1.3 Dose-response relationships

1.3 Dose-response relationships : 

1.3 Dose-response relationships What is there that is not a poison? All things are poison. It is solely the dose that makes the poison Paracelsus (1493-1541)

Slide 2: 

COMMONWEALTH OF AUSTRALIA Copyright Regulations 1969 WARNING This material has been copied and communicated to you by or on behalf of the University of New South Wales pursuant to Part VA of the Copyright Act 1968 (the Act). The material in this communication may be subject to copyright under the Act. Any further copying or communication of this material by you may be the subject of copyright or performers’ protection under the Act. Do not remove this notice. 2

Advance Reading : 

Advance Reading Shibamoto, pp 1-6 Hughes, pp 29-38 NLM Toxicology Tutor "Dose and Response" http://sis.nlm.nih.gov/enviro/toxtutor/Tox1/a21.htm Dose-Response Relationships In Toxicology http://pmep.cce.cornell.edu/profiles/extoxnet/TIB/dose-response.html Reference : Deshpande ch 6 3

Keywords : 

Keywords Associative relationships Causal relationships Ceiling effect Cumulative dose-response graph Dose-response relationship ED50 Effective dose (ED) Efficacy Epidemiological studies Frequency dose-response graph Lethal dose (LD) LD50 Margin of safety No Observable Effect Level (NOEL) Normal distribution Potency Potent Subthreshold doses TD50 Threshold dose Threshold Limit Value (TLV) Toxic dose (TD) 4

Learning Objectives 1 : 

Learning Objectives 1 Understand the quantitative relationship between toxicant exposure and induced effects. Describe frequently encountered toxic effects. Interpret frequency (normal distribution) and dose - response curves. Understand threshold effects with dosage increase. Transition from no effect to toxic effect. 5

Learning Objectives, 2 : 

Learning Objectives, 2 Understand effective dose, margin-of-safety and the relationship of effective vs. toxic dose. Examine the use of actual data for no observed effect and lowest observed effect in risk assessments. Summarize effective, lethal and toxic doses. Understand a linearised multi-stage model for non-threshold responses 6

What is a dose? : 

What is a dose? Dose is the amount of a substance administered at one time. Dosage is the amount per unit weight of the exposed individual. Exposure is characterized by Number of doses Frequency of dosing The total period of time for the exposure . E.g. Arsenic at low doses over a period enhances the body’s ability to detoxify As. 7

Quantifying the dose : 

Quantifying the dose Dose: gram (g) is the standard unit but mg is typical of most exposures in toxicology. Dosage: mg (dose)/kg (body weight, bw)/day (duration) mg/kg/d Exposures are quantified in relation to the medium. mg/L in water. mg/kg in food. mg/m3 in air. Variation in units is common (ppm, ppb) Sometimes moles are the appropriate units 8

Key Concepts : 

Key Concepts Dosage Response is a mathematical relationship (positive slope). Increased dose gives increase in observable response. Response clinical or biochemical/molecular biology? Debate on whether molecular response is a manifestation of toxicosis. Causal relationship. Observable responses. Statistical management of variability of individual responses. Species, genetics, age, sex confound results 9

Responses (Toxic Effects) 1 : 

Responses (Toxic Effects) 1 Local or systemic response. Inflammation. Necrosis. Cell or tissue death. Enzyme inhibition. Biochemical pathway interruption. Cholinesterase impaired Competitive; non-competitive. Biochemical uncoupling. Heptachlor phenol Interference with phosphate molecule synthesis (ATP) Fluroacetate interrupts Krebs cycle. Alkylation of DNA 10

Responses (Toxic Effects) 2 : 

Responses (Toxic Effects) 2 Lethal synthesis. Toxicant incorporation into a biochemical pathway. Lipid peroxidation. Free radical oxidation of fatty acids leading to cell death. Lipid peroxidation attack lipid bilayers in cells. Covalent binding. Of electrophilic reactive metabolites to nucleophillic macromolecules. 11

Responses (Toxic Effects) 3 : 

Responses (Toxic Effects) 3 Receptor interaction. Modification of normal biological effects mediated by the receptor. Environmental estrogens Immune-mediated hypersensitivity reactions. Antigenic chemicals resulting in allergic reaction. Immuno-suppression. Dioxin and other chlorinated hydrocarbons Increased susceptibility to infectious agents and tumorigenesis 12

Responses (Toxic Effects) 4 : 

Responses (Toxic Effects) 4 Neoplasia. Aberrant cell division and tissue growth. Neoplasms: tumorigenesis, oncogenesis. Malignant neoplasms: carcinogenesis. A liver with multiplesecondary cancer deposits thedocrecommends.com/livercancer/ 13

Responses (Toxic Effects) 5 : 

Responses (Toxic Effects) 5 Genotoxic interaction. Chemical interaction with DNA possibly leading to heritable change. Clastogenic (chromosomal) effects. Mutagenic (base pair in DNA) effects. Developmental and reproductive toxicity. Adverse effects on conception, and structure and function of the conceptus If it results in a deformed or damaged fœtus, this response is called teratogenesis 14

Types of toxic responses: Idiosyncratic : 

Types of toxic responses: Idiosyncratic Genetically determined sensitivity or resistance to toxicity Usually lack of enzymes / factor involved in metabolism – genetic defect Primaquine (oxidative anti-malarial drug) - 10% black males / erythrocyte G-6-P dehydrogenase / hemolytic anemia caused by primaquine Glucose-6-phosphate dehydrogenase deficiency, the most common enzyme deficiency worldwide Nitrites in water - blue babies. Nitrate to nitrite in the blood May lack NADH-methaemoglobin reductase / methemoglobinaemia. 15

Types of toxic responses: Allergic : 

Types of toxic responses: Allergic Immunological mediated response (memory) Requires sensitizing exposure May involve chemical/protein complex (hapten) Atypical dose response Small doses most effective Large dose tolerance Ts cells (suppressor T lymphocytes) modulates overactive immune response. Contact dermatitis; anaphylaxis Pollens, pesticides, sulphur and SO2, penicillin http://www.allergyfacts.org.au/ 16

Types of toxic responses: Dose-response : 

Types of toxic responses: Dose-response Quantitative analysis of incremental dose increase and occurrence of toxic end effect Responses follow normal (Gaussian) frequency distribution http://www.johncon.com/john/correspondence/020217114704.27107.html 17

Gaussian response to the same dose : 

Gaussian response to the same dose Mild Response Severe Population representation of variability Number of individuals Resistant individuals Minimal effect Majority of individuals Average effect Sensitive individuals Maximal effect 18

Normal distribution parameters : 

Normal distribution parameters Mean + one SD = 68.3 % population Mean + two SD = 95.5 % population Mean + three SD = 99.7 % population Frequency (red)curveconverted to cumulativegives sigmoid (green) dose-response curve Two curves normalised to Max frequency = 1 or to total cumulative effect = 1 19

Dose-response curve : 

Dose-response curve The cumulative curve is used to show data Y-axis: Response % (lethality, toxic response, effective drug dose) X-axis: Dose (mg) Dose may be on alinear or a log scale No response belowthreshold Ceiling effect: no difference once all individuals are affected Resistant individuals Threshold Sensitive individuals Ceiling effect 100% 20

Fitting dose-response curves : 

Fitting dose-response curves Non-linear regressionhttp://www.curvefit.com/se_and_ci_.htm http://www.dtreg.com/logistic.htm 21

Observed effects: beginning of curve : 

Observed effects: beginning of curve NOEL: No observable effects level NOAEL: No observable adverse effects level SNARL: Suggested no adverse response level LOEL: Lowest observable effect level TLV: Threshold limit value 22

NOAEL and LOAEL : 

NOAEL and LOAEL Two terms often encountered are No Observed Adverse Effect Level (NOAEL) and Low Observed Adverse Effect Level (LOAEL). They are the actual data points from human clinical or experimental animal studies. http://aquaticpath.umd.edu/appliedtox/module1-dose.html 23

Slide 24: 

The NOAEL, LOAEL, NOEL, and LOEL have great importance in the conduct of risk assessments. 24

Mortality: Lethal dose : 

Mortality: Lethal dose LD50 50% of subjects die http://www.cals.ncsu.edu/course/ent425/tutorial/toxicity.html 25

Median lethal dose LD50 : 

Median lethal dose LD50 Interpretation Often used to compare toxicity between different chemicals Only measures lethality Best for quantal data (good dose response set) Best for acute exposure. Mode of exposure must be specified E.g. acute oral, acute dermal, chronic oral, chronic dermal LD50 tells nothing about slope – homogeneity or heterogeneity of response Specific ally quantify one aspect of the chemicals 26

Pesticide toxicity and LD50 : 

Pesticide toxicity and LD50 Which compound is LEAST toxic to a human applicator (based on acute oral LD50)? 27

Shape and slope : 

Shape and slope The slope of a dose-response curve gives a good indication of how a target population will respond to a toxicant.   When all members of the population react in a similar way, their dose-response curve is quite steep (Figure A).   The population is said to be homogeneous.   On the other hand, if some members of the population are much more sensitive than others (the population is heterogeneous), then the dose-response curve is flatter, like Figure B. http://www.cals.ncsu.edu/course/ent425/tutorial/toxicity.html 28

Slide 29: 

Based on what you know about the process of natural selection, which of these populations would have the greatest potential for developing resistance to this pesticide?   Explain your reasoning. 29

Other parameters : 

Other parameters ED: effective dose Pharmaceuticals EC: effective concentration Pharmaceuticals in vivo Often in blood Environmental toxicology LC: lethal concentration Environmental toxicology TDLo: Lowest published toxic dose (animal clinical studies, industrial exposure, accidental or deliberate poisoning) TCLo: Lowest published toxic concentration 30

Therapeutic Index - TI : 

Therapeutic Index - TI Ratio of dose to produce toxic effect to dose to produce desired effect TI = LD50/ED50 The larger the ratio, the greater the safety (e.g. 10) Slope of dose response is important Often used for pharmaceuticals The same theory applies to food additives 31

Therapeutic Index and Margin of Safety : 

Therapeutic Index and Margin of Safety The Therapeutic Index (TI) is used to compare the therapeutically effective dose to the toxic dose. The TI is a statement of relative safety of a drug. It is the ratio of the dose producing toxicity to the dose needed to produce the desired therapeutic response. The common method used to derive the TI is to use the 50% dose-response points. For example, if the LD50 is 200 and the ED50 is 20 mg, the TI would be 10 (200/20). A clinician would consider a drug safer if it had a TI of 10 than if it had a TI of 3. http://aquaticpath.umd.edu/appliedtox/module1-dose.html 32

Toxic doses : 

Toxic doses Toxic Doses (TDs) indicate doses that cause adverse toxic effects. The usual dose estimates are listed: http://aquaticpath.umd.edu/appliedtox/module1-dose.html 33

Effective and toxic doses : 

Effective and toxic doses The knowledge of the effective and toxic dose levels aids the toxicologist and clinician in determining the relative safety of pharmaceuticals. Two dose-response curves are presented for the same drug, one for effectiveness and the other for toxicity. In this case, a dose that is 50-75% effective does not cause toxicity whereas a 90% effective dose may result in a small amount of toxicity. http://aquaticpath.umd.edu/appliedtox/module1-dose.html 34

Test yourself : 

Test yourself What can you say about the relationship between these two compounds? http://aquaticpath.umd.edu/appliedtox/module1-dose.html 35

Margin of safety (MOS) : 

Margin of safety (MOS) The use of the ED50 and LD50 doses to derive the TI may be misleading as to safety, depending on the slope of the dose-response curves for therapeutic and lethal effects. To overcome this deficiency, toxicologists often use another term to denote the safety of a drug - the Margin of Safety (MOS). MOS applies to something that has a beneficial effect at lower concentrations and a toxic effect at higher concentrations Note, we are talking about lethal effect, not just some observed toxicity. http://aquaticpath.umd.edu/appliedtox/module1-dose.html 36

Margin of safety : 

Margin of safety The MOS is usually calculated as the ratio of the dose that is just within the lethal range (LD01) to the dose that is 99% effective (ED99). The MOS = LD01/ED99. A physician must use caution in prescribing a drug in which the MOS is less than 1. 37

Margin of safety : 

Margin of safety Accounts more for slope differences than just observing the LD50 MOS = LD1/ED99 Neither TI or MOS works for chemicals with no beneficial effect or for repeated dose trials Some therapies are very toxic (chemotherapy) - justified if the disease is lethal MOS for food additives is much higher than for drugs 38

Slide 39: 

Risk of deficiency and toxic effects of essential trace elements (ETE) or vitamins www.inchem.org/documents/ehc/ehc/ehc228.htm 39

Carcinogen risk assessment : 

Carcinogen risk assessment Linearized Multistage Model Assumes non-threshold effect. Linear extrapolation through zero threshold dose from upper confidence level of lowest dose that caused cancer in animal study. Analysis results in a cancer slope factor that can be used to predict cancer risk at a specific dose http://www.epa.gov/waterscience/standards/academy/supp/health/page18.htm 40

Linearised multistage model : 

Linearised multistage model Dose-response curve based on experimental results. Dose is on a log scale Lowest dose that caused cancer Linear extrapolation through zero dose to upper confidence level on the lowest dose that caused cancer Determine slope factor of the linear extrapolation http://www.epa.gov/waterscience/standards/academy/supp/health/page18.htm 41

Other Models for Risk Assessment : 

Other Models for Risk Assessment One hit model (cancer) Assumes a molecular event with cellular response. Multi hit model (cancer) Assumes multiple events prior to cellular activation. Probit model Linearization transformation that assumes log normal distribution. PB PK - Physiologically based pharmacokinetic model Uses intensive pharmacokinetic and mechanistic data. Most used for lead exposure, especially for children 42

Transformation of Variables : 

Transformation of Variables Allows better (simpler) analysis of data at points of interest such as LD50. Analysis of sigmoid curves where they are changing rapidly is very inaccurate. Transformation into an approximate normally distributed variable. Examples (rj = dead animals; nj = total animals) Probit transformation. Based on Gaussian (Bell) curve. Probit (rj/nj) = -1 (rj/nj) Useful in acute lethality tests. Logit transformation. Log odds of a quantal response. Logit (rj/nj) = ln [(rj/nj)/1 - (rj/nj)] Weibull transformation. Exponential model used in modeling multistage processes. 43

Slide 44: 

PROBIT TRANSFORMATION NOT GIVEN IN 2009 SLIDES 44 TO 48 OMITTED 44

Probit Transformation 1 : 

Probit Transformation 1 Probability units  probits Convert % response to units of deviation from the mean or normal equivalent deviations (NEDs). Hence the NED for a 50% response is 0. Probit approach adds 5 to avoid negatives. 45

Probit transformation 2 : 

Probit transformation 2 46

Probit Transformation 3 : 

Probit Transformation 3 Perform log10 transformation of the dose. Assumes log normal distribution. Produces an approximately linear relationship. Allows linear regression analysis. Maths for probit analysis - if you really want it – can be found at http://www.ats.ucla.edu/stat/sas/dae/probit.htm See Deshpande p133 for good summary 47

Probit Transformation 4 : 

Probit Transformation 4 Log normal distribution Dose (mg/kg) Mortality frequency (%) 48

Probit unit transformation : 

Probit unit transformation Graphing probit response against log dose gives a straight line. 49 49

Summary: transformations of dose-response curve : 

Summary: transformations of dose-response curve Normal frequency distribution Arithmetic dose to log dose Frequency data changed to cumulative Probability of response to normal equivalent deviations NED NED = Standard deviations from mean NED to probit Probit =NED + 5 50

Dose-response curve summary : 

Dose-response curve summary Major Parameters Median Lethal Dose - LD50 Other LDs, TDs or EDs Slope Thresholds System saturation of dose to response Comparative toxicity Risk assessment and managing risk/benefit analysis. 51

Slide 52: 

Acknowledgement Greg Möller. University Idaho. 52