Dose-Response Modeling: Past, Present, and
Future
Rory B. Conolly, Sc.D.
Center for Computational Systems Biology
& Human Health Assessment
CIIT Centers for Health Research
(919) 558-1330 - voice
rconolly@ciit.org - e-mail
SOT Risk Assessment Specialty Section, Wednesday, December 15, 2004
1
Outline
• Why do we care about dose response?
• Historical perspective
– Brief, incomplete!
• Formaldehyde
• Future directions
2
Perspective
• This talk mostly deals with issues of cancer risk
assessment, but I see no reason for any formal
separation of the methodologies for cancer and non
cancer dose-response assessments
– PK
– Modes of action
– Tumors, reproductive failure, organ tox, etc.
3
Response
Typical high dose rodent data – what do
they tell us?
Dose
4
Response
Not much!
Dose
Interspecies
5
Response
Possibilities
Dose
Interspecies
6
Response
Possibilities
Dose
Interspecies
7
Response
Possibilities
Dose
Interspecies
8
Response
Possibilities
Dose
Interspecies
9
Benzene Decision of 1980
• U.S. Supreme Court says that exposure standards must
be accompanied by a demonstration of “significant
risk”
– Impetus for modeling low-dose dose response
10
1984 Styrene PBPK model
(TAP, 73:159-175, 1984)
A physiologically based description of
the inhalation pharmacokinetics of
styrene in rats and humans
John C. Ramseya and Melvin E. Andersenb
a
Toxicology Research Laboratory, Dow Chemical USA, Midland, Michigan
48640, USA
b Biochemical Toxicology Branch, Air Force Aerospace Medical Research
Laboratory (AFAMRL/THB), Wright-Patterson Air Force Base, Ohio 45433,
USA
11
Biologically motivated computational models
(or)
Biologically based computational models
• Biology determines
– The shape of the dose-response curve
– The qualitative and quantitative aspects of interspecies
extrapolation
• Biological structure and associated behavior can be
– described mathematically
– encoded in computer programs
– simulated
12
Biologically-based computational models:
Natural bridges between research and risk
assessment
Experiments to understand
mechanisms of toxicity and
extrapolation issues
Computational
models
Risk
assessment
13
Garbage in – garbage out
• Computational modeling and laboratory experiments
must go hand-in-hand
14
Response
Refining the description with research on
pharmacokinetics and pharmacodynamics
(mode of action)
Dose
Interspecies
15
Response
Refining the description with research on
pharmacokinetics and pharmacodynamics
(mode of action)
Dose
Interspecies
16
Response
Refining the description with research on
pharmacokinetics and pharmacodynamics
(mode of action)
Dose
Interspecies
17
Response
Refining the description with research on
pharmacokinetics and pharmacodynamics
(mode of action)
Dose
Interspecies
18
Formaldehyde nasal cancer in rats:
A good example of extrapolations across
doses and species
19
20
1980 - First report of formaldehyde-induced
tumors
Swenberg JA, Kerns WD, Mitchell RI, Gralla EJ,
Pavkov KL
Cancer Research, 40:3398-3402 (1980)
Induction of squamous cell carcinomas of the rat
nasal cavity by inhalation exposure to
formaldehyde vapor.
21
Formaldehyde bioassay results
60
Kerns et al., 1983
Monticello et al., 1990
40
30
20
10
0
0
0.7
2
6
10
Tumor Response
(%)
50
15
Exposure Concentration (ppm)
22
Mechanistic Studies and Risk Assessments
23
What did we know in the early ’80’s?
• Formaldehyde is a carcinogen in rats and mice
• Human exposures roughly a factor of 10 of
exposure levels that are carcinogenic to rodents.
24
1982 – Consumer Product Safety
Commission (CPSC) voted to ban ureaformaldehyde foam insulation.
25
1983 - Formaldehyde cross-links DNA
with proteins - “DPX”
Casanova-Schmitz M, Heck HD
Toxicol Appl Pharmacol 70:121-32 (1983)
Effects of formaldehyde exposure on the
extractability of DNA from proteins in the rat
nasal mucosa.
26
DPX
RESPIRATORY
EPITHELIUM
MUCUS
CH2
(FORMALDEHYDE
IN AIR)
CH2
CH2
27
1984 - Risk Assessment Implications
Starr TB, Buck RD
Fundam Appl Toxicol 4:740-53 (1984)
The importance of delivered dose in
estimating low-dose cancer risk from
inhalation exposure to formaldehyde.
28
1985 – No effect on blood levels
Heck, Hd’A, Casanova-Schmitz, M, Dodd,
PD, Schachter, EN, Witek, TJ, and Tosun, T
Am. Ind. Hyg. Assoc. J. 46:1. (1985)
Formaldehyde (C2HO) concentrations in the
blood of humans and Fisher-344 rats
exposed to C2HO under controlled
conditions.
29
1987 – U.S. EPA cancer risk assessment
• Linearized multistage (LMS) model
– Low dose linear
– Dose input was inhaled ppm
– U.S. EPA declined to use DPX data
30
Summary: 1980’s
• Research
– DPX – delivered dose
– Breathing rate protects the mouse (Barrow)
– Blood levels unchanged
• Regulatory actions
– CPSC ban
– US EPA risk assessment
31
Key events during the ’90s
• Greater regulatory acceptance of mechanistic data for
risk assessment (U.S. EPA)
• Cell replication dose-response
• Better understanding of DPX (Casanova & Heck)
• Dose-response modeling of DPX (Conolly, Schlosser)
• Sophisticated nasal dosimetry modeling (Kimbell)
• Clonal growth models for cancer risk assessment
(Moolgavkar)
32
1991 – US EPA cancer risk assessment
• Linearized multistage (LMS) model
– Low dose linear
– DPX used as measure of dose
33
1991, 1996 - regenerative cellular
proliferation
Monticello TM, Miller FJ, Morgan KT
Toxicol Appl Pharmacol 111:409-21 (1991)
Regional increases in rat nasal epithelial cell
proliferation following acute and subchronic
inhalation of formaldehyde.
34
Normal respiratory epithelium
in the rat nose
35
Formaldehyde-exposed respiratory epithelium
in the rat nose (10+ ppm)
36
Dose-response for cell division rate
7.00E-04
(Raw data)
6.00E-04
5.00E-04
4.00E-04
3.00E-04
2.00E-04
0
1
2
3
4
5
6
7
ppm formaldehyde
37
DPX submodel – simulation of rhesus
monkey data
DPX dose-response for Rhesus monkey
3
DPX (pmol/mm)
10
10
-1
-2
Vmax: 91.02. pmol/mm3/min
Km: 6.69 pmol/mm3
kf: 1.0878 1/min
10
10
-3
Tissue thickness
ALWS: 0.5401 mm
MT: 0.3120 mm
NP: 0.2719 mm
-4
1
2
3
4
5
6
7
PPM
38
Summary: Dose-response inputs to the
clonal growth model
• Cell replication
– J-shaped
• DPX
– Low dose linear
39
CFD Simulation of Nasal Airflow
(Kimbell et. al)
40
2-Stage clonal growth model
(MVK model)
Division
(aN)
(aI)
Normal
cells (N)
Initiated
cells (I)
Cancer
cell
(delay)
Death/
differentiation
(bI)
(bN)
Tumor
41
Dose-response for cell division rate
7.00E-04
5.5000E-04
(Hockey stick transformation)
(Raw data)
5.0000E-04
6.00E-04
4.5000E-04
5.00E-04
4.0000E-04
3.5000E-04
4.00E-04
3.0000E-04
3.00E-04
2.5000E-04
2.00E-04
2.0000E-04
0
1
2
3
4
5
ppm formaldehyde
6
7
0
1
2
3
4
5
6
ppm formaldehyde
42
7
Simulation of tumor response in rats
43
CIIT clonal growth cancer risk assessment
for formaldehyde
(late ’90’s)
• Risk assessment goal
– Combine effects of cytotoxicity and mutagenicity to
predict the tumor response
44
1987 U.S. EPA
Inhaled ppm
Cancer model
(LMS)
Tumor response
45
1991 U.S. EPA
Inhaled ppm
Tissue dose
(DPX)
Cancer model
(LMS)
Tumor response
46
1999 CIIT
Inhaled ppm
CFD modeling
Tissue dose
Cell killing
Mutagenicity
(DPX)
Cell proliferation
Cancer model
(Clonal growth)
Tumor response
47
Formaldehyde: Computational fluid dynamics
models of the nasal airways
F344 Rat
Rhesus Monkey
Human
48
Human assessment
Inhaled
formaldehyde
exposure scenario
CFD nasal
dosimetry model
single-path lung
dosimetry model
cell replication
in control rats
cells at risk in
respiratory tract
respiratory tract
tumor data
(control only)
site-specific flux into
respiratory tract epithelium
cell replication
dose-response (rat)
mode of action
dose-response
submodels
DPX dose-response
prediction (scale-up
from rat and monkey)
2-STAGE
CLONAL
GROWTH
MODEL
2-STAGE
CLONAL
GROWTH
MODEL
maximum likelihood
estimation of baseline
parameter values
human tumor
incidence
49
Baseline calibration against human lung
cancer data
50
DPX and direct mutation
• Direct mutation is assumed to be proportional to the
amount of DPX:
mutation  KMU  DPX
• Is KMU big or small?
51
Grid search
52
Optimal value of KMU is zero
53
Upper bound on KMU
54
Calculation of the value of KMU
• Grid search
• Optimal value of KMU was zero
– Modeling implies that direct mutation is not a
significant action of formaldehyde
• 95% upper confidence limit on KMU was estimated
55
Human risk modeling
Division
(aN)
(aI)
Normal
cells (N)
Initiated
cells (I)
Cancer
cell
(delay)
Death/
differentiation
(bI)
(bN)
Tumor
56
Final model: Hockey stick and 95% upper
confidence limit on value of KMU
95% UCL on KMU
DPX dose-response for Rhesus monkey
3
DPX (pmol/mm)
10
10
-1
-2
Vmax: 91.02. pmol/mm3/min
Km: 6.69 pmol/mm3
kf: 1.0878 1/min
10
10
-3
Tissue thickness
ALWS: 0.5401 mm
MT: 0.3120 mm
NP: 0.2719 mm
-4
1
2
3
4
5
6
7
PPM
5.5000E-04
5.0000E-04
4.5000E-04
4.0000E-04
3.5000E-04
3.0000E-04
2.5000E-04
2.0000E-04
0
1
2
3
4
5
6
7
57
Predicted human cancer risks
(hockey stick-shaped dose-response for cell
replication; optimal value for KMU)
Optimal value of KMU
KMU = 0.
DPX dose-response for Rhesus monkey
5.5000E-04
10
-1
5.0000E-04
3
DPX (pmol/mm)
4.5000E-04
4.0000E-04
3.5000E-04
10
Vmax: 91.02. pmol/mm3/min
Km: 6.69 pmol/mm3
kf: 1.0878 1/min
10
3.0000E-04
-2
-3
Tissue thickness
ALWS: 0.5401 mm
MT: 0.3120 mm
NP: 0.2719 mm
2.5000E-04
10
2.0000E-04
0
1
2
3
4
5
6
7
-4
1
2
3
4
5
6
7
PPM
58
“Negative risk” using raw dose-response
for cell replication
DPX dose-response for Rhesus monkey
10
-1
3
DPX (pmol/mm)
95% UCL on KMU
10
-2
Vmax: 91.02. pmol/mm3/min
Km: 6.69 pmol/mm3
kf: 1.0878 1/min
10
10
-3
Tissue thickness
ALWS: 0.5401 mm
MT: 0.3120 mm
NP: 0.2719 mm
-4
1
2
3
4
5
6
7
PPM
7.00E-04
6.00E-04
5.00E-04
4.00E-04
3.00E-04
2.00E-04
0
1
2
3
4
5
6
7
59
Make conservative choices when faced with
uncertainty
• Use hockey stick-shaped cell replication
• Use a 95% upper bound on the dose-response for the
directly mutagenic mode of action
– Statistically optimal model has 0 (zero) slope
• Risk model predicts low-dose linear risk.
• Optimal, data based model predicts negative risk at
low doses
60
Summary: CIIT Clonal Growth Assessment
• Either no additional risk or a much smaller level of
risk than previous assessments
• Consistent with mechanistic database
– Direct mutagenicity
– Cell replication
61
Summary: CIIT Clonal Growth Assessment
• International acceptance
–
–
–
–
–
Health Canada
WHO
MAK Commission (Germany)
Australia
U.S. EPA (??)
• Peer-review
62
IARC 2004
• Classified 1A based on nasopharyngeal cancer
• Myeloid leukemia data suggestive but not sufficient
– Concern about mechanism
– British study negative
• Reclassification driven by epidemiology
• In my opinion inadequate consideration of regional
dosimetry
63
Whole
nose
Anterior
nose
nasopharynx
64
IARC: hazard characterization vs. doseresponse assessment
65
Formaldehyde summary
•
•
•
•
•
Nasal SCC in rats
Mechanistic studies
Risk Assessments
Implications of the data
IARC
66
The future
67
Outline
•
•
•
•
•
Long-range goal
Systems in biological organization
Molecular pathways
Data
Example
– Computational modeling
– Modularity
68
Long-range goal
• A molecular-level understanding of dose- and timeresponse behaviors in laboratory animals and people.
– Environmental risk assessment
– Drug development
– Public health
69
Levels of biological organization
Populations
Descriptive
Organisms
(systems)
Tissues
(systems) Mechanistic
Cells
(systems)
Organelles
Molecules
(systems)
(systems)
70
Levels of biological organization
Populations
Organisms
Tissues
Cells
Organelles
Today
Molecules
(systems)
71
Molecular pathways
72
Segment polarity genes in Drosophila
Albert & Othmer, J. Theor Biol. 223, 1 – 18, 2003
73
ATM curated
Pathway from
Pathway Assist®
74
Approach
• Initial pathway identification
– Static map
• Existing data
• New data
• Computational modeling
– Dynamic behavior
– Iterate with data collection
75
Initial pathway identification
• Use commercial software that can integrate data from
a variety of sources (Pathway Assist)
– Scan Pub Med abstracts to identify “facts”
– Create pathway maps
– Incorporate other, unpublished data
• Quality control
– Curate pathways
76
Computational modeling
• To study the dynamic behavior of the pathway
• Analyze data
– Are model predictions consistent with existing data?
• Make predictions
– Suggest new experiments
– Ability to predict data before it is collected is a good
test of the model
77
DNA damage and cell cycle checkpoints
(a) G1/S Checkpoint
(b) G2/M Checkpoint
78
p21 time-course data and simulation
Experimental data
79
Mutation Fraction Rate
Mutations dose-response and model
prediction
model calculated values
IR
(Redpath et al, 2001)
80
Data
81
Tissue dosimetry is the “front end” to a
molecular pathway model
(Fat)
Air-blood
interface
Liver
Venous
blood
Rest of Body
82
Gain-of-function and loss-of-function
screens to study network structure
• Selectively alter behavior of the network
– Loss-of-function
• SiRNA
– Gain-of-function
• full-length genes
• Look for concordance between lab studies and the
behavior of the computational model
– Mimic gain-of-function and loss-of-function changes in
the computer
83
Example
• Skin irritation
• MAPK, IL-1a, and NF-kB computational “modules”
• High throughput overexpression data to characterize
IL-1a – MAPK interaction with respect to NF-kB
84
Skin Irritation
Chemical
Dead cells
Tissue damage
Tissue
damage
Nerve
Endings
A cascade of inflammatory
responses (cytokines)
Epidermis
(keratinocytes)
Dermis
(fibroblasts)
Blood vessels
•
Study on the dose response of the skin cells to inflammatory cytokines contributes
to quantitative assessment of skin irritation
85
Modular Composition of IL-1 Signaling
IL-1
IL-1R
Secondary messenger
Constitutive NF-kB
downstream
NF-kB
module
MAPK
Extracellular
Intracellular
IL-1 specific
top module
Others
IL-6, etc.
Transcriptional factors
86
Top IL-1 Signaling Module
P
MyD88
TRAF6
P
IRAK
TAB1
TAK1
TAB2
P
P
TRAF6
Self-limiting
mechanism
IkK
IkK P
NF-kB module
Degraded
IRAK
Cytoplasm
IRAK gene
Nucleus
87
Top Module Simulation
• IL-1 receptor number and ligand binding
parameters from human keratinocytes
• Other parameters constrained by reasonable
ranges of similar reactions/molecules, and
tuned to fit data
TAK1*
IRAKp
Increasing IRAKp
degradation
Time (hrs)
Time (hrs)
88
Constitutive NF-kB Signaling Module
Input signal
IkK P
IkK
P
IkB
P
NF kB
IkB
NF kB
NF kB
IkB
Degraded
IkB
NF kB
Cytoplasm
Negative
feedback
IkB gene
NF kB
IkB
NF kB
IL-6 gene
Nucleus
89
NF-kB Module Simulation
• Parameters from existing NF-kB model
(Hoffmann et al., 2002) and refined to fit
experimental data in literature
+
Add constant
input signal
IkB
IL-6
_
NF-kB
Time (hrs)
Smoothened
oscillations
Longer
delay
Time (hrs)
90
The IB–NF-B Signaling Module: Temporal Control and Selective Gene Activation
Alexander Hoffmann, Andre Levchenko, Martin L. Scott, David Baltimore
Science 298:1241 – 1245, 2002
6 hr
91
MAPK intracellular signaling cascades
92
http://www.weizmann.ac.il/Biology/open_day/book/rony_seger.pdf
Growth factor
PKC
MAPKKK
AA
MAPKK
PLA2
MAPK
MKP
93
MAPK time-course and bifurcation after a
short pulse of PDGF
Growth factor
PKC
MAPKKK
AA
MAPKK
PLA2
MAPK
MKP
Input pulse
94
IL-1 MAPK crosstalk and NFkB activation
IL-1
IL-1R
MyD88
P
IRAK
P
TAB2
IRAK gene
TAB1
TRAF6
TAK1
IRAK
IRAK
MAP3K1
P
Degraded
P
IκK
IκK
NFκB module
NFκB-dependent
transcription
95
Fold Induction
Gain-of-function screen
45
40
35
30
25
20
15
10
5
0
0.001
0 ng MAP3K1
10 ng MAP3K1
30 ng MAP3K1
0.01
0.1
1
10
[IL-1a] ng/ml
96
Model prediction
97
Future directions
• Computational modeling and data collection at higher
levels of biological organization
– Cells
• Intercellular communication
– Tissues
– Organisms
• NIH initiatives
• Environmental health risk, drugs ==> in vivo
98
Summary
• Biological organization and systems
• Molecular pathways
– identification
– Computational modeling
• Data
– Gain-of-function
– Loss-of-function
• Skin irritation example
– 3 modules
– Crosstalk
– Targeted data collection
99
Acknowledgements
• Colleagues who worked on the clonal growth risk
assessment
– Fred Miller, Julian Preston, Paul Schlosser, Julie
Kimbell, Betsy Gross, Suresh Moolgavkar, Georg
Luebeck, Derek Janszen, Mercedes Casanova, Henry
Heck, John Overton, Steve Seilkop
100
Acknowledgements
• CIIT Centers for Health Research
–
–
–
–
Rusty Thomas
Maggie Zhao
Qiang Zhang
Mel Andersen
• Purdue
– Yanan Zheng
• Wright State University
– Jim McDougal
• Funding
– DOE
– ACC
101
End
102