Do You Order Pharmacogenetic Testing? Why?

This work may not be copied, distributed, displayed, published, reproduced, transmitted, modified, posted, sold, licensed, or used for commercial purposes. By downloading this file, you are agreeing to the publisher’s Terms & Conditions.

Do You Order Pharmacogenetic Testing? Why?

Joseph F. Goldberg, MDa,*

The first question posed in the title of this article, whether pharmacogenetic testing should be ordered, arises with increasing frequency at psychiatric conferences and in discussions among peers. The second question, why to order it, is posed less often, apart from broad declarations that "pharmacogenetic testing can help guide treatment." How, exactly? And when? Seldom does the idea of pharmacogenetic testing incur the usual skepticism with which psychiatrists size up purported treatment advances. Perhaps this is due to high hopes that, despite its nontrivial cost, such testing enhances care and also demonstrates the biological underpinnings of our prescribing practices. While the ideals of personalized medicine are linked closely with pharmacogenetics, most notably in cancer, the enthusiasm with which many clinicians now embrace pharmacogenetic testing as a clinically useful tool, given the present state of knowledge, warrants a critical and dispassionate appraisal.

The proposition that genetics may play a role in drug response is tested at the molecular level by pharmacogenetics and at the macro level by examining familiality of drug response. Countless practitioners seize upon historical information that a patient’s relative—however distant—"responded" to a particular treatment and conclude that a similar outcome is therefore expectable for the proband. Literature to support or refute this presumption is scant, with only a few notable exceptions: concordance rates of about two-thirds have been shown with lithium responsivity between bipolar probands and their affected first-degree relatives,1 and similar concordance rates have been reported for antidepressant response to fluvoxamine among major depression patients and their affected first-degree relatives.2 The paucity of data certainly does not negate the possible heritability of drug response, but rather serves as a reminder that assumptions about a genetic basis for treatment response are more speculative than factual.

Separate from the question of whether drug response is a heritable trait is whether drug effects can be predicted from single-nucleotide polymorphisms (SNPs) for genes involved in the presumptive neural circuitry associated with a given psychiatric disorder. Here too, data are lacking to establish that pharmacodynamic effects broadly constitute a genetically mediated phenomenon. Some authorities3 point out 3 core considerations around pharmacogenetic testing: (1) analytic validity (eg, whether a test accurately detects allelic variants); (2) clinical validity (ie, whether a particular genetic profile causes clinically relevant effects); and (3) clinical utility (ie, whether testing results meaningfully alter treatment outcomes). The first of these issues is hampered in part because, except for the Roche AmpliChip CYP450 Test, no commercially available pharmacogenetic tests are approved or regulated by the US Food and Drug Administration. This article focuses mainly on the latter two of these concepts, along with testing intended to help predict drug benefits ("efficacy pharmacogenetics"4) versus adverse effects ("safety pharmacogenetics"4), particularly in major depression.

Evidence to support commercial claims that pharmacogenetic tests can identify treatment responsiveness is modest and indirect, stemming mainly from open-label, nonrandomized industry-sponsored studies showing improvements from baseline in symptoms, quality of life, or patient (or doctor) satisfaction, when clinicians are given results from a panel of pharmacokinetic gene variants (eg, cytochrome P450 [CYP]) and putative pharmacodynamic gene SNPs (eg, related to serotonin or dopamine receptor functioning).5-7 Particularly given the many clinical, psychosocial, and other factors that influence depression outcomes, it is difficult to attribute drug response to genotype guidance when study designs lack sham-guidance control groups, information about actual medications and dosages, drug adherence, diagnostic standardization and reliability, possible ascertainment bias, control for confounding factors (racial, demographic, or clinical characteristics or cotherapies), or uniformity of study psychiatrists’ experience and expertise. CYP2D6 poor metabolizers (PMs) have been shown from post hoc analyses in venlafaxine trials to have less robust improvement on depression severity scales,8 although rating scales may not neatly differentiate physical symptoms of depression from adverse drug effects.

In major depression, a large meta-analysis found that the l (long) (versus s [short]) variant of the serotonin transporter gene (SLC6A4) only modestly predicted selective serotonin reuptake inhibitor (SSRI) response or remission, with caveats (findings applied to older white women with late illness onset).9 Underpowered and unreplicated, findings from candidate gene association studies such as these are at best preliminary and lack generalizability. In far larger genome-wide association studies (GWAS) of depression, replicated findings of putative susceptibility loci that achieve genome-wide significance are virtually nonexistent. Indeed, pharmacogenomic GWAS for SSRI response in depression have yielded no findings of genome-wide significance (eg, Biernacka et al10). Clinicians who may be unfamiliar with such study design limitations in genetics are vulnerable to accepting manufacturers’ claims at face value that a test will "double response rates" based on statistically nonsignificant findings from commercially sponsored, underpowered studies.

Current pharmacogenetic practice guidelines address pharmacokinetic more than pharmacodynamic considerations; they provide information about the likelihood of poor drug response in CYP2D6 or CYP2C19 ultrarapid metabolizers and a greater adverse effect burden among PMs due to decreased clearance.11,12 They offer helpful suggestions about compensatory dosing adjustments for ultrarapid or poor metabolizers but make no formal recommendations to practitioners about whether and when pharmacogenetic testing should be ordered as part of routine care. Nor has its cost-effectiveness been demonstrated in the treatment of major depression.13 On the basis of currently available data, the Centers for Disease Control and Prevention Evaluation of Genomic Applications in Practice and Prevention (EGAPP) Working Group found "insufficient evidence to support a recommendation for or against use of CYP testing in adults beginning SSRI treatment for nonpsychotic depression. In the absence of supporting evidence, and with consideration of other contextual issues, the EGAPP group discourages use of CYP testing for patients beginning SSRI treatment until further clinical trials are completed" (https://www.cdc.gov/genomics/gtesting/EGAPP/recommend/CYP450.htm).

The tides turn, somewhat, when discussion shifts from efficacy pharmacogenetics to safety pharmacogenetics. Genetically poor metabolizers of drugs that are substrates for CYP2D6 or CYP2C19 (about 5%-10% of whites and more variable numbers of nonwhites) are more susceptible to adverse effects. They may also poorly convert prodrugs that are metabolized by hepatic enzymes (eg, codeine) to their biologically active metabolites (eg, morphine). (However, most psychiatric prodrugs, such as lisdexamfetamine, become activated not by hepatic metabolism but by gastrointestinal or peripheral hydrolytic enzymes.) Does knowing the genotype of a "side effect-prone" patient, who might have a PM genotype, change their management? Perhaps, but would a clinician not simply use low doses in such patients and avoid potent inhibitors of their metabolic enzymes as a general rule? Moreover, poor drug tolerability is not fully explainable by genotype; in one study of venlafaxine for major depression, one-quarter of patients with normal ("extensive metabolizer") CYP2D6 genotypes resembled PM phenotypes, while adverse effects were 7 times more prominent in patients with normal than PM genotypes.14

Pharmacogenetic studies have very provisionally identified a handful of adverse effects that may be associated with particular candidate gene SNPs, which are possibly relevant (if replicated) for anticipating phenomena such as hyperprolactinemia or extrapyramidal effects from antipsychotics, nausea or sexual dysfunction with serotonergic antidepressants, and weight gain from atypical antipsychotics. The methylene tetrahydrofolate reductase gene (MTHFR) has a known functional polymorphism (C677T), coding for the enzyme needed to transport folic acid across the blood-brain barrier in order to enable CNS serotonin synthesis, but it is unknown whether oral supplemental l-methylfolate is "indicated" for MTHFR C677T poor or intermediate metabolizers—a hypothesis for which supportive data are preliminary15 and unreplicated. If genetic variants contribute to adverse (or beneficial) drug effects, one must bear in mind that most such outcomes represent complex, non-Mendelian traits influenced by multiple genes that exert small effects. How much those effects account for observed pharmacodynamic outcomes, relative to the influence of nongenetic factors, remains unknown.

A handful of specific genetic variants have been identified that pose a significant hazard if left unidentified; most notably, an elevated risk for Stevens-Johnson syndrome from carbamazepine among certain Asian groups with the HLA-B*1502 genotype (accompanied by a manufacturer’s boxed warning16). High iloperidone or valbenazine doses pose a greater cardiac risk for QT prolongation in unrecognized CYP2D6 PMs. The quest to identify genetic markers to predict clozapine-induced agranulocytosis (eg, Goldstein et al17) is ongoing.

A recent survey of ASCP members found that about one-third reported ever obtaining pharmacogenetic testing in at least 1 patient, most often to affirm what they already suspected about sensitivities to adverse effects; less than 10% felt that test results were useful to guide treatment.18 Meanwhile, a systematic review of pharmacogenetic testing in psychiatry concluded that "antidepressant pharmacogenetics have not produced any knowledge applicable to routine clinical practice yet."19(p62) Like most if not all putative biomarkers in psychiatry, pharmacogenetics holds promise and importance mainly for translational investigators. As a research tool, it could help identify possible endophenotypes to refine nosology and diagnostics. In years to come, it could help define personalized medicine with the same aspirational optimism as in oncology. Until then, psychiatrists who want to be at the "cutting edge" might glean more from following the literature rather than their patient’s genome panel.

Published online: August 8, 2017.

Potential conflicts of interest: None.

Funding/support: None.

REFERENCES

1. Grof P, Duffy A, Cavazzoni P, et al. Is response to prophylactic lithium a familial trait? J Clin Psychiatry. 2002;63(10):942-947. PubMed doi:10.4088/JCP.v63n1013

2. Franchini L, Serretti A, Gasperini M, et al. Familial concordance of fluvoxamine response as a tool for differentiating mood disorder pedigrees. J Psychiatr Res. 1998;32(5):255-259. PubMed doi:10.1016/S0022-3956(98)00004-1

3. de Leon J. Pharmacogenetic tests in psychiatry: from fear to failure to hype. J Clin Psychopharmacol. 2016;36(4):299-304. PubMed doi:10.1097/JCP.0000000000000522

4. Roses AD. Pharmacogenetics and drug development: the path to safer and more effective drugs. Nat Rev Genet. 2004;5(9):645-656. PubMed doi:10.1038/nrg1432

5. Brennan FX, Gardner KR, Lombard J, et al. A naturalistic study of the effectiveness of pharmacogenetic testing to guide treatment in psychiatric patients with mood and anxiety disorders. Prim Care Companion CNS Disord. 2015;17(2):doi:10.4088/PCC.14m017175. PubMed doi:10.4088/PCC.14m01717

6. Hall-Flavin DK, Winner JG, Allen JD, et al. Utility of integrated pharmacogenomic testing to support the treatment of major depressive disorder in a psychiatric outpatient setting. Pharmacogenet Genomics. 2013;23(10):535-548. PubMed doi:10.1097/FPC.0b013e3283649b9a

7. Hall-Flavin DK, Winner JG, Allen JD, et al. Using a pharmacogenomic algorithm to guide the treatment of depression. Transl Psychiatry. 2012;2:e172. PubMed doi:10.1038/tp.2012.99

8. 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. 2010;71(11):1482-1487. PubMed doi:10.4088/JCP.08m04773blu

9. 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):239-258. PubMed doi:10.1016/j.euroneuro.2011.10.003

10. Biernacka JM, Sangkuhl K, Jenkins G, et al. The International SSRI Pharmacogenomics Consortium (ISPC): a genome-wide association study of antidepressant treatment response. Transl Psychiatry. 2015;5:e553. PubMed doi:10.1038/tp.2015.47

11. Hicks JK, Bishop JR, Sangkuhl K, et al; Clinical Pharmacogenetics Implementation Consortium. Clinical Pharmacogenetics Implementation Consortium (CPIC) Guideline for CYP2D6 and CYP2C19 Genotypes and Dosing of Selective Serotonin Reuptake Inhibitors. Clin Pharmacol Ther. 2015;98(2):127-134. PubMed doi:10.1002/cpt.147

12. Hicks JK, Swen JJ, Thorn CF, et al; Clinical Pharmacogenetics Implementation Consortium. Clinical Pharmacogenetics Implementation Consortium guideline for CYP2D6 and CYP2C19 genotypes and dosing of tricyclic antidepressants. Clin Pharmacol Ther. 2013;93(5):402-408. PubMed doi:10.1038/clpt.2013.2

13. Rosenblat JD, Lee Y, McIntyre RS. Does pharmacogenetic testing improve clinical outcomes for major depressive disorder? a systematic review of clinical trials and cost-effectiveness studies. J Clin Psychiatry. 2017;78(6):720-729. PubMed doi:10.4088/JCP.15r10583

14. Preskorn SH, Kane CP, Lobello K, et al. Cytochrome P450 2D6 phenoconversion is common in patients being treated for depression: implications for personalized medicine. J Clin Psychiatry. 2013;74(6):614-621. PubMed doi:10.4088/JCP.12m07807

15. Jain R, Papakostas GI, Shelton RC, et al. Personalized therapy of adjunctive l-methylfolate to selective serotonin reuptake inhibitor-resistant major depressive disorder. Poster presented at the College of Psychiatric and Neurologic Pharmacists Annual Meeting; April 29-May 2, 2012; Tampa, FL.

16. Tegretol [package insert]. East Hanover, NJ: Novartis; 2017.

17. Goldstein JI, Jarskog LF, Hilliard C, et al. Clozapine-induced agranulocytosis is associated with rare HLA-DQB1 and HLA-B alleles. Nat Commun. 2014;5:4757. PubMed doi:10.1038/ncomms5757

18. Goldberg JF, Freeman MP, Balon R, et al. The American Society of Clinical Psychopharmacology survey of psychopharmacologists’ practice patterns for the treatment of mood disorders. Depress Anxiety. 2015;32(8):605-613. PubMed doi:10.1002/da.22378

19. Fabbri C, Porcelli S, Serretti A. From pharmacogenetics to pharmacogenomics: the way toward the personalization of antidepressant treatment. Can J Psychiatry. 2014;59(2):62-75. PubMed