Dr Cliff Elcombe Biomedical Research Centre Ninewells Hospital

1
Dr Cliff ElcombeBiomedical Research
CentreNinewells Hospital Medical
SchoolUniversity of Dundee

• Risk Perception, Risk Assessment and the Role of
  Mechanisms

2
(No Transcript)
3
(No Transcript)
4
(No Transcript)
5
(No Transcript)
6
(No Transcript)
7
Hazard and Risk

• Hazard
• Intrinsic property of the chemical
• Risk
• Hazard x exposure (dose and time)

8
Hazard and Risk

• Hazard is the potential for harm
• Risk is the chance (probability) that harm will
  actually occur

9
Traumatic (accidental) Injuries and Deaths USA
1985

• 60 million people were injured
• 9 million people sustained disabling injuries
• 92,500 people were killed
• 45,600 deaths were highway-related
• 11,600 deaths were work-related
• 20,500 deaths occured at home
• 19,000 deaths were other public accidents
• resulting in 543 million restricted activity
  days
• Population of USA was 230 million

10
The Average Americans Chances of Dying this Year
of
Heart disease
1 in 300
Cancer
1 in 700
Motor vehicle accident
1 in 4,200
Suicide
1 in 9,300
Homocide
1 in 12,000
Fire
1 in 31,000
Electrocution
1 in 230,000
Tornado
1 in 3.3 million
Living near nuclear plant
1 in 5 million
2 quarts of "TCE water" (5 ppb)
1 in 70 million
11
Concentration Analogies
One Part Per Million is

• one automobile in bumper-to-bumper traffic from
  Cleveland to San Francisco
• one pancake in a stack four miles high
• 1 inch in 16 miles
• one minute in two years
• one ounce in 32 tons
• one cent in 10,000
• 0.0001 (or 10,000 ppm equals 1)

12
Concentration Analogies
One Part Per Billion is

• one 4-inch hamburger in a chain of hamburgers
  circling the earth at the equator two-and-a-half
  times (4x10e9 inches)
• one kernel of corn in a 45-foot high, 16-foot
  diameter silo
• one sheet in a roll of toilet paper stretching
  from New York to London
• one second of time in 32 years

13
Concentration Analogies
One Part Per Trillion is

• one square foot of floor tile on a kitchen floor
  the size of Indiana
• one drop of detergent in enough dishwater to fill
  a string of railroad tank cars ten miles long
• one square inch in 250 square miles
• one mile on a 2-month journey at the speed of
  light

14
Concentration Analogies
One Part Per Quadrillion is

• one postage stamp on a letter the size of
  California and Oregon
• one human hair out of all the hair on all the
  heads of all the people in the world
• one mile on a journey of 170 light years

15
Perceived Risk Ratings
16
A Selection of Natural Carcinogens

• anise
• apples
• bananas
• basil
• brocolli
• brussel sprouts
• cabbage
• carrots
• cauliflower
• celery
• cinnamon
• cloves
• cocoa
• comfrey tea
• fennel
• grapefruit juice

• honey dew melon
• horseradish
• kale
• mushrooms
• mustard
• nutmeg
• orange juice
• parsely
• parsnips
• peaches
• black pepper
• pineapples
• radishes
• raspberries
• tarragon
• turnips

17
Carcinogenic Risk of "Alar- Contaminated" Apple
Juice
15 pints of apple juice per day has a
carcinogenic risk equivalent to
1 mushroom (15g) per day
1/3 of a peanut butter sandwich
100g celery
100g cabbage
1/100 pint (6.5ml) of beer
1/2 teaspoon (2.5ml) of wine
9.6 litres

15 minutes in a swimming pool
18
The Stages of Risk Assessment
Hazard Identification
Risk Estimation
Risk Evaluation
Risk Management
19
Hazard Identification

• Human Studies
• Short Term Tests and Structure Activity
• Animal Studies

20
Hazard Identification

• Human Studies
• epidemiology
• retrospective
• low statistical power
• uncertain exposure estimates
• volunteer studies
• prospective
• usually not ethical

21
Hazard Identification
Short Term Tests and Theoretical
Approaches
surrogate measurements
qualitative
22
Hazard Identification
Animal Studies
high to low dose extrapolation
route of exposure
23
Animal Studies in Toxicological Evaluation

• Acute Toxicity ("LD50")
• rats/mice
• Irritancy/corrosivity/sensitization
• guinea pigs/rabbits
• Mutagenicity, clastogenicity
• in vivo and in vitro
• Subacute/subchronic toxicity
• rats/mice/dogs

• Carcinogenicity
• rats and mice
• Reproductive toxicology
• rats/mice/rabbits
• teratology
• foetal toxicity
• multigeneration

24
Risk Estimation
Safety Factor Approach
Mathematical Models
25
Risk Estimation - Safety Factor Approach

• ADI NOEL / SF
• ADI acceptable (allowable) daily intake
• NOEL no observable effect level
• SF safety factor

26
Hypothetical Dose-response
35
30
25
20
Tumour Incidence ()
15
10
5
0
0
200
400
600
800
1,000
1,200
Dose (mg/kg)
27
Application of a 100-fold safety factor to
convert a NOAEL in animals into an ADI for humans
10 fold
10 fold
Human variability
Species differences
28
Risk Estimation- Mathematical Models

• Virtually Safe Dose
• Linear Extrapolation
• One-hit Models
• Multi-hit Models
• Multistage Models
• Weibull Model
• Physiologically-based Pharmacokinetic Models

29
Linearised Model
1,000,000
100,000
10,000
1,000
Risk per 1,000,000
100
10
1
1
100
1,000
10,000
100,000
10
Dose (ppm in diet)
VSD ("virtually safe dose")
30
Metabolic Saturation!
31
Interspecies Comparisons

• Scaling Factors
• Mechanisms of Toxicity

32
Interspecies Comparisons

• Scaling Factors need to account for
• absorbtion, distribution, metabolism and
  excretion
• size
• lifespan
• Fraction of Diet Scaling
• Body Weight Scaling
• Surface Area Scaling
• PB - PK

33
The Exposure-Dose-Response Paradigm for
Carcinogens and Toxicants
oral inhalation dermal
Exposure
Blood concentration
Tissue dose of toxic moiety
metabolic activation/deactivation accumulation/exc
retion
Toxic moiety-target interaction
eg. transcriptional activation, cofactor
depletion, mutation, enzyme inhibition, etc.
acute, eg. cell death subacute, eg. organ
growth chronic, eg. cancer
Toxic response
34
Biologically-based Model Linking Mechanisms of
the Exposure-dose-response Continuum
Exposure
PB-PK
Mechanisms
Models
Tissue Dose
Toxicant-target
Interaction
Mechanisms
Models
Toxicant-tissue Interaction
Tissue-response
Mechanisms
Models
Toxic Response
35
Physiologically-based Pharmacokinetic Model for
Volatile Organic Chemicals
36
The Use Of Physiologically-based Pharmacokinetic
Modelling
Extrapolation beyond the experimental range
Extrapolation between different routes of exposure
Extrapolation between different time frames
Prediction of
data
Use of above in risk assessment
37
Biologically-based Cancer Response Model
Mutation
Mutation
Initiated
Neoplastic
Normal



Cell
Cell
Cell
D0
D1
38
Physiologically-based Pharmacokinetic Models

• Based on the physiology of the organism
• Instead of compartments defined by experimental
  data, these models use actual organs and tissue
  groups with weights and blood flows taken from
  the literature
• Instead of composite rate constants obtained by
  fitting the data, actual physicochemical and
  biochemical constants are used
• These models can predict the behaviour of the
  experimental time course without being based on
  it
• ADVANTAGE Quantitative extrapolation beyond the
  experimental range

39
Risk Evaluation and Management

• What level of risk is acceptable?
• how safe is safe enough?
• risk-benefit analysis
• perception of relative risk
• technical considerations
• socioeconomics
• politics

40
Summary
41
Mechanisms of Hazard and Risk Assessment
42
Classification of Carcinogens
Carcinogen
Non-genotoxic
Genotoxic
Cytotoxic
Non-cytotoxic
Endocrine
Direct mitogen
disruptor
43
Liver Growth Carcinogens

• Hyperplasia
• Stimulate cell proliferation (acute and/or
  chronic)
• Inhibit apoptosis
• Hypertrophy
• Organelle proliferation
• SER
• Peroxiosmes

44
Characteristics of the Peroxisome Proliferation
Phenomenon in Rats

• Hepatomegaly
• Proliferation of peroxisomes and smooth
  endoplasmic reticulum
• Induction of peroxisomal fatty acid oxidising
  enzymes
• Induction of CYP 4A1
• Stimulation of replicative DNA synthesis
• Inhibition of apoptosis
• Hepatocellular tumours in long term studies

45
What is the Mechanism of Hepatocarcinogenesis?

• Receptor-mediated
• Reactive oxygen/oxidative stress
• Stimulation of S-phase
• Inhibition of apoptosis
• ??????????
• Biological plausibility?

46
Chemicals Eliciting Peroxisome Proliferation in
Rats and/or Mice
47
Model of Peroxisome Proliferator Action
Peroxisome Proliferation, Growth Regulation and
Hepatocarcinogenesis
48
PPARa Knockout Mouse

• No hypolipidaemia
• No hepatomegaly
• No peroxisome proliferation
• No peroxiosomal enzyme induction
• No CYP4A induction
• No stimulation of DNA synthesis
• No inhibition of apoptosis
• No hepatocellular tumours

49
(No Transcript)
50
Diethylhexyladipate Toxicology

• Very low acute toxicity
• Not irritant/corrosive/sensitizer
• Non-mutagenic
• No reproductive effects
• Hepatocarcinogen in mice not rats
• Potential Human Carcinogen?

51
In Vivo/In Vitro Extrapolation
Animal in vitro
Human in vitro
Animal in vivo
Human in vivo
?
52
Metabolism of Diethylhexyladipate (DEHA)
HOOC(CH2)4COOH

HOOCCHCH2CH2CHCH3
HOOCCHCH2CH2CH2CH3
CH2CH3
CH2CH3
OH
HO0CCHCH2CH2CCH3
O
CH2CH3
53
Peroxisome Proliferation in Mouse Hepatocyte
Cultures - DEHA and Metabolites
3,000
EHA
2,500
2,000
EH
MEHA
1,500
Peroxisome Proliferation ( control)
1,000
OH-EHA
Keto-EHA
500
EHdiA
AA
0
0
1,200
200
400
600
800
1,000
Concentration (microMolar)
54
Species Differences in Response - EHA
3,000
Mouse
2,500
2,000
1,500
Peroxisome Proliferation ( control)
1,000
Rat
Human
500
Guinea pig
Marmoset
0
0
200
400
600
800
1,000
1,200
Concentration (microMolar)
55
Dose-response to Peroxisome Proliferators
Liver growth
Tumours
Response
1
3
2
Dose
Threshold for early events
Tumour threshold
56
Summary

• Hazard and risk are not the same
• Perception of risk does not necessarily relate to
  actual risk
• Risk assessment should not be a purely
  mathematical exercise
• Mechanisms of toxicity should play a major role
  in risk assessment

57
Good Web Sites for Risk Assessment
http//www.sis.nlm.nih.gov/ToxTutor.html
http//ruby.fgcu.edu/Courses/Twimberley/IDS3920/ma
in.html
Epidemiology for Journalists
http//www.facsnet.org/tools/ref_tutor/epidem/inde
x.php3 http//www.facsnet.org/tools/ref_tutor/risk
/index.php3
General Site for Risk Assessment Information
http//riskcenter.doe.gov/whatisrisk/riskassessmen
t.cfm