pk/pd of non-steroidal drugs (nsaids)
TRANSCRIPT
PK/PD of Non-Steroidal Drugs (NSAIDs)
Pierre-Louis Toutain,
Ecole Nationale Vétérinaire
INRA & National veterinary School of Toulouse, France
Wuhan 18/10/2015
Anti-inflammatory drugs
Corticosteroids NSAIDs
Why to investigate NSAIDS
• All domestic species suffer pain and controlling pain is a priority issue for veterinary pharmacologists
• Inflammation is a major source of pain
–Acute (e.g. infectious) or chronic (e.g. osteoarthritis)
• To determine an adequate dosage regimen
–Efficacy
–Safety
» Selectivity (COX1 vs. COX2)
A review
Why to investigate PK/PD for NSAIDS
Why PKPD for NSAIDs
• Doses for NSAIDs were difficult to establish in veterinary medicine and historically wrong
• The case of aspirin in emblematic
1795: Rev Edward Stone described the antipyretic properties of the willow
1897
Aspirin was not properly used in veterinary medicine
• Apparent no efficacy in large domestic species
• Very toxic in cats
• Clinicians were actually unable to find the appropriate dose because they extrapolated the human dose to the veterinary species using the allometric paradigm
The Lloyd E. Davis’ paper on salicylate (1972) A non-linear PK with huge interspecific differences
37h
8.6h
5.9h
1.0h
0.8h
T1/2h
Time
Plasma salicylate 37h
8.6h
5.9h
1.0h
0.8h
T1/2h
Time
Plasma salicylate
A double log plot of salycilate half-life in different species
Body Weight (KG)
Ha
lf-l
ife
(h)
The Lloyd E. Davis’ paper (1972)
“the present data indicate the futility of extrapolating dose and dosage regimens from one species to another, as has been done in the past, in the treatment of domestic animals”
The Lloyd E. Davis’ paper (1972)
“We believed that information relevant to the biotransformation and rates of disappearance from blood of several drugs in a series of large domestic
animals might prove of value”
Mechanism of action of NSAIDs
•1982 Nobel Prize for Medicine for his research on mechanism of action of NSAID (prostaglandins).
Cyclo-oxygenases (COXs)
Physiology Inflammation
COX1: constitutive
macrophage / other cells
COX2: inducible
TXA2 PGI2 PGE2
Platelets endothelium Kidney
stomach
Protease PG others
médiators
Inflammation
Side effects Therapeutic effects
Physiological function
Two consequences of the knowledge of the COXs
• Possibility to develop preferential or even selective NSAIDs (i.e. COX-2 inhibitors)
• To use in vitro/ex vivo data as surrogates of clinical endpoints to approximate dosage of NSAIDS
NSAIDs selectivity
What is selectivity: COX-1 vs COX-2
COX2
COX1
0.3 3 (µg/mL)
effect
Emax
EC50, COX-1 EC50, COX-2
102
1
50
50
CoxIC
CoxICRatio
Robenacoxib selectivity
-20
0
20
40
60
80
100
0.001 0.01 0.1 1 10 100 1000
% i
nh
ibit
ion
Robenacoxib concentration (µM)
Fitted COX 1
Fitted COX 2
Observed COX-1
Observed COX-2
Ratio EC50=140
Test systems used to determine NSAIDS selectivity
In vitro test to determine COX selectivity
•Numerous in vitro test systems
–purified/recombinant enzymes,
–cultures of intact cells
–target species whole blood (the so-called whole blood assay).
The whole blood assay
The whole blood assay
• Freshly drawn, heparinized whole blood is incubated with calcium ionophore A23187
–A23187 activates COX1 and stimulate the production of thromboxane B2 (TxB2) by platelets .
– TxB2 concentration is measured by immunoassay,
• When blood is incubated with E. lipopolysaccharide (LPS), COX2 is induced
– aspirin had no effect on LPS-induced TxB2, but inhibited TxB2 production by ionophore A23187 through acetylation of pre-existing COX1.
–Also measurement of PGE2
Semrad et al. Flunixin meglumine given in small doses: pharmacokinetics and
prostaglandin inhibition in healthy horses. Am. J. Vet. Res.1985, 46(12): 2474-
2479
Thromboxane B2 inhibition as function of flunixin concentration
Selectivity of veterinary NSAIDS
• COX-1 preferential
– Ketoprofen
– Vedaprofen
– flunixine
Non selective
– Aspirin
– Ibuprofene
– phenylbutazone
COX-2 preferential
Carprofene
Meloxicam
Ac. Tolfenamique
COX-2 selective
Firocoxib
Robenacoxib
Cimicoxib
Mavacoxib
Ex vivo/in Vivo model of inflammation
A.J. Higgins et al. Development of equine models of inflammation. Vet. Rec. 1987, suppl.120(22) 517-
522
Tissue Cage Model of Acute Sterile Inflammation (P. Lees)
• Implantation of perforated tissue cages at subcutaneous sites (4 per animal)
• Internal volume
– 35 ml (calf, camel, horse) = 15 ml (pig) =10 ml (sheep, goat)
• After >30 days, stimulation of granulation tissue by intracaveal injection 0.5 ml 1% carrageenan solution
• Withdrawal at pre-determined times of inflamed fluid (exudate)
• Withdrawal of non-inflamed fluid from control tissue cages (transudate)
Data for PK/PD modelling
Dependent variable for PK/PD modelling
• It can be useful to distinguish between drug:
–action
–effect
–response
NSAID blood
concentration
COX
inhibition
Inhibition
of PGE2
production
Suppression
of lameness
ACTION EFFECT RESPONSE
Mechanistic
interest
Mechanistic interest
Biomarker
Clinical outcome
(clinical or surrogate
end points)
Action vs. Effect vs. response for NSAIDs
Biomarkers and surrogates to compute a dose
Demonstrate
therapeutic response Confirming
Drug development
Screening Biomarkers
Surrogate
Field clinical outcome
Local temperature
Pain modulation
Binding affinity
COX inhibition
PGs production
Lameness
NSAID
Wellbeing/Demeanor
EC50 response EC50 response >> EC50 effect
EC50 in vivo effect EC50 action
whole blood assay
Which dependent variable for PK/PD modeling ?
NSAID plasma
concentration Inhibition
of COX
Inhibition of
PGE2
production
Suppression
of lameness
Requires 90-95% PGE2 inhibition
A first estimation of the dose using the EC90 of COX2 inhibition
𝑫𝒐𝒔𝒆 =𝑪𝒍𝒆𝒂𝒓𝒂𝒏𝒄𝒆 × 𝑬𝑪𝟗𝟎 𝒇𝒐𝒓 𝑪𝑶𝑿𝟐 𝒊𝒏𝒉𝒊𝒃𝒊𝒕𝒊𝒐𝒏
𝑩𝒊𝒐𝒂𝒗𝒂𝒊𝒍𝒂𝒃𝒊𝒍𝒊𝒕𝒚
In vivo determination of the dose
An example of application of PK/PD to determine a dosage regimen for a
NSAID in cat
Step 1: selection of an appropriate inflammatory model
Inflammation model: requirements
• Reversible for ethical reasons
• Time development appropriate to display a sustained inflammation window over 2-3 days during which the NSAID can be tested without the confounding effect of the spontaneous resolution of the inflammation
As for a conventional dose titration, PK/PD investigations generally require a relevant
experimental model (here a kaolin inflammation model)
Possibility to perform PK/PD in patient
• Kaolin = clay mineral
inflammatory reaction composed of neutrophilic
polymorphonuclear leukocytes, macrophages
• 500 mg kaolin/paw : significant increase in body and
skin temperature and paw volume with no signs of
clinical remission before day 4
Kaolin Inflammation model development
Clinical relevance
feasibility (e.g. vertical force exerted by the inflamed
limb)
metrological performance in healthy animals
Responsiveness to drug administration
end points evaluated: body temperature, skin temperature,
paw volume, pain (digital pressure), game (willingness to
play) and locomotion scores, walking distance, locomotion
tests
End point selection
“Clinical endpoints
Paw volume, skin temperature,
body temperature
• To measure the vertical forces, a corridor of walk is used with a force plate placed in its center.
• The cat walks on the force plate on leach.
Video
Measure of vertical forces exerted on force
plate
• The measure of vertical force and video control are recorded
Vertical forces (Kg)
Video
Measure of vertical forces
exerted on force plate
Surrogate end points: locomotion tests
descending, climbing and
creeping time
Measure of pain with analgesiometer
• Cat is placed in a Plexiglas box.
• A light ray is directed to its paw to create a thermal stimulus.
• The time for the cat to withdraw its paw of the ray is measured.
withdrawal time of the paws (second)
Video
Step 3 validation of the model
• Repeatability and reproducibility of the different measurements
• Spontaneous time-development of the inflammation
Reproducibility of end point measurements End points Coefficient of Variation (%)
Repeatability Time effect Day effect
Body temperature (0.03) 0.25 0.07
Skin temperature
difference (spot 2) 24.91 18.51
Left paw skin
temperature (spot 2) (1.24) 1.67 0.83
Right paw skin
temperature (spot 2) (1.24) 1.02 0.95
Left paw volume 1.85 0.09 0.91
Right paw volume 2.41 0.19 0.54
Descending time 13.24 3.36 6.31
Climbing time 25.85 5.25 17.97
Creeping time 9.17 1.24 3.68
Time development of the inflammation
Follow up of mean paw volumes
-10
0
10
20
30
40
50
60
70
80
90
100
-1 0 1 2 3 4 5 6 7 8 9
Days after kaolin injection
Increase
in
paw
volu
me (
%)
Treated cats
Control cats
500 mg
Kaolin
0.3 mg/kg
Meloxicam
Time development of the inflammation
-10
0
10
20
30
40
50
60
70
80
90
100
-1 0 1 2 3 4 5 6 7 8
Incr
ease
in
paw
volu
me
(%)
Days after kaolin injection
Follow up of mean paw volumes
Control cats
500 mg
Kaolin
Time development of the inflammation
Follow up of mean locomotion score
0
1
2
3
4
5
-1 0 1 2 3 4 5 6 7 8 9
Days after kaolin injection
Lo
co
mo
tio
n s
co
re
Treated cats
Control cats
0.3 mg/kg
Meloxicam
500 mg
Kaolin
Step 4: measurement of the drug effect
38.0
38.5
39.0
39.5
40.0
40.5
41.0
0.0 0.5 1.5 2.5 3.5 4.5 6.0 7.0 8.0 10.0 12.0 23.4
Re
cta
l te
mp
era
ture
(°C
)
Time after robenacoxib administration (h)
Follow-up of mean rectal temperature
Results: body temperature
2 mg/kg
Robenacoxib
0
1
2
3
4
5
0.0 0.6 1.5 2.5 3.5 4.6 6.1 8.1 10.1 12.1 23.5
Lo
co
mo
tio
n s
co
re
Time after robenacoxib administration (h)
Follow-up of mean locomotion score
Results: locomotion score
2 mg/kg
0
5
10
15
20
25
30
35
40
45
50
55
60
0.0 0.6 1.6 2.6 3.6 4.6 6.2 8.2 10.1 12.2 23.5
Clim
bin
g tim
e (
s)
Time after robenacoxib administration (h)
Follow-up of mean climbing time
Results: climbing time 2 mg/kg
0
2
4
6
8
10
12
14
16
18
20
22
0.0 0.8 1.7 2.7 3.7 4.8 6.3 8.3 10.3 12.3 23.7
Wit
hd
raw
al t
ime
(s
)
Time after robenacoxib administration (h)
Follow-up of mean paw withdrawal time
Results: Pain as withdrawal time
2 mg/kg
Step 5: modelling
Turnover model Indirect response model
Principle for model building
• 𝑶𝒃𝒔𝒆𝒓𝒗𝒆𝒅 𝒓𝒆𝒔𝒑𝒐𝒏𝒔𝒔𝒆 = 𝑰𝑵𝑷𝑼𝑻 − 𝑶𝑼𝑻𝑷𝑼𝑻
𝑹𝒆𝒄𝒕𝒂𝒍 𝒕𝒆𝒎𝒑𝒆𝒓𝒂𝒕𝒖𝒓𝒆 = 𝑻𝒉𝒆𝒓𝒎𝒐𝒈𝒆𝒏𝒆𝒔𝒊𝒔 − 𝑻𝒉𝒆𝒓𝒎𝒐𝒍𝒚𝒔𝒊𝒔
Circadian rhythm
Lipopolysaccharide
Stimulation or inhibition
NSAID
Stimulation or inhibition
The turn over model (indirect effect model)
where dR/dt represents the rate of variation in the response
variable (R). Kin is the rate of input and Kout•R is the rate of
loss; the model assumes that the measured response is being
formed at a zero-order constant rate (Kin) but disappears in a
first-order manner (Kout).
𝒅𝑹
𝒅𝑻= 𝑲𝒊𝒏 − 𝑲𝒐𝒖𝒕 × 𝑹
The 4 basic equations
Inhibition Kin
Inhibition Kout
Stimulation Kin
Stimulation Kout
SC50 125.2 ng/mL
SD50 = 1.85 mg/kg/24h
Robenacoxib: antipyretic effect
38
39
39
40
40
41
0 2 4 6 8
Time (h)
Bo
dy
te
mp
era
ture
(°C
)
0
200
400
600
800
1000
1200
1400
1600
1800
Co
nc
en
tra
tio
ns
(n
g/m
L)
𝒅𝑹
𝒅𝑻= 𝑲𝒊𝒏 − 𝑲𝒐𝒖𝒕 𝟏 +
𝑺𝒎𝒂𝒙 × 𝑪𝒏
𝑺𝑪𝟓𝟎𝒏 + 𝑪𝒏
IC50 42.8 ng/mL
ID50 = 0.63 mg/kg/24h
Robenacoxib: AI effect (climbling)
3
8
13
18
23
28
33
0 2 4 6 8 10
Time (h)
Clim
bin
g t
ime
(s
)
0
200
400
600
800
1000
1200
1400
1600
1800
Co
nc
en
tra
tio
ns
(n
g/m
L)
𝒅𝑹
𝒅𝑻= 𝑲𝒊𝒏 𝟏 −
𝑰𝒎𝒂𝒙+𝑪𝒏
𝑰𝑪𝟓𝟎𝒏 +𝑪𝒏 -𝑲𝒐𝒖𝒕 × 𝑹
IC50 40.0 ng/mL
ID50 = 0.59 mg/kg/24h
Robenacoxb : analgesic effect
0
200
400
600
800
1000
1200
1400
1600
1800
0
10
20
30
40
50
60
70
80
90
100
0 2 4 6 8 10 12
Co
nc
en
tra
tio
ns
(n
g/m
L)
Pa
in (
%)
Time (h)
𝒅𝑹
𝒅𝑻= 𝑲𝒊𝒏 𝟏 −
𝑰𝒎𝒂𝒙+𝑪𝒏
𝑰𝑪𝟓𝟎𝒏 +𝑪𝒏 -𝑲𝒐𝒖𝒕 × 𝑹
Step 6: simulations
Simulated dose-response: Robenacoxib: analgesic effect
-250
-200
-150
-100
-50
0
50
100
0 4 8 12 16 20 24
Time (h)
Pain
sco
re (
%) 0.1 mg/kg
0.2 mg/kg
0.3 mg/kg
0.4 mg/kg
0.5 mg/kg
1 mg/kg
Dosage interval and effectiveness
Simulations Robenacoxib: once vs. twice a day
Mean effect 32 % Mean effect 52 %
Simulated time course of pain
0
10
20
30
40
50
60
70
80
90
100
0 4 8 12 16 20 24
Time (h)
Pa
in (
%)
5 mg/kg
2 x 2.5 mg/kg
5 mg/kg split in 12
Mean effect 96 %
Paw inflammation model
Freund adjuvant arthritis in horse
Carpitis
Carpitis
Modèle de carpite à l'adjuvant de Freund
• Endpoints
– Stride lenght
– Others
PK / PD: flunixine
Time (h)
Co
nc
en
trati
on
(µ
g/m
l)
Str
ide
le
ng
th (
cm
)
Time (h)
Co
nc
en
trati
on
(µ
g/m
l)
Str
ide
le
ng
th (
cm
)
Time (h)
Co
nc
en
trati
on
(µ
g/m
l)
Str
ide
le
ng
th (
cm
)
Co
nc
en
trati
on
(µ
g/m
l)
Str
ide
le
ng
th (
cm
)
8
0
16
0 4 8 12 16 20 hours
Str
ide
le
ng
ht
(cm
)
1
0.5
2
DOSE mg/kg
Dose effct relationship for flunixin as
obtained from a PK/PD model
12
14
8
4
0 0 4 8 12 16 20 24
heures
Stride length (cm)
1.25
1.0
1.5 2 4 8
DOSE mg/kg
PK/PD: Phenylbutazone
8
0 0 4 8 12 16 20 24
hours
Stride length (cm)
Flunixine 1mg / kg
PBZ 4mg / kg
16
20
PK/PD : PBZ vs flunixine
PD parameters for different NSAIDs
PD parameters Efficacy Potency Sensitivity
Drugs Emax (cm) EC50
(µg/mL)
Slope
PBZ 13.6 3.6 >5
Flunixin 22.8 0.93 >5
Meloxicam 27.4 0.19 >5
Application of PK/PD to determine a dosage regimen for NSAIDs
PBZ
Flunixin
Meloxicam
Ketoprofen
Meloxicam
Nimesulide
Tolfenamic acid
COXIB
Meloxicam
Coxib
Ketoprofen
Tolfenamic acid
Conclusion
• PK/PD is a powerful tool to determine a dosage regimen for NSAIDS –Dose & dosing interval
• Require appropriate in vivo models
• Require modelling & simulation