.
CCPR: How about MTHFR and COMT? There’s a lot of marketing, especially for L-methylfolate and primary care doctors getting a test from 23 and Me to look at COMT. Dr. Besterman: A decade ago, during the candidate gene era of psychiatric genetics, we didn’t have much information about the genetic architecture of psychiatric disease, and researchers would guess which genes might be related to psychiatric illness based on neurobiology. The evidence supporting their role in either risk for genetic disorders or drug response in psychiatric treatment is still limited, and ordering genetic tests for MTHFR and COMT is not particularly helpful for our psychiatric patients. That’s not to say that those genes have absolutely no role in the neurobiology of psychiatric illness, but there is insufficient evidence to say testing for changes in those genes will alter our care in any way.
CCPR: You mentioned that the test results from different companies might be inconsistent. Dr. Besterman: There was a study a few years ago by Bousman and colleagues. He took patient samples and sent them to different direct-to-consumer laboratories and got their medication recommendations. There was a lot of inconsistency between the companies, not only in the specific genotypes, but also in the metabolic phenotypes—meaning whether they thought a given patient was a poor, intermediate, or rapid metabolizer. And the medications that they were putting into green, yellow, and red boxes were also very different (Bousman CA et al, Pharmacogenet Genomics 2017;27(11):387–393). There needs to be standardization in the field. In fact, the FDA started sending cease-and-desist letters to some of these companies last year. The field is undergoing some evolution, and that’s probably a healthy thing.
CCPR: Do you have any quick guidelines that we might use? Dr. Besterman: Absolutely. Visit the ISPG website at www.ispg.net. They have helpful educational resources.
CCPR: Is it possible to pack all of this into a bottom-line message for our colleagues? Dr. Besterman: Learn more about genetics, including the hopes but also the current limitations. It is a growing area of medicine and is going to permeate all fields, including psychiatry, more and more with time. There are many resources out there. I’ve co-written some reviews regarding what psychiatrists need to know about genetics that are freely available on PubMed (www.ncbi.nlm.nih.gov/pmc/articles/PMC5371713). In addition to the ISPG, there’s also the National Neuroscience Curriculum Initiative or NNCI (www.nncionline.org), which is an incredible education resource with many online modules that take you step by step through easily digestible scenarios, including genetics.