ADVANCES

Life Sciences Reports

Biomarkers in
BIOMARKERS IN CLINICAL DEVELOPMENT

Clinical Development:
Implications for Personalized
Medicine and Streamlining R&D
Author: Richard Fisler, Beachhead Consulting
Contributing Author: Olivia Scaros, PharmD
MARCH 2005

CHA CAMBRIDGE HEALTHTECH ADVISORS

Life Sciences Reports
Biomarkers in Clinical Development:
Implications for Personalized Medicine
and Streamlining R&D
By Richard Fisler, Beachhead Consulting

Contributing Author: Olivia Scaros, PharmD

Report #47, published in March 2005 by Cambridge Healthtech Advisors.

• www.chadvisors.com • Reproduction prohibited i

ii
 Biomarkers in Clinical Development:

Implications for Personalized Medicine

and Streamlining R&D
By Richard Fisler, Beachhead Consulting

Contributing Author: Olivia Scaros, PharmD

Expert Interviews
Michael T. Stocum, Personalized Medicine Partners LLC; David Lester, PhD, Pfizer Inc.; Anonymous, Merck; Felix
Frueh, MD, FDA Center for Drug Evaluation and Research; Aaron Kantor, PhD, SurroMed LLC; Robert N.
McBurney, PhD, BG Medicine; Rick Ludwig, PhD, Indiana Center for Applied Protein Sciences; Rudy Potenzone,
PhD & Dr. Richard Chen, Ingenuity Systems

About the Authors

Richard Fisler is a partner with Beachhead Consulting (www.beachhead.com), a firm specializing in technology
evaluation, strategic planning, and market assessment in the life science industry. Prior to his work at Beachhead, he
brought a wide variety of high-technology platforms to market through positions in the microarray and live-cell
microscopy industries. Additionally, he has spent 11 years in management and engineering in the medical imaging
field. Mr. Fisler can be contacted at [email protected].

Olivia Scaros graduated from the University of Illinois in 1988 with a Doctor of Pharmacy degree. She has been
employed by various pharmaceutical firms, including Sandoz Pharmaceuticals, Bayer Corporation, and Pfizer Inc.,
both as an employee and a consultant. For the past 13 years, Dr. Scaros’ main focus has been medical writing,
including study reports, clinical protocols, articles, and other projects for the pharmaceutical industry.

For more information about published CHA Advances Life Sciences Reports, visit www.chadvisors.com or call Cindy
Ohlman at 781-547-0202.

A Cambridge Healthtech Advisors publication © 2005 by Cambridge Healthtech Advisors (CHA).

This report cannot be duplicated without prior written permission from CHA.

Every effort is made to ensure the accuracy of the information presented in CHA Reports. Much of this
information comes from public sources or directly from company representatives. We do not assume any
liability for the accuracy or completeness of this information or for the opinions presented.

Cambridge Healthtech Advisors, 1000 Winter Street, Waltham, MA 02451

Phone: 781-547-0200 • Fax: 781-547-0100 • www.chadvisors.com

• www.chadvisors.com • Reproduction prohibited iii

iv • www.chadvisors.com • Reproduction prohibited
Executive Summary
The “one drug fits all” paradigm of drug discovery has been the
standard for many years, but the pharmaceutical industry is
experiencing a shift toward “the right drug at the right dose in the right
patient” approach, also known as personalized medicine. Biomarkers
have the potential to play a critical role in personalized medicine. This
report will look at the current state of biomarker development and
application, with a close look at the technologies and how they are
being deployed from research to the clinic. A biomarker discovery
discussion will include the areas of theranostics, proteomics,
pharmacogenomics, and molecular imaging. The FDA’s view on
validated and unvalidated biomarkers will be examined, as well as the
influence of pharmacogenomics on new drug therapies and biomarker
detection devices. Examples of successful biomarker programs will be
discussed. Interviews with industry leaders are presented to specify
current and future biomarker programs. Deals involving biomarkers
between pharmaceutical companies and biotech and non-profit
companies are detailed and discussed. Finally, technologies that are
required to discover and screen biomarkers are discussed, particularly
with regard to their impact on both biomarker identification and
application in wide-scale clinical use.

One goal of biomarker usage in clinical research is to expedite the drug

development process to produce drug therapies as efficiently as possible,
while maintaining the safety profile. Biomarkers have been used for
decades, from monitoring blood pressure to lipid levels.

Biomarkers can be influential in every phase of drug development, from

drug discovery and preclinical evaluations, through each phase of
clinical trials and into post-marketing studies (Figures 1.1 and 1.3).

• www.chadvisors.com • Reproduction prohibited v

Executive Summary

Figure 1.1. Phases of the Drug Development Process That

Are Impacted by Biomarkers

Source: Reprinted from Disease Markers, Volume 18, AB Kantor,

“Comprehensive phenotyping and biological marker discovery,” pages
91–97, Copyright 2002, with permission from IOS Press.

Figure 1.3. Value Proposition of Biomarkers Throughout the

Therapeutic Development and Application Phases

Source: Beachhead Consulting

vi • www.chadvisors.com • Reproduction prohibited

 Biomarkers in Clinical Development: Implications for Personalized Medicine and Streamlining R&D

Protein biomarkers and corresponding tests can be used to predict and

monitor drug response. This enables the stratification of patients into
groups that are most likely to respond to a certain drug treatment
regimen with minimal side effects. A diagnostic test that can increase
the clinical utility of a drug and reduce the risks and costs associated
with developing and marketing that drug creates synergy, which leads
to improved disease management.

Technologies used to identify and measure biomarkers are as diverse

and numerous as the biomarkers themselves. From the traditional in
vitro analyses of gene patterns, gene expression, protein expression, and
metabolite quantification, to the in vivo measurement of biological
processes in both animal and human subjects using functional imaging
technologies, the goal of these technologies is to correlate the
biomarker to clinical data. One way to describe these technologies is by
measuring the throughput (or measure of data transmission) as low,
medium, or high. The list below gives examples of some of the more
commonly used technologies available today:

•Pharmacogenomics
Microarrays

•Proteomics
2D-PAGE
Isotope-coded affinity tags
MALDI
LC-MS/MS
Imaging MS
Free flow electrophoresis
Protein arrays
Affinity-based MS techniques
Tissue arrays

•Metabolomics
2D-PAGE
Mass spectrometry
Protein arrays – Suspension and solid-support based
Tissue arrays

• www.chadvisors.com • Reproduction prohibited vii

Executive Summary

•Systems Biology
Protein pathway mapping
Modeling and predicting biological response

•Molecular Imaging
Computed Tomography (CT)
Magnetic Resonance Imaging (MRI)
Positron Emission Tomography (PET)
Single-Photon Emission Computed Tomography
(SPECT)
Biophotonic imaging

All of these technologies have greatly impacted the drug discovery

process. Figure 2.2 demonstrates the flow of information from the clinic
into discovery to enable biomarker discovery and application.

The approach to biomarker discovery and development varies within

the pharmaceutical industry. These approaches are discussed for several
pharmaceutical companies, including Pfizer, Bristol-Myers Squibb,
Roche, and Novartis. Biotechnology companies, such as SurroMed, are
also profiled, and a table listing deals made between the two industries
is included. Interviews were conducted with individuals from
Personalized Medicine Partners, Pfizer, Merck, SurroMed, BG
Medicine, Indiana Center for Applied Protein Sciences, Ingenuity
Systems, and the FDA; their insights are presented.

Many of today’s therapies that have been brought to market by

biomarker technologies are discussed, including Iressa, Herceptin,
Gleevec, Amevive, and Enbrel. Diagnostic tools associated with these
medications, including DakoCytomation’s HercepTest (Herceptin) and
Ventana Medical System’s VentanaDx c-Kit Test (Gleevec), which are
essential to the drug’s success, are examined.

The FDA’s view on biomarker research and development is highlighted

in this report. The FDA has issued guidelines to help guide the industry
with regard to biomarkers and the approval process. The agency is
encouraging pharmaceutical companies to share their
pharmacogenomic data so that both the agency and the pharmaceutical
industry can benefit from this learning process.

viii • www.chadvisors.com • Reproduction prohibited

 Biomarkers in Clinical Development: Implications for Personalized Medicine and Streamlining R&D

Figure 2.2. How Technologies Can Better Connect Discovery

and Clinical Research

Source: Beachhead Consulting

The cost to discover and develop a drug is increasing dramatically;

however, the number of approved new drug products is on the decline.
Drug manufacturers are desperately searching for ways to expedite the
drug discovery process while decreasing the expense. They are turning
to the area of biomarkers as one possible solution to this problem.
Patient-enrichment strategies use biomarkers to identify certain patient
populations that are more likely to respond to the drug therapy or to
avoid specific adverse events.

• www.chadvisors.com • Reproduction prohibited ix

Executive Summary

x • www.chadvisors.com • Reproduction prohibited

Table of Contents
CHAPTER 1
INTRODUCTION............................................................................1
1.1. Scope of Report ....................................................................................1
1.2. Overview of Biomarkers ......................................................................1
Applications in Clinical Research ................................................2
Beneficial Impacts of Biomarkers ..................................................4
1.3. Biomarkers: Definitions and Taxonomy ............................................8
1.4. The Role of Biomarkers in Drug Development ................................9
1.5. Risks Associated with Biomarker Usage ..........................................13
1.6. FDA’s Perspective on Biomarkers in Clinical Development ..........13
Accelerated Approval Provisions ................................................14
FDA’s Guidelines on Pharmacogenomic Markers ......................14
Industry’s Response to FDA’s Request for Pharmacogenomic
Data ..............................................................................................16

CHAPTER 2
FROM DISCOVERY TO CLINIC ................................................19
2.1. Pharmacogenomics ..............................................................................19
2.2. Technologies Used for Biomarker Discovery and Application ......21
Microarrays predict likelihood of breast cancer metastases ........22
Specific genes predict risk of Alzheimer’s disease ......................23
Proteomics ..........................................................................................24
2D-PAGE ....................................................................................25
Isotope-coded affinity tags ..........................................................25
MALDI ........................................................................................26
LC-MS/MS ..................................................................................26
Imaging MS..................................................................................27
Free-flow electrophoresis ............................................................28
Protein arrays ..............................................................................28
Affinity-based MS techniques for the identification of
biomarkers ....................................................................................30
Tissue Arrays ................................................................................31
Metabolomics ......................................................................................34

• www.chadvisors.com • Reproduction prohibited xi

Table of Contents

Paradigm Genetics’ focus on metabolomic biomarkers of liver

damage ........................................................................................35
Systems Biology ..................................................................................36
Molecular Imaging ..............................................................................38
Tomography-based imaging technologies....................................39
Emission-based tomography methods..........................................39
Biophotonic imaging ..................................................................41

CHAPTER 3
BIOMARKERS: APPLICATIONS IN CLINICAL
RESEARCH AND MEDICAL DIAGNOSTICS ........................43
3.1. Imaging Technologies ........................................................................44
3.2. Theranostics ........................................................................................49
Herceptin and DakoCytomation’s HercepTest ..........................50
Gleevec and Ventana Medical System’s VentanaDx c-Kit Test 51
Erbitux..........................................................................................52
Iressa and Mutations in the EGFR Gene ....................................53
Amevive and Enbrel ....................................................................53

CHAPTER 4
SUCCESSFUL BIOMARKER PROGRAMS WITHIN
INDUSTRY ......................................................................................55
4.1. Pfizer Global Research and Development ........................................55
4.2. Roche ..................................................................................................56
4.3. Bristol-Myers Squibb ..........................................................................58
4.4. Novartis ..............................................................................................61
4.5. SurroMed ............................................................................................63

CHAPTER 5
BUSINESS OUTLOOK AND CONCLUSION ..........................67
5.1. Biomarker Research Within the Pharmaceutical Industry ............67
5.2. Expert Interviews................................................................................75
5.3. Company Profiles..............................................................................110
Affymetrix ........................................................................................110
Beckman Coulter, Inc. ......................................................................111
BG Medicine Inc. ............................................................................112
BioMarker Pharmaceuticals, Inc.......................................................113
Ciphergen Biosystems, Inc. ..............................................................114
Clinical MicroArrays, Inc.................................................................116
Gene Logic Inc. ..............................................................................118
High Throughput Genomics ............................................................120
MDS Pharma Services ......................................................................121
ParAllele BioScience ........................................................................122
SurroMed ..........................................................................................124
Xenogen ............................................................................................125

xii • www.chadvisors.com • Reproduction prohibited

 Biomarkers in Clinical Development: Implications for Personalized Medicine and Streamlining R&D

References ..................................................................................................127

Glossary ......................................................................................................131

Index..............................................................................................................135

• www.chadvisors.com • Reproduction prohibited xiii

Table of Contents

FIGURES
Figure 1.1. Phases of the Drug Development Process That Are Impacted
by Biomarkers ....................................................................................................3
Figure 1.2. Effective Profitability of Approved and Released
Pharmaceuticals ..................................................................................................5
Figure 1.3. Value Proposition of Biomarkers Throughout the Therapeutic
Development and Application Phases ..............................................................7
Figure 1.4. Reasons Cited for Compound Failure ........................................11
Figure 1.5. The Biomarker Research and Development Process ................12
Figure 2.1. Linkage of a Basic Systems Biology Research Cycle with
Drug Discovery and Treatment ......................................................................38
Figure 2.2. How Technologies Can Better Connect Discovery and
Clinical Research..............................................................................................42

TABLES
Table 1.1. Biomarker/Surrogate Endpoints That Have Aided Drug
Development ....................................................................................................10
Table 1.2. Classification of Biomarkers as Described by the FDA ............15
Table 2.1. Comparative Throughput of Biomarker Technologies................21
Table 2.2. Gene Expression as a Prognostic Tool in Breast Cancer............22
Table 2.3. Genes for Early-Onset Alzheimer’s Disease ..............................23
Table 2.4. Genes for Late-Onset Alzheimer’s Disease ................................24
Table 2.5. Comparison of Proteomic Technologies Used for Biomarker
Discovery ..........................................................................................................33
Table 2.6. Comparison of Advantages and Disadvantages of Selected
Molecular Imaging Technologies ....................................................................41
Table 3.1. Pharmacological MRI Studies Involving Psychiatric and
Neurological Conditions ..................................................................................45
Table 5.1. Selected Pharmaceutical Company Partnerships ........................69

xiv • www.chadvisors.com • Reproduction prohibited

CHAPTER 1
INTRODUCTION

1.1. Scope of Report

The “one drug fits all” paradigm of drug discovery has been the
standard for many years. However, the pharmaceutical industry is
experiencing a shift toward “the right drug at the right dose in the right
patient” approach, also known as personalized medicine. Biomarkers
have the potential to play a critical role in personalized medicine. This
report will examine the current state of biomarker development and
application, with a close look at the technologies and how they are
being deployed from research to the clinic. A biomarker discovery
discussion will include the areas of theranostics, proteomics,
pharmacogenomics, and molecular imaging. The FDA’s view on
validated and unvalidated biomarkers will be examined, as well as the
influence of pharmacogenomics on new drug therapies and biomarker
detection devices. Examples of successful biomarker programs will be
discussed. Interviews with industry leaders are presented to specify
current and future biomarker programs. Deals involving biomarkers
between pharmaceutical companies and biotech and non-profit
companies are detailed and discussed. Finally, technologies that are
required to discover and screen biomarkers are discussed, particularly
with regard to their impact on both biomarker identification and
application in wide-scale clinical use.

1.2. Overview of Biomarkers

With so many new molecular entities entering the drug development
pipeline, more powerful discovery and screening technologies are much
needed to streamline the process to make therapeutic agents available
to patients as efficiently and safely as possible. One approach to achieve
this goal is to avail oneself of the multitude of analytical tools that can
assess biological parameters known as biomarkers.

• www.chadvisors.com • Reproduction prohibited 1

Introduction

Biomarkers represent the real entrée into the world of personalized

medicine. While a lot of hype and speculation has occurred in the press
about the availability of the $1,000 genome, modern medicine is a long
way from every individual having their own personal genome, or even
their own personal genotype, on a CD or implantable “biochip” that
can be used to predict disease and diagnose the appropriate therapeutic.

Biomarkers have been used for decades, primarily to diagnose and

prescribe the appropriate treatment. Blood pressure, for example, is a
biomarker that drove the development and use of antihypertensive
compounds, a market that reached $9.2 billion in the United States in
2004. Similarly, LDL cholesterol levels have driven the entire statin
market to $10.5 billion in U.S. sales in 2004, led by Pfizer’s Lipitor.

Applications in Clinical Research

Pharmaceutical companies are currently panning for new types of
biomarkers, in a targeted pharmacological gold rush. Driven by
examples coming from a small number of compounds, the industry is
looking to biomarker programs to predict clinical failure and success,
often through limiting the targeted patient population for the designed
therapeutic. So-called “patient-enrichment strategies” are under
development in an effort to create a set of biological measurements to
refine the clinical population—all geared toward a better success rate
for the drug.

In clinical trials, the measurement of biomarkers can help explain

empirical results by noting the effects of interventions on molecular
and cellular pathways and relating these to clinical responses. In this
way, biomarkers provide researchers an understanding of the differences
in clinical response that may be influenced by uncontrolled factors
(The Biomarkers Definitions Working Group, 2001).

Predicting clinical outcomes based on biological measurements is the

goal of all biomarker discovery. Successfully establishing biomarkers as
surrogate endpoints is believed to add substantial value to the modern
pharmaceutical development process, thereby accurately predicting the
success or failure of compounds before they enter the costly phases of
drug development.

Although a biomarker may be discovered in early phase research, the

ability to adequately screen for this biomarker in the context of
preclinical and clinical studies may not always be possible. The

2 • www.chadvisors.com • Reproduction prohibited

 Biomarkers in Clinical Development: Implications for Personalized Medicine and Streamlining R&D

technical feasibility of method validation and biomarker qualification

needs to be thoroughly investigated before the development plan is
finalized. Molecular imaging technologies are seen as potential methods
to successfully translate between research, preclinical, and human trials.
Other measurement technologies are applicable; however, they suffer
from a variety of issues, including specificity, ability to measure in vivo,
sensitivity, and screening costs.

Biomarkers can be beneficial to many aspects of drug development.

Biomarkers can help identify patient populations, alter
pathophysiologic mechanisms, and achieve clinical outcomes. The
ability of a biomarker to predict a patient’s response to a drug is the
challenge. Biomarkers may also help regulators approve new drug
products faster and more effectively. The goal of all of these factors, in
effect, is to provide new drug products to the patients as expediently
and safely as possible.

Biomarkers can be influential in every phase of drug development, from

drug discovery and preclinical evaluations through each phase of
clinical trials and into post-marketing studies (Figure 1.1).

Figure 1.1. Phases of the Drug Development Process That

Are Impacted by Biomarkers

Source: Reprinted from Disease Markers, Volume 18, AB Kantor,

“Comprehensive phenotyping and biological marker discovery,” pages
91–97, Copyright 2002, with permission from IOS Press.

• www.chadvisors.com • Reproduction prohibited 3

Introduction

Beneficial Impacts of Biomarkers

Attrition rates in drug development are alarming. Among the ten
largest pharmaceutical companies, during the period 1991–2000,
attrition rates were as follows:

• 38% of the drugs dropped out in Phase I due to safety/blood

levels
• 60% of those remaining failed in Phase II due to basic
efficacy failures
• 40% of the remaining candidates failed in Phase III, again
due to efficacy failures
• 23% of those that made it through the clinic failed to be
approved by the FDA

That translates to about an 11% success rate from starting in the clinic
(Kola and Landis, 2004). The question is, can biomarkers have an
Biomarkers must impact on attrition rates?
also drive the
released Patients rely on innovations such as biomarkers to maintain their
pharmaceutical health and well-being and to aid in fighting ever-more-complex
diseases. The pharmaceutical market is highly competitive, providing
into a more
incentives for pharmaceutical companies to be the first to bring a new
profitable innovation to the market. As shown in Figure 1.2, not all released
proposition for the drugs result in profits for the drug makers. In fact, 70% of released
manufacturer. compounds are less than break-even propositions. This does not take
into consideration the number of compounds that fail to be released.

Overall, innovations like biomarkers do not merely have to save time

and money in the development cycle; they must also drive the released
pharmaceutical into a more profitable proposition for the manufacturer.
This can be accomplished in a number of ways: the compound can be
significantly more effective than its competitors; the compound can
demand a higher price based on its value and effectiveness; or the
compound must be effective for its intended target market.

4 • www.chadvisors.com • Reproduction prohibited

 Biomarkers in Clinical Development: Implications for Personalized Medicine and Streamlining R&D

Figure 1.2. Effective Profitability of Approved and Released

Pharmaceuticals

Source: Pharmaceutical Research and Manufacturers of America

(PhRMA). Pharmaceutical Industry Profile 2004 (Washington, DC:
PhRMA, 2004).

Biomarker-based drug development can offer both faster development

time as well as higher efficacy based on a selected patient population. It
can, however, also be used competitively to allow companies to pursue
a leading edge over existing compounds.

Some therapeutic areas are more amenable to biomarker usage than

others. Neurology, cardiology, and oncology appear to be three
therapeutic areas generating the most biomarker research. Dr. David
Lester, New York Site Head for Worldwide Clinical Technology at
Pfizer, Inc., states, “There are therapeutic areas where biomarkers are
used more, and there are more traditional biomarkers. The
cardiovascular area is dominated by biomarkers, whereas an area like
oncology has recently exploded with biomarker research as we are
really beginning to understand the importance of the population and
the disease itself. So it is different for each therapeutic area.”

• www.chadvisors.com • Reproduction prohibited 5

Introduction

An executive at Merck expands on this: “Merck’s interest in biomarkers

is across the therapeutic areas of our current investigation, although we
recognize that there are areas of greater tractability of biomarkers in
some therapeutic areas that are better than in others, which factors into
the kind of effort we apply. For example, in neuroscience, many of the
traditional, circulating blood biomarkers are fewer and farther in
between, but an experimental model may be feasible and could be used
instead of a biomarker in that case. Also, we endeavor to make sure
that we have our biomarkers represented in the earliest drug
development studies as possible.”

By focusing as early in the process as possible, pharmaceutical

companies realize that increased profitability comes from “killing off”
bad compounds as early as possible. A lot is discussed in the press and
at conferences about clinical biomarkers that predict a patient’s
response to a drug. This is not hype, but it does represent only a small
percentage of the biomarker potential.

Novartis’ Gleevec (imatinib) targets an enzyme called Bcr-Abl, which

leukemia cells use to proliferate. Gleevec attaches to the cancerous cells
and stops them from growing and spreading. In the case of Gleevec, the
molecular translocation called the Philadelphia chromosome defines
chronic myelogenous leukemia (CML). The Philadelphia chromosome
produces a specific tyrosine kinase enzyme, Bcr-Abl, which is the target
of Gleevec. Clinicians use this mutation as a biomarker that determines
the level of receptor expression to better identify those patients who
will respond to therapy.

Gleevec is Novartis’ second-biggest product, with sales of $1.1 billion in

the first nine months of 2004. But in some patients, perhaps 12%, the
cancer cells mutate just enough to be resistant to Gleevec.

Bristol-Myers Squibb is targeting patients whose response to Gleevec

declines over time. Its drug is known by the experimental name BMS-
354825. According to a Bristol-Myers Squibb press release in December
2004, “BMS-354825 is a rationally designed oral investigational agent
that inhibits five tyrosine kinase proteins, including Bcr-Abl, the
protein that accounts for abnormal cell growth in CML, and SRC,
proteins that may play a role in imatinib resistance.”

During a trial, 31 of 36 patients with advanced CML who had not been
helped by Gleevec had a complete hematologic response to BMS-

6 • www.chadvisors.com • Reproduction prohibited

 Biomarkers in Clinical Development: Implications for Personalized Medicine and Streamlining R&D

354825, meaning their bodies stopped producing leukemia cells. This

translates to an 86% remission rate.

BMS-354825 affects a different enzyme called SRC, pronounced “sark.”

(Novartis is reported to be working on its own “Super-Gleevec.”) In
this case, Novartis has a clear advantage with its released compound,
which is specific to a set of patients with a particular biomarker.
Through further study of additional biomarkers, thereby widening the
net, Bristol-Myers Squibb may be able to address part of Novartis’
billion-dollar market.

Figure 1.3. Value Proposition of Biomarkers Throughout the

Therapeutic Development and Application Phases

Source: Beachhead Consulting

As shown in Figure 1.3, the implementation and success of biomarkers

is not limited only to the pharmaceutical industry. From the drug
companies and their partner biotechnology companies, through to the
patients and the insurance industry, better and more successful
pharmaceuticals imply better healthcare and increased profitability for
the various stakeholders involved.

• www.chadvisors.com • Reproduction prohibited 7

Introduction

1.3. Biomarkers: Definitions and Taxonomy

A biological biomarker, as defined by the Biomarkers Definitions
Working Group, is a characteristic that is objectively measured and
evaluated as an indicator of normal biological processes, pathogenic
processes, or pharmacologic responses to therapeutic intervention. A
molecular biomarker is an early sign of change in an organism’s
physiological state, such as adaptation, stress, or injury, due to
environmental factors or disease (The Biomarkers Definitions Working
Group, 2001). For example, it is possible to tell that animals have been
exposed to the toxic metal cadmium by measuring their levels of a
specific molecule that binds to cadmium, the protein called
metallothionein. Increased levels of metallothionein, as well as
increased expression of the gene that leads to elevated protein levels,
are called molecular biomarkers of exposure. Changes in molecules such
as these are sensitive and specific, making them useful sentinels of an
organism’s exposure to a specific environmental agent.
A surrogate
endpoint is Other molecular changes indicate progression of a disease process. For
expected to predict example, hemoglobin is the protein that carries oxygen in red blood
cells. A form of hemoglobin called hemoglobin 1AC is a biomarker of
clinical benefit
diabetes. As blood glucose levels increase in people with adult-onset
based on diabetes, the levels of this form of hemoglobin in their blood increase
epidemiologic, accordingly, providing a diagnostic marker of the progression of disease.
therapeutic, Measurements of hemoglobin 1AC are considered a molecular
biomarker of effect (Center for Environmental Health Sciences at
pathophysiologic, Dartmouth). A surrogate endpoint is a biomarker that is intended to
or other scientific substitute for a clinical endpoint (a characteristic or variable that
evidence. reflects how a patient feels, functions, or survives). A surrogate
endpoint is expected to predict clinical benefit (or lack of benefit or
harm) based on epidemiologic, therapeutic, pathophysiologic, or other
scientific evidence. Biomarkers can be validated (those for which
evidence has established that a drug-induced effect on the surrogate
predicts or results in the desired effect on the clinical outcome of
interest) or unvalidated (a surrogate that is “reasonably likely” to
predict the clinical benefit of interest, but for which there is not
sufficient evidence to establish this).

8 • www.chadvisors.com • Reproduction prohibited

 Biomarkers in Clinical Development: Implications for Personalized Medicine and Streamlining R&D

The Biomarkers and Surrogate Endpoint Working Group agreed on a

classification system for biomarkers based on their differences.

Type 0: Markers of the natural history of a disease which

correlate longitudinally with known clinical indices, such as
symptoms over the full range of disease states
Type I: Markers which capture the effects of an intervention
in accordance with the mechanism of action of the drug,
even though the mechanism might not be known to be
associated with clinical outcome
Type II: Markers that are considered surrogate endpoints
because change in that marker predicts clinical benefit (The
Biomarkers Definitions Working Group, 2001)

As a Type II biomarker (or surrogate endpoint) must be relevant to

both the mechanism of action of the drug and the pathophysiology of
the disease, this type of biomarker would most likely have the greatest
impact on reducing both the overall time and cost of drug There are many
development. However, Type II biomarkers are much more difficult to
examples of
develop than Type 0 or Type I.
established
biomarkers used
1.4. The Role of Biomarkers in Drug Development today in the drug

There are many examples of established biomarkers used today in the development
drug development process. Blood pressure is an accepted surrogate process.
endpoint (Type II) for antihypertensive agents, as it predicts
cardiovascular disease, heart failure, stroke, and kidney failure.
Cholesterol has long been recognized as a surrogate endpoint for
reduced mortality. Discovered more recently, a protein known as C-
reactive protein (CRP) has been shown to be a predictor of heart
disease progression. Statin drugs lower CRP, and these are the same
drugs that are used to lower cholesterol. The studies have shown that
lower levels of CRP were linked to a slower progression of
atherosclerosis and fewer heart attacks and death (Ridker et al., 2005).
Several classes of agents for bone mineral density have shown to have
good correlation with fracture rates; however, conflicting data in the
literature preclude the use of these markers as surrogates for any efficacy
endpoints in clinical practice. Table 1.1 lists biomarkers already used in
drug development.

• www.chadvisors.com • Reproduction prohibited 9

Introduction

Table 1.1. Biomarker/Surrogate Endpoints That Have Aided Drug Development

Biomarker/surrogate endpoint Type of drug Clinical endpoint

Blood pressure Antihypertensives Stroke, atherosclerosis, heart failure

Cholesterol LDL-lowering statins Coronary artery disease, heart attacks

Viral RNA Antiretroviral agents Survival, decrease in infections

HbA1C, glucose Antidiabetic agents Diabetic neuropathy

CD4+ T cells Antiretroviral agents, Sustained reduction in viral RNA

cytokines

Intraocular pressure Antiglaucoma agents Preservation of peripheral vision

Bone mineral density (BMD) Antiosteoporotic agents Fracture rate

MRI scans Agents for treatment of MS Decrease in rate of progression disease

CT scans for tumor size Anticancer agents Survival

Source: Reprinted from Disease Markers, Volume 18, CD Lathia, “Biomarkers and surrogate
endpoints: how and when might they impact drug development?” Pages 83–90, Copyright 2002, with
permission from IOS Press.

Validated molecular biomarkers have been instrumental in diagnosing

disease and have begun to assume a greater role in drug discovery and
development. Biomarkers can greatly enhance the objective of
providing more efficacious and safer drugs in an expedient manner.
However, the part the biomarker will play needs to be taken into
consideration very early in the drug development process, while the
therapeutic agent is still being identified and conceptualized.

From 1991 to 2000, the pharmaceutical industry realized the

importance of studying the metabolism of an investigational compound
in the earlier stages of drug design, thereby producing a dramatic
decrease in attrition rates due to pharmacokinetic failures (Figure 1.4;
Frank and Hargreaves, 2003).

10 • www.chadvisors.com • Reproduction prohibited

 Biomarkers in Clinical Development: Implications for Personalized Medicine and Streamlining R&D

Figure 1.4. Reasons Cited for Compound Failure

Source: Frank R, Hargreaves R. “Clinical biomarkers in drug discovery and

development.” Nature Reviews Drug Discovery. 2003;2:566–580. © 2003
Nature Publishing Group.

In order to meet the requirements set forth for pharmacokinetic

parameters, in vitro screens for absorption and metabolism have been
validated by subsequent correlation with clinical measurements. By
using this type of innovation and foresight, biomarkers can have a
major impact on attrition rates in the pharmaceutical arena.

The use of mechanism-based biomarkers (biomarkers whose activity is

mediated through the theoretical disease mechanism of action) in drug
discovery and development can assist in making cost-effective and
efficacious decisions. Their use in later phases of drug development can
be instrumental with regard to dose selection and disease prognosis.
After careful investigation of the disease under study, the identified
biomarker needs to represent a critical mechanistic process of the
disease progression and be impacted by the appropriate therapeutic
intervention. False positive results occur when it is assumed that the
biomarker is an integral part of the disease process, when in fact it is
associated in a minor way (Colburn, 2003). Biomarkers can be
categorized into three distinct compartments, based on their
contribution to the logic of the clinical plan. Although they seem to

• www.chadvisors.com • Reproduction prohibited 11

Introduction

parallel the three phases of drug development, the objective is to

deploy them as early as possible, first to confirm hitting the target and
then to test two concepts—namely, that hitting this target alters the
pathophysiological mechanism, and altering this mechanism affects
clinical status (Frank and Hargreaves, 2003) (Figure 1.5).

Figure 1.5. The Biomarker Research and Development

Process

Source: Beachhead Consulting

Another important consideration in choosing a biomarker is the

technical feasibility of method validation and biomarker qualification.
Method validation is a process of assessing the assay or measurement
performance characteristics of the biomarker. Qualification is the
evidentiary process of linking a biomarker with biological processes and
clinical endpoints. (The Development of Biomarkers for Decision-
Making in the Development and Regulatory Evaluation of New Drugs.
A Discussion Paper by the Biomarker and Genomics Working Groups,
PhRMA.)

12 • www.chadvisors.com • Reproduction prohibited

 Biomarkers in Clinical Development: Implications for Personalized Medicine and Streamlining R&D

1.5. Risks Associated with Biomarker Usage

There are several inherent risks associated with biomarker usage that
need to be investigated carefully before the decision is made to
implement the drug development plan. These include:

1. The drug affects the biomarker but does not affect the clinical
outcome. In this case, the biomarker is non-specific. If such a
biomarker is chosen in early phase clinical development, the
pharmaceutical company could end up wasting a lot of money on
clinical development that relies on an inappropriate biomarker.

2. The drug affects the biomarker and clinical outcome to a different

extent. In this case, there will be some correlation between the
biomarker and clinical outcome, but the biomarker will not be able to
fully account for the effect on clinical outcome. If the chosen
biomarker accounts for a small portion of the clinical benefit, the
pharmaceutical company could make a wrong decision to discontinue Risks associated
the development of a good drug.
with biomarker
usage need to be
3. The biomarker may be associated with only an aspect of the effects
investigated
on clinical outcome. For example, quinidine was found to suppress
cardiac arrythmias, leading to normalization of sinus rhythm. Quinidine carefully before
treats the arrhythmia, but has not been found to decrease the incidence implementing the
of sudden death associated with arrythmias; in fact, no antiarrythmic drug development
drugs have shown this. However, it also caused premature deaths
(Lathia, 2002). Like many other antiarrhythmic drugs, quinidine can plan.
provoke new lower-chamber or ventricular arrhythmias of a particular
type known as torsades de pointes. Torsades is a life-threatening
arrhythmia and can result in fainting spells, cardiac arrest, or sudden
death.

1.6. FDA’s Perspective on Biomarkers in Clinical

Development
Although current law and regulations permit the Food and Drug
Administration to base the approval of a therapeutic agent on an
unvalidated biomarker, interpreting the data of the surrogate marker as
a primary measure makes the approval process much more difficult. On
the other hand, the use of biomarkers to obtain information in early
phases of drug development is considered appropriate and
noncontroversial.

• www.chadvisors.com • Reproduction prohibited 13

Introduction

Accelerated Approval Provisions

Typically, approval has been based on well-controlled clinical trials,
which show the drug has a beneficial effect that is directly and
obviously related to the patient’s clinical status. With the advent of
HIV-related diseases in the early 1990s, it was felt that delaying
approval of products due to inability to complete trials of reasonable
duration or size was inappropriate. The Agency adopted new
regulations designed to hasten approval of important new therapies,
known as the Accelerated Approval provisions. This provision
included verbiage that drug approvals could be based on a surrogate
marker in lieu of clinical outcome. The relevant portion of the
regulation is as follows:

“The United States Food and Drug Administration (FDA) may grant
marketing approval for a new drug product on the basis of adequate and
well-controlled clinical trials establishing that the drug product has an
effect on a surrogate endpoint that is reasonably likely, based on
epidemiologic, therapeutic, pathophysiologic, or other evidence, to
FDA has set forth predict clinical benefit, or on the basis of an effect on a clinical
endpoint other than survival or irreversible morbidity.”
specific guidelines
to categorize The FDA furthermore required that drugs given accelerated approval
pharmacogenomic based on surrogate markers undergo post-launch confirmatory studies. It
is important to recognize that the above regulation applies to the use of
markers based on
unvalidated biomarkers in clinical trials—not validated biomarkers
their degree of (Katz, 2004).
validation.

FDA’s Guidelines on Pharmacogenomic Markers

The FDA has set forth specific guidelines to categorize
pharmacogenomic markers based on their degree of validation; namely,
the greater the degree of validation, the greater the FDA’s submission
requirements (Table 1.2).

14 • www.chadvisors.com • Reproduction prohibited

 Biomarkers in Clinical Development: Implications for Personalized Medicine and Streamlining R&D

Table 1.2. Classification of Biomarkers as Described by the FDA

Exploratory Probable Valid Known Valid “Regulatory

Biomarker Biomarker Biomarker Biomarker”

A biomarker based on A biomarker that has A biomarker that is not A biomarker being used to
general exploratory or not reached the status of being used as a support scientific
research information, a known valid regulatory biomarker, arguments made by the
such as broad gene biomarker because, for but for which there is sponsor about drug dosing,
expression screening, or example, the supporting widespread agreement in safety, patient selection, or
collection of sera or data has not been the scientific effectiveness; or that the
tissue samples, and that independently community about its sponsor proposes to
has not reached the replicated or is not biological significance describe in the drug label;
status of a probable conclusive. and which is measured or that are essential to
valid biomarker. in an analytical system achieve the dosing, safety,
with well-established or effectiveness described
performance in the drug label, or that
characteristics. will be used for decision
making in any clinical
trial or in an animal trial
used to support safety.

IND Voluntary data Voluntary data Abbreviated report Full report and data
submission (not used submission (not used and voluntary data submission mandatory
for FDA decision for FDA decision submission
making) making)

Existing Voluntary data Abbreviated report Abbreviated report Full report and data
NDA submission (not used OR synopsis, and and voluntary data submission mandatory
for FDA decision voluntary data submission
making) submission

New Synopsis, and Abbreviated report Abbreviated report Full report and data
NDA voluntary data and voluntary data and voluntary data submission mandatory
submission submission submission

Source: Cambridge Healthtech Advisors. “The FDA Seizes the Initiative: Implications of the Draft
Guidance on Pharmacogenomics Data Submissions.” November 2003.

• www.chadvisors.com • Reproduction prohibited 15

Introduction

By working closely with the pharmaceutical industry, the FDA has

developed this set of guidelines to provide the pharmaceutical industry
with a decisive framework on which to structure clinical studies
utilizing pharmacogenomic biomarkers.

Furthermore, new genetic biomarker discoveries, along with their

pharmacogenomic-based drug therapies, bring a different set of
challenges to the FDA and the pharmaceutical industry. Genomic
information allows the pharmaceutical industry to target a specific
patient population that is more likely to respond to the drug therapy, or
to avoid individuals who are likely to develop specific adverse events,
in their clinical studies. This patient-enrichment strategy will reduce
clinical study costs and accelerate the drug development process.

The FDA is encouraging pharmaceutical companies to utilize

pharmacogenomic data in their clinical investigations and to share this
data, as described in the guidelines set forth in November 2003:

“It is important for the FDA to have a role in the evaluation of

pharmacogenomic tests, both to ensure that evolving FDA policies are
based on the best science and to provide public confidence in the field.
It is also important that FDA policy facilitate, not impede, the use of
pharmacogenomic tests during drug development and, to the extent
possible, encourage open and public sharing of data and information on
pharmacogenomic test results.” (FDA Guidance for Industry,
Pharmacogenomic Data Submissions. November 2003)

Industry’s Response to FDA’s Request for Pharmacogenomic Data

Drug manufacturers are using pharmacogenomic information to better
understand their clinical compounds throughout the phases of clinical
studies, but are reluctant to share this information with the FDA for
fear that questions arising during the review may delay the approval of
the product.

Another reason the industry may be hesitant to provide the FDA with
this information relates to the technologies used in obtaining the
genomic information. Pharmaceutical companies are worried the FDA
may require the company to co-develop a diagnostic test to accompany
the new biomarker in order to bring the product to market. The FDA is
currently working on guidelines surrounding devices. With the
emergence of more genomic biomarkers that can identify high- and
low-responding patients, the development of the diagnostic tool that

16 • www.chadvisors.com • Reproduction prohibited

 Biomarkers in Clinical Development: Implications for Personalized Medicine and Streamlining R&D

can identify which patients should be treated with these therapies is

what the regulators are seeking.

In July 2004, drug manufacturers and FDA officials from CDER and the
Center for Devices and Radiological Health participated in a workshop
entitled “Co-development of Drug, Biological, and Device Products.”
Manufacturers voiced their concerns regarding drug-device
development. One participant noted that a “partner product” approach,
where different companies may manufacture the therapy and the
device, might be a good approach. They stated that the FDA needed to
be flexible in its guidelines, and consider different scenarios where
either the diagnostic test or the therapy could develop first, or both
develop simultaneously. “What I’ve seen lately from the FDA is very
encouraging in terms of their recognition that companion products are
going to be a part of the future and are already appearing now. They are
going to be more and more the norm in the future and we need to ask
how we proactively develop that data during the clinical development
activities that surround the pharmaceutical and leverage that data later, FDA has begun to
as the drug approaches the marketplace. The FDA has this initiative,” consider whether a
states Michael Stocum, Managing Director of Personalized Medicine
therapy already
Partners LLC, and former Director Business Development and Alliance
Management at GlaxoSmithKline’s Human Biomarkers Center. approved should
require labeling

As the guidelines are still under construction, the FDA has begun to based on new
consider whether new drug applications and therapies already approved pharmacogenomic
should require labeling based on new pharmacogenomic information information
and testing associated with it (Wechsler, 2004). For example, Strattera
associated with it.
(atomoxetine HCl, Eli Lilly) is a therapy for Attention-
Deficit/Hyperactivity Disorder (ADHD). During clinical studies, it was
found that poor metabolizers (PMs) of CYP2D6 had a 10-fold higher
area under the curve (AUC) and a 5-fold higher peak concentration to
a given dose of Strattera compared to extensive metabolizers (EMs). As
approximately 7% of the Caucasian population is PMs, the agency
recommended that wording be inserted into the labeling regarding this
issue (Package insert for Strattera, Eli Lilly and Company).

• www.chadvisors.com • Reproduction prohibited 17

Introduction

18 • www.chadvisors.com • Reproduction prohibited