Download An In VitroAlternative For Predicting Systemic Toxicity - In

An In Vitro Alternative For Predicting
Systemic Toxicity
By: James McKim, Ph.D., DABT
Chief Science Officer
What is Systemic Toxicity and What do We Want
From Alternative Methods?
 Toxicity that occurs after a chemical is absorbed into
general circulation
•
Acute systemic toxicity
•
Subacute systemic toxicity
•
Subchronic systemic toxicity
•
Chronic systemic toxicity
Single dose, and short exposure time
Intrinsic toxicity of a chemical, LD50, organ effects
Repeated-dose study, typically 14 day
Information on toxicity following repeated exposure
Helps establish doses for subchronic studies
Repeated-dose, typically 28 and 90 days
Organ specific effects
Establish NOAEL and LOAEL
Regulatory implications FDA and EPA
The Goal is to Predict Human Systemic Toxicity
Most value = Subchronic
Realistically
What data are required for risk assessment
NOAEL, LOAEL
RfD, Benchmark dose
Ideally
Building an In Vitro Model to Predict Repeat Dose
Systemic Toxicity is Too Complex to Achieve
Within the Next 10 Years!
Most models
focus on
predicting
organ specific
toxicity
BARRIER to SUCCESS
Alternative approach is
to identify a plasma concentration
that results in general systemic toxicity
Proprietary to CeeTox, Inc.
Relating Plasma Concentration to Toxicity
 Start with pharmaceuticals
• Intended for internal exposure at higher doses
• Environmental chemicals typically not intended to be
ingested, exposure to lower doses, but for longer times
 Develop a means to estimate the plasma
concentration where general toxicity would
occur
• Develop a relationship between concentration response
curves in vitro and Cmax in vivo at the dose where toxicity
was observed in rat 14-day repeated oral dose studies
 Clinical chemistry
 Histopathology
The Relationship Between In Vivo Plasma
Concentration (Bioavailability) and Toxicity
Provides the Basic Model
JL Stevens (2006) Chem Res Toxicol 19, 1393-1401
Toxicity increases with potency = target
Cell models that lack target show toxicity = chemistry
Proprietary to CeeTox, Inc.
Developing the Hypothesis and Testing it
When the in vivo Cmax for toxicity is identified
at steady state for the lowest dose that results
in toxicity, and compared to in vitro
concentration response curves for toxicity, a
relationship can be identified that enables the
prediction of the in vivo plasma concentration
where general toxicity would be expected to
occur.
Building the Predictive Model Based on
Empirical Data Sets
• Acquired in vivo data for pharmacokinetic parameters
• Developed concentration response data in vitro
• Based on multiple endpoint analysis developed a
relationship between in vivo and in vitro data sets
• Tested the model with small number of drugs
• Validated model in large blinded study
Focus on Cell Biology and Physical Chemical
Properties Improves the Model
Biochemical Function
Membrane Integrity
Mitochondrial Function
Cell Proliferation
Redox-State
Oxidative Stress
Apoptosis
Pharmacology
CNS receptors
Cardiovascular receptors and ion
channels
Physical-Chemical Properties
Solubility
Partition coefficients
Pka
Protein binding
Metabolic stability
Metabolic activation
Transporter interaction
Pharmacokinetic Parameters
Clearance
Bioavailability
Volume of distribution
Selecting a Cell Model That Consistently
Provides Accurate Data
 Rat hepatoma (H4IIe) Cell line
• Retains oxidative metabolism
• Low constitutive metabolism
• Accepts a wide range of serum concentrations
Analysis of In Vitro Concentration-Response to
In Vivo Plasma Levels
 Multiple endpoints improve interpretation
• Membrane integrity
• Cell number or mass
• Mitochondrial function
 Analysis and comparison to in vivo plasma Cmax
 Development of weighting factors for each
endpoint
 Result was an estimate of the plasma
concentration where toxicity would be
expected to occur (Ctox)
Multiple Endpoints Are Essential For Correct
Interpretation of In Vitro Data
CMPD H4IIE 24HR 20050817
Cell#
MemTox
MTT
ATP
GSH
140
120
% Control
100
80
60
40
20
0
1
10
100
FIG. A: Exposure Concentration (µM)
1000
Determining Weighting Factors
 Membrane integrity
• Highest weight = cell death
• Concentration response and potency
• Time adjustment
 Mitochondrial function
• Moderate weighting
• Concentration response and potency
• Time adjustment
Evaluating Assumptions Against Rat
In Vivo Data
100
R2 = 0.85
80
60
Keto (rat)
40
20
0
0
20
40
60
80
100 120
Predicted Concentration (uM)
In Vitro Toxicity Data Correlates With Adverse
Effects in Animals and Humans
CHLOROQUINE H4IIE 24HR 20041006
140
Cell#
MemTox
MTT
ATP
Ctox = 5µM
120
Rat actual Ctox = 8 µM,
Human actual Ctox = 1-3 µM
100
80
KETOCONAZOLE H4IIE 24HR
140
60
120
20
100
% Control
40
0
0.1
1
10
100
Exposure Concentration (µM)
1000
Cell#
MemTox
MTT
ATP
METHOTREXATE H4IIE 24HR
20060118, 20060329
Ctox = 1
140
120
Rat actual Ctox = 0.7 µM, Human actual Ctox = 0.88 µM
60
40
Rat Actual Ctox = 60 µM,
Human actual Ctox = 30 µM
20
0
1
10
100
Exposure Concentration (µM)
80
60
40
20
0
0.001
80
0.1
100
0.1
10
FIG. A: Exposure Concentration
1000
Cell#
MemTox
MTT
ATP
Ctox = 55 µM
1000
Metabolic Activation Identifies False Negative
Results
 Tacrine = liver
 Flutamide = liver
 Acetaminophen = liver
% GSH relative
GSH Depletion relative to no drug control after 3.5 hr incubation with
induced microsomes
140.0
120.0
100.0
80.0
60.0
40.0
20.0
0.0
No Drug
Flutamide
NEM
Acetaminophen
Tacrine
Test Article
Full Reaction
-NADPH reaction
Ctox (µM)
Determining Relevance to Human
Systemic Toxicity
100
10
Azathioprine
1
1e-8
1e-7
1e-6
1e-5
1e-4
1e-3
max therapeutic [plasma] (log M)
Ctox = 5 µM
MTPC = 8 µM In vitro margin of safety <= 1
Proprietary to CeeTox, Inc.
Azathioprine: Inhibits Purine Synthesis and
Produces Toxicity Under Clinical Conditions
AZATHIOPRINE H4IIE 24HR 20051012
AZATHIOPRINE H4IIE 24HR 20051012
% Control
100
120
40
GSH % Control
120
80
60
100
80
30
60
20
40
40
20
20
10
0
0
1
10
100
0
1
1000
10
100
FIG. B: Exposure Concentration (µM)
FIG. A: Exposure Concentration (µM)
AZATHIOPRINE H4IIE 24HR 20051012
Fluorescence Units
500
Extremely Toxic
Ctox = < 5 µM
Caspase 3
400
300
200
100
0
1
GSH
ISO
50
140
Cell#
MemTox
MTT
ATP
10
100
1000
FIG. C: Exposure Concentration (µM)
Proprietary to CeeTox, Inc.
1000
8-Isoprostane (pg/ml)
140
Ctox (µM)
There is Good Concordance Between Predicted Toxicity
Validated to Rat Data and Human Clinical Data
100
Thioridazine
10
Azathioprine
1
1e-8
1e-7
1e-6
1e-5
1e-4
1e-3
max therapeutic [plasma] (log M)
THIORIDAZINE H4IIE 24HR 20041006
140
Cell#
MemTox
MTT
ATP
120
100
80
60
40
20
0
0.1
1
10
100
Exposure Concentration (µM)
1000
Ctox = 11 µM
MTPC = 6 µM In vitro margin of safety = 1.83
Drug Potency Can Influence Plasma
Concentration and Toxicity
Ctox (µM)
Valproic acid
100
10
Paroxetine
Chloroquine
Thioridazine
Azathioprine
1
1e-8
1e-7
1e-6
1e-5
1e-4
1e-3
max therapeutic [plasma] (log M)
Antidepressant Paxil (SSRI) Considered Safe Under
Appropriate Use: Rare Incidences of Hepatotoxicity
Cell#
MemTox
MTT
ATP
Paroxetine H4IIE 24HR 20051108
140
120
100
80
60
40
20
0
1
10
100
FIG. A: Exposure Concentration
Ctox = 15 µM
MTPC = 0.3 µM
1000
In Vitro Margin of Safety = 50
Predicting Bioavailability of a New Drug and
Maximum Therapeutic Plasma Concentration
• In Vitro CLint (human microsomes)
• Scaled in vitro CLint
• Calculate bioavailability (%f)
– f=1–E
– E = fu x CLint / QH
Kuhnz and Gieschen (1998) Drug Metab Disp 26, 1120.
What is Systemic Toxicity and What do We Want
From Alternative Methods?
 Toxicity that occurs after a chemical is absorbed into general
circulation
 Acute systemic toxicity
 Single dose, and short exposure time
 Intrinsic toxicity of a chemical, LD50, organ effects
 Subacute systemic toxicity
 Repeated-dose study, typically 14 day
 Information on toxicity following repeated exposure
 Helps establish doses for subchronic studies
 Subchronic systemic toxicity
 Repeated-dose, typically 28 and 90 days
 Organ specific effects
 Establish NOAEL and LOAEL
 Regulatory implications FDA and EPA
 Chronic systemic toxicity
Predicting a Rat Acute Oral LD50 Dose
 Collaborative research effort with LOREAL Paris
 Integrative or systems biology + Computational
Toxicology
• Multiple endpoints related to cell health
• Receptor binding data (pharmacology)
• Physical Chemical parameters
• Approach has been evaluated with more than 200
chemicals
• Posters at the World Congress in Rome 2009, SOT 2010
 Results show that an integrated in vitro approach
can provide good estimates of an LD50
Addition of Pharmacology and Physical Chemistry
Properties Improves Model Performance
Performance of the standard model at the
LD50 threshold of 500 mg/kg
Performance of the integrated model at the
LD50 threshold of 500 mg/kg
REDUCTION OF THE FALSE NEGATIVE RATE (yellow area)
Conclusions
 There is a robust model capable of identifying acute and
subchronic systemic toxicity (Today)
 Chemicals can be evaluated for systemic toxicity
potential
 High level of concordance with in vivo toxicity
 An acute LD50 can be determined
Future Improvements to the Model
 Develop a drug database with animal preclinical data
and in vitro toxicity data
 Determine a calculation for NOAEL and LOAEL
 Broaden the chemical space to include non-
pharmaceutical agents and test the model
CeeTox Overview of Services
… a wide range of in vitro toxicology and safety assessment services
 Systemic Toxicity Services
–
‒
CTOX Panel®
Acute Toxicity Screen
 In vitro LD50 (acute tox)
 Dermal Toxicity





Percutaneous Absorption
Corrosion
Irritation
Sensitization
Photo Toxicity
 Ocular Toxicity
 Endocrine Disruption





Estrogen receptor (ER) binding
Estrogen transcriptional activation
Androgen receptor (AR) binding
Steroidogenesis
Aromatase
 Organ Specific Toxicity
Models
–


Heart (CardioTox®)
Kidney
Liver
 CYP Induction/Inhibition
 Drug / Drug Interaction
 Target / Pathway Assays













Acute Inflammatory Response
Apoptosis
Cell Proliferation
Cell Viability
Glutathione Homeostasis
Hepatobiliary Toxicity
Kidney Collecting Duct Toxicity
Kidney Distal Tubular Toxicity
Kidney Proximal Tubule Toxicity
Lipidosis
Membrane Integrity
Metabolism
Mitochondrial Function