Transcript Pharmacogenetics: Clinical Implications

Pharmacogenomics:
The Promise of Personalized
Medicine
Christina Aquilante, Pharm.D.
Assistant Professor
Department of Pharmaceutical Sciences
School of Pharmacy
University of Colorado at Denver and Health
Sciences Center
Objectives
Provide an overview of pharmacogenomics and
its clinical relevance
Discuss clinically-relevant examples of:
– Drug metabolism pharmacogenomics
– Drug target pharmacogenomics
Discuss the challenges facing
pharmacogenomic studies and the movement of
pharmacogenomics into clinical practice
Discuss pharmacogenomics from the FDA and
pharmaceutical industry perspective
Interindividual Variability in Drug
Response
Disease
Asthma
Hypertension
Solid Cancers
Depression
Diabetes
Arthritis
Schizophrenia
Drug Class
Rate of Poor Response
Beta-agonists
40-75%
Various
30%
Various
70%
SSRIs, tricyclics
20-40%
Sulfonylureas, others
50%
NSAIDs, COX-2 inhibitors 30-60%
Various
25-75%
Factors Contributing to
Interindividual Variability in Drug
Disposition and Action
Age
Race/ethnicity
Weight
Gender
Concomitant Diseases
Concomitant Drugs
Social factors
GENETICS
PERSONALIZED
MEDICINE
“We wish to suggest a
structure for the salt of
[DNA].
This structure has novel
features which are of
considerable biological
interest.”
Human Genome Project
Determine the sequence of the 3 billion
nucleotides that make up human DNA
Characterize variability in the genome
Identify all the genes in human DNA
The Era of Genomic Medicine:
– Improve prediction of drug efficacy or toxicity
– Improve the diagnosis of disease
– Earlier detection of genetic predisposition to disease
Newsweek
June 25, 2001
“…pharmacogenetics
promises to target
treatment to a patient’s
genetic profile…”
Genetics or Genomics?
Pharmacogenetics
– Study of how genetic differences in a SINGLE
gene influence variability in drug response
(i.e., efficacy and toxicity)
Pharmacogenomics
– Study of how genetic (genome) differences in
MULTIPLE genes influence variability in drug
response (i.e., efficacy and toxicity)
Current Concept of
Pharmacogenomics
Roden DM et al. Ann Intern Med 2006; 145:749-57
Potential of Pharmacogenomics
All patients with same diagnosis
1
2
Non-responders
and toxic
responders
Treat with alternative
drug or dose
Responders and patients
not predisposed to toxicity
Treat with
conventional
drug or dose
Clinical Relevance
Can we predict who will derive an optimal
response?
Can we predict who will have a toxicity?
– Host (patient) genotype determines optimal
drug therapy approach
– Disease (pathogen) genotype determines
optimal drug therapy approach
DNA is Information
DNA
ENGLISH
A, T, G, C
Abcdefg….xyz
Codon
Word
Gene
Sentence
Chromosome
Chapter
Genome
Book
Composition of the Human
Genome
Mutation/Polymorphism
Unit of genetic code
Coding sequence (exons)
Gene (exons and introns)
Chromosome
Human genome
1 bp
3 bp
3,000 bp
50,000 bp
150,000,000 bp
3,000,000,000 bp
The Foundation of Pharmacogenomics:
Differences in the Genetic Code
Between People
Mutation: difference in the DNA code that
occurs in less than 1% of population
– Often associated with rare diseases
Cystic fibrosis, sickle cell anemia, Huntington’s disease
Polymorphism: difference in the DNA code that
occurs in more than 1% of the population
– A single polymorphism is less likely to be the main
cause of a disease
– Polymorphisms often have no visible clinical impact
Single Nucleotide Polymorphisms
(SNP)
Pronounced “snip”
Single base pair difference in the DNA
sequence
– Over 2 million SNPs in the human genome
Other polymorphisms:
– Insertion/deletion polymorphisms
– Gene duplications
– Gene deletions
Consequences of Polymorphisms
May result in a different amino acid or stop
codon
May result in a change in protein function
or quantity
May alter stability of mRNA
No consequence
Genetics Terminology
Alleles = different
DNA sequences at a
locus
– Codon 389 1-AR
Arg (0.75)
Gly (0.25)
Genotype = pair of
alleles a person has
at a region of the
chromosome
– Codon 389 1-AR
Arg389Arg
Arg389Gly
Gly389Gly
Genetics Terminology
Phenotype: outward manifestation of a
trait
Linkage: measure of proximity of 2 or
more polymorphisms on a single
chromosome
– Polymorphisms in close proximity tend to be
co-inherited
– Regions of linked polymorphisms are known
as haplotypes
Haplotype Map
For specific locations in the genome, a
small number of SNP patterns
(haplotypes) can account for 80-90% of
entire human population
International HapMap Project:
– Identifying common haplotypes in four
populations from different parts of the world
– Identifying “tag” SNPs that uniquely identify
these haplotypes
Pharmacogenomics
DRUG
TARGETS
DRUG
TRANSPORTERS
PHARMACODYNAMICS
PHARMACOKINETICS
Variability in
Efficacy/Toxicity
Johnson JA. Trends in Genetics 2003: 660-666
DRUG
METABOLIZING
ENZYMES
Drug Metabolism
Pharmacogenomics
Evidence of an inherited basis for drug
response dates back in the literature to the
1950s
– Succinylcholine: 1 in 3000 patients
developed prolonged muscle relaxation
Monogenic
Phenotype to genotype approach
Drug Metabolizing Enzymes
Examples of Drug Metabolism
Pharmacogenomics
NEJM 2003; 348: 529-537
Examples of Drug Metabolism
Pharmacogenomics
NEJM 2003; 348: 529-537
Warfarin and CYP2C9
Widely prescribed anticoagulant drug used to
prevent blood clots
Narrow range between efficacy and toxicity
Large variability in the dose required to achieve
therapeutic anticoagulation
– Doses vary 10-fold between people
CYP2C9 is the enzyme responsible for the
metabolism of warfarin
SNPs exist in CYP2C9 gene that decrease the
activity of the CYP2C9 metabolizing enzyme
CYP2C9 Polymorphisms and
Warfarin Dose
Warfarin dose is affected by CYP2C9 genotype
*2 and *3
are SNPs
Gage BF et al.
Thromb Haemost
2004; 91: 87-94
CYP2C9 Genotype and Bleeding Events
Compared to wildtype, CYP2C9 variants
had a higher risk of
serious or lifethreatening bleeds
WT
Variant
Hazard Ratio of 3.94
during the first 3
months of follow-up
Hazard Ratio of 2.39
for the entire follow-up
period
Higashi et al. JAMA 2002; 287
Challenges Facing Warfarin
Pharmacogenomics
Despite the strong association between
CYP2C9 genotype and warfarin dose,
CYP2C9 genotype accounts for only a
small portion of the total variability in
warfarin doses (~10-20%)
Need to determine other genetic and
non-genetic factors that contribute to
interindividual variability in warfarin doses
CYP2D6 Polymorphisms
CYP2D6 is responsible for the metabolism of a
number of different drugs
– Antidepressants, antipsychotics, analgesics,
cardiovascular drugs
Over 100 polymorphisms in CYP2D6 have been
identified
Based on these polymorphisms, patients are
phenotypically classified as:
– Ultrarapid metabolizers (UMs)
– Extensive metabolizers (EMs)
– Poor metabolizers (PMs)
CYP2D6 Phenotypes
NEJM 2003; 348:529
Roden DM et al. Ann Intern Med 2006; 145:749-57
CYP2D6 Polymorphisms and
Psychiatric Drug Response
Increased rate of adverse effects in poor
metabolizers due to increased plasma
concentrations of drug:
– Fluoxetine (Prozac) death in child attributed
to CYP2D6 poor metabolizer genotype
– Side effects of antipsychotic drugs occur more
frequently in CYP2D6 poor metabolizers
– CYP2D6 poor metabolizers with severe
mental illness had more adverse drug
reactions, increased cost of care, and longer
hospital stays
CYP2D6 and Codeine
Codeine requires activation by CYP2D6 in
order to exert its analgesic effect
Due to genetic polymorphisms, 2-10% of
the population cannot metabolize codeine
and are resistant to the analgesic effects
Interindividual variability exists in the
adequacy of pain relief when uniform
doses of codeine are given
Strattera® (Atomoxetine)
Treatment of attention deficit hyperactivity
disorder
– CYP2D6 poor metabolizers have 10-fold higher
plasma concentrations to a given dose of
STRATTERA compared with extensive
metabolizers
– Approximately 7% of Caucasians are poor
metabolizers
– Higher blood levels in poor metabolizers may
lead to a higher rate of some adverse effects of
STRATTERA
CYP2C19 and Proton Pump
Inhibitors
Proton pump inhibitors are used to treat
acid reflux and stomach ulcers
Ulcer cure rates using omeprazole and
amoxicillin by CYP2C19 phenotype:
Cure Rate
– Rapid metabolizers
– Intermediate metabolizers
– Poor metabolizers
28.6%
60%
100%
Furuta, T. et. al. Ann Intern Med 1998;129:1027-1030
Roche AmpliChip: FDA-Approved
Roche AmpliChip P450 Test
The Roche AmpliChip CYP450 Test is intended
to identify a patient's CYP2D6 and CYP2C19
genotype from genomic DNA extracted from a
whole blood sample.
Information about CYP2D6 and CYP2C19
genotype may be used as an aid to clinicians in
determining therapeutic strategy and treatment
dose for therapeutics that are metabolized by
the CYP2D6 or CYP2C19 gene product.
Thiopurine-S-Methyltransferase (TPMT)
Thiopurine drugs are used to treat cancer
– Acute lymphoblastic leukemia
TPMT is important for metabolizing thiopurines
– azathioprine, mercaptopurine (6-MP)
Polymorphisms in the TPMT gene result in
decreased TPMT enzyme activity
Decreased TPMT activity predisposes
individuals to severe, life-threatening toxicities
from these drugs
Variability in TPMT Activity
% of Subjects
Enzyme Activity Levels in 300 Caucasian
Patients
100
90
80
70
60
50
40
30
20
10
0
low
medium
TPMT Enzyme Activity
high
Genotype-Guided 6-MP Dosing
Pharmacogenomics 2002;3(1):89-98.
6-Mercaptopurine Prescribing
Information
There are individuals with an inherited deficiency
of the enzyme thiopurine methyltransferase
(TPMT) who may be unusually sensitive to the
myelosuppressive effects of mercaptopurine and
prone to developing rapid bone marrow
suppression following the initiation of treatment.
Substantial dosage reductions may be required to
avoid the development of life-threatening bone
marrow suppression in these patients.
Imuran Prescribing Information
TPMT genotyping or phenotyping can be used to identify
patients with absent or reduced TPMT activity.
Patients with low or absent TPMT activity are at an increased
risk of developing severe, life-threatening myelotoxicity from
IMURAN if conventional doses are given.
Physicians may consider alternative therapies for patients
who have low or absent TPMT activity (homozygous for nonfunctional alleles). IMURAN should be administered with
caution to patients having one non-functional allele
(heterozygous) who are at risk for reduced TPMT activity that
may lead to toxicity if conventional doses are given. Dosage
reduction is recommended in patients with reduced TPMT
activity.
TPMT and Thioguanines
Clinical implications:
– Genetic testing for TPMT is routine practice at
some cancer centers for protocols involving
thiopurine drugs
– CLIA approved test available
– Implications for cancer, transplant, rheumatoid
arthritis, lupus, dermatology, and Crohn’s
disease treatment
Drug Target
Pharmacogenomics
Drug Target Pharmacogenomics
Direct protein target of drug
– Receptor
– Enzyme
Proteins involved in pharmacologic response
– Signal transduction proteins or downstream proteins
Polymorphisms associated with
disease risk
– “Disease-modifying” polymorphisms
– “Treatment-modifying” polymorphisms
POLYGENIC
Complexity of Drug Effect
Assessing Phenotype in Drug Target
Pharmacogenomics
Depression—Symptom rating scales
– Indirect measure of drug response
– Inter-rater reliability
Hypertension—Blood pressure
– Minute to minute and diurnal variability
– Influence of environmental factors (e.g. lack of
rest before measurement)
Diabetes—Blood glucose
– Diurnal variation in blood glucose
– Influence of environmental factors (e.g.
diet/exercise)
Comparison
Drug Metabolism Pgx
Polymorphisms often lead
to non-functional or
absent proteins
Distinct phenotypes
– Bimodal/trimodal
distribution
Phenotypes are easily
measured
– Drug concentration
– In vitro catalytic activity
Drug Target Pgx
Polymorphisms usually
don’t result in lack of
protein function
– “Subtle” effects
Differences in
phenotypes are smaller
Measurement of
phenotypes is difficult
– Imprecise and variable
Failure to consider
haplotypes
Johnson JA and Lima JJ. Pharmacogenetics 2003; 13:525-534
Examples of Drug Target
Pharmacogenomics
Evans WE. NEJM 2003; 348:538-48
Examples of Disease or Treatment
Modifying Pharmacogenomics
Evans WE. NEJM 2003; 348:538-48
Beta-blockers and
Hypertension (HTN)
HTN is the most prevalent chronic disease in the
US and a contributor to morbidity and mortality
Beta-blockers are first-line agent in the treatment
of HTN
Marked variability in response to beta-blockers
– 30-60% of patients fail to achieve adequate blood
pressure lowering with beta-blockers
Common beta-blockers used in HTN:
– Metoprolol
– Atenolol
Beta-1 Adrenergic Receptor
Codon 49 SerGly
Codon 389
ArgGly
Podlowski, et al. J Mol Med 2000;78:90.
Beta-1 Receptor Polymorphisms
and Response to Metoprolol
Johnson JA et al. Clin Pharmacol Ther 2003; 74:44-52
Beta-2 Adrenergic Receptor
Polymorphisms and Response to
Albuterol in Asthma
Hyperreactivity of the airways is the hallmark of
asthma
Airway smooth muscle contains beta-2 receptors
that produce broncodilation
Albuterol is a beta-2 agonist that is used in the
treatment of asthma
– Produces smooth muscle cell relaxation and
bronchodilation
Forced expiratory volume in 1 second (FEV1)
– Phenotypic measure of response
Beta-2 Polymorphisms and
Response to Albuterol
•Single 8 mg albuterol dose
•Albuterol-evoked increases
in FEV1 were higher and
more rapid in Arg16
homozyotes compared with
Gly carriers
• Codon 16 polymorphism is
a determinant of
bronchodilator response to
albuterol
Lima JJ et al. Clin Pharmacol
Ther 1999; 65: 519-25
Lima JJ. Clin Pharmacol Ther 1999; 65:519-25
VKOR and Warfarin
Warfarin works by inhibiting Vitamin K Epoxide
Reductase (VKOR)
VKOR helps recycle vitamin K which is
important in proper functioning of clotting
factors
By inhibiting VKOR, warfarin alters the vitamin
K cycle and results in the production of inactive
clotting factors
Polymorphisms exist in the gene for VKOR
(VKORC1)
VKORC1 Genotype and Warfarin
Dose Requirements
Mean Warfarin Dose (mg per week)
70
46 mg/wk
60
33 mg/wk
50
40
21 mg/wk
30
20
10
0
G/G
G/A
VKORC1 3673 Genotype
A/A
Warfarin Pharmacogenomics
CYP2C9 SNPs account for a small amount of
variability in warfarin doses (~10%)
VKORC1 SNPs explain a larger portion of
variability in warfarin doses (~20-25%)
Almost 50% of variability in warfarin doses can
be explained by a combination of factors:
– VKORC1 SNPs
– CYP2C9 SNPs
– Non-genetic factors (age, weight, concomitant drugs,
concomitant disease states)
Warfarin Pharmacogenomics
Current Status:
– Develop and validate dosing algorithms to
that include VKORC1 genetic information,
CYP2C9 genetic information, and non-genetic
factors (e.g., age, weight, concomitant drugs,
concomitant disease states)
– Test if dosing warfarin based on genotype is
better than the “usual” care approach
Abacavir Hypersensitivity
Antiretroviral used for treatment of HIV
5% of patients experience hypersensitivity
reactions to the drug
– Hypersensitivity is fatal in rare cases
Hypersensitivity reaction starts with severe GI
symptoms, followed by fever and rash
Discontinuation of drug reverses symptoms
Re-challenge of abacavir in hypersensitive
individuals can result in life-threatening low
blood pressure or death
Lancet 2002;359:727-32.
Abacavir Hypersensitivity
Hypersensitivity typically believed to be an
immunologic reaction
Hypersensitivity might be genetically
linked, and thus predictable
Major histocompatibility proteins (MHC)
investigated because of known links in
other immune responses and allergic
reactions
Genetics of Abacavir Hypersensitivity
Western HIV Australia HIV Cohort Study
• Patients with the HLA-B*5701 variant were 117 times more
likely to be hypersensitive to abacavir than those who did
not have the variant
• 13 patients had 3 linked genetic variants (*5701, DR7, DQ3)
and all patients were abacavir hypersensitive
•All abacavir hypersensitive patients were Caucasian, therefore
studies in other racial groups are needed
Disease Risk Polymorphisms
Polymorphisms can predispose individuals to a
disease or increase the risk for disease
If a drug with a known adverse effect is given to
a person with a genetic susceptibility to that
adverse effect, there is an increased likelihood
for that adverse effect
Clotting Factor Polymorphisms, Blood
Clots, and Oral Contraceptive Pills
Polymorphisms exist in clotting factor genes
Oral contraceptive pills alone are associated with
an increased risk of blood clots
Women who have clotting factor polymorphisms
are at an even greater risk for blood clots if they
receive oral contraceptive pills
Heterozygotes
Oral Contraceptive Pills and Blood Clots
Patients on OCP who are homozygous for Factor V Leiden have
50 to100-fold increased risk of VTE
Martinelli I. Pharmacogenetics 2003; 13:589-594
A Different Aspect of
Drug Target Pharmacogenomics
In all of the examples thus far, the drug target
has been a human protein
The gene encoding that human protein has
generally NOT undergone mutation during the
patient’s life
However, in the areas of infectious diseases and
cancer, the drug target is often a non-human
protein
– Cancer: Tumor DNA
– Infectious Diseases: Viral genotype
e.g., HIV, HBV, HCV
HER-2 Protein and Herceptin
Herceptin (trastuzumab):
– Metastatic breast cancer
– Targets tumor cells that overexpress the human epidermal
growth factor receptor 2
(HER2) protein
– Best response attained in
women who over-express the
HER2 protein
– HER-2 over-expression in
breast cancer cells should be
done before patients receive
the drug
Herceptin: Prescribing Information
HERCEPTIN (Trastuzumab) as a single agent is
indicated for the treatment of patients with
metastatic breast cancer whose tumors
overexpress the HER2 protein and who have
received one or more chemotherapy regimens
for their metastatic disease.
HERCEPTIN should be used in patients whose
tumors have been evaluated with an assay
validated to predict HER2 protein
overexpression
Hepatitis C
Interferon--2b and ribavirin are used to treat
patients with hepatitis C virus
Different mutations exist in the hepatitis C virus
Knowledge of a person’s hepatitis C genotype
may help play a role in the therapeutic decisionmaking process
HIV and Antiretroviral Drugs
Resistance to antiretroviral agents hinders
the management of HIV disease
In the virus, mutations occur which confer
drug resistance
Knowledge of viral genotype (or
phenotype) can help guide the selection of
antiretroviral therapy
Challenges Facing the Field of
Pharmacogenomics
Multiple studies, but literature is inconclusive in
some instances
Genetics accounts for an insufficient percentage
of response variability for a given drug
Few studies documenting genotype-guided
therapy is better than the “usual care” approach
Few “point-of-care” tests available to determine
a person’s genetic make-up or protein
expression
Potential Reasons for
Discrepancies
Inadequately powered studies
Studying different drug response phenotypes
Studying different patient populations
(differences in allele frequencies)
Problems precisely measuring phenotype
Subtlety of functional effects of polymorphisms
Focus on single SNPs instead of haplotypes
Failure to consider the complexity of drug
response
Johnson JA and Lima JJ. Pharmacogenetics 2003; 13:525-534
Pharmacogenetics and
Pharmacogenomics Knowledge Base
(PharmGKB)
Publicly accessible knowledge base
– www.pharmgkb.org
Goal: establish the definitive source of
information about the interaction of genetic
variability and drug response
1. Store and organize primary genotyping data
2. Correlate phenotypic measures of drug response
with genotypic data
3. Curate major findings of the published literature
4. Provide information about complex drug pathways
5. Highlight genes that are critical for understanding
pharmacogenomics
Role of Pharmacogenomics in the
Drug Development Process
80% of products that enter the development
pipeline FAIL to make it to market
Pharmacogenomics may contribute to a
“smarter” drug development process
– Allow for the prediction of efficacy/toxicity during
clinical development
– Make the process more efficient by decreasing the
number of patients required to show efficacy in clinical
trials
– Decrease costs and time to bring drug to market
Pharmacogenomic Paradigm in the
Drug Development Process
Proportion of patients showing
poor or no response
Current Options
Options with Pharmacogenomics
High
Low
Abandon drug
before market
Continue trials safely
by excluding at-risk pts
Continue clinical trials
to market
Optimize clinical trials,
making them
smaller and shorter
Personalized Medicine and the
Pharmaceutical Industry
Targeted Therapies:
– Herceptin: treatment of HER2 positive metastatic
breast cancer
– Gleevec: treatment for patients with Philadelphia
chromosome-positive chronic myeloid leukemia
– Erlotinib: treatment for non-small cell lung cancer
Most effective in epidermal growth factor receptor positive
tumors
– Maraviroc (not approved): treatment for HIV
Studies have incorporated a screening process for different
receptors that HIV uses to gain access to cells
– Iloperidone (not approved): schizophrenia treatment
Company identified a genetic marker that predicts a good
response to the drug
Pharmacogenomic Paradigm in the
Pharmaceutical Industry
Efficacy prediction
Common side effect
prediction
Rare side effect
prediction
Market expansion
FDA and Pharmacogenomics
Traditionally, industry has been hesitant to
submit pharmacogenomic data due to fears of:
– Delays in drug development
– Request for additional clinical studies
– Potentially put clinical trials on hold
FDA published: “Draft Guidance for Industry:
Pharmacogenomic Data Submission” in 2003.
(currently under revision)
Set criteria for Voluntary Genomic Data
Submission (VGDS)
http://www.fda.gov/cder/guidance/5900dft.pdf
Pharmacogenomics Information in
the Published Literature
Zineh I et al. Ann Pharmacother. 2006; 40: 639-44
Pharmacogenomics Information in
FDA-Approved Prescribing Information
Zineh I et al. Pharmacogenomics J; 2004: 1-5
Moving Pharmacogenomics to
Clinical Practice
Document Pgx superiority:
Pgx-guided versus usual care
Documenting sufficient variability
to predict clinical utility
Studies that mimic clinical practice
Proof-of-concept clinical studies
In vitro functional studies
Identify sequence variability in candidate genes
Johnson JA. Trends in Genetics 2003; 19: 660-66
The Future of Pharmacogenomics
Genome wide approach versus candidate gene
approach
Thousands of SNPs
Thousands of patients
Replication studies
Sophisticated databases housing
pharmacogenomic information
Drug selection and dosing algorithms
incorporating non-genetic and genetic
information
Point of care genetic testing
"Here's my
sequence..."
The New Yorker
“Personalized medicine:
elusive dream or imminent reality?
In summary: it is both.”
Larry Lesko, Director of the FDA
Office of Clinical Pharmacology
and Biopharmaceutics
Clin Pharmacol Ther; 2007: 807-816