woirkshop on alternative methods - ankara, 12-13 november 2007 1 agnieszka kinsner and pilar prieto...

Woirkshop on Alternative Methods - Ankara, 12-13 November 2007 1

Agnieszka Kinsner and Pilar Prieto

Key area – Systemic toxicity


• Rationale:Rationale: Knowledge of the effects of acute and chronic exposure to unknown compounds is necessary for hazard and risk assessment

• The aim of this key area:The aim of this key area: to identify and validate in vitro tests relevant for target organ (e.g. lung, liver, kidney) and target system-specific toxicities (e.g. the haemopoietic system and the immune system) to be incorporated into testing strategies for the estimation of human systemic toxicity

• The final goal: The final goal: to provide cheaper, more ethical and more scientifically-based testing strategies.

Key area - Systemic toxicity


ECVAM’s activities in the area of systemic toxicity cover:

• Acute systemic toxicity

• Haematotoxicology

• Immunotoxicology

• Chronic systemic toxicity

• Pulmonary toxicity

• Biological barriers (intestinal and blood-brain barrier)



Acute oral toxicity - OECD test guidelines

• Deletion of TG 401 in 2000• Adoption of TG420, TG423 and TG425 by OECD represents a

significant reduction in the number of animals used per test: from ~45 to ~8 animals per substance

• Recommendations: to take into account the in vitro methods for the determination of the starting dose

OECD TG420: Acute Oral Toxicity - Fixed Dose Procedure

OECD TG423: Acute Oral toxicity - Acute Toxic Class Method

OECD TG425: Acute Oral Toxicity - Up-and-Down Procedure


Strategy to Replace Acute Toxicity Testing

Registry of Cytotoxicity

MEIC

-4

-3

-2

-1

0

1

2

3

-6 -4 -2 0 2 4log IC50 [mmol/l]

log

LD

50

[mm

ol/k

g]

In vitro cytotoxicity test:

Relatively good correlation (~70%)

Certain number of misclassifications

Further needs:

Improve the in vitro - in vivo correlation by evaluating existing outliers in order to introduce further parameters (ADE, metabolism, organ specificity).

ICCVAM/ECVAM validation study


Title: Optimisation and prevalidation of an in vitro test strategy for

predicting human acute toxicity

Integrated Project of the Sixth Framework Programme of the European

Commission

35 Partners from 13 European states: Universities, SME, Research

Institutes, Industries, Foundations, JRC

Start: January 2005; End: December 2009


Analysis andAnalysis andintegration of integration of

in vitro/in vivoin vitro/in vivo datadata

Iterative amendment of the testing strategy

Generation of an in vivo database and establishment of a depository of reference

compounds

Generation of an in vitrodatabase

WP 6WP 6 Role of metabolism

RoleRole ofof ttarget organ toxicity

• WP 7.1 – neurotoxicity

• WP 7.2 – nephrotoxicity

• WP 7.3 – hepatotoxicity

New cell systems, newendpoints

Technical optimisation of the amended test strategy

Prevalidationof the testing strategy

WP 5WP 5 Role of ADE

Alerts and correctors Alerts and correctors in toxicity screening in toxicity screening

WP 1WP 1 WP 2WP 2

WP 4WP 4

WP 3WP 3

WP 5WP 5

WP 6WP 6

WP 7WP 7

WP 8WP 8

WP 9WP 9





compounds












WP 1WP 1 WP 2WP 2

WP 4WP 4

WP 3WP 3

WP 5WP 5

WP 6WP 6

WP 7WP 7

WP 8WP 8

WP 9WP 9


97 reference chemicals

human data

kinetics (in vitro, in silico)

animal in vivo data

in vitro data


WP1: Selection of reference chemicals and collection WP1: Selection of reference chemicals and collection

of of in vivoin vivo data data

• 97 reference chemicals were selected within a wide range of acute toxicity and generic uses

• Generation of the in vivo database (animal and human)


Distribution of SD for log-transformed LD50 (rat oral studies: 62 chemicals)

log-transformed SD <0.5 for majority of chemicals,i.e., excluding 5 chemicals exhibiting extreme variability

Interpolation: overall median log-transformed SD ~0.2 (0.17 for rat, 0.21 for mouse)

Range comparable to intervals (log-scale)defining EU/GHS toxicity classifications/categories

WP1: Evaluation of WP1: Evaluation of in vivoin vivo animal data – animal data – variabilityvariability

Inter-species comparison: rat vs mouse mean LD50 (n=40): highly correlated (near overlap with line of identity)

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4

0

5

10

15

SD of log-transformed LD50 (rat, oral)

nu

mb

er o

f ch

emic

als

20


The database contains human acute toxicity data from a

single poisoning, consisting of:• sub-lethal blood concentrations • lethal blood concentrations• post-mortem blood concentrations

WP1: Evaluation of WP1: Evaluation of in vivoin vivo human data – human data – calc. of LC50 valuescalc. of LC50 values


Example: Acetaminophen approximate LC0 and LC100 and LC50

LC50 = (3.35+ 3.40)/2= 3.37 in microM Converted to M LC50=-2.63

LC100 = 3.40LC0 = 3.35

WP1:WP1: Estimation of LC50 humanEstimation of LC50 human

1,80

2,00

2,20

2,40

2,60

2,80

3,00

3,20

3,40

3,60

3,80

0 10 20 30 40 50 60 70 80 90 100

log

(Let

hal b

lood

con

cent

ratio

n m

ikro

M)

time (h)

Investigation: 01-acetaminophen-lethalRaw Data Plot

Lethal blood concentratio~

1

2

3

4

5

6 7

8

9

1011

12

13

14

15

16

1718

19

20

21

22

23

24

2526

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

4445

4647

48

49

50

51

52

53

54

55

56

57

58

59

60

61

62 63

64

65

66

67

68

69

7071

7273

74

75

.






compounds












WP 1WP 1 WP 2WP 2

WP 4WP 4

WP 3WP 3

WP 5WP 5

WP 6WP 6

WP 7WP 7

WP 8WP 8

WP 9WP 9





compounds












WP 1WP 1 WP 2WP 2

WP 4WP 4

WP 3WP 3

WP 5WP 5

WP 6WP 6

WP 7WP 7

WP 8WP 8

WP 9WP 9


• Assessment of basal cytotoxicity in:– BALB/3T3 (NRU)

– NHK (NRU)

– HL-60 (ATP)

– HepG2 (NRU, total protein)

– Fa32 (NRU, total protein)

• Generation of an in vitro database for 97 selected reference chemicals

WP2: Generation of WP2: Generation of in vitroin vitro basal basal cytotoxicity datacytotoxicity data






compounds












WP 1WP 1 WP 2WP 2

WP 4WP 4

WP 3WP 3

WP 5WP 5

WP 6WP 6

WP 7WP 7

WP 8WP 8

WP 9WP 9





compounds












WP 1WP 1 WP 2WP 2

WP 4WP 4

WP 3WP 3

WP 5WP 5

WP 6WP 6

WP 7WP 7

WP 8WP 8

WP 9WP 9


• 6 basal cytotoxicity tests: similar information i.e. similar

ranking

• The validated 3T3/NRU seems to be the best candidate

WP3: Evaluation of in vitro cytotoxicity dataWP3: Evaluation of in vitro cytotoxicity data


-5

-4

-3

-2

-1

-4 -3 -2 -1

YVar

(log(

mea

n LD

50m

ol/k

g)ra

t)

YPred[1](log(mean LD50mol/kg)rat)

Richards12var28june07clean-MSj+phys+in vivo-del.M6 (PLS)YPred[Last comp.](log(mean LD50mol/kg)rat)/YVar(log(mean LD50mol/kg)rat)

RMSEE = 0,784482

acetaminop

acetylsali

atropine scaffeine

carbamazep

cyclohexim

diazepam

digoxin

isopropyl

malathion

mercury (I

pentachlor

phenobarbi

sodium lau sodium val

5-fluorour

benzene

tert-butyl

acrylaldehcadmium (I

phenanthre

pyrene

1,2,3,4-tepentachlor

hexachloro

amiodarone

(±)-verapa

rifampicin

tetracycli

orphenadri

nicotinelindane

D-amphetammethadone

ethanol

parathion

dichlorvos

physostigm

glufosinat

cis-diammi

diethylene

diquat dib

ochratoxin

cyclospori

17a-ethyny sodium flu

paraquat d

glycerol

dimethylfo

amitriptyl

ethylene g

methanol

phenol

sodium chlxylene

potassium

lithium su

theophyllihaloperidopropranolo

arsenic tr

thioridazi

thallium sw arfarin

chloroformisoniazid

dichlorome

hexachloro

chloroquin

5,5-diphen

chloramphe

potassium

chloral hy

2,4-dichlo

meprobamat

strychnine

glutethimimaprotilin

disopyrami

chlormethi

quinidine

procainami

codeine

chlorproma

sodium sel

acetonitrisodium bicformaldehy

(-)-epinep

y=1*x-0,01449R2=0,457

SIMCA-P+ 11 - 2007-06-29 08:53:49

Plot observed rat vs predicted LD50From in vitro 3T3/NRU, PLS regression analysis

Log

LD

50 (

mol

/kg

b.w

.) r

at,

exp

.

Log LD50 (mol/kg b.w.), predicted with 3T3/NRU


-8

-7

-6

-5

-4

-3

-2

-1

-6 -5 -4 -3 -2 -1

YVar

(hum

an lo

gLC5

0cor

r+-,2

(log

M))

YPred[1](human logLC50corr+-,2 (log M))

Djurönäsdata.M2 (PLS)YPred[Last comp.](human logLC50corr+-,2 (log M))/YVar(human logLC50corr+-,2 (log M))

RMSEE = 0,900934

acetaminophenacetylsalicylic

atropine sulfat

caffeine

carbamazepine

colchicine

diazepam

digoxin

isopropyl alcoh

malathion

mercury (II) ch

pentachlorophen

phenobarbital

sodium valproat

5-fluorouracil

cadmium (II) ch

amiodarone hydr(±)-verapamil h

rifampicine

orphenadrine hy

nicotine

lindanemethadone hydro

ethanol

glufosinate-amm

cis-diamminepla

diquat dibromid

cyclosporine A

17a-ethynylestr

sodium fluoride

paraquat dichlo

dimethylformami

iron II sulfate

amitriptyline h

ethylene glycolmethanol

phenol

sodium chloride

potassium cyani

lithium sulfate

theophylline

propranolol hydarsenic trioxidthioridazine hythallium sulpha

warfarin

isoniazid

chloroquine dip

5,5-diphenylhydchloramphenicol

potassium chlor

chloral hydrate

2,4-dichlorophe

meprobamate

sodium pentobar

strychnine

glutethimide

maprotiline

disopyramide

diphenhydramine

chlormethiazole

quinidine sulfa

procainamide hy

codeine

chlorpromazine

paraldehyde

sodium selenate

acetonitrile

y=1*x-0,02498R2=0,7104 sodium bicarbon

SIMCA-P+ 11 - 2007-06-30 10:49:21

Chemicals with poor human data

Plot observed LC50 humans vs predictedin vitro variables

Log

hu

man

LC

50

Log predicted LC50






compounds












WP 1WP 1 WP 2WP 2

WP 4WP 4

WP 3WP 3

WP 5WP 5

WP 6WP 6

WP 7WP 7

WP 8WP 8

WP 9WP 9





compounds












WP 1WP 1 WP 2WP 2

WP 4WP 4

WP 3WP 3

WP 5WP 5

WP 6WP 6

WP 7WP 7

WP 8WP 8

WP 9WP 9


WP4: Cytokine secretionWP4: Cytokine secretion

plasma

PBMC

Ficoll

Granulocytes & macrophages

PBMC


WP4: HaematopoiesisWP4: Haematopoiesis






compounds












WP 1WP 1 WP 2WP 2

WP 4WP 4

WP 3WP 3

WP 5WP 5

WP 6WP 6

WP 7WP 7

WP 8WP 8

WP 9WP 9





compounds












WP 1WP 1 WP 2WP 2

WP 4WP 4

WP 3WP 3

WP 5WP 5

WP 6WP 6

WP 7WP 7

WP 8WP 8

WP 9WP 9


• Measurement of the transport across the intestinal

barrier and the blood-brain barrier using in vitro

models and neuronal networks (n=21)

• Measurement of protein binding, microsomal

stability, lipophilicity (n=42)

• Measurement (n=3) and modelling of free

concentration of compounds in the in vitro systems.

• Generic biokinetic model for the interpretation of in

vitro toxic concentrations in relation to the in vivo

acute toxic dose – under developmenta

bs

orp

tio

n

sp

ecia

l b

arr

iers

Concentration at target

protein binding

eli

min

ati

on

: ex

cre

tio

n

an

d m

eta

bo

lism

Free

concentration

ab

so

rpti

on

sp

ecia

l b

arr

iers

Concentration at target

protein binding

eli

min

ati

on

: ex

cre

tio

n

an

d m

eta

bo

lism

Free

concentration

WP5: Role of ADE (in vitro/inWP5: Role of ADE (in vitro/in silico)silico)


72% overall accuracy72% overall accuracy

H = High ; HIA > 80 %

M = Moderate ; HIA < 20-70 %

P = Poor ; HIA < 20 %

Caco-2 Caco-2computer

Papp 10-6cm/s < 1= Poor (P)

Papp 10-6cm/s< 1 - 10 = Moderate (M)

Papp 10-6cm/s > 10 = High (H)

WP5: Oral absorptionWP5: Oral absorption


73% overall accuracy

Log BB > -0.7 Poor (P)

-0.7 < Log BB < -0.3 Moderate (M)

Log BB > -0.3 High (H)

WP5: Blood-brain barrierWP5: Blood-brain barrier

Luminal compartment(Blood)

Abluminal compartment (Brain)Coated microporous membrane


WP6: RoleWP6: Role of metabolismof metabolism

IC50hepat/IC50 HepG2

Compound Cas no P15 P23 P31 Bayer Mean Comparison hepatocyes vs HepG2Reported bioactivation

Atropine sulfate 5908-99-6 >1E-03 0,06 0,53 0,01 0,20 More toxic to hepatocytesthan to HepG2 YESMercury II 7487-94-7 0,04 0,18 0,75 0,18 0,29 More toxic to hepatocytesthan to HepG2 NOPentachlorophen 87-86-5 0,97 0,04 0,84 1,28 0,78 Slightly more toxic to hepatocytes than to HepG2 YESRifampicine 13292-46-1 0,85 0,56 1,18 0,67 0,82 Slightly more toxic to hepatocytes than to HepG2 NOTetracycline HCl 64-75-5 >1E-03 0,31 0,06 1,13 0,50 Slightly more toxic to hepatocytes than to HepG2 NOOrphenadrine HCl 341-95-5 1,34 1,55 0,25 0,56 0,93 Similar toxicity to hepatocytes than to HepG2 NODiazepam 439-14-5 1,25 1,50 1,24 0,85 1,21 Similar toxicity to hepatocytes than to HepG2 NOMalathion 121-75-5 1,46 1,46 >=1E-03 >1E-03 1,46 Similar toxicity to hepatocytes than to HepG2 YESAmiodarone HCl 1951-25-3 1,35 1,02 1,10 1,54 1,25 Similar toxicity to hepatocytes than to HepG2 NOSLS 151-2-3 1,63 0,42 1,69 1,45 1,30 Similar toxicity to hepatocytes than to HepG2 NODigoxin 20830-75-5 908,72 ?? >=1000 >=1000 Less toxic to hepatocytes than to HepG2 NO(±)-Verapamil HCl 152-11-4 8,85 2,75 0,31 1,70 3,40 Less toxic to hepatocytes than to HepG2 NO

HepatomaHepatocyte

IC50A IC50B

IC50A <IC50A < IC50B

BioactivatedBioactivated

0

100

AA BB

Cyt

oto

xici

ty (

%)

Concentration

0

100

AA BB

0

100

AA BB

Concentration

IC50ANot BioactivatedNot BioactivatedIC50A = IC50B

Concentration

0

100

AA BB

0

100

AA BB

Cyt

oto

xici

ty (

%)


WP7.1: WP7.1: NeurotoxicitNeurotoxicityy

• Basal cytotoxicity• General cell physiology (energy status,

glycolytic activity, Ca2+ homeostasis, cell and mitochondrial membrane potential, oxidative stress (ROS)

• Neurochemistry– Voltage operated ion channels – Receptor function – Neurotransmitter synthesis/degradation – Neurotransmitter uptake – Neurotransmitter release – Global electrical activity

Human neuroblastoma SH-SY5Y cell line

Serum-free aggregating rat brain cell cultures

Primary cultures of rat cerebellar granule cells


ions

BARRIER FUNCTION

• Grown on permeable supports• Current across epithelium• Rate of flux of ions• Electrical resistance

REMS MACHINE

↓TER ----------- ↑Permeability

Loss of barrier function

WP7.2: NephrotoxicityWP7.2: Nephrotoxicity Cells: LLC-PK1

•TER: sensitive indicator of nephrotoxicity

•TER: greater sensitivity for nephrotoxic chemicals

•Compounds requiring metabolism (diethylene glycol) did not show toxicity at concentrations used


Concentration

Eff

ect

(%

)

0

100 AA BB CC

IC50(A) < IC50 (B) ≈ IC50(C): “hepatotoxic” (bioactivable) → alertalert

IC50(A) ≈ IC50 (B) < IC50(C): “hepatotoxic” → alertalert

IC50(A) ≈ IC50 (B) ≈ IC50(C): no hepatotoxic → no alertno alert

WP7.3 HepatotoxicityWP7.3 Hepatotoxicity

IC50CIC50BIC50A

Non-hepatic cellHepatomaHepatocyte

IC50CIC50BIC50A

Non-hepatic cellHepatomaHepatocyte






compounds












WP 1WP 1 WP 2WP 2

WP 4WP 4

WP 3WP 3

WP 5WP 5

WP 6WP 6

WP 7WP 7

WP 8WP 8

WP 9WP 9





compounds












WP 1WP 1 WP 2WP 2

WP 4WP 4

WP 3WP 3

WP 5WP 5

WP 6WP 6

WP 7WP 7

WP 8WP 8

WP 9WP 9Last year of the projectLast year of the projectLast year of the projectLast year of the project


The Colony Forming Unit-Granulocyte/Macrophage (CFU-Colony Forming Unit-Granulocyte/Macrophage (CFU-GM) AssayGM) Assay for predicting acute neutropenia in humans has been scientifically validated

A validation study on CFU-GM assay withCFU-GM assay with rat progenitorrat progenitor cells is ongoing, aiming to increase the accuracy of determining the maximum permissible exposure limit for xenobiotics.

Pessina A. et al. Application of the CFU-GM assay to predict acute drug-induced neutropenia: an international blind

trial to validate a prediction model for the maximum tolerated dose (MTD) of myelosuppressive xenobiotics. Toxicol

Sci. 2003 75(2):355-67.

Haematotoxicology


A multi-laboratory study has been carried out to evaluate the most promising endpoints for immune-suppression and immune-suppression and immunotoxicityimmunotoxicity

As a follow-up, a validation study is currently ongoing, with the aim of evaluating reproducibility and predictivity of a set of in vitro assays to detect immunotoxicity, to be used as animals replacement

Carfi' M et al. In vitro tests to evaluate immunotoxicity: a preliminary study.Toxicology. 2007 229(1-2):11-22.

Immunotoxicology


In vitro models

Model 1: Primary culture bovine brain capillary endothelial cells co-cultured with rat astrocytes

Model 2: Porcine primary endothelial cells cultured in hydrocortisone-conditioned medium

Model 3: 4-days BBB culture of bovine brain capillary endothelial cells in conditioned medium

Model 4: Cell line HCMEC/D3

Model 5: MDCKmdr-1 cells

Study on evaluation of the performance of five selected in vitro blood-brain barrier evaluation of the performance of five selected in vitro blood-brain barrier modelsmodels for predicting absorption/uptake into the brain

Transport experiments (Papp values) in 5 labs using 16 coded reference chemicals; 3 independent experiments

Barrier models: blood-brain barrier


Barrier models – intestinal barrier

Prevalidation of in vitro models for the prediction of gastrointestinal absorption

High Absorption

Intermediate absorption

Low absorption

Isopropanol (reference)

Cimetidine (reference)

Atenolol (reference)

Ethylen glycol (chemical)

Paraquat (pesticide)

Cupric sulphate (chemical)

Sodium Valproate (drug)

Paracetamol (drug)

Colchicine (drug)

Acetylsalicylic acid (drug)

- Two testing laboratories

- Two in vitro models: Caco-2, TC7 clone

- Permeability assays (Paap): at non toxic concentration during 60 min, samples taken at 15, 30, 45 and 60 min

10 Reference Chemicals


Marie Curie Research Training Network that aims to elucidate basic molecular, cellular and tissue-related mechanisms involved in the generation of various acute and chronic pulmonary diseasesacute and chronic pulmonary diseases.

Two in vitro models are under investigation at ECVAM: human airway epithelial cell line Calu-human airway epithelial cell line Calu-33 and the human alveolar type cell line NCI human alveolar type cell line NCI H441H441

Pulmonary toxicity


• Haematotoxicology [email protected]

• Immunotoxicology

• Chronic systemic toxicity [email protected]

• Pulmonary toxicity

• Biological barriers

• Acute systemic toxicity [email protected]

[email protected]



In these 2 classes 97% of all new

industrial chemicals

Toxicity prevalence of new industrial Chemicals in the EU

Chemicals (%) LD50 (mg/Kg)

0 < 25

3 > 25 – 200

21 > 200 – 2000

76 > 2000

Evaluation of the 3T3 NRU assay for the prediction of non-toxic compounds

woirkshop on alternative methods - ankara, 12-13 november 2007 1 agnieszka kinsner and pilar prieto...

Documents