Variability in Antidepressant Treatment of Major Depressive Disorder

Major depressive disorder (MDD) is a common problem and is associated with significant morbidity, mortality, and cost.⁷ Based on 2014 National Survey on Drug Use and Health (NSDUH) data, the prevalence of major depression in Americans was 6.6%, which led to severe impairment for 65.5%.⁸ In Veterans, the prevalence of MDD is as high as 13.5%.⁹ Antidepressants are a mainstay of treatment for MDD, including the second-generation medications citalopram, sertraline, fluoxetine, escitalopram, paroxetine, venlafaxine, duloxetine, bupropion, and mirtazapine. But rates of remission are low and variable, with approximately 11-30% of patients remitting, even after one year of treatment.^10-13 Even in the Sequenced Treatment Alternatives to Relieve Depression (STAR*D) Trial, it took more than 50 weeks to obtain a remission rate of 67%.¹⁴ In this study, all participants were initially treated with citalopram, and those who did not achieve remission were offered up to 4 separate additional clinical treatment approaches until remission was obtained. Thus, even in this large, state-of-the-art clinical trial, a large proportion of patients did not achieve remission. This is concerning as treatment that falls short of remission is associated with continued disabling symptoms, higher rates of depression relapse and recurrence, poorer work productivity, more impaired psychosocial functioning, higher levels of health care use, and potentially higher risk for suicide.¹⁰

To guide the choice of antidepressants, clinicians have typically used clinical factors thought to be associated with variable treatment response, including the character of depressive symptoms (eg, atypical, melancholic, and comorbid anxiety), family history, the possibility of drug-drug interactions, renal and hepatic function, medical and psychiatric comorbidity, nutritional status, nature of prior response to medication, and personal preference. However, few data are available to show that consideration of these factors improve remission rates or reduce adverse events. Therefore, clinicians match patients with specific antidepressants via a prolonged trial-and-error process. This delays clinical improvement and potentially increases the risks and cost of depression treatment. There remains a need for identification of additional strategies to better personalize a pharmacological approach to further improve antidepressant effectiveness. A variety of biomarkers, including pharmacogenomic approaches, have been studied to improve antidepressant prescribing.

Role of Genetic Factors in Antidepressant Treatment Variability

Genetic variation has long been explored as a potential contributor to individual differences in antidepressant treatment outcome. Pharmacogenetics/pharmacogenomics is “the use of genomic information to help predict how an individual might respond to a particular drug, to identify individuals who might experience an adverse reaction to taking a drug, or to assist in selecting the optimal dosage of a drug.”¹⁵ Antidepressant pharmacogenomic research has largely focused on genes involved in the pharmacokinetic action of medications. Pharmacokinetics is the study of the time course of medications through the body and involves absorption, presystemic elimination, drug distribution, and elimination. The most widely studied genetic differences in pharmacokinetics involve the cytochrome p450 (CYP) liver enzyme system (eg, CYP2D6, CYP2C19). CYP450 variants are believed to alter drug metabolizing enzyme (DME) function and change the normal rate of antidepressant metabolism and resulting plasma drug levels.¹⁶ CYP450 variants that reduce enzyme function may cause poor or intermediate metabolism levels, which may lead to higher than expected active drug levels or reduced prodrug levels. CYP450 variants that increase enzyme function may cause ultra-rapid metabolism, which may lead to lower than expected active drug levels or increased prodrug levels. ABCB1, which encodes the P-glycoprotein which regulates transport across the blood-brain barrier, has also been studied. Genetic differences in pharmacodynamics, which is the study of the time course and intensity of pharmacological effects of drugs, have been less-extensively studied. Studies of serotonin receptors (eg, 5-HTT, 5-HT1A, 5-HTR2A), the serotonin transporter, and dopamine receptor (DAT1) have been studied. Finally, the mediation of inflammatory response to antidepressant drugs (eg, interleukin-1β)¹⁷ may impact both pharmacokinetics and pharmacodynamics.

Understanding the extent to which genetic variation is meaningfully associated with individual differences in antidepressant treatment outcomes continues to be a challenge, however. Reviews and meta-analyses of studies that evaluated the potential interaction between antidepressant and genetic variants on remission, response, and harms have emphasized the following key limitations of the evidence base: (1) individual studies failed to control for known clinical and environmental confounders, (2) small sample sizes, and (3) heterogeneity across studies in the direction and magnitude of effect sizes and with regard to patients' clinical characteristics (eg, depression type, severity, etc), patient health behaviors such as food intake, medication regimen (eg, different types and classes, monotherapy versus combination therapy), and outcome measurement methods (eg, remission versus response, different end time points, different rating scales, etc).^16-24

Characteristics and Regulation of Antidepressant Pharmacogenomic Testing Resources

Despite the challenges of understanding the association between genetic variation and antidepressant treatment outcomes, many tests for detecting genetic variants are now available for clinical use.^25,26 Antidepressant pharmacogenomic testing resources vary widely in which gene variants or alleles are genotyped, how the testing is performed and regulated, and how they report results. At minimum, the majority of testing resources include some CYP2D6 and CYP2C19 variants.²⁶ But the testing resources differ in which and how many CYP2D6 and CYP2C19 variants are included and if and how many other genes variants or alleles are genotyped.²⁵

Regulation of pharmacogenomic tests differs depending on whether they are developed and performed by a single, central laboratory or sold to multiple labs. Some pharmacogenomic tests are sold as kits – “a group of reagents used in the processing of genetic samples that are packaged together and sold to multiple labs” – and these are regulated by the US Food and Drug Administration (FDA) as Class II devices and subject to the 510(k) marketing clearance process.^15,27 The 510(k) is “a premarket submission made to the FDA to demonstrate that the device to be marketed is at least as safe and effective, that is, substantially equivalent, to a legally marketed device.”²⁸ The Roche AmpliChip® Test ‒ which genotypes 28 CYP2D6 variants and 3 CYP2C19 variants ‒ is the first and only pharmacogenomic testing platform with 510(k) marketing clearance.²⁷ The majority of pharmacogenomic tests are Laboratory-Developed Tests (LDTs) that are developed and performed by a single, central laboratory. LDTs have historically not been assessed before being offered for clinical use and have only been regulated by the Centers for Medicare and Medicaid Services (CMS) with regard to whether laboratories' practices for using them are in compliance with the Clinical Laboratory Improvement Amendments of 1988 (CLIA).^15,20,25 However, in 2010, the FDA proposed initiating premarket evaluation of pharmacogenomic LDTs' analytical and clinical validity as well.²⁹ In October 2014, FDA released for public comment a Draft Regulatory Guidance document and 9-year implementation time-frame.^15,30,31 Main concerns about the FDA LDT oversight proposal include that the anticipated added cost and time involved in compliance with new regulations could potentially impede advances and negatively impact patient care.^32,33

Format for reporting results also varies widely across pharmacogenomic testing. Sources of variation include (1) whether they are drug-focused or gene-focused,²⁶ (2) how much detail is provided about therapeutic implications, categorization of interaction, and clinical impact, and (3) whether or not consultation with a professional genetic counselor and/or a pharmacist is available to help treating clinician interpret the results. For example, the interpretive report for the GeneSight test organizes the results by gene-drug interaction category and clinical impact and highlights these features at the top of the report. For gene-drug interaction category, the GeneSight report stratifies specific antidepressant and antipsychotic medications into 3 color-coded categories: (1) green “use as directed” bin to indicate little or no gene-drug interaction, (2) yellow “use with caution” bin to indicate moderate gene-drug interaction, and (3) red “use with caution with more frequent monitoring” bin to indicate more severe gene-drug interaction.³ Footnotes supply the details of the therapeutic implications of the gene-drug interactions for the yellow and red categories and the gene results are in a table at the bottom of the report. In contrast, the Genecept Assay interpretive report organizes the results by gene result and provides much more detailed information about therapeutic implications for each gene result. Gene-drug interaction is also stratified by 3 categories: (1) a green check mark for “no known gene-drug interaction”, (2) a blue lightbulb for “therapeutic options”, and (3) an orange caution symbol for “use caution” with related therapies to indicate medications that may require a dose adjustment or have a higher risk of side effects of inefficacy. Compared to the GeneSight report, the Genecept report guidance on clinical impact is more general, referring to classes of antidepressants and antipsychotics, rather than listing specific antidepressants.³⁴ It is unclear if and how such differences in pharmacogenomic testing results format may affect the accuracy of their interpretation and use.²⁵

Guidelines and Labeling for Use of Pharmacogenomics in Antidepressant Treatment

With the increased availability of pharmacogenomic testing, its clinical use is increasing. This has led to development of antidepressant FDA product labeling and clinical practice guidelines from various professional genetic societies on the use of pharmacogenomic testing.^25,27 A 2014 review by Drozda et al provides a detailed analysis of information contained in FDA labeling of antidepressant drugs and consensus guidelines.²⁵ At the time of the Drozda review, the number of neuropsychiatric medications with FDA labeling listing a CYP450 variant metabolizer status as an important biomarker was 27 for CYP2D6 and 3 for CYP2C19.

Guidance on using existing CYP2D6 and/or CYP2C19 genotyping results to inform antidepressant selection and dosing is also provided by a few professional genetic societies (see appendix A in supplemental materials).^25,27,35-38 These guidelines vary in when they were released, the antidepressant medications they address, their development methods, and how they assess the level of evidence and/or classification of recommendations. The most up-to-date genotyping-based dosing guidelines (late 2015 to early 2016) are from the Clinical Pharmacogenetics Implementation Consortium (CPIC) and available on the Pharmacogenomics Knowledge Base (PharmGKB website).³⁷ CPIC guidelines provide metabolism implications and therapeutic recommendations for various antidepressant drugs and generally classify the recommendations as moderate to strong.

When Should Antidepressant Pharmacogenomic Testing Be Performed?

Despite availability of guidance on how to use already available pharmacogenomic testing results, uncertainty remains on when and for whom pharmacogenomic testing should be performed. Professional genetic society guidelines vary on whether and how they address when to perform pharmacogenomic testing to inform antidepressant treatment.^25,27,35-40 The earliest guideline, released in 2007 by the Evaluation of Genomic Applications in Practice and Prevention (EGAPP) Working Group,³⁸ which was based on an evidence review¹⁶ by the Agency for Healthcare Research and Quality's Duke Evidence-based Practice Center, concludes that CYP2D6 testing for SSRI treatment for depression was not recommended at that time. More recent guidelines have identified some “important indications” for combining genotyping with therapeutic drug monitoring³⁹ and have suggested “clinicians consider FDA drug labeling recommendations about genetic testing for some specific drugs,” but with the caveat that evidence remains inconclusive about possible clinical utility of CYP450 in psychiatry.⁴⁰ Guidelines specific to the treatment of patients with depressive disorders either do not reference pharmacogenomic testing at all,^41,42 or mention it only briefly as an area for future research.^43-45

The conservative nature of existing formal guidance about the clinical utility of pharmacogenomic testing is likely due at least in part to the fact that it was developed prior to the emergence of the first of several recent studies that have compared the use of pharmacogenomic data to guide antidepressant treatment selection for depressive disorders to treatment as usual. Although various articles^{19,20,26,27,46} have briefly reviewed findings from some of the recent studies, they have not formally critically appraised the complete body of available evidence. The purpose of this report is to conduct an evidence brief on the comparative effectiveness, harms, and cost-effectiveness of pharmacogenomics-guided antidepressant treatment versus usual care for major depressive disorder, to inform an April 2016 ORD planning meeting for development of a clinical study to implement precision medicine in mental health (PMH).

Polygene Panel of CYP2D6, CYPC19, CYP1A2, SLC6A4, HTR2A (GeneSight)

Compared to usual care, GeneSight-guided care has not yet been shown to significantly improve remission or response in a randomized trial.^3,4,46 The only completed double-blind, randomized trial provided low-strength evidence that 10 weeks of GeneSight-guided care does not significantly improve remission rates (HAM-D ≤ 7: 20% vs 8%; RR 2.40, 95% CI 0.51 to 11.21) or response (≥ 50% HAM-D improvement: 36% vs 21%; RR 2.14, 95% CI 0.56 to 7.69).³ This trial was conducted in the outpatient clinics of Pine Rest Christian Mental Health Services in Grand Rapids, Michigan and involved 51 patients with major depressive disorder, with a mean baseline HAM-D of 21, who had failed a mean of 4 previous psychiatric medication trials. The applicability of these findings to Veterans is unclear, however, as the mean age was 49 years, only 20% of patients were male, and the types of antidepressants medications used were not reported. Another weakness of this study is that it did not report on adverse effects.

Results from 2 previous open-label nonrandomized studies^53,54 were less informative than the findings of the RCT.³ These studies were also short-term (8 weeks), did not evaluate adverse effects, and included mostly females in their mid-forties with unknown comorbidities, and had more methodological limitations. Although these open-label nonrandomized studies found that GeneSight-guided care significantly improved response, defined as a 50% or greater decreased from baseline in HAM-D17 score (ESP-pooled: 40% vs 23%; RR 1.73, 95% CI 1.09 to 2.73), there is a high likelihood that this was due to increased expectations in the GeneSight-guided group based on their knowledge that their medication selection was being guided by DNA testing and, in the case of the Hall-Flavin 2013 study, fewer previously failed psychiatric medication trials in the guided group (4.7 vs 3.6; P = 0.021).⁵³ Also, because groups were not matched on psychiatric and medical comorbidities, concomitant medications, medication adherence, and health and lifestyle characteristics, important differences could exist that could have confounded the effects of the GeneSight guiding. When data from the double-blind RCT³ and these 2 open-label nonrandomized studies^53,54 were combined in a meta-analysis,⁴ the improved response with GeneSight-guided care reached statistical significance (RR 1.71; 95% CI 1.17 to 2.49). However, the described specific weaknesses of these open-label, nonrandomized studies also seriously limit the validity of the meta-analysis.

We identified 3 ongoing clinical trials assessing the efficacy of GeneSight-guided management of depressive disorders (NCT02189057, NCT02466477, NCT02109939). All studies are double-blind RCTs that are expected to address some gaps in the existing evidence by increasing precision with larger sample sizes and providing longer follow-up (see appendix E in supplemental materials).⁴⁶ The studies are expected to be completed between 2015 and 2018.

Polygene Panel of ABCB1, ABCC1, CYP2C19, CYP2D6, UGT1A1 (CNSDose, Not Commercially Available)

Compared to usual care, 12 weeks of CNSDose-guided antidepressant treatment improves remission (HAM-D ≤ 7; 72% vs 28%; RR 2.52, 95% CI 1.71 to 3.73; moderate SOE) and reduces the proportion of patients taking sick leave (usual care = 15% vs guided = 4%; RR 1.13, 95% CI 1.01 to 1.25; low SOE) and intolerability (having an event where patient needed to reduce the dose or stop their antidepressant: usual care = 15% vs guided = 4%; RR 1.13, 95% CI 1.01 to 1.25; low SOE) in patients with baseline HAM-D score of 25 taking various second-generation antidepressants.¹ Supporting evidence comes from one randomized trial of 148 adults with MDD conducted in Australia. The main strength of this study is its high internal validity due to its use of robust methodology. However, the main weakness of this study is its limited applicability to the VA population. This study excluded smokers and patients with other active comorbid psychiatric disorders and ended up with a population of mostly employed females in their early forties. Average number of MDD episodes was 2, with average duration of 8.55 months, but number of previously failed antidepressant trials was not reported, nor was current number of antidepressant medication or other types of concomitant treatment. A replication study of CNSDose has been completed and its findings have been submitted for peer review (author correspondence, Australian and New Zealand Clinical Trials Registry ID # ACTRN12613001135707). Its results are expected to be more applicable to the typical VA MDD population as its eligibility criteria were more inclusive, allowing psychiatric comorbidities such as PTSD and smoking.

ABCB1 Genotyping (Codes for P-glycoprotein, Not Commercially Available)

Compared to usual care treatment, 5 weeks of ABCB1 genotyping-guided antidepressant treatment improves remission (improved remission (HAM-D < 10): 83.6% vs 62.1%; X² (1) = 6.596, P = 0.005; ESP calculated-RR 1.33; CI 1.06 to 1.72; low SOE). Response (50% reduction in HAM-D), quality of life, functional status, and side effects and tolerability were not reported. Supporting evidence comes from one observational pilot study of 116 adults with MDD and bipolar disorder conducted in Germany.² The study has several weaknesses that limit its applicability to Veterans. First, the HAM-D remission cut-off ( < 10) is not considered complete remission according to typical HAM-D scoring methods, so remission may be overestimated in this study. Also, all treatment occurred while patients were hospitalized, and clinicians were given antidepressant plasma levels weekly in addition to the ABCB1 genotyping. It is unclear how antidepressant plasma monitoring impacted care in this study and it has limited applicability to VA care as antidepressant plasma levels are not routinely monitored in the VA. The mean age in the sample (46.59 to 48.53) was somewhat younger than the mean age of US Veterans currently enrolled in care with depression (57.2) and some patients had bipolar disorder, which is a relative contraindication to treatment with antidepressants. The study population was predominantly female, with an average number of depressive episodes of 4.24, duration of current episode of 25 weeks, and 1.3 antidepressant trials during recent admission in the experimental group, compared to 2.43 depressive episodes, 39.2 weeks for the current episode of depression, and 0.98 antidepressant trials during the recent admission for the comparison group. These were not statistically significantly different.

CYP2D6, CYP2C19, SLC6A4, CACNA1C, DRD2, COMT, MTHFR (Genecept)

There is insufficient evidence to draw conclusions about the comparative benefits and harms of Genecept-guided care versus usual care in patients with depressive disorders. This is because we found no completed studies that evaluated this comparison. In a single-group before-after study of 685 adults with mood and anxiety disorders, 38% achieved remission (Quick Inventory of Depression Symptoms – QIDS-SR < 5) and 39% achieved response (≥ 50% reduction in QIDS-SR score) after 3 months of Genecept-guided care.⁵⁵ However, this type of study design – lacking an usual care control group – generally does not provide reliable evidence of the specific effects of an intervention as distinct from what may have naturally occurred over time regardless of the intervention.

We identified one ongoing double-blind randomized controlled trial of 8 weeks of Genecept-guided versus usual care in adults with MDD that expected to assess response, remission, and safety outcomes (NCT02634177). This study is expected to complete in October of 2016 and should provide more relevant and higher-quality evidence with which to evaluate the clinical utility of Genecept.

CYP2D6 AND CYP2C19

There is insufficient evidence to draw conclusions about the clinical utility of CYP2D6 and CYP2C19 genotyping for MDD. The only study we found was a single-group before-after study of less than or equal to 100 adults receiving antipsychotics/antidepressants for unspecified reasons.⁵⁶ Twelve weeks after genotypic information was considered in their drug therapy, “several” of the patients reported they were “much improved.” But because this study lacks an usual care control group and its data was not presented in a scientifically rigorous manner – lacking detail on patient characteristics, completeness of the data, etcetera – it does not provide clearly relevant or reliable evidence on the clinical utility of CYP2D6 and CYP2C19 genotyping for MDD.

Cohort Studies

No study has prospectively compared the cost-effectiveness of pharmacogenomics-guided care versus usual care specifically in patients with depressive disorders. We identified one prospective cohort study⁶³ and one retrospective cohort study⁶² that compared actual healthcare costs between pharmacogenomics-guided versus usual care, but their applicability to patients with depressive disorders is likely very limited. This is because they both focused on use of psychotropic medications for any diagnosis. In the prospective study, only 29% of the neuropsychiatric medication use was for CNS disorders and among that only 14% was for major depression.⁶³ In the retrospective study, only 29% of patients had depressive disorders.⁶² Neither study performed subgroup analyses of cost in the depression subgroups. Other main limitations of both studies are that (1) they lacked measurement of antidepressant medication clinical benefits and harms and (2) there is a high risk of residual confounding as patients were matched only on demographics and the presence of a CNS diagnosis, which does not account for depression severity or subtype. The prospective study had additional limitations of (1) lacking measurement of other outpatient and inpatient healthcare costs and (2) not accounting for comorbidity. In the prospective study, compared to usual care, GeneSight-guided care (CYP2D6, CYPC19, CYP1A2, SLC6A4, HTR2A) reduced total medication costs by $1035.60 over one year (P=0.007).⁶³ But, the majority of savings related to non-CNS medications ($714.24; 69%), with only $321.36 (31%) annual savings for CNS medications. In the retrospective study, Genecept-guided care reduced overall pharmacy and outpatient care-related costs by 9.5% over 4 months, or $562.⁶²

Modeling Studies

Table 2 below summarizes the characteristics and findings from 6 modeling studies.^16,57-61 Pharmacogenomic test type varied across studies, including the GeneSight polygene panel (CYP2D6, CYPC19, CYP1A2, SLC6A4, HTR2A),⁵⁷ 5-HTTLPR,^58,59 HTR2A,⁶⁰ and CYP450 polymorphisms.^16,61 A key limitation of all modeling studies is their reliance on inferences of likely outcomes rather than directly linking interventions to actual observed outcomes. Also, the findings of these studies have questionable relevance to the VA setting, as in some cases the treatment regimens were poorly characterized and the base-case parameters used in the models may have been more favorable than the average characteristics of Veterans with depression.

Table 2

Summary of Cost-effectiveness Findings from Modeling Studies.

Five of 6 modeling studies found limited cost-effectiveness, only under certain circumstances. The exception was that, compared to treatment as usual, the GeneSight polygene panel was found to have a high probability (94.5%) of being cost-effective at the Willingness-To-Pay (WTP) threshold of $50,000/Quality-Adjusted Life-Years (QALYs) (Table 2).⁵⁷ This finding was robust to variation in input parameters (95% confidence intervals), with 74.7% of 10,000 simulations showing that Genesight-guided care was more effective and more cost-saving than treatment as usual. This study used a Markov state-transition analysis to evaluate QALYs, cumulative direct (ie, depression and non-depression drugs, inpatient and outpatient physician treatment, psychotherapy, and other costs) and indirect (ie, productivity and absenteeism) costs, and cost per QALY gained. Relative benefit ratio of GeneSight on response rate, catch-up year, and starting age were the top 3 input parameters that had the greatest influence on incremental costs, with ranges that included cost-increasing scenarios. First, for the test effect on response rate, its reliability is important because sensitivity analyses ranked it as having the largest effect on incremental costs. In this analysis, the test effect on response rate was estimated at 1.71 based on 3 clinical studies of GeneSight. However, as discussed above, these studies involved populations that were younger and included more females than in the Veteran population and for which psychiatric and medical comorbidities, concomitant medications, medication adherence, and health and lifestyle characteristics were not well-characterized.^3,53,54 But at least one of the studies³ excluded individuals with comorbid substance abuse or dependence. Therefore, it is unclear how applicable that 1.71 relative benefit of GeneSight on response is to a VA population. The Probabalistic Sensitivity Analysis of the lower and upper bound of the 95% confidence interval around the test effect (1.17 to 2.49) includes both increased total cost of $815 to an increased savings of $8543. Because we don't know what the test effect on response rate actually is in Veterans, we can't determine the cost implications within this continuum. Second, this analysis included the time-related parameter of “catch-up year,” that assumed that it would take 3 years for the usual care group's response rate to “catch up” to the Genesight-guided group. Sensitivity analyses found that if the usual care group caught up faster, within one year, the use of the GeneSight test to guide care became cost-increasing ($491 higher). Finally, compared to the mean age of getting tested at 44 years of age used in the model, the higher mean age of 57 years in Veterans with depression⁹ may be expected to reduce cost-effectiveness. Age had the third-highest impact on the incremental costs. The Probabalistic Sensitivity Analysis of the lower and upper bound of the 95% confidence interval around age (18 to 82 years) includes both increased total cost savings of $3835 to a higher incremental cost of $2500, and we don't know where along that continuum the mean Veteran age of 57 falls. Although each of these cases led to cost-increasing scenarios, Incremental Cost-Effectiveness Ratio (ICER) values were noted to remain below WTP threshold of $50,000/QALYs. However, it was unclear how the combination of lower response benefits, lower catch-up year, and higher age – as potentially characteristic of Veterans – may have affected ICER values. On the other hand, compared to the $2500 cost of the GeneSight test, the likely lower cost of pharmacogenomic testing in the VA – particularly if a process is adapted for returning results to Veterans from their already available Million Veteran Program genetic analysis – may be expected to improve cost-effectiveness. For these reasons, there remains a need for further cost-effectiveness analysis in more representative samples to better determine the true benefits in Veterans with depression.

MDD GUIDELINES

GENETIC TESTING GUIDELINES

Database: Ovid MEDLINE

Database: Ovid MEDLINE(R) <1946 to February Week 4 2016>, Ovid MEDLINE(R) In-Process & Other Non-Indexed Citations <March 03, 2016>

Search Strategy:

1. pharmacogenetic.mp. or exp Pharmacogenetics/ (12208)
2. pharmacogenomic.mp. (1946)
3. 1 or 2 (13046)
4. antidepressant.mp. or exp Antidepressive Agents/ (137858)
5. exp Depression/ or depression.mp. (302917)
6. major depressive disorder.mp. or exp Depressive Disorder, Major/ (28690)
7. 5 or 6 (311056)
8. 3 and 4 and 7 (335)
9. limit 8 to (english language and humans) (296)

Database: Ovid MEDLINE(R) without Revisions <1996 to February Week 4 2016>

Search Strategy:

1. pharmacogenomic.mp. or Pharmacogenetics/ (9631)
2. systematic review.mp. (53368)
3. Depression/ or depression.mp. (178021)
4. meta analysis.mp. or Meta-Analysis/ (83682)
5. 2 or 4 (115800)
6. 1 and 3 and 5 (7)

Database: Cochrane Central Register of Controlled Trials (February 2016)

1. pharmacogenetic.mp. or exp Pharmacogenetics/ (544)
2. pharmacogenomic.mp. (122)
3. 1 or 2 (637)
4. antidepressant.mp. or exp Antidepressive Agents/ (13539)
5. exp Depression/ or depression.mp. (35805)
6. major depressive disorder.mp. or exp Depressive Disorder, Major/ (4502)
7. 5 or 6 (36737)
8. 3 and 4 and 7 (51)
9. limit 8 to (english language and humans) [Limit not valid; records were retained] (46)

Database: PsychINFO (March 2016)

1. pharmacogenetic.mp. or exp Pharmacogenetics/ (694)
2. pharmacogenomic.mp. (140)
3. 1 or 2 (801)
4. antidepressant.mp. or exp Antidepressive Agents/ (28844)
5. exp Depression/ or depression.mp. (250924)
6. major depressive disorder.mp. or exp Depressive Disorder, Major/ (14871)
7. 5 or 6 (252352)
8. 3 and 4 and 7 (114)
9. limit 8 to (english language and humans) [Limit not valid in PsycINFO; records were retained] (108)

DATA ABSTRACTION OF INCLUDED PRIMARY STUDIES

QUALITY ASSESSMENT OF INCLUDED PRIMARY STUDIES

STRENGTH OF EVIDENCE FOR INCLUDED STUDIES

Download PDF (56K)

REFERENCES

1.

Department of Veterans Affairs and Department of Defense: The Management of MDD Working Group. VA/DoD Clinical Practice Guideline for the Management of Major Depressive Disorder. 2016.

2.

American Psychiatric Association. Practice guideline for the treatment of patients with major depressive disorder. 2010.

3.

Bauer M, Whybrow PC, Angst J, Versiani M, Möller H-J. World Federation of Societies of Biological Psychiatry (WFSBP) Guidelines for Biological Treatment of Unipolar Depressive Disorders, Part 2: Maintenance treatment of major depressive disorder - update 2015. The World Journal of Biological Psychiatry. 2015;16:76–95. [PubMed: 25677972]

4.

National Institute for Health and Care Excellence. Depression in adults: recognition and management. 2009. [PubMed: 31990491]

5.

Bauer M, Pfennig A, Severus E, Whybrow PC, Angst J, Möller H-J. World Federation of Societies of Biological Psychiatry (WFSBP) guidelines for biological treatment of unipolar depressive disorders, part 1: update 2013 on the acute and continuation treatment of unipolar depressive disorders. The World Journal of Biological Psychiatry. 2013;14(5):334–385. [PubMed: 23879318]

6.

Hicks J, Swen J, Thorn C, et al. Clinical Pharmacogenetics Implementation Consortium guideline for CYP2D6 and CYP2C19 genotypes and dosing of tricyclic antidepressants. Clin Pharmacol Ther. 2013;93(5):402–408. [PMC free article: PMC3689226] [PubMed: 23486447]

7.

National Academy of Clinical Biochemistry. Laboratory Analysis and Application of Pharmacogenetics to Clinical Practice. 2009.

8.

Hiemke C, Baumann P, Bergemann N, et al. AGNP consensus guidelines for therapeutic drug monitoring in psychiatry: update 2011. Pharmacopsychiatry. 2011;44(6):195–235. [PubMed: 21969060]

9.

International Society of Psychiatric Genetics. Genetic Testing Statement: Genetic Testing and Psychiatric Disorders. 2014. [March, 2016]. http://ispg​.net/genetic-testing-statement/

10.

de Leon J, Armstrong SC, Cozza KL. Clinical guidelines for psychiatrists for the use of pharmacogenetic testing for CYP450 2D6 and CYP450 2C19. Psychosomatics. 2006;47(1):75–85. [PubMed: 16384813]

11.

Drozda K, Müller DJ, Bishop JR. Pharmacogenomic testing for neuropsychiatric drugs: current status of drug labeling, guidelines for using genetic information, and test options. Pharmacotherapy: The Journal of Human Pharmacology and Drug Therapy. 2014;34(2):166–184. [PMC free article: PMC3939793] [PubMed: 24523097]

12.

PharmGKB. Dosing Guidelines - Clinical Pharmacogenetics Implementation Consortium. 2015-2016. [March, 2016]. https://www​.pharmgkb​.org/view/dosing-guidelines​.do?source=CPIC .

13.

Singh AB. Improved Antidepressant Remission in Major Depression via a Pharmacokinetic Pathway Polygene Pharmacogenetic Report. Clinical Psychopharmacology and Neuroscience. 2015;13(2):150–156. [PMC free article: PMC4540033] [PubMed: 26243841]

14.

Winner JG, Carhart JM, Altar CA, Allen JD, Dechairo BM. A prospective, randomized, double-blind study assessing the clinical impact of integrated pharmacogenomic testing for major depressive disorder. Discov. 2013 Nov;16(89):219–227. [PubMed: 24229738]

15.

Breitenstein B, Scheuer S, Pfister H, et al. The clinical application of ABCB1 genotyping in antidepressant treatment: a pilot study. CNS Spectr. 2014 Apr;19(2):165–175. [PubMed: 23880209]

16.

Hall-Flavin DK, Winner JG, Allen JD, et al. Using a pharmacogenomic algorithm to guide the treatment of depression. Transl Psychiatry. 2012;2:e172. [PMC free article: PMC3565829] [PubMed: 23047243]

17.

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 Oct;23(10):535–548. [PubMed: 24018772]

18.

Fagerness J, Fonseca E, Hess GP, et al. Pharmacogenetic-guided psychiatric intervention associated with increased adherence and cost savings. The American journal of managed care. 2014;20(5):e146–156. [PubMed: 25326929]

19.

Winner JG, Carhart JM, Altar CA, et al. Combinatorial pharmacogenomic guidance for psychiatric medications reduces overall pharmacy costs in a 1 year prospective evaluation. Current medical research and opinion. 2015 [PubMed: 26086890]