AMENDMENT NO. 1 TO FORM S-3

Approximate date of commencement of proposed sale to the public: As soon as possible after this Registration Statement becomes effective.

If the only securities being registered on this Form are being offered pursuant to dividend or interest reinvestment plans, please check the following box. o

If any of the securities being registered on this Form are to be offered on a delayed or continuous basis pursuant to Rule 415 under the Securities Act of 1933, other than securities offered only in connection with dividend or interest reinvestment plans, check the following box. o

If this Form is filed to register additional securities for an offering pursuant to Rule 462(b) under the Securities Act, please check the following box and list the Securities Act registration statement number of the earlier effective registration statement for the same offering. o

If this Form is a post-effective amendment filed pursuant to Rule 462(c) under the Securities Act, check the following box and list the Securities Act registration statement number of the earlier effective registration statement for the same offering. o

If delivery of the prospectus is expected to be made pursuant to Rule 434, please check the following box. o

The Registrant hereby amends this Registration Statement on such date or dates as may be necessary to delay its effective date until the Registrant shall file a further amendment which specifically states that this Registration Statement shall thereafter become effective in accordance with Section 8(a) of the Securities Act of 1933 or until the Registration Statement shall become effective on such date as the Commission, acting pursuant to said Section 8(a), may determine.

The information in this prospectus is not complete and may be changed. We may not sell these securities until the registration statement filed with the Securities and Exchange Commission is effective. This prospectus is not an offer to sell these securities and it is not soliciting an offer to buy these securities in any state where the offer or sale is not permitted.

The shares are quoted on the Nasdaq National Market under the symbol “OSIP.” On November 3, 2004, the last reported sale price of the shares on the Nasdaq National Market was $64.345 per share.

Investing in the common stock involves risks that are described in the “Risk Factors” section beginning on page 9 of this prospectus.

The underwriters may also purchase up to 825,000 additional shares from OSI Pharmaceuticals, Inc. at the public offering price, less the underwriting discount, within 30 days from the date of this prospectus to cover overallotments.

Neither the Securities and Exchange Commission nor any state securities commission has approved or disapproved of the shares being offered by this prospectus, or determined if this prospectus is truthful or complete. Any representation to the contrary is a criminal offense.

In this prospectus, “OSI,” “the Company,” “we,” “us,” “the issuer,” “the registrant,” and “our” refer to OSI Pharmaceuticals, Inc. and its consolidated subsidiaries. You should rely only on the information contained or incorporated by reference in this prospectus. We have not, and the underwriters have not, authorized any other person to provide you with different information. If anyone provides you with different or inconsistent information, you should not rely on it. We are not, and the underwriters are not, making an offer to sell these securities in any jurisdiction where the offer or sale is not permitted. You should assume that the information appearing in this prospectus is accurate only as of the date on the front cover of this prospectus. Our business, financial condition, results of operations, and prospects may have changed since that date.

The following is a summary of our business. This summary does not contain all of the information that you should consider before deciding to invest in our common stock. We urge you to read this entire prospectus carefully, including the “Risk Factors” section of this prospectus and the documents identified in the “Incorporation of Certain Documents by Reference” section of this prospectus.

We are a leading biotechnology company primarily focused on the discovery, development and commercialization of high quality oncology products that both extend life and improve the quality-of-life for cancer patients worldwide. Our flagship product, Tarceva^TM (erlotinib), is an oral, once-a-day, small molecule inhibitor of the epidermal growth factor receptor (HER1/ EGFR). Tarceva^TM is the first EGFR inhibitor, and the first non-chemotherapy agent, to demonstrate a survival benefit in advanced non-small cell lung cancer (NSCLC) and also in pancreatic cancer, two forms of cancer widely recognized as amongst the toughest treatment challenges facing oncologists. On July 30, 2004, we completed our new drug application (NDA) filing with the U.S. Food and Drug Administration (FDA) for monotherapy Tarceva^TM use in the treatment of all NSCLC patients who have failed at least one prior chemotherapy regimen. Based upon the FDA’s prior designation of Fast Track status and priority review status for Tarceva^TM, the FDA has until January 30, 2005 (otherwise known as the PDUFA date) to take action on the NDA filing. We believe that the FDA is rapidly proceeding with its review of the NDA, and we anticipate receipt of approval from the FDA for the NSCLC indication in advance of the PDUFA date. We intend to file a supplemental NDA (sNDA) for Tarceva^TM for the treatment of front-line pancreatic cancer patients (in combination with gemcitabine) in 2005. In addition, our partner, Roche, has completed a regulatory filing under the centralized process in the European Union (EU) for Tarceva^TM in NSCLC. Beyond Tarceva^TM, we have a balanced pipeline of oncology drug candidates that includes signal transduction inhibitors, apoptosis (also known as programmed cell death) inducers and a next-generation cytotoxic chemotherapy agent. We market and promote Novantrone® (mitoxantrone concentrate for injection) for approved oncology indications in the United States, and we market and distribute Gelclair®, a bioadherent oral gel for the relief of pain associated with oral mucositis, a frequent side-effect of chemotherapy, in North America. We also have an independently operated diabetes and obesity subsidiary, Prosidion Ltd., which is based in the United Kingdom. Prosidion’s lead clinical candidate, PSN9301, an inhibitor of dipeptidyl peptidase IV (DP-IV), is in Phase II clinical trials for the treatment of type 2 diabetes. PSN9301 was acquired by Prosidion from Probiodrug AG in July 2004. Behind PSN9301, Prosidion has an emerging pipeline of diabetes and obesity drug candidates.

Tarceva^TM. Tarceva^TM is a small molecule drug designed to target the HER1/ EGFR signaling pathway. The EGFR gene is known to be over-expressed, mutated or amplified in approximately 40% to 60% of the approximately 1.3 million new cases of cancer diagnosed in the United States each year. According to the American Cancer Society, lung cancer is the leading cause of cancer related deaths in the United States each year with an estimated 160,000 deaths in 2004. Patients with NSCLC, the initial target indication of Tarceva^TM, account for approximately 80% of these deaths. The American Cancer Society also estimates that 31,000 cancer patients in the United States will die from pancreatic cancer in 2004. Effective treatment of both NSCLC and pancreatic cancer remains a major unmet need.

Currently, front-line combination cytotoxic chemotherapy treatment of NSCLC only provides a median survival of approximately ten months. Patients may subsequently be treated with cytotoxic chemotherapy in the second-line setting; however, these agents have significant side-effects. In addition, patients may be treated with another EGFR inhibitor, Iressa®, which was approved for the third-line setting based on response rate data only. Iressa® has yet to demonstrate a survival benefit. Currently, only approximately 46,000 people a year suffering from NSCLC receive second and/or third-line treatments, usually with limited success. Many other patients either decline treatment or have treatment withheld due to the side-effects of cytotoxic chemotherapy or a perception that further treatment would be of limited benefit. Tarceva^TM has also demonstrated a survival benefit in pancreatic cancer. A Phase III study of the agent in combination with gemcitabine improved survival as compared with gemcitabine alone. Since

EGFR is implicated in a significant portion of both NSCLC and pancreatic cancer cases, and because we consider Tarceva^TM to be a novel, relatively well-tolerated and differentiated EGFR inhibitor, we believe that, upon approval, Tarceva^TM will represent a valuable treatment option for NSCLC patients after failure of initial therapy and for front-line pancreatic cancer patients when used in combination with gemcitabine.

In April 2004, we announced that Tarceva^TM met its primary endpoint of improving overall survival and its key secondary endpoints of progression-free survival and objective tumor response rate in a 731-patient randomized, double-blind placebo controlled Phase III trial in NSCLC patients, referred to as the BR.21 study. Tarceva^TM also delayed the deterioration of selected lung cancer symptoms in the BR.21 study. The trial compared Tarceva^TM to placebo in the treatment of patients with advanced NSCLC following the failure of first or second-line chemotherapy. The data demonstrates a 42% improvement in median survival and a 45% improvement in the one-year survival rate relative to placebo. The results revealed a survival benefit in essentially all subsets of patients examined, including males, smokers and patients with squamous cell carcinoma histology (subsets that, consistent with previous studies with EGFR inhibitors, had a relatively low rate of tumor response in our study), as well as females, non-smokers and patients with adenocarcinoma (subsets with higher rates of tumor response).

Recent publications have shown a strong correlation of tumor response with a group of newly-identified EGFR mutations in lung cancer which are clustered in patients who are non-smokers or have tumors with adenocarcinoma histology. These publications claim that the clinical benefit observed for EGFR inhibitors in NSCLC may be restricted to patients whose tumors have these EGFR mutations. The BR.21 study results clearly show, however, that tumor response is not always a good surrogate for survival benefit and that the improvement in overall survival seen in our BR.21 study cannot be explained by the reported incidence (approximately 10%) of these mutations. A survival benefit was seen in our pancreatic cancer study despite the fact that there was no difference in tumor response rates between the two arms in the study and that publications in scientific literature indicate that the mutations may be largely confined to lung cancer. This supports our belief that a much broader group of patients than those achieving a tumor response derive a meaningful survival benefit from treatment with Tarceva^TM.

We believe our BR.21 results are particularly noteworthy in that they demonstrate a meaningful, broad-based clinical benefit in a very advanced population of lung cancer patients. The cytotoxic chemotherapy agents, Taxotere® and Alimta®, showed similar survival results in a recent Phase III study comparing the two drugs; however, these agents exhibited a severe side-effect profile and were tested in a less advanced patient population. In contrast, Tarceva^TM has a relatively benign side-effect profile and the BR.21 study enrolled second and third-line patients, many of whom were in poorer overall health.

The table below summarizes key results from our BR.21 study. Approximately 50% of the patients in the study had failed one prior regimen of chemotherapy and the other half had failed at least two prior regimens of chemotherapy. In addition, a significant proportion of the patients entered the study with poor performance status, including 25% with performance status 2 and 9% with performance status 3. The Eastern Cooperative Oncology Group (ECOG) ranks patients’ performance status from 0 to 5 according to the level of daily activity and symptoms of disease, with 0 representing normal activity and 5 representing a patient who has died. It is unusual for patients with performance status 3 to be included in clinical trials.

Summary Data for the Tarceva^TM Phase III Study in Second and Third-Line NSCLC


Note:	The data is based on our statistical analysis of data from the BR.21 study.



(1)	Hazard ratio is a statistical measure of the difference in overall survival between the study drug group and the control group. A hazard ratio of less than 1 indicates a reduction in the risk of death; for example, a hazard ratio of 0.73 represents a 27% reduction in the risk of death and a 37% improvement in overall survival.


(2)	P-value is a statistical measure of significance. A p-value of <0.05 indicates a statistically significant difference.


(3)	Data is adjusted for stratification factors prior to randomization.


(4)	Objective tumor response rate represents the sum of the percentage of patients who exhibited a partial response (PR) or a complete response (CR).

The safety profile observed in the BR.21 study was relatively benign compared to cytotoxic chemotherapy and was consistent with that seen in prior Tarceva^TM studies with the most common side-effects consisting of rash (75%) and diarrhea (54%). We believe that the combination of survival benefit with a relatively benign side-effect profile positions Tarceva^TM as an important treatment option for oncologists treating advanced NSCLC patients. Tarceva^TM also demonstrated statistically significant delays in the time-to-symptom deterioration of the key lung cancer symptoms of cough, dyspnea and pain in the BR.21 study, as well as avoiding the severe drug related toxicities associated with chemotherapy, such as neutropenia, thrombocytopenia and alopecia.

Tarceva^TM was granted a Fast Track designation by the FDA in September 2002 for the second and third-line NSCLC indication, and in January 2004, we initiated the rolling submission of our NDA under provisions available with Fast Track status. Tarceva^TM was also granted priority review status as well as Pilot 1 status, which required the FDA to initiate a six-month review of each unit of the submission either as it was received or upon granting of Pilot 1 status, whichever was later. On July 30, 2004, we completed our NDA filing with the FDA for monotherapy Tarceva^TM use in the treatment of all NSCLC patients who have failed at least one prior chemotherapy regimen. Based upon the priority review status, the FDA has until January 30, 2005 (otherwise known as the PDUFA date) to take action on the NDA filing. We believe that the FDA is rapidly proceeding with its review of the NDA, and we anticipate receipt of approval from the FDA for the NSCLC indication in advance of the PDUFA date. We believe that Tarceva^TM will be the first EGFR inhibitor to receive full approval from the FDA. We anticipate launching Tarceva^TM within five business days of receipt of approval and our partner, Genentech, will finalize and announce pricing for the United States at the time of launch. We are responsible for manufacturing and supply in the United States. A supply chain has been established and sufficient inventory is on hand in

advance of a potential launch of Tarceva^TM. Our partner, Roche, filed for registration under the centralized process in the EU using the positive results of our BR.21 study in the third quarter of calendar 2004 and, assuming a successful review, we anticipate that Roche will launch Tarceva^TM in the second half of calendar 2005. We plan to discuss with the FDA our intent to submit an sNDA for Tarceva^TM for the pancreatic cancer indication with the FDA in the near future.

In January 2001, we signed co-development and marketing agreements with Genentech and Roche, forming a global alliance for the development and commercialization of Tarceva^TM in order to maximize its potential value. Under these agreements, we will receive milestone payments upon successive regulatory filings and approvals, an equal profit share on sales in the United States with Genentech, and a royalty on sales outside of the United States from Roche. We recently entered into agreements with Genentech with respect to promotion, marketing and manufacturing responsibilities in the U.S. market. Genentech will have the lead responsibility for the marketing and promotion of Tarceva^TM in the United States. However, we will co-promote the product and field at least 25% of the combined U.S. sales force.

The successful execution of a strategy to expand Tarceva^TM indications and grow the product post launch is our clear priority. We intend to broaden the use of Tarceva^TM to earlier stage lung cancer patients, both in the first-line and adjuvant setting. The first part of this strategy has been initiated with an ongoing randomized Phase II trial evaluating monotherapy Tarceva^TM against chemotherapy in patients who have received no prior chemotherapy and have poor performance status. Additionally, we intend to expand the use of Tarceva^TM to other disease settings in oncology. The positive results from our pancreatic cancer Phase III trial are an endorsement of our belief that Tarceva^TM will have broad utility in treating cancer. To date, tumor responses have been documented in Phase II studies for monotherapy Tarceva^TM in bronchioalveolar cell carcinoma, glioblastoma multiforme, head and neck cancer, hepatocellular carcinoma, breast cancer and ovarian cancer. Approximately 100 clinical trials of Tarceva^TM, including both registration oriented and publication only studies, are either ongoing or planned. We also intend to explore the use of Tarceva^TM in combination with other targeted therapies, particularly the anti-angiogenic antibody Avastin®. Data from Phase II studies has revealed promising indications of activity for the combination of Tarceva^TM and Avastin® in renal cell carcinoma and in advanced NSCLC patients with adenocarcinoma. Genentech has also implemented a Tarceva^TM expanded access program for patients with second and third-line NSCLC who have failed standard chemotherapy. Roche has begun a similar expanded access program in Europe.

Marketed Products and Balanced Pipeline. Beyond Tarceva^TM, we believe that we can realize significant future value by investing in our growing oncology franchise and in our second disease area business unit, Prosidion. Our promotional efforts for Novantrone® and Gelclair® have enabled us to establish a core commercial organization, which includes approximately 50 sales representatives and managers. We believe this commercial capability will enhance our ability to pursue additional in-licensing, acquisition and co-promotion opportunities. Our pipeline behind Tarceva^TM is comprised of targeted therapies and a next-generation cytotoxic chemotherapy agent as summarized in the chart below.


(1)	Being developed by Pfizer Inc. for which we are entitled to royalties if commercialized and to which we have development rights if Pfizer chooses to halt development for any reason.

Behind this development pipeline, we have invested in establishing an active discovery research program focused on small molecule targeted therapies, including advanced discovery projects targeting the development of agents that act as insulin-like growth factor receptor (IGF-1R) inhibitors and agents that can inhibit both phosphoinositide-dependent kinase (PDK) and EGFR. We will also continue to explore the potential of the selective apoptotic antineoplastic drug (SAAND) discovery platform we acquired from Cell Pathways, Inc. Based on historical data for the prototype molecule Aptosyn® in familial adenomatous polyposis and early prostate cancer, we have continued development of the more potent follow-on molecule, OSI-461, despite halting our Aptosyn® development efforts.

We consider expansion into a second disease area to be an important part of our strategy for long-term value creation. To this end, in July 2004 our diabetes and obesity subsidiary, Prosidion, acquired the type 2 diabetes clinical candidate, PSN9301, and its associated intellectual property estate from the German company, Probiodrug AG. PSN9301 is an oral, small molecule inhibitor of DP-IV, which is recognized as an important target in diabetes. DP-IV cleaves and inactivates glucogen-like peptide 1 (GLP-1), an important mediator of blood glucose levels. Inhibition of DP-IV leads to prolonged GLP-1 activity and inhibitors of DP-IV have demonstrated significant effects on blood glucose in clinical trials.

Novartis, Inc., Merck & Co., Inc. and Bristol-Myers Squibb Company (BMS) all have clinical development programs targeting DP-IV. In addition to composition of matter claims for PSN9301, the intellectual property estate acquired from Probiodrug AG includes issued U.S. methods-of-use claims that have been non-exclusively licensed to Novartis and Merck, among others, for milestones and royalties. Prosidion also anticipates initiating clinical trials for two diabetes candidates, PSN105 (a glucokinase activator) and PSN357 (a glycogen phosphorylase inhibitor), in the first half of calendar 2005.

Our Strategy for Future Growth. Our objective going forward is to build upon Tarceva^TM’s significant market potential and to capitalize on the experienced management team and the comprehensive set of capabilities from discovery to commercialization that we have established over the last several years in order to create a premier biotechnology organization and drive value creation for our stockholders. To accomplish this, we intend to:



	•	maximize Tarceva^TM’s value with a timely registration and an effective launch in the United States, the EU and Japan, and an effective product growth strategy;


	•	establish our position as a premier oncology franchise by advancing our pipeline, reinforcing our commercial presence, validating our research capabilities and actively pursuing in-licensing and acquisition opportunities; and

	•	diversify our business through continued investment in Prosidion to grow a second business unit focused on diabetes and obesity in order to help drive long-term growth.

We were incorporated in the State of Delaware in March 1983. Our corporate headquarters are located at 58 South Service Road, Suite 110, Melville, New York 11747, and our telephone number is (631) 962-2000.

The following table presents a summary of our historical consolidated financial information, which has been derived from our audited and unaudited consolidated financial statements. You should read this information in conjunction with our consolidated financial statements and related notes and the information under “Selected Consolidated Financial Data” and “Management’s Discussion and Analysis of Financial Condition and Results of Operations,” included or incorporated by reference in this prospectus.

Unless otherwise indicated, all information in this prospectus assumes the underwriters do not exercise their overallotment option to purchase up to 825,000 shares of common stock.

The number of shares of our common stock that will be outstanding after this offering is based on 43,586,747 shares outstanding as of September 30, 2004 and excludes:


	•	825,000 additional shares of common stock that the underwriters have a right to purchase from us within 30 days from the date of this prospectus to cover overallotments;


	•	4,843,279 shares of common stock issuable upon exercise of options outstanding as of September 30, 2004 at a weighted average exercise price of $36.59 per share;


	•	2,998,800 shares of common stock issuable upon the conversion of our 3 1/4% Convertible Senior Subordinated Notes due 2023 (2023 Notes) outstanding as of June 30, 2004;

	•	1,584,973 shares of common stock issuable to former Cell Pathways’ stockholders in the event an NDA is accepted for filing with the FDA by June 12, 2008 for either of the two clinical candidates acquired from Cell Pathways in June 2003;


	•	1,370,733 shares of common stock issuable upon the exercise of stock options not outstanding, but reserved for issuance as of September 30, 2004;



	•	675,571 shares reserved for issuance under our stock purchase plans as of September 30, 2004; and



	•	33,882 shares of common stock issuable upon exercise of outstanding and unexercised warrants as of September 30, 2004 which were assumed by us in connection with the Cell Pathways acquisition.

This offering involves a high degree of risk. You should carefully consider the risks and uncertainties described below and the other information in this prospectus, including the consolidated financial statements and the related notes incorporated by reference into this prospectus, before deciding to invest in shares of our common stock. If any of the following risks or uncertainties actually occurs, our business, prospects, financial condition and operating results would likely suffer, possibly materially. In that event, the market price of our common stock could decline and you could lose all or part of your investment.

We have had net operating losses since our inception in 1983. We expect to continue to incur operating losses over the next several years as a result of our expenses for the continued development of Tarceva^TM and our other clinical products and research programs and the expansion of our commercial operations. We cannot predict when our business will become profitable. We do not expect to achieve profitability for at least three years following the launch of Tarceva^TM.

At June 30, 2004, our accumulated deficit was $642.8 million. Our net losses were $47.3 million and $137.2 million for the three and nine months ended June 30, 2004, respectively, and $75.1 million and $132.4 million for the three and nine months ended June 30, 2003, respectively. We expect an increase in our net operating loss for the fourth quarter of fiscal 2004, as compared to the prior year period. This increase is primarily as a result of an in-process research and development charge and an impairment of intangible asset charge recorded in the fourth quarter of fiscal 2004. Our net losses were $181.4 million, $218.5 million and $23.8 million for fiscal years 2003, 2002 and 2001, respectively. Our net loss for fiscal 2003 included an in-process research and development charge of $31.5 million related to the acquisition of Cell Pathways in June 2003. Our net loss for fiscal 2002 included an in-process research and development charge of $130.2 million related to the acquisition of certain assets from Gilead Sciences, Inc. in December 2001.

Assuming that we are successful in receiving approval to market Tarceva^TM, our ability to successfully penetrate the market and generate sales of Tarceva^TM may be limited by a number of factors, including the following:


	•	Certain of our competitors in the HER1/ EGFR field, namely AstraZeneca plc and BMS/ ImClone Systems Inc., have already received regulatory approvals for and have begun marketing similar products in the United States, the EU, Japan and other territories, which may result in greater physician awareness of their products as compared to Tarceva^TM.

	•	Information from our competitors or the academic community indicating that current products or new products are more effective than Tarceva^TM could, if and when it is generated, impede our market penetration or decrease our existing market share.

	•	Physicians may be reluctant to switch from existing treatment methods, including traditional chemotherapy agents, to Tarceva^TM.

	•	The price for Tarceva^TM, which will be set by Genentech in the United States after consultation with us and by Roche outside of the United States, as well as pricing decisions by our competitors, may have an effect on our Tarceva^TM-derived revenues.

	•	Our Tarceva^TM-derived revenues may diminish if third-party payors, including private health coverage insurers and health maintenance organizations, do not provide adequate coverage or reimbursement for Tarceva^TM.

All of our potential products must undergo extensive regulatory approval processes in the United States and other countries. These regulatory processes, which include pre-clinical testing and clinical trials of each compound to establish safety and efficacy, can take many years and require the expenditure of substantial resources. The FDA and the other regulatory agencies in additional markets which are material to us and our collaborative partners, including the European Agency for the Evaluation of Medicinal Products (EMEA) and the Japanese Ministry of Health, may delay or deny the approval of our potential products. None of our potential products have yet received governmental approval.

A drug candidate cannot be marketed in the United States until it has been approved by the FDA. Once approved, the drug can only be marketed for the indications and claims approved by the FDA. The FDA also has the authority, when approving a product, to impose significant limitations on the product in the nature of warnings, precautions and contra-indications that could negatively affect the profitability of a drug. The ability to market and sell a drug product outside of the United States is also subject to stringent and, in some cases, equally complex regulatory processes that vary depending on the jurisdiction.

We have completed our filing with the FDA of an NDA for Tarceva^TM in NSCLC and Roche has completed a filing with the EMEA for Tarceva^TM under the centralized process for registration in the EU. The FDA or foreign regulatory authorities may not approve our NDA for Tarceva^TM or any other drug candidates that we develop, in a timely manner or at all. Delays or rejections may be encountered during any stage of the regulatory process based upon the failure of the clinical data to demonstrate compliance with, or upon the failure of the product to meet, a regulatory agency’s requirements for safety, efficacy and quality. Any such delay could have a negative effect on our business. Once approved, drugs, as well as their manufacturers, are subject to continuing and ongoing review, and discovery of previously unknown problems with these products or the failure to adhere to manufacturing or quality control requirements may result in restrictions on their distribution, sale or use, or their withdrawal from the market. Furthermore, once a drug is approved, if we fail to comply with FDA regulations prohibiting promotion of off-label uses and the promotion of products for which marketing clearance has not been obtained, the FDA, or the Office of the Inspector General of the U.S. Department of Health and Human Services, or state Attorneys General could bring an enforcement action against us that would inhibit our marketing capabilities as well as result in significant penalties.

We are responsible for manufacturing and supplying Tarceva^TM in the United States under the terms of a Manufacturing and Supply Agreement entered into with Genentech in 2004. We rely on two third-party suppliers to manufacture the API, erlotinib, for Tarceva^TM. We also currently rely on a single manufacturer to formulate the Tarceva^TM tablets. We are presently seeking another manufacturer to serve as an alternative provider of Tarceva^TM tablets. If our relationships with any of these manufacturers terminate or if they are unable to meet their obligations, we will need to find other sources of supply. Such alternative sources of supply may be difficult to find on terms acceptable to us or in a timely manner, and, if found, would require FDA approval which could cause delays in the availability of erlotinib and ultimately Tarceva^TM tablets, which, in turn, could negatively impact our Tarceva^TM-derived revenues.

Furthermore, the manufacturing of our products is, and will continue to be, subject to current good manufacturing practices regulations prescribed by the FDA or other standards prescribed by the appropriate regulatory agency in the country of use. If our manufacturers, including the current manufacturers of erlotinib and Tarceva^TM tablets, do not comply with all applicable regulatory standards, they may not be permitted to manufacture Tarceva^TM or any other product for commercial sale. If this occurs, we might not be able to identify another third-party manufacturer on terms acceptable to us or in a timely manner, or such other third-party manufacturer may not receive FDA approval in a timely

manner or at all. Any of the foregoing could cause delays in the availability of our products, including erlotinib and/or Tarceva^TM tablets, which would negatively impact our revenues. If we fail to meet our manufacturing obligations, our partner, Genentech, also has the right to take over supply of Tarceva^TM in the United States.

Tarceva^TM is being developed in an alliance under co-development and marketing agreements with Genentech and Roche. The development program is managed by us, Genentech and Roche under a global development committee. If Tarceva^TM receives regulatory approval, under the alliance, Genentech will lead the marketing efforts in the United States and Roche will market the drug in the rest of the world. Recently, we signed an amendment to our collaboration agreement with Genentech to provide us with the right to co-promote Tarceva^TM in the United States and signed a Manufacturing and Supply Agreement with Genentech that clarified our role in supplying Tarceva^TM for the U.S. market.

The OSI/ Genentech collaboration agreement continues until the date on which neither we nor Genentech is entitled to receive a share of the operating profits or losses on any products resulting from the collaboration, that is, until the date that we and Genentech mutually agree to terminate the collaboration or until either party exercises its early termination rights as described as follows. The OSI/ Genentech collaboration agreement is subject to early termination in the event of certain customary defaults, such as material breach of the agreement and bankruptcy. Since January 8, 2003, Genentech has had the right to terminate the OSI/ Genentech collaboration agreement with six months’ prior written notice. The provisions of the amendment allowing us to co-promote are also subject to termination by Genentech upon a material breach of the amendment by us which remains uncured or upon a pattern of nonmaterial breaches which remains uncured.

The OSI/ Roche agreement continues until the date on which we are no longer entitled to receive a royalty on products resulting from the development of Tarceva^TM, that is, until the date of expiration or revocation or complete rejection of the last to expire patent covering Tarceva^TM or, in countries where there is no valid patent covering Tarceva^TM, on the tenth anniversary of the first commercial sale of Tarceva^TM in that country. The OSI/ Roche agreement is subject to early termination in the event of certain customary defaults, such as material breach of the agreement and bankruptcy. In addition, since July 31, 2003, Roche has had the right to terminate the agreement on a country-by-country basis with six months’ prior written notice. Since July 31, 2003, we also have had the right to terminate the agreement on a country-by-country basis if Roche has not launched or marketed a product in such country under certain circumstances.

If we do not maintain a successful collaborative partnership with Genentech and Roche for the co-development and commercialization of Tarceva^TM, we may be forced to focus our efforts internally to commercialize Tarceva^TM. This would require greater financial resources and would result in us incurring greater expenses and may cause a delay in launch and market penetration while we continue to build our own commercial operation or seek alternative collaborative partners.

If Novantrone® or Tarceva^TM or any of our other current or future marketed products become the subject of problems, including those related to, among others:


	•	efficacy or safety concerns with the products, even if not justified;

	•	unexpected side-effects;

	•	regulatory proceedings subjecting the products to potential recall;


	•	publicity affecting doctor prescription or patient use of the product;

	•	pressure from competitive products; or

	•	introduction of more effective treatments;

our revenues from such marketed products could decrease. For example, efficacy or safety concerns may arise, whether or not justified, that could lead to the recall or withdrawal of such marketed products. In the event of a recall or withdrawal of a product such as Novantrone®, our revenues would significantly decline.

Sales of our products will depend, in part, upon the extent to which the costs of our products will be paid by health maintenance, managed care, pharmacy benefit and similar reimbursement sources, or reimbursed by government health administration authorities, private health coverage insurers and other third-party payors. Such third-party payors continue to aggressively challenge the prices charged for healthcare products and services. Additionally, federal and state governments have prioritized the containment of healthcare costs, and drug prices have been targeted in this effort. If these organizations and third-party payors do not consider our products to be cost-effective, they may not reimburse providers of our products, or the level of reimbursement may not be sufficient to allow us to sell our products on a profitable basis.

After a Congressional debate regarding the formula by which Medicare reimbursement for intravenous oncology products is rendered to oncologists, the levels of reimbursement to oncologists for intravenous oncology products like Novantrone® were lowered, effective January 2004. Under the new Medicare Prescription Drug Improvement and Modernization Act of 2003, Medicare benefits will be primarily provided through private entities that will attempt to negotiate price concessions from pharmaceutical manufacturers, which may increase pressure to lower prescription drug prices. The Act also includes other cost containment measures for Medicare in the event Medicare cost increases exceed a certain level, which may also impose limitations on prescription drug prices. These changes in Medicare reimbursement could have a negative impact on our revenues derived from sales of Novantrone®.

Serono is responsible for the manufacture and supply of Novantrone®. Under our agreement with Serono, we do not have the obligation nor the right to manufacture Novantrone®. These obligations and rights are held solely by Serono. If Serono is delayed in or restricted from supplying the product, we would be directly affected in that any such delay or restriction would impede us from selling the product. Without the sales of Novantrone®, our revenues would decrease.

As part of our business strategy, we plan to identify and acquire product candidates and approved products for markets that we can reach through our commercial operations. We may not be able to acquire rights to additional products or product candidates on acceptable terms, if at all. If we fail to obtain, develop and successfully commercialize such additional product candidates and approved products, we may not achieve expectations of our future performance.

Successful research and development of pharmaceutical products is high risk. Most products and development candidates fail to reach the market. Our success depends on the discovery of new drugs that we can commercialize. It is possible that our potential oncology products and diabetes and obesity products may never reach the market for a number of reasons. They may be found ineffective or may cause harmful side-effects during pre-clinical testing or clinical trials or fail to receive necessary regulatory approvals. We may find that certain products cannot be manufactured on a commercial scale basis and, therefore, they may not be economical to produce. Our products could also fail to achieve market acceptance or be precluded from commercialization by proprietary rights of third parties. We have a number of product candidates in various stages of development and do not expect them to be commercially available for a number of years, if at all. Our candidates that are in clinical trials will still require significant research and development and regulatory approvals before we or our collaborative partners will be able to market them.

We face significant competition from industry participants that are pursuing similar products and technologies that we are pursuing and are developing pharmaceutical products that are competitive with our products and potential products. Some of our industry competitors have greater capital resources, larger overall research and development staffs and facilities, and a longer history in drug discovery and development, obtaining regulatory approval and pharmaceutical product manufacturing and marketing than we do. With these additional resources, our competitors may be able to respond to the rapid and significant technological changes in the biotechnology and pharmaceutical industries faster than we can. Our future success will depend in large part on our ability to maintain a competitive position with respect to these technologies. Rapid technological development, as well as new scientific developments, may result in our compounds, products or processes becoming obsolete before we can recover any of the expenses incurred to develop them. For example, changes in our understanding of the appropriate population of patients who should be treated with a targeted therapy like Tarceva^TM may limit the drug’s market potential if it is subsequently demonstrated that only certain subsets of patients should be treated with the targeted therapy.

In the course of product development, we engage CROs to conduct and manage clinical studies and to assist us in guiding our products through the FDA review and approval process. For example, we collaborated with the National Cancer Institute of Canada’s Clinical Trial Group based at Queens University, Ontario, in connection with our Tarceva^TM Phase III trials. Because we have engaged and intend to continue to engage CROs to help us obtain market approval for our drug candidates, many important aspects of this process have been and will be out of our direct control. If the CROs fail to perform their obligations under our agreements with them or fail to perform clinical trials in a satisfactory manner, we may face delays in completing our clinical trials, as well as commercialization of one or more drug candidates. Furthermore, any loss or delay in obtaining contracts with such entities may also delay the completion of our clinical trials and the market approval of drug candidates.

The nature of our business exposes us to potential liability risks inherent in the testing, manufacturing and marketing of drug candidates and products. If any of our drug candidates in clinical trials or our marketed products harm people or allegedly harm people, we may be subject to costly and damaging product liability claims. A number of patients who participate in trials are already critically ill when they

enter a trial. The waivers we obtain may not be enforceable and may not protect us from liability or the costs of product liability litigation. While we currently maintain product liability insurance that we believe is adequate, we are subject to the risk that our insurance will not be sufficient to cover claims. There is also a risk that adequate insurance coverage will not be available in the future on commercially reasonable terms, if at all. The successful assertion of an uninsured product liability or other claim against us could cause us to incur significant expenses to pay such a claim, could adversely affect our product development and could cause a decline in our product revenues. Even a successfully defended product liability claim could cause us to incur significant expenses to defend such a claim, could adversely affect our product development and could cause a decline in our product revenues.

The development of some of our products may require the use of technology developed by third parties. The extent to which efforts by other researchers have resulted or will result in patents and the extent to which we or our collaborative partners are forced to obtain licenses from others, if available, on commercially reasonable terms is currently unknown. If we or our collaborative partners must obtain licenses from third parties, fees must be paid for such licenses, which would reduce the revenues and royalties we may receive on commercialized products.

We hold numerous U.S. and foreign patents and have many pending applications for additional patents. We intend to continue to seek patent protection for or maintain as trade secrets all of the commercially promising product candidates that we have discovered, developed or acquired. Our success depends, in part, on our ability and our collaborative partners’ ability to obtain and maintain patent protection for new product candidates, maintain trade secret protection and operate without infringing the proprietary rights of third parties. As with most biotechnology and pharmaceutical companies, our patent position is highly uncertain and involves complex legal and factual questions. Without patent and other similar protection, other companies could offer substantially identical products for sale without incurring the sizeable discovery and development costs that we have incurred. Our ability to recover these expenditures and realize profits upon the sale of products could be diminished. The process of obtaining patents can be time-consuming and expensive with no certainty of success. Even if we spend the necessary time and money, a patent may not issue or it may insufficiently protect the technology it was intended to protect. We can never be certain that we were first to develop the technology or that we were the first to file a patent application for the particular technology because most U.S. patent applications are confidential until a patent issues, and publications in the scientific or patent literature lag behind actual discoveries. If our pending patent applications are not approved for any reason or if we are unable to receive patent protection for additional proprietary technologies that we develop, the degree of future protection for our proprietary rights will remain uncertain. Furthermore, third parties may independently develop similar or alternative technologies, duplicate some or all of our technologies, design around our patented technologies or challenge our issued patents.

Our processes and potential products may conflict with patents that have been or may be granted to competitors, academic institutions or others. As the biotechnology and pharmaceutical industries expand and more patents are filed and issued, the risk increases that our product candidates may give rise to a declaration of interference by the U.S. Patent and Trademark Office, to administrative proceedings in foreign patent offices or to claims of patent infringement by other companies, institutions or individuals. These entities or persons could bring legal proceedings against us seeking substantial damages or seeking to enjoin us from testing, manufacturing or marketing our products. If any of these actions were successful,

we may also be required to cease the infringing activity or obtain the requisite licenses or rights to use the technology that may not be available to us on acceptable terms, if at all. Any litigation, regardless of the outcome, could be extremely costly to us.

Changes in our operating plans, increased expenses or other events described in this “Risk Factors” section may require us to seek additional debt or equity financing. Financing may not be available on acceptable terms, or at all, and our failure to raise capital when needed could negatively impact our growth, financial condition and results of operations. Additional financing would most likely be dilutive to the holders of our common stock, and debt financing, if available, may involve significant cash payment obligations and covenants that restrict our ability to operate our business.

We grant stock options to our employees and other individuals as part of our overall compensation plan which, upon vesting, are exercisable for common stock. In addition, we have issued convertible debt which may be converted into common stock as well as contingent value rights which, upon the occurrence of certain events, may be exchanged for common stock. We are not able to estimate when, if ever, the stock options or convertible debt will be exercised or converted into common stock or when, if ever, shares will be issued in connection with the contingent value rights, but any such conversion or issuance would almost certainly dilute stockholder value.

Further, if some or all of such shares are registered and sold into the public market over a short time period, the price of our stock is likely to decline, as the market may not be able to absorb those shares at the prevailing market prices.

As a result of the issuance of our 2023 Notes, our long-term debt represented by these notes was $150.0 million as of September 30, 2004. Our 2023 Notes significantly increased our interest expense and related debt service costs. Interest on these notes accrues at the rate of 3.25% per annum. This amounts to interest payments of $2.4 million due and payable on the 2023 Notes semi-annually on March 8 and September 8 of each year on the outstanding amount of the notes. Cumulative interest payments of $95.1 million are scheduled to be paid between September 8, 2004 and September 8, 2023 on the 2023 Notes.


	•	make it more difficult for us to obtain any necessary financing in the future for working capital, capital expenditures, debt service requirements or other purposes; and

	•	make us more vulnerable in the event of a downturn in our business.

We currently are not generating sufficient cash flow to satisfy the annual debt service payments on the notes. If we are unable to satisfy our debt service requirements, we will default on our 2023 Notes, and we would face liquidity problems as a result.

When the stock prices of companies in the Nasdaq Biotechnology Index fall, our stock price will most likely fall as well. The stock price of biotechnology and pharmaceutical companies, including our stock price, has been volatile and may remain volatile for the foreseeable future. From October 1, 2001 through

September 30, 2002, the range of our stock price was between $50.94 and $11.50, and the range of the Nasdaq Biotechnology Index was between 978.42 and 397.36. From October 1, 2002 through September 30, 2003, the range of our stock price was between $38.34 and $12.84, and the range of the Nasdaq Biotechnology Index was between 801.40 and 442.09. From October 1, 2003 through November 3, 2004, the range of our stock price was between $98.70 and $24.47, and the range of the Nasdaq Biotechnology Index was between 851.44 and 622.01. The following factors, among others, some of which are beyond our control, may also cause our stock price to decline:


	•	fluctuations in operating results;

	•	announcements of technological innovations or new therapeutic products by others;

	•	negative or neutral clinical trial results;

	•	developments concerning strategic alliance agreements;

	•	unanticipated clinical efficacy or safety results from our competitors’ products;

	•	changes in government regulation, including pricing controls;

	•	delays with the FDA in the approval process for Tarceva^TM and other clinical candidates;

	•	developments in patent or other proprietary rights;

	•	public concern as to the safety of our products;

	•	future sales of substantial amounts of our common stock by existing stockholders; and

	•	comments by securities analysts and general market conditions.

In addition, historically, our stock price has been affected by technological, clinical and regulatory developments in the HER1/ EGFR field, including developments with respect to the products of our main competitors in the field. It is possible that future developments concerning our competitors’ products as well as further research and clinical results of targeted therapies in general and the HER1/ EGFR field in particular could have an impact on our stock price due to Tarceva^TM’s classification as a targeted therapy in the HER1/ EGFR field.

If our stock price falls, our stockholders may not be able to sell their stock when desired or at desirable prices. Further, you may not be able to resell your shares at or above the public offering price.

We have had a shareholder rights plan, commonly referred to as a “poison pill,” since January 1999. The purpose of the shareholder rights plan is to protect stockholders against unsolicited attempts to acquire control of us that do not offer a fair price to our stockholders as determined by our board of directors. Under the plan, the acquisition of 17.5% or more of our outstanding common stock by any person or group, unless approved by our board of directors, will trigger the right by our stockholders to acquire additional shares of our common stock, and, in certain cases, the stock of the potential acquiror, at a bargain purchase price, thus significantly increasing the acquisition cost to a potential acquiror. The shareholder rights plan may have the effect of dissuading a potential hostile acquiror from making an offer for our common stock at a price that represents a premium to the then current trading price. Our certificate of incorporation and by-laws contain certain additional anti-takeover protective devices. For example,


	•	no stockholder action may be taken without a meeting, without prior notice and without a vote; solicitations by consent are thus prohibited;

	•	special meetings of stockholders may be called only by our board of directors;


	•	nominations by stockholders of candidates for election to the board of directors at our annual meeting of stockholders must be made at least 45 days prior to the date on which we first mailed our proxy materials for the prior year’s annual meeting of stockholders; and

	•	our board of directors has the authority, without further action by the stockholders, to fix the rights and preferences, and issue shares, of preferred stock. An issuance of preferred stock with dividend and liquidation rights senior to the common stock and convertible into a large number of shares of common stock could prevent a potential acquiror from gaining effective economic or voting control.

Further, we are subject to Section 203 of the Delaware General Corporation Law which, subject to certain exceptions, restricts certain transactions and business combinations between a corporation and a stockholder owning 15% or more of the corporation’s outstanding voting stock for a period of three years from the date the stockholder becomes a 15% stockholder. In addition to discouraging a third party from acquiring control of us, the foregoing provisions could impair the ability of existing stockholders to remove and replace our management and/or our board of directors.

The purchase price of the shares of common stock offered by this prospectus are substantially higher than the unaudited pro forma net tangible book value of our outstanding common stock. Investors who purchase shares in this offering will therefore experience immediate and substantial dilution in the net tangible book value of their investment. Subsequent sales of common stock could also have the effect of lowering our stock price, thereby increasing the number of shares issuable and consequently further diluting our outstanding common stock. These sales could have an immediate negative effect on the market price of our common stock.

The perceived risk associated with the possible sale of a large number of shares of our common stock could cause some of our stockholders to sell their stock, which might cause the price of our stock to decline. In addition, actual or anticipated downward pressures on our stock price due to actual or anticipated sales of stock could cause some institutions or individuals to engage in short sales of our common stock, which may cause the price of our common stock to decline.

This prospectus and the documents incorporated by reference in this prospectus include forward-looking statements. We have based these forward-looking statements on our current expectations and projections about future events. These forward-looking statements include, among other things, statements relating to:


	•	our anticipated business strategies;

	•	our and our partners’ ability to obtain regulatory approval of drug candidates;

	•	our ability to conduct clinical trials;

	•	our relationships with third parties, including manufacturers, CROs, collaborative partners and suppliers;

	•	our pending and anticipated clinical trials;

	•	our intention to introduce new product candidates;

	•	our acquisition of and integration of businesses, assets and liabilities;

	•	anticipated sales commissions and revenues;

	•	anticipated trends in our businesses;

	•	sufficiency of resources to fund operating and capital requirements;

	•	operating cash burn rates; and

	•	future capital expenditures.

The forward-looking statements included in this prospectus or in the documents incorporated by reference in this prospectus are subject to risks, uncertainties and assumptions about us. Our actual results of operations may differ materially from the forward-looking statements as a result of, among other things, the success or failure of our clinical trials, the speed at which our clinical trials progress, the success of our competitors in developing products equal or superior to ours and the timing of their development of such products, the success of our collaborative relationships and the other reasons described under “Risk Factors.” Except for our ongoing obligations to disclose material information under the federal securities laws, we undertake no obligation to publicly update or revise any forward-looking statement, whether as a result of new information, future events or otherwise, except as required by law. In light of these risks, uncertainties and assumptions, the forward-looking events discussed in this prospectus might not occur.

For these forward-looking statements, we claim the protection of the safe harbor for forward-looking statements contained in Section 21E of the Securities Exchange Act of 1934, as amended (Exchange Act) and Section 27A of the Securities Act of 1933, as amended (Securities Act).

We estimate that we will receive net proceeds from this offering of $333.7 million ($383.9 million if the underwriters exercise the overallotment option in full), based on an assumed public offering price of $64.345 per share and after deducting underwriting discounts and commissions and estimated offering expenses payable by us. We anticipate that the net proceeds from this offering will be used primarily for:



	•	the ongoing development and commercialization of Tarceva^TM in the United States;


	•	our research and development activities;

	•	strategic acquisitions and in-licensing opportunities;

	•	working capital requirements; and

	•	general corporate purposes.

We have not identified the amounts we plan to spend on each of these areas or the timing of such expenditures, and we will have significant discretion in the use of any net proceeds. The amounts actually expended for each purpose may vary significantly depending upon numerous factors. Prior to using the proceeds for the purposes described above, we plan to invest the net proceeds of this offering in short-term, investment-grade interest-bearing securities.

Our common stock is traded on the Nasdaq National Market under the symbol “OSIP.” The following is the range of high and low sales prices by fiscal quarter for our common stock from the first quarter of fiscal 2002 through November 3, 2004 as reported on the Nasdaq National Market:

As of August 31, 2004, there were approximately 949 holders of record of our common stock.

We have never declared or paid cash dividends on our common stock and do not anticipate declaring or paying any dividends in the foreseeable future.


	•	On an actual basis;
	•	On a pro forma basis to give effect to the full redemption on July 19, 2004 of our outstanding 4% Convertible Senior Subordinated Notes due 2009 (2009 Notes). The aggregate principal amount outstanding of the 2009 Notes was $160 million, and such 2009 Notes were converted into 3.2 million shares of our common stock. The remaining balance of the debt issuance costs of $3.7 million was reclassified to additional paid-in capital. Cash, cash equivalents and investment securities has been reduced by $6.4 million to reflect the payment of the remaining portion of the guaranteed interest on the 2009 Notes; and

	•	On a pro forma as adjusted basis to adjust the pro forma information to give effect to the sale of 5,500,000 shares of common stock offered by us in this offering, based on an assumed public offering price of $64.345 per share and after deducting the underwriting discounts and commissions and estimated offering expenses payable by us.

The number of shares of common stock that will be outstanding after this offering does not include:


	•	825,000 additional shares of common stock that the underwriters have a right to purchase from us within 30 days from the date of this prospectus to cover overallotments;

	•	5,177,161 shares of common stock issuable upon exercise of options outstanding as of June 30, 2004 at a weighted average exercise price of $35.66 per share;

	•	2,998,800 shares of common stock issuable upon the conversion of our 2023 Notes outstanding as of June 30, 2004;
	•	1,584,973 shares of common stock issuable to former Cell Pathways’ stockholders in the event an NDA is accepted for filing with the FDA by June 12, 2008 for either of the two clinical candidates acquired from Cell Pathways in June 2003;

	•	1,379,658 shares of common stock issuable upon the exercise of stock options not outstanding, but reserved for issuance as of June 30, 2004;


	•	680,730 shares reserved for issuance under our stock purchase plans as of June 30, 2004; and


	•	33,882 shares of common stock issuable upon exercise of outstanding and unexercised warrants as of June 30, 2004 which were assumed by us in connection with the Cell Pathways acquisition.

Our reported net tangible book value as of June 30, 2004 was $11.7 million, or $0.29 per share of common stock, based upon 40,030,398 shares outstanding as of this date. Net tangible book value per share is calculated by subtracting our total liabilities from our total tangible assets, which equals total assets less intangible assets, and dividing this amount by the number of shares of common stock outstanding as of June 30, 2004. The following discussion of dilution, and the table qualifying it, assumes the issuance by us of 3.2 million shares of our common stock and an increase in net tangible book value of $160.0 million in connection with the conversion by holders of all of our 2009 Notes in July 2004 occurred as of June 30, 2004.

Assuming the sale by us of 5,500,000 shares of common stock offered in this offering at an assumed public offering price of $64.345 per share and after deducting the underwriting discounts and commissions and estimated offering expenses payable by us, our net tangible book value as of June 30, 2004 would have been $505.5 million, or $10.37 per share of common stock. This represents an immediate increase in the net tangible book value of $6.40 per share to our existing stockholders and an immediate dilution in the net tangible book value of $53.97 per share to new investors.

The selected financial information set forth below as of September 30, 2003 and 2002 and for the three-year period ended September 30, 2003 has been derived from our audited consolidated financial statements, which are included in our annual report on Form 10-K for the year ended September 30, 2003 and are incorporated by reference in this prospectus. The selected financial information set forth below as of September 30, 2001, 2000 and 1999 and for the years ended September 30, 2000 and 1999 has been derived from our audited consolidated financial statements, and are not incorporated by reference in this prospectus. The selected financial information set forth below for the nine months ended June 30, 2004 and 2003 and as of June 30, 2004 has been derived from our unaudited consolidated financial statements included in our quarterly report on Form 10-Q for the period ended June 30, 2004 and are incorporated by reference in this prospectus. The financial information shown below should be read in conjunction with the consolidated financial statements and accompanying notes incorporated by reference in this prospectus and “Management’s Discussion and Analysis of Financial Condition and Results of Operations” incorporated by reference in this prospectus.


(1)	The fiscal 1999 consolidated financial statements include the acquisition of Cadus Pharmaceutical Corporation’s research business for $2.2 million in cash, including an $806,000 charge to operations for in-process research and development acquired; a gain of $3.3 million on the sale of our Anaderm Research Corporation stock to Pfizer; and a $535,000 charge to operations for the estimated costs of closing our facilities in North Carolina.

(2)	The fiscal 2000 consolidated financial statements include a $3.5 million technology access fee received upon the execution of a collaborative research and license agreement with Tanabe Seiyaku Co., Ltd.; non-cash compensation charges of approximately $6.8 million and deferred compensation of approximately $8.8 million associated with options issued to an employee and consultants; net proceeds of approximately $53 million from a private placement of common stock; a $3.7 million gain resulting from the sale of our diagnostics business, including the assets of our wholly-owned subsidiary, OSDI, Inc., to The Bayer Corporation; and a charge to operations of $700,000 representing the cost of a license to use and practice certain of Cadus’s technology and patents. (See notes 5(c), 11(a) and 18 to the consolidated financial statements included in our annual report on Form 10-K for the fiscal year ended September 30, 2003.)

(3)	The fiscal 2001 consolidated financial statements include a cumulative effect of the change in accounting principle of $2.6 million relating to the adoption of SAB No. 101; the acquisition of certain assets from British Biotech plc for $13.9 million; $25 million in upfront fees received upon the execution of collaboration agreements with Genentech and Roche; net proceeds of approximately $404 million from a public offering of common stock in November 2000; the sale of newly-issued shares of common stock to Genentech and Roche for an aggregate purchase price of $35 million each; and a charge to operations of $5.1 million for the estimated cost of closing our Birmingham, England and Tarrytown, New York facilities. (See notes 1(b), 3(d), 5(a), 11(f), 11(g), 17(a) and 17(b) to the consolidated financial statements included in our annual report on Form 10-K for the fiscal year ended September 30, 2003.)

(4)	The fiscal 2002 consolidated financial statements include the acquisition of certain assets from Gilead for approximately $175.7 million in cash and common stock; the receipt of $4.5 million from the phase-down of our collaboration with Anaderm, of which $1.8 million was recognized as revenue in accordance with SAB No. 101; the issuance of $200.0 million of convertible senior subordinated notes for net proceeds of approximately $192.9 million; and the early retirement of $40.0 million aggregate principal amount of convertible senior subordinated notes resulting in a net gain of approximately $12.6 million. (See notes 3(b), 5(b) and 10(b) to the consolidated financial statements included in our annual report on Form 10-K for the fiscal year ended September 30, 2003.)


(5)	The fiscal 2003 consolidated financial statements include the acquisition of the marketing and promotion rights to Novantrone® for approved oncology indications in the United States for approximately $45.0 million in cash; the acquisition of Cell Pathways for approximately $55.0 million in common stock, contingent value rights and cash; the issuance of $150.0 million of convertible senior subordinated notes for net proceeds of approximately $145.1 million and the purchase of 503,800 shares of our common stock for $19.0 million. (See notes 2, 3(a) and 10(a) to the consolidated financial statements included in our annual report on Form 10-K for the fiscal year ended September 30, 2003.)

(6)	Includes sales commissions which represent commissions earned on the sales of the drug, Novantrone®, in the United States for oncology indications pursuant to a Co-Promotion Agreement dated March 11, 2003 with Ares Trading S.A., an affiliate of Serono. Sales commissions from Novantrone® on net oncology sales are recognized in the period the sales occur based on the estimated split between oncology sales and multiple sclerosis sales of Novantrone®, as determined on a quarterly basis by a contracted external third party. (See note 1(b) to the consolidated financial statements included in our annual report on Form 10-K for the fiscal year ended September 30, 2003.) In June 2004, we called for the full redemption of the outstanding $160.0 million of the 2009 Notes. All of the holders of the 2009 Notes converted their notes into shares of our common stock prior to the redemption date of July 19, 2004. As a result of these conversions, on July 19, 2004, we issued 3.2 million shares of our common stock and paid the remaining portion of the guaranteed interest of $6.4 million. Under the terms of the 2009 Notes, the note holders were guaranteed the payment of interest for the first three years through February 1, 2005. The note holders became fully entitled to this guaranteed interest on June 18, 2004, the date we called for redemption of the 2009 Notes, and accordingly we accrued an additional interest charge of $3.7 million in the quarter ended June 30, 2004. Upon conversion of the 2009 Notes, the remaining balance of the debt issuance costs of $3.7 million was reclassified to additional paid-in capital in July 2004. Since the 2009 Notes were converted to common stock, the 2009 Notes have been reflected as a non-current liability on the accompanying consolidated balance sheet data as of June 30, 2004. (See note 17 to the unaudited consolidated financial statements included in our quarterly report on Form 10-Q for the quarter ended June 30, 2004.)

We are a leading biotechnology company primarily focused on the discovery, development and commercialization of high quality oncology products that both extend life and improve the quality-of-life for cancer patients worldwide. Our flagship product, Tarceva^TM, is an oral, once-a-day, small molecule inhibitor of the HER1/EGFR. Tarceva^TM is the first EGFR inhibitor, and the first non-chemotherapy agent, to demonstrate a survival benefit in advanced NSCLC and also in pancreatic cancer, two forms of cancer widely recognized as amongst the toughest treatment challenges facing oncologists. On July 30, 2004, we completed our NDA filing with the FDA for monotherapy Tarceva^TM use in the treatment of all NSCLC patients who have failed at least one prior chemotherapy regimen. Based upon the FDA’s prior designation of Fast Track status and priority review status for Tarceva^TM, the FDA has until January 30, 2005 (otherwise known as the PDUFA date) to take action on the NDA filing. We believe that the FDA is rapidly proceeding with its review of the NDA, and we anticipate receipt of approval from the FDA for the NSCLC indication in advance of the PDUFA date. We intend to file an sNDA for the treatment of front-line pancreatic cancer patients (in combination with gemcitabine) in 2005. In addition, our partner, Roche, has completed a regulatory filing under the centralized process in the EU for Tarceva^TM in NSCLC. Beyond Tarceva^TM, we have a balanced pipeline of oncology drug candidates that includes signal transduction inhibitors, apoptosis inducers and a next-generation cytotoxic chemotherapy agent. We market and promote Novantrone® (mitoxantrone concentrate for injection) for approved oncology indications in the United States, and we market and distribute Gelclair®, a bioadherent oral gel for the relief of pain associated with oral mucositis, a frequent side-effect of chemotherapy, in North America. We also have an independently operated diabetes and obesity subsidiary, Prosidion, which is based in the United Kingdom. Prosidion’s lead clinical candidate, PSN9301, an inhibitor of DP-IV, is in Phase II clinical trials for the treatment of type 2 diabetes. PSN9301 was acquired by Prosidion from Probiodrug AG in July 2004. Behind PSN9301, Prosidion has an emerging pipeline of diabetes and obesity drug candidates.

Our strategy is to build upon Tarceva^TM’s significant market potential and to capitalize on the experienced management team and the comprehensive set of capabilities from discovery to commercialization that we have established over the last several years in order to create a premier biotechnology organization and drive value creation for our stockholders. To accomplish this, we intend to:



	•	maximize Tarceva^TM’s value with a timely registration and an effective launch in the United States, the EU and Japan, and an effective product growth strategy;


	•	establish our position as a premier oncology franchise by advancing our pipeline, reinforcing our commercial presence, validating our research capabilities and actively pursuing in-licensing and acquisition opportunities; and

	•	diversify our business through continued investment in Prosidion to grow a second business unit focused on diabetes and obesity in order to help drive long term growth.

Maximize Tarceva^TM’s value. Our immediate focus is on the timely registration and launch of Tarceva^TM in the key global markets of the United States, the EU and Japan. Our primary indication is second and third-line NSCLC; however, we intend to submit an sNDA to the FDA for the pancreatic cancer indication during 2005. On July 30, 2004, we filed our NDA for Tarceva^TM as a monotherapy for the treatment of patients with advanced NSCLC for whom chemotherapy has failed. Tarceva^TM was designated Fast Track status in September 2002 and, in June 2004, Tarceva^TM was granted priority review status and Pilot 1 status which required the FDA to initiate a six-month review of each unit of the submission as it was received, or upon granting of Pilot 1 status, whichever was later. On July 30, 2004, we completed our NDA filing with the FDA for monotherapy Tarceva^TM use in the treatment of all NSCLC patients who have failed at least one prior chemotherapy regimen. Based upon the priority review status, the FDA has until January 30, 2005 (otherwise known as the PDUFA date) to take action on the NDA filing. We believe that the FDA is rapidly proceeding with its review of the NDA, and we anticipate receipt of approval from the FDA for the NSCLC indication in advance of the PDUFA date. We believe that Tarceva^TM will be the first EGFR inhibitor to receive full approval from the FDA. We anticipate

launching Tarceva^TM within five business days of receipt of approval and our partner, Genentech, will finalize and announce pricing for the United States at the time of launch. We anticipate that Roche will launch Tarceva^TM for NSCLC in the EU, assuming an approval by the EMEA, in the second half of calendar 2005 and that Chugai Pharmaceutical Co., Ltd., Roche’s subsidiary in Japan, will pursue a timely registration for NSCLC in Japan.

We have responsibility for manufacturing and supply of Tarceva^TM in the United States. We, along with our supply chain of third-party providers, have established sufficient inventory to be ready to launch the product within five business days of receipt of FDA approval. While Genentech has primary responsibility for commercialization of Tarceva^TM in the United States and Roche has responsibility for commercialization outside of the United States, we are actively engaged in supporting the activities of our partners in order to ensure readiness for a timely market launch of Tarceva^TM. We will also co-promote the product in the United States and field at least 25% of the combined U.S. sales force.

Our strategy to rapidly grow and expand Tarceva^TM post-launch has three main themes:



	•	Expand the use of Tarceva^TM to earlier stage NSCLC patients, both in the front-line and adjuvant settings. We believe that the survival data from our BR.21 study is similar to the survival data from the recent Phase III study comparing the cytotoxic chemotherapy agents, Taxotere® and Alimta®, and that the side-effect profile for Tarceva^TM is considerably more benign than these cytotoxic chemotherapy agents. We have, therefore, filed for approval to use Tarceva^TM in NSCLC patients who have failed at least one prior regimen of chemotherapy, and we believe that Tarceva^TM will be used in second and third-line patients. To expand on this, we intend to move directly to registration-oriented studies in both the front-line and adjuvant settings. The first part of this program has been initiated with an ongoing randomized Phase II trial evaluating monotherapy Tarceva^TM against chemotherapy in patients who have received no prior chemotherapy and have poor performance status.


	•	Expand the use of Tarceva^TM to other oncology disease settings. Tarceva^TM is designed to target the HER1/ EGFR signaling pathway. The EGFR gene itself is known to be over-expressed, mutated or amplified in a significant portion of the approximately 1.3 million new cases of cancer diagnosed each year in the United States. The positive results from our pancreatic cancer Phase III trial are an endorsement of our belief that Tarceva^TM will have broad utility in a wide variety of disease settings beyond NSCLC. To date, tumor responses have been documented in Phase II studies for monotherapy Tarceva^TM in bronchioalveolar cell carcinoma, glioblastoma multiforme, head and neck cancer, hepatocellular carcinoma, breast cancer and ovarian cancer. We will continue programs to broaden the use of Tarceva^TM to additional indications.

	•	Develop all-targeted therapy combinations of Tarceva^TM with other targeted agents, particularly the anti-angiogenic antibody Avastin®. Targeted therapies like Tarceva^TM and Avastin® are designed to deliver a treatment benefit without the significant toxicities evident with the use of cytotoxic chemotherapy agents. We believe that combinations of these novel targeted therapies have the potential to deliver enhanced efficacy without the severe cumulative toxicities associated with widely used combinations of cytotoxic chemotherapy. This program is already underway and has revealed promising indications of activity in single arm Phase II studies, using combinations of Tarceva^TM and Avastin® in renal cell carcinoma and in NSCLC patients with adenocarcinoma. We will also explore combinations of Tarceva^TM with other oral targeted therapies that offer patients the important benefits of ease-of-use and convenience in addition to the potential for enhanced efficacy with more benign side-effects.

To pursue these strategic themes, we will continue to execute, together with our alliance partners, Genentech and Roche, a development program comprised of both registration oriented and publication studies designed to expand the indications and market potential for Tarceva^TM.

Establish our position as a premier oncology franchise. Over the past several years we have assembled through acquisitions and internal investment the necessary components of a high quality oncology franchise operating from drug discovery through commercialization. A key goal for us is to raise

awareness of the value of our franchise by advancing our pipeline, validating our research and exploiting our broad capabilities to effectively pursue in-licensing, acquisition and partnering opportunities. We have a diverse pipeline of oncology drug candidates which includes signal transduction inhibitors, apoptosis inducers and a next-generation cytotoxic agent. We plan to add to our pipeline both from our internal research efforts and through third-party transactions, such as partnering, in-licensing and product acquisitions. We intend to continue to focus our research efforts on two main areas of cancer drug discovery, namely the discovery of modulators of signal transduction pathways that either (i) drive cancer cell proliferation or (ii) prevent apoptosis in cancer cells. In order to effectively manage the risks inherent in biotechnology research and development and to complement our internal research efforts, we believe that we must continue to aggressively manage our pipeline and actively explore in-licensing and acquisition initiatives. We expect these activities to add to our pipeline of oncology products and clinical candidates and strengthen our growing presence in oncology.

We believe that we are also well positioned to enhance our commercial presence by competing for co-promotion and product acquisition opportunities. Our marketed products, Novantrone® and Gelclair®, have allowed us to establish a core commercialization group, which includes approximately 50 sales representatives and managers. We believe that our commercial group and research and development capabilities make us a highly attractive partner in oncology.

Diversify our business through continued investment in Prosidion. We believe that expansion into a second disease area is an important part of our strategy for long term value creation, and will ultimately be important in establishing ourselves as a premier biotechnology organization. We, therefore, consider it important that we continue to invest in our majority-owned diabetes and obesity subsidiary, Prosidion, based in Oxford, United Kingdom. To this end, we recently announced the acquisition by Prosidion of the type 2 diabetes clinical candidate PSN9301 and its associated intellectual property estate from Probiodrug AG. PSN9301 is an oral, small molecule inhibitor of DP-IV, which is recognized as an important target in diabetes. In addition to composition of matter claims for PSN9301, the acquired intellectual property estate includes issued U.S. methods-of-use claims, covering the inhibition of DP-IV as a target in diabetes that have been non-exclusively licensed to Novartis and Merck, among others, for milestones and royalties. Prosidion also anticipates initiating clinical trials for two diabetes candidates, PSN105, a glucokinase activator, and PSN357, a glycogen phosphorylase inhibitor, in the first half of calendar 2005.

We believe that we have the integrated capabilities and the strength and depth of management to accomplish our goal of establishing ourselves as a premier biotechnology company. In the past several years, we have demonstrated a proven capability to execute complex, large scale development and registration programs, manage intricate partnerships, execute transactions, integrate acquisitions, and build new capabilities, such as our commercial group, in a manner that positions us well for future growth.

Tarceva^TM was discovered jointly by us and Pfizer in the course of a long-standing discovery collaboration between us and Pfizer that terminated in 2001. We gained full development and marketing rights to Tarceva^TM in June 2000, when the U.S. Federal Trade Commission (FTC) ordered Pfizer to divest its rights to Tarceva^TM to us as a result of an antitrust finding upon the FTC’s review of Pfizer’s merger with the Warner-Lambert Company.

Tarceva^TM is an oral, once-a-day, small molecule drug designed to inhibit the receptor tyrosine kinase activity of the product of the HER1/ EGFR gene. HER1/ EGFR is a key component of the HER signaling pathway, which plays a role in the regulation of growth in many normal cells. EGFR inhibitors were designed to arrest the growth of tumors (cytostasis); however, under certain circumstances EGFR inhibition can lead to apoptosis which in turn would result in tumor shrinkage. The HER1/ EGFR gene is over-expressed, mutated or amplified in approximately 40% to 60% of all cancers and contributes to the abnormal growth signaling in these cancer cells. A frequently occurring mutation of the HER1/ EGFR gene called EGFRVIII is also found in many tumors, including glioblastoma multiforme and NSCLC, and recently publications in scientific literature have correlated tumor response in lung cancer patients treated with EGFR inhibitors with additional newly identified mutations. There is a strong scientific rationale and

a substantial potential market for EGFR inhibitors. We believe that Tarceva^TM is likely to have utility in many oncology disease settings. However, the initial focus of the program has been on NSCLC and pancreatic cancer since effective treatment of both of these forms of cancer remains a major unmet need. According to the American Cancer Society, lung cancer is the leading cause of cancer-related deaths in the United States each year with an estimated 160,000 deaths in 2004. Patients with NSCLC account for approximately 80% of these deaths. The American Cancer Society also estimates that 31,000 cancer patients in the United States will die from pancreatic cancer in 2004.

In order to help us accomplish the goals of a registration program focused on NSCLC and pancreatic cancer and to ensure the optimal competitive positioning of Tarceva^TM, we entered into a co-development and commercialization alliance with Genentech and Roche in January 2001. Since the inception of our alliance, we have implemented a global development strategy for Tarceva^TM with our partners. This strategy was a broad-based approach that implemented simultaneous clinical programs designed to result in a registration with the FDA. This plan included a single agent Phase III trial for second and third-line NSCLC patients as well as combination trials with existing cytotoxic chemotherapy regimens for front-line use in pancreatic cancer and NSCLC. Currently, there are approximately 100 investigator-sponsored studies and National Cancer Institute/ Cancer Therapy Evaluation Program Studies ongoing or planned in the Tarceva^TM program. These studies are exploring monotherapy and combination uses of Tarceva^TM in various tumor types, with a variety of treatment modalities, such as radiation, and in the adjuvant setting.

In April 2004, we announced that Tarceva^TM met all its key predetermined study endpoints (including overall survival, progression-free survival, and objective tumor response) in a 731-patient randomized, double-blind placebo controlled Phase III trial, known as the BR.21 study. Tarceva^TM also delayed the deterioration of selected lung cancer symptoms in the BR.21 study. The trial compared Tarceva^TM to placebo in the treatment of patients with advanced NSCLC who had previously received one or two prior chemotherapy regimens. Patients were randomized to the Tarceva^TM or placebo arm in a 2:1 ratio, respectively. Approximately 50% of the patients in the study had failed one prior regimen of chemotherapy and the other half had failed at least two prior regimens. In addition, a large proportion of the patients entered the study with poor performance status (25% with performance status 2 and 9% with performance status 3). This international study was conducted by the National Cancer Institute of Canada’s Clinical Trial Group in collaboration with our own clinical development team.

Patients receiving Tarceva^TM had a median survival of 6.7 months versus 4.7 months for patients receiving placebo, a 42% improvement. A hazard ratio of 0.73 and a p-value of <0.001 were determined for comparisons of overall survival, indicating a highly statistically significant treatment benefit. In addition, 31.2% of patients receiving Tarceva^TM in the study were alive at one year versus 21.5% in the placebo arm, a 45% improvement. The results of the study also revealed a treatment benefit in virtually all subsets of patients examined. A treatment effect (hazard ratio of <1) was seen in males, smokers and patients with squamous cell carcinoma histology (subsets that, consistent with previous studies with EGFR inhibitors, had a relatively low rate of tumor response in our study), as well as in females, non-smokers and patients with adenocarcinoma histology (subsets with higher rates of tumor response). Recent publications have shown a correlation of tumor response with a group of newly identified EGFR mutations which are clustered in patients who are non-smokers or have tumors with adenocarcinoma histology. The BR.21 study results clearly show that tumor response is not always a good surrogate for survival benefit and that the improvement in overall survival seen in our BR.21 study cannot be explained by the reported incidence (10%) of these mutations. The patient population was also unusual in that patients with ECOG performance status 3 were included in the study.

Summary Data for the Tarceva^TM Phase III Study in Second and Third-Line NSCLC

Note: The data is based on our statistical analysis of data from the BR.21 study.



(1)	Hazard ratio is a statistical measure of the difference in overall survival between the study drug group and the control group. A hazard ratio of less than 1 indicates a reduction in the risk of death; for example, a hazard ratio of 0.73 represents a 27% reduction in the risk of death and a 37% improvement in overall survival.


(2)	P-value is a statistical measure of significance. A p-value of <0.05 indicates a statistically significant difference.


(3)	Data is adjusted for stratification factors prior to randomization.



(4)	Objective tumor response rate represents the sum of the percentage of patients who exhibited a PR or a CR.

The safety profile observed in the BR.21 study was relatively benign compared to cytotoxic chemotherapy and was consistent with that seen in prior Tarceva^TM studies with 75% of patients receiving Tarceva^TM exhibiting rash versus 17% in the placebo group and 54% of patients receiving Tarceva^TM experiencing diarrhea versus 18% for placebo. Most of these events were mild to moderate. In this large placebo controlled study, severe pulmonary events, including potential cases of interstitial lung disease, were rare and generally equally distributed between the Tarceva^TM and placebo arms. We believe this combination of survival benefit with a relatively benign side-effect profile positions Tarceva^TM as an important treatment option for oncologists treating advanced lung cancer patients who have failed front-line chemotherapy. In addition to the survival benefit, Tarceva^TM also demonstrated in the BR.21 study statistically significant delays in the time-to-symptom deterioration of key lung cancer symptoms. Statistically significant improvements were observed for cough (adjusted p-value = 0.04), pain (adjusted p-value = 0.04) and dyspnea (adjusted p-value = 0.03) in the Tarceva^TM arm versus the placebo arm.

In September 2004, we announced that Tarceva^TM also met its primary endpoint of improving overall survival in a 569-patient randomized, double-blind placebo controlled Phase III trial in front-line pancreatic cancer patients with locally advanced or metastatic disease. This study was also conducted by the National Cancer Institute of Canada’s Clinical Trial Group in collaboration with our own clinical development team. The trial compared a combination of Tarceva^TM and the chemotherapy agent gemcitabine (the only recently approved agent for the treatment of pancreatic cancer) with gemcitabine plus placebo. The data demonstrates a 23.5% improvement in overall survival (a hazard ratio of 0.81 and a p-value of 0.025) for the Tarceva^TM arm compared to the placebo arm. Median and projected one-year survival in the Tarceva^TM plus gemcitabine arm were 6.4 months and 24%, respectively, compared to 5.9 months and 17%, respectively, in the gemcitabine plus placebo arm. A statistically significant improvement in progression-free survival was also observed. A preliminary analysis of the safety data did not reveal any unexpected safety signals beyond that seen in the prior use of Tarceva^TM in both monotherapy and combination settings. As expected, rash and diarrhea were the principal Tarceva^TM-related side effects in the study. The results were noteworthy in that Tarceva^TM was used at a lower dose (100 mg) for the majority of patients treated in the study (521 patients received 100 mg per day of Tarceva^TM or placebo and 48 patients received 150 mg per day of Tarceva^TM or placebo) and Tarceva^TM was used in combination with a chemotherapy agent. We believe that the data demonstrates that Tarceva^TM is likely to have broad utility beyond the initial lung cancer indication and will also have profound implications for our understanding of this new class of drugs.

Earlier Phase I and Phase II trials of Tarceva^TM in NSCLC, head and neck cancer and ovarian cancer demonstrated that the drug possessed activity as a single agent and was relatively well-tolerated with manageable side-effects, principally, a reversible rash and generally mild diarrhea. Tumor responses have been documented in Phase II studies for monotherapy Tarceva^TM in bronchioalveolar cell carcinoma, glioblastoma multiforme, breast cancer and hepatocellular carcinoma. Our Phase III clinical trial program also included two front-line Phase III combination trials in NSCLC that had been initiated for business reasons in order for us to be competitive with a similar clinical trial program with a competitor’s EGFR product, Iressa®, in this setting. On October 1, 2003, we announced that the two front-line Phase III studies of Tarceva^TM plus standard chemotherapy in metastatic NSCLC did not meet their primary endpoints of improving overall survival. These results were widely anticipated based on the competitor’s previously announced failure of Iressa® in this front-line setting in August 2002. This failure of combination trials of Tarceva^TM with chemotherapy in NSCLC suggested that the mechanism of action of Tarceva^TM may be antagonistic with that of certain cytotoxic chemotherapy regimens.

The market for NSCLC therapeutics is competitive with multiple treatment options available in the market today. These therapeutic options include cytotoxic chemotherapy agents, targeted therapeutics and radiation and surgery. These available options are often ineffective or have severe side-effects. For example, front-line combination cytotoxic chemotherapy treatment of NSCLC only provides a median survival of approximately ten months and is accompanied by severe side-effects. Patients subsequently also may be treated with cytotoxic chemotherapy agents, such as Taxotere® or Alimta®, in the second-line setting. While these agents have previously shown similar survival results to Tarceva^TM, they exhibit a more severe side-effect profile than those seen in the Tarceva^TM trial. Furthermore, only approximately 46,000 people a year suffering from NSCLC receive second and/or third-line treatments, usually with limited success, while many other patients either decline treatment or have treatment withheld due to the side-effects of cytotoxic chemotherapy or a perception that further treatment would be of limited benefit.

The table below summarizes the results announced within the last 18 months of three large, randomized, placebo controlled Phase III trials in advanced NSCLC, consisting of our BR.21 study, a study comparing Taxotere® to Alimta®, and our study comparing Taxotere® plus Aptosyn® to Taxotere® plus placebo. The table is presented for comparative purposes only and does not purport to represent all existing second and third-line NSCLC trials after failure of initial therapy.


(1)	OSI data from our Phase III study for Aptosyn® unblinded on June 11, 2004.

(2)	Tarceva^TM Median Survival in performance status 0-1: 8.2 months

Although Tarceva^TM has not been tested in a direct head-to-head study with either Taxotere® or Alimta®, the table shows that Tarceva^TM produced similar survival data to Taxotere® and Alimta® in this comparison of contemporary studies. This was despite the fact that Tarceva^TM was tested in a more advanced patient population, both in terms of number of prior regimens and overall performance status, which makes the ability to show a survival benefit in absolute terms more challenging. Tarceva^TM demonstrated statistically significant delays in the time-to-symptom deterioration of the key lung cancer symptoms of cough, dyspnea and pain in the BR.21 study, as well as avoiding the severe drug related toxicities associated with chemotherapy, such as neutropenia, thrombocytopenia and alopecia. We therefore believe that Tarceva^TM will provide oncologists with an attractive option in the treatment of NSCLC patients who have failed at least one prior chemotherapy regimen.

Patients may also be treated with another EGFR inhibitor, Iressa®, which was recently approved for the third-line setting based on response rate data only. Iressa® has yet to demonstrate a survival benefit. Recent scientific publications have also identified a subset of NSCLC patients, predominantly females with tumors that typically have adenocarcinoma histology, who possess mutations in their EGFR gene. This subset of patients has now been shown to be very responsive to treatment with EGFR targeted agents like Tarceva^TM and Iressa® which are both potent inhibitors of this mutated form of the receptor. Data suggests that patients with lung tumors possessing these mutations may constitute the majority of patients seen to have a tumor response when treated with these agents and some investigators have hypothesized that the clinical benefits observed for EGFR inhibitors may be restricted to patients whose tumors have these EGFR mutations. However, our BR.21 study clearly shows that tumor response is not always a good surrogate for survival benefit and that the improvement in overall survival cannot be explained by the reported incidence (approximately 10%) of these mutations. For example, patients treated with Tarceva^TM in our BR.21 study who had tumors with squamous cell carcinoma histology had only a 3.8% tumor response rate but a 49% improvement in survival (hazard ratio 0.67) whereas patients in the study with tumors of adenocarcinoma histology had a similar improvement in survival of 41% (hazard ratio 0.71) but a 13.9% response rate. Based on this data, we and others believe that a second group of patients that express other forms of EGFR will also benefit from EGFR treatment at higher doses of drug even if this is not manifested by high response rates. The pancreatic cancer trial supports this belief in that a survival benefit was seen in the study despite the fact that there was no difference in the percentage of patients achieving a partial response between the Tarceva^TM plus gemcitabine arm and the gemcitabine plus placebo

arm and that publications in the scientific literature indicate that the mutations may be largely confined to lung cancer. Based on published studies, Tarceva^TM at 150mg/day produces greater drug exposure in patients’ blood than 700mg/day of Iressa®, which is approved for use only at 250mg/day. We believe that dose and drug exposure will be important components to the activity of EGFR targeted agents for this larger second group of patients and that the ability to safely dose these agents at higher dose levels will be critical to their effective use for these patients. Tarceva^TM is currently the only EGFR inhibitor to have demonstrated a survival benefit in NSCLC patients at a dose accompanied by a relatively mild side-effect profile.

Antibody products are also under development which target the EGFR pathway and have demonstrated improved anti-cancer activity when used in conjunction with existing treatment and chemotherapy regimens. One of these, Erbitux^TM, has been approved for advanced colorectal cancer in the United States and is marketed by BMS and ImClone for this indication. Erbitux^TM also recently demonstrated a survival benefit in combination with radiotherapy in a Phase III study in the treatment of squamous cell carcinoma of the head and neck. However, to date clinical studies in lung cancer have yet to demonstrate robust activity for Erbitux^TM and it is significantly more expensive than the small molecule inhibitors. Furthermore, we believe these antibodies may be less likely than the tyrosine kinase inhibitors to effectively inhibit mutated forms of HER1/ EGFR. They also require delivery via intravenous infusion and are relatively difficult and expensive to produce. In contrast to these agents, we believe that small molecule inhibitors of the tyrosine kinase activity, such as Tarceva^TM, should be effective against either mutant or over-expressed forms of HER1/ EGFR, are convenient once-a-day oral therapies and are relatively easy and inexpensive to manufacture.

Tarceva^TM was granted a Fast Track designation from the FDA in September 2002 for the second and third-line NSCLC indication. In January 2004, we initiated the rolling submission of our NDA with the FDA for the use of Tarceva^TM in this indication. Following the release of the positive results of the BR.21 study, the FDA informed us that Tarceva^TM had been granted a priority review designation and also Pilot 1 status, a new FDA initiative aimed at reducing drug approval times for agents with Fast Track status. Tarceva^TM is among the first group of drugs to be accepted into the new Pilot program. The Pilot 1 program introduces a six-month review period for each submitted reviewable unit of the rolling NDA. Unlike the Fast Track designation which does not require the FDA to begin review of the sections at the time they are submitted, the Pilot 1 program requires the FDA to complete its review of each section within six months following the submission of such section, or from the date of granting of Pilot 1 status, whichever is later. On July 30, 2004, we completed our NDA filing with the FDA for monotherapy Tarceva^TM use in the treatment of all NSCLC patients who have failed at least one prior chemotherapy regimen. Based upon the priority review status, the FDA has until January 30, 2005 (otherwise known as the PDUFA date) to take action on the NDA filing. We believe that the FDA is rapidly proceeding with its review of the NDA, and we anticipate receipt of approval from the FDA for the NSCLC indication in advance of the PDUFA date. We believe that Tarceva^TM will be the first EGFR inhibitor to receive full approval from the FDA. We anticipate launching Tarceva^TM within five business days of receipt of approval and our partner, Genentech, will finalize and announce pricing for the United States at the time of launch. The NDA was prepared in the International Committee of Harmonization-approved common technical dossier format which can facilitate registration in the EU and Japan. Roche filed the EU application under the centralized process for the NSCLC indication in the third quarter of calendar 2004 and we anticipate that Roche will launch Tarceva^TM in the NSCLC indication in the EU, assuming an approval by the EMEA, in the second half of calendar 2005 and that Chugai, Roche’s subsidiary in Japan, will pursue a timely registration in Japan. We plan to discuss the submission of an sNDA for the pancreatic cancer indication with the FDA in the near future.

Novantrone®. We market and promote Novantrone® for approved oncology indications in the United States and receive commissions from Serono on net oncology sales in this market. Novantrone® is an anthracenedione used as an intravenous chemotherapy agent. Novantrone® is approved by the FDA for the

treatment of acute non-lymphocytic leukemia (ANLL), which includes myelogenous, promyelocytic, monocytic and erythroid acute leukemias, and the relief of pain associated with advanced hormone refractory prostate cancer (HRPC). Novantrone® is also used extensively by oncologists for the treatment of non-Hodgkin’s lymphoma which is not an approved indication in the United States. The drug is also approved for certain advanced forms of multiple sclerosis, a key strategic area for Serono. The drug was licensed by Serono from Amgen, Inc., and we signed a co-promotion agreement with Serono to market the drug for its cancer indications in March 2003. Serono is continuing to market Novantrone® for the multiple sclerosis indication and records all U.S. sales in all indications. The patent for Novantrone® expires in April 2006. We estimate that we will receive in excess of $33 million in sales commission revenue for Novantrone® in fiscal year 2004.

Gelclair®. We market and distribute Gelclair® in the oncology setting in the United States. Gelclair® was cleared for sale as a device by the FDA in 2002. Gelclair® is a bioadherent oral gel that provides relief for pain associated with oral lesions, including oral mucositis, a debilitating side-effect often seen in cancer patients undergoing radiation treatment or chemotherapy. An estimated 320,000 cancer patients undergoing chemotherapy or radiotherapy develop oral mucositis every year.

On October 15, 2004, we announced that it would be necessary to record an impairment charge related to the intangible asset for exclusive distribution rights to Gelclair®. In performing our recoverability test, we determined that the total of the expected future undiscounted cash flows directly related to the Gelclair® asset was less than the carrying value of the Gelclair® asset. As a result, we determined that an impairment charge of approximately $24.6 million, which represented the full-unamortized balance of the Gelclair® intangible asset, was necessary and will be recorded during the quarter ended September 30, 2004.

The impairment charge is non-cash and will not result in future cash expenditures. The impairment charge resulted from both the recent discontinuance of discussions with a replacement dental partner, and slower than originally expected sales growth in the oncology marketplace following the re-launch of the product in October 2003. In addition to the impairment charge related to the Gelclair® intangible asset, we also expect to record a provision for obsolete inventory, for both inventories on hand and additional committed inventory purchases, that we consider to be in excess of forecasted future demand based on the expiration dates of the product. We are currently in the process of quantifying the impact of the potential inventory write off, which we expect to be material.

We have established a core commercial group of approximately 80 people, which includes approximately 50 sales representatives and managers covering the major territories in the United States. All of our sales representatives have considerable experience in the pharmaceutical industry, and most have ample experience with oncology products. We intend to market all future products directly in the United States but we may partner with other pharmaceutical companies to support our products in territories outside of the United States.

On August 5, 2004, we announced to our employees a proposal to consolidate our U.K. based oncology research and development activities into our New York locations by November 30, 2004. This decision was based on the need to prioritize the expansion of our commercial operation infrastructure and increase our level of investment in both translational research and our diabetes and obesity subsidiary, Prosidion.

This consolidation will primarily affect our Oxford facility where the consolidation may result in the reduction of our work force by approximately 90 employees. Following the consolidation, the only operations remaining at the Oxford facility will be those related to our international clinical trials group and our diabetes and obesity subsidiary, Prosidion. We currently estimate annual savings of $10 million to $15 million in operating expenses as a result of this consolidation.

Cancer remains a major healthcare concern with approximately 1.3 million Americans diagnosed with various solid tumors, lymphomas and leukemias every year. In total, it is estimated that the overall direct medical costs for cancer in the United States for 2002 were in excess of $60 billion. The worldwide total cancer market has been estimated to be $14 to $15 billion and is expected to grow as new products, which offer safer and more effective treatment options, based upon an improved understanding of the genetic basis of human cancer, begin to enter the market. Traditionally, development of anti-cancer drugs has resulted in products which generally kill rapidly dividing cells. Although these products, called cytotoxic drugs, are effective in killing rapidly dividing cancer cells, they usually interfere directly and non-selectively with normal processes in the cell associated with DNA replication and cell division. Since these cell division processes occur routinely in healthy tissues, the cytotoxic drugs are limited in their utility by their serious side-effects, such as disruption of the blood, immune and gastrointestinal systems.

Our approach to cancer therapy includes three diversified areas in an attempt to improve the available drug treatment options for cancer patients: signal transduction inhibitors that target either aberrant cancer cell growth or the induction of apoptosis in cancer cells that no longer respond to these tightly regulated processes, and next-generation cytotoxics. The aberrant cancer cell growth and apoptosis inducer programs are targeted therapy approaches focused on the exploitation of our rapidly growing understanding of the genetic basis for cancer in order to develop drugs that directly target the genetic abnormalities present in human cancers or treat their consequences. These areas are the focus of our pre-clinical research efforts. As these new targeted therapies emerge in clinical testing, they may be used independently, in combination with other targeted drugs or in combination with cytotoxic chemotherapy drugs, in an attempt to maximize the anti-cancer benefit by using so-called drug cocktails.

We have also previously acquired drug candidates in the cytotoxic area where we believe these new cytotoxic agents represent improvements in activity or technological innovations over existing drugs. For example, OSI-7904L is a liposomal formulation of a thymidylate synthase inhibitor (TSI), a known class of cytotoxic agents. Liposomal formulations are technological innovations that are designed either to improve targeting of the cytotoxic agent to the tumor or to change the exposure profile of the drug molecule, thus improving the therapeutic index, the drug benefits versus its toxic side-effects.

Our drug discovery efforts in targeted therapies were conducted in partnership with Pfizer from 1986 to 2001. Tarceva^TM was jointly discovered as part of this alliance. Pfizer is continuing to develop three other clinical stage targeted therapies from this alliance, the funded discovery phase of which concluded in April 2001. If Pfizer is successful in commercializing any of these drug candidates, we will receive a royalty from Pfizer on the sales of these drugs. If Pfizer chooses to discontinue development of any of these drug candidates, we will have the right to pursue development of them.

The chart below summarizes our marketed products and the status of our research and development pipeline, including Tarceva^TM.


(1)	Being developed by Pfizer for which we are entitled to royalties if commercialized and to which we have development rights if Pfizer chooses to halt development for any reason.

OSI-7904L. OSI-7904L is a liposomal formulation of the TSI, which was licensed from GlaxoSmithKline and acquired by us as part of the acquisition of Gilead’s oncology business in 2001. TSIs are a class of cytotoxic chemotherapy agents. The leading TSI used today is 5-FU, a generically available TSI which is extensively used in many tumor types, notably colorectal cancer. Clinical studies with 5-FU have shown that long, continuous infusions of the drug have shown better activity than more typical intravenous bolus infusion regimens. The goal of our OSI-7904L program is to mimic this effect with a single short infusion of the liposomal formulation of our TSI molecule. OSI-7904L was formulated in liposomes with a goal of extending its pharmacokinetic, or drug exposure, half-life and improving its therapeutic index. We are developing OSI-7904L primarily for gastrointestinal tract cancers. OSI-7904L

demonstrated promising activity in pre-clinical testing for the potential treatment of various solid tumors. Data from the Phase I study indicated that the liposomal formulation has extended the drug exposure profile of OSI-7904L in patients’ blood. Our Phase II single-agent study of OSI-7904L for gastric and gastroesophageal junction cancer in patients who have received no prior chemotherapy has moved forward to its second stage. The study, having met the initial requirement of achieving at least three objective responses in the first 18 evaluable patients has completed accrual of 53 patients in Europe and the United States. The primary endpoint of this trial is response rate. This open-label, non-randomized study was initiated in October 2003. We also broadened the OSI-7904L development program by initiating two additional studies evaluating the use of OSI-7904L in combination with the chemotherapy agents cisplatin and oxaliplatin commonly used in the treatment of gastrointestinal malignancies. Milestone and royalty payments are due to GlaxoSmithKline upon successful development of this product.

OSI-930. OSI-930 is a tyrosine kinase inhibitor that acts as a potent co-inhibitor of the receptor tyrosine kinases c-kit and VEGFR, and is designed to target both cancer cell proliferation and blood vessel growth (angiogenesis) in selected tumors. It is the first de novo development candidate to emerge from our discovery research operation since we realigned the focus of our core business towards oncology research in the fall of 2002. We have continued to advance OSI-930 through pre-clinical development and anticipate initiating a clinical program in the first half of calendar 2005. The mutated Kit receptor is directly involved in tumor progression in the majority of gastrointestinal stromal tumors and certain leukemias, and over-expressed normal Kit is thought to play a role in small cell lung cancer. The inhibition of the tyrosine kinase activity of Kit is expected to result in reduced cancer cell proliferation and increased cellular apoptosis in tumor types driven by Kit, thus resulting in inhibition of tumor growth. In addition to inhibiting Kit activity, OSI-930 is also capable of inhibiting the receptor tyrosine kinase called KDR. KDR, or vascular endothelial growth factor receptor-2 (VEGFR-2), is present on endothelial cells and is the key mediator of blood vessel growth in response to the angiogenic growth factor VEGF. The pathway is believed to be the single most important mechanism for recruitment of new blood vessels in nearly all solid tumors; hence, inhibition of this pathway should impact the growth and metastases of a wide range of angiogenesis-dependent malignancies. While the combination of Kit and KDR inhibition would be expected to offer the greatest therapeutic benefit to patients bearing Kit expressing solid tumors, the KDR component is considered an attractive target for all solid tumors. At this year’s American Association of Cancer Research meeting we presented data showing that OSI-930 analogs were able to inhibit the growth of human tumor xenografts. We believe that OSI-930 may represent a less promiscuous split kinase inhibitor when compared to other known inhibitors of Kit/ KDR driven tumors and other promiscuous kinase inhibitors in development (such as Pfizer’s SU11248).

SAANDs Platform. The SAANDs platform that we acquired from Cell Pathways in June 2003 consisted of two clinical candidates, Aptosyn® and OSI-461, designed to induce apoptosis through the sustained activation of protein kinase G, and associated intellectual property. Aptosyn®, the prototype product was originally developed by Cell Pathways to treat familial adenomatous polyposis. Following rejection of an NDA by the FDA in September 2000, Cell Pathways initiated a large scale Phase III program comparing a combination of Aptosyn® and Taxotere® versus Taxotere® and placebo in second-line NSCLC. The 610-patient trial was based primarily on promising pre-clinical data in orthotopic rat models, and on June 11, 2004, we announced that this Phase III trial failed to meet its primary endpoint of improving patient survival. Survival in the Taxotere® plus Aptosyn® arm (median survival = 6.9 months; one-year survival rate = 30.7%) was essentially indistinguishable from survival in the Taxotere® plus placebo arm (median survival = 6.9 months; one-year survival rate = 29.5%). Since our acquisition of Cell Pathways in June 2003, we have anticipated a negative outcome for this trial due to the absence of clinical data for the compound in advanced NSCLC. To date there have been some indications of activity for Aptosyn® in pre-cancerous settings; however, we do not believe Aptosyn® will have utility in advanced cancers and have halted development of the agent as an anti-cancer drug. Like Aptosyn®, OSI-461 is an inhibitor of cGMP phosphodiesterases which leads to sustained activation of the intracellular signaling protein, protein kinase G, and subsequent stimulation of apoptosis through the c-jun kinase pathway. OSI-461 also has effects on tubulin and microtubular biology in cells which is a known mechanism of action for other anti-cancer agents. OSI-461 is more potent than Aptosyn® in in-vitro assays but has proven to be

difficult to dose to therapeutic levels in animal models. Following the completion of the acquisition of Cell Pathways in June 2003, we completed a detailed review of the OSI-461 program and concluded that further dose optimization studies would be required before committing to a full Phase II program. In February 2004, we expanded an ongoing Phase I dose escalating and pharmacokinetic trial of OSI-461 in patients with advanced solid tumors. This study has been amended to allow us to explore the possibility that administering OSI-461 with food may increase drug exposure levels achievable in humans following oral dosing of OSI-461. Data from this study is expected in the first quarter of calendar 2005. In July 2002, Cell Pathways also commenced a Phase II trial of OSI-461 in the non-cancerous area of inflammatory bowel disease at doses we consider to be sub-optimal. We recently received inconclusive results from this study which we need to further analyze in order to determine how best to proceed with this compound in this indication.

In fiscal 2003, we refocused our pre-clinical research efforts into two areas in which we believe we can build a competitive presence in cancer drug discovery. These areas relate to the discovery of targeted therapies focused on two core biological processes important in both normal and cancer cell regulation, namely signal transduction pathways that either (i) drive cancer cell proliferation or (ii) prevent apoptosis in cancer cells. The dysfunctional regulation of these two processes is a key element in the progression of normal cells to the cancerous state. Within these areas, we have focused our efforts on three key signaling axes described by the central signal transduction gene products that make up these pathways. These pathways are thought to be critical in driving either cancer cell proliferation or in protecting cancer cells from undergoing apoptosis.

More advanced discovery projects include co-inhibitors of PDK and EGFR and those targeting IGF-1R inhibitors. Recent studies have indicated that NSCLC patients who respond to EGFR inhibitors show both loss of EGFR phosphorylation as well as loss of protein kinase B (PKB) phosphorylation. PKB is another important regulatory protein on the EGFR signaling axis. In contrast, NSCLC patients who are not responsive to EGFR inhibitors, show a loss of EGFR phosphorylation, but do not show the same degree of reduction in PKB phosphorylation to that observed in the responding patients. This data suggests that the PKB pathway is activated in an EGFR independent fashion in the non-responding NSCLC patients. Similar data has been observed in NSCLC cell lines as well as in other tumor types. We believe the PKB activation is dependent upon phosphorylation by phosphoinositide-dependent kinase-1 (PDK-1). PDK-1 is a serine/threonine kinase ubiquitously expressed in human tissues. We have postulated that a dual PDK-1/ EGFR inhibitor would be expected to have a broader range of antitumor activity in NSCLC patients when compared with a selective EGFR inhibitor, such as Tarceva^TM, and have an ongoing drug discovery project designed to identify a dual inhibitor of PDK-1 and EGFR.

IGF-1R is a receptor tyrosine kinase that stimulates proliferation, enables oncogenic transformation, and suppresses apoptosis. It is an excellent strategic fit within our core strategy. Inhibitors of IGF-1R are expected to have broad utility in oncology since the over-expression of IGF-IR and/or its ligands (IGF-I and IGF-II) or the down-regulation of ligand binding proteins (IGFBP) occurs in numerous human malignancies including lung, colon, breast, prostate, brain and skin cancers. Correlations with increased risk and poor prognosis have been established. In addition, signaling through the IGF system has also been implicated in protecting tumor cells from apoptosis induced by a number of anti-cancer treatments such as EGFR inhibitors (e.g., Tarceva^TM), the anti-HER2/erbB2 antibody Herceptin® and cytotoxic agents. To be competitive in the IGF-1R space, we have initiated a research collaboration that uses structure based design as an enabling technology to identify proprietary IGF-1R inhibitors for us. The collaboration has significantly improved our understanding of the structure activity relationship within our lead chemical series and we anticipate identifying a development candidate in the first quarter of calendar 2005. We believe that an IGF-1R inhibitor should be useful both as a single agent, and in the potentiation of other molecularly targeted therapeutic agents and cytotoxic agents.

Our approach to discovering drugs is focused on the discovery and development of small molecule pharmaceutical products that, typically, would be taken either orally by a patient as a pill, capsule or suspension or intravenously as is common for many cancer products. Our drug discovery platform

constitutes an integrated set of technologies and capabilities covering every major aspect of pre-clinical research and pre-clinical and clinical development. We have built a fully-integrated drug discovery platform in order to accelerate the process of identifying and optimizing high-quality, small molecule drug candidates. Our core technologies and capabilities include (i) signal transduction, protein kinases, gene transcription and other assay systems, (ii) automated high throughput screening, (iii) an extensive library of proprietary small molecule compounds, (iv) medicinal and automated combinatorial chemistry, (v) in vivo pharmacology, pharmacokinetics and pharmaceutical development capabilities and (vi) core clinical project management and regulatory affairs units. We currently employ approximately 160 scientists in our pre-clinical research activities.

In order to be competitive in the oncology drug discovery field we have endeavored to build a network of alliances designed to fill gaps in our expertise as well as enhance our existing understanding of the drug discovery process. This is based upon a philosophy that we should access complementary technology and expertise on the outside where and when this is appropriate. We have made external investments in three key areas: (i) target identification and validation, (ii) lead optimization, and (iii) surrogate biomarkers translational research.

Early in fiscal 2003 we established a research collaboration with Cold Spring Harbor Laboratories (CSHL) to utilize RNA interference to identify and validate new drug targets. CSHL has a patent application covering the use of viral delivery of RNA hairpins into cells. An extensive gene library has been generated with the expectation that the size of this library will cover the entire genome. The entire gene library will be screened against genetically defined cancer cell lines and will identify those specific gene products that may modulate apoptosis and/or drive tumor cell proliferation. Additional research within OSI will help define potential drug targets.

We also initiated a research collaboration with EiRX Therapeutics of Cork, Ireland (EiRX) in June 2003. Using appropriate growth factors and reagents, EiRX is able to manipulate and test the apoptosis process in primary cells associated with normal and disease conditions. EiRX has been able to demonstrate that several of the genes they have identified are present in cancer cells, and are functional in blocking cell death. We entered into our agreement with EiRX to evaluate a set of 12 genes, whose use as cancer targets is covered within a filed patent application. We have selected four new targets for entry into our drug discovery efforts.

We have approached lead optimization in two ways. First, we have enhanced the quality of our chemical library through two major deals in the last year. We have acquired 130,000 new compounds from Array Biopharma, Inc. which are of high purity, are well characterized and are considered highly “drug-like.” Second, we became a subscriber to a library of structures available from Albany Molecular Research, Inc. representing a variety of chemical building blocks. These transactions have improved the diversity of our chemical library and, as a result, we are confident that our high throughput screening efforts are more likely to identify promising drug-like leads for our ongoing project teams.

We have also enhanced our capabilities in lead optimization through investments in structure based drug design. We have an ongoing alliance with Structural Genomix, Inc. (SGX), in which SGX will provide and will continue to provide crystal structures of up to 12 of our drug targets over a collaboration period ending February 1, 2005, unless extended. SGX has received milestone payments for successfully co-crystallizing target with inhibitor. To date, we have made significant progress in a number of our projects and we view this collaboration as being a successful entry for OSI into the field of structure based drug design.

The optimal clinical development of novel targeted therapies requires assays and reagents (surrogate markers) that are capable of identifying whether a specific patient tumor will respond to the therapy under investigation. Going forward, we intend to invest in this area in order to:


	•	identify surrogate biomarkers that predict cellular sensitivity and tumor response to OSI compounds of interest; and

	•	determine the pathways that are influenced/impacted through RNA interference of select OSI drug targets.

To ensure that our lead compounds are active against the target of interest, they are profiled in pharmacodynamic assays. This allows us to develop surrogate biomarkers of drug activity in vivo and will enable and support our future clinical development. This process ensures that our lead compounds have retained their anticipated mechanism of action in vivo.

Translational research is designed to bridge our research knowledge base into the clinic and the marketplace. The current emphasis of our translational research programs is on Tarceva^TM and a series of collaborations and studies are ongoing (including, for example, our Phase II dose-to-rash study in NSCLC) designed to improve our understanding of how best to use Tarceva^TM clinically.

After identifying a suitable drug candidate, a molecule is advanced toward clinical trials and enters the investigational new drug (IND) track phase, in which toxicological, scale-up synthesis and clinical development strategy are addressed. The IND track phase typically takes up to one year. An IND application is reviewed by the FDA or its foreign equivalent prior to the commencement of clinical studies. A drug is typically first assessed for its safety and pharmacokinetic properties. After these Phase I trials, drugs are tested for preliminary efficacy in Phase II trials to demonstrate initial activity and confirm safety in humans prior to the initiation of extensive Phase III trials designed to collect the safety and efficacy data necessary to support a filing of an NDA with the FDA or similar marketing application authorization overseas. We currently employ over 165 physicians, scientists and clinical operations specialists who are responsible for generating pre-clinical data required for IND submission and managing clinical trials and the associated regulatory affairs effort to support submissions and interactions with the FDA and other regulatory agencies around the world.

The entire drug discovery and development process typically takes over a decade and is subject to significant risk and attrition. A significant majority of drug candidates which enter clinical trials fail to result in a successful product. We have, therefore, adopted a research strategy that manages a portfolio of product opportunities, adding, through in-licensing, lead compounds at various stages of the process in order to help mitigate the risks inherent in these efforts.

We currently rely on third-party manufacturers to manufacture all of our marketed products and late stage product candidates. Under our collaboration agreement with Genentech, we are responsible for the manufacture and supply of erlotinib, the API, and Tarceva^TM tablets for pre-clinical and clinical trials and for the supply of commercial quantities of Tarceva^TM tablets for sales within the United States. We entered into a Manufacturing and Supply Agreement with Genentech, effective as of June 4, 2004, to define each party’s rights and responsibilities in connection with such supply. Under our collaboration agreement with Roche, Roche has elected to take responsibility for the supply of Tarceva^TM tablets for sales outside of the United States.

Erlotinib is manufactured in a three-step process with high yield. Sumitomo Chemical Co., Ltd. (formerly known as Sumika Fine Chemicals Co. Ltd.) and Dinamite Dipharma S.p.A are our manufacturers of erlotinib used for commercial supplies. Both of these manufacturers have manufactured API for Tarceva^TM clinical trials. We contracted with Schwarz Pharma AG to manufacture Tarceva^TM tablets and placebo product for clinical supplies and are engaging Schwarz to manufacture Tarceva^TM tablets for commercial supplies. We are also evaluating the capability of another manufacturer to serve as

an alternative provider of Tarceva^TM tablets. Clinical supplies of Tarceva^TM tablets are currently stored, labeled, packaged and distributed by Cardinal Health Clinical Services, and we are engaging Cardinal Health Packaging Services for labeling and secondary packaging services for commercial supplies of Tarceva^TM tablets before their subsequent distribution to Genentech or a storage facility designated by Genentech. We expect to enter into long term supply agreements with our API and tablet manufacturers. All manufacturers of erlotinib and Tarceva^TM tablets are required to comply with current good manufacturing practices. We have produced sufficient quantities of Tarceva^TM tablets to conduct our ongoing clinical trials, and we have a supply chain in place with inventory on hand to support launch.

In connection with our acquisition of Cell Pathways in June 2003, we acquired the exclusive marketing and distribution rights to Gelclair® in North America. The manufacturing rights and obligations were held by Sinclair and subsequently licensed to Helsinn in July 2003.

In connection with our purchase of certain oncology assets from Gilead in December 2001, we entered into a manufacturing agreement covering products acquired from Gilead. During the one-year transition period, Gilead manufactured and supplied us with the API for preparation of OSI-7904L drug product. We have transitioned the manufacture of the API to new manufacturers. Gilead will produce for us liposomal formulations of OSI-7904L at its manufacturing facility in San Dimas, California to support our ongoing clinical trial activities.

In 2002, we made the decision to focus our initial research and development efforts into the oncology area and to either shut down or divest research programs in other areas with the exception of our research assets in the diabetes and obesity arena. These programs were transferred to a newly created U.K. based subsidiary, Prosidion, in January 2003. Prosidion is registered in England and Wales under registered number 4600121 and is our majority-owned subsidiary.

As part of the formation of Prosidion, we transferred and/or licensed our diabetes assets to the subsidiary and we transferred to Prosidion our employees who were focusing on diabetes research. In 2003, we hired Anker Lundemose, M.D., Ph.D. as Chief Executive Officer of Prosidion, established a separate board of directors and committed approximately $10 million in funding to support Prosidion’s operations. In August 2003, Prosidion completed a transaction with H. Lundbeck A/ S’s subsidiary, Synaptic Pharmaceutical Corporation, to license co-exclusive rights to a novel G-protein coupled receptor (GPCR) target for the treatment of diabetes and obesity. In March 2004, Prosidion signed an agreement with Tanabe in order to obtain U.S. and European rights to the OSI/ Tanabe joint technology in glucokinase activators and to gain freedom to operate in all other areas previously covered by the OSI/ Tanabe alliance. Both Tanabe and Synaptic are minority shareholders of Prosidion as a result of these transactions.

On July 26, 2004, Prosidion acquired a platform of DP-IV technology from Probiodrug AG for approximately $35 million in cash plus future milestones. The milestone payments are payable upon the successful development of PSN9301, the lead DP-IV inhibitor acquired from Probiodrug, which is in Phase II clinical trials for the treatment of type 2 diabetes. These milestone payments are payable in January 2007 and/or January 2010 unless the development of PSN9301 is terminated prior to such dates. Probiodrug AG, based in Halle (Saale), Germany, had pioneered much of the research and development that has led to the characterization of DP-IV as one of the most important targets in diabetes drug development today. Included in the assets acquired by Prosidion is a portfolio of medical use patents around the target. This portfolio includes issued and pending patents with claims covering DP-IV as a target, and use of combinations of DP-IV with other oral anti-diabetes drugs such as Metformin. Merck, Novartis and Ferring B.V. are all non-exclusive licensees to the estate. Future potential milestones and royalties arising from this intellectual property have been transferred to Prosidion. Prosidion also entered into a research agreement with Probiodrug under which Prosidion will provide funding of certain research aimed at discovering backup compounds to PSN9301 and discovering the potential therapeutic use of the related target glucose-dependent insulinotropic peptide receptor.

In April 2004, we announced that our board of directors approved an investment of up to an additional $40 million in Prosidion. The first installment of $10 million was invested at a cost of $10 per

share bringing our ownership position in the subsidiary on a fully diluted shares outstanding basis to 81%. In order to finance the Probiodrug AG transaction and fund the current ongoing research and development efforts at Prosidion, we increased our investment commitment to $60 million in Prosidion. Following the closing of the Probiodrug acquisition and our increased investment in Prosidion to finance the transaction, our ownership of Prosidion increased to approximately 96%.

DP-IV cleaves and inactivates GLP-1, an important mediator of blood glucose levels. Inhibition of DP-IV leads to prolonged GLP-1 activity and inhibitors of DP-IV have demonstrated significant effects on mean blood glucose and post-prandial blood glucose and HbA_1C levels, a reference marker widely used in the monitoring of diabetes patients, in pre-clinical and clinical trials. The field is competitive with the Novartis DP-IV inhibitor, LAF237, currently entering Phase III trials and compounds from Merck and Bristol-Myers Squibb in Phase II trials. PSN9301 is an orally active, competitive inhibitor of DP-IV that is designed as a shorter acting inhibitor. PSN9301 has been shown to lower glucose in type 2 diabetics in early clinical trials.

In addition to PSN9301, Prosidion has two molecules in the late stages of pre-clinical development which are scheduled to enter clinical trials in the first half of calendar 2005. PSN105 is an oral, small molecule activator of glucokinase and PSN357 is an oral, small molecule inhibitor of glycogen phosphorylase. Both glucokinase and glycogen phosphorylase are targets for therapeutic intervention in diabetes. The molecules have demonstrated efficacy in animal models and successfully completed acute toxicological profiling.

On January 8, 2001, we entered into an alliance with Genentech and Roche for the global co-development and commercialization of Tarceva^TM. We received upfront fees of $25 million related to this alliance, and Genentech and Roche each purchased $35 million of our common stock at $75.66 per share. We are also entitled to up to $92 million upon the achievement of certain milestones under the terms of the alliance. We have entered into separate agreements with both Genentech and Roche with respect to the alliance, as well as a Tripartite Agreement.

Under the Tripartite Agreement, we agreed with Genentech and Roche to optimize the use of each party’s resources to develop Tarceva^TM in certain countries around the world and share certain global development costs on an equal basis; to share information generated under a global development plan; to facilitate attainment of necessary regulatory approval of Tarceva^TM for commercial marketing and sale in the world; and to work together on such matters as the parties agree from time to time during the development of Tarceva^TM. We, as well as Genentech and Roche, may conduct clinical and pre-clinical activities for additional indications for Tarceva^TM not called for under the global development plan, subject to certain conditions. The Tripartite Agreement will terminate when either the OSI/ Genentech collaboration agreement or the OSI/ Roche agreement terminates.

Under the OSI/ Genentech collaboration agreement, we agreed to collaborate in the product development of Tarceva^TM with the goals of obtaining regulatory approval for commercial marketing and sale in the United States of products resulting from the collaboration, and, subsequently, supporting the commercialization of the product. Consistent with the development plan and with the approval of a joint steering committee, we agree with Genentech as to who will own and be responsible for the filing of drug approval applications with the FDA other than the first NDA, which we own and are responsible for filing, and the first supplemental NDA, which we have the option to own and be responsible for filing. Genentech has primary responsibility for the design and implementation of all product launch activities and the promotion, marketing and sales of all products resulting from the collaboration in the United States, its territories and Puerto Rico.

We have certain co-promotion rights under the OSI/ Genentech collaboration agreement which were defined in an amendment to the agreement effective as of June 4, 2004. Pursuant to this amendment, we will co-promote Tarceva^TM using a sales force that will be equal to or greater than 25% of the combined OSI/ Genentech sales force. We will share equally in the operating profits or losses on products resulting from the collaboration. Under the OSI/ Genentech collaboration agreement, we granted to Genentech a

royalty-free non-transferable (except under certain circumstances), non-sublicensable (except under certain circumstances), co-exclusive license under our patents and know-how related to Tarceva^TM to use, sell, offer for sale and import products resulting from the collaboration in the United States, its territories and Puerto Rico. In addition, Genentech granted to us a royalty-free non-transferable (except under certain circumstances), non-sublicensable (except under certain circumstances), co-exclusive license to certain patents and know-how held by Genentech to use, make, have made, sell, offer for sale and import products resulting from the collaboration in the United States, its territories and Puerto Rico. We have primary responsibility for patent filings for the base patents protecting Tarceva^TM and, in addition, we have the right, but not the obligation, to institute, prosecute and control patent infringement claims relating to the base patents.

In connection with our collaboration with Genentech, Genentech will recognize all U.S. sales of Tarceva^TM and associated costs and expenses. We will recognize revenues from our alliance with Genentech, which will consist of our share of the pretax profits (loss) generated from the sales of Tarceva^TM in the United States. We also will recognize manufacturing revenue from the sale of inventory to Genentech for commercial sales of Tarceva^TM in the United States, reimbursement from Genentech of our Tarceva^TM-related sales force expenses. We will receive royalties on sales of Tarceva^TM outside of the United States by Roche and up to $92 million in non-refundable milestone payments upon the achievement of certain milestones relating to regulatory submissions and approval, certain of which have already been received. We expect to recognize revenue upon the achievements of each of these related milestones.

The OSI/ Genentech collaboration agreement continues until the date on which neither we nor Genentech are entitled to receive a share of the operating profits or losses on any products resulting from the collaboration, that is, until the date that we and Genentech mutually agree to terminate the collaboration or until either party exercises its early termination rights. The OSI/ Genentech collaboration agreement is subject to early termination in the event of certain customary defaults, such as material breach of the agreement and bankruptcy. The provisions of the amendment allowing us to co-promote are also subject to termination by Genentech upon a material breach by us of the amendment which remains uncured or upon a pattern of nonmaterial breaches which remains uncured. In addition, since January 8, 2003, Genentech has had the right to terminate the OSI/ Genentech collaboration agreement with six months’ prior written notice.

Effective June 4, 2004, we entered into a Manufacturing and Supply Agreement with Genentech that defined each party’s responsibilities with respect to the manufacture and supply of clinical and commercial quantities of Tarceva^TM. Under certain circumstances, if we fail to supply such clinical and commercial quantities, Genentech has the right, but not the obligation, to assume responsibility for such supply. The Manufacturing and Supply Agreement will terminate upon the termination of the OSI/ Genentech collaboration agreement.

Under the OSI/ Roche agreement, we granted to Roche a license to our intellectual property rights with respect to Tarceva^TM. Roche is collaborating with us and Genentech in the product development of Tarceva^TM and is responsible for future marketing and commercialization of Tarceva^TM outside of the United States in certain territories as defined in the agreement. The grant is a royalty-bearing, non-transferable (except under certain circumstances), non-sublicensable (except under certain circumstances), sole and exclusive license to use, sell, offer for sale and import products resulting from the development of Tarceva^TM worldwide, other than the territories covered by the OSI/ Genentech collaboration agreement. In addition, Roche has the right, which it has exercised, to manufacture commercial supplies of Tarceva^TM for its territory, subject to certain exceptions. Roche will pay us certain milestone payments and royalty payments on sales of products resulting from the collaboration. We have primary responsibility for patent filings for the base patents protecting Tarceva^TM and, in addition, we have the right, but not the obligation, to institute, prosecute and control patent infringement claims relating to the base patents. The OSI/Roche agreement continues until the date on which we are no longer entitled to receive a royalty on products resulting from the development of Tarceva^TM, that is, until the date of expiration or revocation or complete rejection of the last to expire patent covering Tarceva^TM or, in countries where there is no valid patent covering Tarceva^TM, on the tenth anniversary of the first commercial sale of Tarceva^TM in that country, or

until either party exercises early termination rights. The OSI/Roche agreement is subject to early termination in the event of certain customary defaults, such as material breach of the agreement and bankruptcy. In addition, since July 31, 2003, Roche has had the right to terminate the agreement on a country-by-country basis with six months’ prior written notice. Since such time, we also have had the right to terminate the agreement on a county-by-country basis if Roche has not launched or marketed a product in such country under certain circumstances.

Pfizer is continuing to develop three drug candidates which arose from our collaborative drug discovery program in targeted therapies for cancer, all of which are in clinical trials. These programs are focused on developing drugs which are orally available, potent inhibitors of key protein tyrosine kinase receptors involved in signal transduction and angiogenesis. Angiogenesis is the process of blood vessel growth and is induced by solid tumors which require nutrients that enable growth. We believe that the ability to safely and effectively inhibit this process represents an intriguing opportunity in cancer drug development. Under our alliance with Pfizer, we discovered two compounds in this area. CP-547,632 targets VEGFR and is in Phase II and Phase I trials, and CP-868,596 targets PDGFR and is in Phase I trials. An additional candidate from the Pfizer program, CP-724,714, is a potent and selective small molecule inhibitor of the HER2/erbB2 receptor tyrosine kinase, and is in Phase I clinical trials. Over-expression of HER2/erbB2 oncogenes has been demonstrated to correlate with aggressive cancer growth particularly in metastatic breast cancer. Approximately 25% to 30% of all women with metastatic breast cancer over-express HER2/erbB2.

Patents and other proprietary rights are vital to our business. Our policy is to protect our intellectual property rights through a variety of means, including applying for patents in the United States and other major industrialized countries. We also rely upon trade secrets and improvements, unpatented proprietary know-how and continuing technological innovations to develop and maintain our competitive position. In this regard, we seek restrictions in our agreements with third-parties, including research institutions, with respect to the use and disclosure of our proprietary technology. We also enter into confidentiality agreements with our employees, consultants and scientific advisors.

Patents issued in the United States, Europe, Japan, and 20 other countries, cover composition of matter for the Tarceva^TM compound itself, processes for its preparation, and pharmaceutical compositions containing Tarceva^TM. These patents expire in 2015. Patent applications are being pursued, seeking protection in and outside the United States, for polymorphic, anhydrous, hydrate, and certain salt forms of Tarceva^TM, as well as for processes and important intermediate chemicals in the manufacture of Tarceva^TM. The polymorphic patent, when it issues, should provide patent exclusivity for Tarceva^TM through 2020. Further, patent protection for methods of use of Tarceva^TM are being sought.

Patents directed to the OSI-7904L compound have issued in the United States, Japan, and Europe. Patent protection is also being sought, with a composition of matter and method-of-use patent issuing earlier this year in the United States, for liposomal formulations of OSI-7904L. From our acquisition of Cell Pathways, we have a patent estate to methods of identifying compounds that participate in a specific apoptotic pathway. Aptosyn® and OSI-461 are each covered by U.S. and foreign patents to methods of treatment using the compounds. We have rights to a U.S. patent covering a method-of-use of Novantrone® for solid tumors, expiring April 11, 2006. A patent application has been allowed in the United States covering the composition of Gelclair®.

We have assembled a strong gene transcription patent portfolio and have a non-exclusive out-licensing program for our gene transcription patent estate. Currently, we have licensed this technology to Aurora Biosciences Corporation (assigned to Vertex Pharmaceuticals Incorporated), Pharmacia Corp., R.W. Johnson Pharmaceutical Research Institute, Wyeth, BASF Corporation and Merck. We also have non-exclusive licenses from Cadus (seven U.S. patents and additional U.S. and foreign applications) and Wyeth (four U.S. patents and additional foreign applications) to a portfolio of patents and applications

Prosidion has acquired rights to U.S. patents and foreign patents and patent applications covering methods of treatment of diabetes through the administration of a DP-IV inhibitor, as well as patents and patent applications covering composition of matter for specific small molecule DP-IV inhibitors including the clinical candidate PSN9301, which is in Phase II clinical trials, as part of its asset acquisition from Probiodrug AG. Merck and Novartis have taken licenses to some of these method-of-use patents and patent applications, from which Prosidion may accrue certain milestone payments and royalties. One of these patents has been challenged in Europe and, in May 2004, in a ruling by the European Patent Office in Munich, one of the DP-IV method-of-use patents was revoked. Prosidion is appealing this decision.

The pharmaceutical and biotechnology industries are intensely competitive. We face, and will continue to face, intense competition from organizations such as large pharmaceutical companies, biotechnology companies and academic and research institutions. We face significant competition from fully-integrated pharmaceutical companies, as well as numerous smaller companies, which possess extensive drug discovery programs in oncology and diabetes, are pursuing the same or similar technologies as those that comprise our technology platforms and are pursuing pharmaceutical products or therapies that are directly competitive with ours, including developing novel small molecule pharmaceuticals. Most of the major pharmaceutical organizations competing with us have greater capital resources and larger overall research and development staff and facilities than we do in drug research and development, regulatory affairs and pharmaceutical product manufacturing and marketing.

With respect to our cancer drug discovery and development programs, other companies have potential drugs in clinical trials to treat diseases in the same areas for which we are seeking to discover and develop drug candidates. These competing drug candidates may be further advanced in clinical development than our potential products within our small molecule programs, and may result in effective, commercially successful products. At least three competitors, AstraZeneca, BMS/ ImClone/ Merck KGaA and Amgen/ Abgenix, Inc., also have substantial clinical development programs for the same target as our flagship product, Tarceva^TM. AstraZeneca has received approval for its anti-EGFR, small molecule drug in the United States (accelerated approval, contingent studies required by FDA), Switzerland, Japan and other territories. However, AstraZeneca’s drug has not demonstrated a survival benefit. BMS/ImClone/Merck KGaA’s anti-EGFR antibody, Erbitux™, has been approved (as single agent or in combination with irinotecan) in the United States and in the EU and Switzerland for the treatment of metastatic colon carcinoma in patients who are refractory or have developed intolerance to irinotecan based chemotherapy. In addition to agents that target EGFR, Taxotere® and Alimta® are chemotherapy agents that are approved for use in second-line NSCLC. Taxotere® and Alimta® are manufactured and distributed by Aventis and Eli Lilly and Company, respectively.

With the acquisition of the co-promotion rights for Novantrone® in the oncology arena in the United States and the marketing and distribution rights for Gelclair® in North America, we are facing competition in their respective areas of use. Novantrone® is mainly used for the treatment of pain associated with advanced HRPC, ANLL and non-Hodgkin’s lymphoma which is not an approved indication in the United States. A key competitor in HRPC is Taxotere®, which the FDA approved for use in combination with prednisone in May 2004 for HRPC based on a survival benefit for Taxotere®-based regimens compared to the better tolerated Novantrone® plus prednisone. In ANLL, Novantrone® competes against a variety of generic products including idarubicin and daunorubicin. In non-Hodgkin’s lymphoma, the reference standard is the CHOP regimen consisting of four generic agents: cyclophosphamide, doxorubicin, vincristine and prednisone. In addition, Rituxan®, marketed by Genentech/ IDEC Pharmaceuticals, Inc., is used extensively in non-Hodgkin’s lymphoma both as a single-agent and in combination with CHOP or other chemotherapies including Novantrone®. Gelclair® is a bioadherent oral gel for the relief of pain associated with oral mucositis. Key competitors include a myriad of products often blended in the dispensing pharmacy, none of which have been specifically approved for this indication, such as Xylocaine®, Benadryl® Elixir®, Carafate®, Orabase®, and over-the-counter and prescription analgesics.

OSI-7904L, our next-generation cytotoxic drug candidate, is designed to improve upon products of similar mechanism already in the market and available generically. We must therefore clearly differentiate the activity or safety of our molecule if we are to successfully register this drug and compete in the marketplace. OSI-7904L, which is currently in Phase II trials, is a TSI designed to compete with generic 5-FU, as well as Xeloda® by Roche, an oral TSI.

Companies with related research and development activities also present significant competition for us. Research efforts with respect to gene sequencing and mapping are identifying new and possibly superior target genes than our target genes. In addition, alternative drug discovery strategies, such as monoclonal antibodies, may prove more effective than those pursued by us. Furthermore, competitors may have access to more diverse compounds than we do for testing by virtue of larger compound libraries.

With respect to Prosidion’s drug discovery and development programs for diabetes and obesity, other companies have potential drugs in clinical trials to treat diseases in the same areas for which we are seeking to discover and develop drug candidates. These competing drug candidates may be further advanced in clinical development than our potential products within our small molecule programs, and may result in effective, commercially successful products. At least two competitors, Merck and Novartis, have advanced clinical development programs directly competitive with Prosidion’s lead clinical candidate, PSN9301, an inhibitor of DP-IV for treating diabetes which is in Phase II clinical trials. DP-IV inhibitors are designed to prevent cleaving and inactivation of GLP-1, an important mediator of blood glucose levels. There are no DP-IV inhibitors that are currently marketed, although Amylin Pharmaceuticals, Inc. has filed an application with the FDA to market a GLP-1 product, Exenatide^TM, for the treatment of diabetes. Prosidion must therefore clearly distinguish the profile of PSN9301 if we are to successfully register this product and compete in the marketplace. Prosidion’s other clinical candidates for diabetes, a glucokinase activator, PSN105, and a glycogen phosphorylase inhibitor, PSN357, are currently in late pre-clinical development and anticipated to enter clinical development in the first half of calendar 2005. At least two competitors, Roche and Pfizer, have, at various times, announced similar research and development activities.

We believe that our ability to compete successfully will be based upon, among other things, our ability to create and maintain scientifically advanced technology, attract and retain scientific and clinical personnel possessing a broad range of expertise, obtain patent protection or otherwise develop and protect proprietary products or processes, compete for premium in-licensing products, conduct clinical trials, obtain required government approvals on a timely basis and commercialize our products on a profitable basis.

We and our collaborative partners are subject to, and any potential products discovered and developed by us must comply with, comprehensive regulation by the FDA in the United States and by comparable authorities in other countries. These national agencies and other federal, state, and local entities regulate, among other things, the pre-clinical and clinical testing, safety, effectiveness, approval, manufacture, quality, labeling, distribution, marketing, export, storage, record keeping, advertising and promotion of pharmaceutical and diagnostic products.

The process required by the FDA before pharmaceutical products may be approved for marketing in the United States generally involves:


	•	pre-clinical laboratory and animal tests;

	•	submission to the FDA of an IND application, which must be in effect before clinical trials may begin;

	•	adequate and well controlled human clinical trials to establish the safety and efficacy of the drug for its intended indication(s);

	•	FDA compliance inspection and/or clearance of all manufacturers;

	•	submission to the FDA of an NDA; and


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

Pre-clinical tests include laboratory evaluation of product chemistry and formulation, as well as animal studies, to assess the potential safety and efficacy of the product. Certain pre-clinical tests must comply with FDA regulations regarding current good laboratory practices. The results of the pre-clinical tests are submitted to the FDA as part of an IND, to support human clinical trials and are reviewed by the FDA, with patient safety as the primary objective, prior to the IND commencement of human clinical trials.

Clinical trials are conducted according to protocols that detail matters such as a description of the condition to be treated, the objectives of the study, a description of the patient population eligible for the study and the parameters to be used to monitor safety and efficacy. Each protocol must be submitted to the FDA as part of the IND. Protocols must be conducted in accordance with FDA regulations concerning good clinical practices to ensure the quality and integrity of clinical trial results and data. Failure to adhere to good clinical practices and the protocols may result in FDA rejection of clinical trial results and data, and may delay or prevent the FDA from approving the drug for commercial use.

Clinical trials are typically conducted in three sequential Phases, which may overlap. During Phase I, when the drug is initially given to human subjects, the product is tested for safety, dosage tolerance, absorption, distribution, metabolism and excretion. Phase I studies are often conducted with healthy volunteers depending on the drug being tested; however, in oncology, our primary area of focus, Phase I trials are more often conducted in cancer patients. Phase II involves studies in a limited patient population, typically patients with the conditions needing treatment, to:


	•	evaluate preliminarily the efficacy of the product for specific, targeted indications;

	•	determine dosage tolerance and optimal dosage; and

	•	identify possible adverse effects and safety risks.

Pivotal or Phase III adequate and well-controlled trials are undertaken in order to evaluate efficacy and safety in a comprehensive fashion within an expanded patient population for the purpose of registering the new drug. The FDA may suspend or terminate clinical trials at any point in this process if it concludes that patients are being exposed to an unacceptable health risk or if they decide it is unethical to continue the study. Results of pre-clinical and clinical trials must be summarized in comprehensive reports for the FDA. In addition, the results of Phase III studies are often subject to vigorous statistical analysis. This data may be presented in accordance with the guidelines for the International Committee of Harmonization that can facilitate registration in the United States, the EU and Japan.

FDA approval of our own and our collaborators’ products is required before the products may be commercialized in the United States. FDA approval of an NDA will be based, among other factors, on the comprehensive reporting of clinical data, risk/benefit analysis, animal studies and manufacturing processes and facilities. The process of obtaining NDA approvals from the FDA can be costly and time consuming and may be affected by unanticipated delays.

There is an NDA review process referred to as the Fast Track program which is designed to expedite the approval of new drugs that are intended to treat serious or life-threatening conditions and that demonstrate the potential to address unmet medical needs. Applicants that meet the relevant acceptance criteria may receive Fast Track designation. The benefits of a Fast Track designation include the ability to schedule meetings to seek FDA input into development plans, the option of submitting an NDA in sections on a rolling basis rather than submitting all components simultaneously, and the option of requesting evaluation of studies using surrogate endpoints (accelerated approval). Tarceva^TM was granted a Fast Track designation for the advanced NSCLC indication.

In addition, the FDA has recently instituted a new trial program, the Pilot 1 program, which provides for the review of a limited number of pre-submitted portions, or reviewable units, of an applicant’s marketing application before submitting the complete application. The Pilot 1 program only applies to certain new drug or biological products that have received Fast Track designation, have been the subject of an end-of-Phase II and/or a pre-NDA meeting, and have demonstrated significant promise as a

therapeutic advance in clinical trials. As part of the Pilot 1 program, the FDA agrees to complete reviews of an applicant’s reviewable units within six months of the date of the submission of such reviewable unit or from granting of Pilot 1 status, whichever is later, and to provide early feedback in the form of information requests and discipline review letters. Tarceva^TM was granted Pilot 1 status and is one of the first drugs to be entered into the program.

Among the conditions for NDA approval is the requirement that the prospective manufacturer’s procedures conform to current good manufacturing practices, which must be followed at all times. In complying with this requirement, manufacturers, including a drug sponsor’s third-party contract manufacturers, must continue to expend time, money and effort in the area of production, quality assurance and quality control to ensure compliance. Domestic manufacturing establishments are subject to periodic inspections by the FDA in order to assess, among other things, compliance with current good manufacturing practices. To supply products for use in the United States, foreign manufacturing establishments also must comply with current good manufacturing practices and are subject to periodic inspection by the FDA or by regulatory authorities in certain countries under reciprocal agreements with the FDA.

Both before and after market approval is obtained, a product, its manufacturer and the holder of the NDA for the product are subject to comprehensive regulatory oversight. Violations of regulatory requirements at any stage, including after approval, may result in various adverse consequences, including the FDA’s delay in approving or refusal to approve a product, withdrawal of an approved product from the market and the imposition of criminal penalties against the manufacturer and NDA holder. In addition, later discovery of previously unknown problems may result in restrictions on the product, manufacturer or NDA holder, including withdrawal of the product from the market. Furthermore, new government requirements may be established that could delay or prevent regulatory approval of our products under development.

In addition to regulations enforced by the FDA, we must also comply with regulations under the Occupational Safety and Health Act, the Environmental Protection Act, the Toxic Substances Control Act, the Resource Conservation and Recovery Act and other federal, state and local regulations. For example, sales, marketing and scientific/educational grant programs must comply with the Medicare-Medicaid Anti-Fraud and Abuse Act, as amended, the False Claims Act, also as amended, the privacy provisions of the Health Insurance Portability and Accountability Act (HIPAA) and similar state laws. Pricing and rebate programs must comply with the Medicaid rebate requirements of the Omnibus Budget Reconciliation Act of 1990, as amended, and the Medicare Prescription Drug Improvement and Modernization Act of 2003. If products are made available to authorized users of the Federal Supply Schedule of the General Services administration, additional laws and requirements may apply. All of these activities are also potentially subject to federal and state consumer protection and unfair competition laws. In addition, our research and development activities involve the controlled use of hazardous materials, chemicals and various radioactive compounds the handling and disposal of which are governed by various state and federal regulations.

In addition to regulations in the United States, we are subject to various foreign regulations governing clinical trials and the commercial sales and distribution of our products, including Tarceva^TM. We must obtain approval of a product by the comparable regulatory authorities of foreign countries before we can commence clinical trials or marketing of the product in those countries. The requirements governing the conduct of clinical trials, product licensing, pricing and reimbursement and the regulatory approval process all vary greatly from country to country. Additionally, the time it takes to complete the approval process in foreign countries may be longer or shorter than that required for FDA approval. Foreign regulatory approvals of our products are necessary whether or not we obtain FDA approval for such products. Finally, before a new drug may be exported from the United States, it must either be approved for marketing in the United States or meet the requirements of exportation of an unapproved drug under Section 802 of the Export Reform and Enhancement Act or comply with FDA regulations pertaining to INDs.

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

We believe that our success is largely dependant upon our ability to attract and retain qualified personnel in the scientific and technical fields. As of November 3, 2004, we, including Prosidion, have 504 employees worldwide, 363 of whom are in the United States. Of the 504 employees, 156 are primarily involved in research activities, 149 are primarily involved in development activities, 81 are primarily involved in sales and marketing, with the remainder engaged in executive and administrative capacities. Although we believe that we are appropriately sized to focus on our mission, we intend to add personnel with specialized expertise, as needed.

We believe that we have been successful to date in attracting skilled and experienced scientific and business professionals. We consider our employee relations to be good. However, competition for personnel is intense and we cannot assure that we will continue to be able to attract and retain personnel of high caliber.

We currently lease three facilities in New York, one located at 58 South Service Road, Melville, New York, consisting of approximately 37,000 square feet; one located at One Bioscience Park Drive, Farmingdale, New York, consisting of approximately 53,000 square feet; and one located at 106 Charles Lindbergh Boulevard, Uniondale, New York, consisting of approximately 30,000 square feet. The Melville facility houses our principal executive, commercial, finance, legal and administrative offices. The Farmingdale facility houses our U.S. drug discovery and pre-clinical laboratories. In August 2003, we consolidated our employees from the Uniondale facility into the Farmingdale facility. We are currently in the process of attempting to either assign or sublease our rights under the lease agreement for the Uniondale facility.

We currently lease three facilities in Boulder, Colorado, one located at 2860 Wilderness Place, consisting of approximately 60,000 square feet; one located at 2900 Center Green Court South, consisting of approximately 10,000 square feet; and one located at 2970 Wilderness Place, consisting of approximately 26,000 square feet. The Boulder facilities house our clinical research, regulatory affairs and drug development personnel. We are currently in the process of attempting to either assign or sublease our rights under the lease agreement for the facility at 2900 Center Green Court South.

In June 2003, in connection with our acquisition of Cell Pathways, we acquired a lease to a facility in Horsham, Pennsylvania, consisting of approximately 40,000 square feet. In May 2004, we subleased our rights under the lease agreement for this facility.

Our wholly owned subsidiary, OSI Pharmaceuticals (UK) Limited, currently leases two facilities, one located at Windrush Court, Watlington Road, Oxford, England, consisting of approximately 88,000 square feet, and another located at Isis House, Watlington Road, Oxford, England, consisting of approximately 34,000 square feet. The Oxford facilities house research and development laboratories and administrative offices as well as our Prosidion subsidiary. Following the consolidation of our U.K. based oncology research and development activities into our New York locations, the only operations remaining at the Oxford facility will be those related to our international clinical trials group and Prosidion.

Set forth below is a biographical description of each executive officer and director based on information supplied by such executive officer and director:

Robert A. Ingram was appointed Chairman of our Board effective January 1, 2003. Mr. Ingram serves as Vice Chairman Pharmaceuticals at GlaxoSmithKline, and previously served as the Chief Operating Officer and President of Pharmaceutical Operations. He began his career in the pharmaceutical industry as a sales representative for the company that would later become Merrell Dow Pharmaceuticals. He advanced rapidly through sales management at Merrell Dow and into government and public affairs. He left Merrell Dow in 1985 as Vice President of Public Affairs to become Vice President of Government Affairs at Merck. In 1988, he was promoted to President of Merck Frosst Canada, Inc. In 1990, Mr. Ingram left Merck to join Glaxo Inc., Glaxo plc’s U.S. subsidiary, as Executive Vice President of Administrative and Regulatory Affairs and assumed a series of increasingly responsible positions, including Group Vice President. He was named Executive Vice President in January 1993, President and Chief Operating Officer in June 1993, President and Chief Executive Officer in March 1994, and Chairman in January 1999. As an Executive Director of Glaxo Wellcome plc, Mr. Ingram held responsibility for operations in North America and Latin America. He was appointed to the global company’s board in May 1995. In October 1997, he became Chief Executive of Glaxo Wellcome with responsibility for worldwide business operations, and added the position of Chairman to his responsibilities. Mr. Ingram graduated from Eastern Illinois University with a B.S. degree in Business Administration. He serves on the Board of Directors of the Wachovia Corporation, Nortel Networks, Lowe’s Companies, Edwards Lifesciences Corporation, Valeant Pharmaceuticals International, and Misys plc (non-executive Director). In addition to his professional responsibilities, Mr. Ingram was asked by former U.S. President George H. Bush to form and chair the CEO Roundtable on Cancer. He also currently serves as Chairman of the Board of Trustees of the American Cancer Society Foundation, and is a member of numerous other civic and professional organizations. Mr. Ingram is also a frequent speaker at industry, pharmacy and government seminars.

Colin Goddard, Ph.D., was appointed our Chief Executive Officer in October 1998. He also served as Chairman of our Board from August 2000 to January 2003. He served as our President from September 1997 to September 2000; Executive Vice President and Chief Operating Officer from September 1996 to September 1997; Vice President, Research Operations from April 1995 to September 1996; Vice President, Research Operations, Pharmaceutical Division from December 1993 to April 1995; Director, Pharmaceuti-

cal Operations from April 1993 to December 1993; Director, Drug Discovery from April 1992 to April 1993; and Program Manager, Drug Discovery from April 1991 to April 1992. Dr. Goddard joined us as a scientist in January 1989. Dr. Goddard was instrumental in the development of our oncology franchise and fully integrated drug discovery and development operations and has led our corporate development, acquisition and financing efforts over the last seven years. Before joining us, Dr. Goddard spent four years at the National Cancer Institute in Bethesda, Maryland. Dr. Goddard is a member of the American Association of Cancer Research. Dr. Goddard trained as a cancer pharmacologist in Birmingham, U.K. receiving his Ph.D. from the University of Aston, Birmingham, U.K. in September 1985. Dr. Goddard has been our director since October 1998.

Gabriel Leung was appointed our Executive Vice President and President, Oncology Business in May 2003. Prior to joining us, Mr. Leung was Group Vice President of Global Prescription Business at Pharmacia Corp. and was a member of the CEO’s Operating Committee from May 2001 to April 2003. He headed Pharmacia’s Global Oncology Franchise where his responsibilities included medical affairs, marketing and sales worldwide in over 80 countries. Mr. Leung also co-chaired the Oncology Development Committee, which oversaw all the oncology R&D projects and portfolio strategies. Prior to Pharmacia, Mr. Leung was at Bristol-Myers Squibb where he led the growth of Taxol® and Paraplatin® into the then first and second chemotherapeutic agents in the United States. Mr. Leung is a pharmacist, trained at the University of Texas at Austin where he earned his B.S. degree with High Honors. He attended graduate school at the University of Wisconsin-Madison where he earned his M.S. degree in Pharmacy, with concentration in pharmaceutical marketing. Mr. Leung is an active member of C-Change, a national initiative chaired by former U.S. President George H. Bush and Mrs. Barbara Bush with the goal of reducing cancer mortality and incidence in the United States. Under this initiative, Mr. Leung co-chairs a special task force to design a new R&D paradigm to help expedite oncology drug discovery and development.

Nicole Onetto, M.D., was appointed our Executive Vice President and Chief Medical Officer, Oncology in May 2003. She served as Executive Vice President, Oncology from January 2002 to April 2003. Dr. Onetto served approximately two years with Gilead as Senior Vice President Medical Affairs from November 2000 to December 2001 and as Vice President Medical Affairs from July 1999 to November 2000. Prior to the merger of Gilead and NeXstar Pharmaceuticals, Inc., Dr. Onetto was Vice President Medical Affairs for NeXstar. At NeXstar, she was responsible for the strategy/implementation and coordination of all clinical trials worldwide bringing many potential products into development. From January 1995 to May 1997, she served as Senior Director Medical Affairs for the European oncology division of Bristol Myers Squibb. During her European assignment at Bristol Myers Squibb, she was responsible for all the coordination of all European clinical trials from Phase I to Phase IV for all oncology products. From July 1991 to January 1995, Dr. Onetto was Director, Clinical Cancer Research for Bristol Myers Squibb based in the United States, and she was the International Project Leader for Taxol®. She was responsible for the filing of the initial NDA for Taxol® and several supplemental NDAs. Before this appointment, she held positions at Immunex Research and Development Corp. and Hoechst Canada, Inc. Dr. Onetto studied at the University of Paris where she received her Bachelor’s degree. She holds a Doctorate in medicine and is qualified in pediatrics and hematology oncology and has a Master of Science from the University of Montreal.

Robert L. Van Nostrand was appointed our Vice President and Chief Financial Officer in December 1996, having previously served as Vice President, Finance and Administration. Mr. Van Nostrand has served as our Treasurer since March 1992 and Secretary from March 1995 to January 2004. Mr. Van Nostrand joined us as Controller and Chief Accounting Officer in September 1986. Mr. Van Nostrand currently serves as a Board member of the New York Biotechnology Association. Prior to joining OSI, Mr. Van Nostrand was in a managerial position with the accounting firm of Touche Ross & Co. (now Deloitte & Touche). Mr. Van Nostrand holds a B.S. degree in Accounting from Long Island University, New York, and he completed advanced management studies at the Wharton School, Philadelphia, Pennsylvania. He is a Certified Public Accountant.

Neil Gibson, Ph.D., was appointed our Vice President, Research in October 2002. Prior to this, Dr. Gibson served as our Vice President of U.S. Research from August 2001 to October 2002, and as Senior Director of Cancer Discovery from January 2001 to August 2001. Prior to joining us, Dr. Gibson served as Director of Cancer Research at Bayer in West Haven, Connecticut from May 1997 until January 2001. Prior to May 1997, Dr. Gibson served as a Senior Research Investigator in Pfizer’s cancer discovery group. Dr. Gibson enjoyed a successful academic career in cancer research, holding various positions at the University of Southern California, the AMC Cancer Research Center in Denver, Colorado, Fox Chase Cancer Center in Philadelphia, Pennsylvania and the National Cancer Institute in Bethesda, Maryland. Dr. Gibson has served on the National Cancer Institute’s Experimental Therapeutics Study Section and has been actively involved with the American Association of Cancer Research. Dr. Gibson received his Ph.D. in cancer pharmacology from the University of Aston in Birmingham, U.K.

Robert L. Simon was appointed our Vice President, Global Regulatory Affairs and CMC in January 2002. Mr. Simon served with Gilead as Vice President Global Regulatory Affairs from July 2000 to December 2001. Mr. Simon served as Vice President Worldwide Regulatory Affairs at Bristol Myers Squibb from November 1997 to July 2000. At Bristol Myers Squibb he was responsible for all CMC regulatory activities worldwide for both marketed products and new drug registration. From January 1987 to October 1997, Mr. Simon held various other regulatory affairs positions at Bristol Myers Squibb. He was responsible for the filings of numerous U.S. INDs, NDAs, AADAs and SNDAs, as well as a variety of international dossiers. During his career in regulatory affairs, he was instrumental in the creation of an electronic CMC dossier system capable of providing the CMC sections of worldwide registrational dossiers simultaneously. In addition, he successfully negotiated the approval of the CMC section of the only NDA that required an environmental impact statement. Among other achievements, Mr. Simon established, with the aid of the FDA, an efficient procedure for communication to help expedite the review and approval of the CMC sections of NDAs that are the subject of important new therapies. Mr. Simon holds a Bachelors degree in Chemistry from California State University and has had Executive Management training from the Levinson Institute. He also helped co-found the Regulatory Sciences Section of the American Association of Pharmaceuticals Scientists.

Barbara A. Wood, Esq., was appointed our Vice President and General Counsel in April 2001 and Secretary in January 2004. Prior to joining us, Ms. Wood was a partner at Squadron, Ellenoff, Plesent and Sheinfeld, LLP, a New York law firm which is now part of Hogan & Hartson, LLP, where she commenced her legal career in September 1987. While at Squadron, Ms. Wood specialized in mergers and acquisitions, licensing and securities law matters. She holds a Bachelor of Arts degree in classics and economics from Connecticut College and a law degree from Columbia Law School where she was a Harlan Fiske Stone Scholar.

Edwin A. Gee, Ph.D., our director since November 1985, served as President, Chairman of the Board, Director and Chief Executive Officer of International Paper Company from 1978 until his retirement in April 1985. Prior to 1978, Dr. Gee was a Senior Vice President, member of the Executive Committee and Director of E.I. du Pont de Nemours and Company. Dr. Gee served as a member of the Board of Directors of Biocryst Pharmaceuticals, Inc. until May 2004. Dr. Gee is also past Director of Salomon Brothers Fund, Inc., the Salomon Brothers Investors Fund, Inc. and the Salomon Brothers Capital Fund, Inc. He is also a past Director of Cambrex Corporation, Bethlehem Steel Corporation, American Home Products and Air Products and Chemicals. Dr. Gee served as our Executive Officer and was Chairman of our Board from April 1987 until March 1990. From March 1990 to December 1997, Dr. Gee was Chairman of the Board, but no longer served as one of our officers. Dr. Gee has served as a consultant to us since October 1996. As of December 1997, Dr. Gee was elected our Chairman Emeritus of the Board.

Michael G. Atieh joined our Board in June 2003 and was appointed Chairman of our Audit Committee in October 2003. Most recently, he was Group President of Dendrite International. From October 2000 to July 2001, he was Senior Vice President and Chief Financial Officer of Dendrite. Mr. Atieh began his career in July 1975 at Arthur Young & Company (now Ernst & Young). In July 1981, Mr. Atieh joined Merck where, from July 1981 to April 1994, he served in a variety of roles including Director of Accounting Standards, Director of Accounting, Director of Investor Relations, Vice

President, Government Relations, Treasurer, and Vice President, Public Affairs. From April 1994 to December 1998, Mr. Atieh was at the Merck-Medco Managed Care Division of Merck with his last position as Senior Vice President, Sales and Business Development. From January 1999 to October 2000, he was Vice President and General Manager-Medicare Business Initiative of Merck’s U.S. Human Health Division. Mr. Atieh is currently a member of the Board of Directors and the Audit Committee of ACE Limited.

G. Morgan Browne was Chief Financial Officer of CSHL from January 2001 until his retirement in December 2003 and was Administrative Director from June 1985 to December 2000. Prior to June 1985, Mr. Browne provided management services to a series of scientifically based companies, individually and as an Associate of Laurent Oppenheim Associates, Industrial Management Consultants. He was Chairman and Director of Specialty Composites Corp., Newark, Delaware and Vice President Finance and a Director of Lunn Industries, Inc., Wyandanch, New York. Mr. Browne is a graduate of Yale University. He is presently a Director of Harris & Harris Group, Inc. Mr. Browne currently serves on the committee which administers our 401(k) Savings and Investment Plan. Mr. Browne became our director in March 1993.

Daryl K. Granner, M.D., is a professor of Molecular Physiology and Biophysics and of Internal Medicine at Vanderbilt University. Dr. Granner served as Chairman of Molecular Physiology/ Biophysics at Vanderbilt University from July 1984 to August 1998. From July 1970 to June 1984, he was a professor of Internal Medicine and Biochemistry at the University of Iowa, where he directed the Division of Endocrinology and Diabetes and the Iowa Diabetes Center. Dr. Granner directs the Vanderbilt Diabetes Center and is an acknowledged authority in the mechanism of insulin action and the pathophysiology of diabetes mellitus. He has served on numerous national advisory panels. Dr. Granner also served as a scientific consultant to us from January 1992 to December 2002. Dr. Granner has been providing consulting services to OSI’s subsidiary, Prosidion, and, since August 2003, has been the Chairman of Prosidion’s Scientific Advisory Board. Dr. Granner serves on the board of directors of Prosidion and, pursuant to his grant of founder shares, is a shareholder of Prosidion. Dr. Granner has been our director since September 1996.

Walter M. Lovenberg, Ph.D., was an Executive Vice President and member of the Board of Directors of Marion Merrell Dow Inc. from September 1989 through August 1993. Dr. Lovenberg served as President of the Marion Merrell Dow Research Institute from September 1989 to August 1993 and Vice President from September 1986 through August 1989. Dr. Lovenberg has received the Fulbright-Hayes Senior Scholar Award, the Public Health Service Superior Service Award and the Third International Award for Research on Adult Diseases. Dr. Lovenberg has been the President of Lovenberg Assoc. since 1994. Dr. Lovenberg currently serves as a member of the Board of Directors of Inflazyme Pharmaceuticals, Ltd., and the Scientific Advisory Board of Guilford Pharmaceuticals, Inc. Dr. Lovenberg served as a member of the Board of Directors of Xenometrix, Inc. from May 1992 to March 2001 and Chief Executive Officer of Helicon Therapeutics, Inc. from July 1997 to December 1999. Dr. Lovenberg is also a Director of the following private companies: Merrimack Pharmaceutics, Inc., Quantum Bio, Inc. and Proquest Pharmaceuticals, Inc. Dr. Lovenberg served as a consultant to us from October 1993 to December 2002. Dr. Lovenberg became our director in March 1994.

Viren Mehta is the founder and managing member of Mehta Partners, LLC, providing investment, and strategic and financial advice to the global pharmaceutical and biotechnology industries, since January 1998. Mehta Partners was a strategic and financial advisor to us from April 1995 to December 2002. Dr. Mehta was a partner of Mehta and Isaly from July 1989 to December 1997. He was also a part of the strategic planning team of the International Division of Merck. Dr. Mehta obtained his Doctor of Pharmacy from the University of Southern California and his MBA in International Finance and Marketing from UCLA. Dr. Mehta advises investors and senior managers in the pharmaceutical and biotechnology industry. Dr. Mehta became our director in November 1999.

Herbert Pinedo, M.D., Ph.D., is Professor of Medical Oncology, and President of the VUMC-Cancer Center-Amsterdam which now includes the Institute of Drug Development and the Institute of Cancer Screening. Dr. Pinedo’s work focuses on translational research, in particular, drug resistance, angiogenesis and immunology. The Cancer Center has a formal collaboration with the John Hopkins Oncology Center, School of Medicine. Dr. Pinedo has received numerous international awards including the prestigious Josef

Steiner award. Dr. Pinedo is a member of numerous foundations and boards including the Dutch Cancer Society. He currently serves on the Scientific Advisory Boards of a number of pharmaceutical companies. He is a member of the British Royal Society of Medicine and The Royal Dutch Academy of Science and Arts, where he is Chairman of the Board of the Medical Division. Dr. Pinedo founded the New Drug Development Office — Oncology, which coordinates early clinical trials with anticancer agents internationally. He was the first President of the Federation of European Cancer Societies, and Past President to the European Society of Medical Oncology. Dr. Pinedo is the co-founder of the Annals of Oncology and The Oncologist and is the Co-Editor of Current Opinion in Anticancer Drugs. He serves on numerous editorial boards including Clinical Cancer Research and Journal of Clinical Oncology. Dr. Pinedo has authored more than 600 peer reviewed international publications and more than 120 chapters, invited papers or proceedings. Dr. Pinedo has been decorated by the Netherlands Queen with the prestigious Knight of the order of the Netherlands Lion. Dr. Pinedo was elected to our board on June 1, 2004.

Sir Mark Richmond, Ph.D., is an emeritus senior fellow of University College, London. From November 1993 to February 1996, Sir Mark served as the Head of Research and Special Assignments, Research Directorate, at Glaxo Research & Development, Ltd. From August 1981 to October 1990, he was the Vice Chancellor of the University of Manchester, and served as the Chairman of the Science and Engineering Research Council, the leading government funded agency supporting academic research in the United Kingdom from October 1990 to November 1993. Sir Mark is a non-executive director of a number of biotechnology companies in the U.S. and Europe, including Genentech, a public company. He is also a consultant in the biotechnology and pharmaceutical industries. Sir Mark became our director in April 2000.

John P. White is a partner of Cooper & Dunham LLP, a New York City law firm specializing in patent, trademark and related intellectual property matters, and has been associated with the firm since February 1977. Mr. White is a member of numerous professional organizations, both legal and scientific, and has written and lectured extensively on the subject of legal protection for biotechnology. Mr. White has been our director since May 1985.

The following discussion is a general summary of the material U.S. federal income and estate tax consequences of the ownership and disposition of the common stock applicable to “Non-U.S. Holders.” As used herein, a Non-U.S. Holder means a beneficial owner of the common stock that is not a U.S. person or a partnership for U.S. federal income tax purposes, and that will hold shares of the common stock as capital assets (i.e., generally, for investment). For U.S. federal income tax purposes, a U.S. person includes:


	•	an individual who is a citizen or resident of the United States;

	•	a corporation (or other business entity treated as a corporation) created or organized in the United States or under the laws of the United States or of any political subdivision thereof;

	•	an estate the income of which is includible in gross income regardless of source; or

	•	a trust, if such trust (A) is subject to the primary supervision of a court within the United States and one or more U.S. persons have the authority to control all substantial decisions of such trust, or (B) otherwise has elected to be treated as a U.S. domestic trust.

If a partnership holds shares of the common stock, the U.S. federal income tax treatment of a partner in the partnership will generally depend on the status of the partner and the activities of the partnership.

This summary is intended for general information only and does not consider specific facts and circumstances that may be relevant to a particular Non-U.S. Holder’s tax position. This summary does not consider U.S. state and local or non-U.S. tax consequences. It also does not consider Non-U.S. Holders subject to special tax treatment under the U.S. federal income tax laws (including banks and insurance companies, dealers in securities, persons that hold common stock as part of a “straddle,” “hedge,” “conversion transaction” or other risk-reduction transaction, controlled foreign corporations, passive foreign investment companies, foreign personal holding companies, companies that accumulate earnings to avoid U.S. federal income tax, foreign tax-exempt organizations, former U.S. citizens or residents, and persons who hold or receive common stock as compensation). This summary is based on provisions of the U.S. Internal Revenue Code of 1986, as amended (Code), applicable Treasury regulations, administrative pronouncements of the U.S. Internal Revenue Service (IRS) and judicial decisions, all as in effect on the date of this prospectus, all of which are subject to change, possibly on a retroactive basis, and different interpretations. No ruling has been or will be sought from the IRS with respect to any of the U.S. federal tax consequences described below, and there can be no assurance that the IRS will not take a contrary position.

The following summary is included herein as general information only. Accordingly, each prospective Non-U.S. Holder is urged to consult its own tax advisor with respect to the U.S. federal, state, local or non-U.S. income, estate and other tax consequences of holding and disposing of the common stock.

For purposes of this discussion, dividend income and gain on the sale or other taxable disposition of the common stock will be considered to be “U.S. trade or business income” if such income or gain is (i) effectively connected with the conduct by a Non-U.S. Holder of a trade or business within the United States and (ii) in the case of a Non-U.S. Holder that is eligible for the benefits of an income tax treaty with the United States, attributable to a permanent establishment maintained by the Non-U.S. Holder in the United States. Generally, U.S. trade or business income is not subject to U.S. federal withholding tax (provided the Non-U.S. Holder complies with applicable certification and disclosure requirements); instead, U.S. trade or business income generally is subject to U.S. federal income tax on a net income basis at regular U.S. federal income tax rates in the same manner as a U.S. person. Any U.S. trade or business income received by a Non-U.S. Holder that is a corporation also may, under specific circumstances, be subject to a “branch profits tax” at a 30% rate, or at a lower rate prescribed by an applicable income tax treaty.

As discussed under “Dividend Policy,” we do not currently expect to pay distributions on the common stock. In the event that distributions are paid by us, however, such distributions generally will constitute dividends for U.S. federal income tax purposes to the extent paid from our current or accumulated earnings and profits (as determined under U.S. federal income tax principles). Except as discussed above under “U.S. Trade or Business Income,” a Non-U.S. Holder generally will be subject to U.S. federal withholding tax at a 30% rate, or at a reduced rate prescribed by an applicable income tax treaty, on any dividends received in respect of the common stock. If the amount of a distribution exceeds our current and accumulated earnings and profits, such excess first will be treated as a tax-free return of capital to the extent of the Non-U.S. Holder’s tax basis in the common stock, and thereafter will be treated as capital gain, as discussed below under “Dispositions of Common Stock.” In order to obtain a reduced rate of U.S. federal withholding tax under an applicable income tax treaty, a Non-U.S. Holder will be required to provide a properly executed IRS Form W-8BEN certifying its entitlement to benefits under the treaty. A Non-U.S. Holder of our common stock that is eligible for a reduced rate of U.S. federal withholding tax under an income tax treaty may obtain a refund or credit of any excess amounts withheld by filing an appropriate claim for a refund with the IRS. A Non-U.S. Holder should consult its own tax advisor regarding its possible entitlement to benefits under an income tax treaty.

The U.S. federal withholding tax does not apply to dividends that are U.S. trade or business income, as defined above, paid to a Non-U.S. Holder who provides a properly executed IRS Form W-8ECI, certifying that the dividends are effectively connected with the Non-U.S. Holder’s conduct of a trade or business within the United States.

Except as discussed below under “Information Reporting and Backup Withholding Requirements,” a Non-U.S. Holder generally will not be subject to U.S. federal income or withholding tax in respect of any gain on a sale or other disposition of common stock unless:


	•	the gain is U.S. trade or business income, as defined above;

	•	the Non-U.S. Holder is an individual who is present in the United States for 183 or more days in the taxable year of the disposition and meets other conditions; or

	•	we are or have been a “U.S. real property holding corporation” (USRPHC) under section 897 of the Code at any time during the shorter of the five-year period ending on the date of disposition and the Non-U.S. Holder’s holding period for the common stock.

In general, a corporation is a USRPHC if the fair market value of its “U.S. real property interests” equals or exceeds 50% of the sum of the fair market value of its worldwide real property interests and its other assets used or held for use in a trade or business. If we are determined to be a USRPHC, the U.S. federal income and withholding taxes relating to interests in USRPHCs nevertheless will not apply to gains derived from the sale or other disposition of the common stock by a Non-U.S. Holder whose shareholdings, actual and constructive, at all times during the applicable period, amount to 5% or less of the common stock, provided that the common stock is regularly traded on an established securities market. We do not believe that we currently are a USRPHC, and we do not anticipate becoming a USRPHC in the future. However, no assurance can be given that we will not be a USRPHC, or that our common stock will be considered regularly traded, when a Non-U.S. Holder sells its shares of common stock.

Shares of the common stock owned or treated as owned by an individual who is a Non-U.S. Holder at the time of death will be included in the individual’s gross estate for U.S. federal estate tax purposes, and may be subject to U.S. federal estate tax, unless an applicable estate tax treaty provides otherwise.

We must annually report to the IRS and to each Non-U.S. Holder any dividend income that is subject to U.S. federal withholding tax, or that is exempt from such withholding tax pursuant to an income

tax treaty. Copies of these information returns also may be made available under the provisions of a specific treaty or agreement to the tax authorities of the country in which the Non-U.S. Holder resides. Under certain circumstances, the Code imposes a backup withholding obligation (currently at a rate of 28%) on certain reportable payments. Dividends paid to a Non-U.S. Holder of common stock generally will be exempt from backup withholding if the Non-U.S. Holder provides a properly executed IRS Form W-8BEN or otherwise establishes an exemption.

The payment of the proceeds from the disposition of common stock to or through the U.S. office of any broker, U.S. or foreign, will be subject to information reporting and possible backup withholding unless the owner certifies as to its non-U.S. status under penalties of perjury or otherwise establishes an exemption, provided that the broker does not have actual knowledge or reason to know that the holder is a U.S. person or that the conditions of any other exemption are not, in fact, satisfied. The payment of the proceeds from the disposition of common stock to or through a non-U.S. office of a non-U.S. broker will not be subject to information reporting or backup withholding unless the non-U.S. broker has certain types of relationships with the United States (U.S. related person). In the case of the payment of the proceeds from the disposition of common stock to or through a non-U.S. office of a broker that is either a U.S. person or a U.S. related person, the Treasury regulations require information reporting (but not the backup withholding) on the payment unless the broker has documentary evidence in its files that the owner is a Non-U.S. Holder and the broker has no knowledge to the contrary. Non-U.S. Holders should consult their own tax advisors on the application of information reporting and backup withholding to them in their particular circumstances (including upon their disposition of common stock).

Backup withholding is not an additional tax. Any amounts withheld under the backup withholding rules from a payment to a Non-U.S. Holder will be refunded or credited against the Non-U.S. Holder’s U.S. federal income tax liability, if any, if the Non-U.S. Holder provides the required information to the IRS.

The following description of our capital stock and certain provisions of our certificate of incorporation and bylaws is a summary and is qualified in its entirety by the provisions of our certificate of incorporation and bylaws.

Our authorized capital stock consists of 200,000,000 shares of common stock and 5,000,000 shares of preferred stock.

As of September 30, 2004, there were 43,586,747 shares of our common stock outstanding. The holders of common stock are entitled to one vote for each share held of record on all matters submitted to a vote of the stockholders. The holders of common stock are not entitled to cumulative voting rights with respect to the election of directors, and as a consequence, minority stockholders are not able to elect directors on the basis of their votes alone. Subject to preferences that may be applicable to any shares of preferred stock issued in the future, holders of common stock are entitled to receive ratably such dividends as may be declared by our board of directors out of funds legally available therefor. In the event of a liquidation, dissolution or winding up of our company, holders of the common stock are entitled to share ratably in all assets remaining after payment of liabilities and the liquidation preference of any then outstanding preferred stock. Holders of common stock have no preemptive rights and no right to convert their common stock into any other securities. There are no redemption or sinking fund provisions applicable to the common stock.

Our board of directors has the authority, without further action by the stockholders, to issue up to 5,000,000 shares of preferred stock in one or more series and to fix the rights, preferences, privileges and restrictions thereof, including dividend rights, conversion rights, voting rights, terms of redemption, liquidation preferences, sinking fund terms and the number of shares constituting any series or the designation of such series, without any further vote or action by stockholders. The issuance of preferred stock could adversely affect the voting power of holders of common stock and the likelihood that such holders will receive dividend payments and payments upon liquidation and could have the effect of delaying, deferring or preventing a change in control. We have no present plan to issue any shares of preferred stock.

Certain of our security holders have registration rights in respect of their shares. Under the agreements between us and the holders of such registration rights, we are required to maintain the effectiveness of certain registration statements covering their shares.

Our board of directors adopted a shareholder rights plan, pursuant to which one Series SRPA Junior Participating Preferred Stock Purchase Right was issued as a dividend for each outstanding share of common stock.

These rights entitle the holder to buy one one-thousandth of a share of Series SRPA Junior Participating Preferred Stock upon a triggering event as discussed below.

The rights become exercisable upon the occurrence of a triggering event, such as the announcement by a potential acquirer of the intention to initiate a tender offer that would result in the acquisition of 17.5% or more of the outstanding shares of our company or the actual acquisition of 17.5% or more of the outstanding shares of our company by any person or group of affiliated or associated persons.

Upon the actual acquisition of 17.5% or more of the outstanding common stock of our company by a person or group, the rights held by all holders other than the acquiring person or group will be modified

automatically to be rights to purchase shares of common stock (instead of rights to purchase preferred stock) at 50% of the then market value of such common stock. Furthermore, such rightholders will have the further right to purchase shares of common stock at the same discount if our company merges with, or sells 50% or more of its assets or earning power to, the acquiring person or group or any person acting for or with the acquiring person or group. If the transaction takes the form of a merger of our company into another corporation, these rightholders will have the right to acquire at the same percentage discount shares of common stock of the acquiring person or other ultimate parent of such merger party.

We can redeem the rights at any time before (but not after) a person has acquired 17.5% or more of our common stock, with certain exceptions. The rights will expire on August 31, 2010 if not redeemed prior to such date.

Our rights agreement may have anti-takeover effects. The rights may cause substantial dilution to a person or group that attempts to acquire us. Accordingly, the existence of the rights may deter acquirors from making takeover proposals or tender offers. However, the rights are not intended to prevent a takeover, but rather are designed to enhance the ability of our board to negotiate with an acquiror on behalf of all of the stockholders. In addition, the rights should not interfere with a proxy contest.

Our certificate of incorporation and bylaws discourage certain types of transactions involving an actual or potential change in control of our company which might be beneficial to our company or its stockholders. Our certificate of incorporation requires that any action required or permitted to be taken by our stockholders must be effected at a duly called annual or special meeting of the stockholders and may not be effected by a consent in writing. Our bylaws allow special meetings of stockholders to be called only by a majority vote of our board of directors and impose advance notice requirements and procedures for the submission by stockholders of nominations for our board of directors and stockholder proposals.

The Transfer Agent and Registrar for our common stock is The Bank of New York. The address is Stock Transfer Administration, 101 Barclay Street, 11th Floor East, New York, NY 10286, and its telephone number is (212)815-3881.

We intend to offer the shares of common stock through the underwriters named below. Merrill Lynch, Pierce, Fenner & Smith Incorporated is acting as representative of the underwriters named below. Subject to the terms and conditions described in an underwriting agreement among us and the underwriters, we have agreed to sell to the underwriters, and the underwriters severally have agreed to purchase from us, the number of shares listed opposite their names below.

The underwriters have agreed to purchase all of the shares sold under the purchase agreement if any of these shares are purchased. If an underwriter defaults, the purchase agreement provides that the purchase commitments of the non-defaulting underwriters may be increased or the purchase agreement may be terminated.

We have agreed to indemnify the underwriters against certain liabilities, including liabilities under the Securities Act, or to contribute to payments the underwriters may be required to make in respect of those liabilities.

The underwriters are offering the shares, subject to prior sale, when, as and if issued to and accepted by them, subject to approval of legal matters by their counsel, including the validity of the shares, and other conditions contained in the purchase agreement, such as the receipt by the underwriters of officers’ certificates and legal opinions. The underwriters reserve the right to withdraw, cancel or modify offers to the public and to reject orders in whole or in part.

The representative has advised us that the underwriters propose initially to offer the shares to the public at the public offering price on the cover page of this prospectus and to dealers at that price less a concession not in excess of $ per share. The underwriters may allow, and the dealers may reallow, a discount not in excess of $ per share to other dealers. After the offering, the public offering price, concession and discount may be changed.

The following table shows the public offering price, underwriting discount and proceeds before expenses to us. The information assumes either no exercise or full exercise by the underwriters of their overallotment option.

The expenses of the offering, not including the underwriting discount, are estimated at $ and are payable by us.

We have granted an option to the underwriters to purchase up to 825,000 additional shares at the public offering price less the underwriting discount. The underwriters may exercise this option for 30 days from the date of this prospectus solely to cover any overallotments. If the underwriters exercise this option,

We, our directors and our executive officers have agreed, with certain exceptions, not to sell or transfer any common stock for 90 days after the date of this prospectus without first obtaining the consent of Merrill Lynch. Specifically, these directors and officers have agreed not to directly or indirectly:


	•	offer, pledge, sell or contract to sell any common stock;

	•	sell any option or contract to purchase any common stock;

	•	purchase any option or contract to sell any common stock;

	•	grant any option, right or warrant for the sale of any common stock;

	•	otherwise dispose of or transfer any common stock;

	•	file, or cause to be filed, any registration statement under the Securities Act of 1933, as amended, with respect to any common stock; or

	•	enter into any swap or any other agreement or any transaction that transfers, in whole or in part, directly or indirectly, the economic consequence of ownership of any common stock, whether any such swap or transaction is to be settled by delivery of common stock or other securities, in cash or otherwise.

This lockup provision applies both to our common stock and to any securities convertible into or exchangeable or exercisable for our common stock.

Our shares of common stock are traded on the Nasdaq National Market under the symbol “OSIP.”

Until the distribution of our common stock offered hereby is completed, the SEC rules may limit the underwriters from bidding for or purchasing our common stock. However, the representative may engage in transactions that stabilize the price of our common stock, such as bids or purchases that peg, fix or maintain that price.

The underwriters may purchase and sell our common stock in the open market. These transactions may include short sales, stabilizing transactions and purchases to cover positions created by short sales. Short sales involve the sale by the underwriters of a greater number of shares than they are required to purchase in the offering.

“Covered” short sales are sales made in an amount not greater than the underwriters’ option to purchase additional shares from us in the offering. The underwriters may close out any covered short position by either exercising their option to purchase additional shares or purchasing shares in the open market. In determining the source of shares to close out the covered short position, the underwriters will consider, among other things, the price of shares available for purchase in the open market as compared to the price at which they may purchase shares through the overallotment option. “Naked” short sales are any sales in excess of such option. The underwriters must close out any naked short position by purchasing stock in the open market. A naked short position is more likely to be created if the underwriters are concerned that there may be downward pressure on the price of our common stock in the open market after pricing that could adversely affect investors who purchase in the offering. Stabilizing transactions consist of various bids for or purchases of common stock made by the underwriters in the open market prior to the completion of the offering.

The underwriters may also impose a penalty bid. This occurs when a particular underwriter repays to the underwriters a portion of the underwriting discount received by it because the representative has repurchased shares sold by or for the account of such underwriter in stabilizing or short covering transactions.

Similar to other purchase transactions, the underwriters’ purchases to cover the syndicate short sales may have the effect of raising or maintaining the market price of our common stock or preventing or retarding a decline in the market price of our common stock. As a result, the price of our common stock may be higher than the price that might otherwise exist in the open market.

Neither we nor any of the underwriters make any representation or prediction as to the direction or magnitude of any effect that the transactions described above may have on the price of the common stock. In addition, neither we nor the underwriters make any representation that the underwriters will engage in these transactions or that these transactions, once commenced, will not be discontinued without notice.

In connection with this offering, underwriters and selling group members may engage in passive market making transactions in our common stock on the Nasdaq National Market in accordance with Rule 103 of Regulation M under the Exchange Act during a period before the commencement of offers or sales of our common stock and extending through completion of the distribution. A passive market maker must display its bid at a price not in excess of the highest independent bid of that security. However, if all independent bids are lowered below the passive market maker’s bid, that bid must then be lowered when specified purchase limits are exceeded.

Merrill Lynch will be facilitating Internet distribution for this offering to certain of its Internet subscription customers. Merrill Lynch intends to allocate a limited number of shares for sale to its online brokerage customers. An electronic prospectus is available on the Website maintained by Merrill Lynch. Other than the prospectus in electronic format, the information on the Merrill Lynch Website is not a part of this prospectus.

The underwriters and their affiliates have provided investment and commercial banking and financial advisory services from time to time for us in the ordinary course of business for which they have received customary fees. Any of the underwriters or their respective affiliates may in the future engage in investment banking or other transactions of a financial nature with us or our affiliates, including the provision of advisory services and the making of loans to us or our affiliates, for which they would receive customary fees or other payments.

Saul Ewing LLP, Philadelphia, Pennsylvania, will pass upon the validity of the shares of common stock offered in this prospectus by us. Shearman & Sterling LLP, New York, New York, will pass upon certain legal matters on behalf of the underwriters.

The consolidated financial statements of OSI Pharmaceuticals, Inc. and subsidiaries as of September 30, 2003 and 2002, and for each of the years in the three-year period ended September 30, 2003, have been incorporated by reference in this prospectus and in the registration statement in reliance upon the report of KPMG LLP, independent accountants, also incorporated by reference in this prospectus, and upon the authority of said firm as experts in accounting and auditing. The audit report covering our September 30, 2003 consolidated financial statements reflects the full adoption of the provisions of Statement of Financial Accounting Standards (“SFAS”) No. 142, “Goodwill and Other Intangible Assets” in 2003; the provisions of SFAS No. 141, “Business Combinations” and SFAS No. 142, for acquisitions consummated on or after July 1, 2001; and the early adoption of the provisions of Statement of Financial Accounting Standards No. 145 “Rescission of FASB Statements No. 4, 44 and 64, Amendment of FASB Statement No. 13, and Technical Corrections” relating to the classification of the effect of early debt extinguishments in 2002.

We have filed a registration statement on Form S-3, of which this prospectus is a part, and related exhibits with the Securities and Exchange Commission (SEC) as required by the Securities Act. The registration statement contains additional information about us and our common stock. We also file annual and quarterly reports, proxy statements and other information with the SEC. You may read and copy the registration statement or any other document we file with the SEC at the SEC’s Public Reference Room at 450 Fifth Street, N.W., Washington, D.C. 20549. Please call the SEC at 1-800-SEC-0330 for further information on the Public Reference Room.

The SEC also maintains a website that contains reports, proxy and information statements, and other information that we have filed electronically. The SEC’s website is located at http://www.sec.gov. We also maintain our own website which is located at http://www.osip.com. The information on our website is not incorporated by reference in this prospectus and should not be considered a part of this prospectus. We have included our website address as an inactive textual reference only.

The SEC allows us to incorporate into this prospectus information that we file with the SEC in other documents, which means that we can disclose important information to you by referring you to those documents. The information incorporated by reference is an important part of this prospectus. Any statement contained in a document which is incorporated by reference is automatically updated and superseded by such information contained in this prospectus. In addition, information that we file with the SEC after the date of this prospectus and prior to the completion of the offering of the shares offered hereby will update and supersede the information confirmed in this prospectus and the incorporated filings. We incorporate by reference in this prospectus the following documents we have filed with the SEC:


	1. our annual report on Form 10-K for the fiscal year ended September 30, 2003, filed with the SEC on December 2, 2004;


	2. our current reports on Form 8-K, filed with the SEC on October 2, 2003, April 26, 2004, June 7, 2004, June 9, 2004, June 28, 2004, July 6, 2004, September 21, 2004 and October 21, 2004;


	3. our quarterly report on Form 10-Q for the quarter ended December 31, 2003, filed with the SEC on February 3, 2004;

	4. our quarterly report on Form 10-Q for the quarter ended March 31, 2004, filed with the SEC on May 14, 2004;

	5. our quarterly report on Form 10-Q for the quarter ended June 30, 2004, filed with the SEC on August 16, 2004;

	6. our proxy statement, dated February 4, 2004, for our 2004 annual meeting of stockholders, filed with the SEC on January 28, 2004; and

	7. the description of our common stock, which is registered under Section 12 of the Exchange Act in our registration statement on Form 8-A, filed on November 21, 1986 (No. 0-15190) including any amendments or reports filed for the purpose of updating such description.

All documents we have filed with the SEC under Sections 13(a), 13(c), 14 or 15(d) of the Exchange Act after the date of this prospectus prior to the completion of this offering (other than respective filings or portions of filings that are furnished, under applicable SEC rules, rather than filed) will be deemed incorporated by reference in this prospectus and will become a part of this prospectus. To receive a free copy of any of the documents incorporated by reference in this prospectus call or write Robert L. Van Nostrand, Vice President and Chief Financial Officer, OSI Pharmaceuticals, Inc., 58 South Service Road, Melville, New York 11747, telephone (631)962-2108. We will not send exhibits to the documents unless those exhibits have been specifically incorporated by reference in this prospectus.

The following expenses incurred in connection with the sale of the securities being registered will be borne by OSI. Other than the SEC registration fee, the amounts stated are estimates.

Section 145 of the General Corporation Law of Delaware empowers a corporation to indemnify any person who was or is a party or is threatened to be made a party to any threatened, pending or completed action, suit or proceeding, whether civil, criminal, administrative or investigative, by reason of the fact that he or she is or was a director, officer, employee or agent of the corporation or another enterprise if serving such enterprise at the request of the corporation. Depending on the character of the proceeding, a corporation may indemnify against expenses (including attorney’s fees), judgments, fines and amounts paid in settlement actually and reasonably incurred in connection with such action, suit or proceeding if the person indemnified acted in good faith and in a manner he or she reasonably believed to be in or not opposed to the best interests of the corporation, and, with respect to any criminal action or proceeding, had no reasonable cause to believe his or her conduct was unlawful. In the case of an action by or in the right of the corporation, no indemnification may be made in respect to any claim, issue or matter as to which such person shall have been adjudged to be liable to the corporation unless and only to the extent that the Court of Chancery or the court in which such action or suit was brought shall determine that despite the adjudication of liability, but in view of all the circumstances of the case, such person is fairly and reasonably entitled to indemnity for such expenses that the court shall deem proper. Section 145 further provides that to the extent a director or officer of a corporation has been successful in the defense of any action, suit or proceeding referred to above, or in defense of any claim, issue or matter therein, he or she shall be indemnified against expenses (including attorney’s fees) actually and reasonably incurred by him or her in connection therewith.

OSI’s Certificate of Incorporation provides that OSI shall, to the fullest extent authorized by the General Corporation Law of Delaware, indemnify any person, or the legal representative of any person, who is or was a director, officer, employee or agent of OSI or another enterprise if said person served such enterprise at the request of OSI. The Certificate of Incorporation also provides that any amendment to the General Corporation Law of Delaware shall only be applicable to the extent any such amendment permits OSI to provide broader indemnification rights than said law permitted OSI to provide prior to such amendment. The Certificate of Incorporation further provides that in the case of an action, suit or proceeding initiated by the indemnified person, OSI shall indemnify the person only if such action, suit or proceeding was authorized by OSI’s Board of Directors. The Certificate of Incorporation also contains a provision eliminating the liability of directors of OSI to OSI or its stockholders for monetary damages for breach of fiduciary duty except under certain specified circumstances. The Certificate of Incorporation also permits OSI to maintain insurance to protect itself and any director, officer, employee or agent against any liability whether or not OSI would have the power to indemnify such persons under the General

The following is a list of exhibits filed as part of the registration statement:


*	Previously filed as an exhibit to the registration statement (file no. 333-119229) filed on September 23, 2004.

Insofar as indemnification for liabilities arising under the Securities Act of 1933 may be permitted to directors, officers and controlling persons of the registrant pursuant to the foregoing provisions, or otherwise, the registrant has been advised that in the opinion of the Securities and Exchange Commission such indemnification is against public policy as expressed in the Act and is, therefore, unenforceable. In the event that a claim for indemnification against such liabilities (other than the payment by the registrant of expenses incurred or paid by a director, officer or controlling person of the registrant in the successful defense of any action, suit or proceeding) is asserted by such director, officer or controlling person in connection with the securities being registered, the registrant will, unless in the opinion of its counsel the matter has been settled by controlling precedent, submit to a court of appropriate jurisdiction the question whether such indemnification by it is against public policy as expressed in the Act and will be governed by the final adjudication of such issue.

The undersigned registrant hereby undertakes that, for purposes of determining any liability under the Securities Act of 1933, each filing of the registrant’s annual report pursuant to Section 13(a) or Section 15(d) of the Securities Exchange Act of 1934 (and, where applicable, each filing of an employee benefit plan’s annual report pursuant to section 15(d) of the Securities Exchange Act of 1934) that is incorporated by reference in the registration statement shall be deemed to be a new registration statement relating to the securities offered therein, and the offering of such securities at that time shall be deemed to be the initial bona fide offering thereof.


	(i) for purposes of determining any liability under the Securities Act of 1933, the information omitted from the form of prospectus filed as part of this registration statement in reliance upon Rule 430A and contained in a form of prospectus filed by the registrant pursuant to Rule 424(b)(1) or (4), or 497(h) under the Securities Act of 1933 shall be deemed to be part of this registration statement as of the time it was declared effective.

	(ii) for the purpose of determining any liability under the Securities Act of 1933, each post-effective amendment that contains a form of prospectus shall be deemed to be a new registration statement relating to the securities offered therein, and the offering of such securities at that time shall be deemed to be the initial bona fide offering thereof.

Pursuant to the requirements of the Securities Act of 1933, as amended, the registrant certifies that it has reasonable grounds to believe that it meets all of the requirements for filing on Form S-3 and has duly caused this Amendment No. 1 to be signed on its behalf by the undersigned, thereunto duly authorized, in the City of Melville, State of New York, on November 4, 2004.

Pursuant to the requirements of the Securities Act of 1933, as amended, this amendment has been signed by the following persons in the capacities and on the dates indicated.


Delaware		13-3159796
(State or other jurisdiction of incorporation)		(I.R.S. Employer Identification No.)


	Per Share	Total

Public offering price	$	$
Underwriting discount	$	$
Proceeds, before expenses, to OSI Pharmaceuticals, Inc.	$	$



												Nine Months Ended
			Years Ended September 30,									June 30,

			2001			2002			2003			2003			2004


												(Unaudited)

			(In thousands, except per share data)
Consolidated Statement of Operations Data:
	Revenues		$	26,022		$	21,816		$	32,369		$	20,086		$	29,773

	Expenses:
		Cost of product sales		—			—			157			55			2,262
		Research and development		56,038			102,202			102,642			76,297			76,178
		Acquired in-process research and development		—			130,200			31,451			31,290			—
		Selling, general and administrative		16,033			28,146			70,532			41,496			65,279
		Amortization of intangibles		742			1,255			9,300			4,666			13,986

	Loss from operations		$	(46,791	)	$	(239,987	)	$	(181,713	)	$	(133,718	)	$	(127,932	)

	Other income(expense) — net			25,661			7,904			356			1,331			(9,250	)
	Gain on sale of diagnostic business			—			1,000			—			—			—
	Gain on early retirement of convertible senior subordinated notes — net			—			12,604			—			—			—

	Loss before cumulative effect of accounting change		$	(21,130	)	$	(218,479	)	$	(181,357	)	$	(132,387	)	$	(137,182	)

	Cumulative effect of the change in accounting for the recognition of upfront fees			(2,625	)		—			—			—			—

	Net loss		$	(23,755	)	$	(218,479	)	$	(181,357	)	$	(132,387	)	$	(137,182	)

	Basic and diluted net loss per common share:
		Loss before cumulative effect of change in accounting policy	$	(0.62	)	$	(6.07	)	$	(4.87	)	$	(3.62	)	$	(3.50	)
		Cumulative effect of change in accounting policy	$	(0.08	)		—			—			—			—

		Net loss	$	(0.70	)	$	(6.07	)	$	(4.87	)	$	(3.62	)	$	(3.50	)

	Weighted average number of shares of common stock outstanding			33,852			35,978			37,249			36,618			39,173


										As of
	As of September 30,									June 30,

	2001			2002			2003			2004

										(Unaudited)

	(In thousands)
Consolidated Balance Sheet Data:
Cash, cash equivalents and investment securities (unrestricted and restricted)	$	551,479		$	476,277		$	404,147		$	321,903
Working capital		533,761			445,596			379,598			299,384
Total assets		591,689			579,044			591,502			484,022
Total liabilities		41,857			199,936			373,445			372,983
Accumulated deficit		(105,744	)		(324,223	)		(505,580	)		(642,762	)
Total stockholders’ equity		549,831			379,108			218,057			111,039


Common stock offered by OSI		5,500,000 shares


Common stock to be outstanding after this offering		49,086,747 shares



Use of proceeds		We estimate that we will receive net proceeds from this offering of $333.7 million ($383.9 million if the underwriters exercise the overallotment option in full), after deducting underwriting discounts and commissions and estimated offering expenses payable by us. We anticipate that the net proceeds from this offering will be used for the ongoing development and commercialization of Tarceva^TM; research and development; strategic acquisitions and in-licensing opportunities; working capital; and general corporate purposes.


Risk factors		See the “Risk Factors” section included in this prospectus as well as any risk factors incorporated by reference into this prospectus for a discussion of factors you should carefully consider before deciding to invest in shares of our common stock.

Nasdaq National Market symbol		OSIP


Spencer W. Franck, Jr. Saul Ewing LLP 1200 Liberty Ridge Drive, Suite 200 Wayne, Pennsylvania 19087 (610) 251-5082		Marc M. Rossell Danielle Carbone Shearman & Sterling LLP 599 Lexington Avenue New York, New York 10022-6069 (212) 848-4000


	Page

Prospectus Summary		1
Risk Factors		9
Special Note Regarding Forward-Looking Statements		18
Use of Proceeds		19
Price Range of Common Stock		20
Dividend Policy		20
Capitalization		21
Dilution		22
Selected Consolidated Financial Data		23
Business		26
Management and Directors		50
Certain U.S. Federal Tax Considerations for Non-U.S. Holders		55
Description of Capital Stock		58
Underwriting		60
Legal Matters		63
Experts		63
Where You Can Find More Information		63
Incorporation of Certain Documents by Reference		64


	Tarceva^TM		Placebo		Hazard
	N = 488		N = 243		Ratio(1)			P-value(2)

Median Survival/Hazard Ratio		6.7 months		4.7 months		0.73	(3)		<0.001	(3)
One-Year Survival Rate		31.2%		21.5%		—			—
Median Progression-Free Survival/Hazard Ratio		2.2 months		1.8 months		0.59	(3)		<0.001	(3)
Objective Tumor Response Rate(4)		9%		<1%		—			—


	*We have incurred losses since our inception, and we expect to incur losses over the next several years, which may cause the value of our common stock to decrease.*


	We, together with our alliance partners Genentech and Roche, may not be able to market or generate sales of Tarceva^TM to the extent anticipated.


	If government agencies do not grant us or our collaborative partners required approvals for any of our potential products, including Tarceva^TM, in a timely manner or at all, we or our collaborative partners will not be able to distribute or sell our products currently under development.


	We are responsible for the supply of Tarceva^TM in the United States. Since we have no commercial manufacturing facilities, we are dependent on two suppliers for the active pharmaceutical ingredient (API) for Tarceva^TM and a single supplier for the tableting of Tarceva^TM in the United States.


	If we do not maintain our co-development and marketing alliance with Genentech and Roche for Tarceva^TM, the marketing and sale of Tarceva^TM may be compromised or delayed.


	If any of our current or future marketed products, including Novantrone® or Tarceva^TM, were to become the subject of problems related to their efficacy, safety, or otherwise, or if new, more effective treatments were to be introduced, our revenues from such marketed products could decrease.


	*If we do not receive adequate third-party reimbursement for the sales of our products, we may not be able to sell such products on a profitable basis.*


Merrill Lynch & Co.	Morgan Stanley


	Bear, Stearns & Co. Inc.


	Lazard


	The Medicare Prescription Drug Improvement and Modernization Act of 2003 materially changes the Medicare reimbursement guidelines for intravenous and oral oncology products. Such changes may negatively impact our revenues for Novantrone®.


	If Serono S.A. does not fulfill its obligations for manufacturing and supplying Novantrone®, we may not be able to continue the marketing and distribution of the product which could cause our revenues to decrease.


	One aspect of our business strategy depends on our ability to identify and acquire product candidates and marketed products, which if not met, may prevent us from achieving expected future performance.


	Although we have potential products that appear to be promising at early stages of development and in clinical trials, none of our potential products may reach the commercial market for a number of reasons.


	If our competitors succeed in developing products and technologies that are more effective than our own, or if scientific developments change our understanding of the potential scope and utility of our products, then our products and technologies may be rendered less competitive.


	Our reliance on third parties, such as clinical research organizations (CROs), may result in delays in completing, or a failure to complete, clinical trials if they fail to perform under our agreements with them.


	*The use of any of our potential products in clinical trials and the sale of any approved products exposes us to liability claims.*


	*If we or our collaborative partners are required to obtain licenses from third parties, our revenues and royalties on any commercialized products could be reduced.*


	*If we cannot successfully protect, exploit or enforce our intellectual property rights, our ability to develop and commercialize our products will be severely limited.*


	*If other companies claim that we infringe on their intellectual property rights, we may be subject to costly and time-consuming litigation and delays in product introduction.*


	*Our ability to raise capital in the future may be limited, which could adversely impact our growth.*


	We have outstanding options, convertible debt, contingent value rights and warrants, the exercise, conversion or exchange of which could dilute stockholder value and cause our stock price to decline.


	Our outstanding indebtedness increased substantially with the issuance of convertible senior subordinated notes in September 2003, and we may not be able to make the required payments on these notes when due and therefore may face liquidity problems.


	If the market price of our common stock, similar to other biotechnology companies, remains highly volatile, then our stockholders may not be able to sell their stock when desired or at desirable prices.


	Our governance documents and state law provide certain anti-takeover measures which will discourage a third party from seeking to acquire us and may impede the ability of stockholders to remove and replace our board of directors and, therefore, our management.


	*New investors will experience immediate and substantial dilution and this offering could lead to a decline of the price of our common stock.*


		High		Low

Fiscal Year 2005
	First Quarter (through November 3, 2004)	$	66.47	$	60.51
Fiscal Year 2004
	First Quarter	$	34.19	$	24.47
	Second Quarter		43.26		29.41
	Third Quarter		98.70		33.94
	Fourth Quarter		70.41		50.71
Fiscal Year 2003
	First Quarter	$	22.74	$	14.04
	Second Quarter		17.39		12.84
	Third Quarter		37.30		13.05
	Fourth Quarter		38.34		29.15
Fiscal Year 2002
	First Quarter	$	50.94	$	31.91
	Second Quarter		49.05		35.11
	Third Quarter		39.45		20.52
	Fourth Quarter		33.81		11.50



			As of June 30, 2004

									Pro Forma
			Actual			Pro Forma			As Adjusted


			(Unaudited, in thousands,
			except per share data)
Cash, cash equivalents and investment securities (unrestricted and restricted)			$	321,903		$	315,503		$	649,275

Long-term debt:
	4% Convertible Senior Subordinated Notes due 2009			160,000			—			—
	3 1/4% Convertible Senior Subordinated Notes due 2023			150,000			150,000			150,000
Stockholders’ equity:
	Preferred stock, $0.01 par value; 5,000 shares authorized; no shares issued or outstanding actual and as adjusted			—			—			—
	Common stock, $0.01 par value; 200,000 shares authorized; 41,474 shares of issued actual; 44,674 shares of issued pro forma; 50,174 shares of issued pro forma as adjusted			415			447			502
	Additional paid-in capital			778,328			934,573			1,268,290
	Deferred compensation			(330	)		(330	)		(330	)
	Accumulated deficit			(642,762	)		(642,762	)		(642,762	)
	Accumulated other comprehensive income			839			839			839
	Treasury stock, at cost 1,443 shares			(25,451	)		(25,451	)		(25,451	)

		Total stockholders’ equity		111,039			267,316			601,088

		Total capitalization	$	421,039		$	417,316		$	751,088


Assumed public offering price per share				$	64.345
	Net tangible book value per share as of June 30, 2004	$	3.97
	Increase per share attributable to new investors		6.40

As adjusted net tangible book value per share after the offering					10.37

Dilution per share to new investors				$	53.97



		Tarceva^TMvs												Taxotere® Plus Aptosyn® vs.
		Best Supportive Care						Taxotere® vs.						Taxotere
		(BSC)						Alimta®						(610 patients)(1)
		(731 patients)						(571 patients)
														Taxotere®Plus
		Tarceva^TM			BSC			Taxotere®			Alimta®			Aptosyn®			Taxotere®
Study		(N = 488)			(N = 243)			(N = 288)			(N = 283)			(N = 304)			(N = 306)

Performance Status:
	0-1		65	%		68	%		88	%		89	%		87	%		87	%
	2		26	%		23	%		12	%		11	%		11	%		10	%
	3		9	%		9	%		0	%		0	%		0	%		0	%
	Missing		0	%		0	%		0	%		0	%		2	%		3	%
Prior Regimens:
	1		50	%		50	%		100	%		100	%		74	%		75	%
	³2		50	%		50	%		0	%		0	%		23	%		22	%
Response Rate (Complete Response & Partial Response)			8.9	%		<1	%		8.8	%		9.1	%		9.2	%		7.2	%
Survival:
	1 year Survival		31.2	%		21.5	%		29.7	%		29.7	%		30.7	%		29.5	%
	Median Survival (Months)		6.7	(2)		4.7			7.9			8.3			6.9			6.9