Transcript Diapositive 1 - Physiologie et Thérapeutique Ecole Véto Toulouse

NATIONAL
VETERINARY
SCHOOL
TOULOUSE
Bioequivalence :
some challenges and issues
P.L. Toutain
National Veterinary School Toulouse France
Informal CVMP/CMDv Paris 2008
CVMP CMD Paris 2008 - 1
[PHENYTOIN]
µg/mL
Change in phenytoin excipients results in
epidemic toxicity
WEEKS
* Bochner F, et al. Proc Aust Assoc Neurol 1973;9:165-70
CVMP CMD Paris 2008 - 2
Non-bioequivalence of various
trademarks of enrofloxacin in cow
Sumano & al 2001 Dtsch tierärztl Wschr 108 281-320
CVMP CMD Paris 2008 - 3
Aims of the presentation
Give an overview on the concept of bioequivalence
Discuss some technical points in relationship with the CVMP guideline
CVMP CMD Paris 2008 - 4
Bioequivalence :
Definition and assumptions
CVMP CMD Paris 2008 - 5
Bioequivalence : Definition
• Two medicinal products are bioequivalent if
their bioavailabilities (rate and extent) after
administration in the same molar dose are
similar to such degree that their effect and
safety will be essentially the same (E.U. for
man)
CVMP CMD Paris 2008 - 6
Basic assumptions for
bioequivalence
CVMP CMD Paris 2008 - 7
Bioequivalence :
The basic assumption
• “Similar” overall plasma exposure
 same effects (efficacy, safety)
–is it always true ?
• Classical objections
–Plasma concentration is not
biophase concentration
–there is no (univocal) relationships
between exposure and effect !
CVMP CMD Paris 2008 - 8
Relevance of plasma concentrations
was challenged for ivermectins
CVMP CMD Paris 2008 - 9
Basic assumption to bioequivalence
Is there an univocal relationship between
exposure and effect ? yes
DOSE
yes
Plasma concentrations
Systemic Effects
driven by plasma
concentrations
Yes
yes
yes
Local effects
not driven by plasma
concentrations
Yes
Plasma concentrations
Yes/No ?
CVMP CMD Paris 2008 - 10
Why to use bioavailability
outcomes to demonstrate
bioequivalence
CVMP CMD Paris 2008 - 11
Basic assumption to bioequivalence
Similar plasma concentration profile  same effect ?
Why ?
Effect
Drug property (efficacy)
Effect =
Emax
Emax  Dose
ED50 + Dose
ED50
Dose
Hybrid drug and formulation properties (Potency)
CVMP CMD Paris 2008 - 12
Basic assumption to bioequivalence
Drug property
ED50 =
Clearance  EC50
Bioavailability
Formulation property
CVMP CMD Paris 2008 - 13
Basic assumption to bioequivalence
• Similar plasma concentration profile 
same effect?
Effect =
substance properties
Emax  Dose
Clearance  EC50 + Dose
F%
Formulation properties
CVMP CMD Paris 2008 - 14
Bioequivalence
vs.
Bioavailability
CVMP CMD Paris 2008 - 15
Bioavailability vs Bioequivalence
A
?
=
B
injection
Analytical approach
in vivo approach
injection
"the cattle column"
HPLC column
Pharmaceutical equivalence
In vivo equivalence
Both HPLC and animal "column" need to be reproducible
CVMP CMD Paris 2008 - 16
Bioequivalence vs. Bioavailability
Inference from a trial
- Bioavailability
No generalization from a subgroup of animal to the
population
- Bioequivalence
If B.E. is demonstrated in a particular subgroup of
animals, conclusion should be extended to the
whole population unless there is an interaction
between formulation and a constitutional factor
CVMP CMD Paris 2008 - 17
Generics: advantages and
disadvantages in terms of
public health
CVMP CMD Paris 2008 - 18
Generics:
Public health perspective
• to promote access to medicines through
(inter alia) generic competition
– generic competition is desirable as it may
contribute to significant drug prices decrease
– Financial advantage for health care providers
• But what about public health perspective
for veterinary drugs??
CVMP CMD Paris 2008 - 19
Generic competition for drugs availability:
Q1: is it a good news for veterinary
medicine especially for antibiotics?
CVMP CMD Paris 2008 - 20
In: Clinical infectious deseases 2005 41 114-117
CVMP CMD Paris 2008 - 21
Correlation between community use and the
number of trade names for oral-use agents for 6
antibacterial classes in EU
High consumption countries
Nb of trade names
Low consumption countries
Nb of trade names
CVMP CMD Paris 2008 - 22
Generic competition for drugs availability:
Q2: Is it a good veterinary practice to
encourage the use of old antibiotics rather
than new ones?
CVMP CMD Paris 2008 - 23
Is it a good veterinary practice to encourage the
use of old antibiotics rather new ones?
• Traditionally, from a public health perspective,
veterinarians are encouraged not to employ new
drugs, but rather to use the older antibiotics.
• The recommendation whether to choose older
rather than new antibiotics was recently
challenged on an epidemiological basis
(Amyes et al., 2007) and shown to be flawed
for quinolones, cephalosporins and
carbapenems.
CVMP CMD Paris 2008 - 24
. For
three antibiotic classes (quinolones, cephalosporins
and carbapenems), it was observed that the less active
drugs could be worse at hastening the spread of
resistance than more active drugs in the same class.
This led the authors to qualify the (WHO) stratagem of
recommending the use of old antibiotics as part of
microbiological folklore.
CVMP CMD Paris 2008 - 25
Is it a good veterinary practice to encourage the
use of old antibiotics rather new ones?
• Old antibiotics were developed long time
ago and their dosage regimens need to be
reassessed (e.g. tetracyclines; quinolones
etc.).
CVMP CMD Paris 2008 - 26
Technical issues addressed in the presentation
• Bioequivalence: the a priori bioequivalence range
• Types of bioequivalence trials
•
•
•
•
•
•
– PK, PD vs. clinical trials
– Average, population bioequivalence & individual
bioequivalence
Interspecies extrapolation
Bioequivalence and drug residues
Male vs. female in bioequivalence trial
Single dose vs. multiple doses
Data analysis: the sequence effect
GLP vs. field conditions
CVMP CMD Paris 2008 - 27
Bioequivalence :
The a priori
bioequivalence range
CVMP CMD Paris 2008 - 28
A priori bioequivalence range
• These are the two limits ( 1, 2 ) between
which the 90% CI interval of the
difference or the ratio of the two products
should be located in order to accept
average B.E.
•To be defined by the clinician
CVMP CMD Paris 2008 - 29
Decision procedures in bioequivalence trials
BE not
accepted
1
- 20%
2
BE not
accepted
the 90 % CI of the
difference or of the ratio
BE accepted
+20%
µt
Mean of ref
formulation
CVMP CMD Paris 2008 - 30
A priori Bioequivalence range:
For AUC-difference or AUC-ratio
General rule: The 90% confidence interval for
this metrics should lie within an a priori
acceptance interval:
± 20 % of the reference product (for the
difference, untransformed)
-20, +25% for the ratio (i.e. 0.8-1.25 for Lntransformed)
CVMP CMD Paris 2008 - 31
A priori Bioequivalence range:
for Cmax-difference or Cmax-ratio
0.70 - 1.30 (untransformed) or 0.70 - 1.43
(Ln-transformed)
Wider interval is accepted because Cmax is
a single observed extreme concentration
CVMP CMD Paris 2008 - 32
PK/PD relationship to discuss
bioequivalence acceptance criteria
Effect
•A 20% difference
for PK has no
impact on efficacy
•Typically for safe
drugs
∆ = 20%
Exposure
CVMP CMD Paris 2008 - 33
Effect
PK/PD relationship to discuss
bioequivalence acceptance criteria
•A 20% difference
for PK may have an
impact on efficacy
•Typically for
unsafe drugs
∆ = 20%
Exposure
CVMP CMD Paris 2008 - 34
The case of antiepileptic drugs
CVMP CMD Paris 2008 - 35
A priori Bioequivalence range for
drug with a narrow therapeutic index
0.90 - 1.10 (Untransformed)
0.90 - 1.11 (Ln-transformed)
CVMP CMD Paris 2008 - 36
A priori Bioequivalence range for
drug with a large therapeutic index
• If there are no animal safety or efficacy
concerns, is it possible to accept a larger
interval (e.g. in case of highly variable drugs)?
• Therapeutics vs. equity considerations
• Concern about the possibility of study-bystudy alteration of approval acceptance
criteria
CVMP CMD Paris 2008 - 37
Types of bioequivalence trials
Order of preference
1.
2.
3.
4.
Pharmacokinetic
In vitro studies
Pharmacodynamic*
Clinical*
* Not considered in draft 4
CVMP CMD Paris 2008 - 38
Clinical or pharmacological endpoint
studies to demonstrate BE for
systemically acting drugs
• Motivations
– Blood drug concentration is not quantifiable
– Excessive variability in the blood
concentration/time profile
– Manufacturer/ regulators consider that blood
level profiles are irrelevant (or less relevant
than a clinical endpoint)
CVMP CMD Paris 2008 - 39
Types of Bioequivalence trial
Pharmacodynamic endpoints
Test
Reference
Effect
100 %
Response A
50 %
T and R are not
bioequivalent
Systemic exposure
AUC
CVMP CMD Paris 2008 - 40
Types of Bioequivalence trial
Pharmacodynamic endpoints
Test
Reference
Effect
100 %
Response B
50 %
T and R are
bioequivalent
Systemic exposure
AUC
CVMP CMD Paris 2008 - 41
Average
vs.
population bioequivalence
vs.
individual bioequivalence
CVMP CMD Paris 2008 - 42
Different types of bioequivalence
• Average (ABE) : mean
• Population (PBE) : prescriptability
• Individual (IBE) : switchability
CVMP CMD Paris 2008 - 43
Average bioequivalence
reference
test
Same mean
CVMP CMD Paris 2008 - 44
Average bioequivalence
Average B.E. refers to the location parameters
Average B.E. may not be sufficient to
guarantee that an individual patient could be
switched from a reference to a generic
formulation
(e.g., more than 50% of subjects may be
outside the B.E. range when the average B.E.
is actually demonstrated!!!!)
CVMP CMD Paris 2008 - 45
Average bioequivalence
• Addresses only mean (center of
distribution) but not variability
(shape of distribution)
• Does not address switchability
CVMP CMD Paris 2008 - 46
Prescribability
• Refers to the clinical setting in which a
practitioner prescribes a drug product to a
patient for the first time
• he has no information on his patient
• the prescriber needs to know the
comparability of the 2 or n formulations in
the population
population bioequivalence
CVMP CMD Paris 2008 - 47
Population bioequivalence
Population dosage regimen
Yes
No
Pigs that eat less:
Possible
underexposure
Pigs that eat more
Possible
overexposure
CVMP CMD Paris 2008 - 48
Individual bioequivalence
test
reference
NO
animal-by-formulation
interaction
YES
Address switchability
“Test” and “reference” are bioequivalent if the individual
subject means and variabilitites are sufficiently similar
with regard to AUC and Cmax
CVMP CMD Paris 2008 - 49
Switchability
• Refers to the clinical setting in which a
practitioner transfers a patient from one drug
product to another
• We have information on the response of the
patient to a particular formulation and
clinicians have titrated the dose to reach a
particular goal
• issue for drug of critical therapeutic
categories, for elderly, debilitated patients etc.
CVMP CMD Paris 2008 - 50
The types of bioequivalence
Average
Population
Individual
Pioneer
Test
Only
guarantees
on the mean
Guarantees an
overall distribution
(mean and
variance)
Test of no interaction
between patient and
formulation guarantees
an individual BE
CVMP CMD Paris 2008 - 51
Individual bioequivalence
• The clinical relevance of a subject-byformulation interaction has not clearly
been demonstrated
–e.g.: a pH-specific excipient effect
associated with certain diazepam
formulations result in producing
unequivalence when administered to
individuals with elevated gastric pH (like
elderly)
CVMP CMD Paris 2008 - 52
?
Generic 2
Generic 1
yes
yes
?
yes
Pioneer
???
Generic 3
Other reference
medicinal product
CVMP CMD Paris 2008 - 53
Bioequivalence:
interspecies extrapolation
CVMP CMD Paris 2008 - 54
Bioequivalence:
interspecies extrapolation
• A major obstacle to generic drug approval is
the current need to submit in-vivo BE study
data for each major species
– BE cannot be extrapolated across target animal
species
• Associated scientific question
– Is there a potential “species-by-formulation”
interaction (that is different from a species effect)
• No: BE demonstration in a single species is appropriate
• Yes : BE should be demonstrated in each target species
CVMP CMD Paris 2008 - 55
CVMP CMD Paris 2008 - 57
Digestive tract:
major interspecies differences
Assumption of a no species-by-formulation interaction is
not reasonable for oral formulation
What about IM, SQ…
CVMP CMD Paris 2008 - 58
Two SQ IVM formulation differed in the rate of drug absorption in
pigs but not in calves
CVMP CMD Paris 2008 - 59
Bioequivalence
and the
problem of drug residues
CVMP CMD Paris 2008 - 60
CVMP BE guideline (draft 4)
• When bioequivalence data are used to
substantiate an extrapolation of a withdrawal
period between formulations the entire 90%
confidence interval for the ratio must be within
the 80-125 % acceptance limits for both AUC
and Cmax.
– My comment: To substantiate = To establish???
• If broader limits are used, then residue data to
confirm the withdrawal period is required unless
its absence can be justified.
CVMP CMD Paris 2008 - 61
Bioequivalence and the
problem of drug residues
• Bioequivalence studies in foodproducing animals are not
acceptable in lieu of residues
data: why?
CVMP CMD Paris 2008 - 62
Definitions and statistics
associated to (average)
bioequivalence and withdrawal
time are fundamentally different
CVMP CMD Paris 2008 - 63
Bioequivalence and withdrawal time
• Bioequivalence is related to a confidence interval for a
parameter (e.g. geometric mean AUC-ratio for 2
formulations)
• Withdrawal time is related to a tolerance limit (quantile 95%
EU or 99% in US) and it is define as the time when the
upper one-sided 95% tolerance limit for residue is below the
MRL with 95% confidence
• The fact to guarantee that the 90% confidence interval for
the AUC-ratio of the two formulations lie within an
acceptance interval of 0.80-1.25 does not guarantee that
the upper one-sided 95% tolerance limit for residue is below
the MRL with 95% confidence for both formulations
CVMP CMD Paris 2008 - 64
Bioequivalence and withdrawal time
Formulation A
Formulation B
AUCA = AUCB
A and B are BE
Concentration
Mean curve
Mean curve
Individuals
individuals
Time
CVMP CMD Paris 2008 - 65
Bioequivalence and withdrawal time
Formulation A
WTA < WTB
Formulation B
Concentration
MRL
WTA
Time
WT
CVMP CMD Paris
B 2008 - 66
Bioequivalence and withdrawal time
• Withdrawal times are generally much longer than
the time for which plasma concentrations were
measured for BE demonstration
Pionner
WT for
the generic
Generic
WT for
the pionner
LOQ
BE
WT
CVMP CMD Paris 2008 - 67
Who is affected by an inadequate
statistical risk associated to a WT
• It is not a consumer safety issue
• It is the farmer that is protected by the
statistical risk associated to a WT
–It is the risk, for a farmer, to be controlled
positive while he actually observe the WT.
–When the WT is actually observed, at
least 95% of the farmers in an average of
95% of cases should be negative!!!
CVMP CMD Paris 2008 - 68
Bioequivalence:
test subjects
CVMP CMD Paris 2008 - 69
Bioequivalence: test subjects
• Some issues on the selection of test
subjects
–healthy or diseased animals?
• Possible interaction between health
studies and formulation
–sex: both male and female?
CVMP CMD Paris 2008 - 70
Male vs. female in
bioequivalence trial
• Issue: do we need to select animals of
both sexes (male and female) to test
bioequivalence?
• there were long debates during the writing
of version 2 of the EMEA guidelines
because sex effect was confused with sexby-formulation interaction !
CVMP CMD Paris 2008 - 71
Sex bioavailability and bioequivalence
A sex effect
AUC
Sex effect
Frequent in human medicine because BW is not considered !
Moxidectin in dog (male > female)
CVMP CMD Paris 2008 - 72
Sex bioavailability and bioequivalence
Interaction sex/formulation
(A vs. B)
A B
B
A B
BE
A B
BE
A
not BE
CVMP CMD Paris 2008 - 73
Sex, bioavailability and bioequivalence
• Question: do we need to test both sexes?
–Bioavailability
yes : possible sex effect (actually few examples in
veterinary medicine but frequent in human
medicine because BW is not taken into account
for dosage regimen)
–Bioequivalence
no : interaction formulation*sex unlikely
see: Chen ML et al Pharmacokinetic analysis of bioequivalence
trials: implication for sex related issues in clinical pharmacology
and biopharmaceutics. Clin. Pharmacol. 2000, 68: 510-521
CVMP CMD Paris 2008 - 74
Gender representation in trials
• US congress enacts legislation to require
that a clinical trial must be “designed and
carried out in a manner sufficient to provide
for a valid analysis of whether the variables
being studied in the trial affect women…
differently than other subjects in the trial”
CVMP CMD Paris 2008 - 75
Single dose vs. multiple doses
steady state studies
CVMP CMD Paris 2008 - 76
Study design in general
(CVMP BE Guideline Draft 4)
• Regarding single dose versus multiple dose
studies, single dose studies are preferred as
the potential to detect a difference in rate of
absorption is lower if the drug is accumulated.
• Multiple dose designs should be justified and
could be considered if e.g. problems of
sensitivity of analytical method preclude
sufficiently precise plasma concentration
measurements after single dose administration.
CVMP CMD Paris 2008 - 77
Single dose vs. multiple dose
steady state studies
• Two bio-inequivalent formulations
(single dose) may become
bioequivalent under steady-state
conditions
CVMP CMD Paris 2008 - 78
Single dose vs. multiple dose steady state studies
2 products that are
not bioequivalent
after a single dose
may appear to be
bioequivalent in a
multiple dose
administration
K01=0.1 vs. 0.05h-1 single dose administration
0.7
Formulation1
0.6
0.5
Formulation2
FFormulation
2ormulation2
0.4
1
0.3
2
0.2
0.1
0.0
0
50
100
150
200
250
300
Time (h)
K01=0.1 vs 0.05h-1. Multiple doses administrations
2.5
Formulation1
2.0
1.5
1.0
1
2
Formulation2
0.5
0.0
0
50
100
150
Time (h)
200
250
300
CVMP CMD Paris 2008 - 79
Data analysis
The sequence effect
CVMP CMD Paris 2008 - 80
Bioequivalence :
experimental design
- 2x2 crossover
groups
or
sequences
periods
1
2
1
A
B
2
B
A
- other crossover
e.g. : AB, BA, AA, BB ( BALAAM design )
CVMP CMD Paris 2008 - 81
Why to perform an ANOVA
• To validate the cross-over design and
especially to detect a sequence effect
• To estimate the residual which is
required for the two one-sided test
procedures
CVMP CMD Paris 2008 - 82
What does the new guideline say
CVMP BE guideline (draft 4)
• Data analysis has to be given in detail.
• Formulation, period, sequence and animalnested-in-sequence effects should be
investigated in the analysis of variance and
the results should be presented.
• If any effect is found significant, its
relevance should be discussed.
CVMP CMD Paris 2008 - 83
The 2x2 cross-over design
The period effect
• Not desirable
• Does not invalidate a cross-over design
• Origin : e.g. enzymatic induction by the
substance, environment,
• Period effect = equal carryover effect
CVMP CMD Paris 2008 - 84
The carryover effect
• The direct drug effect is the effect that
a drug produces during the period in
which the drug is administered
• The carryover effect is the drug effect
that persists after the end of the
dosing period ("memory effect")
CVMP CMD Paris 2008 - 85
The carryover effect
Origin: a too short washout period
• The washout period is the rest period between 2
treatment periods
• The duration depends on the drug
• Should be long enough to avoid a carryover effect of
PK or PD origin
– PD origin: local tolerance if the same site of injection is
selected for the 2 periods, enzymatic inhibition or induction
etc.
CVMP CMD Paris 2008 - 86
Equal vs. unequal carry-over effect
Period 1
Period 2
A
B
B
A
Period 1
Period 2
A
B
B
A
Equal carryover
effect gives a
period effect
Unequal carry-over effect
gives a sequence effect that
is totally confounded in a 2x2
crossover design with a
formulation-by-period
interaction
CVMP CMD Paris 2008 - 87
The 2x2 cross-over design
The sequence effect
• Not desirable
• invalidate a cross-over design
• Origin: residual drug amount for one of the 2 tested
formulations, enzymatic induction by a vehicle
associated with one of the tested formulations etc.,
• Sequence effect = unequal carryover
CVMP CMD Paris 2008 - 88
The sequence (or the differential
carryover) effect
• If the carryover effects are unequal,
no unbiased estimate exists for the
direct effects from both periods
• Note: for the sequence effect the
nominal risk is not qualified in the
CVMP draft (10%??)
CVMP CMD Paris 2008 - 89
The conduct of
bioequivalence studies
CVMP CMD Paris 2008 - 90
The conduct of bioequivalence studies
(CVMP BE Guideline Draft 4)
•
“Bioequivalence studies should be conducted under Good
Laboratory Practice (GLP)”.
–
BUT also
•
For two products, pharmacokinetic equivalence (i.e.
bioequivalence) is established if the rate and extent of
absorption of the active compound investigated under
identical and appropriate experimental conditions only differ
within acceptable predefined limits
•
Are GLP conditions always appropriate
conditions???
–
The case of Pour-on formulations
CVMP CMD Paris 2008 - 91
Pour-on administration
•Cattle lick (auto- and allo-licking).
It is a natural behavior in cattle
CVMP CMD Paris 2008 - 92
Ivermectin disposition after a pour-on
formulation administration
500 µg/kg Pour-on
ingestion 58 - 87%
55 %
Oral absorption
22 %
Plasma
overall bioavailability
IVM
13 % of the
dose
FECES
7 - 14 % of the
dose absorbed
by skin
F total = 18 - 68 %
CVMP CMD Paris 2008 - 93
Pour-on administration
• Pour-on formulations are skin formulations under GLP
conditions but oral formulation under field
conditions!!!
• Possible behavior-by-formulation interaction
(palatability) and BE in GLP conditions does not
guarantee BE in field conditions!!!
CVMP CMD Paris 2008 - 94
Regulatory concerns raised by
licking
• How to explain that none of the so-called
GLP trials revealed the phenomenon?
• Do you think that here GLP is synonymous
to good science?
• Population kinetics (in field conditions)
would have immediately evidenced this
phenomenon
CVMP CMD Paris 2008 - 95
Concerns raised by this social
drug exposition
•
•
•
•
Efficacy
Safety
Environment
Regulation (GLP study?)
CVMP CMD Paris 2008 - 96
CONCLUSION
CVMP CMD Paris 2008 - 97