Getting the Dose Right: Why Are Genes Important

1
Getting the Dose Right Why Are Genes Important?
University of Maryland AAPS ChapterBaltimore,
MarylandNovember 30, 2004

• Lawrence J. Lesko, PhD, FCPDirector, Office of
  Clinical Pharmacology and Biopharmaceutics,
  CDERFood and Drug AdministrationRockville,
  Maryland

2
What I Will Do in This Presentation

• Ultimate goal of drug development providing
  safe therapies with meaningful clinical benefit
• Most promising impact of genomics applying
  biomarkers to inform therapeutic decisions

Drug Gene Interactions ? Optimal Dosing
3
Getting the Dose Right For Whom?

• Pharmaceutical company
• Drug approval
• FDA
• Acceptable benefit/risk
• Physician
• Relief of patient symptoms
• Patient
• Causes no harm

4
Which Dose to Get Right?

• Starting (effective) dose
• Minimize likelihood of ineffective treatments
• Incremental (titrated) dose
• Minimize time to optimal dose with AEs
• Maximum (tolerated) dose
• Minimize the risk of toxicity

Tools preclinical in vitro-in vivo data,
biomarkers, adequate range of human doses, shape
of exposure-response curves, modeling and
simulation, patient biomarker-clinical outcome
links
5
We Can and Should Do Better Finding the Optimal
Dose
FDA Critical Path Document Challenges the
traditional drug development process 1 in 4
approved drugs undergoes relabeling because of
inadequate dosing
6
The Major Challenge Variability in Dose-Response
The Practice of Medicinein 1892
If it were not for the great variability among
individuals, medicine might as well be a science
and not an art
7
Modern Clinical Trial Methodologies Determines
Dose Based on Populations
Efficacy Higher probability of clinical
improvement in controlled patient population,
compared with placebo, using selected, relevant
clinical endpoints
8
Regulatory Review Delicate Balance Between
Benefit and Risk

• Systemic approach - assess the benefit/risk
  under the conditions of use specified in the
  label
• Benefit magnitude of response in target
  population vs. placebo
• Risk - probability and consequences of AE

9
Drug Approval The Usual Dose for Intended
Population
Trial and Error
Symptoms Diagnosis TreatmentDosing
NonspecificBroadNonselectiveOne-size-fits-all
Phenotype
10
Result 1 Unpredictable Variability in Efficacy
Spear et al. Trends in Molecular Medicine, 7201
(2001).
11
Result 2 Unpredictable Variability in Safety
12
Example I Dose-Response Curves for Penicillin
EFFICACY
SAFETY
Probability
of Maximum
13
Dose-Response Curves for 6MP
EFFICACY
SAFETY
Probability
of Maximum
14
Blunt Instrument in Drug Development Special
Population Studies
15
Working Definition of Special Populations
Intrinsic Factors
Extrinsic Factors

• Age
• Body weight
• Gender
• Genotype
• Organ dysfunction

• Diet
• Drug interactions
• Food interactions
• Herbal products
• Smoking habit

Adapted from ICH Guideline E5 Ethnic Factors
inthe Acceptability of Foreign Clinical Data,
1998
16
What Makes Them So Special?
1. Expectations that there will be differences
in pharmacokinetics (sometimes pharmacodynamics)
2. At risk for adverse events (sometimes loss of
effectiveness) because of PK differences3.
Patient groups in which the target disease
frequently occurs (e.g., incidence of epilepsy in
the elderly) 4. Underserved in terms of
inclusion in late clinical stages of drug
development programs
17
Typical Results Special Population Studies
70
60
50
40
Mean Change in AUC
30
20
10
0
Elderly
Gender
Renal (S)
Fluoxetine
Verapamil
Renal (M)
Propranolol
Ergotamine
Hypertension
Moclobemide
Ketoconazole
18
Controversial Decisions..What Do These Results
Mean?
19
Quantitative Methods Facilitate Debate and
Regulatory Decision-Making
D/R Efficacy
Extensive PK Special Pop
M O D E L
D/R Safety
Probability ofBenefit
Probability ofAdverse Event
20
Individualization of Treatment
Emphasis should not be focused on population
averages, but rather on providing prescribers
with the tools to determine the most effective
and safest dosage for individual patients with a
minimum of trial and error
Dr. Gerhard Levy inClinical Pharmacokinetics 34
323-323, 1998
21
Did We Miss the Genetic Era?

• Pharmacogenetics (PGt) Study of
  interindividual variation in DNA sequence
  related to drug absorption and disposition
  (PK) and or drug action (PD) including
  polymorphic variation in genes that encode the
  functions of transporters, metabolizing enzymes,
  receptors and other proteins.

22
Back to the Future Individualized Dosing of 6MP

• Approved over 50 years ago for use in children
  with ALL to maintain remission
• dosing is major determinant of outcome
• 6MP widely used off-label in adults
• Clearance varies up to 10-fold
• 6MP deactivated by TPMT to 6MeMP
• Mutation on TPMT gene on chromosome 6 resulting
  in loss of activity
• Shunting of 6MP to pathway leading to toxic
  buildup of 6-TGN metabolites

23
Pharmacogenetics Polymorphism in TPMT

• Mutations include 2, 3A and 3C
• 1300 are homozygous no activity
• 11100 are heterozygous partial activity
• Genotypes at risk for severe bone marrow
  depression and potential fatalities with usual
  dose of 50 mg/m2
• Dosing
• reduce dose 90 for homozygous deficient and
  50 for heterozygous deficient

24
Genomic Biomarkers to Inform Dosing
2400 children are diagnosed with Acute
Lymphoblastic Leukemia per year
- When starting therapy- First week of therapy-
Overt toxicity
Krynetski and Evans, Amer J Hum Gen 63(1), 11-16,
1998
25
Labeling Regulations
If evidence is available to support the safety
and effectiveness of the drug only in selected
subgroups of the larger population with a
disease, the labeling shall describe the evidence
and identify specific tests needed for selection
and monitoring of patients who need the
drug. - 21 CFR 201.57
26
Irinotecan

• Approved in 1996 for refractory patients with
  metastatic colorectal cancer
• Doses of 300-350 mg/m2 every 3 weeks
• Tumor response rate of 12-15 and prolongs
  survival
• Causes severe diarrhea and neutropenia in 20-35
  of patients (2nd or 3rd line therapy)
• Prevalence of fatal events 5

Rougier et al, Lancet 3521407-1412, 1998Saltz
et al, NEJM 343905-914, 2000
27
Pharmacogenetics UGT Polymorphism

• Irinotecan converted to SN-38 which is
  inactivated by UGT glucuronidation
• UGT1A128 is a variant allele with reduced gene
  expression and glucuronidation
• homozygous UGT1A128 (7/7 genotype) has 2-4 fold
  lower glucuronidation than wild-type (6/6
  genotype)
• increased exposure to SN-38

Ando et al, Cancer Res 606921-6926, 2000Iyer et
al, Pharmacogenomics J 243-47, 2002,
28
Safety Pharmacogenomics of Irinotecan

• Prevalence of grade 4 neutropenia in 59 patients
  was 9.5
• 7/7 genotype (6) ? 50
• 6/7 genotype (24) ? 12.5
• 6/6 genotype (29) ? 0
• Should UGT genotyping be used to identify cancer
  patients predisposed to severe toxicity?

Innocenti et al, J Clin Oncology 221-7, 2004
29
Current Understanding of PGx and Neutropenia
Based on data from Innocenti et al (2004)
30
Potential of UGT Testing to Inform Dosing
All Patients with Same Diagnosis (10 risk)
31
Individualized Medicine Diagnostics
In the very near future, primary care physicians
will routinely perform genetic tests before
writing a prescription because they will want to
identify the poor responders
Dr. Francis Collins, HGRI DirectorAAFP Annual
Meeting, 1998
32
Enter the Genomic Era..

• Pharmacogenomics (PGx) Study of
  interindividual variation in whole genome or
  candidate gene single nucleoside polymorphism
  (SNP) maps, haplotype markers, and alteration in
  gene expression or inactivation that may be
  correlated with pharmacological function or
  therapeutic response

33
Example Variability in Dose Response to Iressa
(Gefitinib)

• A tyrosine kinase inhibitor that targets a tumor
  protein, epidermal growth factor receptor (EGFR)
• Approved for advanced non-small cell lung cancer
  by FDA on 5 May 2003
• overall US response rate only 1 in 10 (n
  216)
• response rate 25-30 in Japan
• other substantial subset differences in US
• women and adenocarcinoma 17, men and smokers
  5
• males and females with more dramatic response
  (median 7 mos)

34
Overall Response Rate to Iressa Was 10 in
Overall Population in RCT
Responders Individuals with a higher
probability of clinical improvement, compared
with placebo, in selected clinical endpoints that
have varying degrees of clinical relevance (e.g.,
tumor shrinkage)
35
Significant Safety Concern with Iressa

• Incidence of interstitial lung disease (n
  23,000)
• 2 in Japanese patients (approved in July 2002)
  and 0.3 in patients outside Japan
• 1/3 of ILD patients died
• Consequences of exposing non-responders to Iressa
  are significant
• A genomic solution to the problem of variability
  in response to Iressa would be very beneficial
• can genomic biomarkers identify responders and
  facilitate individualization

36
Molecular Mechanism Underlying Iressa Sensitivity
Activating Mutations in the Epidermal Growth
Factor Receptor Underlying Responsiveness of
Non-Small Cell Lung Cancer to Gefitinib from the
laboratories of Dr. Daniel Haber, NEJM, 350 (21),
May 3, 2004
37
Results of Iressa Genomic Study

• Specific mutations on EGFR gene correlated with
  clinical response
• deletions or amino acid substitutions around the
  ATP binding site of Iressa
• increased EGFR signaling and susceptibility to
  inhibition
• Mutations identified in 8 of 9 responders
• lung cancer cells with mutations are 10 times
  more responsive than normal cells
• mutations much more common in tumor cells from
  Japanese patients
• Mutations not identified any of 7 non-responders

38
Hypothetical Distribution of Response To Iressa
39
Molecular Mechanism Underlying Iressa Sensitivity
Activating Mutations in the Epidermal Growth
Factor Receptor Underlying Responsiveness of
Non-Small Cell Lung Cancer to Gefitinib from the
laboratories of Dr. Daniel Haber, NEJM, 350 (21),
May 3, 2004
40
Pharmacogenomics Identifies Biomarkers for
Diagnostic Tests

• Identify responders and begin treatment earlier
  to reduce disease progression
• Exclude nonresponders and avoid toxicity in those
  who do not benefit from the drug
• Lift the financial burden (2000 per month) from
  patients who receive no benefit
• Enrich (stratify) clinical trials of tyrosine
  kinase inhibitors for other types of cancer (e.g.
  solid tumors, gliomas)

41
Seismic Shift in the Drug Dosing Paradigm
42
Final Thought Good Drugs Go Bad But They
Dont Have To..
Variability is the law of life, and as no two
faces are the same, so no two bodies are alike,
and no two individuals react alike,and behave
alike under the abnormal conditions which we
know as disease.
Sir William Osler (1849-1919)
43
What I Did in This Presentation

• Ultimate goal of drug development providing
  safe therapies with meaningful clinical benefit
  getting the dose right
• Most promising impact of genomics applying
  biomarkers to inform therapeutic decisions
  diagnostics to inform dosing decisions

44
Thanks for your attention leskol_at_cder.fda.gov