Table of Contents

UNITED STATES

SECURITIES AND EXCHANGE COMMISSION

WASHINGTON, D.C. 20549

FORM 10-K

(Mark One)

For the fiscal year ended December 31, 2006

OR

For the transition period from                      to                     

Commission file number: 001-32979

THRESHOLD PHARMACEUTICALS, INC.

(Exact name of registrant as specified in its charter)

(650) 474-8200

(Registrant’s telephone number, including area code)

Securities registered pursuant to Section 12(b) if the Act:

Securities registered pursuant to Section 12(g) of the act:    None

Indicate by check mark whether the registrant is a well-known seasoned issuer, as defined in Rule 405 of the Securities Act. Yes  ¨  No  x

Indicate by check mark whether the registrant is not required to file reports pursuant to Section 13 or Section 15(d) of the Act. Yes  ¨  No  x

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

Indicate by check mark if disclosure of delinquent filers pursuant to Item 405 of Regulation S-K is not contained herein, and will not be contained, to the best of registrant’s knowledge, in definitive proxy or information statements incorporated by reference in Part III of this Form 10-K or any amendment to this Form 10-K. Yes  ¨  No  x

Indicate by check mark whether registrant is a large accelerated filer, an accelerated filer or a non accelerated filer. See definition of “accelerated filer and large accelerated filer” in Rule 12b-2 of the Exchange Act (Check One):

Indicate by check mark whether the registrant is a shell company (as defined in Rule 12b-2 of the Exchange Act). Yes  ¨  No  x

The aggregate market value of the voting stock held by non-affiliates of the registrant based upon the closing price of the Common Stock on the Nasdaq National Market on June 30, 2006 was $77,975,552 Shares of Common Stock held by each executive officer and director and by each person or group who owns 5% or more of the outstanding Common Stock at June 30, 2006 have been excluded. Exclusion of such shares should not be construed to indicate that any such person possesses the power, direct or indirect, to direct or cause the direction of the management or policies of the registrant or that such person is controlled by or under common control with the registrant.

On February 28, 2007 there were 37,407,750 shares of the registrant’s common stock outstanding.

Documents incorporated by reference: Portions of the Proxy Statement for Registrant’s Annual Meeting of Stockholders to be held May 16, 2007, or the Proxy Statement, are incorporated herein by reference into Part III.

Table of Contents

Threshold Pharmaceutica ls, Inc.

TABLE OF CONTENTS

2

Table of Contents

PART I

This annual report on Form 10-K, including the sections entitled “Business,” “Risk Factors,” and “Management’s Discussion and Analysis of Financial Condition and Results of Operations,” contains forward-looking statements within the meaning of Section 27A of the Securities Act of 1933, as amended, and Section 21E of the Securities Exchange Act of 1934, as amended. We may, in some cases, use words such as “project,” “believe,” “anticipate,” “plan,” “expect,” “estimate,” “intend,” “should,” “would,” “could,” “potentially,” “will,” or “may,” or other words that convey uncertainty of future events or outcomes to identify these forward-looking statements. Forward-looking statements may include statements about:

There are a number of important factors that could cause actual results to differ materially from the results anticipated by these forward-looking statements. These important factors include those that we discuss in this annual report on Form 10-K under the caption “Risk Factors.” You should read these factors and the other cautionary statements made in this annual report on Form 10-K as being applicable to all related forward-looking statements wherever they appear in this annual report on Form 10-K. If one or more of these factors materialize, or if any underlying assumptions prove incorrect, our actual results, performance or achievements may vary materially from any future results, performance or achievements expressed or implied by these forward-looking statements. We undertake no obligation to publicly update any forward-looking statements, whether as a result of new information, future events or otherwise, except as required by law. Unless the context requires otherwise, in this annual report on Form 10-K the terms “Threshold,” “Threshold Pharmaceuticals,” “we,” “us” and “our” refer to Threshold Pharmaceuticals, Inc. Threshold Pharmaceuticals, Inc., our logo and Metabolic Targeting are our trademarks. Other trademarks, trade names and service marks used in this annual report on Form 10-K are the property of their respective owners.

Overview

We are a biotechnology company focused on the discovery and development of drugs based on Metabolic Targeting, an approach that targets fundamental differences in metabolism between normal and certain diseased

3

Table of Contents

cells. Two of our product candidates are designed to utilize Metabolic Targeting through the potential targeting of the increased uptake of glucose in cancer cells relative to most normal cells. These product candidates, glufosfamide and 2-deoxyglucose (“2DG”), share certain structural characteristics with glucose but act instead as poisons when taken up by a cancer cell. Our other product candidate, TH-302, and the other compounds our scientists are creating and testing in our laboratories, use Metabolic Targeting by targeting the decreased blood supply and oxygenation of most tumor tissues relative to normal tissue. These compounds are relatively non-toxic when oxygen is present, as in healthy tissues, but undergo a chemical conversion in the presence of low levels of oxygen that converts them into toxic compounds that may kill cancer cells. This pipeline of drug candidates is designed to target tumor cells selectively, and we believe that our drugs could be more efficacious and less toxic to healthy tissues than conventional drugs, and thereby provide significant improvement over current therapies.

Our clinical focus is on product candidates for the treatment of cancer. We have three product candidates for which we have exclusive worldwide marketing rights:

We also are working to discover novel drug candidates that will specifically target cancer cells and are actively seeking to in-license other promising compounds or programs.

For the treatment of cancer, we believe that our product candidates, based on Metabolic Targeting, can be applied to the treatment of many solid tumors and will have the potential to increase the effectiveness of existing therapies significantly. Metabolic Targeting provides the opportunity to treat slowly dividing tumor cells that generally evade traditional chemotherapy and radiation therapies and ultimately contribute to relapse. We believe that our focus on Metabolic Targeting, combined with our expertise in medicinal chemistry and drug development, provides us with the capability to identify, discover and develop novel therapies for treating cancer.

Our product candidates are focused on treating patients with significant unmet medical needs. Cancer is the second leading cause of death in the United States after cardiovascular disease. Many cancers, such as pancreatic, lung and liver cancer, currently have few effective treatments and very low survival rates.

Metabolic Targeting for Cancer

Metabolic Targeting is a therapeutic approach that targets fundamental differences in energy metabolism between normal and certain diseased cells. Cells generate energy needed for survival in two ways: the citric acid cycle and glycolysis. The citric acid cycle is a highly efficient process that provides the majority of cellular

4

Table of Contents

energy under normal conditions. Oxygen is essential for energy production through the citric acid cycle. Glycolysis is much less efficient in producing energy than the citric acid cycle. Unlike the citric acid cycle, oxygen is not required for glycolysis, and cells that rely primarily on glycolysis for energy production consume large quantities of glucose. Some diseased cells, including cancer cells rely predominantly or exclusively on glycolysis for their energy needs. When these cells shift energy production to glycolysis, they must increase the levels of the proteins needed to transport and metabolize glucose. Metabolic Targeting takes advantage of these metabolic differences to selectively target certain diseased cells.

Cancer cells require large amounts of glucose for energy production and growth. This increased consumption of glucose has two causes: the process of a normal cell becoming a rapidly dividing cancer cell; and the exposure of a cell to the low oxygen conditions, called hypoxia, within regions of most solid tumors. First, when cells become cancerous, they require more energy and the level of proteins needed for glucose transport and metabolism increases. Second, as a tumor grows, it rapidly outgrows its blood supply, leaving portions of the tumor with regions where the oxygen concentration is significantly lower than in healthy tissues. As a consequence, tumor cells in these hypoxic zones rely on glycolysis for energy production and therefore further increase the levels of proteins responsible for glucose transport and metabolism.

We are focused on developing new cancer therapies by targeting the uptake and metabolism of glucose by cells. In one application of Metabolic Targeting, we use a cancer-killing drug linked to glucose designed to take advantage of the potential for increased glucose intake by cancer cells, thereby delivering the drug more selectively to cancer cells. In another application of Metabolic Targeting, we use compounds that interfere with specific steps of glycolysis. Because cancer cells are believed to have increased dependence on glycolysis to survive, these compounds should substantially reduce energy production, leading to cell death.

We believe that our product candidates may prove effective for treating rapidly dividing cancer cells, because these cells require large amounts of energy and thus metabolize more glucose than do normal cells. Glufosfamide is designed to target the increased glucose intake by these cells by linking a cancer-killing drug to glucose. Our product candidate 2-DG targets glucose metabolism directly and provides the opportunity to increase the effectiveness of current therapies that treat the rapidly dividing cells in the tumor by reducing energy production in those cells. Radiation therapy, as well as the vast majority of chemotherapy drugs, kills cells by damaging DNA or affecting DNA synthesis to prevent cell replication. However, highly energy-dependent DNA repair mechanisms can repair damage to a cell’s DNA. The balance between the extent of DNA damage and the efficiency of cellular DNA repair thus largely determines the effectiveness of therapy. 2-DG, which reduces cellular energy production, is believed to inhibit these repair mechanisms, shifting the balance from repair to damage, and may increase the efficacy of current treatments. Furthermore, cancer cells become resistant to many conventional chemotherapy drugs by a highly energy-dependent process that pumps these drugs out of the cell, reducing their effect. Interference with cellular energy production can disrupt this multidrug resistance, resulting in increased chemotherapy drug accumulation within the cell. We believe our 2-DG product candidate could therefore increase the effectiveness of chemotherapy drugs by interfering with cellular energy production.

In addition to treating rapidly dividing cancer cells, we believe that compounds based on Metabolic Targeting provide the opportunity to kill cancer cells within hypoxic regions, which are poorly treated by current therapies that primarily target the rapidly dividing cells. Cell proliferation in hypoxic regions is greatly inhibited due to insufficient blood supply, leading to insufficient nutrient supply and a lack of oxygen. Slowly dividing cells within the hypoxic region also undergo genetic changes that, as they accumulate in cells, can lead to the development of still more aggressive tumor cells that are resistant to therapy. Following treatment with radiation or chemotherapy, rapidly dividing cells in the vicinity of blood vessels are destroyed, providing room for these more aggressive cells from hypoxic regions to gain access to blood vessels and oxygen. These cells, which may have become resistant to treatment, are then able to grow and proliferate, ultimately contributing to relapse. Thus, current cancer therapies leave the slowly proliferating cells in the hypoxic zones largely untreated while our product candidates are designed to kill these slowly dividing cells by targeting either their increased glucose transport or glucose metabolism. Our hypoxic-activated compounds are designed to specifically target these oxygen-deficient cells within tumors.

5

Table of Contents

Our Product Development Programs

The following table summarizes the status of our current and ongoing product development programs:

Glufosfamide

Market Opportunity

Pancreatic Cancer

The American Cancer Society estimates that 37,170 patients will be diagnosed with pancreatic cancer in the United States in 2007, and approximately 33,370 patients will die from the disease. Only 15-20% of newly diagnosed patients are eligible for surgery, which is typically followed by radiation and chemotherapy. Patients with inoperable pancreatic cancer are treated with radiation and chemotherapy, or in the case of advanced disease, chemotherapy alone as the advantages of radiation are reduced. Gemcitabine is the standard of care for the first-line therapy of advanced metastatic pancreatic cancer. Tarceva was recently approved as a combination therapy with gemcitabine for the first line treatment of pancreatic cancer. Eli Lilly reported worldwide sales of Gemzar (gemcitabine) for all indications to be over $1.3 billion in 2005.

Ovarian Cancer

The American Cancer Society estimates that 22,430 women will be diagnosed with ovarian cancer in the United States in 2007, and approximately 15,280 women will die from the disease. Ovarian cancer is the eighth most common cancer among women, and is the fifth most common cause of cancer death among women. Virtually all newly diagnosed patients undergo surgery, which is typically followed by radiation and chemotherapy. Almost half of all ovarian cancers do not respond at all because of an innate resistance to platinum-based drugs.

Small-cell Lung Cancer

The American Cancer Society estimates that 213,380 people will be diagnosed with lung cancer in the United States in 2007, and approximately 160,390 people will die from the disease. Small cell lung cancer is less common than non-small cell lung cancer. About 15 to 20 percent of all lung cancers are the small cell type. This cancer usually starts in the bronchi near the center of the chest but has often spread outside of the lung by the time of diagnosis. Small cell lung cancer is strongly associated with a history of cigarette smoking.

6

Table of Contents

Soft Tissue Sarcoma

The American Cancer Society estimates that 9,220 people will be diagnosed with soft tissue sarcoma in the United States in 2007, and approximately 3,560 people will die from the disease. Soft tissue sarcomas is a rare and diverse form of cancer originating in various soft tissues such as fat, muscle, nerve, vascular tissue and other connective tissues. Soft tissue sarcomas patients are treated with surgery whenever possible with or without radiation and chemotherapy. Radiation and chemotherapy alone or a combination is also used for advance or recurrent disease or used when surgery is not possible.

Current Therapies for Cancer

Many different approaches are used in treating cancer, including surgery, radiation and drugs or a combination of these approaches. Drugs used to treat cancer include chemotherapeutics, hormones and immune-based therapies. Traditionally, strategies for designing cancer therapies have focused on killing cancer cells that exhibit rapid division and growth, and most conventional cancer drugs have been evaluated and optimized using cellular and animal models that reflect rapid cell growth. However, most solid tumors are actually composed of both rapidly and slowly dividing cells. Conventional cancer treatments are not designed to target the slowly dividing cells found in large portions of solid tumors and therefore rarely succeed in killing all cancerous cells. These slowly dividing cells, which can evade treatment, often contribute to relapse.

Another disadvantage of current cancer therapies that target rapidly dividing cells is their toxic side effects. Because rapidly dividing cells are also found in many healthy tissues, particularly the gastrointestinal tract, bone marrow and hair follicles, nearly all conventional chemotherapy drugs cause severe side effects, such as diarrhea and reduction in blood cell production, which may lead to bleeding, infection and anemia, as well as other side effects, such as hair loss. Likewise, radiation generally cannot be administered without causing significant damage to healthy tissue surrounding a tumor. The toxic and potentially fatal side effects of chemotherapy and radiation therapy are controlled by carefully balancing dose and dosing schedules to minimize toxicity to healthy cells and maximize cancer cell death. Unfortunately, achieving such a balance may permit rapidly dividing cancer cells to survive treatment, resulting in inadequate therapy.

Pancreatic Cancer

With respect to pancreatic cancer, current therapies have limited efficacy. In gemcitabine’s Phase 3 registrational trial, median survival was 5.7 months, and no patient survived beyond two years. In this study, patients treated with 5-flurouracil, or 5-FU, the previous standard of care, had a median survival of 4.2 months, and no patient survived beyond two years during the study. None of the 126 patients treated in both arms of this study achieved a confirmed objective response as measured by tumor shrinkage. Erlotinib was recently approved in first-line pancreatic cancer in combination with gemcitabine with only about two weeks improvement in median survival and 23% improvement in overall survival.

Ovarian Cancer

Treatment of ovarian cancer is rarely curative. The age-adjusted death rate from ovarian cancer has remained unchanged over the past 20 years. Platinum based therapy is the most widely used chemotherapy to treat ovarian cancer, but some women develop resistance to it. When women develop resistance, it is very difficult to treat and to cure them. The current standards of care in treating platinum-resistant ovarian cancer are a variety of single agent and combination regimens including topotecan, anthracyclines such as doxorubicin, gemcitabine, cyclophosphamide, vinorelbine, hexamethylmelamine, ifosfamide and etoposide.

Small-cell Lung Cancer

Although small cell lung cancer (SCLC) is highly responsive to chemotherapy, the responses are short-lived. Response rate for first-line therapy for etoposide and cisplatin or CAV (cyclophosphamide, doxorubicin and vincristine), ranges from 65% to 90% but nearly all patients relapse in less than 12 months. Patients who

7

Table of Contents

relapse more than two to three months after initial therapy are termed sensitive and are considered for second-line therapy. The only approved treatment for second-line sensitive SCLC is topotecan. In a Phase 3 trial in patients with relapse sensitive disease, topotecan induced response rate of 24%, median survival of 25 weeks, and a median time to progression of 13 weeks. Other chemotherapeutic agents used in treating recurrent, sensitive small cell lung cancer include a variety of single agent and combination regimens including topotecan, cyclophosphamide, doxorubicin, vincristine, irinotecan, ifosfamide and cisplatin.

Soft Tissue Sarcoma

The outcome of chemotherapy for advanced soft tissue sarcomas may be influenced by multiple factors, including type of sarcoma, tumor burden, tumor grade, metastatic pattern, performance status and type and intensity of treatment itself. Results of first-line chemotherapy in adult advanced soft tissue sarcoma remain disappointing. The most active chemotherapy agents against soft tissue sarcomas, doxorubicin and ifosfamide, have demonstrated a relatively consistent single-agent activity yielding response rates of 10% to 25%.

Development Plan for Glufosfamide

Our lead product candidate for cancer, glufosfamide, is a small molecule in clinical development for the treatment of pancreatic and a variety of other cancers. Glufosfamide combines the active part of an approved alkylator, a member of a widely used class of chemotherapy drugs, isophosphoramide, with a glucose molecule. Because of its glucose component and a tumor cell’s increased need for glucose, glufosfamide may be preferentially transported into tumors compared to most normal tissues. Most cancers and isolated cancer cell lines over-express the family of glucose transporters due to the increased energy requirement needed to feed uncontrolled proliferation of cancer cells. While the functional role of many of the glucose transporters is not well established, it has been shown that malignant tumors, including soft tissue sarcomas, pancreatic, ovarian, lung, colorectal, breast and bladder cancer tumors express more glucose transporters and are assumed to undergo enhanced glucose metabolism. Inside cells, the linkage between glucose and the alkylator is thought to be cleaved to release the active drug. With glucose as the major side product, glufosfamide has fewer side effects than other drugs in its class, which are known to cause hemorrhagic cystitis, a serious condition characterized by severe bladder bleeding, unless another protective drug is co-administered.

We believe that the potential unique mechanism of action of glufosfamide, its advantage of generating less toxic metabolites, and demonstrated activity in combination with gemcitabine in animal studies make it well-positioned to potentially replace conventional alkylating agents used in combination with gemcitabine. We are developing glufosfamide for pancreatic cancer based on activity seen in previous clinical trials, a known increase in glucose uptake in pancreatic cancer cells and the extreme hypoxia in tumors of this type. We believe it may offer an improvement over conventional therapies for the indications where activity has been observed.

Glufosfamide has also shown activity against other tumor types in animal models. We believe it may offer an improvement over conventional therapies for the indications where activity has been observed or where other drugs in its class are active. Based on human clinical data, the activity of approved alkylators and our understanding of the mechanism of action of glufosfamide, we believe that ovarian, small cell lung and soft tissue carcinoma represent the most promising indications.

Prior Clinical Trials

Glufosfamide has been evaluated in two Phase 1, five Phase 2 clinical trials and one Phase 3 trial that together enrolled over 500 patients with a variety of advanced-stage cancers. In the two Phase 1 trials, escalating doses of glufosfamide were administered to 72 patients with solid tumors not amenable to established treatments. Although Phase 1 trials are designed primarily to assess safety, tumor shrinkage was observed in patients with breast cancer, non-small cell lung cancer, pleural mesothelioma, renal cell carcinoma and cancers of unknown primary origin.

8

Table of Contents

The five Phase 2 studies of glufosfamide were multi-center studies to evaluate tumor response in patients with locally advanced or metastatic pancreatic cancer, relapsed non-small cell lung cancer, a type of brain cancer called glioblastoma, locally advanced or metastatic colon cancer not amenable to surgery and relapsed metastatic breast cancer. Glufosfamide was well tolerated and showed anti-tumor activity against breast, colon, and pancreatic cancers, marginal activity against non-small cell lung cancer and no activity for the treatment of glioblastoma.

In the Phase 2 trial in patients with advanced pancreatic cancer, two of 34 patients achieved a partial response (defined as 30% or greater tumor diameter shrinkage) and 11 of 34 patients achieved stable disease (defined as less than 30% tumor diameter shrinkage and less than 20% growth in tumor diameter). Overall median survival with glufosfamide was estimated at 5.6 months, and two-year survival was estimated at 9%. The preliminary results of this study, published in the European Journal of Cancer in November 2003, reported a median survival of 5.3 months.

In the Phase 1 and Phase 2 trials, glufosfamide was generally well tolerated, with few drug-related serious adverse events. In particular, glufosfamide’s adverse effects on bone marrow and the kidneys were generally reversible without requiring treatment of the side effects. Only 1% of patients developed severe lowering of the blood platelets, which help to stop bleeding. Toxicity to the kidney, as measured by serum elevation in waste products normally excreted by the kidney, was severe in only 1% of patients. Nausea and vomiting is the most common side effect of glufosfamide treatment. There have been no reports of hemorrhagic cystitis in patients treated with glufosfamide.

These Phase 1 and Phase 2 trials of glufosfamide were conducted by ASTA Medica Oncology, which was subsequently acquired by a subsidiary of Baxter International, Inc. We exclusively licensed worldwide rights to the compound and associated clinical data from Baxter.

Ongoing Clinical Trials

Pancreatic Cancer

We have been developing glufosfamide as a single agent for the second-line treatment of metastatic pancreatic cancer, and in combination with gemcitabine for the first-line treatment of inoperable, locally advanced and/or metastatic pancreatic cancer.

In August 2006, we completed enrollment in a pivotal Phase 3 trial of glufosfamide for the treatment of patients with metastatic pancreatic cancer who have failed treatment with gemcitabine. This two-arm trial, initiated in September 2004, has evaluated approximately 300 previously-treated patients with metastatic pancreatic cancer who receive glufosfamide (4500mg/m2) once every 3 weeks or best supportive care, because there is no approved second-line treatment for pancreatic cancer. Best supportive care includes all medical or surgical interventions that a pancreatic cancer patient should receive to palliate the cancer but excludes treatment with systemic therapies intended to kill the cancer cells. The primary endpoint of this trial was overall survival as measured by time from randomization to death. The timing of the final analysis was therefore event-driven and was conducted after the 261st death had occurred. In addition, the trial investigated the potential efficacy of glufosfamide as determined by response rate, duration of response and progression-free survival, pain score, as well as safety. This trial was conducted under a Special Protocol Assessment. The special protocol assessment is a process that allows for official FDA evaluation of the proposed design of a Phase 3 clinical trial. It provides trial sponsors with an agreement on trial design and analysis required to support a new drug application submission, if the study is performed according to the special protocol assessment and meets its primary endpoint and is statistically persuasive. In addition, glufosfamide for the treatment of second-line pancreatic cancer was granted Fast Track designation by the FDA in 2004, which provides for expedited regulatory review for new drugs that demonstrate the potential to address unmet medical needs for the treatment of serious or life-threatening conditions. In September 2006, we received orphan drug designation for glufosfamide from the FDA.

On February 26, 2007, we announced the results of our Phase 3 trial in patients with metastatic pancreatic cancer who had relapsed after gemcitabine chemotherapy. While the overall survival in patients in the

9

Table of Contents

glufosfamide arm was 18% higher compared to those who received best supportive care alone, the result was not statistically significant. The primary efficacy comparison of overall survival was based on 261 deaths and did not reach statistical significance (p=0.19); the hazard ratio of glufosfamide to BSC was 0.85 (95% confidence interval of 0.66 to 1.08). The median survival of patients who were treated with glufosfamide was 105 days versus 84 days for the patients who received BSC.

No new or unexpected safety signals were observed. Adverse events, including renal toxicity and hematologic toxicity, were similar to those observed in previous clinical trials of glufosfamide. The most common drug-related toxicities in the glufosfamide-treated patients were nausea and vomiting. The full analysis of the clinical trial is ongoing. When completed, the results may be discussed with the FDA to seek guidance and agreement of a revised regulatory strategy towards approval of glufosfamide in patients with pancreatic cancer.

In December 2005, we completed the Phase 1 portion of a Phase 1/2 dose-escalation study of glufosfamide in combination with gemcitabine for the treatment of advanced solid tumors and pancreatic cancer. The primary objective of the Phase 1 portion of the trial was to evaluate safety and to determine the maximum tolerated dose of glufosfamide when administered in combination with gemcitabine. The 19 patients in this portion of the trial received the standard dose of gemcitabine (1000mg/m²) weekly for three of every four weeks and one of four doses of glufosfamide (1500 mg/m², 2500 mg/m², 3500 mg/m² or 4500 mg/m²) administered once every 4 weeks. The maximum tolerated dose was established at 4500 mg/m². No unanticipated adverse events based on previous experience with glufosfamide administered as a single agent were observed. Glufosfamide in combination with gemcitabine was shown to be well tolerated, no significant interaction between glufosfamide and gemcitabine was shown in the pharmacokinetics analysis and the dose of 4500 mg/m² of glufosfamide in combination with gemcitabine was reached. This dose is the dose that is being used in both the Phase 2 stage of this trial of glufosfamide in combination with gemcitabine for first-line treatment of pancreatic cancer and in our recently completed Phase 3 trial of glufosfamide for the second-line treatment of pancreatic cancer.

In January 2006, we began the Phase 2 stage of this clinical trial, which is evaluating patients with locally advanced and/or metastatic pancreatic adenocarcinoma previously untreated with chemotherapy. Patients receive the standard dose of gemcitabine plus glufosfamide. In addition to safety, the trial is investigating the efficacy of glufosfamide in combination with gemcitabine as determined by response rate, duration of response, progression-free survival, overall survival, six- and twelve-month survival and change in serum tumor marker levels (CA19-9). Patients in the trial receive the standard dose of gemcitabine (1000mg/m²) weekly for three of every four weeks and 4500mg/m² of glufosfamide administered once every four weeks.

In December 2006, we announced top-line results from the Phase 2 clinical trial of glufosfamide in combination with gemcitabine for the treatment of advanced pancreatic cancer. Glufosfamide was generally well tolerated in combination with gemcitabine with no new unexpected adverse events. In the Phase 2 clinical trial, 29 patients were treated, of which 28 patients with pancreatic adenocarcinoma previously untreated with chemotherapy were evaluated for response. Overall, 5 patients achieved a confirmed partial response and one other patient achieved an unconfirmed partial response for a response rate of 21%. In addition, 10 patients (36%) experienced stable disease. Objective response was assessed radiologically after every two cycles of therapy. A partial response is characterized as a decrease in size by 30% of the sum of the longest diameters of target lesions, the absence of progression of all non-target lesions and no new lesions. Preliminary analysis of the safety data in this Phase 2 glufosfamide and gemcitabine combination trial suggests the incidence of treatment-related nephrotoxicity may be slightly higher than what was observed in previous experience with either of these agents used individually. This clinical trial remains ongoing. Final survival and safety results will be reported at the completion of the trial, estimated to occur during the third quarter of 2007.

Even though our immediate efforts are focused on pancreatic cancer, the results of Phase 1 and Phase 2 clinical trials suggest that glufosfamide may also be useful for the treatment of other cancers. Based on human clinical data, the activity of approved alkylators and our understanding of the mechanism of action of glufosfamide, we believe that breast, small cell lung, ovarian and colon cancers, as well as lymphomas and sarcomas, represent the most promising indications.

10

Table of Contents

Ovarian Cancer

In January 2007, we initiated patient enrollment for a Phase 2 clinical trial evaluating the dosing, safety and activity of glufosfamide in patients with platinum-resistant ovarian cancer. The clinical trial will evaluate two dosing schedules of glufosfamide, a once weekly schedule and the schedule currently utilized in pancreatic cancer trials which is once every three weeks. The trial will explore the administration of slightly higher aggregate doses utilizing the weekly schedule as compared to every three week dosing. All patients may receive up to six 21-day cycles. Overall, 45 women with ovarian cancer who have previously relapsed after up to four prior chemotherapy regimens including one or two prior platinum-containing regimens and who have demonstrated resistance to their last platinum-containing regimen will be enrolled in the Phase 2, open-label, clinical trial at various sites in the U.S.

In addition to safety, the trial is investigating the efficacy of glufosfamide as determined by response rate, duration of response and progression-free survival based on changes in the serum tumor marker level CA-125 and based on tumor assessments and overall survival.

Small-cell Lung Cancer

In February 2007, we initiated patient enrollment for a Phase 2 clinical trial evaluating the efficacy and safety of glufosfamide in patients with recurrent, sensitive small cell lung cancer. Approximately 50 patients with extensive recurrent sensitive small cell lung cancer, who have progressed at least 60 days after completing chemotherapy, are planned to enroll in the Phase 2, open-label, clinical trial at various sites in the United States, Ukraine and Russia. All patients are to receive 5000 mg/m2 of glufosfamide every three weeks for up to six cycles. The primary efficacy endpoint of the trial is objective response rate. The secondary endpoints of the trial will evaluate duration of response, progression-free survival, overall survival, time to response and various safety and pharmacokinetic parameters. The study will also evaluate the effects of glufosfamide on lung cancer symptoms utilizing the Lung Cancer Symptom Scale (LCSS).

The clinical trial will utilize a two stage design to ensure there is an adequate response rate to justify complete enrollment. The first stage will enroll 21 patients and, at the end of this stage, the trial will be stopped if fewer than three patients have a response. If three or more responses are observed, an additional 29 patients will be enrolled. Tumor response will be evaluated at baseline and every six weeks using the Response Evaluation Criteria In Solid Tumors (RECIST).

2DG

2DG, our product candidate for the treatment of solid tumors, is in a Phase 1 trial. 2DG is an orally administered small molecule that employs Metabolic Targeting to treat solid tumors by directly inhibiting glycolysis. Because tumor cells in general, and those in hypoxic zones in particular, are dependent on glycolysis for survival, tumor cells are particularly sensitive to the effect of 2DG. This compound is a synthetic glucose analog that distributes selectively to tumor tissue because of metabolic changes related to increased glucose consumption. Because tumor cells exhibit increased levels of glucose transport proteins, these cells actively transport 2DG into the cells. Once inside the cell, 2DG interferes with cellular mechanisms for generating energy by competing with glucose for key enzymes in glycolysis. The in vivo efficacy of 2DG has been studied in mouse and rat models of certain cancers, including sarcomas, adenocarcinomas, leukemias, melanomas and bladder, colon and breast tumors. In particular, treatment with 2DG, alone and in combination with other chemotherapy resulted in increased lifespan or a reduction in tumor growth in many of these models.

We are conducting a Phase 1 trial of daily 2DG as a single agent and in combination with docetaxel to evaluate the safety, blood levels and maximum tolerated dose in patients with solid tumors. Animal studies suggest that 2DG and docetaxel may work together to kill cancer cells with greater efficacy than either drug alone, without increased risk of side-effects. We are developing 2DG based on its specificity for targeting tumor cells and extensive human safety data, as well as demonstrated animal efficacy that we and our collaborators at the University of Miami published in Cancer Research in January 2004.

11

Table of Contents

2DG has been administered in clinical trials to approximately 600 people principally to evaluate the hormonal and metabolic effects of glucose deprivation. Collectively, these studies have shown that single intravenous doses of 2DG as high as 200 mg/kg do not cause any serious adverse events. Although these support the safe use of 2DG in humans, we have not yet obtained human safety data for the cumulative dose of oral administration of 2DG we intend to use in our clinical trials. The FDA may require such data before allowing us to proceed into pivotal clinical trials that will support approval.

We launched a Phase 1 clinical trial of 2DG in January 2004. This trial is a dose-escalation study to determine the safety, blood levels and maximum tolerated dose of daily oral doses of 2DG given alone or in combination with docetaxel. The study is intended to enroll up to 50 patients with previously treated refractory advanced solid tumors. The study is designed to evaluate the effect of 2DG alone and in combination with docetaxel on tumor growth, and provide a preliminary assessment of efficacy, as assessed by computer tomography. Initial data from this study, reported at American Society of Clinical Oncology, or ASCO, 2005, suggest that 2DG is well tolerated when administered daily for one week every other week, and we intend to evaluate 2DG administered daily, the schedule we believe will ultimately give 2DG the best opportunity to demonstrate efficacy in this setting. We expect to complete enrollment in this study in the first half of 2007, with top line results expected in the third quarter of 2007.

Provided our safety study yields favorable results, we may initiate at least one Phase 2 study that will be a randomized, blinded, multiple-dose study designed to evaluate the safety and efficacy of 2DG given continuously in combination with chemotherapy. We will choose indications and appropriate combination therapies for our Phase 2 program based on the results of the ongoing Phase 1 trial.

TH-302

We also have a preclinical candidate (TH-302) that is a hypoxically activated prodrug for the potential treatment of solid tumors. TH-302, discovered by Threshold, is a novel drug candidate that is activated under the low oxygen conditions typical of cancer cells in the hypoxic regions of tumors. The hypoxic regions of tumors are known to be the most difficult to treat with standard therapies. TH-302 combines a 2-nitroimidazole oxygen sensing trigger with a masked DNA crosslinker. Upon activation in oxygen deficient zones, TH-302 releases an active toxin, poisoning the hypoxic zone of the tumor. TH-302 has demonstrated dramatic antitumor effects in several animal models. If the preclinical results are supportive, Threshold plans to file an Investigational New Drug (“IND”) application for TH-302 with the FDA in the first half of 2007.

Discontinuation of TH-070 Program

In July 2006, we announced we were discontinuing development of TH-070 for benign prostatic hyperplasia (“BPH”) based on safety and efficacy results of the Phase 2 and Phase 3 trials.

The Phase 2 randomized, placebo controlled, double-blind trial, which was initiated in July 2005, enrolled men with moderate to severe BPH. After a two-week placebo run-in period, 216 patients were randomized to receive placebo or one of four doses of TH-070 (5mg, 25mg, 50mg, 150mg) daily for four weeks and to be followed off therapy for three months. The primary objectives of this study were to determine the dose-response relationship of TH-070 with respect to symptomatic improvement as measured by International Prostate Symptom Score (I-PSS) and to evaluate other efficacy endpoints and the safety at the different doses.

The Phase 3 randomized, placebo controlled, double-blind trial, which was initiated in August 2005, also enrolled men with moderate to severe BPH. After a two-week placebo run-in period, 567 patients were randomized to receive placebo or one of two doses of TH-070 (50mg or 150mg) daily for twelve weeks and to be followed off therapy for one additional month. The primary objective was to evaluate the efficacy of TH-070 compared to placebo as measured by I-PSS. Dosing in this trial was prematurely discontinued at the same time that we announced the partial clinical hold in the U.S. TH-070 program due to certain adverse events relating to elevated liver enzymes.

12

Table of Contents

The interim analysis of the Phase 2 data did not demonstrate a clear dose response in IPSS at four weeks of treatment. The mean I-PSS change from baseline as measured following placebo run-in to one month of treatment ranged from -2.1 to -2.5 across the five dose groups, including the placebo control.

The interim analysis of the Phase 3 data did not demonstrate a statistically significant difference in I-PSS between either of the two drug dose groups (50mg and 150mg) and placebo. The mean I-PSS change from baseline as measured following the placebo run-in to four weeks of treatment ranged from -1.9 to -2.9 and to twelve weeks of treatment ranged from -4.4 to -5.4. There was no statistically significant difference in any of the secondary endpoints with the exception of change in prostate specific antigen (“PSA”) which did show statistical significance at certain time points.

The interim safety results from the Phase 2 and Phase 3 trials include seven cases of myalgia and four cases of testicular pain. Across all TH-070 clinical trials, there were 15 patients who had elevations in liver enzymes (as defined by elevations greater than three times the upper limit of normal), two of whom were in the placebo group. Six of the patients with elevated liver enzymes were deemed to have experienced serious adverse events.

As a result of discontinuing the TH-070 program, in August 2006, we adopted a plan to reduce our operating expenses. The plan included eliminating 29 full-time employees, or approximately a 35% reduction in staff affecting all areas of the Company, and to reduce other expenses.

Discovery Research

We have research programs focused on targeting the hypoxic microenvironment of solid tumors. Solid tumors possess chaotic and insufficient blood flow resulting in regions which are starved for oxygen, or hypoxic. These extremely low oxygen conditions are not found in normal tissue and these hypoxic zones are found in virtually all solid tumors. The hypoxic zones of tumors are known to be resistant to standard chemotherapeutics and to radiation therapy. Tumor hypoxia correlates with poor prognosis in cancer patients and represent a significant unmet medical need. The general nature of hypoxia in solid tumors offers the possibility for cancer therapeutics which are useful in many indications and hence a large market opportunity.

Our most advanced efforts targeting these regions are the design and development of novel cytotoxic prodrug compounds. A prodrug is an inactive compound that is converted in the human body by enzymatic processes that result in the formation of an active drug. The prodrug concept is well established in chemotherapy, but it was initially only employed to modify the pharmacokinetic properties of compounds through non-specific activation processes. Only more recently has the concept been put to use in the design of agents that are selectively activated in tumor tissues through specific activation processes.

Our prodrugs have two distinct parts, a toxic portion and an attached trigger molecule. To prevent general toxicity, the trigger molecule masks the toxin until the prodrug is activated by low oxygen concentration in the hypoxic zones of solid tumors. Once activated, the toxin kills cells in its vicinity. We have designed prodrugs that are triggered only at the very low oxygen levels found in these hypoxic regions. Our experiments indicate that we can achieve a greater than 100-fold difference in cytotoxicity between cells in normal oxygen levels and hypoxic cells. We have identified a clinical candidate, TH-302, which is highly selective and produces a conventional DNA cross-linking toxin upon activation. Hypoxically activated prodrugs of other toxin classes are being pursued. Lead compounds have demonstrated promising in vitro data, and additional characterization, evaluation and optimization of these compounds is currently underway.

Our expertise includes broad capabilities in lead synthesis, assay development and in vitro and in vivo compound evaluation. Our medicinal chemistry expertise allows us to turn initially promising compounds generated by our chemists into drug candidates. We believe that our research focus combined with our medicinal chemistry expertise provide us with the capacity to identify, discover and develop novel therapies.

13

Table of Contents

During the years ended December 31, 2006, 2005 and 2004, we spent $46.3 million, $36.0 million and $16.3 million, respectively, on research and development activities.

Our Strategy

Our goal is to create a leading biotechnology company that develops and commercializes drugs based on Metabolic Targeting with an initial focus on cancer. Key elements of our strategy are to:

Manufacturing and Supply

The production of glufosfamide, 2DG and TH-302 employs small molecule organic chemistry procedures that are standard for the pharmaceutical industry. We currently rely on contract manufacturers for the manufacture of active pharmaceutical ingredient, or API, and final drug product of glufosfamide, 2DG, TH-302 and any other products or investigational drugs for development and commercial purposes. We intend to continue to use our financial resources to accelerate the development of our product candidates rather than diverting resources to establish our own manufacturing facilities.

Our initial supplies of glufosfamide were prepared by a subsidiary of Baxter International, Inc. and were used to initiate our current clinical trials. We are currently using glufosfamide API and drug product that were manufactured by other suppliers and we believe this material will be sufficient to complete our currently ongoing Phase 2 trials. We are in the process of identifying and qualifying an additional vendor to manufacture glufosfamide API. If we experience unexpected delays, or if the API or finished product does not meet specifications, we may experience a significant delay in completion of existing trials and the initiation of additional trials.

While we have sufficient supply of 2DG API, our existing supply of clinical trial material may not be sufficient for our ongoing clinical trials through 2007. We are currently in the process of manufacturing additional 2DG drug product, but if we are not successful we may experience a significant delay in our 2DG clinical program.

14

Table of Contents

We are currently using a contract manufacturer to manufacture TH-302 API and are in the process of final drug product formulation for the initiation of our planned Phase 1 trial in 2007. If we are not successful in manufacturing sufficient quantities of TH-302 API and drug product, we may experience a significant delay in our TH-302 clinical program.

We will need to enter into additional agreements for additional supplies of each of our product candidates to complete clinical development and/or commercialize them. These products will need to satisfy all cGMP manufacturing requirements, including passing product specifications. Our inability to satisfy these requirements could delay our clinical programs.

Sales and Marketing

We currently intend to build our own sales force to market our cancer drugs and to maintain commercial rights to our cancer products in the United States and potentially in Europe. Because the United States cancer market is relatively concentrated, we believe we can effectively target it with a small specialized sales force. We may pursue strategic collaborations to commercialize or co-promote our products for cancer in other territories and on a worldwide basis for indications treated by large physician populations. We currently have no marketing, sales or distribution capabilities. In order to commercialize any of our drug candidates, we must develop these capabilities internally or through collaborations with third parties.

License and Development Agreements

Glufosfamide License

In August 2003, we entered into an agreement with Baxter International, Inc., and Baxter Healthcare S.A., or together, Baxter, for the licensing and development of glufosfamide. Under this agreement, we have an exclusive worldwide license and/or sublicense under Baxter’s patent rights, proprietary information, and know-how relating to glufosfamide to develop and commercialize products containing glufosfamide for the treatment of cancer. Baxter’s patent rights include one issued United States patent and 24 foreign counterparts related to glufosfamide, as well as one foreign patent related to its manufacture. Baxter has agreed to provide us with all of its information related to glufosfamide, including animal study data.

In consideration for our licenses under this agreement, we paid an upfront license fee of $100,000 and development milestone payments of $100,000 and $1.3 million. We are obligated to make certain additional development milestone payments, with the next such payment of $1.0 million due in connection with the filing of a new drug application with the FDA for glufosfamide. Future milestone payments in connection with the development of glufosfamide and United States and foreign regulatory submissions and approvals could equal $8.0 million, and sales-based milestone payments could total up to $17.5 million. Following regulatory approval, we will be obligated to pay up to mid-single digit royalties to Baxter based on sales of glufosfamide products.

This agreement remains in effect until terminated by either party. We may terminate the agreement at will upon 60 days prior written notice to Baxter. Baxter may terminate this agreement if we:

15

Table of Contents

Glufosfamide Asian Development Agreement

In November 2004, we entered into a Development Agreement with MediBIC Co. Ltd. MediBIC is a publicly traded Japanese biotechnology company focused on developing therapeutic compounds in partnership with non-Japanese biotechnology firms and providing consulting services in the design, management, and data analysis of clinical trials using pharmacogenomic platforms developed internally and in collaboration with other companies. By working with MediBIC, we believe that we will be able to develop glufosfamide in Asian countries more quickly than by undertaking such efforts on our own or with other third parties. Pursuant to this agreement, we agreed with MediBIC on a development plan for glufosfamide for the treatment of pancreatic cancer in certain Asian countries, including Japan, South Korea, India, China, Taiwan and Hong Kong. We have also received an exclusive, royalty-free license to MediBIC’s know-how for the manufacture, sale, and distribution of glufosfamide products for the treatment of cancer worldwide. In connection with the Development Agreement, we granted to MediBIC a non-exclusive license to use our confidential information relating to glufosfamide for the limited purpose of preparing the development plan and any associated marketing plans as authorized under the Development Agreement, and a non-exclusive license to use our confidential information for the time necessary for MediBIC to perform its obligations under the development plan.

Under this agreement, in December 2004 we received an upfront payment of $4.75 million to support the development of glufosfamide in the Asian countries covered by the agreement and, under a separate but related agreement, an option payment of $250,000. We are responsible for all development activities and MediBIC has no other funding obligations. We have agreed to pay MediBIC a percentage of net sales or net revenues from the sales of glufosfamide products for the treatment of cancer by us or third parties in the Asian countries covered by the agreement. We may also be required to pay MediBIC a percentage of up-front or milestone payments we receive from any third-party sublicensee of ours for the development of a glufosfamide product for the treatment of cancer in those Asian countries.

We may terminate the agreement at any time by making certain payments to MediBIC ranging from $7.0 million to $15.0 million, depending on the stage of development of the glufosfamide product. Otherwise, the agreement will continue until the expiration of the last-to-expire patent in a country in the Asian countries covered by the agreement that is owned or controlled by us and claims glufosfamide, its use for the treatment of cancer or a process to make such compound in such country.

2DG License

In November 2002, we entered into an exclusive license agreement with Dr. Theodore J. Lampidis and Dr. Waldemar Priebe. This agreement gives us exclusive worldwide rights to international patent application US01/07173, to all of its United States counterpart and priority applications, and any United States and foreign patents and patent applications that claim priority from such applications. Two United States patents and one foreign patent licensed under this agreement have been issued. These patents and the related applications cover the treatment of cancer with 2DG or certain other glycolytic inhibitors, alone or in combination with certain other cancer drugs.

In consideration for this license, we have reimbursed Drs. Lampidis and Priebe for patent costs and will bear all future patent costs incurred under this agreement. We are also obligated to make certain milestone payments, including milestone payments of up to $700,000 in connection with the filing and approval of a new drug application, or NDA, for the first product covered by the licensed patents, as well as royalties based on sales of such products. This license terminates upon the last to expire issued patent covering the technology licensed under it. We have the right to terminate the license at will upon written notice to Drs. Lampidis and Priebe.

The United States government funded research conducted by Drs. Lampidis and Priebe and, therefore, the research is subject to certain federal regulations. For example, under the “march-in” provisions of the Bayh-Dole Act, which governs the transfer of technology developed under federal grants and contracts, the government may have the right under limited circumstances to grant licenses to the technology.

16

Table of Contents

Patents and Proprietary Rights

Our policy is to patent the technologies, inventions and improvements that we consider important to the development of our business. As of December 31, 2006, we owned four issued United States and three issued foreign patents, 34 pending United States patent applications (including both provisional and utility patent applications); 13 international, or PCT, patent applications; and 90 pending foreign national patent applications; and held exclusive commercial rights to one issued United States patent and 24 issued foreign counterparts of this patent, and one additional foreign patent relating to our glufosfamide product candidate; and to one issued United States patent and one issued foreign patent and two foreign applications and two United States continuation counterpart applications (one of which was issued in January 2007) of these issued patents and one United States provisional patent application relating to our 2DG product candidate or certain other glycolytic inhibitors. Fifty-nine of the 137 pending US, PCT, and foreign national patent applications and 3 of the 7 issued US and foreign patents owned by us as of December 31, 2006, relate to our now discontinued TH-070 program; many of the pending applications and some of the issued patents may be abandoned to reduce patent expenses.

Intellectual Property Related to Glufosfamide

Our glufosfamide product candidate is covered by one issued United States patent and 24 issued foreign counterpart patents, as well as one issued foreign patent relating to a method for its manufacture. These patents are owned by Baxter, which has exclusively licensed them to us. The major European market counterparts of the United States patent expire in 2009, and the United States patent expires in 2014. Under the Hatch-Waxman Act in the United States, and similar laws in Europe, there are opportunities to extend the term of a patent for up to five years. Although we believe that our glufosfamide product candidate will meet the criteria for patent term extension, there can be no assurance that we will obtain such extension. Based on our current clinical timeline, if such an extension were obtained, then we expect that it would be for approximately three years or less in the United States. We also own one United States patent application and seven counterpart foreign patent applications describing methods for the identification of patients likely to be most responsive to glufosfamide therapy. We also own one United States patent application and seven counterpart foreign patent applications and three international patent applications describing the use of glufosfamide, alone or in combination with other cancer drugs, including gemcitabine, to treat pancreatic cancer, including gemcitabine-resistant pancreatic cancer and certain other types of cancer. In addition, we own two United States provisional patent applications, one for a new unit dose form and the other for a new method of synthesis, relating to our glufosfamide product candidate.

Intellectual Property Related to 2DG

Our 2DG product candidate is protected by two issued United States patents and one foreign claiming methods for treating breast and other specific cancer with 2DG in combination with either paclitaxel or docetaxel or certain other cancer drugs, as well as two United States continuation patent applications and two foreign counterpart patent applications claiming the use of 2DG or certain other glycolytic inhibitors alone or in combination with certain other drugs to treat cancer. The term of any patent that issues on these applications is not expected to lapse until 2020, assuming patent term extension is not available. We have licensed exclusive commercial rights to these patents from the inventors. In addition, we have