ISIS PHARMACEUTICALS INC - 10-K Annual Report

Isis Pharmaceuticals, Inc.
(Exact name of Registrant as specified in its charter)

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 ý No o

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 this Form 10-K. o

Indicate by check mark whether the registrant is an accelerated filer (as defined in Rule 12(b)-2 of the Securities Exchange Act of 1934). Yes ý No o

The approximate aggregate market value of the voting common stock held by non-affiliates of the registrant, based upon the last sale price of the common stock reported on the National Association of Securities Dealers Automated Quotation National Market System was $227,642,048 as of June 30, 2003.*

The number of shares of voting common stock outstanding as of March 3, 2004 was 55,883,687.

Portions of the registrant's definitive Proxy Statement filed on or about April 15, 2004 with the Securities and Exchange Commission in connection with Registrant's annual meeting of stockholders to be held on May 26, 2004, is incorporated by reference into Part III of this Report. The Exhibit Index (Item No. 15) located on pages 59 to 63 incorporates several documents by reference as indicated therein.

This Form 10-K contains forward-looking statements regarding our business and the therapeutic and commercial potential of our technologies and products in development. Any statement describing our goals, expectations, intentions or beliefs is a forward-looking statement and should be considered an at-risk statement. Such statements are subject to certain risks and uncertainties, particularly those risks or uncertainties inherent in the process of developing technology and systems used to identify infectious agents, discovering, developing and commercializing drugs that can be proven to be safe and effective for use as human therapeutics, and in the endeavor of building a business around such products and services. Actual results could differ materially from those discussed in this Form 10-K. Factors that could cause or contribute to such differences include, but are not limited to, those discussed in this Form 10-K including those identified in the section of Item 1 entitled "Risk Factors". As a result, you should not rely on these forward-looking statements.

Vitravene® is a registered trademark of Novartis AG. Taxotere® is a registered trademark of Aventis Pharmaceuticals, Inc. GeneTrove® and Ibis Therapeutics® are registered trademarks of Isis Pharmaceuticals, Inc. Affinitak™ is a trademark of Eli Lilly and Company. HepaSense™ is a trademark of HepaSense Ltd. Orasense™ is a trademark of Orasense Ltd. Macugen™ is a trademark of Eyetech Pharmaceuticals, Inc. Alnylam® is a registered trademark of Alnylam Pharmaceuticals, Inc. InterfeRx™ is a trademark of Alnylam Pharmaceuticals, Inc.

We are a biopharmaceutical company exploiting proprietary RNA-based drug discovery technologies to identify and commercialize novel drugs to treat important diseases. RNA, or ribonucleic acid, is a molecule that provides to a cell the information the cell needs to produce proteins, including those proteins involved in disease. Interference with RNA can keep the body from producing proteins that are involved in disease. We have a strong proprietary position in RNA-based drug discovery technologies. With our primary technology, antisense, we create inhibitors, or oligonucleotides, designed to hybridize, or bind, with high specificity to their RNA target and modulate the production of proteins associated with diseases. We also use our antisense technology in collaborations with pharmaceutical companies to identify and prioritize attractive gene targets for their drug discovery programs. We are a leader in exploiting RNA as a target for drugs. Within our Ibis program, we are expanding on our RNA expertise by creating a platform technology designed to rapidly and accurately identify a broad range of organisms in a single test. Our ongoing development of this technology and a biosensor system related to this technology have been funded primarily by agencies of the United States Government.

We used our antisense technology to commercialize our first product, Vitravene®. Vitravene demonstrates our ability to meet Food and Drug Administration, or FDA, and European regulatory requirements and to commercially manufacture antisense drugs. We currently have 11 antisense products in our development pipeline with 10 in human clinical trials designed to assess safety and efficacy. Our products in development address numerous therapeutic areas with major market potential, including inflammatory, viral, metabolic, dermatological, and cardiovascular diseases, and cancer. We are expanding the therapeutic opportunities for antisense drugs by developing a variety of formulations to enhance patient convenience and compliance. In addition, we are advancing antisense drugs using second-generation chemistry. Physicians may be able to dose our second-generation drugs, which represent over half of our drugs in development, as infrequently as once per month. We are also making progress on developing oral formulations of our second-generation drugs. Recent clinical trial data showed the potential feasibility of oral solid dosage forms for antisense drugs. This oral formulation platform may increase the commercial competitiveness of any antisense drugs we may develop and broaden their applicability.

Affinitak™, formerly LY900003 or ISIS 3521, which we licensed to Eli Lilly and Company, or Lilly, in 2001, is our most advanced product in development. In March 2003, we announced the results of our Phase III clinical trial of Affinitak to treat patients with non-small cell lung cancer, which were not sufficient to support a single study new drug application. Lilly and we completed an analysis of the data from this trial and presented a summary of the findings at the 39th Annual Meeting of the American Society of Clinical Oncology in June 2003. In a second Phase III study, Lilly is continuing to follow enrolled patients. Lilly and we will make a decision about the future development of Affinitak pending a review upon completion of the second Phase III trial, which most likely will occur in the second half of 2004.

We are conducting two Phase III clinical trials for another product, ISIS 2302, or alicaforsen, in an inflammatory bowel disease known as Crohn's disease. We are conducting these trials in North America and Europe. We expect to report data from these clinical trials in the second half of 2004. We also have Phase II programs ongoing for five additional products.

Our GeneTrove® program uses our antisense technology as a tool to provide important information about the function of genes and has automated the initial steps in our antisense drug

discovery process. Our current focus is to use GeneTrove information to direct our own drug discovery research and that of our antisense drug discovery partners, like Lilly and Amgen.

Within our Ibis program we have invented a platform technology that has the potential to revolutionize the identification of organisms that cause infectious diseases. Through a project called Triangulation Identification for Genetic Evaluation of Risks, or TIGER, we have applied our proprietary technologies to develop a biological sensor, or biosensor, designed to rapidly and simultaneously identify a broad range of infectious organisms in a sample, including organisms that are newly-emerging, genetically altered and unculturable. We have successfully demonstrated proof-of-principle of the TIGER biosensor with the identification of a variety of bacteria and viruses in both environmental and human clinical samples. To date, we have received grants and contracts representing potential funding of up to $55 million from agencies of the U.S. Government, including the Defense Advanced Research Projects Agency, or DARPA, the Centers for Disease Control and Prevention, or CDC, the U.S. Army Medical Research Institute of Infectious Diseases, or USAMRIID, the United States Navy, and the Federal Bureau of Investigation, or FBI, among others.

We have a broad patent portfolio covering our technologies. We have rights to more than 1,300 issued patents, which we believe represents the largest antisense and RNA-oriented patent estate in the pharmaceutical industry. Our intellectual property is a strategic asset that we are exploiting to generate near-term revenue and that we expect will also provide us with revenue in the future. The principal purpose of our intellectual property portfolio is to protect our inventions in RNA-based drug discovery. Our intellectual property estate also enables us to expand our pipeline by granting partners limited access to antisense technology through licenses we grant them. Licensing partnerships may include antisense drug discovery collaborations such as those we have with Lilly and Amgen, as well as GeneTrove functional genomics agreements like those we have with Amgen, Chiron and Pharmacia. In addition, we have licensed our functional genomics patents to companies like Chiron, Amgen, Sequitur and atugen AG. We also license our non-antisense patents, as we did to Eyetech Pharmaceuticals, Inc, or Eyetech. In December 2001, we licensed several chemistry patents to Eyetech for the development of Macugen™, a drug designed to treat age-related macular degeneration, or AMD, that Eyetech is co-developing with Pfizer. In 2003, Pfizer and Eyetech reported encouraging Phase III data for Macugen that Eyetech says will serve as the basis for a new drug application, or NDA, with commercialization of the drug expected in 2005. Assuming successful commercialization of Macugen, we have the opportunity to earn future milestone payments and royalties that could be substantial to us. To date, we have generated more than $35 million in license and royalty fees related to our patent portfolio.

We incorporated in California in 1989, and in 1991 we reincorporated as a Delaware corporation. Our principal offices are in Carlsbad, California. In November 2003, we established Isis Pharmaceuticals Singapore Pte Ltd, a wholly-owned subsidiary in Singapore.

From our progress in antisense we have developed a robust pipeline of promising new drugs and efficient genomics tools that unlock value from gene sequence data. Our Ibis program has the potential to become an important new approach to the identification and treatment of infectious diseases.

Proteins are essential, working molecules in a cell. Almost all human diseases result from inappropriate protein production or improper protein activity. Scientists use traditional drug discovery methods to design drugs to interact with the proteins in the body that are supporting or causing a disease. Antisense technology is different from traditional drug discovery because it specifically targets

disease-causing proteins before the body produces them. We design our antisense drugs, or antisense inhibitors, to act earlier in the disease process than traditional drugs and to interrupt the production of disease causing proteins without disrupting proteins responsible for the body's normal functioning.

Genes contain the information necessary to produce proteins. A gene is made up of bases, or nucleotides: Adenine, Thymine, Guanine, and Cytosine, commonly known as A, T, G and C, which are linked together to form a two-stranded structure that resembles a twisted ladder, known as DNA or deoxyribonucleic acid. The nucleotides on one side of the ladder bind weakly to complementary nucleotides on the other strand according to specific rules; for example, A pairs with T and G pairs with C, creating the ladder's rungs. This highly specific nucleotide pairing is called hybridization. The sequence or order of these nucleotides establishes the cell's recipe for making proteins.

When a cell transcribes information from DNA into messenger RNA, or mRNA, the two complementary strands of the DNA partly uncoil. One strand acts as a template and information stored in the DNA strand is copied into a complementary mRNA. mRNA then carries the information to cellular structures called ribosomes, the cell's factories for manufacturing proteins. The ribosome reads the encoded information, the mRNA's nucleotide sequence, and in so doing, strings together amino acids to form a specific protein. This process is called translation. Antisense technology interrupts the cell's protein production process by preventing the RNA instructions from reaching the ribosome, thus inhibiting the production of the protein. The mRNA sequence of nucleotides that carries the information for protein production is called the "sense" strand. The complementary nucleotide chain that binds specifically to the sense strand is called the "antisense" strand.

We use the information contained in mRNA to design chemical structures, called antisense oligonucleotides, which resemble DNA and RNA and are the complement of mRNA. These potent antisense oligonucleotides, or antisense drugs, inhibit the production of disease-causing proteins. Antisense drugs can selectively inhibit one protein among a closely related group of proteins because antisense drugs interact selectively with the specific RNA and not with RNA of the closely related members of the group. It is easier to differentiate between closely related proteins at the RNA sequence level than by binding to the protein itself, as traditional drugs do. As a result, we can design antisense drugs that selectively inhibit the disease-causing member of the group without interfering with those members of the group necessary for normal bodily functions. This unique specificity means that antisense drugs may be far less toxic than traditional drugs, because we can design them to minimize the impact on unintended targets.

Further, the design of antisense compounds is less complex, more rapid and more efficient than traditional drug design directed at protein targets. Traditional drug design requires companies to identify a small molecule that will interact with protein structures to affect the disease-causing process. Since predicting which small molecules will do this has proven to be difficult, traditional drug discovery involves testing hundreds of thousands of small molecules for their ability to interfere with protein function. As a result, traditional drug discovery is a labor intensive, low probability endeavor. In contrast, we design our antisense compounds to bind to mRNA structures through well understood processes. We can design prototype antisense drugs as soon as we identify the sequence for the mRNA receptor.

We are the leader in the discovery and development of this exciting new class of therapeutic compounds. Our proprietary technology to discover and characterize novel antisense inhibitors has enabled our scientists to modify the properties of our antisense drug candidates for optimal use with particular targets and thus to produce a broad proprietary portfolio of compounds applicable to many disease targets. Further, over the past decade, our scientists have made great advancements in chemistries, which we call our second-generation antisense drugs. Second-generation drugs may have increased potency, stability, oral bioavailability and an improved side effect profile. We have also made significant progress in developing new formulations of antisense drugs, like oral, topical cream,

subcutaneous, intravitreal, aerosol and enema, of antisense drugs that further expand the potential for antisense technology.

In our functional genomics program, GeneTrove, we use antisense technology to generate information about the function of genes and to determine the value of genes as drug discovery targets. This information forms the basis of the first step of our antisense drug discovery program. We use this information to direct our own antisense drug discovery research, and that of our antisense drug discovery partners. We have created inhibitors to thousands of genes, validated many targets and dissected numerous disease pathways. Additionally, we have created libraries of antisense inhibitors to identify novel gene function. Our GeneTrove program has enhanced our own antisense drug discovery efforts and our patent portfolio through custom target validation collaborations and intellectual property licenses while generating near-term revenue for us.

Through our Ibis program, we have expanded on our RNA-based drug discovery and development expertise to create a platform technology designed to rapidly and accurately identify a broad range of organisms in a single test, and to develop small molecule antibacterial and antiviral drugs that bind to RNA. To accomplish these tasks, our scientists integrate functional genomics, bioinformatics and RNA-focused chemistry programs with novel high-throughput, mass spectrometry-based screening methods.

We are collaborating with San Diego-based Science Applications International Corporation, or SAIC, on a multi-year project funded by DARPA for the ongoing development of our TIGER technology and biosensor. This project combines our expertise in microbial genome sequence analysis and advanced mass spectrometry technology with SAIC's advanced signal processing capabilities. As of December 31, 2003, we had been awarded funding of $13.3 million related to this collaboration with SAIC, of which $13.3 million had been billed and $11.9 million had been collected. During 2003, we successfully demonstrated proof-of-principle of the TIGER biosensor with the identification of a variety of bacteria and viruses in both environmental and human clinical samples. In March 2004, we were awarded a two-year contract under this collaboration for additional funding of up to $19.5 million to further the ongoing development of TIGER, including the design and manufacture of a fully integrated, self-contained system, called TIGER 2.0. We are designing this instrument to perform the analysis of samples in an automated manner, and expect to initially deploy it for use by our government partners, including the CDC.

In September 2003, we received a three-year grant for up to $6.0 million from the CDC to develop and apply our TIGER technology to the surveillance of human infectious disease in the U.S. Using the grant from the CDC, we expect to develop and provide TIGER technology for CDC projects focused on emerging human infectious diseases. We believe our work to automate the TIGER biosensor and deploy the system for specific applications for the CDC and DARPA advances our TIGER technology toward commercialization. We retain full commercial rights to the TIGER technology while our government partners have access to the technology to meet their specific needs.

Since Ibis' inception, we have received significant financial support from government-funded grants and contracts to use Ibis' technology to assist in national defense. In March 2002, we transitioned our government-sponsored research program to discover novel broad-spectrum antibacterial drugs for biological warfare defense to USAMRIID. USAMRIID awarded us a three-year, $2.4 million contract to advance our work in this area.

In addition to DARPA, USAMRIID and the CDC, we also have research relationships with other government entities including the United States Navy and the FBI. To date, we have received government contracts and grants representing potential funding of up to $55 million. During 2003,

revenue generated from agencies of the U.S. Government comprised 20% of our total revenue, including approximately 16% of our total revenue from our collaboration with SAIC.

We have successfully developed the first antisense drug to reach the market, Vitravene, for CMV retinitis, which is available through our partner, Novartis Opthalmics AG.

We have designed our drugs in development to treat a variety of health conditions, including inflammatory, viral, metabolic, cardiovascular and dermatological diseases, and cancer, and we are studying them in intravenous, subcutaneous, topical cream, enema and oral formulations. Intravenous and subcutaneous formulations are commonly grouped together and referred to as parenteral forms of administration. The following table lists our approved product and each of our products under development, its target, disease indication and development status, as well as our commercial rights.

Product(1)	Target	Potential Disease Indication(s)	Development Status(2)	Commercial Rights
Vitravene (I)	Antiviral	CMV Retinitis	Available in the U.S..	Isis/Novartis Ophthalmics AG(3)
Affinitak™ (formerly LY900003 or ISIS 3521) (P)	PKC-alpha	Cancer—Non-Small Cell Lung Cancer	Phase III	Lilly
Alicaforsen (ISIS 2302) (P)	ICAM-1	Crohn's Disease	Phase III	Isis
Alicaforsen (ISIS 2302) (E)	ICAM-1	Ulcerative Colitis, Pouchitis	Phase II	Isis
ISIS 14803 (P)	Hepatitis C virus	Hepatitis C	Phase II	Isis
ISIS 104838 (P,O)*	TNF-alpha	Rheumatoid Arthritis	Phase II	Isis
ISIS 104838 (P, T)*	TNF-alpha	Psoriasis	Phase II	Isis
OGX-011 (ISIS 112989) (P)*	Clusterin	Cancer—Solid Tumors	Phase I/II	Isis/OncoGenex
ISIS 113715 (P)*	PTP-1B	Diabetes	Phase I	Isis
ATL 1102 (ISIS 107248)*	VLA-4	Multiple Sclerosis, Inflammatory Diseases	Phase I	Antisense Therapeutics Limited
ISIS 301012 (P)*	apoB-100	Cardiovascular	Phase I	Isis
LY2181308 (P) *	Survivin	Cancer	Preclinical	Lilly

The following section provides more detailed descriptions of our approved product and those products in clinical development and the disease indications they target. We also have a significant research program which we expect to yield additional development candidates in the future.

Individuals with suppressed immune systems, such as those with AIDS resulting from the HIV virus, are susceptible to opportunistic infections caused by CMV. Among patients with AIDS, CMV

retinitis is the primary cause of blindness. Currently approved drugs for CMV retinitis are ganciclovir, foscarnet, cidofovir and fomivirsen.

Vitravene, or fomivirsen—In August 1998, the FDA approved Vitravene to treat CMV retinitis in AIDS patients. Vitravene is an antisense compound that we discovered and developed. Novartis Ophthalmics AG, the eye health unit of life sciences leader, Novartis AG, and our worldwide distribution partner for this drug, launched Vitravene in November 1998. New anti-HIV drugs, particularly protease inhibitors and combination treatment regimens, have prolonged survival in HIV-infected individuals. Over the last several years, this has resulted in a decline in mortality from AIDS, accompanied by a decline in the incidence of many opportunistic infections, including CMV retinitis. As a result, Novartis AG currently offers Vitravene on a limited basis in the U.S.

We, together with our partners, are pursuing the development of antisense drugs for the treatment of a variety of cancers. In our clinical trials, we have observed evidence of activity of our drugs. In addition, patients tolerated our compounds well, with none of the serious side effects, such as bone marrow or immune system suppression, gastrointestinal distress or hair loss, associated with standard cancer chemotherapies.

Affinitak—Affinitak, which we licensed to Lilly in August 2001, is our antisense compound in Phase III clinical development for non-small cell lung cancer. Affinitak inhibits the production of one particular isotype, the alpha isotype, of protein kinase C, or PKC. PKC alpha is a member of a family of proteins that are associated with both normal and abnormal cell growth. PKC alpha has been shown to be involved in cancer cell growth and maintenance. In preclinical studies, we have been able to specifically inhibit the production of the PKC-alpha isotype without inhibiting the production of other isotypes, thus allowing the inhibition of an isotype believed to be involved in abnormal cell growth without more broadly affecting all the different PKC isotypes.

In 2003, we announced the results of our Phase III clinical trial of Affinitak in combination with traditional cancer chemotherapy drugs to treat patients with non-small cell lung cancer. In this 616-patient trial, we observed no difference in a primary log-rank analysis of overall survival, the primary endpoint of the trial, of those patients who received Affinitak plus the chemotherapy regimen of carboplatin and paclitaxel compared to those patients who received the chemotherapy alone. Further, we performed additional analysis of the data using other standard statistical methods or tests. We observed that survival of the Affinitak treated patients was greater than that of the patients who received the chemotherapy alone. Another potentially important observation from the trial was that those patients who completed the prescribed course of therapy, six cycles, experienced a survival benefit compared to those patients who did not complete therapy. This result suggests that the duration of treatment with Affinitak may contribute to improved survival. Additionally, in those patients who completed the prescribed course of therapy, results favored the Affinitak group across multiple secondary endpoints, including time to disease progression, time to treatment failure and duration of remission.

The addition of Affinitak to carboplatin and paclitaxel was well tolerated. There were no increases in severe toxicities or toxicity related deaths in patients receiving Affinitak, compared to those receiving the chemotherapy alone. The most common side effects among patients in the study were fatigue and nausea. Patients in the study receiving Affinitak in combination with the chemotherapy had a higher rate of moderate thrombocytopenia, nausea and vomiting. Further, because clinicians administer Affinitak via continuous intravenous infusion, Affinitak treated patients had a higher incidence of catheter-related infections.

Lilly and we completed an analysis of the data from this trial and presented a summary of the findings at the 39th Annual Meeting of the American Society of Clinical Oncology in June 2003. In a

second Phase III study, Lilly is continuing to follow enrolled patients. Lilly and we will make a decision about the future development of Affinitak pending a review upon completion of the second Phase III trial, which most likely will occur in the second half of 2004.

OGX-011—OGX-011, also called ISIS 112989, is a second-generation antisense inhibitor of clusterin, which we are co-developing and commercializing with OncoGenex Technologies Inc., or OncoGenex, a Canadian oncology-focused research and development company. We have designed OGX-011 to inhibit the secretory protein clusterin, which acts as a cell-survival protein that is over-expressed in response to tumor killing strategies, like chemotherapy, hormone ablation and radiation therapy. Based on analysis of human tumor tissue, clusterin is over-expressed in several cancers, including prostate, breast, renal, bladder, non-small cell lung and ovarian. By inhibiting clusterin, clinicians intend to enhance the effects of drug therapies in the treatment of these cancers.

In preclinical animal studies, scientists from both OncoGenex and Isis, in collaboration with the Prostrate Center at Vancouver General Hospital, demonstrated OGX-011 improved the potency of traditional chemotherapies more than ten-fold in prostate cancer, without compromising safety. These studies also show that OGX-011, when combined with other cancer treatments in preclinical model systems, may significantly improve tumor reduction and delay disease progression in prostate, lung, bladder and renal cancer. These findings support the continued development of OGX-011 in combination with standard chemotherapy and other agents.

OncoGenex and we initiated a Phase I/II program of OGX-011 in patients with prostate cancer in December 2002. This Phase I trial is evaluating OGX-011 in combination with hormone ablation therapy prior to surgical removal of the prostate. In April 2003, Oncogenex and we initiated a second Phase I/II trial to evaluate OGX-011 in combination with TAXOTERE® in various solid tumors. We expect to report data from both clinical trials in the first half of 2004. OGX-011 is our first second-generation antisense anti-cancer drug in clinical trials.

LY2181308—We licensed our preclinical anti-cancer candidate, LY2181308, formerly ISIS 23722, to Lilly in 2002, as part of the expansion of our Lilly research collaboration into cancer. The compound targets survivin, which plays a role in cancer cell death, or apoptosis. Survivin is one of the most abundantly expressed proteins in cancers. Our researchers and collaborators have shown that inhibiting expression of survivin by LY2181308 inhibits the growth of cancer cells. Since normal cells in the body do not express survivin, we expect that this drug will have fewer side effects than traditional chemotherapy.

In April 2003, we earned a $1.5 million milestone from Lilly in the development of LY2181308 as part of our research collaboration oncology expansion with them. LY2181308 is the first compound from this partnership that Lilly has selected for clinical development. We expect Lilly to initiate Phase I trials of LY2181308 during 2004.

ISIS 2503—Substantial evidence exists supporting a direct role for ras gene products in the development and maintenance of human cancer. Ras proteins are involved in passing information between cells. Ras, in both normal and mutated forms, is associated with abnormal cell growth and, as such, is associated with cancer. Cell culture and animal models have shown that ISIS 2503, a potent selective inhibitor of Harvey ras, or H ras, inhibits abnormal cell growth.

During 2002, we completed Phase I studies that demonstrated that ISIS 2503 was well tolerated and reported no serious side effects. We also observed evidence of activity. These results supported continued development of ISIS 2503 in Phase II trials for the treatment of pancreatic and breast cancer; however, during 2003 we decided to discontinue development of ISIS 2503 in order to focus our development resources on other drugs in our pipeline. We will continue to pursue partnership opportunities for ISIS 2503.

Our research and development efforts in the therapeutic area of inflammatory diseases focus on identifying and developing antisense inhibitors to proteins such as intercellular adhesion molecule 1, or ICAM-1, another adhesion molecule called CD49d, which is a subunit of VLA-4, and tumor necrosis factor-alpha, or TNF-alpha. Researchers believe that these proteins are involved in inflammatory diseases.

Alicaforsen (ISIS 2302)—The most advanced compound in our cell adhesion program selectively inhibits ICAM-1 gene expression. ICAM-1 is a member of the adhesion molecule family. Over-expression of ICAM-1 occurs in a wide variety of inflammatory disorders, such as rheumatoid arthritis, asthma, psoriasis and inflammatory bowel diseases. Experts believe that ICAM-1 contributes to the pathology of these diseases and conditions. We are currently evaluating alicaforsen in two Phase III studies for the treatment of Crohn's disease. Additionally, we are conducting Phase II studies of alicaforsen enema formulation for the treatment of ulcerative colitis. According to the Crohn's and Colitis Foundation of America, up to one million people have inflammatory bowel disease, with occurrences evenly split between Crohn's disease and ulcerative colitis, or UC. According to the European Federation of Crohn's and UC Associations, inflammatory bowel disease affects a similar number of people in Europe.

ISIS 104838—ISIS 104838 is a second-generation antisense inhibitor of TNF-alpha and the first product from our proprietary second-generation chemistry to enter the clinic. TNF-alpha, or tumor necrosis factor alpha, is a naturally occurring cytokine that is implicated in the development and progression of many inflammatory, infectious and autoimmune diseases, including rheumatoid arthritis and psoriasis. TNF-alpha is involved in bone and cartilage absorption, facilitates inflammation and inhibits bone formation. Most patients with rheumatoid arthritis have high levels of TNF-alpha.

ISIS 107248—ISIS 107248, or ATL 1102, is a second-generation antisense inhibitor of CD49d, which is a subunit of VLA-4. Studies in animal models have demonstrated that inhibition of VLA-4 has a positive effect on a number of inflammatory diseases like multiple sclerosis. In December 2001, we licensed ISIS 107248 to Antisense Therapeutics Limited, or ATL. Under our agreement with ATL, we completed the required preclinical studies for ISIS 107248 and manufactured the bulk drug for human clinical trials at ATL's expense. ATL is responsible for the future clinical development and the commercialization of the drug. In August 2003, ATL initiated a Phase I clinical trial of ISIS 107248 for the treatment of multiple sclerosis. We expect that ATL will report final data from this trial mid-year 2004, and plan to initiate a Phase II trial in patients with multiple sclerosis by the end of 2004.

HCV represents a major public health challenge. This potentially deadly disease affects the liver and can eventually cause liver cirrhosis and death. HCV affects an estimated four million people in the United States, and will kill 10,000 to 12,000 people in the United States each year. Physicians attempt to eradicate this virus from chronically infected individuals by using Interferon-alpha therapy, either alone or in combination with the drug ribavirin. According to data from the National Institute of Health, patients with genotypes 2 and 3 are two to three times more likely to respond to interferon based therapy than patients with genotype 1. Genotype 1 is the most common genotype in the United States. Better, safer and more effective treatments for HCV are urgently needed, as current therapies have limited efficacy and potentially serious side effects.

ISIS 14803—Our antisense inhibitor of HCV, ISIS 14803, may represent a significant therapeutic advance in treating this serious viral epidemic. We designed ISIS 14803 to inhibit the replication of HCV.

Two clinical trials of ISIS 14803 have demonstrated activity in patients with HCV, and a third clinical trial is in process. Isis presented data from two single agent studies that showed ISIS 14803 is active in drug resistant, genotype 1 HCV patients, the most difficult-to-treat segment of the HCV patient population. In a four-week Phase I/II clinical trial initiated in August 2003, which we designed to evaluate both the safety and efficacy of ISIS 14803 in patients with HCV, ISIS 14803 demonstrated dose-dependent antiviral activity, decreasing the level of virus in the blood in patients with drug resistant chronic HCV. All patients in the clinical study had the most common and drug resistant form of HCV, genotype 1, and all but one patient had failed previous interferon-based therapy. Flu-like symptoms, headache and fatigue were the most common side effects observed in the trials. In October 2003, we reported the final results of a 43-patient single agent Phase II clinical trial that demonstrated promising antiviral activity by producing up to 3.8 log dose-dependent reductions in plasma virus levels in five of 17 patients with HCV.

In May 2003, we initiated a 30-patient Phase II clinical trial to assess the benefits of adding ISIS 14803 to standard treatments for HCV. We designed the trial to explore the activity of ISIS 14803 in combination with standard HCV treatments based on cumulative experience with the drug as a single agent. We plan to announce initial results from this study in the second half of 2004.

We are pursuing the discovery and development of antisense drugs for metabolic diseases such as diabetes and obesity. These chronic diseases affect millions of people and represent significant areas of unmet medical need. We believe that our second-generation antisense drugs will have properties that will make them attractive therapies.

ISIS 113715—ISIS 113715 is our second-generation antisense inhibitor of the PTP-1B gene for type 2 diabetes. According to the American Diabetes Association, diabetes affects more than 18 million people and type 2 diabetes constitutes 90 percent of those cases. An antisense inhibitor of PTP-1B represents a new approach to the treatment of diabetes. For years, pharmaceutical companies interested in diabetes research have actively pursued phosphatases, such as PTP-1B, as part of traditional small molecule drug discovery efforts. However, due to structural similarities among closely related enzymes, it is often difficult to identify small molecule drugs with the degree of specificity that the antisense approach can obtain.

The preclinical studies of ISIS 113715 demonstrated compelling activity in multiple diabetic animal models and suggested activity as an insulin sensitizer without causing hypoglycemia and while reducing cholesterol and weight gain.

In May 2003, we initiated a Phase I clinical trial to assess the safety and pharmacokinetic profile of several doses of ISIS 113715 by parenteral administration in 20 healthy volunteers. In September 2003, we reported the results from this Phase I study, which demonstrated improved glucose tolerance and increased insulin sensitivity in all patients who received ISIS 113715, and that ISIS 113715 was well tolerated.

Based on the results of the Phase I study, we are initiating a Phase II clinical trial to further evaluate the drug's ability to regulate blood sugar levels in patients with type 2 diabetes. We plan to initiate enrollment by mid-2004 and report data from this clinical trial in late 2004 or early 2005.

ISIS 13650—ISIS 13650 is an inhibitor of c-raf kinase for the treatment of diabetic retinopathy and age-related macular degeneration. Diabetic retinopathy is an ocular complication of diabetes. The incidence of these conditions continues to grow with the advancing age of the U.S. population. In preclinical studies, antisense inhibition of c-raf kinase is associated with a reduction of neovascularization, or growth of blood vessels, which can obstruct vision. During 2002, ISIS 13650, a second-generation antisense product, was in preclinical development. During 2003, we decided not to

invest any further in the development of ISIS 13650. This action is part of our broader decision to pursue ophthalmic research through partnerships rather than using our own resources.

Cardiovascular disease is the leading cause of death in the United States, according to the National Institutes of Health. Researchers have shown a strong correlation between high cholesterol levels and subsequent cardiovascular diseases. Statistics from the American Heart Association show more than 100 million American adults have high cholesterol levels.

ISIS 301012—ISIS 301012 is the newest compound that we have added to our pipeline. It targets apoB-100, a molecule that has been of great interest to the industry, yet has long been considered "undruggable" by traditional small molecule approaches. ApoB-100 is a protein that plays a pivotal role in the production of low-density lipoprotein (LDL), the "bad" cholesterol. In preclinical studies, ISIS 301012 reduced total cholesterol, very low-density lipoproteins (VLDL), low-density lipoproteins (LDL), and triglyceride levels, all of which are keys to managing heart disease. We initiated a double-blind, placebo-controlled, dose escalation Phase I study in late 2003. The goal of this trial is to assess the safety, tolerability and pharmacokinetic profile of ISIS 301012, and its ability to reduce several components of cholesterol that are important in the management and prevention of cardiovascular disease. We plan to report data from this study in the second half of 2004.

We combine our core technology programs in medicinal chemistry, RNA biochemistry, and molecular and cellular biology with molecular target-focused drug discovery efforts to design drug candidates. The goal of our target-based research programs is to identify antisense and small molecule drug candidates to treat diseases for which there are substantial markets and for which there is a need for better drugs. In addition, our research programs focus on identifying next-generation compounds to serve as backup compounds to our current products in development and to our development candidates. Our Ibis program focuses on developing mass spectrometry-based technology for identifying and treating infectious diseases.

Our core technology programs can support multiple target-based antisense research programs without significantly increasing costs. Through these programs, we can efficiently explore numerous disease targets and identify lead compounds to advance into preclinical development. We are currently pursuing antisense and small molecule drug discovery programs focused on various anti-viral and anti-bacterial targets, inflammatory disease targets, cardiovascular disease targets and other key molecular targets that might play critical roles in cancer and metabolic diseases like diabetes and obesity.

We are pursuing three early-stage antisense mechanisms, including RNA interference, or RNAi, micro-RNA, and alternative splicing, through research collaborations and partnerships like those we have with Ercole Biotech, Inc., or Ercole, and Alnylam Pharmaceuticals, Inc., or Alnylam.

RNAi is an antisense mechanism that involves using a small interfering RNA, or siRNA, an antisense inhibitor, as a method to target an mRNA sequence. With siRNA, the cell recruits a protein group called RISC to prevent the production of a disease-causing protein. We have a strong intellectual property position in RNAi methodology and are pursuing opportunities to license these patents to companies specializing in RNA interference as a therapeutic method.

Micro-RNAs are an emerging class of drug targets and a new area for drug discovery. Micro-RNAs are small RNA molecules that appear to have critical functions in controlling the process of gene expression. Micro-RNAs can serve as drug targets or as drugs themselves. Researchers estimate that there are approximately 250-300 known micro-RNA molecules in humans.

Modulation of alternative splicing seeks to control the process by which a single gene can lead to several proteins. To be converted into proteins, genes must be initially copied into a pre-mRNA. This pre-mRNA then undergoes splicing, a process where stretches of RNA are deleted and the remaining RNA strand is then linked back together. Splicing is necessary for a messenger RNA, or mRNA, to be read by the cell to produce a particular protein.

Our strategy is to use alliances with other companies and equity-based financing to increase our financial resources, reduce risk, and retain an appropriate level of ownership of products currently in development. Through alliances with major pharmaceutical companies and biotechnology companies, we can obtain funding, expand existing programs, learn of new technologies, and gain additional expertise in developing and marketing products.

We are focused on establishing new partnerships and on advancing and building upon existing relationships. We currently have agreements with more than a dozen partners. These span all four areas of our business: antisense drug discovery and development, GeneTrove, Ibis and our intellectual property estate. The following is a list of our business development highlights for 2003 and early 2004.

In August 2001, we entered into a broad strategic relationship with Lilly that has four key components:

To date, Lilly has paid or committed to pay us more than $200.0 million in funding through 2005, including the $100.0 million loan, the $75.0 million equity investment, the $25.0 million in Affinitak up-front license fees, and amounts for the remaining Affinitak development costs. Assuming the success of multiple products from our collaboration with Lilly, we have the opportunity to earn additional future milestones and royalties from Lilly that could be substantial to us.

In September 2002, we further expanded our relationship with Lilly by agreeing to manufacture Affinitak during the product launch period for Lilly. Through this agreement we upgraded and expanded our manufacturing facility, including the addition of a new state-of-the-art manufacturing suite. Lilly provided us with funding in the form of a $21.2 million loan to build the new suite. In June 2003, we reached a mutually beneficial renegotiation of our manufacturing relationship with Lilly. Lilly waived repayment of the $21.2 million manufacturing loan it provided us to build the new manufacturing facility. Lilly also agreed to allow us to use the facility to manufacture other drugs. In exchange, we released Lilly from its obligations to purchase additional product from us and its obligation to pay for the costs of maintaining an idle manufacturing suite.

In April 2003, we earned a $1.5 million milestone from Lilly in the development of LY2181308, the antisense inhibitor of survivin, as part of the research collaboration oncology expansion. LY2181308 is the first compound from the partnership to be selected for clinical development by Lilly. We expect Lilly to initiate Phase I trials of LY218308 during 2004.

In July 2003, we expanded the cancer research component of our antisense drug discovery collaboration with Lilly to include multiple antisense mechanisms, such as RNAi and alternative splicing, as well as alternative chemistries, such as Peptide Nucleic Acid, or PNA. We are currently jointly developing antisense drugs with Lilly that work through an RNAi mechanism or use our PNA chemistry as potential follow-on drugs to LY2181303.

Our relationship with Lilly has historically provided several revenue sources, including research funding related to their $100.0 million research loan to us, development milestones, and revenue related to Affinitak. During 2003, 2002 and 2001, we generated revenue from our relationship with Lilly totaling $30.9 million, $45.4 million and $14.5 million, respectively, which comprised 62%, 57%, and 27%, respectively, of our total revenue during those same periods.

In December 2001, we entered into a three-year collaboration with Amgen to discover new antisense drugs. Amgen has the right to develop and commercialize antisense drugs resulting from the collaboration. If drugs from the collaboration are successful, we will receive milestone payments upon key clinical and commercial achievements, as well as royalties on sales of any products resulting from the collaboration. In August 2002 and February 2003, we earned progress-related research milestones under this drug discovery collaboration.

In May 2003, we and Ercole Biotech, Inc. initiated a multi-year collaboration to discover antisense drugs that regulate alternative RNA splicing. As part of the collaboration, we cross-licensed our respective splicing-related intellectual property. We are combining our alternative splicing expertise with Ercole to discover antisense drugs that regulate alternative RNA splicing. As part of this collaboration, we granted Ercole a license to our Bcl-x molecule and certain of our chemistry patents. In addition, we took an equity ownership position in Ercole, with the initial funding in the form of convertible debt, which the companies anticipate will convert into securities that Ercole issues in its next venture capital financing. We also have the option to make an additional equity investment in Ercole.

In June 2003, we initiated a collaboration with ITRI to identify antisense candidates targeting the coronavirus associated with SARS. We conducted the research in exchange for an upfront payment of $1.0 million, milestone payments, and the potential for future funding. In December 2003, we achieved two milestones in our antisense drug discovery partnership with ITRI, for which we received a total of $1.0 million. The milestones related to the identification of second-generation antisense drugs that inhibit SARS virus replication and the successful completion of preclinical studies evaluating aerosol and parenteral delivery of antisense drugs as specified under the agreement.

In November 2003, we received a grant of up to $8.0 million over three years from the Singapore Economic Development Board which will fund, in part, the broadening of two of our RNA-based drug discovery and development programs: micro-RNA drug discovery and antisense drug discovery targeting the coronavirus associated with SARS. In connection with this grant, we established Isis Pharmaceuticals Singapore Pte Ltd, a wholly-owned subsidiary of Isis Pharmaceuticals, Inc.

In March 2004, we entered into a strategic alliance with Alnylam to develop and commercialize RNAi therapeutics. Under the terms of the agreement, we exclusively licensed to Alnylam our patent estate relating to antisense motifs and mechanisms and oligonucleotide chemistry for double-stranded RNAi therapeutics in exchange for a $5 million technology access fee, participation in fees for Alnylam's partnering programs, as well as future milestone and royalty payments. We will retain rights to a limited number of RNAi therapeutic targets and all rights to single-stranded RNAi therapeutics. In addition, Alnylam and we will share the proceeds of any licenses Alnylam grants under its previously announced InterfeRx™ program that include sublicenses to our patents. We also made a $10 million equity investment in Alnylam and have agreed to provide Alnylam with access to our resources for development and commercialization of RNAi therapeutics, including process development, bioanalytic methods, quality control, and manufacturing.

In turn, Alnylam nonexclusively licensed us its patent estate relating to antisense motifs and mechanisms and oligonucleotide chemistry to research, develop and commercialize single-stranded RNAi therapeutics and to research double-stranded RNAi compounds. We also received a license to develop and commercialize double-stranded RNAi drugs targeting a limited number of therapeutic targets on either an exclusive or co-exclusive basis depending on the target. If we develop or commercialize an RNAi-based drug using Alnylam's technology, we will pay Alnylam milestones and royalties. As part of the collaboration, each party granted the other party a nonexclusive cross license to its respective patent estate relating to antisense motifs and mechanisms and oligonucleotide chemistry for microRNA therapeutics.

We are focused primarily on discovering and commercializing single-stranded antisense drugs that work through a variety of mechanisms, including RNAi and RNase H. This alliance strengthens Alnylam's position in double-stranded RNAi and gives Isis the ability to share in Alnylam's success in this area.

An important aspect of our business model is to selectively extend our expertise and intellectual property position in antisense technology to industry partners that are interested in developing antisense therapeutics. In return for providing companies with access to our technology, we receive an ownership interest in the resulting products and/or in the companies. This provides us with the opportunity to create a much broader antisense pipeline than we could afford to develop on our own while minimizing our financial obligations. We have implemented this integral component of our strategy through our partnerships with major pharmaceutical companies and with ATL, OncoGenex and Santaris Pharma A/S, or Santaris, formerly Pantheco A/S, or Pantheco. Our partnerships with OncoGenex, ATL and Santaris represent our ability to broaden the reach of antisense technology in emerging companies globally. We believe we will have more of these opportunities that, when combined with our own antisense drug pipeline, will allow us to participate in the establishment of a new sector of the pharmaceutical industry based on antisense technology.

ATL 1102, or ISIS 107248, has been demonstrated to have positive effects in animal models for the treatment of certain inflammatory diseases such as multiple sclerosis. In December 2001, we licensed ISIS 107248 to ATL, an Australian company publicly traded on the Australian Stock Exchange. We were responsible for completing the required preclinical studies for ISIS 107248 and for manufacturing the drug for human clinical trials at ATL's expense. ATL agreed to undertake the future clinical development and commercialization of the drug. In August 2003, ATL initiated a Phase I clinical trial of ISIS 107248 for the treatment of multiple sclerosis, and will report final results in mid-2004. ATL plans to initiate a Phase II study in the second half of 2004. In addition, we are participating with ATL in a five-year antisense drug discovery and development collaboration. ATL will pay us for access to our antisense expertise and for research and manufacturing services we provide them during the collaboration. Additionally, ATL is obligated to pay us royalties on any antisense drugs discovered and developed within the partnership. We currently own approximately 11% of ATL's equity and hold options for additional shares. If all of ATL's outstanding options, including ours, were exercised, our ownership in ATL would be approximately 14%.

In November 2001, we established a drug development collaboration with OncoGenex Technologies Inc., a Canadian oncology-focused research and development company, to co-develop and commercialize OGX-011, or ISIS 112989, an anti-cancer antisense drug candidate. OGX-011 combines OncoGenex's proprietary antisense position in inhibitors to the target clusterin, with our proprietary

second-generation antisense chemistry. We conducted preclinical toxicology and pharmacokinetic studies of OGX-011 during 2002. We also manufactured OGX-011 for preclinical and Phase I/II studies. OncoGenex has responsibility to perform Phase I clinical trials to assess the safety of OGX-011 in combination with hormone ablation therapy in men with localized prostate cancer and to perform Phase I/II clinical trials in combination with standard chemotherapy in patients with solid tumors known to express clusterin, including non-small cell lung, prostate, breast, renal, and ovarian cancers. In December 2002, OncoGenex and we announced the initiation of a Phase I clinical trial of OGX-011 in patients with prostate cancer. In April 2003, OncoGenex and we initiated a second phase I clinical study of OGX-011 in combination with TAXOTERE® in patients with solid tumors. We expect to report data from both clinical trials in the first half of 2004. OGX-011 is our first second-generation antisense anti-cancer drug in human clinical trials.

In September 2003, OncoGenex and we expanded our antisense drug development partnership to include the development of the second-generation antisense anti-cancer drug candidate, OGX-225. OncoGenex has responsibility for the preclinical and clinical development of the drug. OncoGenex issued us OncoGenex securities as payment for an upfront fee. In addition, OncoGenex agreed to provide to us milestone payments for key clinical and regulatory achievements and royalties on product sales. As of December 31, 2003, our ownership interest in OncoGenex was less than 10%.

In November 1998 and September 2000, we entered into license agreements with Santaris, formerly Pantheco. We amended the agreements in May 2003. Under the terms of the amended and restated license agreement, we licensed our novel antisense chemistry, Peptide Nucleic Acid, or PNA, to Santaris on a limited exclusive basis to develop products. The license restricts Santaris to a limited number of molecular targets that are subject to our approval. Santaris has agreed to pay us royalties on any products developed under the license.

As part of our original license agreements with Pantheco, we received shares of Pantheco common stock. In May 2003, Pantheco and Cureon A/S merged to form Santaris. Prior to the merger, we purchased additional shares of Pantheco for $55,000 as a result of antidilution provisions related to Pantheco's stock. After the merger and as of December 31, 2003, our ownership interest in Santaris was less than 10%.

In 1997, we entered into an agreement with Novartis Ophthalmics AG, formerly CIBA Vision Corporation, granting them exclusive worldwide distribution rights for Vitravene, an antisense compound that we discovered and developed. The terms of the agreement provided for us to receive $20.0 million in pre-commercial fees and milestones. As of December 31, 2001, we had received the full $20.0 million of these pre-commercial fees and milestones. In August 1998, the FDA approved Vitravene to treat CMV retinitis in AIDS patients. Novartis Opthalmics AG launched Vitravene in November 1998. Due to the low incidence of CMV retinitis among patients with AIDS, Novartis AG currently offers Vitravene on a limited basis in the U.S.

Our GeneTrove program uses our antisense technology as a tool to provide important information about the function of genes and has automated the initial steps in our antisense drug discovery process. Our current focus is to use GeneTrove information to direct our own drug discovery research and that of our antisense drug discovery partners, like Lilly and Amgen. We also offer antisense-based gene function information and license our antisense based functional genomics patents to pharmaceutical

company partners that are evaluating the genes as targets for their own drug discovery programs. During 2003, we entered into a target validation agreement with Pfizer, Inc., in which Pfizer obtained access to our antisense inhibitors and acquired a license to specific patents within our intellectual property estate for use in its internal antisense-based functional genomics program.

We have entered into numerous contracts and grants with various government agencies to complete research and development work for defense against biological warfare attacks and threat scenarios. To date, our Ibis program has been awarded government contracts and grants representing potential funding of up to $55.0 million. The contracts and grants include a multi-year contract with DARPA, a three-year contract with USAMRIID, a three-year grant from the CDC and a one-year grant from the FBI.

In March 1999, we established a long-term research-scale antisense inhibitor supply agreement with Integrated DNA Technologies, Inc., or IDT. IDT is a leading supplier of antisense inhibitors used in research. Additionally, we further solidified our intellectual property leadership position in antisense technology by broadening our license to certain antisense patents from IDT. In this long-term supply agreement, IDT agreed to manufacture research-scale antisense inhibitors and research reagents to our specifications. We paid IDT $5.0 million toward our future purchase of antisense inhibitors. As of December 31, 2003, the balance of our prepayment was approximately $4.3 million. In December 2001, we expanded our existing licensing agreement with IDT on certain patents that are useful in functional genomics and in making certain antisense drugs. The expanded license allows us to exclusively sublicense this intellectual property for functional genomics purposes. We have paid IDT $4.2 million through December 31, 2003 and expect to pay IDT an aggregate of $700,000 over the next two years for the license.

In May 2001, we entered into an agreement with Hybridon under which we acquired an exclusive license to all of Hybridon's antisense chemistry and delivery patents and technology. Hybridon retained the right to practice its licensed antisense patent technologies and to sublicense it to collaborators under certain circumstances. In addition, Hybridon received a non-exclusive license to our suite of RNase H patents. In exchange for the license to Hybridon's antisense patents, we paid $15.0 million in cash and agreed to pay Hybridon $19.5 million in our common stock before May 2003. In return for access to our patents, Hybridon agreed to pay us $6.0 million in Hybridon common stock before May 2004. In September 2001 and October 2001, we issued to Hybridon 357,143 shares of our common stock valued at $5.0 million and 500,000 shares of our common stock valued at $10.0 million, respectively. In May 2002, Hybridon issued to us 1,005,499 shares of its common stock valued at $1.3 million and paid us $700,000 in cash. In August 2002, Hybridon and we cancelled the remaining reciprocal financial obligations related to this agreement. The cancellation of the obligations resulted in a decrease to our carrying value for the license in the amount of $500,000.

In March 2001, we amended a non-exclusive Patent License Agreement, which we entered into with Molecular Biosystems, Inc. in September 1992. The amendment provided us with a fully paid-up

license to certain patents and patent applications in exchange for a one-time payment to Molecular Biosystems of $1.0 million.

In December 2001, we licensed to Eyetech Pharmaceuticals, Inc., a publicly-held company, certain of our patents necessary for Eyetech to develop, make and commercialize Macugen, a non-antisense compound intended for use in the treatment of ophthalmic diseases. Eyetech paid us a $2.0 million upfront fee and agreed to pay us milestone and royalty payments in exchange for non-exclusive, worldwide rights to the intellectual property licensed from us. In December 2002, Eyetech entered into an agreement with Pfizer to develop and commercialize Macugen. In 2003, Pfizer and Eyetech reported encouraging Phase III data for Macugen that Eyetech says will serve as the basis for an NDA, with commercialization of the drug expected in 2005. Assuming successful commercialization of Macugen, we have the opportunity to earn future milestone payments and royalties that could be substantial to us.

In October 2000, we licensed some of our novel chemistry patents to Roche, a business unit of Roche Diagnostics, for use in the production of Roche's diagnostic products. The royalty-bearing license grants Roche non-exclusive worldwide access to some of our proprietary chemistries, in exchange for initial and ongoing payments from Roche to us.

In the past, except for small quantities, it was generally expensive and difficult to produce chemically modified oligonucleotides, like the antisense drugs we use in our research and development programs. As a result, we have dedicated significant resources to develop ways to improve manufacturing capacity. Since we can use variants of the same nucleotide building blocks and the same type of equipment to produce our oligonucleotide compounds, we found that the same techniques we used to efficiently manufacture one oligonucleotide drug could help improve the manufacturing processes for many other oligonucleotide drugs. By developing several proprietary chemical processes to scale up our manufacturing capabilities, we have greatly reduced the cost of producing oligonucleotide compounds. For example, we have significantly reduced the cost of raw materials through improved yield efficiency, while at the same time increasing our capacity to make the compounds. Through both our internal research and development programs and collaborations with outside vendors we may achieve even greater efficiency and further cost reductions.

In September 2002, we expanded our relationship with Lilly by agreeing to manufacture Affinitak during the product launch period for Lilly. Through this agreement we upgraded and expanded our manufacturing facility, including the addition of a new state-of-the-art manufacturing suite. Lilly provided us with funding in the form of a $21.2 million loan to build the new suite. In June 2003, we reached a mutually beneficial renegotiation of our manufacturing relationship with Lilly. Lilly waived repayment of the $21.2 million manufacturing loan it provided us to build the new manufacturing facility. Lilly also agreed to allow us to use the facility to manufacture other drugs. In exchange, we released Lilly from its obligations contained in the supply agreement for Affinitak, including the obligation to purchase additional product from us and the obligation to pay for the costs of maintaining an idle manufacturing suite.

In addition, we have contractual obligations to manufacture clinical trial materials and/or commercial supply for Amgen, ATL, Lilly, Novartis and OncoGenex. We believe we have sufficient manufacturing capacity to meet our current and future obligations under existing agreements with our

partners for commercial, research and clinical needs as well as meet our current internal research and clinical needs. We believe that we have, or will be able to develop or acquire, sufficient supply capacity to meet our anticipated needs. We also believe that with reasonably anticipated benefits from increases in scale and improvements in chemistry, we will be able to manufacture antisense compounds at commercially competitive prices.

Our success will depend, in part, on our ability to obtain patent protection for our products in the United States and other countries. We file applications, as appropriate, for patents covering our products and processes. As of March 3, 2004, we owned or had exclusively licensed more than 1,300 issued patents worldwide. Patents issued to us, applied for by us or exclusively licensed by us cover the following types of inventions, processes and products:

Extensive regulation by United States and foreign governmental authorities governs our manufacture and potential sale of therapeutics. In particular, pharmaceutical products are subject to rigorous preclinical and clinical testing and other approval requirements by the FDA in the United

States under the Federal Food, Drug and Cosmetic Act and by comparable agencies in most foreign countries. Various federal, state and foreign statutes also govern or influence the manufacture, safety, labeling, storage, record keeping and marketing of such products. State, local, and other authorities also regulate pharmaceutical manufacturing facilities.

In conjunction with obtaining approval of Vitravene, we successfully passed the manufacturing pre-approval inspection by the FDA and European regulatory authorities. Approval of each new drug will require a rigorous manufacturing pre-approval inspection by regulatory authorities.

In addition to regulations enforced by the FDA, we are also subject to regulation under the Occupational Safety and Health Act, the Environmental Protection Act, the Toxic Substances Control Act, the Resource Conservation and Recovery Act and other present and potential future federal, state and local regulations.

We fund our Ibis program primarily through contracts or subcontracts with agencies of the U.S. Government. As a result, we must comply with various government regulations, including the Federal Acquisition Regulations, or FARs, and agency regulations supplemental to the FARs; the Truth in Negotiations Act, which requires certification and disclosure of all cost and pricing data in connection with certain contract negotiations; and laws, regulations and executive orders restricting the use and dissemination of information classified for national security purposes and the export of certain products and technical data.

For many of their applications, antisense-based drugs, as well as Ibis small molecules, will compete with existing therapies for market share. In addition, a number of companies are pursuing the development of oligonucleotide-based technology and the development of pharmaceuticals utilizing this technology. These companies include specialized pharmaceutical firms and large pharmaceutical companies acting either independently or together with biopharmaceutical companies.

Vitravene and our other products under development address numerous markets. The diseases targeted by our drugs for which we may receive regulatory approval will determine our competition. For certain of our products, an important factor in competition may be the timing of market introduction of competitive products. Accordingly, the relative speed with which we can develop products, complete the clinical trials and approval processes and supply commercial quantities of the products to the market is an important competitive factor. We expect to compete among products approved for sale based on a variety of factors, including, among other things, product efficacy, safety, reliability, availability, price and patent position.

A number of factors have affected the market for Vitravene, our antisense drug for CMV retinitis. Anti-HIV drugs that were introduced prior to Vitravene's approval, have prolonged survival in HIV-infected individuals. This has resulted in a decline in mortality from AIDS, accompanied by a decline in the incidence of many opportunistic infections, including CMV retinitis.

We currently have two drugs in Phase III trials. We licensed Affinitak, our antisense drug for non-small cell lung cancer, to Lilly in August 2001. Under our agreement with Lilly, Lilly is responsible for the commercialization of Affinitak. If future studies support commercialization, we expect that physicians would use Affinitak in combination with current standard chemotherapy regimens for non-small cell lung cancer. As such, we expect that it will be complementary to existing drugs for the treatment of non-small cell lung cancer rather than directly competitive. Our second drug in Phase III trials is alicaforsen, which we are studying in patients with Crohn's disease. Alicaforsen will likely compete with Johnson & Johnson's drug, Remicade, which is approved for the treatment of Crohn's disease and rheumatoid arthritis.

As of March 3, 2004 we employed 457 individuals, of whom 174 held advanced degrees. A significant number of our management and professional employees have had prior experience with pharmaceutical, biotechnology or medical product companies. Collective bargaining agreements do not cover any of our employees, and management considers relations with its employees to be good.

The following set forth certain information regarding our executive officers as of March 3, 2004:

Name	Age	Position
Stanley T. Crooke, M.D., Ph.D.	58	Chairman of the Board, President and Chief Executive Officer
B. Lynne Parshall, J.D.	48	Director, Executive Vice President, Chief Financial Officer and Secretary
C. Frank Bennett, Ph.D.	47	Vice President, Antisense Research
Richard K. Brown, Ph.D.	51	Vice President, Business Development
David J. Ecker, Ph.D.	49	Isis Vice President and President, Ibis Therapeutics
Arthur A. Levin, Ph.D.	50	Vice President, Development
Patricia Lowenstam	57	Vice President, Human Resources and Operations
Karen Lundstedt	39	Vice President, Investor Relations and Corporate Communications
John McNeil	39	Vice President, Informatics
Aron F. Stein, Ph.D.	45	Vice President, Regulatory Affairs and Quality Assurance

STANLEY T. CROOKE, M.D., Ph.D.
Chairman of the Board, President and Chief Executive Officer

Dr. Crooke was a founder of Isis and has been its Chief Executive Officer and a director since January 1989. He served as our President from January 1989 to May 1994, and was elected Chairman of the Board in February 1991. SmithKline Beckman Corporation, a pharmaceutical company, employed Dr. Crooke from 1980 until January of 1989, where his titles included President of Research and Development of SmithKline and French Laboratories. He serves as a director of Antisense Therapeutics Ltd., a biopharmaceutical company, Axon Instruments, Inc., a developer and manufacturer of novel high-technology devices and software for drug discovery, and EPIX Medical, Inc., a developer of magnetic resonance imaging contrast agents. Dr. Crooke is also an adjunct professor of pharmacology at the University of California, San Diego, and San Diego State University.

B. LYNNE PARSHALL, J.D.
Director, Executive Vice President, Chief Financial Officer, and Secretary

Ms. Parshall has served as a director of Isis since September 2000. She has served as our Executive Vice President since December 1995, our Chief Financial Officer since June 1994, and our Secretary since November 1991. From February 1993 to December 1995, she was a Senior Vice President of Isis, and from November 1991 to February 1993, she was a Vice President of Isis. Prior to joining Isis, Ms. Parshall practiced law at Cooley Godward LLP, counsel to Isis, where she was a partner from 1986 to 1991. Ms. Parshall is also a member of the Licensing Executives Society and a member of the American, California and San Diego bar associations.

Dr. Bennett has served as our Vice President, Antisense Research since June 1995. From March 1993 to June 1995, he was Director, Molecular Pharmacology, and from May 1992 to March 1993, he was an Associate Director in our Molecular and Cellular Biology department. Prior to

joining Isis in 1989, Dr. Bennett was employed by SmithKline and French Laboratories in various research positions.

Dr. Brown joined Isis in June 2001 as President of the GeneTrove program and has been our Vice President, Business Development since January 2003. Prior to joining Isis, Dr. Brown was President of Irori, a company that develops, manufactures and markets combinatorial chemistry and medicinal chemistry products to the pharmaceutical industry. He joined Irori in 1996 and served as President from 1998 to June 2001.

Dr. Ecker was a founder of Isis and has served as our Vice President & Managing Director of Ibis Therapeutics, a program of Isis Pharmaceuticals, since June 1995. In 2001 he assumed the role of President of the program. He served as our Vice President, Biology from July 1993 to June 1995, as our Executive Director, Molecular and Cellular Biology from February 1993 to July 1993, and as our Director, Molecular and Cellular Biology from February 1989 to February 1993. From 1984 until February 1989, he was employed by SmithKline and French Laboratories in a variety of research positions.

Dr. Levin has served as our Vice President, Development since 1995. Prior to joining Isis, Dr. Levin worked at Hoffmann-La Roche Inc. where he was Research Leader in their Investigative Toxicology Department managing the Nuclear Receptor Research Group. During his tenure at Hoffman-LaRoche, Dr. Levin also established and supervised laboratories dedicated to the research of mechanisms of toxicity, biochemical toxicology and toxicokinetics.

Ms. Lowenstam has served as our Vice President, Human Resources since January 1995. She joined Isis in August 1992 as our Director, Human Resources and served in that capacity until January 1995. Prior to joining Isis, she held senior management positions in Human Resources with Quotron Systems, Inc., Citicorp, Zale Corporation, and the May Company.

KAREN LUNDSTEDT
Vice President, Investor Relations and Corporate Communications

Ms. Lundstedt has served as our Vice President, Investor Relations and Corporate Communications since April 2000. Ms. Lundstedt joined Isis in August 1999 as our Executive Director, Investor Relations and Corporate Communications. From September 1991 until joining Isis, Ms. Lundstedt held various management positions at Dura Pharmaceuticals, a specialty respiratory pharmaceutical and pulmonary drug delivery company.

Mr. McNeil has served as our Vice President, Informatics since October 1999. Mr. McNeil joined Isis in October 1997 as our Director, Informatics. Prior to joining Isis, Mr. McNeil was President of

John McNeil & Co., Inc., and held various positions at SAIC in San Diego from 1989 to 1997, including Manager of the Laboratory Sensors and Automation division.

Dr. Stein has served as our Vice President, Regulatory Affairs and Quality Assurance since August 2002. Prior to joining Isis, Dr. Stein was Divisional Vice President of Medical and Regulatory Affairs, Hospital Products Division for Abbott Laboratories from September 1999 to August 2002, and Vice President in charge of Regulatory Affairs and Quality Assurance for SEQUUS Pharmaceuticals, Inc. from April 1997 to September 1999.

Investing in our securities involves a high degree of risk. In addition to the other information in this Form 10-K, you should carefully consider the risks described below before purchasing our securities. If any of the following risks actually occur, our business could be materially harmed, and our financial condition and results of operations could be materially and adversely affected. As a result, the trading price of our securities could decline, and you might lose all or part of your investment.

If we or our partners fail to obtain regulatory approval for our products, we will not be able to sell them.

We and our partners must conduct time-consuming, extensive and costly clinical trials to show the safety and efficacy of each of our drug candidates before a drug candidate can be approved for sale. We must conduct these trials in compliance with U.S. Food and Drug Administration regulations and with comparable regulations in other countries. If the FDA or another regulatory agency believes that we or our partners have not sufficiently demonstrated the safety or efficacy of our drug candidates, it will not approve them or will require additional studies, which can be time consuming and expensive and which will delay commercialization of a drug candidate. We and our partners may not be able to obtain necessary regulatory approvals on a timely basis, if at all, for any of our drug candidates. Failure to receive these approvals or delays in these approvals could prevent or delay commercial introduction of a product and, as a result, could negatively impact our ability to generate revenue from product sales. In addition, following approval of a drug candidate, we and our partners must comply with comprehensive government regulations regarding how we manufacture, market and distribute products. If we fail to comply with these regulations, regulators could force us to withdraw a drug candidate from the market or impose other penalties or requirements that also could have a negative impact on our financial results.

We have only introduced one commercial product, Vitravene. We cannot guarantee that any of our other drug candidates will be safe and effective, will be approved for commercialization or that our partners or we can successfully commercialize these drug candidates.

If the results of clinical testing indicate that any of our drugs under development, including Affinitak and alicaforsen, are not suitable for commercial use, or if additional testing is required to demonstrate suitability, we may need to abandon one or more of our drug development programs.

Drug discovery and development has inherent risks, including the risk that molecular targets prove not to be important in a particular disease, the risk that compounds that demonstrate attractive activity in preclinical studies do not demonstrate similar activity in human beings, and the risk that a compound is not safe or effective for use in humans. Antisense technology in particular is relatively new and unproven. We are applying most of our resources to create safe and effective drugs for human use. Any of the risks described above could prevent us from meeting this goal. In the past, we have invested in

clinical studies of drug candidates, including some that remain in our pipeline, that have not resulted in proof of efficacy against targeted indications.

In March 2003, we reported the results of our Phase III clinical trial of Affinitak in patients with late stage non-small cell lung cancer. In this trial, Affinitak, when added to carboplatin and paclitaxol, failed to demonstrate improved survival sufficient enough to support an NDA filing. A similar result could occur with the Affinitak trial Lilly is currently conducting as well as the trials for our other drugs. In 2004, we expect to report the results of our Phase III clinical trials of Alicaforsen in patients with active Crohn's disease. If any of our drugs in clinical studies do not show sufficient efficacy in patients with the targeted indication, it could negatively impact our development and commercialization goals for this and other drugs and our stock price could decline.

If the market does not accept our products, we are not likely to generate revenues or become profitable.

Our success will depend upon the medical community, patients and third-party payors accepting our products as medically useful, cost-effective and safe. We cannot guarantee that, if approved for commercialization, doctors will use our products to treat patients. We currently have one commercially available product, Vitravene, a treatment for cytomegalovirus, or CMV, retinitis in AIDS patients, which addresses a small market. Our partners and we may not successfully commercialize additional products.

The degree of market acceptance for any of our products depends upon a number of factors, including:

Based on the profile of our drug candidates, physicians, patients, patient advocates, payors or the medical community in general may not accept and use any products that we may develop.

If any of our collaborative partners fail to fund our collaborative programs or develop or sell any of our products under development, or if we cannot obtain additional partners, we may have to delay or stop progress on our drug development programs.

We have entered into collaborative arrangements with third parties to develop certain product candidates. We enter into these collaborations in order to:

If any of our partners fail to develop or sell any drug in which we have retained a financial interest, our business may suffer. These collaborations may not continue or result in commercialized drugs. Our collaborators can terminate their relationships with us under certain circumstances, some of

which are outside of our control. Examples of terminated collaborations include the termination in 2002 of our HepaSense and Orasense collaborations with Elan and the termination of our collaboration with Merck to develop ISIS 113715.

We are collaborating with Lilly to develop Affinitak, our most advanced drug candidate, with Lilly funding Affinitak's development. Lilly could decide to discontinue its funding of Affinitak at any time. The results of our recently completed Phase III clinical trial for Affinitak, the market potential of Affinitak or negative results from Lilly's Phase III clinical trial for Affinitak could influence Lilly's decision to discontinue funding of future Affinitak activities.

Other drug candidates in our development pipeline are being developed and/or funded by corporate partners, including Antisense Therapeutics Limited, OncoGenex Technologies Inc. and Lilly. In addition, since its inception, Ibis has received significant financial support from various government agencies to use its technology to develop our sensor to identify infectious agents. We have received significant financial support from U.S. Government-funded grants and contracts for our Ibis program and the development of our TIGER biosensor. The U.S. Government can unilaterally terminate these contracts and grants at its convenience at any time, even if we have fully performed our obligations. If any of these pharmaceutical company or government partners stopped funding and/or developing these products, our business could suffer.

Certain of our partners are pursuing other technologies or developing other drug candidates either on their own or in collaboration with others, including our competitors, to develop treatments for the same diseases targeted by our own collaborative programs. Competition may negatively impact a partners' focus on and commitment to our drug candidate and, as a result, could delay or otherwise negatively affect the commercialization of a drug candidate.

Historically, corporate partnering has played a key role in our strategy to fund our development programs and to add key development resources. We plan to continue to rely on additional collaborative arrangements to develop and commercialize our products. However, we may not be able to negotiate additional attractive collaborative arrangements, and, even if negotiated, the collaborative arrangements may not be successful.

In addition, the disappointing results of our first Affinitak Phase III trial could cause our existing partners to reevaluate their commitment to our drug discovery platforms or could impair our ability to attract new collaborative partners. If any of our collaborative partners withdraw their resources or if we cannot continue to secure additional collaborative partners, our revenues could decrease and the development of our drug candidates could suffer.

We have incurred losses, and our business will suffer if we fail to achieve profitability in the future.

Because drug discovery and development and research services require substantial lead-time and money prior to commercialization, our expenses have exceeded our revenue since we were founded in January 1989. As of December 31, 2003, our accumulated losses were approximately $555.6 million. Most of the losses resulted from costs incurred in connection with our research and development programs and from general and administrative costs associated with our operations. Most of our revenue has come from collaborative arrangements, with additional revenue from interest income and research grants and the sale or licensing of patents. We currently derive our current product revenue solely from sales of Vitravene. This product has limited sales potential. We expect to incur additional operating losses over the next several years, and these losses may increase if we cannot increase or sustain revenue. We may not successfully develop any additional products or services, or achieve or sustain future profitability.

We may not successfully develop or derive revenues from our business based on our TIGER biosensor to identify infectious organisms.

Our TIGER biosensor is subject to the risks inherent in developing tools based on innovative technologies. Our product is at an early stage of development and requires additional research and development prior to marketing. If our potential customers fail to purchase our biosensor due to competition or other factors, or if we fail to develop applications that lead to market acceptance, we could lose our investment in this technology and our TIGER business could fail to meet our business and financial objectives.

If we fail to obtain timely funding, we may need to curtail or abandon some of our programs.

All of our product candidates are still undergoing clinical trials or are in the early stages of research and development. All of our products under development will require significant additional research, development, preclinical and/or clinical testing, regulatory approval and a commitment of significant additional resources prior to their commercialization. Based on our current operating plan, we believe that our available cash, cash equivalents and short-term investments as of December 31, 2003, combined with investment income and committed contractual cash payments will be sufficient to meet our anticipated requirements for at least the next 36 months. If we do not meet our goals to commercialize our drug products and research services or to license our proprietary technologies, we may need additional funding in the future. Our future capital requirements will depend on many factors, such as the following:

If we need additional funds, we may need to raise them through public or private financing. Additional financing may not be available at all or on acceptable terms. If we raise additional funds by issuing equity securities, the shares of existing stockholders will be diluted and their price, as well as the price of our other securities, may decline. If adequate funds are not available, we may have to cut back on one or more of our research, drug discovery or development programs or obtain funds through arrangements with collaborative partners or others. These arrangements may require us to give up rights to certain of our technologies, product candidates or products.

If we cannot manufacture our products or contract with a third party to manufacture our products at costs that allow us to charge competitive prices to buyers, we will not be able to market products profitably.

If we successfully commercialize any of our drug candidates, we may be required to establish large-scale commercial manufacturing capabilities. In addition, as our drug development pipeline increases and matures, we will have a greater need for clinical trial and commercial manufacturing capacity. We have limited experience manufacturing pharmaceutical products of the chemical class represented by

our drug candidates, called oligonucleotides, on a commercial scale for the systemic administration of a drug. There are a small number of suppliers for certain capital equipment and raw materials that we use to manufacture our drugs, and some of these suppliers will need to increase their scale of production to meet our projected needs for commercial manufacturing. Further, we must continue to improve our manufacturing processes to allow us to reduce our product costs. We may not be able to manufacture at a cost or in quantities necessary to make commercially successful products.

Also, manufacturers, including us, must adhere to the FDA's current Good Manufacturing Practices regulations which the FDA enforces through its facilities inspection program. We and our contract manufacturers may not be able to comply or maintain compliance with Good Manufacturing Practices regulations. Non-compliance could significantly delay our receipt of marketing approval for potential products or result in FDA enforcement action after approval that could limit the commercial success of our potential product.

If the results of our Phase III trials for alicaforsen are positive and we fail to secure a marketing and distribution partner for this product, our commercialization efforts for alicaforsen may be harmed or delayed.

We have limited personnel with experience in marketing, selling and distributing products. We expect to depend on third parties to commercialize alicaforsen if our Phase III trials for alicaforsen are positive and we receive marketing approval. If we are unable to reach agreements with suitable third parties, we may fail to meet our business objectives for alicaforsen. We may not successfully establish a collaboration or be able to make alternative arrangements. Moreover, a collaboration or other arrangement we secure may not succeed.

If we fail to compete effectively, our products will not contribute significant revenues.

Our competitors are engaged in all areas of drug discovery throughout the world, are numerous, and include, among others, major pharmaceutical companies and specialized biopharmaceutical firms. Other companies are engaged in developing antisense technology or unique methods of identifying infectious organisms. Our competitors may succeed in developing drug candidates or technologies that are more effective than any drug candidates or technologies that we are developing. These competitive developments could make our products obsolete or non-competitive.

Many of our competitors have substantially greater financial, technical and human resources than we do. In addition, many of these competitors have significantly greater experience than we do in conducting preclinical testing and human clinical trials of new pharmaceutical products and in obtaining FDA and other regulatory approvals of products for use in health care. Accordingly, our competitors may succeed in obtaining regulatory approval for products earlier than we do. We will also compete with respect to marketing and sales capabilities, areas in which we have limited or no experience.

If we cannot protect our patents or our proprietary rights, others may compete more directly against us.

Our success depends to a significant degree upon our ability to continue to develop and secure intellectual property rights to proprietary products and services. However, we may not receive issued patents on any of our pending patent applications in the United States or in other countries. In addition, the scope of any of our issued patents may not be sufficiently broad to provide us with a competitive advantage. Furthermore, our issued patents or patents licensed to us may be successfully challenged, invalidated or circumvented so that our patent rights would not create an effective competitive barrier.

Intellectual property litigation could be expensive and prevent us from pursuing our programs.

It is possible that in the future we may have to defend our intellectual property rights. In the event of an intellectual property dispute, we may be forced to litigate to defend our rights or assert them against others. Disputes could involve litigation or proceedings declared by the U.S. Patent and Trademark Office or the International Trade Commission. Intellectual property litigation can be extremely expensive, and this expense, as well as the consequences should we not prevail, could seriously harm our business.

If a third party claims that our products or technology infringe their patents or other intellectual property rights, we may have to discontinue an important product or product line, alter our products and processes, pay license fees or cease certain activities. We may not be able to obtain a license to such intellectual property on favorable terms, if at all. There are many patents issued or applied for in the biotechnology industry, and we may not be aware of patents or applications held by others that relate to our business. This is especially true since patent applications in the U.S. are filed confidentially. Moreover, the validity and breadth of biotechnology patents involve complex legal and factual questions for which important legal issues remain unresolved.

If we do not progress in our programs as anticipated, the price of our securities could decrease.

For planning purposes, we estimate the timing of a variety of clinical, regulatory and other milestones, like when a certain product candidate will enter the clinic, when we will complete a clinical trial, or when we will file an application for marketing approval. We base our estimates on present facts and a variety of assumptions. Many of the underlying assumptions are outside of our control. If we do not achieve milestones when we expect to, investors could be disappointed and the price of our securities would likely decrease. For example, since the data from our Phase III trial for Affinitak were not sufficiently positive to support a single study NDA, we now must wait for the results of Lilly's ongoing Phase III Affinitak trial before we reevaluate whether the data are sufficiently positive to support filing an NDA for Affinitak. We expect results from this second Phase III trial in the second half of 2004.

If Macugen does not achieve marketing approval or its commercial success does not meet our expectations, we will not receive milestone and royalty payments.

As part of our license agreement with Eyetech, we are entitled to receive milestones and royalty payments. However, if Eyetech does not achieve these milestones or receive marketing approval for Macugen, or if Eyetech receives marketing approval for Macugen but fails to commercialize Macugen as expected, we may not receive these payments, or derive the expected value.

The loss of key personnel, or the inability to attract and retain highly skilled personnel, could make it more difficult to run our business and reduce our likelihood of success.

We are dependent on the principal members of our management and scientific staff. We do not have employment agreements with any of our executive officers. The loss of our management and key scientific employees might slow the achievement of important research and development goals. It is also critical to our success that we recruit and retain qualified scientific personnel to perform research and development work. We may not be able to attract and retain skilled and experienced scientific personnel on acceptable terms because of intense competition for experienced scientists among many pharmaceutical and health care companies, universities and non-profit research institutions. Failure to succeed in specific clinical trials, including the results of our first Phase III Affinitak trial, may make it more challenging to recruit and retain qualified scientific personnel.

We depend on third parties in the conduct of our clinical trials for our product candidates and any failure of those parties to fulfill their obligations could adversely affect our development and commercialization plans.

We depend on independent clinical investigators, contract research organizations and other third party service providers in the conduct of our clinical trials for our product candidates and expect to continue to do so in the future. We rely heavily on these parties for successful execution of our clinical trials, but do not control many aspects of their activities. For example, the investigators are not our employees. However, we are responsible for ensuring that each of our clinical trials is conducted in accordance with the general investigational plan and protocols for the trial. Third parties may not complete activities on schedule, or may not conduct our clinical trials in accordance with regulatory requirements or our stated protocols. The failure of these third parties to carry out their obligations could delay or prevent the development, approval and commercialization of our product candidates.

If the price of our securities continues to be highly volatile, this could make it harder for you to liquidate your investment and could increase your risk of suffering a loss.

The market price of our common stock, like that of the securities of many other biopharmaceutical companies, has been and is likely to continue to be highly volatile. These fluctuations in our common stock price may significantly affect the trading price of our securities. During the 12 months preceding December 31, 2003, the market price of our common stock has ranged from $2.50 to $8.05 per share. Many factors can affect the market price of our securities, including, for example, fluctuations in our operating results, announcements of collaborations, clinical trial results, technological innovations or new drug products being developed by us or our competitors, governmental regulation, regulatory approval, developments in patent or other proprietary rights, public concern regarding the safety of our drugs and general market conditions.

Provisions in our certificate of incorporation, other agreements and Delaware law may prevent stockholders from receiving a premium for their shares.

Our certificate of incorporation provides for classified terms for the members of our board of directors. Our certificate also includes a provision that requires at least 66 ?% of our voting stockholders to approve a merger or certain other business transactions with, or proposed by, any holder of 15% or more of our voting stock, except in cases where certain directors approve the transaction or certain minimum price criteria and other procedural requirements are met.

Our certificate of incorporation also requires that any action required or permitted to be taken by our stockholders must be taken at a duly called annual or special meeting of stockholders and may not be taken by written consent. In addition, only our board of directors, chairman of the board or chief executive officer can call special meetings of our stockholders. We also have implemented a stockholders' rights plan, also called a poison pill, which could make it uneconomical for a third party to acquire our company on a hostile basis. These provisions, as well as Delaware law and other of our agreements, may discourage certain types of transactions in which our stockholders might otherwise receive a premium for their shares over then current market prices, and may limit the ability of our stockholders to approve transactions that they think may be in their best interests. In addition, our board of directors has the authority to fix the rights and preferences of and issue shares of preferred stock, which may have the effect of delaying or preventing a change in control of our company without action by our stockholders.

If registration rights that we have previously granted are exercised, then the price of our securities may be negatively affected.

We have granted registration rights in connection with the issuance of our securities to Elan International Services, Ltd., Eli Lilly and Company, and Reliance Insurance Company. In the aggregate, these registration rights cover approximately 4,166,667 shares of our common stock which are currently outstanding and additional shares of our common stock which may become outstanding upon the conversion of outstanding convertible securities. If these holders exercise their registration rights, it will bring additional shares of our common stock into the market, which may have an adverse effect on the price of our securities.

As of March 3, 2004, we occupied approximately 218,000 square feet of laboratory and office space, including 6,888 square feet of manufacturing area built to meet Good Manufacturing Practices. We are primarily located in six buildings in Carlsbad, California. We own three of these buildings and, as of December 31, 2003, these buildings secured approximately $6.4 million of our debt. We lease three of the buildings under lease agreements, of which two leases will expire in 2007 and one will expire in 2010. In February 2003, we completed an expansion of our manufacturing facility to upgrade our existing manufacturing suite and add a second state of the art manufacturing suite.

Item 5. Market For Registrant's Common Equity and Related Stockholder Matters.

Our common stock is traded publicly through the Nasdaq National Market under the symbol "ISIS." The following table presents quarterly information on the price range of our common stock. This information indicates the high and low sale prices reported by the Nasdaq National Market. These prices do not include retail markups, markdowns or commissions.

	HIGH		LOW
2003
First Quarter	$	7.55	$	2.50
Second Quarter	$	6.85	$	3.50
Third Quarter	$	8.05	$	4.55
Fourth Quarter	$	7.07	$	5.20
2002
First Quarter	$	22.40	$	14.07
Second Quarter	$	18.00	$	6.76
Third Quarter	$	11.86	$	6.10
Fourth Quarter	$	11.00	$	6.00

As of March 3, 2004, there were approximately 1,084 stockholders of record of our common stock. We have never paid dividends and do not anticipate paying any dividends in the foreseeable future.

The following table sets forth information regarding outstanding options and shares reserved for future issuance under our equity compensation plans as of December 31, 2003.

We adopted the 2000 Broad-Based Equity Incentive Plan, or the 2000 Plan, to provide our employees, officers, directors and consultants an opportunity to benefit from increases in the value of our common stock through the granting of nonstatutory stock options, stock bonuses and rights to purchase restricted stock. At the time we adopted the 2000 Plan, we were not required to seek the approval of our stockholders. The Board has delegated administration of the 2000 Plan to the Compensation Committee of the Board, and the Compensation Committee has delegated administration of the 2000 Plan to the Non-Management Stock Option Committee with respect to certain option grants to employees who are not our executive officers. The Board has the power to construe and interpret the 2000 Plan and, subject to the provisions of the 2000 Plan, to select the persons to whom stock awards are to be made, to designate the number of shares to be covered by each stock award, to establish vesting schedules, to specify the exercise price and the type of consideration to be paid to us upon exercise or purchase.

As of December 31, 2003, the 2000 Plan had 5,737,000 shares reserved for issuance, options to purchase an aggregate of 5,198,000 shares have been granted and were outstanding under the 2000 Plan, options to purchase an aggregate of 253,000 shares have been exercised under the 2000 Plan, and 539,000 shares remained available for grant thereunder.

Options granted under the 2000 Plan generally have a term of ten years, have an exercise price equal to the fair market value at the time of grant, can only be exercised with a cash payment and vest at the rate of 25% per year after the first year and then at the rate of 2.08% per month thereafter during the optionee's employment or service as a consultant or term as an affiliate. Options granted pursuant to the April 2003 stock option exchange program as discussed in Item 7, "Management's Discussion and Analysis of Financial Condition and Results of Operations," and in the Notes to Consolidated Financial Statements, expire on December 31, 2008 and vested 33.34% on January 1, 2004 and then at the rate of 2.78% per month during the optionee's employment or service as a consultant or term as an affiliate.

If any change is made in the common stock subject to the 2000 Plan, or subject to any stock award, without the receipt of consideration by us (through merger, consolidation, reorganization, recapitalization, reincorporation, stock dividend, dividend in property other than cash, stock split, liquidating dividend, combination of shares, exchange of shares, change in corporate structure or other transaction not involving the receipt of consideration by us), we will adjust the 2000 Plan appropriately in the class(es) and maximum number of securities subject to the 2000 Plan, and we will adjust the outstanding stock awards appropriately in the class(es) and number of securities and price per share of common stock subject to such outstanding stock awards. Our Board will make such adjustments, and its determination will be final, binding and conclusive. We will not treat the conversion of any of our convertible securities as a transaction without receipt of consideration.

In the event of our dissolution or liquidation, all outstanding stock awards will terminate immediately prior to such event.

then any surviving corporation or acquiring corporation will assume any stock awards outstanding under the 2000 Plan or will substitute similar stock awards (including an award to acquire the same consideration paid to the shareholders in the transaction for those outstanding under the 2000 Plan). In the event any surviving corporation or acquiring corporation refuses to assume such stock awards or to substitute similar stock awards for those outstanding under the 2000 Plan, then with respect to stock awards held by participants whose continuous service has not terminated, we will accelerate the vesting of such stock awards in full (and, if applicable, the time during which such stock awards may be exercised) and the stock awards will terminate if not exercised (if applicable) at or prior to such event. With respect to any other stock awards outstanding under the 2000 Plan, such stock awards will terminate if not exercised (if applicable) prior to such event. In addition, as of December 31, 2003, approximately 5,198,000 stock awards granted under the 2000 plan will be accelerated in full if a transaction described above occurs, even if the surviving corporation assumes such award.

We make available on our web site, www.isispharm.com, our 10-K, 10-Q's, 8-K's and amendments thereto, as soon as reasonably practical after we file such materials with the Securities and Exchange Commission. Any information that is included on or linked to our Internet site is not a part of this report or any registration statement that incorporates this report by reference.

Item 6. Selected Consolidated Financial Data (in thousands, except per share amounts):

	Years Ended December 31,
	2003		2002		2001		2000		1999
Consolidated Statement of Operations Data:
Revenue (includes amounts for R&D, licensing and royalties)	$	49,990		$	80,179		$	53,273		$	37,255		$	33,925
Research and development expenses	$	116,963		$	124,074		$	83,741		$	57,014		$	66,413
Net loss applicable to common stock	$	(95,690	)	$	(73,302	)	$	(75,131	)	$	(54,699	)	$	(59,645	)
Basic and diluted net loss per share	$	(1.73	)	$	(1.35	)	$	(1.70	)	$	(1.48	)	$	(2.08	)
Shares used in computing basic and diluted net loss per share		55,463			54,480			44,109			37,023			28,703

	Years Ended December 31,
	2003		2002		2001		2000		1999
Consolidated Balance Sheet Data:
Cash, cash equivalents and short-term investments	$	215,504		$	289,353		$	312,018		$	127,262		$	52,839
Working capital	$	194,004		$	244,230		$	280,569		$	118,568		$	44,213
Total assets	$	334,942		$	438,683		$	417,061		$	183,256		$	103,107
Long-term debt and capital lease obligations, less current portion	$	213,397		$	192,893		$	125,710		$	102,254		$	87,254
Accumulated deficit	$	(555,583	)	$	(459,893	)	$	(386,591	)	$	(311,460	)	$	(256,761	)
Stockholders' equity	$	67,178		$	155,477		$	223,099		$	66,366		$	869

Item 7. Management's Discussion and Analysis of Financial Condition and Results of Operations

Since our inception in 1989, we have pioneered the science of antisense for the development of a new class of drugs. We have designed antisense drugs to treat a wide variety of diseases. Due to their gene selectivity, antisense drugs have the potential to be highly effective and less toxic than traditional drugs. We have made significant progress in understanding capabilities of antisense drugs in treating disease. We have developed new chemistries and novel formulations to enhance the potency and utility of antisense drugs, and we have successfully turned our expertise into 11 antisense drugs currently in all phases of clinical development. Our drugs in development treat a variety of health conditions, including inflammatory, viral, metabolic, dermatological and cardiovascular diseases, and cancer. We are studying these drugs in intravenous, subcutaneous, topical cream, enema, aerosol, and oral formulations, and we are advancing antisense drugs using second-generation chemistry. We achieved marketing clearance for the world's first antisense drug, Vitravene (fomivirsen) in 1998.

Affinitak, formerly LY900003 or ISIS 3521, which we licensed to Eli Lilly and Company, or Lilly, in 2001, is our most advanced product in development. In March 2003, we announced the results of our Phase III clinical trial of Affinitak to treat patients with non-small cell lung cancer, which were not sufficient to support a single study new drug application. Lilly and we completed an analysis of the data from this trial and presented a summary of the findings at the 39th Annual Meeting of the American Society of Clinical Oncology in June 2003. In a second Phase III study, Lilly is continuing to follow enrolled patients. Lilly and we will make a decision about the future development of Affinitak pending a review upon completion of the second Phase III trial, which most likely will occur in the second half of 2004.

We are conducting two Phase III clinical trials for another product, ISIS 2302, or alicaforsen, in an inflammatory bowel disease known as Crohn's disease. These trials are being conducted in North America and Europe. We expect to report data from these clinical trials in the second half of 2004. We also have Phase II programs ongoing for five additional products.

Our Ibis program has invented a platform technology that has the potential to revolutionize the identification of infectious diseases. Through a project called Triangulation Identification for Genetic Evaluation of Risks, or TIGER, we have applied our proprietary technologies to develop a biological sensor to identify a broad range of infectious organisms in a sample, including organisms that are newly-emerging, genetically altered and unculturable. We have successfully demonstrated proof-of-principle of the TIGER biosensor with the identification of a variety of bacteria and viruses in both environmental and human clinical samples. To date, our Ibis program has received awards and contracts of up to $55.0 million from agencies of the U.S. Government, including DARPA, CDC, USAMRIID, the United States Navy, and the FBI, among others.

We are pursuing three early-stage antisense mechanisms, including RNA Interference, or RNAi, micro-RNA, and alternative splicing through research collaborations and partnerships.

We prepare our consolidated financial statements in conformity with accounting principles generally accepted in the United States of America. As such, we are required to make certain estimates, judgments and assumptions that we believe are reasonable, based upon the information available to us. These judgments involve making estimates about the effect of matters that are inherently uncertain and may significantly impact our quarterly or annual results of operations and financial condition. We discuss the development, selection and disclosure of such estimates with our audit committee each quarter. There are specific risks associated with these critical accounting policies that we describe in the following paragraphs. For all of these policies, we caution that future events rarely develop exactly as expected, and that best estimates routinely require adjustment. The significant accounting policies, which we believe are the most critical to aid in fully understanding and evaluating our reported financial results, require the following:

We follow the provisions as set forth by current accounting rules, which primarily include Staff Accounting Bulletin No. 101, or SAB 101, "Revenue Recognition in Financial Statements," SAB 104, "Revenue Recognition," and Financial Accounting Standards Board Emerging Issue Task Force No. 00-21, or EITF 00-21, "Accounting for Revenue Arrangements with Multiple Deliverables."

We generally recognize revenue when we have satisfied all contractual obligations and we are reasonably assured of collecting the resulting receivable. We are often entitled to bill our customers and receive payment from our customers in advance of recognizing the revenue under current accounting rules. We include in deferred revenue on the balance sheet any revenues which we are entitled to bill or have collected in advance of recognizing the associated revenue.

We often enter into collaborations where we receive nonrefundable up-front payments for prior or future expenditures. We recognize revenue related to up-front payments ratably over the period of the contractual arrangements as we satisfy our performance obligations. Occasionally, we are required to estimate the period of a contractual arrangement or our performance obligations when the agreements we enter into do not clearly define such information. Should different estimates prevail, revenue recognized could be materially different. We have made estimates of our continuing obligations on several agreements, including our collaborations with ATL, Amgen, Chiron, ITRI, Lilly and the Singapore EDB.

As part of our Lilly alliance, in 2001 Lilly provided us a $100.0 million interest free loan to fund the research collaboration. We take quarterly drawdowns against this loan and discount the amounts to

their net present value by imputing interest on the amount at 20%, which represented market conditions in place at the time we entered into the loan. As of December 31, 2003, we had drawn down $73.8 million on this loan. We are accreting the loan up to its face value over its term by recording interest expense. The difference between the cash received and the present value of the loan represents value Lilly gave to us to help fund the research collaboration. We account for this difference as deferred revenue and recognize it as revenue over the period of contractual performance.

Our collaborations often include contractual milestones. When we achieve these milestones, we are entitled to payment, as defined by the underlying agreements. We generally recognize revenue related to milestones upon completion of the milestone's performance requirement, as long as we are reasonably assured of collecting the resulting receivable, and we are not obligated to future performance related to the achievement of the milestone. We recognized revenue during 2003 related to milestones achieved under our agreements with Lilly, Amgen, and ITRI.

We generally recognize revenue related to the sale of our inventory as we ship or deliver drugs to our partners. To date, in two instances, we completed the manufacturing of drugs, but our partners asked us to deliver the drug on a later date. Under these circumstances, we ensured that our obligation was complete under the terms of the manufacturing agreement in place, and title had transferred to the customer, before we recognized the related revenue.

We often enter into agreements to license our proprietary patent rights on an exclusive or non-exclusive basis in exchange for license and/or royalty fees. We generally recognize as revenue immediately those licensing and royalty agreements we enter into for which we have no future performance obligations and are reasonably assured of collecting the resulting receivable.

We sometimes enter into revenue arrangements that contain multiple deliverables. In these cases, we recognize revenue from each element of the arrangement as long as we are able to determine a separate value for each element, we have completed our obligation to deliver or perform on that element, and we are reasonably assured of collecting the resulting receivable.

We account for our investments in marketable securities in accordance with current accounting rules as set forth by SFAS 115, "Accounting for Certain Investments in Debt and Equity Securities." We carry these investments at fair market value based upon market prices quoted on the last day of the fiscal quarter. We record unrealized gains and losses as a separate component of stockholders' equity, and include gross realized gains and losses in investment income.

In addition to our investments in cash and cash equivalents, we also have equity investments in privately-and publicly-held biotechnology companies. We hold ownership interests of less than 20% in each of the respective entities. In determining if and when a decrease in market value below our cost is other-than-temporary in our equity positions, we examine historical trends in the stock price, the financial condition, near term prospects of the issuer, and our current need for cash. When we determine that a decline in value is other-than-temporary, we recognize an impairment loss in the period in which the other-than-temporary decline occurs. During the first quarter of 2003, we recorded a non-cash loss of $2.4 million related to the impairment of our equity investments in ATL and Hybridon. We recorded these charges based on declines in market value of the equity investments, as compared to their initial valuations, which we determined to be other-than-temporary.

We assess the value of our long-lived assets, which include property and equipment, patent costs, and licenses acquired from third parties, under the provisions set forth by SFAS 144, Accounting for the Impairment or Disposal of Long-Lived Assets, or SFAS 144. We evaluate our long-lived assets for impairment on at least a quarterly basis. During this process, we review our property and equipment

listings, pending domestic and international patent applications, domestic and international issued patents, and licenses we have acquired from other parties. To determine if any impairment is present, we consider the following, among other factors:

In the event that we determine that impairment exists where we had previously determined that it did not exist, we may need to make a material adjustment to our consolidated financial statements. To date, we have experienced no significant impairment of our long-lived assets.

We include in inventory material costs and related manufacturing costs for drugs that we manufacture for our partners under contractual terms and that we use primarily in our clinical development activities and drug products. We expense these costs when we deliver our drugs to partners, or as we use these drugs in our own clinical trials. We reflect our inventory on the balance sheet at the lower of cost or market value under the first-in, first-out method. We review inventory periodically and reduce our carrying value of items considered to be slow moving or obsolete to their estimated net realizable value. We consider several factors in estimating the net realizable value, including shelf lives of raw materials, alternative uses for our drugs and clinical trial materials and historical write-offs. In the second quarter of 2003, we reduced the carrying value of our raw materials related to Affinitak to zero.

We maintain accrued liabilities related to unbilled costs for ongoing preclinical studies and clinical trials. These costs primarily relate to third-party clinical management costs, laboratory costs and analysis, toxicology studies and investigator grants, among other costs. We have multiple drugs in concurrent preclinical studies and clinical trials at several clinical sites throughout the world. We expect that at any given time we will have liabilities outstanding for our preclinical and clinical development costs related to products or services for which our service providers have not yet billed us. In order to ensure that we have adequately provided for ongoing preclinical and clinical development costs during the period in which we incur such costs, we maintain an accrual to cover these costs. We update our estimate for this accrual on at least a quarterly basis. The assessment of these costs is a subjective process that requires judgment. The ultimate settlement of these costs may differ materially from the amounts we have accrued in our consolidated financial statements.

As disclosed in Note 5 of Notes to the Consolidated Financial Statements, we record net deferred tax assets. We record a valuation allowance to offset the net deferred tax assets because we are uncertain that we will realize these net tax assets. When and if circumstances warrant, we will assess the likelihood that our net deferred tax assets will more likely than not be recovered from future

taxable income and record an appropriate reversal to the valuation allowance. Because we have had net operating losses since inception, we have established a 100% valuation allowance for our net deferred tax asset. As of December 31, 2003, our deferred tax assets and valuation allowance totaled approximately $235.0 million.

Total revenue for the year ended December 31, 2003 was $50.0 million, compared to $80.2 million for 2002. The decrease of $30.2 million was primarily due to the reduction in revenue associated with the clinical development of Affinitak and the conclusion of Elan's participation in the Orasense and HepaSense collaborations in late 2002. We did not have revenue associated with Orasense and HepaSense in 2003 as a result of the conclusion of Elan's participation in the joint ventures. New revenue sources not present in 2002 offset, in part, the decreases in total revenue in 2003.

Under the category research and development revenue under collaborative agreements, for the year ended December 31, 2003, we earned $49.5 million, compared to $67.8 million for 2002. The decrease of $18.3 million was primarily due to the reduction in revenue associated with the clinical development of Affinitak. In March 2003, we announced the results of our Phase III clinical trial of Affinitak to treat patients with non-small cell lung cancer, which were not sufficient to support a single- study new drug application. New sources of revenue not present in 2002 slightly offset the previously described decreases in revenue. New sources of revenue from our Antisense Drug Discovery collaborations included:

New revenue sources from our Ibis program resulted from the expansion of work to continue the advancement of our TIGER technology, which included work for government agencies such as the CDC, FBI, and DARPA. Our work for DARPA continues to be a collaborative effort with SAIC and accounted for approximately 16% and 6% of our revenue in 2003 and 2002, respectively. Our ability to maintain revenue at current levels will depend on new revenue sources and expansion of existing revenue sources in 2004.

Research and development revenue from affiliates consisted of revenue associated with our two collaborations with Elan, Orasense and HepaSense. Elan concluded its participation in the HepaSense and Orasense collaborations in late 2002, in conjunction with its restructuring efforts. As a result, we reacquired product rights to ISIS 14803 for hepatitis C and the oral formulation of ISIS 104838 from the HepaSense and Orasense joint ventures, respectively. We did not earn revenue from these affiliates in 2003. During 2002, we recognized $8.9 million and $3.0 million as revenue from Orasense and HepaSense, respectively. We do not expect to earn revenue related to these collaborations in 2004.

Our revenue from licensing activities and royalties was $523,000 for the year ended December 31, 2003, compared with $417,000 in 2002. We do not anticipate that revenue from licensing activities and royalties will comprise a significant portion of our revenue in 2004.

Total operating expenses were $129.0 million and $131.0 million for the years ended December 31, 2003 and 2002, respectively. The decrease of $2.0 million was primarily due to our implementation of an expense reduction plan during the second quarter of 2003, the completion of our development activities for Affinitak, and the termination of our Genetrove database product offering in November 2002. Increased clinical development expenses for many of our other products and increased costs in the Ibis program partially offset the decrease in operating expenses. Increased expenses of $3.9 million and $430,000 for compensation related to stock options and restructuring activities, respectively, further offset the decrease in operating expenses. In order to analyze and compare our results of operations to other similar companies, we believe that it is important to exclude compensation expense or benefit related to stock options from operating expenses because it is based on the variability of our stock price rather than operations. We also believe that it is important to exclude restructuring activities because these costs are directly related to isolated events.

For the year ended December 31, 2003, we reported total research and development expenditures of $117.0 million, compared to $124.1 million reported in 2002. Our research and development expenses consist of costs for antisense drug discovery, antisense drug development, our Ibis program and related R&D Support costs. The antisense drug discovery costs include costs associated with our GeneTrove program. The decrease of $7.1 million in 2003 over 2002 was primarily due to the termination of our GeneTrove product offering and reorganization of the GeneTrove program in November 2002, our implementation of an expense reduction plan in April 2003, and a decrease in our total Affinitak related expenses. Increased clinical development expenses for many of our other products and our increased efforts in the Ibis program offset, in part, our decreases in research and development expenditures. We expect that total research and development spending will increase in 2004 to support our ongoing research and development activities.

Identifying what a gene does, called gene functionalization, and then determining whether a specific gene is a good target for drug discovery, called target validation, are the first steps in our drug discovery process for our researchers. We use this information in our internal drug discovery process and, through our GeneTrove program, we sell these services and the resulting information to pharmaceutical and biotechnology companies in collaborations. GeneTrove is an integral component of our antisense drug discovery group. As such, GeneTrove shares many of its resources including people, equipment and facilities with the rest of our antisense drug discovery group. In November 2002, we terminated our GeneTrove database product offering and reorganized the GeneTrove program. We incurred a one-time charge of approximately $1.4 million associated with this restructuring in the fourth quarter of 2002. The cost savings from the termination of our GeneTrove database has been offset, in part, by increases in gene functionalization and target validation services for our in-house drug discovery programs and our existing partners. Our current focus is to use GeneTrove information to direct our own drug discovery research and that of our antisense drug discovery partners, like Lilly and Amgen. Antisense drug discovery is also the function within Isis that is responsible for advancing antisense core technology. Through the efforts of our scientists in the antisense drug discovery group, we have produced second-generation antisense drugs that have been shown to have increased potency, increased stability, an improved side effect profile and the potential for oral administration. With more

than a decade focused on learning the capabilities of antisense technology and how these compounds behave in the body, our scientists have learned the organs and tissues in humans to which antisense therapy is effectively directed. Using this knowledge, we have strategically focused our research programs on those sites in the body that accept antisense readily like the liver, kidney, fat tissue and bone marrow. These targets expand the current therapeutic scope of antisense research into new disease categories, including obesity and cardiovascular disease. The work of our scientists has given us the opportunity to enter into important drug discovery relationships with industry leaders like Lilly and Amgen.

As we expand our research programs into new sites in the body and new disease categories, we would expect to see our expenses for antisense drug discovery increase. We anticipate that our existing relationships with Lilly and Amgen combined with our collaborations with ITRI, the Singapore EDB and Ercole and prospective new partners, will continue to help fund our many research programs, as well as contribute to the advancement of the science by funding core antisense technology research.

Antisense drug discovery costs for the year ended December 31, 2003 totaled $36.8 million compared to $38.9 million for 2002. The decrease of $2.1 million in 2003 over 2002 was principally a result of our planned expense reductions started in the second quarter of 2003 and the termination of our GeneTrove database product offering and reorganization of the GeneTrove program in November 2002. These decreases were offset in part by the increase in expenses to support our research collaborations with Lilly, Amgen, and ITRI.

Our development activities reflect our efforts to advance our drugs through the various stages of preclinical, or animal studies, and human clinical trials. The development plans for our drugs are subject to numerous uncertainties like obtaining regulatory approval, market availability and successfully obtaining funding, which may affect our research and development expenditures and capital resources. Prior to starting clinical trials, we test our potential product candidates in numerous preclinical studies to identify disease indications for which they may be candidates. Once we have established that a preclinical drug candidate has met certain clinical requirements and we have filed an Investigational New Drug Application, or IND, with the FDA, we may initiate clinical trials for that drug. It may take several years to complete clinical trials, with the length varying substantially according to the complexity, novelty and intended use of the product candidate. The following timelines represent our estimate of typical completion times for clinical trials we generally conduct: Phase I—one year, Phase II—one to two years, and Phase III—two to four years. However, the duration and the cost of clinical trials may vary significantly depending on a variety of factors including a trial's protocol, the number of patients in the trial, the duration of patient follow-up, the number of clinical sites in the trial, and the length of time required to enroll suitable patient subjects. Although we spend a considerable amount of time planning our clinical trials, often we are required to alter from our plan. For example, we may need to alter the number of patients in a trial or extend the duration of patient follow-up. Any required deviation from our plan may require us to incur additional expenditures.

We may conduct multiple clinical trials on a drug candidate, including multiple clinical trials for the variety of indications we may be studying. Furthermore, as we obtain results from trials we may elect to discontinue clinical trials for certain drug candidates in certain indications in order to focus our resources on more promising drug candidates or indications. Generally, a late stage Phase III trial is substantially more expensive than early stage trials, such as Phase I or Phase II. Currently we have 11 drug candidates in various stages of development, including two drugs in Phase III clinical trials. In March 2003, we announced the results of our Phase III clinical trial of Affinitak for the treatment of non-small cell lung cancer. The results were not sufficient to support a single study new drug application. In a second Phase III study, Lilly is continuing to follow enrolled patients. Lilly and we will make a decision about the future development of Affinitak pending a review upon completion of the second Phase III trial, which most likely will occur in the second half of 2004. Affinitak related

expenses decreased in 2003 compared to 2002. We have two Phase III trials of alicaforsen, or ISIS 2302, in people with active Crohn's disease. We are conducting one of these studies in North America and the other in Europe. These studies are evaluating the safety and efficacy of alicaforsen. We also have Phase II programs ongoing for five additional products, Phase I programs ongoing for three additional products, and preclinical studies ongoing for one project. If we partner a drug, it may affect the size of a trial, its timing, its total cost and the timing of the related cost. For example, under our licensing agreement with Lilly for Affinitak, we received an upfront license payment that helped fund the costs we incurred during 2001 for our Phase III trial. In addition, during 2002 and 2003, Lilly reimbursed us for our costs related to our development of Affinitak.

Development expenditures totaled $45.1 million and $55.3 million for the years ended December 31, 2003 and 2002, respectively. The decrease of $10.2 million was primarily due to planned expense reductions started in the second quarter of 2003 driven by a reduction in clinical development activity related to Affinitak. These decreases were offset, in part, by increased clinical development expenses for our other products, mainly alicaforsen for Crohn's disease and several products in Phase II and earlier stages of development.

The planned expense reductions began in April 2003 when we initiated a restructuring effort in response to the disappointing results from our Phase III trial of Affinitak. As a result, we had a small reduction in our workforce, which primarily represented positions that were in support of the commercialization and manufacture of Affinitak. We completed our development activities for Affinitak in 2003. As a result, expenditures related to Affinitak decreased from $22.6 million in 2002 to $5.1 million in 2003. The majority of the 2003 expenditures occurred in the first quarter of 2003.

Our second drug in Phase III clinical trials, alicaforsen for Crohn's disease, had development expenditures totaling $6.5 million for the year ended December 31, 2003, compared to $4.8 million for the same period of 2002. The increase of $1.7 million was the result of an increase in the number of patients undergoing treatment in our two Phase III trials. These trials were initiated in November 2001 and June 2002.

Expenditures related to our other products in development totaled $29.3 million for the year ended December 31, 2003, compared to $23.1 million for 2002. The increase of $6.2 million in 2003 over 2002 was mainly the result of increased expenses for Phase II trials of alicaforsen in ulcerative colitis, the Phase I trials for ISIS 113715 for type 2 diabetes and preclinical studies and initiation of a Phase I trial of ISIS 301012 for cardiovascular disease.

Expenditures in our Ibis program have historically included costs for scientists, equipment costs, laboratory supplies, chemicals and highly specialized information technology consultants to advance the research and development of anti-infectives. Costs for these items have increased since Ibis expanded the application of its technology to develop a biosensor to identify a broad range of infectious organisms in a sample, including organisms that are, newly-emerging, genetically altered, or unculturable.

Since its inception, Ibis has received significant financial support from various government agencies to advance its technology for the identification and treatment of infectious disease. From June 2000 through June 2002, Ibis also received funding from its collaboration with Pfizer, which ended in June 2002 in accordance with the terms of the agreement.

Ibis expenditures for the year ended December 31, 2003 totaled $9.9 million, compared to $8.3 million for 2002. The increase of $1.6 million was primarily related to Ibis' continued performance obligations under its multi-year government contracts with DARPA through our relationship with SAIC, awarded in October 2001, and USAMRIID, awarded in March 2002, and new performance obligations under Ibis' agreement with the CDC, awarded in September 2003, and with various other government agencies. Included as Ibis expenditures are fixtures and equipment accounted for as pass-through costs

that Ibis purchased for government agencies under the contractual terms of their agreements. We expect our costs for Ibis to increase as we continue to expand this business.

In our research and development expenses, we include support costs such as rent, building and equipment repair and maintenance, utilities, depreciation of laboratory equipment and facilities, amortization of our intellectual property, information technology costs, procurement costs and waste disposal costs. We call these costs R&D Support costs. Generally these costs represent approximately 17% to 22% of our total annual research and development expenses.

R&D Support costs for the year ended December 31, 2003 totaled $25.2 million, compared to $21.6 million for 2002. The increase of $3.6 million was primarily due to increases in our research and development efforts related to our efforts to prepare for the manufacture and commercialization of Affinitak in the first quarter of 2003, our increased efforts related to our government contracts and increased depreciation resulting from the completion of laboratory build-outs. Our planned expense reductions initiated in April 2003 offset, in part, the increases in our research and development efforts noted above.

General and administrative expenses include corporate costs required to support our company, our employees and our stockholders. These costs include personnel and outside costs in the areas of business development, legal, human resources, investor relations and finance. Additionally, we include in general and administrative expenses such costs as rent, repair and maintenance of buildings and equipment, depreciation, utilities, information technology and procurement costs that we need to support the corporate functions listed above.

General and administrative expenses for the year ended December 31, 2003 totaled $9.3 million compared to $8.5 million for 2002. This $800,000 increase was primarily a result of an increase in employees and related benefits in the first quarter of 2003. A reduction in the number of employees and a corresponding reduction in expense as a result of the restructuring in April 2003 offset, in part, the increases noted above. We expect general and administrative expenses to increase in 2004 to support our ongoing R&D activities.

Compensation expense for the year ended December 31, 2003 was $913,000, compared to compensation benefit of $3.0 million for the year ended December 31, 2002. Compensation expense for 2003 primarily consisted of expense for stock options associated with the employee stock option exchange program we initiated in April 2003. We accounted for options affected by the 2003 option exchange program as variable stock options in accordance with Accounting Principles Board Opinion No. 25, or APB 25, and Financial Accounting Standards Board Interpretation No. 44, or FIN 44. APB 25 and FIN 44 required us to account for these exchange options as variable stock options. Variable stock options can result in significant increases and decreases in compensation expense, as a result of the variability of our stock price. We also recorded nominal expense in 2003 related to stock options granted in prior years to consultants, and we accounted for these options in accordance with Emerging Issues Task Force Abstract No. 96-18, or EITF 96-18. The compensation benefit in 2002 represented the reversal of previously recorded compensation expense for stock options accounted for as variable stock options associated with the option exchange program we offered to non-officer employees in January 2000. At December 31, 2002, either employees exercised or we had cancelled all of the variable stock options associated with the 2000 option exchange program.

In April 2003, we initiated a restructuring in response to disappointing results from the first Phase III trial of Affinitak. As a result, we had a small reduction in our workforce, which primarily represented positions that were in support of the commercialization and manufacture of Affinitak. Consequently, we incurred a restructuring charge of $1.8 million during the second quarter of 2003. Isis completed the utilization of the reserve related to this restructuring in the fourth quarter of 2003.

In November 2002, we announced the termination of our GeneTrove database product offering and the reorganization of our GeneTrove program. As a result, we reduced our workforce by approximately 25 people. The restructuring plan included the write-down of certain intellectual property valued at $605,000. As a result of this plan, we incurred a one-time charge of $1.4 million for restructuring activities in the fourth quarter of 2002. We did not recognize any additional GeneTrove restructuring related charges for the year ended December 31, 2003, and completed utilization of the reserve related to the GeneTrove restructuring in the fourth quarter of 2003.

In late 2002, Elan concluded its participation in the Orasense and HepaSense collaborations in conjunction with its restructuring efforts. As a result, we regained all rights to ISIS 104838, the compound that Elan and we were developing within Orasense. We continue to develop certain of our oral delivery technology within Orasense. We also regained all rights to ISIS 14803, the compound that Elan and we were developing within HepaSense.

We did not have equity in loss of affiliates for the year ended December 31, 2003. As mentioned previously, we used the equity method of accounting for our investments in Orasense and HepaSense in 2002. As a result, we recognized 80.1% of the total loss reported by Orasense and HepaSense as equity in loss of affiliates during 2002. Our equity in loss of affiliates for the year ended December 31, 2002 consisted of $9.5 million for Orasense and $6.5 million for HepaSense.

Investment income for the years ended December 31, 2003 and 2002 was $5.1 million and $8.5 million, respectively. The $3.4 million decrease in investment income in 2003 compared to 2002 was primarily due to our lower cash, cash equivalents and short-term investments balances in 2003 compared to 2002, and a decline in interest rates as a result of then current market conditions.

Interest expense for the year ended December 31, 2003 was $18.7 million compared to $16.6 million for the same period in 2002. The $2.1 million increase in interest expense in 2003 compared to 2002 was primarily due to a higher average debt balance during 2003 than during 2002. The increase in debt compared to 2002 related primarily to additional drawdowns under our debt agreement with Lilly. A decrease in the average interest rate on our debt offset, in part, the effect of a higher average debt balance. The decrease in the average debt interest rate was primarily due to the retirement in May 2002 and July 2002 of higher interest rate debt with proceeds from the issuance, in May 2002, of our 5¹/2% convertible notes due in 2009 and to the retirement, in the fourth quarter of 2003, of higher interest rate debt with proceeds from our $32.0 million term loan from Silicon Valley Bank secured in December 2003. The debt retired in the fourth quarter of 2003 consisted of convertible partner debt that carried interest rates ranging from 8.5% to 12%. The new term loan bears interest at the prime

rate, which was 4.0% at December 31, 2003. We can convert the interest rate to a fixed interest rate at our option at any time at the then-applicable prime rate plus 1.25%.

In 2003, $7.5 million of the $18.7 million in interest expense did not require cash payment. This represents the accrual of interest expense related to the $100.0 million loan Lilly has made available to us to fund the research collaboration.

For the year ended December 31, 2002, we reported a $2.3 million loss on prepayment of debt, which represented amounts related to unamortized issuance costs, unamortized warrants and prepaid interest, on the prepayment of approximately $74.0 million of our 14% Senior Subordinated Notes. No such prepayment occurred in 2003.

In July 2002, we used $14.7 million in cash to prepay $19.7 million of 12% convertible debt Elan held. As a result, we reported a $5.0 million gain on prepayment of debt for the year ended December 31, 2002. No such prepayment occurred in 2003.

For the years ended December 31, 2003 and 2002, we reported a net loss of $95.0 million and $72.2 million, respectively. Our net loss applicable to common stock was $95.7 million for the year ended December 31, 2003, and $73.3 million for 2002, included $694,000 and $1.1 million, respectively, of accreted dividends on preferred stock. The decrease in accreted dividends in 2003 from 2002 was the result of the August 2002 conversion of 120,150 shares of Series A Convertible Preferred Stock into 656,674 shares of our common stock using a conversion price of $12.54 per share. Included in the conversion was approximately $2.1 million in preferred stock dividends accrued in prior years. The increase in the net loss applicable to common stock was primarily a result of the increase in loss from operations, a non-cash loss on investments of $2.4 million related to the other-than-temporary impairment of our investments in ATL and Hybridon and the absence of a net gain on debt extinguishment recorded in 2002.

At December 31, 2003, our net operating loss carryforward for federal income tax purposes was approximately $312.1 million. The net operating loss, research credit carryforwards, and capitalized research expense make up the majority of our deferred tax assets. We will use the net operating loss and research credits, and realize the benefit of these deferred tax assets, if we become profitable. We fully reserved all of our deferred tax assets, as their realization is uncertain. Our research credit carryforward and capitalized research expense for federal income tax purposes was approximately $21.9 million and $57.0 million, respectively, as of December 31, 2003. Our federal net operating loss and research credit carryforwards will begin expiring in 2007 and 2004, respectively, unless previously utilized. Our net operating loss and tax credit carryforwards will be subject to an annual limitation regarding utilization against taxable income in future periods, due to "change of ownership" provisions of the Tax Reform Act of 1986. We believe that such limitation will not have a material adverse impact on the benefits that may arise from our net operating loss and tax credit carryforwards. However, there may be additional limitations arising from any future changes in ownership that may have a material adverse impact on us.

Total revenue for the year ended December 31, 2002 was $80.2 million, compared to $53.3 million for 2001. The increase of $26.9 million was primarily a result of increased research and development revenue under collaborative agreements. The most significant contributor was our strategic alliance with Lilly. A decrease in revenue from licensing and royalty revenue in 2002 from that reported in 2001 partially offset the increase in total revenue.

Under the category research and development revenue under collaborative agreements, for the year ended December 31, 2002, we earned $67.8 million, compared to $40.4 million for 2001. The increase of $27.4 million was a result of several collaborations in place during 2002 which were in effect for only a part of 2001 or not in place during 2001. Our Lilly alliance, which we entered into in August 2001, significantly contributed to the increase in 2002 over 2001. Other sources of revenue present in 2002 but only in part or not at all in 2001 included certain GeneTrove agreements and our Ibis biological warfare defense research programs with DARPA and USAMRIID. In addition, during 2002 we earned a milestone under each of our antisense drug discovery collaborations with Amgen and Merck. The June 2002 termination of our Ibis collaboration with Pfizer and a related decrease in earned milestones in 2002 compared to 2001 partially offset the increase in research and development revenue under collaborative agreements.

Research and development revenue from affiliates consisted of revenue associated with our two collaborations with Elan, Orasense and HepaSense. Elan concluded its participation in the HepaSense and Orasense collaborations in late 2002, in conjunction with its restructuring efforts. During 2002, we recognized $8.9 million and $3.0 million from Orasense and HepaSense, respectively, as revenue. During 2001, we recognized $5.4 million and $5.2 million as revenue from Orasense and HepaSense, respectively. The increase in revenue from the Orasense collaboration was primarily due to the progression of ISIS 104838 into later stages of clinical development. The decrease in revenue from the HepaSense collaboration was primarily due to the conclusion of the collaboration during the second half of 2002.

Our revenue from licensing activities and royalties was $417,000 for the year ended December 31, 2002, compared with $2.3 million in 2001. The decrease of $1.9 million was primarily related to one-time revenue recorded in 2001 associated with a $2.0 million license fee paid to us by Eyetech Pharmaceuticals, Inc., a publicly traded company.

Total operating expenses were $131.0 million and $99.4 million for the years ended December 31, 2002 and 2001, respectively. The increase of $31.6 million was primarily due to a $40.3 million increase in research and development activities to support the various products we had in development during 2002, and our research collaborations with Lilly and Amgen. In addition, we had $1.4 million in expenses related to restructuring activities during 2002. A decrease of $2.5 million in general and administrative expenses, and a $7.6 million decrease in compensation related to stock options, offset the increase in operating expenses.

For the year ended December 31, 2002, we reported total research and development expenditures of $124.1 million, compared to $83.7 million reported in 2001. The $40.4 million increase in 2002 over 2001 was primarily due to our investment in the various products we had in development during 2002, including three Phase III clinical trials, costs associated with our Lilly and Amgen research collaborations, and costs associated with increased gene functionalization and target validation activities in support of our numerous GeneTrove collaborations.

Our research and development expenses consisted of costs for antisense drug discovery, including GeneTrove, antisense drug development, our Ibis program and related R&D Support costs.

Antisense drug discovery costs for the year ended December 31, 2002 totaled $38.9 million compared to $20.9 million for 2001. The increase was principally a result of a full year's activity related to the Lilly research collaboration, the September 2002 expansion of the Lilly research collaboration to include oncology, and the advancement of research related to our Amgen research collaboration.

Development expenditures totaled $55.3 million and $37.9 million for the years ended December 31, 2002 and 2001, respectively. The increase of $17.4 million reflected the expansion and advancement of our pipeline. During 2002, we had 13 products in development including two products, Affinitak and alicaforsen for Crohn's disease, in Phase III clinical trials and six products in Phase II clinical trials. Expenditures related to Affinitak in 2002 totaled $22.6 million, compared to $11.5 million in 2001. The increase of $11.1 million in 2002 over 2001 was a result of delivery of Affinitak drug to Lilly for use in clinical trials, expenses for our Phase III trial of Affinitak and the advancement of our various Phase II trials for Affinitak.

Our second drug in Phase III clinical trials, alicaforsen for Crohn's disease, had development expenditures totaling $4.8 million for the year ended December 31, 2002, compared to $3.4 million for the same period of 2001. The increase of $1.4 million for the year ended December 31, 2002 over the same period in 2001 was a result of our two Phase III trials initiated in November 2001 and June 2002, in the United States and Europe, respectively.

Expenditures related to our other products in development totaled $23.1 million in 2002 compared to $16.2 million in 2001. The increase of $6.9 million in 2002 over 2001 was a result of the development of the other antisense products in our pipeline, including expenses related to the initiation of our Phase II clinical trials for alicaforsen in ulcerative colitis, ISIS 14803 and ISIS 104838 in 2002.

Ibis expenditures for the year ended December 31, 2002 totaled $8.3 million, compared to $6.5 million in 2001. The increase of $1.8 million was primarily related to Ibis' performance obligations under its multi-year government contracts with DARPA, awarded in October 2001, and USAMRIID, awarded in March 2002.

R&D Support costs for fiscal year 2002 totaled $21.6 million, compared to $18.4 million for 2001. The increase of $3.2 million was a direct result of increases in our research and development efforts. While we work to control R&D Support costs, we expect that they will increase as direct research and development costs increase.

General and administrative expenses for the year ended December 31, 2002 totaled $8.5 million, compared to $11.1 million for the same period of 2001. This $2.6 million decrease was the result of certain costs previously included in general and administrative expenses, which we determined more accurately reflected research and development efforts.

In 2002, we reversed $3.0 million of expense previously recorded in 2001 related to variable stock options. For the year ended December 31, 2001, we reported $4.6 million as compensation expense related to stock options. The compensation related to stock options was primarily a result of an option exchange program we offered to non-officer employees in January 2000. Accounting Principles Board Opinion No. 25 and Financial Accounting Standard Board Interpretation No. 44 required us to account for these exchange options as variable stock options. Variable stock options can result in significant increases and decreases in compensation expense, as a result of the variability of our stock price. The decrease of $7.6 million in 2002 over 2001 was primarily a result of the 70% decrease in the price of our common stock as of December 31, 2002 compared to its price at December 31, 2001. In addition, we account for stock options granted to consultants in accordance with EITF 96-18, which contributed nominally to the expense recorded in 2002 and 2001. At December 31, 2002, either employees exercised or we cancelled all of the options associated with the 2000 option exchange program.

In November 2002, we announced the termination of the GeneTrove database product offering and the reorganization of the GeneTrove program. As a result, we reduced our workforce by approximately 25 people. The restructuring plan included the write-down of certain intellectual property valued at $597,000. As of December 31, 2002, we incurred a charge of $1.4 million for restructuring activities. There were no restructuring charges in 2001.

Equity in loss of affiliates for the year ended December 31, 2002 was $16.0 million compared to $18.8 million for the year ended December 31, 2001. As mentioned previously, we used the equity method of accounting for our investments in Orasense and HepaSense. As a result, we recognized 80.1% of the total loss reported by Orasense and HepaSense as equity in loss of affiliates. As of December 31, 2002, our equity in loss of affiliates from Orasense and HepaSense totaled $9.5 million and $6.5 million, respectively. In comparison, our equity in loss of affiliates as of December 31, 2001 was $10.3 million for Orasense and $8.3 million for HepaSense.

For the years ended December 31, 2002 and 2001, investment income was $8.5 million and $6.4 million, respectively. Although our ending cash balance decreased slightly in 2002 compared to 2001, our average cash balance was significantly higher, which favorably affected our investment income. In addition, in 2001 we realized a loss in an equity investment of approximately $515,000, which reduced our investment income for 2001.

Interest expense for the year ended December 31, 2002 was $16.6 million compared to $15.2 million for the same period in 2001. Interest expense increased by $1.4 million in 2002 over 2001, primarily as a result of interest accrued on the May 1, 2002 issuance of $125.0 million of 5% convertible subordinated notes and increased borrowings under the $100.0 million loan that Lilly made available to us to fund the research collaboration. The reduction in interest expense from the prepayment in May 2002 of our 14% Senior Subordinated Notes and the prepayment in July 2002 of a

portion of our Elan convertible notes for Orasense and HepaSense offset the aforementioned increases in interest expense.

In 2002, $8.6 million of the $16.6 million in interest expense, which was accrued under various long-term debt agreements, did not require cash payments. The long-term debt agreements with deferred interest and principal payments included our $100.0 million loan from Lilly and the remaining portion of our Elan line of credit for Orasense.

For the year ended December 31, 2002 and 2001, we reported a net loss of $72.2 million and $73.8 million, respectively. Our net loss applicable to common stock was $73.3 million for the year ended December 31, 2002, and $75.1 million in 2001, and included $1.1 million and $1.3 million of accreted dividends on preferred stock as of December 31, 2002 and 2001, respectively. The decrease in accreted dividends in 2002 from 2001 was the result of the August 2002 conversion of 120,150 shares of Series A Convertible Preferred Stock into 656,674 shares of our common stock using a conversion price of $12.54 per share. We included approximately $2.1 million in preferred stock dividends accrued in prior years in the conversion.

We have financed our operations with revenue from research and development under collaborative agreements and from affiliates. Additionally, we have earned licensing and royalty revenue from the sale or licensing of our intellectual property. We have also financed our operations through the sale of our equity securities and the issuance of long-term debt. From our inception through December 31, 2003, we have approximately $400.5 million in revenue from contract research and development and the sale and licensing of our intellectual property. Since we were founded, we have raised net proceeds of approximately $589.1 million from the sale of equity securities. We have borrowed approximately $357.6 million under long-term debt arrangements to finance a portion of our operations of which $250.6 million remains outstanding at December 31, 2003.

At December 31, 2003, we had cash, cash equivalents and short-term investments of $215.5 million and working capital of $194.0 million. In comparison, we had cash, cash equivalents and short-term investments of $289.4 million and working capital of $244.2 million as of December 31, 2002. The decrease in our cash, cash equivalents and short-term investments, and working capital was primarily due to cash used to fund operating activities, capital purchases, additions to our patent portfolio, and repayment of $25.6 million in debt, including principal and accrued interest, to Elan, Abbott Laboratories, and Boehringer Ingleheim International BmbH, or BI, for $8.1 million, $1.0 million, and $16.5 million, respectively. Drawdowns in 2003 totaling $26.3 million from the $100.0 million loan provided by Lilly to fund our joint research collaboration, combined with net proceeds received from a $32.0 million term loan we secured from Silicon Valley Bank in December 2003 partially offset the decrease in cash. We expect that during 2004, we will finance our operations primarily from payments

generated from collaborative agreements and $21.3 million in drawdowns on the $100.0 million interest-free loan Lilly made available to us to fund our joint research collaboration.

As of December 31, 2003, our debt and other obligations totaled $250.6 million, compared to $233.8 million at December 31, 2002. Our debt and other obligations include current and long-term deferred contract revenue and contractual obligations that represent our payment obligations. The increase in our debt and other obligations was primarily due to additional draw downs from the $100.0 million interest-free loan from Lilly, which we discount to their present value by imputing interest on the amount at 20% and accrete to their face value over their terms by recording interest expense, and by securing a $32.0 million term loan from Silicon Valley Bank. We used the proceeds from the term loan to retire convertible partner debt from Elan and BI. The retired convertible partner debt was due from 2003 to 2005 and carried interest rates from 8.5% to 12%. The term loan bears interest at the prime rate, which was 4.0% at December 31, 2003. We can convert the interest to a fixed rate at our option at any time at the then-applicable prime rate plus 1.25%. The increase in our debt and other obligations in 2003 was offset by Lilly's waiver on repayment of the $21.2 million manufacturing loan it provided to us to build the Affinitak manufacturing facility, of which we had $15.4 million outstanding at December 31, 2002. In addition, we repaid principal and interest related to certain of our partner debt and certain of our capital leases. We expect that capital lease obligations will increase over time to fund capital equipment acquisitions required for our growing business. We will continue to use lease financing to maintain our working capital as long as the terms remain commercially attractive. Based on our current operating plan with reasonable assumptions for new sources of revenue and cash, we believe that our available cash, cash equivalents and short-term investments as of December 31, 2003, when combined with investment income and committed contractual cash payments from our partners, will be sufficient to meet our anticipated requirements for at least the next 36 months. Due to the uncertainties in our business discussed in this section, and elsewhere in this report, including "Risk Factors" beginning on page 26, this may not be the case. In addition, we may choose to, or prevailing business conditions may require us to, obtain additional financing from time to time. We may choose to raise additional funds through public or private financing, licensing and contractual agreements or other arrangements. Additional funding, if needed, may not be available on terms favorable to us. Furthermore, any additional equity financing may be dilutive to our shareholders, and debt financing, if available, may involve restrictive covenants. If we choose to obtain funding through licensing and other contractual agreements, those agreements may require us to relinquish our rights to certain of our technologies or products. Our failure to raise capital when needed would harm our business, financial condition and results of operations.

The following table summarizes our contractual obligations as of December 31, 2003. The table provides a breakdown of when obligations become due. We provide a more detailed description of the major components of our debt in the paragraphs following the table:

	Payments Due by Period (in millions)
Contractual Obligations (selected balances described below)	Total		Less than 1 year		1-3 years	3-5 years		After 5 years
Standard Operating Debt	$	38.4		$	6.2	$	18.6		$	13.6	$	—
Convertible Partner Debt	$	80.1		$	6.4	$	73.7		$	—	$	—
5¹/2% Convertible Subordinated Notes	$	125.0		$	—	$	—		$	—	$	125.0
Capital Leases and Other Obligations	$	7.1		$	3.9	$	3.2		$	—	$	—
Operating Leases	$	12.3		$	2.7	$	5.6		$	2.7	$	1.3

Our contractual obligations consist primarily of partner debt and publicly traded convertible debt that we can repay on favorable terms with equity. In addition, we also have standard operating debt, capital leases and other obligations. Convertible partner debt at December 31, 2003 included: 1) the interest-free loan Lilly made available to us to fund the joint research collaboration and 2) the remaining portion of the convertible debt provided by BI, associated with the collaborative agreement

between the two companies that we repaid in January 2004. Our standard operating debt includes a $32.0 million term loan from Silicon Valley Bank. Capital leases and other obligations consist of $6.3 million and $860,000 for capital leases and other obligations, respectively.

In December 2003, we secured a $32.0 million term loan from Silicon Valley Bank to retire our existing debt to BI and Elan. We amortize the term loan over sixty months. The term loan requires equal monthly payments of principal plus accrued interest, and bears interest at the prime interest rate, which was 4.0% at December 31, 2003. The loan is secured by substantially all of our operating assets, excluding intellectual property, real estate, and certain equity investments. The loan is subject to certain liquidity requirements, including a requirement that we maintain a minimum balance in an account at Silicon Valley Bank at all times equal to the outstanding balance of the loan. The loan is convertible to a fixed interest rate at our option at any time at the then-applicable prime rate plus 1.25%. We used the proceeds from the loan to pay off existing debt to Elan of $5.1 million plus accrued interest and to BI of $22.6 million plus accrued interest, of which $6.4 million plus accrued interest was paid off subsequent to December 31, 2003. The carrying value of the term loan at December 31, 2003 was $32.0 million.

In August 2001, Lilly made available to us a $100.0 million interest-free loan to fund the joint research collaboration between the two companies. The loan is interest-free and is repayable, at our option, in cash or common stock at $40 per share at the end of four years. The term of the loan provides for quarterly draw downs by us. As of December 31, 2003, we had drawn down $73.8 million of the $100.0 million available. We are accounting for this loan using an imputed interest rate of 20%, consistent with market conditions in place at the time the loan was agreed to. We carry the net present value of the drawdowns as a long-term obligation and record interest expense over the term of the loan. The difference between the cash received and the present value of the loan represents value given to us by Lilly to help fund the research collaboration, and we account for the difference as deferred revenue related to the collaboration and recognize it as revenue over the period of performance. As of December 31, 2003, the balance in long-term obligations was $53.0 million and the balance in deferred revenue was $20.8 million.

During 1999 and 2000, in conjunction with the HepaSense and Orasense joint ventures, Elan made available to us credit facilities of up to $30.4 million, evidenced by convertible promissory notes. The terms of these notes provided for interest at 12% annually, with maturities through January 2006. The notes were convertible into our common stock at anytime by either party, at a rate determined by the underlying agreements. In July 2002, we prepaid $19.7 million of the then-outstanding debt with $14.7 million in cash, resulting in a gain of approximately $5.0 million for the year ended December 31, 2002. As of December 31, 2002, there was $7.2 million outstanding under the Orasense credit facility. We used the proceeds from our term loan from Silicon Valley Bank which we secured in December 2003, to repay these borrowings in full. We can no longer borrow funds against the Orasense and HepaSense credit facilities.

During 1997 and 1996, we borrowed a total of $22.6 million under a $40.0 million line of credit made available pursuant to the terms of our collaborative agreement with BI. Borrowings under the line of credit bore interest at the seven year U.S. interbanking rate plus 2.0%, determined at the time each advance was made, and ranged from 8.4% to 8.5%. Principal borrowings were repayable in cash or our common stock, at our option. We repaid $16.2 million, plus interest in the fourth quarter of 2003, and we repaid the remaining principal installment of $6.4 million, plus interest, in January 2004, using the proceeds from our Silicon Valley Bank term loan.

In May 2002, we completed a $125.0 million convertible debt offering, which raised proceeds of approximately $120.9 million, net of $4.1 million in issuance costs. The subordinated notes bear interest at 5¹/2%, which is payable semi-annually, and mature in May 2009. Upon maturity, holders of the subordinated notes can convert the notes into shares of common stock at a conversion price of $16.625 per share. At December 31, 2003 and 2002, the principal outstanding on the notes was $125.0 million.

In addition to contractual obligations, we had outstanding purchase orders as of December 31, 2003 for the purchase of services, capital equipment and materials as part of our normal course of business.

We plan to continue to make expenditures to expand our research and development activities including our preclinical and clinical product development. We plan to continue to enter into more collaborations with partners to provide for additional revenue to us and we may be required to incur additional cash expenditures related to our obligations under many of the new agreements we may enter into. For example, we recently made a $10 million equity investment in Alnylam as part of our strategic alliance with them. We currently intend to use our cash and short-term equivalents to finance our activities. However, we may also pursue other financing alternatives, like issuing additional shares of our common stock, issuing debt instruments, refinancing our existing debt, or securing lines of credit. Whether we use our existing capital resources or choose to obtain financing will depend on various factors, including the future success of our business, the prevailing interest rate environment and the condition of financial markets generally.

In March 2004, we entered into a two-year contract with SAIC to further the development of our TIGER biosensor to identify infectious agents in biological warfare attacks. The contract provides for up to $19.5 million in funding by DARPA.

In March 2004, we entered into a strategic alliance with Alnylam to develop and commercialize RNAi therapeutics. Under the terms of the agreement, we exclusively licensed to Alnylam our patent estate relating to antisense motifs and mechanisms and oligonucleotide chemistry for double-stranded RNAi therapeutics in exchange for a $5 million technology access fee, participation in fees for Alnylam's partnering programs, as well as future milestone and royalty payments. As part of our agreement, we also made a $10.0 million equity investment in Alnylam. In turn, Alnylam nonexclusively licensed us its patent estate relating to antisense motifs and mechanisms and oligonucleotide chemistry to research, develop and commercialize single-stranded RNAi therapeutics and to research double stranded-RNAi compounds. We also received a license to develop and commercialize double-stranded RNAi drugs targeting a limited number of therapeutic targets on either an exclusive or co-exclusive basis depending on the target. If we develop or commercialize an RNAi-based drug using Alnylam's technology, we will pay Alnylam milestones and royalties.

We are exposed to changes in interest rates primarily from our investments in certain securities classified as short-term investments and, secondarily, from our long-term debt arrangements. We invest our excess cash in highly liquid short-term investments that are typically held for the duration of the term of the respective instrument. We do not utilize derivative financial instruments, derivative commodity instruments or other market risk sensitive instruments, positions or transactions to manage exposure to interest rate changes. Accordingly, we believe that, while the securities we hold are subject to changes in the financial standing of the issuer of such securities, we are not subject to any material risks arising from changes in interest rates, foreign currency exchange rates, commodity prices, equity prices or other market changes that affect market risk sensitive instruments.

We filed our consolidated financial statements and supplementary data required by this item as exhibits hereto, and listed them under Item 15(a)(1) and (2), and incorporated them herein by reference.

Item 9. Changes in and Disagreements with Accountants on Accounting and Financial Disclosure.

There has been no reported disagreements on any matter of accounting principles or procedures or financial statement disclosure in 2003 with our Independent Auditors.

As of the end of the period covered by this Annual Report on Form 10-K, we carried out an evaluation, under the supervision and with the participation of our management, including our Chief Executive Officer and Chief Financial Officer, of the effectiveness of the design and operation of our disclosure controls and procedures, as defined in Rules 13a-15(e) and 15d-15(e) under the Securities Exchange Act of 1934, as amended. Based on this evaluation, our Chief Executive Officer and Chief Financial Officer concluded that our disclosure controls and procedures were effective as of the end of the year ended December 31, 2003.

An evaluation was also performed under the supervision and with the participation of our management, including our Chief Executive Officer and Chief Financial Officer, of any change in our internal control over financial reporting that occurred during our last fiscal quarter and that has materially affected, or is reasonably likely to materially affect, our internal control over financial reporting. That evaluation did not identify any change in our internal control over financial reporting that occurred during our latest fiscal quarter and that has materially affected, or is reasonably likely to materially affect, our internal control over financial reporting.

We maintain disclosure controls and procedures that are designed to ensure that information required to be disclosed in our Exchange Act reports is recorded, processed, summarized and reported within the time periods specified in the SEC's rules and forms, and that such information is accumulated and communicated to our management, including our Chief Executive Officer and Chief Financial Officer, as appropriate, to allow timely decisions regarding required disclosure. In designing and evaluating the disclosure controls and procedures, management recognizes that any controls and procedures, no matter how well designed and operated, can provide only reasonable and not absolute assurance of achieving the desired control objectives.

We incorporate by reference the information required by this Item with respect to Directors and Audit Committee financial expert by reference from the information under the caption "Election of Directors" and "Audit Committee", respectively, contained in our definitive Proxy Statement (the "Proxy Statement"), which we will file on or about April 15, 2004 with the Securities and Exchange Commission in connection with the solicitation of proxies for our 2003 Annual Meeting of stockholders to be held on May 26, 2004.

We incorporate by reference the required information concerning our Code of Ethics from the information under the caption "Code of Ethics" contained in the Proxy Statement. We have filed our Code of Ethics as an exhibit to this Report on Form 10-K.

Item 1, Part I of this Report contains the required information concerning our Executive Officers. We incorporate by reference the information required by this Item concerning compliance with Section 16(a) of the Securities Exchange Act of 1934, as amended, from the information under the caption "Section 16(a) Beneficial Ownership Reporting Compliance" contained in the Proxy Statement.

We incorporate by reference the information required by this item to the information under the caption "Executive Compensation" and "Compensation Committee Interlock and Insider Participation" contained in the Proxy Statement.

We incorporate by reference the information required by this item to the information under the captions "Security Ownership of Certain Beneficial Owners and Management" and "Equity Compensation Plan Information" contained in the Proxy Statement.

We incorporate by reference the information required by this item to the information under the caption "Certain Relationships and Related Transactions" contained in the Proxy Statement.

We incorporate by reference the information required by this item to the information under the caption "Ratification of Selection of Independent Auditors" contained in the Proxy Statement.

We submitted the consolidated financial statements required by this item in a separate section beginning on page F-1 of this Report.

We omitted these schedules because they are not required, or are not applicable, or the required information is shown in the consolidated financial statements or notes thereto.

On November 4, 2003, we filed a report on Form 8-K for the announcement of our third quarter results and the related press release dated November 4, 2003. We furnished this information under Item 12 of Form 8-K, "Results of Operations and Financial Condition."

On January 5, 2004, we filed a report on Form 8-K for the announcement of data from a Phase 2 clinical trial which demonstrate that ISIS 104838, and antisense TNF-alpha inhibitor, produced a statistically significant disease response in patients with rheumatoid arthritis.

On February 10, 2004, we filed a report on Form 8-K for the announcement of our financial results for the year ended December 31, 2003 and the related press release dated February 10, 2004. We furnished this information under Item 12 of Form 8-K, "Results of Operations and Financial Condition."

Pursuant to the requirements of Section 13 or 15(d) of the Securities Exchange Act of 1934, the Registrant has duly caused this report on Form 10-K to be signed on its behalf by the undersigned, thereunto duly authorized on the 12th day of March, 2004.

KNOW ALL MEN BY THESE PRESENTS, that each person whose signature appears below constitutes and appoints Stanley T. Crooke and B. Lynne Parshall, or any of them, his or her attorney-in-fact, each with the power of substitution, for him or her in any and all capacities, to sign any amendments to this Report, and to file the same, with exhibits thereto and other documents in connection therewith, with the Securities and Exchange Commission, hereby ratifying and confirming all that each of said attorneys-in-fact, or his or her substitute or substitutes, may do or cause to be done by virtue hereof.

Pursuant to the requirements of the Securities Exchange Act of 1934, this report has been signed below by the following persons on behalf of the Registrant and in the capacities and on the dates indicated.

Signatures	Title	Date
/s/ STANLEY T. CROOKE, M.D., PH.D. Stanley T. Crooke, M.D., Ph.D.	Chairman of the Board, President, and Chief Executive Officer (Principal executive officer)	March 12, 2004
/s/ B. LYNNE PARSHALL, J.D. B. Lynne Parshall, J.D.	Director, Executive Vice President, Chief Financial Officer and Secretary (Principal financial and accounting officer)	March 12, 2004
/s/ SPENCER R. BERTHELSEN, M.D. Spencer R. Berthelsen, M.D.	Director	March 12, 2004
/s/ CHRISTOPHER F. O. GABRIELI Christopher F. O. Gabrieli	Director	March 12, 2004
/s/ FREDERICK T. MUTO, J.D. Frederick T. Muto	Director	March 12, 2004
/s/ JOHN C. REED, M.D., PH.D John C. Reed, M.D., Ph.D.	Director	March 12, 2004

Exhibit Number		Description of Document
3.1		Amended and Restated Certificate of Incorporation filed June 19, 1991.(1)
3.2		Certificate of Amendment to Restated Certificate of Incorporation filed April 9, 2001.(19)
3.3		Bylaws.(19)
4.1		Certificate of Designation of the Series A Convertible Preferred Stock.(11)
4.2		Certificate of Designation of the Series B Convertible Preferred Stock.(14)
4.3		Certificate of Designation of the Series C Junior Participating Preferred Stock.(17)
4.4		Specimen Common Stock Certificate.(1)
4.5		Specimen Series A Preferred Stock Certificate.(18)
4.6		Specimen Series B Preferred Stock Certificate.(18)
4.7		Form of Right Certificate.(17)
4.8		Purchase Agreement between the Registrant and Reliance Insurance Company for 14% Senior Subordinated Discount Notes due November 1, 2007 and Warrants for Common Stock dated October 24, 1997 (with certain confidential information deleted).(6)
4.9		First Supplement to Purchase Agreement between the Registrant and Reliance Insurance Company for 14% Senior Subordinated Discount Notes due November 1, 2007 and Warrants for Common Stock dated May 1, 1998 (with certain confidential information deleted).(7)
4.11		Subscription, Joint Development and Operating Agreement, dated April 20, 1999 among the Registrant, Elan Corporation, plc, Elan International Services, Ltd. and Orasense Ltd. (with certain confidential information deleted), together with the related Securities Purchase Agreement, Convertible Promissory Note, Warrant to Purchase Shares of Common Stock, Registration Rights Agreement and License Agreements.(12)
4.13		Subscription, Joint Development and Operating Agreement dated January 14, 2000 among the Registrant, Elan Corporation, plc, Elan International Services, Ltd. and HepaSense, Ltd. (with certain confidential information deleted), together with the related Securities Purchase Agreement, Convertible Promissory Note, Warrant to Purchase Shares of Common Stock, Registration Rights Agreement and License Agreements.(14)
4.14		Securities Purchase Agreement, dated August 17, 2001, between the Registrant and Eli Lilly and Company.(20)
4.15		Registration Rights and Standstill Agreement, dated August 17, 2001, between the Registrant and Eli Lilly and Company.(20)
4.16		Loan Agreement, dated August 17, 2001, between the Registrant and Eli Lilly and Company.(20)
4.17		Registration Rights Agreement, dated May 1, 2002, among the Registrant, UBS Warburg LLC, Robertson Stephens, Inc., Needham & Company, Inc., and Roth Capital Partners, LLC.(16)
4.18		Indenture, dated as of May 1, 2002, between the Registrant and Wells Fargo Bank Minnesota, National Association, as Trustee, with respect to the $125,000,000 5¹/2% Convertible Subordinated Notes due 2009.(16)

		Page
Report of Ernst & Young LLP, Independent Auditors		F-2
Consolidated Balance Sheets at December 31, 2003 and 2002		F-3
Consolidated Statements of Operations for the years ended December 31, 2003, 2002 and 2001		F-4
Consolidated Statements of Stockholders' Equity for the years ended December 31, 2003, 2002 and 2001		F-5
Consolidated Statements of Cash Flows for the years ended December 31, 2003, 2002 and 2001		F-6
Notes to Consolidated Financial Statements		F-7

We have audited the accompanying consolidated balance sheets of Isis Pharmaceuticals, Inc. as of December 31, 2003 and 2002, and the related consolidated statements of operations, consolidated stockholders' equity, and consolidated cash flows for each of the three years in the period ended December 31, 2003. These consolidated financial statements are the responsibility of the Company's management. Our responsibility is to express an opinion on these consolidated financial statements based on our audits.

We conducted our audits in accordance with auditing standards generally accepted in the United States. Those standards require that we plan and perform the audit to obtain reasonable assurance about whether the consolidated financial statements are free of material misstatement. An audit includes examining, on a test basis, evidence supporting the amounts and disclosures in the consolidated financial statements. An audit also includes assessing the accounting principles used and significant estimates made by management, as well as evaluating the overall consolidated financial statement presentation. We believe that our audits provide a reasonable basis for our opinion.

In our opinion, the consolidated financial statements referred to above present fairly, in all material respects, the consolidated financial position of Isis Pharmaceuticals, Inc. at December 31, 2003 and 2002, and the consolidated results of its operations and its cash flows for each of the three years in the period ended December 31, 2003, in conformity with accounting principles generally accepted in the United States.

San Diego, California
January 26, 2004, except for Note 12, as to which the date is March 11, 2004.

ISIS PHARMACEUTICALS, INC.

CONSOLIDATED BALANCE SHEETS

(in thousands, except share data)

ISIS PHARMACEUTICALS, INC.

CONSOLIDATED STATEMENTS OF OPERATIONS

(in thousands, except for per share amounts)

ISIS PHARMACEUTICALS, INC.
CONSOLIDATED STATEMENTS OF STOCKHOLDERS' EQUITY
Years Ended December 31, 2003, 2002 and 2001
(in thousands)

ISIS PHARMACEUTICALS, INC.

NOTES TO CONSOLIDATED FINANCIAL STATEMENTS

December 31, 2003

The consolidated financial statements include the accounts of Isis and its wholly-owned subsidiary, Isis Pharmaceuticals Singapore Pte Ltd. There were no current operations or results of operations for the wholly-owned subsidiary for the year ended December 31, 2003.

Isis Pharmaceuticals was incorporated in California on January 10, 1989. In conjunction with its initial public offering, Isis Pharmaceuticals was reorganized as a Delaware corporation, as Isis Pharmaceuticals, Inc. ("Isis" or the "Company"), in April 1991. Isis was organized principally to develop human therapeutic drugs using antisense and combinatorial technology.

Isis follows provisions of Statement of Financial Accounting Standards (SFAS) No. 128 Earnings per Share. Isis computes basic loss per share by dividing the net loss applicable to common stock by the weighted average number of common shares outstanding during the period ("Basic EPS method"). Isis computes diluted earnings (loss) per common share using the weighted-average number of common and dilutive common equivalent shares outstanding during the period ("Diluted EPS method"). Diluted common equivalent shares of 17.1 million at December 31, 2003 consisted of shares issuable upon exercise of stock options, warrants, convertible debt and conversion of preferred stock. As Isis incurred a loss in the years ended December 31, 2003, 2002 and 2001, Isis did not include diluted common equivalent shares in the computation of diluted net loss per share because the effect would be antidilutive.

Contract revenue consists of non-refundable research and development funding and Isis records contract revenue as earned based on the performance requirements of Isis' collaborative research and development contracts. Isis recognizes contract fees for which no further performance obligations exist when Isis receives the payments or when Isis is reasonably certain it can collect the receivable. Isis records payments received in excess of amounts earned as deferred contract revenue. The company expenses research and development costs as incurred. For the years ended December 31, 2003, 2002 and 2001, research and development costs of approximately $30.2 million, $68.3 million and $59.2 million, respectively, were related to collaborative research and development arrangements.

Isis recognizes revenue when Isis satisfies all contractual obligations and Isis is reasonably certain it can collect the receivable.

Isis recognizes research and development revenue under collaborative agreements as it incurs the related expenses, up to contractual limits. Isis defers payments received under these agreements that relate to future performance and records revenue as Isis earns it over the specified future performance period. Isis recognizes revenue that relates to nonrefundable, upfront fees over the period of the

contractual arrangements as Isis satisfies its performance obligations. Isis recognizes revenue that relates to milestones, under existing arrangements, upon completion of the milestone's performance requirement. Isis recognizes revenue from arrangements entered into subsequent to June 30, 2003 in accordance with Emerging Issues Task Force Issue No. 00-21 ("EITF 00-21") Accounting for Revenue Arrangements with Multiple Deliverables. This issue addresses the timing and method of revenue recognition for revenue arrangements that include the delivery of more than one product or service. Isis sometimes enters into revenue arrangements that contain multiple deliverables. In these cases, Isis recognizes revenue from each element of the arrangement as long as Isis can determine a separate value for each element, Isis has completed its obligation to deliver or perform on that element, and Isis is reasonably assured of collecting the resulting receivable. Isis records revenue from federal research grants during the period in which it incurs the related expenditures. Isis recognizes revenue from product sales as it ships the products. Isis has implemented the provisions of Staff Accounting Bulletin No. 104 ("SAB 104"), which was issued in December 2003. SAB 104 updates portions of the interpretive guidance included in Topic 13 of the codification of Staff Accounting Bulletin No. 101 in order to make this interpretive guidance consistent with current authoritative accounting and auditing guidance and SEC rules and regulations. SAB 104 provides interpretation on selected revenue recognition issues and when revenue is properly recognizable. Revenue should not be recognized until it is realized or realizable and earned. It must meet the following criteria: 1) persuasive evidence of an arrangement exists, 2) delivery occurred or services were rendered, 3) the seller's price to the buyer is fixed or determinable and 4) collectibility is reasonably assured. This statement has not had a material impact on the Company's operating results and financial position.

As part of Isis' alliance with Eli Lilly and Company ("Lilly") in August 2001, Lilly provided Isis a $100.0 million interest free loan to fund the research collaboration. As of December 31, 2003, Isis had drawn down $73.8 million on the $100.0 million loan. Isis discounted the $73.8 million loan to its net present value by imputing interest on the amount at 20%, which represented market conditions in place at the time Isis entered into the loan. Isis accretes the loan up to its face value over its term by recording interest expense. The difference between the cash received and the present value of the loan represents value Lilly gave to Isis to help fund the research collaboration. Isis accounts for this value as deferred revenue and recognizes it as revenue over the period of performance.

In late 2002, Isis terminated its HepaSense and Orasense collaborations with Elan Corporation plc ("Elan") and as a result, Isis no longer earns revenue from these collaborations.

Isis recognizes licensing and royalty revenue immediately, if collectibility is reasonably assured, for arrangements in which Isis is not required to provide services in the future.

Financial instruments that potentially subject Isis to concentrations of credit risk consist primarily of cash equivalents, short-term investments and receivables. Isis places its cash equivalents and certain of its short-term investments with high credit-quality financial institutions. Isis invests its excess cash primarily in auction and money market instruments, and municipal and floating rate bonds. Isis and its audit committee established guidelines relative to credit ratings, diversification and maturities that seek to maintain safety and liquidity.

Isis considers all liquid investments with maturities of ninety days or less when purchased to be cash equivalents. Isis' short-term investments have initial maturities of greater than ninety days from date of purchase. Isis classifies its securities as "available-for-sale" in accordance with SFAS 115, Accounting for Certain Investments in Debt and Equity Securities. Isis carries these investments at fair market value with any unrealized gains and losses recorded as a separate component of stockholders' equity. Fair value is based upon market prices quoted on the last day of the fiscal quarter. Isis uses the specific identification method to determine the cost of debt securities sold. Isis includes gross realized gains and losses in investment income and these amounts have not been material. During the first quarter of 2003, Isis recorded a non-cash loss of $2.4 million related to the impairment of its equity investments in Antisense Therapeutics Limited ("ATL") and Hybridon, Inc. ("Hybridon"). This charge reflected the then-current market climate and was associated with the decline in market value of the equity investments from their initial valuations and Isis determined the decline in value to be other-than-temporary. In the second, third and fourth quarters of 2003, Isis recorded net unrealized gains related to its equity investments in ATL and Hybridon as a separate component of stockholders' equity. This reflected the increase in the market value of the investments since the first quarter of 2003.

Isis' inventory includes drugs with alternative uses that are used primarily for its clinical development activities and drug products it manufactures for its partners under contractual terms. Isis states its inventory at the lower of cost or market, with cost determined under the first-in, first-out method. Isis reviews inventory periodically and reduces the carrying value of items considered to be slow moving or obsolete to their estimated net realizable value. In the second quarter of 2003, Isis reduced the carrying value of its raw materials related to Affinitak to zero.

Inventory includes the following categories as of December 31, 2003 and 2002 (net realizable value in thousands):

	December 31,
	2003		2002
Raw materials	$	1,526	$	10,186
Work in process		9,920		904
Finished goods		2,549		—

	$	13,995	$	11,090

The composition of inventory among raw materials, work-in-process and finished goods fluctuates from period-to-period based on the nature and timing of Isis' manufacturing activities in response to product requirements to support clinical trials and partner collaborations.

Property, plant and equipment is stated at cost and consists of the following (in thousands):

	December 31,
	2003			2002
Land	$	1,163		$	1,163
Buildings and improvements		29,859			37,870
Equipment and computer software		49,108			58,274
Furniture and fixtures		3,356			2,224

		83,486			99,531
Less accumulated depreciation		(48,696	)		(40,437	)

	$	34,790		$	59,094

During 2003, Isis reduced to zero approximately $21.2 million in buildings, building improvements and equipment related to the debt repayment waiver Lilly granted Isis for amounts Lilly loaned Isis to fund the construction of a new manufacturing suite dedicated to the commercial production of Affinitak.

Depreciation of property, plant and equipment is provided on the straight-line method over estimated useful lives as follows:

Building		31.5 years
Building improvements		15 years
Manufacturing facilities		10 years
Equipment		5 years
Computer software		3 years
Furniture and fixtures		5 years

Isis obtains licenses from third parties and capitalizes the cost related to exclusive licenses. Isis' license from Hybridon comprises the majority of the license balance as of December 31, 2003 and 2002. Isis amortizes capitalized licenses over their estimated useful life or term of the agreement, which for current licenses is between six years and 15 years. Accumulated amortization related to licenses was $7.5 million and $5.0 million at December 31, 2003 and 2002, respectively. Based on existing licenses, estimated amortization expense related to licenses is $2.5 million for each of the years ending December 31, 2004, 2005, 2006, 2007 and 2008.

Isis capitalizes costs consisting principally of outside legal costs and filing fees related to obtaining patents. Isis reviews costs regularly to determine that they include costs for patent applications Isis is pursuing. Isis evaluates costs related to patents that the company is not actively pursuing for impairment and writes off any patents, if appropriate. Isis amortizes patent costs over their estimated useful lives of 10 years, beginning with the date the patents are issued. The weighted average remaining

life of issued patents was 6.5 years and 7.0 years at December 31, 2003 and 2002, respectively. Accumulated amortization related to patents was $4.4 million and $3.2 million at December 31, 2003 and 2002, respectively. Based on existing patents, estimated amortization expense related to patents is as follows (in thousands):

Years Ending December 31,	Amortization (in thousands)
2004	$	1,207
2005	$	1,191
2006	$	1,137
2007	$	1,024
2008	$	937

Isis uses the equity method of accounting to account for its investments in 50% or less owned companies over which it has the ability to exercise significant influence. Isis also accounted for investments in its joint ventures, Orasense and HepaSense, using the equity method of accounting. At December 31, 2003 and 2002, Isis had the following investments accounted for using the equity method:

In April 1999, Isis and Elan formed Orasense, Ltd., a Bermuda limited company. In January 2000, Isis and Elan formed HepaSense, Ltd., a Bermuda limited company. Each joint venture was owned 80.1% by Isis and 19.9% by Elan. While Isis owned 80.1% of the outstanding common stock of Orasense and HepaSense, throughout the respective terms of the collaborations related to the joint ventures, Elan and its subsidiaries retained significant minority investor rights that were considered "participating rights" as defined in EITF 96-16. Accordingly, Isis accounted for its investment in each joint venture under the equity method of accounting for the years ended December 31, 2002 and 2001. In 2002, Elan concluded its participation in both the Orasense and HepaSense collaborations. See Note 6—Research and Development Collaborative Arrangements and Licensing Agreements for more detail on Isis' investment in Orasense and HepaSense.

Isis has determined the estimated fair value of its financial instruments. The amounts reported for cash, accounts receivable, accounts payable and accrued expenses approximate the fair value because of their short maturities. Isis reports its investment securities at their estimated fair value based on quoted market prices of comparable instruments.

Isis periodically evaluates carrying values of long-lived assets including property, plant and equipment and intangible assets, when events and circumstances indicate that these assets may have been impaired. Isis adopted SFAS 144, Accounting for the Impairment of Long-Lived Assets, and this standard has not had a material impact on the results of operations. For the year ended December 31, 2002, Isis recorded an impairment charge of $605,000 for the write-down of intellectual property related to the restructuring of the GeneTrove program.

The preparation of consolidated financial statements in conformity with accounting principles generally accepted in the United States requires management to make estimates and assumptions that affect the amounts reported in the consolidated financial statements and accompanying notes. Actual results could differ from those estimates.

In January 2000, Isis offered non-officer employees an opportunity to exchange certain of their existing out-of-the-money stock options for new options with exercise prices at the then-current market value. These options are required to be accounted for as variable stock options in accordance with Financial Interpretation No. 44 ("FIN 44"), Accounting for Certain Transactions Involving Stock Compensation—an Interpretation of APB Opinion No. 25. Isis reported the resulting compensation expense in the statement of operations. Variable stock options can result in significant increases and decreases in compensation expense, subject to the variability of Isis' stock price. As of December 31, 2002, optionholders had exchanged all of these options, or the options had expired.

In April 2003, Isis implemented an employee stock option exchange program ("2003 option exchange program") to maintain one of Isis' key assets, its employee base, in a manner that was sensitive to shareholder interests. The 2003 option exchange program allowed employees during the offering period, which began on April 8, 2003 and ended on May 8, 2003, to surrender options, granted prior to January 5, 2002, which had higher exercise prices, in exchange for a lesser number of options, which had lower exercise prices. Employees exchanged 2.2 million options having a weighted-average exercise price of $14.89 for 1.0 million options having an exercise price of $5.15. The new options vest over three years beginning on January 1, 2003 and expire on December 31, 2008. Isis accounts for the affected options, until all these options have been exercised or cancelled, using variable accounting consistent with the provisions of APB 25 and FIN 44. As a result, Isis recorded compensation expense of approximately $886,000 in 2003 and will continue to account for the affected options using variable accounting.

Isis has adopted the disclosure-only provision of SFAS 123, Accounting for Stock-Based Compensation ("SFAS 123"). Accordingly, Isis has not recognized compensation expense, except primarily for compensation expense related to the affected options from the 2000 and 2003 option exchange programs, for the Isis stock option plans and the employee stock purchase plan ("ESPP"). Had Isis determined compensation expense consistent with SFAS 123, Isis would have reported the following proforma amounts for net loss and basic and diluted net loss per share (in thousands, except per share amounts):

	2003			2002			2001
Net loss applicable to common stock—as reported	$	(95,690	)	$	(73,302	)	$	(75,131	)
Net loss applicable to common stock—pro forma	$	(101,099	)	$	(95,329	)	$	(83,830	)
Basic and diluted net loss per share—as reported	$	(1.73	)	$	(1.35	)	$	(1.70	)
Basic and diluted net loss per share—pro forma	$	(1.82	)	$	(1.75	)	$	(1.90	)

For purposes of proforma disclosures, Isis estimated, the fair value of each option grant and ESPP purchase rights on the date of grant using the Black-Scholes option pricing model with the following weighted-average assumptions:

	Stock Options		ESPP
	2003		2002		2001		2003		2002		2001
Risk—free interest rate	4.3	%	3.8	%	4.5	%	4.3	%	3.8	%	4.5	%
Dividend yield	0	%	0	%	0	%	0	%	0	%	0	%
Volatility	69.5	%	78.7	%	79.1	%	49.5	%	80.9	%	85.9	%
Expected Life	6.2 years		5.4 years		5.7 years		6 months		6 months		6 months

The weighted average fair value of options granted was $5.70 for 2003, $11.34 for 2002, and $12.14 for 2001. The weighted average fair value of the ESPP purchase rights was $4.46, $5.60, and $8.45 for 2003, 2002 and 2001, respectively.

SFAS 130, Reporting Comprehensive Income ("SFAS 130") requires Isis to display comprehensive income (loss) and its components as part of Isis' full set of consolidated financial statements. The measurement and presentation of net income (loss) did not change. Comprehensive income (loss) is comprised of net income (loss) and certain changes in equity that are excluded from net income (loss). Specifically, SFAS 130 requires unrealized holding gains and losses on Isis' available-for-sale securities, which Isis reported separately in stockholders' equity, to be included in accumulated other comprehensive income (loss). Comprehensive income (loss) for the years ended December 31, 2003, 2002 and 2001 has been reflected in the Consolidated Statements of Stockholders' Equity.

In December 2003, the Securities Exchange Commission ("SEC") issued Staff Accounting Bulletin No. 104 ("SAB 104"), Revenue Recognition, which updates portions of the interpretive guidance included in Topic 13 of the codification of Staff Accounting Bulletin No. 101 ("SAB 101") Revenue Recognition in Financial Statements, in order to make this interpretive guidance consistent with current authoritative accounting and auditing guidance and SEC rules and regulations. SAB 104 provides interpretation on selected revenue recognition issues and when revenue is properly recognizable. Revenue should not be recognized until it is realized or realizable and earned. It must meet the following criteria: 1) persuasive evidence of an arrangement exists, 2) delivery occurred or services were rendered, 3) the seller's price to the buyer is fixed or determinable and 4) collectibility is reasonably assured. The adoption of SAB 104 has not had a material impact on Isis' operating results and financial position.

In May 2003, the FASB issued SFAS No. 150, "Accounting for Certain Financial Instruments with Characteristics of both Liabilities and Equity." The statement is effective for financial instruments entered into or modified after May 31, 2003, and otherwise is effective at the beginning of the first interim period beginning after June 15, 2003. This statement establishes standards for how an issuer classifies and measures certain financial instruments with characteristics of both liabilities and equity. It requires that an issuer reclassify certain instruments previously classified as equity as a liability. The adoption of this Statement did not have an impact on Isis' consolidated financial statements.

In January 2003, the FASB issued FASB Interpretation No. 46 ("FIN 46"), Consolidation of Variable Interest Entities, an Interpretation of ARB No. 51, which addresses consolidation by business enterprises of variable interest entities ("VIE"s) either: (1) that do not have sufficient equity investment at risk to permit the entity to finance its activities without additional subordinated financial support, or (2) in which the equity investors lack an essential characteristic of a controlling financial interest. In December 2003, the FASB completed deliberations of proposed modifications to FIN 46 ("Revised Interpretations") resulting in multiple effective dates based on the nature as well as the creation date of the VIE. VIEs created after January 31, 2003, but prior to January 1, 2004, may be accounted for either based on the original interpretation or the Revised Interpretations. However, the Revised Interpretations must be applied no later than Isis' quarter ended March 31, 2004. VIEs created after January 1, 2004 must be accounted for under the Revised Interpretations. Special Purpose Entities ("SPE"s) created prior to February 1, 2003 may be accounted for under the original or Revised Interpretation's provisions no later than the quarter ended December 31, 2003. Non-SPEs created prior to February 1, 2003, should be accounted for under the Revised Interpretation's provisions no later than Isis' quarter ended March 31, 2004. As of December 31, 2003, Isis had a collaborative arrangement with a development stage biopharmaceutical company developing drugs based on RNA splicing technology. Isis considers this entity to be a VIE under the provisions of FIN 46. During 2003 and early 2004, Isis advanced $500,000 and $250,000, respectively, in cash in exchange for a promissory note convertible into equity, pursuant to a qualified financing, as defined by the underlying agreement. To date, Isis has contributed the substantial portion of this entity's funding. Due to the uncertainty in ultimately realizing any value from the consideration paid, Isis expensed the payments when made. As a result, Isis expects that the adoption of FIN 46 will not have a material impact on its operating results and financial position.

Isis invests its excess cash in U.S. Government securities and debt instruments of financial institutions and corporations with strong credit ratings. Isis has established guidelines relative to diversification and maturities that maintain safety and liquidity. These guidelines are periodically reviewed and modified to maximize trends in yields and interest rates without compromising safety and liquidity. The following table summarizes the contract maturity of debt securities held by Isis as of December 31, 2003:

Isis has an ownership interest of less than 20% each in two public and two private companies it conducts business with, and accounts for them under the cost method of accounting according to APB 18. The companies include ATL and Hybridon, which are publicly-traded, and Santaris Pharma A/S ("Santaris") and OncoGenex Technologies, Inc. ("OncoGonex"), which are privately-held. In determining if and when a decrease in market value below Isis' cost is other-than-temporary in its equity positions, Isis examines historical trends in stock price, the financial condition and near term prospects of the issuer, and Isis' current need for cash. When Isis determines that a decline in value is other-than-temporary, Isis recognizes an impairment loss in the period operating results to the extent of

the decline. See Note 1—Organization and Significant Accounting Policies for discussion of impairment loss incurred in 2003.

The following is a summary of Isis' investments accounted for as available-for-sale securities (in thousands):

		Available-for-Sale Securities
		Cost		Unrealized Gains (Losses)		Estimated Fair Value
December 31, 2003
U.S. Treasury securities and obligations of U.S. government agencies		$	90,728	$	142	$	90,870
U.S. corporate debt securities			90,905		612		91,517
Corporate equity securities			3,424		2,722		6,146

	Total	$	185,057	$	3,476	$	188,533

Less current portion included in short-term investments			181,633		754		182,387
Less current portion included in other current assets			2,049		2,722		4,771

Non-current portion included in long-term investments		$	1,375	$	—	$	1,375

		Available-for-Sale Securities
		Cost		Unrealized Gains (Losses)			Estimated Fair Value
December 31, 2002
U.S. Treasury securities and obligations of U.S. government agencies		$	89,607	$	758		$	90,365
U.S. corporate debt securities			96,202		930			97,132
Corporate equity securities			5,045		(2,296	)		2,749

	Total	$	190,854	$	(608	)	$	190,246

Less current portion included in short-term investments			185,809		1,688			187,497
Less current portion included in other current assets			4,475		(2,296	)		2,179

Non-current portion included in long-term investments		$	570	$	—		$	570

		December 31,
		2003			2002
Standard Operating Debt		$	38,392		$	5,702
5¹/2% Convertible Subordinated Notes			125,000			125,000
Convertible Partner Debt
	Eli Lilly		73,750			62,843
	Elan		—			7,219
	Boehringer Ingleheim		6,376			22,576
	Abbott		—			1,019

Total Convertible Partner Debt			80,126			93,657
Capital Leases and Other Obligations			7,116			9,469

Total			250,634			233,828
Less: Current Portion			(16,477	)		(21,435	)
Less: Eli Lilly Debt Classified as Deferred Revenue			(20,760	)		(19,500	)

Total Long-Term Obligations		$	213,397		$	192,893

In December 2003, Isis obtained a $32.0 million term loan from Silicon Valley Bank. The term loan is secured by substantially all of Isis' operating assets, excluding intellectual property, real estate, and certain equity investments. The term loan bears interest at the prime rate (4% at December 31, 2003), is payable in equal monthly payments of principal and interest, matures in December 2008, and is convertible at the election of Isis to a fixed rate at the then-applicable prime rate plus 1.25%. The term loan is subject to certain liquidity and other covenants, including a requirement that Isis maintain a minimum balance in an account at the lending bank at all times equal to the outstanding balance of the loan. Isis used the proceeds of the loan to pay off the Elan and Boehringer Ingelheim ("BI") partner debt in late 2003 and early 2004. The carrying value of this loan at December 31, 2003 was $32.0 million, which approximated fair value.

In December 2002, Isis obtained a credit facility evidenced by promissory notes of up to $6.7 million from a bank to refinance two existing notes. The loan is secured by Isis' real property and bears interest at the prime rate plus 0.5% (4.5% at December 31, 2003), is payable in monthly payments of principal and interest, with the final principal payment due on December 2006. Isis borrowed its final installment of $998,000 on this loan in January 2003. The carrying value of this loan at December 31, 2003 and 2002 was $6.4 million and $5.7 million respectively, which approximated fair value.

In May 2002, Isis completed a $125.0 million convertible debt offering, which raised proceeds of approximately $120.9 million, net of $4.1 million in issuance costs. Isis includes the issuance costs in the balance sheet under Deposits and Other Assets and is amortizing these issuance costs to interest expense over the life of the debt. The subordinated notes bear interest at 5¹/2%, which is payable

semi-annually, and mature in May 2009. The notes are convertible into shares of common stock at a conversion price of $16.625 per share. At December 31, 2003 and 2002, the principal and accrued interest outstanding on the notes was $125.0 million and $1.1 million, respectively. The fair value of the subordinated notes was $101.2 million and $90.1 million as of December 31, 2003 and 2002, respectively.

Isis used a portion of the net proceeds from the convertible notes to prepay amounts outstanding on its 14% Senior Subordinated Notes, and certain obligations under its credit facilities with Elan. These early debt retirements resulted in a $2.3 million loss on prepayment of 14% Notes and a $5.0 million gain on prepayment of 12% Notes, and are included as such on the accompanying Consolidated Statements of Operations for the year ended December 31, 2002.

In September 2002, pursuant to a manufacturing agreement, Lilly provided to Isis a $21.2 million loan to fund the construction of a new manufacturing suite dedicated to the commercial production of Affinitak. The movable equipment purchased for the manufacturing suite secured the loan, which bore interest at prevailing market rates. In June 2003, as part of a renegotiation of the manufacturing relationship, Lilly waived repayment of the loan. As a result of the repayment waiver, Isis reduced to zero the values of the $21.2 million loan and property and equipment totaling $21.2 million related to the new manufacturing suite. As of December 31, 2002, $15.4 million was outstanding on this loan, which approximated fair value. There was no outstanding balance on this loan as of December 31, 2003.

In August 2001, Lilly made available to Isis a $100.0 million loan facility to fund joint research between the two companies. The loan is interest-free and payable, at Isis' option, in cash or common stock at $40 per share in August 2005. The loan facility provides for quarterly draw-downs by Isis in varying amounts. As of December 31, 2003 and 2002, Isis had drawn down $73.8 million and $47.5 million, respectively. Isis accounts for this loan using an imputed interest rate of 20%, consistent with market conditions in place at the time the loan agreement was entered into. Isis carries the net present value of the draw-downs as a long-term obligation and records interest expense over the term of the loan. The difference between the cash received and the present value of the loan represents value Lilly gave Isis to help fund the research collaboration. Isis accounts for this difference as deferred revenue and recognizes it as research and development revenue under collaborative agreements over the period of performance. At December 31, 2003 and 2002, the balance in long-term obligations related to this loan was $53.0 million and $28.0 million, respectively, and the balance in deferred revenue was $20.8 million and $19.5 million, respectively.

During 1999 and 2000, in conjunction with the HepaSense and Orasense joint ventures, Elan made available to Isis credit facilities of up to $30.4 million, evidenced by convertible promissory notes. The terms of these notes provided for interest at 12% annually, with maturities through January 2006. The notes were convertible into the common stock of Isis at anytime by either party, as defined by the underlying agreements. In July 2002, Isis prepaid $19.7 million of the then-outstanding debt with $14.7 million in cash, resulting in a gain of approximately $5.0 million for the year ended December 31, 2002. As of December 31, 2002, there was $7.2 million outstanding under the Orasense credit facility.

Isis paid these borrowings in full from the proceeds of the company's term loan with Silicon Valley Bank in December 2003. Isis can no longer borrow funds against these credit facilities.

During 1997 and 1996, Isis borrowed a total of $22.6 million under a $40.0 million line of credit made available pursuant to the terms of its collaborative agreement with BI. Borrowings under the line of credit bore interest at the seven year U.S. interbanking rate plus 2.0%, determined at the time each advance was made, and ranged from 8.36% to 8.46%. Principal borrowings were repayable in cash or Isis common stock, at the option of Isis. In October 2003, Isis repaid the first installment of $8.3 million, plus interest. Isis repaid the remaining principal installments of $7.9 million, plus interest, and $6.4 million, plus interest, in December 2003 and January 2004, respectively, using the proceeds of the company's term loan with Silicon Valley Bank.

In September 1999, Isis borrowed $1.0 million from Abbott for use as contribution toward costs associated with Abbott's design and construction of a facility for commercial scale manufacturing of oligonucleotides. The debt had an interest rate equal to 2% over the Citibank prime rate (6.75% at December 31, 2002), with interest payable annually. Borrowings under this facility at December 31, 2002 were $1.0 million which approximated fair value. In January 2003, Isis repaid all borrowings under this facility.

At December 31, 2003 and 2002, Isis had approximately $6.3 million and $8.3 million outstanding, respectively, under various capital equipment leases which bear interest at rates ranging from 6.22% to 10.34% and mature at various dates through 2006. At December 31, 2003 and 2002, Isis had approximately $860,000 and $1.2 million, respectively, under various contractual obligations, of which $700,000 and $1.1 million, respectively, represented amounts due to Integrated DNA Technologies, Inc. ("IDT") as part of the supply agreement entered into in December 2001. See Note 6—Research and Development Collaborative Arrangements and Licensing Agreements for more detail on Isis' agreement with IDT.

Annual debt and other obligation maturities at December 31, 2003 are as follows (in thousands):

Isis leases equipment and certain office and lab space under non-cancelable operating and capital leases with terms through July 2010. Two of the building leases have two extension options for five

years each. Annual future minimum payments under capital and operating leases as of December 31, 2003 are as follows (in thousands):

	Operating Leases		Capital Leases
2004	$	2,738	$	3,867
2005		2,765		2,474
2006		2,802		454
2007		1,705		—
2008		993		—
Thereafter		1,311		—

Total minimum payments	$	12,314	$	6,795

Less amount representing interest				(539	)

Present value of future minimum payments			$	6,256
Less current portion				(3,467	)

Long-term portion			$	2,789

Rent expense for the years ended December 31, 2003, 2002, and 2001 was $3.2 million, $2.9 million, and $2.3 million, respectively. Cost of equipment under capital leases at December 31, 2003 and 2002 was $20.7 million and $18.7 million, respectively. Accumulated depreciation of equipment under capital leases at December 31, 2003 and 2002 was approximately $13.6 million and $10.3 million, respectively.

Isis is authorized to issue up to 15,000,000 shares of "blank check" Preferred Stock. As of December 31, 2003, outstanding preferred stock consisted of Series B Convertible Exchangeable 5% Preferred Stock. Series C Junior Participating Preferred Stock is designated but not outstanding.

In August 2002, the holder of the Company's Series A Convertible Exchangeable Preferred Stock exercised its option to convert the Series A shares into Isis common stock. The transaction converted the outstanding 120,150 shares of Series A Convertible Preferred Stock into 656,674 shares of Isis common stock using a conversion price of $21.54 per share. Included in the conversion was approximately $2.1 million in preferred stock dividends accrued in prior periods.

At December 31, 2003 and 2002, Isis had 16,620 shares authorized, of which 12,015 shares were issued and outstanding, of Series B Convertible Exchangeable 5% Preferred Stock. The shares have a term of six years and are convertible into Isis' common stock at $14.00 per share, which was 125% of the average closing price of Isis common stock for the 60 trading days ending two business days prior to June 30, 2002. In the event of a significant transaction, the shares are convertible into Isis' common stock at 125% of the average closing price of Isis' common stock for the 60 trading days ending two

business days prior to the significant transaction. The Preferred Stock is also exchangeable for the ownership Isis holds in HepaSense and bears a mandatory pay-in-kind dividend of 5.0% per year based on the original issue price per share, compounded semi-annually, payable only upon conversion into Isis' common stock or cash.

In December 2000, Isis adopted a Preferred Share Purchase Rights Plan (the "Plan"). The Plan provides for a dividend distribution of one preferred stock purchase right (a "Right") for each outstanding share of Isis common stock, par value $0.001 per share (the "Common Shares"), held of record at the close of business on January 10, 2001, and on each subsequently issued share of Isis common stock. The Rights are not currently exercisable. Under certain conditions involving an acquisition or proposed acquisition by any person or group holding 20% or more of Isis' common stock, the Rights permit the holders (other than the 20 percent holder) to purchase one one-hundredth of a share of Series C Junior Preferred Stock, par value $0.001 per share (the "Preferred Shares"), at a price of $85 per one one-hundredth of a Preferred Share (the "Purchase Price"), subject to adjustment. Each one one-hundredth of a share of Preferred Shares has designations and powers, preferences and rights, and the qualifications, limitations and restrictions which make its value approximately equal to the value of a Common Share. Under certain conditions, Isis' Board of Directors may redeem the Rights in whole, but not in part, at a price of $0.001 per Right.

At December 31, 2003 and 2002, Isis had 100,000,000 shares authorized, of which 55,557,253 and 55,215,785 were issued and outstanding, respectively.

In June 1989 and as amended, Isis' Board of Directors adopted, and the stockholders subsequently approved, a stock option plan that provides for the issuance of non-qualified and incentive stock options for the purchase of up to 10,200,000 shares of common stock to its employees, directors, and consultants. The plan also includes provisions for the issuance of stock pursuant to restricted stock purchases and bonuses. Typically, options expire 10 years from the date of grant. Options granted after December 31, 1995 vest over a four year period, with 25% exercisable at the end of one year from the date of the grant and the balance vesting ratably thereafter. Options granted before January 1, 1996 generally vested over a five-year period. At December 31, 2003, a total of 2,689,000 options were outstanding, options to purchase 2,515,000 shares were exercisable, and 2,791,000 shares were available for future grant.

In January 2000, Isis adopted the 2000 Broad-Based Equity Incentive Plan (the "2000 Plan"), which provides for the issuance of non-qualified stock options for the purchase of up to 3,990,000 shares of common stock to its employees, directors, and consultants. In May 2002, the Board of Directors increased the 2000 Plan by 2,000,000 shares, authorizing up to 5,990,000 shares of common stock under the 2000 Plan for issuance to employees, directors, and consultants. Typically options expire 10 years from the date of grant. Options granted under this plan generally vest over a four-year period,

with 25% exercisable at the end of one year from the date of the grant and the balance vesting ratably thereafter. Options granted under this plan, pursuant to the April 2003 stock option exchange program expire on December 31, 2008 and vested 33.34% on January 1, 2004 and then at the rate of 2.78% per month during the optionee's employment or service as a consultant, employee or director. At December 31, 2003, a total of 5,198,000 options were outstanding, 1,928,000 shares were exercisable, and 539,000 shares were available for future grant.

In September 2001, Isis' Board of Directors adopted, and the stockholders subsequently approved, the 1992 Non-Employee Directors' Stock Option Plan, which provides for the issuance of non-qualified stock options to Isis' non-employee directors. The name of the resulting new plan is the 2002 Non-Employee Directors' Stock Option Plan, and it has an aggregate of 803,000 shares of common stock reserved for issuance. Options under this plan expire 10 years from the date of grant. Options granted become exercisable in four equal annual installments beginning one year after the date of grant. At December 31, 2003, a total of 324,000 options were outstanding, 169,000 of the shares issued under this plan were exercisable and 460,000 shares were available for future grant.

The following table summarizes stock option activity for the years ended December 31, 2001 through December 31, 2003 (in thousands, except per share data):

The following table summarizes information concerning currently outstanding and exercisable options (in thousands, except contractual life and exercise price data):

In 2000, Isis' Board of Directors adopted, and the stockholders subsequently approved, the 2000 Employee Stock Purchase Plan and Isis reserved 200,000 shares of common stock for issuance thereunder. In each of the subsequent years, an additional 200,000 shares of common stock were reserved for the 2000 Employee Stock Purchase Plan, resulting in a total of 800,000 shares authorized in the plan. The plan permits full-time employees to purchase common stock through payroll deductions (which cannot exceed 10% of each employee's compensation) at the lower of 85% of fair market value at the beginning of the offering period or the end of each six-month purchase period. During 2003, 307,070 shares were issued under this plan to employees at prices ranging from $4.40 to $5.60 per share. At December 31, 2003, 200,138 shares were available for purchase under this plan.

In 2002, Isis issued warrants to purchase 6,304 shares of common stock to Elan for the achievement of a development milestone related to the HepaSense joint venture between Isis and Elan. As of December 31, 2003, all of the warrants remained outstanding at an exercise price of $59.48 per share. The warrants expire April 25, 2007.

In 2000, Isis issued warrants to purchase 14,881 shares of common stock to Elan as part of the joint venture collaboration between Isis and Elan to form HepaSense. As of December 31, 2003, all of the warrants remained outstanding at an exercise price of $50.40 per share. The warrants expire May 1, 2008.

In 1999, Isis issued warrants to purchase 215,000 shares of common stock to Elan as part of the joint venture collaboration between Isis and Elan to form Orasense. As of December 31, 2003, all of

the warrants remained outstanding at an exercise price of $24 per share. The warrants expire April 19, 2004.

In 1997 and 1998, Isis issued warrants to purchase 500,000 and 300,000 shares of common stock, respectively, in conjunction with a private debt financing agreement. As of December 31, 2003, all of the warrants remained outstanding at an exercise price of $25 per share. The warrants expire November 1, 2004. See Note 3 for further details.

As of December 31, 2003, total common shares reserved for future issuance was approximately 13,238,000.

Significant components of Isis' deferred tax assets as of December 31, 2003 and 2002 are shown below (in thousands). Isis had recognized valuation allowances of $235.0 million and $197.4 million for 2003 and 2002, respectively, to offset the net deferred tax assets as realization of such assets is uncertain.

		2003			2002
Deferred tax assets:
	Capitalized research expense	$	86,827		$	17,780
	Net operating loss carry forwards		110,652			135,688
	Research and development credits		30,095			24,634
	Deferred revenue		9,573			19,573
	Other, net		5,180			617

	Total deferred tax assets		242,327			198,292
Deferred tax liabilities:
	Intangible Assets		(7,361	)		(921	)

	Total deferred tax liabilities		(7,361	)		(921	)

	Total net deferred tax assets		234,966			197,371
	Valuation allowance for deferred tax assets		(234,966	)		(197,371	)

	Net deferred tax assets	$	—		$	—

At December 31, 2003, approximately $6.5 million of the valuation allowance for deferred tax assets related to stock option deductions which, when recognized, will be allocated directly to additional paid-in capital.

At December 31, 2003, Isis had federal and California tax net operating loss carryforwards of approximately $312.1 million and $24.9 million, respectively. Isis also had federal and California research credit carryforwards of approximately $21.9 million and $11.9 million, respectively. The difference between the tax loss carryforwards for federal and California purposes was attributable to the capitalization of research and development expenses for California tax purposes and a required 50% to 60% limitation on the utilization of California loss carryforwards. The federal tax loss carryforwards and the research credit carryforwards will begin expiring in 2007 and 2004, respectively, unless previously utilized. The California tax loss carryforwards will begin expiring in 2004, unless utilized.

Pursuant to Sections 382 and 383 of the Internal Revenue Code, annual use of Isis' net operating loss and credit carryforwards may be limited due to cumulative changes in ownership of more than 50%. Isis believes that changes in ownership have occurred, but believes that such limitations will not have a material impact upon the utilization of the carryforwards.

6. Research and Development Collaborative Arrangements and Licensing Agreements

In November 2003, Isis received a grant of up to $8.0 million over three years from the Singapore Economic Development Board ("Singapore EDB"), which will fund, in part, the broadening of two of Isis' RNA-based drug discovery and development programs: micro-RNA drug discovery and antisense drug discovery targeting the coronavirus associated with Severe Acute Respiratory Syndrome, or SARS. In connection with this grant, Isis established Isis Pharmaceuticals Singapore Pte Ltd, a wholly-owned subsidiary of Isis Pharmaceuticals, Inc. Isis recorded no revenue during 2003 related to this grant, and expects to begin earning revenue from this grant in the last half of fiscal 2004.

In June 2003, Isis initiated a collaboration with the Industrial and Technology Research Institutes of Taiwan, ("ITRI") to identify antisense candidates targeting the coronavirus associated with SARS. The collaboration entitled Isis to an upfront payment, milestone payments, and the potential for future funding. The milestones related to the identification of second-generation antisense drugs that inhibit SARS virus replication and the successful completion of preclinical studies evaluating aerosol and parenteral delivery of antisense drugs as specified under the agreement. During 2003, Isis recorded revenue under this collaboration of $2.0 million, comprised of $1.0 million for an up-front payment and $1.0 million related to the achievement of certain milestones, as described above.

In November 1988 and September 2000, Isis entered into license agreements with Santaris, a privately-held company, formerly Pantheco, a privately-held company. The agreement was amended in May 2003. Under the terms of the amended and restated license agreements, Isis licensed its novel antisense chemistry, Peptide Nucleic Acid, or PNA, to Santaris on a limited exclusive basis to develop products. The license is restricted to a limited number of molecular targets that are subject to Isis' approval. Santaris has agreed to pay Isis royalties on any products developed under the license.

As part of its original license agreements with Pantheco, Isis received shares of Pantheco common stock. In May 2003, Pantheco A/S and Cureon A/S merged to form Santaris Pharma A/S. Prior to the merger, Isis purchased additional shares of Pantheco for $55,000 as a result of antidilution provisions related to Pantheco's stock. After the merger and as of December 31, 2003, Isis' ownership interest in Santaris was less than 10%. Isis used the equity method of accounting for its investment in Pantheco in 2001, when Isis' ownership was in excess of 20%, and, accordingly, recorded $267,000 under equity in loss of affiliates during 2001. Isis did not record equity in loss of affiliates for 2003 and 2002 related to this investment, as Isis' ownership was less than 20%. After the merger and as of December 31, 2003, Isis' ownership interest in Santaris was less than 10%. Isis' balance sheet at December 31, 2003 and

2002 includes a long-term investment at the lower of carrying value or fair market value of $625,000 and $570,000, respectively, related to this equity investment.

In May 2003, Isis and Ercole Biotech, Inc. ("Ercole") initiated a multi-year collaboration to discover antisense drugs that regulate alternative RNA splicing. As part of the collaboration, the two parties cross-licensed their respective splicing-related intellectual property. Ercole received a license to Isis's Bcl-x molecule and certain of Isis' chemistry patents. During 2003, Isis advanced $500,000 in cash in exchange for a promissory note convertible into equity, pursuant to a qualified financing, as defined by the underlying agreement. Due to the uncertainty in ultimately realizing any value from the consideration paid, Isis expensed the payments during 2003, and included them in research and development expense on the accompanying consolidated statements of operations.

In December 2001, Isis licensed its compound, ISIS 107248, an antisense inhibitor, to Antisense Therapeutics, Ltd., Inc. ("ATL"), a publicly-traded company listed on the Australian Stock Exchange. Isis and ATL are participating in a five-year antisense drug discovery and development collaboration, under which ATL pays Isis research fees, reimbursing Isis for costs related to preclinical work on ISIS 107248, and may purchase drugs from Isis. In connection with this collaboration, Isis received 30.0 million shares of ATL common stock upon completion of ATL's initial public offering ("IPO"), representing an initial ownership percentage of approximately 14%, and options to purchase an additional 20.0 million shares of ATL common stock, which expire in 2008. Isis valued its initial ownership at $2.8 million, and recognizes revenue on this amount over the term of the agreement. For the years ended December 31, 2003, 2002 and 2001, Isis recorded revenue of $811,000, $3.7 million and $46,000 respectively, related to this collaboration. As of December 31, 2003, Isis' ownership percentage in ATL, including 10.3 million shares Isis purchased subsequent to shares it acquired in the IPO, was approximately 11%. Isis' balance sheet at December 31, 2003 and 2002 includes a short-term investment at fair market value of $3.6 million and $1.5 million, respectively, related to this equity investment.

Pursuant to a November 2001 drug development collaboration, Isis agreed to co-develop and commercialize the anti-cancer antisense drug candidate, OGX-011, or ISIS 112989, with OncoGenex, a privately-held Canadian oncology-focused research and development company. Pursuant to a September 2003 drug development collaboration, Isis also agreed to license certain technology and collaborate on the initial development of the anti-cancer antisense drug candidate, OGX-225, a second-generation drug, with OncoGenex. Under the terms of the collaboration, during 2003 OncoGenex paid Isis an up-front fee and Isis acquired an ownership interest in OncoGenex of less than 10%. Isis recorded the upfront fee as revenue during 2003. Isis recorded no revenue from OncoGenex during 2002 and 2001.

In December 2001, Isis entered into a three-year collaboration with Amgen, Inc. ("Amgen") to discover new antisense drugs. Amgen has the right to develop and commercialize antisense drugs

resulting from the collaboration. Under the terms of the agreement, Isis is entitled to receive milestone payments upon key clinical, research and commercial achievements, as well as royalties on sales of any products resulting from the collaboration. During 2003 and 2002, Isis recorded revenue of $1.9 million and $3.6 million, respectively, related to progress research milestones under this drug discovery collaboration.

In August 2001, Isis entered into a broad strategic relationship with Lilly that has four key components:

Second, Isis licensed to Lilly rights to Affinitak, an antisense drug which Lilly is testing in a Phase III trial for the treatment of non-small cell lung cancer. Lilly paid Isis $25.0 million in up-front fees for Affinitak and reimbursed Isis for Isis' Affinitak development costs. During 2003, 2002 and 2001, Isis recorded revenue of $11.1 million, $31.9 million and $9.2 million, respectively, related to the reimbursement of Affinitak costs.

Third, Isis initiated with Lilly a four-year antisense drug discovery collaboration in the areas of metabolic and inflammatory diseases and a related collaboration to determine the function of up to 1,000 genes. In 2002, Lilly and Isis expanded this collaboration to include oncology and the license of LY2181308, formerly called ISIS 23722, Isis' antisense inhibitor of survivin, which is the first compound under the collaboration which Lilly has selected for clinical development. During 2003, Isis recorded revenue of $1.5 million under this collaboration related to a development milestone for LY2181308.

Fourth, Lilly committed to lend Isis, interest-free, up to $100.0 million over a four-year period to fund its obligations under the drug discovery collaboration. Isis can repay this loan at its option in either cash or Isis common stock, at a fixed conversion price of $40 per share. During 2003, 2002 and 2001, Isis recorded revenue related to the research collaboration loan of $7.5 million, $3.2 million, and $407,000, respectively and corresponding interest expense during 2003, 2002, and 2001, of $7.5 million, $3.2 million, and $407,000, respectively.

In September 2002, Isis further expanded its relationship with Lilly by agreeing to manufacture Affinitak during the product launch period for Lilly. In connection with this agreement, Isis upgraded and expanded its current manufacturing facility and added a new manufacturing suite. Lilly provided Isis with $21.2 million to build the new suite which Isis initially intended to use for the commercial manufacture of Affinitak. In June 2003, Lilly waived repayment of the $21.2 million manufacturing loan and agreed to allow Isis to use the facility to manufacture other drugs. In exchange, Isis released Lilly from its obligation to purchase additional product from Isis and its obligation to pay for the costs of maintaining an idle manufacturing suite.

In May 2001, Isis agreed to license to Merck & Co, Inc. ("Merck"), Isis' preclinical Type 2 diabetes antisense drug candidate, ISIS 113715. In exchange for the license, Isis received an upfront payment. In addition, Isis received development funding and earned a development milestone. During 2002 and 2001, Isis recorded revenue of $840,000 and $4.4 million, respectively, under this agreement. Isis reacquired full product rights to ISIS 113715 from Merck in December 2002. As a result, no revenue was recorded for the year ended December 31, 2003.

In June 1998, Isis entered into a research collaboration with Merck to discover small molecule drug candidates to treat patients infected with HCV. The collaboration, which concluded in May 2003 in accordance with its terms, provided that Merck would pay Isis' annual research support and milestone payments. Within the collaboration, Isis and Merck designed, synthesized, and evaluated novel compounds that Merck screened in its proprietary assays for identifying HCV replication inhibitors. Merck had the right to commercialize any drugs arising from the collaboration, and Isis retains the right to use technology developed in the collaboration in its antisense program. During 2003, 2002, and 2001, Isis recorded revenue of $900,000, $2.2 million and $3.6 million, respectively, from Merck under the terms of this agreement.

During 1999 and 2000, Isis and Elan, formed HepaSense, a joint venture to develop an antisense drug to treat patients chronically infected with the Hepatitis C virus, or HCV, and Orasense, a joint venture to develop technology for the formulation of oral drugs. Isis and Elan provided development and manufacturing services to HepaSense and Orasense during the joint ventures through late 2002, when Elan concluded its participation in both collaborations. Isis regained all rights to ISIS 14803, the compound that HepaSense had been developing, and ISIS 104838, the compound Orasense had been developing. Isis has continued to develop its oral delivery technology initially developed within Orasense.

During 2002 and 2001, Isis recorded revenue of $3.0 million and $5.2 million, respectively, from HepaSense, and $8.9 million and $5.4 million, respectively, from Orasense, which Isis included in the consolidated statements of operations as research and development revenue from affiliates. Isis recorded no revenue from HepaSense or Orasense during 2003. During 2002 and 2001, Isis recorded $6.5 million and $8.3 million, respectively, as equity in the loss of HepaSense, and $9.5 million and $10.3 million, respectively, as equity in the loss of Orasense. There was no equity in the loss of HepaSense or Orasense during 2003. At December 31, 2002, the balance sheet included $6.2 million of contracts receivable relating to Orasense and a funding obligation due to Orasense of $5.2 million. Isis fully funded Orasense for this obligation in 2003 and Orasense paid Isis amounts due to it. As of December 31, 2003, Isis had no receivable or funding obligation related to Orasense.

In July 1997, Isis and Novartis Ophthalmics AG, ("Novartis"), formerly CIBA Vision Corporation, entered into an agreement granting Novartis exclusive worldwide distribution rights for Vitravene or fomivirsen. Under the terms of the agreement, Isis manufactures and sells Vitravene to Novartis, who is responsible for worldwide sales and marketing. During 2002, Isis recorded revenue of $2.5 million related to a milestone payment from Novartis, and $293,000 in revenue related to the shipment of Vitravene. Isis recorded no revenue related to the sales of Vitravene during 2003 and 2001.

Through its Ibis program, Isis continues to collaborate with San Diego-based Science Applications International Corporation, or SAIC, on a multi-year project funded by the Defense Advanced Research Projects Agency, or DARPA, for the ongoing development of Isis' Triangulation Identification for Genetic Evaluation of Risks ("TIGER") technology. This project combines Isis' expertise in microbial genome sequence analysis and advanced mass spectrometry technology with SAIC's advanced signal

processing capabilities. As of December 31, 2003, Isis had been awarded funding of $13.3 million related to this collaboration with SAIC, of which $13.3 million had been billed and $11.9 million had been collected.

In September 2003, Isis received a three-year grant for up to $6.0 million from the Centers for Disease Control, or CDC, to develop and apply its TIGER technology to the surveillance of human infectious disease in the U.S. Using the grant from the CDC, Isis expects to develop and provide TIGER technology for CDC projects focused on emerging human infectious disease.

In March 2004, Isis entered into a two-year contract with SAIC to further the development of Isis' TIGER biosensor to identify infectious agents in biological warfare attacks. The contract provides for up to $19.5 million in funding by DARPA.

Since inception, Isis has received financial support from government-funded grants and contracts to use its technology to assist in national defense. In March 2002, Isis transitioned its government-sponsored research program to discover novel broad-spectrum antibacterial drugs for biological warfare defense to the U.S. Army Medical Research Institute for Infectious Diseases, or USAMRIID. USAMRIID awarded Isis a three-year, $2.4 million contract to advance Isis' work in this area.

In addition to DARPA, USAMRIID and the CDC, Isis also has research relationships with other government entities including the United States Navy and the Federal Bureau of Investigation. During 2003, 2002, and 2001, Isis recorded revenue generated directly or indirectly from agencies of the U.S. Government of $9.5 million, $5.9 million, and $2.9 million, respectively.

In December 2001, Isis licensed to Eyetech Pharmaceuticals, Inc., ("Eyetech"), a publicly-held company, certain of its patents necessary for Eyetech to develop, make and commercialize Macugen, a non-antisense compound intended for use in the treatment of ophthalmic diseases. Eyetech paid Isis a $2.0 million upfront fee and agreed to pay milestone and royalty payments in exchange for non-exclusive, worldwide rights to the intellectual property licensed from Isis. In December 2002, Eyetech entered into an agreement with Pfizer to develop and commercialize Macugen. In 2003, Pfizer and Eyetech reported encouraging Phase III data for Macugen that Eyetech believes will serve as the basis for a new drug application with commercialization of the drug expected in 2005. Assuming successful commercialization of Macugen, Isis has the opportunity to earn future milestone payments and royalties.

In March 1999, Isis established a long-term research scale antisense inhibitor supply agreement with IDT, which provides for IDT to manufacture research scale antisense inhibitors and research reagents to Isis' specifications. Isis initially paid IDT $5.0 million toward future purchases of supplies and in December 2001 expanded its licensing agreement on certain antisense patents to allow Isis to exclusively sublicense these patents for functional genomics purposes. The agreement also eliminated milestone payments and reduced royalty rates Isis was to pay IDT for commercialized second-generation antisense drugs. In addition, Isis has paid IDT $350,000, $350,000, and $3.5 million in 2003, 2002 and 2001, respectively, and will pay IDT $700,000 over the next two years for the license. At December 31, 2003 and 2002, Isis' balance sheet reflected a deposit of $4.3 million and $4.5 million, respectively, and a licensing asset, net of amortization, of $4.2 million and $4.5 million, respectively.

Pursuant to a license agreement with Hybridon in May 2001 and amended August 2002, Isis acquired an exclusive license to all of Hybridon's antisense chemistry and delivery patents and technology. Hybridon retained the right to practice its licensed antisense patent technologies and to sublicense it to collaborators under certain circumstances. In addition, Hybridon received a non-exclusive license to Isis's suite of RNase H patents. Under the terms of the amended agreement, Isis paid Hybridon $15.0 million in cash and issued to Hybridon approximately 357,000 shares of Isis common stock valued at $5.0 million and 500,000 shares of Isis common stock valued at $10.0 million. Hybridon made a cash payment to Isis of $700,000 and issued to Isis approximately 1.0 million shares of its common stock valued at $1.3 million. Isis' balance sheet at December 31, 2003, 2002 and 2001 reflected a licensing asset, net of amortization, of $23.2 million, $25.1 million and $27.4 million, respectively, and a short-term investment at fair market value of $1.1 million and $704,000, respectively, related to this agreement.

A relatively small number of customers accounted for a significant percentage of Isis' revenue in the fiscal years ended December 31, 2003, 2002 and 2001. In 2003, 2002 and 2001, one significant customer accounted for approximately 62%, 57% and 27% of revenue, respectively. During 2003, Isis derived approximately 20% of its revenue directly or indirectly from agencies of the U.S. Government, including approximately 16% of revenue from one significant customer. Contract receivables from this customer comprised approximately 49% of contract receivables at December 31, 2003.

In April 2003, Isis initiated a restructuring in response to disappointing results from the first Phase III trial of Affinitak. As a result, Isis had a small reduction in its workforce, which primarily represented positions that were in support of the commercialization and manufacture of Affinitak. Consequently, Isis incurred a restructuring charge of approximately $1.8 million during the second quarter of 2003. Isis completed the utilization of the reserve related to this restructuring in the fourth quarter of 2003.

In November 2002, Isis announced the termination of the GeneTrove database product offering and the resulting reorganization of the GeneTrove program. As a result, Isis reduced its workforce by approximately 25 people. The restructuring plan also provided for the write-down of certain intellectual property. As a result of this plan, Isis recognized restructuring related charges of approximately $1.4 million as operating expenses in the fourth quarter of 2002. Isis did not recognize any additional GeneTrove restructuring related charges for the year ended December 31, 2003 and completed utilization of the reserve related to this restructuring in the fourth quarter of 2003.

The following table summarizes the balance of the accrued restructuring reserve related to GeneTrove and Affinitak (in thousands):

Isis has an employee 401(k) salary deferral plan, covering all domestic employees. Employees may make contributions by withholding a percentage of their salary up to the IRS annual limit ($12,000 and $14,000 in 2003 for employees under 50 years old and over 50 years old, respectively). Isis made approximately $463,000 and $413,000 in matching contributions for the years ended December 31, 2003 and 2002, respectively.

Due to the significant minority investor rights retained by Elan and its subsidiaries, Isis accounted for its investment in Orasense under the equity method of accounting through 2002. At inception, Elan granted to Orasense a license to its intellectual property for $15.0 million. The term of the license was three years and amortization expense related to this license was $1.3 million for the year ended December 31, 2002. Through December 2002, Orasense incurred research and development expenses, performed by Elan and Isis on Orasense's behalf, in the course of its product development. In conjunction with its restructuring efforts, Elan concluded its participation in the Orasense collaboration effective December 31, 2002. As a result, Isis regained all rights to ISIS 104838, the compound that Elan and Isis were developing within Orasense. Isis has continued to develop its oral delivery technology within Orasense.

The following table presents summary financial information for Orasense as of and for the years ended (in thousands, except per share amounts):

Due to the significant minority investor rights retained by Elan and its subsidiaries, Isis accounted for its investment in HepaSense under the equity method of accounting through 2002. At inception, Elan granted to HepaSense a license to its intellectual property for $15.0 million. The term of the license was 3 years and amortization expense related to this license totaled $5.0 million for the year ended December 31, 2002. HepaSense incurred research and development expenses, performed by Elan and Isis on HepaSense's behalf, in the course of its product development. In conjunction with its restructuring efforts, Elan concluded its participation in the HepaSense collaboration in 2002 and Isis regained all rights to ISIS 14803, the compound that Elan and Isis were developing within HepaSense. As a result of the collaboration termination, there was no activity during the year ended December 31, 2003.

The following table presents summary financial information for HepaSense as of and for the years ended (in thousands, except per share amounts):

The following financial information reflects all normal recurring adjustments, which are, in the opinion of management, necessary for a fair statement of the results of the interim periods.

Summarized quarterly data for the years ended December 31, 2003, and 2002 are as follows (in thousands, except per share data).

	First Quarter		Second Quarter		Third Quarter		Fourth Quarter
2003 Quarters
Revenues	$	13,980		$	15,017		$	11,294		$	9,699
Operating expenses		32,892			34,536			30,361			31,179
Loss from operations		(18,912	)		(19,519	)		(19,067	)		(21,480	)
Net loss		(24,321	)		(23,097	)		(22,203	)		(25,375	)
Accretion of dividends on preferred stock		(171	)		(172	)		(175	)		(176	)
Net loss applicable to common stock	$	(24,492	)	$	(23,269	)	$	(22,378	)	$	(25,551	)
Basic and diluted net loss per share(1)	$	(0.44	)	$	(0.42	)	$	(0.40	)	$	(0.46	)

	First Quarter		Second Quarter		Third Quarter		Fourth Quarter
2002 Quarters
Revenues	$	17,959		$	20,061		$	20,300		$	21,859
Operating expenses		27,677			32,400			37,809			33,106
Loss from operations		(9,718	)		(12,339	)		(17,509	)		(11,247	)
Net loss		(17,972	)		(20,865	)		(17,576	)		(15,829	)
Accretion of dividends on preferred stock		(335	)		(335	)		(222	)		(168	)
Net loss applicable to common stock	$	(18,307	)	$	(21,200	)	$	(17,798	)	$	(15,997	)
Basic and diluted net loss per share(1)	$	(0.34	)	$	(0.39	)	$	(0.33	)	$	(0.29	)

In March 2004, Isis entered into a strategic alliance with Alnylam to develop and commercialize RNAi therapeutics. Under the terms of the agreement, Isis exclusively licensed to Alnylam its patent estate relating to antisense motifs and mechanisms and oligonucleotide chemistry for double-stranded RNAi therapeutics in exchange for a $5.0 million technology access fee, participation in fees for Alnylam's partnering programs, as well as future milestone and royalty payments. As part of the agreement, Isis also made a $10.0 million equity investment in Alnylam. In turn, Alnylam nonexclusively licensed Isis its patent estate relating to antisense motifs and mechanisms and oligonucleotide chemistry to research, develop and commercialize single-stranded RNAi therapeutics and to research double stranded-RNAi compounds. Isis also received a license to develop and commercialize double-stranded RNAi drugs targeting a limited number of therapeutic targets on either

an exclusive or co-exclusive basis depending on the target. If Isis develops or commercialize an RNAi-based drug using Alnylam's technology, Isis will pay Alnylam milestones and royalties.

ý	ANNUAL REPORT PURSUANT TO SECTION 13 OR 15(d) OF THE SECURITIES EXCHANGE ACT OF 1934
For the fiscal year ended December 31, 2003
o	TRANSITION REPORT PURSUANT TO SECTION 13 OR 15(d) OF THE SECURITIES EXCHANGE ACT OF 1934
For The Transition Period From to

Delaware (State or other jurisdiction of incorporation or organization)		33-0336973 (IRS Employer Identification No.)
2292 Faraday Ave., Carlsbad, CA 92008 (Address of principal executive offices, including zip code)
760-931-9200 (Registrant's telephone number, including area code)
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

Plan Category	Number of Shares to be Issued Upon Exercise of Outstanding Options		Weighted Average Exercise Price of Outstanding Options		Number of Shares Remaining Available for Future Issuance
Equity compensation plans approved by stockholders	3,013,000	(a)	$	9.90	3,251,000	(c)
Equity compensation plans not approved by stockholders	5,198,000	(b)	$	7.94	539,000

Total	8,211,000		$	8.66	3,790,000

4.19		Form of 5¹/2% Convertible Subordinated Note due 2009.(16)
10.1		Form of Indemnification Agreement entered into between the Registrant and its Directors and Officers with related schedule.(1)
10.2*		Registrants 1989 Stock Option Plan, as amended.(6)
10.3*		Registrants 1992 Non-Employee Directors Stock Option Plan, as amended.(4)
10.4*		Registrants Employee Stock Purchase Plan.(10)
10.5		Form of Employee Assignment of Patent Rights.(1)
10.6*		Registrants 2000 Broad-Based Equity Incentive Stock Option Plan and related form of option agreement.(10)
10.7		Agreement between the Registrant and CIBA Vision Corporation (now Novartis Ophthalmics AG) dated July 10, 1997 (with certain confidential information deleted).(5)
10.8		Amendment No. 2 to the Agreement between the Registrant and CIBA Vision Corporation dated September 14, 1998 (with certain confidential information deleted).(8)
10.9		Imperial Bank Note Secured by Deed of Trust dated March 24, 1997 in the amount of $6,000,000, together with the related Deed of Trust and Assignment of Rents dated March 24, 1997.(5)
10.10		Imperial Bank Note Secured by Deed of Trust dated March 24, 1997 in the amount of $3,706,620, together with the related Deed of Trust and Assignment of Rents dated March 24, 1997.(5)
10.11		Asset Purchase Agreement between the Registrant and Gen-Probe Incorporated dated December 19, 1997 (with certain confidential information deleted).(6)
10.12		Research Collaboration and License Agreement between Merck & Co., Inc. and the Registrant dated June 1, 1998 (with certain confidential information deleted).(7)
10.13		Patent Rights Purchase Agreement between the Registrant and Gilead Sciences, Inc., dated December 18, 1998 (with certain confidential information deleted).(9)
10.14		Rights Agreement dated as of December 8, 2000 between the Registrant and American Stock Transfer & Trust Company.(17)
10.15		Master Agreement between the Registrant and Hybridon, Inc., dated May 24, 2001 (with certain confidential information deleted).(19)
10.16		Development and License Agreement, dated August 14, 2001 between the Registrant and Eli Lilly and Company (with certain confidential information deleted).(20)
10.17		Subcontract Agreement, dated October 25, 2001 between the Registrant and Science Applications International Corporation.(21)
10.18		Master Agreement dated October 30, 2001 between the Registrant and Antisense Therapeutics Limited.(24)
10.19		Collaboration and License Agreement dated October 30, 2001 between the Registrant and Antisense Therapeutics Limited (with certain confidential information deleted).(24)
10.20		Clinical Supply Agreement dated October 30, 2001 between the Registrant and Antisense Therapeutics Limited (with certain confidential information deleted).(24)

10.21		Stock Purchase Agreement dated October 30, 2001 between the Registrant and Antisense Therapeutics Limited.(24)
10.22		Collaboration and Co-development Agreement, dated November 16, 2001 between the Registrant and OncoGenex Technologies Inc.(22)
10.23		Oligonucleotide Manufacturing and Supply Agreement dated December 4, 2001 between the Registrant and Integrated DNA Technologies, Inc. (with certain confidential information deleted).(24)
10.24		Amended and Restated IDT-Isis Licensing Agreement dated December 4, 2001 between the Registrant and Integrated DNA Technologies, Inc. (with certain confidential information deleted).(24)
10.25		Collaboration Agreement dated December 11, 2001 between the Registrant and Amgen Inc.(24)
10.26		License Agreement dated December 31, 2001 between the Registrant and Eyetech Pharmaceuticals, Inc. (with certain confidential information deleted).(25)
10.27		Amendment No. 1 to Securities Purchase Agreement dated January 14, 2000, between the Registrant and Elan International Services, Ltd. (with certain confidential information deleted).(27)
10.28		Letter Agreement dated April 24, 2002 between the Registrant and Reliance Insurance Company.(26)
10.29		Amended and Restated Collaboration Agreement dated June 17, 2002, between the Registrant and Eli Lilly and Company (with certain confidential information deleted).(27)
10.30		Settlement, Release, and License Grant Agreement dated September 6, 2002, between the Registrant and Sequitur, Inc. (with certain confidential information deleted).(28)
10.31		Amended and Restated License Agreement among the Registrant, Orasense Ltd. and Elan Corporation Plc. dated October 24, 2002 (with certain confidential information deleted).(30)
10.32		Amended and Restated License Agreement among the Registrant, Orasense Ltd. and Elan Corporation Plc. dated October 24, 2002 (with certain confidential information deleted).(30)
10.33		Amended and Restated Subscription, Joint Development and Operating Agreement among the Registrant, Elan Corporation, Plc., Elan International Services, Ltd. and Orasense Ltd., dated October 24, 2002 (with certain confidential information deleted).(30)
10.34		Termination Agreement among the Registrant, Elan Corporation, Plc., Elan Pharma International Limited, Elan International Services, Ltd. and HepaSense Ltd., dated November 5, 2002.(30)
10.35		Registrant's restated 10b5-1 Trading Plan.(29)
10.36		Registrant's 2002 Non-Employee Directors' Stock Option Plan.(31)
10.37		Registrant's Form of 2002 Non-Employee Directors' Stock Option Agreement.(31)
10.38		Waiver and Release Agreement dated June 5, 2003 between the Registrant and Eli Lilly and Company (with certain confidential information deleted). (32)

		ISIS PHARMACEUTICALS, INC.
		By:		/s/ STANLEY T. CROOKE, M.D., PH.D. Stanley T. Crooke, M.D., Ph.D. Chairman of the Board, President and Chief Executive Officer (Principal executive officer)

/s/ MARK B. SKALETSKY Mark B. Skaletsky		Director		March 12, 2004
/s/ JOSEPH H. WENDER Joseph H. Wender		Director		March 12, 2004

10.39		Amendment Number One to Development and License Agreement dated June 5, 2003 between the Registrant and Eli Lilly and Company (with certain confidential information deleted). (32)
10.40		Initial Collaboration Agreement dated June 23, 2003 between the Registrant and Industrial Technology Research Institutes (with certain confidential information deleted). (32)
10.41		Form of Severance Agreement dated April 2003 entered into between the Registrant and its executive officers and certain key employees, together with related schedule. (32)
10.42		Grant letter dated September 29, 2003 from the Centers for Disease Control and Prevention (with certain confidential information deleted). (33)
10.43		Amendment No. 1 to Isis Pharmaceuticals Inc. 2000 Employee Stock Purchase Plan. (33)
10.44		Loan and Security Agreement dated December 15, 2003 between the Registrant and Silicon Valley Bank, including the related negative pledge agreement.
10.45		Grant letter dated October 31, 2003 from the Singapore Economic Development Board to ISIS Pharmaceuticals Singapore Pte Ltd (with certain confidential information deleted)
10.46		Development agreement dated on November 12, 2004 between the Registrant and RoboDesign International, Inc. (with certain confidential information deleted)
14.1		Registrant's Code of Ethics and Business Conduct
21.1		List of Subsidiaries for the Registrant.
23.1		Consent of Ernst & Young LLP, Independent Auditors.
24.1		Power of Attorney. Reference is made to page 57.
31.1		Certification by Chief Executive Officer Pursuant to 18 U.S.C. Section 1350 as Adopted Pursuant to Section 302 of the Sarbanes-Oxley Act of 2002. (from Q3 03)
31.2		Certification by Chief Financial Officer Pursuant to 18 U.S.C. Section 1350 as Adopted Pursuant to Section 302 of the Sarbanes-Oxley Act of 2002. (from Q3 03)
32.1		Certification Pursuant to 18 U.S.C. Section 1350 as Adopted Pursuant to Section 906 of the Sarbanes-Oxley Act of 2002.
99.2		Form of Confidentiality Agreement.(11)

				December 31,
				2003			2002
ASSETS
Current assets:
	Cash and cash equivalents			$	33,117		$	101,856
	Short-term investments				182,387			187,497
	Contracts receivable				2,657			14,906
	Inventory				13,995			11,090
	Other current assets				7,405			4,831

		Total current assets			239,561			320,180
Property, plant and equipment, net					34,790			59,094
Licenses, net					28,363			30,749
Patents, net					22,374			18,904
Deposits and other assets					8,479			9,186
Long-term investments					1,375			570

			Total assets	$	334,942		$	438,683

LIABILITIES AND STOCKHOLDERS' EQUITY
Current liabilities:
	Accounts payable			$	3,720		$	5,524
	Accrued compensation				4,149			3,330
	Accrued liabilities				6,527			6,794
	Amount due to affiliates				—			5,193
	Current portion of long-term obligations				16,477			21,435
	Current portion of deferred contract revenue				14,684			33,674

		Total current liabilities			45,557			75,950
5¹/2% convertible subordinated notes					125,000			125,000
Long-term obligations, less current portion					88,397			67,893
Long-term deferred contract revenue, less current portion					8,810			14,363
Stockholders' equity:
	Series B Convertible Exchangeable 5% Preferred stock, $.001 par value; 16,620 shares authorized, 12,015 issued and outstanding at December 31, 2003 and 2002				12,015			12,015
	Accretion of Series B Preferred stock dividends				2,560			1,866
	Common stock, $.001 par value; 100,000,000 shares authorized 55,557,253 and 55,215,785 shares issued and outstanding at December 31, 2003 and 2002, respectively				56			55
	Additional paid-in capital				604,948			602,101
	Deferred compensation				(294	)		(59	)
	Accumulated other comprehensive income (loss)				3,476			(608	)
	Accumulated deficit				(555,583	)		(459,893	)

		Total stockholders' equity			67,178			155,477

			Total liabilities and stockholders' equity	$	334,942		$	438,683

		Preferred stock
		Preferred stock								Common stock									Accumulated other comprehensive income/(loss)
Description							Dividend Accretion			Common stock			Additional paid in capital			Deferred compensation						Accumulated deficit			Total Stockholders' Equity
Description		Shares		Amount			Dividend Accretion			Shares	Amount		Additional paid in capital			Deferred compensation						Accumulated deficit			Total Stockholders' Equity
Balance at December 31, 2000		132		$	24,030		$	1,634		40,086	$	40	$	352,854		$	(858	)	$	126		$	(311,460	)	$	66,366
Comprehensive Loss
	Net loss applicable to common stock	—			—			—		—		—		—			—			—			(75,131	)		(75,131	)
	Change in unrealized gains and (losses)	—			—			—		—		—		—			—			534			—			534

Comprehensive loss		—			—			—		—		—		—			—			—			—			(74,597	)

Dividends accrued on preferred stock		—			—			1,299		—		—		—			—			—			—			1,299
Deferred compensation		—			—			—		—		—		3,960			(3,960	)		—			—			—
Issuances of common stock, net of offering costs		—			—			—		12,760		13		218,213			—			—			—			218,226
Options exercised and Employee stock purchase plan		—			—			—		904		1		7,231			—			—			—			7,232
Compensation relating to the granting of options		—			—			—		—		—		—			4,573			—			—			4,573

Balance at December 31, 2001		132		$	24,030		$	2,933		53,750	$	54	$	582,258		$	(245	)	$	660		$	(386,591	)	$	223,099

Comprehensive Loss
	Net loss applicable to common stock	—			—			—		—		—		—			—			—			(73,302	)		(73,302	)
	Change in unrealized gains and (losses)	—			—			—		—		—		—			—			(1,268	)		—			(1,268	)

Comprehensive loss		—			—			—		—		—		—			—			—			—			(74,570	)

Dividends accrued on preferred stock		—			—			1,060		—		—		—			—			—			—			1,060
Deferred compensation		—			—			—		—		—		(3,188	)		3,188			—			—			—
Options exercised and employee stock purchase plan		—			—			—		683		1		5,139			—			—			—			5,140
Compensation benefit relating to the granting of options		—			—			—		—		—		—			(3,002	)		—			—			(3,002	)
Issuance of common stock										126				3,750												3,750
Conversion of preferred stock into common stock		(120	)		(12,015	)		(2,127	)	657		—		14,142			—			—			—			—

Balance at December 31, 2002		12		$	12,015		$	1,866		55,216	$	55	$	602,101		$	(59	)	$	(608	)	$	(459,893	)	$	155,477

Comprehensive Loss
	Net loss applicable to common stock	—			—			—		—		—		—			—			—			(95,690	)		(95,690	)
	Change in unrealized gains and (losses)	—			—			—		—		—		—			—			4,084			—			4,084

Comprehensive loss		—			—			—		—		—		—			—			—			—			(91,606	)

Dividends accrued on preferred stock		—			—			694		—		—		—			—			—			—			694
Deferred compensation		—			—			—		—		—		1,148			(1,148	)		—			—			—
Options exercised and employee stock purchase plan		—			—			—		341		1		1,699			—			—			—			1,700
Compensation relating to the granting of options		—			—			—		—		—		—			913			—			—			913

Balance at December 31, 2003		12		$	12,015		$	2,560		55,557	$	56	$	604,948		$	(294	)	$	3,476		$	(555,583	)	$	67,178

2004	$	13,010
2005		80,609
2006		12,145
2007		6,690
2008		6,924
Thereafter		125,000

Total	$	244,378

		Number of Shares		Price Per Share	Weighted Average Price Per Share
Outstanding at December 31, 2000		7,671		$3.57 to $18.63	$	9.21
	Granted	1,893		$8.38 to $26.65
	Exercised	(817	)	$3.57 to $18.00
	Terminated	(527	)	$6.19 to $21.50

Outstanding at December 31, 2001		8,220		$3.75 to $26.65	$	9.88
	Granted	2,197		$6.87 to $22.19
	Exercised	(505	)	$4.00 to $17.88
	Terminated	(656	)	$5.38 to $24.17

Outstanding at December 31, 2002		9,256		$3.75 to $26.65	$	11.34
	Granted	2,724		$3.12 to $ 7.85
	Exercised	(35	)	$4.00 to $6.81
	Terminated	(3,734	)	$3.75 to $26.65

Outstanding at December 31, 2003		8,211		$3.12 to $26.65	$	8.66

	Options Outstanding
	Options Outstanding				Options Exercisable
		Weighted Average Remaining Contractual Life			Options Exercisable
Range of Exercise Price	Number Outstanding As of 12/31/03	Weighted Average Remaining Contractual Life	Weighted Average Exercise Price		Number Exercisable As of 12/31/03	Weighted Average Exercise Price
$3.12 - $4.98	519	4.73	$	4.11	285	$	3.99
$5.01 - $5.99	1,384	5.74	$	5.23	73	$	5.51
$6.02 - $6.95	2,754	7.00	$	6.77	1,703	$	6.78
$7.03 - $8.90	476	7.90	$	8.08	219	$	8.08
$9.01 - $9.95	829	7.08	$	9.58	590	$	9.59
$10.00 - $10.97	375	6.32	$	10.32	315	$	10.30
$11.00 - $12.94	1,009	5.56	$	12.23	868	$	12.33
$13.00 - $14.98	206	3.95	$	13.69	197	$	13.66
$15.00 - $17.95	312	7.72	$	16.45	158	$	16.29
$18.01 - $20.21	98	5.57	$	18.61	74	$	18.48
$21.05 - $26.65	249	7.99	$	21.25	130	$	21.25

	8,211				4,612

				GeneTrove Severance Cost for Involuntary Employee Terminations			GeneTrove Write- Down of Intellectual Property			Affinitak Severance Cost for Involuntary Employee Terminations			Total
Balance at December 31, 2001				$	—		$	—		$	—		$	—
	Reserve established				768			605			—			1,373
		Utilization of reserves:
			Cash		(379	)		—			—			(379	)
			Non-cash		—			(605	)		—			(605	)

Balance at December 31, 2002				$	389		$	—		$	—		$	389

	Reserve established				—			—			1,803			1,803
		Utilization of reserves:
			Cash		(389	)		—			(1,803	)		(2,192	)
			Non-cash		—			—			—			—

Balance at December 31, 2003				$	—		$	—		$	—		$	—

			December 31,
			2003			2002
Balance Sheet:
Assets
	Cash and cash equivalents		$	6		$	9

		Total assets	$	6		$	9

Liabilities and Stockholders' Equity
	Amounts due to affiliates		$	—		$	6,537
	Common stock, $1.00 par value; 12,000 authorized, issued and outstanding at December 31, 2003 and 2002			12			12
	Additional paid-in capital			50,129			39,388
	Accumulated deficit			(50,135	)		(45,928	)

		Total liabilities and stockholders' equity	$	6		$	9

Results of Operations:
	Revenue		$	—		$	—
	Research and development expenses			4,207			10,660
	Amortization of acquired license			—			1,250

		Total operating expenses		4,207			11,910

		Net loss	$	(4,207	)	$	(11,910	)