4
focus
Focus
Introduction to biotech drugs
Author
Dr Anita M OConnor, Senior Director Biopharmaceuticals
Development and Regulatory Services, MDS Pharma
Services, North Carolina, US
Key words
Biologicals, Biopharmaceuticals, Biosimilars, Recombinant DNA,
Immunoassays, Bioassays, Pharmacokinetics, Immunogenicity, Antibody
drugs, Cellular proliferation assays, Carcinogenicity, Xenograft studies,
Immunohistochemistry studies
Abstract
This article provides an overview of drugs produced by biotechnology,
often called biotech drugs. When these drugs were first submitted
to the regulatory authorities in the 1980s, it was recognised at an
early stage that new approaches were needed for their development
programmes. Challenges specific to biotech drugs include their potential
for immunogenicity, the need for novel bioanalytical assays and the
requirement to use animal models in preclinical studies in which the drug
is pharmacologically active. These issues, among others, present regulatory
challenges that are unique to large molecules.
Definition and manufacturing
Biotech is the term used for biotechnology or products produced by
biotechnology. These drugs are also called biologicals, biotech drugs,
biological drugs, or biopharmaceuticals. True biotech products are
manufactured in live biological systems known as expression systems.
They are generated using recombinant DNA technology, and the
expression cells are amplified in bacterial or mammalian cell culture.
Ultimately the desired product is produced in large scale fermentation
vessels.The end stage fermentation product, a heterogeneous mixture,
is then purified in a series of processing steps to remove extraneous
host cell and other contaminants to produce the drug product. Some
of the newest production systems now being used are transgenic
animals (eg, goats, cattle) to serve as bioreacters.1
During the manufacturing process, many immunoassays and
bioassays must be developed.These are quantitative assays specifically
designed for each step in quality control, product characterisation,
stability testing, comparability analysis, in-process controls, and the
potency or functional activity of the drug candidate.2 The latter,
measured by a potency or bioassay, is required by regulatory
authorities to establish the biological activity of each drug lot, and
must be finalised prior to marketing approval.Thus, the manufacturing
process and its accompanying immunoassays and bioassays are
Regulatory Rapporteur Vol 6, No 1, January 2009
unique to drug and sponsor and essentially define the marketed
drug. This is the crux of the debate over biosimilars (ie, generic
biopharmaceuticals): copies of biotech drugs can be produced but
structurally complex biological drugs wont be identical copies, at
least with existing technology. Attempts to produce generic versions
of biotech products are therefore referred to as biosimilars in
an attempt to recognise the structural differences that may exist
between the original and biosimilar product.
Physical characteristics
Biopharmaceuticals are physically very different from small molecule
drugs.The latter are generally sized at 1 kDa or less and the former are
generally greater than 30 kDa.3 We call them high molecular weight
drugs as opposed to low molecular weight drugs (ie, small molecule or
traditional drugs). Examples of biotech products are: monofunctional
antibodies (ie, one target), bifunctional antibodies (ie, two targets),
antibody fragments, peptides, fusion proteins, plasma derived proteins,
growth factors, cytokines, various ligands and receptors, vaccines,
nucleic acids, gene therapies, cell therapies, and engineered or modified
live (or dead) tissues. Note that this is not an exhaustive list. There are
unknown types and numbers of biotech products that failed at some
stage of development. For reasons of confidentiality we will only know
about these drugs if they are brought back into development.
Route of administration
The route of administration of a biotech drug is very different
from a traditional drug taken as a pill or capsule, and each drug is
developed with a unique route of administration. These drugs are
mainly given intravenously, subcutaneously or intramuscularly. There
are also biotech drugs given to patients by intrathecal, intraarticular,
intravitreal and inhalation routes.They cannot be given orally because
they would be degraded in the gastrointestinal tract. However, the
delivery of biotech drugs orally is an active area of research.
Most biotech drugs are given in the clinic but for some chronic
indications the trend is to develop subcutaneous versions so that they
can be self-administered at home with an auto injector device. For
example, many biotech drugs indicated for rheumatoid arthritis (RA)
now have auto injectors. Rebif (interferon beta-1a), a biotech drug used
to treat multiple sclerosis, is also available with an auto injector device.
Immunogenicity
A unique characteristic of biotech drugs is that the immunogenicity of
the drug is always evaluated with immunoassays both in animal toxicity
studies and in clinical trials. Although biotech drugs are not always
immunogenic, we tend to assume they will be because the human
immune system is designed to generate antibodies to foreign or non
self biological molecules, particularly foreign proteins. We usually look for
non-specific antibodies (those that bind to any part of the drug) and
neutralising antibodies to the drug which bind to the business end (ie, the
www.topra.org

part that binds to its target). Neutralising antibodies are more important
to us because they can diminish or negate the effect of the drug and
alter its pharmacokinetics (eg, increase clearance) and pharmacodynamics.
Neutralising antibodies can diminish or eliminate the drugs therapeutic
effect. Self-generated antibodies to a biotech drug have been known
to react with the endogenous version of the protein and have serious
clinical consequences (eg, Epogen (epoetin alpha)). In a preclinical
animal toxicity study, neutralising antibodies can obviate any toxicities
and complicate interpretation of the study results. In animal studies we
can sometimes dose through the antibody effect, a phenomenon that
also occurs in humans. This means antibodies are generated and then
diminish with time and continued dosing. For some marketed drugs,
there is an inverse relationship between dose and antibody generation to
the drug.4 Note that non-neutralising and neutralising antibodies can be
generated to a biotech drug with no clinical consequences (for example
see the prescribing information for Intron A (interferon alpha-2b)). Also,
although animals may generate antibodies to biotech drugs in preclinical
experiments, this result cannot be automatically extrapolated to humans.
Immunoassays must be developed by the sponsor to detect
antibodies to the drug candidate in the serum of animals and humans.
A screening assay is usually first developed to determine if antibodies
are present and, if positive, additional assays are done for confirmation
of antibody presence and to determine if they are neutralising.
Species specificity
Another term that is frequently bandied about in the biotech drug
world is species specificity. This means that you cant take your drug
candidate and inject it into a rodent or nonrodent species in your
toxicology studies and assume this is going to be acceptable to the
regulatory agencies. In the 1980s when the interferons were being
developed, companies assumed they could use a small molecule drug
development programme for any new biotech drug. They injected
rodents with recombinant interferons and nothing happened, not
because they were not toxic, but because they have little or no
pharmacological activity in rodents. The receptors either dont exist or
are different enough from the human receptors to cause poor binding
of the drug to its putative receptor.To evaluate the toxicity of biologics,
the drug candidate must be shown to have pharmacological activity in
the animal model chosen for preclinical studies. This has necessitated
using primate models for many biopharmaceuticals, because biotech
drug receptors are not necessarily found in lower species.
But what if the drug candidate only binds to the human or
chimpanzee receptor? These types of drugs are more challenging
to develop. Some biotech drugs have been tested in chimpanzees
but the data obtained are limited to pharmacokinetic and clinical
chemistry, because necropsies are no longer permitted. In fact,
experimentation with chimpanzees is now banned in many regions,
notably the EU.
One approach for these drug programmes is to develop the
analogous molecule in a lower species. An anti-TNF mouse version of
Remicade (infliximab) (eg, cV1q) was developed and used for repeat
dose and reproductive toxicology studies in mice. Another example is
Raptiva (efalizumab), for which a surrogate antibody that binds CD11a
in mice was developed (muM17) and used in repeat dose toxicology,
www.topra.org
Focus
5
reproductive toxicology, and toxicokinetic preclinical studies.
The ICH (International Conference on Harmonisation) guidance
(ICH S6) for preclinical development of biopharmaceuticals states
that safety evaluation should normally include two relevant species
(ie, two species in which the drug is pharmacologically active).
However, when only one relevant species can be identified (or when
the biological activity of the biopharmaceutical is well understood)
it is acceptable to do toxicity testing in only one species.5 Note that
the regulatory authorities will require studies demonstrating degree
of binding of the drug to its receptor or epitope in various species to
support the choice of species for toxicity testing.
Preclinical studies
Another difference between small molecule drugs and large
molecule drugs is that the traditional toxicological studies designed
for chemicals do not always work well for biotech drugs. The two-
year rodent bioassay or carcinogenicity study, a classic study used
to evaluate carcinogenicity of chemicals, small molecule drugs, and
food additives, is never required by regulators for biotech drugs.5
The reasoning is that protein drugs do not cross the nuclear
membrane and interact with DNA. And, even if they did, there are
technical problems with this study because many biologics are not
pharmacologically active in rats and mice. Alternatively, if they do have
pharmacological activity you may encounter immunogenicity issues.
Also, there may be technical problems with dosing rodents with
large molecules repeatedly for the lifetime of the animal. If a biotech
drug in development has proliferation as a mechanism of action
there are alternative approaches to glean information on the mitotic
characteristics of the drug on normal cells and tumour cells such as
cellular proliferation assays, short term carcinogenicity studies using
transgenic animal models, assessment of the biopharmaceuticals
receptor on tumour cells, and xenograft studies (effects of drug
on tumours implanted in mice). Sponsors have also looked for
oncological diseases in chronic repeat dose toxicology studies.
Other studies not required by regulatory authorities are genotoxicity
studies for similar reasons (no putative interaction of the drug with DNA).
Also, most genotoxicity studies rely on cytotoxicity for validity, an effect
that is not feasible for a biologic drug because of their large size.6 For many
biotech drugs, genetic toxicology studies were done, especially for the
early drugs and mostly results were negative, spurious, or false positives.
A positive result for a biotech drug in a genotoxicity battery of tests may
be due only to exaggerated pharmacological response6 a common result
of any toxicity study for a biopharmaceutical. An exception to this rule
is if the drug has a free linker or is an immunoconjugate (eg, has a small
molecule drug attached to it). Note that before the ICH S6 document
was published, stating that genotoxicity tests were not required, many
biotech drugs were subjected to the standard battery of tests. This is
reflected in the labelling of many of the early drugs.
The toxicology studies developed for chemicals, drugs, and food
additives have been modified for biologic drugs. Safety pharmacology
is generally built into the repeat dose studies. Reproductive toxicology
studies are done in cynomolgus monkeys if feasible and if the drug
is pharmacologically active in this species (instead of the classic rat
model). Single dose studies using highly exaggerated human doses
Regulatory Rapporteur Vol 6, No 1, January 2009
6
focus
Focus
are often not feasible for large molecules and thus are not always
done for biopharmaceuticals. Examples of marketed biotech drugs
for which no single dose studies were submitted to FDA are Avastin
(bevacizumab), Enbrel (etanercept), and Simulect (basiliximab).
A study that is unique to antibody drugs is the tissue cross-
reactivity study, also called an immunohistochemistry study. Sponsors also various types of antibody fragments being developed as drugs.8
have to do tissue cross-reactivity studies for all antibody products
submitted to regulatory authorities. These studies are de facto for
antibody drugs but are also done for fusion protein drugs to study
nontarget binding in tissues and to support the selection of species
for toxicology studies. The US FDA antibody guidance document7
states that animal and human tissues should be fixed and stained
after exposure with the antibody drug candidate. Staining appears
at sites of antibody binding and can be very specific, nonexistent, or
ubiquitous. The goal is to determine if the drug candidate will bind to
other sites than its target. Off target binding is a safety concern.
Pharmacokinetics
The pharmacokinetics of biotech drugs are very different from small
molecule drugs.Whereas the half-lives of small molecule drugs are minutes
or hours, the half-lives of biologicals, antibody drugs for example, are days
or weeks; others have short half-lives but prolonged pharmacological
effects (eg, Epogen (epoetin alfa) ). In the small molecule world, regulators
look for safety problems around the time of Cmax in animals and in humans.
But, for biotech products, we are more concerned about total exposure
(AUC) instead of Cmax because these drugs will exhibit toxicities at any part
of the time versus concentration curves. The absorption and distribution
of biotech drugs can be more complicated than small molecule drugs,
which is reflected in the pharmacokinetics.4 The physical structure of a
biotech drug, such as an antibody, helps define the pK profile. For example,
antibody drugs that are engineered as fragments will have relatively short
half-lives compared with intact antibody drugs.4
Examples of biotech drugs and their targets
Antibody drugs are the largest class of biotech drugs and oncology and
autoimmune diseases are the most prevalent indications. These drugs
have a range of highly selective targets. Herceptin (trastuzumab) targets
the HER2 receptor on breast cancer tumours. Rituxan (rituximab)
targets the CD20 ligand (antigen) on plasma cells (B cells) and kills the
B cell population, both malignant and normal cells. It does not eradicate
very immature (progenitor) B cells so this cell population regenerates
after the end of the therapy cycle. Avastin (bevacizumab) is an antibody
drug that binds to VEGF, blocking a protein needed for growth of blood
vessels to tumours (angiogenesis).Remicade (infliximab) binds to tumour
necrosis factor (TNF) and blocks an immunostimulatory pathway
implicated in RA. TNF is a cytokine associated with the pathology of
autoimmune disease. Erbitux (cetuximab) is an antibody that blocks the
EGF receptor. A recently approved drug, Soliris (eculizumab), has a very
unique target: C5 of the complement cascade.
These drugs are just the first generation of antibodies to these
targets. There are second and third generation anti-CD20 drugs
being developed by the original sponsor. The sponsor of Remicade
has a second generation anti-TNF antibody in clinical trials. Likewise,
a promising new antibody is being developed by the sponsor of
Regulatory Rapporteur Vol 6, No 1, January 2009
Herceptin that targets both the HER2 and the HER3 receptors.
Besides second and third generation antibodies which have one
target, there are now antibodies in development with two targets,
meaning that they target two types of receptors on two different
cells (ie, BiTE antibodies that link T cells to cancer cells). There are
Some antibody drugs carry a payload or immunotoxin or
radioactive isotope to kill malignant cells. Examples of drugs of this
type are Bexxar (tositumomab + tositumomab carrying radioactive
iodine (I131)), Mylotarg (gemtuzumab ozogamicin) , which carries a
chemotherapy agent called calicheamicin, Ontak (denileukin diftitox)
which carries diphtheria toxin, and Zevalin (ibritumomab tiuxetan)
which carries two radioisotopes ( Y90 and I111).
Fusion proteins are another class of biotech drugs. Orencia
(abatacept), indicated for various autoimmune diseases, is an example of a
fusion protein that targets the CD80/86 ligands and blocks CD28, a critical
immunostimulatory pathway implicated in RA. Enbrel (etanercept), another
drug indicated for autoimmune diseases, is a fusion protein that targets
TNF. Amevive (alefacept) is a dimeric glycosylated fusion protein that binds
to the CD2 ligand on T lymphocytes and is indicated for psoriasis.
The interferon class of biotech drugs was one of the first classes
of biotech drugs to be developed in the 1980s. Examples of interferon
drugs are Avonex (interferon beta-1a), Betaseron (interferon beta-
1b), Intron A (interferon alfa-2b), Rebif (interferon beta-1a), and
Roferon-A (interferon alfa-2a).
Some of the interferon drugs are pegylated, meaning that a large
polymer (eg, polyethylene glycol, PEG) is attached to the interferon
molecule to prolong its half-life (ie, PEG-Intron (pegylated interferon alfa-
2b), Pegasys (pegylated interferon alfa-2A). There are also recombinant
copies of interleukins, another type of cytokine, on the market. Examples of
interleukin drugs are Proleukin (aldesleukin ) and Neumega (oprelvekin),
recombinant forms of IL-2 and IL-11 respectively.
Another large class of biotech drugs is the replacement enzymes.
These drugs are given to patients who have low or nonexistent levels
of a critical enzyme needed to prevent a life-threatening disease.
Marketed drugs for replacement enzyme therapy are: Cerezyme
(imiglucerase), Elaprase (idursulfase), Fabrazyme (agalsidase beta),
Naglazyme (galsulfase), and Myozyme (alglucosidase alfa).
Lastly, the drugs that stimulate progenitor cells to produce
cells lost in chemotherapy are another large class of biotech drugs.
Examples of these drugs are: Epogen (epoetin alfa or recombinant
human erythropoietin), Leukine (sargramostim or GM-CSF), and
Neupogen (filgrastim or G-CSF). Thus, the targets of these therapies
are the immature or progenitor cells for red blood cells (Epogen) and
white blood cells (Leukine, Neupogen).
Regulatory guidance for biotech drug development
There are regulatory guidances that are unique to biopharmaceuticals;
most address manufacturing and preclinical studies. For preclinical
development of biotech drugs, see the ICH S6 document. This was
written by an expert committee of the International Conference for
Harmonization by representatives of the US, EU, and Japan in the
1990s. A reassessment of this document began in 2008 by an ICH
expert committee, and is expected to be finalised in a couple of
www.topra.org

years. Another guidance document specific to the antibody class of
biotech drugs is the 1997 document on manufacture and testing of
monoclonal antibodies.7 A new CHMP guidance on antibody drugs
is soon to be published.9 There is much guidance on immunoassay
and bioassay development. The best sources of regulatory guidance
documents on development of biotech drugs are the FDA (http://
www.fda.gov/) and the EMEA (http://www.emea.europa.eu/)
websites.
Conclusion
Biotech drugs have been under development for 25 years but this is still a
new and evolving field. The first generation of drugs is being replaced by
second and third generation drugs. The future of biotech drugs includes
easier and more comfortable routes of administration for patients and
caregivers, new and improved targets, and more sophisticated engineering.
References
1 P Williams. Methods of Production of Biopharmaceutical Products and
Assessment of Environmental Impact in Preclinical Safety Evaluation of
Biopharmaceuticals, a Science-Based Approach to Facilitating Clinical Trials,
J Cavagnaro (Ed), Wiley, 21-41, 2008.
Biotech and biologicals is EU law keeping up?
London, 28 November, 2008
Reviewed by Dr Sarah Roberts, Senior Director Regulatory
Affairs, MDS Pharma Services,Wokingham, UK.
The Association of the British Pharmaceutical Industry (ABPI) held
its latest in a series of breakfast briefings at its offices in Whitehall,
London in November. The hour-long presentation, by Marjan
Noor and David Nickless from international law firm Howrey LLP,
considered whether EU law was keeping up with the development
of biotech and biological products.
Biologicals now account for an increasing share of most pharma
companies product pipeline. However, to a greater extent than
their chemical counterparts, they give rise to many unresolved legal
difficulties. For example, how much utility must a patent disclose for
an isolated gene sequence before the sequence is deserving of patent
monopoly? Are biologicals appropriately regulated with innovator
data adequately protected and does regulation of biosimilars go far
enough? On biotech inventions in the wider sense, where must the
line be drawn on morality?
The briefing covered the critical areas of biologicals development
associated with legal difficulties including the role of patents in
biologicals, the European regulation of biologicals (including data
exclusivity issues and SmPC protection), the biosimilars approach and
morality of biotech inventions.
www.topra.org
Focus
7
2 P Chamberlain. The role of immunoassays and bioassays in
biopharmaceutical development: an overview for regulatory specialists,
Regulatory Rapporteur, February 2006.
3 J Cavagnaro. The Principles of ICH S6 and the Case-by-Case Approach in
Preclinical Safety Evaluation of Biopharmaceuticals, a Science-Based Approach
to Facilitating Clinical Trials, J Cavagnaro (Ed), Wiley, 45-65, 2008.
4 W Wang, E Q Wang, and J P Balthasar. Monoclonal Antibody
Pharmacokinetics and Pharmacodynamics, Clinical Pharmacology &
Therapeutics, 84: 548-558, 2008.
5 ICH S6 Guidance: Preclinical Safety Evaluation of Biotechnology-Derived
Pharmaceuticals (http://www.fda.gov).
6 D Jacobson-Kram, H Ghantous. Genetic Toxicology Testing of
Biopharmaceuticals in Preclinical Safety Evaluation of Biopharmaceuticals,
a Science-Based Approach to Facilitating Clinical Trials, J Cavagnaro (Ed),
Wiley, 337-341, 2008.
7 Points to Consider in the Manufacture and Testing of Monoclonal
Antibody Products for Human Use. FDA/CBER, February 1997 (http://
www.fda.gov).
8 M P Deonarain. Recombinant antibodies for cancer therapy, Expert
Opin. Biol. Ther., 8:1123-1141, 2008.
9 Guideline on Production and Quality Control of Monoclonal Antibodies and
Related Substances. CHMP Draft, April 2007 (http://www.emea.europa.eu/).
The presentation detailed the legislation and applicable guidance
involved and summarised the case law associated with each area.
This allowed attendees to clearly understand the current thinking
associated with these critical areas of development.
Of particular interest was the information surrounding the emotive
area of morality issues associated with stem cell patents.The presentation
highlighted the differences in thinking between the European Patent
Office (EPO) and the UK Intellectual Property Office (UKIPO) relating
to the use of human embryos for stem cells research.The decision from
the EPO from the most recent patent application (Wisconsin Alumni
Research Foundation (WARF)/Thomson referral) which was issued the
day before the briefing was analysed. The result showed that European
patent law still prohibits the patenting of human stem cell cultures whose
preparation necessarily involves the destruction of human embryos. In
contrast, whilst the UKIPO will not grant patents relating to processes
of obtaining stem cells from human embryos or to human totipotent
stem cells, it will grant patents relating to human pluripotent stem cells
(as they do not have the potential to develop into an entire organism). It
will be interesting to see if this more lenient approach from the UKIPO is
maintained in light of the WARF/Thomson decision made by the EPO.
With more than 25 attendees from pharma companies, clinical
research organisations and law firms, the ABPI breakfast meetings continue
to be a popular forum.The briefing covered a broad range of topic areas
and was of interest both to those with an extensive background in
biologicals and to newcomers to this exciting and challenging field.
Regulatory Rapporteur Vol 6, No 1, January 2009