MolDX 11/2015 Article on CYP Evidence and Rationale

https://www.cms.gov/medicare-coverage-database/details/article-details.aspx?articleId=54748&ver=3&ContrId=353&ContrVer=1&LCDId=36311&name=353*1&UpdatePeriod=654&IsPopup=y&

• CMS Article A54748
• Released 11/30/2015
• Effective 10/15/2015

Article Text:

MolDX recognizes the abundant literature available regarding the pharmacokinetic and pharmacodynamic CYP association with drugs, and that experts, including the FDA, mention and recommend CYP genotyping. However, the MolDX LCD CYP2D6, CYP2C19, CYP2C9 and VKORC1 Genetic Testing limits coverage to specific populations where the evidence supports genotyping will lead to proven clinical management changes and result in improved patient outcomes.

Although MolDX does not dispute the associations between certain genotypes and drug responses or reactions, there is very limited guidance in the published literature regarding whom to test and when. The majority of guidelines reference dosing recommendations without commentary on when the genetic testing is appropriate. In fact, the most recent CPIC guideline addressing CYP2D6 and CYP2C19 and SSRIs states that “clinical variables that may influence SSRI therapy as well as genotyping cost-effectiveness are beyond the scope of this document” [Hicks 2015].

The guideline continues to state there are many influencers, including drug interactions and clinical factors, that should be taken into account when prescribing and dosing medications. For this reason clinical utility of pharmacogenomic testing is difficult to establish. The authors are careful to recommend dosing based on existing CYP2D6 and/or CYP2C19 genotype results, rather than recommend pre-prescribing genetic testing. 

MolDX agrees that, if available, this information may be considered as one component in the decision regarding which medication to use or which dose to prescribe. However; that does not establish clinical utility for the test in the Medicare population.

In the consideration of prospective pre-medication genotyping across the entire Medicare population, further studies addressing clinical practicality, logistics, and cost-effectiveness are necessary to establish clinical utility. The MolDX LCD addresses the clinical indications for genotyping. The policy does not define what to do with that information once it is known. In addition, the FDA labels recommend genotypes prior to use, suggesting that this information may be useful but not necessary. 

Medicare is required to use public health dollars towards interventions that are proven to be effective. At this time there is limited data to support the clinical utility of CYP2D6 and CYP2C19 genotyping beyond the indications outlined in the LCD.


The following is a comprehensive review of CYP genotyping evidence. Each section identifies cited literature to support coverage and the reason MolDX supports the citation or the reason MolDX did not find the evidence compelling.

CYP2D6
The evidence supports CYP2D6 testing for patients with Gaucher disease eligible for treatment with eliglustat (Cerdelga), and for patients being treated with pimozide at doses over 0.05mg/kg/day in children or above 4 mg/day in adults. In poor CYP2D6 metabolizers, pimozide doses should not exceed 0.05mg/kg/day in children or 4 mg/day in adults and doses should not be increased earlier than 14 days. Because it will impact medical management in this defined population, CYP2D6 genotyping is medically necessary for persons being treated with pimozide at doses over 0.05mg/kg/day in children or above 4 mg/day in adults.

CYP2C19
Although the literature indicates allele frequency variation among various ethnic groups, genetic testing must have proven clinical utility in order to create a health disparity or harm. Coverage is provided for a specific indication [patients with ACS undergoing PCI who are initiating or reinitiating Clopidogrel (Plavix) therapy], and is not restricted on the basis of race or ethnic background. The initiation of CYP2C19 genotyping is not considered to have clinical utility outside the given indication at this time.

There are discordant results in the literature regarding the effect of the CYP2C19*17 allele, possibly due in part to the linkage disequilibrium that exists between the *17 and *2 alleles [Scott et al 2013]. In addition, all participants in the Cresci et al (2014) study cited had an acute coronary infarction (ACI). Many of these individuals would have been considered for treatment with clopidogrel, and therefore would likely meet coverage for CYP2C19 genotyping under the current recommended coverage criteria [patients with ACS undergoing PCI who are initiating or reinitiating Clopidogrel (Plavix) therapy].

There is ongoing debate regarding the effect of CYP2C19 genotype on the protection from major adverse cardiovascular events. To date, many of the published studies supporting the clinical use of CYP2C19 genotyping when considering use of clopidogrel have been retrospective meta- analyses and not large, prospective studies examining the association between CYP2C19 genotype and clinical efficacy of clopidogrel. At this time, there have been a number of studies published in this area, but the evidence for guiding clopidogrel dosing and treatment on the basis of CYP2C19 genotype is inconclusive [Samer et al 2013].

The 2013 update of CPIC’s Guidelines for CYP2C19 genotype and clopidogrel therapy does not recommend widespread adoption of CYP2C19-guided antiplatelet therapy, and the FDA boxed warning does not require genetic testing prior to initiation of clopidogrel therapy [Scott et al 2013]. Current data does not support the use of CYP2C19 genotyping to guide treatment in clinical scenarios other than ASC patients undergoing PCI [Scott et al 2013].

CYP2C19 and antidepressants
In the LCD referenced [LCD 35437], GeneSight® testing is only to be reimbursed in certain clinical scenarios as a gene panel. Specifically: “GeneSight® testing may only be ordered by licensed psychiatrists contemplating an alteration in neuropsychiatric medication for patients diagnosed with major depressive disorder (MDD)(in accordance with DSM IV/V criteria) who are suffering with refractory moderate to severe depression (as defined by the 17-item Hamilton Rating Scale for Depression (HAM-D17) score of 14 or greater) after at least one prior neuropsychiatric medication failure.”

Coverage of this specific gene panel for this specific clinical scenario does not support broader coverage for singular CYP2C19 testing. The literature cited in the LCD references prospective trials that support the use of the GeneSight® panel as a whole. Specific effects of CYP2C19 alone cannot be extracted from this data.

Sertraline
The paper referenced in support of CYP2C19 genotype-based dosage recommendations for sertraline is a single case report [Lorenzini 2012]. There is insufficient data to warrant CYP2C19 genotyping to determine health outcomes or adverse drug reactions in treatment with this antidepressant at this time. The CPIC suggests initiating therapy with the starting dose recommended for CYP2C19 extensive and intermediate metabolizers for which a genotype is not necessary [Hicks 2015]. This recommendation is based on data outside of randomized trials. The remaining CPIC recommendations did not have enough high quality evidence to make a strong recommendation for dose modification based on genotype [Hicks 2015].

Imipramine/desipramine
The paper referenced in support of CYP2C19 genotype-based dosage recommendations for these medications is a single case report [Lorenzini 2012].

Citalopram and escitalopram
CYP2C19 PMs and IMs have been reported to show differences in utilization of these medications. “However, this does not appear to result in differences in side effects, as there is a large therapeutic range. Therefore, a dose adjustment is not considered necessary” [Samer et al 2013]. The CPIC suggests initiating therapy with the starting dose recommended for CYP2C19 extensive and intermediate metabolizers, for which a genotype is not necessary [Hicks 2015]. This recommendation is based on data outside of randomized trials. The remaining CPIC recommendations did not have enough high quality evidence to make a strong recommendation for dose modification based on genotype [Hicks 2015].

2C19 and Proton Pump Inhibitors (“PPIs”)
Although Ogawa (2010) addressed the drug-drug interaction profiles of proton pump inhibitors, relying on the pharmacokinetic properties of PPIs, a more recent review of clopidogrel and PPI interactions by Juel (2014), states that “the pharmacodynamic studies support an interaction, whereas the clinical evidence, which is mainly based on nonrandomized, observational studies and secondary analyses of randomized trials, is conflicting.”

At this time, there is insufficient data to warrant CYP2C19 genotyping to determine health outcomes or adverse drug reactions in treatment with proton pump inhibitors. There are no professional society guidelines or recommendations available to address when CYP2C19 genotyping should be performed in the context of PPI use.

CYP2C19 Summary
Additional research is needed regarding the association between CYP2C219 genotype and medications/indications. The current available literature is retrospective and often includes small sample sizes. There is a potential that CYP2C19 genotype-based dosing could be utilized for indications beyond the current recommendations in the future, if large prospective studies reveal a true benefit of genotyping in other clinical scenarios or specific populations.

At this time, there have been a number of studies published in this area, but the evidence for guiding clopidogrel dosing and treatment on the basis of CYP2C19 genotype is incomplete [Samer et al 2013]. 

CYP2D6 
CYP2D6 variants and the prevalence of PMs and UMs varies across populations. According to L Lerena et al [2014], the CYP2D6*4 allele frequency is higher in Caucasians, CYP2D6*10 in East Asians, CYP2D6*41 and duplication/multiplication of active alleles in Middle Easterners, CYP2D6*17 in Black Africans and CYP2D6*29 in African Americans, than in other ethnic groups. Africa and Asia are under- represented in this area relative to the total number of their inhabitants, so further studies in these regions are warranted.

Current literature highlights the importance of carrying out pharmacogenomic research in American Indian/Alaskan Native populations and show that extrapolation from other populations is not appropriate [Fohner et al 2013]. Overall, genotype predicted poor metabolizers (gPMs) and metabolic predicted poor metabolizers (mPMs) are more frequent among Caucasians, and genotype predicted ultra-metabolizers (gUMs) are more frequent among Middle Easterners and Ethiopians [L Lerena et al 2014]. Cespedos-Garro et al [2014] report that the frequencies of UMs and PMs for CYP2D6 vary widely across the mestizo, Amerindian and Afro-Caribbean Costa Rican populations.

Future research in this population should be oriented to identify new CYP2D6 variants through sequencing methods, as well as to determine CYP2D6 phenotype, in order to establish the phenotype- genotype relationship. Finally, further studies involving genetic markers of ancestry are needed in the Costa Rican population. [Céspedes-Garro et al, 2014].

This variation across populations and racial backgrounds of CYP2D6 variants is not sufficient evidence that CYP2D6 status is clinically useful or proven for the drugs that it metabolizes outside of the indications in the LCD. There are currently no clinical recommendations for carrying out pharmacogenomic CYP2D6 testing in particular populations over others.

Individual CYP2D6 genotype in adults was noted to be associated with negative response to codeine. The efficacy and toxicity, including severe or life-threatening toxicity after normal doses of codeine has been linked to an individual’s CYP2D6 genotype [Crews 2014]. According to Prows et al [2014], neither obstructive apnea nor predicted CYP2D6 phenotype was associated with sedation risk. Their results provide evidence that multiple factors are associated with codeine-related ADRs and support the FDA recommendation to avoid codeine's routine use following tonsillectomy in children.

The revised 2014 CPIC guidelines were reviewed and the recommendations are based on robust clinical data to advise the interpretation of CYP2D6 genotype [Crews 2015]. However these guidelines lack guidance on the populations that benefit from testing. Additionally, FDA warnings relate to the highest risk patients (breastfed neonates and children) and recommend against the use of codeine regardless of CYP2D6 genotype. Therefore, it still remains unclear which specific populations and in which potential scenarios there would truly be benefit from CYP2D6 recommendation to include codeine use as an indication for CYP2D6 genotyping is not yet proven for large-scale population application, regardless of age.

CYP2D6 and prodrugs: Tramadol
Lassen et al [2015] carried out a systematic review of the current knowledge on all possible genetic factors that might influence the metabolism or clinical efficacy of tramadol. Only the effect of CYP2D6 polymorphisms on the metabolism of tramadol and the consequent effect on pain relief has been thoroughly studied and sufficiently established as clinically relevant [Lassen et al 2015]. It is shown in the literature that CYP2D6 ultra- rapid metabolizers are more likely to experience the adverse effects of codeine and tramadol [Lassen et al 2015]. In addition, opioid analgesics that do not rely on CYP2D6 for therapeutic activity, such as morphine and hydromorphone, may therefore, be a better alternative to codeine and tramadol, with the limitation that these drugs have their own set of adverse reactions [Haufroid and Hantson, 2015].

Although CYP2D6 genotype has been shown to be clinically relevant to tramadol metabolism, research is ongoing regarding which population of patients will benefit from genotyping. Current literature is conflicting about the use of CYP2D6 testing in the clinical prescribing of tramadol. For instance, Whirl-Carrillo et al [2012] recommends that CYP2D6 UMs be treated with a dose decrease of 30%. Others report that clinical translation of opioid genetics is premature because many important pain and addiction phenotype factors contribute [Somogyi et al 2015]. Finally, the FDA includes pharmacogenomic information in their labeling for tramadol in relation to poor metabolizers only; no comment on ultra-metabolizers. 

Codeine
Per Crews et al [2014], the FDA warnings relate to the highest risk patients (breastfed neonates and children) and recommend against the use of codeine regardless of CYP2D6 genotype, suggesting lack of clinical utility of CYP2D6 testing. However, the FDA’s Table of Pharmacogenomic Biomarkers in Drug Labeling, states that codeine should only be used in specific populations and should be avoided in a patient identified as a CYP2D6 ultrarapid metabolizer (i.e., a CYP2D6 activity score of >2.0); the choice of an alternative analgesic should be made to avoid the risk of severe toxicity associated with a “normal” dose of codeine [FDA 2015]. It may also be preferable to use an analgesic other than the CYP2D6 substrate tramadol in ultrarapid metabolizers [CPIC Guideline 2015].

According to the updated CPIC Guideline [2015], a standard starting dose of codeine, as recommended in the product label, is warranted in patients with an extensive metabolizer phenotype (i.e., a CYP2D6 activity score of 1.0 to 2.0). Likewise, a standard starting dose of codeine is warranted in patients with an intermediate metabolizer phenotype (i.e., a CYP2D6 activity score of 0.5); these patients should be monitored closely for less than optimal response and should be offered an alternative analgesic if required. If the CYP2D6 substrate tramadol is selected as alternative therapy in intermediate metabolizers, close monitoring should be carried out because of the possibility of low response. If clinical genotyping identifies a patient as a CYP2D6 poor metabolizer (i.e., a CYP2D6 activity score of 0), current evidence suggests that the use of codeine be avoided because of the possibility of lack of effect, and that an alternative analgesic should be used. That is, it may be preferable to use an analgesic other than the CYP2D6 substrate tramadol in poor metabolizers. Per the guideline, there is insufficient evidence in the literature to recommend a higher dose of codeine in poor metabolizers, especially given that adverse effects do not differ between poor metabolizers and extensive metabolizers.

Although the evidence suggests that CYP2D6 genotyping may be useful, no guidelines have defined the populations or clinical scenarios in which testing is medically necessary.

Oxycodone
Personalized oxycodone dosing is a new tool for clinicians treating chronic pain patients requiring oxycodone. By expressing a patient's CYP2D6 phenotype pharmacokinetically, a clinician (at least theoretically) can improve the safety and efficacy of oxycodone and decrease the risk for iatrogenically induced overdose or death [Linares et al 2015]. Evidence is limited in this area and additional research is needed on the clinical utility and potential defined target populations for use of CYP2D6 testing related to oxycodone use. The FDA does not currently include pharmacogenomic information in their labeling for Oxycodone.

Hydrocodone
Stauble et al [2014] show that hydromorphone is generated at substantially different rates, dependent on CYP2D6 genotype and suggest that pain relief will vary with CYP2D6 genotype. The authors suggest that inability to metabolize hydrocodone to hydromorphone as seen in the poor metabolizers should alert the clinician to consider alternative medications for managing pain postoperatively. Evidence is limited in this area and additional research needed on the clinical utility of CYP2D6 testing and oxycodone use. The FDA does not currently include pharmacogenomic information in their labeling for hydrocodone.

CYP2D6 and Antidepressants
CYP2D6 poor metabolizers are generally more prone to adverse effects from antidepressants. Among them, the four drugs with the highest level of evidence are amitriptyline, nortriptyline, venlafaxine and fluoxetine. Further data are needed, however, for doxepin and paroxetine, while citalopram adverse effects seem definitely less influenced by CYP2D6 genetic polymorphisms [Haufroid and Hantson, 2015]. According to the prescribing information, CYP2D6 PMs should receive 75 % of the average long-acting aripiprazole dose and pimozide doses >4 mg/day should not be prescribed without CYP2D6 genotyping [Spina et al 2015]. Studies of multiple genes have produced some positive results in groups of patients, but genetic testing does not yet seem to be applicable to choosing medications for a specific patient [Dubovsky 2015].

Rexulti (brexpiprazole)
The brexpiprazole FDA label does not require CYP2D6 genotyping, again suggesting that this information is useful but not necessary in the prescription and dosing of brexpiprazole.

Although tricyclic antidepressants have similar pharmacokinetic properties, there is limited data regarding CYP2D6 genotyping and other TCAs

Venlafaxine (Effexor)
Berm et al [2015] reports on methods of a new clinical trial entitled “Effects and cost-effectiveness of pharmacogenetic screening for CYP2D6 among older adults starting therapy with nortriptyline or venlafaxine: study protocol for a pragmatic randomized controlled trial (CYSCEtrial).” The trial is not yet complete.

PM/IM patients responded to venlafaxine differently than expected according to genotype. However, the UM patient responded to a dosage higher than the usual therapeutic range and without developing side effects, suggesting an association between CYP2D6 gene duplication and the therapeutic efficacy of venlafaxine. The CYP2D6 genotyping may thus provide clinicians with a potential explanation for those patients requiring greater doses of CYP2D6 substrates in order to obtain the same therapeutic efficacy. [Rolla et al 2014]

After venlafaxine discontinuation, Ferriera et al [2014] found there was rapid improvement, with regression of the radiological abnormalities and normalization of the LVEF. This was an important case of drug-induced cardiopulmonary toxicity. The circumstantial intake of inhibitors of the CYP2D6 isoenzyme and the presence of a CYP2D6 slow metabolism phenotype might have resulted in the toxic accumulation of venlafaxine and the subsequent clinical manifestations [Ferreira et al 2014].

The FDA label for venlafaxine does not require CYP2D6 genotyping and there is limited guidance on which populations might benefit from this testing.

CYP2D6 and neuroleptics/psychiatric drugs
Neuroleptic Overview

Although MolDX does not dispute the evidence of an association between CYP2D6 genotype and drug metabolism, this association is not easily translated into broad guidelines for genetic testing across the entire Medicare population. There are limited publications addressing neuroleptics and CYP2D6. MolDX cannot recommend testing for these indications without evidence to support proven changes in medical management for a defined population.

Aripiprazole, Haloperidol, Risperidone, Pimozide
CYP2D6 genotype does affect dose-corrected concentrations of the antipsychotics: for example, median dose- corrected concentrations were 56%, 86%, and 400% higher in predicted poor metabolizers versus extensive metabolizers for aripiprazole (P = 0.004), haloperidol (P > 0.001), and risperidone (P < 0.001), respectively; however, a multiple regression analysis showed that only 4% to 17% of the variation in these concentrations could be explained by CYP2D6 status. CYP2D6 polymorphisms do affect serum levels to a limited extent [Van der Weide 2014].

Aripiprazole
Additional research is needed and specific guidelines on who should be tested for CYP2D6 polymorphisms prior to aripiprazole use have not been defined.

Haloperidol
Additional research is needed and specific guidelines on who should be tested for CYP2D6 polymorphisms prior to haloperidol use have not been defined. 

Risperidone 
Genetic polymorphisms of CYP2D6 play an important role in risperidone, 9- hydroxyrisperidone and active moiety plasma concentration variability, which were associated with common side effects. These results highlight the importance of a personalized dosage adjustment during risperidone treatment [Vandenberghe et al 2015]. In patients with ASD treated with risperidone, a CYP2D6 phenotype may be associated with response to treatment and development of ADRs [Youngster et al 2015]. However, additional research is needed and specific guidelines on who should be tested for CYP2D6 polymorphisms prior to risperidone use have not been defined.

2D6 and perphenazine
Additional research is needed and specific guidelines on who should be tested for CYP2D6 polymorphisms prior to haloperidol use have not been defined.

Thioridazine
This drug was discontinued by the manufacturer in 2005.

Atomoxetine
The higher systemic exposure of the active atomoxetine moieties in CYP2D6*10/*10 individuals may increase the risk of concentration-related adverse events of atomoxetine, although this has not yet been clinically confirmed [Byeon et al 2015]. Differences in pharmacokinetic parameters as well as clinical treatment outcomes across CYP2D6 genotype groups have resulted in dosing recommendations within the product label, however clinical studies supporting the use of genotype guided dosing are currently lacking [Brown et al 2015] and there are no guidelines indicating which populations are appropriate for testing.

2D6 and Movement disorders-Tetrabenazine
Genetic testing of the CYP2D6 gene is considered medically necessary to guide medical treatment and/or dosing for individuals for whom initial therapy is planned with tetrabenazine doses greater than 50 mg/day, or re-initiation of therapy with doses greater than 50 mg/day.

Psychotropic-related movement disorders
Bakker et al [2015] reviewed evidence from pooled data derived from meta- analyses that showed small odds ratios between tardive dyskinesia (TD) and genes encoding DRD2, DRD3, BDNF, COMT, 5-HTR2A, CYP2D6, and MnSOD (p- value < 0.05). GWAS for TD and Parkinsonism have revealed new genes. However, they were unable to glean any clinically useful advice. New pharmacogenetic research and diagnostic systems are likely to create new opportunities [Bakker et al 2015], but there is currently insufficient evidence and guidance surrounding which populations might benefit from testing.

CYP2D6 and beta blockers
There are no current studies on the clinical utility of CYP2D6 testing prior to prescribing beta blockers. Rietveld et al [2015] presented a case that suggests a potential familial liability for metoprolol- induced psychosis. Pharmacokinetic mechanisms are hypothesized to mediate this familial liability through genetic variation in the CYP2D6 genotype. A family history of psychotic symptoms after treatment with beta-blockers may be taken into account when prescribing this beta-blocker [Rietveld et al 2015].

CYP2D6 and adjuvant hormonal therapy in postmenopausal women
In a cohort of 225 breast cancer patients receiving adjuvant tamoxifen, poor metabolizers had a shorter 2-year relapse free survival and a significantly shorter time-to- recurrence as compared to EM patients [Goetz 2007]. Although it is known that in CYP2D6 PM patients, increasing the standard tamoxifen dose two-fold or three-fold raises endoxifen concentrations to levels similar to those of patients with EM phenotype [Martinez de Dueñas et al 2014], no prospective trial has yet shown adjuvant trial data, and CYP2D6 genetic testing is not routinely incorporated into clinical practice. [Sacco et al 2015]

Province et al [2-14] approached the tamoxifen controversy by performing a global meta- analysis of available clinical and CYP2D6 genetic data of tamoxifen-treated breast cancer patients. By strict clinical and genotype criteria, reduced CYP2D6 metabolism is associated with a higher risk of recurrence (as measured by IDFS) in tamoxifen-treated women. However, the heterogeneity observed across sites contributing data to the ITPC points to the likely influence of critical confounding factors unlikely to be controllable in global retrospective studies. Although CYP2D6 is a predictor of IDFS in a subset of patients treated with tamoxifen, the lack of an effect in the entire heterogeneous study population suggests that prospective studies are necessary to finally establish whether genotype-guided selection of hormonal therapy improves clinical outcomes of women with ER-positive breast cancer [Province et al 2014].

Prospective testing of tamoxifen metabolism as gauged by CYP2D6 genotype and serum endoxifen levels is feasible [Ruddy et al 2013]. Future studies of tamoxifen metabolism and efficacy should consider including measurement of serial endoxifen levels. Although clinical evidence at present is insufficient to warrant routine CYP2D6 or endoxifen testing, some clinicians and patients did utilize this predefined algorithm to inform clinical decisions regarding optimal adjuvant endocrine therapy [Ruddy et al 2013].

CYP2C9
CYP2C9*3 and CYP2C9*2 are carried by approximately 35% of Caucasians, but are relatively rare in Asian and African populations (Kirchheiner, 2005). US Hispanics are underrepresented in pharmacogenetic studies.

A higher prevalence of CYP2C9*3 was observed in US Hispanic versus non-Hispanic populations (Claudio-Campos et al., 2015). CYP2C9 *3 and *4 are rare in sub-Saharan populations accounting of <1% of defective alleles and become increasingly more common with greater European influence. CYP2C9*8 is more common in African-Americans and has been used alone, or in combination with CYP2C9 *5, *6, and *11 to predict metabolic phenotype. Most research to-date has focused on the *3 and *4 alleles and additional studies are needed related to additional allelic variation in non-white populations (Guilherme, 2013). CYP2C9 *2 and *3 alleles seem to increase in carrier rate from East to West Asia. Most Asian countries are underrepresented in pharmacogenomic research and additional data from these underrepresented countries will be beneficial to assist in understanding pharmacogenetic dosing of warfarin in this population (Gaikwad, 2014).

CYP2C9 and Vitamin K antagonists
The American College of Medical Genetics and the American College of Chest Physicians (Flockhart 2008, Ansell 2008) recommend that Pharmacogenetic testing for warfarin treatment should not be performed. This stance has not been revised. We recommend continuing to follow ACMG Practice Guidelines at this time due to the limited number of large, prospective clinical trials showing consistent findings.

The FDA placed a warning label on warfarin to include information on CYP2C9 and VKORC1 genotypes as predictors of dose response in 2007 and then in 2010 the warning was revised to include specific dosage recommendations based on genotype. The FDA has not yet mandated this pharmacogenetic testing. Ong et al (2012) cites the major limitation in translation into clinical implementation to be the lack of adoption by major professional societies such as the American College of Chest Physicians and ACMG which assign concern related to the lack of evidence from large-scale prospective randomized trials as noted above. At the time of this publication, three such trials were noted to be underway: COAG (now complete), GIFT (currently recruiting) and EU-PACT (now complete).

Franchini et al. (2014) performed a meta-analysis of all published studies available up until March 2014 evaluating the influence of CYP2C9 and VKORC1 genotypes on Warfarin and other VKA dose requirements. 2,812 total patients were analyzed in the pharmacogenetic arm and 1,401 in the control arm with the primary endpoints including incidence of major bleeding, thrombosis and death during the initiation of warfarin and other VKA therapy. Of greatest note, there was a mean 52.5% reduction of major bleeding episodes in the pharmacogenetic arm (RR =0.47, 95% CI, 0.23-0.96; P = 0.040). Franchini et al. concludes by suggesting a genotype-guided approach appears to lower the risk of severe bleeding in the initial period of VKA anticoagulation and noting that prospective trials based upon clinical rather than on laboratory outcomes are needed to firmly establish the efficacy of a pharmacogenetic approach in guiding initial VKA dose requirements.

The Challengers reference Mega et al. (2015) who published their prospective, randomised, double-blind study data (ENGAGE AF- TIMI 48) including 14,348 genotyped patients, 4,833 of whom were taking warfarin. 2,982 were classified as normal responders, 1,711 as sensitive responders, and 140 as highly sensitive responders. Key findings include sensitive and highly sensitive responders spending greater proportions of time over- anticoagulated in the first 90 days of treatment and had increased risk of bleeding with warfarin (median 2.2% normal, IQR 0-20.2; 8.4% sensitive,0-25.8; and 18.3% highly sensitive, 0-32.6; ptrend<0.0001). While this study does provide evidence of clinical utility, it is a single study and additional studies of this kind are needed to confirm the findings. Of note, this study was funded by Daiichi Sankyo, a large pharmaceutical company.

The results from the COAG trial were published in August 2015 (Kasner et al., 2015). This prospective randomized trial included 1015 participants randomized to either pharmacogenetically or clinically guided warfarin dosing algorithms in either an inpatient or outpatient setting. This study found no difference in time for first therapeutic INR or maintenance dose, no evidence of interaction between pharmacogenetically versus clinically guided therapy and location of initiation for these outcomes. The primary conclusion, “The warfarin dosing algorithms performed similarly for subjects who initiated warfarin as inpatients and outpatients, regardless of whether dosing was pharmacogenetically or clinically guided.”

Large-scale prospective trials are the next step in confirming clinical utility for CYP2C9- guided VKA therapy. At this time, the small number of trials of this kind have been inconsistent in their findings. Additionally, key professional societies have not recommended the use of CYP2C9 genotyping as part of the dosing algorithm for VKA use, especially in the Warfarin-naive population. We therefore cannot agree with the Challengers that CYP2C9 genotyping is currently necessary to guide VKA therapy.

CYP2C9 and phenytoin
CYP2C9 mediates phenytoin clearance. However, as Franco et al. (2015) notes, “the clinical value and cost-effectiveness of CYP2C9 genotyping in improving the safety of phenytoin therapy, however, have not been clearly established and require formal testing in well-designed prospective studies.”

Progress towards potential coverage in this area includes the presence of professional societies developing guidelines/recommendations. Caudle et al (2014) published the Clinical Pharmacogenetics Implementation Consortium (CPIC) guidelines for CYP2C9 and HLA-B genotype and phenytoin dosing. The authors summarized that the recommended phenytoin maintenance dosage does not need adjustment for extensive metabolizers and available evidence does not clearly indicate the amount of dose reduction needed for intermediate and poor metabolizers. Therefore, for intermediate metabolizers, at least 25% reduction of the recommended starting maintenance dose, and at least 50% reduction for poor metabolizers was recommended for consideration by this consortium. Due to high variability in pediatrics, the authors comment that phenytoin therapeutic recommendations based on CYP2C9 genotype in this population are difficult. Additionally, other genes have been implicated in the metabolism of phenytoin which may modify the predicted CYP2C9-mediated effects, thus suggesting that a greater understanding of genotype by genotype interaction is necessary in the metabolism of this medication.

Reconsideration of CYP2C9 genotyping for phenytoin dosing awaits further prospective studies and additional direction from other professional societies.

2C9 and sulfonylureas
Samer et al. (2013) provides limited supporting data for the necessity of CYP2C9*3 evaluation in guiding sulfonylurea dosage for treatment of type 2 diabetes. While this review does highlight the studies noted by the Challengers, the authors go on to state, “even if there is a clear [pharmacokinetic] relation between sulfonylureas and CYP2C9 polymorphisms, no adaptation of dosage is recommended because of a mild risk of hypoglycaemia and the possibility to monitor glucose plasma levels.”

CYP-mediated metabolism of glyburide was evaluated in vitro by Varma et al. (2014) who suggested CYP2C9 function has a minimal effect on the systemic exposure, implying discrepancy in the contribution of CYP2C9 to glyburide clearance.

A more recent meta-analysis performed by Maruthur et al. (2014) evaluated the pharmacogenetics of type 2 diabetes which included 2 studies directly pertaining to sulfonylureas and CYP2C9 interaction. One of these studies observed a greater mean change from baseline in HbA1c at 6 months; however, the sample size for the variant diplotype was only n = 2. Key discussion points posed by the authors include the need for well-designed studies to confirm significant pharmacogenetic interactions in order for genotype-guided therapy to be incorporated into clinical practice. “In this systematic review, we find evidence of biologically plausible pharmacogenetic interactions for metformin, sulfonylureas, ...., but these results require confirmation in future studies to determine if an individual’s genetic information can be used to individualize the choice of prediabetes and diabetes pharmacologic management.”

As with concerns over other CYP2C9-mediated medications, there is a lack of well-designed prospective studies to establish clinical utility for the use of CYP2C9 genotyping for sulfonylureas in any defined population.

2C9 and Angiotensin II receptor antagonists
Overall, there is insufficient and inconsistent published literature regarding clinical validity and utility of CYP2C9- modulated dosing of angiotensin II receptor antagonists. Variation in metabolism rate and drug clearance does not directly correlate to clinical applicability.

Cabaaleiro et al. (2013) evaluated 246 healthy volunteers from seven single-dose clinical trials with CYP2C8 variant alleles as well as CYP2C9*2 and CYP2C9*3 showing a higher half-life of losartan in CYP2C9*3 allele carriers over volunteers with wild- type genotype. The authors drew conclusions to suggest CYP2C9 and CYP2C8 are relevant to the pharmacokinetics of losartan and valsartan, but not to candesartan or telmisartan. Though this study demonstrates genotype-mediated drug clearance differences, the clinical utility of this finding was not reviewed. Likewise, Bae et al (2012) evaluated the effects of 2C9*1/*3 and *1/*13 genotypes on the pharmacokinetics of losartan and its active metabolite, E-3174. While the CYP2C9*1/*3 subjects showed lower oral clearance and longer half-life than 2C9*1/*1, the authors note the clinical effects of losartan may not be reduced by CYP2C8*1/*3 and CYP2C9*1/*13. 

2C9 and NSAIDs
The inconsistent and controversial association between NSAID-related gastrointestinal bleeding complication and 2C9 genotype prevents actionable recommendations for genotype- mediated dosing at this time.

The Challengers reference Somer et al. (2013) regarding the review of CYP2C9 polymorphism related to gastrointestinal bleeding with NSAID therapy. While the authors of this meta-analysis highlight the small number of studies which have found a significantly higher risk of bleeding with NSAID therapy associated with CYP2C9*3 and CYP2C9*2 genotype, the presence of additional retrospective studies is also highlighted finding no relationship between NSAID-induced gastric ulceration and CYP2C9 genotype. Additionally, in vivo evidence regarding this association is controversial and genotype-based dosing is not recommended by the authors.

Additional research is also needed into the potential combinatorial effect of CYP2C9 and CYP2C8 (or other variants) as results from retrospective studies have not aligned with expected outcomes from in vivo studies (Rodrigues, 2005). A more complete understanding of this gene-gene interaction is necessary.

REFERENCES:

Bae JW, Choi CI, Lee HI, Lee YJ, Jang CG, Lee SY. Effects of CYP2C9*1/*3 and *1/*13 on the pharmacokinetics of losartan and its active metabolite E-3174. Int J Clin Pharmacol Ther. 2012 antidepressants. Clin Toxicol (Phila). 2015 Jul;53(6):501- 10. doi: 10.3109/15563650.2015.1049355. Epub 2015 May 22.

Hicks JK, Bishop JR, Sangkuhl K et al. Clinical Pharmacogenetics Implementation Consortium (CPIC) Guideline for CYP2D6 and CYP2C19 Genotypes and Dosing of Selective Serotonin Reuptake Inhibitors. Clin Pharmacol Ther. 2015 Aug;98(2):127-34. doi: 10.1002/cpt.147. Epub 2015 Jun 29.

Juel J, Pareek M, and Jensen SE. The clopidogrel-PPI interaction: an updated mini-review. Curr Vasc Pharmacol. 2014;12(5):751-7.

Kasner SE, Wang L, French B, Messe SR, Horenstein R, Mohler ER, Muldowney JA, Ellenberg J, Kimmel SE. The impact of inpatient versus outpatient initiation on early warfarin dosing. Am J Cardiovasc Drugs. 2015 Aug;15(4):267-74.

Kirchheiner J, Brockmoller J. Clinical consequences of cytochrome P450 2C9 polymorphisms. Clin Pharmacol Ther. 2005;77(1):1-16.

Mega JL, Walker JR, Ruff CT, Vandell AG, Nordio F, Deenadayalu N, Murphy SA, Lee J, Mercuri MF, Giugliano RP, Antman EM, Braunwald E, Sabatine MS. Genetics and the clinical response to warfarin and edoxaban: findings from the randomised, double-blind ENGAGE AF- TIMI 48 trial. Lancet. 2015 Jun;385(9984):2280-7.

Lassen D, Damkier P, Brøsen K. The Pharmacogenetics of Tramadol. Clin Pharmacokinet. 2015 Aug;54(8):825-36. doi: 10.1007/s40262-015-0268-0.

Linares OA, Daly D, Linares AD, Stefanovski D, Boston RC. Personalized oxycodone dosing: using pharmacogenetic testing and clinical pharmacokinetics to reduce toxicity risk and increase effectiveness. Pain Med. 2014 May;15(5):791-806. doi: 10.1111/pme.12380. Epub 2014 Feb 12.

LLerena A, Naranjo ME, Rodrigues-Soares F, Penas-LLedó EM, Fariñas H, Tarazona-Santos E. Interethnic variability of CYP2D6 alleles and of predicted and measured metabolic phenotypes across world populations.

Expert Opin Drug Metab Toxicol. 2014 Nov;10(11):1569-83. doi: 10.1517/17425255.2014.964204.

Local Coverage Determination (LCD): MoPath: GeneSight® Assay for Refractory Depression (L35437). Original effective date: 10/24/14.

Lorenzini K, Calmy A, Ambrosioni J, Assouline B, Daali Y, Fathi M, Rebsamen M, Desmeules J, Samer CF. Serotonin syndrome following drug-drug interactions and CYP2D6 and CYP2C19 genetic polymorphisms in an HIV-infected patient. AIDS. 2012 Nov 28;26(18):2417-8.

Maruthur NM, Gribble MO, Bennett WL, Bolen S, Wilson LM, Balakrishnan P, Sahu A, Bass E, Kao WH Clark JM. The pharmacogenetics of type 2 diabetes: a systematic review. Diabetes Care. 2014;37(3):876-86.

Monte AA, Heard KJ, Campbell J, Hamamura D, Weinshilboum RM, Vasiliou V. The effect of CYP2D6 drug-drug interactions on hydrocodone effectiveness. Acad Emerg Med. 2014 Aug;21(8):879-85. doi: 10.1111/acem.12431. Epub 2014 Aug 24.

Nakamura A, Mihara K, Nemoto K, Nagai G, Kagawa S, Suzuki T, Kondo T. Lack of correlation between the steady-state plasma concentrations of aripiprazole and haloperidol in Japanese patients with schizophrenia. Ther Drug Monit. 2014 Dec;36(6):815-8. doi: 10.1097/FTD.0000000000000082.

Ong FS, Deignan JL, Kuo JZ, Bernstein KE, Rotter JI, Grody WW, Das K. Clinical utility of pharmacogenetic biomarkers in cardiovascular therapeutics: a challenge for clinical implementation. Pharmacogenomics. 2012 Mar;13(4):465-75.

Province MA, Goetz MP, Brauch H, Flockhart DA, Hebert JM, Whaley R, Suman VJ, Schroth W, Winter S, Zembutsu H, Mushiroda T, Newman WG, Lee MT, Ambrosone CB, Beckmann MW, Choi JY, Dieudonné AS, Fasching PA, Ferraldeschi R, Gong L, Haschke-Becher E, Howell A, Jordan LB, Hamann U, Kiyotani K, Krippl P, Lambrechts D, Latif A, Langsenlehner U, Lorizio W, Neven P, Nguyen AT, Park BW, Purdie CA, Quinlan P, Renner W, Schmidt M, Schwab M, Shin JG, Stingl JC, Wegman P, Wingren S, Wu AH, Ziv E, Zirpoli G, Thompson AM, Jordan VC, Nakamura Y, Altman RB, Ames MM, Weinshilboum RM, Eichelbaum M, Ingle JN, Klein TE; International Tamoxifen Pharmacogenomics Consortium. CYP2D6 genotype and adjuvant tamoxifen: meta-analysis of heterogeneous study populations. Clin Pharmacol Ther. 2014 Feb;95(2):216-27. doi: 10.1038/clpt.2013.186. Epub 2013 Sep 23.

Prows CA, Zhang X, Huth MM, Zhang K, Saldaña SN, Daraiseh NM, Esslinger HR, Freeman E, Greinwald JH, Martin LJ, Sadhasivam S. Codeine-related adverse drug reactions in children following tonsillectomy: a prospective study. Laryngoscope. 2014 May;124(5):1242-50. doi: 10.1002/lary.24455. Epub 2013 Nov 13.

Rietveld L, van der Hoek T, van Beek MH, Schellekens AF. Familial liability for metoprolol- induced psychosis. Gen Hosp Psychiatry. 2015 Jun 25. pii: S0163-8343(15)00153-X. doi: 10.1016/j.genhosppsych.2015.06.016. [Epub ahead of print] 

Rodrigues AD. Impact of CYP2C9 genotype on pharmacokinetics: are all cyclooxygenase inhibitors the same? Drug Metab Dispos. 2005 Nov;33(11):1567-75.

Rolla R, Gramaglia C, Dalò V, Ressico F, Prosperini P, Vidali M, Meola S, Pollarolo P, Bellomo G, Torre E, Zeppegno P. An observational study of Venlafaxine and CYP2D6 in clinical practice. Clin Lab. 2014;60(2):225-31.

Ruddy KJ, Desantis SD, Gelman RS, Wu AH, Punglia RS, Mayer EL, Tolaney SM, Winer EP, Partridge AH, Burstein HJ. Personalized medicine in breast cancer: tamoxifen, endoxifen, and CYP2D6 in clinical practice. Breast Cancer Res Treat. 2013 Oct;141(3):421-7. doi: 10.1007/s10549-013-2700-1. Epub 2013
Sep 24.

Sacco K, Grech G. Actionable pharmacogenetic markers for prediction and prognosis in breast cancer. EPMA J. 2015 Jul 22;6(1):15. doi: 10.1186/s13167-015-0037-z. eCollection 2015.

Samer CF, Ing Lorenzini K, Rollason V, Daali Y, Desmeules JA. Applications of CYP450 testing in the clinical setting. Mol Diagn Ther. 2013 Jun; 17(3):165-184.

Scott SA, Sangkuhl K, Stein CM, Hulot J-S, Mega JL, Roden DM, Klein TE, Sabatine MS, Johnson JA, Shuldiner AR. Clinical Pharmacogenetics Implementation Consortium Guidelines for CYP2C19 Genotype and Clopidogrel Therapy: 2013 Update. Clin Pharmacol Ther. 2013 Sep;94(3):317-23. Epub 2013 May 22.

Somogyi AA, Coller JK, Barratt DT. Pharmacogenetics of opioid response. Clin Pharmacol Ther. 2015 Feb;97(2):125-7. doi: 10.1002/cpt.23. Epub 2014 Dec 9.

Spina E, de Leon J. Clinical applications of CYP genotyping in psychiatry. J Neural Transm. 2015 Jan;122(1):5-28. doi: 10.1007/s00702-014-1300-5. Epub 2014 Sep 9. 

Stauble ME, Moore AW, Langman LJ, Boswell MV, Baumgartner R, McGee S, Metry J, Jortani SA Hydrocodone in postoperative personalized pain management: pro- drug or drug? Clin Chim Acta. 2014 Feb 15;429:26-9. doi: 10.1016/j.cca.2013.11.015. Epub 2013 Nov 21.

Van der Weide K, van der Weide J. The Influence of the CYP3A4*22 Polymorphism and CYP2D6 Polymorphisms on Serum Concentrations of Aripiprazole, Haloperidol, Pimozide, and Risperidone in Psychiatric Patients. J Clin Psychopharmacol. 2015 Jun;35(3):228-36. doi: 10.1097/JCP .0000000000000319.

Vandenberghe F1, Guidi M, Choong E, von Gunten A, Conus P, Csajka C, Eap CB Genetics- Based Population Pharmacokinetics and Pharmacodynamics of Risperidone in a Psychiatric Cohort. Clin Pharmacokinet. 2015 Jul 1. [Epub ahead of print]

Whirl-Carrillo M, McDonagh EM, Hebert JM, Gong L, Sangkuhl K, Thorn CF, Altman RB, Klein TE. Pharmacogenomics knowledge for personalized medicine. Clin Pharmacol Ther. 2012 Oct;92(4):414-7. doi: 10.1038/clpt.2012.96.

Yee MM, Josephson C, Hill CE, Harrington R, Castillejo MI, Ramjit R, Osunkwo I. Cytochrome P450 2D6 polymorphisms and predicted opioid metabolism in African American children with sickle cell disease. J Pediatr Hematol Oncol. 2013 Oct;35(7):e301- 5. doi: 10.1097/MPH.0b013e31828e52d2.

Youngster I1, Zachor DA, Gabis LV, Bar-Chaim A, Benveniste- Levkovitz P, Britzi M, Soback S, Ziv-Baran T, Berkovitch M. CYP2D6 genotyping in paediatric patients with autism treated with risperidone: a preliminary cohort study. Dev Med Child Neurol. 2014 Oct;56(10):990-4. doi: 10.1111/dmcn.12470. Epub 2014 May 1