BUSINESS

Overview

        We are a clinical-stage biopharmaceutical company focused on the discovery, development and commercialization of novel small molecule therapeutics for oncology, inflammatory diseases and other unmet medical needs. We have built our product candidate portfolio through internal discovery and the in-licensing of novel cancer therapeutics. We are advancing our product candidates through in-house research and development efforts and strategic collaborations with leading pharmaceutical and biopharmaceutical companies. We believe the quality and breadth of our product candidate pipeline, platform technology, strategic collaborations and scientific team will enable us to become a fully integrated biopharmaceutical company with a diversified portfolio of novel therapeutics for major diseases.

        We are advancing three proprietary oncology product candidates, SNS-595, SNS-032 and an Aurora kinase inhibitor. All three are novel inhibitors of the cell division process, known as cell-cycle inhibitors, intended for the treatment of cancer. We are currently conducting two Phase I clinical trials with our lead product candidate, SNS-595, and we expect to commence up to two Phase II clinical trials in various types of solid tumors and one Phase Ib clinical trial in certain leukemias in the second half of 2005. In addition, in 2006 we intend to commence Phase II clinical trials to evaluate SNS-595 as a stand-alone therapy and Phase Ib clinical trials to evaluate SNS-595 in combination with standard treatments in additional tumor types. In preclinical studies, SNS-595 has demonstrated broad anti-tumor activity. We plan to commence a Phase I clinical trial with our second most advanced product candidate, SNS-032, in the second half of 2005. This compound was in-licensed from BMS in April 2005. In addition, we expect to continue preclinical studies and to select a development candidate for our other product candidate, an Aurora kinase inhibitor, in 2005 and to file an IND and commence Phase I clinical trials in 2006. We believe that small molecule Aurora kinase inhibitors have the potential to limit the growth of multiple tumor types without causing significant peripheral neuropathy. We have worldwide development and commercialization rights to SNS-595, SNS-032 and our Aurora kinase inhibitors program. We may in the future enter into collaborations to maximize the commercial potential of these programs.

        We currently have five strategic collaborations with three leading pharmaceutical and biopharmaceutical companies from which, as of December 31, 2004, we had received an aggregate of approximately $55 million in the form of stock purchase proceeds, fees and loans. We have two separate collaborations with Biogen Idec in oncology and in inflammatory diseases. The oncology program with Biogen Idec is focused on developing multiple kinase inhibitors, including our Raf kinase inhibitor, for which we have an option to co-develop and co-promote on a worldwide basis. We also work with Biogen Idec on the development of small molecule inhibitors of TNF. We are collaborating with Johnson & Johnson PRD on the development of product candidates for the treatment of inflammatory diseases from our Cathepsin S inhibitors program. We have two separate collaborations with Merck to develop therapeutics for Alzheimer's disease and for viral diseases. In the last year, each of our current collaboration partners has either extended its existing collaboration or entered into a new collaboration with our company. We believe that our strategic collaborations will enable us to leverage and expand our internal development capabilities, manage our cash expenditures and diversify risk across our pipeline.

        We have developed a proprietary method of discovering drugs in pieces, or fragments. We call this fragment-based discovery approach "Tethering." We combine Tethering with other drug discovery tools, such as structure-based design and medicinal chemistry, to discover and develop novel therapeutics for major diseases. Tethering allows us to screen drug fragments based on binding properties rather than function, which we believe enables us to identify compounds that may not be discovered through conventional methods of drug discovery. We believe that this capability allows us to efficiently design

43

product candidates that bind to sites or regions on a specific protein not readily accessed by other discovery methods. Tethering is applicable to most proteins, and we have used Tethering on over 15 different protein targets to date.

Strategy

        We are focused on discovering, developing and commercializing small molecule therapeutics for oncology, inflammatory diseases and other unmet medical needs. The key elements of our strategy are as follows:

•

Focus on small molecules with differentiated therapeutic benefits.     We discover and develop novel small molecule drugs designed to inhibit a specific molecular target, such as Aurora or Raf kinase. We also develop small molecule drugs, such as SNS-595 and SNS-032, that we believe have desirable properties or that target disease pathways not fully exploited by existing therapeutics. We believe either approach may result in drugs with improved therapeutic benefits as compared to existing drugs.



•

Maximize the value of our pipeline of product candidates through internal development and strategic collaborations.     We are advancing a diversified portfolio of novel therapeutics through internal development and strategic collaborations. We plan to retain all or a portion of the U.S. commercial rights to some of our novel cancer therapeutics. We may enter into future collaborations for late-stage development and commercialization of these cancer therapeutics. We intend to continue to collaborate with leading pharmaceutical and biopharmaceutical companies on other programs at the discovery or early research stage to leverage and expand our internal development capabilities, manage our cash expenditures and diversify risk across our pipeline. We have retained the economic rights to our three most advanced programs, SNS-595, SNS-032 and our Aurora kinase inhibitors, while also entering into multiple strategic collaborations for other programs.



•

Expand our portfolio of product candidates through our internal drug discovery engine and in-licensing.     We intend to leverage our proprietary fragment-based Tethering drug discovery approach, in combination with other drug discovery tools, such as structure-based design and medicinal chemistry, to expand our portfolio of product candidates for oncology, inflammatory diseases and other unmet medical needs. We may further augment our internal discovery efforts by in-licensing novel therapeutics. For example, we in-licensed SNS-595 and SNS-032, and discovered our Aurora kinase inhibitors through internal drug discovery efforts. We believe that this combination approach will enable us to accelerate the expansion of our portfolio of product candidates.

44

Overview of Cancer Market and Therapeutics

Market

        Cancer is the second leading cause of death in the United States, with 570,000 deaths and 1.4 million new cases estimated in 2005, according to the American Cancer Society. Cancers can be divided broadly into two groups: solid tumor cancers that affect organs in the body, such as the lungs and colon; and hematological, or blood-borne, cancers, such as leukemia. The American Cancer Society estimates that solid tumor cancers accounted for approximately 509,000 cancer-related deaths in 2004 and will account for approximately 1.3 million, or 92%, of new cases diagnosed in 2005.

45

Existing Drug Treatments

        Cancer is characterized by uncontrolled cell growth. Cell growth and function are controlled by proteins that communicate and relay signals within cells. In normal cell proliferation, when a cellular signaling pathway is activated, or "on," it sends a signal telling the cell to grow and divide. When a component of a signaling pathway is mutated, the signal may not turn "off" or it may be constantly "on," causing the cell to continuously reproduce itself, resulting in a tumor.

        The goals of cancer therapy are to cure the patient and, in the absence of cure, to improve the quality of life and extend the life expectancy of the patient. The most common form of pharmaceutical treatment for cancer is cytotoxic therapeutics, which are designed to target and kill rapidly proliferating cells. This process is referred to as cell-cycle arrest, which leads to cell death. Cytotoxic drugs include irinotecan, doxorubicin, taxanes and other inhibitors of cellular proliferation. In addition, therapies designed to hit a specific molecular target, such as Gleevec and Tarceva, may be used in combination with or as alternatives to cytotoxic therapies.

        Due to the genetic diversity among tumors, a combination of drug therapies is generally used to treat any given tumor type, and many patients progress rapidly through all available therapies. Despite the introduction of a number of new therapeutics over the last few years, there is significant demand for new drugs that, by themselves or in combination with existing therapies, can significantly improve the quality of life and extend the life expectancy of cancer patients.

        Although cytotoxic therapies are widely used, their mechanism of action targets all proliferating cell populations, not just cancer cells, and therefore may result in significant side effects, including immune system compromise known as myelosuppression, nausea, vomiting, diarrhea, sores in the mouth and the digestive tract known as mucositis, hair loss, peripheral nerve cell death known as peripheral neuropathy and damage to the heart known as cardiotoxicity. Cytotoxic drugs may have a narrow "therapeutic window" between efficacy and toxicity, which means there is only a small variance between a therapeutic dose and a toxic dose. Proper dosing is a challenge with drugs that have a narrow therapeutic window because dosing below the therapeutic window results in the patient receiving a sub-therapeutic exposure to the drug, whereas dosing above the therapeutic window results in exposing the patient to a toxic level of the drug. Dosing can be particularly challenging with cytotoxic drugs because a number of these have demonstrated highly variable PK. PK variability results in differences in drug exposure among patients and, in some cases, in the same patient even at the same dose. As a result, it is difficult for physicians to determine the proper therapeutic dose for a patient at any particular time, and patients are frequently under-dosed or over-dosed.

        Notwithstanding their limitations, it is widely believed that cytotoxic therapeutics will continue to be a mainstay of cancer therapy. We believe significant commercial opportunities exist for cytotoxic drugs that act by a novel mechanism with a more manageable side effect profile, both as single-agent therapies in patients with resistant, or refractory, disease or in combination with current established therapies. We believe that our SNS-595 program is well positioned to take advantage of these opportunities.

        We also believe there is a need for novel molecularly-targeted therapies that can be used in combination with, or as alternatives to, cytotoxic therapies to improve the outcome of cancer treatment. Molecularly-targeted therapies may have fewer unwanted side effects because they are less likely to affect cell activity unrelated to cancer. We believe that these targeted therapeutics, such as Gleevec, Tarceva and Velcade, are likely to capture an increasing share of the cancer market and even contribute to its growth. For example, Novartis reported 2004 sales of Gleevec at $1.6 billion, a 45% increase from 2003. While these targeted therapeutics have demonstrated clinical benefit in some patients, there remains a significant unmet medical need in cancer patients with other tumor types or resistant tumors. We believe that SNS-032 and our Aurora and Raf kinase inhibitors, which also target specific molecules, are well positioned to benefit from these trends in cancer therapy.

46

SNS-595 Program

        SNS-595 is a novel cytotoxic drug that we believe represents a new class of anti-tumor drugs. SNS-595 is in a chemical class known as naphthyridine analogues. Although naphthyridine analogues have been used as antibiotics and have been demonstrated to be safe in human treatment, no members of this chemical class have been approved for the treatment of cancer. SNS-595 is a broadly active cell-cycle inhibitor that we believe works in a different way than any other cancer therapy, known as a novel mechanism of action, and induces the arrest of proliferating cancer cells immediately before the cell divides into two new cells, leading to cell death. We obtained worldwide development and commercialization rights to SNS-595 from Dainippon through a license agreement in 2003.

Opportunity

        We believe SNS-595 has the following characteristics:

•

broad anti-tumor activity as demonstrated in multiple animal models;



•

favorable PK profile that may enable more predictable dosing within the therapeutic window;



•

ability for physicians to combine with other cancer therapies with minimal drug-drug interactions;



•

novel mechanism of action, which may result in activity in multi-drug-resistant cancers;



•

convenient intravenous dosing that can be completed in a matter of minutes in an outpatient setting and on a schedule consistent with existing chemotherapy schedules;



•

anti-tumor activity through oral administration in animal models;



•

absence of cardiotoxicity or peripheral neuropathy observed in animals and humans to date; and



•

relative ease and availability of manufacturing.

Clinical Trials

        In June 2004, we began the first of two Phase I clinical trials evaluating SNS-595 in groups of patients with advanced solid tumors. In these clinical trials, we are exploring doses and schedules of administration in preparation for Phase II clinical trials designed to evaluate initial clinical efficacy. As of March 31, 2005, we had administered 101 cycles of treatment to a total of 35 patients in our Phase I clinical trials. As of the date of this prospectus, we have observed stable disease lasting at least six cycles in four patients and one minor response in an ovarian cancer patient. Based on our Phase I clinical trials to date, dose limiting toxicity appears to be myelosuppression in the absence of any other dose limiting toxicities. We are conducting these clinical trials at three leading medical centers and expect to treat 30 to 40 patients in each trial.

        The first Phase I clinical trial of SNS-595 is a single-dose administration followed by a 21-day observation period constituting one cycle. Patients participating in this trial may receive up to six cycles of treatment. We plan to complete this trial in the first half of 2005.

        The second Phase I clinical trial, which began in October 2004, is evaluating the administration of three weekly doses of SNS-595 followed by a 14-day observation period in each 28-day cycle. Patients participating in this trial may receive up to six cycles of treatment. We plan to complete this trial by the end of the third quarter of 2005.

        Patients from both Phase I clinical trials whose disease stabilizes while on treatment or who exhibit a partial or complete response after six cycles of treatment may participate in a continuation trial following the completion of the initial trial.

47

Preclinical Studies

        SNS-595 has been the subject of numerous preclinical studies conducted by us and Dainippon. In 2000, Dainippon reported data from studies of SNS-595 in various mouse models. In these models, SNS-595 demonstrated superior therapeutic activity compared to several widely used cytotoxic drugs. Specifically, SNS-595 showed 98.9% to 99.3% inhibition of the growth of established tumors in three different solid tumor mouse models when compared to control animals. Statistical analysis showed that the chance that these differences in tumor size following treatment with SNS-595 were not due to the effect of the drug was less than 5%, known as a probability, or p-value, of less than 0.05. In some cases, complete tumor regressions were observed. In these models, a number of marketed cytotoxic drugs, including paclitaxel and irinotecan, had significantly less activity than SNS-595. In other mouse models involving the study of human tumors in mice with compromised immune systems, SNS-595 has shown broad activity in all models tested and inhibited the growth of established tumors by more than 80% in 14 of 17 tumor lines evaluated, including several drug-resistant tumor lines. This data is statistically significant with p-values of less than 0.05. While we do not know whether SNS-595 will demonstrate comparable activity in humans, we believe that these data are encouraging and support further development.

        SNS-595 has also demonstrated excellent PK properties in animals and humans to date. SNS-595 demonstrated, in all species tested, precise and reproducible PK results and low inter-individual PK variability. We believe this is because levels of SNS-595 in the blood are less affected by the metabolic processes than many other cytotoxic drugs. As a result, we expect to see similar drug exposures across all patients treated with SNS-595, a consistent adverse event profile and fewer unexpected toxicities.

        The potential toxicity of single and repeated doses of SNS-595 has been explored in a number of animal studies that suggest the mechanism-based dose-limiting toxicities in humans receiving SNS-595 may be similar to some of those observed in approved cytotoxic agents, including temporary toxicity to bone marrow cells, the gastrointestinal system and other systems with rapidly dividing cells. These toxicities are usually reversible and can be adequately managed by experienced medical oncology practitioners.

Development Plan

        We plan to commence up to two Phase II clinical trials in various types of solid tumors and one Phase Ib clinical trial in certain leukemias in the second half of 2005. We currently plan to design these trials to evaluate SNS-595 as a stand-alone therapy in patients with small cell and non-small cell lung cancers and acute myelocytic leukemia. In addition, in 2006 we intend to commence Phase II clinical trials to evaluate SNS-595 as a stand-alone therapy and Phase Ib clinical trials to evaluate SNS-595 in combination with standard treatments in additional tumor types. We may modify the tumor types selected for these Phase II and Phase Ib clinical trials based on the anti-tumor activity observed in animal models, tumor types showing responses in our Phase I clinical trials and strategic regulatory and market considerations.

SNS-032 Program

        SNS-032 is a novel targeted inhibitor of certain cyclin-dependent kinases including CDK2, CDK7 and CDK9. Kinases are proteins found in cells that are critical in the communication and relay of signals to promote cell growth or function. Alterations in several proteins that control CDK2 have been shown to be associated with poor prognosis and survival in several cancer types, including breast, ovarian and lung cancers. We believe that prolonged or repeated exposure to SNS-032 will inhibit CDK2, as well as CDK7 and CDK9, and that this inhibition can be a beneficial cancer treatment. We also believe that SNS-032 may be beneficial in addition to or in combination with other cytotoxic chemotherapeutic agents that act by other mechanisms in the cell cycle. We obtained worldwide rights

48

to develop and commercialize SNS-032 for diagnostic and therapeutic applications from BMS under a license agreement executed in April 2005.

Opportunity

        We believe that SNS-032 has the following characteristics:

•

novel target and novel mechanism of action;



•

opportunity to be first-in-class inhibitor of CDK2, CDK7 and CDK9;



•

broad anti-tumor activity as demonstrated in preclinical models of rodent and human tumors;



•

toxicology consistent with its proposed activity as a cytotoxic anticancer drug;



•

potential for both oral and IV formulations;



•

potential for additive or synergistic activity when SNS-032 is administered in combination with cytotoxic chemotherapeutic agents that act by other mechanisms;



•

relative ease and availability of manufacturing; and



•

complementary to our existing programs.

Clinical Trials

        BMS has conducted three Phase I dose-escalation clinical trials evaluating the safety and tolerability of SNS-032 at three different dosing regimens in approximately 125 patients with refractory solid tumors. The first Phase I clinical trial was initiated in August 2001 and evaluated the administration of a one-hour IV infusion every three weeks. The second Phase I clinical trial was initiated in March 2002 and evaluated the administration of a 24-hour IV infusion every three weeks. The third Phase I clinical trial was initiated in July 2002 and evaluated the administration of a one-hour IV infusion every week, with the third dose given orally to measure the availability of the drug in the body when given orally, known as oral bioavailability. We believe that BMS did not complete the second and third clinical trials.

        In these clinical trials, SNS-032 was generally well tolerated and demonstrated toxicity consistent with a cytotoxic anticancer agent. Observed toxicities included QTc prolongation, decreased white blood cell count and liver toxicity. While no objective tumor responses were observed, a number of patients experienced stable disease and minor responses. We believe that patient exposure to the drug may have been suboptimal or inadequate to obtain tumor responses and that more frequent or more dose-intensive regimens will improve the likelihood of achieving responses with SNS-032 as a stand-alone agent.

Preclinical Studies

        In previous preclinical studies conducted by BMS, SNS-032 was shown to be a broadly active inhibitor of the proliferation of tumor cell lines. We believe that the observed cell death caused by this inhibitor is the result of cell cycle arrest. In addition, SNS-032 demonstrated broad anti-tumor activity in multiple mouse and human tumor models, including breast, ovarian, colorectal and skin cell cancers. SNS-032 has also shown synergistic activity in preclinical models when combined with currently approved anti-tumor drugs, including Gemcitabine and Cisplatin.

        BMS also conducted a series of PK and metabolism studies with SNS-032. SNS-032 demonstrated predictable drug exposure between species and was shown to be orally bioavailable in mice, rats, dogs and humans. SNS-032 was found to be a weak inhibitor of human drug metabolizing enzymes,

49

suggesting that the potential for negative side effects resulting from drug-drug interactions when combined with other therapeutics could possibly be low.

        The toxicology of SNS-032 is consistent with the toxicology of other cytotoxic drugs. The dose-limiting toxicities in pre-clinical studies were myelosuppression and gastro-intestinal toxicity.

Development Plan

        We plan to commence a new Phase I clinical trial in the second half of 2005. We currently plan to design this study to evaluate the safety and tolerability of frequent, repeated exposure to SNS-032 in patients with refractory solid tumors, such as lung, breast and ovarian tumors, and in chronic lymphocytic leukemia. Patients would receive an escalating dose of SNS-032 as an IV infusion to assess the safety and tolerability of a dose and schedule that we would plan to use in subsequent Phase II clinical trials in a variety of tumors.

        Concurrently, we intend to develop an oral formulation of SNS-032, which may prove to be more convenient for patients, and to discover additional CDK inhibitors as alternative drug candidates.

Other Oncology Kinase Programs

        We are applying Tethering in several programs to discover and develop other novel kinase inhibitors for the treatment of cancer. Kinases are proteins found in cells that are critical in the communication and relay of signals to promote cell growth or function. In normal cell proliferation, when a cellular signaling pathway is activated, or "on," it sends a signal telling the cell to grow and divide. When a component of a signaling pathway involving a kinase is mutated, the signal may not turn "off" and thus may be constantly "on," causing the cell to continuously reproduce itself. This unregulated growth is a principal characteristic of cancer cells. It is widely believed that the signaling pathway plays an integral role in the growth of some tumor types. We believe that by inhibiting a kinase in a specific overactive pathway, the pathway can be turned "off," restoring normal signaling.

Technology/Scientific Overview

        There is significant pharmaceutical industry interest in kinases as key points of intervention for treating disease. Many strategies are used in the pharmaceutical industry to discover kinase inhibitors for drug development, including high-throughput functional screening of kinase-regulated pathways.

        There are 518 known human kinases, many of which are known to be involved in cancer and other diseases. Most small molecule kinase inhibitors bind to a main site common among many kinases, thereby affecting multiple kinases rather than only those involved in the targeted disease. As a result, it has been difficult to discover small molecule kinase inhibitors that bind only to the kinase involved in the targeted disease process and not to other kinases. The ability to target a single kinase is referred to as specificity. Binding without specificity may lead to unwanted toxicity problems.

        A small but growing number of compounds, including Gleevec, inhibit the targeted kinase with greater specificity by binding not only the main site but also a nearby region called the variable binding region. Each kinase has a different variable binding region. As a result, we believe inhibitors that also bind to the variable binding region will bind with greater specificity to the kinase of interest, and we use Tethering to identify these kinase inhibitors. We believe that these specific inhibitors may be associated with reduced toxicity.

Aurora Kinase Inhibitors Program

        Our Aurora kinase inhibitors program is our most advanced kinase inhibitors program. We believe that Aurora kinase inhibitors represent an area of interest in the pharmaceutical industry and within the cancer treatment community.

50

        Aurora kinases are key enzymes involved in cell growth and division and play an essential role in the abnormal growth and proliferation of tumor cells. Aurora kinases are known to be overexpressed in a number of tumor types, including colon cancer, breast cancer and leukemia. Aurora kinase inhibition retards cell proliferation and limits tumor growth by initiating programmed cell death. Because Aurora kinase is more highly expressed in active cells rather than resting cells, we believe that Aurora kinase therapies may selectively target cancer cells, which are actively dividing, over cells in a normal or resting state, which may lead to reduced toxicities.

        The goal of this program is to develop novel Aurora kinase inhibitors that exhibit broad activity in tumors with superior drug-like properties that do not cause significant peripheral neuropathy. We are conducting preclinical studies, including animal studies. One compound in our lead series was dosed on a once-a-week schedule for three weeks in colon and pancreatic solid tumor mouse models. In these models, our compound caused 80% or greater inhibition of tumor growth in ways consistent with inhibition of Aurora A and B with minimal observed toxicity. These results were statistically significant, with a p-value of less than 0.05. We believe these results are comparable to those observed in the same tumor models for VX-680, an Aurora inhibitor in development by Vertex Pharmaceuticals and Merck. We are currently using the data from these preclinical experiments to optimize the activity and PK properties of our Aurora kinase inhibitors. We anticipate selecting a development candidate in 2005 with the goal of filing an IND and commencing Phase I clinical trials in 2006. We have worldwide rights to commercialize any drugs resulting from our Aurora kinase inhibitors program.

Raf Kinase Inhibitors Program

        We are developing our Raf kinase inhibitors program in collaboration with Biogen Idec. We believe that Raf kinase inhibitors represent an area of interest in the pharmaceutical industry and within the cancer treatment community.

        Raf kinase is an enzyme in the Ras pathway, a signaling pathway important to cell proliferation. The Ras pathway is believed to be abnormally activated in many human cancers by various mechanisms. In approximately 15% of human cancers, a Ras gene is activated by mutation. We believe that several inhibitors of kinases in the Ras pathway have shown evidence of clinical activity in clinical trials.

        The goal of this program is to develop Raf kinase inhibitors with improved pharmaceutical properties as compared to other Raf kinase inhibitors in development. We applied Tethering to discover highly specific and potent Raf kinase inhibitors. We and Biogen Idec are conducting preclinical studies and anticipate selecting a development candidate in 2006. We expect Biogen Idec to file an IND and commence Phase I clinical trials in 2007. We have an option to co-develop and co-promote any drugs developed through this program on a worldwide basis.

Other Kinase Inhibitors Programs

        As part of our collaboration with Biogen Idec, we are applying Tethering to discover novel small molecule leads that inhibit up to five additional oncology kinase targets. We and Biogen Idec are working together on the identification, optimization and development of inhibitor drugs for these kinases. We are also working on the identification and optimization of kinase inhibitor drugs outside of our collaboration with Biogen Idec.

Inflammatory Diseases Programs

Cathepsin S Inhibitors Program

        Cathepsin S is an enzyme involved in the activation of T-cells. Inappropriate activation of T-cells may lead to some inflammatory diseases, such as asthma, rheumatoid arthritis, multiple sclerosis,

51

psoriasis and Crohn's disease. In collaboration with Johnson & Johnson PRD, we are applying Tethering to discover small molecule inhibitors of Cathepsin S. We intend to develop these inhibitors into drugs for the treatment of major inflammatory diseases. We believe that small molecule Cathepsin S inhibitors would have the advantages of a novel mechanism of action, ease of oral administration and ease of manufacturing. Johnson & Johnson PRD holds worldwide rights to commercialize any drugs resulting from this program.

TNF Family Inhibitors Program

        TNF and related cytokine family members are proteins involved in inflammatory diseases, including rheumatoid arthritis, psoriasis and Crohn's disease. While inhibition of TNF has been validated as a treatment for inflammatory diseases, all of the currently marketed TNF inhibitors are injectable biologics. We believe that small molecule inhibitors would have advantages, such as ease of oral administration and ease of manufacturing, but discovery of small molecule TNF inhibitors has been historically challenging. In collaboration with Biogen Idec, we are applying Tethering to discover small molecule TNF family inhibitors. Examples of TNF inhibitors include Enbrel and Humira, which had reported sales of approximately $1.3 billion and $600 million, respectively, for the nine month period ended September 30, 2004. Biogen Idec holds worldwide rights to commercialize any drugs resulting from this program.

Other Programs

BACE Inhibitors for Alzheimer's Disease

        In collaboration with Merck, we are applying Tethering to identify and optimize inhibitors of BACE, an important enzyme target in Alzheimer's disease. BACE, or beta-secretase, is involved in the formation of A-beta peptide, which is the predominant substance in the plaques found in the brains of Alzheimer's patients and believed to contribute to their disease. Merck holds worldwide rights to commercialize any drugs resulting from this collaboration and is responsible for advancing these compounds into lead optimization, preclinical studies and clinical trials.

Anti-Viral Inhibitors Program

        We are collaborating with Merck to identify small molecule inhibitors of an anti-viral target by a novel mechanism. We are providing Merck with a series of small molecule compounds that we derived from Tethering that target a specific viral protein. Merck holds worldwide rights to commercialize any drugs resulting from this collaboration and is responsible for advancing these compounds into lead optimization, preclinical studies and clinical trials.

Rationale For Fragment-Based Drug Discovery

        We have developed a proprietary fragment-based drug discovery approach, called Tethering, that we use in combination with other drug discovery tools, such as structure-based design and medicinal chemistry, to discover and develop novel therapeutics for major diseases.

Limitations of Existing Drug Discovery Approaches

        Pharmaceutical discovery often begins with the hypothesis that a target protein in the body is involved in a certain disease and that compounds that block or inhibit the action of that target will provide therapeutic benefit. The search for these compounds typically starts by screening a collection of molecules to find "hits" that inhibit the target function. These hits are then improved through medicinal chemistry to create more advanced molecules that are tested in animal models of the disease to determine whether they provide therapeutic benefit. Molecules that test positively in animal models are optimized to have the necessary properties to become drugs and are ultimately tested in human clinical trials.

52

        Combinatorial chemistry has expanded the size of compound collections, and advances in automated high-throughput screening, or HTS, have enabled the screening of million-compound libraries. Despite these advances, HTS is limited by the number of complex, fully formed compounds that can practically be made and stored in a collection. Even very large collections represent only a small fraction of the compounds that could be made. Thus, new approaches to searching the vast potential diversity of chemical compounds are highly desired. Another challenge with HTS is that it typically identifies compounds that bind to the main binding site of kinases. It can be difficult to find molecules that bind to the variable binding region of kinases using HTS. We believe that kinase inhibitors that bind to the variable binding region have significant potential therapeutic benefit.

        We believe fragment-based drug discovery offers an alternative strategy for identifying drug-like compounds that bind to proteins. In contrast with HTS where compounds are identified by measuring activity, fragment-based discovery involves the identification of multiple drug fragments by measuring binding. Individual fragments that bind to nearby sites on the protein are identified and combined to form a drug-like compound. By first determining which fragments interact with the protein in the area of interest and then combining them, we believe there is a higher likelihood of identifying novel compounds that will bind to the target. A key challenge of fragment-based drug discovery is how to identify which fragments bind to the protein. Fragments typically bind weakly and can be difficult to detect using conventional methods.

Our Drug Discovery Platform

        Tethering is our proprietary fragment-based drug discovery approach that we believe overcomes the limitations of existing conventional and fragment-based discovery methods. Tethering enables us to identify weak-binding fragments that would otherwise be difficult to detect. In Tethering, we first expose a target protein having a sulfur-containing amino acid named cysteine to a collection of specially designed fragments that also contain sulfur. Those fragments capable of binding to the target near the cysteine form a reversible chemical link known as a disulfide bond that stabilizes the binding of the fragment to the target protein. The formation of the disulfide bond results in an increase in the weight of the protein that allows us to identify the fragment using mass spectrometry. Although the disulfide bond helps to stabilize the weakly binding fragment to the protein, fragments that do not naturally bind to the protein are not detected.

        We create drug-like compounds by combining multiple fragments through a process called "Tethering with extenders." Once we have identified an initial fragment that binds to the target protein, we use this initial fragment as the basis for an extender to search for a companion fragment by the same process used to discover the initial fragment. Subsequently, the initial and companion fragments are combined into a single molecule and the attachment to the protein is removed. This ultimately generates a soluble drug-like compound that can be optimized through medicinal chemistry. In Tethering with extenders, the surface of the target protein is used as a mold to construct its own inhibitor. This approach can be applied to most protein targets.

53

1

A pool of fragments is screened against a protein target.

2

A fragment binds to the protein target.

3

Following a second screening, a new fragment binds to the protein target and the initial fragment, forming a drug-like compound.

4

The drug-like compound is released from the protein target for further optimization.

        By screening and identifying individual fragments that bind to the target and only combining the fragments that bind, we are able to significantly expand the number of drug-like compounds that can be evaluated.

        Our Tethering approach has formed the basis of all of our collaborations to date. Our collaboration partners have been attracted to our company and our Tethering approach because of its potential to identify novel drug-like compounds that are difficult to detect through other means. We have successfully applied Tethering to over 15 different protein targets, including various kinases, to produce drug-like compounds. We optimize the drug-like compound to produce drug candidates by integrating Tethering with medicinal chemistry and structure-based drug design and by introducing pharmacology, including absorption, distribution, metabolism and excretion tests, early in the drug discovery process.

Strategic Collaborations

        We currently have five strategic collaborations with three leading pharmaceutical and biopharmaceutical companies. Each collaboration is target specific and involves personnel from both our company and our collaboration partner working together. These alliances are designed to enable us to leverage and expand our internal development capabilities, manage our cash expenditures and diversify risk across our pipeline. Through our strategic collaborations, we are able to pursue more programs than we could fund on our own. As of December 31, 2004, we had received an aggregate of approximately $55 million in the form of stock purchase proceeds, fees and loans from our collaboration partners. We believe that approximately 175 scientists currently work on our programs and programs partnered with our company. Approximately one-half of these scientists are our employees and of those, approximately 50% are funded through our collaborations.

        In forming each of our strategic collaborations, we have agreed not to conduct certain research, independently or with any commercial third party, that is on the same target as that covered by the collaboration agreement. Some of our collaborations also significantly restrict our ability to utilize intellectual property derived from a collaboration for a purpose outside of the collaboration.

54

Biogen Idec (formerly Biogen, Inc)—TNF Family Research Collaboration

        In December 2002, we entered into a collaboration with Biogen Idec to apply Tethering to discover and develop small molecule modulators of up to four members of the TNF trimeric cytokine super-family plus up to two additional un-named targets. The primary focus of the program is to discover small molecule inhibitors of TNF, an important cytokine involved in the regulation of the human immune system for which only protein drugs, such as Enbrel, are available. During the research term of the agreement, we and Biogen Idec agreed to work together exclusively to identify and develop small molecule pharmaceutical compounds that inhibit a member of the TNF family as well as other collaboration targets, with the exception that either party may collaborate with a third party on a Phase II clinical trial or later stage compound against a collaboration target. Our exclusivity obligation continues for an additional year after the end of the research term. Biogen Idec holds worldwide rights to commercialize any drugs resulting from this program.

        Pursuant to this agreement we received a $3.0 million upfront technology access fee. In addition, Biogen Idec made a $6.0 million equity investment in our company. The agreement also provides for a maintenance fee payable to us of $357,500 per quarter, starting in April 2004 and continuing until the end of the initial research term, and a $4.0 million credit facility from which we may make 10 quarterly draws of $400,000 each, of which as of December 31, 2004 we had drawn an aggregate of $3.2 million. Both parties agreed to dedicate resources as provided in the research plan. To date, we have received payments totaling $7.6 million under this collaboration, including $3.2 million in loan proceeds. The initial research term is for 30 months, after which Biogen Idec has the option to extend the research term for one year upon payment of a technology maintenance fee and a commitment to provide research funding.

        We granted Biogen Idec a worldwide non-exclusive license to our intellectual property relating to Tethering with respect to specific collaboration targets and an exclusive license to our portion of the collaboration intellectual property for the commercialization of a small molecule product arising from the collaboration. Biogen Idec is required to pay up to $60.5 million in pre-commercialization milestones per compound, assuming the compound is approved for multiple indications, as well as royalty payments based on product sales. Royalty rates payable to us may be reduced if Biogen Idec is required to license additional intellectual property related to Tethering from one or more third parties in order to commercialize a collaboration product. Rights to collaboration products revert to us with a reverse royalty to Biogen Idec if Biogen Idec fails to use commercially reasonable and diligent efforts during development and commercialization of such products. Royalty obligations under the agreement continue on a country-by-country and product-by-product basis until the later of the date on which no valid patent claim relating to a product exists or 10 years from the date of first sale of the product.

        Our agreement with Biogen Idec will continue for so long as a product arising from the collaboration is the subject of an active development project or for so long as there is an obligation to pay royalties under the agreement. The agreement may be terminated earlier by Biogen Idec without cause upon 30 days' written notice, in which case each party will have exclusive rights to the compounds it solely invented during the collaboration and non-exclusive rights to jointly invented compounds, Biogen Idec will remain obligated to pay milestones and royalties to us, and we will owe a modest royalty to Biogen Idec. The agreement may also be terminated by either party for uncured breach or bankruptcy of the other party. If Biogen Idec terminates the agreement in connection with our breach or bankruptcy, it retains its licenses from us but receives a reduction in its milestones and royalty obligations. If we terminate the agreement for Biogen Idec's breach or bankruptcy, we will receive exclusive licenses from Biogen Idec and be obligated to make modest royalty payments to Biogen Idec.

55

Biogen Idec—Kinase Research, Development and Commercialization Collaboration

        In August 2004, we entered into a collaboration agreement with Biogen Idec to discover, develop and commercialize small molecule inhibitors of Raf kinase and up to five additional targets. The primary focus of the program is to discover small molecule inhibitors of kinases that play a role in oncology indications or in the regulation of the human immune system. During the research term, we and Biogen Idec agreed to work together exclusively to develop pharmaceutical compounds against collaboration targets with the exception that either party may collaborate with a third party on a Phase II clinical trial or later stage compound against a collaboration target. Our exclusivity obligation continues for an additional year after the end of the research term. We also agreed not to develop or commercialize any compound active against a collaboration target that is the subject of the agreement.

        Pursuant to this agreement, we received a $7.0 million upfront technology access fee. In addition, Biogen Idec made a $14.0 million equity investment in our company. To date, we have received payments totaling $22.7 million under this collaboration, including the $14.0 million equity investment. The initial research term is four years, and both parties agreed to dedicate the research personnel provided in the research plan. Biogen Idec has the option to extend the research term for up to two additional one-year periods upon payment of an additional technology access fee and a commitment to provide research funding. Biogen Idec will bear all costs related to this program for all targets through at least the completion of Phase I clinical trials, after which we have the right to participate in the co-development and co-promotion of up to two targets.

        We granted Biogen Idec a worldwide non-exclusive license to our intellectual property relating to Tethering with respect to specific collaboration targets and an exclusive license to our portion of the collaboration intellectual property for the commercialization of small molecule compounds that have a specified activity against collaboration kinases arising from the collaboration. Biogen Idec is required to pay up to $60.5 million in pre-commercialization milestones per target as well as royalty payments depending on product sales. Royalty payments may be increased if we exercise our option on co-development and co-promotion rights. Royalty rates payable to us will be reduced if Biogen Idec is required to license additional intellectual property related to Tethering from one or more third parties in order to commercialize a collaboration product. Rights to collaboration products revert to us with a reverse royalty to Biogen Idec if Biogen Idec fails to use commercially reasonable and diligent efforts during development and commercialization of co-funded products. If we do not exercise our co-funding option for a product directed at a target selected for further collaborative work, then Biogen Idec may pursue such target on its own. We also have a non-exclusive license, with the right to obtain an exclusive license, from Biogen Idec under joint collaboration intellectual property to develop and commercialize products against other kinase targets. We will owe royalty payments to Biogen Idec for sales of any such products. Royalty obligations under the agreement continue on a country-by-country and product-by-product basis until the later of the date on which no valid patent claim relating to a product exists or 10 years from the date of first sale of the product.

        Our agreement with Biogen Idec will continue for so long as a product arising from the collaboration is the subject of an active development project or for so long as there is an obligation to pay royalties under the agreement. The agreement may be terminated earlier by Biogen Idec without cause at any time before the second anniversary of the agreement upon six months' written notice or immediately upon written notice and payment of a termination fee. After the second anniversary of the agreement, Biogen Idec may terminate the agreement without cause upon 90 days' written notice. Either party may also terminate the agreement for the other party's uncured breach or bankruptcy. If Biogen Idec terminates the agreement early without cause or we terminate due to Biogen Idec's breach or bankruptcy, all co-funded products not approved for sale prior to termination will revert to us, and we will receive a reduction in the royalties we owe to Biogen Idec. If Biogen Idec terminates the agreement early due to our breach or bankruptcy, Biogen Idec will receive a reduction in the royalties

56

it owes to us. Many of the parties' other product rights are not substantially affected by early termination.

Johnson & Johnson PRD—Research, Development and Commercialization Collaboration

        In May 2002, we entered into a collaboration agreement with Johnson & Johnson PRD to discover, develop and commercialize small molecule inhibitors of Cathepsin S, an enzyme that is important in regulating an inflammatory response. During the period of the research term plus two years, we and Johnson & Johnson PRD agreed to work together exclusively to develop pharmaceutical compounds against Cathepsin S. Johnson & Johnson PRD retains the sole right to determine whether a product candidate enters development. Johnson & Johnson PRD holds worldwide rights to commercialize any drugs resulting from this program.

        The agreement provides for payment by Johnson & Johnson PRD to us of a technology access fee and research funding. To date, we have received payments totaling $5.0 million under this collaboration. The initial research term was two years, and Johnson & Johnson PRD had the option to extend the research term for up to two additional one-year periods with the same level of research funding. Johnson & Johnson PRD exercised its first option to extend the research term through May 2005, and in December 2004, Johnson & Johnson PRD extended the research term further to December 31, 2005. We do not expect that this collaboration will be extended beyond 2005.

        We granted Johnson & Johnson PRD a worldwide non-exclusive license to our intellectual property relating to Tethering on Cathepsin S and an exclusive license under the collaboration intellectual property for the commercialization of small molecule products arising from the collaboration. Patents and patent applications arising from the collaboration are owned by our company. Johnson & Johnson PRD is required to pay research and development milestones of up to $24.5 million well as royalty payments depending on product sales. Royalty rates payable to us may be reduced if Johnson & Johnson PRD is required to license additional intellectual property related to Tethering from one or more third parties in order to commercialize a collaboration product. Royalty obligations under the agreement continue on a country-by-country and product-by-product basis until the later of the date on which no valid patent claim relating to a product exists or 10 years from the date of first sale of the product.

        Our agreement with Johnson & Johnson PRD will continue for so long as a product arising from the collaboration is the subject of an active development project or for so long as there is an obligation to pay royalties under the agreement. Johnson & Johnson PRD may terminate the agreement earlier without cause after the end of the research term and upon six months' written notice, and either party may terminate the agreement earlier for the other party's uncured breach or bankruptcy. All early terminations extinguish Johnson & Johnson PRD's licenses from us. If Johnson & Johnson PRD terminates the agreement early without cause or if we terminate due to Johnson & Johnson PRD's breach or bankruptcy, Johnson & Johnson PRD will grant us certain exclusive licenses and transfer its regulatory filings to us, and we will be obligated to pay modest royalties to Johnson & Johnson PRD in return.

Merck—BACE Research, Development and Commercialization Collaboration

        In February 2003, we entered into a license and collaboration agreement with Merck to discover, develop and commercialize small molecule inhibitors of BACE, or beta secretase, an enzyme that is believed to be important for the progression of Alzheimer's disease. During the period of the research term plus one year, we and Merck agreed to work together exclusively to develop a pharmaceutical compound against the collaboration target, with the exception that Merck may acquire from a third party a compound that satisfies development candidate criteria specified in the agreement. Merck holds worldwide rights to commercialize any drugs resulting from this collaboration.

57

        The agreement provides for payment by Merck to us of a technology access fee and research funding. To date, we have received payments totaling $11.2 million under this collaboration. The initial research term is three years and both parties agreed to dedicate the resource funding provided in the research plan. Merck has the option to extend the research term for an additional one-year period with the same level of research funding.

        We granted Merck a worldwide, non-exclusive license to our intellectual property relating to Tethering on BACE and an exclusive license to a composition of matter patent and future intellectual property relating to BACE. Merck is required to pay research and development milestones of up to $90.3 million as well as royalty payments depending on product sales. Royalty rates payable to us may be reduced if Merck is required to license additional intellectual property from one or more third parties in order to commercialize a collaboration product or if a third party markets a version of the collaboration product. Royalty obligations under the agreement continue on a country-by-country and product-by-product basis until the later of the date on which no valid patent claim relating to a product exists or 12 years from the date of first sale of the product. We retain the right to develop and commercialize non-pharmaceutical products containing compounds arising from the collaboration. We would owe Merck a royalty based on sales of any such products.

        Our agreement with Merck will continue for so long as a product arising from the collaboration is the subject of an active development project or for so long as there is an obligation to pay royalties under the agreement. The agreement may be terminated earlier by Merck at any time after the third anniversary of the agreement upon three months' written notice, or may be terminated by either party for the other party's uncured breach or bankruptcy.

Merck—Anti-viral License and Research Collaboration

        In July 2004, we entered into a license and collaborative research agreement with Merck that allows Merck to discover and develop small molecule drugs against an enzyme target for treating viral infections. During the period from the beginning of the research term until the time that Merck ceases activities against the enzyme target, we agreed not to work with any third party on compounds that inhibit the enzyme target. Merck holds worldwide rights to commercialize any drugs resulting from this collaboration.

        The agreement provides for a payment by Merck to us of an upfront technology access fee and annual license fees. To date, we have received $2.6 million under this collaboration. The initial research term is three years and may be extended for one year upon mutual agreement of the parties. Merck may end the research term in January 2006 upon 90 days' written notice.

        We assigned to Merck small molecule compounds related to the viral target and our interest in research program patents and to compounds that act on the target through our inhibition mode. Merck owns all intellectual property generated in the course of performing the research except for improvements related to Tethering, which we own. Merck is required to pay pre-commercialization milestones of up to $22.1 million as well as royalty payments based on product sales. Royalty rates payable to us may be reduced if Merck is required to license additional intellectual property from one or more third parties in order to commercialize a collaboration product. Merck may also reduce its royalty payments to us if the product is not covered by a patent or if a third party markets a competitive product. Royalty obligations under the agreement continue on a country-by-country and product-by-product basis until the later of the date on which no valid patent claim relating to a product exists or 12 years from the date of first sale of the product.

        Our agreement with Merck will continue for so long as a product arising from the collaboration is the subject of an active development project or for so long as there is an obligation to pay royalties under the agreement. The agreement may be terminated earlier by Merck on the 18-month anniversary

58

of the agreement upon 90 days' written notice, or may be terminated by either party for the other party's uncured breach or bankruptcy.

Dainippon Pharmaceutical

        In October 2003, we entered into a licensing agreement with Dainippon Pharmaceutical in which we obtained a worldwide exclusive license, including the right to sublicense, to SNS-595 and related compounds.

        The agreement provides to Dainippon an upfront payment and milestone payments of up to $10.7 million for starting Phase II clinical testing, Phase III clinical testing, and for filing NDAs and receiving regulatory approval in the United States, Europe and Japan for cancer treatment. If SNS-595 is approved for a cancer indication in the United States, Europe or Japan, milestone payments become payable to Dainippon. If SNS-595 is approved for a non-cancer indication, additional milestone payments become payable to Dainippon.

        The agreement also provides for royalty payments to Dainippon at rates that are based on total annual net sales. We may reduce our royalty payments to Dainippon if a third party markets a competitive product or we must pay royalties for third party intellectual property rights necessary to commercialize SNS-595. Royalty obligations under the agreement continue on a country-by-country and product-by-product basis until the later of the date on which no valid patent claims relating to a product exist or 10 years from the date of the first sale of the product.

        If we discontinue seeking regulatory approval and/or sale of the product in a region, we are required to return to Dainippon its rights to the product in that region. The agreement may be terminated by either party for the other party's uncured breach or bankruptcy.

Bristol-Myers Squibb Company

        In April 2005, we entered into a licensing agreement with BMS in which we obtained worldwide exclusive and non-exclusive diagnostic and therapeutic licenses, including rights to sublicense, to SNS-032 and any related compounds that are active against CDK-1, -2, -4, -7 and -9 and are covered by licensed intellectual property.

        The agreement provides to BMS an upfront payment of $8 million payable in our Series C-2 preferred stock, and milestone payments totaling up to $78 million for beginning Phase I, Phase II and Phase III clinical testing, and for filing NDAs and receiving regulatory approval in the United States, Europe and Japan as well as for achieving certain commercial milestones. Milestone payments are distributed among IV and oral formulations and various cancer indications. We may pay some of these milestone payments in equity or a mixture of cash and equity, rather than entirely in cash.

        The agreement also provides for royalty payments to BMS at rates that are based on total annual net sales. Royalty obligations under the agreement continue on a country-by-country basis until the later of (1) expiration of all patents that are owned by us or exclusively licensed to us (whether by BMS or a third party) that cover a licensed product, (2) 10 years following the first commercial sale of a licensed product or (3) expiration of all applicable data exclusivity with respect to a licensed product.

        After completion of any Phase II clinical trial with SNS-032 or other licensed product under a U.S. IND, should we desire to sublicense our rights under the agreement, BMS will have the first right to negotiate with us for such sublicense. If Sunesis and BMS do not reach agreement within a designated length of time, then we are free to sublicense to any third party provided the financial terms are not less favorable than those offered to BMS. We cannot grant a sublicense to any third party before the completion of such Phase II clinical trial unless we receive BMS's consent.

59

        The agreement may be terminated by BMS for our uncured breach (other than a diligence breach) or bankruptcy. BMS may terminate this agreement on a country by country basis for our uncured failure to use commercially reasonable efforts to develop and/or commercialize at least one licensed compound or licensed product in a particular country or territory. Further, if such uncured failure occurs in certain countries, BMS may terminate this agreement as to entire designated territories. BMS may also terminate this agreement if we develop or market a competitive product within certain designated time frames. We may terminate this agreement with respect to a specific licensed product in a particular country without cause but with a specified notice period. We may also terminate this agreement for BMS's uncured breach.

Intellectual Property

        We patent the technology, inventions and improvements that we consider important to the development of our business. As of the date of this prospectus, we owned or had exclusive rights to 59 issued U.S. and foreign patents and 112 pending U.S. and foreign patent applications. Forty-three issued patents and seven pending applications relate to SNS-595, which cover compositions of matter and method of use in oncology and formulations. The issued patents are generally due to expire in 2015. The U.S. composition of matter patent is due to expire on October 6, 2015, and most of its foreign counterparts are due to expire on June 6, 2015. Two pending U.S. applications in each of our Aurora and Raf kinase programs relate to composition of matter, methods of use in oncology and other kinase-mediated diseases and formulations. We intend to seek patent term extension that may be available, including under the Hatch-Waxman Act, which provides up to five years of patent extension. Four issued patents relate to SNS-032. The U.S. composition of matter patent covering SNS-032 is due to expire on October 21, 2018 and most of its foreign counterparts are due to expire on December 7, 2020. Four issued patents, which will expire between 2018 and 2021, and 55 pending applications relate to Tethering. The remaining patents and applications relate to other aspects of our technology or other drug discovery programs that we are no longer actively pursuing.

        Our ability to build and maintain our proprietary position for our technology and drug candidates will depend on our success in obtaining effective claims and enforcing those claims once granted. The patent positions of biopharmaceutical companies like ours are generally uncertain and involve complex legal and factual questions for which important legal principles remain unresolved. No consistent policy regarding the breadth of patent claims has emerged to date in the United States. The patent situation outside the United States is even more uncertain. We do not know whether any of our patent applications or those patent applications that we license will result in the issuance of any patents. The patents we own or license and those that may issue in the future, may be challenged, invalidated or circumvented, and the rights granted under any issued patents may not provide us with proprietary protection or competitive advantages.

        We may not be able to develop patentable products or be able to obtain patents from pending patent applications. Even if patents are issued, they may not be sufficient to protect the technology and drug candidates owned by or licensed to us. These current patents and patents that may issue in the future may be challenged, invalidated, infringed or circumvented, and the rights granted in those patents may not provide proprietary protection or competitive advantage to us. Patent applications filed before November 29, 2000 in the United States are maintained in secrecy until patents issue. Later filed U.S. applications and patent applications in most foreign countries generally are not published until at least 18 months after they are filed. Scientific and patent publication often occurs long after the date of the scientific discoveries disclosed in those publications. Accordingly, we cannot be certain that we were the first to invent the subject matter covered by any patent application or that we were the first to file a patent application for any inventions.

        Our commercial success depends on our ability to operate without infringing patents and proprietary rights of third parties. We cannot determine with certainty whether patents or patent

60

applications of other parties may materially affect our ability to conduct our business. The existence of third party patent applications and patents could significantly reduce the coverage of patents owned by or licensed to us and limit our ability to obtain meaningful patent protection. If patents containing competitive or conflicting claims are issued to third parties and these claims are ultimately determined to be valid, we may be enjoined from pursuing research, development or commercialization of products, or be required to obtain licenses to these patents or to develop or obtain alternative technology.

        We may need to commence litigation to enforce any patents issued to us or to determine the scope and validity of third party proprietary rights. Litigation would result in substantial costs, even if the eventual outcome is favorable to us. An adverse outcome in litigation could subject us to significant liabilities to third parties and require us to seek licenses of the disputed rights from third parties or to cease using the technology if such licenses are unavailable.

        We also rely on trade secrets to protect our technology, especially where we do not believe patent protection is appropriate or obtainable. However, trade secrets are difficult to protect.

        We seek to protect our proprietary information by requiring our employees, consultants, contractors and other advisers to execute nondisclosure and assignment of invention agreements upon commencement of their employment or engagement. Agreements with our employees also prevent them from bringing the proprietary rights of third parties to us. We also require confidentiality or material transfer agreements from third parties that receive our confidential data or materials. There can be no assurance that these agreements will provide meaningful protection, that these agreements will not be breached, that we will have an adequate remedy for any such breach, or that our trade secrets will not otherwise become known or independently developed by a third party.

Government Regulation

U.S. Government Regulation

        FDA and comparable regulatory agencies in state and local jurisdictions and in foreign countries impose substantial requirements upon the clinical development, manufacture, marketing and distribution of drugs. These agencies and other federal, state and local entities regulate research and development activities and the testing, manufacture, quality control, safety, efficacy, labeling, storage, record keeping, approval, advertising and promotion of our drug candidates and drugs.

        In the United States, FDA regulates drugs under the Federal Food, Drug, and Cosmetic Act, or FFDCA, and implementing regulations. The process required by FDA before our drug candidates may be marketed in the United States generally involves the following:

•

completion of extensive preclinical laboratory tests, preclinical animal studies and formulation studies, all performed in accordance with FDA's good laboratory practice, or GLP, regulations;



•

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



•

performance of adequate and well-controlled clinical trials to establish the safety and efficacy of the product candidate for each proposed indication;



•

submission of a new drug application, or NDA, to FDA;



•

satisfactory completion of an FDA pre-approval inspection of the manufacturing facilities at which the product candidate is produced to assess compliance with current good manufacturing practice, or cGMP, regulations; and



•

FDA review and approval of the NDA prior to any commercial marketing, sale or shipment of the drug.

61

        The testing and approval process requires substantial time, effort and financial resources, and we cannot be certain that any approvals for our drug candidates will be granted on a timely basis, if at all.

        Preclinical tests include laboratory evaluation of product chemistry, formulation and stability, as well as studies to evaluate toxicity in animals. The results of preclinical tests, together with manufacturing information and analytical data, are submitted as part of an IND application to FDA. The IND automatically becomes effective 30 days after receipt by FDA, unless FDA, within the 30-day time period, raises concerns or questions about the conduct of the clinical trial, including concerns that human research subjects will be exposed to unreasonable health risks. In such a case, the IND sponsor and FDA must resolve any outstanding concerns before the clinical trial can begin. Our submission of an IND, or those of our collaboration partners, may not result in FDA authorization to commence a clinical trial. A separate submission to an existing IND must also be made for each successive clinical trial conducted during product development, and FDA must grant permission before each clinical trial can begin. Further, an independent institutional review board, or IRB, for each medical center proposing to conduct the clinical trial must review and approve the plan for any clinical trial before it commences at that center and it must monitor the study until completed. FDA, the IRB or the sponsor may suspend a clinical trial at any time on various grounds, including a finding that the subjects or patients are being exposed to an unacceptable health risk. Clinical testing also must satisfy extensive Good Clinical Practice, or GCP, requirements and regulations for informed consent.

Clinical Trials

        For purposes of NDA submission and approval, clinical trials are typically conducted in the following three sequential phases, which may overlap:

•

Phase I clinical trials are initially conducted in a limited population to test the drug candidate for safety, dose tolerance, absorption, metabolism, distribution and excretion in healthy humans or, on occasion, in patients, such as cancer patients. In some cases, particularly in cancer trials, a sponsor may decide to run what is referred to as a "Phase Ib" evaluation, which is a second safety-focused Phase I clinical trial typically designed to evaluate the impact of the drug candidate in combination with currently approved drugs.



•

Phase II clinical trials are generally conducted in a limited patient population to identify possible adverse effects and safety risks, to determine the efficacy of the drug candidate for specific targeted indications and to determine dose tolerance and optimal dosage. Multiple Phase II clinical trials may be conducted by the sponsor to obtain information prior to beginning larger and more expensive Phase III clinical trials. In some cases, a sponsor may decide to run what is referred to as a "Phase IIb" evaluation, which is a second, confirmatory Phase II clinical trial that could, if positive and accepted by FDA, serve as a pivotal clinical trial in the approval of a drug candidate.



•

Phase III clinical trials are commonly referred to as pivotal trials. When Phase II clinical trials demonstrate that a dose range of the drug candidate is effective and has an acceptable safety profile, Phase III clinical trials are undertaken in large patient populations to further evaluate dosage, to provide substantial evidence of clinical efficacy and to further test for safety in an expanded and diverse patient population at multiple, geographically dispersed clinical trial sites.

        In some cases, FDA may condition approval of an NDA for a drug candidate on the sponsor's agreement to conduct additional clinical trials to further assess the drug's safety and effectiveness after NDA approval. Such post-approval trials are typically referred to as Phase IV clinical trials.

62

New Drug Application

        The results of drug candidate development, preclinical testing and clinical trials are submitted to FDA as part of an NDA. The NDA also must contain extensive manufacturing information. Once the submission has been accepted for filing, by law FDA has 180 days to review the application and respond to the applicant. The review process is often significantly extended by FDA requests for additional information or clarification. FDA may refer the NDA to an advisory committee for review, evaluation and recommendation as to whether the application should be approved. FDA is not bound by the recommendation of an advisory committee, but it generally follows such recommendations. FDA may deny approval of an NDA if the applicable regulatory criteria are not satisfied, or it may require additional clinical data or an additional pivotal Phase III clinical trial. Even if such data are submitted, FDA may ultimately decide that the NDA does not satisfy the criteria for approval. Data from clinical trials are not always conclusive and FDA may interpret data differently than we or our collaboration partners interpret data. Once issued, FDA may withdraw drug approval if ongoing regulatory requirements are not met or if safety problems occur after the drug reaches the market. In addition, FDA may require testing, including Phase IV clinical trials, and surveillance programs to monitor the effect of approved products that have been commercialized, and FDA has the power to prevent or limit further marketing of a drug based on the results of these post-marketing programs. Drugs may be marketed only for the approved indications and in accordance with the provisions of the approved label. Further, if there are any modifications to the drug, including changes in indications, other labeling changes, or manufacturing processes or facilities, we may be required to submit and obtain FDA approval of a new NDA or NDA supplement, which may require us to develop additional data or conduct additional preclinical studies and clinical trials.

Fast Track Designation

        FDA's fast track program is intended to facilitate the development and to expedite the review of drugs that are intended for the treatment of a serious or life-threatening condition for which there is no effective treatment and which demonstrate the potential to address unmet medical needs for the condition. Under the fast track program, the sponsor of a new drug candidate may request FDA to designate the drug candidate for a specific indication as a fast track drug concurrent with or after the filing of the IND for the drug candidate. FDA must determine if the drug candidate qualifies for fast track designation within 60 days of receipt of the sponsor's request.

        If fast track designation is obtained, FDA may initiate review of sections of an NDA before the application is complete. This rolling review is available if the applicant provides and FDA approves a schedule for the submission of the remaining information and the applicant pays applicable user fees. However, the time period specified in the Prescription Drug User Fees Act, which governs the time period goals FDA has committed to reviewing an application, does not begin until the complete application is submitted. Additionally, the fast track designation may be withdrawn by FDA if FDA believes that the designation is no longer supported by data emerging in the clinical trial process.

        In some cases, a fast track designated drug candidate may also qualify for one or more of the following programs:

•

Priority Review.     Under FDA policies, a drug candidate is eligible for priority review, or review within a six-month time frame from the time a complete NDA is accepted for filing, if the drug candidate provides a significant improvement compared to marketed drugs in the treatment, diagnosis or prevention of a disease. A fast track designated drug candidate would ordinarily meet FDA's criteria for priority review. We do not know whether any of our drug candidates will receive a priority review designation, or if a priority designation is received, that review or approval will be faster than conventional FDA procedures, or that FDA will ultimately grant drug approval.

63

•

Accelerated Approval.     Under FDA's accelerated approval regulations, FDA is authorized to approve drug candidates that have been studied for their safety and efficacy in treating serious or life-threatening illnesses and that provide meaningful therapeutic benefit to patients over existing treatments based upon either a surrogate endpoint that is reasonably likely to predict clinical benefit or on the basis of an effect on a clinical endpoint other than patient survival. In clinical trials, surrogate endpoints are alternative measurements of the symptoms of a disease or condition that are substituted for measurements of observable clinical symptoms. A drug candidate approved on this basis is subject to rigorous post-marketing compliance requirements, including the completion of Phase IV or post-approval clinical trials to validate the surrogate endpoint or confirm the effect on the clinical endpoint. Failure to conduct required post-approval studies, or to validate a surrogate endpoint or confirm a clinical benefit during post-marketing studies, will allow FDA to withdraw the drug from the market on an expedited basis. All promotional materials for drug candidates approved under accelerated regulations are subject to prior review by FDA.

        When appropriate, we and our collaboration partners intend to seek fast track designation, accelerated approval or priority review for our drug candidates. We cannot predict whether any of our drug candidates will obtain a fast track or accelerated approval designation, or the ultimate impact, if any, of the fast track or the accelerated approval process on the timing or likelihood of FDA approval of any of our drug candidates.

        Satisfaction of FDA regulations and approval requirements or similar requirements of foreign regulatory agencies typically takes several years and the actual time required may vary substantially based upon the type, complexity and novelty of the product or disease. Typically, if a drug candidate is intended to treat a chronic disease, as is the case with some of the drug candidates we are developing, safety and efficacy data must be gathered over an extended period of time. Government regulation may delay or prevent marketing of drug candidates for a considerable period of time and impose costly procedures upon our activities. FDA or any other regulatory agency may not grant approvals for new indications for our drug candidates on a timely basis, or at all. Even if a drug candidate receives regulatory approval, the approval may be significantly limited to specific disease states, patient populations and dosages. Further, even after regulatory approval is obtained, later discovery of previously unknown problems with a drug may result in restrictions on the drug or even complete withdrawal of the drug from the market. Delays in obtaining, or failures to obtain, regulatory approvals for any of our drug candidates would harm our business. In addition, we cannot predict what adverse governmental regulations may arise from future United States or foreign governmental action.

Other Regulatory Requirements

        Any drugs manufactured or distributed by us or our collaborators pursuant to FDA approvals are subject to continuing regulation by FDA, including recordkeeping requirements and reporting of adverse experiences associated with the drug. Drug manufacturers and their subcontractors are required to register their establishments with FDA and certain state agencies, and are subject to periodic unannounced inspections by FDA and certain state agencies for compliance with ongoing regulatory requirements, including cGMPs, which impose certain procedural and documentation requirements upon us and our third-party manufacturers. Failure to comply with the statutory and regulatory requirements can subject a manufacturer to possible legal or regulatory action, such as warning letters, suspension of manufacturing, seizure of product, injunctive action or possible civil penalties. We cannot be certain that we or our present or future third-party manufacturers or suppliers will be able to comply with the cGMP regulations and other ongoing FDA regulatory requirements. If our present or future third-party manufacturers or suppliers are not able to comply with these requirements, FDA may halt our clinical trials, require us to recall a drug from distribution, or withdraw approval of the NDA for that drug.

64

        FDA closely regulates the post-approval marketing and promotion of drugs, including standards and regulations for direct-to-consumer advertising, off-label promotion, industry-sponsored scientific and educational activities and promotional activities involving the Internet. A company can make only those claims relating to safety and efficacy that are approved by FDA. Failure to comply with these requirements can result in adverse publicity, warning letters, corrective advertising and potential civil and criminal penalties. Physicians may prescribe legally available drugs for uses that are not described in the drug's labeling and that differ from those tested by us and approved by FDA. Such off-label uses are common across medical specialties, and and cancer therapy. Physicians may believe that such off-label uses are the best treatment for many patients in varied circumstances. FDA does not regulate the behavior of physicians in their choice of treatments. FDA does, however, impose stringent restrictions on manufacturers' communications regarding off-label use.

Foreign Regulation

        In addition to regulations in the United States, we will be subject to a variety of foreign regulations governing clinical trials and commercial sales and distribution of our future products. Whether or not we obtain FDA approval for a product, we must obtain approval of a product by the comparable regulatory authorities of foreign countries before we can commence clinical trials or marketing of the product in those countries. The approval process varies from country to country, and the time may be longer or shorter than that required for FDA approval. The requirements governing the conduct of clinical trials, product licensing, pricing and reimbursement vary greatly from country to country.

        Under European Union regulatory systems, marketing authorizations may be submitted either under a centralized or mutual recognition procedure. The centralized procedure provides for the grant of a single marking authorization that is valid for all European Union member states. The mutual recognition procedure provides for mutual recognition of national approval decisions. Under this procedure, the holder of a national marking authorization may submit an application to the remaining member states. Within 90 days of receiving the applications and assessment report, each member state must decide whether to recognize approval.

        In addition to regulations in Europe and the United States, we will be subject to a variety of foreign regulations governing clinical trials and commercial distribution of our future products.

Manufacturing

        We outsource the manufacture of SNS-595 to third-party contract manufacturers. The active pharmaceutical ingredient of SNS-595 is manufactured by a single-source supplier through a 13-step convergent synthesis in which two intermediates are manufactured in a parallel process and then combined and deprotected in the final two steps. The active pharmaceutical ingredient is then formulated and vials are filled and finished by a different third party manufacturer. The active pharmaceutical ingredient is classified as a toxic substance, which limits the number of suppliers qualified to manufacture it. We have a sufficient supply of both the active pharmaceutical ingredient for SNS-595 and finished product to conduct our current and planned clinical trials of SNS-595 through the fourth quarter of 2005. We expect that additional active pharmaceutical ingredient and finished product will be manufactured, tested and released by the third quarter of 2005.

        We will outsource the manufacture of SNS-032 to third-party contract manufacturers. As part of our agreement with BMS, we acquired enough of the active pharmaceutical ingredients of SNS-032 for at least our planned Phase I clinical trial for SNS-032. However, before we are able to commence this trial, we must convert the active pharmaceutical ingredients into finished product. We expect that a sufficient supply of finished product to conduct our planned clinical trial of SNS-032 will be manufactured, tested and released by the third quarter of 2005.

65