Advisory report Assessment & Improvement of IT Services / IT Service Management Nynke de Vries.
Data Requirements for Developing IVIVE Models · Data Requirements for Developing IVIVE Models...
Transcript of Data Requirements for Developing IVIVE Models · Data Requirements for Developing IVIVE Models...
Data Requirements for Developing IVIVE Models Nynke I. Kramer, PhD Institute for Risk Assessment Sciences Utrecht University [email protected]
Conflict of Interest Statement
I disclose that I have no conflicts of interest
Overview
What data is needed for IVIVE models?
In vitro assays in risk assessment QIVIVE: an integrated approach Parameters in PBPK models Toxicokinetics and toxicodynamics assays Challenges translating in vitro effect concentrations
In Vitro Assays
Toxicodynamics
Toxicokinetics concentration % o
f liv
ing
cells
EC50
In Vitro Assays in Risk Assessment
Toxicity Testing in 21st Century – National Research Council (2007), Toxicity Testing: A Vision and a
Strategy, DOI: 10.17226/11970
From hazard identification to hazard characterization
Quantitative In Vitro-In Vivo Extrapolation (QIVIVE) – Estimating environmental chemical exposures producing
target tissue exposures in humans equivalent to those associated with effects in in vitro toxicity tests
QIVIVE Louisse et al. (2010) Tox. Sci. 118, 470
In vitro toxicodynamics
Mouse embryonic stem cell test
In vitro toxicokinetics Biotransformation kinetics in hepatocytes
alkoxyacetic acid formation
Blood
Fat
Liver
Slowly perfused tissue
Rapidly perfused tissue
GI tract
Blood
Fat
Liver
Slowly perfused tissue
Rapidly perfused tissue
Input: BMC values meta-bolite alkoxyacetic acid
Output: predicted oral embryotoxic dose parent glycol ethers Evaluation of IVIVE model with in
vivo embryotoxic doses
Integrated Approach
PBPK, reverse dosimetry Exposure assessment Sensitivity analysis In silico tools (e.g. read-across, QSAR, QSPR) AOP In vitro tools, KE / TK Margin of exposure Tiered/iterative process
Steps in IVIVE Model Development
Potential Target Tissue
PBPK Model
In vivo Human Toxicty Estimate
In vitro Dynamics
In Vitro Kinetics
QSAR QSPR
Metabolite ID
Target tissue response
In vivo exposure profiles
Nature of Toxicity
• Hepatic clearance • Intestinal uptake /
metabolism • Renal clearance • Partitioning
Reverse Dosimetry
From Miyoung Yoon Yoon et al. 2012, Crit Rev Toxicol. 42, 633
Bottom-up approach
Parameterization PBPK for IVIVE
– Physiological/Anatomical Tissue blood flow or
respiration rate (Q, e.g. L/hr) Tissue volumes (V) Glomerular filtration rate
Well characterized: Brown
et al. (1997) Toxicol. Indust. Health 13, 407. Tissues
Liver
Veno
us b
lood
Arte
rial b
lood
QtxCin
QtxCin
QtxCout
QtxCout
QtxCin
QtxCin
QtxCin
QtxCout
Fat QtxCout QtxCin
metabolism
dAt/dt = Vt·dC/dt = Qt · (Cin – C/P) Cout = Cfree = Ctissue/P dAm/dt = R = Cfree·Vt·K = (Vmax·Cfree)/(KM + Cfree)
Gut/lung/skin
Parameterization PBPK for IVIVE
– Physicochemical Tissue-blood partition
coefficients (P = Ctissue/Cfree) Protein binding constants Tissue clearance/elimination
rates (K, Vmax, KM) Absorption (bioavailability)
In silico tools – QSPR estimating
partitioning/protein binding – Challenge: metabolic
clearance parameter (QSAR limited, mostly qualitative, METEOR, OECD Toolbox, Multicase) In vitro tools
Tissues
Liver
Veno
us b
lood
Arte
rial b
lood
QtxCin
QtxCin
QtxCout
QtxCout
QtxCin
QtxCin
QtxCin
QtxCout
Fat QtxCout QtxCin
metabolism
dAt/dt = Vt·dC/dt = Qt · (Cin – C/P) Cout = Cfree = Ctissue/P dAm/dt = R = Cfree·Vt·K = (Vmax·Cfree)/(KM + Cfree)
Gut/lung/skin
Toxicokinetics Input
In silico tools Tier 1
Passive diffusion down C gradient
Diffusion into tissue Non-saturable protein binding Well-stirred tissue Whole-body D perfusion-limited
Bessems et al. (2014) Reg. Toxicol. Pharmacol. 68, 119 Peyret and Krishnan (2011) SAR QSAR Environ. Res. 22, 129 Liao et al. (2007) Risk Anal. 27, 1223 Wetmore (2015) Toxicol. 332, 94
J = Papp*SA*C
Toxicokinetics Input
In vitro tools Bessems et al. (2014) Reg. Toxicol. Pharmacol. 68, 199
Absorption Oral (Caco-2, PAMPA) Dermal (OECD TG 428, Skin PAMPA) Inhalation (Head space model for VOS, Calu-3)
Distribution Partition coefficients Protein binding Permeation coefficients
Metabolism Human microsomes,
(cryopreserved) primary hepatocytes, HepaRG, HepG2
Excretion See distribution & human
physiology, headspace model, e.g. cell lines RPTEC/TERT1
In Vitro Tools for Distribution Parameters
Method Advantages Disadvantages Equilibrium Dialysis Widely used Time consuming,
membrane binding
Ultracentrifugation No binding to apparatus
Expensive apparatus
Ultrafiltration
Fast and inexpensive
Filter binding, equilibrium shift
Heringa et al. (2003) TAC 22: 575-587
Solid Phase Microextraction
Measuring Kcompartment
SPME
• Measuring free conc. • Nd-SPME
Amount fiber << Amount medium
])[1(1
1
AKKV
VF
aff
mf
⋅+⋅
+=
Cf Cfree
Cb
Ka
Kf
)(
)(
eqfree
eqff C
CK =
)(
)(
infree
eqff C
CK =
well cells
medium fiber
serum protein
Increasing serum concentration
In Vitro Tools for Metabolism Parameters
• Tissue slices • Primary hepatocytes • Hepatic cell lines • Microsomes, cytosol • Recombinant CYPs
• Parent chemical disappearance vs. metabolite generation • Dose-response (Vmax, Km) vs. activity
Toxicodynamics Input
Concentration-effect relationships
AOP ‘Omics’
Interaction with target
Effect tissue/ pathway
Effect organism
AOP framework MIE > KE > AO
Toxicodynamics (TD) Chemical agnostic (biological pathway)
Parameterization Challenges
Variability in in vitro readout Poorly defined / Uncertainties
– Chemical applicability domain – Biological applicability domain/link effect with adversity – In vitro effective dose
Good Modeling Practice – Formal verification process useful
Good Cell Culture Practice – maximise reproducibility, reliability, credibility, acceptance and
proper application of in vitro results – minimum standards in cell and tissue culture
Price and Coecke (2011) in Cell Culture Techniques, Springer, DOI 10.1007/978-1-61779-077-5_1 Loizou et al. (2008) Reg Toxicol. Pharmacol. 50, 400
In Vitro Dose Metrics
plastic binding
free in medium
protein binding
evaporation
cell binding
target free in cell metabolism CELL
MEDIUM
PLASTIC (well)
plastic binding
free in medium
protein binding
cell binding
target free in cell metabolism CELL
MEDIUM
PLASTIC (well)
Cell Assay Components …
… determining target concentration in vitro Medium components (e.g. serum, 3D matrix) Headspace/air flow Well plate dimensions Cell density Exposure time Temperature pH Metabolic capacity Transporter expression
Physicochemical Properties …
…determining target concentration in vitro Kd /LogP/LogD7.4/ KOW
H pKa Solubility Reactivity Transporter affinity Metabolic enzyme affinity
0.91 ± 0.08%
27.50 ± 5.21%
51.49 ± 6.00 %20.10 ± 2.43%
0.96 ± 0.04%
18.99 ± 5.82%
14.80 ± 4.17%
65.75 ± 7.45 %
5.61 ± 1.82% 0.96 ± 0.10%
10.55 ± 3.21%
82.88 ± 3.96 %
Medium Plastic Loss Cells
1.25%
2.5%
5%
Kramer et al. (2012) Chem. Res. Toxicol. 25, 436
Medium Constituents
Kramer et al. (2012) Chem. Res. Toxicol. 25, 436; Kramer (2010) PhD thesis, Utrecht University
3.00 3.50 4.00 4.50 5.00 5.50 6.00 6.502.50
2.75
3.00
3.25
3.50
3.75
4.00
Log KOW
Log
K s (m
3 /mol
)Nominal EC50: 73 µM 105 µM 187 µM
Serum: 1.25% 2.5% 5%
Free EC50: 5 µM 5 µM 6 µM Free: 7% 5% 3%
Polycyclic aromatic hydrocarbons
Medium Components
Kramer et al. (2012) Chem. Res. Toxicol. 25, 436; Kramer (2010) PhD thesis, Utrecht University
3.00 3.50 4.00 4.50 5.00 5.50 6.00 6.502.50
2.75
3.00
3.25
3.50
3.75
4.00
Log KOW
Log
K s (m
3 /mol
)Nominal EC50: 73 µM 105 µM 187 µM
Serum: 1.25% 2.5% 5%
Free EC50: 5 µM 5 µM 6 µM Free: 7% 5% 3%
Polycyclic aromatic hydrocarbons
Medium Components
• Stronger partitioning than Endo and Goss (2011) Chem. Res. Toxicol. 24, 2293, DeBruyn and Gobas (2007) Environ. Toxicol. Chem. 26, 1803 (neutral chemicals)
• Proteins vs. lipids • Neutral vs. ionized
Well Plate Dimensions
5% 2% 0% FBS
Kramer et al. (2012) Chem. Res. Toxicol. 25, 436 Kramer (2010) PhD Thesis, Utrecht University
3.00 4.00 5.00 6.00 7.00-4.00
-3.50
-3.00
-2.50
-2.00
-1.50
-1.00
-0.50
Log KOW
Log
K p (m
) Polycyclic aromatic hydrocarbons
Well Plate Dimensions
5% 2% 0% FBS
Kramer et al. (2012) Chem. Res. Toxicol. 25, 436 Kramer (2010) PhD Thesis, Utrecht University
3.00 4.00 5.00 6.00 7.00-4.00
-3.50
-3.00
-2.50
-2.00
-1.50
-1.00
-0.50
Log KOW
Log
K p (m
) Polycyclic aromatic hydrocarbons
Dependent on: • Chemical (ionization) • Concentration • Time EU FP7 Predict-IV • Kramer et al. 2015 Toxicol. In Vitro 30, 217
Headspace
0.0 0.5 1.0 1.5 2.0 2.5 3.0
-20
0
20
40
60
80
100
120
Alamar BlueCFDA-AMNeutral Red
Log 1,2,4-trichlorobenzene (concentration in µM)
% o
f con
trol
Recovery from medium after 48h
Conventional dosing: 11% Continuous dosing: 105%
EC50 (µM) Conventional: 135 Continuous total: 38 Continuous free: 11 Fathead Minnow: 16
Bound to Serum Constituents
70%
Nominal Cont. total
Cont. free
Kramer et al. (2010) Chem. Res. Toxicol. 23, 1806
Exposure Time
Gülden et al. (2010) Free Radical Bio Med 49, 1298 1800 uM*min = EC50 (see also Gulden et al. (2015) Toxicology 335, 35, time to incipient EC50 is chemical
dependent)
Cell Density
1 10 100
0
20
40
60
80
100
120
Dieldrin (µM)
% o
f via
ble
cells
15 million cells/mL31 million cells/mL62 million cells/mL123 million cells/mL
4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0
4.04.55.05.56.06.57.07.58.0
Log KOW
Log
Klip
id
Gülden et al (2001) Toxicol. In Vitro 15: 233
Jonker et al (2007) Environ. Sci. Technol. 41: 7363
Cell-Associated Concentration Groothuis et al., manuscript in preparation
Cell-Associated Concentration Groothuis et al., manuscript in preparation
• Chemicals acting by narcosis (basal cytotoxicity, accumulation in cell membranes)
• Passive vs. active uptake/efflux • Time
Repeated Dosing
Area under the curve or Cumulative dose (irreversible mechanism?)
Wilmes et al. (2013) J Proteomics 79: 180 Kramer et al. (2015) Toxicol. In Vitro 30: 217
Modeling Kinetics In Vitro
Austin et al (2002) Drug Metab. Disp. 30: 1497-1503 & Austin et al (2005) Drug Metab. Disp. 33: 419-425
Gülden et al (2003) Tox. 189: 211-222 & Gülden et al (2006) Tox. in Vitro 20: 1114-1124
Kramer et al (in preparation) & Armitage et al. (2014) Environ. Sci. Technol.: 48, 9770−9779
Zaldivar Comenges et al (in press) Toxicol. In Vitro & Stadnicka et al. (2014) PloS One 9(3): e92303
Modeling Kinetics In Vitro
F = 1 1 + Ks [S] + Kp [P] + Kc [C] + Ka [A]
HEADSPACE
evaporation (Ka)
PLASTIC (well)
free in cellmetabolismtarget
cell binding (Kc)
free in medium
plastic binding (Kp) protein binding (Ks)
MEDIUM
CELL
HEADSPACE
evaporation (Ka)
PLASTIC (well)
free in cellmetabolismtarget
cell binding (Kc)
free in medium
plastic binding (Kp) protein binding (Ks)
MEDIUM
CELL
m
aKowKowKow
VV
THCPS
F⋅++++
=−−−
3144.8][10][10][101
170.3log25.194.6log97.029.0log37.0
Modeling Kinetics In Vitro
Serum 0% 2% 5% Measured free
21% 8% 5%
Modeled free
32% 9% 5%
Measured in cells
10% 4% 2%
Modeled in cells
14% 5% 3%
RTgill-W1 exposed to phenanthrene
RTgill-W1
Conclusions
Data needed for IVIVE model dependent on – Required precision – Chemical – Exposure
Physiological vs. chemical-specific parameters Tiered, integrated approach Partitioning parameters from in silico tools Clearance parameters from in vitro tools Define applicability domains tools for parameterization Consider in vitro kinetics as well as in vivo kinetics
Thank you
Questions? – At this time, please limit questions to those specific to this
presentation, and save general questions for the Roundtable Discussion with all the presenters.
References
Armitage et al. (2014) Environ. Sci. Technol. 48, 9770−9779 Bessems et al. (2014) Reg. Toxicol. Pharmacol. 68, 119–139 Brown et al. (1997) Toxicol. Indust. Health 13, 407-484 Groothuis et al. (2015) Toxicol. 332, 30-40 Louisse et al. (2010) Toxicol. Sci. 118, 470–484 National Research Council (2007), Toxicity Testing: A Vision and
a Strategy, The National Academies Press, DOI: 10.17226/11970 Yoon et al. (2012) Crit. Rev. Toxicol. 42, 633-652 Peyret and Krishnan (2011) SAR QSAR Environ. Res. 22, 129-
169
Acknowledgements
Research funding: EU FP6 AcuteTox, EU FP7 Predict-IV, Unilever, CEFIC-LRI
Members of the Toxicology Division of UU IRAS incl. Bas Blaauboer, Joop Hermens, Floris Groothuis, Steven Droge, and Chiel Jonker
Miyoung Yoon, ScitoVation