Transcript Models for proconvulsant drugs and precautions for

Workshop
Models for proconvulsant drugs
and precautions for entry into man
Jean-Pierre MARTINOLLE
Safety Pharmacology Advisory Group
|
1
SEIZURE / CONVULSION
Facts …
Phase
Preclinical
Phase III/
Marketing
Post-Marketing
Post-Marketing
Causes of attrition
Serious ADRs
Causes of attrition
Causes of attrition
ADRs on label
Serious ADRs
Withdrawal from sale
ABPI (2008)
Car (2006)
Sibille et al. (1998)
ABPI (2008)
Olson et al. (2000)
BioPrint® (2006)
Budnitz et al. (2006)
Stevens & Baker
(2008)
156 CDs stopped
88 CDs stopped
1,015 subjects
63 CDs stopped
82 CDs stopped
1,138 drugs
21,298 patients
47 drugs
Cardiovascular
24%
27%
9%
35%
21%
36%
15%
45%
Hepatotoxicity
15%
8%
7%
29%
21%
13%
0%
32%
Haematology/BM
3%
7%
2%
3%
4%
16%
10%
9%
Nervous system
12%
14%
28%
2%
21%
67%
39%
2%
Source:
Sample size:
●
Phase I-III
Causes of attrition
Information:
●
Phase I
Nervous systems AEs responsible in 22% of cases for trial halted,
delayed, development stopped, 39% of labelling restriction and for
25% of withdrawal from sales (2010)
Nervous systems AEs range from
●
●
●
High concern: seizure liability, neuropathy, brain phospholipidosis, abuse
liability, ocular toxicity, suicidality
Medium concern: sedation/sleep disorders, motor/coordination effects,
cognition, dizziness, auditory, psychiatric, somatosensory, anxiety…
Low concern: headache, autonomic, appetite, nausea…
Thundiyil et al, J Med Toxicol 2011
Adapted from Redfern WS et al. SOT 2010 Poster 1081
|
2
DrugCite 2013
SEIZURE / CONVULSION
Definition / Risk
●
SEIZURE
●
●
●
Seizure
abnormal brain electrical
Sensitivity Previous Undetected
discharge, possibly leading to
Off Target
Pharmacological
Seizure
sudden, involuntary, time-limited
History
class
Disease
alterations in behavior, motor
On Target
History
Population
activity, autonomic function,
Facilitator
Compound
related
consciousness or sensation
Direct
CNS activity
related
convulsant
recorded through EEG
Healthy
Patient
SEIZURE
BBB penetration
RISK
Pgp
Chemical
Indication Co-disease
MRP, BCRP properties
Administration
a violent and involuntary
related
muscular activity (as a
Brain
Co-medication
consequence of brain seizure) of
concentration
one or several muscles which are
Route
part of an organ, or of limbs, or
Cmax
generalized to all the body
Acute
Single dose
detected through OBSERVATION
Metabolite
Seizure
CONVULSION
●
●
●
Unknown
Known
Structural
class
DRUG-INDUCED SEIZURE
●
●
Repeated dose
1% (world) with epilepsy, 3-10% (world ) will experience seizures in their life
10% people with natural seizure tendency if a stimulus (lower threshold)
Everyone could have a seizure (particular set of circumstances)
1.2% of seizures at emergency department
6% of new seizures are directly drug-related, 9% of status epilepticus are directly
drug-related
2% of mortality when it occurs, 5% of further status epilepticus when it occurs
Chronic
Seizure
Kindling
|
3
Preclinical Predictivity
●
Seizure is undetectable without sophisticated methods like EEG or bio-imaging
●
Convulsion is tricky to detect if only based on observation
● Rare event
● Rarely occuring in all species tested
● Often resolving quickly thus not seen (if limited time of observation)
● Not always leading to death thus no alert
Then chance of detection generally increases with the muliplication of studies and species
but can lead to a late discovery of the risk !
RISK TYPE
POTENTIAL MECHANISMS
Following single administration
BBB crossing, Primary/secondary pharmacology

Following single and repeated
administration, with no aggravation
BBB crossing, Primary/secondary pharmacology …

Following single and repeated
administration, with aggravation
BBB crossing/BBB altered, Primary/secondary pharmacology
Potential neurotoxicity
Kindling / Brain adaptative changes
Brain accumulation / ADME effects

BBB crossing/BBB altered, Primary/secondary pharmacology
Co-factors
ADME effects
Brain adaptative changes

Following repeated administration but not
after single administration
Observed in specific population (age,
gender, pathology)
Observed with co-medication
Observed with treatment withdrawal
PRECLINICAL
PREDICTIVITY

|
4
What are the chances to detect seizure / convulsion
during preclinical ?
Classical
Preclinical package
Possible detection
Limitations/ Comments
CNS Gross behavior (FOB)
rodent
 mainly clear convulsion not seizure,
improved detection when IV route
Only males-rodent-young adults
Single admin. / Limited observations
CNS Locomotor activity
rodent
 did not detect convulsion nor seizure
except if during administration
Only males-rodent-young adults
Single admin. / Very limited observations
Detect CNS properties possibly in relation
to seizure (stimulants, sedative …)
CARDIO Telemetry
RESPI assay, rodent/non rodent
 did not detect convulsion nor seizure
except if during administration
Only young adults
Single admin. / Very limited observations
Possibly video in Telemetry non rodent to
explain death
General TOX (acute, DRF, REG)
rodent / non rodent
 mainly clear convulsion not seizure,
improved detection when IV route
after single/repeated administration
+ brain/histology
Only young adults
Very limited observations / Rare video to
explain unknown lethality
Unadapted histopathology (generally no
brain structural changes visible)
Genotox package
Reprotox package
 did not detect convulsion nor seizure
Not designed for
except if during administration
No chance to detect a seizure / Low chance to observe a convulsion if no dedicated study
|
5
SEIZURE / CONVULSION
Improving Preclinical Risk assessment
●
Target
●
●
●
Drug class
●
●
●
use of SAR approach, in silico tools
Target population
●
●
●
●
specific CNS activity
any compound on the market, any withdrawal
Drug structure
●
●
specific localisation in the brain
specific CNS role in particular in relation to seizure
Drug class Black list
Alcohols, glycols
Anesthetics, local
Anesthetics, general
Antibiotics
Anticonvulsants
Anticholinergics
Antidepressants, cyclic
Antidepressants, other
Antifungals
Antihistaminics
Antineoplasics
Antiparasitics
Antivirals
Asphyxiants
Cardiovascular
Drugs of abuse/withdrawal
Hypoglycemics
Immunosuppressives
Muscle relaxants
Mushrooms derivatives
Neuroleptics
Neuromuscular blockers
NSAIDs
Opioids
Radiography contrastant
Sedative reversal agents
Sympathomimetics
Vaccines…
Healthy volunteers (undetected seizure sensitivity,
unknown history of seizure)
Patients (undetected seizure sensitivity, unknown
history of seizure, specific disease history,
targeted pathology, BBB altered, poly-medication,
age (elderly or <5 years))
Brain penetration
Secondary pharmacology
|
6
SEIZURE / CONVULSION
Secondary pharmacology black list
Receptors

























Adrenergics α1A antago
Adrenergics α2A antago
Adrenergics α1B ago
Adrenergics b1 ago
Adrenergics b2 ago
Adenosine A1 antago
Adenosine A2 antago
AMPA ago
Benzodiazepine BZD antago
Cannabinoids CB1 antago
Channels Na (+)
Channels K (-)
Channels Ca (+) (L, T, N, P/Q)
CRF1 ago
Dopamine D1 ago
Dopamine D2 antago
Endothelin ETA ago
GABA A antago
GABA B ago
GABA B antago
Galanin GalR1/2 antagonist
Glutamate mGlu1/5 ago
Glutamate mGlu2/3/4/7/8 antago
Glycine (stryc sens) antago
Glycine (stryc insens) ago
























Other
Histamine H1
Histamine H2 antago
Histamine H3 ago
Kainate ago
Muscarinic M1 ago
NCX1 reverse mode (+)
NCX3 reverse mode (+)
Nicotinic a7 ago
Nicotinic a4b2 ago
NMDA ago
NPY2/5 antagonist
Oxytocin antagonist
Opioid µ ago
Opioid K antago
Opioid d ago
Opioid NOP ago
P2X7 ago
Serotonine 5-HT1A antago
Serotonine 5-HT1B ago
Serotonine 5-HT2C ago
Serotonine 5-HT3 antago
Serotonine 5-HT7 ago
Sigma 1 ago
Vasopressin V1a agonist















Acetylcholinesterase (-)
Adenosine kinase (-)
PDE3/4/7 inhibitors
Cytokine, Growth factors (+/-)
Glutamate synthetase (-)
Glutamate deshydrogenase (+)
Melatonin MT1/2 ago/antago
NO synthase inhibitor
Phosphate activated glutaminase (+)
Transporters GLU
Transporters Glutamine
Transporters GABA
Transporters DA
Transporters 5-HT
Transporters NE
strong evidence
evidence
unclear evidence
|
7
SEIZURE/CONVULSION LIABILITY
GABAA receptor patch-clamp
GABA A RECEPTOR PATCH-CLAMP (electrophysiology) is aimed at
quantitatively determinate the property and potency of the tested
compound on GABA A receptor (ligand-gated Cl- channel). Many
convulsants modulate GABA A receptor and inhibition of GABA A
transmission induces convulsions
●
Pros
●
●
●
●
●
●
●
High scientific reliability demonstrated during validations with various references
(GABA, pentylenetetrazole…)
Excellent sensitivity (>85%) and predictivity (>85%)
Functional test
High through-put screening (manual / automate)
Limited compound needs (5 mg)
Cell transfected with rat/human clone, possibly use of iPS in the future
Relative Current (%)
●
Results: agonist, antagonist, modulator, EC50, IC50
Cons/Limitations
●
●
Solubility
Many other receptors/channels implicated
100
80
60
40
20
100
0.01
0.1
1
10
Pentylenetetrazole (mM)
1.00
0.80
I/IMax
●
0.60
0.40
0.20
0.1
1
GABA (µM)
|
10
8
SEIZURE/CONVULSION LIABILITY
Brain slice electrophysiology
●
BRAIN SLICE ELECTROPHYSIOLOGY (hippocampus, cortex) is aimed at
determinate drug-induced changes in local brain electrical activity
(extracellular recording, focal EEG) using a MEA (microelectrode array)
●
●
●
●
●
●
●
Dentate
Gyrus
CA3
High scientific reliability demonstrated with various references (GABA,
pentylenetetrazole, psychostimulants…)
Excellent sensitivity (86%) and predictivity (100%)
Rapid functional test with mechanistic approaches
Limited compound needs (30 mg)
Various slices: cortex, hippocampus, possibly use of iPS layers in the future
Cons/Limitations
●
●
●
CA1
Results: spontaneous spiking (rate, seizure-like discharges…) and evoked EPSP (area
under the curve, slope, amplitude…)
Pros
●
Hippocampal slice
Microelectrode array
Solubility
Translation from focal to generalized seizure
Medium/Low through-put for screening
AUC analysis
Spiking
Spike sorting
Stages:
HTL
PreCan |
9
Mechanistic
SEIZURE/CONVULSION LIABILITY
Zebrafish convulsion assay
●
●
ZEBRAFISH (ZF) CONVULSION ASSAY is aimed at quantitative detection of
convulsion-like locomotor patterns (velocity>20 mm/s) by videotracking of ZF larvae
(Ethovision XT) and qualitative scoring of behavioral alterations using video
recording
●
Results: safe level and convulsion threshold (µM)
Pros
●
●
●
●
●
High scientific reliability for convulsants with discrimination of psychostimulants
Good sensitivity (77%) and sufficient predictivity (63%)
In vivo test (integrated model)
Medium throughput test / Limited compound needs (50mg, n=16 larvae/group)
Wide and staged range of concentrations from pharmacology to MTD (5-6 concentrations)
Stages:
HTL
LTC
PreCan
Mechanistic
Cons/Limitations
●
●
Solubility
Interpretation as regards blood or brain concentrations
Control
2.5mM
5mM
10mM
15mM
25mM
Contr
ol
2.5mM
5mM
10mM
15mM
——— Control
Total Distance Moved (mm)
High Velocity
25mM
——— PTZ, 2.5mM
900
distance moved (mm)
●
——— PTZ, 5 mM
800
——— PTZ, 10 mM
700
***
600
***
500
***
400
*** ***
***
***
***
******
300
***
200
——— PTZ, 15 mM
***
***
***
***
***
*
100
——— PTZ, 25 mM
***
**
***
***
** **
*
40-45
45-50
*
***
***
*
0
0-5
5-10
10-15
15-20
20-25
25-30
30-35
35-40
50-55
55-60
Time (min)
|
10
SEIZURE/CONVULSION LIABILITY
PTZ convulsion assay in rodents
●
PENTYLENETETRAZOLE (PTZ) CONVULSION ASSAY is aimed at mimicking a
seizure-sensitive population (Ph1 safety of any abnormal « healthy volunteers ») by
quantitative detection of any proconvulsant effects in naive rodents (before PTZ) then
in non-naive rodents brought to a sub-convulsant level (with a subconvulsant PTZ
dose).
●
●
Results: safe level and convulsion threshold i.e. dose (mg/kg) and blood (eventually brain)
level (ng/mL) with myoclonus, clonic, clonico-tonic or tonic convulsions / per se and +PTZ
effects)
Pros
●
●
●
●
●
●
Stages:
PreCan
Candidate
Preclinical
High scientific reliability for convulsants
In vivo test (integrated model)
Medium throughput test / Quite limited compound needs (100 mg, n=8/group, 3 doses)
IV route to exacerbate Cmax effects or any Clinical route / rat or mouse / other proconvulsant
agents can be used to test any other mechanism (strychnine, pilocarpine…)
Kindling protocols available
Cons/Limitations
●
PK interaction with PTZ
|
11
SEIZURE/CONVULSION liability
EEG telemetry in rats
EEG TELEMETRY in freely-moving rats (implanted with
supradural cortical electrodes) + video-recording to detect /
discriminate convulsions for quantitative detection of number /
frequency / delay of abnormal EEG morphologies (spike-wave
trains, seizures and/or convulsion…) / pre-ictal changes
●
●
Results: safe level, seizure and convulsion threshold i.e. dose
(mg/kg) and blood (eventually brain) level (ng/mL) with abnormal
EEG morphologies
Pros
●
●
●
●
●
●
Spectral analysis
with Theta Waves
in EBs period
High scientific reliability for convulsants + discrimination of other
CNS agents
Excellent sensitivity (100%) and predictivity (100%)
In vivo test (integrated model) = gold standard
IV route to exacerbate Cmax effects or any Clinical route
Kindling protocols available
Cons/Limitations
●
●
●
●
Low throughput test
Medium compound needs (300 mg, 3 dose-groups, n=6 rats/group)
Artefacts, ambiguous interpretation
Number of recording leads
EEG
EBs
Stages:
PreCan
Candidate
Preclinical
350
Number of Spike Trains (1st hour post dosing)
●
300
250
200
150
100
50
0
10
20
30
40
PTZ (mg/kg IP), n= 6 rats
|
12
SEIZURE/CONVULSION LIABILITY
Adapted FOB test in rodents
●
ADAPTED FOB ASSAY is aimed at quantitative and qualitative detection of any
convulsant effects in naive rodents that may be missed in a routine FOB
procedure due to limited duration of observation. Animals are observed
individually in an open-field for an extended period of time after administration
(1-3 hour) followed by additional kinetic observations during the first 6-8h
●
●
●
Stages:
PreCan
Candidate
Preclinical
Results: safe level and convulsion threshold i.e. dose (mg/kg) and blood (eventually
brain) level (ng/mL) with number of animals versus controls presenting tremors,
myocloni, clonic, clonico-tonic, tonic convulsion or mortality
Pros
●
●
●
●
●
●
Good scientific reliability
Improved sensitivity vs classic FOB
IV route to exacerbate Cmax effects or any Clinical route
Antidote can easily be tested
Administration schedules can be tested
Kindling protocols
Cons/Limitations
●
Medium compound needs (300-1000 mg, 3 dose-groups, n=8 rats/group)
|
13
SEIZURE/CONVULSION LIABILITY
Videorecording assay in rats
●
VIDEORECORDING is aimed at quantitative and qualitative detection of any
convulsant effects in naive rats that may have been missed in others studies.
Animals are individually videorecorded in a Phenotyper cage 24h/day (IR
camera, Ethovision XT, The Observer) with no limitation in duration. Following
death discovery, video can be rewound back to the animal death and causes,
concommittant or preceding clinical signs can be visualized and diagnosed
●
●
●
Optime code
SNX
Results: number of death and preceding causes observed (diagnostic)
Pros
●
●
●
Good scientific reliability
Diagnostic tool
Kindling protocols
Cons/Limitations
●
Low throughput test
|
14
SEIZURE LIABILITY
Strategy
Preprog
Lead
Identification
Program
Lead Optimization
• Target assessment
• Secondary
Pharmacology
(short panel)
Pre Cd
Program
Candidate
Candidate Selection
Preclinical
Development
Predicting
• Secondary
Pharmacology
(extended panel)
Derisking
Understanding
• Derisking Off-Target
pharmacology (IC50, ago/antago)
- GABA A
- Other targets
•TOX acute preliminary
rodent / non rodent
• Preliminary PK Rodent
(non rodent)
• Brain slices rodent
• Zebrafish convulsion test
• EEG rodent
• PTZ convulsion test
• TOX DRF rodent
• TOX DRF rodent
• Follow-up test (Brain slices, EEG,
Videotracking, adapted FOB,
antidote search)
|
15
Evolving models
●
In silico
●
●
●
Electrophysiology using human stem cells or iPS
●
●
●
●
patch clamp
microelectrode array
High content analysis
Bioimaging
●
●
Lack of specific in silico models predicting effects on
CNS receptors / channels linked with seizure or convulsion
Lack of integrated in silico models predicting directly
seizure or convulsion
mainly neurotoxicity
Biomarkers
●
●
results evolving from epilepsy research …
but mainly changes in biomarkers as a consequence of a seizure (not prediction)
|
16
Questions to the audience
●
Preclinical translation from preclinical to clinical
● Confidence in predictivity
● Gaps
● Improvements
●
Clinical evaluation
● Confidence on available tools
● Gaps
● Improvements
|
17