Transcript BPA - Physiologie et Thérapeutique Ecole Véto Toulouse (ENVT)

BPA and conjugated -BPA:
Pharmacokinetics across species
Pierre-Louis Toutain
National Veterinary School of Toulouse, France
NIEHS BPA Grantee research Meeting
January 28-29,2013
Research Triangle Park, North Carolina, USA.
1
Goal of the presentation
• How PK data across different animal
species can contribute to predict and
understand the internal exposure to the
bioactive form of BPA in humans
2
Exposure
External
Internal
Oral
Inhalation
Skin
Urine
Plasma
Mass information
Concentration
Total BPA
BPA
Risk
management
options
To determine what has
been actually
absorbed
(≠bioavailable)
To discuss safety issues
To extrapolate between species
and from in vitro to in vivo
3
A fundamental relationship
• This relationship is not a model but results of
definitions:
– For a linear kinetic , clearance is a scaling parameter
between a dose and plasma concentration
4
Body (plasma, blood) clearance
• The most important PK parameter
• Unknown in man
• Should be estimated by a
modeling approach
5
The most simple model for
Interspecies extrapolation
Log parameter
Allometric relationship
LogY  a  b  LogBW
Log BW
6
Allometric estimation of plasma
clearance in man
30mL/kg/min
Cho et al 2002 Xenobiotica 32 925-934
About 23.8 mL/kg/min
Doerge et al 2012 Tox letters
7
Allometric estimation of plasma clearance in man
ClBPA (L/min)=0.0440 x BW0.8645
BPA clearance (L/min)
10
1
0.1
Man
0.01
ClBPA = 24.7 mL/(kg.min)
or 103L/h for a 70kg BW
0.001
Prediction interval [9.3-66.0]
0.0001
0.01
0.1
Collet et al Unpublished results
1
10
100
Body weight (kg)
1000
8
An order of magnitude for BPA plasma
clearance in man is of 25ml/kg/min
• Q2: is it a high or low clearance?
9
Physiological Interpretation of
BPA body clearance
• Interpretation of body clearance consists of
calculating an overall extraction ratio
10
Physiological Interpretation of
BPA body clearance
• Interpretation of body clearance consists of
calculating an extraction ratio
An extraction ratio of 0.31 is high
(typical value for drugs is <0.05)
11
What is the origin of this high
plasma clearance and why this
question?
• High Hepatic or high renal
clearance?
• If BPA has a high hepatic clearance,
a low bioavailability is expected for
the oral route (gavage)
12
Clearances are additive
Negligible
To predict the maximal oral bioavailability
• Theoretically very easy to
determine without BPA
administration;
Renal clearance of BPA is very low
Renal clearance was less than 1% in the
5 investigated species after an IV
administration
Hepatic BPA clearance in man:
in vitro estimation
Clh was estimated at 22 mL/kg/min in
humans i.e. to exactly the same value as
from allometry.
Interpretation of a high
hepatic clearance to predict
the extent of an hepatic firstpass effect and of a maximal
possible oral bioavailability
16
The hepatic portal system
17
Presystemic metabolism
First pass effect
Gut
(oral route)
Lumen
Gut Wall
Portal
vein
Liver
To site of
measurement
Hepatic Metabolism
BPA Gut wall
Metabolism:
To feces
NO
YES
BPA absorption, first pass effect and bioavailability
for the oral route
Liver
GIT
BPA
Fraction of
the dose that
is absorbed
Fmax = 1 - Eh
Eh
Fraction eliminated by a first pass effect
Fraction non absorbed
Fmax = 1 - Eh
Hepatic first-pass effect
Maximal oral bioavailability (Fmax)
Bioavailability of BPA across species
0.9%
•
•
•
•
0.6%
1.3%
3.0%
4.1%
BPA IV: 5mg/kg; BPA oral: 100 mg/kg
Simultaneous BPA and BPA-G quantification in plasma and in urine samples were
performed with an Acquity ultra performance liquid chromatography (UPLC®) coupled to a
Xevo triple quadrupole mass spectrometer (Waters, Milford, MA, USA) (Lacroix et al 2011).
LOQ of BPA from 1 10 ng/mL in plasma and 10 to 25 ng/mL in urine
Data analysis: NCA
Collet et al unpublished results
21
Estimation the oral bioavailability of BPA in
rhesus monkey (data from Doerge et al 2010)
Both IV and oral data
F%=0.9±0.58%
22
BPA
bioavailability vs. absorption
Bioavailability vs. absorption
• Absorption: movement of drug from the
site of administration into the blood
which drains the site of administration
• Bioavailability refers to the amount of
drug which actually gains the access
to the systemic (arterial) circulation
Why it is important to assess the
extent of oral BPA absorption?
• To validate that the urinary data can be
used to estimate the BPA daily intake of
BPA
Computation of the absorbed
fraction vs. the bioavailability
Assuming that the drug is only eliminated hepatically (i.e. that
there is no or minimal renal elimination of the parent substance
as for BPA) the fraction of drug that is absorbed is uniquely
determined by the AUC ratio of metabolite
26
Absorbed fraction of BPA after an
oral administration across species
98.4
•
•
•
•
63.7
75.67
41
98
BPA IV: 5mg/kg; BPA oral: 100 mg/kg
Simultaneous BPA and BPA-G quantification in plasma and in urine samples were
performed with an Acquity ultra performance liquid chromatography (UPLC®) coupled to a
Xevo triple quadrupole mass spectrometer (Waters, Milford, MA, USA) (Lacroix et al 2011).
LOQ of BPA-G from 20 to 50 ng/mL in plasma and of 1000ng/mLin urine
Non Comp analysis (sheep, pig, dog) and comp modeling (rat, mouse)
Collet et al unpublished results
27
Absorbed fraction of BPA after an
oral administration in man
• Human subjects were orally administered d-6-bisphenol A (5 mg in
toto).
• d16-Bisphenol A glucuronide was the only metabolite of d16bisphenol A detected in urine and blood samples, and
concentrations of free d16-bisphenol A were below the limit of
detection both in urine (6 nM) and blood samples (10 nM).
• The applied doses were completely recovered in
urine as d16-bisphenol A glucuronide.
What are the predicted plasma
concentrations of BPA in the general
population:
a Monte Carlo Simulation
29
What are the plasma concentrations of
BPA in the general population
30
Pg/mL
Average daily intake: about 0.05µg/kg/day
31
The 2005–2006 National Health and
Nutrition Examination Survey (NHANES)
were used to estimate daily BPA intakes
F from 1 to 5%
• From allometry:
24.7±4.94ml/kg/min
Plasma BPA concentrations (pg/ml)
Monte Carlo Simulation
33
Average plasma concentration: 0.12pg/mL; percentile 90: 0.26pg/mL or 1.1pM
Predicted plasma BPA concentrations in
the US population: a sensitivity analysis
• Daily intake is the
most influential
variable in the
prediction of plasma
BPA level.
• And Plasma
clearance the less
influential
34
IV Plasma clearance variability
• Many possible factors of variability
• No population information but intersubject
variability cannot generate a difference of
several order of magnitude between
subjects
– An example: effect of obesity
35
Influence of obesity on the disposition of BPA
Obese after feeding
Control after feeding
Time (h post administration)
Parameters
units
Control
Obese
Change%
Clearance
mL/kg/min
105
81
-22%
Vss
mL/kg
2984
3965
+32%
MRT
min
29
50
+72%
36
Are we sure of a systematic low oral
bioavailability in man?
• A low oral bioavailability has been actually measured
in all investigated species
• But BPA is generally administered by gavage allowing
a full first-pass effect .
•
•
•
This is the first report of serum BPA concentrations in an animal model exposed to BPA
via the diet.
Bolus administration underestimates bioavailable serum BPA concentrations in
animals—and presumably humans—than would result from dietary exposure.
Exposure via diet is a more natural continuous exposure route than oral bolus exposure
and is thus a better predictor of BPA concentrations in chronically exposed animals and
humans.
37
Environ Health Perspect 119:1260–1265 (2011).
The 3 segments of the digestive tract in
terms of first-pass effect
Buccal cavity
No
first-pass
effect
Small intestine/large bowel
Full First pass-effect
Rectal
Limited
first-pass effect
38
Buccal/Sublingual
administration to bypass the
hepatic first-pass effect?
Buccal/Sublingual absorption
• sublingual (SL) = under tongue
The surface area of the oral mucosa (200 cm2) is relatively small
compared with the gastrointestinal tract (350 000 cm2) and skin (20 000
cm2).
• Scintigraphic imaging revealed that
clearance of the formulation from the oral
cavity was rapid, with a mean T50%
clearance of 0.86±0.46 min, and T80%
clearance of 2.75±1.52 min
41
42
Selection of an animal model:
Keratinized vs. non-keratinized oral mucosa
• Non keratinized oral mucosa:
– dogs, rabbits, pigs, and Rhesus monkeys are
acceptable models, yielding permeability values
similar to those found for humans.
• Keratinized oral mucosa
Yes
– Pose a significant barrier to intra-oral absorption that impact
its correlation to that in humans.
– the heavily keratinized rat or hamster oral mucosa tends to
underestimate absorption from the non-keratinized human
oral mucosa.
No
43
The dog was selected as a model
• 6 beagle dogs
• Routes of administration: IV, sublingual and gavage
• Dose tested: high (5mg/kg) in ethanol and low (50µg/kg)
with water as vehicle delivered over 10 mins towards the
floor of the mouth
• Sampling at the jugular vein
44
Results for a representative dog
Plasma BPA (ng/ml)
10000.0
BPA 5mg/kg
1000.0
BPA 0.05mg/kg
100.0
10.0
1.0
0
100
200
300
400
Plasma BPAG (ng/ml)
100000.0
100.0
10000.0
1000.0
10.0
100.0
10.0
1.0
1.0
0
0
100
200
300
20
40
60
80
100
120
400
Time (min post administration)
Sublingual # IV
45
Results for the 6 dogs
BPA (0.05mg/kg) i.e the TDI
Plasma BPA (ng/ml)
BPA (5mg/kg) i.e the NOAEL
10000
10000
Sampling at the jugular vein
1000
1000
100
100
10
10
1
1
2
4
8
15
30
60
90 120
2
4
8
15 30 60 90 120
Time post administration (min)
For a dose of 1µg/kg, plasma concentration should be about 1ng/mL
46
Bioavailability of BPA by sublingual
administration
Oral
20mg/kg
Sublingual
5mg/kg
Sublingual
50µg/kg
Bioavailability %
(from BPA AUCs)
0.72
±0.28
81
± 29
>100%
Absorption%
(From BPA-gluc AUCs)
54
±19
81
±18
90
±26
47
The concentration ratio of BPAG to BPA in
plasma:
IV, sublingual or oral administration
• The concentration ratio of BPAG to BPA in plasma was approximately
100-fold lower following sublingual administration than after oral
dosing allowing to easily distinguishing the two pathway of absorption.
ORAL
SL
IV
30 min post ad
48
Buccal exposure
BPA was found in all paper
currencies at concentration
ranging up to 82.7µg/g
Crown restoration of a molar may release
13µg BPA in the average case scenario or
30mg BPA in the worst case scenario both
49
after 24h
Hand-to-Mouth exposure
• Average indoor hand-to-mouth behavior ranged from 6.7 to 28.0
contacts/hour,
– with the lowest value corresponding to the 6 to <11 year olds and the highest
value corresponding to the 3 to <6 month olds.
•
Average outdoor hand-to-mouth frequency ranged from 2.9 to 14.5
contacts/hour,
–
with the lowest value corresponding to the 6 to <11 year olds and the highest
value corresponding to the 6 to<12 month olds.
50
Volume of
distribution
Absorption
Clearance
bioavailability
Half-life
BPA
accumulation
51
BPA Systemic
exposure
BPA and BPA-gluc disposition :
Time dependency
60 daily SQ administrations of BPA to pregnant ewes (5mg/kg/day does not lead to
BPA or BPA-Gluc accumulation
No evidence of accumulation
Viguie et al
The case of the very late terminal phase: an
issue for the question of a possible BPA
bioaccumulation in a (small) deep compartment
Very late terminal phase
Bioaccumulation
due to the very late
terminal phase
In case of a very late terminal phase , bioaccumulation
can occur in a deep compartment while plasma
53
concentration profile remains
unchanged
BPA distribution:
the case of fetus
24 IV infusion of BPA 2 mg/(kg.d)
Analytical technic measuring selectively BPA, BPA-G and BPA-S
54
Materno-fetal transfer of BPA
24 IV infusion of BPA 2 mg/(kg.d)
Maternal Compartment
Fetal Compartment
PLACENTA
BPA Internal (blood)
Exposure:
6 ng/mL
BPA Internal
(blood) Exposure:
40 ng/mL
2%
BPA-G and BPA-S trapped in
amniotic fluid
98%
As in man, BPA exposure lower in fetus than in mother but higher
exposure for BPA-G in the fetus
Feto-maternal transfer of BPA
24 IV infusion of BPA 5 mg/(kg.d)
Maternal Compartment
Fetal Compartment
PLACENTA
BPA Internal Exposure:
25 ng/mL
BPA Internal Exposure:
4 ng/mL
91%
Total BPA
Clearance :
250 mL/(kg.min)
9%
BPA-G and BPA-S trapped in
amniotic fluid
Placenta eliminated 90% of fetal BPA
Feto-maternal transfer of BPA-G (120 days)
IV infusion of BPA-G 3.54 mg/(kg.d)
Maternal Compartment
PLACENTA
Fetal Compartment
BPA-G Internal
Exposure:
4530 ng/mL
BPA and BPA-G
never detected
BPA-G?
BPA was never
detected
(LOQ=1ng/mL)
No reactivation of BPA-G into BPA
observed despite a very high dose
of infused BPA-G in the fetus
Principles to extrapolate results
(effects) between species
58
The question of the safe dose in man should be
established and discussed through the determination
of a safe plasma concentration
Range1/1000
Range1/2
Toxicokinetics (satellite groups) is in order
59
Same profile (CL/F
similar) but both plasma
clearance and F are
higher in mouse than in
monkey (X3-4)
60
Comparison of effective plasma
concentrations in different species
• Only free plasma concentration is active
• Plasma protein binding is a factor of
confusion
Fu is the confounding factor (0.06 in man; 0.05 in sheep; 0.1 in lamb fetus)
61
The dose effect relationship
and PKPD consideration
62
Response
What is the shape of the dose-effect
relationship for BPA?
• [Drug]
63
What is the shape of the dose-effect
relationship
64
Conclusion
Plasma
BPA=1ng/mL
(4.38nM)
Probable
NO
Possible
YES
65
Thanks for your attention
The BPA team
S.Collet & T.Corbel
V.Gayrard
N Hagen
C Viguié
M. Lacroix & S.Puel
Answers to 10 questions
67
Q1:what we really know and don’t know:
Plasma clearance
– We have no direct measurement of plasma BPA
clearance but only indirect estimates (allometry).
– BPA body clearance is (very) high.
• We ignore population distribution of this parameter
(subpopulations?)
– We know that the renal BPA clearance is very low
– BPA clearance is mainly hepatic
• Thus a high first-pass effect is expected for the
oral route (gavage)
68
Q2:what we really know and don’t know:
Absorption & bioavailability
• Extent of oral absorption is high but
bioavailability (different of absorption) for
oral route (gavage) is (very) low
– Sublingual route of BPA absorption escape to
a the hepatic first-pass effect and can lead to
relevant BPA plasma concentrations
– Rodents are not a good model for sublingual
absorption
69
Q3:what we really know and don’t know:
half-life
• Currently, reported BPA half-life is short thus no
accumulation possible
– No experimental evidence of BPA accumulation in sheep
• However a very late terminal phase cannot be
exclude (it is a question of LOQ) with an
accumulation in a deep compartment without
alteration of BPA plasma concentration
70
Q4:what we really know and don’t know:
linearity
• BPA and BPA-gluc disposition are linear
(dose proportionality)
– Possibility to extrapolate between doses for
exposure
• No time dependency
71
Q5:what we really know and don’t know:
Plasma concentration
• Plasma concentration is the driving force
controlling biophase (receptors) concentrations
• With the current available data (a high plasma
clearance, a low oral bioavailability and a low
daily intake), predicted plasma concentration
should be very low (less than 1pg/mL).
If plasma concentration can actually
reach few ng/mL (no contamination) ,
one or several of the aforementioned
statements is (are) totally wrong
72
Q6: Are humans really different from other species
(rodents) with regard to pharmacokinetics and
metabolism
• No qualitative relevant differences
– Allometric law apply
– The case of enterohepatic recycling in rodents
• Some quantitative differences
All species and all route of BPA administration
can provide relevant results to discuss systemic
effects in man as long as plasma concentrations
in the test system are monitored and used to
bridge results with the human plasma exposure
73
Q8: Can one really use a urine
measurement to develop a PK model of
what the human exposure is?
• Yes.
• Urine is the main pathway of BPAconjugates elimination in man
• Urine information is a mass information
and mass balance concepts apply
(determination of the daily BPA absorbed
(not bioavailable) dose.
74
Q9:Value of biomonitoring
data
• How to explain that several studies using a
variety of different analytical technique
have measured free, unconjugated BPA
concentration in humam serum at levels
ranging from 0.2 to 20ng/mL?
– Analytical bias that overestimate BPA in
plasma ?
– Biomonitoring (sampling conditions) bias that
underestimate total BPA in urine?
75
Q10: If data doesn’t match the model
which do you throw out…the data or
the model?
• The model should be rejected
• But
– A sound model can help to detect flawed data
(bad analytical technics), unplausible results
etc
• Quotes by Box: ‘all models are wrong; the
practical question is how wrong do they
have to be to not be useful’.
76