srlf dabor - paris descartes€¦ · page 1 what can we learn from analysis of the toxicokinetic /...

What can we learn from analysis of the

toxicokinetic / toxicodynamic relationships

in acute poisonings?

Bruno Mégarbane, MD, PhD

Réanimation Médicale et Toxicologique,

INSERM U705 - CNRS, UMR 7157, Université Paris 7,

Hôpital Lariboisière, Paris, France

Epidemiology of acute poisonings

Acute poisoning: a leading cause of hospital admission and mortality in youths.

A poisoning should be considered as severe if : • Life-threatening symptoms occur, including hemodynamic instability, heart dysrhythmia or conduction disturbances, coma, seizures, respiratory failure or alveolar hypoventilation • The patient has been exposed to a large amount of toxicant requiring a close monitoring • The patient is more vulnerable (co-morbidities, elderly or infants)

French Society of Critical Care Medicine Réanimation 2006

Acute poisoning, a dynamic process

Increasing symptoms

Possible sequellae

Possible death

time T0

Favorable outcome Delay

24 to 72 h

exposition Danel V, 2005

• Diagnosis in medical toxicology relies on the clinical approach (history + toxidrome) . • Management of poisoned patients is mainly based on supportive treatments and antidotes if available. • For emergent management, the routine biological tests are more important than the toxicological analysis.

Place of toxicological analysis

French Society of Critical Care Medicine Réanimation 2006

• Ethanol poisoning

• Ethylene glycol

• Fluorhydric acid

• Cyanide

• Anti-vitamin K

• Organophosphates

• Iron salts

Intoxication Biological tests Toxicological analysis

Serum glucose

Arterial pH and anionic gap

Serum calcium

Lactate

Prothrombin time

Red cell cholinesterases

Blood iron concentration

Plasma ethanol

Plasma ethylene glycol

Blood fluoride

Blood cyanide - - -

Place of toxicological biochemistry in a situation of emergency

Signal

measurement

Interpretation

- Diagnostic

- Prognostic

- Thérapeutical

Interest

of analysis

Sampling Blood/plasma

Urine

Gastric liquid

But also:

Salive

Sweat

Hair

Detection

Separation +

Detection

Quantification

What is a toxicological analysis?

x

x x x

x

Urine Blood

Prognostic value of plasma concentration measurement in the case of injuring toxicants

Acetaminophen poisonings: Prescott normogram

Paraquat poisonings: Proudfoot and Scherrmann normograms

Concentration Target

Temps

Apnea Coma Shock

Toxicodynamics Toxicokinetics

Dual aspects of any toxic phenomenon

Mégarbane B. Expert Opin Drug Metab Toxicol 2008

Definitions …

Intoxication: interaction between a toxicant / target cells (biological mechanisms)

Concentration at the target tissue

(Ir)reversible cellular modifications

Clinical signs and symptoms

Toxicokinetics: Concentration = f (Time)

Toxicodynamics: Effect = f (Dose)

Dynetics: Effect = f (Time)

TK/TD relationship : Effect = f (Time-course concentration)

PK/PD relationships … in pharmacology

Analgesics

Hypnotics

Anticonvulsive drugs

Cardiovascular

Steroids

Antibiotics

Anticancer drugs

Fields of application :

How the drug acts on the organism …

How the organism acts on the drug …

PK/PD relationships … in toxicology

More difficult because: - The ingested dose and ingestion date are generally unknown - Measurement of blood concentrations is not routinely performed

- A reversible PD effect is not directly attributable to the toxicant and easily measurable in clinical practice

Signs:

• Physicians

• Nurses

• Researchers

Analysis:

• LCHP

• LCMS

• GCMS

PK/PD:

• Data base

• Mathematical

analysis

Severity Blood Conc

Functional toxicant – Injuring toxicant

t

x

x x

x

x

I- Functional toxicants : Severity factors

II- Injuring toxicants : Prognosis factors

t

x

x

x

x

x

Severity Blood Conc x

The concept of “biophase”

Definition: The virtual compartment containing the receptors that interact with the active compound

Each active compound should reach its target site to exhibit its effects No simple relationship between concentrations in blood and biophase.

Corresponds to the delay between the occurrence of the effects and evolution of the concentrations

The onset of effects depends on: Distribution according to the specificities of the target tissue (anatomy,

perfusion, permeability, …) Metabolism if the ingested compound is inactive. Cellular events (receptor affinity, transduction, post-receptor events, …)

Clockwise hysteresis Effect decreases more quickly than concentrations

Anticlockwise hysteresis Effect decreases more slowly than concentrations

Hypothetical PK/PD relationships showing clockwise or anticlockwise hysteresis

When connected in chronological order, the manner in which the data points appear allows one to define a hysteresis loop, meaning that the time courses of concentration and effect are out of phase.

100

125

150

175

200

225

0 2,5 5 7,5 10 12,5 15

Ex

pir

ato

ry t

ime

(% o

f b

as

al

va

lue

)

Plasma pralidoxime concentrations (mg.L-1)

Concentration = 7,1 mg.L-1

in elimination phase = inefficiency

Theoric limit

Concentration = 7,3 mg.L-1

in resorption phase = efficience

Houzé P. Tox Sci 2010

PK/PD relationships for pralidoxime treatment of OP-related respiratory depression in rats Correlation between toxicity and blood concentrations

in acute ethanol poisoning

0.9 – 2.5 g/l: euphoria - excitation

1.8 - 3 g/l: confusion

2.5 – 4.5 g/l: stupor, coma

> 4.5 g/l: respiratory depression,

death

Are you drunk?

Correlation between toxicity and plasma concentrations in acute theophyline poisonings

10-20 mg/l (55-110 µmol/l): Therapeutic level

20-40 mg/l: Minor toxicity

40-100 mg/l: Moderate toxicity

> 100 mg/l: Severe toxicity

> 120 mg/l: Risk of convulsions (50%)

EXPOSURE

VARIABILITIES

PHARMACODYNAMICS (Receptors : number and affinity)

PHARMACOKINETICS (A.D.M.E)

GENETIC POLYMORPHISM

PHYSIOLOGY (Age, sex, weight, …)

PATHOLOGY (Renal, liver, and cardiac insufficiencies, burnings, …)

CHRONOBIOLOGY

ENVIRONMENT OBSERVANCE TOLERANCE DIFFRENCES BETWEEN

THE ACTIVE CONCENTRATIONS

AND THE EFFECTS

Sources of individual variability

Effecting proteins represent the origin of the variability

750 (368 cloned)

10n

40 a

2 a

30 a

Receptors

Signal transducing proteins and other pharmacological targets

Metabolism enzymes

Circulating transport proteins

Membrane transport proteins

PD (TD) PK (TK)

a) Number implicated in chemotherapy

Phase I Phase II

Gene polymorphism in the enzymes implicated in the drug metabolism

Guengerich FP, Mol Interv 2003

Metabolic ratio (log)

Gene polymorphism of CYP 2D6

Rapid

Metabolizers

(85-90 %)

Slow

Metabolizers

(7-10 %)

Ultra-rapid metabolizers

(1-7 %)

Clinical consequences of gene polymorphism of CYP 2D6

CYP 2D6 (MR)

CODEINE CODEINE- 6- GLUCURONIDE

10%

MORPHINE

MORPHINE 3 and 6- GLUCURONIDE

NORCODEINE

CYP 3A4

UGT

CODEINE CODEINE - 6- GLUCURONIDE

NORCODEINE

CYP 3A4

Slow metabolizer

Phenotype

without

analgesic

pattern

CYP 2D6 (MR)

CODEINE CODEINE- 6- GLUCURONIDE

NORCODEINE

CYP 3A4

Rapid metabolizer

Phenotype

with

hyperproduction of

morphine

Case report 1 -

Male 62 yrs, Chronic lymph leukemia

Bilateral pneumonia, fever, cough

D1: ceftriaxone + clarithromycine + voriconazole + codeine (25 mg x 3/d)

D4: consciousness impairment (GCS: 6), with myosis

Naloxone 0,4 mg/h: normal consciousness

Biology

– Creatinine: 182 µmol/L

Pharmaceutical concentrations:

– Codeine 114 µg/L (N: 13-75)

– Morphine 80 µg/L (N: 1-4)

– G6M 136 µg/L (N: 1-13)

Genotyping:

– CYP2D6 ultra-rapid

Gasche et al. NEJM 2004

Case report 2 -

Normal pregnancy and birth

Postpartum pain:

– Codeine 120 mg/d

– Paracetamol 500 mg/d

D7: sleepness, reduced milk intake

D13: death of the new-born

Autopsy:

– No congenital disease

– Morphine in blood 70 µg/L (N: 10-12)

– Paracetamol in blood : 5.9 µg/L

Maternal gentoyping:

– heterozygote CYP2D6*2x2 (ultrarapid)

– UGT2B7*2 (ultrarapid))

Koren et al. Lancet 2007

Elkalioubie A. Eur J Clin Pharmacol 2011

Near-fatal tramadol cardiotoxicity in a CYP2D6 ultrarapid metabolizer

- Ultrarapid metabolizer phenotype suggested by tramadol/M1 metabolic ratio

- Heterozygous for duplicated wt allele predictive of CYP2D6 ultrarapid metabolizer

phenotype

+ Ketoconazole at inhibitory concentration of CYP3A/CYPB6 (200 ng/ml)

Tramadol

O-desmethyltramadol (M1)

N-desmethyltramadol (M2)

CYP2D6 CYP2B6/CYP3A4

Steps of the pharmacokinetics

ABSORPTION

DISTRIBUTION

METABOLISM

EXCRETION

Enterocyte

Endothelial cell (BBB)

Hepatocyte

Renal

epithelium

Phase 1: CYP,…

Phase 2: UGT, SULT, GST, NAT…

M

M

M-OH

M-O-R

Phase 0

Phase 3

T

T

M-O-R

INTESTINE (enterocyte)

FECES ORAL EXPOSURE Secretion Absorption

LIVER (hepatocyte)

BILE

Efflux

Uptake

Excretion

BBB (brain

microcapillary endothelial cell)

Efflux Uptake

BRAIN

KIDNEY (tubule proximal cell)

URINE

Secretion

Reabsorption

Filtration

From JM SCHERRMANN-Comprehensive Med Chem 2007

PHARMACOKINETIC ROLE OF TRANSPORTERS

ABC (ATP Binding Cassette) SUPERFAMILY 48 human genes; 9 drug transporters

Two superfamilies involved in drug transport

http://nutrigene.4t.com/human abc.htm

Paulsen, FEBS Lett 1998 ; J Mol Biol 1998

• SLC (Solute Linker Carrier) SUPERFAMILY 362 mammalian genes; ≈ 30 drug transporters

http://www.gene.ulc.ac.uk/nomenclature

INFLUX

EFFLUX

They share commun Substrates and Inhibitors with similar ranges of active concentrations (from µM to mM)

They share similar regulation pathways from gene to protein expression (induction, repression of genes...)

They are sensitive to genetic polymorphisms, physiological and disease states

They follow Michaelis-Menten kinetics like metabolism enzymes

Common Properties between Enzyme and Transporter Proteins

Subfamilies ABCA ABCB ABCC ABCD ABCE ABCF ABCG

Members 12 11 12 4 1 3 5

Important isoforms for PK

? ABCB1

(P-gp)

ABCCs

(MRPs) ? ? ? ABCG2

(BCRP)

N = 48 in humans, 7 subfamilies

Superfamily of ABC (ATP-Binding Cassette) transportors

P-gp: P-glycoprotéine

MRPs: Multidrug resistance-associated proteins

BCRP: Breast cancer resistance protein

P-glycoprotein substrates Anticancer agents

– Doxorubicine

– Vincristine

– Irinotecan

– Paclitaxel

– Inhibiteurs tyrosine kinase..

Anticoagulants – Dabigatran, …

Cardiovascular drugs – Digoxine

– Quinidine

– ivabradine

Antiproteases – Indinavir

– Nelfinavir

– Saquinavir

Anti-diarrhea drugs – Loperamide

Antibiotics – Erythromycine, …

Immunosuppressive agents – Cyclosporine – Tacrolimus

Steroids – Dexamethasone

Calcium channel blokers – Verapamil

Adrenergic antagonists – Talinolol

H1 Anti-histamine drugs – fexofenadine

H2 Anti-histamine drugs – Cimetidine

Antidepressants – Amitriptyline

Others – Colchicine – Phenytoine

Transporters expressed at the blood-brain barrier

Brain extracellular liquid

Astrocyte

Pericyte

Endothelial cell

Blood

MRP4

P-gp

BCRP

Basal membrane

Mechanisms of buprenorphine toxicity

BUP

Pharmacological effect

N-BUP

Central respiratory toxicity

Alhaddad. Crit Care Med 2012

Pharmacologic

or Genetic

P-gp inhibition

Pharmacongenetics of PK effectors for opioids

In red: proteins presenting functional SNP

Drug Transport Metabolism Metabolites Active

metabolite

Morphine P-glycoprotein UGT2B7 MM6G, M3G yes (M6G)

Codéine CYP2D6 Morphine yes

Methadone P-glycoprotein CYP3A4 EDDP no

Fentanyl P-glycoprotein CYP3A4 Norfentanyl no

Tramadol P-glycoprotein CYP2D6 M1, M2 yes (M1)

Oxycodone P-glycoprotein CYP2D6 oxymorphone yes

Modeling of individual TK-TD relationships

D --> Cp (t), Ct (t) --> Ce (t) --> E (t)

if Cp(t) and E(t) simultaneous ==> similar Ce(t) and Cp(t) kinetics

if Cp(t) and E(t) with delay ==> different Ce(t) and Cp(t) kinetics

• Mono- and multi-compartment modeling

• Kinetics of the effects E(t) according to the measured concentration (Ce, Cp ou Ct)

Linear model: E = S.C + E0

Semi-log model: E = S.logC + E0

E max model: E = Emax.C/Ec50 + E0

Sigmoidal model: E = Emax.Cen/EC50n + Cen

Cmax

Slope of toxicity

Threshold of toxicity

Emax

a E50

Plateau

EC50

Sigmoidal modeling of individual TK/TD relationships

Description of a TK/TD relationship in drug-related coma

Patient 1

0255075100125

3

6

9

12

15

Plasma ethanol concentration (mmol/l)

Gla

sg

ow

Co

ma S

co

re

2 Sigmoid curve

1 Functional toxicant

4 calculation of C50 and Hill coefficient

3 Plateau of effects

1 g/l = 21.7 mmol/l

Patient 17

2030405060708090100110120130140150160

3

6

9

12

15


Gla

sgo

w C

om

a S

core

2 Lower C50 1

Move to the right

Findings suggestive of a drug-drug interaction

Patient 9

405060708090100110120

3

6

9

12

15


Gla

sg

ow

Co

ma S

co

re

2

More elevated C50

1 Move to the left

3

Lower plateau

Findings suggestive of a pharmacodynamic tolerance TK/TD relationships in acute ethanol poisonings Patient 1

0255075100125

3

6

9

12

15


Gla

sgow

Com

a S

core

Patient 9

405060708090100110120

3

6

9

12

15


Gla

sgow

Com

a S

core

Patient 4

1020304050607080

3

6

9

12

15


Gla

sgo

w C

om

a S

core

Patient 6

0102030405060

3

6

9

12

15


Gla

sgow

Com

a S

core

Patient 14

0102030405060708090100

3

6

9

12

15


Gla

sgow

Com

a S

core

Patient 17

2030405060708090100110120130140150160

3

6

9

12

15


Gla

sgow

Com

a S

core

Patient 18

405060708090100110120

3

6

9

12

15


Gla

sgow

Com

a S

core

Patient 20

0102030405060708090100

3

6

9

12

15


Gla

sgow

Com

a S

core

Patient 1

0255075100125

3

6

9

12

15


Gla

sgow

Com

a S

core

Mégarbane B. Clin Toxicol 2007 (abstract)

TK/TD relationships in acute meprobamate poisonings

Mégarbane B. Clin Tox 2004 (abstract)

TK/TD relationships in acute phenobarbital poisonings


Description of the sigmoidal models to represent

TK/TD relationships in psychoactive drug poisonings

Hill coefficients C50

14.9 [5.7-75.9] 61.2 mmol/l [38.9-96.3]

10.1 [2.8-49.2] 384 µmol/l [116-772]

4.8 [2.5-11.8] 257 µmol/l [160-442]

Ethanol

Meprobamate

Phenobarbital 050100150200250300350400

3

6

9

12

15

Plasma GHB concentration (mg/l)

Gla

sg

ow

Co

ma S

co

re

Case report TK/TD relationship in GHB intoxication

EC50: 189 mg/l Hill coefficient: 9.2 R2: 0.96

M, 26 years Ethanol 36.1 mmol/l GHB 324 mg/l Unique consumption

Mégarbane B. Clin Toxicol 2007 (abstract)

0 5 10 15 20 25 30 35

100

1000

R,S-methadone

R-methadone

S-methadone Naloxone infusion rate

0.1

1

10

100

Time after ingestion (h)

Pla

sma

met

had

on

e co

nce

ntr

atio

ns

(ng

/ml)

Nalo

xon

e infu

sion

rate

(mg

/h)

0 250 500 750 1000 1250 15000.0

0.1

0.2

0.3

0.4

0.5

R,S-methadone concentration (ng/ml)

Nalo

xo

ne i

nfu

sio

n r

ate

(mg

/h)

0 100 200 300 400 500 600 7000.0

0.1

0.2

0.3

0.4

0.5

R-methadone concentration (ng/ml)

Nalo

xo

ne i

nfu

sio

n r

ate

(m

g/h

)

Mégarbane B. Crit Care 2007

Modeling of respiratory effects in methadone overdose

E0 0 Emax 0.78 µg/kg/min EC50 23.9 µmol/l Hill coefficient 1.79 R2 0.99

Mégarbane B. Intensive Care Med 2006

Modeling of cardiac toxicity in venlafaxine poisoning

Hypothesis: Slow CYP 2D6 metabolizer

Therapeutic range: 0.25-0.85 µmol/l

TK/TD variability according to the TD parameter used in flecainide poisonings

0.500.751.001.251.50

0.0

2.5

5.0

7.5

10.0 Patient 2

Flecainide concentration (mg/l)

Dob

utam

ine

infu

sion

(µg

/kg/

min

)

Catecholamine infusion rate QRS duration


Insulin self-poisonings: interests of TK-TD relationships to analyze the needs in dextrose infusion

Patient 1

05001000150020002500

0

5

10

15

Plasma insulin concentration (mIU/l)

Glu

co

se i

nfu

sio

n r

ate

(g

/h)

Patient 2

050100150200250

0

10

20

30


Glu

co

se i

nfu

sio

n r

ate

(g

/h)

Patient 3

02505007501000

0

50

100

150


Glu

co

se i

nfu

sio

n r

ate

(g

/h)

Patient 4

02500500075001000012500

0

10

20

30

40


Glu

co

se i

nfu

sio

n r

ate

(g

/h)

Patient 5

01000200030004000500060007000

0

10

20

30

40

50

60


Glu

co

se i

nfu

sio

n r

ate

(g

/h)

Patient 6

050100150200250

0

5

10

15


Glu

co

se i

nfu

sio

n r

ate

(g

/h)

Mégarbane B. Crit Care 2007

TK/TD using blood concentrations TK/TD using plasma concentrations

Chloroquine poisonings: interests of measuring blood concentrations in comparison to plasma concentrations

0 25 50 75 100 125 1500.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

Blood chloroquine concentration (µmol/L)

Epin

ephr

ine

infu

sion

rat

e (m

g/h)

R2 = 0.96

0 10 20 30 40 500.0

0.5

1.0

1.5

2.0

Blood chloroquine concentration (µmol/L)

Epin

ephr

ine

infu

sion

rat

e (m

g/h) R2 = 0.91

0 1 2 3 4 5 6 7 8 90.0

0.5

1.0

1.5

2.0

Plasma chloroquine concentration (µmol/L)

Epi

neph

rine

infu

sion

rat

e (m

g/h)

0 5 10 15 20 250

1

2

3

4

Plasma chloroquine concentration (µmol/L)

Epin

ephr

ine

infu

sion

rat

e (m

g/h)

R2 = 0.05

R2 = 0.17

Patient 1

Patient 2


Interests of plasma lactate concentrations in acute cyanide poisonings

Baud F, Borron SW, Mégarbane B. Crit Care Med 2002

40 µM 100 µM

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - F E D C B A

0

Severity

Blood concentrations

Interests of PK/PD relationships to understand the individual response to the treatments

- Is required when data are derived from studies with different designs and with great numbers of individuals with sparse and unbalanced data. - Allows estimation of population mean values of PK parameters, their inter-individual and intra-individual variability, as well as residual variability. - May test the influence of pertinent factors that could explain inter-individual variability including age, use of charcoal, and co-ingestions (CYP metabolism). - The Bayesian approach allows estimation of individual PK parameters from population parameters and concentrations/effects measured in the patient. - However, usually requires numerous statistical assumptions (for random and fixed effects distribution), estimation methods (algorithms & approximations), and softwares (NONMEM, PKBUGS/WinBUGS, S-PLUS, and MLwiN).

Population-based PK/PD methodology

Mégarbane B. Clin Tox2010

Goodness-of-fit plots : population predictions (PRED) versus observations (DV) and population-weighted residuals (WRES) versus predictions

Chloroquine poisoning: Design of a PK/PD model


Comparison between the 10th (dash line), 50th (full line), and 90th (dash line) percentile obtained from 1,000 simulations and the observed data (DV, points)

Chloroquine poisoning: Evaluation of the final PK/PD model

PK/PD population modelling in chloroquine poisonings


Chloroquine poisonings: Simulated probability over time for having an epinephrine

infusion rate >3 mg/h


QT prlongation following citalopram poisonings

Isbister G. BJCP2005

Risk reduction in escitalopram-related QT prolongtation with charcoal

No charcoal Charcoal

Van Gorp F. Br J Clin Pharm 2011

Moderate reduction in the risk of abnormal QT interval with SDAC

Dose-related lengthening of the QT interval lagging the increase in drug

concentration.

The effects of decontamination procedures on venlafaxine pharmacodynamics in overdose

Isbister G. BJCP2011

Conclusions

To help establishing the positive and differential diagnosis;

To understand the clinical presentation and evolution of poisoned patients;

To suggest and identify a mechanism of individual variability in response to a given toxicant;

To understand the limits and identify the patients that may benefit from any toxicokinetic treatment (ex: GI decontamination);

To explain the interests of an antidote (if weak therapeutic index) or determine its efficient dose (if weak toxicity slope).

Is the study of TK/TD relationships interesting in acute poisonings?

srlf dabor - paris descartes€¦ · page 1 what can we learn from analysis of the toxicokinetic /...

Documents