GENAISSANCE PHARMACEUTICALS INC - 10-K Annual Report

Back to GetFilings.com

- --------------------------------------------------------------------------------
- --------------------------------------------------------------------------------
SECURITIES AND EXCHANGE COMMISSION
WASHINGTON, D.C. 20549
--------------------------

FORM 10-K

FOR ANNUAL AND TRANSITION REPORTS
PURSUANT TO SECTIONS 13 OR 15(d) OF THE
SECURITIES EXCHANGE ACT OF 1934

(MARK ONE)

/X/ ANNUAL REPORT PURSUANT TO SECTION 13 OR 15(d) OF THE
SECURITIES EXCHANGE ACT OF 1934

FOR THE FISCAL YEAR ENDED: DECEMBER 31, 2002
OR

/ / TRANSITION REPORT PURSUANT TO SECTION 13 OR 15(d) OF THE
SECURITIES EXCHANGE ACT OF 1934

FOR THE TRANSITION PERIOD FROM ________________ TO ________________

COMMISSION FILE NO. 000-30981

--------------------------

GENAISSANCE PHARMACEUTICALS, INC.

(Exact name of registrant as specified in its charter)

DELAWARE 06-1338846
(State or other jurisdiction of (I.R.S. Employer Identification No.)
incorporation or organization)

FIVE SCIENCE PARK 06511
NEW HAVEN, CONNECTICUT (Zip Code)
(Address of principal executive offices)

Registrant's telephone number, including area code: (203) 773-1450

--------------------------

Securities registered pursuant to Section 12(b) of the Act: NONE

Securities registered pursuant to Section 12(g) of the Act: COMMON STOCK, $.001
PAR VALUE
(Title of each class)

Indicate by check mark whether the registrant (1) has filed all reports
required to be filed by Section 13 or 15(d) of the Securities Exchange Act of
1934 during the preceding 12 months (or for such shorter period that the
registrant was required to file such reports), and (2) has been subject to such
filing requirements for the past 90 days. Yes /X/ No / /

Indicate by check mark if disclosure of delinquent filers pursuant to Item
405 of Regulation S-K is not contained herein, and will not be contained, to the
best of registrant's knowledge, in definitive proxy or information statements
incorporated by reference in Part III of this Form 10-K or any amendment to the
Form 10-K. / /

Indicate by check mark whether the registrant is an accelerated filer (as
defined in Exchange Act Rule 12b-2). Yes / / No /X/

The aggregate market value of voting Common Stock held by non-affiliates of
the registrant (without admitting that any person whose shares are not included
in such calculation is an affiliate) on June 30, 2002 was $26,843,669 based on
the last reported sale price of the Common Stock on the Nasdaq consolidated
transaction reporting system.

Number of shares of the registrant's class of Common Stock outstanding as of
March 28, 2003: 22,846,897.

Documents Incorporated By Reference:

Items 10, 11, 12, 13 and 15 of Part III (except for information required
with respect to our executive officers which is set forth under "Executive
Officers" in Item 1A of Part I of this report) have been omitted from this
report, since we expect to file with the Securities and Exchange Commission, not
later than 120 days after our fiscal year end, a definitive proxy statement. The
information required by Items 10, 11, 12, 13 and 15 of this report, which will
appear in the definitive proxy statement, is incorporated by reference into
Part III of this report.

- --------------------------------------------------------------------------------
- --------------------------------------------------------------------------------

ITEM 1. BUSINESS

COMPANY OVERVIEW

Our goal is to create personalized medicines through the integration of
pharmacogenomics into drug development and marketing. Pharmacogenomics is the
study of how an individual's genetic inheritance affects the body's response to
a drug. We discover inherited differences, or genomic markers, that exist in
human genes. We use our technological capabilities and methods and our clinical
genetics development skills to identify the genomic markers that appear to
define a patient population that responds best to a medication and has a
superior safety profile. We market our technology and our genomic markers to the
pharmaceutical and biotechnology industry as a means to improve the development,
marketing and prescribing of drugs.

We were incorporated in Delaware on February 22, 1992 and changed our name
to Genaissance Pharmaceuticals, Inc. on March 18, 1997. Our principal executive
offices are located at Five Science Park, New Haven, Connecticut 06511. Our
telephone number is (203) 773-1450. Our website is located at
http://www.genaissance.com.

INDUSTRY OVERVIEW

DEVELOPMENT AND MARKETING OF DRUGS

The pharmaceutical, biotechnology and healthcare industries face intense
pressure to become more productive and deliver more cost effective healthcare.
Two of the pharmaceutical and biotechnology industry's most challenging issues
are the high cost and low success rate of developing drugs and the need to
differentiate approved drugs in highly competitive markets. At the same time,
healthcare providers and payers are spending a growing proportion of their
resources on prescription drugs.

The drug development process is costly and subject to a high failure rate.
The average cost of developing a drug is estimated to be $500 million, including
the cost of unsuccessful drug candidates. Even with recent technological
advances, including advances in areas such as genomics, the failure rate of
clinical trials remains very high. In the United States, only one in five drug
candidates that enters clinical trials reaches the market. Seventy percent of
the drug candidates that enter clinical trials successfully complete phase I,
33% complete phase II, 25% complete phase III and only 20% achieve regulatory
approval. The decision to enter phase III, the most costly phase of clinical
trials, is generally based upon the results obtained from the limited number of
individuals, often fewer than 200, typically studied in phases I and II. The
typical patient population in phase III is between 1,000 and 5,000 individuals,
and the average amount of money spent in a single phase III clinical trial is
estimated to be greater than $40 million.

Approved drugs often face intense competition. The period of market
exclusivity for the first drug in a new therapeutic class is typically much
shorter today than it was a few years ago. Consequently, marketing expenditures
have increased rapidly as companies attempt to maintain or increase market
share. For example, in 2001, U.S. based pharmaceutical companies spent over
$19 billion on promotional activities, including sales representatives, product
samples and journal advertising. Marketing departments are also under pressure
to maximize the revenue generated from approved products in order to meet
corporate-wide revenue and earnings goals. In addition, the major pharmaceutical
companies, which have reported an average increase in earnings of 14% over the
past five years, now face competition from generic drugs for several of their
blockbuster drugs that generated over $40 billion in sales in 2001. Thus, in
order to maintain revenue growth rates and profitability, pharmaceutical
companies must both improve the success rate of clinical trials and
differentiate their drugs in a crowded market place.

According to the National Institute for Health Care Management, sales of
prescription drugs in the United States increased 18.2% in 2001 from
$148.2 billion in 2000 to $175.2 billion. Retail spending

2

on outpatient prescription drugs has nearly doubled since 1997. In an attempt to
contain the rising cost of drug expenditures, healthcare providers and payers
face the difficult task of deciding which drugs should be prescribed to specific
patients and are suitable for reimbursement. Healthcare providers make these
decisions using medical outcome studies and economic benefit factors but they do
not have any knowledge of which individual patients are most likely to benefit
from a specific drug, if at all. Thus, healthcare providers and patients would
benefit from using drugs that are targeted for a patient population that would
have the best drug response and safety profile, and, thus, allow for more
appropriate and safer intervention.

POPULATION GENOMICS

The medical community generally acknowledges that most drugs work more
effectively for some patients than for other patients. The pharmaceutical and
biotechnology industry often poorly understands this variability in patient
response. Consequently, pharmaceutical and biotechnology companies may
unnecessarily discontinue further drug development, fail to obtain regulatory
approval for promising drug candidates, or, even if a drug obtains approval, be
unable to market an approved drug effectively or to obtain approval for third
party reimbursement.

Scientists have known for a long time that genomic differences influence how
patients respond to drugs. However, pharmaceutical and biotechnology companies
generally have not considered genomic differences between patients in developing
and implementing clinical trials or in the marketing of approved drugs. If, in
clinical trials, pharmaceutical and biotechnology companies were able to use
genomic markers to identify patient populations that would have different drug
responses, they could improve the drug development and marketing process. For
example, pharmaceutical and biotechnology companies could use the genomic
markers, which are identified in phase I and phase II clinical trials as being
predictive of a clinical outcome, to determine the size of the patient
population that would likely benefit from the drug under development. They would
also know the size of the clinical group needed for a phase III clinical trial
to obtain statistically significant data to support the clinical development
program. In addition, if pharmaceutical and biotechnology companies could
identify the patients most likely to have a side effect, they could more closely
monitor these patients or eliminate them from participating in clinical trials
and receiving the drug and, hence, increase the safety profile of a drug. The
pharmaceutical and biotechnology companies would, therefore, have a better
understanding of the cost required to complete the development of a drug and the
likely economic return on their investment before proceeding to a phase III
clinical trial. In addition, if pharmaceutical and biotechnology companies could
use genomic markers to predict a drug response, they would be able to improve
the marketing of their drugs by identifying those patient populations for which
particular drugs are likely to be most effective with the least likelihood of
having an adverse reaction. Furthermore, healthcare providers and payers would
likely benefit economically from predictive information that would enable a
physician to prescribe the most appropriate and safest medication at the
earliest possible time.

Population genomics is the analysis of genomic variation within groups of
people. The genomic blueprint each person inherits from his or her biological
parents determines differences, such as height, hair color and eye color. As
scientists better understand variation at the molecular or genomic level, they
are more certain that an individual's response to a drug is dependent upon that
individual's unique DNA sequence and that more than one gene is probably
involved in a drug response. Scientists know that every drug generally
interacts, directly and indirectly, with a variety of different proteins
produced by different genes. Therefore, in order to predict a specific drug
response, scientists must analyze genomic variation in multiple genes.

3

SINGLE NUCLEOTIDE POLYMORPHISMS AND HAPLOTYPES

At the DNA level, genomic variation occurs mainly as a result of variation
at a single position in the DNA sequence, commonly referred to as a single
nucleotide polymorphism or SNP. Geneticists historically studied genetic
variation by analyzing the inheritance of traits within an extended family.
Classical population geneticists coined the term haplotype to describe the
physical organization of genetic variation as it occurs in an individual. The
haplotype is the standard for measuring genetic variation. At the molecular
level, a haplotype consists of multiple, individual SNPs that are organized into
one of the limited number of combinations that actually exists as units of
inheritance in humans. Each haplotype contains significantly more information
than individual, unorganized SNPs. As a result, clinicians need fewer patients
to define a patient population with a different drug response if they use
haplotypes rather than individual, unorganized SNPs.

In October 2002, an international consortium, composed of non-profit
biomedical research groups and private companies in Japan, the United Kingdom,
Canada, China and the United States, initiated an effort to create a genome-wide
haplotype map. This new venture is aimed at speeding the discovery of genes that
are related to common illnesses, such as asthma, cancer, diabetes and heart
disease. The consortium expects that the International HapMap Project, which
will use as its starting material the more than 2.8 million SNPs that are
already in a public database, will take three years to complete at an estimated
cost of $100 million. The resulting haplotype map could contain anywhere from
300,000 to 600,000 SNPs. Using this genome-wide haplotype map, researchers may
only identify a genomic region that is involved in causing a disease or drug
response rather than the specific gene that is responsible. Researchers are,
thus, likely to need a database of gene haplotypes to determine which of the
genes that are contained within the identified genomic region is actually the
gene responsible for causing the disease or drug response.

Through the use of gene haplotypes, together with sophisticated software
programs, drug developers could identify, with statistical accuracy in a small
population of the size commonly seen in phases I and II of clinical trials, the
genomic variation that defines different patient populations with different drug
responses and, therefore, make better informed decisions on whether or not to
enter phase III clinical trials. In addition, defining the patient population
with the desired clinical outcome would enable drug developers to improve the
marketing of their drugs by identifying those patients for whom particular drugs
are likely to have the best safety profile and be most effective. Furthermore,
healthcare providers and payers would likely benefit economically from
predictive information that would enable a physician to prescribe the
appropriate medication at the earliest possible time.

REGULATORY ENVIRONMENT

On January 31, 2003, the FDA announced a broad initiative, whose purpose is
to help make innovative medical technology available sooner and to reduce the
costs of developing safe and effective medical products while maintaining the
FDA's traditional high standards of consumer protection. The FDA included, in
its initiative, the use of pharmacogenomics during drug development. The FDA
indicated that within six months the agency would issue draft guidance on when
and how to submit pharmacogenomic information to the FDA during drug
development. The FDA said that this guidance would facilitate the exploratory
use of pharmacogenomic screening during drug development and would clarify when
such information would be considered part of the evaluation of drug safety. In
addition, the FDA stated that the agency would hold a workshop in 2003 on issues
that are involved in the co-development of a pharmacogenomic test and a drug and
that within 18 months the agency would issue guidance on the regulatory pathway
for such combinations through the agency's Center for Drug Evaluation and
Research and the Center for Devices and Radiological Health.

4

THE GENAISSANCE SOLUTION

We have developed a combination of technologies and expertise, which we call
our HAP-TM- Technology, that allows population genomics to be integrated into
the development, marketing and prescribing of new and existing medicines.

The key components of our HAP-TM- Technology are:

- a database of highly informative, proprietary measures of genomic
variation, or haplotypes, which we call HAP-TM- Markers, for
pharmaceutically relevant genes;

- a proprietary informatics system, which we call
DECOGEN-Registered Trademark-, including unique algorithms, for defining
patient populations with different drug responses;

- a cost effective, efficient process for measuring genomic variation in
clinical DNA samples, which we call HAP-TM- Typing; and

- clinical genetics development skills.

We designed our HAP-TM- Technology to permit pharmaceutical and
biotechnology companies to use population genomics in a variety of ways for drug
development and commercialization.

DRUG DEVELOPMENT. We designed our HAP-TM- Technology to improve the success
rate of drugs in clinical trials by:

- assessing efficiently the genomic variation among the patients involved in
a clinical trial, thereby permitting pharmaceutical and biotechnology
companies to incorporate genomic variation information into all decisions
required during the course of a clinical trial;

- creating better informed, or "smarter," clinical trials through the design
of protocols, which result in the inclusion of those patients most likely
to benefit, with a superior safety profile, from the proposed therapeutic
product;

- facilitating earlier "go/no-go" decisions on whether to proceed to the
next phase of clinical trial testing, which should result in a more
efficient use of clinical resources; and

- reducing the size and, hence, the cost of late-stage clinical trials by
enrolling a group of patients, who are most likely to respond to a drug
and are least likely to suffer an adverse reaction.

DRUG MARKETING AND PRESCRIBING. We also designed our HAP-TM- Technology to
help maximize the value of an approved drug by:

- identifying which of our HAP-TM- Markers define the patient population
with the best response and with a superior safety profile;

- integrating genomic variation information into marketing strategies to
sustain and enhance a market leading position or to address problems such
as poor market penetration, competitive pricing issues, safety, risk of
therapeutic substitution, and limited patent life; and

- targeting new markets and obtaining approval for new indications.

Our HAP-TM- Technology could also be useful for improving the drug discovery
process through the selection and validation of drug targets. In addition,
pharmaceutical and biotechnology companies could incorporate data obtained
during clinical trials into the drug discovery process to develop second-
generation drugs. If widely adopted, our HAP-TM- Technology could enable the
healthcare system to personalize treatment based upon an individual's unique
genome.

5

OUR STRATEGY

Our strategy consists of the following:

TO GENERATE CURRENT REVENUES FROM COMMERCIALIZING OUR HAP-TM- TECHNOLOGY IN
VARIOUS MARKETS. The specific revenue sources would include:

- license fees from granting access to our database of HAP-TM- Markers for
pharmaceutically relevant genes;

- HAP-TM- Typing fees for measuring the genomic variation present in DNA
samples;

- Study fees for identifying the genomic markers responsible for defining
populations with different drug response or different traits; and

- revenues from homebrew diagnostic tests, which we offer that define
populations with different drug response or different traits.

The different markets include:

- pharmaceutical companies, including those in regional markets, such as
Japan;

- biotechnology companies;

- diagnostic companies;

- government and academic groups; and

- consumer markets, such as nutraceuticals.

TO ESTABLISH CONTRACTUAL RIGHTS FOR ROYALTIES FROM THE SALE OF DRUGS AND
DIAGNOSTIC TESTS. The revenue sources would include developing, with partners,
drug response pharmacogenomic tests from the intellectual property created in
our HAP-TM- Partnership program or in our internal programs. Under our current
strategy, and for the foreseeable future, we do not expect to develop or market
pharmaceutical or diagnostic products on our own. However, we seek royalties
from the sale of drugs and diagnostics, which use our HAP-TM- Markers in a
diagnostic test that:

- defines drug safety or efficacy or

- is sold in combination with the use of a drug.

Our commercialization programs include the following:

COMMERCIALIZE OUR HAP-TM- TECHNOLOGY THROUGH OUR HAP-TM- PARTNERSHIP
PROGRAM. We have developed a program intended to provide pharmaceutical and
biotechnology companies access to our HAP-TM- Technology and our clinical
genetics development expertise throughout each phase of drug development and
marketing. Potential partners are offered access to our proprietary HAP-TM-
Markers, our DECOGEN-REGISTERED TRADEMARK- Informatics System, our HAP-TM-
Typing facility, which we use to measure HAP-TM- Markers from individual patient
samples, and our clinical genetics development expertise. In return, we seek
annual subscription fees and payments for our collaborative contributions to
specific drug development or marketing projects and for our HAP-TM- Typing
services. In addition, we expect license fees, as well as milestone and royalty
payments, for the commercial use of our HAP-TM- Markers. While we expect to
retain all intellectual property rights in HAP-TM- Markers discovered during the
partnership, we offer our partners an option to obtain multiple exclusive
licenses for the use of particular HAP-TM- Markers for the development of
therapeutic and diagnostic products within specified drug classes and for
particular disease indications. We currently have a HAP-TM- Partnership program
with six pharmaceutical and biotechnology companies (AstraZeneca UK Limited;
Biogen, Inc.; Johnson & Johnson Pharmaceutical Research & Development, a
division within the Johnson & Johnson family of companies; Millennium
Pharmaceuticals, Inc.; Pfizer Inc.; and Pharmacia & Upjohn Company). We are in
discussions and

6

negotiations with additional pharmaceutical and biotechnology companies to enter
into our HAP-TM- Partnership program, including pharmaceutical companies in such
regional markets as Japan, where we are using Intec Web and Genome Informatics
Corporation as our sales representative. We cannot, however, assure you that we
will be successful in these discussions and negotiations.

COMMERCIALIZE WITH PARTNERS THE INTELLECTUAL PROPERTY THAT WE DEVELOP WITH
OUR HAP-TM- TECHNOLOGY. We currently have two programs from which we expect to
have intellectual property to commercialize with our partners.

STRENGTH TRIALS (STATIN RESPONSE EXAMINED BY GENETIC HAP-TM- MARKERS). We
applied our HAP-TM- Technology and our clinical genetics development skills to
the statin class of drugs, which doctors use to treat patients with high
cholesterol and lipid levels and who are, therefore, at risk for cardiovascular
disease. This market is highly competitive with multiple approved products
seeking to gain increased market share. Currently, the market is approximately
$18 billion worldwide and experts estimate that the market is growing at the
rate of about 20% per year.

We examined five statins in our two prospective STRENGTH clinical trials.
The goal of these two trials was to identify which of our HAP-TM- Markers define
a patient population that has the best therapeutic response to one or more of
these statins with a superior safety profile. In April 2001, we began enrollment
for the first of our two STRENGTH trials, in which patients were randomized and
then enrolled into one of four treatment arms, in which a patient was treated
with one of four statins. The four drugs used were atorvastatin (sold by
Pfizer Inc. as Lipitor-Registered Trademark-), cerivastatin (sold by Bayer AG as
Baycol-Registered Trademark-), pravastatin (sold by Bristol-Myers Squibb Company
as Pravacol-Registered Trademark-) and simvastatin (sold by Merck as
Zocor-Registered Trademark-). Following the withdrawal of cerivastatin from the
market in August 2001, we discontinued treating all of the patients who were
taking this product. In October 2001, we began enrollment in our second STRENGTH
trial, in which enrolled patients received lovastatin (sold by Merck as
Mevacor-Registered Trademark-). The trial protocol for both STRENGTH trials was
to treat each patient for eight weeks with the lowest dose recommended on the
drug label followed by then treating each patient for eight weeks with the
highest dose recommended on the drug label. At least 149 patients were enrolled
into each treatment arm. Enrollment was completed for STRENGTH I in July 2001
with the last patient receiving the last examination in November 2001.
Enrollment was completed for STRENGTH II in November 2001 with the last patient
receiving the last examination in March 2002.

In January 2003, we signed an agreement with Bayer HealthCare LLC to
commercialize exclusively the diagnostic rights and non-exclusively the drug
product development rights from our STRENGTH trials. We are in discussions with
pharmaceutical companies to commercialize additional drug product development
and marketing rights from these trials, but we cannot assure you that we will be
successful in these discussions.

CARING STUDY (CLOZAPINE HAP-TM- MARKER DISCOVERY STUDY). In February 2002,
we initiated a study designed to identify which of our HAP-TM- Markers define
the patients who are most likely to develop agranulocytosis, a potentially
life-threatening depletion of white blood cells, if treated with clozapine.
Clozapine, which is no longer under patent protection, is a highly effective
therapeutic for treating certain patients with schizophrenia. During the first
six months of treatment with clozapine, a patient must undergo weekly blood
monitoring, a requirement that results in poor patient compliance. We believe
that clozapine's third-line therapy status and the requirement for repeated
blood testing are the primary reasons that the market share for all clozapine
products is only approximately $350 million in the United States. Sales of
antipsychotic drugs in the United States reached an estimated $5.2 billion in
2000 and are expected to exceed $8 billion by 2005. As of March 3, 2003, we have
obtained blood samples from 23 patients who took clozapine and developed
agranulocytosis. We have also obtained blood samples from 12 patients who took
clozapine and developed granulocytopenia (a significant reduction in the number
of white cells) as well as 21 matched control samples from patients who took
clozapine without experiencing any significant decrease in white blood cell
counts.

7

We plan to use these patient samples with our HAP-TM- Technology to identify
the HAP-TM- Markers that define the population most likely to develop this
adverse reaction. If we are successful in identifying these genomic markers, we
plan to seek a suitable partner to commercialize these HAP-TM- Markers either
through a diagnostic test or a diagnostic test associated with a drug.

COMMERCIALIZE OUR HAP-TM- TECHNOLOGY IN NUMEROUS MARKETS. We are now
offering our HAP-TM- Technology, including our sequencing and genotyping
services, to potential partners in additional markets, including research groups
in government and academia, and to companies involved in the sale of
nutraceuticals.

PURSUE STRATEGIC ACQUISITIONS. We continually evaluate opportunities that
may provide us with, among other things, intellectual property, key personnel,
capabilities that could augment our recurring revenues or technologies that will
enhance and complement our HAP-TM- Technology. From time to time, we intend to
pursue acquisitions, which we believe will meet these goals.

ASTHMA CLINICAL STUDY AS A PROOF OF PRINCIPLE

To demonstrate how the pharmaceutical and biotechnology industry and
healthcare providers could use our HAP-TM- Technology, we conducted a clinical
study to determine whether we could use HAP-TM- Markers or SNPs to define an
asthma patient population that responded to the drug albuterol (sold by
GlaxoSmithKline as Ventolin-Registered Trademark-), a standard treatment for
persons with asthma. We conducted the study in collaboration with Dr. Stephen
Liggett, a member of our scientific advisory board and a professor of medicine
at the University of Cincinnati Medical Center. We examined genomic variation in
the target of the drug, the (BETA)(2)-adrenergic receptor ((BETA)(2)-AR).
Dr. Liggett recruited 121 asthmatic individuals for clinical treatment and made
a large number of standard pulmonary measurements, before and after he treated
the patients with albuterol. The response to the drug differed significantly
from patient to patient and the drug was clinically effective in only 40% of
these patients as measured by generally accepted clinical criteria. Dr. Liggett
and his staff drew blood samples and extracted the DNA. We did the HAP-TM-
Typing and, using our DECOGEN-REGISTERED TRADEMARK- Informatics System,
identified specific pairs of HAP-TM- Markers in the (BETA)(2)-AR receptor gene
that were carried by patients that exhibited a positive drug response and
specific pairs of HAP-TM- Markers that were carried by patients that exhibited a
poor drug response. By contrast, no individual SNP was found to have such
predictive power in this study. This study, which was similar in sample size
typically used for a phase II clinical trial, showed that a patient's response
to albuterol correlated, in a statistically significant manner, with specific
HAP-TM- Markers.

We published a more detailed description of this study in a peer reviewed
scientific journal in September 2000. In February 2003, we received notice from
the U.S. Patent and Trademark Office of its allowance of a patent application
filed by us, which contains claims based on these findings. In addition, we
worked with a company that has a FDA approved, DNA-based diagnostic platform and
verified the SNP assays they developed for defining the predictive HAP-TM-
Markers. This proof of principle diagnostic assay and our clinical study
demonstrate that a pharmacogenomic test for drug response can be developed for
point of care use with technology currently available.

OUR TECHNOLOGY

OVERVIEW

Our process for discovering HAP-TM- Markers eliminates the need, which is
inherent in family-based approaches, to find and then test large numbers of
families or related individuals to determine genomic variation. Initially, we
discover individual SNPs by high-throughput sequencing of DNA samples of
unrelated and related individuals that are representative of the individuals who
constitute the major pharmaceutical markets of the world. We use our proprietary
algorithms to organize the SNPs into

8

HAP-TM- Markers. Using these algorithms, we found that the number of actual
HAP-TM- Markers per gene is significantly less than the theoretically large
number of ways in which SNPs could be organized.

Our DECOGEN-REGISTERED TRADEMARK- Informatics System contains a number of
components. Our HAP-TM- Database contains our HAP-TM- Markers, including
information about their sequence, frequency and distribution. Our
DECOGEN-REGISTERED TRADEMARK- Informatics System also contains a proprietary
collection of algorithms and a search engine that correlates a patient's HAP-TM-
Markers with a particular response to a drug. To handle large amounts of
information, we developed a proprietary tool, which we call RuleFinder-TM-, that
allows us to identify rapidly potential associations between clinical endpoints
and genetic variation. Then using standard analytical tools, we have been able
to determine with statistical accuracy the correlation between HAP-TM- Markers
and drug response in a small population of the size commonly seen in phase I and
phase II of a clinical trial.

HAP-TM- Typing is our process for measuring which HAP-TM- Marker pairs are
present in a patient's DNA sample. Our HAP-TM- Typing facility uses proprietary
software, robotics and Sequenom's MassARRAY-TM- platform to determine, on a
high-throughput basis, which two HAP-TM- Markers for a gene are present in a
patient's DNA sample. We integrate the resulting data into our
DECOGEN-REGISTERED TRADEMARK- Informatics System to search for a correlation
with a patient's drug response. We have a customized 8,000 square foot facility,
dedicated to HAP-TM- Typing, which became licensed to do diagnostic testing in
April 2002 by the Connecticut Department of Public Health under the U.S.
Clinical Laboratory Improvement Amendments of 1988 (CLIA).

The following outlines the components of our HAP-TM- Technology and how we
use our HAP-TM- Technology to define a patient population with a specific drug
response.

GENE SELECTION

Our goal is to discover HAP-TM- Markers for pharmaceutically relevant genes.
We prioritize these genes for HAP-TM- Marker discovery based upon the needs of
our HAP-TM- Partnership program partners. We obtain genomic information relevant
for gene selection from publicly available sources and from proprietary
databases. We have discovered HAP-TM- Markers for genes that are, or will likely
become, drug targets; are associated with drug target pathways; are involved in
how drugs modify cell communication or regulate other genes; and are involved in
the metabolic process by which the body absorbs a drug and breaks it down.

INDEX REPOSITORY

We constructed an Index Repository, a collection of diverse DNA samples, to
discover the SNPs that are present in genes. We designed our Index Repository to
contain genomic information that would be representative of the people who
constitute the major pharmaceutical markets of the world; to aid in the quality
control analysis of the SNPs we discover; and to facilitate the organization of
SNPs into HAP-TM- Markers.

To build our Index Repository, we recruited over 650 individuals whose
parents and grandparents came from specified geographical regions. We obtained
personal information from each individual, including sex, date of birth, and
general medical information, as well as a detailed family history and drew blood
samples so that we could create continually multiplying cells from the white
cells present in the blood. The resulting cells, called permanent cell lines,
provide us with a supply of DNA from which to discover SNPs. We store frozen
samples of each cell line at multiple locations to ensure that all of these cell
lines are available in the future. To supply sufficient DNA for the production
process, we routinely grow the cell lines in our cell culture facility. We
employ quality control procedures that permit each DNA sample to be
unambiguously matched to its corresponding cell line. We store all of the
information about a cell line in our proprietary HAP-TM- Database that is a
component of our DECOGEN-REGISTERED TRADEMARK- Informatics System.

9

DISCOVERING SNPS

We use a subset of our Index Repository to discover SNPs. We employed
principles of population statistics to determine the minimum number of unrelated
individuals that we needed to have a 99% probability of detecting a SNP or
HAP-TM- Marker that occurs in at least 5% of the general population or in at
least 10% of a population from a specific geographical region.

We sequence individual samples of DNA so that we can accurately determine
the frequency of a SNP in the population. Our procedure allows us to detect SNPs
that are present at lower frequencies than if we were to analyze a mixture of
DNA from different individuals, as is done by some companies. We sequence 93
individual, human DNA samples, or 186 individual genomes, from our Index
Repository in the following genomic regions for each selected gene: the region
responsible for controlling when a gene is active, the control region; the
regions containing coding information that is found in the protein product of
the gene, the coding regions; the boundaries between the genomic regions
containing coding information and those interspersed regions that do not contain
coding information, the non-coding regions; and the region at the end of a gene
immediately after the last region containing coding information.

Our sequencing process is highly automated, from picking the regions to be
sequenced through loading the samples onto one of our sequencing machines. We
have also developed a proprietary laboratory information management system to
track genes as they progress through the production pipeline. We use this system
to monitor the overall quality of data we produce to ensure that the sequencing
process is operating according to our established standards. The sequence
information undergoes two forms of quality control analysis. We use electronic
procedures and established population genomic principles to identify and
validate that a SNP exists at a given position.

HAP-TM- MARKERS AND HAP-TM- DATABASE

Geneticists use the term haplotype to describe how SNPs are organized on a
chromosome. Typically, geneticists study the inheritance of genetic variability
in extended families in order to determine haplotypes. We do not need to conduct
family studies to discover haplotypes. Rather than relying on family studies, we
have developed an entirely computerized process for discovering haplotypes. Our
proprietary method works because we analyze a large number of individual samples
and we have members of extended families in our sample set. We have validated
the accuracy of our computerized process by conventional family studies and
molecular techniques. We use our proprietary computational methods and
algorithms to determine how the SNPs in a gene are organized on each of the two
chromosomes in each sample we sequence from our Index Repository. We use the
term HAP-TM- Marker, derived from haplotype, to describe the organization of
SNPs we find for a gene.

Our computerized process assigns a confidence value to each HAP-TM- Marker
we discover. If the HAP-TM- Markers we discover for a gene fall below a defined
confidence level, we subdivide the gene into regions. We reexamine each region
until we identify HAP-TM- Markers that meet our acceptance level. We then enter
each HAP-TM- Marker into our proprietary HAP-TM- Database. We also enter other
relevant population information, such as the distribution and frequency of each
HAP-TM- Marker among people from different geographical regions. We also
include, in our HAP-TM- Database, other genomic markers that others have
identified and are available in public databases.

As of March 3, 2003, we had processed in excess of 7,000 pharmaceutically
relevant genes through our production process and deposited their HAP-TM-
Markers and associated information into our HAP-TM- Database. All of the nearly
500 current drug targets, with identified genomic structure, have gone through
our production process. To date, we have found an average of approximately 18
SNPs per gene. There are generally two possible forms of a SNP that are found at
a site of genomic variation. Therefore, these 18 SNPs could theoretically be
organized into 2(18) or 262,144 potential HAP-TM- Markers.

10

Using our proprietary algorithms, we found that these SNPs are organized into an
average of only approximately 19 HAP-TM- Markers per gene.

THE DECOGEN-REGISTERED TRADEMARK- INFORMATICS SYSTEM

We have constructed a proprietary informatics system, called
DECOGEN-REGISTERED TRADEMARK-(), which contains the proprietary database of
population information for our Index Repository and our proprietary HAP-TM-
Database of HAP-TM- Markers. These databases can accommodate information from a
variety of populations, including individuals suffering from a specific disease
and patients in clinical trials, as well as associated data, such as detailed
medical histories, including responses to drugs. The portal to these databases
is the DECOGEN-REGISTERED TRADEMARK- Informatics System's search engine, which
we designed with an intuitive, graphical user interface so that drug development
clinicians can easily manage their data to find a correlation between HAP-TM-
Markers and a drug response.

Our DECOGEN-REGISTERED TRADEMARK- Informatics System can use either
qualitative or quantitative clinical measurements as a clinical endpoint to
search for a correlation with our HAP-TM- Markers. The informatics system has
the ability to exchange information with standard software packages used in the
pharmaceutical industry and additional tools are also available within the
system to help in the design and operation of clinical trials.

HAP-TM- TYPING

We use the term HAP-TM- Typing to describe the process of determining which
HAP-TM- Marker is present for each of the two versions of each gene in a
patient's clinical sample. We designed our HAP-TM- Typing capabilities to
support our HAP-TM- Partnership program partners as well as to provide
pharmacogenomics support services to pharmaceutical, biotechnology and
diagnostic companies, to government and academic groups and to other entities.
The first step in searching for a clinical correlation is to do HAP-TM- Typing
on clinical samples. Our DECOGEN-REGISTERED TRADEMARK- Informatics System
contains a proprietary computational tool that determines the minimal number and
combination of variable sites, which we must analyze in order to identify, with
high confidence, the two HAP-TM- Markers that are present for each gene in a
clinical sample of DNA. This proprietary tool exploits an established genetic
principle. That is, the presence of a given form of genomic variation at one
position can be highly predictive of the form of genomic variation present at
another site in a gene. This predictability reduces the complexity of the
information needed to identify a HAP-TM- Marker in a genomic sample. We can
determine this predictability, however, only if we already know the haplotype or
the organization of SNPs in a gene. Our HAP-TM- Markers contain this needed
information.

OUR COLLABORATIONS

Through December 31, 2002, we have entered into the following licenses and
collaborations:

ASTRAZENECA UK LIMITED

Effective November 29, 2001, we entered into a three-year agreement with
AstraZeneca UK Limited, in which AstraZeneca gained limited access to our
HAP-TM- Technology to investigate associations between our HAP-TM- Markers and
disease susceptibility, in exchange for a specified, onetime payment. We granted
AstraZeneca a perpetual exclusive license to use those HAP-TM- Markers that are
shown to have a predictive association with a certain disease, for discovering,
developing, manufacturing, marketing and selling of AstraZeneca drugs. We also
granted AstraZeneca a perpetual, co-exclusive license, with us, to use the
predictive HAP-TM- Markers for discovering, developing, manufacturing, marketing
and selling prognostic products used in connection with the sale or prescription
of AstraZeneca drugs. In exchange for these license grants, AstraZeneca granted
us

11

options, which expire in 2011, to obtain licenses under its intellectual
property for making, using, marketing and selling prognostic and diagnostic
products that detect these predictive HAP-TM- Markers.

BD (BECTON, DICKINSON AND COMPANY)

Effective December 18, 2002, we entered into a training and license
agreement with BD. Under the terms of the agreement, we acquired a
non-exclusive, non-transferable license to BD's proprietary BDProbeTec-TM- ET
platform and Strand Displacement Amplification Technology. The license is fully
paid up for our internal research and development activities, which are limited
to the United States, is royalty-bearing for selling products and services
world-wide for genotyping HAP-TM- Markers in certain fields of use. BD will
provide equipment, certain reagents and training on the development of tests.

BIOGEN, INC.

Effective December 21, 2001, we entered into an agreement with
Biogen, Inc., in which Biogen gained non-exclusive access to selected HAP-TM-
Markers from our HAP-TM- Database solely for research and development purposes.
We receive payments based upon the HAP-TM- Markers that Biogen selects.

Effective January 31, 2002, we entered into a second agreement with
Biogen, Inc., in which we are applying our HAP-TM- Technology to a study of the
pharmacogenetic basis of variability in response to
Amevive-Registered Trademark- (alefacept), a biologic developed for the
treatment of adults with moderate-to-severe chronic plaque psoriasis who are
candidates for systemic therapy or phototherapy. We granted Biogen an exclusive,
fee-bearing license to use our HAP-TM- Markers, which are shown to be predictive
of response to this biologic, to develop diagnostic tests for use in connection
with marketing Amevive-Registered Trademark-. We also granted Biogen an
exclusive option, for a limited time, to acquire an exclusive license to use
particular HAP-TM- Markers for developing and commercializing other products.
Biogen agreed to pay us a non-refundable initial fee, research funding and
milestone payments based upon the achievement of predetermined goals, as well as
payments for the commercial use of our HAP-TM- Markers in conjunction with the
sale of Amevive-Registered Trademark-. The agreement will automatically
terminate after a defined number of years. Biogen received certain early
termination rights and either party may terminate the agreement early if the
other party breaches the agreement. In December 2002, we amended the agreement
to define additional work that we would perform as part of the original research
plan and for which Biogen agreed to pay us additional research funding.

JOHNSON & JOHNSON PHARMACEUTICAL RESEARCH & DEVELOPMENT, A DIVISION OF
JANSSEN PHARMACEUTICA, N.V.

Effective November 22, 2000, we entered into a collaboration agreement with
Janssen Research Foundation, referred to as J&J PRD, under which we granted J&J
PRD a non-exclusive license to our HAP-TM- Technology in exchange for the
payment of annual subscription fees and other fees described below. We installed
our DECOGEN-REGISTERED TRADEMARK- Informatics System at one of their sites. We
are collaborating with J&J PRD in research projects to identify HAP-TM- Markers
associated with a patient's response to certain J&J PRD drugs. For each of the
first two years of the agreement, we received a minimum fee for providing
HAP-TM- Typing services and we continue to be paid for providing our HAP-TM-
Typing services as part of this collaboration. For the first three research
projects, we defined product license fees, milestone and royalty payments for
drug and diagnostic products that result from these research projects. The
agreement will automatically terminate after three years. Either party may
terminate the agreement early if the other party breaches the agreement. In
November 2002, we amended the agreement with Johnson & Johnson Pharmaceutical
Research & Development, the successor to the Janssen Research Foundation. Under
the terms of the amendment, we granted J&J PRD exclusive commercial licenses to
use HAP-TM- Marker associations with certain drugs in exchange for specified
fees.

12

PHARMACIA & UPJOHN COMPANY

Effective December 12, 2002, we entered into an agreement with Pharmacia &
Upjohn Company, in which Pharmacia received a non-exclusive license to selected
HAP-TM- Markers from our HAP-TM- Database solely for internal research purposes
and to specific components of our DECOGEN-REGISTERED TRADEMARK- Informatics
System, one of which is a proprietary algorithm to build haplotypes, and access
to our HAP-TM- Typing services for genotyping their clinical samples from a
specific project. We installed the DECOGEN-REGISTERED TRADEMARK- Informatics
System at one of their sites. We receive payments based upon the HAP-TM- Markers
that Pharmacia selects and for genotyping their clinical samples and annual fees
for access to our DECOGEN-REGISTERED TRADEMARK- Informatics System. We expect to
complete the genotyping in 2003.

PFIZER INC

Effective August 31, 2001, we entered into a one-year agreement with
Pfizer Inc., in which Pfizer gained non-exclusive access to selected data from
our HAP-TM- Database. We receive payments based upon the HAP-TM- Markers that
Pfizer selects. In May 2002 and February 2003, we amended the agreement to
extend the terms of the agreement first through February 2003 and now through
August 31, 2004, respectively.

INTEC WEB & GENOME INFORMATICS CORPORATION

Effective February 4, 2002, we entered into a two-year agreement with Intec
Web and Genome Informatics Corporation, referred to as Intec W&G, in which we
appointed Intec W&G as a non-exclusive, authorized sales representative with
responsibility for the Japanese market. We agreed to pay Intec W&G a fixed
commission on all payments, excluding royalties, we receive from agreements
concluded through Intec W&G with Japanese companies. To date, Intec W&G has not
brokered any agreements for us.

VISIBLE GENETICS, A PART OF BAYER HEALTHCARE LLC

Effective November 21, 1996, we granted to Visible Genetics, Inc. a
worldwide, exclusive license to our patented technology relating to the coupled
amplification and sequencing, or CAS, of DNA for diagnostic use. This technology
is not part of our HAP-TM- Technology. Under the terms of the agreement, Visible
Genetics paid us a one-time licensing fee and continues to pay us royalties
based on global sales of products using the licensed technology. Visible
Genetics incorporated the CAS technology in its TruGene-TM- HIV diagnostic kit,
which they designed to perform pharmacogenomic analysis of HIV and to customize
HIV and AIDS therapy for particular patient sub-groups. The FDA granted market
clearance for the TruGene-TM- HIV diagnostic kit on September 26, 2001. In
March 2000, we amended the agreement to, among other things, reduce the amount
of royalties payable under the agreement and expand the field of the license to
the research products market. In return for the reduction of royalties and
broadening of the field, Visible Genetics paid us an additional one-time fee of
$2 million. The term of the agreement extends until the last of the patents
covered by the agreement expires. Either party may terminate the agreement early
if the other party breaches the agreement, and we can terminate the agreement
early if Visible Genetics fails to make any payments. In October 2002,
Leverkusen Bayer through its Bayer Corporation completed the acquisition of
Visible Genetics, Inc., which is now part of the Diagnostics Division of Bayer
HealthCare LLC, and assumed the obligations and rights of Visible Genetics under
this agreement.

SEQUENOM

Effective May 28, 2000, we entered into a three-year collaboration agreement
with Sequenom, Inc., under which we committed to use Sequenom's MassARRAY-TM-
system as our exclusive equipment platform for high-throughput SNP analysis in
our HAP-TM- Typing facility. In return, Sequenom provided

13

equipment and supplies, as well as ongoing access to information about new
technology and products in development by Sequenom. In addition, we had the
option to be a test site for these new technologies and products. The agreement
required us to purchase a minimum number of MassARRAY-TM- systems and allowed
for predetermined pricing of consumables. The agreement would have automatically
terminated after three years. In November 2002, we terminated the collaboration
agreement and also received notice from Sequenom that we could use purchased
Sequenom products to provide commercial services, as part of our
pharmacogenomics support services, to third parties, which are not part of our
HAP-TM- Partnership program, without any additional payment or compensation due
to Sequenom for such use.

Through March 11, 2003, we have entered into the following additional
licenses and collaborations:

BAYER AG AND BAYER HEALTHCARE LLC

Effective January 15, 2003, we entered into a research collaboration and an
exclusive license agreement with Bayer AG and with Bayer Healthcare LLC through
its Diagnostics Division to develop pharmacogenomic markers of drug safety and
efficacy for a defined drug category and for certain disease fields. Under the
terms of the agreement, each party has contributed portions of intellectual
property derived from its respective programs. We will receive funding to apply
our HAP-TM- Technology to Bayer's clinical samples. Bayer will receive exclusive
rights to develop and market diagnostic tests based on the results of the
collaboration. We are entitled to receive royalties and rights to perform these
diagnostic tests in our CLIA-licensed diagnostic laboratory. There are mutual
royalty provisions for any pharmaceutical drugs derived from the collaboration.
The collaboration will terminate at the end of defined safety and efficacy
studies. The agreement will terminate upon the expiration of all licenses and
other granted rights and of the obligation to pay royalties. Either party may
terminate the agreement early if the other party breaches the agreement.

MILLENNIUM PHARMACEUTICALS, INC.

Effective January 7, 2003, we entered into a multi-year agreement with
Millennium Pharmaceuticals, Inc., under which we granted Millennium a
non-exclusive license to our HAP-TM- Technology in exchange for the payment of
annual subscription fees. Millennium granted us certain rights to support their
DNA biomarkers and pharmacogenomic efforts. Millennium has an option to pay
specified annual fees to extend the agreement beyond the defined expiration
date. Either party may terminate the agreement early if the other party breaches
the agreement.

WAYNE STATE UNIVERSITY

Effective March 11, 2003, we entered into an agreement with Wayne State
University (WSU) to support WSU's research contract with the National Institute
of Child Health and Human Development's Perinatology Research Branch (PRB),
which is located at the WSU School of Medicine in Detroit, Michigan. Under the
agreement, WSU gained access to specific HAP-TM- Markers and obtained a limited
license to use our DECOGEN-Registered Trademark- Informatics System. We will
develop assays for the selected HAP-TM- Markers and provide high-throughput
genotyping on clinical samples provided by WSU and the PRB. We receive a license
fee and other payments from WSU. Either party may terminate the agreement early
if the other party breaches the agreement. We expect to complete the genotyping
in 2003.

INTELLECTUAL PROPERTY

We rely on patents, trade secrets, non-disclosure agreements, copyrights and
trademarks to protect our proprietary technologies and information. In addition,
our goal is to license to third parties certain components of our intellectual
property that is peripheral to our core products and services.

14

As of March 3, 2003, our patent portfolio included a total of eight issued
patents, which we own or for which we are the exclusive licensee. Our issued
patents and pending patent applications include those for:

- HAP-TM- Markers for pharmaceutically important genes;

- correlations between specific HAP-TM- Markers and response to albuterol
and statins;

- components of our DECOGEN-REGISTERED TRADEMARK- Informatics System,
including the process for assembling HAP-TM- Markers and for determining
clinical associations; and

- processes for integrating personalized medicines into the healthcare
system.

We also rely upon unpatented trade secrets and improvements, unpatented
know-how and continuing technological innovation to develop and maintain our
competitive position. We generally protect this information with reasonable
security measures, including confidentiality agreements that provide that all
confidential information developed or made known to others during the course of
the employment, consulting or business relationship shall be kept confidential
except in specified circumstances. Agreements with employees provide that all
inventions conceived by the individual while employed by us are our exclusive
property.

COMPETITION

There is significant competition among entities attempting to use genomic
variation data and informatics tools to develop and market new and existing
medicines. We expect the intensity of the competition to increase. We face, and
will continue to face, competition from pharmaceutical, biotechnology and
diagnostic companies, both in the United States and abroad. Several entities are
attempting to identify and assemble SNP and haplotype databases for use as a
measure of genomic variation. These databases are based on various technologies
and approaches, including the sequencing of either cDNA or genomic DNA and a
genome-wide approach or a candidate gene approach. In addition, some of these
entities are providing or intend to provide informatics tools for integrating
the use of SNPs and haplotypes into the drug development process. These entities
include, among others, Perlegen Sciences, Celera Genomics Group and the
International HapMap Project. In addition, numerous pharmaceutical companies are
developing internal capabilities for identifying and utilizing gene variation
data. In order to compete successfully against existing and future entities, we
must demonstrate the value of our HAP-TM- Technology and that our informatics
technologies and capabilities are superior to those of our competitors. Many of
our competitors have greater resources, gene variation discovery capabilities
and informatics development capabilities than we do. Therefore, our competitors
may succeed in identifying gene variation and applying for patent protection
more rapidly than we do.

We expect that our ability to compete will be based on a number of factors,
including:

- the ability of our partners to develop and commercialize therapeutic and
diagnostic products based upon our HAP-TM- Technology;

- our ability to attract and retain partners;

- our ability to commercialize our intellectual property;

- our ability to attract and retain qualified personnel;

- our ability to protect against unauthorized use of our HAP-TM- Technology
under various intellectual property laws and contractual obligations; and

- our ability to secure sufficient resources to fund our HAP-TM- Technology
commercialization programs.

15

GOVERNMENT REGULATION

Regulation by governmental entities in the United States and other countries
will be a significant factor in the development, manufacturing and marketing of
any product that our partners or we develop. Various federal and, in some cases,
state statutes and regulations govern or influence the manufacturing, safety,
labeling, storage, record keeping and marketing of human therapeutic and
diagnostic products. The extent to which these regulations may apply to our
partners or us will vary depending on the nature of the product. Currently, the
FDA does not require companies seeking product approvals to provide data
regarding the correlation between therapeutic response and genomic variation.

Virtually all of the pharmaceutical products developed by our partners will
require regulatory approval by governmental agencies prior to commercialization.
In particular, the FDA and similar health authorities in foreign countries will
impose on these products an extensive regulatory review process before they can
be marketed. This regulatory process typically involves, among other
requirements, preclinical studies, clinical trials, and often post-marketing
surveillance of each compound. This process can take many years and requires the
expenditure of substantial resources. Delays in obtaining marketing clearance
could delay the commercialization of any therapeutic or diagnostic products
developed by our partners, impose costly procedures on our partners' activities,
diminish any competitive advantages that our partners may attain and lessen our
potential royalties. Any products our partners develop may not receive
regulatory approval in a timely fashion or at all.

The FDA regulates human therapeutic and diagnostic products in one of three
broad categories: drugs, biologics or medical devices. Products developed using
our technologies could potentially fall into any of these three categories.

The FDA generally requires the following steps for pre-market approval of a
new drug or biologic product:

- preclinical laboratory and animal tests;

- submission to the FDA of an investigational new drug application, which
must become effective before clinical trials may begin;

- adequate and well-controlled human clinical trials to establish the safety
and efficacy of the product for its intended indication;

- submission to the FDA of a new drug application, or NDA, if the FDA
classifies the product as a new drug, or a biologic license application,
or BLA, if the FDA classifies the product as a biologic; and

- FDA review of the NDA or BLA in order to determine, among other things,
whether the product is safe and effective for its intended uses.

The FDA classifies medical devices, which include diagnostic products, as
class I, class II or class III, depending on the nature of the medical device
and the existence in the market of any similar devices. Class I medical devices
are subject to general controls, including labeling, premarket notification and
good manufacturing practice requirements. Class II medical devices are subject
to general and special controls, including performance standards, postmarket
surveillance, patient registries and FDA guidelines. Class III medical devices
are those which must receive premarket approval, or PMA, by the FDA to ensure
their safety and effectiveness, typically including life-sustaining,
life-supporting, or implantable devices or new devices which have been found not
to be substantially equivalent to currently marketed medical devices. It is
impossible to say at this time, which of these categories, will apply to any
diagnostic product incorporating our technologies.

16

Before a new device can be introduced into the U.S. market, it must, in most
cases, receive either premarket notification clearance under section 510(k) of
the Food, Drug, and Cosmetic Act or approval pursuant to the more costly and
time-consuming PMA process. A PMA application must be supported by valid
scientific evidence to demonstrate the safety and effectiveness of the device,
typically including the results of clinical trials, bench tests, laboratory and
animal studies. A 510(k) clearance will be granted if the submitted information
establishes that the proposed device is "substantially equivalent" to a legally
marketed class I or class II medical device or a class III medical device for
which the FDA has not called for PMAs. While less expensive and time-consuming
than obtaining PMA clearance, securing 510(k) clearance may involve the
submission of a substantial volume of data, including clinical data, and may
require a lengthy substantive review.

Even if regulatory clearance is obtained, a marketed product and its
manufacturer are both subject to continuing review. Discovery of previously
unknown problems with a product may result in withdrawal of the product from the
market, which could reduce our revenue sources and hurt our financial results.
Violations of regulatory requirements at any stage during the process, including
preclinical studies and clinical trials, the review process, post-marketing
approval or in manufacturing practices or manufacturing requirements, may result
in various adverse consequences to us, including:

- the FDA's delay in granting marketing clearance or refusal to grant
marketing clearance of a product;

- withdrawal of a product from the market; or

- the imposition of civil or criminal penalties against the manufacturer and
holder of the marketing clearance.

Generally, similar regulatory requirements apply to products intended for
marketing outside the United States.

We use clinical samples of blood from individuals in developing our
intellectual property consisting of HAP-TM- Markers and HAP-TM- Marker
associations. In some cases, a clinical research organization, or CRO, with
which we have a contract, collects these blood samples, plus personal and
medical information about each individual. In other cases, we collect DNA plus
personal and medical information without the assistance of a CRO. Our CRO
prepares, subject to our approval, the sample collection protocol and the
patient informed consent form, as well as identifying the clinical sites, which
collect the samples. The individual clinical sites recruit the patients for each
clinical study and, following the study protocol, explain and obtain the signed
and witnessed informed consent documents from each patient. The informed consent
form includes the patient's authorization to use the patient's blood sample and
data derived from it for developing commercial products. Our contract with the
CRO requires an independent institutional review board to approve the study
protocol, the patient informed consent form, and the transmission of the samples
to us. Either we do not know the identity or we have in place procedures to
maintain the confidentiality of any of the individuals from whom we receive
clinical samples. We believe that these procedures comply with all applicable
federal, state and institutional regulations.

While the FDA does not currently regulate our HAP-TM- Typing facility, CLIA
defines standards that constitute good clinical laboratory practice. Although
this is a federal law, each state is responsible for administering the statute.
In April 2002, the Connecticut Department of Public Health inspected our
laboratory and issued a CLIA license for our HAP-TM- Typing facility, which can
now provide diagnostic test results in support of therapeutic or medical
interventions. As a CLIA licensed diagnostic laboratory, inspectors from the
Connecticut Department of Public Health can inspect our laboratory at any time
to insure that we are in compliance with CLIA.

RESEARCH AND DEVELOPMENT

For the years ended December 31, 2002, 2001 and 2000, we spent approximately
$23.9 million, $46.3 million and $27.4 million, respectively, on research and
development activities.

17

HUMAN RESOURCES

As of March 3, 2002, we had 104 full-time employees, 83 of whom were engaged
in research and development activities, 8 of whom were engaged in business
development and 13 of whom conducted general and administrative functions. Of
the 83 employees engaged in research and development activities, 36 were engaged
in industrial genomics, 33 were engaged in bioinformatics, software development
and information technology, 6 were engaged in medical affairs activities, 5 were
engaged in population genomics and 3 were engaged in intellectual property.
Thirty-one of our employees hold Ph.D. and/or M.D. degrees and 17 hold other
advanced degrees.

None of our employees are covered by a collective bargaining agreement, and
we consider our relations with our employees to be good.

ITEM 1A. EXECUTIVE OFFICERS

Set forth below is certain information regarding our current executive
officers, including their respective ages as of March 3, 2003:

KEVIN RAKIN. Mr. Rakin was appointed Chief Executive Officer, in addition
to President, in August 2002. He cofounded Genaissance and has served as a
Director since 1995. Mr. Rakin has served as the Company's President since
October 2000. From 1995 through 2002, Mr. Rakin also served as Chief Financial
Officer and, from January 1997 to October 2000, as our Executive Vice President.
Prior to 1998, Mr. Rakin was also a Principal at the Stevenson Group, a
consulting firm, where he provided financial and strategic planning services to
high-growth technology companies and venture capital firms. Prior to this,
Mr. Rakin was a manager with Ernst & Young's entrepreneurial services group.
Mr. Rakin holds a B.S. in business and a M.S. in finance from the University of
Cape Town and a M.B.A. from Columbia University.

GUALBERTO RUANO, M.D., PH.D. Dr. Ruano was appointed Vice Chairman and
Chief Scientific Officer in August 2002. From 1995 through August 2002, he
served as Chief Executive Officer. Dr. Ruano has served as a Director since
1995. Prior to cofounding Genaissance, he was involved in developing genetic
mapping products and services for BIOS Laboratories, Inc. and engaged in
research at Yale University, where he focused on haplotyping technologies for
profiling genome diversity stemming from population genetics. Dr. Ruano holds a
B.A. in biophysics from The Johns Hopkins University and a M.D. and a Ph.D. in
population genetics from Yale University, where he was a fellow of the Medical
Scientist Training Program and the Ford Foundation.

GERALD F. VOVIS, PH.D. Dr. Vovis was appointed Executive Vice President, in
addition to Chief Technology Officer, in April 2002. He has served as our Chief
Technology Officer since October 2000. From October 2000 to April 2002,
Dr. Vovis was a Senior Vice President and, from April 1999 to October 2000, was
our Senior Vice President of Genomics. From 1980 to 1999, he was affiliated with
Genome Therapeutics Corporation, a genomics company, most recently as Senior
Vice President of Scientific Affairs. Dr. Vovis has twenty-three years of
experience in the management of genetic research

18

and in the development and management of collaborative research programs with
pharmaceutical and biotechnology companies. Dr. Vovis holds a B.A. in chemistry
from Knox College and a Ph.D. in biology from Case Western Reserve University.

RICHARD S. JUDSON, PH.D. Dr. Judson has been our Senior Vice President of
Medical Affairs and Informatics since August 2002. From April 2000 to
August 2002, Dr. Judson was our Senior Vice President of Informatics and, from
November 1999 to April 2000, our Vice President of Informatics. He joined
Genaissance in February 1999 as Associate Director, Bioinformatics. From
January 1997 to February 1999, Dr. Judson served as Group Leader in the
Bioinformatics Department of CuraGen Corporation, a genomics company, where he
was responsible for developing software for protein-protein interactions and DNA
sequence analysis. From January 1990 to December 1996, he served as Senior
Member of the Technology Staff at Sandia National Laboratories, leading modeling
projects in several areas including computational drug design, protein modeling
and sequence analysis. He holds a B.A. in chemistry and physics from Rice
University and a M.A. and a Ph.D. in chemistry from Princeton University.

JOSEPH KEYES. Mr. Keyes was appointed Chief Financial Officer and Vice
President in July 2002. Previously, he served as Executive Director of Finance
since joining the Company in March 2001. From March 2000 until joining
Genaissance, Mr. Keyes served as Vice President of Finance at DSL.net. From
June 1999 to February 2000, he served as Vice President, Finance and Chief
Financial Officer at Heritage Consumer Products. Previous to that, Mr. Keyes was
the Group Controller for United States Surgical Corporation from 1992 to 1999.
He holds a B.S. in business administration from Bryant College. Mr. Keyes is a
certified public accountant.

ITEM 2. PROPERTIES

Our executive offices and laboratories are located at Five Science Park, New
Haven, Connecticut. We lease nearly 72,000 square feet of space, under a lease
expiring on September 30, 2006, which we may extend for 10 years. We believe
that our current facilities and the space available to us under a lease
extension are suitable to meet our current requirements and that suitable
additional space will be available on commercially reasonable terms, if
required.

ITEM 3. LEGAL PROCEEDINGS

We are not a party to any material legal proceedings.

ITEM 4. SUBMISSION OF MATTERS TO A VOTE OF SECURITY HOLDERS

No matters were submitted to stockholders for a vote during the fourth
quarter of 2002.

PART II

ITEM 5. MARKET FOR THE REGISTRANT'S COMMON STOCK AND RELATED STOCKHOLDER MATTERS

Our common stock is currently quoted on the Nasdaq National Market under the
symbol "GNSC."

19

Our common stock began trading on August 1, 2000 and the high and low
closing sale prices, as reported on the Nasdaq National Market for the periods
indicated, were as follows:

As of March 14, 2003, there were approximately 279 holders of record of our
common stock.

We have never paid cash dividends on our common stock and we do not
anticipate paying any cash dividends in the foreseeable future. We currently
intend to retain future earnings, if any, for use in our business.

20

ITEM 6. SELECTED FINANCIAL DATA

The following selected financial data should be read in conjunction with our
financial statements and related notes and "Management's Discussion and Analysis
of Financial Condition and Results of Operations" appearing elsewhere in this
Annual Report on Form 10-K. The selected balance sheet data set forth below, as
of December 31, 2002, and the statements of operations data for the year ended
December 31, 2002, are derived from our financial statements, which have been
audited by PricewaterhouseCoopers LLP, independent public accountants, and are
included elsewhere in this Annual Report on Form 10-K. Financial statements for
fiscal years 1998 through 2001 were audited by Arthur Andersen LLP (Andersen)
which has ceased operations. A copy of the report previously issued by Andersen
on our financial statements as of December 31, 2000 and 2001 and for each of the
three years in the period ended December 31, 2001 is included elsewhere in this
Annual Report on Form 10-K. Such report has not been reissued by Andersen. The
historical results are not necessarily indicative of the results we expect for
future periods. This data is in thousands, except per share data.

21

ITEM 7. MANAGEMENT'S DISCUSSION AND ANALYSIS OF FINANCIAL CONDITION AND RESULTS
OF OPERATIONS

THE FOLLOWING DISCUSSION AND ANALYSIS OF OUR FINANCIAL CONDITION AND THE
RESULTS OF OPERATIONS SHOULD BE READ IN CONJUNCTION WITH THE "SELECTED FINANCIAL
DATA" AND OUR FINANCIAL STATEMENTS AND RELATED NOTES APPEARING ELSEWHERE IN THIS
ANNUAL REPORT ON FORM 10-K.

THIS ANNUAL REPORT ON FORM 10-K CONTAINS FORWARD-LOOKING STATEMENTS WITHIN
THE MEANING OF SECTION 21E OF THE SECURITIES EXCHANGE ACT OF 1934, AS AMENDED.
FOR THIS PURPOSE, STATEMENTS CONTAINED HEREIN THAT ARE NOT STATEMENTS OF
HISTORICAL FACT MAY BE DEEMED TO BE FORWARD-LOOKING STATEMENTS. WITHOUT LIMITING
THE FOREGOING, THE WORDS "BELIEVES," "ANTICIPATES," "PLANS," "EXPECTS" AND
SIMILAR EXPRESSIONS ARE INTENDED TO IDENTIFY FORWARD-LOOKING STATEMENTS. THESE
FORWARD-LOOKING STATEMENTS INVOLVE RISKS AND UNCERTAINTIES AND ARE NOT
GUARANTEES OF FUTURE PERFORMANCE. ACTUAL RESULTS MAY DIFFER MATERIALLY FROM
THOSE INDICATED IN SUCH FORWARD-LOOKING STATEMENTS AS A RESULT OF CERTAIN
FACTORS INCLUDING, BUT NOT LIMITED TO, THOSE SET FORTH UNDER THE HEADING
"FACTORS AFFECTING FUTURE OPERATING RESULTS".

OVERVIEW

Since our inception, we have incurred significant operating losses and, as
of December 31, 2002, we had an accumulated deficit of $193.8 million. The
majority of our operating losses have resulted from costs we incurred developing
our HAP-TM- Technology, in our clinical trials and administrative costs
associated with operations. We expect to dedicate a significant portion of our
resources for the foreseeable future to service our HAP-TM- Partnership program
and STRENGTH program partners, our pharmacogenomics support services customers,
our CARING program and to maintain our HAP-TM- Technology. To date, our revenue
has been primarily from licensing and service fees from our agreements with
AstraZeneca UK Limited, Biogen, Inc., Gene Logic, Inc., J&J PRD and
Pfizer, Inc., as well as a sublicensing agreement with Visible Genetics, Inc.
and, to a lesser extent, government grants.

In August 2002, we announced a restructuring and cost reduction program to
revise our business focus and to better align our operating cost structure with
our current and projected partner needs and projected revenues from our current
and projected partners. The cost reduction program included a realignment of
management responsibilities, a reduction in our workforce and a decision to seek
partners for all internal product development programs. The workforce was
reduced by 20 percent to 110 employees in the third fiscal quarter, with the
majority of the workforce reductions occurring in the DNA sequencing facility
and related informatics support. We incurred a charge for severance and related
costs of approximately $200,000, which was recorded in operating results in the
third fiscal quarter of 2002. All severance obligations associated with this
workforce reduction had been paid by December 31, 2002.

As a result of the restructuring and cost reduction program, we expect
future operating expenses to decrease from prior levels, primarily in research
and development expenses. In addition to reducing expenses, our plans and
projections reflect a measurable reduction in our negative operating cash flow.
These planned reductions in our negative operating cash flow assume significant
year over year increases in revenues. There can be no assurance that revenues
will continue to increase or that there will continue to be a decrease in the
level of cash used to fund operations. If we are not successful in increasing
revenues or reducing expenses, as planned, we may not be able to maintain our
operations at planned levels.

During the second quarter of 2002, we recorded a $6.0 million charge for the
impairment of fixed assets. This charge relates to sequencing equipment,
computer hardware and software and leasehold improvements in our DNA sequencing
facility that we determined needed to be reviewed for potential impairment. As a
result of our review, we determined that the carrying value of the assets was in
excess of discounted future cash flows to be generated by the asset group and we
recorded a write-down of

22

$6.0 million. We continue to review our fixed assets for potential impairment
and, based upon our current cash flow projections, have determined that there
has not been an additional impairment of our fixed assets.

In November 2002, we announced that we had amended our collaboration
agreement with a customer which accounted for 46% of our revenue for the year
ended December 31, 2002. The amendment clarifies the intellectual property
rights granted pursuant to the agreement. This agreement expires in the fourth
quarter of 2003 unless an extension is negotiated. A collaboration agreement
with a customer that accounted for 25% of our revenue for the year ended
December 31, 2002, expired in the fourth quarter of 2002.

In December 2002, we announced that we received a notice from General
Electric Capital Corporation (GE), claiming that an event of default had
occurred under our lease agreement as a result of an alleged material adverse
change in our business. We also received a notice from Finova Capital
Corporation (Finova), stating that as a result of the default claim by GE, there
was a cross-default under our agreement with Finova. Both lessors declared that
all principal and future interest obligations were immediately due and payable.
In March 2003, we settled the claim with GE and simultaneously amended our lease
agreement (GE Amendment). In connection with the GE Amendment, and as additional
security for our payment obligations to GE, we delivered to GE a $2 million
irrevocable letter of credit and agreed to additional covenants. It is expected
that the letter of credit will be adjusted on a quarterly basis commencing
October 1, 2003 based on a decrease in outstanding amounts due to GE, as
defined. In connection with the amendment, GE retroactively rescinded their
default claim. The GE Amendment and letter of credit has been retroactively
reflected in the accompanying 2002 financial statements. We have not entered
into a settlement agreement with Finova. Accordingly, we have recorded an
additional charge to interest expense to reflect the remaining interest and
principal currently due and payable under the Finova capital lease agreement.

We enter into discussions from time to time regarding the acquisition of or
strategic investment in other businesses or technologies. We are currently in
negotiations with respect to the acquisition of all or substantially all of the
assets of a company with technologies we believe are complementary to ours. We
cannot assure you if we will be successful in these negotiations.

CRITICAL ACCOUNTING POLICIES AND ESTIMATES

Our financial statements have been prepared in accordance with accounting
principles generally accepted in the United States of America (GAAP). Our
significant accounting policies are described in Note 1 to the Financial
Statements. The preparation of financial statements in accordance with GAAP
requires management to make estimates and assumptions that affect the amounts
reported in the Financial Statements and accompanying footnotes. We base our
estimates on historical experience and on various other assumptions that we
believe to be reasonable under the circumstances, the results of which form our
basis for making judgments about the carrying values of assets and liabilities
that are not readily apparent from other sources. Actual results may differ from
estimates under different conditions, sometimes materially. Critical accounting
policies and estimates are defined as those that are both most important to the
portrayal of our financial condition and results and require management's most
subjective judgments.

Our critical accounting policies are as follows:

- Revenue recognition

- Valuation of long-lived assets

- Research and development expenses

23

REVENUE RECOGNITION. We recognize license and service revenue from our
agreements with third parties. The revenue includes fees we receive for the
license of our HAP-TM- database and proprietary software, milestone payments
based on the achievement of certain goals and services fees for providing
specific data on genetic variation. Upfront, non-refundable fees received in
connection with a collaboration agreement are deferred and amortized into
revenue over the term of the agreement. License revenues are recognized ratably
over the access period of the agreement. Revenue derived from the achievement of
a milestone is recognized when the milestone is achieved, provided that the
milestone is substantive and a culmination of the earnings process has occurred.
Service fees are recorded as the services are performed. Revenues derived from
the achievement of milestones or recognition of related work when performed
under the terms of a contract, as well as the access period of the license
agreement, may cause our operating results to vary considerably from period to
period.

VALUATION OF LONG-LIVED ASSETS. We assess the impairment of long-lived
assets whenever events or changes in circumstances indicate that the carrying
value of the assets may not be recoverable. Factors which could trigger an
impairment review, include:

- a significant adverse change in the extent or manner in which a long-lived
asset is being used;

- a significant adverse change in the business climate that could affect the
value of a long-lived asset; and

- a significant decrease in the market value of assets.

When determining whether the carrying value of the long-lived assets is
recoverable we make certain estimates and assumptions regarding the undiscounted
cash flows and discounted cash flows expected to be generated by the assets. We
determine the discounted future cash flow using a discount rate determined by
our management to be commensurate with the risk inherent in our current
business. A change in our estimates or the manner in which we use the asset may
cause results to vary from period to period.

RESEARCH AND DEVELOPMENT EXPENSES. We record research and development
expenses when they are incurred or, in the case of clinical trial expenses,
based upon information we receive from third party CROs. Research and
development expenses include the following major types of costs: salaries and
benefits, material and reagent costs, research license fees, clinical trial
expenses, depreciation and amortization of lab equipment and leasehold
improvements and building and utility costs related to research space. Vendor
contractual costs consist primarily of consulting arrangements and certain
clinical trial expenses. We expense clinical trial costs as incurred based on
information we receive from third parties and estimates that we make. Our
estimates may change as additional information becomes available which could
cause results to vary from period to period.

RESULTS OF OPERATIONS

YEARS ENDED DECEMBER 31, 2002 AND 2001

Revenue consists primarily of proceeds received in connection with the
licensing of our HAP-TM- Technology, service revenue and sublicensing of
patents. Revenue increased to $8.1 million in 2002 from $5.3 million in 2001.
The increase in license revenues is attributable to the commercialization of our
HAP-TM- Technology, including agreements entered into with Biogen during 2002,
AstraZeneca during 2001 and J&J PRD during 2000. Revenue from Gene Logic, Inc.,
J&J PRD and Pfizer, Inc. accounted for 82% and 94% of our total revenue in
fiscal 2002 and 2001, respectively. Revenue from each of these customers
accounted for 10% or more of our total revenue for 2002 and 2001, respectively.
Revenue for fiscal 2002 includes a payment received from J&J PRD in the fourth
quarter for achieving a milestone. We are recognizing the annual license and
subscription fees over the term of the agreements and the service fees as the
services are performed. Future milestone and royalty payments, when and if
received, will be recognized when earned. Revenue also includes the

24

amortization, over the remaining life of the sublicensed patent, of upfront
payments received in connection with the sublicensing of a patent.

Research and development expenses consist primarily of payroll and benefits
for research and development personnel, materials and reagent costs, costs
incurred in connection with clinical trials, depreciation and maintenance costs
for equipment used for HAP-TM- Marker discovery and HAP-TM- Typing, and
facility-related costs. We expense our research and development costs as
incurred. Research and development costs decreased to $23.9 million in 2002 from
$46.3 million in 2001. The decrease in research and development expenses in 2002
is primarily attributable to a decrease of approximately $9 million in
expenditures related to our clinical trials, a decrease of approximately
$5 million in material and reagent costs associated with the discovery of new
HAP-TM- Markers, a decrease of approximately $2 million in payroll and related
costs and a decrease of approximately $1 million in technology license fees. The
decrease in clinical trial expenses is primarily due to our STRENGTH Trials,
which were substantially completed by December 31, 2001. The decrease in
material and reagent costs is primarily due to a reduction in DNA sequencing
consistent with the removal of the majority of our ABI
Prism-Registered Trademark- 3700 DNA Analyzers from production in the quarter
ended June 30, 2002. The decrease in payroll and related costs is primarily due
to our work force reduction in connection with our restructuring program. We
expect to continue to devote substantial resources to research and development
expenses in the near future as we continue to service our HAP-TM- Partnership
program and STRENGTH program partners, our pharmacogenomic support services
customers, our CARING program and to maintain our HAP-TM- Technology. We expect
research and development expenses to decrease modestly in 2003 as a result of
the restructuring program that we initiated in August 2002.

General and administrative expenses consist primarily of payroll and
benefits for executive, business development, finance, public affairs and other
administrative personnel, as well as facility related costs and outside
professional fees incurred in connection with corporate development, general
legal and financial matters. General and administrative expenses decreased to
approximately $8.8 million in 2002 from $11.9 million in 2001. The decrease in
general and administrative expenses in 2002 is primarily due to a $2 million
decrease in consulting and professional service fees and a general reduction in
expenses as part of our cost reduction program initiated during 2002.

The $6.0 million impairment of fixed assets charge relates to sequencing
equipment, computer hardware and software and leasehold improvements in our DNA
sequencing facility. During the quarter ended June 30, 2002, our management
determined that certain conditions had arisen during the quarter that triggered
the need for a review of our long-lived assets for potential impairment. These
conditions included, but were not limited to, the overall business climate in
which we operate and a significant change in the manner in which we were
utilizing our DNA sequencing facility and related assets. In particular, during
the quarter ended June 30, 2002, we determined that our sequencing production
capacity significantly exceeded our forecasted demand for the foreseeable
future, which resulted in our decision to remove from production the majority of
our ABI Prism-Registered Trademark- 3700 DNA Analyzers, the primary assets of
the group. Accordingly, we performed an impairment review on our sequencing
long-lived assets. As a result of our review, we determined that the carrying
value of the assets was in excess of the projected undiscounted cash flows to be
generated by the asset group. To determine the amount of the impairment charge,
we compared the carrying value of the applicable fixed assets to their fair
value. We determined the fair value of the fixed assets by discounting expected
future cash flows using a discount rate determined by our management to be
commensurate with the risk inherent in our current business. As a result of our
analysis, we determined that the carrying value of the assets was in excess of
discounted future cash flows to be generated by the asset group and we recorded
a write-down of $6.0 million. The impairment charge has been allocated to the
individual assets on a pro-rata basis. The revised carrying value of the assets
is being depreciated over the average remaining life of the primary assets of
the group.

25

Interest income decreased to approximately $1.0 million in 2002 from
$3.9 million in 2001. The decrease is the result of higher cash, cash equivalent
and short-term investment balances in 2001 and the much lower interest rates on
investments during 2002.

Interest expense increased to approximately $3.5 million in 2002 from
$2.6 million in 2001. The increase is primarily due to recording additional
interest expense during the quarter ended December 31, 2002, in connection with
certain events of default. In December 2002, we received a notice from GE
claiming that an event of default had occurred under our lease agreement as a
result of an alleged material adverse change in our business. Because of the
alleged default, GE declared that all principal and future interest obligations
were immediately due and payable. GE also filed a complaint in the Superior
Court of the State of Connecticut, demanding payment of all amounts due under
the lease agreement. We also received a notice from Finova, stating that as a
result of the default claim by GE, there had been a cross-default under our
agreement with Finova. Finova declared that all principal and future interest
obligations were immediately due and payable. In March 2003, we settled the
claim with GE and simultaneously amended our lease agreement (GE Amendment). In
connection with the amendment, GE retroactively rescinded their default claim
and both parties have withdrawn from any legal action. As we have not entered
into a settlement agreement with Finova, we have recorded an additional charge
to interest expense to reflect the fact that all future interest and principal
is currently due and payable under the Finova capital lease agreement. If, in
the future, we enter into a settlement or amend our agreement with Finova it may
result in the reversal, in the period of amendment or settlement, of the
remaining interest expense recognized in the fourth quarter. The status of the
capital lease obligations is discussed further in "Liquidity and Capital
Resources".

State income tax expense (benefit) decreased to an expense of $35,000 in
2002 from a benefit of $4.1 million in 2001. The benefit represents a net tax
benefit from the State of Connecticut as a result of legislation which allowed
companies to receive cash refunds from the State at a rate of 65% of their
incremental research and development tax credit, as defined, in exchange for
foregoing the carryforward of the research and development tax credit. The State
of Connecticut rescinded the legislation in 2002.

YEARS ENDED DECEMBER 31, 2001 AND 2000

Revenue consists primarily of proceeds received in connection with the
licensing of our HAP-TM- Technology, service revenue and sublicensing of
patents. Revenue increased to $5.3 million in 2001 from $753,000 in 2000. The
increase in license revenues is attributable to the commercialization of our
HAP-TM- Technology, including an agreement entered into with Pfizer during 2001,
as well as agreements entered into with J&J PRD and Gene Logic, Inc. during
2000. Revenue from each of J&J PRD, Gene Logic, Inc. and Pfizer, Inc. accounted
for 10% or more of our total revenue in fiscal 2001. J&J PRD, Gene Logic, Inc.
and Visible Genetics, Inc. each accounted for 10% or more of our total revenue
in 2000. We are recognizing the annual license and subscription fees over the
term of the agreements and the service fees as the services are performed.
Future milestone and royalty payments, when and if received, will be recognized
when earned. Revenue also includes the amortization, over the remaining life of
the sublicensed patent, of upfront payments received in connection with the
sublicensing of a patent.

Research and development expenses consist primarily of payroll and benefits
for research and development personnel, materials and reagent costs, costs
incurred in connection with clinical trials, depreciation and maintenance costs
for equipment used for HAP-TM- Marker discovery and HAP-TM- Typing, and
facility-related costs. Research and development costs increased to
$46.0 million in 2001 from $25.7 million in 2000. Research and development
expenses include stock based and other non-cash compensation charges of $351,000
and $1.7 million for 2001 and 2000, respectively, for options granted to
employees, scientific advisory board members and consultants. The overall
increase in research and development expenses in 2001 is primarily attributable
to an increase in expenditures

26

related to our clinical trials, the expansion of our HAP-TM- Typing process, and
the increase in resources dedicated to supporting and improving our proprietary
DECOGEN-Registered Trademark- Informatics System. The increase in expenses
includes a $7.9 million increase in costs related to our clinical trials, a
$2.7 million increase in payroll and related costs, a $2.3 million increase in
technology licenses and a $4.6 million increase in depreciation expense. The
decrease in stock based compensation is primarily due to our decision to vest
fully all unvested options previously granted to scientific advisory board
members in March 2000, which resulted in a one-time expense of approximately
$1.4 million.

General and administrative expenses consist primarily of salary and related
costs for executive, business development, finance, public affairs and other
administrative personnel, as well as facility related costs and outside
professional fees incurred in connection with corporate development, general
legal and financial matters. General and administrative expenses increased to
approximately $11.8 million in 2001 from $8.8 million in 2000. General and
administrative expenses include stock based and other non-cash compensation
charges of $147,000 and $3.6 million for 2001 and 2000, respectively for options
granted to employees, directors and consultants. The increase in general and
administrative expenses in 2001 is primarily due to increased salary and related
costs as a result of the expansion of our business development activities and
the additional costs of operating as a public company for a full fiscal year.
The increase in general and administrative expenses is partially offset by a
decrease in stock based compensation due to the recording of approximately
$2.9 million of expense during 2000 which related to a stock purchase agreement
between two officers and a former executive officer which we recognized as
compensatory and accordingly recognized non-cash compensation based on the
increase in the fair value of the stock through the date of our initial public
offering.

Sublicensing royalty expense represents royalty obligations incurred by us
on sublicensing fees that we receive. Sublicensing royalty expense decreased to
$54,000 in 2001 from $530,000 in 2000. This decrease relates primarily to a
nonrefundable cash payment received in 2000 in connection with an amendment to a
patent sublicensing agreement. We elected to recognize this expense in 2000 as
paid.

Interest income (expense), net decreased to approximately $1.3 million in
2001 from $2.8 million in 2000. The decrease is the result of higher cash, cash
equivalent and short-term investment balances in 2000 as a result of proceeds
received from our sale of Series B and Series C preferred stock in February and
March 2000, respectively, and our initial public offering in August 2000. The
decrease is also the result of an increase in interest expense as a result of
higher capital lease and other debt obligations to fund the acquisition of
equipment and partially fund the expansion of our facilities.

State income tax benefit of $4.1 million represents a net tax benefit from
the State of Connecticut as a result of recent legislation which allows
companies to receive cash refunds from the State at a rate of 65% of their
incremental research and development tax credit, as defined, in exchange for
foregoing the carryforward of the research and development tax credit.

LIQUIDITY AND CAPITAL RESOURCES

We have financed our operations primarily through the private and public
sale of common and preferred stock, government research grants, payments under
licensing agreements, loans and capital leases. From inception through
December 31, 2002, we have received aggregate gross proceeds of approximately
$163.1 million from issuance of common and preferred stock. In addition, through
December 31, 2002, we have received $4.5 million of government grant funding and
$17.8 million from license and service fees, royalties and research contracts.
We also have received $26.2 million from capital lease financing arrangements
and $8.2 million from other loans. Through December 31, 2002, we have acquired
$41.8 million of property and equipment. These assets were largely financed
through capital lease financing arrangements and other loans.

We expect to continue to finance our operations in the short-term from cash
we received in 2000 from the sale of our common and preferred stock and revenue
from our HAP-TM- Partnership program

27

partners and our pharmacogenomics support services customers. Our business
strategy depends on, among other things, entering into partnership agreements
with pharmaceutical and biotechnology companies. If we are unsuccessful in
marketing our partnership programs and pharmacogenomic support services, we may
not generate sufficient revenues to sustain our operations at planned levels.

Cash used in operations for the year ended December 31, 2002 was
$18.5 million compared with $40.1 million for the same period in 2001. The cash
used in operations for the year ended December 31, 2002 resulted primarily from
a net loss of $33.1 million and a $4.9 million decrease in accounts payable and
accrued liabilities, partially offset by $14.2 million of non-cash charges for
depreciation and amortization expense and impairment of fixed assets, a
$2.6 million increase in deferred revenue and a $2.0 million decrease in other
current assets including the receipt, in September 2002, of $1 million from the
State of Connecticut for the sale of research and development tax credits.

Cash provided by investing activities was $18.8 million in 2002 compared
with $2.5 million in 2001. In 2002, we used cash to purchase $70,000 of property
and equipment and received proceeds of $28.5 million from the liquidation of
investments in marketable securities and used cash to purchase $9.6 million of
marketable securities.

Cash used in financing activities was $9.0 million in 2002 compared to
$6.4 million in 2001. During 2002, we repaid debt of approximately $6.9 million
and were required to provide cash collateral of $2.1 million in connection with
the letter of credit issued as part of the GE Amendment.

Our contractual cash obligations as of December 31, 2002, are as follows:

- ------------------------

* Reflects all lease obligations as being current. However, as a result of the
March 2003 GE Amendment, we anticipate paying the outstanding GE lease
obligation under the original maturity schedule through February, 2005.

Long-term debt consists primarily of three financing agreements with
Connecticut Innovations, Inc. (CII), a stockholder of the Company, used to
finance certain leasehold improvements and other costs associated with our
facility expansion. Each agreement provides for interest only for a certain
period with principal payments, based on a 120 month amortization, beginning in
April 2001 through October 2002, and with final balloon payments due in
March 2009 through June 2011. Borrowings under the agreements bear interest at
6.5% and are secured by the related leasehold improvements. The Company was in
compliance with all debt covenants as of December 31, 2002 and no cross default
provisions exist in the agreements.

Capital lease obligations, related to equipment, consist principally of
arrangements with four equipment leasing companies. The leases have terms
ranging from two to four years with installments originally scheduled to end
between August 2003 and October 2004 and which bear interest at rates ranging
from 8.15% to 12.46%. The majority of the capital lease arrangements allow for
the purchase of the related equipment at the completion of the lease term, as
defined in the agreements, and certain of the agreements require the purchase of
the equipment at the completion of the lease term. In

28

December 2002, the Company announced it received a notice from GE claiming that
an event of default had occurred under the lease agreement as a result of an
alleged material adverse change in the Company's business. Because of the
alleged default, GE declared that all principal and future interest obligations
were due and payable immediately. GE also filed a complaint in the Superior
Court of the State of Connecticut, demanding payment of all amounts due under
the lease agreements. The Company also received a notice from Finova, stating
that as a result of the alleged default claim by GE, there was a cross-default
under the Company's agreement with Finova. In connection with a cross-default,
Finova declared that all principal and future interest obligations were due and
payable immediately. In March 2003, the Company and GE agreed to an amendment to
their existing lease agreement. In connection with this amendment, GE has
rescinded and withdrawn its default claim and both parties have withdrawn from
any legal action. The lease amendment requires certain additional covenants as
well as a $2 million letter of credit, which is collateralized by cash. Because
of the outstanding claim of default by Finova, the Company recorded an
additional charge to interest expense during 2002. This charge represents all
future interest payments due under the terms of the agreement with Finova. As of
December 31, 2002, the Company has classified as current the GE and Finova lease
obligations, as a result of continuing material adverse change provisions within
the GE agreement and as a result of the Finova default claim.

We lease our operating facilities located in New Haven, Connecticut. The
lease agreements require annual lease payments of $816,000 per year increasing
to $1.1 million per year over the original term which expires in 2006. We have
two five-year renewal options to extend the lease agreements beyond the initial
term. We are recording the expense associated with the lease on a straight-line
basis over the expected term of the lease. In addition to the operating lease
agreements for our current facility, we also have operating leases for various
office equipment.

In addition, we periodically enter into agreements with third parties to
obtain exclusive or non-exclusive licenses for certain technologies. The terms
of certain of these agreements require us to pay future royalty payments and
certain milestone payments based on product sales or sublicense income generated
from applicable technologies, if any. The amount of such payments will depend
upon successful commercialization of applicable technologies, if any. The future
minimum payments (assuming non-termination of these leases) are included in the
minimum license obligations above.

Capital expenditures are not expected to exceed $500,000 to $1 million for
each of fiscal 2003 and 2004.

Our cash requirements will vary depending upon a number of factors, many of
which are beyond our control, including:

- the demand for our HAP-TM- Technology;

- the efforts and success of our HAP-TM- Partnership program;

- the commercialization of intellectual property derived from our
associations;

- the level of competition we face;

- our ability to maintain our HAP-TM- Technology; and

- our ability to manage effectively our operating expenses.

On December 31, 2002, cash, cash equivalents and short-term investments
totaled $34.2 million, which includes $2.1 million of restricted cash, compared
to approximately $59.7 million at December 31, 2001. Our cash reserves are held
in interest-bearing, high-grade corporate bonds and money market accounts. We
believe that our existing cash reserves will be sufficient to fund our expected
net losses, debt obligations and capital expenditures for at least 12 months. To
execute our business plan, we will need to grow our revenues significantly each
year and we may need to seek

29

additional funding through public or private equity offerings, debt financings
or commercial partners. We cannot assure you that we will obtain additional
partners or capital funding on acceptable terms, if at all.

INCOME TAXES

We have not generated any taxable income to date and, therefore, have not
paid any federal income taxes since inception. On December 31, 2002, we had
available unused net operating loss carryforwards of approximately
$121.1 million and $120.1 million, which may be available to offset future
federal and state taxable income, respectively. Use of our federal and state net
operating loss carryforwards, which will begin to expire in 2007 and 2003,
respectively, may be subject to limitations. The future utilization of these
carryforwards may be limited due to changes within our current and future
ownership structure as defined within the income tax code. We have recorded a
full valuation allowance against our deferred tax asset, which consists
primarily of net operating loss carryforwards, because of uncertainty regarding
its recoverability, as required by Statement of Financial Accounting Standards
No. 109 ACCOUNTING FOR INCOME TAXES.

RECENT ACCOUNTING PRONOUNCEMENTS

In June 2002, the Financial Accounting Standards Board (FASB) issued
Statement of Financial Accounting Standards No. 148, ACCOUNTING FOR STOCK-BASED
COMPENSATION--TRANSITION AND DISCLOSURE, AN AMENDMENT OF FASB STATEMENT
NO. 123. SFAS No. 148 amends SFAS No. 123, ACCOUNTING FOR STOCK-BASED
COMPENSATION, to provide alternative methods of transition for a voluntary
change to the fair value-based method of accounting for stock-based employee
compensation. In addition, SFAS No. 148 amends the disclosure requirements of
SFAS No. 123 to require prominent disclosures in both annual and interim
financial statements about the method of accounting for stock-based employee
compensation and the effe