Basic Science / THE INTEGRATION OF MOLECULAR DIAGNOSTICS WITH THERAPEUTICS
The Integration of Molecular Diagnostics
With Therapeutics
Implications for Drug Development and Pathology Practice
Jeffrey S. Ross, MD,1,2 and Geoffrey S. Ginsburg, MD, PhD2
Key Words: Personalized medicine; Molecular diagnostics; Pharmacogenomics; Pharmacogenetics; Toxicogenomics; Cancer predisposition;
Productivity
DOI: 10.1092/VMLL66Y5KHQ35KUE
Abstract
It is widely anticipated that during the next 5 years
the molecular diagnostic industry will continue to grow
at double-digit pace to meet increasing demand for
personalized medicine. A wide variety of drugs in late
preclinical and early clinical development are being
targeted to disease-specific gene and protein defects that
will require coapproval of diagnostic and therapeutic
products by regulatory agencies. An increasingly
educated public will demand more information about
their predisposition for serious diseases and how these
potential illnesses can be detected in an early stage
when they can be arrested or cured with new therapies
custom-designed for their individual clinical status. To
respond to this demand, major pharmaceutical
companies will partner with diagnostics companies or
develop their own in-house capabilities that will permit
efficient production of more effective and less toxic
integrated personalized medicine drug and test products.
For clinical laboratories and pathologists, this
integration of diagnostics and therapeutics represents a
major new opportunity to emerge as leaders of the new
medicine, guiding the selection, dosage, route of
administration, and multidrug combinations and
producing increased efficacy and reduced toxicity of
pharmaceutical products.
26 Am J Clin Pathol 2003;119:26-36
26 DOI: 10.1092/VMLL66Y5KHQ35KUE
Downloaded from http://ajcp.oxfordjournals.org/ by guest on July 5, 2016
Molecular pathology is in a state of rapid evolution
featuring continuous technology developments and new clin-
ical opportunities for drug selection, predicting efficacy and
toxicity, and monitoring disease outcome.1-3 The approvals of
trastuzumab (Herceptin, Genentech, South San Francisco,
CA) for the treatment of HER-2/neu overexpressing breast
cancer and imatinib (Gleevec, Novartis Pharmaceuticals, East
Hanover, NJ) for the treatment of chronic myelogenous
leukemia featuring a bcr/abl translocation and gastroin-
testinal stromal tumors with selective c-kit oncogene–acti-
vating mutations have brought to the diagnostic laboratory an
expanding role for the testing of patients to determine their
eligibility to receive these new therapies.4,5 The introduction
of targeted therapeutics into clinical practice has created
major opportunities for further development of the molecular
pathology. As shown in ❚Table 1❚, the molecular diagnostics
industry is a market greater than $3 billion, which currently is
growing at approximately 25% per year. The worldwide
pharmaceuticals annual market is estimated to exceed $1.1
trillion by 2010. It is widely held that, during the next 5 to 10
years, the clinical application of molecular pathology and
molecular diagnostics will further revolutionize the drug
discovery and development process; customize the selection,
dosing, and route of administration of existing and new thera-
peutic agents; and truly personalize medical care.6,7
Many so-called blockbuster drugs show only limited
efficacy in as many as 70% of treated patients8 ❚Table 2❚. The
current roster of phenotypically derived drugs that often treat
only the symptoms of diseases but do not cure them is no
longer acceptable to the public. Moreover, adverse drug reac-
tions caused by the failure to predict toxicity in individuals
and toxic drug-drug interactions now account for 100,000
© American Society for Clinical Pathology
 Basic Science / REVIEW ARTICLE

❚Table 1❚
Estimated Worldwide Molecular Diagnostics Markets*

2000 2005 2010

By diagnostic technique
DNA probes 240 400 850
Polymerase chain reaction–based 1,450 2,200 4,350
Fluorescence in situ hybridization 600 850 2,350
Arrays 400 1,300 3,300
Others 450 750 1,150
Total 3,140 5,500 12,000
By clinical setting
Infections 2,080 2,920 5,100
Cancer 337 900 1,900
Genetic tests 320 792 2,007
Food supply testing 110 198 400
Others 293 690 1,593
Total 3,140 5,500 11,000
* Modified from Jain KK. Molecular Diagnostics: Technologies, Markets, Companies. Basel, Switzerland: Jain PharmaBiotech Publications; 2002. Data are given in millions of
US dollars.

Downloaded from http://ajcp.oxfordjournals.org/ by guest on July 5, 2016

❚Table 2❚
Examples of the Variable Efficacy of Phenotypically
Prescribed Drugs
Frequency of Absent or
Drug Class Incomplete Efficacy (%)
Angiotensin-converting enzyme inhibitors 10-30
Beta blockers 15-25
Antidepressants 20-50
Statins 30-70
Beta2-agonists 40-70
patient deaths, 2 million hospitalizations, and $100 billion in
health care costs in the United States every year.9-12 As many
as 20% to 40% of people receiving pharmaceutical agents
may be receiving the wrong drug.9-11 The discovery of the
human genome and the subsequent expansion of proteomics
research combined with emerging technologies such as func-
tional imaging, biosensors, and sophisticated computational
biology are producing unprecedented changes in today’s
health care. A variety of new approaches to drug use have
introduced genetically prescribed medications ❚Table 3❚.
Moreover, modern medicine now is driven by rapid commu-
nication and electronic information sharing, which has
created more informed and demanding consumers.
Personalized Medicine
Personalized medicine includes the concepts that for a
given disease, the rate of progression of the disease for each
person is unique and each person responds in a unique way to
drugs.12 In its broadest sense, personalized medicine includes
the detection of disease predisposition, screening and early
disease diagnosis, assessment of prognosis, pharmacogenomic
measurements of drug efficacy and risk of toxic effects, and
the monitoring of the illness until the final disease outcome is
known ❚Figure 1❚. Emerging genomic and proteomic tech-
nologies and information are resulting in the molecular
subclassification of disease as the basis for diagnosis, deter-
mining the prognosis, and therapeutic selection. In the past,
medicine relied on nonspecific clinical signs for diagnosis
and empiric treatment strategies. For example, expanded
knowledge of the molecular basis of cancer has shown that
significant differences in gene expression patterns can guide
therapy for a variety of solid tumors and hematologic malig-
nant neoplasms.13,14 In cardiovascular disease, genetic hetero-
geneity has been identified in the long QT syndrome, and
choice of therapy (sodium channel blocker vs K+ channel
blocker vs beta blocker) now is determined by the genetic
etiology of the syndrome.15-17
Goals of Personalized Medicine
The ultimate goals of personalized medicine are to
take advantage of a molecular understanding of disease to
optimize drug development and to direct preventive
resources and therapeutic agents at the right population of
people while they are still well. The goals of personalized
medicine in drug development include the following: (1)
the selection of optimal drug targets; (2) the selection of
optimal drug dosage; (3) the selection and monitoring of
patients for shorter, less expensive, advanced clinical
trials; (4) the ability to predict which individuals will
respond to drugs at high rates and which individuals will
be less likely to have toxic effects; (5) the reduction of the
overall cost of drug development and an increase in drug
value; and (6) ultimate improvement and provision of

© American Society for Clinical Pathology Am J Clin Pathol 2003;119:26-36 27

2 DOI: 10.1092/VMLL66Y5KHQ35KUE 27
Ross and Ginsburg / THE INTEGRATION OF MOLECULAR DIAGNOSTICS WITH THERAPEUTICS

❚Table 3❚
Genetically Prescribed Medications

Target Drug Disease Action

Cytochrome P-450s Warfarin Coagulation disorders Toxicity

TPMT Mercaptopurine Cancer Toxicity
CETP HMG-CoA reductase inhibitors Atherosclerosis Efficacy
APOE-4 Tacrine Alzheimer Efficacy
HER-2 receptor Trastuzumab Cancer Efficacy
bcr-abl Imatinib Cancer Efficacy
Estrogen receptor Hormone replacements Osteoporosis Efficacy
BRCA2 Tamoxifen Cancer Efficacy
Transcriptional profiles Taxanes, anthracyclines, platinins Ovarian cancer, leukemia, breast cancer Efficacy

APOE-4, apolipoprotein E, epsilon 4; BRCA, breast cancer; CETP, cholesteryl ester transfer protein; HMG-CoA, 3-hydroxy-3-methylglutaryl coenzyme A; TPMT, thiopurine S-
methyltransferase.

more effective health care for all individuals whether well Disease outcome

Disease Management

y
or in the early or late stages of illness.

og

Symptom Severity
ol
Genetic variants can be used to predict the predisposi- Monitoring

th

Downloaded from http://ajcp.oxfordjournals.org/ by guest on July 5, 2016

pa
tion of a person for future disease development such as

ar
Pharmacogenomic

ul
cancer ❚Table 4❚. Preventive therapies currently in use for a

ec
ol
Clinical

M
variety of diseases include selective estrogen receptor modu- Prognostic
symptoms
lators for patients at risk for breast cancer 18 and
Normal Screening Subclinical
osteoporosis19 and statins for people at risk for developing
health disease process
coronary artery disease.20 Predisposition

Time

Productivity in the Pharmaceutical

Industry and Personalized Medicine
By applying the principles of personalized medicine, it
is possible to enhance substantially the productivity of drug
discovery and development. Through the identification of the
right gene, the right pathway, and the right target in the
pathway, the right drug can be developed to treat the right
patient. Estimated potential reductions by 20% in the
number of new compounds tested in phase 2 and 3 human
clinical trials, by 50% in the number of patients in phase 2
trials, by 10% in the number of patients in phase 3 trials, and
by 20% in the length of phase 3 trials are well within reach.8
The net result of biomarker-assisted drug development is an
overall reduction of compound failures in clinical trials and
on the market and a cost savings of perhaps as much as $500
million for each drug launched.
Pharmacogenetics and Single
Nucleotide Polymorphisms
An individual’s response to a drug is the complex inter-
action of both genetic and nongenetic factors. Genetic vari-
ants in the drug target itself, disease pathway genes, and drug
metabolizing enzymes all may be used as predictors of drug
efficacy or toxicity. More than 1 million genetic markers
❚Figure 1❚ Personalized medicine. Personalized medicine
includes the detection of disease predisposition, screening
and early disease diagnosis, prognosis assessment,
pharmacogenomic measurements of drug efficacy and risk of
toxic effects, and the monitoring of the illness until the final
disease outcome is known.
known as single nucleotide polymorphisms (SNPs) have
become available for genotyping and phenotyping stud-
ies.21-24 The use of a variety of techniques for cloning and
sequencing and an evolving number of informatics
approaches leading to the successful mining of high-quality
sequence variations have revealed numerous loci that seem
to have substantial potential to generate clinically useful data
for patient management. On a regular basis, novel geno-
typing strategies are emerging that use a variety of tech-
niques, including oligonucleotide genomic arrays,21,22 gel
and flow cytometry, classic gene sequencing, mass spec-
troscopy,23 and microarray or gene chips,24 all designed to
increase the rate of data generation and analysis. In 1999, 10
companies and the Wellcome Trust formed a consortium to
discover clinically relevant SNPs. To date, more than 2.1
million SNPs have been deposited into the SNP Consor-
tium’s public database (http://www.snp.cshl.org). A high-
resolution SNP map will expedite the identification of genes

28 Am J Clin Pathol 2003;119:26-36 © American Society for Clinical Pathology

28 DOI: 10.1092/VMLL66Y5KHQ35KUE
 Basic Science / REVIEW ARTICLE

❚Table 4❚
Cancer Predisposition and Tumor Suppressor Genes

Gene Name Designation Location Disease Association

Adenomatous polyposis coli APC 5q21 Familial polyposis coli

Breast cancer 1 BRCA1 17q21 Familial breast cancer
Breast cancer 2 BRCA2 13q12 Familial breast cancer
E-cadherin CDH1 16q22 Familial stomach cancer; lobular breast cancer (also
determining prognosis in epithelial cancers)
Cyclin-dependent kinase inhibitor 1C CDKN1C 16q22 Beckwith-Wiedemann syndrome; Wilms tumor;
rhabdomyosarcoma
p16 Cyclin-dependent kinase inhibitor 2A CDKN2A 11p15 Familial melanoma; bladder cancer; other epithelial tumors
(MTS 1; INK4A)
Mitogen-activated protein kinase 4 MAP2K4 17p11 Pancreatic, breast, colorectal cancer
Multiple endocrine neoplasia type 1 (type 2 MEN1 11q13 Pituitary adenoma; parathyroid adenoma; islet cell
associated with the ret oncogene) carcinoma; carcinoids
Hereditary nonpolyposis colorectal cancer type 1 hMSH2 2p22 Colorectal cancer; ovarian cancer; endometrial cancer
Hereditary nonpolyposis colorectal cancer type 1 hMLH1 3p21 Colorectal cancer; ovarian cancer; endometrial cancer
Neurofibromatosis type 1 NF1 17q11 Neurofibroma; glioma; pheochromocytoma; acute
myelogenous leukemia
Neurofibromatosis type 2 NF2 22q12 Schwannoma; meningioma; ependymoma
Phosphatase and tensin homologue PTEN 10q23 Cowden syndrome; glioma; prostate cancer; endometrial cancer
Retinoblastoma RB1 13q14 Retinoblastoma; osteosarcoma; small cell lung cancer;

Downloaded from http://ajcp.oxfordjournals.org/ by guest on July 5, 2016

breast cancer
Serine/threonine kinase 11 STK11 19p13 Peutz-Jeghers syndrome
TP53 p53 17p13 Li-Fraumeni syndrome; inactive in >50% of human malignant
neoplasms
Tuberous sclerosis 1 TSC1 (KIAA023) 9q34 Hamartoma; renal cell carcinoma; angiomyolipoma
Tuberous sclerosis 2 TSC2 16p13 Hamartoma; renal cell carcinoma; angiomyolipoma
von Hippel-Lindau VHL 3p26 Hemangioma; renal cell carcinoma; pheochromocytoma
Wilms tumor 1 WT1 11p13 Genitourinary dysplasia; familial Wilms tumor

for complex diseases such as asthma, diabetes mellitus,
atherosclerosis, and psychiatric disorders. In oncology, SNP
technology has focused on detecting the predisposition for
cancer, predicting toxic responses to drugs, and selecting the
best individual and combination anticancer drugs.
Cancer Predisposition Testing
SNP genotyping and gene sequencing (see “Predisposi-
tion, Screening, and Early Diagnosis of Disease”) have
revealed a variety of familiar cancer predisposition
syndromes (Table 4) based on single and multiple gene vari-
ants.24-27 Genotyping has been introduced widely for the
detection of familial cancers of the breast, ovary, and colon;
melanoma; and multiple endocrine neoplasia.26,28-30
Prediction of Drug Toxicity
One of the earliest applications of SNP genotyping in
cancer management was the discovery of variations in drug
metabolism associated with genomic variations in drug-
metabolizing enzymes such as the cytochrome system.31-33
Other detoxification pathways associated with SNP-based
variations in drug metabolism have been described,32-34 but,
in comparison with the cytochrome gene system, these
markers have not achieved substantial clinical usefulness.
The potential clinical value of the pharmacogenetics
approach for predicting drug toxicity will be revealed as
more candidate polymorphisms are discovered.
Prediction of Drug Efficacy
The application of genotyping strategies to predict anti-
cancer drug efficacy has emerged in a variety of clinical
settings.35-39 Genotype resistance testing of HIV isolates has
demonstrable clinical usefulness and provides a way to assist
therapeutic decision making for patients whose HIV RNA
levels are rising.40 In colorectal cancer, pretreatment geno-
typing on peripheral blood samples is being used to select
therapy based on the prediction of resistance associated with
certain genetic polymorphisms.39
Pharmacogenomics
Pharmacogenomics can best be defined as the applica-
tion of whole genome technologies (eg, gene and protein
expression data) for the prediction of the sensitivity or
resistance of an individual’s disease to a single drug or
group of drugs.41,42
Transcriptional Profiling and Genomic Microarrays
The development of printed and spotted genomic
microarrays has permitted the rapid accumulation of new
information about gene mutation and expression in human
malignant neoplasms.43-50 Microarrays can be used to deter-
mine gene mutations and SNPs and to provide rapid
screening information about messenger RNA expression. In

© American Society for Clinical Pathology Am J Clin Pathol 2003;119:26-36 29

2 DOI: 10.1092/VMLL66Y5KHQ35KUE 29
Ross and Ginsburg / THE INTEGRATION OF MOLECULAR DIAGNOSTICS WITH THERAPEUTICS

transcriptional profiling, the relative expression levels of
messenger RNA of both characterized and novel genes
from a tumor can be compared with the gene expression
status of a reference sample. Reference samples can
include cell lines, normal tissues, and cancer precursor
lesions. 43-50 Transcriptional profiling has the ability to
generate hundreds of thousands of data points, and it
requires sophisticated and complex information systems for
accurate and useful data analysis. Complementary DNA
(cDNA) microarrays were introduced during the mid-
1990s51 and have achieved widespread use for the expres-
sion profiling of human clinical samples and in drug and
biomarker target discovery. The microarray technique has
generated a wealth of new information in the fields of
leukemia and lymphoma and solid tumor classification,
drug and biomarker target discovery, and pharmacoge-
nomic drug efficacy testing43-50 ❚Image 1❚. Oligonucleotide
of anticancer drugs has been used as a guide to anticancer
drug therapy. 50-56 By using cDNA microarrays, several
groups have reported their success at discovering gene
expression that can be linked to resistance and responsive-
ness to standard-of-care chemotherapy.57,58 During the next
several years, the ability of this approach to personalize the
treatment of newly diagnosed cancer patients with individu-
alized selection and dosage of chemotherapeutic agents will
be tested on a large scale.
Pharmacodynamics
In addition to the pharmacogenetic and pharmacoge-
nomic tests, a series of bioassays, gene expression profiles,
and tissue-based biomarkers have emerged to guide the
dosage, timing, and route of administration for novel thera-

Downloaded from http://ajcp.oxfordjournals.org/ by guest on July 5, 2016

microarrays have been used mostly to detect SNPs and
individual gene point mutations and for profiling differ-
ences in gene expression between samples from human
tumors and reference test tissues. The oligonucleotide
microarray technique is, however, well-regarded as a
method of automated and robotic sampling of the genome
in search of SNPs and other clinically relevant polymorphisms.52
Transcriptional Profiling and Pharmacogenomics
The hierarchical clustering technique of data analysis from
transcriptional profiling of clinical samples known to have
responded or been resistant to a single agent or combination
peutics. These tests are used to speed the preclinical and early
clinical development of drugs by enhancing the achievement
of the ideal therapeutic dose while avoiding dose-related toxic
effects. Recent examples of these novel assays include the
blood-based bioassay of the proteasome designed to guide the
use of the proteasome inhibitor MLN-341 in multiple
myeloma and solid tumors59-61 and the immunohistochemical
determination of the epidermal growth factor receptor
(EGFR) target and cell proliferation markers in skin biopsy
specimens of patients receiving the anti-EGFR small mole-
cule tyrosine kinase inhibitor geftinib (Iressa, AstraZeneca
Pharmaceuticals, Manchester, England).62

❚Image 1❚ Breast cancer complementary DNA (cDNA) microarray results evaluated by hierarchical gene cluster analysis for defining
specific gene expression signatures. Hierarchical clustering algorithm permits the clustering of individual tumor profiles on the
basis of their similarities to their coexpression with the estrogen receptor alpha gene. Each row represents a tumor and each
column a single gene. Red indicates up-regulation; green, down-regulation; and black, no change in relative gene expression.

30 Am J Clin Pathol 2003;119:26-36 © American Society for Clinical Pathology

30 DOI: 10.1092/VMLL66Y5KHQ35KUE

Toxicogenomics
Toxicogenomics is the study of gene expression
patterns using high-throughput microarrays, automated
reverse transcriptase–polymerase chain reaction, nuclear
magnetic resonance, and proteomic strategies designed to
detect up-regulation and down-regulation of genes associ-
ated with the risk of drug toxic effects.63-66 Toxicogenomic
markers for adverse effects may influence selection and
optimization of lead compounds before human studies. For
example, inherited forms of the long QT syndrome can be
caused by high-affinity drug blockade associated with
mutations in the HERG potassium channel regulatory
gene.17,67 The limitations and uncertainties of gene expres-
sion profiling associated with data mining constraints on
mechanistic and predictive toxicology have been empha-
sized, and the clinical value of toxicogenomics is mostly
unrealized.66
Targeted Therapies
The integration of diagnostics and therapeutics is real-
ized in the discovery and development of targeted therapies.
Targeted therapy has emerged in new cancer treatment
strategies as evidenced in the approvals of trastuzumab and
imatinib. 4,5 The features of the ideal anticancer target
designed to maximize efficacy and minimize toxic effects
include the following: (1) The target is crucial for the exis-
tence of the malignant phenotype, not significantly
expressed in vital organs and tissues, a biologically relevant
molecular feature, measurable in readily obtained clinical
samples in a reproducible manner, and clearly correlated
with clinical outcome. (2) Interruption of, interference with,
or inhibition of the target yields a clinical response in a
substantial proportion of patients whose tumors express the
target. (3) Responses in patients whose tumors do not
express the target are minimal.
In 1999, the US Food and Drug Administration
approvals of trastuzumab and HercepTest (DAKO, Carpin-
teria, CA) for patients with HER-2/neu overexpressing
breast cancers marked the first simultaneous regulatory
review of a targeted therapeutic with its companion diag-
nostic test.4,13,68,69 A variety of modifications of the HER-
2/neu testing platform have been introduced and are
designed to achieve the best prediction of trastuzumab
response ❚Image 2❚. This integration of drug selection based
on a diagnostic test result remains among the foremost of
targeted therapy strategies currently available for cancer
patients. Another recent example of targeted therapy in
oncology in which a diagnostic test specifically directs
therapy is the detection of either the bcr/abl translocation in
© American Society for Clinical Pathology
Basic Science / REVIEW ARTICLE
patients with chronic myelogenous leukemia or activating
mutations in the c-kit tyrosine kinase receptor in patients
with gastrointestinal stromal tumors for the selection of the
small molecule drug, imatinib.5,70,71 A substantial number
of small molecule drugs in current clinical trials are targeted
to a variety of growth factor receptors, angiogenesis
promoters, cell cycle regulators, and invasion or metastasis
biomarkers.72-74 Similarly, a variety of antitumor antibodies
have been targeted specifically to cell surface antigens and
designed to deliver cytotoxic radioisotopes75 and cellular
toxins.76 During the next 5 years, the impact of the targeted
approach in which diagnostic tests guide the clinical selec-
tion of anticancer drugs will become known.
Integrated Diagnostics and Therapeutics
in Nonmalignant Diseases
Downloaded from http://ajcp.oxfordjournals.org/ by guest on July 5, 2016
A wide variety of existing and novel diagnostic tests
have been integrated into the management strategy of many
major nonmalignant diseases. A number of groups have
studied patients with rheumatoid arthritis to learn whether
germline-based SNP detection or biomarker assays
performed on serum or synovial fluid could predict the
development of erosive disease and guide the selection of
therapy.77,78 Patients with inflammatory bowel disease have
been tested for biomarkers of disease severity as a method
of selecting local vs systemic drug therapy.79,80 In addition to
classic measurements of known risk factors, new strategies
for predicting the risk of early-onset coronary atheroscle-
rosis have emerged that use well-established traditional
biomarkers such as serum C-reactive protein levels in new
ways81 and novel biomedical imaging and molecular diag-
nostic tests.82,83 Similarly, new integrations of genotyping
and molecular diagnostics are widening the knowledge base
and influencing the management of people with diabetes
mellitus.84 Finally, the widest integration of molecular diag-
nostics in current clinical practice is in the selection and
monitoring of therapy for infectious diseases. Viral load
testing for people infected with HIV or the hepatitis C virus
is the most frequently performed nucleic acid–based test in
molecular diagnostics laboratories.85 In addition, the use of
antiretroviral drug resistance testing has become the stan-
dard of care for managing patients with AIDS.86,87
Predisposition, Screening, and Early
Diagnosis of Disease
In addition to their direct roles in drug development
and clinical use, diagnostic tests are integrated with thera-
peutics in a variety of additional applications. Molecular
Am J Clin Pathol 2003;119:26-36 31
3 DOI: 10.1092/VMLL66Y5KHQ35KUE 31
Ross and Ginsburg / THE INTEGRATION OF MOLECULAR DIAGNOSTICS WITH THERAPEUTICS

A B

❚Image 2❚ HER-2/neu testing in breast cancer. A, HER-2/neu

Downloaded from http://ajcp.oxfordjournals.org/ by guest on July 5, 2016

C
gene amplification detected by fluorescence in situ hybridization
(Ventana Inform System, Ventana, Tucson, AZ). B, Immuno-
histochemical analysis using the HercepTest system (DAKO,
Carpinteria, CA) with continuous membranous 3+ positive
immunostaining for HER-2/neu protein (×400). C, HER-2/neu
gene amplification detected by chromogenic in situ
hybridization (Zymed System, Zymed, South San Francisco,
CA) (×400).

diagnostic tests are performed widely to detect cancer 150

Relative Fluorescence

130
predisposition,28-30,88 to screen for malignant and nonmalig- 110
nant diseases,89-92 and to facilitate diagnosis of early and 90
70
recurrent disease ❚Figure 2❚ and ❚Image 3❚ when therapy can 50
be most effective.93,94 High-throughput proteomic profiling 30
10
has emerged as a potential breakthrough in early disease –10
–30
detection and therapy monitoring.95-98 A recent serum-based – 50
1 5 10 15 20 25 30 35 40
study using the surface-enhanced laser desorption ionization
CT Number of PCR Cycles
technique combined with self-learning, computer pattern
recognition software successfully detected early stage
ovarian cancer with high sensitivity and specificity. 99 ❚Figure 2❚ Micrometastasis detection of breast cancer in a
Biomedical and function imaging including computed sentinel axillary lymph node biopsy specimen. Real-time
tomography, magnetic resonance imaging, and positron polymerase chain reaction (PCR; TaqMan, Applied
emission tomography scanning also have contributed major Biosystems, Foster City, CA) detection of cytokeratin 19
advances in the early detection of disease ❚Image 4❚ and in messenger RNA after 25 cycles of amplification (red)
the monitoring of therapeutic response.100 normalized to internal control dye (blue).

32 Am J Clin Pathol 2003;119:26-36 © American Society for Clinical Pathology

32 DOI: 10.1092/VMLL66Y5KHQ35KUE
 Basic Science / REVIEW ARTICLE

A B

❚Image 3❚ Micrometastasis detection of breast cancer in a sentinel axillary lymph node biopsy specimen. A, Section of half of a

Downloaded from http://ajcp.oxfordjournals.org/ by guest on July 5, 2016

lymph node showing no evidence of tumor cells (H&E, ×400). B, In the same lymph node, a single aggregate of 2 malignant
cells beneath the surface in the subcapsular sinusoid is shown (immunohistochemical stain for cytokeratin 19, ×200).

Day 0 Day 2 Day 4

❚Image 4❚ Metastatic prostate cancer in cervical lymph nodes. Radiolabeled indium scan using the J591 monoclonal antibody,
which detects the external domain of prostate-specific membrane antigen. The scan confirms that the initial clinical finding of a
neck mass was a cervical metastasis from prostate cancer. This patient subsequently had a significant (>50%) reduction in
serum prostate-specific antigen level in response to a single dose of yttrium 90 cytotoxic radio-conjugate of J591.

Conclusion of these challenges are being addressed. As patients continue

The traditional trial-and-error practice of medicine is
eroding progressively in favor of more precise marker-assisted
diagnosis and safer and more effective molecularly guided
treatment of disease. For the pharmaceutical industry, there
seems to be an equally desirable outcome of this approach:
increased efficiency, productivity, and product lines. The
diagnostics industry has an unprecedented opportunity for
integration, increased value, and commercial opportunities
for molecularly derived tests. The realization of the integra-
tion of diagnostics with therapeutics and the transition to
personalized medicine are not without challenges, yet many
to take on more and more of the burden of their own health
and well-being, educational forums must be developed for
patients and providers alike to understand the complex
nature of the genomic and proteomic information that is
driving biomarker-based drug development and the future
introduction of integrated diagnostics and therapeutics.
From the 1Department of Pathology and Laboratory Medicine,
Albany Medical College, Albany, NY; and 2Division of Molecular
Medicine, Millennium Pharmaceuticals, Cambridge, MA.
Address reprint requests to Dr Ross: Dept of Pathology and
Laboratory Medicine, Albany Medical College, Albany, NY 12208.

© American Society for Clinical Pathology Am J Clin Pathol 2003;119:26-36 33

3 DOI: 10.1092/VMLL66Y5KHQ35KUE 33
Ross and Ginsburg / THE INTEGRATION OF MOLECULAR DIAGNOSTICS WITH THERAPEUTICS
Acknowledgments: We are indebted to F. Symmans, MD, and
L. Pustzai, MD, M.D. Anderson Cancer Center, Houston, TX; and
N. Bander, MD, New York Presbyterian Hospital, New York, NY,
for their collaborative efforts.
References
1. Keesee SK. Molecular diagnostics: impact upon cancer
detection. Expert Rev Mol Diagn. 2002;2:91-92.
2. Poste G. Molecular diagnostics: a powerful new component of
the healthcare value chain. Expert Rev Mol Diagn. 2001;1:1-5.
3. Leonard DG. The present and future of molecular diagnostics.
Mol Diagn. 2001;6:71-72.
4. Vogel CL, Cobleigh MA, Tripathy D, et al. Efficacy and safety
of trastuzumab as a single agent in first-line treatment of
HER2-overexpressing metastatic breast cancer. J Clin Oncol.
2002;20:719-726.
5. O’Dwyer ME, Druker BJ. STI571: an inhibitor of the BCR-
ABL tyrosine kinase for the treatment of chronic
myelogenous leukaemia. Lancet Oncol. 2000;1:207-211.
6. Amos J, Patnaik M. Commercial molecular diagnostics in the
US: the Human Genome Project to the clinical laboratory.
Hum Mutat. 2002;19:324-333.
7. Bottles K. A revolution in genetics: changing medicine,
changing lives. Physician Exec. 2001;27:58-63.
8. Halapi E, Hakonarson H. Advances in the development of
genetic markers for the diagnosis of disease and drug response.
Expert Rev Mol Diagn. 2002;2:411-421.
9. Bordet R, Gautier S, Le Louet H, et al. Analysis of the direct
cost of adverse drug reactions in hospitalised patients. Eur J
Clin Pharmacol. 2001;56:935-941.
10. Suh DC, Woodall BS, Shin SK, et al. Clinical and economic
impact of adverse drug reactions in hospitalized patients. Ann
Pharmacother. 2000;34:1373-1379.
11. du Souich P. In human therapy, is the drug-drug interaction or
the adverse drug reaction the issue? Can J Clin Pharmacol.
2001;8:153-161.
12. Ginsburg GS, McCarthy JJ. Personalized medicine:
revolutionizing drug discovery and patient care. Trends
Biotechnol. 2001;19:491-496.
13. Ross JS, Fletcher JA. The HER-2/neu oncogene in breast
cancer: prognostic factor, predictive factor, and target for
therapy. Oncologist. 1998;3:237-252.
14. Morgan GJ, Pratt G. Modern molecular diagnostics and the
management of haematological malignancies. Clin Lab
Haematol. 1998;20:135-141.
15. Zareba W, Moss AJ, Schwarts PJ, et al, for the International
Long Q-T Syndrome Registry Research Group. Influence of
genotype on clinical course of the long-QT syndrome. N Engl
J Med. 1998;339:960-965.
16. Shimizu W, Antzelevitch C. Sodium channel block with
mexiletine is effective in reducing dispersion of repolarization
and preventing torsade de pointes in LQT2 as well as LQT3
models of the long-QT syndrome. Circulation. 1997;96:2038-
2047.
17. Schwartz PJ, Priori SG, Locati EH, et al. Long QT syndrome
patients with mutations of the SCN5A and HERG genes
have differential responses to Na+ channel blockade and to
increases in heart rate: implications for gene-specific therapy.
Circulation. 1995;92:3381-3386.
34 Am J Clin Pathol 2003;119:26-36
34 DOI: 10.1092/VMLL66Y5KHQ35KUE
18. Jordan VC. Progress in the prevention of breast cancer:
concept to reality. J Steroid Biochem Mol Biol. 2000;74:269-277.
19. Doren M, Samsioe G. Prevention of postmenopausal
osteoporosis with oestrogen replacement therapy and
associated compounds: update on clinical trials since 1995.
Hum Reprod Update. 2000;6:419-426.
20. Crouch MA. Effective use of statins to prevent coronary heart
disease. Am Fam Physician. 2001;63:309-320.
21. Taylor JG, Choi EH, Foster CB, et al. Using genetic variation
to study human disease. Trends Mol Med. 2001;7:507-512.
22. Carlson CS, Newman TL, Nickerson DA. SNPing in the
human genome. Curr Opin Chem Biol. 2001;5:78-85.
23. Griffin TJ, Smith LM. Single-nucleotide polymorphism
analysis by MALDI TOF mass spectrometry. Trends Biotechnol.
2000;18:77-84.
24. Kallioniemi OP. Biochip technologies in cancer research. Ann
Med. 2001;33:142-147.
25. Relling MV, Dervieux T. Pharmacogenetics and cancer
therapy. Nat Rev Cancer. 2001;1:99-108.
26. Borg A. Molecular and pathological characterization of
Downloaded from http://ajcp.oxfordjournals.org/ by guest on July 5, 2016
inherited breast cancer. Semin Cancer Biol. 2001;11:375-385.
27. Renegar G, Rieser P, Manasco P. Pharmacogenetics: the Rx
perspective. Expert Rev Mol Diagn. 2001;1:255-263.
28. Jass JR. Familial colorectal cancer: pathology and molecular
characteristics. Lancet Oncol. 2000;1:220-226.
29. Weber W, Estoppey J, Stoll H. Familial cancer diagnosis.
Anticancer Res. 2001;21:3631-3635.
30. Calzone KA, Biesecker BB. Genetic testing for cancer
predisposition. Cancer Nurs. 2002;25:15-25.
31. Ingelman-Sundberg M. Genetic susceptibility to adverse
effects of drugs and environmental toxicants: the role of the
CYP family of enzymes. Mutat Res. 2001;482:11-19.
32. Ingelman-Sundberg M. Pharmacogenetics: an opportunity for
a safer and more efficient pharmacotherapy. J Intern Med.
2001;250:186-200.
33. Steimer W, Potter JM. Pharmacogenetic screening and
therapeutic drugs. Clin Chim Acta. 2002;315:137-155.
34. Roses AD. Pharmacogenetics. Hum Mol Genet. 2001;10:2261-
2267.
35. Diasio RB, Johnson MR. The role of pharmacogenetics and
pharmacogenomics in cancer chemotherapy with 5-
fluorouracil. Pharmacology. 2000;61:199-203.
36. Gold B. DNA genotyping. Adv Clin Chem. 2001;36:171-234.
37. Linder MW, Valdes R Jr. Genetic mechanisms for variability
in drug response and toxicity. J Anal Toxicol. 2001;25:405-413.
38. Danesi R, De Braud F, Fogli S, et al. Pharmacogenetic
determinants of anti-cancer drug activity and toxicity. Trends
Pharmacol Sci. 2001;22:420-426.
39. Pullarkat ST, Stoehlmacher J, Ghaderi V, et al. Thymidylate
synthase gene polymorphism determines response and toxicity
of 5-FU chemotherapy. Pharmacogenomics J. 2001;1:65-70.
40. Durant J, Clevenbergh P, Halfon P, et al. Drug-resistance
genotyping in HIV-1 therapy: the VIRADAPT randomised
controlled trial. Lancet. 1999;353:2195-2199.
41. Rusnak JM, Kisabeth RM, Herbert DP, et al.
Pharmacogenomics: a clinician’s primer on emerging
technologies for improved patient care. Mayo Clin Proc.
2001;76:299-309.
42. Hess P, Cooper D. Impact of pharmacogenomics on the
clinical laboratory. Mol Diagn. 1999;4:289-298.
© American Society for Clinical Pathology

43. Harkin DP. Uncovering functionally relevant signaling
pathways using microarray based expression profiling.
Oncologist. 2000;5:501-507.
44. Snijders AM, Meijer GA, Brakenhoff RH, et al. Microarray
techniques in pathology: tool or toy? Mol Pathol. 2000;53:289-
294.
45. Raetz EA, Moos PJ, Szabo A, et al. Gene expression profiling:
methods and clinical applications in oncology. Hematol Oncol
Clin North Am. 2001;15:911-930.
46. Polyak K, Riggins GJ. Gene discovery using the serial analysis
of gene expression technique: implications for cancer
research. J Clin Oncol. 2001;19:2948-2958.
47. Lewis F, Maughan NJ, Smith V, et al. Unlocking the archive
gene expression in paraffin-embedded tissue. J Pathol.
2001;195:66-71.
48. Maughan NJ, Lewis FA, Smith V. An introduction to arrays.
J Pathol. 2001;195:3-6.
49. Alizadeh AA, Ross DT, Perou CM, et al. Towards a novel
classification of human malignancies based on gene
expression patterns. J Pathol. 2001;195:41-52.
50. Ramaswamy S, Golub TR. DNA microarrays in clinical
oncology. J Clin Oncol. 2002;20:1932-1941.
51. Schena M, Shalon D, Davis RW, et al. Quantitative
monitoring of gene expression patterns with a complementary
DNA microarray. Science. 1995;270:467-470.
52. Zanders ED. Gene expression analysis as an aid to the identifi-
cation of drug targets. Pharmacogenomics. 2000;1:375-384.
53. Weinstein JN. Pharmacogenomics: teaching old drugs new
tricks [editorial]. N Engl J Med. 2000;343:1408-1409.
54. Chicurel ME, Dalma-Weiszhausz DD. Microarrays in
pharmacogenomics: advances and future promise.
Pharmacogenomics. 2002;3:589-601.
55. Los G, Yang F, Samimi G, et al. Using mRNA expression
profiling to determine anticancer drug efficacy. Cytometry.
2002;47:66-71.
56. Slonim DK. Transcriptional profiling in cancer: the path to
clinical pharmacogenomics. Pharmacogenomics. 2001;2:123-136.
57. Dougherty ER, Barrera J, Brun M, et al. Inference from
clustering with application to gene-expression microarrays.
J Comput Biol. 2002;9:105-126.
58. Fryer RM, Randall J, Yoshida T, et al. Global analysis of gene
expression: methods, interpretation, and pitfalls. Exp Nephrol.
2002;10:64-74.
59. Lightcap ES, McCormack TA, Pien CS, et al. Proteasome
inhibition measurements: clinical application. Clin Chem.
2000;46:673-683.
60. Adams J. Development of the proteasome inhibitor PS-341.
Oncologist. 2002;7:9-16.
61. Elliott PJ, Ross JS. The proteasome: a new target for novel
drug therapies. Am J Clin Pathol. 2001;116:637-646.
62. Albanell J, Rojo F, Averbuch S, et al. Pharmacodynamic
studies of the epidermal growth factor receptor inhibitor
ZD1839 in skin from cancer patients: histopathologic and
molecular consequences of receptor inhibition. J Clin Oncol.
2002;20:110-124.
63. Hamadeh HK, Bushel P, Paules R, et al. Discovery in
toxicology: mediation by gene expression array technology.
J Biochem Mol Toxicol. 2001;15:231-242.
64. Fielden MR, Zacharewski TR. Challenges and limitations of
gene expression profiling in mechanistic and predictive
toxicology. Toxicol Sci. 2001;60:6-10.
© American Society for Clinical Pathology
Basic Science / REVIEW ARTICLE
65. Burchiel SW, Knall CM, Davis JW II, et al. Analysis of
genetic and epigenetic mechanisms of toxicity: potential roles
of toxicogenomics and proteomics in toxicology. Toxicol Sci.
2001;59:193-195.
66. Hamadeh HK, Amin RP, Paules RS, et al. An overview of
toxicogenomics. Curr Issues Mol Biol. 2002;4:45-56.
67. Mitcheson JS, Chen J, Lin M, et al. A structural basis for
drug-induced long QT syndrome. Proc Natl Acad Sci U S A.
2000;97:12329-12333.
68. Slamon DJ, Leyland-Jones B, Shak S, et al. Use of
chemotherapy plus a monoclonal antibody against HER2 for
metastatic breast cancer that overexpresses HER2. N Engl J
Med. 2001;344:783-792.
69. Lohrisch C, Piccart M. HER2/neu as a predictive factor in
breast cancer. Clin Breast Cancer. 2001;2:129-137.
70. Druker BJ. STI571 (Gleevec) as a paradigm for cancer
therapy. Trends Mol Med. 2002;8(4 suppl):S14-S18.
71. Joensuu H. Treatment of inoperable gastrointestinal stromal
tumor (GIST) with imatinib (Glivec, Gleevec). Med Klin.
2002;97(suppl 1):28-30.
Downloaded from http://ajcp.oxfordjournals.org/ by guest on July 5, 2016
72. Baselga J. Targeting the epidermal growth factor receptor with
tyrosine kinase inhibitors: small molecules, big hopes
[editorial]. J Clin Oncol. 2002;20:2217-2219.
73. Fox SB, Gasparini G, Harris AL. Angiogenesis: pathological,
prognostic, and growth-factor pathways and their link to trial
design and anticancer drugs. Lancet Oncol. 2001;2:278-289.
74. Herbst RS, Langer CJ. Epidermal growth factor receptors as a
target for cancer treatment: the emerging role of IMC-C225
in the treatment of lung and head and neck cancers. Semin
Oncol. 2002;29:27-36.
75. McDevitt MR, Barendswaard E, Ma D, et al. An alpha-
particle emitting antibody ([213Bi]J591) for
radioimmunotherapy of prostate cancer. Cancer Res.
2000;60:6095-6100.
76. Sievers EL, Linenberger M. Mylotarg: antibody-targeted chemo-
therapy comes of age. Curr Opin Oncol. 2001;13:522-527.
77. Piecyk M, Anderson P. Signal transduction in rheumatoid
arthritis. Best Pract Res Clin Rheumatol. 2001;15:789-803.
78. Williamson AA, McColl GJ. Early rheumatoid arthritis: can
we predict its outcome? Intern Med J. 2001;31:168-180.
79. Watts DA, Satsangi J. The genetic jigsaw of inflammatory
bowel disease. Gut. 2002;50(suppl 3):III31-III36.
80. Satsangi J. Genetics of inflammatory bowel disease: from
bench to bedside? Acta Odontol Scand. 2001;59:187-192.
81. Patel VB, Robbins MA, Topol EJ. C-reactive protein: a
“golden marker” for inflammation and coronary artery disease.
Cleve Clin J Med. 2001;68:521-524.
82. Feinstein SB, Voci P, Pizzuto F. Noninvasive surrogate markers
of atherosclerosis. Am J Cardiol. 2002;89:31C-43C.
83. Stein E. Laboratory surrogates for anti-atherosclerotic drug
development. Am J Cardiol. 2001;87:21A-26A.
84. McCarthy MI, Hattersley AT. Molecular diagnostics in
monogenic and multifactorial forms of type 2 diabetes. Expert
Rev Mol Diagn. 2001;1:403-412.
85. Berger A, Preiser W. Viral genome quantification as a tool for
improving patient management: the example of HIV, HBV,
HCV and CMV. J Antimicrob Chemother. 2002;49:713-721.
86. Vandamme AM, Houyez F, Banhegyi D, et al. Laboratory
guidelines for the practical use of HIV drug resistance tests in
patient follow-up. Antivir Ther. 2001;6:21-39.
Am J Clin Pathol 2003;119:26-36 35
3 DOI: 10.1092/VMLL66Y5KHQ35KUE 35
Ross and Ginsburg / THE INTEGRATION OF MOLECULAR DIAGNOSTICS WITH THERAPEUTICS
87. Buss N, Cammack N. Measuring the effectiveness of
antiretroviral agents. Antivir Ther. 2001;6:1-7.
88. Russo A, Zanna I, Tubiolo C, et al. Hereditary common
cancers: molecular and clinical genetics. Anticancer Res.
2000;20:4841-4851.
89. von Knebel Doeberitz M. New molecular tools for efficient
screening of cervical cancer. Dis Markers. 2001;17:123-128.
90. Ross JS. Economic, regulatory, and practice issues in
molecular pathology and diagnostics. Am J Clin Pathol.
1999;112(1 suppl 1):S7-S10.
91. Sakallah SA. Molecular diagnostics of infectious diseases:
state of the technology. Biotechnol Annu Rev. 2000;6:141-161.
92. Pang CP. Molecular diagnostics for cardiovascular disease.
Clin Chem Lab Med. 1998;36:605-614.
93. Lacroix J, Doeberitz MK. Technical aspects of minimal
residual disease detection in carcinoma patients. Semin Surg
Oncol. 2001;20:252-264.
36 Am J Clin Pathol 2003;119:26-36
36 DOI: 10.1092/VMLL66Y5KHQ35KUE
94. Bernard PS, Wittwer CT. Real-time PCR technology for
cancer diagnostics. Clin Chem. 2002;48:1178-1185.
95. Jain KK. Applications of proteomics in oncology.
Pharmacogenomics. 2000;1:385-393.
96. Lee KH. Proteomics: a technology-driven and technology-
limited discovery science. Trends Biotechnol. 2001;19:217-222.
97. Herrmann PC, Liotta LA, Petricoin EF III. Cancer
proteomics: the state of the art. Dis Markers. 2001;17:49-57.
98. Srinivas PR, Srivastava S, Hanash S, et al. Proteomics in early
detection of cancer. Clin Chem. 2001;47:1901-1911.
99. Petricoin EF, Ardekani AM, Hitt BA, et al. Use of proteomic
patterns in serum to identify ovarian cancer. Lancet.
2002;359:572-577.
100. Phelps ME. PET: the merging of biology and imaging into
molecular imaging. J Nucl Med. 2000;41:661-681.
Downloaded from http://ajcp.oxfordjournals.org/ by guest on July 5, 2016
© American Society for Clinical Pathology