REVIEWS

T h e r a p e u T i c r e s i s Ta n c e

CYP2D6 and tamoxifen:

dnA matters in breast cancer
Janelle M. Hoskins*, Lisa A. Carey and Howard L. McLeod*

Abstract | Tamoxifen is the most widely used anti-oestrogen for the treatment of
hormone-dependent breast cancer. The pharmacological activity of tamoxifen is dependent
on its conversion by the hepatic drug-metabolizing enzyme cytochrome P450 2D6 (CYP2D6)
to its abundant metabolite, endoxifen. Patients with reduced CYP2D6 activity, as a result of
either their genotype or induction by the co-administration of drugs that inhibit CYP2D6
function, produce little endoxifen and seem to derive inferior therapeutic benefit from
tamoxifen. Here we review the existing data that relate CYP2D6 genotypes to response to
tamoxifen and discuss whether the analysis of the CYP2D6 genotype might be an early
example of a pharmacogenetic tool for optimizing breast cancer therapy.
Adjuvant treatment
Refers to additional treatment,
which is usually given after
surgery, in cases in which all
detectable disease has been
removed but there remains a
statistical risk of relapse.

*UNC Institute for

Pharmacogenomics and
Individualized Therapy,

Division of Haematology
and Oncology, and the

Division of Pharmacotherapy
and Experimental
Therapeutics, University of
North Carolina, Chapel Hill,
27599, North Carolina, USA.
Correspondence to H.L.M.
e-mail: hmcleod@unc.edu
doi:10.1038/nrc2683

It is well established that the clinical response to many

drugs varies widely among individuals1,2. In a group of
patients taking the same drug at the same dose for the
same disease indication, some will have a therapeutic
response and some a partial response, and others can
experience adverse events that are severe and, in rare
cases, life-threatening. Many factors can affect the efficacy and safety of a drug, including age, gender, genetics, environment (for example, co-administered drugs or
herbs), disease aetiology and renal and hepatic function.
Pharmacogenetics is the study of how genetic variations
affect the disposition of drugs, including their metabolism and transport and their safety and efficacy. The
number of genetic variants that have been demonstrated
by clinical trials to predict drug toxicity or efficacy is
increasing, and some of this knowledge is now being
applied in the clinic to tailor therapies to individuals on
the basis of their genetic makeup. It is hoped that this
will improve the safety and efficacy of treatments.
Breast cancer is a major public health issue, with
more than one million new cases observed around the
world in 2002 (Ref. 3). Breast cancers that express
the oestrogen receptor (ER) are referred to as ER+ and
are often dependent on oestrogen for growth. Selective
ER modulators inhibit oestrogen binding to ERs, reducing or eliminating oestrogen-driven proliferation of ER+
tumours, and are therefore effective therapies for the
treatment of breast cancer.
Tamoxifen, a selective ER modulator, is the most
widely used anti-oestrogen therapy for premenopausal and postmenopausal women with metastatic breast
cancer, for adjuvant treatment of primary breast cancer

and as a chemopreventive agent for women with a high

risk of developing breast cancer (BOXeS 1,2). Recent data
support the requirement for tamoxifen to be converted
to the abundant metabolite endoxifen for biological
activity 4. This conversion is catalysed by the polymorphic enzyme cytochrome P450 2D6 (CYP2D6). Among
healthy Europeans, 610% are deficient in CYP2D6
metabolism, which is an inherited trait 5. These individuals convert tamoxifen to endoxifen poorly 6 and
therefore may not derive full therapeutic benefit from
tamoxifen therapy 7. The use of CYP2D6 genotype
information to guide tamoxifen therapy therefore represents an early example of a pharmacogenetic tool for
optimizing anticancer efficacy. This Review presents
evidence for an association between tamoxifen and
CYP2D6 metabolizer status and discusses potential
clinical applications of these observations.

Oestrogen and breast cancer

ERs are members of the nuclear receptor family of ligand-dependent transcription factors. Mammals express
two ERs: ER (encoded by ESR1) and ER (encoded by
ESR2)8. The receptors have different biological functions
and are widely expressed, but they differ in their ligandbinding properties. The two proteins have different tissue expression patterns, although they are coexpressed in
many cell types including normal and neoplastic breast
tissue. ER is well established as a key transcriptional
regulator of both tissues, but the role of ER is not well
understood. This Review focuses on ER because of its
more established role in breast cancer, and we will now
refer to it as ER.

576 | AuguST 2009 | VoluME 9

www.nature.com/reviews/cancer
2009 Macmillan Publishers Limited. All rights reserved

REVIEWS
at a glance
The selective oestrogen receptor modulator tamoxifen is the most widely used
antioestrogen for the treatment of hormone-dependent breast cancer.
Hepatic, drug-metabolizing cytochrome P450s (CYPs) catalyse the oxidation of
tamoxifen to several metabolites. The metabolites, endoxifen and
4-hydroxytamoxifen, have greater binding affinities for oestrogen receptors and
suppress cell proliferation more effectively than tamoxifen does. Plasma
concentrations of endoxifen are considerably higher than those of
4-hydroxytamoxifen, suggesting that endoxifen is the main pharmacologically active
species of tamoxifen in vivo. The conversion of tamoxifen to endoxifen is
predominantly catalysed by CYP2D6.
Many polymorphisms in CYP2D6 have been identified. In Caucasian populations,
610% of people inherit two alleles containing polymorphisms and/or a gene
deletion, leading to no protein expression or the expression of a protein with no
CYP2D6 enzymatic activity. These individuals have impaired metabolism of CYP2D6
substrates and are called poor metabolizers of CYP2D6. Some drugs, such as the
antidepressants fluoxetine or paroxetine, are potent inhibitors of CYP2D6 and can
confer a poor metabolizer phenotype on individuals with normal CYP2D6 activity.
The findings of pharmacokinetic studies indicate that the conversion of endoxifen is
reduced in poor metabolizers of CYP2D6, either by genotype or by co-prescribed
fluoxetine or paroxetine, which are commonly prescribed to manage hot flashes.
Recent data suggest that poor metabolizers of CYP2D6 do not derive as much
benefit from tamoxifen therapy as other patients do; however, some studies have
yielded conflicting results.
The analysis of CYP2D6 genotype might represent an early example of a
pharmacogenetic tool for optimizing breast cancer therapy; however, the findings of
larger, well-designed studies that support the current data are necessary before a
change in clinical practice is advocated.

Haem-thiolate enzymes
The collective name given to a
class of haemoproteins in
which a thiolate group
(typically from a cysteine
residue) is the axial ligand of
haem iron.

Microscopy studies show that ERs reside in the

nucleus and in small pools in the plasma membrane
and cytoplasm. ERs function in the nucleus as transcriptional regulators of oestrogen-responsive genes9
and are activated by oestrogen binding to their ligandbinding domain (fIG. 1). This induces their phosphorylation by kinases including the MAPKs ERK1, ERK2
and p38MAPK, the cyclin-dependent kinases CDK2 and
CDK7, Src, protein kinase A, ribosomal protein S6
kinase-1 (KS6A1) and the Akt kinases. This phosphorylation alters their conformation and triggers dimerization and the recruitment of co-activator proteins, such
as nuclear receptor co-activator 1 (NCoA1; also known as
SRC1), glutamate receptor-interacting 1 (gRIP1) and
nuclear receptor co-activator 3 (NCoA3), to the oestrogenbound ER complex 9,10. The DNA-binding domain of the
complex then binds to oestrogen response elements in
the promoter region of oestrogen-responsive genes, such
as those encoding insulin receptor substrate 1 (IRS1) and
cyclin D1 (RefS 9,10), and initiates their transcription.
ER in the nucleus can also indirectly regulate the transcription of target genes by proteinprotein interactions
with other transcription factors, such as SP1 and activator protein 1 (AP1), which regulate cell division, angiogenesis and survival. This can lead to sustained breast
cancer growth and progression10. These functions are
considered the classic or genomic actions of ER. Nongenomic activities of ER or membrane-initiated steroid
signalling that occur in the cytoplasm or cell membrane
have also been described, and these activities include
the mobilization of intracellular Ca2+, stimulation of

adenylyl cyclase activity and cyclic AMP production10.

The activation of extranuclear ERs leads to the activation of tyrosine kinases, such as epidermal growth factor
receptor (EgFR), insulin-like growth factor and ERBB2,
and the activation of downstream signalling cascades that
result in cell proliferation and survival9,10.

Tamoxifen metabolism
oral tamoxifen citrate has been the mainstay of endocrine
therapy of breast cancer for the past 25 years and has been
approved for the treatment of many ER+ breast cancerrelated indications (BOX 1). Tamoxifen binds to the ligand-binding domain of an ER and blocks the binding of
oestrogen (fIG. 1). This prevents conformational changes
of the ER that are required for its association with coactivators and leads to the preferential recruitment of
co-repressor proteins, such as nuclear receptor corepressor 1 (NCoR1), which block the transcriptional
activation functions of the ER and subsequent tumour
growth10,11.
The metabolism of tamoxifen is complex and involves
hepatic phase I and II enzymes (fIG. 1). Tamoxifen is
metabolized by hepatic cytochrome P450s (CYPs) to
produce its major, primary metabolites in the plasma:
N-desmethyltamoxifen and 4-hydroxytamoxifen, which
are principally formed by the CYP3A4 (and CYP3A5)
and CYP2D6 enzymes, respectively 12,13. oxidation
of these metabolites results in the formation of the
abundant and pharmacologically active metabolite,
4-hydroxy-N-desmethyltamoxifen (endoxifen)13.
The anti-oestrogen activities of endoxifen and
4-hydroxytamoxifen are similar in terms of their binding
affinities to ER and ER, suppression of ER-dependent
proliferation of breast cancer cells and modulation of
ER-mediated global gene expression1417. A recent study
showed that tamoxifen and 4-hydroxytamoxifen stabilize ER in breast cancer cells4, and endoxifen reduces
ER protein levels by targeting it for degradation by
proteasomes. Taken together, these findings, as well
as the clinical observation that endoxifen plasma concentrations are around 510-fold higher than those of
4-hydroxytamoxifen, suggest that endoxifen is probably the crucial metabolite responsible for the in vivo
pharmacological activity of tamoxifen14,16,18.
cYp2D6 and tamoxifen
CYP2D6 background. CYPs are membrane-bound
haem-thiolate enzymes involved in the oxidative, perioxidative and reductive metabolism of various molecules. More
than 50 CYP genes, each of which encodes a different
CYP protein product, have been identified in humans19.
CYP2D6 is predominantly expressed in the liver and is
involved in the metabolism of many commonly prescribed
drugs, including antidepressants, anti-arrhythmics, antipsychotics, -blockers and tamoxifen5. CYP2D6 is located
on chromosome 22q13.1, and polymorphisms in this gene
can significantly affect enzymatic activity.
To date, more than 75 CYP2D6 variant alleles have
been reported20 (see CYP2D6 allele nomenclature website). Many polymorphisms seem to be silent and result
in alleles that express proteins with normal CYP2D6

NATuRE REVIEwS | CanCer

VoluME 9 | AuguST 2009 | 577

2009 Macmillan Publishers Limited. All rights reserved

REVIEWS
Box 1 | clinical use of tamoxifen
Tamoxifen has been approved by the US Food and Drug Administration for several therapeutic and preventive options in
breast cancer. Indications include metastatic, adjuvant therapy for oestrogen receptor-positive (ER+) and progesterone
receptor-positive disease, and chemoprevention in high-risk women11,56,57.
Most (75%) breast cancers are ER+ and, of those cases, most (65%) are also progesterone receptor-positive. By contrast,
35% of breast cancers are oestrogen receptor-negative (ER) but progesterone receptor-positive. The progesterone
receptor gene is activated by oestrogen in normal reproductive tissues and breast cancer, and patients with breast
cancer may benefit from anti-oestrogen therapy and are therefore treated with tamoxifen.
Tamoxifen is effective in approximately 70% of ER+ and/or progesterone receptor-positive tumours; however, 3050%
of patients relapse58.
Tamoxifen is well tolerated by most patients. The most common adverse effects experienced are the symptoms of
menopause, including hot flashes and atrophic vaginitis, which are more commonly reported in premenopausal than in
postmenopausal women. Rarely, patients can experience cataracts and vascular-related thrombotic events, including
stroke, deep vein thrombosis and pulmonary emboli. They may also develop certain cancers including endometrial cancer
and uterine sarcoma59.
The activity of tamoxifen as an oestrogen agonist is beneficial in some tissues, including favourable effects on serum
cholesterol (which was recently shown to be associated with genetic variation in the genes encoding oestrogen
receptors 1 and 2 and not with the cytochrome P450 2D6 (CYP2D6) gene60) and protection against bone loss and
cardiovascular disease. However, in other tissues the oestrogenic effects of long-term tamoxifen use are associated with
life-threatening adverse events, including invasive endometrial cancer and thromboembolic disease59. These adverse
events limit the duration of therapy to 5 years.
Tamoxifen has US Food and Drug Administration approval for administration at doses from 10 mg to 40 mg per day.
Typical doses for advanced breast cancer in women and men are 20 mg or 40 mg per day. For adjuvant therapy,
the typical dose is 20 mg per day for 5 years following tumour resection.
Following oral administration, the elimination of tamoxifen is biphasic and dependent on the cumulative dose61. The
terminal elimination half-life of tamoxifen for a single dose is 57 days, and the time to reach steady state is 34 weeks61.
Tamoxifen is predominantly excreted in the faeces61.

activity, which are known as extensive metabolizer (EM)

alleles (TABLe 1). Null alleles, which are also known as
poor metabolizer (PM) alleles, are less common (TABLe 1)
and carry a gene deletion or polymorphisms that lead
to no protein expression or the expression of a protein
with no CYP2D6 enzymatic activity. A third class of alleles includes polymorphisms that reduce enzyme activity, which are known as intermediate metabolizer (IM)
alleles (TABLe 1). Individuals with two PM alleles have
impaired metabolism of CYP2D6 substrates and are
classified as poor metabolizers5. Individuals with EM/
PM or EM/EM genotypes have normal metabolism of
CYP2D6 substrates and are classified as extensive or
normal metabolizers. IM/IM or IM/PM individuals
have CYP2D6 enzyme activity that is between extensive
and poor metabolizers and are referred to as intermediate metabolizers21. CYP2D6 amplification has also
been observed in all the populations studied. up to 13
copies of CYP2D6 have been observed in otherwise
healthy individuals. gene duplication or multiplication
of alleles with normal activity, which are known as ultrarapid metabolizer (uM) alleles (TABLe 1), are associated

Box 2 | premenopausal and postmenopausal breast cancer

Young women are likely to present with breast cancer at a higher stage than their older
counterparts.
Premenopausal breast cancers are more likely to be oestrogen receptor-negative and
more aggressive than postmenopausal breast cancers.
Adjuvant tamoxifen is the drug of choice for premenopausal women, but recent
research suggests that tamoxifen is inferior to aromatase inhibitors in the adjuvant
setting for postmenopausal women6265.

with ultra-rapid metabolism of CYP2D6 substrates5.

Multiplications have also been reported for IM and PM
alleles, as have alleles with unknown activity 22 (TABLe 1).
The frequencies of PM and IM alleles vary among ethnic
groups and are presented in BOX 3.
CYP2D6 genotype and endoxifen production. Several
pharmacokinetic studies have demonstrated that the
CYP2D6 genotype has an important role in the in vivo
formation of endoxifen from tamoxifen6,14,18. In a study
of 80 patients with newly diagnosed breast cancer, Jin
and colleagues measured the tamoxifen and endoxifen
plasma concentrations following initiation of adjuvant
tamoxifen therapy 6. After 4 months of therapy, patients
with a poor metabolizer genotype (PM/PM) had mean
endoxifen concentrations between fourfold and twofold
lower than patients with EM/EM and EM/PM genotypes,
respectively, which was suggestive of a gene-dose effect.
Reduced-activity alleles of CYP2D6 and tamoxifen
metabolism. As enzymes encoded by IM alleles have
lower catalytic activity than those encoded by EM alleles, it follows that patients who carry IM alleles might
also have impaired conversion of tamoxifen to endoxifen
compared with patients with normal metabolizer genotypes. Indeed, in vitro, CYP2D6.10 (the enzyme product
of the IM allele CYP2D6*10) has a 15-fold lower rate of
conversion of N-desmethyltamoxifen to endoxifen
than has CYP2D6.1 (the product of the EM allele,
CYP2D6*1) 23 . This finding was confirmed by a
pharmacokinetic study of 202 Korean women who
received adjuvant tamoxifen or tamoxifen therapy
for metastatic breast cancer 24. The authors found that

578 | AuguST 2009 | VoluME 9

www.nature.com/reviews/cancer
2009 Macmillan Publishers Limited. All rights reserved

REVIEWS

Glucuronide
UGTs
Tamoxifen N-oxide

FMO3
FMO1

Tamoxifen

UGTs
Glucuronide

CYP2D6

CYP2C9

UGTs

CYP3A4

CYP3A5
CYP2C9

SULT1A1
CYP3A4

UGTs
Endoxifen
SULT1A1

CYP3A5
CYP2D6

N-didesmethyltamoxifen

Sulphate

CYP3A5

CYP2D6
N-desmethyltamoxifen

UGTs

Glucuronide

4-Hydroxytamoxifen

CYP1A2

CYP2D6

CYP3A4

Glucuronide

CYP3A4

CYP2C19
CYP2C19

Liver cell

CYP2B6

UGTs

Glucuronide
Sulphate

Glucuronide

Metabolite E
SULT1A1

Sulphate

Oestrogen
Nucleus

Breast cancer
cell

ER ER Co-activator
ERE

Cell division
Angiogenesis
Tumour growth

Figure 1 | Partial metabolic pathway of tamoxifen and its interaction with oestrogen receptors (ers). The
Nature Reviews
| Cancer
primary pathways of tamoxifen metabolism in the liver are catalysed by cytochrome P450s (CYPs), including
CYP3A4,
CYP3A5, CYP2C9, CYP2C19, CYP1A2, CYP2B6 and CYP2D6, and flavin-containing monooxygenases (FMOs), including
FMO1 and FMO3 (shown in blue)12,13,55. The enzymes that are key for each metabolic pathway are shown in bold. Tamoxifen
metabolism to N-desmethyltamoxifen is catalysed predominantly by CYP3A4 and CYP3A5, and metabolism to
4-hydroxytamoxifen is catalysed mainly by CYP2D6. The formation of these metabolites accounts for ~92% and ~7% of
primary tamoxifen oxidation, respectively13. Both of these metabolites are converted to 4-hydroxy-N-desmethyltamoxifen
(endoxifen)13. Endoxifen formation from N-desmethyltamoxifen is almost exclusively catalysed by CYP2D6, and formation
from 4-hydroxytamoxifen by CYP3A4 and CYP3A5 (Ref. 13). Tamoxifen and its metabolites undergo phase II conjugation
reactions, including glucuronidation and sulphation. In a breast cancer cell (shown in blue), oestrogen binds to the ER in
the nucleus, leading to phosphorylation and dimerization. The complex recruits co-activators and binds to a specific DNA
sequence, called the oestrogen response element (ERE), which is present in oestrogen-responsive genes. Binding of the ER
dimer causes transcriptional activation of these genes. Subsequent translation produces proteins that are important for
cell division, angiogenesis and survival, leading to sustained breast cancer growth and progression. This function is
considered the classic action of ERs. SULT1A1, sulphotransferase 1A1; UGT, uridine diphosphate glucuronosyltransferase.

intermediate metabolizers (IM/IM) had endoxifen plasma

concentrations around twofold lower than those of
patients with normal metabolizer genotypes.
CYP2D6 inhibitors and tamoxifen metabolism. Certain
selective serotonin reuptake inhibitors (SSRIs), such as
fluoxetine and paroxetine, together with the selective
noradrenaline reuptake inhibitors (SNRIs), are commonly
co-prescribed with tamoxifen to alleviate hot flashes, which
are a common and undesirable side effect of tamoxifen25,26.
However, these drugs are also potent inhibitors
of CYP2D6 (Ref. 27). The effect of CYP2D6 inhibitors on
endoxifen formation was assessed in a small, prospective
pharmacokinetic trial of 12 women with breast cancer
who were taking adjuvant tamoxifen. The plasma concentrations of endoxifen decreased by around twofold

after 4 weeks of paroxetine exposure, confirming that

CYP2D6 mediates endoxifen formation. Therefore, use
of CYP2D6 inhibitors such as SSRIs and SNRIs may
negatively affect the efficacy of tamoxifen14.
In a second study, patients with an EM/EM genotype,
who were taking co-medications that are known
to inhibit CYP2D6 (for example, SSRIs), had mean
endoxifen plasma concentrations of almost half the
concentration of patients who were not taking CYP2D6
inhibitors6. The same effect of CYP2D6 inhibitors on
endoxifen plasma concentrations was observed for
patients with EM/PM genotypes, although the effect
was not significant. This supports the idea that low
CYP2D6 activity, through genetic polymorphisms
or drug interactions, leads to low levels of the active
tamoxifen metabolite.

NATuRE REVIEwS | CanCer

VoluME 9 | AuguST 2009 | 579

2009 Macmillan Publishers Limited. All rights reserved

REVIEWS
Table 1 | cytochrome p450 2D6 (CYP2D6) alleles and their effects on cYp2D6 enzyme activity
CYP2D6 alleles

allele designation

enzyme activity

allele abbreviation

*1, *2, *33, *35

Normal or wild type

Normal

EM

*3, *4, *5-*8, *11-*16, *18-*21, Null

*36, *38, *40, *42, *44, *56, *62

No protein, inactive
or negligible

PM

*9, *10, *17, *29, *41, *59

Reduced activity

Decreased

IM

*22-*28, *30-*32, *34, *37,

*39, *43, *45-*55

Unknown activity

Unknown

Not applicable

*1N, *2N, *35N

Multiplication of normal alleles

Increased

UM

*10N, *17N, *29N, *41N

Multiplication of reduced activity alleles

Decreased

IM

*4N, *6N, *36N

Multiplication of null alleles

Inactive or negligible

PM

*43N, *45N

Multiplication of alleles of unknown

activity

Unknown

Not applicable

Duplicated alleles

EM, extensive metabolizer allele; IM, intermediate metabolizer allele; PM, poor metabolizer allele; UM, ultra-rapid metabolizer allele.

A follow-up study of 158 patients with breast

cancer who were receiving adjuvant tamoxifen examined patients for 33 CYP2D6 alleles, including many rare
variants18. Consistent with previous studies, patients who
were taking CYP2D6 inhibitors had lower endoxifen
plasma concentrations than patients who were not taking inhibitors6,14. This study showed that there were
strong concordances between patient genotype and
concomitant CYP2D6 inhibitors and endoxifen plasma
concentrations; however, some variability in endoxifen
plasma concentrations remained unaccounted for. This
residual variability may partly be explained by unidentified common CYP2D6 variants that alter CYP2D6
activity, possibly in the CYP2D6 promoter region,
which could regulate gene expression. Another possible explanation is inter-patient variation in the activity of other enzymes that are involved in the formation
of endoxifen, for example CYP3A4, CYP3A5 and
CYP2C19 (RefS 13,20,28). Patients carrying one or two
copies of CYP2C19*17, which is a CYP2C19 allele with
higher activity than the wild-type allele29, had a better
prognosis than other patients in a study of European
women who were receiving adjuvant tamoxifen 28.
However, in a study of Japanese women taking adjuvant
tamoxifen, recurrence-free survival was similar between
patients with two null CYP2C19 alleles (CYP2C19*2 or
CYP2C19*3) and impaired metabolism of CYP2C19
substrates30 and other patients31. It is therefore unclear
whether CYP2C19 genotype influences the formation of
endoxifen and patient response to tamoxifen.
Endoxifen plasma concentrations are dependent on
the formation of this metabolite and also its clearance by
phase II enzymes, including processes such as sulphation
and possibly glucuronidation32,33. Individual variation
in the activity of phase II enzymes may partly explain
some of the residual variability in endoxifen concentrations. This has been investigated for sulphotransferase
1A1 (SULT1A1) genotypes in patients, but no association with endoxifen concentrations has been demonstrated6,34 (see Supplementary information S1 (table)).
The influence of uridine diphosphate glucuronosyltransferase (ugT) polymorphisms on the glucuronidation of

trans-endoxifen has been recently evaluated in vitro35.

The common ugT2B7 codon 268 polymorphism
(histidine to tyrosine) and the rare ugT1A8 codon
277 polymorphism (cysteine to tyrosine) both altered
O-glucuronidation of trans-endoxifen. These polymorphisms could alter the elimination of endoxifen, thereby
contributing to individual variability in endoxifen plasma
concentrations and patient response to tamoxifen, and
they may warrant further investigation.
The SNRI venlafaxine is a weak inhibitor of
CYP2D6 and does not seem to have the same impact
on endoxifen formation as paroxetine and fluoxetine do36. other antidepressants, such as duloxetine
and bupropion, are moderate inhibitors of CYP2D6,
whereas the SSRIs sertraline, citalopram, escitalopram
and fluvoxamine, as well as the SNRI reboxetine, are
weak or negligible inhibitors of CYP2D6 and are less
likely to interact with tamoxifen than fluoxetine or
paroxetine36. Sertraline and citalopram are used to
treat hot flashes, but the other agents are not routinely
used. The anti-arrhythmic agent quinidine is also a
potent inhibitor of CYP2D6 and might reduce the formation of endoxifen if co-prescribed with tamoxifen37.
The effects of co-prescribed inhibitors of CYP2D6 on
the efficacy of tamoxifen will be addressed later in
this Review.
CYP2D6 genotype and clinical outcome. The clear
effect of CYP2D6 activity, as determined by either
genotype or environment, on tamoxifen pharmacokinetics also translates into an effect on clinical outcome ( TABLe 2; see Supplementary information S2
(table)). goetz et al. 38 used the uS North Central
Cancer Treatment group adjuvant breast cancer trial
(89-30-52 trial), a randomized Phase III clinical trial in
postmenopausal women with resected ER + breast
cancers who were treated with tamoxifen, to evaluate
the hypothesis that CYP2D6 genotype influences the
outcome of tamoxifen therapy. A subset of patients
was genotyped for the most common PM allele in
European populations, CYP2D6*4, and the rarer PM
allele CYP2D6*6. In a univariate analysis, patients with

580 | AuguST 2009 | VoluME 9

www.nature.com/reviews/cancer
2009 Macmillan Publishers Limited. All rights reserved

REVIEWS
Hazard ratio
In a survival analysis, this is the
effect of an explanatory
variable on the hazard risk of
an event.

Allozyme
Variant forms of an enzyme
that are encoded by different
alleles at the same locus.

the PM/PM genotype had worse relapse-free time and

disease-free survival than other patients. However, a
trend could be seen only when using a multivariate
analysis. The findings suggest that poor metabolizers
(women with a PM/PM genotype) may not derive as
much therapeutic benefit from tamoxifen and are at a
higher risk of breast cancer recurrence than extensive
metabolizers.
Findings from other clinical trials have further
confirmed a role for the CYP2D6 genotype in the
activation of tamoxifen and the likelihood of therapeutic benefit from testing for CYP2D6 genotype by
showing that women with intermediate metabolizer
genotypes also derive less therapeutic benefit from
tamoxifen than women with normal metabolizer
genotypes do. CYP2D6 genotype was evaluated as a
prognostic marker in a german study of women diagnosed with ER+ primary invasive breast cancer who
either received adjuvant tamoxifen monotherapy or
did not receive such therapy 28. In support of the findings of goetz et al.7,38, tamoxifen-treated patients with
intermediate or poor CYP2D6 activity (PM/PM, EM/
PM, IM/IM or IM/PM genotypes) had shorter relapsefree times and event-free survivals than patients with

Box 3 | ethnic differences in reduced- and null-activity CYP2D6 alleles

Among the European population, 610% have two poor metabolizer (PM) alleles,
leading to no or negligible cytochrome P450 2D6 (CYP2D6) enzyme activity, and these
individuals consequently have impaired metabolism of CYP2D6 substrates and are
classified as poor metabolizers of CYP2D6. The total frequency of PM alleles in the
Caucasian population is 0.26 (Ref. 21). In a population of Europeans from Germany,
97% of individuals identified as poor metabolizers by a standard phenotyping test
using a probe substrate of CYP2D6 were explained by four PM alleles, CYP2D6*3, *4,
*5 and *6 (Ref. 66). The most common PM allele in this ethnic group is CYP2D6*4,
which is carried by ~75% of European poor metabolizers. Its allelic frequency is 0.20 in
Europeans, 0.010.02 in East Asians (including Koreans, Chinese and Japanese) and
0.020.07 in Africans21. The allele has a substitution at position 1846 of G to A, which
causes a splicing defect that results in the production of a protein with no enzyme
activity67.
CYP2D6 PM alleles are rare in East Asian populations, which explains the low
incidence of CYP2D6 poor metabolizers in this ethnic group (<1%)5. The frequency of
intermediate metabolizer (IM) alleles, namely CYP2D6*10, is higher in East Asian
groups Chinese = 0.56, Koreans = 0.45 and Japanese = 0.38 than in Europeans, for
whom it is <0.02 (RefS 21,68,69). The enzyme product of CYP2D6*10, denoted
CYP2D6.10, carries a proline-to-serine substitution at amino acid 34 (c.100C>T) in its
amino terminus. This allozyme has less stability and substantially lower turnover of
CYP2D6 substrates than the enzyme product of the extensive, or normal, metabolizer
(EM) wild-type allele (CYP2D6*1), CYP2D6.1 (Ref. 70). The higher frequency of
CYP2D6*10 in East Asian groups than in Europeans explains the greater incidence of
intermediate metabolizers in East Asia5.
PM alleles are also less common in individuals of African descent than in Europeans,
and the combined frequency of PM alleles in African populations is 0.050.08 (Ref. 21).
However, IM alleles are more common in African populations. The IM allele CYP2D6*17
carries three coding-region single nucleotide polymorphisms that confer the amino
acid substitutions T107I, R296C and S486T. Its enzyme product, CYP2D6.17, also has
lower turnover of typical CYP2D6 substrates than CYP2D6.170. The CYP2D6*17 allele
occurs at moderate to high frequencies in African populations (frequency = 0.090.34)21.
A second IM allele, CYP2D6*29 (which causes V136I, R296C, V338M and S486T amino
acid substitutions relative to CYP2D6.1), has also been found at a relatively high
frequency in a Tanzanian population (0.20). This frequency combined with that of
CYP2D6*17 gives a total frequency of the African-specific IM alleles in this population
that is reported to be 0.37 (Ref. 71).

normal CYP2D6 metabolism (EM/EM or IM/EM

genotypes) did. In the group that was not treated
with tamoxifen, CYP2D6 genotype did not influence
survival, suggesting that CYP2D6 activity predicts the
outcome of tamoxifen therapy but is not prognostic for
breast cancer. In a study from the Netherlands, patients
with a PM/PM genotype had increased breast cancer
mortality, but they did not show higher mortalities due
to any cancer or cause than patients with the EM/EM
genotype39. Three more studies found that patients with
IM/IM genotypes have a poor response to tamoxifen.
A small prospective Korean study of premenopausal
and postmenopausal patients with metastatic breast
cancer who were taking tamoxifen found that patients
with the IM/IM genotype (which is common in this
ethnic group) had a shorter time to disease progression
than other patients24. Similarly, a Japanese study investigating premenopausal and postmenopausal women
with invasive ER+ or progesterone receptor-positive
breast cancer treated with adjuvant tamoxifen monotherapy for 5 years found that the IM/IM genotype was
predictive of a worse disease recurrence40. A study of
Chinese women treated with adjuvant tamoxifen also
found that, compared with other patients, those with
an intermediate metabolizer genotype had a reduced
disease-free survival41. The results of these studies suggest that the activation of tamoxifen is also impaired
in intermediate metabolizers. In addition, the results
of the Korean study 24 suggest that CYP2D6 metabolic
capacity may be important not only in the adjuvant
treatment of postmenopausal women with early-stage
breast cancer, but also in the metastatic setting.
CYP2D6 genotype might also affect the efficacy
of tamoxifen when the drug is administered as a
chemopreventive agent. The findings of a large pilot
European chemoprevention study suggest that, in this
clinical setting, women with poor metabolizer genotypes were less likely to derive therapeutic benefit from
the drug 42.
goetz et al. also evaluated whether co-prescribed
inhibitors of CYP2D6, such as SSRIs, reduced the
efficacy of tamoxifen7,38. After adjusting for tumour
size and nodal status, the authors found that patients
with low CYP2D6 metabolism (PM/PM and EM/PM
genotypes or EM/EM patients taking a known potent
or moderate CYP2D6 inhibitor) had shorter times to
breast cancer recurrence, as well as relapse-free and
disease-free survivals. However, there was no effect
on overall survival compared with normal metabolizers (patients with an EM/EM genotype who were
not co-prescribed a CYP2D6 inhibitor)7. Although
few poor metabolizers have been studied in each
trial, and the effect of CYP2D6 metabolizer status
on response to tamoxifen-based therapy is moderate ( hazard ratio ~2) and of marginal significance
(p<0.05 to 0.01) (TABLe 2), the findings of these studies further support the observations of pharmacokinetic studies that CYP2D6 genotype has a role in
the activation of tamoxifen. These studies also suggest that patients with decreased CYP2D6 metabolism, whether as a result of genetic or environmental

NATuRE REVIEwS | CanCer

VoluME 9 | AuguST 2009 | 581

2009 Macmillan Publishers Limited. All rights reserved

REVIEWS
Table 2 | Summary of clinical studies that have evaluated the association between CYP2D6 genotype and response to tamoxifen therapy
N

First
author

Tamoxifen
therapy

Menopausal and
tumour status

Subgroup selected
for analysis

Subgroup Outcome Univariate

N
hazard ratio*

Multivariate
hazard ratio*

European populations
Goetz

256 Monotherapy

Postmenopausal
and ER+

190

RFS
DFS

2.71 (1.156.41])|| 1.85 (0.764.52)

2.44 (1.224.90)|| 1.86 (0.913.82)

Schroth28

486 Monotherapy or
no tamoxifen

Postmenopausal
and ER+, or ER

206

RFT
EFS

Not reported
Not reported

2.24 (1.164.33)||
1.89 (1.103.25)||

Goetz

256 Monotherapy

Postmenopausal
and ER+

180

RFS
DFS

Not reported
Not reported

2.69 (1.345.37)#
2.44 (1.274.69)#

108 Not reported**

Not reported

85

BCS

Not reported

4.1 (1.115.9)||

226 + Chemotherapy
or radiation or
no tamoxifen +
chemotherapy or
radiation

Premenopausal, or
postmenopausal
and ER+, or ER

CYP2D6*1/*1

107

DRFS

Not reported

0.91 (0.531.57)

CYP2D6*1/*4 and
CYP2D6*4/*4

47

DRFS

Not reported

0.28 (0.110.74)#

ER+ or ER

Tamoxifen

160

PFS

Not reported

0.67 (0.331.35)

Postmenopausal
and ER+

2 yr tamoxifen

103

RFS

0.87 (0.381.97)

Not reported

5 yr tamoxifen

105

RFS

0.33 (0.081.43)||

Not reported

115

RFS

1.9 (0.84.8)

Not reported

(Ref. 38)

(Ref. 7)

Bilj39
Wegman

44

Nowel43

337 Monotherapy or
+ chemotherapy
or radiation or no
tamoxifen

Wegman45 677 Not reported

Newman

48

115 Monotherapy or
+ chemotherapy
and/or radiation

Tamoxifen
monotherapy, ER+

Premenopausal, or
postmenopausal and
ER+, or ER (familial
breast cancer)

East Asian populations

Lim24

212 + Prior
Premenopausal, or
chemotherapy
postmenopausal
or aromatase
and ER+
inhibitor or none
(metastatic group)

Metastatic disease

21

TDP

3.69
(1.2810.67)||

3.68
(1.2311.04)||

Kiyotani40

67

Premenopausal, or
postmenopausal
and ER+

CYP2D6*1/*1,
CYP2D6*1/*10 and
CYP2D6*10/*10 only

58

RFS

8.67
(0.2419.79)||

10.04
(1.1786.27)||

Xu75

293 Monotherapy or
+ chemotherapy
or radiation or no
tamoxifen

Premenopausal, or
postmenopausal
and ER+, or ER

Tamoxifen

152

DFS

Not reported

4.7 (1.120.0)||

Okishiro31

173 Monotherapy or +
chemotherapy or
goserelin

Premenopausal, or
postmenopausal
and ER+, or ER

Tamoxifen
monotherapy

73

RFS

0.94 (0.342.60)

0.6 (0.181.92)

Monotherapy

Table shown in full in Supplementary information S2 (table). *95% confidence intervals are indicated in brackets. The patient cohort was postmenopausal women
with resected ER+ breast cancer who participated in a North Central Cancer Treatment Group randomized Phase III clinical trial (NCCTG 89-30-52). Adjuvant setting.
||
p <0.05.p <0.10. #p <0.01. **Setting not reported. Adjuvant, metastatic setting. BCS, breast cancer survival; CYP2D6, cytochrome P450 2D6; DFS, disease-free
survival; EFS, event-free survival; DRFS, distant recurrence-free survival; ER+, oestrogen- progesterone-positive tumour; ER, oestrogen- progesterone-negative
tumour; N, number; PFS, progression-free survival; RFS, recurrence-free survival; RFT, relapse-free time; TDP, time to disease progression.

factors, are less likely to derive clinical benefit from

adjuvant tamoxifen therapy than patients with normal
CYP2D6 metabolism.
Counter evidence. The findings of four retrospective
tamoxifen studies are inconsistent with the results of
the studies mentioned above and do not support a
role for CYP2D6 in the metabolism and efficacy of
tamoxifen31,4345 (TABLe 2). A Swedish study of postmenopausal women with ER+ and ER breast cancer

found no difference in distant recurrence-free survival

between patients with EM/PM or PM/PM genotypes
and patients with an EM/EM genotype44. A uS-based
study of women with breast cancer treated with
tamoxifen also found no difference in progressionfree survival and overall survival between patients
with EM/PM or PM/PM genotypes and patients with
normal metabolism (EM/EM genotype)43. As patients
with no CYP2D6 enzyme activity derived benefit
from tamoxifen, the results suggest that its efficacy is

582 | AuguST 2009 | VoluME 9

www.nature.com/reviews/cancer
2009 Macmillan Publishers Limited. All rights reserved

REVIEWS
not dependent on CYP2D6. A second Swedish study
evaluated the genotype association in a larger cohort
of postmenopausal women with breast cancer who
were treated with tamoxifen45. A subgroup of patients
was randomized to receive two different doses of
tamoxifen for 2 years or 5 years. Patients with EM/PM
or PM/PM genotypes tended to have a lower risk of
recurrence when treated with tamoxifen for 5 years,
but not for 2 years, than patients with normal CYP2D6
metabolism (EM/EM genotype). This again suggested
that CYP2D6 is not involved in the activation of
tamoxifen. Furthermore, a Japanese study of patients
with primary breast cancer who had ER+ or progesterone receptor-positive tumours and were treated with
adjuvant tamoxifen found that the IM/IM genotype
was not predictive of recurrence-free survival31.
There is considerable heterogeneity between all of
the clinical studies, making it hard to compare them
(TABLe 2). Furthermore, none of these studies investigated the associations between CYP2D6 genotype,
endoxifen levels and treatment outcome, so the causal
relationship of endoxifen levels to outcome cannot be
established. Although all of the Europe-based studies
genotyped the most common PM allele in this ethnic
group (CYP2D6*4), many did not genotype the rarer
PM or IM alleles. It is therefore probable that most of
the studies underestimated the true incidence of poor
metabolizers. In addition, the concomitant administration of medications that inhibit CYP2D6 activity was
considered in only three studies7,40. Furthermore, the
dose of tamoxifen that was prescribed was not constant
between studies and, in some cases, was not reported
at all7,24,28,40,43,45. Although the efficacy of tamoxifen is
not thought to vary between doses46 this may not be
true for patients who are deficient in CYP2D6 enzyme
activity. Moreover, factors that are known to affect
patient outcome, such as the duration of tamoxifen
therapy and the co-administration of radiotherapy or
other chemotherapeutic agents, were not constant in
the different studies and were not included in multivariate analyses43,44,47. The studies also varied in the
way that they grouped genotypes for analysis, making
it difficult to compare results across the studies. Some
studies compared outcomes between patients with PM/
PM genotypes and other patients38, whereas other studies compared outcomes between patients with one or
more PM alleles and those with extensive metabolizer
genotypes43,45,48.
Although most trials studied only postmenopausal patients, some also included premenopausal
patients24,44. Some studies also included patients with
ER tumours that had been treated with tamoxifen,
although it is not an indicated therapy for this tumour
type43,44. Finally, the grade and stage of patients with
breast cancer varied among the studies, as did the
ethnicity of the patients and the survival outcomes
that were measured. Each of these factors could have
introduced bias to all the studies. Moreover, owing
to the low prevalence of poor metabolizers (610%)
and the limited sample sizes of the studies, few
poor metabolizers were studied in each trial. As the

effect of the CYP2D6 genotype on tamoxifen efficacy

(hazard ratio ~2)7,28 is moderate, many of the studies
may not have had sufficient power to show a difference
in tamoxifen-related efficacy between genotypes.

conclusions
Tamoxifen has been one of the highlights of mechanistic cancer therapeutics over the past 50 years, saving many thousands of lives each year. However, the
variation in response to and toxicity of tamoxifen
has been recognized from the initial development of
the drug.
The identification of genetically based resistance to
tamoxifen therapy is only a first step in the quest
to assure safe and effective anticancer therapy. Crucial
clinical and mechanistic investigations are needed to
understand the basis for breast cancer treatment resistance, the penetrance of the genedrug relationship
and the optimal methods of treating patients with
CYP2D6-mediated resistance.
In october 2006, the uS Food and Drug Administration
Clinical Pharmacology Subcommittee of the Advisory
Committee for Pharmaceutical Sciences reviewed
the relationship between tamoxifen and CYP2D6 and
recommended that the package insert for tamoxifen
be amended to warn postmenopausal women of the
potential heightened risk of treatment failure for patients
with deficient CYP2D6 activity and that certain antidepressants (for example, SSRIs) can interfere with the
bioactivation of tamoxifen.
The in vivo pharmacology and retrospective
clinical trial data make a strong case for the use of
CYP2D6 genotype to guide the selection and dose
of tamoxifen. However, the data do not make genotype assessment mandatory. Ideally, a large, prospective, randomized clinical trial that compares outcomes
between genotype-guided and standard-of-care dosing
of tamoxifen would provide the best basis on which to
make a decision about mandatory genotyping. However,
such a study is unlikely to be performed, owing to the
current lack of financial support and interest from
investigators to better understand how to optimally use
off-patent medications. The results of other large, retrospective tamoxifen pharmacogenetic trials, including those that compare tamoxifen therapy to alternative
hormone therapies, such as aromatase inhibitors (BOX 4),
will clarify the prognostic and predictive relevance
of CYP2D6 testing. Confirmation of the association
between CYP2D6 activity (determined by genotype
or drug interaction) and tamoxifen outcomes by these
studies will help the uS Food and Drug Administration
reach a decision on the label warning for tamoxifen.
These studies will also help to clarify the advice given
to clinicians regarding the use of CYP2D6 genotype to
guide selection of an anti-oestrogen therapy for postmenopausal women with early-stage breast cancer and
how to best treat patients that are intermediate or poor
metabolizers (BOX 4).
It is currently considered prudent for patients with
breast cancer to avoid CYP2D6 inhibitors such as
fluoxetine and paroxetine to alleviate hot flashes and

NATuRE REVIEwS | CanCer

VoluME 9 | AuguST 2009 | 583

2009 Macmillan Publishers Limited. All rights reserved

REVIEWS
to instead receive other antidepressants that are weak
or negligible inhibitors of CYP2D6, such as venlafaxine or citalopram36,49. However, this recommendation
is based on limited evidence that comes largely from
pharmacokinetic studies that demonstrated an association between co-administered CYP2D6 inhibitors
and reduced plasma endoxifen concentrations6,14,18 and
only one trial that showed a heightened risk of recurrence in this patient group7. Regarding the management
of depression, some patients may not tolerate or respond
to SNRIs or SSRIs that do not inhibit CYP2D6. It is
therefore important for oncologists to discuss changing
antidepressants with patients and their psychiatrists.
The frequency of PM and IM alleles varies among ethnic groups, suggesting that there may be ethnic differences
in the response of patients to tamoxifen therapy (BOX 3).
IM alleles, which are more common in East Asian and
African populations than in European populations, have
also been shown to be associated with reduced endoxifen
formation and tamoxifen efficacy in East Asian patients,
and this requires investigation in African populations.
Indeed, East Asian women with the IM/IM genotype have
similar endoxifen plasma concentrations to European
women with PM/PM genotypes (~20 nmol l1)18,24.
As ~20% of East Asians, and 16% of those with African
ancestry, have decreased CYP2D6 metabolism (PM/
PM, IM/IM and PM/IM genotypes), compared with
only 610% of Europeans21, it is possible that more East
Asians and Africans are at risk of not deriving the full
benefit from tamoxifen that most Europeans do. This

is especially important in East Asian countries, where

tamoxifen continues to be an important therapy more
than half of breast cancers in these countries are premenopausal50,51, and tamoxifen remains the standard
endocrine therapy for this indication52.
Individualizing anti-oestrogen therapy could
improve the efficacy of tamoxifen in patients with
breast cancer. The value of monitoring endoxifen
plasma concentrations in this patient population
has yet to be explored, although its clinical utility
may be limited because steady-state metabolite concentrations are not reached until after a month of
continuous therapy 53. Therefore, CYP2D6 genotyping
may initially be more useful than therapeutic drug
monitoring, because information about the CYP2D6
metabolizer status of a patient can be obtained rapidly
before initiating anti-oestrogen therapy. Many companies offer genotyping tests, and interpretation of the
genotype results using the recently published CYP2D6
activity score method54 would allow prediction of
CYP2D6 phenotype with greater certainty. This is
especially important because IM alleles influence the
amount of endoxifen formed, as well as therapeutic outcomes. overall, the current data suggest that
genotype-guided tamoxifen administration may be a
useful part of a comprehensive strategy to optimize
the treatment of breast cancer; however, the findings
of large, well-designed clinical trials that support such
a change in clinical practice are needed before this
change is advocated.

Box 4 | alternative hormonal therapies

Postmenopausal women with early stage oestrogen receptor-positive (ER+) breast cancer who are predicted to not
respond to tamoxifen on the basis of their cytochrome P450 2D6 (CYP2D6) genotype could be offered an alternative
anti-oestrogen therapy, such as one of the third-generation aromatase inhibitors (AIs): letrozole, anastrozole and
exemestane. These compounds prevent the formation of oestrogens in extragonadal tissues by inhibiting the
aromatase enzyme CYP19A1. Adjuvant AIs are modestly more efficacious than adjuvant tamoxifen for postmenopausal
women with early-stage breast cancer that is ER+ and/or progesterone receptor-positive, whether given alone or in
tandem with tamoxifen as part of first-line therapy6265. Because of this increased efficacy and their more favourable
safety profile, AIs have recently replaced tamoxifen as the preferred adjuvant anti-oestrogen therapy for
postmenopausal women59. A recent study estimated that this benefit of AI compared with tamoxifen may be entirely
due to the poor efficacy of tamoxifen in patients with the CYP2D6 poor metabolizer (PM)/PM genotype, as patients
with the extensive metabolizer (EM)/EM genotype obtained the same adjuvant benefits from tamoxifen as they did
from AI72. However, it is important to note that this study estimated the recurrence rates for each genotype on the basis
of the hazard ratios reported by Goetz et al.38, so the heterogeneity of outcomes among the CYP2D6tamoxifen trials
was not considered.
However, AIs can have serious toxicities. The profound oestrogen deprivation state induced by AIs results in
accelerated bone loss, with a higher risk of bone fractures associated with long-term AI therapy than with tamoxifen
therapy. AIs can be accompanied by musculoskeletal adverse events, including anthralgia and myalgia, which can
necessitate their discontinuation57,59. Furthermore, AIs may not provide the same cardioprotective effects as tamoxifen59,
and the greater cost of AIs and their modest additional benefit make them less cost-effective than tamoxifen in the same
context.
AIs are not effective in women who have functioning ovaries; therefore, tamoxifen remains the standard
endocrine therapy for premenopausal patients with ER+ breast cancer52. More than half of breast cancers in Korea,
China and possibly Japan are premenopausal50,51, suggesting that tamoxifen will continue to be an important
therapy in these countries. Premenopausal women with early-stage breast cancer and reduced CYP2D6 activity
may derive more therapeutic benefit from alternative selective ER modulators that do not require CYP2D6. Other
approaches include ovarian suppression with or without AI. A recently reported randomized study of
premenopausal women treated with ovarian suppression plus tamoxifen as opposed to ovarian suppression plus
anastrozole did not detect a difference in outcome73. Regarding chemoprevention, raloxifene has shown activity in
breast cancer prevention, suggesting that this might be a reasonable alternative to tamoxifen in CYP2D6 poor
metabolizer patients74.

584 | AuguST 2009 | VoluME 9

www.nature.com/reviews/cancer
2009 Macmillan Publishers Limited. All rights reserved

REVIEWS
1.
2.
3.
4.

5.

6.

7.

8.
9.

10.

11.
12.
13.

14.

15.

16.

Evans, W. E. & McLeod, H. L. Pharmacogenomics

drug disposition, drug targets, and side effects.
N. Engl. J. Med. 348, 538549 (2003).
Evans, W. E. & Relling, M. V. Moving towards
individualized medicine with pharmacogenomics.
Nature 429, 464468 (2004).
Parkin, d. M., Bray, F., Ferlay, J. & Pisani, P. Global
cancer statistics, 2002. CA Cancer J. Clin. 55,
74108 (2005).
Wu, X. et al. The tamoxifen metabolite, endoxifen, is a
potent antiestrogen that targets estrogen receptor
for degradation in breast cancer cells. Cancer Res. 69,
17221727 (2009).
Ingelman-Sundberg, M. Genetic polymorphisms of
cytochrome P450 2d6 (CYP2D6): clinical
consequences, evolutionary aspects and functional
diversity. Pharmacogenomics J. 5, 613 (2005).
Jin, Y. et al. CYP2D6 genotype, antidepressant use,
and tamoxifen metabolism during adjuvant breast
cancer treatment. J. Natl Cancer Inst. 97, 3039
(2005).
This study found a gene dose effect of CYP2D6
genotype on steady-state endoxifen plasma
concentrations: PM/PM patients had the
lowest concentrations, EM/EM patients had the
highest and EM/PM patients had intermediate
concentrations. It also showed that co-administered
antidepressants that are potent inhibitors of
CYP2D6 reduced the formation of endoxifen.
Goetz, M. P. et al. The impact of cytochrome P450
2d6 metabolism in women receiving adjuvant
tamoxifen. Breast Cancer Res. Treat. 101, 113121
(2007).
This study, which used the same cohort as in
reference 38, found that postmenopausal women
who received tamoxifen for early breast cancer with
decreased metabolism of CYP2D6, by genotype or
enzyme inhibition, had a shorter time to recurrence
and worse relapse-free survival than patients with
EM/EM genotypes who were not taking a CYP2D6
inhibitor.
dahlman-Wright, K. et al. International Union of
Pharmacology. LXIV. Estrogen receptors. Pharmacol.
Rev. 58, 773781 (2006).
Osborne, C. K., Shou, J., Massarweh, S. & Schiff, R.
Crosstalk between estrogen receptor and growth
factor receptor pathways as a cause for endocrine
therapy resistance in breast cancer. Clin. Cancer Res.
11, S865S870 (2005).
Massarweh, S. & Schiff, R. Unraveling the mechanisms
of endocrine resistance in breast cancer: new
therapeutic opportunities. Clin. Cancer Res. 13,
19501954 (2007).
Osborne, C. K. Tamoxifen in the treatment of breast
cancer. N. Engl. J. Med. 339, 16091618 (1998).
dehal, S. S. & Kupfer, d. CYP2d6 catalyzes tamoxifen
4-hydroxylation in human liver. Cancer Res. 57,
34023406 (1997).
desta, Z., Ward, B. A., Soukhova, n. V. & Flockhart,
d. A. Comprehensive evaluation of tamoxifen
sequential biotransformation by the human
cytochrome P450 system in vitro: prominent roles for
CYP3A and CYP2d6. J. Pharmacol. Exp. Ther. 310,
10621075 (2004).
This comprehensive in vitro study used human liver
microsomes and expressed human CYPs to
characterize the oxidation of tamoxifen, showing
that the conversion of tamoxifen to its secondary
metabolite endoxifen occurs through two primary
metabolites and is predominantly mediated by
CYP2D6, CYP3A4 and CYP3A5.
Stearns, V. et al. Active tamoxifen metabolite plasma
concentrations after coadministration of tamoxifen
and the selective serotonin reuptake inhibitor
paroxetine. J. Natl Cancer Inst. 95, 17581764
(2003).
Johnson, M. d. et al. Pharmacological characterization
of 4-hydroxy-N-desmethyl tamoxifen, a novel active
metabolite of tamoxifen. Breast Cancer Res. Treat. 85,
151159 (2004).
This study showed that endoxifen has essentially
equivalent activity to the potent metabolite
4-hydroxytamoxifen with respect to binding to ERs,
ability to inhibit oestrogen-stimulated breast
cancer cell proliferation and the regulation of
oestrogen-responsive genes.
Lim, Y. C., desta, Z., Flockhart, d. A. & Skaar, T. C.
Endoxifen (4-hydroxy-N-desmethyl-tamoxifen) has
anti-estrogenic effects in breast cancer cells with
potency similar to 4-hydroxy-tamoxifen. Cancer
Chemother. Pharmacol. 55, 471478 (2005).

17. Lim, Y. C. et al. Endoxifen, a secondary metabolite of

tamoxifen, and 4-OH-tamoxifen induce similar
changes in global gene expression patterns in MCF-7
breast cancer cells. J. Pharmacol. Exp. Ther. 318,
503512 (2006).
This study investigated the effects of endoxifen and
4-hydroxytamoxifen on gene expression in MCF-7
cells, showing that endoxifen and
4-hydroxytamoxifen have similar effects on global
gene expression patterns and that most of the
affected genes are oestrogen-regulated genes.
18. Borges, S. et al. Quantitative effect of CYP2D6
genotype and inhibitors on tamoxifen metabolism:
implication for optimization of breast cancer
treatment. Clin. Pharmacol. Ther. 80, 6174 (2006).
19. nelson, d. R. et al. Comparison of cytochrome P450
(CYP) genes from the mouse and human genomes,
including nomenclature recommendations for genes,
pseudogenes and alternative-splice variants.
Pharmacogenetics 14, 118 (2004).
20. Ingelman-Sundberg, M., Sim, S. C., Gomez, A. &
Rodriguez-Antona, C. Influence of cytochrome P450
polymorphisms on drug therapies: pharmacogenetic,
pharmacoepigenetic and clinical aspects. Pharmacol.
Ther. 116, 496526 (2007).
21. Bradford, L. d. CYP2D6 allele frequency in European
Caucasians, Asians, Africans and their descendants.
Pharmacogenomics 3, 229243 (2002).
22. Gaedigk, A. et al. Cytochrome P4502d6 (CYP2D6)
gene locus heterogeneity: characterization of gene
duplication events. Clin. Pharmacol. Ther. 81,
242251 (2007).
23. Kusama, M., Hisaka, A., Hibino, Y. & Suzuki, H. The
influence of Asian specific variant, CYP2D6*10 on
in vitro formation of endoxifen, an active metabolite of
tamoxifen. Abstr. 21st Japanese Society for the Study
of Xenobiotics Annual Meeting 21, 229 (2006).
24. Lim, H. S. et al. Clinical implications of CYP2D6
genotypes predictive of tamoxifen pharmacokinetics in
metastatic breast cancer. J. Clin. Oncol. 25,
38373845 (2007).
25. Kimmick, G. G., Lovato, J., McQuellon, R.,
Robinson, E. & Muss, H. B. Randomized, double-blind,
placebo-controlled, crossover study of sertraline
(Zoloft) for the treatment of hot flashes in women with
early stage breast cancer taking tamoxifen. Breast J.
12, 114122 (2006).
26. Loprinzi, C. L. et al. Venlafaxine in management of hot
flashes in survivors of breast cancer: a randomised
controlled trial. Lancet 356, 20592063 (2000).
27. Hemeryck, A. & Belpaire, F. M. Selective serotonin
reuptake inhibitors and cytochrome P-450 mediated
drugdrug interactions: an update. Curr. Drug Metab.
3, 1337 (2002).
28. Schroth, W. et al. Breast cancer treatment outcome
with adjuvant tamoxifen relative to patient CYP2D6
and CYP2C19 genotypes. J. Clin. Oncol. 25,
51875193 (2007).
29. Sim, S. C. et al. A common novel CYP2C19 gene
variant causes ultrarapid drug metabolism relevant for
the drug response to proton pump inhibitors and
antidepressants. Clin. Pharmacol. Ther. 79, 103113
(2006).
30. desta, Z., Zhao, X., Shin, J. G. & Flockhart, d. A.
Clinical significance of the cytochrome P450 2C19
genetic polymorphism. Clin. Pharmacokinet. 41,
913958 (2002).
31. Okishiro, M. et al. Genetic polymorphisms of CYP2D6
10 and CYP2C19 2, 3 are not associated with
prognosis, endometrial thickness, or bone mineral
density in Japanese breast cancer patients treated
with adjuvant tamoxifen. Cancer 115, 952961
(2009).
32. Ogura, K. et al. Quaternary ammonium-linked
glucuronidation of trans-4-hydroxytamoxifen, an
active metabolite of tamoxifen, by human liver
microsomes and UdP-glucuronosyltransferase 1A4.
Biochem. Pharmacol. 71, 13581369 (2006).
33. nishiyama, T. et al. Reverse geometrical selectivity in
glucuronidation and sulfation of cis- and
trans-4-hydroxytamoxifens by human liver UdPglucuronosyltransferases and sulfotransferases.
Biochem. Pharmacol. 63, 18171830 (2002).
34. Gjerde, J. et al. Effects of CYP2D6 and SULT1A1
genotypes including SULT1A1 gene copy number on
tamoxifen metabolism. Ann. Oncol. 19, 5661
(2008).
35. Blevins-Primeau, A. S. et al. Functional significance of
UdP-glucuronosyltransferase variants in the
metabolism of active tamoxifen metabolites. Cancer
Res. 69, 18921900 (2009).

NATuRE REVIEwS | CanCer

36. Spina, E., Santoro, V. & dArrigo, C. Clinically relevant

pharmacokinetic drug interactions with secondgeneration antidepressants: an update. Clin. Ther. 30,
12061227 (2008).
37. newton, d. J., Wang, R. W. & Lu, A. Y. Cytochrome
P450 inhibitors. Evaluation of specificities in the
in vitro metabolism of therapeutic agents by human
liver microsomes. Drug Metab. Dispos. 23, 154158
(1995).
38. Goetz, M. P. et al. Pharmacogenetics of tamoxifen
biotransformation is associated with clinical outcomes
of efficacy and hot flashes. J. Clin. Oncol. 23,
93129318 (2005).
This study showed that postmenopausal women
who received tamoxifen for early breast cancer with
PM/PM genotypes tended to have a higher risk of
disease relapse and a lower incidence of hot
flashes than patients with EM/EM or EM/PM
genotypes.
39. Bijl, M. J. et al. The CYP2D6*4 polymorphism affects
breast cancer survival in tamoxifen users. Breast
Cancer Res. Treat. 3 Feb 2009 (doi:10.1007/s10549008-0272-2).
40. Kiyotani, K. et al. Impact of CYP2D6*10 on
recurrence-free survival in breast cancer patients
receiving adjuvant tamoxifen therapy. Cancer Sci. 99,
995999 (2008).
41. Xu, Y. & Villalona-Calero, M. A. Irinotecan:
mechanisms of tumor resistance and novel strategies
for modulating its activity. Ann. Oncol. 13,
18411851 (2002).
42. Bonanni, B. et al. Polymorphism in the CYP2D6
tamoxifen-metabolizing gene influences clinical effect
but not hot flashes: data from the Italian Tamoxifen
Trial. J. Clin. Oncol. 24, 37083709; author reply
3709 (2006).
43. nowell, S. A. et al. Association of genetic variation in
tamoxifen-metabolizing enzymes with overall
survival and recurrence of disease in breast cancer
patients. Breast Cancer Res. Treat. 91, 249258
(2005).
44. Wegman, P. et al. Genotype of metabolic enzymes and
the benefit of tamoxifen in postmenopausal breast
cancer patients. Breast Cancer Res. 7, R284R290
(2005).
45. Wegman, P. et al. Genetic variants of CYP3A5,
CYP2D6, SULT1A1, UGT2B15 and tamoxifen
response in postmenopausal patients with breast
cancer. Breast Cancer Res. 9, R7 (2007).
46. Bratherton, d. G. et al. A comparison of two doses of
tamoxifen (nolvadex) in postmenopausal women with
advanced breast cancer: 10 mg bd versus 20 mg bd.
Br. J. Cancer 50, 199205 (1984).
47. Effects of chemotherapy and hormonal therapy for
early breast cancer on recurrence and 15-year
survival: an overview of the randomised trials. Lancet
365, 16871717 (2005).
48. newman, W. G. et al. Impaired tamoxifen metabolism
reduces survival in familial breast cancer patients. Clin.
Cancer Res. 14, 59135918 (2008).
49. Lash, T. L. et al. Tamoxifens protection against breast
cancer recurrence is not reduced by concurrent use of
the SSRI citalopram. Br. J. Cancer 99, 616621
(2008).
50. Shin, H. R., Jung, K. W., Won, Y. J., Park, J. G. & 139
KCCR-affiliated Hospitals. 2002 annual report of the
Korean Central Cancer Registry: based on registered
data from 139 hospitals. Cancer Res. Treat 36,
103114 (2004).
51. Yu, K. d. et al. development and trends of surgical
modalities for breast cancer in China: a review of
16-year data. Ann. Surg. Oncol. 14, 25022509
(2007).
52. Parton, M. & Smith, I. E. Controversies in the
management of patients with breast cancer: adjuvant
endocrine therapy in premenopausal women. J. Clin.
Oncol. 26, 745752 (2008).
53. Furlanut, M. et al. Tamoxifen and its main
metabolites serum and tissue concentrations in
breast cancer women. Ther. Drug Monit. 29,
349352 (2007).
54. Gaedigk, A. et al. The CYP2d6 activity score:
translating genotype information into a qualitative
measure of phenotype. Clin. Pharmacol. Ther. 83,
234242 (2008).
55. Parte, P. & Kupfer, d. Oxidation of tamoxifen by
human flavin-containing monooxygenase (FMO) 1 and
FMO3 to tamoxifen-N-oxide and its novel reduction
back to tamoxifen by human cytochromes P450 and
hemoglobin. Drug Metab. Dispos. 33, 14461452
(2005).

VoluME 9 | AuguST 2009 | 585

2009 Macmillan Publishers Limited. All rights reserved

REVIEWS
56. Early Breast Cancer Trialists Collaborative Group.
Tamoxifen for early breast cancer: an overview of
the randomised trials. Lancet 351, 14511467
(1998).
57. Ingle, J. n. Pharmacogenomics of tamoxifen and
aromatase inhibitors. Cancer 112, 695699 (2008).
58. Clarke, R., Leonessa, F., Welch, J. n. & Skaar, T. C.
Cellular and molecular pharmacology of antiestrogen
action and resistance. Pharmacol. Rev. 53, 2571
(2001).
59. Perez, E. A. Safety profiles of tamoxifen and the
aromatase inhibitors in adjuvant therapy of hormoneresponsive early breast cancer. Ann. Oncol. 18
(Suppl. 8), 2635 (2007).
60. ntukidem, n. I. et al. Estrogen receptor genotypes,
menopausal status, and the lipid effects of tamoxifen.
Clin. Pharmacol. Ther. 83, 702710 (2008).
61. Morello, K. C., Wurz, G. T. & deGregorio, M. W.
Pharmacokinetics of selective estrogen receptor
modulators. Clin. Pharmacokinet. 42, 361372 (2003).
62. Coombes, R. C. et al. A randomized trial of exemestane
after two to three years of tamoxifen therapy in
postmenopausal women with primary breast cancer.
N. Engl. J. Med. 350, 10811092 (2004).
63. Howell, A. et al. Results of the ATAC (arimidex,
tamoxifen, alone or in combination) trial after
completion of 5 years adjuvant treatment for breast
cancer. Lancet 365, 6062 (2005).
64. Kaufmann, M. et al. Improved overall survival in
postmenopausal women with early breast cancer after
anastrozole initiated after treatment with tamoxifen
compared with continued tamoxifen: the ARnO 95
Study. J. Clin. Oncol. 25, 26642670 (2007).
65. Thurlimann, B. et al. A comparison of letrozole and
tamoxifen in postmenopausal women with early breast
cancer. N. Engl. J. Med. 353, 27472757 (2005).

66. Sachse, C., Brockmoller, J., Bauer, S. & Roots, I.

Cytochrome P450 2d6 variants in a Caucasian
population: allele frequencies and phenotypic
consequences. Am. J. Hum. Genet. 60, 284295
(1997).
67. Gough, A. C. et al. Identification of the primary gene
defect at the cytochrome P450 CYP2D locus. Nature
347, 773776 (1990).
68. Cai, W. M., Chen, B. & Zhang, W. X. Frequency of
CYP2D6*10 and *14 alleles and their influence on
the metabolic activity of CYP2d6 in a healthy
Chinese population. Clin. Pharmacol. Ther. 81,
9598 (2007).
69. Lee, S. Y. et al. Sequence-based CYP2D6 genotyping in
the Korean population. Ther. Drug Monit. 28,
382387 (2006).
70. Shen, H. et al. Comparative metabolic capabilities and
inhibitory profiles of CYP2d6.1, CYP2d6.10, and
CYP2d6.17. Drug Metab. Dispos. 35, 12921300
(2007).
71. Wennerholm, A. et al. Characterization of the
CYP2D6*29 allele commonly present in a black
Tanzanian population causing reduced catalytic
activity. Pharmacogenetics 11, 417427 (2001).
72. Punglia, R. S., Burstein, H. J., Winer, E. P. & Weeks,
J. C. Pharmacogenomic variation of CYP2d6 and the
choice of optimal adjuvant endocrine therapy for
postmenopausal breast cancer: a modeling analysis.
J. Natl Cancer Inst. 100, 642648 (2008).
73. Gnant, M. et al. Endocrine therapy plus zoledronic
acid in premenopausal breast cancer. N. Engl. J. Med.
360, 679691 (2009).
74. Mahoney, M. C., Bevers, T., Linos, E. & Willett, W. C.
Opportunities and strategies for breast cancer
prevention through risk reduction. CA Cancer J. Clin.
58, 347371 (2008).

586 | AuguST 2009 | VoluME 9

75. Xu, Y. et al. Association between CYP2D6 *10

genotype and survival of breast cancer patients
receiving tamoxifen treatment. Ann. Oncol. 19,
14231429 (2008).

Acknowledgements

This work was supported by the US national Institutes of

Health Pharmacogenetics Research network (Grant U01
GM63340).

DaTaBases
Entrez Gene: http://www.ncbi.nlm.nih.gov/entrez/query.
fcgi?db=gene
CYP2C19 | CYP2D6 | ESR1 | ESR2 | SULT1A1
National Cancer Institute Drug Dictionary: http://www.
cancer.gov/drugdictionary/
quinidine | tamoxifen | venlafaxine
UniProtKB: http://www.uniprot.org
CYP2C19 | CYP2D6 | CYP3A4 | CYP3A5 | EGFR | ER | ER |
ERBB2 | ER | GRIP1 | IRS1 | KS6A1 | NCOA1 | NCOA3 |
NCOR1

FurTher inFOrMaTiOn
Howard L. McLeods homepage: http://ipit.unc.edu
Comprehensive Research on Expressed Alleles in
Therapeutic Evaluation (UNC): http://create.unc.edu/
CYP2D6 allele nomenclature website: http://www.
cypalleles.ki.se/cyp2d6.htm
UNC Institute for Pharmacogenomics and Individualized
Therapy: http://ipit.unc.edu/

suppLeMenTarY inFOrMaTiOn
See online article: S1 (table) | S2 (table)
all linkS are aCTive in The Online PdF

www.nature.com/reviews/cancer
2009 Macmillan Publishers Limited. All rights reserved