Back to GetFilings.com

SECURITIES AND EXCHANGE COMMISSION

Washington, D.C. 20549

FORM 10-K

Commission File No. 0-28298

ONYX PHARMACEUTICALS, INC.

(Exact name of registrant as specified in its charter)

3031 Research Drive
Richmond, California 94806
(510) 222-9700

(Address, including zip code, and telephone number, including area code,
of registrant's principal executive offices)

Securities Registered Pursuant to Section 12(b) of the Act: None

Securities Registered Pursuant to Section 12(g) of the Act:

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

    Indicate by check mark if disclosure of delinquent filers pursuant to Item 405 of Regulation S-K (Section 229.405 of this chapter) is not contained herein, and will not be contained, to the best of registrant's knowledge, in definitive proxy or information statements incorporated by reference in Part III of this Form 10-K or any amendment to this Form 10-K. / /

    The aggregate market value of the voting stock held by nonaffiliates of the Registrant based upon the last trade price of the common stock reported on the Nasdaq National Market on March 28, 2001 was approximately $122,182,000.

    The number of shares of common stock outstanding as of March 28, 2001 was 18,428,102.

DOCUMENTS INCORPORATED BY REFERENCE

    Portions of Onyx's Definitive Proxy Statement filed with the Commission pursuant to Regulation 14A in connection with the 2001 Annual Meeting are incorporated herein by reference into Part III of this report.

    Certain Exhibits filed with Onyx's Registration Statement on Form SB-2 (Registration No. 333-3176-LA), as amended, Onyx's Annual Report on Form 10-K for the years ended December 31, 1997 and December 31, 1999, Onyx's Quarterly Reports on Form 10-Q for the quarters ended June 30, 1996, March 31, 1997, September 30, 1997, March 31, 1999, September 30, 1999, June 30, 2000 and September 30, 2000, Onyx's Current Report on Form 8-K filed on January 26, 1998, Onyx's Current Report on Form 8-K filed on March 1, 2000, and Onyx's Current Report on Form 8-K filed on February 23, 2001, are incorporated by reference into Part IV of this Report.

PART I.

    This Annual Report on Form 10-K contains forward-looking statements within the meaning of Section 27A of the Securities Act of 1933 and Section 21E of the Securities Act of 1934. These statements involve known and unknown risks, uncertainties and other factors that may cause our or our industry's results, levels of activity, or achievements to differ significantly and materially from that expressed or implied by such forward-looking statements. These factors include, among others, those listed under "Additional Business Risks" and elsewhere in this Annual Report.

    In some cases, you can identify forward-looking statements by terminology such as "may," "will," "should," "intend," "expect," "plan," "anticipate," "believe," "estimate," "predict," "potential," or "continue," or the negative of such terms or other comparable terminology.

    Although we believe that the expectations reflected in the forward-looking statements are reasonable, we cannot guarantee future results, events, levels of activity, performance or achievements. We do not assume responsibility for the accuracy and completeness of the forward-looking statements. We do not intend to update any of the forward-looking statements after the date of this Annual Report to conform these statements to actual results, unless required by law.

Item 1.  Business

Overview

    We are developing innovative products for the treatment of cancer. Based on our proprietary virus technology, we are developing our lead product, CI-1042, formerly known as ONYX-015. CI-1042 is a human virus genetically engineered to selectively replicate in and kill cancer cells based on the mutation or loss of function of a specific tumor suppressor gene, the p53 gene. We believe that the mutation or loss of function of the p53 gene is present in the majority of human cancers.

    We have completed two Phase II clinical trials of CI-1042 for the treatment of head and neck cancer. Based on the data from these trials, we have initiated a 300-patient, multi-center Phase III clinical trial of CI-1042 administered by direct injection into tumors, or intratumoral injection, in patients with head and neck cancer. This Phase III clinical trial is being conducted in conjunction with Warner-Lambert Company, a wholly-owned subsidiary of Pfizer, Inc., our collaborator for the development and commercialization of CI-1042.

    We have conducted a Phase I/II clinical trial evaluating CI-1042 administered through an artery leading to the liver, or intrahepatic artery infusion, in patients with colon cancer that has spread to the liver, or liver metastases of colorectal cancer. We are conducting a separate Phase I/II clinical trial evaluating CI-1042 in patients with pancreatic cancer. In addition, a Phase I clinical trial has been completed evaluating CI-1042 in patients who have advanced cancers with tumors in the lung. Based on the results from these trials, we have initiated a Phase II clinical trial evaluating CI-1042 intravenously in patients with liver metastases of colorectal cancer. In addition, we have initiated a Phase II clinical trial in patients with non-small cell lung cancer.

    In October 1999, we entered into a collaboration with Warner-Lambert to develop and commercialize CI-1042 as well as two additional product candidates. These product candidates are human viruses that incorporate anticancer genes, or armed viruses. Pfizer subsequently acquired Warner-Lambert. Under the terms of this collaboration, we have the right to co-promote CI-1042 and the two additional product candidates in the United States and Canada and to share equally in the profits or losses. Moreover, under the terms of the collaboration, Warner-Lambert is responsible for commercializing the products in the rest of the world and is obligated to pay us royalties based on net sales in these markets. Additionally, Warner-Lambert is obligated to fund up to $40 million of development costs related to CI-1042 and to fully fund preclinical development of the two armed viruses.

2

    In addition to CI-1042, we are conducting a number of research and development programs based on our virus technology as well as our small molecule drug discovery efforts. Together with Bayer, we initiated a Phase I clinical trial of a compound that blocks inappropriate growth signals in tumor cells by inhibiting the raf kinase enzyme. We intend to initiate with Bayer a Phase I/II clinical trial in patients with acute myelogenous leukemia in Canada in the first half of 2001 and Phase II clinical trials in various solid tumor types in the second half of 2001.

    Further, we are developing human viruses that selectively replicate in and kill cancer cells based on mutations or loss of function of the retinoblastoma, or RB, tumor suppressor gene. We have shown that these RB-selective therapeutic viruses produce significant tumor shrinkage in animal models. Pending the final results of these ongoing preclinical studies, we intend to file an investigational new drug application, or IND, for a product candidate from our RB gene program in mid-2002. In our armed virus program, we are currently developing two classes of armed viruses: (A) viruses armed with prodrug converting enzymes and (B) viruses armed with genes that activate an immune response to tumors, or cytokines. We are currently developing two armed virus candidates, one from each class, under our collaborative agreement with Warner-Lambert.

Our Product Candidates

    We are developing the product candidates listed below:

CI-1042

    Our lead product, CI-1042, is a human virus that is genetically engineered so that it does not make E1B 55k, a viral protein that binds to a specific tumor suppressor protein, or p53, and blocks its function. As a result, CI-1042 cannot inactivate p53 or its function and, therefore, cannot effectively replicate in normal cells. However, in most cancer cells, p53 has already lost its function through mutation of the p53 gene or other mutations in the p53 pathway. When CI-1042 infects these cancer cells, the virus growth cycle proceeds and the cancer cells are killed. New viruses are then released and

3

infect neighboring cancer cells, killing those cells and perpetuating the selective cancer killing cycle. This process continues, amplifying the therapeutic effect.

    We are currently conducting Phase II and Phase III clinical trials using CI-1042 to treat several cancer types. The American Cancer Society projects that the numbers of new cases and deaths from these types of cancers in the United States for the year 2000 are as follows:

*

Includes cancers of the larynx, tongue, mouth, oral cavity and pharynx.

Head and Neck Clinical Programs

    We have completed two Phase II clinical trials with CI-1042 in head and neck cancer patients. In the first trial, patients with head and neck cancer that is resistant to all therapies, or refractory disease, received intratumoral injections of CI-1042 daily over five consecutive days. In this trial, five of 21 evaluable patients experienced tumor regressions of 50% or more. In two of these cases, the patient experienced a 100% reduction in the size of the injected tumor. In this trial, patients generally tolerated treatment with CI-1042 well. Sixty percent of the patients experienced minor flu-like symptoms.

    In the most recent Phase II clinical trial, patients with head and neck cancer that has progressed following initial treatment with surgery and/or radiation, or recurrent disease, received CI-1042 administered in combination with Cisplatin and 5-Fluorouracil. These two chemotherapeutic drugs are the current standard of care for patients with recurrent head and neck tumors. In this trial, the dosing regimen provided for:

•

daily direct intratumoral injections of CI-1042 on each of days one through five;



•

intravenous dosing of Cisplatin on day one; and



•

intravenous dosing of 5-Fluorouracil on each of days one through five.

    In this trial, 19 of the 30 evaluable patients, or 63%, experienced regression of 50% or more in their injected tumors. Eight of the patients, or 27%, experienced a 100% regression in the size of their injected tumors. We believe these rates compare favorably to past trials involving treatment by chemotherapy alone in which approximately 35% of the patients experienced tumor regressions of greater than 50%, and less than 10% experienced a complete tumor regression. In this trial, the patients generally tolerated the treatment well. Some patients, however, experienced chemotherapy-related gastrointestinal symptoms and injection site pain.

    In a separate analysis of this same trial, we compared the efficacy of CI-1042 plus chemotherapy in the injected tumor to the efficacy of chemotherapy alone in distant uninjected tumors. Eleven patients had more than one tumor, and the investigators injected CI-1042 only in the largest tumor that caused the most severe symptoms. Nine of the 11 tumors injected with CI-1042 responded, while only three of the uninjected tumors in the 11 patients responded. We believe that this trial, which considers the patients as their own control, demonstrates the superiority of CI-1042 plus chemotherapy over chemotherapy alone.

    Together with Warner-Lambert, we initiated a pivotal Phase III clinical trial for the treatment of recurrent head and neck disease. The trial compares intratumoral injection of CI-1042 plus standard

4

chemotherapy, 5-Fluorouracil and Cisplatin, versus standard chemotherapy alone. We are conducting the trial at numerous centers in the United States and Europe and will include approximately 300 patients, half in each arm of the study. Currently we have enrolled patients in ten centers, but because of the limited availability of clinical materials, we have only supplied two sites with CI-1042. The primary objectives, or endpoints, of the trial are to establish improvements in (A) lasting reductions in the size of the treated tumor, also known as durable tumor response, and (B) patient survival without further growth of the treated tumor, also known as progression free survival. Secondary endpoints are to demonstrate higher quality of life and overall survival of patients.

    We are also planning to initiate an open-label Phase II/III clinical trial that will evaluate CI-1042 in patients with refractory head and neck cancer. We expect this trial will include 50 to 100 patients who will receive CI-1042 in combination with chemotherapy with durable tumor response as the primary endpoint. We intend to use this second head and neck clinical trial to support the results of our pivotal Phase III clinical trial, which we are using to obtain regulatory approval from the Food and Drug Administration, or FDA.

Other Clinical Programs

    We have completed a Phase I/II clinical trial evaluating CI-1042 administered through intrahepatic artery infusion in patients with liver metastases of colorectal cancer and pancreatic cancer. In addition, we have completed a Phase I clinical trial evaluating CI-1042 in patients who have advanced cancers with tumors in the lung. The results from these trials described below are at an early stage of development and we may not observe sufficient anticancer activity or clinical benefit in later trials. In addition, we have recently initiated a Phase II clinical trial evaluating CI-1042 administered intravenously in patients with liver metastases of colorectal cancer and a separate Phase II clinical trial evaluating CI-1042 administered intravenously in patients with non-small cell lung cancer.

Liver Metastases of Colorectal Cancer.  Colorectal cancer is the third leading cause of cancer death in the United States. Aside from the standard therapies for treatment of colorectal cancer, currently there is no standard treatment specifically for liver metastases of colorectal cancer, other than surgery. Approximately 5% of patients with liver metastases of colorectal cancer have disease that may be surgically treated.

    We have completed a Phase I/II clinical trial in which CI-1042 is administered through intrahepatic artery infusion, in patients with liver metastases of colorectal cancer. Intrahepatic artery infusion permits simultaneous delivery of CI-1042 to multiple tumor sites within the liver. In this trial, patients received initial cycles of treatment with CI-1042 alone and then received a combination of CI-1042 plus 5-Flourouracil and Leucovorin, the standard chemotherapeutic regimen for colorectal cancer treatment, for subsequent treatment cycles. A total of 33 patients were treated and evaluated. Six patients were treated in the dose escalation phase and 27 patients received the high doses. All patients tolerated the regimen. We are in the process of data analysis from this trial. Preliminary data suggest that for patients receiving the highest dose, survival was prolonged compared with patients receiving lower doses. Patients who received low doses survived a maximum of 250 days. However, patients receiving higher doses had a median survival rate of greater than 250 days with 10 of 27 patients surviving more than one year. We plan to present the final analysis of this trial at future scientific meetings.

Pancreatic Cancer.  Pancreatic cancer is the fifth leading cause of cancer death in the United States. The median survival rate for patients with pancreatic cancer is 4-6 months and the 5-year survival rate is 3%.

    We have reported interim results of an ongoing Phase I/II clinical trial in which we treated patients with pancreatic cancer with CI-1042 by intratumoral injections guided by ultrasound. This technique allows for multiple injections per treatment by guiding a needle with ultrasound down the esophagus and injecting the tumor. Patients who tolerated the initial cycles of treatment with CI-1042 alone then

5

received Gemcitabine, a chemotherapeutic agent that is the standard treatment for pancreatic cancer, in combination with CI-1042 in four subsequent treatment cycles and, thereafter, received Gemcitabine alone. Two additional groups of patients remain to be treated with higher doses in this trial. In 21 evaluable patients, two patients achieved 50% or greater reduction in their tumors. In addition, two patients experienced a 25-50% reduction in their tumors.

Advanced Cancers with Tumors in the Lung.  One of our independent investigators has completed a Phase I clinical trial in 10 patients with advanced cancers with tumors in the lung. The primary objectives of the trial were to determine safety, distribution and biological effects of intravenously administered CI-1042. CI-1042 was administered intravenously in combination with Carboplatin and Taxol, the standard chemotherapeutic treatment of lung cancer. All patients tolerated the treatment well. As follow up to this trial, we have initiated a Phase II clinical trial involving intravenous treatment with CI-1042 in patients with non-small cell lung cancer. Lung cancer is the leading cause of cancer death in the United States. While several new therapies have recently been approved for use, the 5-year survival rate for patients with lung cancer is approximately 14%.

Other Product Opportunities

    In addition to CI-1042, we are also conducting a number of research and development programs based on our virus technology as well as our small molecule drug discovery efforts.

Raf Kinase Inhibitor

    The ras gene and its related biochemical pathway, or the ras signaling pathway, play a key role in cell proliferation. In normal cell proliferation, when the ras signaling pathway is activated, or "on," it sends a signal telling the cell to grow and divide. When a gene in the ras signaling pathway is mutated, the signal may not turn "off," causing the cell to continuously reproduce itself. We believe that the ras signaling pathway plays an integral role in growth of some tumor types, and that inhibiting this pathway could have an effect on the tumor growth. Mutations in the ras gene occur in approximately 30% of all human cancers, including 90% of pancreatic cancer, 50% of colon cancer and some lung cancers.

    With Bayer, we have initiated a Phase I clinical trial of a compound that blocks ras signaling in cells by inhibiting a specific enzyme known as raf kinase. We intend to initiate with Bayer a Phase I/II clinical trial in patients with acute myelogenous leukemia in Canada in the first half of 2001 and Phase II clinical trials in various solid tumor types in the second half of 2001.

RB-Selective Therapeutic Viruses

    We believe that the RB pathway function is abnormal in most cancers. We have engineered human viruses to selectively replicate in and kill cancer cells based on defects in the RB function in these cells. In preclinical studies, we have shown that these viruses produce significant tumor shrinkage in animal models. Pending the final results of these ongoing preclinical studies, we plan to designate a RB-selective therapeutic virus as a clinical candidate and file an IND in mid-2002.

Armed Therapeutic Viruses

    We are developing therapeutic viruses armed with anticancer genes. A number of anticancer chemotherapeutic agents are inactive forms of drugs, or prodrugs, that are converted to their active forms by specific chemical modifications. Specific enzymes, or prodrug converting enzymes, are known to carry out this conversion.

    We have developed a system to arm our therapeutic viruses with genes that produce prodrug converting enzymes. With this strategy, we arm each virus with a single gene that produces an enzyme that converts a prodrug into an anticancer chemotherapeutic drug at the tumor site. As a result, we

6

believe that arming our viruses enables the concentrated delivery of the anticancer chemotherapeutic drug to the tumor, thereby enhancing the therapeutic effect. To date, we have obtained licenses to four prodrug converting enzymes for arming our therapeutic viruses.

    In addition, we are developing a class of viruses armed with genes, particularly genes that encode cytokines, that activate an immune response in the tumor. This immune response may result either from the protein expressed by the gene itself, or from other elements of the human immune system that are activated as a result.

Cell Cycle and Inflammation Programs

    Together with Warner-Lambert, we have identified a number of lead compounds that modulate the activity of key enzymes that regulate the process whereby a single cell replicates itself and divides into two identical new cells, or the cell cycle. Mutations in genes that regulate the cell cycle are present in a majority of human cancers. We are currently evaluating the most advanced lead compound in this program for its effectiveness in animal models of human cancer. We also have identified a lead compound that inhibits an enzyme associated with inflammatory diseases. We expect joint research under these programs to continue until the second half of 2001. Thereafter, if Warner-Lambert develops and markets products from this research, we will receive royalties on their worldwide sales.

Business Strategy

    Our objective is to be a leading developer of novel cancer therapies. We intend to develop and commercialize a broad portfolio of products based primarily on our selectively-replicating virus technology. Elements of our strategy are to:

•

Obtain regulatory and marketing approvals for CI-1042 for head and neck cancer. Together with Warner-Lambert, we have initiated a Phase III clinical trial aimed at gaining regulatory approval of CI-1042 in patients with recurrent disease. We also plan to initiate a Phase II/III clinical trial in patients with refractory disease.



•

Develop CI-1042 for other cancer types. We have studied CI-1042 in Phase I or Phase I/II clinical trials in patients with liver metastases of colorectal cancer, pancreatic cancer and other advanced cancers with tumors in the lung. Together with Warner-Lambert, we have initiated a Phase II clinical trial evaluating CI-1042 in patients with liver metastases of colorectal cancer. In addition, we have initiated a Phase II clinical trial in patients with non-small cell lung cancer. We are also conducting Phase I clinical trials in other cancer types. Based on results from these trials, we expect to initiate further clinical trials in an effort to generate data supporting the therapeutic value of CI-1042 in multiple cancer types.



•

Identify and develop additional virus products. We have designated a RB-selective therapeutic virus candidate and plan to file an IND by mid-2002. In addition, we are developing armed viruses that incorporate prodrug converting enzymes to selectively activate chemotherapeutic drugs at the tumor site and armed viruses incorporating cytokines. In collaboration with Warner-Lambert, we intend to select a cytokine for use in an armed virus by mid-2001.



•

Retain substantial co-promotion rights to our products. We intend to build sales and marketing capabilities to promote our products in the United States and Canada. In establishing new product collaborations, we seek to retain at least 50% of marketing and profit-sharing rights in the United States. In our collaboration with Warner-Lambert for CI-1042 and the two armed viruses, we have retained co-promotion rights as well as the opportunity for equal profit sharing in the United States and Canada. Our collaboration with Bayer offers us co-promotion and profit-sharing rights for any developed product in the United States.



•

Acquire technologies and products that complement our business. We intend to in-license or acquire complementary technologies to enhance our product portfolio and strengthen our market position. Currently, we are focusing on accessing technologies to strengthen our efforts in arming our therapeutic viruses with anticancer genes as well as developing ways to modify our viruses to improve their activity when administered intravenously.

7

Collaborations

Warner-Lambert: CI-1042 and Armed Viruses

    Effective September 1999, we entered into an agreement with Warner-Lambert for the purpose of developing and commercializing CI-1042 and two armed therapeutic viruses. Under the terms of this agreement, we have the right to co-promote CI-1042 and the two armed virus products with Warner-Lambert in the United States and Canada. We also have the right to share equally in resulting profits or losses in these territories. Additionally, Warner-Lambert is responsible for commercializing the products in the rest of the world and is obligated to pay us a royalty on net sales in these markets.

    We will jointly manage the development of these products. Warner-Lambert is primarily responsible for conducting the agreed-upon pivotal Phase III clinical trials of CI-1042 in head and neck cancer and other pivotal trials in various cancer types. We are required to share responsibility for exploratory Phase I and Phase II clinical trials with Warner-Lambert. Warner-Lambert is responsible for the commercial manufacturing of the collaboration products.

    Warner-Lambert is obligated to provide the first $40 million in funding for clinical development of CI-1042. We are obligated to fund 25% of the costs over $40 million to retain our profit sharing and co-promotion rights in the United States and Canada, and Warner-Lambert is obligated to fund the remaining 75%. Warner-Lambert is also required to fully fund the research and preclinical development of the two armed virus product candidates. When a product candidate is selected for clinical development, we will be obligated to pay 25% of the development costs to retain our profit-sharing and co-promotion rights in the United States and Canada for any resulting product. If we choose not to or are unable for any reason to (A) fund our portion of the development costs for CI-1042 or an armed virus product candidate, or (B) maintain our required co-promotion effort, then:

•

we would lose our co-promotion and profit-sharing rights for that product;



•

we would license the product exclusively to Warner-Lambert; and



•

we would receive a royalty on these sales in the applicable market.

    Warner-Lambert has chosen a p53-selective therapeutic virus, armed with a prodrug converting enzyme as a candidate for further product testing. An additional armed virus product candidate will be a virus targeting cancer cells with mutations in the p53 gene or loss of p53 function in the p53 gene armed with one of two cytokines. We intend to select the cytokine candidate by mid-2001. The armed virus products provided for in the agreement are currently in the research stage, and the timing of clinical trials will depend on the results of research and preclinical development activities. We have retained the rights to independently develop and commercialize products based on p53-selective armed viruses other than the two selected in the collaboration. In addition, we retain rights to all other products derived from our therapeutic virus technology.

    In addition to providing ongoing research and development support, Warner-Lambert is obligated to make milestone payments based on product development achievements, and, pursuant to a related agreement, we have the right to cause Warner-Lambert to purchase shares of our common stock from us. If all products covered by the agreement are developed successfully, the milestone payments could exceed $100 million. We exercised our right to cause Warner-Lambert to purchase $5 million of our common stock from us in February 2000. We exercised a similar right during the first quarter of 2001 to sell to Warner-Lambert an additional $5 million of our common stock.

    Warner-Lambert has the right to terminate this agreement, or terminate its efforts directed at select product candidates, at any time on 90 days advance notice. In such event, all product rights, or related product rights in the case of termination directed at select product candidates, would revert to us. These product rights would be royalty-free in the case of CI-1042, and would be royalty-bearing to

8

Warner-Lambert in the case of the armed virus product candidates, if such candidates had entered clinical trials prior to termination.

Warner-Lambert: Cell Cycle

    In May 1995, we entered into a research and development collaborative agreement with Warner-Lambert to discover and commercialize small molecule drugs that restore control of or otherwise intervene in the misregulated cell cycle in tumor cells. Under this agreement, we developed screening tests, or assays, for jointly selected targets, and transferred these assays to Warner-Lambert for screening of their compound library to identify active compounds. Warner-Lambert is responsible for subsequent medicinal chemistry and preclinical investigations on the active compounds. We will receive milestone payments on clinical development and registration and royalties on worldwide sales of the products. Warner-Lambert is obligated to conduct and fund all clinical development, make regulatory filings and manufacture for sale the collaboration compounds. We have agreed to extend research under this agreement through August 2001, but we do not expect to extend the research term under this agreement beyond this date. Thereafter, Warner-Lambert may develop products identified during the research term and, if Warner-Lambert receives FDA approval, we could receive milestone payments and royalties on these marketed products.

Warner-Lambert: Inflammation

    In July 1997, we entered into a three-year research and development collaborative agreement with Warner-Lambert to discover and commercialize small molecule drugs for the treatment of acute and chronic inflammatory disorders. The obligations of the parties are similar to those agreed to under the cell cycle program and each party must dedicate a specified minimum number of researchers to the collaboration. We would receive milestone payments based on the development and registration of any resulting products and would receive royalties on worldwide sales. We have agreed to extend research under this agreement through August 2001, but we do not expect to extend the research term under this agreement beyond this date. Thereafter, Warner-Lambert may develop products identified during the research term and, if Warner-Lambert receives FDA approval, we could receive milestone payments and royalties on these marketed products.

Bayer

    Effective February 1994, we established a research and development collaborative agreement with Bayer to discover, develop and market compounds that inhibit the function, or modulate the activity, of the ras signaling pathway or that appropriately modulate the activity of this pathway to treat cancer and other diseases. We and Bayer concluded collaborative research under this agreement in 1999. Based on this research, we and Bayer identified a development candidate, a compound that inhibits ras signaling in cells by inhibiting raf kinase. In our raf kinase inhibitor program, we have initiated with Bayer a Phase I clinical trial in Germany, the United States, Canada and Belgium. We intend to initiate with Bayer a Phase I/II trial in acute myelogenous leukemia in Canada in the first half of 2001 and additional Phase II trials in the second half of 2001.

    Bayer has paid all the costs of research and preclinical development of this drug candidate. Under our agreement with Bayer, we have the opportunity to co-fund 50% of clinical development costs worldwide, excluding Japan. With the initiation of Phase I clinical trials, we are currently co-funding 50% of clinical development costs. Bayer will fund 100% of development costs in Japan and pay us a royalty on sales. If we continue to co-fund and we exercise our right to co-promote in the United States, we would share equally in profits or losses. If we continue to co-fund but do not co-promote in the United States, Bayer would first receive a portion of the product revenues to repay Bayer for its commercialization infrastructure, before determining our share of profits and losses. In other parts of the world except Japan, Bayer would also receive this preferential distribution. If we elect to share in

9

development costs, Bayer would pay us substantial development milestone payments based on the product's progress through clinical trials. These milestone payments would be repayable to Bayer from our share of profits and royalties. At any time during product development, either company may terminate its participation in development costs, in which case the other party would retain exclusive rights to the product on a royalty-bearing basis. If we do not continue to bear 50% of product development costs, Bayer would retain exclusive, worldwide rights to this product candidate and would pay royalties to us on net sales.

    In addition to the development candidate referred to above, Bayer's chemists are optimizing an additional lead compound identified in the course of the collaborative research. This continuing work could result in an additional development candidate.

Chiron

    Our business began in April 1992 by means of the transfer from Chiron to us of the drug discovery program being conducted at Chiron by Dr. Frank McCormick, our scientific founder, and his research team. Under an agreement between Chiron and us, Chiron has an option through April 2007 to receive an exclusive or co-exclusive royalty-bearing license to any diagnostic and vaccine product candidates we may develop. If Chiron does not exercise its option rights with respect to a particular product candidate, then prior to the completion of Phase II clinical trials, we may seek a third party licensor of that product, subject to a right of first refusal in favor of Chiron, and after the completion of Phase II clinical trials for product candidates, the related option rights of Chiron expire.

    This agreement also includes a mechanism for our making proposals to Chiron for future collaborations. In these proposals we would disclose to Chiron the material information known to us regarding the program and propose a set of commercial terms. If a proposal is made, and we and Chiron do not reach agreement within 60 days after we make the proposal, then we may, within 120 days thereafter, enter into an agreement with a third party on terms no more favorable taken as a whole than the terms that we offered to Chiron. Chiron has advised us that it believes the foregoing provision, in the context of the other provisions of this agreement, requires us to offer gene therapy programs to Chiron pursuant to this mechanism before we license any such program to a third party. We do not agree that these provisions impose this obligation to make proposals to Chiron. However, the resulting uncertainty about the interpretation of this agreement may impede our ability to enter into agreements with other companies for gene therapy products in the absence of a waiver by Chiron.

    Chiron has never exercised any right to receive a product license from us. We executed the agreement with Warner-Lambert concerning CI-1042 and two armed viruses pursuant to a waiver letter from Chiron. We understand that Chiron has recently reduced its research activities in the field of gene therapy. However, it is possible that Chiron will, in the future, assert rights under this agreement, which may impede or delay our ability to enter into collaboration agreements with others.

Marketing and Sales

    We currently have no marketing, sales or distribution capabilities, but we intend to build these capabilities to promote our products in the United States and Canada. Consequently, we have retained co-promotion rights in these territories for most of our products. We also expect to exploit relationships with one or more pharmaceutical companies with established marketing, sales and distribution capabilities and direct sales forces to market our products. We may be unable to establish in-house marketing, sales and distribution capabilities or relationships with third parties, and may not be successful in gaining market acceptance for our products.

10

Manufacturing

    At this time, we do not have internal manufacturing capabilities to supply small or large-scale clinical trials or commercial quantities of our therapeutic virus, nor do we have experience in this manufacturing. Warner-Lambert has the exclusive right to manufacture the products that result from our collaborations with them. To manufacture our products for clinical trials or on a commercial scale, if we are required to or choose to do so, we will have to build or gain access to a manufacturing facility, which will require a significant amount of funds. In our collaborations in small molecule drug discovery, Bayer and Warner-Lambert are obligated to manufacture all such drugs for clinical development and commercialization.

    We use a contract manufacturer, BioReliance Corporation, for the production of CI-1042 for use in Phase I and Phase II clinical trials. We have an agreement with Molecular Medicine LLC to manufacture CI-1042 for use in Phase III clinical trials using a small-scale manufacturing process. Although Molecular Medicine has produced viral products for Phase I and Phase II clinical trials, prior to the commencement of our Phase III clinical trial of CI-1042, Molecular Medicine had not produced Phase III clinical trial materials for us or any other parties.

    To date, we have experienced delays in the production and the release of batches of clinical material manufactured by Molecular Medicine. While we have successfully produced and released batches of materials to support clinical trials, including the Phase III clinical trial, Molecular Medicine has had production problems resulting in failed batches. Our current manufacturing process as well as operational errors at Molecular Medicine may have contributed to those failures. In addition, Molecular Medicine has needed extra time to bring its record keeping into compliance with our and Warner-Lambert's requirements. We have initiated efforts to address problems with the current manufacturing process and to improve the time to release supplies of CI-1042 from Molecular Medicine.

    In January 2001, we entered into a process development and manufacturing agreement for CI-1042 with XOMA (US) LLC. The agreement would allow Warner-Lambert and us to access clinical and commercial supplies of CI-1042 from XOMA. The agreement calls for XOMA to develop a large-scale production process for CI-1042, produce clinical supplies, prepare the manufacturing facility for FDA review and provide commercial supplies to us and Warner-Lambert. Together with XOMA, we are likely to need to improve the existing manufacturing process to successfully produce CI-1042 on a large-scale basis. Although XOMA has produced a number of protein products on a large-scale basis, XOMA does not have experience in large-scale viral manufacturing. XOMA may terminate the contract with 48 months' notice for any reason. Further, if Warner-Lambert asserts its right as the sole manufacturer of CI-1042 for commercial supply, we will have to rely on Warner-Lambert for commercial supply of CI-1042 rather than XOMA, and the timing for the availability of the commercial supply will depend on Warner-Lambert's decision to invest in a commercial manufacturing facility.

    We had initially anticipated that our contract manufacturers would produce sufficient quantities of CI-1042 to supply our clinical trials. However, we have become involved in an increased number of clinical trials including the Phase II intravenous clinical trials and the aggregate demand for drug supply currently exceeds our production rate. If we continue to support the increased number of clinical trials and continue to have the production problems that we have experienced to date, we may suffer delays in the timelines for our clinical trials and we may not meet the commercialization timelines in accordance with our plans. Together with XOMA, we are currently developing plans for process scale-up and commercial manufacturing of CI-1042. While we have made substantial progress in the development of a scalable manufacturing process to supply our Phase III clinical trial, we may not successfully establish a commercial-scale manufacturing process and may not establish a commercial facility in a timely manner.

11

    We are aware of only a limited number of manufacturers who we believe would have the ability and capacity to manufacture this product or any other therapeutic viruses we may develop. Failure of any such third-party manufacturer to comply with state and federal regulations and to deliver the required quantities on a timely basis and at commercially reasonable prices would seriously harm our business, financial condition and results of operations. We, acting alone or with a third party, may not make a successful transition to commercial-scale production of our potential products. Even if we are successful, we may not maintain such production.

Patents and Proprietary Rights

    We believe that patent and trade secret protection is crucial to our business and that our future will depend in part on our ability to obtain patents, maintain trade secret protection and operate without infringing the proprietary rights of others, both in the United States and other countries. In October 1997, we were awarded United States Patent No. 5,677,178 for claims covering the use of CI-1042 for the treatment of functionally p53-deficient cancers. In April 1999, we were awarded European Patent No. 689,447, for claims covering the use of CI-1042 for the treatment of functionally p53-deficient cancers. We were also awarded United States Patent No. 5,801,029, a broad methods of use patent, for claims covering the use of adenoviral mutants that kill functionally RB-deficient tumor cells and the corresponding European Patent No. 931830. Each of these patents includes claims covering armed viruses with anticancer genes. Further, we have made additional filings worldwide that claim adenoviruses that can be used to kill functionally deficient p53 or RB cancer cells, with or without a prodrug converting enzyme. We were also awarded United States Patent No. 5,846,945, for claims covering compositions of matter that consist of CI-1042 and a chemotherapeutic. As of December 31, 2000, we owned or had licensed rights to 39 United States patents and 50 United States patent applications, and generally, foreign counterparts of these filings. We have licensed patents and patent applications covering formulations of viruses, prodrug converting enzymes and other technology useful in the conduct of our business.

    Our existing patent rights may not have a deterrent effect on competitors who are conducting or desire to commence competitive research programs with respect to the biological targets or fields of inquiry that we are pursuing. Our ultimate patent position will depend on our ability to obtain effective patent coverage for the compositions of matter identified in these research programs. Because these programs are at an early stage, we cannot determine whether potential products that we may derive from our drug discovery program may be subject to the patent rights of third parties.

    Generally, patent applications in the United States are maintained in secrecy until patents issue and since publication of discoveries in the scientific or patent literature often lag behind actual discoveries, we are not certain that we were the first to make the inventions covered by each of our pending patent applications or that we were the first to file patent applications for such inventions. The patent positions of biotechnology and pharmaceutical companies are highly uncertain and involve complex legal and factual questions. Therefore, we cannot predict the breadth of claims allowed in biotechnology and pharmaceutical patents, or their enforceability. To date there has been no consistent policy regarding the breadth of claims allowed in biotechnology patents. Third parties or competitors with similar technology may challenge or circumvent our patents or patent applications, if issued, and the rights granted thereunder may not provide proprietary protection or competitive advantages to us with respect to these competitors. Furthermore, others may independently develop similar technologies or duplicate technology that we have developed. Because of the extensive time required for development, testing and regulatory review of a potential product, it is possible that before we commercialize any of our products, any related patent may expire, or remain in existence for only a short period following commercialization, thus reducing any advantage of the patent.

12

    We are aware of pending patent applications that others have filed that may pertain to aspects of our programs. If patents are issued to others containing preclusive or conflicting claims and these claims are ultimately determined to be valid, we may be required to obtain licenses to these patents or to develop or obtain alternative technology. Our breach of an existing license or failure to obtain a license to technology required to commercialize our products may seriously harm our business. We also 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 create substantial costs. If our competitors prepare and file patent applications in the United States that claim technology also claimed by us, we may have to participate in interference proceedings declared by the United States Patent and Trademark Office to determine priority of invention, which could result in substantial cost, 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.

    With respect to CI-1042 and selectively-replicating viruses, we are aware of a patent application filed in the United States, Europe, Japan and Canada by General Hospital Corporation, an affiliate of Massachusetts General Hospital. This patent application is related to research involving a modified herpes simplex virus but it also includes broader claims that, if they were to issue, would cover CI-1042 and other selectively-replicating viruses. We believe, and have received an opinion from outside counsel to the effect, that such broad claims made in the General Hospital patent application are not patentable. Consistent with this opinion is a review of the European patent status of the General Hospital patent application, which shows that the patent examiner is requiring that General Hospital limit the claims to herpes viruses. However, General Hospital may receive broad claims in one or more countries, and we may not successfully challenge these claims. We may not obtain a license if this patent were to issue.

    We have identified United States Patent No. 5,837,520 that covers methods of purification of viral vectors. Canji, Inc. either owns or has licensed the rights to this patent. We may seek a license under this patent from Canji. However, if a license were not available at commercially reasonably terms, or at all, we believe that we could develop purification methods that are not covered by the patent.

    We have identified four European Patent Applications, EPA 415, 731; EPA 657, 539; EPA 657, 540; and EPA 690, 129, that claim enzymes for converting prodrugs to their active forms for treating disease, including cancer, and methods of delivering the enzymes using viruses. Glaxo Wellcome either owns or has licensed the rights to these patent applications. The European Applications were filed in 1989, and as yet, have not been granted. We assume that one or more corresponding United States patent applications have been filed, but none have issued yet. The issuance of any of these patent applications in either Europe or the United States will not prevent us from developing and commercializing CI-1042. However, these patent applications, should they issue, may restrict our ability to incorporate prodrug converting enzymes into viruses. Moreover, we may not be successful in challenging these claims, under any of these patents if they were to issue, and we may not obtain a license to the patent.

    In June 1997, ICT Pharmaceuticals, Inc. notified us of two issued United States Patents, Nos. 4,980,281 and 5,266,464, that ICT Pharmaceuticals believes cover the use of a cell for the screening, testing or pharmacological characterization of new drugs or other substances. Foreign counterparts of the United States patents are pending. ICT Pharmaceuticals has offered us a license to the patents. We have not determined whether to negotiate a license.

    Genetic Therapeutic Inc., owns or has licensed the rights of United States patent No. 5,998,205. This patent covers adenoviruses that replicate in specific tissue types in which replication is controlled by genetic elements whose expression is unique to those tissues. This patent does not cover CI-1042, but may cover a subset of our RB selective viruses whose replication is controlled by a genetic element

13

expressed in nearly all tumor cells regardless of the tissue of origin. If this patent were asserted against us, we may not successfully challenge these claims and we may not obtain a license.

    Introgen, Inc. owns or has licensed the rights of United States Patent No. 6,194,191. This patent covers methods for producing/purifying adenoviruses. Introgen has indicated in its public statements that it is willing to license the patent. However, if a license were not available at commercially reasonably terms, or at all, we believe that we could develop purification methods that are not covered by the patent.

    Together with our licensors, we also rely on trade secrets to protect our combined technology, especially where we do not believe patent protection is appropriate or obtainable. However, trade secrets are difficult to protect. We protect our proprietary technology and processes, in part, by confidentiality agreements with our employees, consultants and collaborators. These parties may breach the agreement, and we may not have adequate remedies for any breach. Our trade secrets may otherwise become known or be independently discovered by competitors. To the extent that we or our consultants or research collaborators use intellectual property owned by others in their work for us, we may have disputes with them or other third-parties as to the rights in related or resulting know-how and inventions.

Government Regulation

    Regulation by government authorities in the United States and other countries will be a significant factor in the manufacturing and marketing of any products that may be discovered or developed by us, or that may arise out of our research. We must obtain the requisite regulatory approval by government agencies prior to commercialization of our products. We anticipate that our products will be subject to rigorous preclinical and clinical testing and premarket approval procedures by the FDA and similar health authorities in foreign countries. Various federal statutes and regulations also govern or influence the manufacturing, testing, labeling, storage, recordkeeping and marketing and promotion of such products.

    The steps ordinarily required before a drug or biological product may be marketed in the United States include:

•

preclinical studies;



•

the submission to the FDA of an IND that must become effective before human clinical trials may commence;



•

adequate and well-controlled human clinical trials to establish the safety and efficacy of the drug or biologic;



•

the submission of a marketing application to the FDA; and



•

FDA approval of the marketing application, including inspection and approval of the product manufacturing facility.

    Preclinical trials involve laboratory evaluation of product chemistry, formulation and stability, as well as animal studies to assess the potential safety and efficacy of each product. Preclinical safety trials must be conducted by laboratories that comply with FDA regulations regarding Good Laboratory Practice. The results of the preclinical trials are submitted to the FDA as part of an IND and are reviewed by the FDA before the commencement of clinical trials. Unless the FDA objects to an IND, the IND will become effective 30 days following its receipt by the FDA. Submission of an IND may not result in FDA clearance to commence clinical trials, and the FDA's failure to object to an IND does not guarantee FDA approval of a marketing application.

14

    Clinical trials involve the administration of the investigational product to humans under the supervision of a qualified principal investigator. In the United States, clinical trials must be conducted in accordance with Good Clinical Practices under protocols submitted to the FDA as part of the IND. In addition, each clinical trial must be approved and conducted under the auspices of an Institutional Review Board, or IRB, and with the patient's informed consent. The IRB will consider, among other things, ethical factors, the safety of human subjects and the possible liability of the institution conducting the clinical trial. The United Kingdom and certain other European and Asian countries have similar regulations.

    The goal of Phase I clinical trials is to establish initial data about safety and tolerance of the investigational product in humans. The goal of Phase II clinical trials is to provide evidence about the desired therapeutic efficacy of the investigational product in limited studies with small numbers of carefully selected subjects. The investigators seek to evaluate the effects of various dosages and to establish an optimal dosage level and dosage schedule. Investigators also gather additional safety data from these studies. The Phase III clinical trial consists of expanded, large-scale, multicenter studies in the target patient population. The goal of these studies is to obtain definitive statistical evidence of the efficacy and safety of the proposed product and dosage regimen.

    We would need to submit all data obtained from this comprehensive development program as a marketing application to the FDA, and to the corresponding agencies in other countries for review and approval, before marketing products. The FDA may elect to present data on our products to one of its advisory committees for review and recommendation before it grants approval. Essentially all of our proposed products will be subject to demanding and time-consuming approval procedures in the countries where we intend to commercialize our products. These regulations define not only the form and content of the development of safety and efficacy data regarding the proposed product, but also impose specific requirements regarding:

•

manufacture of the product;



•

testing;



•

quality assurance;



•

packaging;



•

storage;



•

documentation;



•

recordkeeping;



•

labeling;



•

advertising; and



•

marketing procedures.

    Effective commercialization also requires inclusion of our products in national, state, provincial, or institutional formularies or cost reimbursement systems.

    The FDA must approve our products, including the manufacturing processes and facilities used to produce these products, before the products may be marketed in the United States. The process of obtaining FDA approval can be costly, time consuming and subject to unanticipated delays. The FDA may refuse to approve an application if it believes that applicable regulatory criteria are not satisfied. The FDA may also require additional testing for safety and efficacy of the drug. Moreover, if regulatory approval of a drug product is granted, the approval will be limited to specific indications. Approvals of our proposed products, processes or facilities may not be granted on a timely basis, if at all. Any failure to obtain, or delay in obtaining, such approvals would seriously harm our business,

15

financial condition and results of operations. Moreover, even if regulatory approval is granted, such approval may include significant limitations on indicated uses for which a product could be marketed. In some instances, regulatory approval may be granted with the condition that confirmatory Phase IV clinical trials are carried out. If these Phase IV clinical trials do not confirm the results of previous studies, regulatory approval for marketing may be withdrawn. Failure to comply with FDA and other applicable regulatory requirements may result in, among other things:

•

warning letters;



•

civil penalties;



•

criminal prosecution;



•

injunctions;



•

seizure or recall of products;



•

total or partial suspension of production;



•

refusal of the government to grant approval; or



•

withdrawal of approval of our products.

    In addition to regulations enforced by the FDA, we are subject to regulation under:

•

the Occupational Safety and Health Act;



•

the National Environmental Policy Act;



•

the Toxic Substances Control Act;



•

the Resource Conservation and Recovery Act; and



•

other present and potential future federal, state or local regulations.

    Our potential products may require review by the Recombinant DNA Advisory Committee. In other countries, similar regulations may apply. Our research and development involves the controlled use of hazardous materials and chemicals. Although we believe that our safety procedures for handling and disposing of such materials comply with the standards prescribed by state and federal regulations, we cannot completely eliminate the risk of accidental contamination or injury from these materials. In the event of such an accident, we could be held liable for any damages that result and any such liability could exceed our resources.

    Whether or not we obtain FDA approval, approval of a product by comparable regulatory authorities will be necessary in foreign countries prior to the commencement of marketing of the product in such countries. The approval procedure varies among countries and can involve additional testing. The time required to obtain approval may differ from that required for FDA approval. Although there is now a centralized European Union approval mechanism in place, each European country may nonetheless impose its own procedures and requirements, many of which are time consuming and expensive. Thus, there can be substantial delays in obtaining required approvals from both the FDA and foreign regulatory authorities after the relevant applications are filed. We expect to rely on our collaborators and licensees, along with our own expertise, to obtain governmental approval in foreign countries of drug and biological products discovered by us or arising from our programs.

Competition

    We are engaged in a rapidly changing and highly competitive field. Other products and therapies that currently exist or are being developed will compete with the products we are seeking to develop and market. Some of these competitive products are in clinical trials. In particular, among other trials,

16

Schering-Plough Corporation is conducting a Phase II clinical trial of its p53 gene therapy product in liver metastases of colorectal cancer. Aventis Inc./Introgen Therapeutics, Inc. have initiated a Phase III clinical trial in head and neck cancer with their p53 gene therapy product. ImClone Systems, Inc. is conducting Phase II/III clinical trials of a monoclonal antibody targeting cancer cells with a positive epidermal growth factor receptor in patients with solid tumors including head and neck cancer and liver metastases of colorectal cancer. Matrix Pharmaceuticals, Inc. has completed Phase III clinical trials of Intradose, a gel containing cisplatin and epinephrine in patients with head and neck cancer and a Phase II trial in patients with liver metastases of colorectal cancer. OSI Pharmaceuticals is conducting Phase II/III clinical trials of a small molecule compound targeting cancer cells with positive epidermal growth factor receptors for treatment. Targeted Genetics Corp. is initiating a Phase II clinical trial of a tumor inhibiting gene therapy in combination with radiation therapy for the treatment of head and neck cancer. If approved, the products of these and other competitors now in clinical trials will compete directly with CI-1042. Other companies are developing small molecule drugs that may compete with product candidates identified in our small molecule programs including companies that are developing ras pathway inhibitor.

    Competition from fully integrated pharmaceutical companies and more established biotechnology companies is intense and is expected to increase. Substantially all of these companies have significantly greater financial resources and expertise in research and development, manufacturing, preclinical and clinical testing, obtaining regulatory approvals, and marketing than us. Smaller companies may also compete with us, particularly through collaborations with large pharmaceutical and established biotechnology companies. Many of these competitors have significant products that have been approved or are in development and operate large, well-funded research and development programs. Academic institutions, governmental agencies and other public and private research organizations also conduct research, seek patent protection and seek marketing and research and development collaborations that compete with our programs. These companies and institutions also compete with us in recruiting and retaining highly qualified scientific and management personnel. In addition to the above factors, we will face competition based on:

•

product efficacy and safety;



•

the timing and scope of regulatory approvals;



•

availability of supply;



•

marketing and sales capability;



•

reimbursement coverage;



•

price; and



•

patent position.

Employees

    As of December 31, 2000, we had 135 full-time employees of whom 25 hold Ph.D. or M.D. degrees. Of our employees, 104 are in research and development and 31 are in corporate development, finance and administration. No employee of ours is represented by a labor union and we consider our employee relations to be good.

17

Scientific Advisory Board

    Our Scientific Advisory Board, or SAB, consists of individuals with expertise in many aspects of molecular oncology that advise us and provide critical review of our various development activities. The SAB meets several times a year. In addition, the SAB members consult with and meet informally with us on a frequent basis. Certain SAB members own shares of our common stock and each is compensated for his services. Every member of the SAB has entered into a consulting agreement with us covering the terms of their positions as our consultants and as members of the SAB. The members of our SAB are as follows:

ADDITIONAL BUSINESS RISKS

    In addition to the risks discussed in "Business" and "Management's Discussion and Analysis of Financial Condition and Results of Operations," our business is subject to the business risks set forth below.

If we are not able to demonstrate the effectiveness of CI-1042 in our clinical trials or if our clinical trials are delayed, we may be unable to commercialize CI-1042.

    We have completed Phase II clinical trials designed to obtain safety and effectiveness trend information for CI-1042 for the treatment of head and neck cancer, both as a single agent and in combination with chemotherapy. Based on data from these Phase II clinical trials, we, together with Warner-Lambert, have initiated a Phase III clinical trial designed to obtain effectiveness information for CI-1042 for the treatment of recurrent disease. Historically, many companies have failed to demonstrate the effectiveness of pharmaceutical products in Phase III clinical trials notwithstanding favorable results in Phase II clinical trials. We may fail to demonstrate desired effectiveness levels in our Phase III clinical trial of CI-1042. In addition, we may observe previously unforeseen side effects. We may fail to extend the findings of previous clinical trials in our Phase III clinical trial of CI-1042, including similar tumor response rates, duration of tumor response or safety.

    The process of obtaining FDA and other required regulatory approvals, including foreign approvals, often takes many years and can vary substantially based upon the type, complexity and novelty of the products involved. We have had only limited experience in filing and pursuing applications necessary to gain regulatory approvals. The FDA may not accept the results of our Phase III clinical trial, or accept as sufficient for market approval other elements of the application that we may file for CI-1042. The FDA may require changes in our current trials or may require additional clinical trials which may be extensive, expensive and time-consuming. We cannot market CI-1042 unless we receive regulatory approval.

    In addition, in our clinical trials we treat patients who have failed conventional treatments and who are in advanced stages of cancer. During the course of treatment, these patients may die or suffer adverse medical effects for reasons that may be unrelated to CI-1042. These adverse effects may impact the interpretation of clinical trial results.

18

We may fail to demonstrate that CI-1042 is effective for the treatment of other types of cancer even if CI-1042 is proven effective for the treatment of head and neck cancer.

    We are initially developing CI-1042 for treatment of head and neck cancer, using intratumoral injection. Even if we are successful in developing CI-1042 for this type of cancer, we may not demonstrate that CI-1042 is effective in the treatment of a broader array of cancer types. We have conducted a Phase I/II clinical trial for treatment of liver metastases of colorectal cancer with CI-1042 administered through intrahepatic artery infusion. In addition, we are in the process of completing Phase I/II clinical trials of CI-1042 for treatment of pancreatic cancer. The Phase I/II clinical trial in liver metastases of colorectal cancer is based on a small number of patients, and we may not reproduce the results from these clinical trials in future clinical trials with additional patients. In addition, we may not succeed in our efforts to deliver CI-1042 to tumors through routes other than intratumoral injection. If we are not successful in developing additional routes of administration for CI-1042, or establishing its effectiveness in a broad range of cancer types, CI-1042 may not have a broad commercial use.

We do not have manufacturing expertise or capabilities and are dependent on third parties to fulfill our manufacturing needs, which could result in the delay of clinical trials or regulatory approval.

    We lack the resources, experience and capabilities to manufacture our products on our own. We would require substantial funds to establish these capabilities. Consequently, we are dependent on third parties, including collaborative parties and contract manufacturers, to manufacture our products and product candidates. These parties may encounter difficulties in production scale-up, including problems involving production yields, quality control and quality assurance and shortage of qualified personnel. These third parties may not perform as agreed or may not continue to manufacture our products for the time required by us to successfully market our products. These third parties may fail to deliver the required quantities of our products or product candidates on a timely basis and at commercially reasonable prices. Failure by these third parties could delay our clinical trials and our applications for regulatory approval. If these third parties do not adequately perform, we may be forced to incur additional expenses to pay for the manufacture of our products or to develop our own manufacturing capabilities.

We currently rely on a sole source or limited number of sources for the manufacturing of CI-1042, and if these sources are unable or unwilling to deliver the required quantities, we may not be able to find replacement manufacturers, which could result in a delay in clinical trials or in regulatory approval.

    We currently rely on a sole source contract manufacturer for the supply of CI-1042 for Phase III clinical trials. We are aware of only a limited number of manufacturers who we believe would have the ability and capacity to manufacture this product or any other therapeutic viruses we may develop. Our contract manufacturer has produced multiple batches of material for our Phase III clinical trials, however, our contract manufacturer has experienced production problems resulting in failed batches. We may need to modify the current process to minimize batch failures. In addition, we and Warner-Lambert have held up final release of some of the batches until the manufacturer completes documentation that complies with Warner-Lambert's and our manufacturing standards. As a result, we have limited the number of clinical trial sites that are enrolling patients including sites in the Phase III clinical trial. Ten clinical sites for the Phase III trial have been open for enrollment, however, to date only two sites have been supplied with CI-1042. We will continue to limit the number of clinical sites that are enrolling patients for the Phase III clinical trial until product release is occurring on a routine basis.

    If our current sole source contract manufacturer is unable or unwilling to deliver the required quantities of CI-1042 on a regular basis or terminates our relationship, we will have to rely on XOMA to manufacture CI-1042. We do not expect XOMA to begin supplying us with CI-1042 before the first

19

quarter of 2002. This could delay our clinical trials and our applications for regulatory approval with the FDA. If XOMA fails to supply us with sufficient materials, we may be forced to incur additional expenses to pay for the manufacture of materials using a replacement contract manufacturer, if we can find a replacement manufacturer, or to develop our own manufacturing capabilities which may not occur within a reasonable amount of time or at commercially reasonable rates.

No one has manufactured replicating human viruses on a large scale; if we are unable to develop an effective process to manufacture CI-1042 on a large scale, our clinical trials and regulatory approval would be delayed.

    To date, our contract manufacturers have produced CI-1042 using small scale processes. No one has produced replicating human viruses using a commercial-scale manufacturing process. We have developed a process that we believe will be scaleable; however, we and our contract manufacturer for Phase III clinical trials have experienced production problems using this process that have resulted in failed batches. We may need to make additional process changes and operational changes to make the manufacturing process more reliable and to make the process easier for us to develop into a larger commercial-scale manufacturing process. If we are unable to improve the large scale manufacturing process, we may not increase the number of clinical trials for CI-1042 and the number of sites enrolling patients. If we are unable to expand the number of clinical trials, clinical sites enrolling patients and patients receiving CI-1042, the results from our clinical trials, in particular our Phase III clinical trial, will be delayed and we will not receive regulatory approval without the results from these trials.

XOMA or Warner-Lambert may not be able to produce commercial quantities of CI-1042, which could delay regulatory approval.

    To obtain regulatory approval for CI-1042, we will need to treat a percentage of the patients in our Phase III clinical trial using CI-1042 produced from the same manufacturing process and in the same manufacturing facility that we intend to use following FDA approval. We have recently executed a process development and manufacturing agreement with XOMA to modify the process and to produce large quantities of CI-1042. Under our collaboration agreement with Warner-Lambert, Warner-Lambert has the sole right and responsibility to manufacture or have manufactured all supplies of CI-1042 for all commercial sales. Warner-Lambert has delayed investment in the manufacturing capacity needed to supply clinical trials and the potential commercial supply of CI-1042.

    In conjunction with Warner-Lambert or XOMA, we will need to modify the manufacturing process to produce large quantities of CI-1042 and to lower the cost by improving the efficiency of the process. To modify the manufacturing process and to meet our quality standards for CI-1042, in conjunction with Warner-Lambert or XOMA, we will need to spend a significant amount of time and capital and complete a substantial amount of experimentation. In conjunction with Warner-Lambert or XOMA, we will need to expand and modify existing manufacturing facilities to produce commercial quantities of CI-1042. XOMA does not have experience in large-scale viral manufacturing.

    In addition, XOMA may terminate our process development and manufacturing agreement with us for any reason by providing notice 48 months in advance. Further, if Warner-Lambert asserts its right as the sole manufacturer of CI-1042 for commercial supply, we will have to rely on Warner-Lambert for commercial supply of CI-1042 rather than XOMA and the timing for the availability of the commercial supply will depend on Warner-Lambert's decision to invest in a commercial manufacturing facility.

    If we do not treat patients in our Phase III clinical trial with product from the process manufactured at the new facility, the FDA will most likely require a bridging study to show that the CI-1042 produced from the new process is comparable to CI-1042 produced from our existing manufacturing process at our contract manufacturer's existing facility. Filing of our applications for regulatory approval may be delayed if we:

•

encounter difficulties in modifying the manufacturing process;

20

•

fail to treat patients in our Phase III clinical trial with product from the new manufacturing process; or



•

fail to conduct a bridging study prior to FDA review of our Phase III clinical trial.

We are dependent upon collaborative relationships to develop, manufacture and commercialize our products and to obtain regulatory approval, which could delay the development and commercialization of our products.

    Our strategy for developing, manufacturing and commercializing our products and obtaining regulatory approval depends in large part upon maintaining and entering into collaborative agreements with pharmaceutical companies or other collaborators. We have entered into a number of collaborative agreements with different parties, including research, development and marketing agreements with Warner-Lambert and Bayer. Our agreements with Warner-Lambert also give Warner-Lambert the responsibility of providing commercial manufacturing of our potential products. If we fail to maintain these collaborative relationships or to establish new collaborative relationships, we would need to undertake these research, development, manufacturing and marketing activities at our own expense, which would significantly increase our capital requirements and limit the programs we are able to pursue. Further, we would incur significant delays with the development, manufacture or sale of our products.

    We are subject to a number of risks associated with our dependence upon collaborative relationships, including:

•

the amount and timing of expenditure of resources can vary because of collaborator decisions;



•

business combinations and changes in a collaborator's business strategy may adversely affect the party's willingness or ability to complete its obligations under the collaboration agreement with us;



•

the right of the collaborator to terminate its collaboration agreement with us on limited notice and for reasons outside our control;



•

loss of significant rights to our collaborative parties if we fail to meet our obligations under these agreements;



•

disagreements as to ownership of clinical trial results or regulatory approvals, and the refusal of the FDA to recognize us as holding the regulatory approvals necessary to commercialize our products;



•

withdrawal of support by a collaborator following the development or acquisition by the collaborator of competing products; and



•

disagreements with a collaborator regarding the collaboration agreement or ownership of proprietary rights.

    Due to these factors and other possible disagreements with collaborators, we could suffer delays in the research, development or commercialization of our products or we may become involved in litigation or arbitration, which would be time consuming and expensive.

Pfizer has acquired Warner-Lambert, which could result in the modification, disruption or termination of our collaborative relationship with Warner-Lambert.

    In June, 2000, Pfizer acquired Warner-Lambert. As the parent of Warner-Lambert, Pfizer may not be interested in continuing the multiple research and development and marketing collaboration agreements we have with Warner-Lambert, including the collaboration agreement related to CI-1042, in part because these collaborations may address smaller markets than Pfizer generally seeks to address. Warner-Lambert is focusing on commercializing CI-1042 for indications where CI-1042 is administered

21

intravenously to treat tumors systemically over indications where CI-1042 is administered locally or regionally, such as intratumoral injections or intrahepatic artery infusions. As a result, Warner-Lambert has initiated Phase II clinical trials where CI-1042 is administered intravenously, and these trials have placed increased demands on the available supply of CI-1042 competing for supplies that were intended for the Phase III clinical trial for head and neck cancer. To date, we have allocated the limited CI-1042 supply produced from our sole source supplier to support the Phase II intravenous clinical trials and the pivotal Phase III clinical trial, thereby impeding the progress of all of these clinical trials and limiting the number of patients enrolled in these trials. If the Phase II intravenous clinical trials do not generate sufficient evidence of antitumor activity or Warner-Lambert is unable to conduct intravenous clinical trials because of a limited supply of CI-1042, Warner-Lambert may discontinue our collaboration regarding CI-1042 and other products pursuant to the terms of these agreements.

    Under our collaboration agreement with Warner-Lambert, Warner-Lambert has the sole right and responsibility to manufacture or have manufactured all supplies of CI-1042 for all commercial sales. Warner-Lambert has delayed investment in the manufacturing capacity needed to supply clinical trials and the potential commercial supply of CI-1042. To avoid delays in the development of CI-1042 for this indication, we have entered into an agreement with XOMA to supply CI-1042 for potential commercial use.

    We are currently discussing with Warner-Lambert amending the collaboration agreement to include using XOMA as a commercial manufacturer of CI-1042. If we are unsuccessful in amending the collaboration agreement, we may have to bear the full manufacturing costs from XOMA, including any up-front payments. Further, if Warner-Lambert asserts its right as the sole manufacturer of CI-1042 for commercial supply, we may have to rely on Warner-Lambert for commercial supply of CI-1042 rather than XOMA and the timing for the availability of the commercial supply will depend on Warner-Lambert's decision to invest in a commercial manufacturing facility. Depending on the timing and the nature of the investment, we may experience a delay in completing our Phase III clinical trials and submitting applications to the FDA for regulatory approval.

    As the parent of Warner-Lambert, Pfizer could cause Warner-Lambert to modify, disrupt or terminate the collaborative agreements we originally entered into with Warner-Lambert, subject to the terms of these agreements. Under the agreement relating to CI-1042, Warner-Lambert has the right to terminate the agreement for any reason with 90 days notice, in which case Warner-Lambert is required to return all rights to CI-1042 to us royalty-free.

    If Warner-Lambert terminates our agreement relating to CI-1042, we would lose a significant portion of the $40 million of funding for the development of CI-1042, including the cost of clinical trials, which Warner-Lambert had agreed to provide under the agreement. In addition, we would not have access to Warner-Lambert's sales and marketing capabilities and expertise to commercialize CI-1042 as provided in the agreement. We currently intend to rely on the sales and marketing services provided to us under the agreement. Further, if Warner-Lambert terminates our agreement relating to CI-1042 and does not agree to manufacture CI-1042 for commercial use, we would rely solely on XOMA to provide commercial supply of CI-1042. XOMA has not manufactured large-scale viral products in the past and, together with us, may not successfully establish a commercially feasible manufacturing process.

Chiron may have preferential rights to establish collaborations with us, which may complicate our future collaborative arrangements.

    We were established in April 1992 by means of a transfer from Chiron Corporation to us of the drug discovery program conducted at Chiron by Dr. Frank McCormick, our scientific founder, and his research team. Under the agreement executed at that time, we granted Chiron preferential rights to receive product licenses in the fields of diagnostics and vaccines, and also established a mechanism for

22

our making proposals to Chiron for future collaborations. Chiron has advised us that it believes this mechanism requires us to offer gene therapy programs to Chiron before licensing any of these programs to a third party. We and Chiron have different interpretations of this agreement as it relates to the scope of Chiron's rights. We executed our agreement with Warner-Lambert for the development of CI-1042 and two other virus products pursuant to a waiver letter from Chiron. If Chiron does not grant us further waivers and asserts rights under the April 1992 agreement, or if disputes arise, we may encounter difficulties or delays in entering into future collaborations for other product candidates.

We do not fully understand the biological characteristics of our therapeutic viruses, and their interactions with other drugs and the human immune and other defense systems, which may cause us to fail to demonstrate the safety and effectiveness of our products in clinical trials.

    Therapeutic viruses are novel and we are still determining the biological characteristics of these viruses. For example, in our clinical trials to date, we have achieved the best results when CI-1042 is used in combination with standard chemotherapy drugs, but we are uncertain as to the reasons for and the nature of the interaction of the virus with these drugs. In addition, we are still investigating the response of the human immune system to our therapeutic viruses, and the immune system may play a role in limiting the tumor-killing effect of our therapeutic viruses. We also do not know the extent the human body may clear our therapeutic viruses from circulation in the bloodstream and limit the tumor-killing activity of our therapeutic viruses. Further, we are uncertain as to whether the killing activity of CI-1042 is specific to cells having the abnormal function involving the p53 gene. Moreover, we do not understand all of the many factors that contribute to the formation of each individual patient's cancer. These factors include not only the cancer type, but also the pressures within the tumor and the presence of normal cells and fibrous tissue within the tumor. These factors make each tumor unique. Because of the variety of factors, some cancer patients respond to a particular type of cancer therapy while others do not, even among patients with the same cancer type. The novelty and scientific uncertainties regarding our therapeutic viruses and the uniqueness of human cancers from patient-to-patient increase the risk that we will not successfully develop our product candidates or prove their safety and effectiveness in clinical trials. Even if we succeed in developing our product candidates, our product candidates may not have a therapeutic effect in a broad patient population.

Even if our products are approved, the market may not accept our products.

    Even if our product development efforts are successful and even if the requisite regulatory approvals are obtained, our products may not gain market acceptance among physicians, patients, healthcare payers and the medical community. A number of additional factors may limit the market acceptance of products including the following:

•

rate of adoption by healthcare practitioners;



•

types of cancer for which the product is approved;



•

rate of the products' acceptance by the target population;



•

timing of market entry relative to competitive products;



•

availability of alternative therapies;



•

price of our product relative to alternative therapies;



•

availability of third-party reimbursement;



•

extent of marketing efforts by us and third-party distributors or agents retained by us; and



•

side effects or unfavorable publicity concerning our products or similar products.

23

    If any of our products do not achieve market acceptance, we may lose our investment in that product which may cause our stock price to decline.

We do not have marketing or sales experience or capabilities and are dependent on the efforts of others, which could limit our ability to commercialize our products.

    We intend to enter into agreements with third parties to market and sell most of our products if we receive regulatory approval for a product. We may not be able to enter into marketing and sales agreements with others on acceptable terms, if at all. To the extent that we enter into marketing and sales agreements with other companies, our revenues, if any, will depend on the efforts of others. We also have the right under our collaboration agreements to co-promote our products in conjunction with our collaborators. If we are unable to enter into third-party agreements or if we are exercising our rights to co-promote a product, then we will be required to develop marketing and sales capabilities. We may not successfully establish marketing and sales capabilities or have sufficient resources to do so. If we do not develop marketing and sales capabilities, we may not meet our co-promotion obligations under our collaboration agreements, which could result in our losing these co-promotion rights. If we do develop such capabilities, we will compete with other companies that have experienced and well-funded marketing and sales operations and we will incur additional expenses.

Adverse events in the field of viral gene therapy may negatively affect regulatory approval or public perception of our products, which could delay our clinical trials.

    We depend in part on public acceptance of the use of viruses as therapeutics or as delivery vehicles for gene therapy for the prevention or treatment of human diseases. Public attitudes may be influenced by claims that these therapies are unsafe, and these therapies may not gain acceptance by the public or the media. As a result of negative public reaction to these therapies, the FDA may impose greater regulation, stricter clinical trial oversight and stricter commercial product labeling requirements.

    The media has widely publicized the recent death of a patient at the University of Pennsylvania undergoing viral gene therapy. As a result of this death, the United States Senate held hearings to determine whether additional legislation is required to protect volunteers and patients who participate in gene therapy clinical trials. The Recombinant DNA Advisory Committee, which acts as an advisory body to the National Institutes of Health, has extensively discussed gene therapy clinical trials. This death and any other adverse events in the field of gene therapy that may occur in the future may result in greater governmental regulation of our product candidates and potential regulatory delays relating to the testing or approval of our product candidates.

We have a history of losses and we expect to continue to incur losses.

    As of December 31, 2000, we had an accumulated deficit of approximately $85.6 million. We have incurred these losses principally from costs incurred in our research and development programs and from our general and administrative costs. We have derived no significant revenues from product sales or royalties. We expect to incur significant and increasing operating losses over the next several years as we expand our research and development efforts, preclinical testing and clinical trial and manufacturing activities, including the 2001 manufacturing contract with XOMA. We expect that the amount of operating losses will fluctuate significantly from quarter to quarter as a result of increases or decreases in our research and development efforts, the establishment or termination of collaborations, the timing and amount of collaboration payments under the terms of our collaborative agreements, or the initiation, success or failure of clinical trials.

    We do not expect to generate revenues from the sale of products for the foreseeable future. We expect that substantially all of our revenues for the foreseeable future will result from payments under

24

our collaborative agreements. Our ability to achieve profitability depends upon our success in completing development of our potential products, obtaining required regulatory approvals and manufacturing and marketing our products.

We will need substantial additional funds, and our future access to capital is uncertain.

    We will require substantial additional funds to conduct the costly and time-consuming research, preclinical testing and clinical trials necessary to develop our technology and proposed products, and to establish or maintain relationships with collaborative parties. Our future capital requirements will depend upon a number of factors, including:

•

continued scientific progress in the research and development of our technology programs;



•

the size and complexity of these programs;



•

our ability to establish and maintain collaboration agreements;



•

our ability to manufacture sufficient drug supply for complete clinical trials;



•

progress with preclinical testing and clinical trials;



•

the time and costs involved in obtaining regulatory approvals;



•

the cost involved in preparing, filing, prosecuting, maintaining and enforcing patent claims;



•

competing technological and market developments; and



•

product commercialization activities.

    We may not be able to raise additional financing on favorable terms, or at all. If we are unable to obtain additional funds, we may delay or terminate clinical trials, curtail operations or obtain funds through collaborative and licensing arrangements that may require us to relinquish commercial rights or potential markets or grant licenses that are unfavorable to us.

If we lose our key employees and consultants or are unable to attract or retain qualified personnel, our business could suffer.

    The loss of the services of one or more of our key employees could have an adverse impact on our business. We do not maintain key person life insurance on any of our officers, employees or consultants, other than for our chief executive officer. We depend on our continued ability to attract, retain and motivate highly qualified management and scientific personnel. We face competition for qualified individuals from numerous pharmaceutical and biotechnology companies, universities, and other research institutions. Because of the scientific nature of our business, we are highly dependent on principal members of our scientific and management staff. To pursue our product development plans, we will need to hire additional management personnel and additional qualified scientific personnel to perform research and development, as well as personnel with expertise in clinical testing, government regulation and manufacturing. We may not be successful in hiring or retaining qualified personnel.

We face intense competition and rapid technological change, and many of our competitors have substantially greater managerial resources than we have.

    We are engaged in a rapidly changing and highly competitive field. We are seeking to develop and market products that will compete with other products and therapies that currently exist or are being developed. Many other companies are actively seeking to develop products that have disease targets similar to those we are pursuing. Some of these competitive products are in clinical trials. If approved, the products of these and other competitors now in clinical trials will compete directly with CI-1042.

25

Other companies are devel