Technology Trends in Drug Discovery and

Development: Implications for the Development

of the Pharmaceutical Industry in Australia

Working Paper No. 3

Kim Sweeny

Pharmaceutical Industry Project

Equity, Sustainability and Industry Development
Working Paper Series

March 2002

Centre for Strategic Economic Studies

Victoria University of Technology
PO Box 14428 Melbourne City MC VIC 8001 Australia
Telephone +613 9248 1340
Fax +613 9248 1350

Email: [email protected]
Website: http://www.cfses.com
1. Introduction

The process of developing a new drug to treat an illness is long, costly and uncertain.
A number of studies have tried to estimate the cost, the most quoted figures being
those from the US Pharmaceutical Manufacturers Association (PhRMA) 1 which are
based on work done by DiMasi and others 2 at the Tufts Center in Boston. PhRMA
estimates the cost at US$500 million over a period of 11 years from the initial
research stage to the successful marketing of a new drug. More recent estimates by
DiMasi3 put the average cost at US$802 million spread over 12 years, while the
Boston Consulting Group estimates the cost as $880 million over 15 years. 4 These
estimates are averages and there is significant variation in both time and cost,
depending on the nature of the disease being targeted, the type of drug being
developed and the nature and scope of the clinical trials required to gain regulatory
approval. 5

Because of the political sensitivity of drug prices, especially in the USA, these
estimates of new drug costs have been disputed by a number of organisations. Public
Citizen, a consumer interest group, for instance, estimates average cost at closer to
$100 million6 , with these estimates in turn being disputed by PhRMA. 7

The costly nature of drug development means that pharmaceutical manufacturers must
make large investments in R&D over extended periods of time and draw heavily on
fundamental research carried out in universities and other research organisations. The
pharmaceutical industry has the highest ratio of R&D spending to sales (estimated at
18.5% in 2001) of any industry in the USA8 and this investment in R&D has been
growing by about 13% per year over recent years.

The cost of developing drugs is rising, and this, combined with a perceived decrease
in the productivity of R&D9 , has been the one of the major reasons for the mergers
and acquisitions among pharmaceutical companies over recent years, as they seek to
find and exploit economies of both scale and scope in drug R&D. 10

The evidence on the success of this strategy is somewhat mixed however.

CenterWatch for instance found that research spending and productivity declined
sharply during the three years after a merger11 , while DiMasi points to the success of
companies such as Johnson & Johnson, Lilly and Merck in maintaining relatively
high rates of new product introductions without participating in major mergers or
acquisitions 12 . Gambardella has also pointed to the ability of companies such as
Merck and Lilly to sustain competitive advantage over long periods of time. 13

The Director of Strategic Planning at Astrazeneca, has highlighted for the industry as
a whole, the “widening gap between research and development spending and the
number of new products to actually reach the market”. 14 This declining productivity is
partly due to the fact that all the simple disease targets have been addressed and those
that are left are much more difficult to address from a traditional chemistry
perspective, or their role in disease is not well understood. 15

At the same time as R&D productivity appears to be stalling, there is increasing

pressure from governments and private pharmaceutical purchasing bodies to reduce
the cost of drugs. This has lead companies to pay increasing attention to reducing the

Centre for Strategic Economic Studies 1

cost of developing drugs at each stage of the pipeline from basic research to market.
Their strategies for achieving this are to improve the portfolio of promising drug
candidates, to ensure that ineffective drug candidates are eliminated earlier in the
process, and to reduce the time that successful candidates spend in each stage.

Aside from undertaking mergers, pharmaceutical companies have responded to the

crisis in productivity by developing new capabilities internally, principally through
recruiting new staff and through the acquisition of new platform technologies. They
have also recognised that they will never be able to develop new drugs by relying
solely on their own resources 16 and have entered into a variety of agreements with
biotechnology and other companies that are working on particular disease areas, are
investigating promising drug candidates, or are developing new tools and services.

Paralleling the growth of pharmaceutical companies over the last twenty years has
been the rapid development of the biotechnology industry. Several human
biotechnology companies specialising in the development of new drugs are now large
companies in their own right and they have also engaged in increasingly widespread
alliances with the established pharmaceutical companies, other biotechnology
companies and research groups.

This greater interaction of pharmaceutical and biotechnology companies with outside

organisations has opened up the opportunities for small technology based companies
in Australia to participate more easily in the global pharmaceutical industry, and
provides an avenue for Australian research in the life sciences and allied fields to be
commercialised more effectively.

This paper describes the drug discovery and development pipeline in Section 2 before
a more detailed discussion of the impact of technology on the pipeline stages in
Section 3.

Section 4 identifies the expertise and capabilities in Australia that can contribute at
different stages and where there are deficiencies. Section 5 concludes with a
discussion of where Australia might concentrate its effort to enable a better
participation in the global industry.

Centre for Strategic Economic Studies 2

2. The Drug Discovery and Development Process

2.1 The Drug Discovery Pipeline

The process by which a new drug is brought to market stage is referred to by a

number of names – most commonly as the development chain or “pipeline”, 17 and
consists of a number of distinct stages. The description of the process by the PhRMA
is one of the most commonly used, and a modified version of this is set out in the box
on the next page.

There are various estimates of the cost of each stage of the pipeline. 18 Most show that
clinical trials is the most expensive stage and accounts for at least 40% of costs.

Table 1, for instance reports on R&D spending by pharmaceutical companies in the

USA and shows that Phase I to III clinical trials comprise 29% of cost, Phase IV trials
(post- marketing) make up 12%, with pre-clinical trials being 7%. These estimates do
not include the cost of manufacturing, marketing and distributing the final drug.

Table 1 Allocation of US Pharmaceutical R&D by Function

1999, as percentage of total R&D cost

R&D Function %

Discovery/Basic Research
Synthesis and Extraction 10.0
Biological Screening and Pharmacological Testing 14.2
Preclinical Testing
Toxicology and Safety Testing 4.5
Pharmaceutical Dosage Formulation and Stability 7.3
Clinical Trials
Clinical Evaluation Phases I, II and III 29.1
Clinical Evaluation Phase IV 11.7
Process Development for Manufacturing and Quality Control 8.3
Regulatory: IND and NDA 4.1
Bioavailability 1.8
Other 9.0

Total 100.0

Source: PhRMA Annual Survey 2001 as reported in PhRMA, Pharmaceutical Industry

Profile 2001, p15.

As mentioned in Section 1, there have been a number of estimates of the cost of

developing a new drug. Some of these, such as the study by the Boston Consulting
Group reported in Table 2, have also put a value on the cost of each stage in the
pipeline.

Centre for Strategic Economic Studies 3

 Stages in drug discovery and development19

Discovery/Basic Research
• Synthesis and Extraction – the process of identifying new molecules with
the potential to produce a desired change in a biological system
• Biological Screening and Pharmacological Testing – studies to explore
the pharmacological activity and therapeutic potential of compounds

Preclinical Testing
• Toxicology and Safety Testing – tests to determine the potential risk a
compound poses to humans and the environment, involve use of
animals, tissue cultures or other test systems
• Pharmaceutical Dosage Formulation and Stability – the process of
turning an active compound into a form and strength suitable for human
use

Regulatory Review : IND

• Application to regulatory authority to use compound in human testing. In
the US the compound is then called an Investigational New Drug (IND)

Phase I Clinical Trials

• Testing of a new compound in 20-80 healthy human volunteers to
determine tolerance, pharmacological effects, and absorption,
distribution, metabolism and excretion (ADME) patterns

Phase II Clinical Trials

• Trials in 100-300 patients with the targeted condition to determine
effectiveness in treating disease or medical condition and short term risks

Phase III Clinical Trials

• Trials on 1000-5000 patients to determine clinical benefit and incidence
of adverse reactions

Process Development for Manufacturing and Quality Control

• Engineering and manufacturing design activities to establish capacity to
produce in large volumes and to ensure stability, uniformity and overall
quality

Bioavailability Studies
• Use of healthy volunteers to show that formulation used in trials is
equivalent to product to be marketed

Regulatory Review: NDA

• Application for approval to market a new drug. In the US this is called a
New Drug Application (NDA)

Phase IV
• Post marketing trials to identify undetected adverse effects and long term
morbidity and mortality profile

Centre for Strategic Economic Studies 4

Table 2 Drug Discovery and Development Process
Boston Consulting Group, 2001

Cost Cost Time

US$m % years
Biology
Target Identification 165 18.8 1.0
Target Validation 205 23.3 2.0
Chemistry
Screening 40 4.5 .4
Optimisation 120 13.6 2.7
Development
Preclinical 90 10.2 1.6
Clinical 260 29.5 7.0

Total 880 100.0 14.7

Source: Boston Consulting Group, A Revolution in R&D, November 2001 p12.

An alternative estimate by PAREXEL, based on a composite of sources, assigns more

of the cost to the trials stages but is similar in the amount of time a drug spends at
each stage (Table 3).

Table 3 Drug Discovery and Development Process

PAREXEL, 2001

Years % of cost
Basic research 2.5 4
Discovery 3.0 15
Preclinical development 1.0 10
Phase I 1.5 15
Phase II 2.0 22
Phase III 2.5 31
FDA review and approval 1.5 3

Total 14.0 100.0

Source: PAREXEL, PAREXEL’s Pharmaceutical R&D
Statistical Sourcebook, 2001, p 96.

These estimate, as well as the one by the Tufts Center, include an opportunity cost for
the capital involved, as well as the cost of developing drug candidates that are
ultimately unsuccessful. This recognises the very high rates of failure that occur as
compounds move through the pipeline. A commonly cited ratio 20 is that for every
drug that is finally approved by the regulatory authority for sale, 5 enter Phase I
testing, and 250 enter preclinical testing after 5,000 -10,000 have been tested in the
discovery stage.

CMR International has made more precise estimates of attrition rates at each stage
based on reports from pharmaceutical companies, as shown in Table 4. 21

Centre for Strategic Economic Studies 5

Table 4 Attrition Rates for Compounds, 1998

Start of stage Probability of reaching market

%

Preclinical development 10.3

Phase I 18.4
Phase II 28.1
Phase III 65.8
FDA review and approval 90.6
Source: CMR International survey of 29 pharmaceutical companies in
1998 as reported in PAREXEL, op cit, p 195.

2.2 Types of Drugs

The traditional pharmaceutical industry has its roots in the dye and chemical industry
over 100 years ago. The first pharmaceuticals were based on the somewhat accidental
discoveries that chemicals derived from tars could have beneficial effects on some
human diseases. Experiments were undertaken to create variants of these drugs to see
if they could also be used to treat other diseases, and this approach proved to be very
successful in the discovery and development of new drugs. 22

This success created the dominant approach within the pharmaceutical industry to the
creation of new drugs, namely the synthesis of variants of small molecular weight
compounds as drug candidates. The compounds from which these variants are made
were initially discovered by a combination of accident and luck, but became
increasingly based on systematic attempts to exploit the increasing knowledge base of
chemistry, biology and medicine. Over time, companies and research groups
developed large libraries of compounds, which could be tested for effect against the
disease of interest.

A second approach to drug development was to use the body’s own biological
molecules as disease treatments. This approach had already been pioneered in the
1920s by companies such as Lilly, which developed injectable insulin for the
treatment of diabetes, but is most closely associated with the rise of biotechnology
companies over the past 30 years. These companies sought to identify which naturally
occurring biolo gical molecules are associated with disease and to use newly
discovered biotechnology methods to manufacture these compounds.

The main types of biotechnology drugs to reach the market have been monoclonal
antibodies. These are proteins in the body that are part of the immune system that
fights disease. They are made by a process of generating the antibodies in mice, then
fusing these antibodies with immortal cancer cells, which produce further quantities
of antibody through multiplication. The antibody is then separated from the culture. 23

Other biotechnology based drugs include recombinant proteins such as cytokines,

which are manufactured using recombinant DNA techniques. Here the gene which is
responsible for making the protein is spliced into the genome of a bacteria or other
vector which then produces the entity as it multiplies. The culture is then treated to
remove and purify the entity.

Centre for Strategic Economic Studies 6

Naturally occurring proteins are also made by non-recombinant techniques and
collectively these types of drugs are usually referred to as biopharmaceuticals or
biologicals.

Biologicals have a number of advantages and disadvantages when compared to

synthetic drugs. Because they are naturally occurring compounds, there is less
difficulty in convincing regulators of their safety and efficacy, which means that the
cost of clinical trials is less and they can arrive at market earlier. On the other hand,
their manufacture is more difficult and expensive, which means their price is often
higher than traditional drugs. In addition, it is generally harder to ensure purity in
manufacturing than it is for traditional drugs. The ability to manufacture biologicals in
quantity is currently a serious issue with capacity severely limited around the world. 24
This has lead to efforts to find cheaper and more effective ways of manufacturing
biologicals, such as through the use of genetically engineered crops. 25 Biologicals are
often not suitable or effective for certain diseases. Finally, biologicals are often
destroyed in the digestive system, so can only be administered by injection.

While it is often useful to differentiate established pharmaceutical companies from

biotechnology companies in the pharmaceutical industry, in practice the distinction is
becoming somewhat blurred. The established companies are increasingly turning to
biotechnology techniques and approaches to discover and test new drugs, 26 either
directly or through alliances, while the biotechnology companies are also making
small molecule drugs.

In 2001, for instance, the US Food and Drug Administration (FDA) approved 32 new
medicines of which 24 were small molecule drugs while 8 were biologicals.
Biotechnology companies were responsible for 6 drugs and 6 biologicals, while
pharmaceutical companies contributed 18 drugs and 2 biologicals – including Lilly’s
sepsis treatment Xigris, a recombinant version of human activated protein C. 27

Some of the longer established biotechnology companies are now starting to reach a
size comparable to the traditional pharmaceutical companies – Amgen for instance
has a market capitalisation of about US$80 billion. They are therefore facing exactly
the same pressures as the traditional pharmaceutical companies and this has lead to a
similar wave of mergers and acquisitions as these companies seek to overcome R&D
productivity problems and fill their drug pipelines. 28

Centre for Strategic Economic Studies 7

3. The Impact of Technologies on the Drug Development Process

This section looks at some of the technologies and techniques used by researchers at
different stages of the drug development pipeline as they seek to improve the process
of identifying promising new drug candidates and taking them to market.

As mentioned in Section 2, there are a number of ways of describing the stages of the
pipeline. In January 2001, analysts at the US finance and consulting company,
Lehman Brothers undertook a study of the impact of genomics on the drug
development process. They illustrated the potential contribution of genomics and
other technologies using a diagram, a modified version of which is reproduced as
Figure 1. It shows how these various technologies are employed from initial research
through to the clinical trials stage. 29

Figure 1 Genomics to Clinical Development

Genomics Functional Target Target High through-

genomics identification validation put screening

EST Positional Cascade Knock- Animal Disease Micro- Detection

cloning analysis out models models technology systems

Expression Proteomics Robotics

arrays
Microassays

Bioinformatics Transgenes
Chemistry

Chemical Natural
libraries products

Combinatorial Chemi-
chemistry informatics

Clinical Development Lead Chemical

development candidates optimisation hits

ADME Scale -up Computational Chemi-

chemistry biology informatics

Toxicology Structure based

design

The diagram illustrates the influence on the identification of therapeutic targets of

recent technologies in the general field of genomics, such as bioinformatics and
proteomics. Target validation is undertaken with the use of animal and disease
models. Techniques in chemistry such as combinatorial chemistry and chemi-
informatics are used to generate multiple lead compounds which are tested against the
targets using high through-put screening. Microtechnology, nanotechnology, robotics
and new array techniques are having a major influence on this screening process.
Promising drug candidates arising from this are optimised using techniques from
computational biology and structure based design, before being subject to toxicology
and other preclinical testing. Having passed all these hurdles, a drug candidate is
tested further in trials involving both healthy and sick patients.

Centre for Strategic Economic Studies 8

Many of these technologies are quite recent and subject to rapid change and
development, often in unpredictable ways. Their full potential therefore is uncertain
as is their ultimate importance. Nonetheless the development of each is being actively
pursued by research groups, start- up biotechnology companies and established drug
companies and they are all attracting substantial funding.

The following sections discuss some of these technologies, concentrating on those

that are more established.

3.1 Discovery

The drug discovery process can be described as the identification and validation of a
disease target and the discovery and development of a chemical compound to interact
with that target. This interaction can be to block, promote or otherwise modify the
activity of the target.

The history of drug development over the past century has been the accumulation of
knowledge and techniques that provide a progressively more detailed understanding
of both the target and the compound that could become a drug.

Targets are usually proteins, either those occurring within the human body of in
outside agents such as viruses and other pathogens. The major difficulty faced by drug
researchers is understanding the complex che mical pathways involved in the disease
process in order to find the most appropriate intervention point, and then to discover
or design a compound that modifies the chemical process at that point.

a. Leads/Compounds

Experience gained in the development of drugs combined with insights from rational
drug design and medicinal chemistry over an extended period of time have lead drug
researchers in pharmaceutical companies to concentrate on small molecules with a
molecular weight of less than 500 as the preferred compounds to make new drugs. 30
Larger compounds are more difficult for the body to absorb and are less stable.

However, it has been estimated that the number of possible molecules with a
molecular weight less than 500 is 10200 of which perhaps 1050 might possess drug- like
properties. 31

The pharmaceutical industry has navigated this universe of possibilities by using

insights gained over years of experience. In the process most companies have
amassed large libraries of compounds that could be possible candidates for new drugs.
These libraries combine naturally occurring compounds with those that have been
synthesised. In addition specialist companies have been formed to develop libraries
and to provide services to drug discovery groups either within exis ting large
pharmaceutical companies or other research groups.

The owners of these libraries are continually seeking to add new compounds and
some have engaged on systematic searches among animals and plants for new
molecules. This is often done in conjunc tion with government agencies within

Centre for Strategic Economic Studies 9

countries supporting a large range of unique flora and fauna. Examples of such
efforts in Australia are described in Section 4.

Once a potential disease target has been identified, the action of these compounds can
be tested against it to see which demonstrate some activity with respect to the target.
Even though promising candidates are identified by this screening process, they are
invariably not in a form suitable to be made into a drug. Medicinal chemists take these
candidates and synthesise new analogue compounds by modifying them in ways that
are expected to increase their suitability. This process traditionally was laborious - a
chemist might synthesise one compound a week using traditional techniques, submit
the compound for testing in biological assays, wait for the results, then modify the
design of the compounds and go through the same cycle again.

This was a major rate- limiting step in developing new drugs and has seen remarkable
increases in productivity over the past ten years or so through the use of combinatorial
chemistry linked to high throughput screening.

Combinatorial chemistry is an approach to chemical synthesis that enables the

creation of large numbers of organic compounds by linking chemical building blocks
in all possible combinations. Compounds are synthesised on plastic beads that are
segregated into different containers. In each container, a different chemical building
block is added to the beads. The beads from each container are then divided among a
new set of containers. When the next building blocks are added to each container,
they attach to all the first building blocks at the same time, providing all possible
combinations. 32

The process is highly automated using robots and is multiplicative so a small number
of steps can rapidly produce large libraries of compounds, for instance 390,625
unique compounds (254 ) can be generated after 4 iterations by starting with 25
compounds in 25 containers.

Combinatorial chemistry is used in this fashion for lead identification.

Once these compounds have been tested against potential drug targets, the leading
candidates can be refined further through a similar technique called parallel synthesis,
which produces large number of multiple variants of these candidates. This process
can be characterised as lead optimisation.

In order to test these large libraries of compounds against one or more targets, it has
been necessary to improve the productivity of the screening or assay process in a
similar fashion to that which has occurred with synthesis.

High-throughput screening is a highly automated robotic system that tests small

amounts of large numbers of compounds against potential targets. Protein targets are
prepared in 96-well microplates which are standardised plastic trays with 96 "wells,"
or depressions, for holding small quantities of material. The 96 wells are uniformly
located in 8 rows of 12 wells each. Recent advances have increased the number of
wells to 1536, enabling what is being called ultra high throughput screening.
Compounds are tested using multiple plates in parallel.

Centre for Strategic Economic Studies 10

As an example of the impact of combinatorial chemistry and HTS, Lilly screens about
40,000 compounds a day and expects to increase this to 100,000 when it acquires ultra
HTS technology. Lilly acquired a specialist company called Sphinx Pharmaceuticals
(now Sphinx Laboratories) in 1994 to access these technologies. 33

Specialist companies such as Pharmacopeia in the USA and Tripos in the UK provide
equipment and resources for companies wishing to carry out combinatorial chemistry
and HTS but also offer services in this area as well.

Protein targets often occur in very low amounts so it has been necessary to use
techniques used in semiconductor manufacturing as well as microtechnology to
design and manufacture these microplates and to improve the sensitivity of detection
techniques. This enables picolitre volumes of the target to be used in each test rather
than microlitres.

Most pharmaceutical and biotechnology companies now carry out some form of
combinatorial chemistry and HTS in their laboratories, and the technologies are
relatively mature. 34

The massive increase in the number of compounds and tests however can be very
expensive and there has been some scepticism about the efficacy of these
techniques. 35 Dean et al argue that HTS has not been as “attractive as hoped for two
reasons (1) the numbers of compounds that can be economically screened is small
compared to the chemical space available, and (2) the theoretical coverage of
molecular diversity within the screening set is limited”. 36

To date there appear to have been few drug candidates to emerge from these
processes, although the techniques themselves have not been widely used for very
long. Bristol-Myers Squibb has a compound BMS-201038, a MTP inhibitor in clinical
trials and Merck has a series of molecules that interact with somatostatin receptors. 37
On the other hand, 50% of the drug leads identified by GlaxoSmithKline in 2000
originated from HTS. 38

The combination of combinatorial chemistry with high throughput screening can be

described as a “brute force” or “big, dumb science” approach to testing drug
candidates. Its worth seems to be highly dependent on the quality of the initial library
of compounds, and the intelligence used to select the initial lead candidates and to
guide the screening process. 39

In addition, as Schmid et al. note “While screening large libraries allows one to obtain
leads, it is likely that further chemistry will be required to improve potency, solubility,
pharmacokinetics, and so forth. To do this requires structure activity relationships to
be followed in a learning, systematic, sequential manner. This is the traditional
domain of the medicinal chemist and single compound synthesis”. 40

Because these technologies and their associated computing hardware and software
have enabled massive increases in productivity and the numbers of drugs that can be
tested, their use is increasingly being referred to as the “industrialisation” of drug
discovery. Analogies are often drawn to similar automated processes in the
semiconductor manufacturing industry.

Centre for Strategic Economic Studies 11

A further refinement that has emerged recently is the ability to do at least some of this
work in silico, i.e. using computers to simulate the screening of lead compounds. This
technique depends crucially on knowing the 3D atomic structure of the target protein.
The compounds that can be screened are either existing compounds where a known
structure has been stored on the computer, or virtual collections of compounds whose
structure is generated by the computer. 41

3D structure is usually obtained using X-ray crystallography or nuclear magnetic

resonance (NMR) techniques.

Boston Consulting Group have estimated that in silico technologies can save US$130
million from the US$880 million required to develop a new drug and save 8 months
in development time.

b. Targets

While significant gains have been made in the technologies to generate and test
potential new drugs, the most difficult problem associated with new drug discovery is
the identification and characterisation of the most appropriate target within a disease
pathway. There are likely to be few positive outcomes from screening programs
aimed at non-validated or poorly validated targets.

Drews has estimated that all drugs developed to date address about 500 molecular
targets within the human body, with cell membrane receptors (principally G protein-
coupled receptors) and enzymes accounting for 73%. 42

Genomics seeks to exploit the findings from the sequencing of the human and other
genomes to find new drug targets.

Since the completion of the human genome sequencing programs, it has been
estimated that the human genome consists of a sequence of around 3 billion
nucleotides (the A C G T bases) which in turn probably encode 35,000 – 50,000
genes, although the actual number of genes is still unknown.

Drews estimates that the number of genes implicated in disease, both those due to
defects in single genes and those arising from combinations of genes, is about 1,000.
He only considers the 100-150 diseases that “pose a major medical problem in the
industrial world”. Based on 5 or 10 linked proteins per gene, he proposes that the
number of potential drug targets may lie between 5,000 and 10,000.

While some ge nes have been identified as contributing to disease, these have mainly
been for conditions caused by defects in single genes, and these conditions are
relatively uncommon, accounting for fewer than 2% of diseases. Where there is a
genetic component to disease however, it is more likely to be due to the interaction of
several genes and much less is known about the relationship of these genes to specific
diseases.

One technique that is now being used is to compare the genomes from both healthy
and sick people and to identify where their genomes vary. This is being done through

Centre for Strategic Economic Studies 12

the use of Single Nucleotide Polymorphism (SNP) libraries. SNPs are single
nucleotide variations in the genome sequence (eg an A rather than a T) and are
expected to provide good markers for disease genes.

Most major pharmaceuticals and biotechnology companies now have access to

genomic and SNP databases which they are using to identify suitable gene targets.

Boston Consulting Group argues that the impact of genomics technologies could save
up to US$140 million and 11 months of time on average per new drug, although the
savings are more likely to occur some time after the introduction of these technologies
when researchers have become adept in their use.

Lehman Brothers in associatio n with McKinsey & Co are more cautious. 43 Their
analysis leads them to conclude that “despite the current need for better target
validation through functional genomics, these technologies are unlikely to add value
in the near term. These technologies are simply not yet robust enough to yield truly
validated targets”.

The human genome-sequencing project was only attainable through the parallel
application of automated sequencing equipment, in a manner similar to that being
used in lead identification and optimisation. The result of this was to significantly
reduced the time and cost involved in determining the nucleotide sequences of
genomes. This has lead in turn to a rapid increase in the sequencing of other genomes,
eg for favourite research model systems such as the mouse and fruit fly. The
technology for sequencing is now becoming available for many research groups
around the world and they are applying it to a multitude of plant and animal systems.

As the technology is relatively mature and understood and has become highly
automated, attention has turned to analysing the huge amounts of genetic information
produced from these sequencing projects, particularly the comparison of sequences
among systems, and the use of SNP databases.

More importantly, there has been an increasing realisation that the focus should now
be on the proteins that the genes encode. Proteins are closer to disease processes and
drug action than genes, as most drug targets are proteins. In addition, proteins now
form a significant proportion of drugs, as recombinant proteins such as monoclonal
antibodies.

Proteomics 44 is the study of the proteome i.e. the ensemble of proteins found within
a system (sometimes referred to as structural genomics or functional genomics).
While there are may be some 35,000–50,000 genes in the human genome, they are
responsible for the production of 5000,000–1,000,000 proteins. The link from the
genome to the proteome therefore is not straightforward, there being over 100 known
biochemical post-translationa l modifications such as phosphorylation and
glycosylation. The structure and function of proteins are modified significantly
according to the nature and state of the cells in which they are found and by external
environmental factors. “Whereas the genome is static (aside from occasional
mutations… ) and is determined at conception of the organism, the proteome varies
constantly with the nature and state of the cell, making proteomics a much more
complicated endeavour”. 45

Centre for Strategic Economic Studies 13

Proteomics has been the subject of intense investment and research interest since the
sequencing of the human genome. There are moves to form an international effort
similar to the genome-sequencing project, to understand the structure and function of
all relevant human proteins. 46 In additio n several companies have announced alliances
to sequence all human proteins, although these programs seem to be unrealistically
ambitious. The action of proteins depends crucially on their shape, so there are now
proposals to understand the 3D structure of all proteins using highly automated X-ray
crystallography. 47

The established techniques for the study of proteomics analysis are 2-D gel
electrophoresis for separating proteins and mass spectrometry for analysing them.
These techniques have also been modified and improved through the application of
microtechnology techniques, in a similar fashion to high throughput screening. This
has also lead to the production of protein chip arrays similar to DNA chip arrays.

The importance and potential of proteomics has been underlined by the

Lehman/McKinsey report – “The evolving area of proteomics promises ultimately to
make knowledge of all pathways in the human body available in much the same
manner as the knowledge of genes is becoming available today. This effort, however,
is still in its infancy and current technologies have not been automated in the robust
manner of the DNA sequencers. We expect this automation to happen in the next few
years. Experts we have consulted with have confirmed that the value of proteomics
will far exceed the value of sequencing the human genome”. 48

3.2 Development

The outcome of the discovery phase is a handful of lead candidate compounds that
have shown promising activity against a drug target.

It is often at this stage that promising candidates are patented, with most patents
having a 20-year lifespan. These candidates are subject to further testing for safety
and efficacy firstly in a preclinical development stage and then in clinical trials using
human patients.

Most of the profitability arising from selling a drug occurs while it is being sold as the
sole treatment protected by patent. This window of opportunity will be maximised if
the period between grant of the patent and the drug’s commercial release – i.e. the
development stage of the drug - can be made as short as possible.

As shown in Section 2, the development stage is the most costly in terms of bringing a
drug to market, so technologies or approaches that reduce the time a drug spends in
trials, that reduce the number of patients in trials, or that make the data gathering and
data analysis more efficient, can make a two- fold contribution to a drug’s ultimate
profitability – by reducing development costs and increasing the period over which it
is marketed.

The clinical trials stage is often outsourced to Clinical Research Organisations (CRO),
which undertake the recruitment of patients and clinical research staff, the collection
of data, and the preparation of reports on the trials. While CROs are often subject to

Centre for Strategic Economic Studies 14

criticism, their use is virtually inevitable for smaller drug companies that cannot carry
the cost of in-house expertise.

Preclinical Testing

The preclinical stage is really concerned with whether the compound can be made
into a drug that will treat the disease, is not toxic and has minimum unwanted side
effects.

Toxicity tests are undertaken to show safety while pharmacokinetics testing is done to
provide data on how a drug is absorbed, distributed, metabolised and excreted
(ADME) from the body. These tests have traditionally be done on animals such as
mice, dogs and non-human primates, but fortunately these are increasingly being
replaced by tests using cell cultures, i.e. mammalian cells grown outside the body. In
addition computer systems that simulate these tests have been developed and are
beginning to be deployed.

There is a limit however in the amount of information that can be obtained from
preclinical testing. In some disease, such as depression, bipolar disorder, and
schizophrenia it is not possible to understand the efficacy of a treatment by testing in
animals or cell culture.

Over two thirds of preclinical candidates fail because they cannot be developed into
drugs either because of their toxicity or because of poor ADME properties. 49 Because
of this, methods such as combinatorial lead optimisation that automate and
miniaturise toxicity and ADME testing have been developed. These will enable lead
compounds to be tested earlier in the discovery process, before they enter the more
expensive development stage.

The other principal concern in the preclinical development stage is the

manufacturability of the drug, i.e. how to formulate the compound so that it is stable,
has the correct dosage and is suitable for economic large-scale manufacturing.

Clinical Trials

Clinical trials are used to test the efficacy and safety of new drugs in humans. In
Phase I trials, the drug is administered to a small number (20-80) of healthy
volunteers to test for toxicity and side effects and for correct dosage levels. In Phase II
this is replicated in a larger number (100-300) patients with the disease to be treated,
while in Phase II trials yet larger numbers (1,000-3,000) of patients are used to verify
the efficacy of the drug and to monitor adverse effects during lo nger-term use.

A number of technologies such as pharmacogenomics, bioinformatics, and Internet-

based technologies can and will significantly influence the clinical phases of drug
development, both in terms of better selection of patients and drugs for clinical trials
but also in the more efficient collection and analysis of data from trials.

Pharmacogenomics is based on the recognition that drugs developed for mass

markets will not work for many people who have the disease targeted. Beta blockers
do not work for between 15% and 35% of patients, tricyclic antidepressants have no

Centre for Strategic Economic Studies 15

effect on 20% to 50% of patients, while interferons are of no use to 30% to 70%50 .
Part of the reason for this variation in response among patients is due to differing
genetic makeups. As the genome is better understood, the genetic variation in
response will be correlated with other factors such as drug metabolism and
toxicokinetics to help predict how an individual patient will respond to a given drug in
terms of efficacy and safety. 51

Pharmacogenomics is expected to have a number of positive effects on drug

discovery, development and marketing. It will enable doctors to prescribe the
medicines best suited to a patient’s genetic profile as well as the optimal dose. It will
also enable drug companies to improve the selection of participants in clinical trials as
it will weed out those that will not respond to the drug. It will also rescue drugs that
might have failed in clinical trials because of adverse reactions from very small
groups in the population.

While these effects are largely positive, the possibility of multiple variants of drugs
for multiple sub groups of the population could increase the cost of drug development
and manufacture for pharmaceutical companies if individualised therapies are
demanded by patients or managed care intermediaries.

3.3 Information Technology

Clinical trials involve the collection of a large amount of data from patients by
investigators and other clinical staff. By and large this data is collected on paper
forms that need to be manually processed into a form suitable for submission to
regulatory authorities and for use by clinical researchers.

To maintain historical growth rates and meet market expectations, pharmaceutical and
biotechnology companies are increasing the number of drugs being tested. This has in
turn increased the demand for patients for trials and the number of trials undertaken.

The result is that the amount of data that needs to be collected and analysed is
increasing rapidly. Information technology enables data to be collected and analysed
more efficiently and is being increasingly deployed to improve efficiency and speed.
Call centres and data warehouses, which are used in other industries such as banking
and telecommunications, are being used to assist in patient recruitment and data
analysis. 52

Professor John Houghton covers the impact of information technology on

pharmaceuticals and heath care extensively in a report in this series. 53

Bioinformatics

The advent of combinatorial chemistry in conjunction with high throughput screening

has meant that researchers can quickly generate large volumes of data points. The
application of techniques such as mass spectrometry and X-ray crystallography for
determining the structure of proteins and the generation of nucleotide and SNP data
from genomics research have also contributed to an explosion in the amount of data
generated by researchers in pharmaceuticals and the life sciences. This has created
challenges for the computer industry in storing and managing such data.

Centre for Strategic Economic Studies 16

As a lot of genomics and proteomics research involves comparison of experimental
data with established genomic and proteomics databases, there are also significant
challenges for the computer software industry in enabling quick and accurate searches
within these databases.

These challenges have lead to the creation of a separate discipline within the life
sciences called bioinformatics and most large pharmaceutical and biotechnology
companies now have bioinformatics teams. In addition a rash of start up companies
have been formed to develop technologies and sell information from databases.
Examples include Double Twist, Lion Biosciences, Rosetta Inpharmatics and
Structural GenomiX. The company that sequenced the human genome – Celera
Genomics (now part of PE Corp) is essentially a bioinformatics company.

As computers become more powerful, it is increasingly feasible to simulate various

aspects of the drug discovery and development pipeline in silico rather than undertake
experiments or trials in the real world. This could lead to significant savings in both
time and cost. As knowledge expands, it is becoming more possible to simulate
complex interactions among targets and leads, and among all the proteins involved in
complex pathways within the body.

The complexity of these bioinformatics applications has attracted information

technology providers to the life sciences. IBM for instance is in the process of
building an advanced petaflop supercomputer to tackle Grand Challenge problems in
areas such as protein folding. It is also undertaking other research programs in pattern
discovery, protein structure and structural genomics. 54

Computer companies have entered into alliances such as Hitachi and Oracle with
Myriad Genetics to sequence the human proteome, and IBM with Proteome Systems
to identify and analyse proteins.

Part of the reason for this is that pure bioinformatics companies are having a difficult
time with a business model that relies on selling bioinformatics software and access to
databases. The market is necessarily restricted and reaching saturation as established
companies acquire these capabilities. Bioinformatics companies have therefore
responded by entering the drug discovery arena using their own tools and
information. 55

Centre for Strategic Economic Studies 17

4. Australian Capabilities in Drug Discovery and Development

4.1 Introduction

While Australia has some capabilities in all aspects of the drug discovery and
development process, its strengths have historically been concentrated in only a few
of the stages. Australia is acknowledged for the strength of its basic research in
medicine, biology and biotechnology and has developed a strong presence in clinical
trials (principally in Phase III) as a result of this strength. A recent analysis of
publication citations undertaken for the National Health and Medical Research
Council, for instance, highlights Australia’s expertise in genetics, oncology and
carcinogenesis, haematology, immunology, gastrointestinal and neurological diseases,
and the more general fields of microbiology, parasitology, virology, biochemistry and
clinical chemistry. 56

On the other hand, Australia is relatively weak in areas such as drug related
chemistry.

The sections below review Australia’s capabilities in key aspects of the drug
development process.

4.2 The Australian Pharmaceuticals and Biotechnology Industry

The Australian pharmaceuticals and biotechnology industry consists of the Australian

operations of a range of large multinational pharmaceutical companies, a few
Australian-based pharmaceutical wholesalers and manufacturers and a number of
smaller Australian biotechnology companies.

Many of the Australian subsidiaries of the multinational pharmaceutical companies

have been here for a long time and have well established distribution and marketing
operations. Some such as Merck, Sharpe & Dohme and GlaxoSmithKline have
significant formulation and manufacturing plants while others such as Eli Lilly have
made significant investments in clinical trials. Their research programs in Australia
are largely conducted through Australian university and medical research institutes.
As described below, AstraZeneca has played an important part in developing the
natural library, combinatorial chemistry and high throughout screening capabilities in
Queensland.

The larger Australian operations – Fauldings (now part of Mayne Health), Sigma
Pharmaceuticals and Australian Pharmaceutical Industries, are principally wholesalers
though Fauldings and Sigma manufacture generic drugs both on their own account
and on contract. Their involvement in technology development is generally small.

The Australian biotechnology sector consists of both listed and unlisted companies.
While there has been no exhaustive study of the complete biotechnology sector in
Australia, a number of comprehensive directories have been compiled, usually with
the support of government. Bio-Accent, a biotechnology consulting company, has
prepared directories for Victoria, Queensland and New South Wales, 57 while the
Australian Biotechnology Association has complied an on- line BioDirectory of
organisations involved in Australian biotechnology. 58

Centre for Strategic Economic Studies 18

The listed Australian biotechnology companies are tracked by both Deloitte Touche
Tohmatsu and Deutsche Bank. 59

Appendix One lists the companies in the Deloitte’s index with market capitalisation at
October 2001, as well as their classification from the Deutsche Bank index. Some
companies not appearing in the Deutsche Bank list have allocated to categories
according to their activities. Market capitalisation can vary considerably for a variety
of reasons and may be poor indicator of ultimate worth, especially for early stage
research based companies. This list does not contain Antisense Therapeutics which
was listed in November 2001.

The list is dominated by CSL with a market capitalisation of $7.1 billion. It is

primarily a blood products company with some presence in the vaccine distribution
market and with a portfolio of research projects targeting peptic ulcers, genital warts,
cervical cancer, melanoma, periodontal disease, and glandular fever.

The “Medical Devices” group of 16 companies has a collective capitalisation of about

$6.6 billion, of which Resmed and Cochlear are together worth $5.7 billion. They
make devices for sleep apnoea and profound hearing loss respectively.

The “Research Biotechnology” group includes 42 companies with a total worth of

about $1.7 billion, or an average worth of $41 million. There are 11 biotechnology
companies with a capitalisation greater than $50 million, and 30 with a capitalisation
exceeding $10 million.

Those listed companies whose operations are most closely related to the drug
discovery and development business, therefore are small by international standards,
even allowing for the fact that the cost of doing biotechnology R&D in Australia is
half that in the USA. 60

4.3 Leads

While most focus in Australian biomedical research is on understanding disease

pathways and identifying suitable targets for drugs, a number of organisations are
active in developing libraries of lead compounds and using high throughput screening
to identify promising drug candidates.

Australia has a unique and diverse biota, the country accounting for instance for about
10% of global plant biodiversity. 61

This resource has been recognised by researchers and industry as a potentially

valuable source of drug lead compounds and a number of organisations have
compiled libraries of natural compounds for this purpose.

Astrazeneca has entered into an agreement with the State of Queensland that gives the
company first rights of refusal to develop compounds based on the State’s biota, i.e.
plants and other organisms unique to the State. In return the company is helping the
State to complete its survey of the biota and providing screening facilities at Griffith
University to screen for potential new drug candidates. 62

Centre for Strategic Economic Studies 19

 BioProspect Limited 63 is a listed company based in Western Australia that has a
licence granted by the Western Australian Government giving it access to plant
species collected by the WA Herbarium. It provides profiled plant extracts to drug
discovery companies from this library as well as screening services, in conjunction
with partners such as Southern Cross University and Royal Perth Hospital. The library
has produced compounds with promise as a human sedative and an organic pesticide.

Cerylid Biosciences Ltd was founded in January 2000 when as an offshoot of Amrad
Corporation. It has a number of microbial and plant and marine macro-organism
libraries sourced from a number of Australian States and territories as well as Papua
New Guinea and Sarawak in Malaysia. It offers screening services and bioassay
development using extracts from these libraries. In addition it operates an internal
drug discovery and development program concentrating on drugs for multiple
sclerosis, endometriosis and type I diabetes.

Other companies are working on developing new forms of lead compounds,

Starpharma, for instance, is commercialising new polyvalent compounds called
dendrimers for action against a broad range of viruses and other human diseases,
including HIV/AIDS and cancer.

Discussions with senior managers from Griffith University, Starpharma, the Institute
for Molecular Bioscience and other groups working in the general area of developing
lead compounds, have identified a serious shortage in medicinal chemists in Australia.
Although courses are offered at some tertiary institutions in Australia, they are having
difficulty attracting students, partly because of the poor image of chemistry. This is a
significant bottleneck in the process of identifying lead compounds and converting
them into commercial drugs.

4.4 Targets

Of the “Research Biotechnology” companies listed in Appendix One, IDT

manufactures active ingredients while Clover Corporation manufactures lipid-based
nutrients.

Biotech Capital, Circadian Technologies, Genetic Technologies and Medica Holdings

are essentially investment companies that have supported a range of instrument
companies such as Axon Instruments, Proteome Systems, Optiscan Imaging and X-
Ray technologies, as well as small unlisted drug discovery companies, such as
Alchemia, Antisense Therapeutics, Cytopia and Xenome.

To identify the listed companies working primarily in drug discovery with a

capitalisation greater than $10 million, these other companies were removed from the
list in Appendix One. Table 5 shows these companies, as well as their technology
base and diseases targeted.

This table still includes some companies whose main activities are in biologicals
manufacture, diagnostics, and drug delivery which would reduce the list of pure drug
discovery and development companies still further.

Centre for Strategic Economic Studies 20

These drug discovery companies are primarily targeting disease where there is an
unmet need such as skin cancer, solid tumours, obesity, osteoporosis, HIV/AIDS,
Alzheimer’s disease and other inflammatory diseases.

4.5 Technologies Supporting Drug Discovery

In addition to companies working on drug leads and targets, there is a range of

companies in Australia that produce supporting technologies for drug discovery and
development.

The principal companies in this area are Axon Instruments, Gradipore, and Proteome
Systems.

Centre for Strategic Economic Studies 21

Table 5 Australian Drug Discovery Companies

Company Value* Technology base Diseases targeted

$m
Peptech 342.1 Tumour Necrosis Factor Inflammatory diseases
antibodies
Polyunsaturated fatty acids
Novogen 102.7 Development of isoflavinoids Osteoporosis, inflammatory
diseases
Gropep 83.3 Biologics manufacture Diabetic neuropathy,
In-licensing candidates for venous ulcers, oral
development mucositis, osteoporosis
Amrad 70.2 Virology and cytokines Nerve damage, hepatitis B,
severe pain, cardiovascular
disease, stroke
Metabolic 65.6 Human growth hormone Obesity, type II diabetes
Genesis R&D 59.5 DNA sequencing, transcription Tuberculosis, asthma,
regulators, cytokines psoriasis
Norwood Abbey 53.2 Mainly laser-based drug Immune based diseases
delivery, GnRH analogues
Bresagen 49.0 Interleukin GF, human GF, cell Leukaemia, rheumatoid
therapy arthritis, asthma, solid
tumours
Provalis 38.1 Vaccines Pneumonia, ear infection,
streptococcus
Agenix 37.8 Agen immunoassays, vaccines Medical diagnostics,
vaccines
Panbio 34.1 Development of diagnostics for Dengue fever, Ross River
infectious diseases fever, glandular fever
Autogen 30.3 Genomics for novel therapeutic Obesity, type II diabetes
targets
Peplin 29.0 Pharmaceuticals from plants Skin cancer, solid tumour
cancer
Biota 27.8 Rational drug design Influenza, rhinovirus
Starpharma 27.5 Development of dendrimers STDs, angiogenesis
inhibitors
Progen Industries 22.0 Biologics manufacture, inhibitors Cancer angiogenesis
of carbohydrate-protein inhibitor, anti-thrombotic
interactions inhibitor
Meditech Research 21.8 Hyaluronic acid as anti-cancer Skin cancer, bowel cancer,
drug delivery breast cancer
Anadis 21.5 Bovine colostrum Diarrhoea, osteoporosis, H
pylori
Prana Biotechnology 20.3 Oxidation proteins Alzheimer’s disease
Bionomics Ltd 16.8 Genomics Breast cancer, epilepsy
Solbec 15.9 Steroidal glycosides Cancer, mesothelioma
Pharmaceuticals
Virax Holdings 14.5 Immunotherapy vaccines HIV/AIDS
* Value of shares at October 2001.

Centre for Strategic Economic Studies 22

5. Policy Directions

New and existing technologies are likely to reshape the drug discovery and
development process in the future. Their impact will change the way in which drugs
are discovered, developed and manufactured and this will present opportunities for
Australian companies and researchers to participate in all aspect of the drug discovery
and development pipeline, from basic research through to clinical trials and
marketing.

Unlike the situation in North America and Europe, the Australian biotechnology
community is relatively immature and its companies are small by world standards. In
particular it suffers from a shortage of personnel with management and financial
experience in the pharmaceutical industry. A program to encourage expatriate
personnel with this experience to take up positions in Australian pharmaceutical and
biotechnology companies could alleviate this shortage.

These companies will require considerable nursing by governments, research

institutions and financial organisations for some time.

There is a serious lack of experienced medicinal chemists with the expertise to

convert promising lead compounds into drugs that can be marketed. This is an area
where government can be proactive in encouraging the pharmaceuticals industry in
Australia.

In addition, there is a major deficiency in the preclinical stage of drug development,

forcing companies with promising drug candidates to have the toxicology and ADME
testing done overseas. A group of commercial and research bodies are encouraging
the Commonwealth Government to fund the establishment of preclinical testing units
within existing medical research institutions and this should be supported.

Proteomics is an area in pharmaceuticals which will have major consequences for

drug discovery over the next few years. Australia has a recognised capability in this
field, through companies such as Proteome Systems and Axon Instruments that are
growing strongly and have significant strategic alliances with major companies. There
is intense interest in proteomics and it is a prime candidate for programs of support
from both government and the pharmaceuticals industry.

Centre for Strategic Economic Studies 23

Endnotes
1
As cited in a number of publications eg. PHRMA, “Pharmaceutical Industry Profile 2001”,
p17, 2001.
2
The original publication is DiMasi JA, Hansen RW, Grabowski, HG , Lasagna L, Cost of
innovation in the pharmaceutical industry”, Journal of Health Economics, 1991, 10, p 107-
142.
3
Tufts Centre for the Study of Drug Development, “Tufts Center for the Study of Drug
Development Pegs Cost of a New Prescription Medicine at $802 Million”, Press Release, 30
November 2001.
4
Boston Consulting Group, “A Revolution in R&D How Genomics and Genetics Are
Transforming the Biopharmaceutical Industry”, November 2001.
5
For a good summary of recent estimates of drug development costs see PAREXEL,
PAREXEL’s Pharmaceutical R&D Statistical Sourcebook 2001, p73.
6
Public Citizen, Rx R&D Myths: The Case Against The Drug Industry’s R&D “Scare Card”,
July 2001
7
Ernst&Young LLP, “Pharmaceutical Industry R&D Costs: Key Findings about the Public
Citizen Report”, August 2001.
8
PHRMA, “Pharmaceutical Industry Profile 2001”, p12, 2001.
9
Drews, J, Reyser S, “Innovation deficit in the pharmaceutical industry”, Drug Information
Journal, 1996, 30, p97-108.
10
See for instance,
Cockburn Iain M, Henderson Rebecca M, “Scale and scope in drug development: unpacking
the advantages of size in pharmaceutical research”, Journal of Health Economics, 2001, 20,
p1033-1057.
Henderson Rebecca, Cockburn, Iain, “Scale, scope and spillovers: the determinants of
research productivity in drug discovery”, RAND Journal of Economics, Spring 1996, pp 32-
59.
DiMasi Joseph A, Grabowski Henry G, Vernon J, “R&D Costs, Innovative Output and Firm
Size in the Pharmaceutical Industry”, International Journal of the Economics of Business,
1995, Vol2, No2, p201-219.
11
CenterWatch, “The Effect of Pharma Mergers on Development Pipelines, Productivity, and
R&D Spending”, 2000 as reported in PAREXEL, op cit, p41.
12
DiMasi Joseph A, “New Drug Innovation and Pharmaceutical Industry Structure: Trends in
the Output of Pharmaceutical Firms” Drug Information Journal, 2000, Vol 34, p1169-1194.
13
Gambardella A, Science and Innovation, The US pharmaceutical industry during the 1980s,
Cambridge University Press, 1995, Chapter 4.
14
Jarvis, Lisa, “Productivity of Big Pharma Falls After Consolidation”, Chemical Market
Reporter, April 2, 2001.
15
Schmid EF, James K, and Smith DA, “The Impact of Technological Advances on Drug
Discovery Today”, Drug Information Journal, 2001, Vol 34, pp 41-45.
16
See for instance, Mandel L, “Merck Research Laboratories : Overview and Strategy for
Growth”, August 2001.
17
Lilly calls it the “rocket”.
18
All pipeline estimates
19
Description is based on PhRMA, op cit, p24, APMA, APMA Facts Book 1999-2000, p 16.
and Office of Technology Assessment, Pharmaceutical R&D: Costs, Risks and Rewards,
February 1993, p 4-5.
20
PhRMA, op cit, p24.
21
For other estimates see PAREXEL, op cit, p175-217
22
The history of drug development can be found in American Chemical Society, The
Pharmaceutical Century Ten Decades of Drug Discovery, 2001, at www.pubs.acs.org ,
Drews, J, In Quest of Tomorrow’s Medicines, Springer, 1999, and Drews J, “Drug Discovery
: A Historical Perspective”, Science, Vol 287, 17 March 2000, pp 1960- 1964.

Centre for Strategic Economic Studies 24

23
For further information on monoclonal antibodies see Ezzell, Carol, “Magic Bullets Fly
Again”, Scientific American, October 2001, pp 29-35.
24
Ezzell, Carol op cit.
25
van Brunt, Jennifer, “Molecular Farming’s Factories”, Signals Magazine, 19 February 2002
at www.signalsmag.com.
26
The Chairman of GlaxoSmithKline, Sir Richard Sykes, for instance is quoted as saying “the
future is in molecular genetics, cell biology and the modern sciences”, in Agnew B, “When
Pharma Merges, R&D is the Dowry”, Science, Vol 287, 17 March 2000, pp 1952-1953.
27
“Twelve New Biotech Therapies in 2001”, Signals Magazine, 1 February 2002, at
www.signalsmag.com and Biotechnology Industry Organisation, “Approved Biotechnology
Drugs” at www.bio.org/er/approveddrugs.asp
28
van Brunt, J, “Inflection Point”, Signals Magazine, 5 January 2002 at www.signalsmag.com
29
Lehman Brothers, “The Fruits of Genomics”, January 2001, p12.
30
Low molecular weight is one of the Lipinski rule of 5, as described in eg Schmid EF, James
K, and Smith DA, “The Impact of Technological Advances on Drug Discovery Today”, Drug
Information Journal, 2001, Vol 34, pp 41-45.
31
Dean PM, Zanders ED, Bailey DS, “Industrial-scale, genomics-based drug design and
discovery”, Trends in Biotechnology, Vol 19, No 8, August 2001, p288.
32
Pharmacopeia, Combinatorial Chemistry,
www.pharmacopeia.com/dd/techno/tech_combinchm. See also “Combinatorial chemistry for
drug discovery”, Specialty Chemicals, 2000, p 250-252.
33
www.lilly.com/health/innovation/newdrugs/leads.html.
34
BMJ, “A revolution in drug discovery”, British Medical Journal, Vol 321, 9 September
2000, p 581-582.
35
See for instance Lahana, R, “How many leads from HTS?”, Drug Discovery Today, 1999,
4, pp 447-448; Drews J “Drug Discovery: A Historical Perspective”, Science, Vol 287, 17
March 2000, p 1960-1964; Dean et al, op cit; BMJ, op cit.
36
Dean et al, op cit.
37
BMJ, op cit.
38
Borman, Stu, “Combinatorial Chemistry”, Chemical and Engineering News, 27 August
2001, pp 49-58.
39
Bailey, DS and Brown, D “High-throughput chemistry and structure-based design, survival
of the smartest”, Drug Discovery Today, 2001, 6, pp 57-59
40
Schmid et al, op cit, p 43.
41
Dean et al, op cit, p291-292.
42
Drews J, op cit., p 1961.
43
Lehman Brothers, “The Fruits of Genomics”, January 2001. For another evaluation, see
Harris T, “Genetics, Genomics and Drug Discovery”, Medical Research Review, 2000, No 3,
pp 203 –211; Bumol TF and Watanabe AM, “Genetic Information, Genomic Technologies,
and the Future of Drug Discovery”, JAMA, Vol 285, No 5, February 7, 2001, pp 551- 555.
44
Useful sources of information on proteomics are Banks RE et al, “Proteomics: new
perspectives, new biomedical opportunities”, The Lancet, Vol 356, November 18, 2000, p
1749-1756 ; Burrill & Company, Biotech 2001, 2001
45
Frost and Sullivan, “The Total Proteomics Market”, 2000 as extracted in PAREXEL, op cit,
p 194.
46
The proposed Human Proteomics Initiative is described in O’Donovan, Claire, Apweiler,
Rolf, Bairoch, Amos, “The human proteomics initiative (HPI)”, Trends in Biotechnology, Vol
19, No5, May 2001.
47
See Service, Robert, “Can Celera Do It Again?”, Science, Vol 287, 24 March 2000, pp
2136-2138; “Celera and Syrrx Announce Structural Proteomics Collaboration”, Business
Wire, 12 December 2001; “Myriad, Hitachi, Oracle & Friedl Join Forces to Map The Entire
Human Proteome”, press release from Myriad Genetics, 4 April 2001.
48
Lehman Brothers, op cit, p 9.

Centre for Strategic Economic Studies 25

49
Ratti, E and Trist, D “Continuing evolution of the drug discovery process in the
pharmaceutical industry”, Pure and Applied Chemistry, Vol 73, No 1, pp 67-75, 2001
50
Peakman T and Arlington S, “Puttin g the Code to Work: The Promise of Pharmacogenetics
and Pharmacogenomics”, PriceWaterhouseCoopers, 2001
51
Moos, WH and Stelliou, K, “Pharmacogenomics 2000”, Drug Development Research,
2000, 49, pp 1-3.
52
King, S, “Towards a New Model of Clinical Development”, Scrips Magazine, March 2000,
available at www.pcwglobal.com.
53
Houghton, J, “Information Technology and the Revolution in Health Care”, CSES,
February 2002.
54
Allen, F et al, “Blue Gene : A vision for protein science using a petaflop supercomputer”,
IBM Systems Journal Vol 40, No 2, 200; Waszkowycz, B et al, “Large-scale virtual screening
for discovering leads in the postgenomic era”, IBM Systems Journal, Vol 40, No2, 2001
55
Hoff man, Bryce, “Bioinformatics : Time to Morph”, Signals Magazine, 14 December 2001
at www.signalsmag.com
56
Butler, L and Biglia, B, “Analysing the journal outputs of NHMRC research grants
schemes”, NHMRC, March 2001.
57
BioAccent, “Victorian Biotechnology and Bioscience Company Directory”, November
2001; Department of Innovation and Information Economy, “Biotechnology in Queensland,
Australia”, 2001; Government of New South Wales, “BioFirst NSW Biotechnology Strategy
2001”, 2001
58
At www.biomedoz.com.au .
59
Deloitte Touch Tohmatsu, “Deloitte Biotech Index”, quarterly at www.deloitte.com.au and
Deutsche Bank, “Vital Signs”, weekly.
60
Ernst & Young, Hay Group, Strategic Industry Research Foundation, “Benchmarking Study
of R&D Costs in Selected Segments of Australian Biotechnology”, January 2001.
61
Quinn Ronald J, “High-Throughput Screening in Natural Product Drug Discovery in
Australia Utilising Australia’s Biodiversity”, Drug Development Research 1999,46, pp 250-
254
62
Further information is available at www.astrazeneca.com.au.
63
Further information is available at www.bioprospect.com.

Centre for Strategic Economic Studies 26

APPENDIX ONE

Listed Australian biotechnology companies

Market Capitalisation* Segment
CSL 7,124,815,977 CSL
Resmed Inc 3,312,331,811 Medical Devices
Cochlear 2,425,178,217 Medical Devices
Peptech 342,056,934 Research Biotechnology
Axon Instruments 306,164,421 Medical Devices
IDT 178,927,736 Research Biotechnology
Vita Life Sciences 103,902,978 Medical Devices
Novogen 102,675,357 Research Biotechnology
Polartechnics 100,988,726 Medical Devices
Gropep 83,301,838 Research Biotechnology
MicroMedical Industries 79,874,626 Medical Devices
Gradipore 78,751,018 Medical Devices
Compumedics 78,400,000 Medical Devices
Amrad 70,168,340 Research Biotechnology
Circadian 65,938,589 Research Biotechnology
Metabolic 65,556,766 Research Biotechnology
Genesis Research & Development 59,546,082 Research Biotechnology
Cellestis 53,723,994 Research Biotechnology
Norwood Abbey 53,183,115 Research Biotechnology
Genetic Technologies 51,870,783 Research Biotechnology
Bresagen 48,958,704 Research Biotechnology
Chemeq 41,186,692 Research Biotechnology
Provalis 38,079,467 Research Biotechnology
Agenix 37,774,698 Research Biotechnology
Panbio 34,064,714 Research Biotechnology
Ellex Medical Lasers 33,808,228 Medical Devices
Autogen 30,253,737 Research Biotechnology
SSH Medical 29,581,094 Medical Devices
Peplin 29,000,526 Research Biotechnology
Biota 27,758,889 Research Biotechnology
Starpharma 27,532,529 Research Biotechnology
Biotech Capital 26,400,033 Research Biotechnology
Medica Holdings 26,097,636 Research Biotechnology
Optiscan Imaging 24,971,731 Medical Devices
Progen Industries 21,952,682 Research Biotechnology
Meditech Research 21,829,366 Research Biotechnology
Anadis 21,537,355 Research Biotechnology
Ambri 20,754,949 Medical Devices
Clover Corporation 20,282,780 Research Biotechnology
Prana Biotechnology 20,256,637 Research Biotechnology
Bionomics Ltd 16,789,863 Research Biotechnology
Solbec Pharmaceuticals 15,926,978 Research Biotechnology
Sirtex Medical 14,509,184 Medical Devices
Virax Holdings 14,477,092 Research Biotechnology
Bioprospect 8,930,995 Research Biotechnology
Psivida 8,062,860 Research Biotechnology
Biotron 7,353,000 Research Biotechnology

Centre for Strategic Economic Studies 27

Pharmaction Holdings 7,323,039 Manufacturing
VRI Biomedical 6,810,847 Research Biotechnology
Genesis Biomedical 6,614,590 Medical Devices
Brain Resource Company 5,957,688 Medical Devices
Australian Cancer Technologies 5,425,829 Research Biotechnology
Xcell Diagnostics 4,238,522 Research Biotechnology
Australian Vaccine Technologies 4,187,272 Research Biotechnology
Prima BioMed 4,105,976 Research Biotechnology
Aquacarotene 4,021,512 Research Biotechnology
NSL Health 3,858,206 Medical Devices
Inovax 3,758,601 Distributor
Epitan 3,391,992 Research Biotechnology
Psiron 2,249,318 Research Biotechnology
Pi2 1,235,160 Research Biotechnology
* value of shares at October 2001

Centre for Strategic Economic Studies 28