UNITED STATES

SECURITIES AND EXCHANGE COMMISSION

WASHINGTON, D.C. 20549

FORM 10-K

(Mark One)

for the fiscal year ended December 31, 20022003

OR

for the transition period from ____________ to ____________ .

Commission file number: 0-28386

CELL THERAPEUTICS, INC.

(Exact name of registrant as specified in its charter)

Registrant’s telephone number, including area code: (206) 282-7100

Securities to be registered pursuant to Section 12(b) of the Act:  None

None.

Securities to be registered pursuant to Section 12(g) of the Act:

Common Stock, no par value

(Title of Classes)

Title of each class		Name of each exchange on which registered
Common Stock, no par value		NASDAQ National Market

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

Indicate by check mark whether the registrant is an accelerated filer (as defined in Exchange Act Rule 12b-2). Yesx  No¨

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

Indicate by check mark whether the registrant is an accelerated filer (as defined in Exchange Act Rule 12b-2). Yesx  No¨

The aggregate market value of the voting and non-voting common equity held by nonaffiliates as of December 31, 2002,2003 was approximately $207,823,000, based on the closing price of such shares on the Nasdaq National Market on June 28, 2002, was approximately $137,774,000.30, 2003. Shares of common stock held by each executive officer and director and by each person known to the Company who beneficially owns more than 5% of the outstanding Common Stock have been excluded in that such persons may under certain circumstances be deemed to be affiliates. This determination of executive officer or affiliate status is not necessarily a conclusive determination for other purposes.

The number of outstanding shares of the registrant’s common stock as of February 28, 2003March 1, 2004 was 33,135,892.50,306,447.

DOCUMENTS INCORPORATED BY REFERENCE:REFERENCE

The information required by Part III of this Report, to the extent not set forth herein, will be incorporated herein by reference from the registrant’s definitive proxy statement relating to the annual meeting of shareholders to be held in 2003,2004, which definitive proxy statement or amendment to this annual report shall be filed with the Securities and Exchange Commission within 120 days after the end of the fiscal year ended December 31, 2003 to which this Report relates.

CELL THERAPEUTICS, INC.

TABLE OF CONTENTS

PART IForward Looking Statements

This Form 10-K contains, in addition to historical information, forward-looking statements. These statements relate to our future plans, objectives, expectations, intentions and financial performance, and assumptions that underlie these statements. When used in this Form 10-K, terms such as “anticipates,” “believes,” “continue,” “could,” “estimates,” “expects,” “intends,” “may,” “plans,” “potential,” “predicts,” “should,” or “will” or the negative of those terms or other comparable terms are intended to identify forward-looking statements. These statements involve known and unknown risks, uncertainties and other factors that may cause industry trends or our actual results, level of activity, performance or achievements to be materially different from any future results, levels of activity, performance or achievements expressed or implied by these statements. These factors include those listed under “Factors Affecting Our Operating Results,” “Management’s Discussion and Analysis of Financial Condition and Results of Operations,” and “Business” and elsewhere in this Form 10-K.

Although we believe that expectations reflected in the forward-looking statements are reasonable, we cannot guarantee future results, levels of activity, performance or achievements. We will not update any of the forward-looking statements after the date of this Form 10-K to conform these statements to actual results or changes in our expectations. Readers are cautioned not to place undue reliance on these forward-looking statements, which apply only as of the date of this Form 10-K.

PART I

Item 1.    Business

Overview

We develop, acquire and commercialize novel treatments for cancer. Our goal is to build a leading, vertically-integrated biopharmaceutical company with a diversified portfolio of proprietary oncology drugs. Our research and in-licensing activities are concentrated on identifying new, less toxic and more effective ways to treat cancer. We market TRISENOX^® for the treatment of relapsed or refractory acute promyelocytic leukemia in the U.S. and in the European Union, or EU, and have more than 70 ongoing or planned Cell Therapeutics Inc. (“CTI”) clinical development trials or investigator-sponsored research trials related to potential market expansion for this product. We are developing XYOTAX for the treatment of non-small cell lung cancer, or NSCLC, and ovarian cancer. We are completing enrollment in the second and third of three pivotal phase III trials for the treatment of NSCLC. These three trials are designed to enroll more than 1,600 patients. We are developing pixantrone, a novel anthracycline, for the treatment of non-Hodgkin’s lymphoma, or NHL, and plan to initiate a pivotal phase III trial for relapsed, aggressive NHL in the first half of 2004. We are also developing CT-2106, which is entering phase II trials for the treatment of colorectal, ovarian and small cell lung cancer.

We were incorporated in Washington in 1991. Our principal executive offices are located at 501 Elliott Avenue West, Seattle, Washington 98119. Our telephone number is (206) 282-7100. Our website can be found at www.cticseattle.com. We make available free of charge on our Internet website our annual reports on Form 10-K, quarterly reports on Form 10-Q, current reports on Form 8-K and other filings pursuant to Section 13(a) or 15(d) of the Securities Exchange Act of 1934, and amendments to such filings, as soon as reasonably practicable after each is electronically filed with, or furnished to, the Securities and Exchange Commission, or the SEC.

“CTI,” “TRISENOX”“TRISENOX,” and “XYOTAX” (formerly referred to as PG-TXL) are our proprietary marks of Cell Therapeutics, Inc.marks. All other product names, trademarks and trade names referred to in this Form 10-K are the property of their respective owners.

Our Products

We acquired our lead product called arsenic trioxide, or TRISENOX,^®, in January 2000. We received Food and Drug Administration, or FDA, approval to market TRISENOX in the U.S. in September 2000, and the European Agency for the Evaluation of Medicinal Products, or EMEA, approval to market in the European Community, or EU in March 2002. TRISENOX is

marketed for patients with a type of blood cell cancer called acute promyelocytic leukemia, or APL, who have relapsed or failed standard therapies. In its pivotal trial in patients with relapsed or refractory APL, 70% of the 40 patients experienced complete remission, or CR, following treatment with TRISENOX with 82% achieving a molecular remission. We have received orphan drug designation for TRISENOX from the FDA for APL, multiple myeloma, myelodysplastic syndromes, or MDS, chronic myeloid leukemia, or CML, and acute myeloid leukemia, or AML.AML, chronic lymphocytic leukemia or CLL, and hepatocellular carcinoma, or HCC. In addition, TRISENOX is currently listed in the U.S. Pharmacopeia Oncology Drug Information, or USP DI, under Orphan Product Designation and Approvals in multiple myeloma, MDS, CML, AML, CLL and MDS.HCC. We have also received designation as an orphan medicinal product by

the EMEA under its orphan drug legislation for APL, MDS, and multiple myeloma. Thirty-four TRISENOXMore than 70 clinical and investigator-sponsored trials studying the drug alone or in combination with other therapies are ongoing.planned or ongoing as of February 2004. Preliminary data from ongoing clinical trials have shown encouraging responses in patients with multiple myeloma, MDS, CML, prostate cancer, and neuroblastoma.

We are also developing a new way to deliver cancer drugs more selectively to tumor tissue in order to reduce the toxic side effects to normal organs and tissues and to improve the anti-tumor activity of existing chemotherapy agents. Our technology links, or conjugates, chemotherapy drugs to biodegradable polymers, including polyglutamate. We believe this technology works by taking advantageTwo of the characteristics ofour product candidates, XYOTAX and CT-2106, utilize a unique biodegradable protein polymer to deliver a taxane and a camptothecin, respectively, more selectively to tumor tissue. Because tumor blood vessels to increaseare more porous than those in normal tissue, polymers within a specific size range circulate in the percentage ofbloodstream and get trapped in tumor tissue and accumulate preferentially within the drug administered that actually reachestumor. In the tumor which may increasetissue, the potency and reduce the side effects of a given dose compared to giving thepolymer drug alone. In addition, the conjugates appear to beis taken up by cells. The polymer bound drugs are inactive while circulating in the bloodstream, which may also lower toxicity relativecompared to the active drug substance alone.

Once within a cell, naturally occurring enzymes digest the polymer releasing the chemotherapy drug. Preclinical animal studies and human clinical data to date indicate that our polygulatamate technology may allow more drug to reach the tumor, provide increased efficacy using the same amount of chemotherapy drug and less toxicity at the same or higher equivalent doses of drug, as compared to unlinked cancer drugs.

Our first application of the polymer technology is XYOTAX^™, or CT-2103, which is paclitaxel linked to polyglutamate. Paclitaxel is the active ingredient in Taxol^®, one of the world’s best selling cancer drugs. In animal studies, XYOTAX demonstrated fewer side effects and improved tumor killing-activity when compared to Taxolpaclitaxel alone. SixEight phase I clinical trials, fourthree phase II clinical trials and three phase III clinical trials are currently underway,underway. We also anticipate that the Gynecologic Oncology Group, or GOG, will submit a phase III trial protocol through a Special Protocol Assessment, or SPA, to the FDA to compare XYOTAX as maintenance therapy, administered monthly for 12 months, to no maintenance treatment for ovarian cancer patients who have achieved a complete remission, or CR, following front-line treatment with three additional trials plannedcarboplatin and paclitaxel. The trial is expected to begin in 2003.the first half of 2004. We also initiated development of a novel polyglutamate-camptothecin molecule, or PG-CPT,CT-2106, and filed a U.S. investigational new drug application, or IND, in December 2001. We initiated a phase I clinical trial with PG-CPTCT-2106 in the first quarter of 2002, and plan to initiate another trialphase II trials in 2004.

On January 1, 2004 we completed our acquisition of Novuspharma S.p.A., (“Novuspharma”), an Italian biopharmaceutical company focused on oncology. Prior to its spinout as an independent company in 1998, Novuspharma was the second halfoncology drug development division of 2003.Boehringer Mannheim and part of Hoffmann-La Roche. This acquisition provided us worldwide rights to pixantrone, approximately $92.5 million of cash and cash equivalents and a high-quality drug discovery organization with an extensive track record in cancer drug development. We believe pixantrone has potentially higher activity and lower cardiac toxicity than marketed anthracyclines. Cardiac toxicity is a serious limitation that can result in life-threatening heart failure. The Novuspharma acquisition and the drug candidates we obtained are consistent with our strategy of growth through strategic acquisition and our goal of developing less toxic more effective cancer therapies.

In January 2004, we received a Notice of Allowance from the U.S. Patent and Trademark Office for a patent application that would, upon issuance, extend our U.S. intellectual property position for TRISENOX through 2018. This extension is 11 years beyond the original orphan drug designation for APL that currently expires in 2007. We believe further investments in registration-directed clinical trials for various blood-related cancers and solid tumors could accelerate TRISENOX sales growth and increase the drug’s market potential.

We have identifiedare working on a number of drug targets in Discovery Research. Among these programs are HIF-1 alpha:p300, proteasome inhibition, a novel drug target called lysophosphatidic acid acyltransferase, or LPAAT-ß, that when inhibited has been shown to reduce tumor cell growth in preclinical models.and other similar enzymatic targets. We are in the process of identifying acontinued target validation and lead candidate for clinicaloptimization and may elect to move one of these programs into early development during 2003.in 2005.

The Oncology Market

Overview.Overview.    According to the American Cancer Society, or ACS, cancer is the second leading cause of death in the United States, resulting in over 555,000 deaths annually. The National Cancer Advisory Board reports that more than 8.99.2 million people in the United States have cancer, and it is estimated that one in three American women, and one in two American men will develop cancer in their lifetime. Approximately 1.31.4 million new cases of cancer are diagnosed each year in the United States. The most commonly used methods for treating patients with cancer are surgery, radiation and chemotherapy. Patients usually receive a combination of these treatments depending upon the type and extent of their disease. At the time of diagnosis, 70% of patients have tumors that have already spread to other parts of the body. Therefore, almost all receive systemic therapy such as chemotherapy during the course of their disease.

Despite recent advances in sequencing the human genome and the introduction of new biologic therapies for the treatment of cancer almost all patients with advanced cancer will receive chemotherapy at some point during the treatment of their disease. Four classes of chemotherapy agents, anthracyclines, camptothecins, platinates and taxanes, account for more than 95% of all chemotherapy drug usage. Unfortunately, there are significant limitations and complications associated with radiation and chemotherapythese agents that result in a high rate of treatment failure. The principal limitations of chemotherapy include:

We believe next generation agents which have lower toxicity and/or increased effectiveness of these cornerstone chemotherapy classes addresses a significant unmet market need for the majority of cancer patients in the U.S. Our cancer drug development pipeline includes a next-generation drug candidate for each of the four leading classes of chemotherapeutic agents.

Treatment related toxicities.toxicities.    The majority of current chemotherapy agents kill cancer cells by disrupting the cell division process. Chemotherapy drugs disrupt the process by killing cells once they begin to undergo division and replication.replication process. Although this mechanism often works in cancer cells, which grow rapidly through cell division, non-cancerous cells are also killed because they too undergo routine cell division. This is especially true for cells that line the mouth, stomach and intestines, hair follicles, blood cells and reproductive cells (sperm and ovum). Because the mechanism by which conventional cancer drugs work is not limited to cancer cells, their use is often accompanied by toxicities. These toxicities limit the effectiveness of cancer drugs and seriously impact patients’the patient’s quality of life.

Inability to selectively target tumor tissue.tissue.    When administered, chemotherapy drugs circulate through the bloodstream, reaching both tumor and normal tissues. Normal dividing tissues are generally as sensitive as tumor cells to the killing effects of chemotherapy. Thesechemotherapy and toxic side effects on normal tissues prevent use of higher, potentially more effective, doses of chemotherapy.limit the treatment that can be given to patients with cancer.

Chemotherapy resistance.resistance.    Resistance to the cancer killing effects of conventional chemotherapy drugs is a major impediment to continued effective treatment of cancer. Approximately 70% of all cancer patients undergoing chemotherapy ultimately develop resistance to one or more chemotherapy agents and eventually die from their disease. Because many chemotherapy drugs share similar properties, when a tumor develops resistance to a single drug, it may become resistant to many other drugs as well. Drugs that work differently from existing chemotherapies, and are notless susceptible to the same mechanisms of resistance, couldhave begun to play a veryan important role in treating resistant tumors.

Strategy

Our goal is to become a leading cancer drug company. The following are the key elements of our business strategy:

Products in Development

The following table lists active trials (indicated by a status of “open”), the trials that will be opened to enrollment during the secondfirst quarter of 20032004 (status “2Q2003”“1Q04”) for our products in development, and a phase III trialthe trials that will be opened to enrollment during the thirdsecond quarter of 20032004 (status “3Q2003”“2Q04”) and the trials that will be opened to enrollment during the first half of 2004 (“1H04”). Also listed are the trials that have recently closed to enrollment but for which clinical trial reports are in progress (status “enrollment completed”). In addition to clinical trials that are part of our registration strategy, we also assist clinical investigators who request our help for their independent investigations that advance clinical knowledge of the use of our products. These “Investigator-Sponsored Trials” are indicated on the following table by an “*”.

Product Candidate		Indication/Intended Use		Phase/Status
		Other Leukemia/Lymphoma
		ATO in combinationCombination with ascorbic acid for relapsed/refractory lymphoid malignanciesmalignancies*		II / open
		Single agent in cutaneous T-cell lymphoma*		II / open
		SOLID TUMORS
		ATO in neuroblastomas and other solid tumors in pediatric patientsSingle agent for relapsed/refractory metastatic melanoma*		II / open
		Hepatocellular carcinomaSingle agent for metastatic melanoma*		III / open
		Advanced cancer patientsCombination with renal dysfunctionascorbic acid and temozolomide in melanoma*		II/II / open
		ATO single agent for metastatic melanoma		II/ open
		ATO in combinationCombination with ascorbic acid for metastatic colon cancer refractory to chemotherapychemotherapy*		II / open
		ATO single agent in cutaneous T-cell lymphoma		I / open
		ATO in combination with docetaxel for hormone-refractory prostate cancer		II/ 2Q2003
		ATO single agent for relapsed/refractory metastatic melanoma		II /2Q2003
		ATO in combination with ascorbic acid for hepatocellular carcinoma		II / 2Q2003
		ATO in combination with 5-fluorouracil for gastrointestinal tract tumor		II / 2Q2003
		ATO singleSingle agent for advanced carcinoma of the breastbreast*		II / 2Q2003 open
		ATO singleSingle agent for advanced non-small cell carcinoma of the lunglung*		II / 2Q2003 open
		ATOSingle agent for neuroblastomas and other solid tumors in combinationpediatric patients*		II / open
		Hepatocellular carcinoma		I / open
		Advanced cancer patients with renal dysfunction		I / open
		Combination with DTIC for malignant melanomamelanoma*		I/II / open
		Combination with radiosurgery/radiotherapy for recurrent malignant glioma*		I /II/ open
		Pharmacokinetics in patients with APL		I / 2Q2003open
		Combination with docetaxel for prostate cancer*		II/ 1Q04
XYOTAX		Non-small-cell lung cancer (second-line; “STELLAR 2”)		III / open
(CT-2103)		Non-small-cell lung cancer in combination with carboplatin (front-line PS2; “STELLAR3”)		III / enrollment completed
		Non-small-cell lung cancer (front-line PS2; “STELLAR4”)		III / open
		Ovarian front-line maintenance (GOG)		III /1H04
		(CT-2103)Ovarian³ second relapse (GOG)		II / open
		Advanced solid tumors—Dosing every 3 weeks (UK)Ovarian front-line dose escalation (GOG)		II/II / enrollment completedopen
		Ovarian 1st line in combination with carboplatin		II / open
		Advanced solid tumors in combination with cisplatin		I / open
		Advanced solid tumors in combination with carboplatin		I / open
		Advanced solid tumors, single agent—dosing every week (US)(U.S.)		I / open
		Advanced solid tumors, single agent—dosing every 2 weeks (UK)		I / open
		Advanced solid tumors, single agent—dosing every 3 weeks (US)(U.S.)		I / open enrollment completed
		Non-small-cell lung cancer salvage, single agent		I / open enrollment completed
		Ovarian front-line dose escalation (GOG)Advanced solid tumors—Dosing every 3 weeks (UK)		I/III / 2Q2003 enrollment completed
		Ovarian, fallopian tube, peritoneal carcinoma—salvageAdvanced solid tumors combined with carboplatin		III / enrollment completed
		ColorectalCombination with cisplatin and radiation for esophageal and gastric cancer salvage		III / open
		Non-small-cell lung cancer (high risk patients)		II / open
		Ovarian second relapse (GOG)		II / open
		Ovarian front-line (GOG)		III/ 3Q2003
		Lung cancer, in combination with radiation		I / 2Q2003open

Product Candidate		Indication/Intended Use		Phase/Status
Pixantrone		Aggressive NHL, > 3 relapses, single agent		III / 1H04
		Breast cancer (Cancer Research UK)Relapsed indolent NHL, pixantrone +/- Rituxan		IIII/III / open closing**
		Non-small-cell lung cancer (second line; multinational)Relapsed aggressive NHL, BSHAP		IIIII / open
		Non-small-cell lung cancer in combination with carboplatin (first line PS=2; multinational)Aggressive NHL, front-line, CPOP-R		IIIII / open 2Q04
		Non-small-cell lung cancer (first line PS=2; multinational)Relapsed AML, single agent		III / open
PG-CPT (CT-2106)		Advanced solid tumors		I / open 2Q04
		Relapsed indolent NHL, FND-R		I/II / enrollment completed
		Relapsed aggressive NHL, BSHAP		I / enrollment completed
		Relapsed aggressive NHL, CPOP		I/II / open
		Multiple sclerosis, single agent, RR or PR		I / 1H04
CT-2106		Phase I Q3wk Dosing		I / open
		Phase I Q1wk Dosing		I / open
		Small cell lung cancer		II / 2Q04
		Relapsed ovarian cancer		II / 2Q04
		Relapsed colorectal cancer		II / 1H04

**

As a result of our strategy to conduct a pivotal phase III trial in aggressive NHL, which we believe provides the fastest route to registration for pixantrone, the ongoing phase III trial of pixantrone in indolent NHL has been modified and reduced to a study to support registration and market development.

TRISENOX (arsenic trioxide) injection

We are marketing TRISENOX for the treatment of patients with chemotherapy resistant or relapsed APL. We received FDA approval in this indication in September 2000, and in March 2002, we received EMEA approval in the EU with a subsequent launch in the EU in June 2002. TRISENOX is a highly purified version of arsenic, a natural element. TRISENOX appears to have multiple targets and mechanisms of antileukemic activity: it degrades a protein that causes abnormal levels of immature white blood cells while simultaneously forcing immature cancer cells to self destruct through a process called programmed cell death or apoptosis. Apoptosis is a normal part of a cell’s life cycle. Because cancer is often associated with a malfunction of the normal process of apoptosis, drugs that can induce apoptosis offer the hope of affecting cancer cells more selectively without the typical toxic side effects of conventional treatments. Direct induction of apoptosis represents a new method of killing tumor cells that is different from that of the majority of conventional cancer drugs. As a result, in addition to its use as single agent therapy, TRISENOX may work well when administered in combination with other cancer therapies to produce more durable cancer response rates.

We intend to protect TRISENOX by obtaining orphan drug marketing exclusivity in the U.S. and Europe. When granted orphan drug status, products usually receive seven years of marketing exclusivity in the U.S. and ten years in the EU. If a product with an orphan drug designation subsequently receives the first FDA or EMEA approval for the indication for which it has such designation, the product is entitled to orphan drug marketing exclusivity, meaning that the regulatory agency may not approve any other applications to market the same drug for the same indication, except in certain very limited circumstances, for a period of seven or ten years. We have received U.S. orphan drug marketing exclusivity for TRISENOX in APL and have received U.S. orphan drug designation for TRISENOX for the treatment of multiple myeloma, MDS, CML, AML, CLL and AML.HCC. TRISENOX has received orphan drug designation for the treatment of APL, multiple myeloma, and MDS under the European orphan drug regulation. In addition, we recently received a Notice of Allowance from the U.S. Patent and Trademark Office for a patent application that would, upon issuance, extend our U.S. intellectual property position for TRISENOX through 2018, which is 11 years beyond the original orphan drug designation that currently expires in 2007. We believe further investments in registration directed trials for various blood related cancers and solid tumors such as melanoma, prostate cancer and brain cancer could accelerate TRISENOX sales growth and increase the drug’s market potential. We also plan to pursue orphan designation for other indications. In addition, we have exclusive rights to several patent applications filed by PolaRx Biopharmaceuticals, Inc., or PolaRx, and The Memorial Sloan-Kettering Cancer Center that are directed to

methods of treating a variety of cancers and conditions with TRISENOX, and a patent application from the Samuel Waxman Cancer Research Foundation that is directed to certain formulations of arsenic trioxide.

We commenced sales of TRISENOX in October 2000, and currently market TRISENOX in the U.S. and in the EU. For the year ended December 31, 2003 we recorded $22.1 million in TRISENOX sales, an increase of 94% over 2002. TRISENOX is approved for patients with a type of blood cell cancer called APL who have relapsed or failed standard therapies. However, most TRISENOX sales result from treatment of multiple myeloma and Myelodysplastic Syndromes, or MDS. There are over 15,000 new cases of multiple myeloma diagnosed per year in the U.S. We are currently conducting or plan to conduct more than 70 clinical and investigator-sponsored trials addressing a variety of blood related cancers, including multiple myeloma and MDS, as well as solid tumors.

TRISENOX for Acute Promyelocytic Leukemia.Leukemia or APL.    APL is a malignant disorder of the white blood cells that can occur across all age groups. Based on ACS data, approximately 1,5001,200 to 2,0001,800 patients are diagnosed with APL each year in the United States, with a similar incidence in the EU. Current treatment for newly diagnosed APL patients includes the use of all-trans retinoic acid, commonly called ATRA, in combination with anthracycline chemotherapy. Between 10% to 15% of patients die during front linefront-line therapy, some patients will have long-term toxicity due to anthracycline treatment, and up to 30% of patients who achieve initial remission will eventually relapse. After relapse, the long-term outlook for these patients is poor.

TRISENOX has been investigated in relapsed and refractory APL patients, previously treated with an anthracycline and retinoid regimen, in two open label studies. One was a single investigator clinical, or pilot, trial involving 12 patients and the other was a multicenter, nine-institution study, or pivotal, trial of 40 patients. The pilot trial results and accompanying editorial describing the use of TRISENOX to treat patients with relapsed APL were published in the November 5, 1998 issue ofThe New England Journal of Medicine. The results of this study were confirmed by the pivotal trial that was published in September 2001 inThe Journal of Clinical Oncology. Long term follow up data from the multicenter study werewas presented at the 8^th8th International Symposium on APL in Rome, Italy.Italy in October 2001. The results demonstrated that among the 85% of patients who achieved a complete remission,CR, 82% were confirmed to have a molecular remission using a highly sensitive molecular test. With a median follow up of 30 months, the overall survival estimate for the 52 patients in these two studies is 66%.

Side effects of TRISENOX noted in these studies were generally manageable, usually did not require interruption of therapy, and most patients were treated as outpatients once the serious symptoms of their APL were resolved. The most common side effects

included nausea, cough, fatigue, headache, vomiting, abdominal pain, diarrhea, shortness of breath, leukocytosis (an increase in the number of white blood cells in circulation), hyperglycemia (increased blood sugar), rash, prolongation of the QT interval (an asymptomatic change in electrocardiogram, or EKG)electrocardiogram), edema (water retention), and dizziness.

TRISENOX for Multiple Myeloma.    Multiple myeloma is a malignant disease of the bone marrow that is invariably fatal. According to the Multiple Myeloma Research Foundation, multiple myeloma is the second most common blood cell malignancy, affecting nearly 50,000 people in the United StatesU.S. with over 14,00015,000 new cases reported annually. The disease is initially treated with oral chemotherapy drugs. Once the disease can no longer be controlled with oral drugs, treatments include high dose corticosteroids, high dose chemotherapy, a combination of high dose chemotherapy and stem cell transplants, and recently thalidomide. Approximately 50% of myeloma patients respond to these treatments.

Preclinical studies have suggested that TRISENOX may be able to kill multiple myeloma cells taken from chemotherapy-resistant patients and that the killing may be enhanced when TRISENOX is combined with vitamin C (ascorbic acid), corticosteroids, or other agents used to treat myeloma. Preliminary reports from threetwo clinical studies and a series of case studies using TRISENOX, alone or in combination with other agents such as ascorbic acid, in patients with myeloma who had failed multiple prior therapies showed encouraging responses asresponses. In general, the combinations were well tolerated with no reported grade 4 toxicities. These findings were reported

at the 9^th International Workshop on Multiple Myeloma in May 2003, at the American Society of Clinical Oncology, or ASCO, meeting in June 2003, at the 8^th Congress of the European Hematology Association, or EHA, also in June, and at the American Society of Hematology, or ASH, meeting in December 2002. We2003.We are sponsoring several multicenter trials with TRISENOX used either as a single agent or in combination with corticosteroids, ascorbic acid, or thalidomide for advanced stages of multiple myeloma. TRISENOX has received orphan drug designation from the FDA and the EMEA for this indication.

TRISENOX for Myelodysplastic Syndrome, or MDS.    MDS is a preleukemic condition affecting about 35,00036,000 individuals a year with an annual incidence of 15,000 patients a year. Many patients who develop MDS progress to develop acute leukemia. All patients have a progressive decline in their ability to make blood cells, ultimately resulting in anemia requiring red blood cell transfusions, a low white blood cell count placing them at risk for infections, and a low platelet count making them prone to bleeding. There is no specific approved therapy for this disorder except supportive care and the use of growth factors such as Procritepoetin alfa and Leukine.sargramostim. Reports from threetwo clinical studies, and a series of case studies using TRISENOX, alone or in combination with other agents such as Vitamin C or with dexamethasone, in high and low-risk myelodysplastic syndrome patients showed encouraging responses as reported at ASHthe 7^th Annual International Symposium on MDS in December 2002.2003 as well as at ASCO, at EHA and at ASH. Drug related adverse events were generally manageable and resolved after completion of therapy. Additional trials exploring the activity of TRISENOX, alone or in combination with growth factors, thalidomide and Ara-C have been initiated, and preliminary data is encouraging. Orphan drug designation has been received from both the FDA and the EMEA.

TRISENOX for Chronic Myeloid Leukemia, or CML.    CML is a form of leukemia affecting approximately 16,00026,000 individuals in the U.S. and has an annual incidence of 4,0004,600 patients per year. It is caused by a highly specific chromosomal rearrangement that produces an abnormal fusion gene called the bcr-abl (this is similar to the cause of APL, which results from a different chromosomal rearrangement). A dramatic advance was recently made in the treatment of CML in 2001 with the approval of Gleevec, a new drug that specifically targets and inactivates the bcr-abl gene product. Gleevec can induce durable clinical remissions in a very high percentage of patients with early stage CML. Although it is active in patients with later stages of the disease, termed accelerated phase or blast crisis, the remissions are short-lived as resistance to Gleevec develops. There is a need to identify drugs that will enhance the efficacy of Gleevec in advanced stages of CML and in particular, prevent the emergence of resistance. Two publications indicateData reviewed as part of abstracts presented at the ASH (2002) and at the Advances in Cancer Differentiation Therapy Symposia, Shanghai, China (2000), suggest that TRISENOX may be the idealan agent to use with Gleevec for the following reasons:

Trials exploring the use of TRISENOX in conjunction with Gleevec in both early and later stages of CML are in progress or are about to begin.

TRISENOX for Other Hematologic Malignancies.    A number of other cancers of blood and lymphatic organs are under study with TRISENOX including lymphomas and leukemias. Non-Hodgkin’s lymphomaleukemias, including NHL, which has affected more than 300,000350,000 people in the U.S. and there are approximately 54,000 new cases per year according to the American Cancer Society.ACS. Despite new effective therapies, relatively few patients are cured and additional treatments are needed. Studies are currently in progress to evaluate the activity of TRISENOX as a single agent and in combination with standard therapies for lymphoma.

TRISENOX for Solid Tumors.    Solid tumors include malignancies that develop in various tissues throughout the body, as opposed to hematologic cancers described above. Genitourinary cancers, such as

cervical, renal cell, bladder and prostate, have affected approximately three million patients in the United States,U.S., with over 370,000350,000 new cases diagnosed annually. Preclinical data and preliminary clinical trial results have suggested that TRISENOX may have anti-tumor activity in a number of solid tumors including cancers of the ovary, prostate, bladder, liver, lung and melanoma. EarlyPreliminary data from phase I and II studies show evidence forindicate clinical activity in prostate cancer, melanoma and neuroblastoma. A number of other studies looking at TRISENOX as a single agent and in combination with standard therapy in patients with solid tumors are either underway or planned.

Polyglutamate Drug Delivery Technology

We are also developinglicensed the worldwide exclusive rights to polyglutamate (PG) and related polymers with their applications to anticancer drugs from PG-TXL Company, L.P. in 1998. This technology was originally developed at the M.D. Anderson Cancer Center. The PG technology represents a new way to deliver cancer drugs more selectively to tumor tissue with the goal of reducing the toxic side effects and improving the anti-tumor activity of existing chemotherapy agents. Our technology links cancer drugs to proprietary polymers, such as polyglutamate. Polyglutamate, which we call PG is a biodegradable polymer of glutamic acid, a naturally occurring amino acid. To build a PG drug, we link glutamic acid molecules together to an optimal size.size and then link them to active drug agents. We believe the polymer technology takes advantage of a well-described difference between tumor blood vessels and blood vessels in normal tissues. The blood vessels in tumor tissues are more porous than those in normal tissues, and they are therefore more permeable to large molecules, such as our polymers, that are within a specific size range. As the polymer, carrying its tumor-killing drug, circulates in the bloodstream and passes through the tumor blood vessels, it becomes trapped in the tumor tissue allowing a significantly greater percentage of the anti-cancer drug to accumulate in tumor tissue compared to normal tissue. The toxicity of the chemotherapy drug to normal tissues also may be reduced because the drug appears to beis inactive as long as it is bound to the polymer. Once the polymer backbone is digested in the tumor, the cancer-killing drug is released directly into the cancer tissues.

Based on observations from preclinical animal studies, along with early clinical trial data from phase I and phase II studies, we believe that our polymer-chemotherapy drug conjugates may be able to achieve a number of benefits over existing chemotherapy drugs, including:

In addition, we believe that linking our polymers to existing drugs will yield patentable subject matter and that our polymer-drug conjugates will not infringe any third party patents covering the underlying drug. However, there can be no assurance that we will receive a patent for our polymer conjugates or that we will not be challenged by the holder of a patent covering the underlying drug.

We licensed the worldwide exclusive rights to PG and related polymers with their applications to anticancer drugs from PG-TXL Company, L.P. in 1998. This technology was originally developed at the M.D. Anderson Cancer Center. The initial patent, which issued in November 1999, covers compositions of PG and related polymers coupled with commonly used cancer drugs such as paclitaxel, docetaxel, etoposide, teniposide, or camptothecins. The patented technology also covers formulations of PG-conjugated paclitaxel that also include the use of human serum albumin and conjugation to epothilones.

Our strategy is to use this novel polymer technology to build a portfolio of potentially safer and more effective versions of well-known anti-cancer agents. We believe that our polymer drug development program may lower the risks inherent in developing new drugs because we are linking polymers to well defined and widely used chemotherapy drugs. We are initially focusing our development efforts on applying PG to two of the fastest growing classes of anticancer drugs, taxanes and camptothecins.

XYOTAX (paclitaxel poliglutamex).    XYOTAX, or CT-2103, utilizes a biodegradable protein polymer to deliver the chemotherapy drug paclitaxel more selectively to tumor tissue, for the treatment of non-small cell lung cancer, ovarian cancer and other cancers where taxanes are widely used. Taxanes are one of the best-selling classes of chemotherapy drugs. Paclitaxel, one of two marketed taxanes, is branded as paclitaxel and is approved for the treatment of NSCLC, ovarian cancer and breast cancer, although its widest usage is in lung and ovarian cancers. XYOTAX is PG linked to paclitaxel, the active ingredient in Taxol, one of the world’s best selling cancer drugs. Taxolpaclitaxel. Paclitaxel is difficult to administer because it is formulated in Cremophor, a mixture of castor oil and ethanol, which is extremely irritating to blood vessels and requires surgical placement of a large catheter for administration. It also may cause allergic reactions, and typically requires a minimum of three hours of intravenous infusion. XYOTAX is 80,000 times more water-soluble than paclitaxel, allowing it to be dissolved in 100 mLML of a simple water and sugar based solution and infused over ten minutes. Also, because XYOTAX is water-soluble, its administration does not require routine premedication with steroids and antihistamines to prevent severe allergic reactions; such premedication can be reserved for those patients who show signs of sensitivity during treatment. XYOTAX may also allow delivery of higher doses than can be achieved with paclitaxel.

It is estimated that in the United States, more than 3.8 million people have breast, ovarian, lung and coloncolorectal cancer, with more than 550,000 new cases diagnosed each year in the United States.year. IMS Health reported taxane U.S. sales of approximately $934$881 million, and worldwide sales of roughly $1.7$2.5 billion for 2002,2003, despite the difficulties associated with their administration and their serious dose-limiting toxicities. The majority of taxane use has been in breast, ovarian and lung cancer indications.

XYOTAX has been compared to paclitaxel in numerousmultiple animal studies in animals with a variety of different tumors. These animal studies indicate that XYOTAX has a unique profile resulting in better tolerability and efficacy, both when used by itself as a single agent or in combination with other chemotherapy, radiation therapy, or therapeutic monoclonal antibodies. Specifically:In preclinical animal studies:

Lastly, based on additional preclinical data, XYOTAX is more effective than standard paclitaxel at enhancing the effectiveness of other cancer therapies including chemotherapy and radiation. A recentMarch 2003 report in the International Journal of Radiation Oncology Biology-Physics shows that in a curative, standard radiation model, XYOTAX selectively improved tumor response to radiation by up to eightfold after single-dose or fractionated irradiation. Most importantly, unlike standard paclitaxel, XYOTAX did not sensitize normal

organs such as skin, hair follicles, or the gastrointestinal tract to radiation. A grant from the National Cancer Institute to the MD Anderson Cancer Center and usCTI will support a clinical trial using XYOTAX in sensitive patients undergoing potentially curative radiation for lung cancer.

We chose to initiate clinical trials of XYOTAX in the U.K. because of the Cancer Research Campaign’s, or CRC’s, experience with polymer drug conjugates and because of the ability to perform trials in patients who had not received a taxane. The phase I clinical trial of XYOTAX sponsored by the Cancer Research Campaign, or CRC, has completed patient enrollment. Preliminary data presented by the investigators showedenrollment and has been reported to show that XYOTAX may have a more favorable toxicity profile than expected from equivalent doses of Taxol,paclitaxel, while demonstrating evidence of anti-tumor activity across a variety of types of cancer.cancer including gastric cancer and mesothelioma, tumors that are intrinsically resistant to taxanes.

Based on the preliminary data generated in the phase I CRC trial, and following discussions with a number of opinion leaders and cooperative groups, we initiated an aggressivephase II development program for XYOTAX in the USU.S. and other countries. SixEight phase I clinical trials, fourthree phase II clinical trials and three phase III clinical trials are currently underway, with three additional trials planned this year. Some of these ongoingunderway. Most studies use XYOTAX at doses in excess of the approved dose for Taxol and all are scheduled to useutilize a convenient 10-minute infusion time.

Our registrationdevelopment strategy for XYOTAX is to examine its potential safety and efficacy as single agent therapy or in combination with other chemotherapy drugs in solid tumors. The phase III trials in non-small-cell lung cancer patients are intended to more fully characterizedemonstrate the safety and efficacy of XYOTAX.XYOTAX compared to current treatment standards. Based upon a successful conclusion, to these trials, we anticipate submitting a New Drug Application, or NDA, inat the second halfend of 2004.

In 2002, we announced that the Gynecologic Oncology Group, or GOG plans to conduct a phase III trial of XYOTAX in front-line treatment of ovarian cancer.cancer We anticipate that the GOG will submit a phase III trial protocol through a SPA to the FDA to compare XYOTAX as maintenance therapy, administered monthly for 12 months, to no maintenance treatment for ovarian cancer patients who have achieved a CR following front-line treatment with carboplatin and paclitaxel. This trial is expected to begin in the thirdfirst half of 2004.

Based on our phase I and II data from over 400 patients, we believe that XYOTAX may have less severe side effects, including a reduction in severe neutropenia, allergic reactions and hair loss, and superior anti-tumor activity than marketed taxanes. In a phase II study of XYOTAX for the front-line treatment of advanced NSCLC patients who were either over 70 years old or performance status 2, or PS2, we observed a median survival time of 5.4 months among PS2 patients, which compares favorably to the 2.4 months reported in a separate randomized trial of standard paclitaxel. In the fourth quarter of 2003. We had2002, we initiated athree pivotal XYOTAX phase III ovarian clinical trials. These include two phase III trials of XYOTAX for the front-line treatment of poor PS2, NSCLC patients and one phase III trial for the second-line treatment of NSCLC.

The following table summarizes our ongoing pivotal trials for XYOTAX.

In November 2003 we completed enrollment in STELLAR 3, our first pivotal trial in July 2002, butlung cancer. We anticipate completing enrollment for STELLAR 2 and STELLAR 4, the other two pivotal XYOTAX lung cancer trials, by mid-2004. We anticipate releasing clinical data from STELLAR 3 and STELLAR 4 in the second half

of 2004 (“2H04”) and releasing clinical data from STELLAR 2 by mid-2005. We have since closed thisfollowed the FDA guidelines for SPA on all of our ongoing pivotal trials. Based on the positive outcome of the first of these phase III trials (STELLAR 3) and supporting data from phase II studies, it is anticipated that an NDA will be filed at the end of 2004.

Lung cancer is the most common cause of cancer death in the U.S. and Europe. The ACS estimates that 139,000 new cases of non-small cell lung cancer will be diagnosed in the U.S. in 2004 and approximately 106,000 of these are expected to receive chemotherapy in 2004. Of these 106,000 patients approximately 30,000 are classified as PS2. These patients tolerate chemotherapy poorly and have a significantly shorter median survival than healthier patients. The chemotherapy drug most commonly used to treat NSCLC in the U.S. is paclitaxel. Data from a randomized trial of paclitaxel in NSCLC showed a median survival of approximately 2.4 months in front-line therapy of PS2 patients when administered as a resultsingle agent and 4.7 months when administered in combination with platinum-containing chemotherapy drugs. Approximately 55,000 patients in the U.S. receive second line treatment for lung cancer U.S. annually, for which docetaxel is the most commonly used agent to treat recurrent NSCLC.

In June 1998, we licensed exclusive worldwide rights to paclitaxel linked to polyglutamate, branded as XYOTAX, and to all potential uses of the GOG’s decision to conduct the trial for us.polyglutamate polymer technology.

PG-CPT (polyglutamate camptothecin).    PG-CPT,In October 2003, following a planned safety analysis by an independent Data Monitoring Committee of our three XYOTAX pivotal trials, we reduced the dose of XYOTAX from 235mg/m² to 175mg/m² in the STELLAR 4 trial. This change was based on a small percentage of patients in the study who appeared to develop early (first- or CT-2106,second-cycle) neutropenic-related toxicities when compared to the number of patients in the gemcitabine comparison arm. While the incidence of grade 4 neutropenia observed in our phase II trials of 235mg/m² XYOTAX is substantially lower than that reported in the label for the equivalent dose of paclitaxel (25% vs 50% respectively), when compared to the reported incidence of gemcitabine 6% (label claim), the observed higher occurrence of early neutropenic events in the XYOTAX arm was not surprising. Since the incidence of neutropenia is low at the 175mg/m² dose of XYOTAX (2%) we believe lowering the dose in STELLAR 4 should prevent any increase in early neutropenic related events relative to the gemcitabine comparator arm. The 175mg/m² dose was well tolerated in prior phase II studies in more than 125 patients and resulted in encouraging duration of median survival in the phase II NSCLC study. All patients currently enrolled in STELLAR 4 are continuing their treatment at the lower dose of 175mg/m², while the comparator arm dosages will continue according to the protocol at their approved marketed dose. The recommendation of the Data Monitoring Committee did not impact the STELLAR 3 trial, which completed enrollment last year With this dose adjustment, we do not anticipate there being an effect on the integrity or utility of these pivotal studies for registration since the single-agent dose of 175mg/m² of XYOTAX is standardized across all PS2 patients, including PS2 patients in the STELLAR 2 pivotal trial.

At last year’s European Cancer Conference, meeting, we reported data on a camptothecin linkedphase II study of XYOTAX in the front-line treatment of advanced NSCLC patients who were either over 70 years old and/or PS2. In the first part of the study, 28 patients were treated with XYOTAX every 21 days. Using standard criteria to PG. Camptothecinsassess efficacy, 18 patients (64%) achieved disease control, with two patients (7%) achieving a partial remission and 16 patients (57%) having stable disease. XYOTAX therapy was well tolerated with 50% of patients receiving four or more cycles of therapy and almost 30% of patients receiving six or more cycles. No alopecia or hypersensitivity reactions, which are an importantcommonly seen with standard paclitaxel formulations, were reported. Only one patient experienced grade 4 neutropenia and rapidly growing class of anti-cancer drugs. However, like taxanes, their full clinical benefit is limited by poor solubilityfour patients reported grade 3 neuropathy, which occurred mostly in patients with concomitant progressive disease and significant toxicity. To avert solubility limitations, oral analogs such as Hycamtindisease-related comorbid conditions. We observed a median survival time of 5.4 months among PS2 patients, which compares favorably to the 2.4 months reported in a separate randomized trial of standard paclitaxel.

Ovarian cancer is diagnosed in approximately 26,000 women per year in the U.S. The standard of care for front-line treatment of ovarian cancer is paclitaxel and Camptosar were developed. However, conversioncarboplatin. In the fourth quarter of 2002, we announced

that the GOG agreed to oral dosage forms has been accompanied byconduct a reduction in anti-tumor potency. Despite these limitations, camptothecins are becoming standard drugsphase III trial of XYOTAX in the treatment of advanced colon, lung and ovarian cancer. Worldwide salesWe anticipate the GOG will submit a phase III trial protocol through an SPA to the FDA to compare XYOTAX as maintenance therapy, administered monthly for camptothecins exceeded $970 million12 months, to no maintenance treatment for patients who have achieved CR following front-line treatment with carboplatin and paclitaxel. The planned primary endpoint of the study will be progression free survival with overall survival as a secondary endpoint. A prior study of monthly paclitaxel maintenance for 12 months significantly increased progression-free survival, however, toxicities prevented full dose treatment with paclitaxel for the majority of patients in 2002.this trial. Based on discussion with the GOG, we believe the pivotal trial will begin in the first half of 2004. We are also currently conducting three other phase II clinical trials of XYOTAX in ovarian cancer.

Pixantrone

We are developing pixantrone for the treatment of NHL. Preclinical data and clinical studies in over 175 patients indicate that pixantrone is easy to administer, may exhibit significantly lower potential for cardiac toxicity and may have more potent anti-tumor activity than marketed anthracyclines. We are currently conducting or plan to conduct one phase III trial, three phase II trials, one phase II/III trial in indolent NHL and two phase I trials. The trial in indolent NHL has been modified and reduced to a registration supporting study based on our strategy to conduct a pivotal phase III trial in aggressive NHL, which we believe provides the fastest route to registration for pixantrone. We intend to begin a randomized pivotal phase III trial for relapsed, aggressive NHL during the first half of 2004. We expect to have data from the pivotal trial and, if the trial is successful, we intend to file an NDA for pixantrone in late 2005 or early 2006. We believe pixantrone complements our hematology and oncology expertise and our growing TRISENOX commercial franchise.

Anthracyclines are one of the most potent classes of anti-cancer agents used in front-line treatment of aggressive lymphoma, leukemia and breast cancer. For these diseases, anthracycline-containing chemotherapy regimens can often produce long-term cancer remissions. However, the currently marketed anthracyclines can cause severe, permanent and life threatening cardiac toxicity when administered beyond widely recognized cumulative lifetime doses. This toxicity prevents repeat use of anthracyclines in the 70% of patients who relapse after front-line anthracycline treatment. In addition, the cardiac toxicity of anthracyclines prevents their use in combination with other drugs that also can cause cardiac toxicity. For example, Herceptin and anthracyclines are both used for the treatment of breast cancer, but overlapping cardiac toxicity prevents these agents from being used together. An anthracycline with lower cardiac toxicity could potentially allow the use of both drugs in combination.

We believe a next-generation anthracycline with better ease of administration, greater anti-tumor activity and less cardiac toxicity could gain a significant share of the anthracycline market. We also believe such a drug could expand anthracycline use by allowing repeat anthracycline therapy in relapsed patients and by allowing combination therapy with a broader range of chemotherapy drugs.

Recently, we met with the FDA and discussed the design for a pivotal clinical trial of single-agent pixantrone in the treatment of third-line relapsed aggressive NHL. A SPA package was filed and we expect to initiate a pivotal trial of pixantrone for relapsed NHL in approximately 320 patients in the first half of 2004. We believe such a trial would qualify for accelerated review and approval under current FDA guidelines. Enrollment is expected to take approximately one year and assuming a positive outcome for this trial we would intend to file an NDA in late 2005 or early 2006.

In a phase II trial published in the journal Hematologica in August 2003, among patients with relapsed aggressive NHL who failed a median of two or more prior regimens including prior anthracycline therapy, single-agent pixantrone produced an objective tumor response in 9 of 33 patients (27%) with 5 patients experiencing a complete response. Median duration of response was encouraging, in some cases lasting over 17 months. Pixantrone was well tolerated in this trial with neutropenia being the most frequently reported side effect. Cardiac symptoms were infrequent with only three patients experiencing a decrease of more than 10% of the left ventricular ejection fraction, or LVEF, a marker of cardiac function, which was possibly treatment-related. We believe that the low incidence of cardiac toxicity reported in this trial was encouraging because

patients had previously been exposed to anthracylcines, an equivalent dose of high cumulative doses of doxorubicin.

We have also reported promising clinical data for pixantrone in combination with other agents that are used in the treatment of lymphoma. In one trial, pixantrone was used to replace the standard anthracycline agent doxorubicin as part of the CHOP (a combination-chemotherapy regimen consisting of cyclophosphamide, doxorubicin (an anthracycline), vincristine and predisone which is the standard of care for front-line treatment) regimen. The CHOP regimen is highly active when used in front-line patients, but for patients who relapse following front-line CHOP therapy further treatment with doxorubicin may not be possible due to the high risk of cardiac toxicity. Preliminary results from this trial, which were presented at the 2003 ASH meeting, revealed that of the 22 patients evaluable for a response, 13 patients (59%) had a complete response, with four patients experiencing a partial response. This corresponds to a major objective response rate of 77%. At the highest pixantrone dose (150 mg/m²), all seven patients responded, with six patients (86%) experiencing a complete response. All patients had failed prior CHOP therapy for relapsed NHL. The median prior exposure to anthracyclines in this patient group was 560 mg/m². Despite prior heavy anthracycline exposure, only one patient with a prior history of heart disease reported symptoms of chest pain. A phase II study at the highest dose level studied is ongoing in up to 30 additional patients.

We have also conducted clinical trials for pixantrone using a variant of the ESHAP (a regimen consisting of etoposide, methylprednisdone, cytarabine and cisplatin), termed the BSHAP regimen, a non-anthracycline multi-agent regimen developed as a second line therapy for patients who fail front-line CHOP and who are not able to receive further anthracycline treatment. In this modified regimen, pixantrone replaces etoposide, with a goal to improve efficacy. In this trial, 11 of 18 (61%) of evaluable patients achieved an objective tumor response with six patients (33%) achieving a complete response. No clinically significant cardiac events were observed in this trial and no patient experienced a decrease in LVEF of more than 20%.

Other clinical data suggest the product may also be useful in treating indolent NHL a less rapidly progressive but none the less ultimately fatal form of NHL. In one trial pixantrone was administered in a variation of the FND-R regimen, where pixantrone replaces the anthracycline derivative mitoxantrone. Preliminary results reveal that all six evaluable patients in this trial achieved an objective response. A randomized phase III clinical trial designed to compare the efficacy and safety of pixantrone administered in combination with the monoclonal antibody Rituxan, to Rituxan alone, is ongoing. This trial is currently accruing patients in the U.S. and Europe and is expected to support the marketing of pixantrone, as well as to provide pharmacokinetic and additional safety data in the indolent NHL setting.

In the U.S., aggressive relapsed NHL affects over 160,000 people with approximately 30,000 new cases diagnosed per year. The standard of care for front-line treatment is known as CHOP. This regimen is used either alone or in conjunction with Rituxan, and is able to induce CRs in up to half of patients. However, upon relapse, many patients are unable to undergo an additional course of CHOP therapy due to the risk of cardiac toxicity, and other, less effective chemotherapy regimens are often used for the second-line treatment of relapsed NHL. There are no approved treatments for second or third line treatment for the 73,000 patients with relapsed aggressive NHL in the U.S.

A pilot study in multiple sclerosis is planned to start in the first half of 2004. Preliminary data from this phase I study will be used to guide future business development strategies in 2005.

CT-2106 (polyglutamate camptothecin)

We are also developing a novel polyglutamate-camptothecin molecule, CT-2106. Linking a camptothecin to PGour polyglutamate polymer renders it water soluble, and animal studies suggest that it permits up to 400% more drug to be administered without an increase in toxicity. PG-CPTCT-2106 as a single agent and/or in combination with 5FU showed significantly enhanced anti-tumor activity in several animal models of lung, colon and breast cancer, with up to 500% improvement over the free drug.tumor models. We have optimizedinitiated a polyglutamate camptothecin for clinical development and filed an IND in December 2001. A phase I clinical trial of PG-CPTstudy for this product candidate in 2002. CT-2106 has been administered to 29 patients with a variety of advanced stage cancers was initiated inwith encouraging preliminary safety data with the first quarter of 2002,dose limiting toxicity observed being

neutropenia and thrombocytopenia. Notably patients did not experience severe gastrointestinal or genitourinary toxicity, two side effects common with camptothecin therapy. In 2004, we plan to initiate another trialphase II clinical trials of CT-2106 in combination with standard chemotherapy in colorectal cancer and as a single agent in ovarian cancer and small cell lung cancer.

Camptothecins are an important and one of the fastest growing classes of anti-cancer drugs. However, like taxanes, their full clinical benefit is limited by poor solubility and significant toxicity. Orally delivered analogs, such as topotecan and irinotecan, are soluble, but are less effective in combating tumors. Camptothecins are important drugs in the second halftreatment of advanced colon, lung and ovarian cancers. Worldwide sales for camptothecins were approximately $1.3 billion in 2003.

Preclinical Programs

We are working on a number of drug targets in Discovery Research. Among these programs are bisplatinates, HIF-1 alpha:p300, proteasome inhibition, a novel enzymatic drug target called LPAAT-ß and other cytotoxic and/or antiangiogenic targets. We are in the process of continued target validation and lead optimization and may elect to move one of these programs into early development in 2005. In addition to Discovery Research, preclinical activities are focused on product lifecycle management, including the development of alternative dosage forms and routes of administration for TRISENOX and existing products in the development pipeline.

Research and development is essential to our business. We spent $89.5 million in 2003, $58.8 million in 2002 and $44.7 million in 2001 on Company sponsored research and development activities. Under a licensing agreement with Chugai Pharmaceutical Co., Ltd., we received development expenditure reimbursements of approximately $1.1 million and $1.9 million in 2003 and 2002, respectively, which we recorded as revenue.

Collaboration and Licensing Arrangements

PG-TXL Company, L.P.    On June 30, 1998, we entered into an agreement with PG-TXL Company, L.P. granting us an exclusive worldwide license for the rights to PG-TXL and to all potential uses of PG-TXL Company’s polymer technology. Under the terms of the agreement, we acquired the rights to fund the research, development, manufacture, marketing and sale of anti-cancer drugs developed using this polymer technology. We are obligated to make payments upon the attainment of significant development milestones, as defined in the agreement. We also granted warrants to purchase 350,000 shares of our common stock to PG-TXL Company, L.P., which became exercisable upon our entering a licensing agreement for CT-2103 with Chugai Pharmaceutical Co., Ltd. WeAs of December 31, 2003, we have made paymentsan aggregate of $3.0$5.0 million during 2002 and a $2.0 million payment in 2000. milestone payments.The aggregate amount of future milestone payments we may be required to pay pursuant to the PG-TXL agreement is $15.5 million. These additional payments may become due upon future milestones, such as trial commencements and completions, filings and regulatory approvals.

Chugai Pharmaceutical Co., Ltd.Ltd.    In October 2001, we entered into a licensing agreement with Chugai Pharmaceutical Co., Ltd., or Chugai, for the development and commercialization of XYOTAX. This agreement grants an exclusive license to Chugai to develop and commercialize XYOTAX in several Asian markets. Upon execution of the agreement, Chugai paid us a $3.0 million initial payment. Under the agreement, we may also receive future milestone payments totaling up to $13.0 million upon Chugai’s achievement of certain product development milestones, and we are entitled to receive royalties on product sales in the territories covered under the agreement. We received a $3.0approximately $1.1 million milestone payment and approximately $1.9 million in development expenditure reimbursements from Chugai duringin 2003 and 2002, respectively, as well as a $3.0 million milestone payment in 2002. Chugai has also committed up to $54 million in development expenditures over the course of the licensing agreement. The agreement will terminate on a country-by-country basis upon the earlier to occur of the expiration of the applicable patent rights, if any, in a given country or fifteen years from the date of the first commercial sale of XYOTAX in such country.

Nippon Shinyaku Co. Ltd.Ltd.    In December 2002, we entered into a distribution agreement with Nippon Shinyaku Co. Ltd., or Nippon. This agreement grants certain rights to Nippon to exclusively market and distribute TRISENOX (arsenic trioxide) injection in Japan, South Korea, and Taiwan. Upon execution of the agreement, Nippon paid us a $750,000 initial payment. Under the agreement, we mayreceived and recognized as revenue a $500,000 milestone payment in June 2003 for Nippon’s submission of an NDA in Japan. We are also eligible to receive future milestone payments totaling up to $4.0$3.5 million upon attainment of certain regulatory achievements.

Patents and Proprietary Rights

We dedicate significant resources to protecting our intellectual property.property, which we believe is important to our business. Through our acquisition of PolaRx, we obtained rights to four pending patent applicationsfamilies that, in the aggregate, cover dosage formulations, methods of administration and methods of use for various forms of arsenic trioxide and related compounds. This portfolio includes one allowed U.S. patent application and 35 U.S. and foreign pending patent applications directed to TRISENOX.

We have exclusive rights to threefive issued U.S. patents and 21101 U.S. and foreign pending patent applications relating to our polymer drug delivery technology. There are five issued U.S. patents and 28 pending U.S. and foreign patent applications directed to XYOTAX. Of the five issued U.S. patents, two of them and another 18 pending U.S. and foreign patent applications are directed to CT-2106. Additionally, we have four issued U.S. patents and 57 foreign pending and issued patents directed to pixantrone.

We intend to file additional patent applications when appropriate, with respect to improvements in our core technology and to specific products and processes that we develop. Patents may not issue from any present or future applications or, if patents do issue, such patents may not be issued on a timely basis or claims allowed on issued patents may not be sufficient to protect our technology. In addition, the patents issued to us may be challenged, invalidated or circumvented or the rights granted thereunder may not provide proprietary protection or commercial advantage to us. With respect to such issued U.S. patents or any patents that may issue in the future, they may not effectively protect the technology involved, foreclose the development of competitive products by others or otherwise be commercially valuable.

We have sought and intend to aggressively seek patent protection in the United States, Canada, Mexico, Europe and Japan to protect any products that we may develop. We also intend to seek patent protection or rely upon trade secrets to protect certain of our enabling technologies that will be used in discovering and evaluating new drugs that could become marketable products. However, such steps may not effectively protect the technology involved. To protect any such trade secrets and other proprietary information, we rely on confidentiality and material transfer agreements with our corporate partners, employees, consultants, outside

scientific collaborators and sponsored researchers and other advisors. These agreements may be breached, we may not have adequate remedies for breach or our trade secrets may otherwise become known or independently discovered by competitors. We also have members of our Scientific Advisory Board, our clinical advisors, our consultants and, in most cases, our employees enter into agreements requiring disclosure to us of ideas, developments, discoveries or inventions conceived during employment or consulting and assignment to us of proprietary rights to such matters related to our business and technology.

Manufacturing

We currently use, and expect to continue to be dependent upon, contract manufacturers to manufacture each of our product candidates. We have established a quality control and quality assurance program, including a set of standard operating procedures and specifications, designed to ensure that our products are manufactured in accordance with current Good Manufacturing Procedures, or cGMPs, and other applicable domestic and foreign regulations. These manufacturers may not meet our requirements for quality, quantity or timeliness.

We will need to developinvest in additional manufacturing resources, and may seek to enter into additional collaborative arrangements with other parties that have established manufacturing capabilities or may electcapabilities. It is likely that we

will continue to have arely on third party manufacture of our development and commercial products on a contract basis. WeCurrently, we have agreements with third party vendors to furnish TRISENOX, XYOTAX,pixantrone and PG-CPTCT-2106 drug supply for clinical studies and in the case of TRISENOX, for commercial market demand.

In September 2001, we entered into a supply agreement with Natural Pharmaceuticals, Inc. for paclitaxel, a key starting material for XYOTAX. Under the supply agreement, we purchased paclitaxel at a pre-determined price and willexpect to receive supply through early 2004. We will be dependent upon these third parties to supply us in a timely manner with products manufactured in compliance with cGMPs or similar standards imposed by foreign regulatory authorities where our products are tested and/or marketed. Contract manufacturers may violate cGMPs, and the FDA has recently intensified its oversight of drug manufacturers. The FDA may take action against a contract manufacturer who violates cGMPs. Such actions may include requiring the contract manufacturer to cease its manufacturing activities.

Sales and Marketing

We have developed an experienced sales and marketing infrastructure in the United States to commercialize our portfolio of oncology products. The oncology market is highly concentrated. It is comprised primarily of the approximately 8,500 physicians who order the vast majority of cancer therapeutics, but we sell TRISENOX primarily to pharmaceutical wholesalers and oncology distributors, who in turn sell TRISENOX primarily to hospitals and clinics. We currently are marketing TRISENOX with our direct sales force in the U.S. consisting of one national account manager, five regional business directors, and 3940 field based oncology account managers. We are usingAn additional seven medical science liaisons provide scientific support in the field.

In February 2004, we announced a combinationsignificant expansion of our own regionalEuropean commercial operations by hiring additional sales personnelpersonnel. We have 16 fully trained field-based country managers and contract sales personnel to supportrepresentatives in eight major market countries selling TRISENOX. We believe this experienced sales force will increase European sales of TRISENOX as well as aid in the commercialization outsidepromotion of the U.S.any additional commercial products that we may acquire or develop internally.

Competition

Competition in the pharmaceutical and biotechnology industries is intense. We face competition from a variety of companies focused on developing oncology drugs. We compete with large pharmaceutical companies and with other specialized biotechnology companies.companies, including but not limited to Bristol-Myers Squibb Co., Aventis, American Pharmaceutical Partners, Neopharm Inc. and Sonus Pharmacuticals for XYOTAX, Celgene Corporation, Millennium Pharmaceuticals, Inc., Pharmion Corporation and SuperGen Corporation for TRISENOX and Inex/Enzon Pharmaceuticals Corporation for pixantrone. Many of our existing or potential competitors have substantially greater financial, technical and human resources than us and may be better equipped to develop, manufacture and market products. Smaller companies may also prove to be significant competitors, particularly through collaborative arrangements with large pharmaceutical and established biotechnology companies. Many of these competitors have significant products that have been approved or are in development and operate large, well-funded research and development programs.

We expect to encounter significant competition for the principal pharmaceutical products we plan to develop. Companies that complete clinical trials, obtain required regulatory approvals and commence

commercial sales of their products before their competitorsus may achieve a significant competitive advantage if their products work through a similar mechanism as our products.products and if the approved indications are similar. Accordingly, we do not believe competition is as intense among products that treat cancer through novel delivery or therapeutic mechanisms.mechanisms where these mechanisms translate into a clinical advantage in safety and/or efficacy. A number of biotechnology and pharmaceutical companies are developing new products for the treatment of the same diseases being targeted by us. In some instances, such products have already entered late-stage clinical trials or received FDA approval. However, cancer drugs with distinctly different mechanisms of action are often used together in combination for treating cancer, allowing several different products to target the same cancer indication or disease type. Such combination therapy is typically supported by clinical trials that demonstrate the advantage of combination therapy over that of a single agent treatment.

We believe that our ability to compete successfully will be based on our ability to create and maintain scientifically advanced technology, develop proprietary products, attract and retain scientific personnel, obtain patent or other protection for our products, obtain required regulatory approvals and manufacture and successfully market our products either alone or through outside parties. We will continue to seek licenses with respect to technology related to our field of interest and may face competition with respect to such efforts.

Government Regulation

The research, development, testing, manufacture, labeling, promotion, advertising, distribution, and marketing, among other things, of our products are extensively regulated by governmental authorities in the United States and other countries. In the United States, the FDA regulates drugs under the Federal Food, Drug, and Cosmetic Act, or the FDCA, and its implementing regulations. Failure to comply with the applicable U.S. requirements may subject us to administrative or judicial sanctions, such as FDA refusal to approve pending new drug applications,NDAs, warning letters, product recalls, product seizures, total or partial suspension of production or distribution, injunctions, and/or criminal prosecution.

Drug Approval Process.    None of our drugs may be marketed in the U.S. until the drug has received FDA approval. The steps required before a drug may be marketed in the U.S. include:

Preclinical tests include laboratory evaluation of product chemistry, toxicity, and formulation, as well as animal studies. The conduct of the preclinical tests and formulation of the compounds for testing must comply with federal regulations and requirements. The results of the preclinical tests, together with manufacturing information and analytical data, are submitted to the FDA as part of an IND, which must become effective before human clinical trials may begin. An IND will automatically become effective 30 days after receipt by the FDA, unless before that time the FDA raises concerns or questions about issues such as the conduct of the trials as outlined in the IND. In such a case, the IND sponsor and the FDA must resolve any outstanding FDA concerns or questions before clinical trials can proceed. We cannot be sure that submission of an IND will result in the FDA allowing clinical trials to begin.

Clinical trials involve the administration of the investigational drug to human subjects under the supervision of qualified investigators. Clinical trials are conducted under protocols detailing the objectives of the study, the parameters to be used in monitoring safety, and the effectiveness criteria to be evaluated. Each protocol must be submitted to the FDA as part of the IND.

Clinical trials typically are conducted in three sequential phases, but the phases may overlap or be combined. The study protocol and informed consent information for study subjects in clinical trials must also be approved by an Institutional Review Board for each institution where the trials will be conducted. Study subjects must sign an informed consent form before participating in a clinical trial. Phase I usually involves the initial introduction of the investigational drug into people to evaluate its short-term safety, dosage tolerance, metabolism, pharmacokinetics and pharmacologic actions, and, if possible, to gain an early indication of its

effectiveness. Phase II usually involves trials in a limited patient population to (i) evaluate dosage tolerance and appropriate dosage; (ii) identify possible adverse effects and safety risks; and (iii) evaluate preliminarily the efficacy of the drug for specific indications. Phase III trials usually further evaluate clinical efficacy and test further for safety by using the drug in its final form in an expanded patient population. There can be no assurance that phase I, phase II, or phase III testing will be completed successfully within any specified period of time, if at all. Furthermore, the Company or the FDA may suspend clinical trials at any time on various grounds, including a finding that the subjects or patients are being exposed to an unacceptable health risk.

The FDCA permits FDA and the IND sponsor to agree in writing on the design and size of clinical studies intended to form the primary basis of an effectiveness claim in an NDA application. This process is known as Special Protocol Assessment, or SPA. These agreements may not be changed after the clinical studies begin, except in limited circumstances.

Assuming successful completion of the required clinical testing, the results of the preclinical studies and of the clinical studies, together with other detailed information, including information on the manufacture and composition of the drug, are submitted to the FDA in the form of an NDA requesting approval to market the product for one or more indications. The testing and approval process requires substantial time, effort, and financial resources. The agencies review the application and may deem it to be inadequate to support the registration and we cannot be sure that any approval will be granted on a timely basis, if at all. The FDA may also refer the application to the appropriate advisory committee, typically a panel of clinicians, for review, evaluation and a recommendation as to whether the application should be approved. The FDA is not bound by the recommendations of the advisory committee.

The FDA has various programs, including fast track, priority review, and accelerated approval, that are intended to expedite or simplify the process for reviewing drugs, and/or provide for approval on the basis surrogate endpoints. Generally, drugs that may be eligible for one or more of these programs are those for serious or life-threatening conditions, those with the potential to address unmet medical needs, and those that provide meaningful benefit over existing treatments. We cannot be sure that any of our drugs will qualify for any of these programs, or that, if a drug does qualify, that the review time will be reduced.

Before approving an NDA, the FDA usually will inspect the facility or the facilities at which the drug is manufactured, and will not approve the product unless cGMP compliance is satisfactory. If the FDA evaluates the NDA and the manufacturing facilities as acceptable, the FDA may issue an approval letter, or in some cases, an approvable letter followed by an approval letter. Both letters usually contain a number of conditions that must be met in order to secure final approval of the NDA. When and if those conditions have been met to the FDA’s satisfaction, the FDA will issue an approval letter. The approval letter authorizes commercial marketing of the drug for specific indications. As a condition of NDA approval, the FDA may require postmarketing testing and surveillance to monitor the drug’s safety or efficacy, or impose other conditions.

After approval, certain changes to the approved product, such as adding new indications, making certain manufacturing changes, or making certain additional labeling claims, are subject to further FDA review and approval. Before we can market TRISENOX for additional indications, we must obtain additional approvals from FDA. Obtaining approval for a new indication generally requires that additional clinical studies be conducted. We cannot be sure that any additional approval for new indications for TRISENOX will be approved on a timely basis, or at all.

Post-Approval Requirements.    TRISENOX was approved by the FDA under its accelerated approval process in September 2000. In order to secure this approval, CTIwe committed to completing several post-approval requirements, including the conduct of additional clinical studies. Should CTIwe fail to fulfill these obligations, the FDA may withdraw approval of TRISENOX. In addition, holders of an approved NDA are required to: (i) report certain adverse reactions to the FDA, (ii) comply with certain requirements concerning advertising and promotional labeling for their products, and (iii) continue to have quality control and manufacturing procedures

conform to cGMP after approval. The FDA periodically inspects the sponsor’s records related to safety reporting and/or manufacturing facilities; this latter effort includes assessment of compliance with cGMP. Accordingly, manufacturers must continue to expend time, money, and effort in the area of production and quality control to maintain cGMP compliance. We use and will continue to use third party manufacturers to produce our products in clinical and commercial quantities, and future FDA inspections may identify compliance issues at our facilities or at the facilities of our contract manufacturers that may disrupt production or distribution, or require substantial resources to correct. In addition, discovery of problems with a product after approval may result in restrictions on a product, manufacturer, or holder of an approved NDA, including withdrawal of the product from the market.

TRISENOX was also approved in Europe by way of the centralized process and marketing authorization was granted by the EMEA “under exceptional circumstances”. CTI hascircumstances.” We have agreed to fulfill several post-approval

commitments regarding TRISENOX. In addition, reporting of adverse reactions, compliance with certain requirements concerning advertising and promotional labeling and adherence to cGMP in the area of production and quality control is also required. Not completing these commitments or maintaining adherence to cGMP may result in similar actions as those described above for FDA, including withdrawal of TRISENOX.

Orphan Drug.    The FDA may grant orphan drug designation to drugs intended to treat a “rare disease or condition,” which generally is a disease or condition that affects fewer than 200,000 individuals in the United States. Orphan drug designation must be requested before submitting an NDA. AfterIf the FDA grants orphan drug designation, which it may not, the identity of the therapeutic agent and its potential orphan use are publicly disclosed by the FDA. Orphan drug designation does not convey an advantage in, or shorten the duration of, the review and approval process. If a product which has an orphan drug designation subsequently receives the first FDA approval for the indication for which it has such designation, the product is entitled to orphan exclusivity, meaning that the FDA may not approve any other applications to market the same drug for the same indication, except in certain very limited circumstances, for a period of seven years. Orphan drug designation does not prevent competitors from developing or marketing different drugs for anthat indication.

We have obtained orphan drug market exclusivity from the FDA for TRISENOX to treat patients with drug resistant or relapsed APL. We have also received orphan drug designation for TRISENOX for the treatment of patients with refractory multiple myeloma and MDS, CML, AML, CLL and AML.HCC. However, TRISENOX may not receive an orphan drug marketing exclusivity for any of these indications, or any of our other drug products may not receive orphan drug exclusivity for any indication. Also, it is possible that our competitors could obtain approval, and attendant orphan drug exclusivity, for products that would preclude us from marketing our products for specified indications for some time.

Non-United States Regulation.    Before our products can be marketed outside of the United States, they are subject to regulatory approval similar to that required in the United States, although the requirements governing the conduct of clinical trials, including additional clinical trials that may be required, product licensing, pricing and reimbursement vary widely from country to country. No action can be taken to market any product in a country until an appropriate application has been approved by the regulatory authorities in that country. The current approval process varies from country to country, and the time spent in gaining approval varies from that required for FDA approval. In certain countries, the sales price of a product must also be approved. The pricing review period often begins after market approval is granted. Even if a product is approved by a regulatory authority, satisfactory prices may not be approved for such product.

In Europe, marketing authorizations may be submitted at a centralized, a decentralized or national level. The centralized procedure is mandatory for the approval of biotechnology products and provides for the grant of a single marketing authorization that is valid in all European UnionEU members states. As of January 1995, a mutual recognition procedure is available at the request of the applicant for all medicinal products that are not subject to the centralized procedure. There can be no assurance that the chosen regulatory strategy will secure regulatory approvals on a timely basis or at all.

Environmental Regulation

In connection with our research and development activities, we are subject to federal, state and local laws, rules, regulations and policies governing the use, generation, manufacture, storage, air emission, effluent discharge, handling and disposal of certain materials, biological specimens and wastes. Although we believe that we have complied with these laws, regulations and policies in all material respects and have not been required to take any significant action to correct any noncompliance, we may be required to incur significant costs to comply with environmental and health and safety regulations in the future. Our research and development involves the controlled use of hazardous materials, including, but not limited to, certain hazardous chemicals and radioactive materials. Although we believe that our safety procedures for handling and disposing of such materials comply with the standards prescribed by state and federal regulations, the risk of accidental contamination or injury from these materials cannot be eliminated. In the event of such an accident, we could be held liable for any damages that result and any such liability could exceed our resources.

Employees

As of February 28, 2003,29, 2004, we employed 286372 individuals, including 83 holding261 in the U.S. and 111 in Europe. In the U.S., 76 employees hold doctoral or other advanced degrees.degrees while 55 hold doctoral or other advanced degrees in Europe. Our U.S. employees do not have a collective bargaining agreement. Our European employees are subject to a collective bargaining agreement. We consider our relations with our employees to be good.

Scientific Advisory Board and Clinical Advisors

We have a Scientific Advisory Board that consistsInformation regarding our executive officers is set forth in Item 10 of recognized scientists with expertise in the fields of immunology, cell and molecular biology, and synthetic and medical chemistry. Our Scientific Advisory Board meets with our management and key scientific employees on a semi-annual basis and in smaller groups or individually from time to time on an informal basis. The members assist us in identifying scientific and product development opportunities, reviewing with management the progress of our specific projects and recruiting and evaluating our scientific staff. We also have clinical advisors that assist us from time to time on clinical matters.

The following are members of our Scientific Advisory Board:

Lewis Cantley, Ph.D.,this Report, which information is a noted authority in cellular biochemical signaling pathways that employ phosphatidyl inositol and its metabolites and is the discoverer of one of the most critical enzymes in those pathways, the PI3 Kinase. He is currently Professor of Cell Biology at Harvard Medical School and Chief of the Division of Signal Transduction in the Department of Medicine, Beth Israel Hospital, Boston and the author of over 180 publications.

Edward A. Dennis, Ph.D., is Professor and Vice Chair for Medical Biochemistry and former Chair of the Department of Chemistry and Biochemistry at the University of California, San Diego. He is a noted authority on phospholipases, cell signaling and phospholipid metabolism. Dr. Dennis serves as Chair of the Board of Directors and President of the Keystone Symposia on Cell and Molecular Biology. He has served as President of the Association of Medical and Graduate Departments of Biochemistry and as Chair of the Publications Committee and of the Meetings Committee of the American Society for Biochemistry and Molecular Biology. In 2000, he was the recipient of the prestigious Avanti Award for research in lipid enzymology. In July 2003, he will become editor of the Journal of Lipid Research. He has over 250 publications and numerous patents in the lipid area.

Mike Gelb, Ph.D., is Professor of Chemistry and Adjunct Professor of Biochemistry at the University of Washington. He has developed techniques for structure-based drug design and clinical assays of enzymes to identify genetic defects. He is a noted authority on lipid enzymes. Dr. Gelb is the recipient of the Pfizer Award in Enzyme Chemistry and the ICI Pharmaceuticals Award for Excellence in Chemistry and the author of more than 200 manuscripts.

Edwin Krebs, M.D., is a Professor Emeritus, Department of Pharmacology and Biochemistry, at the University of Washington in Seattle and a Senior Investigator Emeritus at the Howard Hughes Medical Institute. He is a recognized authority on mechanisms of action of second messengers, including protein kinases and phosphorylation reactions. He is the recipient of numerous awards and honors and has authored 297 manuscripts. In 1992, Dr. Krebs was awarded the Nobel Prize in Physiology of Medicine for his work on second messenger pathways.

L. Jackson Roberts, II, M.D., is an internationally recognized authority on the oxidative metabolism of polyunsaturated fatty acids. He is known for having identified PGD2 on the major mast cell lipid mediator and, more recently, for having originated the field of studying non enzymatically generated prostanoids, including the isprostanes and neuroprostanes, and newly discovered products of lipid peroxidation, isofurans. He is currently Professor of Pharmacology and Medicine at Vanderbilt University and is the author of over 283 publications.

Julian Simon, Ph.D., is an Associate Member, Clinical Research and Human Biology Divisions, at the Fred Hutchinson Cancer Research Center in Seattle and is Affiliate Associate Professor of Chemistry at the University of Washington. He is a member of the Experimental Therapeutics Study Section and Special Emphasis Panel of the National Cancer Institute. His research is focused on application of chemical genetics to the discovery of mechanistically novel anticancer agents.

The following are our retained Clinical Advisors:

E. Donnall Thomas, M.D., is the former Associate Director of Clinical Research and presently a Professor Emeritus at the Fred Hutchinson Cancer Research Center, of which he was a founding member. His research has spanned a wide array of fields from radiation biology to developmental immunology, and from cancer causing genes to gene transfer therapies. For his pioneering work in bone marrow transplant, Dr. Thomas was awarded the Nobel Prize for Medicine in 1990. Among the other honors awarded to Dr. Thomas in recognition of his medical research are the American Cancer Society Award for Distinguished Service in Basic Research and the Kettering Prize of the General Motors Cancer Research Foundation. He is a member of the U.S. Academy of Sciences.

Steven Soignet, M.D., is the Vice President and co-founder of the Arcus Group, a healthcare information consulting company. He held a faculty appointment in the Developmental Chemotherapy Service, Memorial Sloan-Kettering Cancer Center, and in the Department of Medicine, Cornell University Medical Center. Dr. Soignet’s research primarily has focused on early phase clinical drug development in both hematologic and solid tumors. He is a member of the American College of Physicians, the American Association of Cancer Research, the American Society of Hematology, and the American Society of Clinical Oncology.

In addition to selected retained experts, an Ad Hoc advisory board approach has been takenincorporated herein by us to avail ourselves to the broadest expertise in a given oncologic disease. We have convened disease specific advisory boards in the U.S. as well as in Europe to take advantage of the differences in clinical practice as well as regulatory requirements between these different territories. This allows us to plan for registration of our drugs in multiple markets.reference.

Factors Affecting Our Future Operating Results

This annual report on Form 10-K contains forward-looking statements that involve risks and uncertainties. Our actual results could differ materially from those anticipated in these forward-looking statements as a result of certain factors, including the risks faced by us described below and elsewhere in this annual report on Form 10-K.

We mayexpect to continue to incur net losses, and we maymight never achieve profitability.

We were incorporated in 1991 and have incurred a net operating loss every year. As of December 31, 2002,2003, we had an accumulated deficit of approximately $340.5 million.$470.5 million, not including losses of Novuspharma. We may never become profitable, even if we are able to commercialize additional products. We will need to conduct significant research, development, testing and regulatory compliance activities that, together with projected general and administrative expenses, we expect will result in substantial increasing operating losses for at least the next severalcouple of years. Even if we do achieve profitability, we may not be able to sustain or increase profitability on a quarterly or annual basis.

If we do not successfully develop additional products, we may be unable to generate additional revenue.significant revenue or become profitable.

We have only one product, TRISENOX, for relapsed or refractory APL that has received marketing approval to date. Our leading drug candidates, TRISENOX for other indications, XYOTAX, pixantrone and PG-CPT,CT-2106, are currently in clinical trials. These clinical trials of the drug candidates involve the testing of potential therapeutic agents, or effective treatments, in humans in three phases to determine the safety and efficacy of the drug

candidates necessary for an approved drug. Many drugs in human clinical trials fail to demonstrate the desired safety and efficacy characteristics.may not be successful. Even if our drugs progress successfully through initial human testing, they may fail in later stages of development. Many drugs in human clinical trials fail to demonstrate the desired safety and efficacy characteristics. A number of companies in the pharmaceutical industry, including us, have suffered significant setbacks in advanced clinical trials, even after reporting promising results in earlier trials. For example, in our first phase III human trial for lisofylline, completed in March 1998, we failed to meet our two primary endpoints, or goals, even though we met our endpoints in two earlier phase II trials for lisofylline. As a result, we are no longer developing lisofylline as a potential product. Many of our drug candidates are still in research and pre-clinical development, which means that they have not

yet been tested on humans. We will need to commit significant time and resources to develop these and additional product candidates. Our product candidates will be successful only if:

We are dependent on the successful completion of these goals in order to generate revenues. The failure to generate such revenues may preclude us from continuing our research and development of these and other product candidates.

We may need to raise additional funds in the future, and they may not be available on acceptable terms, or at all.

We expect that our existing and anticipated capital resources will enable us to maintain our planned operations through at least the first half of 2005. We expect to receive certain grants and subsidized loans from the Italian government and the EU through our Italian branch into which Novuspharma’s operating assets and liabilities will be contributed. However, we may not receive the relevant funding because the grants and subsidies are awarded at the discretion of the relevant authorities.

Beyond the first half of 2005, or if our plans or assumptions change or are inaccurate, we will have to raise additional funds to continue the development of our technologies and complete the commercialization of products, if any, resulting from our technologies. We may raise such capital through public or private equity financings, partnerships, debt financings, bank borrowings or other sources.

Additional funding may not be available on favorable terms or at all. If adequate funds are not otherwise available, we may curtail operations significantly, including the delay, modification or cancellation of research and development programs aimed at bringing new products to market. To obtain additional funding, we may need to enter into arrangements that require us to relinquish rights to certain technologies, drug candidates, products and/or potential markets. To the extent that additional capital is raised through the sale of equity, or securities convertible into equity, you may experience dilution of your proportionate ownership of us.

Our operations in Italy make us subject to increased risk regarding currency exchange rate fluctuations.

As a result of our merger with Novuspharma and our consequent operations in Italy, we are exposed to risks associated with foreign currency transactions insofar as we might desire to use U.S. dollars to make contract payments denominated in Euros or vice versa. As the net positions of our foreign currency transactions might fluctuate, our earnings might be negatively affected. In addition, dataas a result of our merger with Novuspharma, we are exposed to risks associated with the translation of Novuspharma’s Euro-denominated financial results and balance sheet into United States dollars. Our reporting currency will remain as the United States dollar, however, a portion of our consolidated financial obligations will arise in Euros. In addition, the carrying value of some of our assets and liabilities will be affected by fluctuations in the value of the United States dollar as compared to the Euro. Changes in the value of the United States dollar as compared to the Euro might have an adverse effect on our reported results of operations and financial condition.

We may take longer to complete our clinical trials than we expect, or we may not be able to complete them at all.

Before regulatory approval for any potential product can be obtained, we must undertake extensive clinical testing on humans to demonstrate the safety and efficacy of the product, both on its own terms, and as compared to the other principal drugs on the market that have the same therapeutic indication. Although for planning purposes we forecast the commencement and completion of clinical trials, the actual timing of these events can vary dramatically due to a number of factors.

We may not obtain authorization to permit product candidates that are already in the pre-clinical development phase to enter the human clinical testing phase. Authorized pre-clinical or clinical testing may not be completed successfully within any specified time period by us, or without significant additional resources or expertise to those originally expected to be necessary. Many drugs in human clinical trials fail to demonstrate the desired safety and efficacy characteristics. Clinical testing may not show potential products to be safe and efficacious and potential products may not be approved for a specific indication. Further, the results from pre-clinical studies and early clinical trials may not be indicative of the results that will be obtained in later-stage clinical trials. Data obtained from clinical trials are susceptible to varying interpretations. Government regulators and our collaborators may not agree with our interpretation of our future clinical trial results. TheIn addition, we or regulatory authorities may suspend clinical trials at any time on the basis that the participants are being exposed to unacceptable health risks. Completion of clinical trials depends on, among other things, the number of patients available for enrollment in a particular trial, which is a function of many factors, including the number of patients with the relevant conditions, the nature of the clinical testing, the proximity of patients to clinical testing centers, the eligibility criteria for tests as well as competition with other clinical testing programs involving the same patient profile but different treatments.

We have limited experience in conducting clinical trials. We expect to continue to rely on third parties, such as contract research organizations, academic institutions and/or co-operative groups, to conduct, oversee and monitor clinical trials as well as to process the clinical results and manage test requests, which may result in delays or failure to complete trials, if the third parties fail to perform or to meet the applicable standards.

If we fail to commence or complete, or experience delays in any of our present or planned clinical trials, including the Phase III clinical trials of XYOTAX, the Phase II clinical trials of TRISENOX and the Phase II and Phase III clinical trials of pixantrone, our ability to conduct our business as planned could be harmed. Our development costs may increase if we experience any future delays in our clinical trials for XYOTAX, TRISENOX, pixantrone or our other product candidates or if we need to perform more or larger clinical trials than planned. If delays or costs are significant, our financial results and PG-CPT or any of our future drugability to commercialize our product candidates may not be successful.

Many of our drug candidates are still in research and preclinical development, which means that they have not yet been tested on humans. We will need to commit significant time and resources to develop these and additional product candidates. We are dependent on the successful completion of clinical trials and obtaining regulatory approval in order to generate revenues. The failure to generate such revenues may preclude us from continuing our research and development of these and other product candidates.adversely affected.

Even if our drug candidates are successful in clinical trials, we may not be able to successfully commercialize them.

Since our inception in 1991, we have dedicated substantially all of our resources to the research and development of our technologies and related compounds. With the exception of TRISENOX for patients with APL who have relapsed or refractory APL,failed standard therapies, all of our compounds currently are in research or development, and none has been submitted for marketing approval. Our other compounds may not enter human clinical trials on a timely basis, if at all, and we may not develop any product candidates suitable for commercialization.

Prior to commercialization, each product candidate will require significant additional research, development and preclinicalpre-clinical testing and extensive clinical investigation before submission of any regulatory application for marketing approval. The development of anti-cancer drugs, including those we are currently developing, is unpredictable and subject to numerous risks. Potential products that appear to be promising at early stages of development may not reach the market for a number of reasons. Potential productsreasons including that they may:

Our product development efforts or

The occurrence of any of these events could adversely affect the commercialization of our collaborative partners’ efforts may not be successfully completed and we may not obtain required regulatory approvals.products. Any products, if introduced, may not be successfully marketed norand/or may not achieve customer acceptance. If we fail to commercialize products or if our future products do not achieve significant market acceptance, we will not likely generate significant revenues or become profitable.

If we fail to establish and maintain collaborations or if our partners do not perform, we may unable to develop and commercialize our product candidates.

We have entered into collaborative arrangements with third parties to develop and/or commercialize product candidates and are currently seeking additional collaborations. For example, we have entered into an agreement with Chugai Pharmaceutical Co., Ltd. to develop and commercialize XYOTAX in several Asian markets. Additional collaborations might be necessary in order for us to fund our research and development activities and third-party manufacturing arrangements, seek and obtain regulatory approvals and successfully commercialize our existing and future product candidates. If we fail to enter into additional collaborative arrangements or fail to maintain our existing collaborative arrangements, the number of product candidates from which we could receive future revenues would decline.

Our dependence on collaborative arrangements with third parties will subject us to a number of risks that could harm our ability to develop and commercialize products, including that:

The occurrence of any of these events could adversely affect the development or commercialization of our products.

Because we basedbase several of our drug candidates on unproven novel technologies, we may never develop them into commercial products.

We base many of our product candidates upon novel delivery technologies that we are using to discover and develop drugs for the treatment of cancer. This technology hasThese technologies have not been proven. Furthermore, preclinicalpre-clinical results in animal studies may not predict outcomeoutcomes in human clinical trials. Our product candidates may not be proven safe or effective. If this technology doesthese technologies do not work, our drug candidates may not develop into commercial products.

We may not completeface difficulties in achieving acceptance of our clinical trialsproducts in the time expected, which could delay or prevent the commercialization ofmarket if we do not continue to expand our products.sales and marketing infrastructure.

AlthoughWe currently are marketing TRISENOX with our direct sales force. Competition for planning purposesthese individuals is intense, and in the event we forecast the commencement and completion of clinical trials, the actual timing of these events can vary dramatically due to factors such as delays, scheduling conflicts with participating clinicians and clinical institutions and the rate of patient enrollment. Clinical trials involving our product candidates may not commence nor be completed as forecasted. We have limited experience in conducting clinical trials. In certain circumstancesneed additional sales personnel, we rely on academic institutions or clinical research organizations to conduct, supervise or monitor some or all aspects of clinical trials involving our products. In addition, certain clinical trials for our products will be conducted by government-sponsored agencies and consequently will be dependent on governmental participation and funding. We will have less control over the timing and other aspects of these clinical trials than if we conducted them entirely on our own. These trials may not commence or be completed as we expect. They may not be conducted successfully. Failureable to commencehire individuals with the

experience required or complete, or delays in,number of sales personnel we need. In addition, if we market and sell products other than TRISENOX, we may need to further expand our marketing and sales force with sufficient technical expertise and distribution capacity. If we are unable to expand our direct sales operations and train new sales personnel as rapidly as necessary, we may not be able to increase market awareness and sales of our products, which may prevent us from growing our revenues and achieving and maintaining profitability.

If any of our planned clinical trials could delaylicense agreements for intellectual property underlying TRISENOX, XYOTAX, pixantrone or preventany other products are terminated, we may lose our rights to develop or market that product.

We have licensed intellectual property, including patent applications from The Memorial Sloan-Kettering Cancer Center, Samuel Waxman Cancer Research Foundation, Beijing Medical University, The University of Vermont, Hoffman La Roche and others, including the commercializationintellectual property relating to TRISENOX and pixantrone. We have also in-licensed the intellectual property relating to our drug delivery technology that uses polymers that are linked to drugs, known as polymer-drug conjugates, including XYOTAX and CT-2106. Some of our product development programs depend on our ability to maintain rights under these licenses. Each licensor has the power to terminate its agreement with us if we fail to meet our obligations under these licenses. We may not be able to meet our obligations under these licenses. If we default under any license agreements, we may lose our right to market and sell any products and harm our business.based on the licensed technology.

If we fail to protect adequately protect our intellectual property, our competitive position could be harmed.

Development and protection of our intellectual property are critical to our business. If we do not adequately protect our intellectual property, competitors may be able to practice our technologies. Our success depends in part on our ability to:

In particular we believeWhen polymers are linked, or conjugated, to drugs, the results are referred to as polymer-drug conjugates. We are developing drug delivery technology that linking our polymerslinks chemotherapy drugs to existingbiodegradable polymers. For example, XYOTAX is paclitaxel, the active ingredient in Taxol^®, one of the world’s best selling cancer drugs, may yield patentable subject matter.linked to polyglutamate. We do not believe that our polymer-drug conjugates will infringe any valid third-party patents covering the underlying drug. However, we may not receive a patent for our polymerpolymer-drug conjugates and we may be challenged by the holder of a patent covering the underlying drug.

The patent position of biopharmaceutical firms generally is highly uncertain and involves complex legal and factual questions. The U.S.United States Patent and Trademark Office has not established a consistent policy regarding the breadth of claims that it will allow in biotechnology patents. If it allows broad claims, the number and cost of patent interference proceedings in the U.S.United States, and the risk of infringement litigation may increase. If it allows narrow claims, the risk of infringement may decrease, but the value of our rights under our patents, licenses and patent applications may also decrease.

Patent applications in which we have rights may never issue as patents and the claims of any issued patents may not afford meaningful protection for our technologies or products. In addition, patents issued to us or our licensors may be challenged and subsequently narrowed, invalidated or circumvented. Litigation, interference proceedings or other governmental proceedings that we may become involved in with respect to our proprietary technologies or the proprietary technology of others could result in substantial cost to us. Patent litigation is widespread in the biotechnology industry, and any patent litigation could harm our business. Costly litigation might be necessary to protect our orphan drug designations in the United States or EU, which are designations for products meeting criteria based on the size of the potential United States or EU patient population for a drug, respectively, and which entitle that drug to seven years of exclusive rights in the United States market or ten years in the EU market, as applicable, or to protect a patent position or to determine the scope and validity of third party proprietary rights, and we may not have the required resources to pursue any such litigation or to protect our patent rights. AnAny adverse outcome in litigation

with respect to the infringement or validity of any of our patents owned by third parties could subject us to significant liabilities to third parties, require disputed rights to be licensed from third parties or require us to cease using a product or technology.

We also rely upon trade secrets, proprietary know-how and continuing technological innovation to remain competitive. Third parties may independently develop such know-how or otherwise obtain access to our

technology. While we require our employees, consultants and corporate partners with access to proprietary information to enter into confidentiality agreements, these agreements may not be honored.

If any of our license agreements for intellectual property underlying TRISENOX, XYOTAX or any other product are terminated, we may lose our rights to develop or market that product.

Patents issued to third parties may cover our products as ultimately developed. We may need to acquire licenses to these patents or challenge the validity of these patents. We may not be able to license any patent rights on acceptable terms or successfully challenge such patents. The need to do so will depend on the scope and validity of these patents and ultimately on the final design or formulation of the products and services that we develop.

We have licensed intellectual property, including patent applications from The Memorial Sloan Kettering Cancer Center, Samuel Waxman Cancer Research Foundation, Beijing Medical University and others, including the intellectual property directed to arsenic drugs and TRISENOX. We have also in-licensed the intellectual property relating to our polymer drug delivery technology, including XYOTAX. Some of our product development programs depend on our ability to maintain rights under these licenses. Each licensor has the power to terminate its agreement with us if we fail to meet our obligations under those licenses. We may not be able to meet our obligations under these licenses. If we default under any of these license agreements, we may lose our right to market and sell any products based on the licensed technology.

Our products could infringe on the intellectual property rights of others, which may cause us to engage in costly litigation and, if we are not successful, could cause us to pay substantial damages and prohibit us from selling our products.

Although weWe attempt to monitor the patent filings ofthat may be relevant to our competitorsproducts and product candidates in an effort to guide the design and development of our products to avoid infringement, third partiesinfringement. We may not be able to successfully challenge the validity of these patents that have been issued or licensed to us. We mayand could have to pay substantial damages, possibly including treble damages, for past infringement if it is ultimately determined that our products infringe a third party’s patents. Further, we may be prohibited from selling our products before we obtain a license, which, if available at all, may require us to pay substantial royalties. Moreover, third parties may challenge the patents that have been issued or licensed to us. Even if infringement claims against us are without merit, defending a lawsuit takesor if we challenge the validity of issued patents, lawsuits take significant time, may be expensive and may divert management attention from other business concerns.

If we are unable to enter into new licensing arrangements, our future product portfolio and potential profitability could be harmed.

One component of our business strategy is in-licensing drug compounds developed by other pharmaceutical and biotechnology companies or academic research laboratories. Substantially all of our product candidates in clinical development are in-licensed from a third party, including TRISENOX, XYOTAX and pixantrone. Competition for new promising compounds and commercial products can be intense. If we are not able to identify future in-licensing opportunities and enter into future licensing arrangements on acceptable terms, our future product portfolio and potential profitability could be harmed.

We may be unable to obtain the raw materials necessary to produce our XYOTAX product candidate in sufficient quantity to meet demand when and if such product is approved.

We may not be able to continue to purchase the materials necessary to produce XYOTAX, including paclitaxel, in adequate volume and quality. Paclitaxel is derived from certain varieties of yew trees. Supply of paclitaxel is controlled by a limited number of companies. We purchase the majority of the paclitaxel we need from a single vendor. We also purchase the raw material polyglutamic acid from a single source on a purchase order basis. Should the paclitaxel or polyglutamic acid purchased from our sources prove to be insufficient in quantity or quality, or should these relationships terminate, we may not be able to obtain a sufficient supply from alternate sources on acceptable terms, or at all.

Our dependence on third party manufacturers means that we may not have sufficient control over the manufacture of our products.

We do not currently have internal facilities for the cGMP manufacture of any of our development or commercial products. In addition, TRISENOX, our first commercial product, is currently manufactured by a single vendor. In 2002, we began the process of qualifying an additional supplier for our finished product manufacturing for TRISENOX. This additional supplier received FDA approval to manufacture TRISENOX in June 2003. Because we do not directly control our suppliers, these vendors may not be able to provide us with

finished product when we need it. Plans are in place to develop additional manufacturing resources, such as entering into collaborative arrangements with other parties that have established manufacturing capabilities or elect to have other additional third parties manufacture our products on a contract basis.

We will be dependent upon these third parties to supply us in a timely manner with products manufactured in compliance with cGMPs or similar manufacturing standards imposed by foreign regulatory authorities where our products will be tested and/or marketed. While the FDA and other regulatory authorities maintain oversight for cGMP compliance of drug manufacturers, contract manufacturers may at times violate cGMPs. The FDA and other regulatory authorities may take action against a contract manufacturer who violates cGMPs. Such actions may include requiring the contract manufacturer to cease its manufacturing activities.

Another one of our products under development, XYOTAX, has a complex manufacturing process, which may prevent us from obtaining a sufficient supply of drug product for the clinical trials and commercial activities currently planned or underway on a timely basis, if at all.

We are subject to extensive government regulation, including the requirement of approval before our products may be marketed.

Regulatory agencies have approved only one of our products, TRISENOX, for sale in the United States and the EU, to treat patients with a type of blood cancer called APL who have relapsed or failed standard therapies. Before we can market TRISENOX for other indications in the United States, or EU, we must obtain additional FDA approval and/or approval of the EMEA. Our other products are in development, and will have to be approved by the FDA before they can be marketed in the United States and by the EMEA before they can be marketed in the EU. Obtaining FDA or other national regulatory approval requires substantial time, effort and financial resources, and we may not obtain approval on a timely basis, if at all. If the FDA or the EMEA do not approve our developmental products and any additional indications for marketed products in a timely fashion, or does not approve them at all, our business and financial condition may be adversely affected.

In addition, we and our currently marketed product and our product candidates are subject to comprehensive regulation by the FDA and the EMEA. Regulation by the FDA and EMEA begins before approval for marketing is granted and continues during the life of each product. For example, TRISENOX was approved for its current indication by the FDA under its accelerated approval process and by the EMEA “under exceptional circumstances,” and we committed to completing several post-approval requirements to both the FDA and the EMEA, including the conduct of additional clinical studies. If we fail to fulfill these obligations, the FDA or EMEA may withdraw approval of TRISENOX. In addition, the FDA and other regulatory authorities regulate, for example, research and development, including pre-clinical and clinical testing, safety, effectiveness, manufacturing, labeling, advertising, promotion, export, and marketing of our products. Manufacturing processes must conform to cGMPs. The FDA and other regulatory authorities periodically inspect manufacturing facilities to assess compliance with cGMPs. Accordingly, manufacturers must continue to expend time, money, and effort to maintain compliance. Also, a drug may not be promoted for other than its approved indication, and the FDA, EMEA and other regulatory authorities may institute enforcement actions against companies that do so. Our failure to comply with this or other FDA or other regulatory requirements may result in various adverse consequences, including possible delay in approval or refusal to approve a product, recalls, seizures, withdrawal of an approved product from the market, and/or the imposition of civil or criminal sanctions.

Additionally, we are subject to numerous regulations and statutes regulating the manner of selling and obtaining reimbursement for our products that receive marketing approval. For example, federal statutes generally prohibit providing certain discounts and payments to physicians to encourage them to prescribe our product. Violations of such regulations or statutes may result in treble damages, criminal or civil penalties, fines or exclusion of CTI or its employees from participation in federal and state health care programs. Although we have policies prohibiting violations of relevant regulations and statutes, unauthorized actions of our employees or consultants, or unfavorable interpretations of such regulations or statutes may result in third parties or regulatory agencies bringing legal proceedings or enforcement actions against us.

As a result of our merger with Novuspharma, we are required to comply with the regulatory structure of Italy, which could result in administrative challenges.

As a result of our merger with Novuspharma, our operations now need to comply not only with applicable laws of and rules of the United States, including Washington law and the rules and regulations of the Securities and Exchange Commission and the Nasdaq National Market, but also the EU legal system and the Republic of Italy, including the rules and regulations of CONSOB and Borsa Italiana, which collectively regulate companies listed on Italy’s public markets such as the Nuovo Mercato. Conducting our operations in a manner that complies with all applicable laws and rules will require us to devote additional time and resources to regulatory compliance matters. For example, the process of seeking to understand and comply with the laws of each country, including tax, labor and regulatory laws, might require us to incur the expense of engaging additional outside counsel, accountants and other professional advisors and might result in delayed business initiatives as we seek to ensure that each new initiative will comply with all regulatory regimes.

As a result of our merger with Novuspharma, we are subject to new legal duties and additional political and economic risks related to our operations in Italy.

As a result of our merger with Novuspharma, a portion of our business is based in Italy. We are subject to duties and risks arising from doing business in Italy, such as:

These risks related to doing business in Italy could harm the results of our operations.

Uncertainty regarding third party reimbursement and health care cost containment initiatives may limit our returns.

The ongoing efforts of governmental and third party payors to contain or reduce the cost of health care may affect our ability to commercialize our products successfully. Governmental and other third party payors are increasingly attempting to contain health care costs by:

The trend toward managed health care in the United States, the growth of organizations such as health maintenance organizations, and legislative proposals to reform healthcare and government insurance programs could significantly influence the purchase of healthcare services and products, resulting in lower prices and reducing demand for our products. In addition, in almost all European markets, pricing and choice of prescription pharmaceuticals are subject to governmental control. Therefore, the price of our products and their reimbursement in Europe will be determined by national regulatory authorities.

Even if we succeed in bringing any of our proposed products to the market, they may not be considered cost-effective and third party reimbursement might not be available or sufficient. If adequate third party coverage is not available, we may not be able to maintain price levels sufficient to realize an appropriate return on our investment in research and product development. In addition, legislation and regulations affecting the pricing of pharmaceuticals may change in ways adverse to us before or after any of our proposed products are approved for marketing. While we cannot predict whether any such legislative or regulatory proposals will be adopted, the adoption of such proposals could make it difficult or impossible to sell our products. TRISENOX has been reimbursed by third party payors, but there is no guarantee this reimbursement will continue.

We face direct and intense competition from our competitors in the biotechnology and pharmaceutical industries and we may not compete successfully against them.

Competition in the oncology industry is intense and is accentuated by the rapid pace of technological development. We anticipate that we will face increased competition in the future as new companies enter our markets. Our competitors in the United States and elsewhere are numerous and include, among others, major multinational pharmaceutical companies, specialized biotechnology companies and universities and other research institutions. Specifically:

Many of our competitors, either alone or together with their collaborators and in particular, the multinational pharmaceutical companies, have substantially greater financial resources and development and marketing teams

than us. In addition, many of our competitors, either alone or together with their collaborators, have significantly greater experience than we do in developing, manufacturing and marketing products. As a result, these companies’ products might come to market sooner or might prove to be more effective, to be less expensive, to have fewer side effects or to be easier to administer than ours. In any such case, sales of our products or eventual products would likely suffer and we might never recoup the significant investments we are making to develop these product candidates.

If we lose our key personnel or we are unable to attract and retain additional personnel, we may be unable to pursue collaborations or develop our own products.

We are highly dependent on Dr. James A. Bianco, our president and chief executive officer, Dr. Jack W. Singer, our chief medical officer and Silvano Spinelli, our executive vice president of development and managing director of European operations. The loss of any one of these principal members of our scientific or management staff, or failure to attract or retain other key scientific employees, could prevent us from pursuing collaborations or developing and commercializing our products and core technologies. Recruiting and retaining qualified scientific personnel to perform research and development work are critical to our success. There is intense competition for qualified scientists and managerial personnel from numerous pharmaceutical and biotechnology companies, as well as from academic and government organizations, research institutions and other entities. In addition, we will rely on consultants and advisors, including scientific and clinical advisors, to assist us in formulating our research and development strategy. All of our consultants and advisors will be employed by other employers or are self-employed, and will have commitments to or consulting or advisory contracts with other entities that may limit their availability to us.

The integration of Novuspharma’s business and operations will be a challenging, complex, time-consuming and expensive process and may disrupt our business if not completed in a timely and efficient manner.

The challenges involved in the integration of Novuspharma include the following:

We may not succeed in addressing these challenges or any other problems encountered in connection with integration following the merger, which may be exacerbated by the geographic separation of our operations in the United States and in Italy. If management is not able to address these challenges, we may not achieve the anticipated benefits of the merger, which may have a material adverse effect on our business and could result in the loss of key personnel.

Our limited operating experience may cause us difficulty in managing our growth and could seriously harm our business.

As a result of additional trials for TRISENOX for indications other than relapsed or refractory APL and clinical trials currently underway for XYOTAX, pixantrone and our other products in development, we have expanded our operations in various areas, including our management, regulatory, clinical, financial and information systems and other elements of our business process infrastructure. We may need to add additional key personnel in these areas. In addition, as growth occurs, it may strain our operational, managerial and financial resources. We willmay not be able to increase revenues or control costs unless we continue to improve our operational, financial, regulatory and managerial systems and processes, and expand, train and manage our work force.

If we fail to keep pace with rapid technological change in the biotechnology and pharmaceutical industries, our products could become obsolete.

Biotechnology and related pharmaceutical technology have undergone and are subject to rapid and significant change. We expect that the technologies associated with biotechnology research and development will continue to develop rapidly. Our future will depend in large part on our ability to maintain a competitive position with respect to these technologies. Any compounds, products or processes that we develop may become obsolete before we recover any expenses incurred in connection with developing these products.

We face direct and intense competition from our rivals in the biotechnology and pharmaceutical industries and we may not compete successfully against them.

The biotechnology and pharmaceutical industries are intensely competitive. We have numerous competitors in the United States and elsewhere. Our competitors include major, multinational pharmaceutical and chemical companies, specialized biotechnology firms and universities and other research institutions. Many of these competitors have greater financial and other resources, larger research and development staffs and more effective marketing and manufacturing organizations, than we do. In addition, academic and government institutions have become increasingly aware of the commercial value of their research findings. These institutions are now more likely to enter into exclusive licensing agreements with commercial enterprises, including our competitors, to market commercial products.

Our competitors may succeed in developing or licensing technologies and drugs that are more effective or less costly than any we are developing. Our competitors may succeed in obtaining FDA or other regulatory approvals for drug candidates before we do. In particular, we face direct competition from many companies focusing on delivery technologies. Drugs resulting from our research and development efforts, if approved for sale, may not compete successfully with our competitors’ existing products or products under development.

We may need to raise additional funds in the future, and they may not be available on acceptable terms, or at all.

We expect that our existing capital resources and the interest earned thereon will enable us to maintain our planned operations through at least mid 2004. Beyond that time, if our capital resources are insufficient to meet future capital requirements, we will have to raise additional funds to continue the development of our technologies and complete the commercialization of products, if any, resulting from our technologies. We will require substantial funds to: (1) continue our research and development programs, (2) in-license or acquire additional technologies, (3) conduct preclinical studies and clinical trials and (4) launch new drug products. We may need to raise additional capital to fund our operations repeatedly. We may raise such capital through public or private equity financings, partnerships, debt financings, bank borrowings, or other sources. Our capital requirements will depend upon numerous factors, including the following:

Additional funding may not be available on favorable terms or at all. If adequate funds are not otherwise available, we may curtail operations significantly. To obtain additional funding, we may need to enter into arrangements that require us to relinquish rights to certain technologies, drug candidates, products and/or potential markets. To the extent that additional capital is raised through the sale of equity, or securities convertible into equity, you may experience dilution of your proportionate ownership of the company.

Our stock price is extremely volatile, which may affect our ability to raise capital in the future.

The market price for securities of biopharmaceutical and biotechnology companies, including ours, historically has been highly volatile, and the market from time to time has experienced significant price and

volume fluctuations that are unrelated to the operating performance of such companies. For example, during the twelve months ended December 31, 2002, our stock price has ranged from a low of $2.68 to a high of $27.45. Fluctuations in the trading price or liquidity of our common stock may adversely affect our ability to raise capital through future equity financings.

Factors that may have a significant impact on the market price and marketability of our common stock include:

In the past, following periods of volatility in the market price of a company’s securities, securities class action litigation has often been instituted. If a securities class action suit is filed against us, we would incur substantial legal fees and our management’s attention and resources would be diverted from operating our business in order to respond to the litigation.

We may be unable to attain the raw materials necessary to produce our XYOTAX product candidate in sufficient quantity to meet demand when and if such product is approved.

Paclitaxel is derived from certain varieties of yew trees. Supply of yew trees is tightly controlled by a limited number of companies. We cannot be sure that we will be able to continue to purchase the materials necessary to produce XYOTAX in adequate volume and quality. We purchase the majority of the paclitaxel we need from a single vendor. Should the paclitaxel purchased from this source prove to be insufficient in quantity or quality, or should this relationship terminate, there can be no assurance that we will be able to enter into a similar agreement with an alternate source.

Our dependence on third party manufacturers means that we may not have sufficient control over the manufacture of our products.

We currently do not have internal facilities for the manufacture of any of our products for clinical evaluation or commercial production. In addition, TRISENOX, our first commercial product, is currently manufactured by a single vendor. In 2002, we began the process of transitioning our finished product manufacturing for TRISENOX to a new additional supplier. We believe that this additional supplier will receive FDA approval to manufacture

TRISENOX in the second half of 2003. If this approval is not received and we are unable to obtain supply from our current vendor, we may not have sufficient supply to meet our future sales demand. We will need to develop additional manufacturing resources, enter into collaborative arrangements with other parties that have established manufacturing capabilities or elect to have other third parties manufacture our products on a contract basis. We are dependent on such collaborators or third parties to supply us in a timely way with products manufactured in compliance with standards imposed by the FDA and foreign regulatory authorities. The manufacturing facilities of contract manufacturers may not comply with applicable manufacturing regulations of the FDA nor meet our requirements for quality, quantity or timeliness. Another of our products under development, XYOTAX, is complex to manufacture, which may prevent us from obtaining a sufficient supply for the increased clinical trials that are currently planned or underway.

We may face difficulties in achieving acceptance of our products in the market if we do not continue to expand our sales and marketing infrastructure.

We currently are marketing TRISENOX with our direct sales force. Competition for these individuals is intense, and in the event we need additional sales personnel, we may not be able to hire the experience required and number of sales personnel we need. In addition, if we market and sell products other than TRISENOX, we may need to further expand our marketing and sales force with sufficient technical expertise and distribution capacity. If we are unable to expand our direct sales operations and train new sales personnel as rapidly as necessary, we may not be able to increase market awareness and sales of our products, which may prevent us from growing our revenues and achieving and maintaining profitability.

If we lose our key personnel or are unable to attract and retain additional personnel, we may be unable to pursue collaborations or develop our own products.

We are highly dependent on Dr. James A. Bianco, our Chief Executive Officer, and Dr. Jack W. Singer, our Executive Vice President, Research Program Chairman. The loss of these principal members of our scientific or management staff, or failure to attract or retain other key scientific employees, could prevent us from pursuing collaborations or developing our products and core technologies. Recruiting and retaining qualified scientific personnel to perform research and development work are critical to our success. There is intense competition for qualified scientists and managerial personnel from numerous pharmaceutical and biotechnology companies, as well as from academic and government organizations, research institutions and other entities. In addition, we rely on consultants and advisors, including our scientific and clinical advisors, to assist us in formulating our research and development strategy. All of our consultants and advisors are employed by other employers or are self-employed, and have commitments to or consulting or advisory contracts with other entities that may limit their availability to us.

We are subject to extensive government regulation, including the requirement of approval before our products may be marketed.

The FDA has approved only one of our products, TRISENOX, for sale in the United States, for relapsed or refractory APL. Before we can market TRISENOX for other indications, we must obtain FDA approval. Our other products are in development, and will have to be approved by the FDA before they can be marketed in the United States. If the FDA does not approve our products and any additional indications for marketed products in a timely fashion, or does not approve them at all, our business and financial condition may be adversely affected.

In addition, we and our products are subject to comprehensive regulation by the FDA both before and after products are approved for marketing. The FDA regulates, for example, research and development, including preclinical and clinical testing, safety, effectiveness, manufacturing, labeling, advertising, promotion, export, and marketing of our products. Our failure to comply with regulatory requirements may result in various adverse consequences including FDA delay in approving or refusal to approve a product, recalls, withdrawal of an approved product from the market, and/or the imposition of civil or criminal sanctions.

Because there is a risk of product liability associated with our products, we face potential difficulties in obtaining insurance.

Our business exposes us to potential product liability risks inherent in the testing, manufacturing and marketing of human pharmaceutical products, and we may not be able to avoid significant product liability exposure. While we have insurance covering product use in our clinical trials, and currently have product liability insurance for TRISENOX, it is possible that we will not be able to maintain such insurance on acceptable terms or that any insurance obtained will provide adequate coverage against potential liabilities. Our inability to obtain sufficient insurance coverage at an acceptable cost or otherwise to protect against potential product liability claims could prevent or limit the commercialization of any products we develop. A successful product liability claim in excess of our insurance coverage could exceed our net worth.

Uncertainty regarding third party reimbursement and health care cost containment initiatives may limit our returns.

The ongoing efforts of governmental and third party payors to contain or reduce the cost of health care will affect our ability to commercialize our products successfully. Governmental and other third party payors are increasingly attempting to contain health care costs by:

In addition, the trend toward managed health care in the United States, the growth of organizations such as health maintenance organizations, and legislative proposals to reform healthcare and government insurance programs could significantly influence the purchase of healthcare services and products, resulting in lower prices and reducing demand for our products.

Even if we succeed in bringing any of our proposed products to the market, they may not be considered cost-effective and third party reimbursement might not be available or sufficient. If adequate third party coverage is not available, we may not be able to maintain price levels sufficient to realize an appropriate return on our investment in research and product development. In addition, legislation and regulations affecting the pricing of pharmaceuticals may change in ways adverse to us before or after any of our proposed products are approved for marketing. While we cannot predict whether any such legislative or regulatory proposals will be adopted, the adoption of such proposals could make it difficult or impossible to sell our products. TRISENOX has been reimbursed by third party payors, but there is no guarantee this reimbursement will continue.

Since we use hazardous materials in our business, we may be subject to claims relating to improper handling, storage or disposal of these materials.

Our research and development activities involve the controlled use of hazardous materials, chemicals and various radioactive compounds. We are subject to federal, state and local laws and regulations governing the use, manufacture, storage, handling and disposal of such materials and certain waste products. Although we believe that our safety procedures for handling and disposing of such materials comply with the standards prescribed by state and federal regulations, the risk of accidental contamination or injury from these materials cannot be eliminated completely. In the event of such an accident, we could be held liable for any damages that result and any such liability not covered by insurance could exceed our resources. Compliance with environmental laws and regulations may be expensive, and current or future environmental regulations may impair our research, development or production efforts.

We may not be able to conduct animal testing in the future, which could harm our research and development activities.

Certain of our research and development activities involve animal testing. Such activities have been the subject of controversy and adverse publicity. Animal rights groups and other organizations and individuals have attempted to stop animal testing activities by pressing for legislation and regulation in these areas.areas and by disrupting activities through protests and other means. To the extent the activities of these groups are successful, our business could be materially harmed by delaying or interrupting our research and development activities.

BecauseOur stock price is extremely volatile, which may affect our charter documents contain certain anti-takeover provisions and we have a rights plan, it may be more difficult for a third partyability to acquire us, andraise capital in the rights of some shareholders could be adversely affected.future.

The market price for securities of biopharmaceutical and biotechnology companies, including ours, historically has been highly volatile, and the market from time to time has experienced significant price and volume fluctuations that are unrelated to the operating performance of such companies. For example, during the twelve months ended December 31, 2003, our stock price ranged from a low of $5.18 to a high of $15.70. Fluctuations in the trading price or liquidity of our common stock may adversely affect our ability to raise capital through future equity financings.

Factors that may have a significant impact on the market price and marketability of our common stock include:

In the past, following periods of volatility in the market price of a company’s securities, securities class action litigation has often been instituted. If a securities class action suit is filed against us, we would incur substantial legal fees and our management’s attention and resources would be diverted from operating our business in order to respond to the litigation.

Our Restated Articles of Incorporation and Bylawscharter documents contain provisions that may make it more difficult forprevent or delay removal of incumbent management or a third partychange of control.

Provisions of our articles of incorporation and bylaws may have the effect of deterring or delaying attempts by our shareholders to acquireremove or make a bid for us.replace management, proxy contests and changes in control. These provisions could limitinclude:

In addition, as a Washington corporation, we are subject to Washington law, including Chapter 23 of the holdersWashington Business Corporations Act, which prohibits public companies from engaging in some business combinations without the approval of common stock will be subjecta majority of the votes within each voting group entitled to and may be adversely affected by,vote separately on the rights of any holders of preferred stock that may be issued in the future. The issuance of preferred stock, while providing desirable flexibility in connection with possible acquisitions and other corporate purposes,transaction.

These provisions, alone or together, could have the effect of making it more difficult for a third party to acquire,deterring or of discouraging a third party from acquiring, a majority of our outstanding voting stock. We have no present plans to issue any additional shares of preferred stock. In addition, we have adopted a shareholder rights plan that, along with certain provisions of our Restated Articles of Incorporation, may have the effect of discouraging certain transactions involving a change of control of the company.delaying changes in incumbent management, proxy contests or changes in control.

Item 2.    Properties

We lease approximately 68,000 square feet of space at 201 Elliott Avenue West in Seattle, Washington for our laboratory and administrative operations. The lease expires in January 2008, with a one five-year renewal option at the then prevailing market rent. We also lease approximately 110,000 square feet of space at 501 Elliott Avenue West in Seattle, Washington for our executive offices and administrative operations. The lease expires in July 2012. To accommodate the operational requirements of our wholly-owned subsidiaries, Cell Therapeutics (UK) Limited and Cell Therapeutics Corporate Development, Inc., we leased additional space in London, UK and Hillsboro, Oregon, respectively. CTI (Europe), our newly acquired European branch, leases approximately 75,000 square feet of office and laboratory space in Bresso (Milan), Italy. We believe our existing and planned facilities are adequate to meet our present requirements. We currently anticipate that additional space will be available to us, when needed, on commercially reasonable terms.

Item 3.    Legal Proceedings

We are notOn February 10, 2004, Micromet AG (“Micromet”), a partyGerman company, filed a complaint against the Company in federal district court in Washington, asserting that CTI (Europe), the Company’s European Branch (formerly known as Novuspharma S.p.A.), had purportedly breached a contract with Micromet for the development of a drug candidate known as MT-201. The alleged breach is based on the assertion that CTI (Europe) failed to anypay for certain milestones and development expenses owed under the contract. On February 23, 2004, the Company answered the complaint, denying the substance of the allegations, and filing counterclaims for breach of contract and for rescission of the contract based on Micromet’s misrepresentations and failures to disclose material legal proceedings.information. The Company believes that Micromet’s complaint is without merit and intends to vigorously defend against the Micromet action, as well as to seek recovery based upon its counterclaims.

Item 4.    Submission of Matters to a Vote of Security Holders

No matters were submitted to a vote of security holders during the fourth quarter of the year ended December 31, 2002.

(a)

On October 23, 2003, we held a Special Meeting of Shareholders (the “Special Meeting”). Each share of Common Stock was entitled to one vote per share.

PART II

Item 5.    Market for Registrant’s Common Equity and Related Shareholder Matters

Our common stock is traded on the Nasdaq National Market under the symbol “CTIC.” The following table sets forth, for the periods indicated, the high and low reported sales prices per share of the common stock as reported on the Nasdaq National Market.

On March 24, 2003,4, 2004, the last reported sale price of our common stock on the Nasdaq Market was $8.73$8.70 per share. As of March 24, 2003,4, 2004, there were approximately 264255 shareholders of record of our common stock.

Effective January 2, 2004, we commenced the trading of our common stock on the Nuovo Mercato in Italy under the ticker symbol (CTIC).

Dividend Policy

We have not declared or paid any cash dividends on our capital stock since our inception. We currently intend to retain all of our cash and any future earnings to finance the growth and development of our business and therefore do not anticipate paying any cash dividends in the foreseeable future. Any future determination to pay cash dividends will be at the discretion of the Board of Directors and will be dependent upon our consolidated financial condition, results of operations, capital requirements and such other factors as the Board of Directors deems relevant.

Sales of Unregistered Securities

None.

Equity Compensation Plan Information

The following table gives information about our common stock that may be issued upon the exercise of options, warrants, and rights under all of our existing compensation plans as of December 31, 2002,2003, including the 2003 Equity Incentive Plan, 1994 Equity Incentive Plan and the 1996 Employee Stock Purchase Plan.

Cell Therapeutics, Inc. Novuspharma S.p.A. Stock Option Plan

In December 2003, the Board of Directors approved the assumption and amendment and restatement of the Cell Therapeutics, Inc. Novuspharama S.p.A. Stock Option Plan (the “Plan”) in connection with the merger between CTI and Novuspharma. The Plan provides for the grant of nonqualified stock options and restricted stock to certain of our (and our affiliates) officers, employees, members of our Board of Directors, and consultants. The plan administrator determines, on a grant-by-grant basis, what terms and conditions apply to options and restricted stock granted under the Plan (including vesting restrictions). The Plan permits options to be exercised with cash or certain other legal forms of consideration. In the event of our change of control (including our merger with or into another corporation or our sale of substantially all of our assets), the Plan provides that we may determine, in our discretion, that each optionee may vest in his or her option or restricted stock award with respect to any or all of the shares subject to the award (including shares that were unvested prior to the change of control) and that such awards may otherwise be assumed or substituted for by the successor corporation. There are 350,000 shares of common stock reserved under the Plan, and 148,200 shares remain for future issuance.

License Agreement with PG-TXL Company, L.P.

In 1998, we issued warrants to purchase 350,000 shares of our common stock in connection with a license agreement with PG-TXL Company, L.P. These warrants expire in 2008 and have an exercise price of $20.00. We also issued 103,665 restricted share rights to non-employees for which ownership vests upon the achievement of future events.

Warrants Issued to Placement Agents

In 1999, we completed a $10 million private placement of 10,000 shares of Series D convertible preferred stock and warrants to acquire 1,523,810 shares of common stock. In connection with the offering, we issued warrants to purchase shares of common stock to a placement agent. These warrants expire in 2004, and have an exercise price of $2.38.

In 2000, we completed a $40 million private placement of 3,333,334 shares of common stock. In connection with the offering, we issued warrants to purchase shares of common stock to a placement agent and finder. These warrants expire in 2005, and have an exercise price of $13.20.

Research Services Agreement with The Hope Heart Institute

In 2002, we entered into an agreement with The Hope Heart Institute for research services. In connection with this agreement, we issued fully-vested warrants to purchase shares of common stock at an exercise price of $10.00. These warrants expire in 2007.

Item 6.    Selected Consolidated Financial Data

The data set forth below should be read in conjunction with Item 7. “Management’s Discussion and Analysis of Consolidated Financial Condition and Results of Operations” and the Consolidated Financial Statements and Notes thereto appearing at Item 8 of this report.

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

The following discussion should be read in conjunction with the “Selected Financial Data” and the Consolidated Financial Statements and the related Notes included in Items 6 and 8 of this Form 10-K. The following discussion contains forward-looking statements that involve risks and uncertainties. Such statements, which include statements concerning research and development expenses, selling, general and administrative expenses, additional financings and additional losses, are subject to risks and uncertainties, including, but not limited to, those discussed below and elsewhere in this Form 10-K, particularly in “Factors Affecting Our Operating Results,” that could cause actual results to differ significantly from those projected.

Overview

We develop, acquire and commercialize novel treatments for cancer. Our goal is to build a leading, vertically-integrated biopharmaceutical company with a diversified portfolio of proprietary oncology drugs. Our research and in-licensing activities are concentrated on identifying new, less toxic and more effective ways to treat cancer. As of December 31, 2002,2003, we had incurred aggregate net losses of approximately $340.5$470.5 million since inception.inception, not including losses of Novuspharma. We expect to continue to incur significant additional operating losses over the next severalcouple of years from our research and development efforts.

On January 1, 2004, we completed our acquisition of Novuspharma, an Italian-public biopharmaceutical company focused on oncology. At the time of the merger, approximately 15,629,000 shares of CTI common stock were issued based on the conversion of approximately 6,379,000 outstanding Novuspharma ordinary shares multiplied by the fixed exchange ratio of 2.45. The total cost of the merger is estimated to be approximately $196.3 million. This acquisition provided us worldwide rights to pixantrone, approximately $92.5 million of cash and cash equivalents, a high-quality drug discovery organization and staff with an extensive track record in cancer drug development. The Novuspharma acquisition and pixantrone are consistent with our strategy of growth by strategic acquisition and our goal to develop improved cancer therapies.

XYOTAX

In June 1998, we entered into an agreement with PG-TXL Company, L.P. and scientists at the M.D. Anderson Cancer Center, granting us an exclusive worldwide license to the rights to PG-TXL, and to all potential uses of PG-TXL’s polymer technology. PG-TXL is paclitaxel linked to polyglutamate, and is branded as XYOTAX^™. Under the terms of the agreement, we will fund the research, development, manufacture, marketing and sale of drugs developed using PG-TXL’s polymer technology. We will be obligated to make future milestone payments upon the attainment of significant achievements, as defined in the agreement, of up to $15.5 million, and royalty payments on net product sales. As of December 31, 2002,2003, we have made $5.0 million in milestone payments.

In September 2001, we entered into a supply agreement with Natural Pharmaceuticals, Inc. for paclitaxel, a key starting material for our XYOTAX drug candidate. Under the supply agreement, we purchased paclitaxel at a pre-determined price and willexpect to receive supply through early 2004.

In October 2001, we entered into a licensing agreement with Chugai Pharmaceutical Co., Ltd., or Chugai, for the development and commercialization of XYOTAX. This agreement grants an exclusive license to Chugai to develop and commercialize XYOTAX in several Asian markets. Upon execution of the agreement, Chugai paid us a $3.0 million initial payment, which has been recorded as deferred revenue and is being recognized as license revenue over the development period of approximately six years on a straight-line basis. Under the agreement, we received and recognized as revenue a $3.0 million milestone payment during 2002, we may also receive future milestone payments totaling up to $13.0 million upon Chugai’s achievement of certain product development milestones, and we are entitled to receive royalties on product sales in the territories covered under

the agreement. We received and recognized as revenue a $3.0 million milestone payment and approximately $1.9 million in development expenditure reimbursements from Chugai during 2002. Chugai has also committed to incur up to $54.0 million in development expenditures over the course of the licensing agreement. During

In 2002, we initiated a XYOTAX phase III clinical trial for second-line treatment of non-small cell lung cancer,NSCLC, and two additional phase III trials of XYOTAX in the front line treatment of poor performance status patients with non-small cell lung cancer.NSCLC.

In November 2002, we announced that the Gynecologic Oncology Group, or GOG plans to conduct a phase III trial of XYOTAX in front-line treatment of ovarian cancer. ThisThe GOG plans to submit a protocol under the FDA’s special protocol assessment and the trial is expected to begin in late 2003. We had initiated athe first half of 2004. In June 2003, we received fast track designation from the FDA for our XYOTAX pivotal trials in PS2 patients with advance NSCLC.

In November 2003, we completed enrollment in one of our XYOTAX phase III ovarian clinical trial in July 2002, but have since closed this trial as a resultpivotal trials for the front-line treatment of the GOG’s decision to conduct the trial for us.poor performance status patients with non-small cell lung cancer.

TRISENOX

InOn January 7, 2000, we acquired TRISENOX upon our acquisition ofentered into a Merger Agreement to acquire PolaRx Biopharmaceuticals, Inc., or PolaRx, a single productbiopharmaceutical company that owned the rights to TRISENOX.TRISENOX (arsenic trioxide), an anti-cancer compound for which we submitted and received approval for a New Drug Application with the FDA. The aggregateacquisition was accounted for as a purchase pricetransaction. Under the terms of the Merger Agreement, we have made an additional contingent payment of approximately $36.2$4.0 million consisted primarily of fivefor meeting a $10.0 million shares of common stock and included assumed net liabilities of $3.9 million from PolaRx. In connection with the achievement of a $10 millionTRISENOX sales threshold we recorded $4.0in 2002 and are currently obligated to pay $5.0 million in additionalcash for achieving a $20.0 million TRISENOX sales threshold in 2003, recorded as goodwill and a currentshort-term liability as of December 31, 2002, which we will pay in cash in 2003. AnWe are also required to make an additional $5.0 million payout tied to a sales threshold of $20 million in any four consecutive quarters is payable in cash or common stock with a market value of $5 million, within thirty days following the end of the first calendar quarter after the end of the previous four calendar quarter period in which such threshold is achieved. For any calendar year that sales of TRISENOX exceed $40 million, PolaRx shareholders will receive a 2% royalty on total net sales, for that year payable in cash or common stock at the then fair market value of our common stock, payable within thirty days following the endfor any calendar year that sales of the calendar year. The acquisition was accounted for as a purchase transaction. Any additional or contingent payments made to PolaRx shareholders will be considered additional purchase price and will be capitalized as additional goodwill.TRISENOX exceed $40.0 million.

In September 2000, we received marketing approval of our New Drug Application, or NDA by the Food and Drug Administration, or FDA for TRISENOX (arsenic trioxide),. Sales of TRISENOX in the U.S. commenced sales in October 2000 and have recorded cumulative net product sales for TRISENOX of approximately $18.0 million through December of 2002. Inin March 2002, we received from the European Agency for the Evaluation of Medicinal Products, or EMEA approval to market TRISENOX in the European Community, or EU. We commenced the launch and sale of TRISENOX in the EU during the second quarter of 2002. We have recorded cumulative net product sales for TRISENOX of approximately $40.1 million through December 31, 2003. TRISENOX is manufactured primarily by a single vendor and sold through our direct sales force.

In December 2002, we entered into a distribution agreement with Nippon Shinyaku Co. Ltd., or Nippon. This agreement grants an exclusive license to Nippon to market and distribute TRISENOX (arsenic trioxide) injection in Japan, South Korea, and Taiwan. Upon execution of the agreement, Nippon paid us a $750,000 initial payment, which we recorded as deferred revenue and which is being recognized as revenue over the performance period of approximately eighteen months on a straight-line basis. Under the agreement, we received and recognized as revenue a $500,000 milestone payment from Nippon in 2003 and we may also receive additional future milestone payments totaling up to $4.0$3.5 million upon attainment of certain achievements.

Other CompoundsPIXANTRONE

We acquired pixantrone, a novel anthracycline, for the treatment of NHL, through our merger with Novuspharama S.p.A. in January 2004. We are developing pixantrone, and plan to initiate a pivotal phase III trial in relapsed aggressive NHL in the first half of 2004.

OTHER COMPOUNDS

We are developing a novel polyglutamate-camptothecin molecule, or PG-CPT.CT-2106. We filed a U.S. investigational new drug application, or IND, in December 2001 for this compound, and initiated a phase I clinical study in the first quarter of 2002.

We have identified a novel drug target called lysophosphatidic acid acyltransferase, or LPAAT-ß, that when inhibited has been shown2002 and plan to reduce tumorinitiate three phase II trials in small cell growth in preclinical models. We are in the process of identifying a lead candidate for clinical developmentlung cancer, relapsed ovarian cancer and relapsed colorectal cancer during 2003.2004.

Critical Accounting Policies and Estimates

In December 2001,Management makes certain judgments and uses certain estimates and assumptions when applying accounting principles generally accepted in the SEC requestedUnited States in the preparation of our consolidated financial statements. We evaluate our estimates and judgments on an on-going basis and base our estimates on historical experience and on assumptions that all registrants discuss their most “critical accounting policies” in management’s discussionwe believe to be reasonable under the circumstances. Our experience and analysisassumptions form the basis for our judgments about the carrying value of financial conditionassets and liabilities that are not readily apparent from other sources. Actual results of operations. The SEC indicated that a “critical accounting policy” is one which is both important to the portrayal of the company’s financial conditionmay vary from what we anticipate and results and requires management’s most difficult, subjectivedifferent assumptions or complex judgments, often as a result of the need to make estimates about the effect of matters that are inherently uncertain.future could change our reported results. While our significant accounting policies are more fully described in Note 1 to our consolidated financialfinancials statements, included in this report, we believe the following accounting policies are the most critical to be critical:us, in that they are important to the portrayal of our consolidated financial statements and require our most difficult, subjective or complex judgments in the preparation of our consolidated financial statements.

Product Sales

We recognize revenue from product sales when there is persuasive evidence that an arrangement exists, title has passed and delivery has occurred, the price is fixed and determinable, and collectibility is reasonably assured. Product sales are recorded net of an allowance for estimated returns and discounts. Customers may return damaged or expired inventory up to twelve months after the expiration date. In estimating returns, we analyze historical returns, sales patterns, estimated inventory on hand at the distributors and the remaining shelf life of that inventory. In arriving at the accrual for product returns we match the returns to the corresponding production batch to assess the historical trend for returns. Based on this analysis, the estimated return percentage is applied to current period sales. Allowances for returns, discounts and bad debts are netted against accounts receivable.

License Agreement Revenues

We may generate revenue from technology licenses, collaborative research and development arrangements, and cost reimbursement contracts. Revenue under technology licenses and collaborative agreements typically

consists of nonrefundable and/or guaranteed technology license fees, collaborative research funding, and various milestone and future product royalty or profit-sharing payments.

Revenue associated with up-front license fees, and research and development funding payments under collaborative agreements is recognized ratably over the relevant periods specified in the agreement, generally the research and development period. If the time period is not defined in the agreement, we calculate the revenue recognition period based on our current estimate of the research and development period considering experience with similar projects, level of effort and the stage of development. Should there be a change in our estimate of the research and development period, we will revise the term over which the initial payment is recognized. Revenue from substantive at-risk milestones and future product royalties is recognized as earned based on the completion of the milestones and product sales, as defined in the respective agreements. Revenue under cost reimbursement contracts is recognized as the related costs are incurred. Payments received in advance of recognition as revenue are recorded as deferred revenue.

Product SalesIn November 2002, the Emerging Issues Task Force (or “EITF”) of the Financial Accounting Standards Board (or “FASB”) issued EITF 00-21, “Revenue Arrangements with Multiple Deliverables,” which addresses

We recognizecertain aspects of the accounting for arrangements that involve the delivery or performance of multiple products, services and/or rights to use assets. Under EITF 00-21, revenue from product sales whenarrangements with multiple deliverables should be divided into separate units of accounting if the deliverables meet certain criteria, including whether the delivered items have a stand-alone value for the customer and whether there is persuasive evidence that an arrangement exists, delivery has occurred,of fair value of the price is fixed and determinable, and collectibility is reasonably assured. Product sales are recorded netundelivered items. In addition, the total consideration should be allocated among the separate units of an allowance for estimated returns, and discountsaccounting based on historical experience. In estimating returns, we analyze historical returns, sales patterns, current inventory on hand at the distributorstheir fair values and the remaining shelf lifeapplicable revenue recognition criteria should be considered separately for each of that inventory. In arriving at the accrualseparate units of accounting. We have adopted EITF 00-21 as of July 1, 2003, and will apply its provisions for product returnsall revenue arrangements we matchenter into on or after this date. The adoption of EITF 00-21 did not have a material impact on our revenue recognition accounting policies, financial position or result of operations. We will continue to evaluate the returns toimpact of EITF 00-21 as we enter into new revenue arrangements in the corresponding production batch to assess the historical trend for returns. Based on this analysis, the estimated return percentage is applied to current period sales. Allowances for discounts, returns and bad debts are netted against accounts receivable.future.

Inventory

Inventory is stated at the lower of cost or market. Cost is determined using a weighted-average approach that approximates the first-in first-out method. Finished goods inventory consists of our FDA-approvedFDA and EMEA approved pharmaceutical drug, TRISENOX. Prior to FDA approval, the raw material and production costs of TRISENOX were recorded as research and development expense. We also record an allowance for inventory that may expire and become unsaleable due to the expiration of shelf life. In estimating inventory obsolescence reserves, we analyze (i) the shelf life and the expiration date, (ii) sales forecasts and (iii) inventory levels compared to forecasted usage. Judgment is required in determining whether the forecasted sales and usage information is sufficiently reliable to enable us to estimate inventory obsolescence reserve.

Research and Development Expenses

Research and development expenses include related salaries and benefits, clinical trial and related clinical manufacturing costs, contract and other outside service fees, and facilities and overhead costs. Research and development expenses consist of costs incurred for proprietary and collaboration research and development and also include activities such as product registries and investigator and company sponsored clinical trials, production of clinical supplies and regulatory costs associated with product registries.trials. Research and development costs are expensed as incurred. In instances where we enter into agreements with third parties for research, and/or clinical trial, activities,and related clinical trial manufacturing costs, costs are expensed upon the earlier of when non-refundable amounts are due or whenas services are performed. Amounts due under such arrangements may be either fixed fee or fee for service, and may include upfront payments, monthly payments, and payments upon the completion of milestones or receipt of deliverables.

Derivative Financial Instruments

Effective at the beginning of fiscal 2001, we adopted SFAS 133,Accounting for Derivative Instruments and Hedging Activities, as amended. We are subject to risks associated with fluctuations in the LIBOR interest rate from lease payments on our aircraft. Our policy is to hedge a portion of these forecasted transactions through an interest rate swap agreement. This swap agreement has been designated as a cash flow hedge. The portion of the net gain or loss on the derivative instrument that is effective as a hedge is reported as a component of accumulated other comprehensive income or loss in shareholders’ equity (deficit) and is reclassified into earnings in the

same period during which the hedged transaction affects earnings. The remaining net gain or loss on the derivative in excess of the present value of the expected cash flows of the hedged transaction is recorded in earnings immediately. If a derivative does not qualify for hedge accounting, or a portion of the hedge is deemed ineffective, the change in fair value is recorded in earnings. The swap was perfectly effective at December 31, 20022003 and 2001.2002. We do not enter into forward agreements for trading purposes.

Purchase price allocation

The purchase price for Novuspharma S.p.A. was allocated to the tangible and identifiable intangible assets acquired and liabilities assumed based on their estimated fair values at the acquisition date of January 1, 2004. An independent third-party valuation firm was engaged to assist in determining the fair values of in-process research and development, identifiable intangible assets, and certain property, plant, and equipment. Such a

valuation requires significant estimates and assumptions including but not limited to: determining the timing and expected costs to complete the in-process projects, projecting regulatory approvals, estimating future cash flows from product sales resulting from in-process projects, and developing appropriate discount rates and probability rates by project. The Company believes the fair values assigned to the assets acquired and liabilities assumed are based on reasonable assumptions. However, these assumptions may be incomplete or inaccurate, and unanticipated events and circumstances may occur. Additionally, estimates for the purchase price allocation may change as subsequent information becomes available.

Results of Operations

Years ended December 31, 20022003 and 2001.2002.

Product sales.    TRISENOX is our pharmaceutical grade arsenic product that has been approved by the FDA and EMEA to treat patients with relapsed or refractory acute promyelocytic leukemia.APL. We recorded net product sales of approximately $11.4$22.1 million and $6.1$11.4 million for TRISENOX for the years ended December 31, 20022003 and 2001,2002, respectively. The increase in net sales is primarily due to greater demand for our product in 2002. 2003. An increase in net sales in the fourth quarter was due to both an increase in product demand and additional wholesaler purchases resulting from an anticipated price increase that occurred in December 2003.

We expect first quarter sales 2004 may be at or below first quarter sales in 2003 primarily as a result of a technical error made by Center for Medicare Services (CMS) stating a payment rate of $2.81/mg for TRISENOX when administered in a physician’s office versus the correct rateof $32.94/mg. This delayed physicians/patients from receiving accurate approved reimbursement for the product. Following several inquiries from our congressional delegates regarding the CMS error, the correct payment rate was published in early February. While this error will likely contribute to softer than anticipated first quarter sales, based upon recent new patient starts and wholesaler ordering trends we anticipate product demand to increase with overall net sales to increasefor 2004 being substantially higher than 2003. As a result, we expect that fourth quarter sales may be higher than sales in 2003.the first quarter of 2004.

License and contract revenue.    In October 2001, we entered into a licensing agreement with Chugai Pharmaceutical Co., Ltd., or Chugai, for the development and commercialization of XYOTAX. Upon execution of the agreement, Chugai paid us a $3.0 million initial payment, which we recorded as deferred revenue and which is being recognized as revenue over the estimated development period of approximately six years on a straight-line basis. In December 2002, we entered into a distribution agreement with Nippon Shinyaku Co., Ltd., or Nippon, for the distribution and commercialization of TRISENOX. We received $750,000 upon execution of the agreement which we recorded as deferred revenue and which is being recognized as revenue over the performance period of approximately eighteen months on a straight-line basis. For the year ended December 31, 2003, we recognized approximately $2.7 million of license and contract revenue, of which approximately $1.2 million related to cost reimbursements for development expenses received from Chugai in 2003, $1.0 million related to the amortization of the initial payments from Chugai and Nippon, and $0.5 million related to a milestone payment received from Nippon in 2003 for their submission of an NDA in Japan. For the year ended December 31, 2002, we recognized $5.5 million of license and contract revenue, of which $3.0 million related to a milestone payment and $1.9 million for cost reimbursements for development expenses received from Chugai in 2002, and $0.5 million related to the amortization of the initial payments from Chugai and Nippon.

Cost of product sold.    The cost of product sold during the year ended December 31, 2003 and 2002 was approximately $840,000 and $423,000, respectively. Our gross margins have remained consistent. Cost of product sold consists primarily of manufacturing costs, allowances for excess inventory that may expire and become unsaleable, and royalties paid on product sales. We expect product costs in the future to continue to approximate a small percentage of product sales.

Research and development expenses.    Our research and development expenses for compounds under development and discovery research are as follows (in thousands):

Costs for compounds under development include external direct expenses such as principal investigator fees, clinical research organization charges and contract manufacturing fees incurred for preclinical, clinical, manufacturing and regulatory activities associated with preparing the compounds for submissions of new drug applications to the FDA or similar regulatory filings with agencies outside the U.S. Operating costs include our personnel and occupancy expenses associated with developing these compounds. Discovery research costs include primarily personnel, occupancy, and laboratory expenses associated with the discovery and identification of new drug targets and lead compounds. We do not allocate operating costs to the individual compounds under development as our accounting system does not track these costs by individual compound. As a result, we are not able to capture the total cost of each compound. Direct external costs incurred to date for XYOTAX and TRISENOX are $107.8 million and $18.2 million, respectively.

Research and development expenses increased to approximately $89.5 million for the year ended December 31, 2003, from approximately $58.8 million for the year ended December 31, 2002. Costs for our XYOTAX program increased primarily due to increased clinical and manufacturing costs of approximately $27.8 million associated with the set up, initiation and execution of our three phase III clinical trials and other clinical trials. These manufacturing costs were offset in part by a $2.0 million charge for a payment related to the achievement of contractual milestones associated with the completion of a phase II trial and the filing of the first IND application in Japan during the year ended December 31, 2002. TRISENOX costs decreased primarily as a result of a reduction in manufacturing and clinical costs of approximately $1.0 million offset in part by an increase in investigator sponsored trial expenses of approximately $0.6 million. Costs incurred for other compounds decreased primarily due to a $1.0 million milestone payment made in 2002 for the commencement of our phase I clinical trial for CT-2106. Operating expenses increased by approximately $4.6 million primarily due to additional personnel and occupancy costs related to our expanded development plans for XYOTAX, TRISENOX and CT-2106, including employee termination benefits of $0.6 million resulting from a reduction in workforce associated with our merger with Novuspharma. Costs for discovery research increased primarily as a result of $0.9 million in employee termination benefits and $0.8 million in additional occupancy costs. These costs were offset in part by a $0.5 million charge related to the fair value of warrants issued to the Hope Heart Institute in connection with a Sponsored Research Agreement entered into in November 2002 and a $0.4 million decrease in personnel costs. We anticipate increased research and development expenses during 2004 in connection with our clinical development plans for XYOTAX, pixantrone and our other products. Research and development costs may also increase as a result of restructuring charges to be incurred in connection with our election to vacate certain excess laboratory facilities during 2004 due to our integration with Novuspharma.

Our lead drug candidates, XYOTAX, pixantrone and TRISENOX for indications other than relapsed or refractory APL, are currently in clinical trials. Many drugs in human clinical trials fail to demonstrate the desired safety and efficacy characteristics. Even if our drugs progress successfully through initial human testing, they may fail in later stages of development. A number of companies in the pharmaceutical industry, including us, have suffered significant setbacks in advanced clinical trials, even after reporting promising results in earlier trials. Regulatory agencies, including the FDA and EMEA, regulate many aspects of a product candidate’s life cycle, including research and development and pre-clinical and clinical testing. We or regulatory authorities may

suspend clinical trials at any time on the basis overthat the estimatedparticipants are being exposed to unacceptable health risks. Completion of clinical trials depends on, among other things, the number of patients available for enrollment in a particular trial, which is a function of many factors, including the availability and proximity of patients with the relevant condition. We rely on third parties to conduct clinical trials, which may result in delays or failure to complete trials if the third parties fail to perform or meet applicable standards. Many of our drug candidates are still in research and preclinical development, which means that they have not yet been tested on humans. We will need to commit significant time and resources to develop these and additional product candidates.

Our products will be successful only if:

We will be dependent on the successful completion of these goals in order to generate revenues. The failure to generate such revenues may preclude us from continuing our research and development of these and other product candidates. We also enter into collaboration agreements for the development and commercialization of our product candidates. We cannot control the amount and timing of resources our collaborators devote to product candidates, which may also result in delays in the development or marketing of products.

Because of these risks and uncertainties, we cannot accurately predict when or whether we will successfully complete the development of our product candidates or the ultimate product development cost. Based on our current development plans, we anticipate launching XYOTAX for non-small-cell lung cancer in 2005. If this launch is successful, we would expect to receive significant cash inflows in 2005 from this compound.

With the exception of TRISENOX, we anticipate that we will not generate revenue from the sale of commercial drugs for a couple of years, if ever. Without revenue generated from commercial sales, we anticipate that funding to support our ongoing research, development and general operations will primarily come from public or private equity financings, collaborations, milestones and licensing opportunities from current or future collaborators.

Selling, general and administrative expenses.    Selling, general and administrative expenses increased to approximately $55.6 million for the year ended December 31, 2003, from approximately $49.8 million for the year ended December 31, 2002. This increase is primarily attributed to approximately $4.6 million of additional sales and marketing authorization application approval.costs mainly related to TRISENOX as well as increased marketing costs associated with product awareness and medical education for XYOTAX, approximately $3.2 million of additional personnel, operating and occupancy costs associated with supporting our research, development and marketing activities and an increase of approximately $0.6 million in stock-based compensation charges. These costs were offset in part by reductions of approximately $1.4 million in our corporate communications program and approximately $1.6 million in maintenance and operating costs of our leased aircraft. We expect selling, general and administrative expenses to increase during 2004 as a result of our merger with Novuspharma in addition to expenses related to our expanded research, development and commercialization efforts. Additionally, due to the variable accounting treatment of certain stock options, fluctuation in quoted prices for our common stock may result in unpredictable and potentially significant charges or credits to our stock-based compensation.

Goodwill and amortization of acquisition related intangibles.    Amortization for the year ended December 31, 2003 decreased to approximately $1.3 million from approximately $6.7 million for the year ended December 31, 2003, due to a marketing intangible asset that became fully amortized in December 2002.

We are required to perform an annual impairment test, which we will perform in the fourth quarter of each year. During the fourth quarter of 2003, we completed our annual impairment test and determined that our goodwill had not been impaired. The net book value of goodwill will be reviewed for impairment annually and whenever there is an indication that the value of the goodwill may be impaired. Any resulting impairment will be recorded in the income statement in the period it is identified and quantified.

Investment income.    Investment income decreased to approximately $1.9 million for the year ended December 31, 2003 from approximately $4.8 million for the year ended December 31, 2002. This decrease is attributed to lower average cash balances and lower prevailing interest rates on our securities available-for-sale during the year ended December 31, 2003 compared with the year ended December 31, 2002, offset by a reduction of investment premium amortization of approximately $1.0 million.

Interest expense.    Interest expense decreased to approximately $9.3 million for the year ended December 31, 2003 from approximately $11.2 million for the year ended December 31, 2002. In December 2002, we completed an exchange offer for our 5.75% convertible subordinated notes, in which approximately $145.4 million of our 5.75% convertible subordinated notes were tendered in exchange for approximately $85.5 million of our new 5.75% convertible senior subordinated notes. The decrease in interest expense is attributable to the reduction of debt resulting from this exchange and was partially offset by an increase in interest expense due to the issuance of $75.0 million principal amount of our 4% convertible senior subordinated notes in June 2003. We expect interest expense to increase in 2004 due to our 4% convertible senior subordinated notes which will be outstanding during the entire year.

Income Taxes

As of December 31, 2003, we had net operating loss carryforwards of approximately $467.7 million, of which $43.9 million relates to stock option deductions, and research and development credit carryforwards of approximately $15.2 million. The carryforwards begin to expire in 2007. Utilization of stock option deductions will not result in a reduction of tax expense.

Due to rounds of equity financings, and other ownership changes as defined in Section 382 of the Internal Revenue Code of 1986, as amended, or the Code, we incurred “ownership changes” pursuant to the Code. Accordingly, our use of the net operating loss carryforwards is limited to approximately $102.4 million annually for losses incurred prior to September 30, 2000 (which aggregate $209.7 million) and may be subject to additional limitations thereafter. Additionally, all losses incurred prior to March 27, 1997 (which aggregate $75.5 million) are subject to an annual limitation of approximately $4.2 million. To the extent that any single-year loss is not utilized to the full amount of the limitation, such unused loss is carried over to subsequent years until the earlier of its utilization or the expiration of the relevant carryforward period, which is generally 15-20 years. Approximately $13.2 million of pre-March 27, 1997 losses are expected to expire unused in the year 2012.

Years ended December 31, 2002 and 2001.

Product sales.    We recorded net product sales of approximately $11.4 million and $6.1 million for TRISENOX for the years ended December 31, 2002 and 2001, respectively. The increase in net sales is primarily due to greater demand for our product in 2002.

License and contract revenue.    For the year ended December 31, 2002, we recognized approximately $5.5 million of license and contract revenue, of which $3.0 million related to a milestone payment and $1.9 million for cost reimbursements for development expenses received from Chugai, and $0.5 million related to the amortization of the initial payments from Chugai and Nippon.

Cost of product sold.    The cost of product sold during the year ended December 31, 2002 and 2001 was approximately $423,000 and $394,000, respectively. Our gross margins have improved mainly due to lower charges for excess inventory in 2002. Cost of product sold consists primarily of manufacturing costs, allowances for excess inventory that may expire and become unsaleable, and royalties paid on product sales. We expect product costs in the future to continue to approximate a small percentage of revenue.

Research and development expenses.    Our research and development expenses for compounds under development and discovery research are as follows (in thousands):

Costs for compounds under development include external direct expenses such as principal investigator fees, clinical research organization charges and contract manufacturing fees incurred for preclinical, clinical, manufacturing and regulatory activities associated with preparing the compounds for submissions of new drug applications to the FDA or similar regulatory filings with agencies outside the U.S. Operating costs include our

personnel and occupancy expenses associated with developing these compounds. Discovery research costs include primarily personnel, occupancy, and laboratory expenses associated with the discovery and identification of new drug targets and lead compounds. We do not allocate operating costs to the individual compounds under development.

Research and development expenses increased to approximately $58.8 million for the year ended December 31, 2002, from approximately $44.7 million for the year ended December 31, 2001. Costs for our XYOTAX program increased primarily due to approximately $11.9 million of clinical and manufacturing costs associated with the set up and initiation of phase III clinical trials as well as several other clinical trials, approximately $3.3 million of preclinical and quality assurance development costs related to our agreement with Chugai and $2.0 million in milestone payments to PG-TXL Company, L.P. for completion of oura phase II clinical trial and the filing of the first IND application in Japan. These increases were offset in part by a $9.2 million decrease in an equity-based development expense related to the vesting of warrants to purchase 350,000 warrantsshares of common stock upon the achievement of a milestone in 2001. TRISENOX costs increased due to approximately $2.5 million of medical affairs, manufacturing, quality assurance and clinical expenses associated with ongoing clinical trials offset by reduced costs of approximately $.9 million of regulatory and preclinical expenses. Costs incurred for other compounds in 2002 included a $1.0 million milestone payment for the commencement of our phase I clinical trial for PG-CPT, offset by lower expenses due to the discontinued clinical development of CT-2584 during 2001. We also incurred additional personnel and operating expenses of approximately $3.7 million related to our expanded development plans for XYOTAX, TRISENOX and PG-CPT. We anticipate increased research and development expenses in connection with our clinical development plans for XYOTAX and our other products.