Business

Overview

We are a biopharmaceutical company focused on the discovery, development, and commercialization of novel therapies that harness the newly-discovered, natural mechanisms of immunological tolerance. Our goal is to utilize these mechanisms to design therapies that reprogram the immune system in order to treat autoimmune disease patients, organ transplant recipients, and patients suffering from other harmful immune system responses. Our therapies are designed to provide long-term efficacy after a short course of treatment without compromising the immune system's ability to protect the body against invading bacteria, viruses, and fungi.

We have established a pipeline of two monoclonal antibodies in human clinical trials. Our TRX4 antibody is in an investigator-sponsored Phase II trial of 80 patients in Europe with new-onset Type I diabetes. The investigators have reported to us unpublished data that at 6 months following 6 consecutive days of therapy, a significantly higher percentage of the TRX4-treated patients maintained a sustained or improved function of the insulin-producing, or beta, cells of the pancreas as compared to the placebo-treated group. The investigators have also reported to us that this clinical effect was durable 12 months after therapy. We filed an Investigational New Drug, or IND, application that has been accepted by the U.S. Food and Drug Administration, or FDA, for TRX4, and we intend to commence a Phase I clinical trial in mid 2004 to expand our indications to include psoriasis. Our TRX1 antibody is being developed, in collaboration with Genentech, to induce immunological tolerance in transplantation, autoimmune disease, and clinical situations where the immune system attacks a therapeutic protein. We completed a Phase Ia clinical trial of TRX1 in the United Kingdom in July 2003. We have filed an IND that has been accepted by the FDA, and we intend to commence a Phase Ib clinical trial of TRX1 in the United States in patients with hemophilia A in mid 2004. We also received an additional $0.5 million research support payment from Genentech relating to our preclinical study of TRX1 in transplantation.

Our research and development programs are designed to generate new products based on our unique understanding of the mechanisms of immunological tolerance and the cells that mediate tolerance. We have several novel monoclonal antibodies being tested in animal models that have been discovered based on our knowledge of immunological tolerance. We are also developing the TolerMab antibody technology which modifies monoclonal antibodies so they can generate tolerance to themselves. TolerMab antibodies are designed to allow for repetitive dosing in immunocompetent patients with chronic diseases without eliciting the interfering immune system response often encountered with traditional long-term antibody therapy.

The immune system and immunological tolerance

The human immune system protects the body against infection by bacteria, viruses, and fungi predominantly through the action of a group of blood cells collectively called white blood cells or leukocytes. A subset of leukocytes known as T lymphocytes, or T cells, have a surface receptor that recognizes a specific antigen presented to them by other cells of the immune system called antigen presenting cells, or APCs. APCs are found throughout the body and function as sentinels guarding against invasion by pathogenic organisms. APCs continuously

39

sample their surrounding environment for antigens which they process, transport, and present to T cells. When T cells recognize a specific antigen presented by APCs, the APCs also instruct the T cells whether the antigen is dangerous or harmful to the body and should be eliminated through a classical immune response, or whether it is non-threatening and should be preserved through a tolerogenic immune response. In a normal immune system, leukocytes eliminate pathogens and other dangerous antigens without inflicting damage to the body's own healthy cells and tissues.

The Classical Immune Response. A classical immune response is initiated when APCs present antigens to T cells in the presence of alarm signals from the body's own cells and from foreign invaders, such as molecules associated with inflammation and cell death. This type of antigen presentation results in T cell activation and the generation of a subtype of T cells known as T-effector cells. These T-effector cells orchestrate the immunological processes that eliminate the antigen and cells which contain the antigen on their surface.

The Tolerogenic Immune Response. Immunological tolerance is a normal state in which the immune system recognizes an antigen, or tissue bearing an antigen, as non-dangerous and thus prevents a harmful or misdirected attack. A tolerogenic immune response begins through a similar pathway as the classical immune response with antigen presentation by APCs and T cell activation. Without alarm signals, or in the presence of certain tolerance-promoting molecules, antigen presentation by APCs to T cells results in the generation of a newly discovered subtype of T cells known as T-regulatory cells. These T-regulatory cells are potent mediators of immunological tolerance in the body and can actively suppress T-effector cells.

The classical immune response and the tolerogenic immune response can be considered competing and reciprocal immune responses to a specific antigen. When the classical immune response is dominant, T-effector cells mediate the elimination of the antigen and cells which contain the antigen on their surface. In contrast, when the tolerogenic immune response is dominant, T-regulatory cells mediate the preservation of the antigen and cells which contain the antigen on their surface.

The problems associated with immune system diseases and current treatments

A normally functioning immune system makes the appropriate determination between permitting the classical immune response or the tolerogenic immune response to dominate upon presentation of an antigen. In autoimmune disease, T cells are incorrectly instructed to respond to the body's own, or self, antigens that are non-threatening as though they are harmful or dangerous. This leads to the generation of T-effector cells specific for self antigens which dominate over T-regulatory cells, and the immune system harmfully attacks its own tissue. For example, in Type I diabetes, the immune system attacks and destroys the insulin-producing cells of the pancreas. As a result of this attack, the Type I diabetes patient loses the ability to secrete insulin and to control glucose levels, requiring life-long insulin injections. People with Type I diabetes are at an increased risk of visual impairment and blindness, serious foot problems, kidney disease, and damage to the nervous system. Other conditions associated with adverse, harmful, or misdirected immune responses to self tissues include psoriasis, psoriatic arthritis, rheumatoid arthritis, multiple sclerosis, and systemic lupus erythematosis.

Historically, autoimmune diseases have been treated with drugs that suppress the entire immune system. These immunosuppressive drugs must be administered continuously as

40

discontinuation of these drugs results in relapse of disease. These therapies do not specifically target the cause of a disease and can have numerous harmful effects, including rendering the patient susceptible to life-threatening infections, malignancies, and metabolic disorders.

In organ transplant patients, the immune system often rejects the transplanted organ instead of tolerating it as the body's own tissue. Accordingly, transplant patients generally must receive multiple immunosuppressive drugs. This course of therapy causes numerous side effects for transplant recipients, including an increased risk of infection as well as an increased risk of cancer after transplant. In order to further reduce the risk of organ rejection, tremendous efforts are made to match potential donors with recipients. Despite this regimen, approximately 30% of patients experience an acute rejection episode.

Therapeutic proteins are increasingly used to treat diseases, including interferon beta for multiple sclerosis, Factor VIII for hemophilia A, and monoclonal antibodies for various disorders. However, these proteins frequently elicit an immune response that can reduce their efficacy or preclude repeat dosing. Betaseron, an interferon beta product marketed by Schering AG and Berlex Laboratories, has been reported to generate an undesirable immune response in approximately 45% of multiple sclerosis patients. Approximately 30% to 40% of patients with severe hemophilia A develop neutralizing antibodies to Factor VIII replacement therapy. The use of Johnson & Johnson's approved monoclonal antibody, Orthoclone OKT3, to treat transplant rejection has been limited by the immune response against it and first dose reactions such as fever, chills, and pulmonary and gastrointestinal symptoms.

Despite progress in the use of immunosuppressive drugs and therapeutic proteins, there is still a significant, unmet clinical need for more effective and safer therapies for patients with diseases of the immune system.

The TolerRx solution

Our solution is to harness the body's natural tolerance mechanisms to develop therapies that reprogram the immune system and provide a long-term, durable effect after a short course of therapy.

T cells are at the heart of most immune responses. We have focused our research efforts on a new class of T cells, called T-regulatory cells, which can actively suppress the generation and function of T-effector cells. T-regulatory cells are among the most powerful mechanisms used by the body to establish and maintain immunological tolerance. T-regulatory cells, and their interaction with other cells of the immune system, represent a new discovery relating to the function of a normal immune system. We believe that T-regulatory cells provide novel mechanisms and targets for use in the development of new products to treat patients with diseases of the immune system. Additionally, we are developing surrogate markers and prognostic indicators of T-regulatory cell dominance and tolerance which we believe could serve as early efficacy endpoints in clinical development and allow us to improve our clinical outcomes.

Our therapies are designed to change the immune system's perception of autoimmune disease related antigens and transplanted organ antigens so that T-regulatory cells are induced, tolerogenic mechanisms become dominant, and normal immunological defenses against invading pathogens are preserved. As these therapies induce natural mechanisms of the

41

immune system, we believe they will be safe, durable, and effective, and in some cases may provide a cure. For example, we intend to develop therapies for new-onset Type I diabetes that induce the immune system to suppress the T-effector cells that cause injury to the insulin-producing cells in the pancreas and thereby promote tolerance to these insulin-producing cells. In addition, we intend to develop therapies that will allow transplanted organs to be accepted without the chronic use of immunosuppressive drugs.

Our strategy

Our goal is to become the recognized leader in the discovery, development, and commercialization of products to treat patients with diseases of the immune system. Our strategy to achieve this goal is to:

Develop and commercialize our lead products to provide breakthrough therapies that induce immunological tolerance. We are evaluating TRX4 in an investigator-sponsored Phase II clinical trial of 80 patients in Europe with new-onset Type I diabetes. We are also initiating a Phase I clinical trial of TRX4 in patients with psoriasis. In collaboration with Genentech, we are developing our TRX1 monoclonal antibody to induce tolerance in transplantation, autoimmune disease, and clinical situations where the immune system attacks therapeutic proteins or biologic drugs, such as Factor VIII for hemophilia A. We are initiating a Phase Ib clinical trial of TRX1 in patients with hemophilia A. We believe that our lead products will provide breakthrough therapies for immune system diseases by establishing or re-establishing long-lasting immunological tolerance to the disease causing antigens after only a short course of treatment.

Generate new products for our pipeline based on our understanding of T-regulatory cells. We have performed extensive studies comparing T-regulatory cells to T-effector cells in order to identify genes, proteins, and signaling pathways that are unique to or preferentially expressed by each cell population. These genes, proteins, and pathways represent target opportunities for novel immunological therapies. Targets that can inhibit T-effector cells and not T-regulatory cells, or targets that activate T-regulatory cells, could lead to tolerance promoting therapies. Conversely, therapies that block the function of T-regulatory cells and not T-effector cells may be utilized to temporarily disrupt tolerance mechanisms that a solid tumor or chronic viral infection tends to induce, thereby allowing the immune system to remove the cancer cells or virus. We have several monoclonal antibodies directed against our newly discovered targets that are being tested in animal models, and we plan to expand our pipeline by advancing additional products into clinical development.

Leverage collaborative partnerships. Leveraging the resources of collaborative partners is an important element of our strategy to exploit the products and targets being generated from our research and development programs. In these collaborations, we seek to maintain co-development, profit sharing, and commercialization opportunities for our products. In December 2002, we entered into a collaboration agreement with Genentech for the development and commercialization of TRX1. We believe that this collaboration allows us to enhance the development of TRX1 through the funding of research and by providing us with Genentech's expertise in manufacturing, preclinical and clinical development, regulatory affairs, and marketing.

42

Retain commercialization rights to certain products. We intend to maximize the value of our product pipeline by retaining rights to certain products for internal development and commercialization. We have retained all rights for the development and commercialization of TRX4, TolerMab TRX4, TolerMab TRX3, and our internally discovered antibodies.

Exploit knowledge of immunological tolerance to optimize clinical development. We are utilizing our knowledge of T-regulatory cells to identify surrogate markers of tolerance that can serve as early endpoints in our clinical trials. With further development, we believe these surrogate markers will serve as prognostic indicators that will allow us to assess the severity of a patient's condition, adjust a patient's dosing regimen, and monitor a patient's progress towards the establishment of immunological tolerance. We expect that this approach will improve our clinical outcomes and accelerate the development of our drugs.

Commercialize the TolerMab antibody technology internally and through partners. We intend to develop our TolerMab antibody technology to enable long-term, repeated dosing of antibodies in immunocompetent patients with chronic diseases. After further development, we plan to enter into agreements for TolerMab versions of partners' antibodies in exchange for license fees, milestone payments, and royalties. In addition, we intend to use the TolerMab technology to develop products for our internal pipeline, such as TolerMab TRX3 and TolerMab TRX4.

Drug development programs

We currently have two monoclonal antibodies in human clinical trials as follows:

(1)   Investigator-sponsored trial in Europe.

TRX4

TRX4 is a humanized monoclonal antibody that binds to a receptor found on all T cells called CD3, which is involved in normal T cell signaling. TRX4 is designed to block the function of T-effector cells that attack the body and cause autoimmune disease. Because T-effector cells and T-regulatory cells function differently, TRX4 is expected to have a favorable effect on T-regulatory cells, thereby promoting a state of immunological tolerance. We are developing TRX4 to treat patients with Type I diabetes and psoriasis.

CD3 is a well validated target for monoclonal antibody therapeutics. The first monoclonal antibody approved by the FDA was the murine anti-CD3 antibody, Orthoclone OKT3, by Johnson & Johnson. OKT3 is used to deplete T cells in order to reverse organ transplant rejection. However, the use of OKT3 has been limited due to early adverse effects, including fevers, chills, and pulmonary and gastrointestinal symptoms. Additionally, the clinical utility of OKT3 has been limited by the immune response that the patient generates against it. TRX4 is designed to reduce these side effects and adverse responses.

43

Type I diabetes

TRX4 is currently being studied in an investigator-sponsored Phase II clinical trial in Europe in patients with new-onset Type I diabetes. Patients enrolled in the trial must have had no more than 4 weeks of insulin therapy prior to the start of treatment and have the ability to produce some insulin. The patients were randomized such that 40 patients were treated with TRX4 and 40 patients were given a placebo. All patients were treated with insulin as medically necessary. As of March 2003, the last patient enrolled had completed treatment in this randomized, double-blind, placebo-controlled trial. The trial is being conducted by the Juvenile Diabetes Research Foundation Center for Beta Cell Therapy in Europe, and with financial support from the Juvenile Diabetes Research Foundation. It involves the participation of the Belgium Diabetes Registry, the universities in Brussels (VUB-ULB), Louvain (KUL), Antwerp (UIA), the Schwabing Hospital Munich, and Hospital Necker in Paris.

The investigators performed an efficacy and safety analysis at 6 months and 12 months following 6 consecutive days of therapy with TRX4. The investigators have reported to us unpublished data that at 6 months after therapy, a significantly higher percentage of the TRX4-treated patients maintained a sustained or improved function of the beta cells of the pancreas as compared to the placebo-treated group. The investigators have also reported to us that this clinical effect was durable 12 months after therapy. The investigators intend to publish these data in a major medical or scientific journal. We intend to perform an independent audit and analysis of these data and to conduct further clinical trials in patients with new-onset Type I diabetes after discussions with the FDA in 2005.

According to the Juvenile Diabetes Research Foundation, approximately one million people in the United States have Type I diabetes and 30,000 new patients are diagnosed each year, including over 13,000 children. In Type I diabetes, the pancreas produces little or no insulin as a result of the immune system attacking and destroying the beta cells in the pancreas. As a result, Type I diabetes patients require frequent administration of insulin therapy each day to control their blood sugar levels. Maintaining the function of the beta cells allows a Type I diabetes patient to continue to produce some natural insulin and diminishes the patient's need to administer insulin therapy. Published studies have demonstrated that the preservation of beta cell function in patients with Type I diabetes results in easier and better control of blood sugar levels, including a lower risk for hypoglycemia, or low blood sugar levels, one of the major complications of intensive insulin therapy. These studies have also indicated that retention of beta cell function and preservation of natural insulin production is associated with a reduced incidence of diabetes-related chronic complications, including serious eye disease, nervous system, kidney, and vascular disorders.

TRX4 has been designed to treat patients with Type I diabetes by blocking T-effector cells from attacking beta cells. In addition to the data with TRX4 from the investigator-sponsored trial in Europe, there are published studies which also indicate that blocking these T cells early in the disease process with anti-CD3 therapy will decrease injury to the beta cells and preserve the normal function of these cells. In one of those published studies, 75% of treated patients showed improvement or lack of disease progression one year after a single therapeutic course. We believe that TRX4 therapy in new-onset Type I diabetes has the potential to have a durable therapeutic effect, a characteristic that no approved treatment regime can claim in this indication. We expect our therapy to provide efficacy and an improved quality of life for these

44

patients by preserving beta cell function, and thereby decreasing their dependence on insulin injections and preserving normal glucose control.

Psoriasis

We filed an IND in the United States to expand our TRX4 clinical indications by commencing a Phase I clinical trial of TRX4 in patients with psoriasis. In the Phase I psoriasis study, we will evaluate the safety, pharmacokinetic and pharmacodynamic profile of TRX4 in a single-dose, dose escalating study. The IND was accepted by the FDA.

Psoriasis is an immune-mediated, genetic disease of the skin that affects about two percent of the U.S. population. The most severe and disabling cases of psoriasis require systemic medications such as methotrexate and cyclosporine, which have severe side effects including decreased kidney function, high blood pressure, and high cholesterol. Several new treatments for psoriasis include biologics, such as Amevive®, Raptiva®, and Enbrel®, which must be injected or infused into the body rather than taken orally.

Published studies have indicated that anti-CD3 therapy has the potential for therapeutic benefit in psoriasis, including patients with psoriatic arthritis. In one study, anti-CD3 therapy resulted in a statistically significant reduction in the number of tender and swollen joints 30 days and 90 days after therapy. The anti-CD3 therapy also resulted in an improvement in the accompanying skin disease.

Preclinical and clinical safety analysis

In a preclinical toxicology study in chimpanzees, two dose levels of TRX4 were evaluated. One group of animals received a low dose that was approximately the same as the highest anticipated human dose and approximately the same as the total dose administered in the European Type I diabetes trial. The high dose group animals received a dose that was five times greater than the highest anticipated human dose. In the high dose group, TRX4 levels in the blood over time, known as exposure, was estimated to be over 40-fold higher than the exposure that the patients experienced in the European Type I diabetes trial. We believe that this increased exposure and differences in how chimpanzee cells interact with TRX4 compared to how human cells interact with TRX4 caused an exaggerated pharmacological response. This response led to reactivation of a chimpanzee-specific virus similar to the Epstein-Barr virus, or EBV-like virus, which caused lymphoproliferative disease and ultimately the death of the three animals in the high dose group. The chimpanzees in the low dose group experienced an estimated 4-fold greater exposure to TRX4 compared with the patients in the European Type I diabetes trial. These animals showed temporary and expected immunological changes during TRX4 therapy, including some reactivation of the EBV- like virus, and recovered. As a result, we and the FDA have agreed to the laboratory and clinical tests that will be used in our clinical trials with TRX4 in order to monitor and ensure patient safety during dose escalation.

In the European Type I diabetes study, the majority of patients experienced mild to moderate flu-like symptoms related to initial doses of TRX4 that were markedly diminished as compared to the investigators' historical experience with OKT3. These reactions decreased or diminished with continued dosing of TRX4. At approximately two to three weeks after treatment with TRX4, many patients developed an increased level of EBV detectable in their circulation, some of which were coincident with signs and symptoms suggestive of a viral infection, such as sore

45

throat, fever, enlarged lymph nodes, and rash. One individual was hospitalized twice for sore throat, fever, and enlarged lymph nodes and was treated with antibiotics and steroids. However, the signs, symptoms, and EBV increases were transient, resolved in all subjects likely due to an effective anti-viral immune response, and did not recur in any individual, including the one who required hospitalization.

TolerMab TRX4

We are developing a TolerMab form of TRX4 as a potential second generation product. TolerMab TRX4 is in preclinical development. We own all rights for the development and commercialization of TRX4 and TolerMab TRX4.

TRX1

In December 2002, we entered into a collaboration agreement with Genentech for the development and commercialization of products targeting the CD4 molecule, including TRX1. TRX1 is a humanized monoclonal antibody that binds to the CD4 receptor found on both T-effector cells and T-regulatory cells. Because T-effector cells utilize different signaling pathways upon antigen presentation as compared to T-regulatory cells, some of which are associated with the CD4 receptor, TRX1 is expected to block the activation and function of T-effector cells and to favor the dominance of T-regulatory cells. TRX1 is being developed to induce immunological tolerance in transplantation, autoimmune diseases, and in settings requiring the chronic administration of therapeutic proteins.

In July 2003, we completed a single-dose, placebo-controlled, double-blind Phase I trial in the United Kingdom with TRX1. The trial evaluated doses of 1, 5 and 10 mg/kg of TRX1 in normal healthy volunteers. The endpoints of the trial were designed to assess the safety and pharmacokinetic parameters of TRX1. In the study, TRX1 appeared to be well tolerated, did not deplete T cells, and first dose side effects, such as fevers, chills, and hypotension were not observed. At the highest dosing group of 10 mg/kg, a rash was observed in two of three TRX1-treated volunteers at 17 and 20 days after dosing, which required symptomatic therapy to resolve the rash. Rash was not observed at the 1 and 5 mg/kg dosing groups. Based upon the pharmacokinetic profile of TRX1, we believe that the effective tolerizing dose of TRX1 will be lower than 10 mg/kg, thereby reducing the risk of rash.

In nonhuman primates, we have demonstrated that a short course of TRX1 can induce a long-term, antigen-specific tolerance to a normally immunogenic biological protein drug without compromising normal immune function. We believe that TRX1 could be applicable in situations where administration of a biologic protein therapy induces an interfering immune response, such as Factor VIII therapy in hemophilia A, interferon beta therapy in multiple sclerosis, enzyme replacement therapy, and therapies involving repeated administration of antibodies. For example, approximately 30% to 40% of patients with severe hemophilia A develop neutralizing antibodies to Factor VIII, requiring clinical immune tolerance induction strategies involving increased Factor VIII dosing and expenses of approximately $1 million per patient during one year of treatment. In collaboration with Genentech, we are evaluating the potential use of TRX1 to induce tolerance to therapeutic proteins or biologic drugs. We filed an IND to commence a Phase Ib clinical trial in patients with hemophilia A. In the Phase Ib study, we will evaluate the safety, pharmacokinetics and pharmacodynamics of TRX1 in patients

46

who have inhibitors to their Factor VIII replacement therapy and are failing, have failed, or are ineligible for Factor VIII immune tolerance induction therapy in a multi-dose, dose escalating study. The IND was accepted by the FDA.

In addition, we are evaluating TRX1 in nonhuman primates and Genentech is evaluating TRX1 in rodent models as a tolerance induction therapy in transplantation. The objective is to prevent transplant rejection and decrease or eliminate the need for chronic, immunosuppressive, maintenance therapy. From July 2001 to June 2002, approximately 25,000 patients in the U.S. received an organ transplant, approximately 80,000 people were on a waiting list to receive a transplant, and approximately 6,300 people died while waiting for a transplant. Our preclinical studies are examining the parameters required for successful induction of transplantation tolerance and are expected to provide the foundation for the design of human clinical trials using TRX1 in transplantation. We received a $0.5 million research support payment from Genentech in November 2003 related to our nonhuman primate preclinical study in kidney transplantation.

As part of our collaboration with Genentech, Genentech is evaluating TRX1 in preclinical models as a primary therapy for patients with autoimmune diseases. The objective will be to induce tolerance to autoimmune disease causing antigens.

TRX2

TRX2 is a humanized monoclonal antibody that binds to the CD8 receptor found on a subset of T cells. TRX2 is currently in preclinical development in a renal transplantation study in nonhuman primates as combination therapy with TRX1. We believe that TRX2 could be synergistic with TRX1 for the induction of immunological tolerance in organ transplantation due to the role that T cells bearing the CD8 receptor have in the early rejection of transplants. TRX2 is designed to deplete these T cells at the time of transplant so that they do not cause rejection of the transplant before immunological tolerance can be established by TRX1. In addition to the current agreement with Genentech that covers uses of TRX1 in combination therapy with other therapeutics including TRX2, we have also granted Genentech a right of first negotiation for TRX2 and any rights we may own or control that solely relate to anti-CD8 antibodies if we decide to seek a partner for such products.

TolerMab antibody technology

One of the problems associated with monoclonal antibody drugs is that the patient's immune system may incorrectly respond to the antibody as dangerous and eliminate it from the body, thereby decreasing its effectiveness. While humanized and fully human antibodies generate less of an immune response than mouse or rat antibodies, published data regarding certain antibodies currently marketed or under development indicate that the body will still generate an immune response to these antibodies. We believe that this immunogenicity will continue to be recognized as a significant clinical issue. For example, the fully human antibody Humira®, marketed by Abbott Laboratories, has been reported to elicit a 12% rate of immunogenicity in rheumatoid arthritis trials, and previously, it has been reported that the humanized antibody MEDI-507 from MedImmune had a 50% rate of immunogenicity in psoriasis patients.

47

Using our TolerMab technology, we intend to modify monoclonal antibodies so that immunological tolerance is induced to the antibodies while maintaining therapeutic benefit. We have two TolerMab antibodies in preclinical development, TolerMab TRX3 and TolerMab TRX4. The TolerMab technology is expected to provide the following benefits:

•  effective, long-term, repeated dosing for the treatment of immunocompetent patients with chronic disease;

•  lower first dose side effects than traditional antibodies that deplete their target cells; and

•  no or lower rates of interfering immune responses.

Our proprietary TolerMab technology involves attaching two small amino acid sequences, or modifiers, to the antibody. Each modifier can move in and out of an antibody binding site. When a modifier blocks a binding site, the antibody is hindered from binding to target sites. However, binding to the target and antibody function occur when a modifier moves out of the binding site. The result is that the binding rate of the antibody to the target is decreased, resulting in an excess of free, non-cell bound antibody rather than cell bound antibody immediately after dosing. The immune system recognizes the high concentration of the unbound antibody as non-dangerous and triggers the induction of immunological tolerance to TolerMab antibodies. We believe that TolerMab antibodies bind to their target and will be effective in the body after this tolerance induction is established. TolerMab antibodies are designed to be used repeatedly without the interference of a neutralizing immune response.

TolerMab TRX3

TolerMab TRX3 is a monoclonal antibody that binds to the CD2 receptor found on T cells and other subsets of lymphocytes. TolerMab TRX3 is in preclinical development for treatment of autoimmune diseases. Activated T cells have elevated levels of CD2 on their surface and have been shown to play an important role in autoimmune and inflammatory diseases such as psoriasis. Inhibiting the function of activated T cells by targeting the CD2 antigen is expected to have therapeutic benefit. Biogen Idec's Amevive®, which was recently approved for the treatment of patients with psoriasis, is a fusion protein that is directed at the CD2 antigen.

TolerMab TRX3 has been designed to have similar therapeutic benefits as other CD2-related therapies but without immunogenicity. In a recent clinical trial of psoriasis patients, MedImmune reported that as many as 50% of patients generated an undesirable immune response to its humanized anti-CD2 monoclonal antibody MEDI-507. We own all rights for the development and commercialization of TolerMab TRX3.

Research and discovery

In our research and discovery programs, we have discovered, isolated, and characterized cells that mediate immunological tolerance. We have developed a unique understanding of how the

48

immune system operates, both normally and in disease states. Our research and development efforts have produced and will continue to develop:

•  a database consisting of over 100 potential targets, including components of signaling pathways, co-stimulatory molecules, cell surface molecules, and cytokines that selectively block or enhance the function of T-effector cells and T-regulatory cells;

•  screening assays, or tests, for the identification of new tolerance promoting therapies;

•  animal models that allow for a direct validation of the potential targets or analysis of the efficacy of tolerance promoting therapies;

•   in vivo surrogate markers and prognostic indicators for tolerance to rapidly identify the most appropriate dosing strategies; and

•  novel products directed to our targets.

We have research and discovery programs to identify novel products involved in immunological tolerance mediated by T-regulatory cells. We believe that products directed to targets found to be associated with T-effector cells rather than T-regulatory cells, or that activate T-regulatory cells, could become tolerance-inducing therapies to treat patients with immune system diseases. We also believe that targets on certain subsets of dendritic cells could lead to tolerance promoting therapies by preventing the development and function of mature dendritic cells that present antigens to T cells in a manner that induces a classical immune response. These therapies could enhance the activity of more immature dendritic cells that present antigens to T cells in a manner that favors tolerogenic immune responses.

The majority of our research to date has been focused on the therapeutic induction of immunological tolerance. However, in some diseases, the immune system fails to identify and eliminate a danger to the body. Certain tumors and viruses, such as hepatitis C and HIV, evade detection and removal by the immune system. These tumors and viruses appear to have defense mechanisms that enable them to induce T-regulatory cells and a tolerogenic immune response instead of a classical immune response. In these diseases, our goal is to discover therapies that block the function of T-regulatory cells rather than T-effector cells in order to disrupt the specific tolerance mechanisms that have been generated and unleash the body's immune system to eliminate the tumor or virus.

Overall, we have identified and are validating potentially relevant targets and have filed several patent applications to protect these discoveries. We have generated multiple monoclonal antibodies and fusion proteins against many of our cell surface targets. We also intend to develop small molecule therapeutics to selected intracellular targets. We will use our development experience to advance these new drug candidates, and we expect to move at least one product into IND-enabling studies by the end of 2004. Three of these new monoclonal antibody products are:

ILT3 Antibody

ILT3 is an inhibitory receptor expressed by dendritic cells and a subset of white blood cells called monocytes. Our gene expression analysis revealed ILT3 expression to be increased on tolerogenic immature dendritic cells that lead to the initiation of a tolerogenic immune response as compared to mature dendritic cells that favor the classical immune response. Our

49

ILT3 antibody is expected to trigger inhibitory signaling, suppress immunogenic antigen presentation, and promote the development of tolerogenic antigen presentation to treat autoimmune disease or prevent transplant rejection.

CD160 Antibody

CD160 is a surface receptor molecule found on a small subset of T-effector cells that have cell killing activity. Our anti-CD160 monoclonal antibody is designed for use as an anti-inflammatory or tolerance-promoting drug to eliminate self-reactive T-effector cells without harming T-regulatory cells.

GITR Antibody

GITR is a surface receptor molecule that has been shown to be involved in inhibiting the suppressive activity of T-regulatory cells and extending the survival of T-effector cells. We believe that stimulation of GITR with a monoclonal antibody will break T-regulatory cell mediated tolerogenic immune responses to tumor and viral antigens allowing the development of anti-tumor and anti-viral T-effector cell classical immune responses. Our GITR antibody is expected to unleash the immune system and permit the T-effector cells to eliminate cancer cells or viruses.

Collaboration with Genentech

In December 2002, we entered into a collaboration agreement with Genentech for the development and commercialization of products that target the CD4 molecule, including our TRX1 monoclonal antibody. To date, we have received a total of $8.5 million from Genentech in the form of a license fee, an equity investment, a milestone payment, and funded research and development. Subject to the achievement of all development and regulatory approval milestones regarding multiple indications for products that target the CD4 molecule, we may receive up to approximately $80 million of additional funding from Genentech. If TRX1 receives marketing approval, Genentech would be responsible for commercialization and would record all revenue related to sales of TRX1, and we would receive royalty payments on net sales of TRX1. In lieu of receiving royalties on sales of TRX1 in the United States, we have the option to participate in a cost and profit sharing relationship and to support the commercialization of TRX1 through our own medical science liaisons who would educate the medical profession about our product, subject to Genentech's approval. We have also granted Genentech a right of first negotiation for products targeting the CD8 molecule on T cells, including our TRX2 monoclonal antibody. Genentech has the right to terminate our collaboration agreement at any time, with or without cause, 90 days after providing written notice. Otherwise, the agreement terminates upon the later of (a) the date when Genentech has made all of the royalty payments it is obligated to make under certain sections of the agreement and (b) if (and only if) we have exercised and continue to participate in the profit sharing option in the United States, the date when Genentech terminates sales of the products in the United States. There are also instances where a party may terminate the agreement for material breach, insolvency, or bankruptcy of the other party. Our collaboration with Genentech, including the receipt of any payments from Genentech, is subject to further risks set forth in "Risk factors" contained elsewhere in this prospectus, in particular, the risks relating to our collaboration with Genentech appearing on pages 8 and 9.

50

Patent and proprietary rights

We will be able to protect our proprietary rights from unauthorized use by third parties only to the extent that our proprietary rights are covered by valid and enforceable patents or trademarks or are effectively maintained as trade secrets. Accordingly, patent, trademark, and other proprietary rights are an essential element of our business.

To date, we have licenses to approximately 95 issued U.S. and foreign patents. Of these licensed patents, five have been granted in the United States, and they expire between 2013 and 2017. In addition, we own or have licenses to approximately 40 pending U.S. and foreign patent applications. Specifically, we have entered into a license agreement with Cambridge University Technical Services Limited (CUTS) with respect to our TRX1 and TRX2 products. The license is in the field of prevention, treatment, and diagnosis of human diseases and is exclusive with respect to the TRX1 and TRX2 antibodies, except for certain non-commercial rights retained by the licensor. We are obligated to pay royalties in each country for the longer of ten years from the date of the license or expiration of the licensed patents. Upon expiration of the royalty term, we retain a non-exclusive royalty free license to the unpatented technology. The license may be terminated by the licensor for material breach including but not limited to insolvency or bankruptcy, failure to make payments when due, and failure to exert reasonable commercial efforts to research, develop, and commercialize product under the license. We are obligated to pay patent costs and have the right to enforce licensed patents.

We have also entered into a license agreement with both the Chancellor Masters and Scholars of the University of Oxford and ISIS Innovation Limited with respect to our TRX1 and TRX2 products. The license is exclusive with respect to the TRX1 and TRX2 antibodies, except for certain non-commercial rights retained by the licensor. We are obligated to pay royalties in each country for the longer of ten years from the date of commercialization or expiration of the applicable patents in a country. Upon expiration of such royalty term, we retain a royalty free license. The license may be terminated by the licensor for certain uncured material breaches including, but not limited to failure to make payments when due, failure to exert reasonable commercial efforts to develop and commercialize the licensed product, and bankruptcy. We are obligated to pay patent costs and have the right to enforce licensed patents.

We have also been granted a license from CUTS with respect to our TRX3 product. The license is an exclusive license in the field of treating and preventing human disease, except for certain non-commercial rights retained by the licensor. We are obligated to pay royalties in each country for the longer of ten years from the date of the license agreement or the date of expiration of the licensed patents in a country. Upon expiration of the royalty term, we retain a non-exclusive license to the licensed unpatented technology. The license may be terminated by the licensor for an uncured material breach including but not limited to failure to make a payment when due, insolvency or bankruptcy, and failure to exert reasonable commercial efforts to research, develop, and commercialize product under the license. We are obligated to pay patent costs and have the right to enforce licensed patents.

We have a license from BTG International Limited (BTG) with respect to our TRX4 product. The license is exclusive as to patents and non-exclusive as to technology, in each case subject to non-commercial rights retained by the inventors. We are obligated to pay royalties on patented products and to also pay minimum royalties after launch of product. BTG is responsible for

51

paying patent costs and we have the right to enforce the patents at our expense. The license may be terminated by BTG for uncured material breach including but not limited failure to make payments when due, failure to exert reasonable efforts to develop product under a development plan, and insolvency and bankruptcy.

As our research and development efforts yield novel targets, small molecule drug candidates, and therapeutic antibodies, we intend to apply for broad composition of matter patents covering such targets, drug candidates, and antibodies. In addition, we intend to file patent applications claiming new methods of treating diseases for our identified drug candidates and antibodies. As we develop new technologies or novel improvements in our existing technologies, we will file for both composition of matter and method of use patents, as appropriate.

We are party to various agreements that give us rights to use technologies developed in collaborative research and development programs. These collaborations are intended to further our understanding of the mechanisms of tolerance induction and maintenance. We believe that this knowledge will lead to the discovery of novel targets, small molecule drug candidates, and therapeutic antibodies. We, either individually or jointly with our collaborators, in accordance with the terms of our agreements, may seek patent protection for these discoveries.

As a general matter, obtaining patents in the biopharmaceutical field is highly uncertain and involves complex legal, scientific, and factual matters. Obtaining a patent in the United States in the biopharmaceutical field can be expensive and can require several years. Failure to receive patents under the applications that we have filed, or that others have filed on our behalf, and may file from time to time could be harmful to us. Our patent filings in the United States may be subject to interference or re-examination proceedings. The defense and prosecution of interference and re-examination proceedings and related legal and administrative proceedings in the United States involve complex legal, scientific, and factual questions.

We also file patent applications outside of the United States. The laws of some foreign countries may not protect our proprietary rights to the same extent as do the laws of the United States. Third parties may attempt to oppose the issuance of our patents in foreign countries by initiating opposition proceedings. Additionally, if an opposition proceeding is initiated against any of our patent filings in a foreign country, that proceeding could have an adverse effect on the corresponding patents that are issued or pending in the United States. If we become involved in any interference, re-examination, opposition, or litigation proceedings in the United States or foreign countries regarding patent or other proprietary rights, those proceedings may result in substantial cost to us, regardless of the outcome. In addition, these proceedings may have a material adverse effect on our ability to develop, manufacture, market, or license our technologies or products, or to maintain or form strategic alliances.

Although we plan to aggressively prosecute our patent applications and defend our patents against third-party infringement, our patent applications may not result in the issuance of patents or, if issued, our patents may be challenged, invalidated, or circumvented. Moreover, our patents, to the extent they are or will be issued, may not provide us protection against competitors with other technologies. Our technologies and potential products may conflict with patents that have been or may be granted to competitors, universities, or others. In particular, patent applications or patents for innovative and broadly applicable technologies,

52

such as our TolerMab antibody platform, are sometimes challenged by third parties as obvious or as obvious extensions of technologies previously developed by those third parties.

As the biopharmaceutical industry expands and more patents are issued, the risk increases that our technologies and potential products may give rise to claims that they infringe the patents of others. Third parties claiming infringement of their proprietary rights could bring legal actions against us claiming damages and seeking to enjoin our commercialization of a product or our use of a technology. For example, we are aware of a third party patent application that would cover the use of a non-mitogenic anti-CD3 antibody for the treatment of autoimmune disease, as well as an issued third party patent covering the ILT3 molecule and the use of antibodies directed at that target. If any actions based on these claims are successful, in addition to any potential liability for damages, we could be required to obtain a license in order to continue to use a technology or to manufacture or market a product, or could be required to cease using those products or technologies. Any claim, with or without merit, could result in costly litigation and divert the efforts and attention of our scientific and management personnel. We may not prevail in any action, and any license required under any patent may not be made available on acceptable terms, if at all.

In addition to patent protection, we may rely on trade secrets, proprietary know-how, and continuing technological advances to develop and maintain our competitive position. To maintain the confidentiality of our trade secrets and proprietary information, all of our employees are required to enter into and adhere to a confidentiality and invention assignment agreement as a condition of employment. Additionally, these agreements require that our employees do not bring to us, or use without proper authorization, any third-party proprietary technology. We also require all of our consultants and collaborators that have access to proprietary property to execute confidentiality and invention assignment agreements in our favor before beginning their relationship with us. While such arrangements are intended to enable us to better control the use and disclosure of our proprietary property and provide for our ownership of proprietary technology developed on our behalf, they may not provide us with meaningful protection for such property and technology in the event of unauthorized use or disclosure.

Manufacturing and supply

We have relied upon Biovest International, Laureate Pharma, and the Therapeutic Antibody Centre of the University of Oxford to produce material for preclinical and clinical testing purposes. We have received sufficient material from these third parties to meet our current needs. Under our collaboration agreement with Genentech, we expect that Genentech will manufacture the TRX1 monoclonal antibody for clinical and commercial use.

Government regulation

U.S. regulatory requirements

The U.S. Food and Drug Administration, or FDA, and comparable regulatory agencies in foreign countries regulate and impose substantial requirements upon the research, development, preclinical and clinical testing, labeling, manufacture, quality control, storage, approval, advertising, promotion, marketing, distribution, and export of pharmaceutical products,

53

including biologics, as well as significant reporting and record-keeping obligations. State governments may also impose obligations in these areas.

In the United States, pharmaceutical products are regulated by the FDA under the federal Food, Drug, and Cosmetics Act, or FDCA, and other laws, including in the case of biologics, the Public Health Service Act. We believe, but cannot be certain, that our products will be regulated as biologics and drugs by the FDA. The process required by the FDA before biologics or drugs may be marketed in the United States generally involves the following:

•  preclinical laboratory and animal tests performed under the FDA's Good Laboratory Practices regulations to assess potential safety and effectiveness;

•  submission and approval of an IND which must become effective before clinical trials may begin in the United States;

•  obtaining approval of Institutional Review Boards to protect the welfare and rights of human subjects in clinical trials;

•  adequate and well-controlled human clinical trials to establish the safety and efficacy of the product in the product's intended use;

•  development of manufacturing processes which conform to FDA-mandated current Good Manufacturing Practices, or cGMPs;

•  laboratory evaluation of the product's formulation and stability; and

•  FDA review and approval of either a Biologics License Application, or BLA, or a New Drug Application, or NDA, prior to any commercial sale or shipment of a product.

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

The results of the preclinical tests, together with initial specified manufacturing information, the proposed clinical trial protocol, and information about the participating investigators, are submitted to the FDA as part of an IND, which must be approved before we may begin human clinical trials. Additionally, an independent Institutional Review Board at each medical site proposing to conduct the clinical trials must review and approve each study protocol and oversee conduct of the trial. An IND becomes effective 30 days after receipt by the FDA, unless the FDA, within the 30-day period, raises concerns or questions about the conduct of the trials as outlined in the IND and imposes a clinical hold. If the FDA imposes a clinical hold, the IND sponsor must resolve the FDA's concerns before clinical trials can begin. Preclinical tests and studies can take several years to complete, and there is no guarantee that an IND we submit based on such tests and studies will become effective within any specific time period, if at all.

Human clinical trials are typically conducted in three sequential phases that may overlap:

•  Phase I: The drug is initially introduced into healthy human subjects or patients and tested for safety and dosage tolerance. Absorption, metabolism, distribution, and excretion testing is generally performed at this stage.

•  Phase II: The drug is studied in controlled, exploratory therapeutic trials in a limited number of subjects with the disease or medical condition for which the new drug is intended to be

54

used in order to identify possible adverse effects and safety risks, to determine the preliminary or potential efficacy of the product for specific targeted diseases or medical conditions, and to determine dosage tolerance and the optimal effective dose.

•  Phase III: When Phase II studies demonstrate that a specific dosage range of the drug is likely to be effective and the drug has an acceptable safety profile, controlled, large- scale therapeutic Phase III trials are undertaken at multiple study sites to demonstrate clinical efficacy and to further test for safety in an expanded patient population.

We cannot be certain that we will successfully complete Phase I, Phase II, or Phase III testing of our products within any specific time period, if at all. Furthermore, the FDA, the Institutional Review Board or we may suspend or terminate clinical trials at any time on various grounds, including a finding that the subjects or patients are being exposed to an unacceptable health risk.

Results of preclinical studies and clinical trials, as well as detailed information about the manufacturing process, quality control methods, and product composition, among other things, are submitted to the FDA as part of a BLA or NDA seeking approval to market and commercially distribute the product on the basis of a determination that the product is safe and effective for its intended use. BLAs are used for products that are regulated as biologics, such as antibodies, and NDAs are used for products that are regulated as drugs, such as synthetic chemicals. Before approving a BLA or NDA, the FDA will inspect the facilities at which the product is manufactured and will not approve the product unless cGMP compliance is satisfactory. If applicable regulatory criteria are not satisfied, the FDA may deny the BLA or NDA or require additional testing or information. As a condition of approval, the FDA also may require post-marketing testing or surveillance to monitor the product's safety or efficacy. Even after a BLA or NDA is approved, the FDA may impose additional obligations or restrictions (such as labeling changes), or even suspend or withdraw product approval on the basis of data that arise after the product reaches the market, or if compliance with regulatory standards is not maintained. We cannot be certain that any BLA or NDA we submit will be approved by the FDA on a timely basis, if at all. Also, any such approval may limit the indicated uses for which the product may be marketed. Any refusal to approve, delay in approval, suspension or withdrawal of approval, or restriction on indicated uses could have a material adverse impact on our business.

Each BLA or NDA must be accompanied by a user fee, pursuant to the requirements of the Prescription Drug User Fee Act, or PDUFA, and its amendments. According to the FDA's fee schedule, effective on October 1, 2003 for the fiscal year 2004, the user fee for an application requiring clinical data, such as a BLA or NDA, is $573,500. The FDA adjusts the PDUFA user fees on an annual basis. PDUFA also imposes an annual product fee for prescription drugs and biologics ($36,080 for the fiscal year 2004), and an annual establishment fee ($226,800) on facilities used to manufacture prescription drugs and biologics. We are not at the stage of development with our products where we are subject to these fees, but they are significant expenditures that will be incurred in the future and must be paid at the time of application submission to FDA.

Satisfaction of FDA requirements typically takes several years. The actual time required varies substantially, based upon the type, complexity, and novelty of the pharmaceutical product, among other things. Government regulation imposes costly and time-consuming requirements

55

and restrictions throughout the product life cycle and may delay product marketing for a considerable period of time, limit product marketing, or prevent marketing altogether. Success in preclinical or early stage clinical trials does not assure success in later stage clinical trials. Data obtained from preclinical and clinical activities is not always conclusive and may be susceptible to varying interpretations that could delay, limit, or prevent marketing approval. Even if a product receives marketing approval, the approval is limited to specific clinical indications. Further, even after marketing approval is obtained, the discovery of previously unknown problems with a product may result in restrictions on the product or even complete withdrawal of the product from the market.

After product approval, there are continuing significant regulatory requirements imposed by the FDA, including record-keeping requirements, obligations to report adverse experiences, and restrictions on advertising and promotional activities. Quality control and manufacturing procedures must continue to conform to cGMPs, and the FDA periodically inspects facilities to assess cGMP compliance. Additionally, post-approval changes in manufacturing processes or facilities, product labeling, or other areas require FDA review and approval. Failure to comply with FDA regulatory requirements may result in enforcement action by the FDA, including product recalls, suspension or revocation of product approval, seizure of product to prevent distribution, impositions of injunctions prohibiting product manufacture or distribution, and civil and criminal penalties. Maintaining compliance is costly and time-consuming. Nonetheless, we cannot be certain that we, or our present or future suppliers or third-party manufacturers, will be able to comply with all FDA regulatory requirements, and potential consequences of noncompliance could have a material adverse impact on our business.

FDA's policies may change, and additional government regulations may be enacted that could delay, limit, or prevent regulatory approval of our products or affect our ability to manufacture, market, or distribute our products after approval. Moreover, increased attention to the containment of healthcare costs in the United States and in foreign markets could result in new government regulations that could have a material adverse effect on our business. Our failure to obtain coverage, an adequate level of reimbursement, or acceptable prices for our future products could diminish any revenues we may be able to generate. Our ability to commercialize future products will depend in part on the extent to which coverage and reimbursement for the products will be available from government and health administration authorities, private health insurers, and other third-party payors. European Union and U.S. government and other third-party payors increasingly are attempting to contain healthcare costs by consideration of new laws and regulations limiting both coverage and the level of reimbursement for new drugs. We cannot predict the likelihood, nature or extent of adverse governmental regulation that might arise from future legislative or administrative action, either in the United States or abroad.

Our activities also may be subject to state laws and regulations that affect our ability to develop and sell our products. We are also subject to numerous federal, state, and local laws relating to such matters as safe working conditions, clinical, laboratory, and manufacturing practices, environmental protection, fire hazard control, and disposal of hazardous or potentially hazardous substances. We may incur significant costs to comply with such laws and regulations now or in the future, and the failure to comply may have a material adverse impact on our business.

56

The FDCA includes provisions intended to facilitate and expedite the development and review of drugs and biological products intended for treatment of serious or life-threatening conditions that demonstrate the potential to address unmet medical needs for such conditions. These provisions set forth a procedure for designation of a drug as a "fast track" product. Concurrent with or after an IND is filed, the sponsor may request designation as a fast track product, and the FDA is required to respond within 60 days.

Hatch-Waxman Act

We intend to take advantage of statutory marketing exclusivity provisions for our products. Under the U.S. Drug Price Competition and Patent Term Restoration Act of 1984, known as the Hatch-Waxman Act, newly approved drugs and indications benefit from a statutory period of non-patent marketing exclusivity. The Hatch-Waxman Act provides five years of marketing exclusivity to the first applicant to gain approval of an NDA for a new chemical entity, meaning that the FDA has not previously approved any other new drug containing the same active ingredient. Hatch-Waxman prohibits an abbreviated new drug application, an ANDA, or an NDA where the applicant does not own or have a legal right of reference to all the data required for approval, to be submitted by another company for another version of such drug during the five year exclusive period. Protection under Hatch-Waxman will not prevent the filing or approval of another NDA, however, the applicant would be required to conduct its own adequate and well-controlled clinical trials to demonstrate safety and effectiveness. The Hatch-Waxman Act also provides three years of marketing exclusivity for the approval of new NDAs with new clinical trials for previously approved drugs and supplemental NDAs, for example, for new indications, dosages, or strengths of an existing drug, if new clinical investigations are essential to the approval. This three year exclusivity covers only the new changes associated with the supplemental NDA and does not prohibit the FDA from approving ANDAs for drugs containing the original active ingredient.

The Hatch-Waxman Act also permits a patent extension term of up to five years as compensation for patent term lost during product development and the FDA regulatory review process. However, patent extension cannot extend the remaining term of a patent beyond a total of 14 years. The patent term restoration period is generally one-half the time between the effective date of an IND and the submission date of the NDA or BLA, plus the time between the submission date of the NDA or BLA and the approval of that application. Only one patent applicable to an approved drug is eligible for the extension and it must be applied for prior to expiration of the patent. The U.S. Patent and Trademark Office, in consultation with FDA, reviews and approves the application for the patent term extension. We will consider applying for patent term extensions for our patents to add patent life beyond the expiration date, depending on the length of clinical trials and other factors involved in the filing of a NDA or BLA.

Best Pharmaceuticals for Children Act

The Best Pharmaceuticals for Children Act, signed into law on January 4, 2002, provides an additional six months of marketing exclusivity for new or marketed drugs for specific pediatric studies conducted at the written request of the FDA. On December 3, 2003, the Pediatric Research Equity Act was signed into law, authorizing the FDA to require pediatric studies for drugs and biological products to ensure the drugs' or products' safety and effectiveness in

57

children. This Act required that new NDAs, BLAs or supplements to NDAs or BLAs contain data assessing the safety and effectiveness for the claimed indication in all relevant pediatric subpopulations. Dosing and administration must be supported for each pediatric subpopulation for which the drug is safe and effective. The FDA may grant deferrals for submission of data, or full or partial waivers.

Foreign regulatory requirements

Outside the United States, our ability to market our products will also be contingent upon receiving marketing authorizations from the appropriate regulatory authorities and compliance with applicable post-approval regulatory requirements. Although the specific requirements and restrictions vary from country to country, as a general matter, foreign regulatory systems include risks similar to those associated with FDA regulation, described above. Under EU regulatory systems, marketing authorizations may be submitted either under a centralized or decentralized procedure. Under the centralized procedure, a single application to the European Medicines Evaluation Agency (EMEA) leads to an approval granted by the European Commission which permits the marketing of the product throughout the EU. The centralized procedure is mandatory for certain classes of medicinal products but is optional for others. For example, all medicinal products developed by means of recombinant DNA technology, gene expression in prokaryotes and eukaryotes, or monoclonal antibody methods must be authorized via the centralized procedure. We assume that the centralized procedure will apply to our products that are developed by means of one or more of such biotechnology processes. The decentralized procedure provides for mutual recognition of nationally approved decisions and is used for products that are not required to be authorized by the centralized procedure and those products for which the centralized procedure is optional but which shall be marketed in select EU member countries only. Under the decentralized procedure, the holder of a national marketing authorization may submit further applications to the competent authorities of the remaining member states which will then be requested to recognize the original authorization based upon an assessment report prepared by the original authorizing competent authority. The recognition process should take no longer than 90 days, but if one member state makes an objection, which under the legislation can only be based on a possible risk to human health, we have the option to withdraw the application from that country or take the application to arbitration by the Committee for Proprietary Medicinal Products (CPMP) of the EMEA. If a referral for arbitration is made, the procedure is suspended, and in the intervening time, the only EU country in which the product can be marketed will be the country where the original authorization has been granted, even if all the other designated countries are ready to recognize the product. The opinion of the CPMP, which is binding, could support or reject the objections or alternatively could reach a compromise position acceptable to all EU countries concerned. Arbitration can be avoided if the application is withdrawn in the objecting country, but once the application has been referred to arbitration, it cannot be withdrawn. The arbitration procedure may take an additional year before a final decision is reached and may require the delivery of additional data.

As with FDA approval, we may not be able to secure regulatory approvals in Europe in a timely manner, if at all. Additionally, as in the United States, post-approval regulatory requirements, such as those regarding product manufacture, marketing, or distribution, would apply to any product that is approved in Europe, and failure to comply with such obligations could have a material adverse effect on our ability to successfully commercialize any product.

58