Biomedical Engineering Reference
In-Depth Information
cannot be reliably identified or assessed in animals
(e.g. headache, blurred vision, tinnitus). However, studies
attempting to address concordance have found species
differences for predicting the effects of drugs in humans.
For small molecule evaluation, both nonhuman primates
and dogs are better predictors of human toxicity than
rodents (
Olson et al., 2000
), with nonhuman primates and
dogs being roughly equivalent. The combination of rodent
and either a dog or nonhuman primate provides the best
concordance, supporting the two-species guidance. It
should be noted that no species or nonclinical toxicity test
is adequate for predicting idiosyncratic toxicity. This is not
surprising since by definition not even humans are adequate
predictors of idiosyncratic toxicity.
Considerations for large molecule (i.e. bio-
pharmaceuticals which include proteins, peptides and
nucleic acids) test species selection differ from those for
small molecules. Toxicity is expected to be an extension of
pharmacological activity, thus the species selected for
nonclinical studies should be pharmacologically relevant.
Receptor or epitope homology to humans occurs more
often in nonhuman primates than in dogs or rodents. Since
large molecules are metabolized to naturally occurring
endogenous substances, it is anticipated that these break-
down products will be well tolerated. An exception may be
if an unknown receptor binding site is exposed on the
catabolized parent molecule. In addition to determining the
drug-induced effects, initial concerns for large molecule
safety/toxicity assessment were targeted towards contami-
nants from production, since large molecules are produced
by genetically manipulated mammalian cells, yeast, or
Escherichia coli in complicated production systems.
Production processes are tightly controlled and monitored,
which has diminished concerns of
easy approach; as it is, production and characterization of
a surrogate is expensive and time consuming. Use of
surrogate molecules and transgenic models may also be
necessary when no pharmacologically relevant species is
identified.
The nonhuman primate species of choice for both safety
assessments of small and large molecules are cynomolgus
macaques, and less commonly rhesus monkeys, with
selected use of marmosets and other nonhuman primate
species. Selection of these species is based on availability,
size, historical usage, appropriateness for the study design
and type of drug, and access to databases. The fidelity of
detection of drug-induced effects positively correlates with
reduction of variability, thus the animal model and test
conditions are tightly controlled to allow intra-animal and
inter-animal effects to be discerned. Source of nonhuman
primates is an important variable to control. The genetics
and phenotypes will vary dependent on country of origin;
Cambodia, China, Indonesia, Mauritius, Philippines, and
Vietnam are common sources for cynomolgus and rhesus
monkeys. It is important within and between studies to use
one species, and in some circumstances to be aware of
animal source in relation to geographical origin. Relative
genetic hetero/homogeneity in typical laboratory nonhuman
primate species, including differences between and within
source groups, and the role genetic factors play in drug
response are only beginning to be defined and are an area of
growing focus in drug development science. Control of the
environment, housing, food, procedures and health are also
critical to control to reduce variability. Nonhuman primates
for use in nonclinical safety studies are generally purpose
bred, but it isn't the law in the USA, and there are still
imports of wild-caught animals, especially Mauritian origin
monkeys.
Nonhuman primates can vary according to country of
origin in their genetic makeup, erythrogram, leukogram,
clinical chemistry profile, disease carrier status, commensal
microflora, and background findings in gross and micro-
scopic evaluation of tissues (Covance, 2010, internal
communication). Monkeys carrying infectious disease
organisms can not only affect a study, but also the health of
the colony (
Sasseville and Mansfield, 2010
).
Drug safety evaluations are generally done only in
healthy animals; however, consideration is occasionally
given to evaluating safety/toxicity effects in animal
models of human disease for specific situations. There are
potential advantages and disadvantages in using disease
models for determining safety (
Dixit and Boelsterli, 2007;
Cavagnaro, 2008
). One disadvantage is the lack of
historical data to assist in determining if a difference from
concurrent control is a drug-induced effect. No animal
model completely predicts toxicity in humans due to
a variety of reasons, including the differences between
human and nonhuman primate already listed. Additionally,
toxicity due to
contaminants.
For large molecules, occasionally chimpanzees or
baboons are the only pharmacologically relevant species.
However, these species are very infrequently used for drug
development safety/toxicity studies for ethical and legal
reasons driven by both moral and animal welfare concerns
(see Chapter 2). ICH S6, which addresses testing of
biotechnology products, specifically does not require use
of the most pharmacologically relevant species since this
may be the chimpanzee or baboon (
International Confer-
ence on Harmonisation, 1997
). Alternate strategies for
safety assessment (e.g. utilization of surrogate molecules
or transgenic animals) must be devised in these cases.
A surrogate molecule is a homologous molecule (i.e. rat
protein for use in a rat) or, for a monoclonal antibody, an
antibody that cross-reacts in a preclinical species to phar-
macologically mimic the parent molecule. While the drug
candidate itself is not tested (which is considered disad-
vantageous), the effects of pharmacological activity and
potential cross-reactivity can be evaluated. This is not an
