Posted: February 26th, 2023

genomics

Value of Supportive Care Pharmacogenomics in Oncology Practice

JAI N. PATEL , a LAUREN A. WIEBE,

b HENRY M. DUNNENBERGER,

b HOWARD L. MCLEOD

aLevine Cancer Institute, Carolinas HealthCare System, Charlotte, North Carolina, USA; bNorthShore University Health System, Evanston, Illinois, USA; cThe DeBartolo Family Personalized Medicine Institute, Moffitt Cancer Center, Tampa, Florida, USA Disclosures of potential conflicts of interest may be found at the end of this article.

Key Words. Supportive care • Pharmacogenomics • Pharmacogenetics • Cancer • Oncology • Symptom management

ABSTRACT

Genomicmedicine provides opportunities to personalize cancer therapy for an individual patient. Although novel targeted therapies prolong survival, most patients with cancer continue to suffer from burdensome symptoms including pain, depres- sion, neuropathy, nausea and vomiting, and infections, which significantly impair quality of life. Suboptimal management of these symptoms can negatively affect response to cancer treat- ment and overall prognosis. The effect of genetic variation on drug response—otherwise known as pharmacogenomics—is

well documented and directly influences an individual patient’s response to antiemetics, opioids, neuromodulators, antidepres- sants, antifungals, and more. The growing body of pharmacoge- nomic data can now guide clinicians to select the safest and most effective supportive medications for an individual patient with cancer from the very first prescription.This review outlines a theoretical patient case and the implications of using pharma- cogenetic test results to personalize supportive care throughout the cancer care continuum.The Oncologist 2018;23:1–9

Implications for Practice: Integration of palliative medicine into the cancer care continuum has resulted in increased quality of life and survival for patients with many cancer types. However, suboptimal management of symptoms such as pain, neuropathy, depression, and nausea and vomiting continues to place a heavy burden on patients with cancer. As demonstrated in this theoretical case, pharmacogenomics can have a major effect on clinical response to medications used to treat these conditions. Recognizing the value of supportive care pharmacogenomics in oncology and application into routine practice offers an objective choice for the safest andmost effective treatment compared with the traditional trial and error method.

INTRODUCTION

Personalization of medicines and careful attention to quality of life (QOL) are increasingly part of expectations for patients with cancer throughout the care trajectory. With the growing com- plexity of both antineoplastic and supportive care, a practicing oncologist has diminishing time to manage each patient’s myr- iad supportive care concerns by trial and error. Suboptimal management of these symptoms compromises potential bene- fits from cancer therapy, disrupts clinic workflow, increases emergency room visits, and affects both patient satisfaction and reimbursement [1–5]. Better tools are needed to make individual, tailored choices easier for busy clinicians every day.

Genetic variation is well documented across the human genome and ultimately affects a patient’s response to medica- tions with regard to efficacy and toxicity. The genome is quickly becoming a pragmatic tool that can assist medical oncologists and palliative medicine providers in the selection of the best supportive care treatments for patients with cancer. Notably, knowledge of pharmacogenetic variants associated with drug response is rapidly evolving. To aid in the use of pharmacoge- netic data, the Clinical Pharmacogenetics Implementation

Consortium (CPIC) develops peer-reviewed guidelines on how to best apply genetic data to modify drug therapy [6, 7]; how- ever, there is also an emerging category of relevant genes not currently covered by CPIC guidelines. CPIC categorizes patients into metabolizer phenotypes based on their genotype (Table 1) and provides specific dosing or therapy selection recommenda- tions for each category. Increasingly in this era of personalized medicine, patients with cancer are expecting their oncologist to use their unique genomes to choose therapy correctly the first time and minimize drug-related toxicities [8].

THE CASE: BARB G.

TheOncologist 2018;23:1–9

www.TheOncologist.com Oc AlphaMed Press 2018

Symptom Management and Supportive Care

Correspondence: Jai N. Patel, Pharm.D., Levine Cancer Institute, Carolinas HealthCare System, 1021 Morehead Medical Drive, Charlotte, North Carolina 28204, USA. Telephone: 980-442-4113; e-mail: Jai.Patel@carolinashealthcare.org Received November 14, 2017; accepted for publication February 21, 2018; published Online First on April 6, 2018. http://dx.doi.org/10.1634/theoncologist.2017-0599

Barb G., a 60-year-old woman, is a new patient in clinic with a breast mass found to be adenocarcinoma. Many of her relatives had extreme reactions to prescription medications, so she researched extensively and wants to do a full pharmacoge- nomic profile, as she heard this kind of testing could inform drug choice and dosing throughout her cancer journey. She hands you her results that show she is a CYP2D6 poor metabo- lizer (PM) and a CYP2C19 ultrarapid metabolizer (UM).

The Oncologist 2018;23:956–964

www.TheOncologist.com

©AlphaMed Press 2018

1549490x, 2018, 8, D ow

nloaded from https://theoncologist.onlinelibrary.w

iley.com /doi/10.1634/theoncologist.2017-0599 by N

ova Southeastern U niversity, W

iley O nline L

ibrary on [21/12/2022]. See the T erm

s and C onditions (https://onlinelibrary.w

iley.com /term

s-and-conditions) on W iley O

nline L ibrary for rules of use; O

A articles are governed by the applicable C

reative C om

m ons L

icense

http://orcid.org/0000-0001-9845-3697

957Patel, Wiebe, Dunnenberger et al.

www.TheOncologist.com ©AlphaMed Press 2018

The plan is neoadjuvant chemotherapy with doxorubicin

and cyclophosphamide. Barb is terrified of nausea and asks you

if the pharmacogenomic test results will direct your decisions

about antiemetic selection. She wants to be sure she is getting

the best supportive care possible.

ANTIEMETIC SELECTION Chemotherapy-induced nausea and vomiting (CINV) is one of the most notorious and debilitating adverse drug effects experi- enced by patients treated with cytotoxic chemotherapy agents [9]. Ineffective control of CINV can lead to patient distress, unacceptable QOL, and treatment noncompliance [10]. Since their advent, serotonin receptor antagonists (5HT3-RA) have been the backbone of CINV prophylaxis and treatment. CYP2D6 is a key metabolic pathway for inactivation of most 5HT3-RAs— particularly ondansetron and palonosetron, the two most widely used 5HT3-RAs. For example, CYP2D6 UMs, who are found in approximately 5% of the white population, degrade ondansetron too rapidly, resulting in ineffective blood levels and thus weak control of CINV [10–13]. Studies showmore epi- sodes of vomiting and higher reported nausea for CYP2D6 UMs receiving ondansetron on equivalent chemotherapy regimens [13, 14].

CPIC guidelines support a change in therapy for patients with known CYP2D6 UM status and planned ondansetron [15]. Granisetron is the only 5HT3-RA that does not involve CYP2D6 in its metabolism; thus, it might be the most reasonable option in a suspected UM [10]. If switching 5HT3-RAs does not have an effect on the poorly controlled nausea and vomiting, most guidelines support the addition of a neurokinin 1-receptor antagonist. The pharmacogenomic test results could be submit- ted to insurance in order to justify nonformulary coverage in a case such as this. Although many polymorphisms exist that might explain patient variability in 5HT3-RA efficacy for acute CINV, only CYP2D6 appears to be clinically actionable. Currently in clinical practice, CYP2D6 genetic testing is readily available and may be used to guide future 5HT3-RA regimen choices because of its consistent clinical data, relatively low cost, and high patient benefit. (See Fig. 1.)

Barb is a CYP2D6 poor metabolizer and is likely to have the

appropriate benefit from ondansetron, which is a mainstay of

your practice. Given that she will have slowed inactivation of

the ondansetron, she might be at a slightly higher risk for side

Table 1. Definition of phenotypes and potential clinical implication on drug response

Phenotypes Definition

Clinical implication

Active drug Prodrug

Ultrarapid metabolizer (UM)

Increased enzyme activity compared with rapid metabolizers

Significantly increased inactivation and reduced response

Significantly increased activation and increased response and side effects

Rapid metabolizer (RM)

Increased enzyme activity compared with normal metabolizers but less than ultrarapid metabolizers

Increased inactivation and reduce response

Increased activation and increased response and side effects

Normal metabolizer (NM)

Fully functional enzyme activity Normal or expected clinical response Normal or expected clinical response

Intermediate metabolizer (IM)

Decreased enzyme activity compared with normal metabolizers but more than poor metabolizers

Reduced inactivation and increased response and side effects

Reduced activation and reduced response

Poor metabolizer (PM)

Little to no enzyme activity Significantly reduced inactivation and increased response and side effects

Significantly reduced activation and reduced response

Clinical implications noted in the table are generally true, but may differ based on the specific gene and drug (e.g. CYP3A5 NMs may require higher tacrolimus doses than PMs since PM is the predominant phenotype and NMs may have sub-therapeutic concentrations).

Figure 1. Pharmacogenetic-driven treatment pathway for chemotherapy-induced nausea and vomiting. CYP2D6 UMs receiving moderate to high emetogenic chemotherapeutic regimens are rec- ommended to receive granisetron as the first-line 5HT3-RA because of increased metabolism or inactivation of other 5HT3-RAs. PMs may require closer and more frequent monitoring for side effects (malaise, constipation, headache, QTprolongation) because of possi- ble supratherapeutic serum levels. Clinical risk factors (emesis with prior chemotherapy, female gender, younger age, lack of a signifi- cant history of alcohol consumption, history of motion sickness, con- current radiation treatment, history of hyperemesis gravidarum, and high dose or highly emetogenic combination chemotherapy regi- mens) should be considered when deciding whether or not to administer a neurokinin 1 receptor antagonist in patients receiving moderate emetogenic chemotherapy or a 5HT3-RA in patients receiving low emetogenic chemotherapy. 1, Monitor closely for 5HT3-RA side effects such as constipation,

low-grade headache, QT prolongation, or malaise because of potentially increased blood levels. 2, If patient is unable to take granisetron or if granisetron is

unavailable, then may consider using high-dose ondansetron. Abbreviations: 5HT3-RA, serotonin receptor antagonist; CYP2D6,

cytochrome P450 2D6; IM, intermediate metabolizer; PM, poor metabolizer; NM, normal metabolizer; UM, ultrarapid metabolizer.

1549490x, 2018, 8, D ow nloaded from https://theoncologist.onlinelibrary.w iley.com /doi/10.1634/theoncologist.2017-0599 by N ova Southeastern U niversity, W iley O nline L ibrary on [21/12/2022]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense

958 Suppor ve Care Pharmacogenomics in Oncology

©AlphaMed Press 2018

effects such as headache or constipation. There is no current

recommendation to reduce the dose of the 5HT3-RA in this set-

ting, but it may be considered in the case of intolerable side

effects, for which she should be closely monitored. If her CINV

requires the addition of intravenous palonosetron, she would

be expected to respond favorably to that as well. Her CYP2D6 PM phenotype suggests that appropriate, effective drug levels

will be present in the serum. Barb tolerates her chemotherapy generally well and has a

favorable response with desired downsizing of the tumor. Next,

she undergoes surgery for removal of the shrinking mass and

calls your nurse the day after discharge from the surgery. She

was given a prescription for Tylenol #3 (acetaminophen con-

taining codeine; Johnson & Johnson, New Brunswick, NJ) and

was instructed to take one tablet every 6 hours maximum. She

mentioned that Tylenol #3 did not help her after an oral surgery a few years ago, so the breast surgeon decided to instead try

tramadol 50 mg every 4 hours because it is not a schedule II

medication and the patient was more comfortable trying this

first. Barb administers tramadol around the clock for 1 week

but tells your nurse that the pain medicine did “absolutely

nothing” and asks her to please help.

OPIOID SELECTION Any practicing oncologist knows that pain is one of the most persistent and burdensome symptoms in patients with cancer, affecting approximately 50% of those with curable cancer and up to 75% with advanced disease. Only one third of patients with cancer in the U.S. achieve significant pain improvement with standard strategies [16]. Known factors associated with ineffective analgesia include geriatric age, minority race, and inadequate clinician assessment [17]; however, there is a grow- ing realization that a patient’s unique genetic makeup could affect clinical response to opioids and thus could be used for drug and/or dose selection. (See Fig. 2.) CYP2D6 is responsible for the activation of codeine, tramadol, oxycodone, and hydro- codone into stronger opioids: morphine, o-desmethyltramadol, oxymorphone, and hydromorphone, respectively [18]. More than 100 CYP2D6 alleles have been identified that may alter enzyme function. Even within an ethnic group, the frequency of the common alleles that result in either reduced function or loss of function are highly variable (15%–41%), thus making generalization of pharmacogenomic phenotype by race highly unreliable in clinical practice [19].

Codeine

The analgesic effect of codeine is mainly attributed to its con- version to morphine mediated by CYP2D6. Morphine has a 200 times higher affinity and 50 times higher intrinsic activity at the m-opioid receptor than codeine itself. Codeine-related deaths have been reported in patients known to be CYP2D6 UMs, now a black-box warning [20–26]. Alternatively, CYP2D6 PMs will find codeine to be an ineffective analgesic given that they have no conversion of codeine to the more active morphine. CPIC guidelines strongly recommend that CYP2D6 UMs and PMs should avoid codeine because of the increased risk of toxicities and lack of analgesic effects, respectively [27].Without pharma- cogenomic testing, astute clinicians might avoid codeine if patients report inefficacy; however, the issue of codeine in CYP2D6 UMs is a real risk of harm without the benefit of formal pharmacogenomic testing.

Oxycodone and Hydrocodone

Although the drugs oxycodone and hydrocodone have some analgesic activity, they are metabolized by CYP2D6 to the much more potent metabolites of oxymorphone and hydromor- phone, respectively. A study of 450 patients with cancer receiv- ing oxycodone demonstrated that plasma concentrations of the more active oxymorphone were up to 11 times higher in patients with rapid metabolism than in those with poor metab- olism at CYP2D6 (p< .0001) [28]. In another study, depending on CYP2D6 metabolism, patients required either 16 (UMs) or 25 (PMs) mg of oxycodone to achieve equal analgesic effect (p5 .005) [29]. Studies have shown that a similar phenomenon occurs when patients are given hydrocodone. CYP2D6 UMs had

Figure 2. Pharmacogenetic-driven treatment pathway for pain management. CYP2D6 UMs and PMs should avoid tramadol, codeine, hydrocodone, and oxycodone. PMs may be at risk for treatment failure because of their inability to convert the parent drug into its more active metabolite. UMs may be at risk for treatment-related side effects because of supratherapeutic con- centrations of active metabolites. Patients with GG genotypes for COMT and/or OPRM1 may require higher morphine equivalents for analgesia. Oxycodone and hydrocodone are also inactivated via CYP3A4; therefore, drugs that inhibit or induce the CYP3A4 pathway should be avoided, when possible. 1, If patient is on a strong CYP2D6 inhibitor, then classify as a

poor metabolizer. 2, If APAP or an NSAID is ineffective for pain, may consider

either increasing dose or progressing to selection from moderate category.

3, If COMT and/or OPRM1 GG genotype, patient may require higher doses or rapid titration for pain relief. 4, May consider methadone in patients unresponsive to stand-

ard pain therapy; refer to pain specialist if necessary. Polymor- phisms in CYP2B6 may alter methadone exposure. Abbreviations: APAP, acetaminophen; CYP2D6, cytochrome

P450 2D6; IM, intermediate metabolizer; NM, normal metabolizer; NSAID, nonsteroidal anti-inflammatory drug; PM, poor metabo- lizer; UM, ultrarapid metabolizer.

959Patel, Wiebe, Dunnenberger et al.

www.TheOncologist.com ©AlphaMed Press 2018

a 10-fold increase in plasma concentrations of hydromorphone compared with patients at the other end of the spectrum (p5 .023), which correlated with pain relief [30]. Finally, in CYP2D6 PMs, opioids that are activated more slowly have less predictable clearance and can result in safety concerns, as the drug and its metabolites are present longer than expected.

Tramadol

Like codeine, tramadol is a prodrug and requires CYP2D6- mediated activation for analgesic activity. Depending on geno- type, the area under the curve of the active metabolite can range from 0 to 235 ng3 hr/mL [31], thus leading to wildly dif- ferent perceptions of clinical efficacy [31–33]. In a prospective study of approximately 300 patients recovering from abdominal surgery, the percent of nonresponders was significantly higher in the PM group (46.7%) compared with the normal metabo- lizer group (21.6%; p5 .005) [33]. Most concerning, tramadol- induced respiratory depression was reported in a CYP2D6 UM patient who also had renal impairment [32]. These data suggest that CYP2D6 is highly informative for consideration of tramadol therapy, similar to the guidelines set forth for codeine by CPIC.

For patients with either ultrarapid or poor CYP2D6 metabo- lism who are prescribed codeine, CPIC guidelines recommend alternative drugs that are not affected by CYP2D6, such as mor- phine. Specifically, tramadol, hydrocodone, and oxycodone are not ideal choices given that they are metabolized by CYP2D6.

A patient like Barb, who is a CYP2D6 PM and previously

failed codeine therapy, will also likely not activate the tramadol

to its active metabolite and will thus miss most of the intended

analgesic effect. A prescription for either morphine or hydro-

morphone would bypass any need for activation and would be

the most appropriate selection in this case. If a practitioner

wished to prescribe either hydrocodone or oxycodone, Barb’s

CYP2D6 PM status predicts that she may require higher doses

than usual for appropriate analgesic effect.

You let the surgeon know that you feel comfortable pre-

scribing morphine based on her pharmacogenomic profile. You

call Barb back and let her know that a prescription for morphine

15 mg immediate release every 4 hours as needed is waiting for

her at the pharmacy, which should be a more effective analge-

sic in her case. Barb ultimately experiences significant pain relief

with morphine. With regard to analgesia, pharmacogenomic testing is guid-

ing drug choice and dose recommendations in an increasingly data-driven way. Beyond the above data on CYP2D6, there are additional ways in which pharmacogenomic testing may affect opioid prescribing in patients with cancer.

Emerging Genes: OPRM1 and COMT

The gene responsible for coding the mu-opioid receptor is OPRM1. Mu receptor activation leads to analgesia and known opioid side effects, including respiratory depression, sedation, euphoria, and decreased gastrointestinal motility [34]. Multiple studies have shown that variation in alleles at this gene result in different clinical responses to opioids. Given altered receptor function, a simple base-pair substitution can lead a patient to require 60%–100% more morphine for equal analgesia than in the average population [9, 35–37]. At the bedside, it may appear that the patient has poor or almost no response to opioids even if they are titrated. These patients are at a real risk of uncontrolled pain, as clinicians may be appropriately hesitant

to escalate opioid doses rapidly without objective genotype- directed information to support an aggressive titration.

Opioid analgesia can also be enhanced by the presence of catecholamines, which are involved in the modulation of pain. Catechol-O-methyltransferase (COMT) is responsible for the metabolism and inactivation of native catecholamines such as dopamine, epinephrine, and norepinephrine. One relatively common base-pair substitution in the coding of COMT reduces the enzyme’s activity by three- to fourfold. This increase in endogenous catecholamines sensitizes patients to opioid ago- nists, lowering the morphine equivalents required for analgesia compared with patients with higher COMT activity, who may require at least doubling of the dose [35, 38–40]. Although the majority of research has studied morphine in this context, it is clear that the mu binding and thus dosing of any opioid will be altered [41–44]. The combined presence of genotypic variations at OPRM1 and COMT result in further complexities in opioid dose selection, which are partially described but undergoing further research at this time [45].

OPRM1 and COMT appear to be promising genotypic markers for determining opioid sensitivity and the dose required for analgesic response. Given the recent institution of manda- tory ceilings on opioid prescription quantities and doses, insur- ers are now less likely to fill the appropriate opioid prescription for patients with severe cancer pain in the setting of these known polymorphisms. Although opioid dose selection and titration should be driven by patient-reported clinical response, these test results may offer an objective measurement to rein- force rapid or slow dose titration and improve clinical care.

Barb now has painful neuropathy from her chemotherapy,

so she is started on gabapentin by a nurse practitioner. Accord-

ing to her known pharmacogenomic profile, there is no altered

metabolism predicted based on her results, so the gabapentin is

escalated to 3,600 mg daily per usual practice. At full dose, there

is no perceivable benefit in her neuropathy, and she begins to

develop mental status changes, so you taper the gabapentin and

consider another medication. Barb’s insurance company states

that she must next try either nortriptyline or amitriptyline for

painful chemotherapy-induced neuropathy. If the tricyclic antide-

pressant fails, only then will her insurance cover duloxetine.

Painful Neuropathy

Approximately 40% of patients treated withmore than one form of chemotherapy will have some form of peripheral neuropathy [46]. The neuropathy can have long-term effects on QOL [47]. The practice guideline by the American Society of Clinical Oncol- ogy (ASCO) for the management of chemotherapy-induced peripheral neuropathy suggests the use of duloxetine, tricyclic antidepressants (TCAs), or gabapentin [48]. Gabapentin metabo- lism is not significantly affected by known pharmacogenetic var- iations. However, duloxetine is inactivated by two liver enzymes, CYP2D6 and CYP1A2, whereas the TCAs have more complex pharmacogenomic considerations with CYP2D6 and CYP2C19.

Amitriptyline is metabolized by CYP2C19 into nortriptyline, whereas both agents require CYP2D6 for metabolism into less active compounds [49]. In a large study, CYP2D6 PMs given TCAs were substantially more likely than patients in the control group to stop the drug because of adverse effects such as dry mouth, dizziness, and cardiac concerns [50]. Alternatively, CYP2D6 UMs have an increased risk of therapeutic failure and

960 Suppor ve Care Pharmacogenomics in Oncology

©AlphaMed Press 2018

discontinuation when treated with amitriptyline [51], likely because they cannot hold sufficient amitriptyline or nortripty- line in the bloodstream to be effective. CPIC recommends avoiding TCAs completely in both PMs and UMs at CYP2D6, as well as avoiding amitriptyline and imipramine in CYP2C19 UMs and PMs [49].

The fact that Barb is a CYP2D6 PM and has ultrarapid

metabolism by CYP2C19 is concerning for the use of amitripty-

line or nortriptyline. Amitriptyline is metabolized to nortriptyline

very quickly by CYP2C19 in a patient like Barb. However, given

that her metabolism at CYP2D6 is slow, the nortriptyline is likely

to reach very high blood levels because of poor removal from

the system. You decide to avoid amitriptyline altogether and try

extremely low doses of nortriptyline, warning her to stop the

medicine at the first sign of any labeled side effects. She toler-

ates the 5 mg of nortriptyline but with no effect on her neurop-

athy. You increase the dose to 10 mg, and 3 days later she stops

the drug with complaints of dry mouth and severe headache.

With Barb’s pharmacogenomic test results in hand, you petition

the insurance company successfully to cover duloxetine. You

know that duloxetine requires some CYP2D6 for inactivation,

and Barb’s genotype would suggest she would be safest and

likely most successful starting at a low dose and titrating up

slowly based on response. Several years later, Barb returns for routine survivorship visit

to your office and admits, “I just feel so wiped out for the last

few days—I can barely get up to the bathroom.” You are paged

by the hematology lab urgently: her complete blood count

shows blasts and profound anemia. After hospital admission,

she is diagnosed with treatment-related acute myeloid leuke-

mia (AML). Given the poor prognosis, she starts standard chem-

otherapy and ultimately undergoes allogeneic bone marrow

transplant. In the post-transplant setting she will be maintained

on voriconazole for antifungal prophylaxis. You place the order

for the antifungal in the electronic medical record, and you get

an immediate prescriber alert that Barb has pharmacogenetic

test results that affect this order.

ANTIFUNGAL SELECTION Voriconazole is an antifungal agent that is used for treatment or prophylaxis of certain fungal infections. Appropriate serum concentrations are critical for effective prevention or treatment of invasive fungal infections (IFIs) [52, 53]. Studies have demon- strated that subtherapeutic voriconazole trough concentrations have been strongly associated with therapeutic failure [54]. Importantly, up to 50% of patients receiving the standard pro- phylactic dose of 200 mg twice daily remain subtherapeutic at steady state [55]. There is a significant association between IFI- related mortality and subtherapeutic initial trough concentra- tions—even when therapeutic blood level monitoring is used to direct subsequent dosing [52, 53, 56].

Importantly, CYP2C19 is responsible for the majority of voriconazole metabolism; thus, polymorphisms in this gene can have a significant effect on serum concentrations [57]. The patients at greatest risk of inadequate drug concentra- tions and thus voriconazole failure are those with rapid CYP2C19 metabolism, which occurs in up to 30%–35% of whites and blacks, such that the drug is removed from the bloodstream too quickly and can never reach therapeutic lev- els [54, 58–65]. Preliminary data show that, in a population of stem cell transplant patients, genotype-guided dosing for vor- iconazole prophylaxis (higher initial doses for CYP2C19 rapid and ultrarapid metabolizers) resulted in zero cases of subther- apeutic initial trough concentrations in this subset of patients compared with 80% in historical controls (p< .001) [66]. Another study showed reduced overall costs with genotype- directed dosing for patients with AML, even when including the tests of genomic analysis [67]. Currently, CPIC recom- mends that patients with rapid, ultrarapid, or poor metabo- lism at CYP2C19 should avoid voriconazole in favor of an alternative antifungal [58] (See Fig. 3.).

Ketoconazole, itraconazole, and isavuconazole clearance is highly dependent on CYP3A4 metabolism, and thus efficacy of these antifungal agents may be prone to variation by individual CYP3A4 genotype. As a start, studies have confirmed that the CYP3A4*22 allele results in significantly lower enzyme activity, impairing the metabolism of common CYP3A4-metabolized drugs [68, 69]. However, additional data are required to navi- gate the interactions between individual genotype and poten- tial CYP3A4-inducers or inhibitors that could be concomitantly administered.

Barb’s pharmacogenomic testing reveals she has ultrarapid

metabolism at CYP2C19—the key enzyme for voriconazole. You

consider starting her voriconazole dose higher, as suggested by

preliminary data from the genotype-directed dosing study.

However, per CPIC guidelines you ultimately decide to avoid vor-

iconazole completely and instead start isavuconazole for

Figure 3. Pharmacogenetic-driven treatment pathway for antifun- gal selection. CYP2C19 PMs, RMs, and UMs should avoid using voriconazole as primary prophylaxis or treatment for fungal infec- tions. CYP2C19 RMs and UMs are at risk of subtherapeutic concen- trations and increased risk of breakthrough fungal infection or lack of efficacy. CYP2C19 PMs are at risk of supratherapeutic concen- trations, which may increase the risk of related side effects. 1, Further dose adjustments or selection of alternative therapy

may be necessary because of other clinical factors, such as drug interactions, hepatic function, renal function, species, site of infec- tion, therapeutic drug monitoring, and comorbidities. 2, Some data suggest that higher initial doses of voriconazole in

CYP2C19 RMs and UMs may overcome subtherapeutic concentra- tions. Abbreviations: CYP2C19, cytochrome P450 2C19; IM, intermedi-

ate metabolizer; NM, normal metabolizer; PM, poor metabolizer; RM, rapid metabolizer; UM, ultrarapid metabolizer.

961Patel, Wiebe, Dunnenberger et al.

www.TheOncologist.com ©AlphaMed Press 2018

prophylaxis, given that this medication does not undergo

CYP2C19-mediated metabolism. After the diagnosis of poor-risk acute leukemia and months

of prolonged hospitalization for the bone marrow transplant,

Barb admits that she has been feeling depressed, losing weight,

and feeling hopeless in the last few weeks. You consult psychia-

try at the start of a long holiday weekend and they will see her

next week. However, she says, “I just want to start feeling bet-

ter as soon as I can—I can’t wait another day.” You feel the

need to start antidepressant therapy sooner, and the electronic

alert reminds you that Barb has prior pharmacogenomic testing

that will influence your decision.

ANTIDEPRESSANT SELECTION At least one quarter of all patients with cancer suffer frommajor depressive disorder. Recognizing this as a major comorbidity, ASCO created guidelines for screening, assessing, and treating depression in patients with cancer [70]. Standard response rates to antidepressants are 30%–50% regardless of what agent is selected [71]. There is a growing recognition that pharmacoge- nomic variationmay help explain some of the low response rates and incidence of adverse effects. Data now clearly justify the clin- ical utility of using an individual patient’s pharmacogenomic pro- file to select the best treatment for depression. (See Fig. 4.)

CYP2C19 plays a major role in the metabolism of citalo- pram, escitalopram, and sertraline. Poor metabolizers at CYP2C19 have been shown to be at increased risk of adverse events, including QT prolongation [72, 73]. Alternatively, UMs have lower plasma concentrations and are more likely to suffer from ineffectively treated depression [74]. CPIC recommends a 50% dose reduction in citalopram, escitalopram, and sertraline for CYP2C19 PMs and avoiding citalopram and escitalopram for CYP2C19 UMs [75]. For CYP2C19 UMs, sertraline can be pre- scribed at the recommended starting dose, but if a patient does not respond clinically, CPIC guidelines suggest considera- tion of an alternative drug not predominantly metabolized by CYP2C19.

Paroxetine and fluvoxamine are primarily metabolized by CYP2D6; thus, PMs are at increased risk of adverse effects, par- ticularly gastrointestinal [76, 77]. CYP2D6 UMs are at risk of poor drug response [78]. CPIC recommends avoiding paroxe- tine in CYP2D6 UMs and PMs and a 25%–50% dose reduction of fluvoxamine in CYP2D6 PMs [75]. Fluoxetine is metabolized by CYP2D6 and CYP2C19; however, there are few data associat- ing specific genetic variants with differences in clinical response to fluoxetine. The U.S. Food and Drug Administration (FDA) label highlights the potential for complicated drug-drug inter- actions in patients with reduced CYP2D6 function taking

Figure 4. Pharmacogenetic-driven treatment pathway for depression. Several antidepressants, including SSRIs, SNRIs, and TCAs, are avail- able to treat depressive symptoms in patients with adequate CYP2D6 and CYP2C19 activity (i.e., NM and IM patients). Patients with CYP2D6 and CYP2C19 variations (i.e., UM and PMs) are at a higher risk for altered antidepressant drug exposure. As such, treatment options become limited in these populations because of potential drug-gene interactions. The newer antidepressants, levomilnacipran, vilazodone, and vortioxetine, are not included on this algorithm but can be used regardless of CYP2C19 and CYP2D6 genotype. However, the maximum recommended daily dose of vortioxetine in CYP2D6 PMs is 10 mg according to the package insert. 1, Strong CYP2D6 inhibitors may result in poor metabolism. 2, Other genetic variants exist that influence response to SSRIs, particularly the serotonin transporter gene, SLC6A4. Reduced response

has been noted in patients carrying the S allele. 3, TCAs are not recommended for first-line therapy because of high incidence of adverse effects. Abbreviations: CYP2C19, cytochrome P450 2C19; CYP2D6, cytochrome P450 2D6; IM, intermediate metabolizer; NM, normal metabo-

lizer; PM, poor metabolizer; SNRI, serotonin and norepinephrine reuptake inhibitor; SSRI, selective serotonin reuptake inhibitor; TCA, tricy- clic antidepressant; UM, ultrarapid metabolizer.

962 Suppor ve Care Pharmacogenomics in Oncology

©AlphaMed Press 2018

fluoxetine [79]. Vortioxetine, a newer antidepressant with multimodal activity, is primarily metabolized by CYP2D6, but also 3A4/5, 2C9, 2C19, 2A6, 2C8, and 2B6 [80], and the FDA label recommends a maximum dose of 10 mg per day in known CYP2D6 PMs [81].

For serotonin and norepinephrine reuptake inhibitors, the evidence associating pharmacogenomic variation with clinical response is weaker than for selective serotonin reuptake inhibitors (SSRIs) but is growing. Venlafaxine is metabolized to multiple active metabolites by CYP2D6 and CYP2C19, although there is not enough evidence yet for a firm guide- line on prescribing [82, 83]. Additionally, genetic variations in serotonin-related genes may influence antidepressant effi- cacy; however, these may be less actionable, as no CPIC guidelines exist for these. For example, patients harboring the S allele for the serotonin transporter gene SLC6A4 may have reduced response to SSRIs. Polymorphisms in the sero- tonin receptor gene HTR2A have been associated with lack of response to SSRIs [84].

Multiple studies have recently been published illustrating the clinical value of multigene pharmacogenetic panels when treating patients with depression. At least four rigorous studies have shown significantly better treatment outcomes for major depressive disorder with pharmacogenomic guidance com- pared with the standard clinical approach [85, 86].

Given that Barb is a CYP2C19 UM, you know that sertra-

line, citalopram, or escitalopram will fail to reach adequate

concentration in the bloodstream and thus are likely to be

ineffective for her depression. Per CPIC guidelines, those med-

icines should be avoided in her case. As a known CYP2D6

PM, Barb could be at risk of excessive side effects if pre-

scribed paroxetine, as it requires CYP2D6 to be broken down

and removed from the blood stream. Safer and more effec-

tive options include desvenlafaxine, low-dose vortioxetine,

mirtazapine, and bupropion. Given that she is losing weight

and her insurance will not cover desvenlafaxine or vortioxe-

tine as first-line therapy, mirtazapine is an appropriate choice

in her case, starting with the lowest dose and titrating based

on clinical response, given that mirtazapine does undergo

some metabolic inactivation via CYP2D6.

CONCLUSION Pharmacogenomic data are important to understand interpa- tient variability in drug response to many supportive oncology medications. Barb’s case presented in this paper demonstrates the possibilities and power from the knowledge of just a few genes that influence the metabolism of many drugs. As these data grow, seemingly exponentially, with ever-cheaper analytic technology, it will soon be the standard of care to perform rou- tine pharmacogenomic testing on all patients with cancer prior to treatment. Ultimately the truest value of these data can only be fully realized when they are implemented into the routine workflowwith care pathways of health care providers and phar- macists on the ground.

As demonstrated in the case above, even two genes can have a major impact on medication management. Beyond CYP2D6 and CYP2C19, there are pharmacogenetic panels commercially available to analyze many more genes with the ability to minimize prescribing by trial and error. In addition

to writing drug and gene guidelines, CPIC creates supplemen- tary informatics resources to assist clinicians. These resources serve as clinical decision support tools to integrate pharma- cogenetic data into the electronic health record at the point of care—when the prescription is written [7]. The value of applying pharmacogenomics downstream, even years after initial testing—as in Barb’s case—depends on clinical decision support tools that are updated in real time to reflect the most recent evidence-based data. Effective integration with oncology workflow is critical and has been achieved at sev- eral prominent institutions [87]. The figures presented in this manuscript represent pharmacogenetic-guided treat- ment algorithms to select the so-called least genetically vul- nerable drug, by avoiding known drug-gene interactions based on presence of pharmacogenetic test results. Although not discussed in detail in this review, it is important to consider the role of pharmacogenomics in determining the magnitude of drug-drug interactions and drug-drug-gene interactions—that is, polymorphisms in a metabolic pathway and inhibition or induction of the same or minor pathway [88]. In fact, a cross-sectional study involving 22,885 patients found that there were approxi- mately 6,900 drug interactions, of which drug-drug-gene, drug-gene, and drug-drug interactions accounted for 22%, 25%, and 53%, respectively [89].

There will always be many demographic, biologic, psycho- logic, and pharmacologic variables that influence medication choice. Pharmacogenetic variation is an increasingly success- ful avenue for making objective choices about the safest and, at times, most effective treatments for patients with cancer. Ultimately, having an individual’s personalized genomic data at the point of care has significant implications for supportive oncology medication management throughout the care tra- jectory and can be integrated to personalize oncology care today.

ACKNOWLEDGMENTS

This clinical review was supported by Admera Health, South Plainfield, New Jersey.

AUTHOR CONTRIBUTIONS Conception/design: Jai N. Patel, Lauren A. Wiebe, Henry M. Dunnenberger, Howard L. McLeod

Provision of study material or patients: Jai N. Patel, Lauren A.Wiebe, Henry M. Dunnenberger, Howard L. McLeod

Collection and/or assembly of data: Jai N. Patel, Lauren A. Wiebe, Henry M. Dunnenberger, Howard L. McLeod

Data analysis and interpretation: Jai N. Patel, Lauren A. Wiebe, Henry M. Dunnenberger, Howard L. McLeod

Manuscript writing: Jai N. Patel, Lauren A. Wiebe, Henry M. Dunnenberger, Howard L. McLeod

Final approval of manuscript: Jai N. Patel, Lauren A. Wiebe, Henry M. Dunnenberger, Howard L. McLeod

DISCLOSURES

Jai N. Patel: Janssen Pharmaceuticals (C/A); Janssen Pharaceuticals, Myriad Genetics (RF), Admera Health (H); Henry M. Dunnenberger: Admera Health (H); Howard L. McLeod: Cancer Genetics, Inc. (SAB); Saladax, Admera Health (C/A); Interpares Biomedicine (OI). The other author indicated no financial relationships. (C/A) Consulting/advisory relationship; (RF) Research funding; (E) Employment; (ET) Expert

testimony; (H) Honoraria received; (OI) Ownership interests; (IP) Intellectual property rights/

inventor/patent holder; (SAB) Scientific advisory board

963Patel, Wiebe, Dunnenberger et al.

REFERENCES

1. Bakitas MA, Tosteson TD, Li Z et al. Early versus delayed initiation of concurrent palliative oncology care: Patient outcomes in the ENABLE III randomized controlled trial. J Clin Oncol 2015;33:1438–1445.

2. Dionne-Odom JN, Azuero A, Lyons KD et al. Benefits of early versus delayed palliative care to informal family caregivers of patients with advanced cancer: Outcomes from the ENABLE III randomized controlled trial. J Clin Oncol 2015; 33:1446–1452.

3. Temel JS, Greer JA, El-Jawahri A et al. Effects of early integrated palliative care in patients with lung and GI cancer: A randomized clinical trial. J Clin Oncol 2017;35:834–841.

4. Temel JS, Greer JA, Muzikansky A et al. Early pal- liative care for patients with metastatic non-small- cell lung cancer. N Engl JMed 2010;363:733–742.

5. El-Jawahri A, LeBlanc T, VanDusen H et al. Effect of inpatient palliative care on quality of life 2 weeks after hematopoietic stem cell transplantation: A randomized clinical trial. JAMA 2016;316:2094–2103.

6. Caudle KE, Klein TE, Hoffman JMet al. Incorpora- tion of pharmacogenomics into routine clinical prac- tice: The Clinical Pharmacogenetics Implementation Consortium (CPIC) guideline development process. Curr DrugMetab 2014;15:209–217.

7. Relling MV, Klein TE. CPIC: Clinical Pharmacoge- netics Implementation Consortium of the Pharma- cogenomics Research Network. Clin Pharmacol Ther 2011;89:464–467.

8. Patel JN. Cancer pharmacogenomics, challenges in implementation, and patient-focused perspec- tives. Pharmgenomics PersMed 2016;9:65–77.

9. Andersen RL, Johnson DJ, Patel JN. Personalizing supportive care in oncology patients using pharmacogenetic-driven treatment pathways. Phar- macogenomics 2016;17:417–434.

10. Trammel M, Roederer M, Patel J et al. Does pharmacogenomics account for variability in control of acute chemotherapy-induced nausea and vomit- ing with 5-hydroxytryptamine type 3 receptor antag- onists? Curr Oncol Rep 2013;15:276–285.

11. He H, Yin JY, Xu YJ et al. Association of ABCB1 polymorphisms with the efficacy of ondansetron in chemotherapy-induced nausea and vomiting. Clin Ther 2014;36:1242–1252.e1242.

12. Sadhasivam S, Zhang X, Chidambaran V et al. Novel associations between FAAH genetic variants and postoperative central opioid-related adverse effects. Pharmacogenomics J 2015;15:436–442.

13. Kaiser R, Tremblay PB, Sezer O et al. Investiga- tion of the association between 5-HT3A receptor gene polymorphisms and efficiency of antiemetic treatment with 5-HT3 receptor antagonists. Pharma- cogenetics 2004;14:271–278.

14. Kaiser R, Sezer O, Papies A et al. Patient-tailored antiemetic treatment with 5-hydroxytryptamine type 3 receptor antagonists according to cyto- chrome P-450 2D6 genotypes. J Clin Oncol 2002;20: 2805–2811.

15. Bell GC, Caudle KE,Whirl-Carrillo M et al. Clini- cal Pharmacogenetics Implementation Consortium (CPIC) guideline for CYP2D6 genotype and use of ondansetron and tropisetron. Clin Pharmacol Ther 2016;102:213–218.

16. Fisch MJ, Lee JW, Weiss M et al. Prospective, observational study of pain and analgesic prescribing in medical oncology outpatients with breast,

colorectal, lung, or prostate cancer. J Clin Oncol 2012;30:1980–1988.

17. Zhao F, Chang VT, Cleeland C et al. Determi- nants of pain severity changes in ambulatory patients with cancer: An analysis from Eastern Coop- erative Oncology Group trial E2Z02. J Clin Oncol 2014;32:312–319.

18. Owusu Obeng A, Hamadeh I, Smith M. Review of opioid pharmacogenetics and considerations for pain management. Pharmacotherapy 2017;37: 1105–1121.

19. Bernard S, Neville KA, Nguyen ATet al. Intereth- nic differences in genetic polymorphisms of CYP2D6 in the U.S. population: Clinical implications. The Oncologist 2006;11:126–135.

20. Ciszkowski C, Madadi P, Phillips MS et al. Codeine, ultrarapid-metabolism genotype, and post- operative death. N Engl JMed 2009;361:827–828.

21. Madadi P, Ciszkowski C, Gaedigk A et al. Genetic transmission of cytochrome P450 2D6 (CYP2D6) ultrarapid metabolism: Implications for breastfeeding women taking codeine. Curr Drug Saf 2011;6:36–39.

22. Madadi P, Joly Y, Avard D et al. Communi- cating pharmacogenetic research results to breastfeeding mothers taking codeine: A pilot study of perceptions and benefits. Clin Pharmacol Ther 2010;88:792–795.

23. Madadi P, Koren G. Pharmacogenetic insights into codeine analgesia: Implications to pediatric codeine use. Pharmacogenomics 2008;9:1267– 1284.

24. Gasche Y, Daali Y, Fathi M et al. Codeine intoxi- cation associated with ultrarapid CYP2D6 metabo- lism. N Engl JMed 2004;351:2827–2831.

25. Shaw KD, Amstutz U, Jimenez-Mendez R et al. Suspected opioid overdose case resolved by CYP2D6 genotyping.Ther DrugMonit 2012;34:121–123.

26. Voronov P, Przybylo HJ, Jagannathan N. Apnea in a child after oral codeine: A genetic variant - an ultra-rapid metabolizer. Paediatr Anaesth 2007;17: 684–687.

27. Crews KR, Gaedigk A, Dunnenberger HM et al. Clinical Pharmacogenetics Implementation Consor- tium guidelines for cytochrome P450 2D6 genotype and codeine therapy: 2014 update. Clin Pharmacol Ther 2014;95:376–382.

28. Andreassen TN, Eftedal I, Klepstad P et al. Do CYP2D6 genotypes reflect oxycodone requirements for cancer patients treated for cancer pain? A cross- sectional multicentre study. Eur J Clin Pharmacol 2012;68:55–64.

29. Stamer UM, Zhang L, Book M et al. CYP2D6 genotype dependent oxycodone metabolism in postoperative patients. PLoS One 2013;8:e60239.

30. Stauble ME, Moore AW, Langman LJ et al. Hydrocodone in postoperative personalized pain management: pro-drug or drug? Clin Chim Acta 2014;429:26–29.

31. Stamer UM, Musshoff F, Kobilay M et al. Con- centrations of tramadol and O-desmethyltramadol enantiomers in different CYP2D6 genotypes. Clin Pharmacol Ther 2007;82:41–47.

32. Stamer UM, St€uber F, Muders T et al. Respira- tory depression with tramadol in a patient with renal impairment and CYP2D6 gene duplication. Anesth Analg 2008;107:926–929.

33. Stamer UM, Lehnen K, H€othker F et al. Impact of CYP2D6 genotype on postoperative tramadol analgesia. Pain 2003;105:231–238.

34. Trescot AM, Datta S, Lee M et al. Opioid phar- macology. Pain Physician 2008;11(suppl 2):S133– S153.

35. Klepstad P, Rakvåg TT, Kaasa S et al. The 118 A>G polymorphism in the human mu-opioid receptor gene may increase morphine requirements in patients with pain caused by malignant disease. Acta Anaesthesiol Scand 2004;48:1232–1239.

36. Chou WY, Yang LC, Lu HF et al. Association of mu-opioid receptor gene polymorphism (A118G) with variations in morphine consumption for analge- sia after total knee arthroplasty. Acta Anaesthesiol Scand 2006;50:787–792.

37. Gong XD, Wang JY, Liu F et al. Gene polymor- phisms of OPRM1 A118G and ABCB1 C3435T may influence opioid requirements in Chinese patients with cancer pain. Asian Pac J Cancer Prev 2013;14: 2937–2943.

38. Lotta T, Vidgren J, Tilgmann C et al. Kinetics of human soluble and membrane-bound catechol O- methyltransferase: A revised mechanism and description of the thermolabile variant of the enzyme. Biochemistry 1995;34:4202–4210.

39. Rakvåg TT, Klepstad P, Baar C et al. The Val158- Met polymorphism of the human catechol-O- methyltransferase (COMT) gene may influence mor- phine requirements in cancer pain patients. Pain 2005;116:73–78.

40. Rakvåg TT, Ross JR, Sato H et al. Genetic varia- tion in the catechol-O-methyltransferase (COMT) gene and morphine requirements in cancer patients with pain. Mol Pain 2008;4:64.

41. Cajanus K, Kaunisto MA, Tallgren M et al. How much oxycodone is needed for adequate analgesia after breast cancer surgery: Effect of the OPRM1 118A>G polymorphism. J Pain 2014;15:1248–1256.

42. Fukuda K, Hayashida M, Ide S et al. Association between OPRM1 gene polymorphisms and fentanyl sensitivity in patients undergoing painful cosmetic surgery. Pain 2009;147:194–201.

43. Hayashida M, Nagashima M, Satoh Y et al. Analgesic requirements after major abdominal sur- gery are associated with OPRM1 gene polymor- phism genotype and haplotype. Pharmacogenomics 2008;9:1605–1616.

44. Matic M, Jongen JL, Elens L et al. Advanced cancer pain: The search for genetic factors correlated with interindividual variability in opioid requirement. Pharmacogenomics 2017;18:1133–1142.

45. Reyes-Gibby CC, Shete S, Rakvåg Tet al. Explor- ing joint effects of genes and the clinical efficacy of morphine for cancer pain: OPRM1 and COMT gene. Pain 2007;130:25–30.

46. Cavaletti G, Zanna C. Current status and future prospects for the treatment of chemotherapy- induced peripheral neurotoxicity. Eur J Cancer 2002; 38:1832–1837.

47. Hershman DL,Weimer LH,Wang A et al. Associ- ation between patient reported outcomes and quantitative sensory tests for measuring long-term neurotoxicity in breast cancer survivors treated with adjuvant paclitaxel chemotherapy. Breast Cancer Res Treat 2011;125:767–774.

48. Hershman DL, Lacchetti C, Dworkin RH et al. Prevention and management of chemotherapy-

964 Suppor ve Care Pharmacogenomics in Oncology

induced peripheral neuropathy in survivors of adult cancers: American Society of Clinical Oncology clini- cal practice guideline. J Clin Oncol 2014;32:1941– 1967.

49. Hicks JK, Sangkuhl K, Swen JJ et al. Clinical phar- macogenetics implementation consortium guideline (CPIC) for CYP2D6 and CYP2C19 genotypes and dos- ing of tricyclic antidepressants: 2016 update. Clin Pharmacol Ther 2016 [Epub ahead of print].

50. Bijl MJ, Visser LE, Hofman A et al. Influence of the CYP2D6*4 polymorphism on dose, switching and discontinuation of antidepressants. Br J Clin Pharmacol 2008;65:558–564.

51. Pe~nas-Lled�o EM, Trejo HD, Dorado P et al. CYP2D6 ultrarapid metabolism and early dropout from fluoxetine or amitriptyline monotherapy treat- ment in major depressive patients. Mol Psychiatry 2013;18:8–9.

52. Troke PF, Hockey HP, Hope WW. Observational study of the clinical efficacy of voriconazole and its relationship to plasma concentrations in patients. Antimicrob Agents Chemother 2011;55:4782–4788.

53. Park WB, Kim NH, Kim KH et al. The effect of therapeutic drug monitoring on safety and efficacy of voriconazole in invasive fungal infections: A randomized controlled trial. Clin Infect Dis 2012;55: 1080–1087.

54. Hamada Y, Seto Y, Yago K et al. Investigation and threshold of optimum blood concentration of voriconazole: A descriptive statistical meta-analysis. J Infect Chemother 2012;18:501–507.

55. Trifilio S, Pennick G, Pi J et al. Monitoring plasma voriconazole levelsmay be necessary to avoid subther- apeutic levels in hematopoietic stem cell transplant recipients. Cancer 2007;109:1532–1535.

56. Miyakis S, van Hal SJ, Ray J et al. Voriconazole concentrations and outcome of invasive fungal infec- tions. Clin Microbiol Infect 2010;16:927–933.

57. Mikus G, Scholz IM,Weiss J. Pharmacogenom- ics of the triazole antifungal agent voriconazole. Pharmacogenomics 2011;12:861–872.

58. Moriyama B, Obeng AO, Barbarino J et al. Clini- cal Pharmacogenetics Implementation Consortium (CPIC) guidelines for CYP2C19 and voriconazole ther- apy. Clin Pharmacol Ther 2016 [Epub ahead of print].

59. Hassan A, Burhenne J, Riedel KD et al. Modula- tors of very low voriconazole concentrations in rou- tine therapeutic drug monitoring. Ther Drug Monit 2011;33:86–93.

60. Hicks JK, Crews KR, Flynn P et al. Voriconazole plasma concentrations in immunocompromised pediatric patients vary by CYP2C19 diplotypes. Phar- macogenomics 2014;15:1065–1078.

61. Owusu Obeng A, Egelund EF, Alsultan A et al. CYP2C19 polymorphisms and therapeutic drug mon- itoring of voriconazole: Are we ready for clinical implementation of pharmacogenomics? Pharmaco- therapy 2014;34:703–718.

62. Pieper S, Kolve H, Gumbinger HG et al. Moni- toring of voriconazole plasma concentrations in immunocompromised paediatric patients. J Antimicrob Chemother 2012;67:2717–2724.

63.Wang T, Zhu H, Sun J et al. Efficacy and safety of voriconazole and CYP2C19 polymorphism for opti- mised dosage regimens in patients with invasive fungal infections. Int J Antimicrob Agents 2014;44: 436–442.

64. Lamoureux F, Duflot T,Woillard JB et al. Impact of CYP2C19 genetic polymorphisms on voriconazole dosing and exposure in adult patients with invasive fungal infections. Int J Antimicrob Agents 2015;47: 124–131.

65. Hamadeh IS, Klinker KP, Borgert SJ et al. Impact of the CYP2C19 genotype on voriconazole exposure in adults with invasive fungal infections. Pharmaco- genet Genomics 2017;27:190–196.

66. Teusink A, Vinks A, Zhang K et al. Genotype- directed dosing leads to optimized voriconazole lev- els in pediatric patients receiving hematopoietic stem cell transplantation. Biol Blood Marrow Trans- plant 2015;22:482–486.

67. Mason NT, Bell GC, Quilitz RE et al. Budget impact analysis of CYP2C19-guided voriconazole prophylaxis in AML. J Antimicrob Chemother 2015; 70:3124–3126.

68. Klein K, Zanger UM. Pharmacogenomics of cytochrome P450 3A4: Recent progress toward the “missing heritability” problem. Front Genet 2013;4: 12.

69. Okubo M, Murayama N, Shimizu M et al. CYP3A4 intron 6 C>T polymorphism (CYP3A4*22) is associated with reduced CYP3A4 protein level and function in human liver microsomes. J Toxicol Sci 2013;38:349–354.

70. Andersen BL, DeRubeis RJ, Berman BS et al. Screening, assessment, and care of anxiety and depressive symptoms in adults with cancer: An American Society of Clinical Oncology guideline adaptation. J Clin Oncol 2014;32:1605–1619.

71. Thase ME, Entsuah AR, Rudolph RL. Remission rates during treatment with venlafaxine or selective serotonin reuptake inhibitors. Br J Psychiatry 2001; 178:234–241.

72. Funk KA, Bostwick JR. A comparison of the risk of QT prolongation among SSRIs. Ann Pharmacother 2013;47:1330–1341.

73.Wang JH, Liu ZQ,WangWet al. Pharmacokinetics of sertraline in relation to genetic polymorphism of CYP2C19. Clin Pharmacol Ther 2001;70:42–47.

74. Huezo-Diaz P, Perroud N, Spencer EP et al. CYP2C19 genotype predicts steady state escitalo- pram concentration in GENDEP. J Psychopharmacol 2012;26:398–407.

75. 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 inhibi- tors. Clin Pharmacol Ther 2015;98:127–134.

76. Suzuki Y, Sawamura K, Someya T. Polymor- phisms in the 5-hydroxytryptamine 2A receptor and CytochromeP4502D6 genes synergistically predict fluvoxamine-induced side effects in Japanese depressed patients. Neuropsychopharmacology 2006;31:825–831.

77. Zourkov�a A, Ceskov�a E, Hadasov�a E et al. Links among paroxetine-induced sexual dysfunctions, gen- der, and CYP2D6 activity. J Sex Marital Ther 2007;33: 343–355.

78. Guzey C, Spigset O. Low serum concentrations of paroxetine in CYP2D6 ultrarapid metabolizers. J Clin Psychopharmacol 2006;26:211–212.

79. Prozac (R) [package insert]. Lilly USA, LLC. India- napolis, IN; January 2017.

80. Spina E, Santoro V. Drug interactions with vorti- oxetine, a new multimodal antidepressant. Riv Psi- chiatr 2015;50:210–215.

81. Brintellix (R) [package insert]. Takeda Pharma- ceuticals America, Inc. Deerfield, IL: September 2013.

82.Waade RB, Hermann M, Moe HL et al. Impact of age on serum concentrations of venlafaxine and escitalopram in different CYP2D6 and CYP2C19 genotype subgroups. Eur J Clin Pharmacol 2014;70: 933–940.

83. Altarelli M, Mancuso AP. Structural biology at the European X-ray free-electron laser facility. Philos Trans R Soc Lond B Biol Sci 2014;369:20130311.

84.Wilkie MJ, Smith G, Day RK et al. Polymor- phisms in the SLC6A4 and HTR2A genes influence treatment outcome following antidepressant ther- apy. Pharmacogenomics J 2009;9:61–70.

85. Altar CA, Carhart J, Allen JD et al. Clinical utility of combinatorial pharmacogenomics-guided antide- pressant therapy: Evidence from three clinical stud- ies. Mol Neuropsychiatry 2015;1:145–155.

86. P�erez V, Salavert A, Espadaler J et al. Efficacy of prospective pharmacogenetic testing in the treat- ment of major depressive disorder: Results of a randomized, double-blind clinical trial. BMC Psychia- try 2017;17:250.

87. Dunnenberger HM, Crews KR, Hoffman JM et al. Preemptive clinical pharmacogenetics imple- mentation: Current programs in five United States medical centers. Annu Rev Pharmacol Toxicol 2014; 55:89–106.

88. Bahar MA, Setiawan D, Hak E et al. Pharmaco- genetics of drug-drug interaction and drug-drug- gene interaction: A systematic review on CYP2C9, CYP2C19 and CYP2D6. Pharmacogenomics 2017;18: 701–739.

89. Hocum BT, White JR Jr, Heck JW et al. Cyto- chrome P-450 gene and drug interaction analysis in patients referred for pharmacogenetic testing. Am J Health Syst Pharm 2016;73:61–67.

Patel,Wiebe, Dunnenberger et al. 9

www.TheOncologist.com Oc AlphaMed Press 2018

964

©AlphaMed Press 2018 964 Suppor ve Care Pharmacogenomics in Oncology 1549490x, 2018, 8, D ow nloaded from https://theoncologist.onlinelibrary.w iley.com /doi/10.1634/theoncologist.2017-0599 by N ova Southeastern U niversity, W iley O nline L ibrary on [21/12/2022]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense

Expert paper writers are just a few clicks away

Place an order in 3 easy steps. Takes less than 5 mins.

Calculate the price of your order

Type of paper needed:

Pages:

You will get a personal manager and a discount.

Academic level:

We'll send you the first draft for approval by at

Total price:

$0.00