Evaluation of Biotechnology-Derived Pharmaceuticals*

fettlepluckyBiotechnology

Dec 1, 2012 (4 years and 6 months ago)

300 views


http://tpx.sagepub.com/
Toxicologic Pathology
http://tpx.sagepub.com/content/27/6/678
The online version of this article can be found at:

DOI: 10.1177/019262339902700610
1999 27: 678Toxicol Pathol
Andrew M. Pilling
Pharmaceuticals
The Role of the Toxicologic Pathologist in the Preclinical Safety Evaluation of Biotechnology-Derived


Published by:

http://www.sagepublications.com
On behalf of:



Society of Toxicologic Pathology
can be found at:Toxicologic PathologyAdditional services and information for





http://tpx.sagepub.com/cgi/alertsEmail Alerts:


http://tpx.sagepub.com/subscriptionsSubscriptions:


http://www.sagepub.com/journalsReprints.navReprints:


http://www.sagepub.com/journalsPermissions.navPermissions:


http://tpx.sagepub.com/content/27/6/678.refs.htmlCitations:

What is This?

- Nov 1, 1999Version of Record >>
by guest on November 30, 2012tpx.sagepub.comDownloaded from
678
The
Role
of
the
Toxicologic
Pathologist
in the Preclinical
Safety
Evaluation of
Biotechnology-Derived
Pharmaceuticals*
ANDREW
M.
PILLING
Molecular
Pathology Group,
Medicines
Safety
Evaluation
Division,
Glaxo Wellcome Research and
Development
Ltd.,
Park
Road,
Ware,
Hertfordshire,
United
Kingdom
*
Address
correspondence
to:
Andrew
M.
Pilling,
Molecular
Pathol-
ogy
Group,
Medicines
Safety
Evaluation
Division,
Glaxo
Wellcome
Re-
search and
Development
Ltd.,
Park
Road,
Ware,
Herts
SG12
ODP,
Unit-
ed
Kingdom;
e-mail:
amp 1069 @ ggr.co.uk.
ABSTRACT
Biotechnology-derived pharmaceuticals,
or
biopharmaceuticals,
represent
a
special
class
of
complex, high-molecular
weight prod-
ucts,
such
as
monoclonal
antibodies,
recombinant
proteins,
and nucleic acids.
With
these
compounds,
it
is
not
appropriate
to
follow
conventional
safety
testing
programs,
and
the
preclinical "package"
for each
biopharmaceutical
needs
to
be
individually designed.
In
addition
to
standard
histopathology,
the
use
of molecular
pathology techniques
is
often
required
either in
conventional animal studies
or
in
in vitro
tests.
In
this
review,
the
safety
evaluation
of
biopharmaceuticals
is
discussed
from the
perspective
of the
toxicologic
pathologist,
and
appropriate
examples
are
given
of
the
use
of
molecular
pathology procedures. Examples
include
the
use
of in situ
hybridization
to
localize
gene
therapy
vectors,
the
assessment
of
vector
integration
into
genomic
DNA
by
the
polymerase
chain
reaction
(PCR),
and
the
use
of
immunohistochemistry
to
evaluate
the
potential cross-reactivity
of
monoclonal
antibodies.
In situ
PCR
techniques
may
allow for
confirmation of
the
germ
cell
localization
of nucleic
acids
and
may
therefore
facilitate the risk
assessment
of
germline
transmission. Increased involvement
with
biopharmaceuticals
will
present
challenging
opportunities
for
the
toxicologic
pathologist
and will allow for much
greater
use
of molecular
techniques,
which
have
a
critical
role
in
the
preclinical
development
of
these
compounds.
Keywords.
Molecular
pathology;
monoclonal
antibodies;
recombinant
proteins;
gene
therapies;
nucleic
acids
INTRODUCTION
In
recent
years,
pharmaceutical
and
biotechnology
companies
have made
a
considerable
investment in the
research and
development
of
products
derived from
bio-
technology
processes
(26).
This has
been facilitated
by
advances
in
recombinant
DNA
technology
and
driven
by
the
possibility
of
producing
novel
therapeutic
strategies
to
treat,
or even
cure,
conditions such
as
cancer
and in-
herited diseases.
Biotechnology-derived pharmaceuticals,
or
biopharmaceuticals,
are
generally high-molecular
weight products
and include
compounds
such
as
mono-
clonal
antibodies
(mAbs),
recombinant
proteins,
and
nu-
cleic
acids
(Table I)
(11).
The
objectives
of the
preclinical
safety
program
for
biopharmaceuticals
are
similar
to
those for
conventional
small
molecules:
to
recognize potential
toxicities,
to
iden-
tify
appropriate
parameters
for clinical
monitoring,
and
to
contribute
to
the
setting
of
human
dosage
(33).
How-
ever,
important
differences
in
study design
exist for bio-
technology products.
First,
the
route
and
frequency
of
dosing
in animal
studies
more
closely parallels
the
pro-
posed
treatment
regime
for
humans,
and
second,
toxicol-
ogy
studies
are
mostly
carried
out
in
biologically
and
pharmacologically
relevant
(responsive)
animal
species.
A
relevant
species
is
one
in
which the
test
material
is
pharmacologically
active,
as
it
will,
for
example,
possess
the
appropriate epitope
(for mAbs)
or
receptor
(for
pro-
teins)
or
will
respond
to
the
promoter
sequence
(for
gene
therapies).
The
use
of
a
relevant
species
allows
for the
evaluation of
any
&dquo;downstream&dquo;
toxicity
that results
from
the
pharmacological
action of
the
drug.
In
some
instances,
nonresponsive species
may
be used
(e.g.,
when
no
relevant
species
is
available
or
when
one
is
assessing
toxicity
that
results from
contaminants).
Because of the
different
natures
of
biopharmaceuticals,
it is
not
appropriate
to
follow
the
pharmaceutical safety
testing
program
traditionally
used for small
molecules,
and the
preclinical
&dquo;package&dquo;
for each
biopharmaceutical
needs
to
be
individually
designed.
(26).
It
is
customary
with
biotechnology products
to
perform
short-
or
medi-
um-term
toxicology
studies in
one
species,
with standard
histopathologic
evaluation.
The
safety
evaluation of bio-
pharmaceuticals
often
requires
the
use
of
specialized
mo-
lecular
pathology techniques,
either in conventional
ani-
mal studies
or
in in
vitro
tests.
Morphologic techniques
that
fall into this
category,
and for which the
role
of
the
toxicologic pathologist
is
paramount,
include
immuno-
histochemistry
(IHC), in
situ
hybridization
(ISH),
and in
situ-polymerase
chain reaction
(IS-PCR)
(25).
The
pa-
thologist
should also
have
a
good
understanding
of the
by guest on November 30, 2012tpx.sagepub.comDownloaded from
679
TABLE
I.-Examples
of
biopharmaceuticals
with
potential
clinical in-
dication(s).
other tools
used
to
support
biotechnology
compounds,
such
as
solution-phase
PCR
and
flow
cytometry.
In this
paper,
the
ways
in
which the
toxicologic
pa-
thologist
may
contribute
to
the
safety
evaluation of
bio-
pharmaceuticals
are
reviewed.
Although
aspects
of
con-
ventional
histopathology
are
addressed,
the
emphasis
is
placed
on
molecular
pathology
techniques.
It is
assumed
that the reader
possesses
a
basic
understanding
of the
principles
of
these
techniques,
so
that their
applications
can
be
fully appreciated.
MONOCLONAL ANTIBODIES
mAb
Technology
was
developed by
Kohler and Mil-
stein
over
20
yr
ago
(36).
Initial
progress
was
slow,
but
within the
last
decade,
significant
advances have been
made
in the
development
of antibodies for the
diagnosis
and
treatment
of human diseases. Mouse-human chime-
ric
or
fully
human recombinant antibodies
are now
avail-
able,
and these antibodies
are
less
immunogenic
and less
toxic in
humans when
compared
with rodent mAbs
(44,
58).
Clinically,
mAbs
can
be used
to
block
a
particular
target
antigen, thereby preventing
&dquo;normal&dquo; interactions
with
other cellular
or
tissue
components.
Alternatively,
they
may
be used
to
target
a
cell
population
for destruc-
tion
via
antibody-dependent
cellular
cytotoxicity
(ADCC)
or
complement-mediated cytolysis
(CMC) (54).
These
destructive
properties
may
be enhanced
by
conjugating
the
mAb
with
a
cytotoxic
agent
(site-directed
chemo-
therapy)
or
toxin
(e.g.,
pseudomonas
toxin for
the
treat-
ment
of
interleukin-expressing
tumors).
Diagnostic
ap-
plications
include localization
of
tumor
metastases
with
radioisotope-labeled
mAbs
(56).
The
toxicologic
pathologist
may
become involved
in
the
preclinical
development
of mAbs
1)
during
in
vitro
or
in vivo
cross-reactivity
tests
and
2)
in the
more
con-
ventional
animal
toxicology
studies.
TABLE
II.-Normal human tissues used
in
cross-reactivity
studies.
In Vitro
Cross-Reactivity
Tests
In vitro tissue
cross-reactivity
screens
play a pivotal
role in the
safety
evaluation
of mAbs
(26).
Although
guidance
on
conducting cross-reactivity
studies is
avail-
able from the U.S. Food
and
Drug
Administration
(FDA)
(20),
this
is
not
intended
to
be
prescriptive,
and much
is
left
to
the discretion of the
investigator.
For
a
test
mAb,
a
panel
of
human tissues is examined
routinely using
IHC
in order
to
ensure
that
specific
Fab
binding
to
the
target
epitope
is
taking place.
A
humanized
test
mAb
must
be
appropriately
labeled
(e.g.,
with
fluorescein
isothiocya-
nate)
so
that it
can
be
detected
in human tissue. A
com-
parison
of
in vitro
cross-reactivity
in tissues from differ-
ent
species
is
also
performed
in
order
to
determine the
most
relevant animal
to
use
for future
toxicology
studies.
Relevant animal
species
are
those
that
express
the
desired
epitope
and that demonstrate
a
cross-reactivity profile
similar
to
that
seen
in human
tissues.
These
comparative
studies
may
involve
tissues from
rodents,
nonhuman
pri-
mates,
and,
occasionally,
fetuses.
For
the human
screen,
it is recommended that
a
large
number
of normal tissues be examined
(Table
II)
and that
specimens
be
acquired
from 3 unrelated donors in order
to
allow
for
any
polymorphisms
that could
account
for
atypical binding
responses
(20).
However,
there
are
dif-
ferences of
opinion
with
regard
to
the
extent
of animal
tissue
screening
necessary:
i.e.,
whether it should
be lim-
ited
to
those
tissues
identified
in the human
panel
or
whether
a
comprehensive
selection of animal tissue is
required
for
adequate comparison
(26).
In
our
laboratory,
the decision has
always
been made
to
examine
a
similar
tissue
list
in animals and humans.
The
procurement
of human tissues is
usually problem-
atic,
as
these tissues
must
be
acquired
with
due
respect
to
the
donor,
the
family,
and
the
law
(7).
Where human
tissues
of unknown
origin
(no
donor
information)
are
used,
it
may
not
be
possible
to
claim full
compliance
with
Good
Laboratory
Practice
(GLP).
In such
cases,
it is
nec-
essary
to
demonstrate that the
study
was
performed
in
compliance
with the
&dquo;spirit
of
GLP,&dquo;
so
that data
gen-
erated
can
be used
to
support
registration
(33).
Frozen
(nonfixed)
surgical
material should
be used
wherever
possible
in order
to
ensure
that
the
antigenic integrity
of
the
tissue is
maintained.
The donors
should be selected
by guest on November 30, 2012tpx.sagepub.comDownloaded from
680
in accordance
with
the
particular
mAb
to
be tested
(e.g.,
if
a
mAb
specific
for
a
lymphocyte
CD
epitope
is
to
be
evaluated,
then it is
prudent
to
avoid
using
tissues
from
immunosuppressed
individuals).
In
order
to
confirm
ac-
ceptable
antigen
preservation
in the
acquired
tissue,
IHC
may
be
performed using
specified
antibodies
that
are
di-
rected
against
appropriate
structural
or
functional
pro-
teins. These
proteins
could be cell surface
markers,
in-
termediate
filaments,
or
secretory
proteins.
The
production
of
relatively large
numbers of
frozen
sections
with
good morphology
can
be
technically
de-
manding, particularly
in the
case
of
small
organs,
such
as
pituitary
and
adrenal
glands.
This is
particularly
true
when
some
of the human tissues
are
postmortem-derived
and
are
of
suboptimal
quality.
It
is advisable
to cut
all
the
sections
in
advance
and
to
be
assured of their
quality
before
proceeding
to
the
cross-reactivity
phase.
In addi-
tion,
the
stability
of
the
target
epitope
should
be
consid-
ered
(e.g.,
the
epitope
could be
damaged by
certain fix-
atives
or
oxidized
by hydrogen peroxide,
if
the
latter is
used in
the
detection
system).
Selecting
the
appropriate
mAb concentrations
to
test
can
be
an
area
of
concern.
FDA
guidelines
state
that &dquo;an-
tibody
affinities
as
well
as
expected
achievable
peak plas-
ma
concentrations
should be
considered when
choosing
the
proper
concentrations for tissue
binding
studies&dquo;
(20).
However,
in
practice,
these
data
are
difficult
to
apply,
and,
in
our
experience,
they
lead
to
the
selection of mAb
concentrations that
are
either
too
high
or
too
low for the
IHC
assay.
Therefore,
we
usually
determine
the
appro-
priate
concentrations
empirically
by
applying
a
range
of
mAb dilutions
to
a
known
positive
tissue
(containing
tar-
get
antigen)
and
negative
control tissue and
by selecting
1
high
and 1 low concentration for
use
in the
screening
process.
It
should be noted
that several other
components
of the
IHC
procedure
will influence
the
signal intensity,
including
the
duration of
incubation with the
test
anti-
body
and the dilutions
and
durations of
application
of
secondary
antibodies
and other
reagents.
These
compo-
nents
also
need
to
be established
empirically using
the
positive
and
negative
control
tissues.
The choice
of
IHC
detection
system
(i.e.,
secondary
antibody,
reporter
molecule,
enzyme,
chromogen,
etc.)
is
at
the discretion of the
pathologist.
Many
of these
sys-
tems
are
currently
in
use
for
experimental
and
diagnostic
purposes,
and
most
are
commercially
available. The
sen-
sitivity
of such
systems
can
vary
considerably,
but those
that
incorporate
biotin
and
streptavidin
are
generally
re-
garded
to
be the
most
sensitive.
It
is
reported
that the
avidin-biotin
complex
(ABC)
technique
provides
supe-
rior
sensitivity
to
the
peroxidase-antiperoxidase
method
(30).
The situation has
recently
become
more
complicat-
ed
with
the
introduction of
methods
involving biotiny-
lated
tyramine
for
signal
amplification
(1).
In
theory,
the
most
sensitive detection
system
should be
employed
for
the
cross-reactivity
screen,
unless
&dquo;good&dquo;
scientific
rea-
sons
can
be
advanced
for
using
less-sensitive
procedures.
Whichever
system
is
selected,
it should be
applied
con-
sistently
to
all
panel
tissues,
although
modifications will
be
necessary
in
organs
with
high endogenous
levels of
biotin,
peroxidase,
or
alkaline
phosphatase.
It is
not
ad-
FIG.
1.-A)
Antibody
panel
used in
a
cross-reactivity study
to
deter-
mine the
specificity
of
binding following
unexpected staining
with
a
test
mAb.
Staining
patterns
consistent with
(B)
Fab
binding
and
(C)
Fc
binding
are
shown.
Nonspecific protein
interactions
(between
mAb and
tissue)
may
generate
various
staining
patterns
that
do
not
conform
to
the above.
visable
to
use
chromogens
that
are
unsuitable for
per-
manent
mounting,
as
this
can cause
problems
during
mi-
croscopy
(particularly
if
peer
review
is
required)
and
be-
cause
stained sections should
be stored
long-term.
In
practice,
we
first
apply
the
test
mAb
to
the
tissue
panel,
with
the
antibody
diluent
being
run
alongside
as
a
control.
The latter
allows
any
staining
that is attributable
to
the
detection
system
to
be
assessed. Positive
and
neg-
ative
control tissues
are
included in the
same run
in order
to
demonstrate
the
adequacy
of the
assay.
Second,
the
specificity
of
any
staining (binding)
that is attributable
to
the
test
antibody
is
investigated.
Several
possibilities
ex-
ist for
the
investigation:
increasing
the
salt concentration
to
disrupt nonspecific
bond formation between the
test
mAb
and
tissue;
using
labeled
versus
unlabeled mAbs in
competitive
binding
assays;
or
using
competition
(inhi-
bition)
assays
with
purified
antigen.
An additional
way
of
assessing specificity,
one
used in
our
laboratory,
is
to
employ
a
panel
of control antibodies and
Fab,
fragments
.
(Fig.
1).
The controls
are
of the
same
immunoglobulin
class
as
the
test
mAb,
but
they
are
not
raised
against
a
mammalian
epitope,
so
they
are
unlikely
to
bind
to
the
.
tissues. This allows Fab
binding
with
the
test
mAb
to
be
.
distinguished
from
Fc
binding
or
nonspecific protein :pro-
f
tein
interaction with
the
tissues. Fab
binding
is confirmed
.
when
a
similar
staining
pattern
is
seen
with
both
the
test
by guest on November 30, 2012tpx.sagepub.comDownloaded from
681
mAb
and
Fab,
fragments generated
from the
test
mAb
combined
with the simultaneous
absence of
staining
with
the control mAb and control
Fab, fragments. Ideally,
any
Fab
binding
identified
by
the
panel
should be
further
in-
vestigated
(e.g.,
with
antigen competition)
in order
to
confirm
that it is attributable
to
the interaction with
the
complementarity determining region,
since
this is where
the
specificity
resides. The
latter
assessment
is
important,
as
cross-reactivity
with
a
non-target
tissue
involving
this
part
of the molecule
may
be
predictive
for in
vivo
tox-
icity.
Increasing
numbers of
tissue
cross-reactivity
screens
using
IHC
are now
being performed
to
support
the
pre-
clinical
development
of mAbs.
Although
IHC is
regarded
as
the
gold
standard
procedure
for the
identification of
cross-reactivity
in
tissues,
it has limitations.
IHC
is
a
morphologic technique,
and
as
is the
case
with
other
his-
tologic
procedures,
its main
purpose,
localization,
is
achieved
at
the
expense
of
sensitivity.
In
the author’s
ex-
perience,
IHC is
less
sensitive
for the
detection of tissue
proteins (epitopes)
than
are
other
procedures,
such
as
flow
cytometry
and
western
blotting,
both of which
are
performed
in solution
phase following
dissociation of the
tissue.
Therefore,
when
one
is
using
IHC,
the lack
of
labeling
of
a
particular antigen
does
not
always
indicate
the
complete
absence of that
antigen
in the
tissue.
For
instance,
the nonlinear
nature
of
IHC has been
demon-
strated
by
the
finding
that,
under standard
conditions,
the
induction of
a
protein (e.g.,
p53)
in
a
tissue
can
appear
as an
absolutely &dquo;all-or-nothing&dquo; phenomenon;
yet,
when
careful
quantitation
is
applied,
the
actual level of induc-
tion of the
protein
is
4-
to
5-fold
(38).
However,
despite
these
limitations,
IHC
is the
only
practicable
way
of
rap-
idly screening
a
wide
range
of tissues
for
cross-reactivity.
Flow
cytometry
is also
routinely
used
to
screen
for
cross-
reactivity
with blood
cells,
and
this
method
accurately
identifies
and
enumerates
those cells
to
which the Fab
region
can
bind.
In
Vivo
Cross-Reactivity
Tests
When
cross-reactivity
of
a
mAb is demonstrated with
non-target
human
tissues,
then
an
in
vivo
investigation
in
a
relevant animal model
may
be
performed.
This
usually
involves
the
administration
of the
test
mAb
to
the
animal
model and the
performance
of IHC
on
the relevant
tissues
for
confirmation of
cross-reactivity.
Animal
Toxicology
Studies
Preclinical
safety testing
of mAbs is carried
out
in
rel-
evant
species,
the choice of
species usually being
made
following
the
in
vitro
studies
(see above).
Knowledge
of
receptor/epitope
distribution
within
the
test
species
may
be
helpful
in
predicting
in vivo
toxicity.
Safety
concerns
with mAbs include their
immunoge-
nicity,
tissue
cross-reactivity,
and effector
function(s).
A
short,
repeat-dose toxicity
assessment,
with
standard
his-
topathologic
evaluation
of
tissues,
is
usually
appropriate
for
mAbs
(26).
Cross-reactivity,
particularly
with
mAbs
with
CMC
or
ADCC
activity,
should
lead
to
more
exten-
sive
preclinical
testing, including
studies
in
more
than
1
animal
species
over
a
range
of
doses
as
well
as
repeat-
dose animal
studies
(20).
It
is
recognized
that animal
models
expressing
the
antigen
of
interest
or
a
closely
re-
lated
highly
cross-reactive
epitope
are
not
always
avail-
able for these
studies.
Therefore,
in
some
cases,
xenograft
or
transgenic
models that
express
the relevant
antigen
may
be used
to
generate
safety
data
(20).
However,
the
use
of these
specialized
models for
safety
evaluation
may
cause
interpretive problems
because of the lack of
famil-
iarity
with
background pathology.
The
use
of homolo-
gous
antibodies
may
also be considered
in the
absence
of
a
relevant
species
or
when neutralization of
activity
is
a
problem.
In
these
situations,
inference of
safety
to
hu-
mans
must
be made
cautiously,
as
biodistribution,
activ-
ity,
and
clearance
may
not
be
analogous
to
the human
mAb
(26).
RECOMBINANT
PROTEINS
Recombinant
proteins
were
the first
biotechnology
compounds
to
be
developed
in the
early
1980s,
with the
production
of
a
number
of
hormones,
including
human
growth
hormone
and
human insulin
(27).
Since
then,
re-
combinant
proteins
have become
increasingly important,
and
numerous
products,
which
have
been derived from
20
different
proteins,
are
already
on
the market
(5).
Re-
combinant
proteins
now
represent
the
largest
group
of
pharmaceutical
products
to
be derived from biotechnol-
ogy
processes
(33).
For
several
years,
the
term
biophar-
maceuticals
seemed
to
be
synonymous
with the
term
re-
combinant
proteins,
but with the
development
of
a
wider
range
of
biotechnology
therapies,
such
terminology
is
no
longer
appropriate.
Recombinant
proteins
are
derived
from various mammalian and
nonmammalian cells
using
a
variety
of
expression
systems
(33).
Since
recombinant
proteins
are
designed
to
be similar
to
the
native human
form,
it
was
initially
assumed that
they
would be well tolerated.
Although
this
assumption
is
correct
for
some
proteins,
others have
proved
to
be
highly
toxic,
and therefore careful
preclinical
assessment
is
required
(47).
The
unanticipated
toxicity
may
be,
in
part,
the
result of the
systemic
administration
of
high
dos-
es
of
factors,
which,
under normal
circumstances,
are
lo-
cally
concentrated in
tissues. The situation is
compound-
ed
by
the
poorly
understood
physiological
roles
of
some
cytokines.
Recombinant
human interferons
and interleukins
act
as
intersignaling
molecules,
which
are
responsible
for mul-
tidirectional communication
among
immune and inflam-
matory
cells.
Biological activity
is
produced through
re-
ceptor-mediated
processes,
and
toxicity
is
generally
con-
sidered
to
be
directly
or
secondarily
related
to
exagger-
ated
pharmacological
effects. It is therefore essential that
these
molecules
are
pharmacologically
active in the
spe-
cies
used
to
assess
toxicity
(26).
As
in the
case
of
mAbs,
it
may
not
be
possible
to
carry
out
preclinical testing
in
a
relevant
species,
in which
case
the
use
of
transgenic
models
or
homologous species
proteins
may
be
consid-
ered.
In
general,
the
use
of
molecular
pathology
techniques
is
not
as
extensive in the
safety
evaluation
of
recombinant
proteins.
Therefore,
in
these
instances,
the role
of
the
toxicologic
pathologist
is
a
more
conventional
one
and
is
by guest on November 30, 2012tpx.sagepub.comDownloaded from
682
largely
restricted
to
the
performance
of
standard
patho-
logical
procedures.
GENE THERAPY
Gene
therapy
is
a
medical intervention based
on
the
deliberate modification of
the
genetic
material of somatic
cells. Cells
may
be
genetically
modified
ex
vivo
for sub-
sequent
administration
to
patients
(a
procedure
termed
somatic
cell
therapy)
or
they
may
be altered in
vivo
by
gene
delivery
(21).
Somatic
cell
therapy
may
involve the
administration of
autologous,
allogeneic,
or
xenogeneic
living
cells,
whereas
in
vivo
methods
usually
include
a
vector
system
to
deliver the
therapeutic
gene
(transgene)
(26).
Vectors
fall into
2
categories:
nonviral
and viral.
Nonviral
vectors
typically
consist
of
DNA
plasmids,
whereas
retroviral
and
adenoviral-based
vectors
have
been the
most
extensively
used viral
systems
in
gene
therapy
(26).
Although
vectors
are
usually
designed
to
be
replication
incompetent,
they
may
still contain
a
small
component
that is
capable
of
replication.
In
the
1970s,
gene
therapy
was
envisaged
as
a
treat-
ment
for
single
gene
disorders,
in which
a
faulty
gene
could be
replaced by
a
normal
copy
(45).
At
the
outset,
investigators studying
the
genetic
correction of human
disease
were
faced
with
many
serious
conceptual
and
technical
problems
(10, 24).
Early
efforts relied
on ex
vivo
methods for
introducing
new
genetic
information
into the
body,
until
pioneering
work
was
performed
that
demonstrated
the
feasibility
of direct
gene
transfer into
tissues
(57).
Since
then,
more in vivo
applications
have
been
pursued.
The
treatment
of
hereditary
diseases
now
represents
only
a
small
part
of the
overall
potential
of
gene
therapy,
and other
important
strategies
are
in
de-
velopment
(43).
Cytotoxic applications,
usually
for
can-
cer
treatment,
involve
the
delivery
of
genes
that encode
cytotoxic
proteins,
the
goal being
that these
genes
will
either
kill the
target
cells
or
make them
vulnerable
to
attacks
by
the
host
immune
system
(23).
Prophylactic
therapy
employs
gene
transfer
to
protect
against
infec-
tious
agents
and is
generally
known
as
&dquo;DNA
vaccina-
tion&dquo;
or
&dquo;gene prophylaxis.&dquo;
This
constitutes
a
new
ap-
proach
to
vaccine
development
(50).
DNA
vaccines
avoid
the risk associated
with
the
administration
of
live,
atten-
uated
organisms
and
lead
to
the
expression
of
antigen
within
host
cells,
a
process
that is
often
required
to
stim-
ulate
lasting immunity
(39).
Plans have
been
developed
to
extend
gene
therapy
trials
to
the fetus
(in
utero
ther-
apy)
in the
event
of
life-threatening
disorders,
for
which
irreparable
organ
damage
in
early
life is
certain
and
for
which
no
satisfactory
treatment
is available
(13, 48).
General
Approach
to
Safety
Evaluation
Safety
studies with
gene
therapy compounds
vary
greatly
in
design, depending
on
the
clinical indication of
the
procedure,
the
type
of
vector
used,
the duration of
gene
expression,
and the mode of
DNA
transfer
(in
vivo
or ex
vivo).
Wherever
possible,
these
products
should be
tested
in relevant
species
(i.e.,
those in which the
biolog-
ical
response
to
therapy
most
closely
mimics the
expected
human
response)
(21).
Essentially,
this
means
choosing
a
species
that
can
respond
to
the
promoter
sequence
(usu-
TABLE
III.-Safety
issues
with the
use
of
gene
therapy
vectors.
ally adjacent
to
the
transgene)
and
that
can
express
mRNA and
protein.
Many
of the
promoters
used
to
date
have
been viral
(e.g.,
cytomegalovirus promoter);
these
promoters
are
activated in
most
mammalian
cells,
and
they
induce
strong transgene
expression.
In these
cases,
the selection of
a
relevant
species
for
toxicology
studies
is
straightforward.
However,
some
promoters
are
more
specific
and
are
only
activated
in
certain cell
types
(i.e.,
the
carcinoembryonic
antigen
promoter,
which is used in
cytotoxic applications
for human colorectal
carcinoma)
(31).
In the latter
situation,
the selection of
an
appropriate
species
for
toxicology
studies is
more
problematic.
With
viral
vectors,
the relevant
species
should also be
selected
based
on
its
susceptibility
to
infection with
the
wild-type
virus
related
to
the
vector
(21).
This is
important,
because
nonsusceptible species
may
not
be
predictive
of
inflam-
matory
responses
to
the
vector
(e.g.,
rhesus
monkeys
have been
used
to
evaluate the
airway delivery
of ade-
noviral
vectors
because
of their
susceptibility
to
infec-
tions with
this
type
of
virus)
(4).
A number of
potential
safety
issues exist
with
gene
therapy
vectors,
both
with
ex
vivo
and
in vivo
treatments
(Table III).
Some of
these issues
can
be monitored
by
standard
histopathology (e.g.,
vector-induced inflamma-
tion).
However,
other
safety
concerns,
such
as
the
inap-
propriate
tissue distribution
of the
vector,
require
the
use
of
more
specialized
assays
to
evaluate the
distribution,
localization,
and
persistence
of
vectors
and
expressed
mRNA/protein
(21).
The
toxicologic
pathologist
will
have
an
important
role in the
generation
of this
infor-
mation,
or
&dquo;biodistribution
data,&dquo;
as
conventional
ab-
sorption,
distribution,
metabolism,
and
excretion
studies
may
not
be relevant for
gene
therapies
(16).
Biodistribution and
Expression
Data
Biodistribution
assays
can
be
divided
into
histologic
techniques,
which
are
performed
on
tissue
sections,
and
nonhistologic procedures,
which
are
carried
out
in
solu-
tion
phase
following
dissociation of
the tissues
(Table
IV).
The
advantage
of
histologic
assays
is that
they
allow
for the
accurate
localization and identification
of
cells
containing
the DNA
or
expressed mRNA/protein.
The
by guest on November 30, 2012tpx.sagepub.comDownloaded from
683
TABLE
IV.-Assays
used
in
gene
therapy safety
studies.
Abbreviations: PCR
=
polymerase
chain
reaction;
RT-PCR =
reverse
transcrip-
tase-polymerase
chain
reaction;
RNase
=
ribonuclease;
ELISA
=
enzyme-linked
immunosorbent
assay.
disadvantage,
in the author’s
experience,
is that the
sen-
sitivity
of
detection,
in
general,
is lower
than that
of
non-
histologic
procedures,
although
this
problem
may
be
re-
solved in the
near
future with
the
use
of
amplification
techniques
such
as
IS-PCR
and
in situ-reverse
transcrip-
tase
(RT)-PCR.
Currently,
the
most
sensitive
means
of
detecting
DNA and mRNA in
tissues is
by
the
use
of
solution-phase
PCR
and
RT-PCR,
respectively.
Unfortu-
nately,
increased
sensitivity
leads
to
an
increased risk of
false-positive
results
as a
result of cross-contamination
(see below).
Other
techniques,
such
as
Southern/northem
blotting
and ribonuclease
(RNase)
protection
assays,
are
not
as
sensitive
as
PCR and
RT-PCR,
although
the RNase
protection
method does allow
mRNA levels
to
be
quan-
tified.
Following
the administration of
a
gene
therapy
vector
by
the
in vivo
route,
its
presence
should
be
confirmed
in
the
target
tissue
and
a
small
survey
of
additional
organs
performed
in
order
to
assess
vector
dissemination
(21).
The
extent
of the
survey
will
depend
on
the
nature
of the
delivered
gene
and
on
the
route
of
administration,
but it
should also include
an
evaluation of
gene
persistence.
At
present,
distribution of the
vector
is
probably
best
as-
sessed
by
solution
phase
PCR,
using
DNA extracted
from
a
range
of
organs.
The
sensitivity
of the
PCR
assay
has
been
a
contentious
issue,
and the
FDA
recently
recom-
mended that it should be
capable
of
detecting
100
copies
(or
less)
of
the
administered
plasmid
or
viral
sequence
per
microgram
of
extracted
DNA
(22).
The
assay
should
include
samples
of
genomic
DNA,
into
which known
amounts
of
target
sequence
(plasmid
or
viral
DNA)
have
been
spiked,
in
order
to
assess
sensitivity
and
to
deter-
mine
if PCR
inhibitors
are
present.
It
is
important
that
significant
numbers of
samples
of
genomic
DNA
are an-
alyzed
from selected
tissues and that
replicate
tests
are
performed.
Following
a
positive
PCR
result,
ISH
can
then
be
attempted
in order
to
determine cellular
locali-
zation.
In addition
to
the
presence
of the
vector,
per
se,
the
degree
and
extent
of
expression
of the
transgene
may
also
be
monitored.
No
guidance
is
currently
available from
regulatory
authorities
as
to
the
precise
circumstances
un-
der
which this
process
should take
place,
and
it
is
usually
performed
at
the discretion of the
investigator. Transgene
expression
may
be assessed
routinely
in
target
tissues
in
order
to
gain
some
indication
regarding
the
efficacy
of
the
procedure.
Transgene
expression
may
also
be
exam-
ined
in
organs
in which the
vector
is shown
to
persist
or
in which
there
is
unexpected toxicity.
For
these
purposes,
the detection of
expressed
protein
rather than mRNA is
preferable,
as
this demonstrates full
pharmacologic
ex-
pression
of the
transgene.
If it is
not
feasible
to
perform
IHC,
then
analysis
for
protein expression
can
be
per-
formed
using
western
blotting
or
the
enzyme-linked
im-
munosorbent
assay
(ELISA).
Both of these
techniques
are
relatively
sensitive,
although
ELISAs
are more
quantita-
tive.
Flow
cytometry
can
be used
to
rapidly analyze
blood cells for the
presence
of
transgene-derived
proteins.
This
procedure
accurately
identifies and
enumerates
cells
containing
the
target
sequence,
although
it
is
less
readily
applied
to
solid tissues. On
some
occasions, however,
the
detection of
protein
may
not
be
possible
(e.g.,
because of
the
unavailability
of
suitable
antibodies
or
because
of
dif-
ficulties
in
discriminating
between
endogenous
and
trans-
gene-derived peptides).
In these
situations,
it
is
necessary
to resort to
the
use
of
techniques
to
detect
transgene-
derived mRNA.
A
significant advantage
of
using
methods
to
detect nucleic
acids
is that labeled
probes
and
primers
are
relatively
quick
and
easy
to
synthesize,
whereas the
production
of
antibodies,
as
tools for
IHC,
is far
more
difficult and time
consuming.
Insertional
Mutagenesis
The random
integration
of
foreign
DNA into the
ge-
nome
of
cells
can
result in insertional
mutagenesis,
and
this is
an
important
issue for
gene
therapy procedures,
as
insertional
mutagenesis
could lead
to tumor
formation
or
disruption
of
normal
gene
expression
(37).
The
extent
of
vector
integration
into the host-cell
genome
should be
investigated
both
when it is intended
and inherent
to
the
method of
expression
(e.g.,
with
retroviral
vectors)
and
also
in
cases
in
which
integration
is
not
intended
(e.g.,
with adenoviral
vectors
or
DNA
plasmids)
(16).
Again,
the
most
appropriate
assays
to
use
in
these
studies
appear
to
be
at
the
discretion
of
the
investigators.
One method
currently being performed
involves the
use
of PCR
to
distinguish
between
integrated
versus
nonintegrated
plas-
mids in
genomic
DNA
preparations
(19).
In
this
assay,
the initial
step
involves isolation of
DNA
from
a
tissue
treated with
test
plasmid.
A restriction
enzyme
such
as
Dpnl,
which cleaves GATC
recognition
sites
only
if ad-
enine nucleotides
on
both strands
are
methylated,
is then
used
to
digest
free
plasmid
in the isolated DNA
sample.
In
contrast,
plasmid
sequence
that has
integrated
into the
genomic
DNA loses
its
bacterial
methylation
pattern
at
replication
and is thus
no
longer
sensitive
to
Dpnl
di-
gestion.
Agarose gel electrophoresis
is then used
to
sep-
arate
free
plasmid
from
host-cell
genomic
DNA. The
ge-
nomic DNA band
is
cut
from the
gel,
and
a
sample
of
this DNA is used in
a
PCR reaction. The
primers
are
selected
to
produce
a
specific
product
in the
presence
of
the
plasmid
sequence.
If
no
integration
has
taken
place,
then
no
product
is
produced.
Meticulous
care
is
required
in order
to
reduce
the
likelihood
that
contamination
with
free
plasmid
produces
a
false-positive
result,
and the
use
of
suitable
positive
and
negative
controls is essential
(42).
by guest on November 30, 2012tpx.sagepub.comDownloaded from
684
Germline
Transmission
Concern
has
recently
been
raised
over
the
possibility
of
germline
transfer
following
the
in vivo
administration
of
gene
therapy
vectors
(26).
Indeed,
the
FDA have
re-
cently
confirmed that
concern
with
germline
spread
is
so
great
that after
successful
animal
studies,
some
research-
ers
have found their human
gene
therapy
clinical
trial
protocols
stalled
over
the
germline
issue.
This is
a
rela-
tively
new
development,
since
traditionally,
human trials
have
been
approved
without
the need
to
examine the
germline
risk
(13).
Preclinical
evidence indicates that fol-
lowing
administration,
viral
vectors
are
widely
distrib-
uted
throughout
the
body
and
are
frequently
detected in
gonads,
where
they
may
persist
for
up
to
3
mo
(18).
The
author
has
also
observed
widespread
distribution
(includ-
ing gonads)
of
a
nonviral
vector
following topical
appli-
cation
to
mucous
membranes. Other data have shown
that
under
the
right
circumstances,
integration
of
adenoviral
DNA
into
germ
cells
can occur
(52).
However,
the overall
risk of
a
deleterious
outcome
as a
result
of
inadvertent
germline
gene
transfer
has
been estimated
to
be
in the
order
of
a one
in
a
billion chances
or
less
(13, 59).
Nev-
ertheless,
concerns over
germline
transfer
are
likely
to
remain
until
sufficient
data
has been
generated
to
allow
the risks
to
be
refined.
Regulatory guidelines
suggest
that &dquo;a
persistent
PCR
signal
for the
vector
will
require
further
investigation
to
exclude
integration
into the host
genome
of
germ
cells&dquo;
(16).
Currently,
however,
there
appears
to
be
no
consen-
sus
as
to
which
investigative
techniques
are
appropriate
in this
situation. The
approach
taken
in this
laboratory
is
to
investigate
all
persistent
gonadal
signals (nominally
of
1-mo duration
or
longer) using
PCR-based
integration
analysis
(as
outlined
above).
This
technique provides
general
information
regarding
the
presence
or
absence of
integration
in the
gonadal
sample, although
in the
future,
the
use
of
histologic
assays may prove
invaluable
in
con-
firming
(or
disproving)
that
a
positive signal
is indeed
attributable
to
germ-cell
localization of the
vector.
The
technique
of
choice
will be
IS-PCR,
which
should allow
localization in
germ
cells
to
be
distinguished
from that
in
stromal tissue
(21).
However,
although
this
procedure
seems
straightforward,
it
is
technically
very
demanding,
and
many
scientists
have
been
unable
to
achieve
repro-
ducible results.
This
situation could
change
with the de-
velopment
of
improved protocols
and
reagents,
but
at
the
present
time,
it is debatable whether in
situ
amplification
techniques
can
confidently
be used in
regulatory
submis-
sions.
Further
possibilities
include the PCR
analysis
of
semen
samples,
collected from
mature
treated
animals,
for
determination
of
vector
presence
and
integration
into
germ
cells
(21).
Immunogenicity
Although
the
immunogenicity
of DNA
vaccines is well
established,
concerns
have been raised
regarding
their
po-
tential
to
induce
deleterious
autoimmune
responses
(3).
This
concern
has
been
heightened by
evidence that
the
plasmid
backbone
of DNA vaccines is bacterial in
origin
and thus has intrinsic
immunostimulatory activity
(35).
Despite
these
concerns,
other
investigations
have
found
no
evidence
of
significant autoantibody production,
and
it has been
concluded
that
conventional
DNA
vaccines
are
not
associated with the induction
of
unsafe
autoim-
mune
sequelae
(41,
53).
Cross- Contamination
One
major
concern,
when
working
with
gene
therapies
and nucleic acid
drugs,
is the issue
of
cross-contamina-
tion.
This is
relatively
common
and involves
the transfer
of
genetic
material
between animals in
different
treatment
groups
or
between tissues
in the
same
treated animal.
Transfer
can occur
at
any
point
in the live animal
phase,
during
necropsies,
and
during handling
of
the tissues
in
the
laboratory. Only
very
small
quantities
of
genetic
ma-
terial
need
be
transferred
to
confound
the PCR
assay,
which
is
a
powerful
amplification
tool and
which is
usu-
ally
nonquantitative
(i.e.,
the
same
positive
result will
be
generated
in
different tissues
despite large
variations in
the
starting quantity
of
target
DNA).
The
consequences
of cross-contamination
are
potentially
serious and
may
lead
to
invalidation of the
study-particularly
when
pos-
itive
PCR
signals
are
obtained in tissues from untreated
(control)
animals. Additional
training
in
necropsy
pro-
cedures
may
be
necessary
for technical
staff,
as
the har-
vesting
of
multiple
tissues
(for
PCR
analysis)
from
large
numbers of animals
on a
single
occasion is
technically
demanding.
Cross-contamination
of control
animals
can
be
minimized
by necropsying
these animals
as
a
group,
in advance of
necropsying
of the treated animals.
During
the
necropsies,
one
set
of
instruments
is used
to
expose
the
relevant
organ/tissue,
and
a
second,
clean
(unused)
set
is
used
to
remove
the
sample.
The whole
procedure
requires
careful
planning, especially
when tissues
are
to
be
removed
from the
same
carcass
to
supply
several
as-
says
(including histology
and
PCR).
Extra
laboratory
fa-
cilities
may
also be
needed,
so
that the tissue
extraction
of
nucleic
acids,
reagent
and material
storage,
and
PCR
can
be carried
out
in
separate
areas.
Particular
care
must
be
taken when
spiking
samples
(e.g.,
with the
test
nucleic
acid)
to
produce
positive
controls. This should be
done
in
an
area
that
is
isolated from the
rest
of the
assay.
ANTISENSE, RIBOZYMES,
AND
APTAMERS
Advances
in
molecular
biology
and
synthetic
chemis-
try
have led
to
novel nucleic acid
drugs
to
inhibit
gene
expression
and
protein
function
(49).
Three
types
of
drugs
that I
will
briefly
discuss
are
oligonucleotides
(an-
tisense
therapy), ribozymes,
and
aptamers.
In
1978,
the
first
experiments
in
which
oligonucleo-
tides
were
used
as
specific
inhibitors of
gene
expression
in
cell
culture
were
described
(60).
Over 20
yr
later,
an-
tisense
therapies using synthetic oligonucleotides
are
be-
ing
assessed
in
clinical trials for diseases such
as
cancer,
inflammation,
and viral infections
(40).
These
therapies
are
based
upon
the
sequence-specific
hybridization
of
an
oligonucleotide
with the mRNA
of
a
target
protein.
The
hybrid
molecule formed is then
degraded
by
nucleases,
although
several other
possibilities
exist
(e.g.,
inhibition
of
mRNA
splicing
or
inhibition of
mRNA
translation)
(40).
Of the
first-generation
oligonucleotides,
the
phos-
by guest on November 30, 2012tpx.sagepub.comDownloaded from
685
phorothioate
class
containing
a
P
=
S
backbone
(PS-oli-
gonucleotides)
have
progressed
the furthest
as
potential
therapeutic
agents.
These
compounds
are
chemically
modified
to
be resistant
to
nucleases and
are
found in all
body
tissues,
but
their
development
has
been limited
by
toxicities that
are
sequence
independent
(i.e.,
the
toxicity
is
attributable
to
the chemical
class) (28, 55).
In
all
spe-
cies
studied,
the
effects include
prolonged
clotting
times
and
renal
degenerative changes. Additionally,
comple-
ment
activation has
been
reported
in
primates
and
diffuse
lymphoid hyperplasia
in rodents
(28).
In
an
attempt
to
overcome
these
problems,
a
second
generation
of
anti-
sense
compounds
has
been
designed,
comprising
mixed
backbone
oligonucleotides
(MBOs).
The
latter contain
segments
of
PS-oligonucleotides
and
appropriately
placed
segments
of
modified
oligodeoxynucleotides
or
oligoribonucleotides.
Recent work
has
shown
reduced
toxicity
of
an
MBO,
as
compared
with
a
PS-oligonucle-
otide,
against
the
same
target
mRNA
(protein
kinase
A)
in CD-1 mice
(2).
Ribozymes
are
small
RNA
structures
that
catalytically
cleave covalent
bonds
in
a
target
RNA
(49).
These mol-
ecules
were
first discovered
in the
Tetrahymena
organ-
ism,
where
they
were
found
acting
as
sequence-specific
endoribonucleases
upon
other
RNA
substrates
(61).
Ri-
bozymes
can
inhibit
gene
expression
in
a
sequence-spe-
cific
manner
and have the
potential
to
be used
therapeu-
tically
to
destroy
harmful
mRNAs
produced
in
cancer
and
viral diseases.
Although
the
amount
of
preclinical
safety
work carried
out to
date is
relatively
small,
ribozyme
products
have
a
promising
safety profile.
A
recent
report
describes the administration
to
rats
of
a
ribozyme
de-
signed
to
cleave
the
mRNA
of
cytochrome
P-450
3A2,
and
the
report
states
that
no
morphologic
evidence of
toxicity
was
observed
(12).
Aptamers
(from
the Latin
aptus,
to
fit)
are
single-
or
double-stranded nucleic acids that
are
capable
of
binding
proteins
or
other
small
molecules
(15,
49).
This
binding
is conducted
by shape recognition
and
is
not
sequence
dependent.
As
therapeutic
agents, aptamers
may
have
a
wide
range
of
protein
targets,
including transcription
fac-
tors,
extracellular
proteins,
and cell surface
molecules.
They
are
also modified
as
phosphorothioates
in
order
to
improve stability,
in
a
way
similar
to
antisense
products.
However,
virtually
no
preclinical safety
data
exist in the
public
domain for these
compounds.
At
the
present
time,
no
specific
guidance
is available
from
regulatory
agencies
regarding
the
preclinical
eval-
uation of
the
3
novel
therapies
discussed
above. Nucleic
acid
drugs
may
be
complexed
with cationic
liposomes
or
artificial
viral
envelopes
in
order
to
facilitate
their deliv-
ery
and
entry
into
the
target
site.
Safety
issues,
similar
to
those
discussed
for
gene
therapies,
would
accompany
the
use
of these
vector
systems.
Molecular
pathology
techniques
may
be of considerable
value
in
safety
studies
with
these
products-for
instance,
when
evaluating
an-
tisense
compounds,
the
use
of
PCR
or
ISH could
be
used
to
demonstrate
changes
in
the
target
mRNA levels in
the
appropriate
cell
population.
In
addition,
IHC is of value
in
assessing
the
cellular distribution of
PS-oligonucleo-
tides
in animal
studies,
as
it is
possible
to
raise antibodies
directly against
these
compounds
(8).
GENERAL
ISSUES
Biopharmaceutical
research will
progress
rapidly
in
the
next
few
years.
As
new
biotechnology-derived
com-
pounds
feed into the
development pipeline,
they
will have
a
considerable
impact
upon
toxicology
in
general
and
on
the role of the
toxicologic pathologist
in
particular.
Fewer
conventional
toxicology
studies,
including carcinogenic-
ity
studies,
will
be
performed,
and, therefore,
the
require-
ment
for standard
histopathology
will be reduced.
How-
ever,
this will be offset
by
the need for
molecular
pa-
thology
assays,
to
which
arena
the
pathologist
can
make
an
important
contribution.
Increased involvement
with
biopharmaceuticals,
with
a
corresponding
increase
in the numbers of molecular
pro-
cedures,
may
pose
challenges
for the
histopathology
de-
partment.
For
the
pathologist,
considerable intellectual ef-
forts
will
be
necessary
to
develop
and maintain
expertise
in
molecular
techniques.
The molecular
assays
described
in this
paper
may
be
appropriate
for
the time
being,
but
techniques
are
likely
to
change
quickly
in this
rapidly
evolving
field. The
pathologist
should continue
to
be
re-
sponsible
for
morphologic
assays,
because
significant
his-
tologic
experience
is
required
for
their
correct
interpre-
tation. The
nonmorphologic techniques
may
be
per-
formed outside the
histopathology department,
but it is
essential
that
the
pathologist
has
a
strong
understanding
of
these
procedures
so
that correlations
can
be made be-
tween
morphologic
and
nonmorphologic techniques.
Fur-
thermore,
the
responsibility
for
harvesting
tissues
at
nec-
ropsies
(to
supply
these
assays)
will
invariably
reside
with the
pathologist.
Difficulties
may
be encountered in
ensuring
that
mo-
lecular
pathology
techniques comply
with GLP
We
prefer
to
contact
the
Quality
Assurance
(QA)
department
at
an
early
stage
(for
discussions)
when
a new
assay
or
pro-
cedure is
being
planned,
as
QA
inspectors
may
have
little
knowledge
or
experience
related
to
molecular
pathology
techniques
and will
usually
value
early
involvement. We
invite
them
to
watch
the
new
procedure,
at
the
develop-
ment
stage,
and
to
ask
for
comments,
before
we com-
mence a
GLP
study.
This
allows the
auditing
process
to
proceed
more
smoothly.
As molecular
techniques
are
complex, frequent adjustments
are
usually
required
as
the
work
progresses,
often
necessitating
last-minute alter-
ations
to
protocols
and
method
sheets. It is worth
remem-
bering
that GLP
does
not
prohibit
these
changes
as
long
they
are
fully
documented,
which
may
involve the fre-
quent
placement
of
explanatory
notes
in the
study
file.
In
the
case
of
in vitro
studies
(e.g.,
cross-reactivity
tests
with
mAbs),
the
pathologist
may
also
be
cast
in
the role
of
study
director,
which
will entail the
assumption
of
addi-
tional
responsibilities,
such
as
protocol
drafting,
record-
ing
of
batch details of
test
materials,
archiving
of
duties,
etc.
If the
pathologist
is
inexperienced
in
this
role,
then
it
pays
for
him
or
her
to
seek advice before
starting
the
work.
Close links with
appropriate
research
groups
are
indis-
pensable
when
planning
molecular
procedures
for
pre-
by guest on November 30, 2012tpx.sagepub.comDownloaded from
686
clinical
development.
In
fact,
regulatory agencies
en-
courage
sponsors
to
obtain
toxicity
data whenever
pos-
sible
while
evaluating
biopharmaceuticals
in
relevant
an-
imal
models
(21).
Research
groups may
have
used
molecular
procedures
to
demonstrate
&dquo;proof
of
concept&dquo;
in
such models.
Any
support
provided
in this
way
is
help-
ful,
since
tight
development
time frames
dictate
that
mo-
lecular
assays
be
validated before
toxicology
studies
commence.
When
developing morphologic
assays
to
sup-
port
the
preclinical development
programs,
the
use
of cell
lines and/or
primary
cell cultures
can
be
a
useful
inter-
mediate
step
prior
to
in
vivo
studies. These
cells
can
be
used both
to
develop
PCR, ISH,
and
IHC
techniques
and
to
act
as
controls for
subsequent in
vivo
experiments.
In
this
review,
emphasis
has been
placed
on
molecular
pathology
assays
that
are
required
to
support
the
devel-
opment
of
biotechnology-derived
products,
although
it
must
be
stressed that conventional
histopathology
will
still
have
an
important
role. The
study
of
biopharmaceu-
ticals is
still
in its
infancy,
and
many
safety
concerns
may
ultimately
prove
to
be of
only
theoretical
importance.
At
this
stage,
however,
confidence
with
these
products
can
only
be achieved
through
continued
safety
assessments.
In
contrast
with
traditional
compounds,
2
prominent
safe-
ty
issues
with
biopharmaceuticals
are
immunogenicity
and contaminant-associated
toxicity.
Many
biopharmaceuticals
intended for humans
are
im-
munogenic
in
animals,
and this
can
limit both the choice
of
preclinical
species
and the duration of
dosing.
Mea-
surement
of antibodies
associated
with the administration
of
these
products
should be
performed
when
conducting
repeat-dose toxicity
studies.
Antibody
responses
should
also be
characterized
(e.g.,
titer,
number of
responding
animals,
neutralizing
or
nonneutralizing),
and
their
ap-
pearance
can
then be
correlated
with
any
toxicologic
changes
(33).
The
potential immunogenicity
of
a
bio-
pharmaceutical
is
a
significant
issue,
since
antibody
bind-
ing
can
partially
or
completely
inhibit
its
biological
ac-
tivity,
affect its
catabolism,
or
alter its distribution and
clearance
(i.e.,
decrease
or
increase the
half-life).
If the
antibody
response
mounted
by
the host
results
in
a
change
in
clearance
rate,
then
exposure
will be
over- or
underestimated
(26).
However,
antibody
formation in it-
self is
not
a reason
for termination of
a
toxicity
study,
particularly
if the antibodies
are
not
neutralizing
or
if
they
do
not
alter
the
pharmacodynamics
of
the
test
compound
(11).
It should be
noted
that the
induction
of
antibody
formation
in animals
is
not
predictive
of
a
potential
for
antibody
formation in humans. Humans
may
develop
se-
rum
antibodies
against
humanized
proteins,
and
frequent-
ly,
the
therapeutic
response
persists
in
their
presence.
When
undertaking
histopathologic
evaluation,
particular
attention should
be
paid
to
the
assessment
of
possible
changes
related
to
the
expression
of
altered
surface
an-
tigens
on
target
cells
(autoimmunity)
and
to
immune
complex
formation
and
deposition
(33).
Some
biotech-
nology
compounds
are
pharmacologically designed
to
stimulate
or
suppress
the immune
system.
In
such
cases,
inflammatory
reactions
(e.g.,
lymphocytic
infiltration
at
the
injection
site)
may
be
indicative of
a
stimulatory
re-
sponse.
The risk
of
viral contamination
is
a
feature
common
to
all
products
derived from cell
lines,
and the
presence
of
these
viral contaminants
can
result in
allergic
reactions
and
other
immunopathological
effects
(32).
The adverse
effects
associated
with nucleic acid contaminants
are
again
theoretical,
but
they
include
potential
integration
into
the host-cell
genome.
The relevance and value of
conducting
classical
car-
cinogenicity bioassays
with
biopharmaceuticals
is
ques-
tionable
(26).
First,
conventional
2-yr
carcinogenicity
studies
may
not
be
possible
because
of the
development
of
neutralizing
antibodies that
render
the
study
valueless.
Second,
these
studies
are
normally
carried
out
in rodent
species,
which
may
not
be
biologically
responsive.
Stan-
dard
carcinogenicity
studies
may
be considered
appro-
priate
for these
products
under certain
conditions,
includ-
ing
in
situations in
which
there is clear
evidence
of
ge-
nomic
integration
or
in
which the
ability
to
support
or
induce cell
proliferation
is demonstrated
(19, 33).
In
vitro
assays may
be
necessary
for the cell
proliferation
work,
in
which molecular
pathology
techniques
will
play
an
important
role
(e.g.,
IHC
and
ISH
for markers
of
cell
proliferation
or
apoptosis).
In
conclusion,
expansion
of the
biotechnology portfo-
lio will
present
opportunities
for the
toxicologic pathol-
ogist
and will allow
for diversification into the
molecular
field,
which
has
previously
been
regarded
as
the
domain
of the
experimental pathologist.
Involvement in
all
phases
of
biopharmaceutical
research
and
development
will be
possible (including
clinical
applications
and
public
health
issues
as
well
as
preclinical
studies).
The
toxicologic
pa-
thologist
is therefore well
placed
to
retain
a
central
po-
sition in the
preclinical
development
of these
compounds.
ACKNOWLEDGMENTS
I thank Richard
Haworth,
Ann
Rowlands,
Ron
Tyler,
Jan
Klapwyk, Rajni
Fagg,
and
David
Tweats
for review-
ing
and
offering
constructive criticisms
on
the
manu-
script.
REFERENCES
1.
Adams JC
(1992).
Biotin
amplification
of
biotin and
horseradish
peroxidase
signals
in
histochemical
stains.
J.
Histochem.
Cytochem.
40: 1457-1463.
2.
Agrawal
S and Zhao
Q
(1998).
Mixed
backbone
oligonucleotides:
Improvement
in
oligonucleotide-induced toxicity in
vivo.
Antisense
Nucleic Acid
Drug
Dev.
8: 135-139.
3. Beard
CW and Mason PW
(1998).
Out
on
the farm with DNA
vaccines.
Nat.
Biotechnol. 16:
1325-1328.
4.
Brody
SL,
Metzger
M,
Danel
C,
Rosenfeld
MA,
and
Crystal
RG
(1992).
Acute
responses
of non-human
primates
to
airway
delivery
of
an
adenovirus
vector
containing
the human
cystic
fibrosis
trans-
membrane
conductance
regulator
cDNA. Hum. Gene Ther. 5 : 821-
836.
5.
Buckel
P
(1996).
Recombinant
proteins
for
therapy.
Trends
Phar-
macol. Sci. 17:
450-456.
6. Burdick
JF,
Pennington
LR,
Smith
W,
Irwin
B,
and Williams GM
(1987).
Reversal
of
progressive
renal
allograft
dysfunction
with
OKT3.
Nephron
46:
52-55.
7. Burnett
R,
Guichard
Y,
and
Barale
E
(1997).
Immunohistochem-
istry
for
light
microscopy
in
safety
evaluation
of
therapeutic
agents:
An
overview.
Toxicology
119: 83-93.
8. Butler
M,
Stecker
K,
and Bennett CF
(1997).
Cellular distribution
by guest on November 30, 2012tpx.sagepub.comDownloaded from
687
of
phosphorothioate
oligodeoxynucleotides
in
normal
rodent tis-
sues.
Lab.
Invest.
77:
379-388.
9.
Caplen
NJ,
Alton
EWFW,
Middleton
PG,
Dorin
JR,
Stevenson
BJ,
Gao
X,
Durham
SR,
Jeffery
PK,
Hodson
ME,
Coutelle
C,
Huang
L,
Porteous
DJ,
Williamson
R,
and
Geddes
DM
(1995).
Liposome-
mediated
CFTR
gene
transfer
to
the
nasal
epithelium
of
patients
with
cystic
fibrosis.
Nat.
Med.
1:
39-46.
10.
Cline
MJ,
Stang
H,
Mercola
K,
Morse
L,
Ruprecht
R,
Browne
J,
and
Salser
W
(1980).
Gene transfer in intact animals.
Nature
284:
422-425.
11.
Dempster
M
(1998).
Designing
non-clinical
safety
evaluation
pro-
grammes :
The
view
of the
toxicologist.
In:
Safety
Evaluation
of
Biotechnologically-Derived
Pharmaceuticals:
Facilitating
a
Sci-
entific Approach,
SA Griffiths
and CE
Lumley
(eds),
Kluwer Ac-
ademic
Publishers,
London, UK,
pp.
7-15.
12.
Desjardins
JP,
Sproat
BS,
Beijer
B,
Blaschke
M,
Dunkel
M,
Gerdes
W,
Ludwig
J,
Reither
V,
Rupp
T,
and
Iversen PL
(1996).
Pharma-
cokinetics of
a
synthetic, chemically
modified hammerhead
ribo-
zyme
against
the
rat
cytochrome
P-450
3A2
mRNA after
single
intravenous
injections.
J.
Pharmacol.
Exp.
Ther.
278: 1419-1427.
13. Editorial
(1999).
Gene
therapy
and the
germline.
Nat.
Med. 5: 245.
14.
Ellington
AD and Conrad R
(1995).
Aptamers
as
potential
nucleic
acid
pharmaceuticals.
Biotechnol.
Annu.
Rev. 1: 185-214.
15.
Ellington
AD
and Szostak
JW
(1990).
In
vitro
selection
of
RNA
molecules
that
bind
specific ligands.
Nature
346: 818-822.
16.
European
Agency
for the
Evaluation of Medicinal
Products,
Com-
mittee
for
Proprietary
Medicinal
Products
(1998).
Safety
studies
for
gene
therapy
products:
Annex
to note
for
guidance
on
gene
therapy product
quality
aspects
in
the
production
of
vectors
and
genetically
modified
somatic cells.
Report
No.
SWP/112/98,
Eu-
ropean
Agency
for the Evaluation of Medicinal
Products,
Com-
mittee
for
Proprietary
Medicinal
Products,
pp.
1-7.
17. Farrell GC
(1991).
Treatment
of chronic
hepatitis
C with
alpha
interferon.
J.
Gastroenterol.
Hepatol.
6:
36-40.
18. Food
and
Drug
Administration
(1997).
Panel
to
investigate possible
germline
transfer
in
gene
therapy
work. FDA Week 3:
1-2.
19.
Food
and
Drug
Administration,
Center
for
Biologics
Evaluation and
Research
(1996).
Points
to
consider
on
plasmid
DNA vaccines
for
preventive
infectious disease indications. Fed.
Reg.
61: 68269.
Available
at:
http://www.fda.gov/cber.
20.
Food and
Drug
Administration,
Center for
Biologics
Evaluation and
Research
(1997).
Points
to
consider
in the manufacture and
testing
of monoclonal
antibody
products
for
human
use.
Fed.
Reg.
62:
9196.
Available
at:
http://www.fda.gov/cber.
21. Food and
Drug
Administration,
Center for
Biologics
Evaluation and
Research
(1998).
Guidance
for
industry:
Guidance for human
so-
matic cell
therapy
and
gene
therapy.
Fed.
Reg.
63: 36413. Avail-
able
at:
http://www.fda.gov/cber.
22.
Food and
Drug
Administration,
Center for
Biologics
Evaluation and
Research
(1999).
General
guidance
available
at:
http://www.nih-gov/
od/orda/3-99sum.htm.
23.
Fox S
(1997).
Gene
therapy
under
the
spotlight.
Pharm.
Forum
8:
5-7.
24.
Friedmann T
(1989). Progress
toward
human
gene
therapy.
Science
244: 1275-1281.
25.
Gillett
NA
and
Chan
CM
(1999).
Molecular
pathology
in
the
pre-
clinical
development
of
biopharmaceuticals.
Toxicol.
Pathol.
27:
48-52.
26.
Griffiths SA and
Lumley
CE
(1998).
Non-clinical
safety
studies
for
biotechnologically-derived pharmaceuticals:
Conclusions
from
an
international
workshop.
Hum.
Exp.
Toxicol. 17:
63-83.
27.
Hansen
B,
Lernmark
A,
Nielsen
JH,
Owerbach
D,
and Welinder B
(1982).
New
approaches
to
therapy
and
diagnosis
of diabetes. Dia-
betologia
22: 61-67.
28.
Henry
SP,
Monteith
D,
and Levin AA
(1997).
Antisense
oligonu-
cleotide
inhibitors for the
treatment
of
cancer:
2.
Toxicological
properties
of
phosphorothioate oligodeoxynucleotides.
Anticancer
Drug
Des. 12:
395-408.
29.
Ho
PT
and
Parkinson
DR
(1997).
Antisense
oligonucleotides
as
therapeutics
for
malignant
diseases.
Semin.
Oncol.
24:
187-202.
30. Hsu
S,
Raine
L,
and
Fanger
H
(1981).
Use of
avidin-biotin-per-
oxidase
complex
(ABC)
in
immunoperoxidase techniques:
A
com-
parison
between
ABC
and
unlabeled
antibody
(PAP)
procedures.
J. Histochem.
Cytochem.
29:
577-580.
31.
Huber
BE,
Richards
CA,
and Austin
EA
(1994).
Virus-directed
enzyme/prodrug
therapy
(VDEPT).
Selectively engineering drug
sensitivity
into
tumors.
Ann.
NY Acad. Sci. 716:
104-114.
32. International Conference
on
Harmonisation of Technical
Require-
ments
for
Registration
of Pharmaceuticals for Human
Use
(1998).
Guidance
on
viral
safety
evaluation of
biotechnology
products
de-
rived from cell lines of human
or
animal
origin. Proceedings of
the
Fourth International
Conference,
Brussels,
1997,
PF
D’Arcy
and
DWG Harron
(eds), Queens
University
Press,
Belfast, Ireland,
pp.
877-908.
33. International Conference
on
Harmonisation of Technical
Require-
ments
for
Registration
of Pharmaceuticals for
Human Use
(1998).
Preclinical
safety
evaluation of
biotechnology-derived pharmaceu-
ticals.
Proceedings
of
the
Fourth
International
Conference,
Brus-
sels, 1997,
PF
D’Arcy
and DWG
Harron
(eds), Queens University
Press,
Belfast, Ireland,
pp.
1045-1056.
34. Irshad
M
(1998).
Nucleic acid
hepatitis
B
virus
(HBV)
vaccine:
Current
status.
Hepatol.
Res.
13:
71-79.
35. Klinman
DM,
Yi
A,
Beaucage
SL,
Conover
J,
and
Krieg
AM
(1996).
CpG
motifs
expressed
by
bacterial
DNA
rapidly
induce
lymphocytes
to secrete
IL-6,
IL-12 and
IFNg.
Proc.
Natl. Acad.
Sci. USA
93:
2879-2883.
36.
Kohler
G and Milstein C
(1975).
Continuous cultures of fused cells
secreting antibody
of
predefined specificity.
Nature
256: 495-497.
37. Kurth
R
(1995).
Risk
potential
of the chromosomal insertion
of
foreign
DNA.
Ann.
NY
Acad.
Sci.
772:
140-151.
38. MacCallum
DE,
Hupp
TR,
Midgley
CA,
Stuart
D,
Campbell
SJ,
Harper
A,
Walsh
FS,
Wright
EG,
Balmain
A,
Lane
DP,
and Hall P
(1996).
The
p53
response
to
ionizing
radiation
in
adult and
devel-
oping
murine tissues.
Oncogene
13:
2575-2587.
39.
Meager
A and Robertson JS
(1998).
Regulatory
and standardization
issues for
DNA and
vectored vaccines.
Curr. Res.
Mol.
Therapeu-
tics 1:
262-265.
40.
Monteith DK and Levin AA
(1999).
Synthetic oligonucleotides:
The
development
of antisense
therapeutics.
Toxicol. Pathol.
27:
8-
13.
41. Mor
G,
Singla
M,
Steinberg
AD,
Hoffman
SL,
Okuda
K,
and Klin-
man
DM
(1997).
Do DNA
vaccines induce
autoimmune disease?
Hum.
Gene
Ther.
8: 293-300.
42. Nichols
WW,
Ledwith
BJ,
Manam
SV,
and
Troilo PJ
(1995).
Po-
tential
DNA
vaccine
integration
into host
cell
genome.
Ann.
NY
Acad. Sci. 772: 30-39.
43. Palu
G,
Bonaguro
R,
and
Marcello
A
(1999).
In
pursuit
of
new
developments
for
gene
therapy
of human diseases.
J.
Biotechnol.
68:
1-13.
44. Peterson
NC
(1996).
Recombinant
antibodies: Alternative
strategies
for
developing
and
manipulating
murine-derived
monoclonal anti-
bodies. Lab.
Anim.
Sci.
46: 8-13.
45.
Rogers
S
(1971).
Gene
therapy:
A
potentially
invaluable aid
to
medicine and
mankind.
Res.
Commun. Chem.
Pathol.
Pharmacol.
2:
587-600.
46. Roth
MS,
Bunn
PA,
and
Foon KA
(1988).
Interferon
therapy
for
lymphoproliferative
disorders.
Cancer Treatment
Res.
38:
231-252.
47.
Ryffel
B
(1996).
Unanticipated
human
toxicology
of recombinant
proteins.
Arch.
Toxicol.
18: 333-341.
48. Schneider H and Coutelle C
(1999).
Commentary:
In
utero
gene
therapy.
The
case
for.
Nat.
Med. 5: 256-257.
49.
Stull
RA and
Szoka FC
(1995).
Antigene,
ribozyme
and
aptamer
nucleic acid
drugs: Progress
and
prospects.
Pharm.
Res. 12:
465-
483.
50.
Tang
D,
DeVit
M,
and
Johnston A
(1992).
Genetic immunization
is
a
simple
method for
eliciting
an
immune
response.
Nature
356:
152-154.
51.
Tang
XB,
Hobom
G.
and Luo
D
(1994).
Ribozyme
mediated de-
struction
of influenza
A
virus in
vitro
and in vivo.
J.
Med. Virol.
42:
385-395.
by guest on November 30, 2012tpx.sagepub.comDownloaded from
688
52. Tsukui
T,
Kanegae
Y,
Saito
I,
and
Toyoda
Y
(1996).
Transgenesis
by
adenovirus-mediated
gene
transfer into
mouse
zona-free
eggs.
Nat. Biotechnol. 14:
982-985.
53. Van Drunen
Little-van
den Hurk
S,
Braun
RP,
Lewis
PJ,
Karvonen
BC,
Baca-Estrada
ME,
and Snider
M
(1998).
Intradermal
immu-
nization
with
a
bovine
herpesvirus-1
DNA
vaccine
induces
protec-
tive
immunity
in cattle. J. Gen. Virol.
79:
831-839.
54. Vaswani
SK
and Hamilton
RG
(1998).
Humanised antibodies
as
potential therapeutic drugs.
Ann.
Allergy
Asthma
Immunol. 81:
105-115.
55.
Wagner
RW
(1995).
The
state
of
the
art
in antisense research. Nat.
Med. 1:
1116-1118.
56. Waldmann TA
(1991).
Monoclonal
antibodies
in
diagnosis
and
therapy.
Science 252:
1657-1662.
57. Wolff
JA,
Malone
RW,
Williams
P,
Chong
W,
Acsadi
G,
Jani
A,
and
Felgner
PL
(1989).
Direct
gene
transfer
into
mouse
muscle
in
vivo. Science 247:
1465-1468.
58.
Yang
XD,
Jia
XC,
Corvalan
JR,
Wang
P,
Davis
CG,
and
Jakobovits
A
(1999).
Eradication of established
tumors
by
a
fully
human
monoclonal
antibody
to
the
epidermal
growth
factor
receptor
with-
out
concomitant
chemotherapy.
Cancer Res.
59: 1236-1243.
59. Ye
X,
Gao
GP,
Pabin
C,
Raper
SE,
and
Wilson
JM
(1998).
Eval-
uating
the
potential
of
germ
line transmission
after
intravenous ad-
ministration
of
recombinant
adenovirus in
the
C3H
mouse.
Hum.
Gene
Ther.
9:
2135-2142.
60.
Zamecnik
PC and
Stephenson
ML
(1978).
Inhibition of
rous sar-
coma
virus
replication
and cell transformation
by
a
specific oligo-
deoxynucleotide.
Proc.
Natl. Acad. Sci. USA
75:
280-284.
61.
Zaug
AJ,
Been
MD,
and
Cech
TR
(1978).
The
Tetrahymena
ribo-
zyme
acts
like
an
RNA
restriction
endonuclease.
Nature
324: 429-
433.
by guest on November 30, 2012tpx.sagepub.comDownloaded from