Genetic Testing for Mental Health Conditions

MEDICAL POLICY POLICY RELATED POLICIES POLICY GUIDELINES DESCRIPTION SCOPE BENEFIT APPLICATION RATIONALE REFERENCES CODING APPENDI HISTORY Genetic Testing for Mental Health Conditions Number 12.04.515 Effective Date May 1, 2016 Revision Date(s) 04/12/16; 07/14/15; 04/24/15; 01/20/15; 12/17/14 Replaces 2.04.110 Policy [TOP] Genetic testing for mutations associated with mental health disorders, including all of the individual component tests listed in Table 1, is considered investigational in all situations, including but not limited to the following: To confirm a diagnosis of a mental health disorder in an affected individual. To predict future risk of a mental health disorder in an asymptomatic individual. In an affected individual to inform the selection or dose of medications used to treat mental health disorders. Genetic testing panels, including but not limited to the following tests, are considered investigational for selecting medications or doses of medications for the treatment of psychiatric or mental health symptoms or disorders: Ally Diagnostics Genetic Testing Panel Alpha Genomics Psychiatry/ADHD Panel Frontier PGx Pharmacogenomic Testing Genecept Assay GeneSight Psychotropic Panel Genetic Technological Innovations Pharmacogenetic Testing Luminex xtag CYP2C19 assay Luminex xtag CYP2D6 assay Mental Health Insight DNA Millennium Pharmacogenetic Testing Molecular Testing Labs Psychotropic Medication Panel PersonaGene PGL Multi-Drug Panel PharmaRisk Basic PharmaRisk Psychiatric Panel Physicians Choice Laboratory Services (PCLS) Pharmacogenetic Testing Primex Expanded Pharmacogenomics Panel Progenity Informed PGx Pharmacogenetic Testing Proove Drug Metabolism Panel Proove Opioid Risk assay STA2R SureGene YouScript Panel

Related Policies 12.04.131 Pharmacogenetic Testing for Pain Management [TOP] Policy Guidelines [TOP] Coding There are no specific CPT codes for genetic testing panels for mental health disorders. There are specific codes for some of the component tests. The remaining tests on the panel that are not currently codified in CPT would be reported with one unit of the unlisted molecular pathology code 81479. CPT 81225 CYP2C19 (cytochrome P450, family 2, subfamily C, polypeptide 19)(e.g., drug metabolism), gene analysis, common variants (e.g., *2, *3, *4, *8, *17) 81226 CYP2D6 (cytochrome P450, family 2, subfamily D, polypeptide 6)(e.g., drug metabolism), gene analysis, common variants (e.g., *2, *3, *4, *5, *6, *9, *10, *17, *19, *29, *35, *41, *1N, *2N, *4N) 81291 MTHFR (5, 10-methylenetetrahydrofolate reductase) (e.g., hereditary hypercoagulability) gene analysis; common variants (e.g., 677T, 1298C) 81401 CYP3A4 (cytochrome P450, family 3, subfamily A, polypeptide 4) (e.g., drug metabolism), common variants (e.g., *2, *3, *4, *5, *6) Description Several commercially available testing panels include genes related to neurotransmitter function and pharmacokinetics of psychiatric drugs. They are intended to be an aid in clinical decision making regarding interventions for psychiatric conditions. [TOP] Background Psychiatric disorders cover a wide range of clinical phenotypes and are generally classified by symptomatology in systems such as the classification outlined in the American Psychiatric Association s Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5). Knowledge of the physiologic and genetic underpinnings of psychiatric disorders is advancing rapidly and may substantially alter the way in which these disorders are classified and treated. Genetic testing could potentially be used in several ways including stratifying patients risks of developing a particular disorder, aiding diagnosis, targeting medication therapy, and optimally dosing medication. Better understanding of these factors may lead to an improved ability to target medications to the specific underlying abnormalities, with potential improvement in the efficiency and efficacy of treatment. Genes Relevant to Mental Health Disorders Mental disorders encompass a wide range of conditions: the DSM-5 includes more than 300 different disorders. However, currently available genetic testing for mental health disorders is primarily related to several clinical situations: 1. Risk stratifying patients for one of several mental health conditions, including schizophrenia and related psychotic disorders, bipolar and related disorders, depressive disorders, obsessive-compulsive and related disorders, and substance-related and addictive disorders.

2. Predicting patients response to, dose requirement for, or adverse effects from one of several medications (or classes of medications) used to treat mental health conditions, including: typical and atypical antipsychotic agents, serotonin and serotonin/norepinephrine reuptake inhibitors (SSRIs), and medications used to treat addiction, such as disulfiram. Panels of genetic tests have been developed and have been proposed for use in the management of mental health disorders. Genes that have been implicated in mental health disorders or their treatments and that are included in currently available panels include the following: Serotonin Transporter (SLC6A4). This gene is responsible for coding the protein that clears serotonin metabolites (5-HT) from the synaptic spaces in the central nervous system (CNS). This protein is the principal target for many of the SSRIs. By inhibiting the activity of the SLC6A4 protein, the concentration of 5-HT in the synaptic spaces is increased. A common polymorphism in this gene consists of insertion or deletion of 44 base pairs in the serotonin-transporter-linked polymorphic region (5-HTTLPR), leading to the terminology of the long (L) and short (S) variants of this gene. These polymorphisms have been studied in relation to a variety of psychiatric and nonpsychiatric conditions, including anxiety, obsessive compulsive disorder, and response to SSRIs. Serotonin Receptor (5HT2C). This gene codes for 1 of at least 6 subtypes of the serotonin receptor that is involved in the release of dopamine and norepinephrine. These receptors play a role in controlling mood, motor function, appetite, and endocrine secretion. Alterations in functional status have been associated with affective disorders such as anxiety and depression. Certain antidepressants, e.g., mirtazapine and nefazodone, are direct antagonists of this receptor. There is also interest in developing agonists of the 5HT2C receptor as treatment for obesity and schizophrenia, but no such medications are commercially available at present. Serotonin Receptor (5HT2A). The 5HT2A gene codes for another subtype of the serotonin receptor. Variations in the 5HT2A gene have been associated with susceptibility to schizophrenia and obsessive-compulsive disorder and response to certain antidepressants. Sulfotransferase Family 4A, Member 1 (SULT4A1). SULT4A1 encodes a protein that is involved in the metabolism of monoamines, particularly dopamine and norepinephrine. Dopamine Receptors (DRD1, DRD2, DRD4). The DRD2 gene codes for a subtype of the dopamine receptor, called the D2 subtype. The activity of this receptor is modulated by G-proteins, which inhibit adenyl cyclase. These receptors are involved in a variety of physiologic functions related to motor and endocrine processes. The D2 receptor is the target of certain antipsychotic drugs. Mutations in this gene have been associated with schizophrenia and myoclonic dystonia. Polymorphisms of the DRD2 gene have been associated with addictive behaviors, such as smoking and alcoholism. The DRD1 gene encodes another G-protein coupled receptor that interacts with dopamine to mediate some behavioral responses and modulate D2 receptor-mediated events. Polymorphisms of the DRD1 gene have been associated with nicotine dependence and schizophrenia. The DRD4 gene encodes a dopamine receptor with a similar structure; DRD4 polymorphisms have been associated with risk-taking behavior and attention deficit hyperactivity disorder. Dopamine Transporter (DAT1 or SLC6A3). Similar to the SCL6A4 gene, this gene product encodes a transporter that mediates the active reuptake of dopamine from the synaptic spaces in the CNS. Polymorphisms in this gene are associated with Parkinson disease, Tourette syndrome, and addictive behaviors. Dopamine Beta-Hydroxylase (DBH). The dopamine beta-hydroxylase protein encoded by this gene catalyzes the hydroxylase of dopamine to norepinephrine. It is primarily located in the adrenal medulla and in postganglionic sympathetic neurons. Variation in the DBH gene has been investigated as a modulator of psychotic symptoms in psychiatric disorders and in tobacco addiction. Gated Calcium Channel (CACNA1C). This gene is responsible for coding of a protein that controls activation of voltage-sensitive calcium channels. Receptors for this protein are found widely throughout the body, including skeletal muscle, cardiac muscle, and in neurons in the CNS. In the brain, different modes of calcium entry into neurons determine which signaling pathways are activated, thus modulating excitatory cellular mechanisms. Associations of polymorphisms of this gene have been most frequently studied in relation to cardiac disorders. Specific polymorphisms have been associated with Brugada syndrome and a subtype of long QT syndrome

(Timothy syndrome). Ankyrin 3 (ANK3). Ankyrins are proteins that are components of the cell membrane and interconnect with the spectrin-based cell membrane skeleton. The ANK3 gene codes for the protein Ankyrin G, which has a role in regulating sodium channels in neurons. Alterations of this gene have been associated with cardiac arrhythmias such as Brugada syndrome. Polymorphisms of this gene have also been associated with bipolar disorder, cyclothymic depression, and schizophrenia. Catechol-O-Methyltransferase (COMT). This gene codes for the COMT enzyme that is responsible for the metabolism of the catecholamine neurotransmitters, dopamine, epinephrine and norepinephrine. COMT inhibitors, such as entacapone are currently used in the treatment of Parkinson disease. A polymorphism of the COMT gene, the Val158Met polymorphism, has been associated with alterations in emotional processing and executive function and has also been implicated in increasing susceptibility to schizophrenia. Methylenetetrahydrofolate reductase (MTHFR). This is a widely studied gene that codes for the protein that converts folic acid to methylfolate. Methylfolate is a precursor for the synthesis of norepinephrine, dopamine, and serotonin. It is a key step in the metabolism of homocysteine to methionine, and deficiency of MTHFR can cause hyperhomocysteinemia and homocysteinuria. The MTHFR protein also plays a major role in epigenetics, through methylation of somatic genes. A number of polymorphisms have been identified that result in altered activity of the MTHFR enzyme. These polymorphisms have been associated with a wide variety of clinical disorders, including vascular disease, neural tube defects,dementia, colon cancer, and leukemia. Gamma-Aminobutyric acid (GABA) A receptor. This gene encodes a ligand-gated chloride channel composed of 5 subunits that responds to GABA, a major inhibitory neurotransmitter. Mutations in the GABA receptor have been associated with several epilepsy syndromes. Mu and Kappa Opioid Receptors (OPRM1 and OPRK1). OPRM1 encodes the mu opioid receptor, which is a G-protein coupled receptor that is the primary site of action for commonly used opioids, including morphine, heroin, fentanyl, and methadone. Polymorphisms in the OPRM1 gene have been associated with differences in dose requirements for opioids. OPRK1 encodes the kappa opioid receptor, which binds the natural ligand dynorphin and a number of synthetic ligands. Cytochrome P450 genes (CYP2D6, CYP2C19, CYP3A4, CYP1A2). These 4 genes code for hepatic enzymes that are members of the cytochrome p450 family and are responsible for the metabolism of a wide variety of medications, including many psychotropic agents. For each of these genes, polymorphisms exist that impact the rate of activity, and therefore the rapidity of elimination of drugs and their metabolites. Based on the presence or absence of polymorphisms, patients can be classified as rapid metabolizers (RM), intermediate metabolizers (IM), and poor metabolizers (PM). P-Glycoprotein Gene. The ABCB1 gene, also known as the MDR1 gene, encodes P-glycoprotein which is involved in the transport of most antidepressants across the blood-brain barrier. ABCB1 polymorphisms have been associated with differential response to antidepressants that are substrates of P-glycoprotein, but not to antidepressants that are not P-glycoprotein substrates. UDP-Glucuronosyltransferase Gene. The UDP-glucuronosyltransferase gene, UGT1A4, encodes an enzyme of the glucuronidation pathway that transforms small lipophilic molecules into water-soluble molecules. Polymorphisms in UGT1A4 have been associated with variation in drug metabolism, including some drugs used for mental health disorders. Commercially Available Genetic Tests Several test labs market either panels of tests or individual tests designed relevant for mental health disorders. Some examples of the specific tests included in panels are summarized in Table 1. The Genecept Assay (Genomind, LLC, Chalfont, PA) is a genetic panel test that includes a range of genetic mutations and/or polymorphisms that have been associated with psychiatric disorders and/or response to psychotropic medication. The test consists of a group of individual genes, and the results are reported separately for each gene. There is no summary score or aggregate results derived from this test. The intent of the test is as a decision aid for treatment interventions, particularly in the choice and dosing of medications. However, guidance

on specific actions that should be taken following specific results of the test is vague. Interpretation of the results and any management changes as a result of the test are left to the judgment of the treating clinician. The STA2R (SureGene Test for Antipsychotic and Antidepressant Response, SureGene, LLC, Louisville, KY) is another genetic panel that provides information about medication response, adverse event likelihood, and drug metabolism. According to the manufacturer s website, the test is recommended for initial medication selection, for patients who have poor efficacy, tolerability, or satisfaction with existing medications, and in cases of severe treatment failure. (1) GeneSight Psychotropic (Assurex Health, Mason, OH) is a genetic panel that provides information about genes that may affect a patient s response to antidepressant and antipsychotic pharmacotherapy. According to the manufacturer s website, following testing, the treating provider receives a report with the most common medications for the patient s diagnosed condition categorized by cautionary level, along with a report of the patient s genetic variants. (2) Details are not provided about the algorithm used by the manufacturer to generate risk levels. The Proove Opioid Risk Panel (Proove Biosciences, Irvine, CA) is a panel to evaluate genes involved in the development of substance abuse or dependence and in response to medical therapy for substance abuse or dependence. Pathway Genomics (San Diego, CA) offers the Mental Health DNA Insight panel, which is a single nucleotide polymorphism-based array test which evaluates a number of genes associated with the metabolism and efficacy of psychiatric medications. AltheaDx (San Diego, CA) offers a number of IDgenetix-branded tests, which include several panels focusing on polymorphisms that affect medication pharmacokinetics for a variety of disorders, including psychiatric disorders. Specific mutations included in the panel were not easily identified from the manufacturer s website. The Ally Diagnostics Genetic Testing Panel (Ally Clinical Diagnostics, Farmers Branch, Texas) is a panel to evaluate genes that may affect a patient s response to medications for the treatment of psychiatric or mental health symptoms or disorders. The panel includes three CYP450 tests (81225, 81226, 81227), vitamin K epoxide reductase (81355), and a non-specific molecular pathology procedure (81401). Molecular Testing Labs Psychotropic Medication Panel (Molecular Testing Labs, Vancouver, WA) offers a genetic testing panel which their website describes as identifying five different categories of patients by the way they metabolize specific drugs. The only specific gene mentioned is CYP2D6. The PharmaRisk Basic Panel Is described by OptimumMeds as a test that analyzes the genes that metabolize many commonly prescribed medications used in all clinical practices including: internal medicine, cardiology, geriatrics, psychiatry and chronic pain management. The test analyzes 55 genetic markers across 4 genes CYP2C19, CYP2C9, CYP2D6 and VKORC1. The PharmaRisk Psychiatric Panel Includes CYP2D6, OPRM1, CYP2C9, COMT, DRD2, CYP2B6, CYP2C19, CYP1A2, UGT2B15. The PGL Multi-Drug Panel Includes CYP2D6, CYP2C9, CYP2C190, CYP1A2, CYP3A4, CYP3A5, SLC6A4, OPRM1, VKORC1, SLCO1B1, SULT24A1, Factor II, Factor V, MTHFR and COM. The Alpha Genomix Psychiatry/ADHD Panel Includes CYP1A2, CYP2C9, CYP2C19, CYP2D6, CYP3A, ADRA2A, and COMT. Genetic Technological Innovations (DBA Vantari Genetics) offers genetic testing for drug metabolism, preconception and pregnancy, inherited conditions and inherited cancer. Their pharmacogenetic panel for drug metabolism includes CYP2C19, CYP2D6, MTHFR and CYP3A4. The PersonaGene panel Uses next generation sequencing to test for metabolism of common drugs for pain management, cardiology, psychiatry and urology. Although this large panel encompassing four specialties, all of the mutations tested are within the CYP450 family. Luminex Offers a genotyping assay which can aid clinicians in determining therapeutic strategy for drugs

metabolized by cytochrome P450. There are three Progenity Informed PGx genetic testing panels ADHD (4 mutations), Depression (7 mutations) and Psychotropic (7 mutations) - and each panel tests a variety of CYP450 genes, MTHFR, etc. In addition to the available panel tests, several labs offer genetic testing for individual genes, including MTFHR, CYP450 genes, and SULT4A1. Table 1: Genetic Panels for Mental Health Disorders Polymorphisms Included Gene Polymorphisms Included in Commercially Available Test Panels Genecept STA2R Assay (SureGene, (Genomind) LLC) GeneSightRx Psychotropic (AssureRx Health) Proove Opioid Risk (Proove Biosciences) SULT4A1 SLC6A4 (serotonin transporter) 5HT2C (serotonin receptor) 5HT2A (serotonin receptor) DRD1 (dopamine receptor) DRD2 (dopamine receptor) DRD4 (dopamine receptor) DAT1 (dopamine transporter) DA beta-hydroxylase CACNA1C (gated calcium channel ) Ankyrin 3 COMT (catechol-omethyltransferase) MTHFR GABA OPRK1 (kappa opioid receptor) OPRM1 (mu opioid receptor) CYP genes CYP2D6 CYP2C19 CYP3A4 CYP1A2 CYP2C9 CYP2B6 UGT1A4 ABCB1 Mental Health DNA Insight Regulatory Status The Genecept Assay, STA2R test, the GeneSight Psychotropic panel and the GeneSight MTFHR tests are

laboratory-developed tests that are not subject to U.S. Food and Drug Administration (FDA) approval. Clinical laboratories may develop and validate tests in-house ( home-brew ) and market them as a laboratory service; such tests must meet the general regulatory standards of the Clinical Laboratory Improvement Act (CLIA). Other examples of these tests are YouScript (Genelex), Proove Drug Metabolism (PROOVEBio), Mental Health Insight DNA (Pathway Genomics), Millennium Pharmacogenetic Testing (Millennium Health), Primex Expanded Pharmacogenomics Panel (PrimexLab) and DNA Test Assay Pain Management Panel (Ally Clinical Diagnostics). Tests include three CYP450 tests (81225, 81226, 81227), vitamin K epoxide reductase (81355), and a nonspecific molecular pathology procedure (81401). Scope [TOP] Medical policies are systematically developed guidelines that serve as a resource for Company staff when determining coverage for specific medical procedures, drugs or devices. Coverage for medical services is subject to the limits and conditions of the member benefit plan. Members and their providers should consult the member benefit booklet or contact a customer service representative to determine whether there are any benefit limitations applicable to this service or supply. This medical policy does not apply to Medicare Advantage. Benefit Application N/A [TOP Rationale This policy was created in November 2013 and has been updated periodically with literature reviews, most recently through April 21, 2015. [TOP] Analytic Validity Genotyping of genes involved in mental health disorders can be done by single-nucleotide polymorphism microarrays, standard Sanger sequencing, or next-generation sequencing methods. Information on analytic validity of commercially-available test panels is lacking. No published studies were identified that specifically evaluated the analytic validity of the test as performed commercially. As a result, it is not possible to determine the analytic validity of the testing process. Clinical Validity Evidence on the clinical validity of genetic testing for mental health disorders consists primarily of genome-wide association (GWS) studies that correlate specific genetic polymorphisms with clinical factors and case-control studies that examine the odds ratio for genetic variants in individuals with a clinical disorder compared with individuals without the disorder. There were no studies of clinical validity identified that evaluated defined groups of patients (e.g., patients with schizophrenia; patients with depression and non-response to serotonin reuptake inhibitors (SSRIs)) and reported the sensitivity and specificity of the panel results for those patients. Therefore it is not possible to estimate the clinical sensitivity and specificity of the test as a diagnostic test for specific patient groups. A comprehensive review of the GWAS and case control studies for all of these genes is beyond the scope of this policy. A 2015 review of meta-analyses examining the association between specific genes and specific mental health disorders reported that 134 genes (206 variants) have been identified as significantly associated risk factors for major depressive disorder, anxiety disorders, attention deficit hyperactivity disorder, schizophrenia, or

bipolar disorder, with 13 genetic variants shared in common between 2 or more disorders.(3) However, some of the representative literature in this area is discussed next. Genes Associated with Increased Disease Risk Serotonin Transporter (SLC6A4) Gene and Risk of Multiple Mental Health Disorders. The SLC6A4 gene that codes for the serotonin transport protein has been studied in relation to a number of psychiatric conditions. Published literature has reported associations between variants in this gene and anxiety, bipolar disorder, and obsessive-compulsive disorder, and drug and alcohol dependence. (4-6) However, these associations have not been reported consistently across studies. In a meta-analysis of 26 studies, Sen et al. reported that the overall association of SLC6A4 variants with anxiety approached, but did not quite reach, statistical significance (p=0.09). (7) In a 2011 study and systematic review/meta-analysis, Minelli et al. also evaluated the association between polymorphisms in the 5-HTTLPR gene and the nearby rs25531 locus and anxiety-related personality traits. (8) In the first part of their study, 287 healthy volunteers underwent 5-HTTLPR genotyping and personality trait assessment. There was no significant association between 5-HTTLPR genotypes and anxiety-related scale score overall, but there was a significant association when the long allele was considered dominant (p=0.02). In the Minelli et al meta-analysis, the authors included studies that evaluated the association between 5-HTTLPR polymorphisms and anxiety-related personality traits. While 50 articles met their inclusion criteria, the meta-analysis used data from 35 articles, after exclusions for insufficient data, significant deviation from Hardy-Weinberg equilibrium, and excessive ethnic heterogeneity. The authors found a significant association between the homozygosity for the 5-HTTLPR short allele and higher scores for anxiety-related traits, but this association was not present when only studies using structured psychiatric screening were included. In 2009 systematic review and meta-analysis, Risch et al. evaluated studies published through March 2009 that assessed the association between polymorphisms in the 5-HTTLPR within the SLC6A4 gene and stressful life events and/or a diagnosis of depression. (9) The authors included 14 studies that had a total of 14,250 participants. In a meta-analysis of published data, there was no association between 5-HTTLPR genotype (homozygous short, homozygous long, or heterogeneous) and depression (weighted odds ratio [OR]; 95% confidence interval [CI], 0.98 to 1.13). There was also no interaction between genotype and the effect of stressful life events on depression (weighted OR=1.01; 95% CI, 0.94 to 1.10). In 2010, Karg et al. reported results from another systematic review and meta-analysis that evaluated the association between 5-HTTLPR polymorphisms and stressful life events and a diagnosis of depression. (10) Using broader search criteria, the authors included 54 studies that had a total of 40,749 patients. In their metaanalysis, conducted using the Liptak-Stouffer z score method to combine studies at the level of significance tests, weighted by study sample size, the authors found a significant association between the presence of the 5- HTTLPR short allele and increased risk of developing depression under stress (p=0.00002). When they confined their analysis to only those studies used in the Risch et al. meta-analysis, there was no significant association between 5-HTTLPR polymorphisms and depression. In 2010, Kiyohara and Yoshimasu reported results from a systematic review and meta-analysis of studies that assessed the association between 5-HTTLPR polymorphisms and depression. (11) The authors included 22 studies, all case-control studies, published through March 2008 that included a total of 7919 patients. Analyses were stratified by ethnicity due to significant between-study heterogeneity in the frequency of the variant 5- HTTLPR allele. In pooled analysis, the homozygous short genotype was significantly associated with depression risk among whites (OR=1.41, 95% CI: 1.15 to 1.72), but not in Asians. SULT4A1 Gene and Risk of Schizophrenia and Related Disorders. Based on a study targeting a polymorphism in the 5 untranslated region of the SULT4A1 gene in 27 families with at least 2 siblings with schizophrenia or schizophrenia spectrum disorder, the SULT4A1 gene has been evaluated as a candidate gene for schizophrenia. Meltzer et al evaluated a panel of patients with schizophrenia or schizoaffective disorder and available DNA to determine the association between 3 SULT4A1 single nucleotide polymorphisms (SNPs) (rs138060, rs138097, and rs138110) and clinical symptoms and quality of life. (12) Among 86 participants included, although all patients had a diagnosis of schizophrenia or schizoaffective disorder, the rs138060 SNP was significantly associated with worse Brief Psychiatric Rating Scale total and anxiety/depression

scores and higher SCALE for the Assessment of Positive Symptoms total scores. In addition, the rs138097 SNP was significantly associated with worse neuropsychological test performance. CACNAIC and ANK3 Genes and Risk of Multiple Mental Health Disorders. The CACNAIC gene has been studied most widely for its association with disorders of cardiac rhythm, such as long QT syndrome and Brugada syndrome. A lesser amount of research has reported associations of polymorphisms of this gene with schizophrenia and bipolar disorder. (13) In 2015, Jiang et al. published a meta-analysis of studies evaluating the association between the CACNA1C SNP rs1006737 and schizophrenia risk in East Asian populations. (14) The authors included 5 case-control studies in East Asian samples including 9432 cases with schizophrenia and 10,661 controls for their primary analysis. A second analysis was conducted for pooled East Asian and European populations, which included 2 additional candidate gene studies and 1 GWAS study in Europeans, for a total of 21,246 cases and 38,072 controls. In East Asian populations, the rs1006737 SNP was significantly associated with risk of schizophrenia (allelic model: pooled OR=1.20 for A allele; p=4.39x10-6). When the European studies were included, there was as stronger association between the CACNAIC polymorphism and schizophrenia risk (allelic model: pooled OR=1.12 for A allele; p=2.40x10-17). Kloiber et al. published results from 2 case-control studies evaluating the association of major depressive disorders with CACNAIC and ANK3. (15) The first population consisted of 720 patients with depression and 542 patients without psychiatric disease. The second population included 827 patients with recurrent depression and 860 patients without psychiatric disease. There were several SNPs on both genes that showed a statistical association with depression on initial analysis, but none of these remained significant after controlling for multiple comparisons. This evidence did not support a strong association between variants of these genes and depression. COMT Genes and Risk of Multiple Mental Health Disorders. For the COMT gene, polymorphisms have been reported to be associated with cognitive function, emotional processing, and other cognitive tasks. (16, 17) However, a more recent meta-analysis found no significant association between COMT genotype and several cognitive phenotypes. (18) In addition, associations with specific psychiatric conditions such as schizophrenia are less certain. (19) Dopamine Receptors and Transporter Genes and Risk of Multiple Mental Health Disorders. The dopamine receptor genes (DRD1, DRD2, DRD4) and the dopamine transporter (DAT1) gene have been associated with mood disorders, schizophrenia, and substance abuse disorders. For the DRD2 gene, a meta-analysis of case control studies that examined the presence of the cys311 polymorphism in patients with schizophrenia and patients without schizophrenia was published by Jonsson et al. (20) A total of 9,152 individuals were included, 3,707 individuals with schizophrenia and 5,363 control patients without schizophrenia. Combined analysis showed a significant association of this allele with schizophrenia (OR=1.43; 95% CI: 1.16 to 1.78; p<0.001). A 2014 meta-analysis which included 13 articles (n=3079 schizophreniz cases, n=3851 controls) reported associations between the DRD2 C957T polymorphism and schizophrenia risk (for C vs T: OR=1.26; 95% CI, 1.09 to 1.45, raw p=0.002; after Bonferroni and Benjamini- Hochberg correction, p=0.005; for CC and CT vs TT: OR=1.47; 95% CI, 1.25 to 1.73; raw p<0.001; after Bonferroni and Benjamini-Hochberg correction, p<0.001). (21) Variants in the DRD2 gene have also shown associations with disorders other than schizophrenia. Zou et al. reported results of a meta-analysis of studies that assessed the association between 3 DRD2 polymorphisms and mood disorders (bipolar disorder and unipolar depression). (22) A total of 2157 cases and 3272 controls from 14 studies were included. A significant association was demonstrated between 1 polymorphism assessed (TaqI A1) and mood disorders (OR=1.84; 95% CI: 1.07 to 3.17; p=0.03). For the DRD4 gene, in another meta-analysis, Lopez Leon et al. reviewed studies that evaluated the association between DRD4 polymorphisms and mood disorders, including unipolar depression and bipolar disorder. (23) Twelve studies that used a patient-control design and reported allele frequencies were included. DRD4

polymorphisms were significantly associated with unipolar depression (p<0.001) and the combined group of unipolar depression and bipolar disorder (p<0.001). For the DRD1 gene, case-control studies have linked polymorphisms to both increased and decreased risk of schizophrenia, (24) along with addictive behaviors including smoking and alcohol dependence. (25, 26) A 2014 meta-analysis of studies evaluating the association between DRD1 polymorphisms and schizophremia risk found that the rs5326 but not the rs4532 SNP was associated with schizophrenia. (27) For the DAT1 dopamine transporter gene (also known as SLC6A3), a number of studies have demonstrated an association between gene polymorphisms and addictive behaviors. For example, in a systematic review and meta-analysis of 5 studies that included 2155 patients, Stapleton et al found that variable number tandem repeat alleles in the 3 untranslated region of the DAT1 gene was associated with greater odds of smoking cessation (overall pooled OR=1.20; 95% CI: 1.01 to 1.43). (28) In another systematic review and meta-analysis, Du et al. found that polymorphisms in the 3 untranslated region of the DAT1 gene were associated with alcoholism with a history of delirium tremens or alcohol withdrawal seizures, although no significant association was seen between polymorphisms and alcoholism in general. (29) In contrast, u and Lin performed a systematic review and metaanalysis of 13 case-control studies evaluating the association between polymorphisms in the 3 untranslated region of the DAT1 gene and alcoholism and found no significant associations. (30) MTHFR Gene and Risk of Depression, Bipolar Disorder, and Schizophrenia The MTHFR gene has been widely studied for non-psychiatric conditions such as hyperhomocysteinemia and thrombophilia. A review of evidence on the association between this gene and thrombophilia is included in a separate Related Policy. For psychiatric disease, Wu et al. performed a meta-analysis of 26 GWAS evaluating the association of MTHFR variants with depression. (31) Overall, there were low-strength associations between numerous MTHFR SNPs and depression, with odds ratios ranging from 1.15 to 1.42. On subgroup analysis, the associations were stronger for Asian populations. In whites, the associations were of marginal significance, and in elderly patients the associations were not statistically significant. In a subsequent meta-analysis, Hu et al. evaluated the association between MTHFR variants and risk of bipolar disorder or schizophrenia. (32) In a meta-analysis of 38 studies, the authors found a significant association between the MTHFR C677T variant and schizophrenia (comparison, TT vs. CT or CC; OR=1.34; 95% CI: 1.18 to 1.53). For bipolar disorder, there was a marginal association between the C677T variant and disease risk (comparison, TT vs. CT or CC; OR=1.26; 95% CI, 1.00 to 1.59). Since the publication of the Wu et al meta-analysis, Bousman et al. conducted a prospective cohort study to evaluate the association between MTHFR genetic variants and prognosis of major depressive disorder. (33) The study included 147 primary care attendees with major depression who underwent genotyping for 2 functional MTHFR polymorphisms (C677T [rs1801133] and A1298C [rs1801131]) and 7 haplotype-tagging SNPs and serial measures of depression. The C677T polymorphism was significantly associated with symptom severity trajectory measured by the Primary Care Evaluation of Mental Disorders Patient Health Questionnaire-9 (p=0.038). The A1298C polymorphism and the haplotype-tagging SNPs were not associated with disease prognosis. In contrast, Lizer et al. conducted a case-control study that included 156 subjects and found no significant differences in the frequency of various MTHFR C667T genotypes between depressed and nondepressed patients. (34) MTHFR mutations have also been associated with schizophrenia and bipolar disorder. Peerbooms et al. conducted a systematic review and meta-analysis of case control studies evaluating associations between the MTHFR SNPs C677T and A1298C and schizophrenia, bipolar disorder, and unipolar depression. (35) The analysis included 24 studies related to schizophrenia, 10 related to bipolar disorder, and 17 related to unipolar depression. The C677T SNP was significantly associated with all disorders combined (OR=1.26 comparing homozygotes; 95% CI: 1.09 to 1.46). The A1298C SNP was significantly associated bipolar disorder (OR=2.03 comparing homozygotes; 95% CI: 1.07 to 3.86). Section Summary The association between mental health disorders and individual gene polymorphisms is an area of active investigation. For tests that are included in currently available genetic testing panels, the largest body of evidence

appears to be related to the role of SLC6A4 and various dopamine receptor gene polymorphisms and multiple mental health disorders. For these and other gene polymorphisms, the association between genetic polymorphisms and disease risks appears to be relatively weak and is not consistently demonstrated across studies. Genes Associated with Medication Pharmacokinetics and Pharmacodynamics Medications are a mainstay of treatment for many mental health disorders. Genetic polymorphisms may alter medications pharmacokinetics (i.e., how medications are absorbed, distributed, metabolized, or excreted) or pharmacodynamics (i.e., medications effects on the body); in turn, inter-individual differences in pharmacokinetics and pharmacodynamics may lead to variability in the clinical effectiveness of medications used to treat mental health disorders. Overview: Pharmacogenetics and Mental Health Disorders Several studies have summarized the associations between multiple candidate genes and single or multiple mental health disorders. Alter et al., in a study funded by Assurex, the manufacturer of the GeneSight Psychotropic panel, conducted a systematic review to assess whether the efficacy and/or adverse effects of 26 antipsychotic and antidepressant medications are associated with polymorphisms in 8 genes: CYP2D6, CYP2C19, CYP2C9, CYP1A2, CYP3A4, 2serotonin receptor genes (HTR2C, HTR2A), and SLC6A4. (36) The authors reviewed 294 studies that met their inclusion criteria. Thirty-two of the studies assessed associations between 5-HTR2C polymorphisms and various aspects of mental health disease. These included drug response, remission, adverse drug reactions, and evaluation of weight gain or metabolic syndrome in patients with psychiatric disorders (most commonly schizophrenia or schizoaffective disorders). Significant associations between at least 1 HTR2C allele and metabolic syndrome were found in 6 of the 7 studies that evaluated metabolic syndrome. Thirty-nine studies assessed the association between 5-HTR2A polymorphisms and adverse events or drug efficacy; 5 of the 10 studies that evaluated antipsychotic-related adverse events found a significant association between 5-HTR2A polymorphisms and adverse drug reactions, including weight gain, tardive dyskinesia, extrapyramidal adverse effects, and antipsychotic-induced Parkinsonism. Seventy-four studies evaluated associations between the SLC6A4 gene and drug response, remission, or adverse events (AEs), most commonly related to the use of SSRIs. Fifty-four studies investigated the most frequently assessed polymorphism (5-HTTLPR long / short ), with 29 studies showing a significant association with drug response or remission. Studies on a number of p450 genes were also assessed and generally included associations of genotype with phenotypic pharmacokinetic measures, including extensive metabolism (EM), intermediate metabolism (IM), poor metabolism (PM), and ultrarapid metabolism (UM) status. The authors conclude that there is substantial evidence of the association between polymorphisms and patient response to psychotropic medications; however, questions remain about how to incorporate testing for polymorphisms into clinical practice. In a study published in 2014 which was not included in the Alter et al. systematic review, Yin et al. assessed the association between SLC6A2, SLC6A3, DRD2, and DRD4 polymorphisms and response to SSRI therapy in a clinical trial of 229 patients undergoing treatment for depression. (37) DRD4 gene polymorphism (the rs1800544 polymorphism) differed significantly between drug responders and nonresponders (p<0.05), with no significant association with response seen for the other genotypes. In a 2014 trial comparing outcomes for patients with depression randomized to either antidepressant therapy or interpersonal counseling (N=137, with 97 in the antidepressant therapy group), SL6A4 genotype (AA genotype and A allele) were associated with response rate to antidepressants in the antidepressant group (p=0.015 and p=0.005, respectively). (38) Dopamine Receptor Genes and Antipsychotic Response. A number of studies have evaluated polymorphisms in the dopamine receptor genes (DRD1, DRD2) and response to treatment for schizophrenia. Zhang et al. reported results from a systematic review and meta-analysis of the association between DRD2 polymorphisms and response to antipsychotic agents among patients with schizophrenia. (39) The authors identified 6 studies that evaluated the role of the -141C Ins/Del polymorphism (N=687 patients). There was a significantly lower response rate to antipsychotics for patients who were Del carriers compared with Ins/Ins groups (pooled OR=0.65; 95% CI: 0.43 to 0.97; p=0.03). Eight studies were identified that evaluated the

association between a different polymorphism (TaqA1) and antipsychotic response (N=748 patients). There was no significant association between the TaqA1 polymorphism and antipsychotic response in pooled analysis. Studies investigating the relationship between polymorphisms in the DRD1 gene and antipsychotic response have not consistently reported a significant association. (24, 40) Serotonin Transporter (SLC6A4) Gene and Antidepressant Response. Polymorphisms in the SLC6A4 gene and the associated 5-HTTLPR region have been associated with variability in response to SSRIs and other antidepressant medications for several different mental health disorders, including depression, bipolar disorder, and generalized anxiety disorder. A number of studies have associated SCL6A4 polymorphisms with antidepressant response. In a 2011 systematic review and meta-analysis, Porcelli et al. evaluated the role of the 5-HTTLPR polymorphisms in predicting antidepressant response. (41) The authors identified 33 publications that compared outcomes after antidepressant use for either major depressive disorder or bipolar disorder, 28 of which were used in an analysis of SSRI response, and 8 in an analysis of other antidepressants. The 5-HTTLPR long allele was associated with remission when homozygous long patients were compared with homozygous short patients (for all antidepressant classes: OR=1.37; 95% CI: 1.09 to 1.72; p=0.007; for SSRIs only OR=1.48; 95% CI: 1.12 to 1.96; p=0.005). Studies on the role of SCL6A4 polymorphisms in antidepressant response that were not included in the Porcelli et al. meta-analysis have had mixed findings. For example, in an analysis of data from 125 patients from a randomized controlled trial comparing the SSRI escitalopram to placebo in the treatment of generalized anxiety disorder in older adults, Lenze et al. evaluated 2 SLC6A4-related polymorphisms, the 5-HTTLPR short/long polymorphism and the rs25531 g/a SNP. (35) Patients who did not have the combination of 5-HTTLPR long/rs25531 had no significant improvement with escitalopram, while those with other haplotypes had moderate improvement. In another prospective study, Seripa et al evaluated the association between SL6A4 polymorphisms and response to treatment with SSRIs (sertraline, paroxetine, or citalopram) in 234 subjects with late-life major depressive disorder.(43) Patients considered to be treatment responders were more likely to have the rs4795541-s allele (0.436 vs 0.321; p=0.023). In an additive regression model predicting treatment response, the single S-allele (0.436 vs 0.321; p=0.023). In an additive regression model predicting treatment response, the single S-allele dose-additive effect was associated with a n OR of 1.74 (95% CI, 1.12 to 2.69). Tomita et al reported that opposite degrees of association between plasma paroxetine concentrations and treatment response on the baiss of 5-HTTLPR genotype among 51 patients with major depressive disorder.(44) Among patients with two short alleles, paroxetine concentration was negatively correlated with improvement in depressive symptoms after 6 weeks, while for patients with 1 or 2 long alleles, paroxetine concentration was positively correlated with improvement in depressive symptoms after 6 weeks. In contrast, in an analysis of data from a randomized trial comparing the SSRI citalopram (n=258) to the norepinephrine uptake inhibitor reboxetine (n=262), Lewis et al. found no differences in treatment response for patients with different 5-HTTLPR genotype. (45) In a regression to predict Beck Depression Inventory Score at 6 weeks following enrollment, the coefficient for the interaction term treatment group and genotype was 0.50 (95% CI: -2.04 to 3.03; p=0.70), indicating no significant moderation of treatment effect by 5-HTTLPR genotype. Research has also evaluated the association between SCL6A4 polymorphisms and antidepressant adverse effects. In a systematic review and meta-analysis, Daray et al. evaluated the role of 5-HTTLPR polymorphisms and antidepressant-induced mania, a complication of antidepressant therapy that can be seen in patients with bipolar disorder. (46) Previous studies had reported that the long and short forms of this gene were associated with different rates of antidepressant-induced mania. In the authors meta-analysis, based on 6 studies that met their inclusion criteria, the short form of the gene was associated with an increased risk of antidepressant induced mania (combined risk ratio=1.35; 95% CI: 1.04 to 1.76). In contrast, in later systematic review and meta-analysis that used more stringent inclusion criteria, Biernacka et al. found no significant association between 5-HTTLPR polymorphisms and antidepressant-induced mania. (47) The SCL6A4 polymorphism has been associated with response to ondansetron, a 5-HT(3) receptor antagonist, among patients with alcohol dependence. (48)

Opioid Receptor Genes and Response to Treatment for Addiction. Several studies have evaluated the role of polymorphisms in the mu opioid receptor gene (OPRM1) and response to the opioid antagonist naltrexone for the treatment of alcohol dependence. Chamorro et al conducted a systematic review and meta-analysis to assess the relationship between the A118G polymorphism in the OPRM1 gene and response to treatment with naltrexone in patients with alcohol dependence. (49) The authors included 6 studies among patients with alcohol dependence. Naltrexone-treated patients who were homozygous for the A allele had a higher rate of relapse than those carrying the G allele (summary OR=1.97; 95% CI: 1.06 to 3.66; p=0.03). Cytochrome p450 Genes. A large amount of research has been conducted on the cytochrome p450 genes, with variants associated with altered drug metabolism for a wide variety of medications. A review of specific associations between these variations and metabolism of some psychiatric medications is discussed below. For selection and/or dosing of all psychiatric medications, genetic testing for cytochrome p450 variants is considered investigational. ABCB1 Gene and Antidepressant Response. Polymorphisms in the ABCB1 gene, which encodes a p-glycoprotein efflux pump which is involved in the transport of various molecules, including antidepressant drugs, across the blood-brain barrier, have been associated with response to antidepressant treatment. In 2015, Breitenstein et al. reported results of a meta-analysis of 16 pharmacogenetic studies (n=2695) evaluating the association between ABCB1 variants and antidepressant treatment outcomes for patients with major depression. (50) Six ABCB1 SNPs were evaluated: rs2032583, rs2235015, rs2235040, rs1045642, rs2032582, and rs1128503. Two SNPs, rs2032583 and rs2235015, were significantly associated with treatment response among inpatients (n=485) after Bonferroni correction (n=485 and p=1.5x10-5 for rs2032583; n=195 and p=3.0x10-4 for rs2235015). Selection or Dosing of Selective Serotonin Reuptake Inhibitors (SSRIs) CYP2D6 and CYP2C19 are primary CYP450 enzymes involved in the metabolism SSRIs. Thus, understanding a patient's metabolizer status might be helpful in choosing an initial SSRI and/or dose that is most likely to be effective. Systematic Reviews In January 2007, an Agency for Healthcare Research and Quality (AHRQ) Evidence-based Practice Center systematically reviewed the evidence on CYP450 testing for adults treated with SSRIs for nonpsychotic depression. (51) Following this commissioned report, the Evaluation of Genomic Applications in Practice and Prevention (EGAPP) Working Group published the following recommendation: The EGAPP Working Group found insufficient evidence to support a recommendation for or against use of CYP450 testing in adults beginning SSRI treatment for non-psychotic depression. In the absence of supporting evidence, and with consideration of other contextual issues, EGAPP discourages use of CYP450 testing for patients beginning SSRI treatment until further clinical trials are completed. (52) Observational Studies A number of prospective and retrospective studies have evaluated the association between CYP450 genotype and response to SSRIs. Several studies have focused on specific drugs, including paroxetine and escitalopram. Gex-Fabry et al. (53) studied paroxetine levels and clinical response in 71 patients with depression who had been genotyped for CYP2D6 and ABCBA polymorphisms. In this prospective observational study, CYP2D6 heterozygous extensive metabolizer phenotype showed a marginal impact on paroxetine levels and no impact on treatment response. Ververs et al., (54) in a cohort study of 74 pregnant women, demonstrated that differences in CYP2D6 genotype caused differential effects on paroxetine plasma concentrations. Extensive metabolizers (EMs) and UMs showed steady decreases in concentrations during the course of pregnancy, with increase in depressive symptoms. Intermediate metabolizers (IMs) and poor metabolizers (PMs) showed an increase in concentrations with no

change in symptoms. It was suggested that knowledge about CYP2D6 genotype would be indispensable in this setting. However, no information on the use or outcome of use of such information was provided. Tsai et al. (55) recently evaluated 100 patients diagnosed with major depressive disorder in an Asian population treated with escitalopram. These investigators evaluated ten alleles involving CYP2D6, CYP2C19, and CYP3A4 and concluded genetic polymorphisms of cytochrome P450 enzymes appeared to influence drug metabolism and treatment response. However, results were variable, and they were unable to provide a confident estimate of the ability of various allelic combinations to predict drug levels or treatment outcomes. Hodgson et al. evaluated the association between CYP450 genotype, antidepressant serum concentration, and treatment response in patients taking escitalopram (N=223) or the tricyclic antidepressant nortriptyline (N=161). (56) Genetic variation in CYP2C19 was significantly associated with serum escitalopram levels, while genetic variation in CYP2D6 was significantly associated with serum nortriptyline and 10-hydroxynortriptyline. However, there was no significant relationship between genotype and treatment response for either medication. Chang et al. evaluated the relationship between CYP2C19 polymorphisms and exposure to escitalopram and citalopram measured by serum/plasma levels in a meta-analysis of 14 studies. (57) Compared with EM homozygotes, citalopram or escitalopram concentrations significantly differed for other metabolizer states, as shown in Table 1. Table 1. (Es)citalopram Serum Concentration by CYP2C19 Genotype Genotype % Change in Serum 95% Confidence p (Es)citalopram Level Interval, % PM/PM (CYP2C19*2 or *3/*3 or *3) 95 40 to 149 <0.001 EM/PM (CYP2C19*1/*2 or *3) 30 4 to 55 <0.05 UM/PM (CYP2C19*17/*2 or *3) 25 1 to 49 <0.05 UM/UM (CYP2C19*17/*17) -36-46 to 27 <0.0001 UM/EM (CYP2C19*17/*1) -14-27 to -1 <0.05 Other studies have evaluated the association of p450 genotype variants with response to multiple antidepressants. Serretti et al., (58) in a retrospective study of 287 patients on antidepressants, demonstrated no association between response and allelic variations for p450 CYP1A2, CYP2C9, CYP2C19 and CYP2D6. Finally, Sim et al. (59) retrospectively studied 1,472 Swedish subjects looking for associations between CYP2C19 polymorphisms and depressive symptoms. They concluded that PMs exhibited a significantly lower level of depressive symptoms than extensive metabolizers (EMs). In the absence of drug-specific treatment outcomes or data related to drug levels, they suggested the need for further investigation of the functional link between CYP2C19 and depressive symptoms to further evaluate this observation. Section Summary Individuals with variants in multiple p450 genes have altered metabolism of SSRI drugs. However, the impact of genetic variants on clinical response and clinical outcomes is less clear, and the evidence is not sufficient to conclude that patients with genetic variants have reduced efficacy of SSRIs. Therefore, the clinical utility of testing for SSRI dose is uncertain. Selection and Dosing of Selective Norepinephrine Reuptake Inhibitors (SNRIs) Selective norepinephrine reuptake inhibitors (SNRIs) are used most commonly as antidepressants. Available agents in the U.S. include venlafaxine, duloxetine, and nefazodone. All of these drugs are metabolized by the cytochrome p450 system, and medication levels vary according to cytochrome p450 status. (60) Some of these agents, such as venlafaxine, are metabolized to an active metabolite by the CYP2D6 enzyme, while other agents, such as duloxetine, are inhibitors of cytochrome p450 activity. Lobello et al. (61) tested patients from 4 randomized controlled trials (RCTs) of venlafaxine versus placebo for CYP2D6 status and correlated genetic status, defined as either EMs or PMs, with response to treatment. There were no significant differences in dose of the drug according to genetic status. In 4 of 5 comparisons, patients who were EMs had a better response to treatment, as determined by depression rating scales. There was also a significantly greater percent of responders in the EM group compared with the PM. There were no differences in

discontinuation of therapy or adverse event rates between the EM and PM group. Waade et al. retrospectively evaluated the association between age, serum levels of venlafaxine and the SSRI escitalopram for different CYP2D6 and CYP2C19 genotype subgroups. (62) The study included 462 serum concentration measurements from 255 patients treated with venlafaxine and 953 serum concentration measurements from 541 patients treated with escitalopram. Patients were divided into 3 CYP2D6 (venlafaxine) or CYP2C19 (escitalopram) phenotype subgroups according to inherited genotype (PMs, heterozygous extensive metabolizers [HEMs], and EMs). An age-related difference (comparing patients <40 years, 40-65 years, and >65 years) was seen for venlafaxine, with a higher mean dose-adjusted serum concentration of venlafaxine for patients older than 65 years compared with those younger than 40 years in the PM group: 18.8 versus 2.4 (p<0.001). There were no significant age-related differences in serum venlafaxine concentration for HEM and EM patients and no association between age and escitalopram concentration regardless of genotype. For duloxetine, the inhibitory effects on cytochrome p450 activity are manifested by higher drug concentrations for other medications metabolized by cytochrome p450, such as tricyclic antidepressants (TCAs) and/or SSRIs. Similarly, other inhibitors of cytochrome p450 such as paroxetine, will increase levels of duloxetine.(63) Atomoxetine HCl is an SNRI that is approved to treat attention-deficit/hyperactivity disorder (ADHD). Atomoxetine, the active moiety, is primarily metabolized by CYP2D6. The therapeutic window for atomoxetine is wide, and dosing is weight-based, initiated at a standard dose per kilogram and adjusted thereafter according to clinical response and adverse effects. At steady state dosing, CYP2D6 PMs have substantially higher atomoxetine plasma concentrations than EMs, although because it is generally well-tolerated across a wide range, adverse effects do not appear to be significantly associated with PMs. (63,64) After titration, mean doses for EMs and PMs also do not differ significantly. (64,65) However, more EM patients discontinued in 1 trial due to lack of efficacy, (66) and PMs improved inattention scores more than EMs in another, (65) perhaps suggesting a need to reexamine recommended dosing limits. FDA decided not to include a recommendation to perform genotyping before prescribing atomoxetine. Dosing directions recommend a low starting dose to be increased to the target dose if well-tolerated. Thus, genotyping for CYP2D6 PMs of atomoxetine is not recommended because the margin of safety is not exceeded and evidence to support guidelines for dosing such that patient outcomes are improved has not been collected. (67,68) Indeed, Ramoz et al. (69) recently reported on 2 independent cohorts of 160 and 105 ADHD children treated for 6 weeks with atomoxetine. Interindividual response to the drug appeared independent of the genetic variants of CYP2D6. The authors did observe drug treatment and genomic associations, but these were found between drug response and a haplotype of the norepinephrine transporter (NET) gene Slc6a2. It was suggested further study be applied to assessment of this region to better manage patients being treated with this drug. Most recently ter Laak et al. (70) evaluated 100 patients treated for ADHD with standard doses of atomoxetine. A neurologist identified 10 of these who, based on late response or adverse effects, were subject to CYP P450 testing. Eight of the 10 were found to have a nonfunctional or less functional 2D6 allele. Four of these children showed improved responses on decreased atomoxetine; 4 were taken off treatment because of initial adverse events. While it is plausible that pretreatment testing could yield improved results, the study was not designed to evaluate the actual effect of testing on treatment outcomes. Section Summary SNRI metabolism is affected by genetic status of cytochrome p450, with the greatest potential clinical effect seen for venlafaxine. For this agent, EMs of CYP2D6 have higher levels of the active metabolite, and genetic status may have an impact on treatment response. A post hoc reanalysis of data from multiple RCTs has correlated treatment response to venlafaxine with genetic status. No studies have yet established that outcomes are improved as a result of genetic testing prior to initiating venlafaxine or other SNRIs. Atomoxetine is a SNRI that is used for ADD. Although it has a wide therapeutic window, there is potential for PMs who require relatively high doses to reach serum levels that may be toxic. However, current recommendations for starting atomoxetine at a low dose and watching closely for adverse effects while titrating higher should minimize the risk of toxicity for PMs. Selection and Dosing of Tricyclic Antidepressants

Nortriptyline and other TCAs are metabolized by the CYP2D6 enzyme. Patients who are PMs will develop serum concentrations of nortriptyline that are 3- to 10-fold higher than patients who are EMs. (71) de Vos et al. studied 678 patients treated with TCAs and reported that EMs had increased metabolism and lower serum levels of amitriptyline and citalopram, but not clomipramine. (72) However, these authors reported that the differences observed were not likely to have clinically important effects. In the study by Hodgson et al previously discussed, CYP2D6 genotype was associated with nortriptyline levels, but not with clinical improvement in 161 patients treated with nortriptyline. (73) It has been reported that patients with TCA overdose may have different risk depending on cytochrome p450 genetic status. (74,75) Simulations and case reports have reported that PMs may be at higher risk for toxic levels of nortriptyline and that toxic levels are maintained for longer periods of time. There are no clinical studies that demonstrate that measuring genetic status improves outcomes for patients who have had a TCA overdose. Section Summary Cytochrome p450 genetic status affects the metabolism and serum levels of multiple TCAs, including nortriptyline, but the clinical impact of these differences in metabolism are not clear. There is some evidence to suggest that patients who are PMs are more prone to toxic levels in the setting of a TCA overdose. There is no evidence available to support that prospective testing of patients treated with TCAs improves outcomes. Selection or Dose of Antipsychotic Drugs Classical antipsychotic agents (e.g., haloperidol, perphenazine, and risperidone) have therapeutic ranges that are often narrow, adverse effects that can be severe, and highly variable clinical responses. Case reports and small studies have reported associations between clinically significant adverse reactions or clinical responsiveness and specific CYP450 genotypes (e.g. CYP2D6, CYP3A4 variants), but most studies are small and results are inconsistent. (76-78) Moreover, plasma concentration of antipsychotic drugs may not be correlated with treatment outcome or adverse effects. (79) Because most patients with schizophrenia take combinations of psychoactive agents for extended periods of time, drug-drug and drug-environmental interactions may influence the CYP450 metabolic phenotype in addition to genotype. For example, carbamazepine, phenytoin, smoking, and alcohol consumption can induce CYP450 activity, whereas caffeine and fluvoxamine are inhibitors of CYP1A2. Some antipsychotic medications are metabolized by multiple CYP450 enzymes, and dominant pathways may vary. Several classical antipsychotic drugs inhibit the CYP450 enzyme required for their metabolism and may render the patient a phenotypic PM, despite an EM genotype. Thus, initial dosing algorithms need to accommodate both genetic influences and other interactions; therapeutic drug monitoring will probably continue to be needed to reflect the metabolic phenotype during ongoing treatment. (80-83) Systematic Reviews Fleeman et al., (83) in a 2010 health technology assessment, reviewed 51 articles on clinical validity of testing for cytochrome P450 in patients with schizophrenia treated with antipsychotic medications. The authors concluded that patients with heterozygous or homozygous mutations for CYP2D6 were at increased risk for tardive dyskinesia (odds ratio, respectively, [OR]: 2.08 and 1.83) and patients with homozygous mutations at increased risk for Parkinsonism (OR: 1.64). However, no published reports on clinical utility were identified. The authors concluded further evidence is required to link phenotype to genotype. This assessment has recently been published in the medical literature. (84) In 2011 Fleeman et al. (84) published a systematic review and meta-analyses of cytochrome p450 testing for use in prescribing antipsychotics in adults with schizophrenia. After search of 2,841 papers, they identified 47 papers that described clinical validity but no published papers on clinical utility of testing. They found no convincing evidence of an association between test results and either drug efficacy or toxicity. Differences when seen (an association, for example, with tardive dyskinesia) were considered too small to be clinically meaningful. Observational Studies Other studies have reported on CYP2D6 polymorphisms and response to risperidone. Jovanovic et al. (85) evaluated the role of CYP2D6 in 83 drug-naïve patients undergoing a first episode of psychosis and treated with risperidone. While significant improvements were observed in positive and general symptoms using this drug, the investigators were unable to identify an association between treatment response and variations in either genetic

or drug concentration findings. Locatelli et al. (86) evaluated CYP2D6 genotypes in 50 patients hospitalized for acute schizophrenia and also treated with risperidone. They found elevations in risperidone plasma levels in patients classified as PMs or IMs based on genotyping. Drug efficacy is not reported, but these authors observed an association between genotype, levels of risperidone, and the occurrence of extrapyramidal syndromes. They were uncertain whether these observations were strong enough to support routine testing as an aid to assessment of drug toxicity and suggested further study was needed. In a study of 76 white adult males with schizophrenia who were being treated with risperidone, Almoguera et al. reported that CYP2D6 phenotype was associated with improved scores on the total and negative symptoms scales of the Positive and Negative Syndrome Scale. (87) Section Summary Individuals with genetic variants in the CYPD6 gene may be at increased risk for adverse effects of antipsychotic drugs, particularly extrapyramidal effects such as tardive dyskinesia. However, the clinical utility of testing is uncertain, since management changes as a result of genetic testing have not been evaluated. Selection or Dosing of Codeine Codeine is metabolized by CYP2D6 to morphine. Enhanced CYP2D6 activity (i.e., in CYP2D6 ultrarapid metabolizers [UMs]) predisposes to opioid intoxication. On August 17, 2007, the U.S. Food and Drug Administration (FDA) issued a warning regarding codeine use by nursing mothers. Nursing infants may be at increased risk of morphine overdose if their mothers are taking codeine and are ultra-rapid metabolizers of codeine. Information about genetic variation and risk of accelerated codeine metabolism is now included in package insert information. The warning was prompted by a 2006 case report concerning an infant who died of morphine overdose. The mother, prescribed codeine for episiotomy pain, was a CYP2D6 UM, with high levels of circulating morphine. Not mentioned in the original case report, but noted in a later publication, is the fact that the mother was also homozygous for the UGT2B7*2 metabolizing enzyme variant, which is believed to also contribute to higher than normal production of active opioids from codeine. (89) Currently, the FDA is not recommending genotyping for any population prior to prescribing codeine because there is only limited information about using this test for codeine metabolism. (90,91) Information is limited to associations of genotype with morphine exposure and adverse effects such as sedation in adults, and association of mothers genotype with morphine exposure in mothers and with infant central nervous system (CNS) depression. Studies have been small, with correspondingly few PMs and UMs for drawing conclusions. Madadi et al. (92) have recently described the use of a pedigree approach to aid in diagnosis, identification of other at-risk family members and simplification of pharmacogenomic analysis. However, they note that for most medical centers, the framework for performing this work may not exist, and its applicability and relevance to general use remain unestablished. In 2013, in response to reports of deaths that have occurred in children with obstructive sleep apnea who received codeine following tonsillectomy and/or adenoidectomy and had evidence of being UMs of codeine due to a cytochrome CYP2D6 polymorphism, FDA added a black box warning to the labeling for codeine listing its use for postoperative pain management in children following tonsillectomy and/or adenoidectomy as a contraindication. The FDA s guidelines state, Routine CYP2D6 genotype testing is not being recommended for use in this setting because patients with normal metabolism may, in some cases, convert codeine to morphine at levels similar to ultra-rapid metabolizers. (93) Section Summary Enhanced CYP2D6 activity is associated with risk of accelerated codeine metabolism with high levels of circulating morphine in rapid metabolizers, which is thought to have contributed to deaths in infants of nursing mothers prescribed codeine and in pediatric patients post-tonsillectomy. There is little evidence about the clinical utility of testing for CYP2D6 genotype. Clinical Utility Studies suggest that there may be a number of genetic variants associated with increased risk of mental health disorders and/or response to specific treatment, although estimates of the magnitude of the increased risk and findings of significance are variable across studies. For the individual tests, results from GWAS and case control studies are insufficient to determine clinical utility. To determine clinical utility, evidence is needed that testing for

variants in these genes leads to changes in clinical management that improve outcomes. Given that many of the available genetic tests for mental health disorders are offered as panels, there are two relevant questions that address the clinical utility of genetic testing for mental health disorders. First, does testing for specific genetic variants lead to changes in management that improve health outcomes? Second, does a testing strategy that relies on a panel of tests lead to improved health outcomes compared with a strategy that relies on testing for variants individually? Does genetic testing for mental health disorders lead to improved health outcomes? Management changes that might be made in response to genetic testing information include selection of specific medications according to test results, discontinuation of medications, and changes in dosing of medications. However, these management changes are not well-defined and may vary according to the judgment of the treating clinician. Additionally, genetic testing could potentially allow more accurate diagnosis of mental health disorders, particularly if a patient s symptoms are consistent with more than one disorder, allowing better targeting of therapy. Currently, there are no specific recommended changes in management that are linked to specific test results, making it difficult to assess whether a particular management change based on test results leads to improvements in health outcomes. Two comparative, nonrandomized studies from the same research group compared clinical outcomes in patients with genetic testing versus patients without genetic testing. In 2013, Hall-Flavin et al. reported results from an open-label, nonrandomized comparative trial to evaluate the effect of providing the GeneSight pharmacogenomics test results and report on the management of psychotropic medications used for major depressive disorder in an outpatient psychiatric practice. (94) Two-hundred twenty seven patients with major depressive disorder were enrolled and grouped consecutively into a guided group (n=113) or unguided group (n=114). All subjects had DNA samples collected and sent for the GeneSight test. Based on results from patients genotypes for CYP2D6, CYP2C19, CYP1A2, SLC6A4, and HTR2A, the test generates a proprietary interpretive report that included recommendations for use as directed, use with caution, or use with caution and with more frequent monitoring for each of 26 antidepressant and antipsychotic agents. Providers for patients in the guided group received the report from the GeneSight test report. Subjects were followed for 8 weeks; 93 patients in the unguided group and 72 patients in the guided group completed follow up. In an analysis of those patients who completed follow up, the authors found a greater reduction in symptoms for the guided group compared with the unguided group for the depression measures used: Hamilton Rating Scale for Depression (HAMD-17; F=22.4, p<0.001)), the Quick Inventory of Depressive Symptomatology Clinician Rated (QIDS-C16; F=29.7, p<0.0001), and the Patient Health Questionnaire (PHQ-9; F=7.07, p=0.002). Patients in the guided group had a higher rate of remission as measured by the QIDS-C16 than the unguided patients (26.4% vs. 12.9%; OR=2.42; 95% CI: 1.09 to 5.39; p=0.03). Patients in the guided group who were initially on a medication that was classified as use with caution and with more frequent monitoring were more likely than those with the same classification in the unguided group to have a medication change or dose adjustment during the study period (93.8% vs 55%, χ2= 6.35; p=0.01). In an earlier nonrandomized pilot study, Hall-Flavin et al. compared outcomes for a group of patients with major depression whose physicians received a GeneSight report to a historical control group of patients who were treated without the GeneSight report. (95) Twenty-six subjects were included in the unguided group and 25 were included in the guided group. At 8 weeks of follow up, patients in the guided group had a reduction in their QIDS-C16 score of 31.2% compared with a 7.25%, reduction in the unguided group (p=0.002), and a reduction in their HAMD17 score of 30.8%, compared with a 18.2% reduction in the unguided group (p=0.04). While both Hall-Flavin et al. studies provide some evidence that a genotype report may be associated with differences in depression treatment outcomes, their limitations, including small sizes, nonrandomized designs, and loss to follow up, make generalization of their results difficult. One small RCT was published in 2012 by Winner et al., but this publication was not likely powered to detect differences in clinical outcomes. This trial evaluated the effect of providing the GeneSight pharmacogenomics test and report on the management of psychotropic medications used for major depressive disorder in a single outpatient psychiatric practice. (96) Fifty-one subjects were enrolled and randomized to a treatment as usual group or a GeneSight testing-guided group. All subjects underwent GeneSight testing and report preparation as described for the Hall-Flavin study previously discussed. At 10 weeks of follow up, treating physicians changed, augmented, or dose-adjusted subjects medication regimens with the same likelihood for the GeneSight group and the treatment as usual group (53% vs. 58% respectively; χ2= 0.19; p=0.66). However, patients in the

GeneSight group who were initially on a medication classified as use with caution and with more frequent monitoring were more likely than those with the same classification in the unguided group to have a medication change or dose adjustment (100% vs. 50% respectively; χ2= 5.09; p=0.02). Depression outcomes, measured by the HAMD-17 score, did not differ significantly at the 10-week follow up between groups. This study s small size may have limited its ability to detect a significant effect. Altar et al. reported the results of pooled analyses of data from the 3 studies previously described (Hall-Flavin et al. [2013], Hall-Flavin et al. [2012], Winner et al. [2012]). (97) Patients who received a red score on the basis of the GeneSight algorithm ( Use with increased caution and with more frequent monitoring ) had less improvement in HAMD-17 scores over 8 weeks compared with patients with yellow scores ( Use with caution ) or green scores ( Use as directed ), or yellow/green for subjects prescribed medications that are CYP2D6 substrates (p=0.001, p=0.01, p=0.002, respectively) and for subjects prescribed medications that are CYP2C19 substrates (p=0.003, p=0.02, p=0.004, respectively). None of the single genes included in the GeneSight panel was individually associated with either positive or negative treatment outcomes. Results of a survey of clinicians who have used the test are reported on the Genomind website. (98) A description of the methodology for this survey is not provided, therefore it is not possible to evaluate such factors as selection of the population or the survey response rate. Survey results were reported for 132 clinicians who used the test in the treatment of 545 patients. Clinicians reported that their treatment decisions were influenced by the test (definitely yes or probably yes) in 87% of cases. For patients in whom decisions were influenced, 76% of the treatment decisions involved a change in medication. Clinicians also reported that confidence in their treatment decisions were increased (definitely yes or probably yes) in 93% of the cases. In 2014, Breitenstein et al. reported results of a small nonrandomized comparative study to assess whether genotyping of the ABCB1 in clinical practice was associated with changes in medication management in outcomes among 58 patients with depression. (99) The authors note that previous studies of the association between ABCB1 genotype antidepressant plasma concentrations and/or treatment outcome in depression have had inconsistent results. In this study, study patients and matched controls were selected from the Munich Antidepressant Response Signature (MARS) project, a naturalistic study designed to identify factors that help to predict and improve treatment response in affective disorders. ABCB1 genotyping was implemented into the study s protocol in 2008, and genotype results were provided to treating physicians with a 1-page letter outlining potential strategies based on genotype (e.g., pay attention to sufficient dosing and consider changing to a medication that is not a substrate of the P-glycoprotein encoded by the ABCB1 gene for subjects who had 2 T alleles of the rs2032583 SNP and 2 G alleles of the rs2235015 SNP). The 58 patients who had ABCB1 genotyping were compared with a matched sample of historical controls who were enrolled before genotyping was implemented. Patients who received ABCB1 genotyping had higher remission rates at the time of hospital discharge (83.6% vs. 62.1%, p=0.005, 1-sided) and lower Hamilton depression scores at the time of hospital discharge (scores extrapolated from graph, 6 vs. 8 (p=0.02, 1-sided). Section Summary A limited number of studies have evaluated clinical outcomes associated with genetic testing panels for mental health disorders, primarily using the GeneSight pharmacokinetic test. These studies provide some evidence that a genotype report may be associated with differences in depression treatment outcomes, however, weaknesses in the studies limit the conclusions that can be drawn. The clinical utility of genetic panels and individual genetic tests for mental health disorders other than the GeneSight test has not been evaluated. Do panels of genetic tests for mental health disorders offer incremental benefits compared with testing for individual genes? A framework for determining the clinical utility of genetic panels is provided in a separate medical policy. (See Related Policies). According to the classification of panels in that policy, the panels evaluated in this policy are classified as a panel intended to assess risk for multiple conditions. The criteria to be used for evaluating panels in this category are as follows: Clinical utility has been established for at least one component of the panel. Test is performed in a CLIA-approved laboratory. Analytic validity of the panel approaches that of direct sequencing Panel offers substantial advantages (efficiency of work-up, cost) over sequential analysis of individual genes.

The impact of ancillary information is well-defined. The genetic testing panels reviewed do not meet the majority of these criteria. Most importantly, clinical utility has not been established for any of the individual tests for their intended purpose in this assay. In addition, the analytic validity of the test is unknown. The testing methods are not well-described, so there is uncertainty as to whether the efficiency of testing is improved by use of this panel. Finally, the impact of ancillary information is not well-defined. It is not known how the results of various tests might be combined to determine overall risk, nor is it known how unexpected results may impact treatment decisions and health outcomes. It is also unknown as to how the results of genetic variants that indicate increased risk for non-psychiatric conditions, such as variants in the CACNA1C gene that may denote an increased risk of cardiac disorders, will impact patient management and outcomes. Ongoing Clinical Trials A search of ClinicalTrials.gov using each test name as a key word identified the following randomized, controlled trials that are currently enrolling patients: PAGE trial (NCT01555021). The Pharmacogenomics for Antidepressant Guidance and Education trial is an RCT that is currently recruiting patients admitted to an inpatient psychiatric facility with treatmentresistant depression, as defined by a failure of at least one prior trial of antidepressant medication. Treatment guided by results of the Genecept Assay will be compared with usual care over a 6-month period. The primary outcome measure is the change in the Quick Inventory of Depressive Symptomatology-Self Report (QIDS-SR) at 6 months. Secondary outcomes include change in treatment decisions based on Genecept results, treatment adherence, and medication adverse effects. Planned enrollment is for 200 patients with an estimated completion date of December 2013. PAGE-1_AG1 trial (NCT01426516). The Pharmacogenomics for Antidepressant Guidance and Education 1 trial is an RCT that is currently recruiting patients with treatment-resistant depression, as defined by a failure of at least one prior trial of antidepressant medication. Treatment guided by results of the Genecept Assay will be compared to usual care over a 6-month period. The primary outcome measure is the change in the Quick Inventory of Depressive Symptomatology-Self Report (QIDS-SR) at 6 months. Secondary outcomes include change in treatment decisions based on Genecept results, quality of life, cost and patient/provider satisfaction. Planned enrollment is for 100 patients with an estimated completion date of December 2014. Impact of GeneSight Psychotropic on Response to Psychotropic Treatment in Outpatients Suffering From a Major Depressive Disorder (MDD) and Having Had an Inadequate Response to at Least One Psychotropic Medication Included in GeneSight Psychotropic (RCT) (NCT02109939). This is a doubleblind RCT designed to compare treatment guided by the GeneSight Psychotropic testing report with treatment as usual among adults with major depression. The primary outcome measure is the HAM-D17 score at 12 weeks postenrollment. Planned enrollment is for 300 subjects with an estimated completion date of February 2016. Table 2. Summary of Key Trials NCT No. Trial Name Planned Completion Date Enrollment Ongoing NCT02109939 a A 12-Week, Randomized, Double-Blind, Controlled Evaluation 300 Feb 2016 Followed by an Open-Label 12-Week Follow-up Period of the Impact of GeneSight Psychotropic on Response to Psychotropic Treatment in Outpatients Suffering From a Major Depressive Disorder (MDD) and Having Had - Within the Current Episode - an Inadequate Response to at Least One Psychotropic Medication Included in GeneSight Psychotropic Unpublished NCT01555021 Pharmacogenomics for Antidepressant Guidance and Education 200 Dec 2014 NCT01426516 a Six-month Study of the Genecept Assay vs. Treatment as Usual to 100 Dec 2014 Evaluate Efficacy of Using Assay Guided Treatment in Outpatient Adults With Treatment Resistant Depression Summary

Panels of multiple genetic tests have been developed to aid the diagnosis and treatment of mental health disorders. Genes included in the panels have shown some association with psychiatric disorders or with the pharmacokinetics of psychotropic medications. The analytic validity of these assays cannot be determined due to a lack of information on the testing methods. The available evidence on clinical validity consists of genome-wide association studies and case-control studies that indicate a correlation between variants of these genes and clinical factors. This evidence shows low-strength associations with a variety of psychiatric and non-psychiatric conditions. Often the evidence for an association is not consistently reported across all studies, and in many cases, there are correlations of the same genetic variants with other non-psychiatric disorders. There are also a range of associations reported for response to certain medications and alterations in pharmacokinetics. Evidence on clinical utility is lacking. Management changes that occur as a result of this assay are ill-defined, with uncertain impact on clinical outcomes. In addition, it is not well -understood how unexpected results or unknown variants are handled and whether these type of results have an impact on diagnostic work-up, treatment decisions, and health outcomes. Due to these deficiencies in the evidence base, genetic testing panels for mental health disorders are considered investigational for all indications. Practice Guidelines and Position Statements None identified. U.S. Preventive Services Task Force Recommendations Not applicable. Medicare National Coverage None. References [TOP] 1. Hu CY, Qian ZZ, Gong FF, et al. Methylenetetrahydrofolate reductase (MTHFR) polymorphism susceptibility to schizophrenia and bipolar disorder: an updated meta-analysis. J Neural Transm. Feb 2015;122(2):307-320. PMID 24938371. 2. Bousman CA, Potiriadis M, Everall IP et al. Methylenetetrahydrofolate reductase (MTHFR) genetic variation and major depressive disorder prognosis: A five-year prospective cohort study of primary care attendees. Am J Med Genet B Neuropsychiatr Genet 2014; 165(1):68-76. 3. Lizer MH, Bogdan RL, Kidd RS. Comparison of the frequency of the methylenetetrahydrofolate reductase (MTHFR) C677T polymorphism in depressed versus nondepressed patients. J Psychiatr Pract 2011; 17(6):404-9. 4. Peerbooms OL, van Os J, Drukker M et al. Meta-analysis of MTHFR gene variants in schizophrenia, bipolar disorder and unipolar depressive disorder: evidence for a common genetic vulnerability? Brain Behav Immun 2011; 25(8):1530-43. 5. Altar CA, Hornberger J, Shewade A et al. Clinical validity of cytochrome P450 metabolism and serotonin gene variants in psychiatric pharmacotherapy. Int Rev Psychiatry 2013; 25(5):509-33. 6. Yin L, Zhang YY, Zhang, et al. TPH, SLC6A2, SLC6A3, DRD2 and DRD4 Polymorphisms and Neuroendocrine Factors Predict SSRIs Treatment Outcome in the Chinese Population with Major Depression. Pharmacopsychiatry. Feb 2 2015. PMID 25642918 7. Matsumoto Y, Fabbri C, Pellegrini S, et al. Serotonin transporter gene: a new polymorphism may affect response to antidepressant treatments in major depressive disorder. Mol Diagn Ther. Oct 2014; 18(5):567-577. PMID 24958631 8. Zhang JP, Lencz T, Malhotra AK. D2 receptor genetic variation and clinical response to antipsychotic drug treatment: a meta-analysis. Am J Psychiatry 2010; 167(7):763-72. 9. Hwang R, Shinkai T, De Luca V et al. Association study of four dopamine D1 receptor gene polymorphisms and clozapine treatment response. J Psychopharmacol 2007; 21(7):718-27. 10. Porcelli S, Fabbri C, Serretti A. Meta-analysis of serotonin transporter gene promoter polymorphism (5-

HTTLPR) association with antidepressant efficacy. Eur Neuropsychopharmacol 2012; 22(4):239-58. 11. Lenze EJ, Goate AM, Nowotny P et al. Relation of serotonin transporter genetic variation to efficacy of escitalopram for generalized anxiety disorder in older adults. J Clin Psychopharmacol 2010; 30(6):672-7. 12. Seripa D, Pilotto A, Paroni G, et al. Role of the serotonin transporter gene locus in the response to SSRI treatment of major depressive disorder in late life. J Psychopharmacol. Mar 31 2015. PMID 25827644 13. Tomita T, Yasui-Furukori N, Nakagami T, et al. The influence of 5-HTTLPR genotype on the association between the plasma concentration and therapeutic effect of paroxetine in patients with major depressive disorder. PLoS One. 2014;9(5):e98099. PMID 24858363 14. Lewis G, Mulligan J, Wiles N et al. Polymorphism of the 5-HT transporter and response to antidepressants: randomised controlled trial. Br J Psychiatry 2011; 198(6):464-71. 15. Daray FM, Thommi SB, Ghaemi SN. The pharmacogenetics of antidepressant-induced mania: a systematic review and meta-analysis. Bipolar Disord 2010; 12(7):702-6. 16. Biernacka JM, McElroy SL, Crow S et al. Pharmacogenomics of antidepressant induced mania: a review and meta-analysis of the serotonin transporter gene (5HTTLPR) association. J Affect Disord 2012; 136(1-2):e21-9. 17. Johnson BA, Ait-Daoud N, Seneviratne C et al. Pharmacogenetic approach at the serotonin transporter gene as a method of reducing the severity of alcohol drinking. Am J Psychiatry 2011; 168(3):265-75. 18. Chamorro AJ, Marcos M, Miron-Canelo JA et al. Association of micro-opioid receptor (OPRM1) gene polymorphism with response to naltrexone in alcohol dependence: a systematic review and metaanalysis. Addict Biol 2012; 17(3):505-12. 19. Breitenstein B, Bruckl TM, Ising M, et al. ABCB1 gene variants and antidepressant treatment outcome: A metaanalysis. Am J Med Genet B Neuropsychiatr Genet. Apr 2 2015. PMID 25847751 20. Matchar DB, Thakur ME, Grossman I, et al. Testing for cytochrome P450 polymorphisms in adults with non-psychotic depression treated with selective serotonin reuptake inhibitors (SSRIs). Evid Rep Technol Assess (Full Rep). Jan 2007(146):1-77. PMID 17764209 21. Recommendations from the EGAPP Working Group: testing for cytochrome P450 polymorphisms in adults with nonpsychotic depression treated with selective serotonin reuptake inhibitors. Genet Med. Dec 2007; 9(12):819-825. PMID 18091431 22. Gex-Fabry M, Eap CB, Oneda B, et al. CYP2D6 and ABCB1 genetic variability: influence on paroxetine plasma level and therapeutic response. Ther Drug Monit. Aug 2008; 30(4):474-482. PMID 18641553 23. Ververs FF, Voorbij HA, Zwarts P, et al. Effect of cytochrome P450 2D6 genotype on maternal paroxetine plasma concentrations during pregnancy. Clin Pharmacokinet. 2009; 48(10):677-683. PMID 19743889 24. Tsai MH, Lin KM, Hsiao MC, et al. Genetic polymorphisms of cytochrome P450 enzymes influence metabolism of the antidepressant escitalopram and treatment response. Pharmacogenomics. Apr 2010; 11(4):537-546. PMID 20350136 25. Hodgson K, Tansey K, Dernovsek MZ, et al. Genetic differences in cytochrome P450 enzymes and antidepressant treatment response. J Psychopharmacol. Feb 2014; 28(2):133-141. PMID 24257813 26. Chang M, Tybring G, Dahl ML, et al. Impact of Cytochrome P450 2C19 Polymorphisms on Citalopram/Escitalopram Exposure: A Systematic Review and Meta-Analysis. Clin Pharmacokinet. Sep 2014; 53(9):801-811. PMID 25154506 27. Serretti A, Calati R, Massat I, et al. Cytochrome P450 CYP1A2, CYP2C9, CYP2C19 and CYP2D6 genes are not associated with response and remission in a sample of depressive patients. Int Clin Psychopharmacol. Sep 2009; 24(5):250-256. PMID 19593158 28. Sim SC, Nordin L, Andersson TM, et al. Association between CYP2C19 polymorphism and depressive symptoms. Am J Med Genet B Neuropsychiatr Genet. Sep 2010; 153B (6):1160-1166. PMID 20468063 29. Bertilsson L. Metabolism of antidepressant and neuroleptic drugs by cytochrome p450s: clinical and interethnic aspects. Clin Pharmacol Ther. Nov 2007; 82(5):606-609. PMID 17898711 30. Lobello KW, Preskorn SH, Guico-Pabia CJ, et al. Cytochrome P450 2D6 phenotype predicts antidepressant efficacy of venlafaxine: a secondary analysis of 4 studies in major depressive disorder. J Clin Psychiatry. Nov 2010; 71(11):1482-1487. PMID 20441720 31. Waade RB, Hermann M, Moe HL, et al. Impact of age on serum concentrations of venlafaxine and escitalopram in different CYP2D6 and CYP2C19 genotype subgroups. Eur J Clin Pharmacol. Aug 2014; 70(8):933-940. PMID 24858822 32. Skinner MH, Kuan HY, Pan A, et al. Duloxetine is both an inhibitor and a substrate of cytochrome P4502D6 in healthy volunteers. Clin Pharmacol Ther. Mar 2003; 73(3):170-177. PMID 12621382 33. de Leon J. The crucial role of the therapeutic window in understanding the clinical relevance of the poor versus the ultrarapid metabolizer phenotypes in subjects taking drugs metabolized by CYP2D6 or CYP2C19. J Clin Psychopharmacol. Jun 2007;27(3):241-245. PMID 17502769 34. Trzepacz PT, Williams DW, Feldman PD, et al. CYP2D6 metabolizer status and atomoxetine dosing in

children and adolescents with ADHD. Eur Neuropsychopharmacol. Feb 2008; 18(2):79-86. PMID 17698328 35. Michelson D, Read HA, Ruff DD, et al. CYP2D6 and clinical response to atomoxetine in children and adolescents with ADHD. J Am Acad Child Adolesc Psychiatry. Feb 2007; 46(2):242-251. PMID 17242628 36. Wernicke JF, Kratochvil CJ. Safety profile of atomoxetine in the treatment of children and adolescents with ADHD. J Clin Psychiatry. 2002; 63 Suppl 12:50-55. PMID 12562062 37. http://www.ehs.lilly.com/msds/msds_atomoxetine_hydrochloride_tablets_and_capsules.html Accessed April 6, 2016. 38. Ramoz N, Boni C, Downing AM, et al. A haplotype of the norepinephrine transporter (Net) gene Slc6a2 is associated with clinical response to atomoxetine in attention-deficit hyperactivity disorder (ADHD). Neuropsychopharmacology. Aug 2009; 34(9):2135-2142. PMID 19387424 39. ter Laak MA, Temmink AH, Koeken A, et al. Recognition of impaired atomoxetine metabolism because of low CYP2D6 activity. Pediatr Neurol. Sep 2010; 43(3):159-162. PMID 20691935 40. Macaluso M, Preskorn SH. CYP 2D6 PM status and antidepressant response to nortriptyline and venlafaxine: is it more than just drug metabolism? J Clin Psychopharmacol. Apr 2011; 31(2):143-145. PMID 21346604 41. de Vos A, van der Weide J, Loovers HM. Association between CYP2C19*17 and metabolism of amitriptyline, citalopram and clomipramine in Dutch hospitalized patients. Pharmacogenomics J. Oct 2011; 11(5):359-367. PMID 20531370 42. Jornil J, Jensen KG, Larsen F, et al. Risk assessment of accidental nortriptyline poisoning: the importance of cytochrome P450 for nortriptyline elimination investigated using a population-based pharmacokinetic simulator. Eur J Pharm Sci. Oct 9 2011; 44(3):265-272. PMID 21854846 43. Maier W, Zobel A. Contribution of allelic variations to the phenotype of response to antidepressants and antipsychotics. Eur Arch Psychiatry Clin Neurosci. Mar 2008; 258 Suppl 1:12-20. PMID 18344045 44. Crescenti A, Mas S, Gasso P, et al. Cyp2d6*3, *4, *5 and *6 polymorphisms and antipsychotic induced extrapyramidal side-effects in patients receiving antipsychotic therapy. Clin Exp Pharmacol Physiol. Jul 2008; 35(7):807-811. PMID 18346175 45. Smoller JW. Incorporating pharmacogenetics into clinical practice: reality of a new tool in psychiatry. Practical issues related to medication selection. CNS Spectr. Mar 2006; 11(3 Suppl 3):5-7. PMID 17760223 46. Panagiotidis G, Arthur HW, Lindh JD, et al. Depot haloperidol treatment in outpatients with schizophrenia on monotherapy: impact of CYP2D6 polymorphism on pharmacokinetics and treatment outcome. Ther Drug Monit. Aug 2007; 29(4):417-422. PMID 17667795 47. Murray M, Petrovic N. Cytochromes P450: decision-making tools for personalized therapeutics. Curr Opin Mol Appendix [TOP] Appendix Table 1. Categories of Genetic Testing Addressed in this Policy Category Addressed Yes No 1. Testing of an affected individual s germline to benefit the individual a. Diagnostic b. Prognostic c. Therapeutic 2. Testing cancer cells from an affected individual to benefit the individual a. Diagnostic b. Prognostic c. Therapeutic 3. Testing an asymptomatic individual to determine future risk of disease 4. Testing of an affected individual s germline to benefit family members 5. Reproductive testing a. Carrier testing: preconception b. Carrier testing: prenatal

c. In utero testing: aneuploidy d. In utero testing: mutations e. In utero testing: other f. Preimplantation testing with in vitro fertilization History [TOP] Date Reason 12/09/13 New Policy. New policy developed with literature review through September 30, 2013. The Genecept assay is investigational for all indications. 05/23/14 Update Related Policies. Add 12.04.509 and removed 12.04.82 as it was deleted. 08/11/14 Annual Review. Policy updated with literature review through April 14, 2014. Policy expanded to include other genetic testing panels for mental health disorders; title of policy changed to Genetic Testing Panels for Mental Health Conditions. Rationale extensively revised. References 1, 2, 7-11, 19-26, 28-8 added. Policy statement changed to indicate that individual genetic tests (as mutations or genetic variations) and genetic testing panels for mental health disorders are investigational. 09/10/14 Minor update. New test added to Policy Statement for genetic testing panels: Primex Expanded Pharmacogenomics Panel. 12/17/14 Minor update. New tests added to investigational Policy Statement for genetic testing panels: Ally Diagnostics Genetic Testing Panel and Molecular Testing Labs Psychotropic Medication Panel. Section reformatted for ease of reading. 01/20/15 Minor update. New tests added to investigational Policy Statement for genetic testing panels: PharmaRisk Basic and PharmaRisk Psychiatric Panel. 02/27/15 Minor update. New test added to investigational Policy Statement for genetic testing panels: Physicians Choice Laboratory Services (PCLS) Pharmacogenetic Testing. Related policies updated with new policy 12.04.131. 03/24/15 Minor update. New test added to investigational Policy statement for genetic testing panels: Frontier Toxicology PGx Pharmacogenomic Testing. 04/24/15 Minor update: Alpha Genomix Psychiatry/ADHD Panel and PGL Multi-Drug Panel added to investigational Policy statement for genetic testing panels. 07/14/15 Annual Review. Policy number changed from 12.04.110 to 12.04.515 due to the addition of several local plan tests to the Policy Statement, Description and Reference section. In this revision, PersonaGene, Progenity PGx Informed and two Luminex panel tests added. Policy updated with literature review through April 21, 2015. Numerous references added. Policy statements changed to clarify which categories of genetic testing the policy addresses; intent of policy statements unchanged. 10/19/15 Update Related Policies. Remove 12.04.509 as it was archived. 04/12/16 Annual review. Added rationale and references for CYP450 for use in review of mental health conditions/medications. References 51-93 added. No change in policy statements. Disclaimer: This medical policy is a guide in evaluating the medical necessity of a particular service or treatment. The Company adopts policies after careful review of published peer-reviewed scientific literature, national guidelines and local standards of practice. Since medical technology is constantly changing, the Company reserves the right to review and update policies as appropriate. Member contra cts differ in their benefits. Always consult the member benefit booklet or contact a member service representative to determine coverage for a specific medical service or supply. CPT codes, descriptions and materials are copyrighted by the American Medical Association (AMA). 2016 Premera All Rights Reserved.