This document provides an overview of safety and toxicology testing for drug development. It discusses key concepts like toxicity, hazard, risk and regulatory guidance. A variety of preclinical safety studies are described that are required to support clinical trials, including general toxicology studies in two species, assessments of genotoxicity, carcinogenicity and reproductive toxicity. The importance of monitoring for safety findings during development is highlighted. The document also covers specialized areas like safety pharmacology and assessments of abuse potential. Overall, the document outlines the extensive non-clinical testing done to evaluate a potential drug's safety profile before human studies.

1. Studocu is not sponsored or endorsed by any college or university
Safety and toxicology - Lecture notes 18
Fundamentals of Drug Discovery and Development (Queen Mary University of London)
Studocu is not sponsored or endorsed by any college or university
Safety and toxicology - Lecture notes 18
Fundamentals of Drug Discovery and Development (Queen Mary University of London)
Downloaded by Guly Chwas (gulychwas@gmail.com)
lOMoARcPSD|31202948

2. Safety and toxicology.
LOs
- Goals of non-clinical safety testing.
- Importance of a therapeutic margin.
- Sub-disciplines of drug safety.
- Regulatory guidance.
- Studies to support clinical trials.
- Monitoring for safety findings.
- Safety pharmacology: CNS effects as safety concerns;
animal test for adverse CNS effects; new approaches to
safety assessment.
- Integrated safety assessment.
Paracelsus ‘All things are poison […] only the dose permits
something not to be poisonous.’
TOXICITY = refers to the inherent adverse effects of a
material.
Hazard = refers to the potential of an inherently adverse
material to cause damage under conditions of the proposed
use.
Risk = a measure of the probability that harm will occur under
defined conditions of exposure to a chemical.
No observed adverse effect level = NOAEL.
Lowest observed adverse effect level = LOAEL. This is the next
dose up from NOAEL.
change in behaviour or a reduction in food intake -> adverse
effect on the animal.
Dose
Hyponatraemia = intoxication after 6L water (diluted her
plasma sodium and potassium ion levels). The woman died of
cardiac arrhythmia.
Downloaded by Guly Chwas (gulychwas@gmail.com)
lOMoARcPSD|31202948

3. Botulinum toxin is the most toxic substance known to man
(lethal when given I.V. at 1-2ng/kg) yet it is used in the
cosmetic industry!
Importance of therapeutic index.
QT interval – used as a marker of drug effect and the risk of Torsades de
Pointes (TdeP).
Dofetilide is a hERG inhibitor is known to cause TdeP in man. This drug
causes harmful effects because it prolongs the QT interval.
Positive benefit to risk
All drugs are toxic at some dose.
A Positive benefit:risk means the drug has clinically relevant efficacy at
a dose that has acceptable side effects (in frequency and severity).
Level of acceptable side effects depends on:
- Disease indication.
- Degree of efficacy.
- Availability of other therapies.
Clinical efficacy is quantified and measured by placebo controlled clinical
trials.
Downloaded by Guly Chwas (gulychwas@gmail.com)
lOMoARcPSD|31202948

4. It is difficult to define whether an adverse event is treatment related.
Defining benefit:risk is often subjective; drugs can be approved in some
countries but not others.
The benefit:risk ratio depends on the disease; new therapies for cancer
have higher tolerance to side effects; this is much lower for non-life-
threatening indications like hay fever.
Regulatory guidance
ICH (international conference on harmonisation of technical
requirements for registration of pharmaceuticals for human
use) – make recommendations towards achieving greater harmonisation
in interpretation + application of technical guidelines + requirements for
pharmaceutical product registration. This reduces duplication of testing
during development of new medicines.
GLP (good laboratory practise).
 Set of principles that provides framework within which lab studies
are planned, performed, monitored, recorded, reported and
archived.
 Helps assure regulatory authorities that the data are a true
reflection of results obtained during the study and can thus be relied
upon when making risk/safety assessments.
Relevance of non-clinical species.
- Choose species that have metabolic similarity to humans.
- Exposure considerations.
- Does the drug bind to the same receptors in the non-clinical species
as it does in humans?
- Ensure species do not have an irrelevant sensitivity e.g. dog?
- Choice often between dogs + monkeys.
Drug safety is a heterogeneous discipline.
Reproductive toxicology = toxicity on reproduction, gametogenesis
through to postnatal development.
Local toxicity = toxicity at the injection site.
Special toxicology = phototoxicity, hypersensitivity, immunotoxicology.
General toxicology
= toxicology of the whole organism.
Endpoints.
Studies are conducted in two species – rodent and non-rodent
(dog/monkey).
Downloaded by Guly Chwas (gulychwas@gmail.com)
lOMoARcPSD|31202948

5. Clinical signs and physical exams: general appearance and behaviour,
locomotion, CNS (tremors/convulsions), respiratory.
Ophthalmology: pre-study and at the end of dosing.
ECG: conducted in non-rodents, to predict risk for arrhythmias.
Body weight/food consumption: quantitative indicators of chronic
toxicity. Adverse effects of a compound are detected early on by changes
to body weight/food consumption (reduced food consumption indicates
toxicity).
Clinical pathology: haematology, clinical chemistry, urinalysis.
Post-mortem investigations: necropsy, organ weight, histopathology.
target organ toxicity.
Target organ.
Toxicants don’t affect all organs to the same extent: one organ/tissue is
often more susceptible.
Which organ is most affected? This should be the target organ.
Can we measure biomarkers in the blood before we observe effects on
humans? E.g. a change in enzyme activity may indicate damage to organ.
Hepatotoxicity.
Damage to the liver is one of the most common pathologies we observe.
- This is because the liver is most important organ in detoxification
and biotransformation. It is there to protect the rest of the body.
- Liver = site for toxication by metabolism.
o First organ to encounter xenobiotic when ingested.
o Enterohepatic recirculation: potential for re-exposure to drug.
- Abnormalities of liver function:
1. Malfunction of liver cells (cirrhosis/necrosis).
2. Obstruction of biliary tract (bile duct stones).
Nephrotoxicity.
Kidneys make up 1% of body weight and receive 20% of cardiac output
flows.
The kidneys are frequently the site of toxic injury due to their central role
in filtration, metabolism and excretion of xenobiotics and/or metabolites.
Processes affected by toxicity:
- Glomerular filtration.
- Tubular re-absorption/tubular secretion.
Other examples: we also look for changes in the immune system (e.g.
bone marrow, thymus, spleen, lymph nodes), respiratory system, heart,
CNS (minor changes can have profound consequences for neurological,
behavioural and related bodily functions).
Pathologies to the eye often result in termination of the compound
(tolerance for change is very low).
Genetic toxicology.
Downloaded by Guly Chwas (gulychwas@gmail.com)
lOMoARcPSD|31202948

6. = toxicity on genetic material.
Hazard ID: screen drug candidates for potential genotoxic activity.
Risk assessment determination: follow-up mechanistic studies or
modified genetic toxicity studies as needed to assess potential for human
risk. This assists with the risk assessment.
Main purpose of genetic toxicology testing = predict carcinogenic
potential early in development as carcinogenic test results are not
available until phase II.
How can DNA be damaged?
X rays and UV -> strand breaks.
Benzopyrenes -> DNA adducts.
Radiation -> double strand break.
Bifunctional alkylating agents -> cross-linking of DNA.
Cellular processes -> oxidative damage.
Mutagenicity and clastogenicity hazards.
Mutagenicity assays detect gene level changes.
- Mutagen = heritable change in the sequence of an organism’s DNA.
Cytogenetic assays detect chromosome level changes.
- Aneugenic or clastogenic changes.
- Aneugenic = causes gain or loss of one or more whole
chromosomes from the normal chromosome number.
- Clastogen = induces chromosome damage resulting in gain, loss or
rearrangement of pieces of chromosomes.
Preclinical package of studies needed to support clinical
trials.
Non-clinical studies to support FIH. (prior to human studies)
Drug metabolism: metabolic profile and prediction of this in humans, PK,
plasma protein binding.
Safety pharmacology: core battery (cardiovascular, respiratory, CNS
effects).
Toxicology: dose ranging studies (to determine appropriate dose levels in
subsequent studies); pivotal study (GLP repeat dose) in rodent and non-
rodent.
Mutagenicity: in vitro (bacteria and mammalian cells).
Local tolerance studies?
Setting a safe starting dose in man.
Volunteers.
Downloaded by Guly Chwas (gulychwas@gmail.com)
lOMoARcPSD|31202948

7. First human dose is a fraction of the highest dose that didn’t result in
toxicity in the most sensitive animal species (NOAEL).
Factors considered in translation from animal to man: metabolic rate/body
surface area; exposure (total drug/free drug).
Traditionally applied rule: 10 (interspecific) x 10 (interindividual) = starting
dose is 100X lower than NOAEL.
Dose-response curve generated.
Phase I clinical trials.
Transition from animal to human testing.
Highly regulated and conducted with great care in specialised institutions.
Initial studies are at low doses selected on the basis of all the non-clinical
data. Doses are increased to a maximum tolerated dose.
Safe procedures, with significant adverse effects very rare.
First dose is a Non-biologically active dose (new approach to dose setting
for Phase I). (MINIMAL OBSERVED BIOLOGICAL EFFECT LEVEL).
Compound failure due to toleration/poor PK.
- Revise testing strategies to improve predictivity in humans.
If compound passes phase i…
 Chronic studies using 2 species (3-12m).
 Carcinogenicity: rat and mice (2y).
 Reproductive toxicology studies.
Toxicity study duration to support clinical trials.
Monitoring for safety findings.
Ability to risk manage safety findings are dependent on:
 Severity.
 Reversibility.
 Safety margin.
 Ability to monitor in humans.
Monitor for effect, for example BP, HR, ECG.
Biomarkers e.g. renal and hepatic changes
- Physiological biomarkers: EEG.
Inability to monitor for the effect (and no biomarkers) can result in
compound termination e.g. changes to the vasculature/vascular injury is
hard to monitor and there are no reliable biomarkers.
Downloaded by Guly Chwas (gulychwas@gmail.com)
lOMoARcPSD|31202948

8. Comparison of safety studies for standard vs cancer
therapeutics.
With cancer patients, we dose them in phase I at pharmacologically active
doses to give some indication of efficacy.
Vaccines.
- Single dose toxicity in at least one species which provides an
adequate safety margin in humans. If toxicity is seen, a
dose/response should be conducted.
- Vaccines requiring multiple doses in humans: repeat dose toxicity in
one species. Incorporate safety pharmacology endpoints in these
studies.
- Reproductive toxicology studies - Effects on fertility not
normally required… embryo/foetal toxicity isn’t normally required
unless intended to be used in women of childbearing age.
Safety pharmacology.
= studies that investigate potential undesirable pharmacodynamic effects
of a substance on physiological functions, related to exposure in the
therapeutic range and above.
There are primary pharmacodynamic, secondary pharmacodynamic and
safety pharmacology studies.
Objectives of safety pharmacology studies:
- Select the best drug target and candidate compounds.
- Prevent serious ADRs.
- Understand concentration-response relationship for any effects on
major physiological systems that may predict adverse events in
humans: primary and secondary pharmacology.
- Safety pharmacology should be used to select the starting dose.
- Generate ‘plausible hypothesis’ to support clinical development.
Downloaded by Guly Chwas (gulychwas@gmail.com)
lOMoARcPSD|31202948

9. - Provide mechanistic understanding of adverse events in clinical
trials – supporting regulatory approval as we can predict which
patients may have AEs and should avoid particular treatment.
Core studies.
CV, pulmonary, CNS, ancillary profiling against a whole range of
compounds (broad binding profile, range of in vitro techniques).
Blood pressure measurements in animals.
Telemetered rat, dog and monkey.
Telemetry – animals operated on by surgeons under controlled anaesthetic
conditions. Can implant transmitters that measure BP, HR, ECG and
remotely transmit this data to receivers. Once surgery is completed and
animal recovered, we can now measure BP over a time frame.
 Robust continuous data.
 Very sensitive marker.
 Surgical intervention.
JET (jacket external telemetry).
 Less invasive – train animal to wear a jacket and put ECG electrodes
on the animal connected to the jacket. The jacket transmits the ECG
data.
 Superficial surgery to collect BP data (transplant BP transmitter).
 Add-on to general toxicology studies.
Tail cuff in toxicology studies
 High variability + low sensitivity (not often used).
Blood pressure is a biomarker for risk…
Small decrease in BP isn’t too much of a worry. Rise in BP is a huge
concern – 5mmHg higher systolic BP is associated with 12-14% increase in
risk of coronary heart disease events and 19-23% increase in risk of
subsequent stroke.
AP prolongation equates to QT prolongation.
QT interval = a measure of how long it takes for electrical activity of the
heart to return to normal, ready for the next contraction.
hERG K+ channel – blocking this channel means K+ leaves the cell more
slowly, resulting in prolongation of the AP (longer QT interval). Electrical
activity in the whole heart is now slowed.
QT prolongation can induce torsade de pointe (TdP). The contraction of
ventricles is no longer coordinated in TdP. This returns back to normal in
most patients. But TdP can also -> ventricular fibrillation + DEATH.
Therefore, hERG K+ blockage can be fatal.
Abuse potential.
- Potential risk for CNS adverse events.
- Dose the drug have positive reinforcing properties? Does cessation
of chronic dosing lead to negative symptoms?
Downloaded by Guly Chwas (gulychwas@gmail.com)
lOMoARcPSD|31202948

10. - Rodent I.V. self-administration models used to assess the reinforcing
properties of the test drug. These models have good concordance
with self-administration (and drug-liking) in humans.
- Dopamine agonists are associated with abuse potential; but rat self-
administration model (cocaine comparisons made) gave increased
confidence that project could progress w/out abuse potential risk.
Drug discrimination.
Assessment of “subjective effects” of a compound.
- Drug-induced interoceptive cues. Does the animal perceive the drug
to be like another one (morphine or cocaine)? Indicates potential for
humans to have this same subjective effect.
- Subjective effects might reinforce drug taking.
- Perceived drug effect associated with one lever; absence of drug
effect associated with opposite lever (lever selection model).
- Drug discrimination doesn’t give direct assessment of abuse but
gives confidence to support abuse potential, and evidence for
similar pharmacology of the test compound with drugs of abuse
(morphine/cocaine).
Drug discrimination methods.
Physical dependence and withdrawal.
= tendency of an active substance to give rise to a need for repeated
doses of the active substance to avoid feeling bad and maintain feeling
good.
Complex disorder – cognitive, behavioural and physiological symptoms.
Morphine administered to rats via Alzet osmotic mini pump. Steady-state
plasma concentration maintained over 14 days. Withdrawal associated
with lower body weight, body temperature and reduced feeding.
Downloaded by Guly Chwas (gulychwas@gmail.com)
lOMoARcPSD|31202948

11. Seizure assessment.
Seizures have low tolerance – very concerning and difficult to measure.
Zebrafish are a valuable animal model for frontloading safety assessment.
Put into 96 well plate, swimming activity measured.
Convulsant compounds cause:
- Increased swimming speed.
- Rapid circling.
These behaviours are associated with abnormal ictical-like electrographic
discharges. The effects can be blocked by antiepileptic drugs.
Moderate throughput zebrafish screen for evaluation of convulsant
liability.
Mouse or rat in vitro brain slice.
 Evoked population spike represents summated neuronal firing.
 Stimulating electrode in one part of the brain slice, and a recording
electrode in a more distant neuron -> electrical activity passes
through the brain.
 Convulsant drugs induce excitation by variety of mechanisms. This
excitation is similar to a convulsion.
Functional observation battery.
Range of observations can be made in animals – autonomic,
neuromuscular, behavioural and sensorimotor. Can measure endpoints
such as those beneath the four headings.
Locomotor assessment.
Two common paradigms:
- Non-acclimated acute open-field activity,
Downloaded by Guly Chwas (gulychwas@gmail.com)
lOMoARcPSD|31202948

12. - Acclimation for 30min prior to open-field activity.
- Subjects returned to cages after dosing. Each rat in individual open
field chamber. 3d matrix of infrared beams or video tracking; photo
beam breaks or video tracking used to quantify movement.
Horizontal activity, distance travelled, line crosses measured over
30-100min.
Pros: easy to use; photocell systems provide quantifiable measurements.
Cons: start-up expense; endpoints not isomorphic to humans.
Rotarod test.
Used to measure the coordination of animals.
Untreated rats evaluated for ability to remain on the rotarod for 60s as
speed of rotation increases.
Latency to fall off rotating rod is measured and indicates motor
coordination + balance.
Generating an integrated safety assessment.
Effect of sildenafil (Viagra) on dog ERG measured.
This drug inhibits PDE5 enzyme,
Observed that some patients reported visual adverse events due to
inhibition of PDE6 enzyme (off-target effect). Objects looked brighter when
they weren’t taking the drug for example. Higher dose -> more frequent
adverse events. Alpha and gamma subunits for PDE6 have been deleted
-> blindness and damage to the eye.
Sildenafil exhibited similar potency across all species – x10-fold selective
for PDE5 enzyme.
Toxicology studies: dogs and rats treated w/high dose of viagra once daily
for 1/2y respectively -> no effects in clinical examinations, no pathological
changes in eye/optic tract.
Downloaded by Guly Chwas (gulychwas@gmail.com)
lOMoARcPSD|31202948

13. Downloaded by Guly Chwas (gulychwas@gmail.com)
lOMoARcPSD|31202948

14. Conclusions.
Safety and toxicology are a complex multi-disciplinary science.
Principles are based on hazard identification and risk management
(severity, reversibility, dose/safety margin, patient population) and
alternative therapies.
Downloaded by Guly Chwas (gulychwas@gmail.com)
lOMoARcPSD|31202948