Thymidylate Synthase: A Critical Target for Cancer Chemotherapy

Comprehensive

Review Thymidylate Synthase: A Critical Target for Cancer Chemotherapy Michal G. Rose, Michael P. Farrell, John C. Schmitz Abstract Thymidylate synthase (TS) is a key enzyme in the synthesis of 2'-deoxythymidine-5'-monophosphate, an essential precursor for DNA biosynthesis. For this reason, this enzyme is a critical target in cancer chemotherapy. As the first TS inhibitor in clinical use, 5-fluorouracil (5-FU) remains widely used for the treatment of colorectal, pancreatic, breast, head and neck, gastric, and ovarian cancers. The reduced folate, leucovorin, has been shown to enhance the activity of 5-FU in colorectal cancer. However, response rates of the combination remain in the 25%-30% range, and much effort has been focused on designing new, more potent TS inhibitors. Raltitrexed is a folate analogue that is approved as first-line therapy for advanced colorectal cancer in Europe, Australia, Canada, and Japan, although it remains an investigational agent in the United States. Pemetrexed is an antifolate analogue that has shown promising activity in several solid tumor types, including mesothelioma. ZD9331, a highly specific TS inhibitor that does not require polyglutamation for its activation, has shown activity in patients with refractory ovarian and colorectal cancer. Capecitabine is an oral fluoropyrimidine carbamate that was designed to generate 5-FU preferentially in tumor cells; this agent was recently approved by the US Food and Drug Administration as first-line therapy for patients with advanced colorectal cancer. As the number of TS inhibitors available for general clinical use increases, further research is needed to elucidate the critical molecular and biochemical elements that determine the efficacy and tumor specificity of each compound. Clinical Colorectal Cancer, Vol. 1, No. 4, 220-229, 2002 Key words: 5-Fluorouracil, Raltitrexed, Pemetrexed, ZD9331, Capecitabine, Nolatrexed, Radiation therapy

Introduction Thymidylate synthase (TS) catalyzes the reductive methylation of 2'-deoxyuridine-5'-monophosphate to 2'-deoxythymidine-5'-monophosphate (dTMP) using 5,10-methylenetetrahydrofolate as the one-carbon methyl donor.1-3 Subsequently, dTMP is phosphorylated by two successive steps to 2'-deoxythymidine-5'-triphosphate (dTTP), an essential precursor for DNA synthesis (Figure 1). This pathway is the sole intracellular de novo source of dTTP. For this reason, TS represents a critical target for cancer chemotherapy.2

mors incorporated uracil into DNA to a significantly greater extent than corresponding normal tissue. With this in mind, Heidelberger and colleagues hypothesized that a chemically modified uracil molecule might be effective in disrupting tumor DNA biosynthesis. Since its synthesis 45 years ago, 5-FU remains an active agent with well-documented antineoplastic activity against many solid tumors, including colorectal, pancreatic, breast, head and neck, gastric, and ovarian cancers.5,6

Activation of 5-FU

5-Fluorouracil First synthesized by Heidelberger et al in 1957,3,4 5fluorouracil (5-FU) is a member of the fluoropyrimidine class of antineoplastic agents. This compound represents the first class of TS inhibitors to be used in the clinic. The design of this compound was based on the observation that rat hepatoma tuDepartment of Medicine, Yale Cancer Center, Yale University School of Medicine and VA CT Cancer Center, VA CT Healthcare System, New Haven, CT Submitted: Nov. 6, 2001; Revised: Jan. 4, 2002; Accepted: Jan. 14, 2002 Address for correspondence: Michal Rose, MD, VA CT Healthcare System, Cancer Center, 111D, 950 Campbell Ave, West Haven, CT 06516 Fax: 203-937-3803; e-mail: [email protected]

220 • Clinical Colorectal Cancer February 2002

The cytotoxic activity of 5-FU is exerted through several mechanisms of action. There is growing evidence that the specific dose, route, and schedule of 5-FU administration may play a critical role in its final mode(s) of action. By itself, 5-FU is inactive and must be converted intracellularly to various nucleotide forms. For example, 5-fluoro-2'-deoxyuridine-5'-monophosphate (FdUMP) is considered to be one of the critical nucleotide metabolites, as it forms a covalent ternary complex with TS in the presence of the reduced folate 5,10-methylenetetrahydrofolate, resulting in inhibition of the enzyme. FdUMP is subsequently phosphorylated by a series of enzymatic steps to 5-fluorodeoxyuridine-5'-triphosphate (FdUTP), which can then be incorporated into DNA, leading to inhibition of DNA synthesis and function. The resultant in-

hibition of TS gives rise to an accuFigure 1 Thymidylate Synthase as a Chemotherapeutic Target mulation of deoxyuridine triphosphate (dUTP), which can be subseCapecitabine Uridine quently misincorporated into DNA, resulting in the formation of singleRNA synthesis and double-strand DNA breaks. 5'-DFUR These DNA-directed mechanisms FUDP 5-FU FUTP FUMP tend to occur following long-term, Uracil-tegafur RR DPD continuous exposure to 5-FU. In contrast, there is growing evidence that when 5-FU is administered by FDHU FdUDP FdUTP FdUMP bolus injection, an RNA-directed cytotoxic effect predominates. With DNA this schedule, 5-FU is preferentially synthesis TS dTMP dUMP metabolized to 5-fluorouridine-5'triphosphate (FUTP), which is then CH2THF Leucovorin TK incorporated into RNA, causing inRaltitrexed, hibition of critical steps in RNA proThymidine Pemetrexed, Nolatrexed, cessing and mRNA translation.7,8 Of SHMT ZD9331, GS7904L note, a meta-analysis of 6 randomized trials showed that continuous 5FU infusion resulted in improved DHFR DHF THF tumor responses and slight increases in overall survival compared to bolus 5-FU administration. Dose- Abbreviations: 5-FU = 5-fluorouracil; 5'-DFUR = 5'-deoxy-5-fluorouridine; CH2THF = 5,10-methylenetetrahydrofolate; limiting toxicities were different be- DHF = dihydrofolic acid; DHFR = dihydrofolate reductase; DPD = dihydropyrimidine dehydrogenase; dTMP = 2'-deoxythymidine-5'-monophosphate; dUMP = 2'-deoxyuridine-5'-monophosphate; FDHU = 5-fluorodihydrouracil; tween these two schedules of admin- FdUDP = 5-fluoro-2'-deoxyuridine-5'-diphosphate; FdUMP = 5-fluoro-2'-deoxyuridine-5'-monophosphate; FdUTP = istration; myelosuppression was 5-fluorodeoxyuridine-5'-triphosphate; FUDP = 5-fluorouridine-5'-diphosphate; FUMP = 5-fluorouridine-5'-monophosphate; more commonly observed with bolus FUTP = 5-fluorouridine-5'-triphosphate; RR = ribonucleotide reductase; SHMT = serine hydroxymethyltransferase; injection, while diarrhea and hand- THF = tetrahydrofolate; TK = thymidine kinase; TS = thymidylate synthase foot syndrome (palmar-plantar erythrodysesthesia [PPE]) were more often experienced with Polymorphism in the TS Gene continuous infusion.9 Recently, a genetic polymorphism in the TS gene has been identified in which a tandem repeat sequence exists in the 5'TS Expression as a Predictive untranslated region. Double (2R) and triple (3R) tandem Marker in Prognosis and Patient repeats in the human TS gene have been identified in patient Outcome tumor specimens, and there is recent evidence that the 3R There are several lines of evidence supporting the concept tandem repeats result in higher levels of TS expression.20 of TS as an important chemotherapeutic target. Several preKawakami et al analyzed 70 samples from patients with colclinical in vitro and in vivo studies have shown an inverse orectal and other gastrointestinal (GI) malignancies for the relationship between the level of TS enzyme activity in tumor presence of these tandem repeats in the TS gene.21 The levels cells and 5-FU sensitivity.6,10 Similar findings have been of TS enzyme activity, as determined by the radioenzymatic extended to the clinical setting, where a strong association beFdUMP-binding assay, were higher in tumor tissues tween the level of TS expression and response to 5-FU–based expressing the homozygous 3R/3R genotype when compared chemotherapy has been observed in patients with breast11 and to the 2R/3R and 2R/2R genotypes. Although the differences colorectal cancer.12 In addition, there is a strong association in TS enzyme activity were not statistically significant, these between the level of TS enzyme inhibition in tumor samples results suggest a potential link between the expression of TS from patients and clinical response to 5-FU–based chemotheraand the number of tandem repeats in the TS gene. Villafranca py. Studies in breast cancer,13 rectal cancer,14,15 and gastric et al demonstrated that patients with rectal cancer who were cancer16,17 have identified an inverse correlation between homozygous for the triple tandem repeat in the TS gene had TS expression at the mRNA and/or protein level in a reduced probability of tumor downstaging and disease-free patient tumor specimens and in disease-free and overall survival than patients who were 2R/2R or 2R/3R.22 Iacopetta survival. In patients with metastatic colorectal cancer,18 et al showed a greater survival benefit from 5-FU–based non–small-cell lung cancer (NSCLC),19 and gastric cancer,16,17 therapy in colorectal cancer patients expressing the 2R/3R or pretreatment levels of TS protein appear to be highly predictive 2R/2R TS genotypes.23 Similar findings were reported by for response to 5-FU–based chemotherapy. Marsh et al who demonstrated that the median survival for

Clinical Colorectal Cancer February 2002

• 221

Thymidylate Synthase: A Target for Chemotherapy Figure 2

Inhibitors of Thymidylate Synthase O

O CH2

HN H2N

N

COOH O

C NH CH

N CH2

CH2

COOH N

HN

C

CH2

CH

COOH

CONH

NH N

H3C

N

CH2 - C = CH

CH3

F

H

N

N

CB3717 ZD9331 (VamidexTM)

S

O CH2

HN H2N

O

COOH

CH3

C NH CH

N

N

CH2

N

O

Raltitrexed (ZD1694, Tomudex®)

O

N

S

N

COOH

OH

O

CH2

CH2

O OH

HO

O

H2N

HO

OH

AG331 O O

H N

CO2H N H

CO2H

H2N

O N

N H

N

S CH3

HN H2N

Pemetrexed (Multitargeted Antifolate, LY231514, AlimtaTM)

N

AG337 (Nolatrexed, ThymitaqTM) O

COOH HN

N CH CH2 NH O

C

CH2

O

CH2

H2C

(CH2)4

CH3

N

O H3C

O F

N

COOH

HN

C

O

N GS7904L (Liposomal GW1843U89)

colorectal cancer patients receiving 5-FU treatment was reduced from 16 months in those expressing a homozygous 2R/2R genotype to 12 months in those with a homozygous 3R/3R genotype.24 At present, the predictive value of TS genotype and the presence of this genetic polymorphism remain to be more clearly elucidated before it can be used as a clinical predictor of response to 5-FU therapy. Moreover, additional molecular-based studies will need to be performed to provide definitive evidence as to the true biologic signifi-

222 • Clinical Colorectal Cancer February 2002

HO

OH Capecitabine (Xeloda®)

cance of these tandem repeat sequences as it relates to TS expression.20,25,26

5-FU and Leucovorin Further support for the importance of TS enzyme inhibition comes from the clinical use of the reduced folate, leucovorin (LV), in combination with 5-FU. Leucovorin is metabolized intracellularly to the reduced folate 5,10-methylenetetrahydrofolate, which forms a ternary complex with the 5-FU

Michal G. Rose et al metabolite, FdUMP, and the target, TS, and in so doing, helps to maintain the enzyme in a maximally inhibited state. Maintaining the enzyme in an inhibited state is critical because the TS-catalyzed reaction provides the essential nucleotide precursors for DNA biosynthesis. Subsequent work confirmed that 5-FU cytotoxicity was significantly enhanced upon addition of LV.27 For well over 15 years, the combination of 5-FU/LV has been considered standard therapy for patients with advanced colorectal cancer. To date, there have been 25 randomized trials comparing the clinical activity of 5-FU/LV to single-agent 5-FU. All of these studies have demonstrated significantly higher response rates with the combination of 5FU/LV when compared to single-agent 5-FU.28 With the exception of one study, however, this combination regimen has not resulted in improved overall survival. The combination of 5-FU/LV has also been extended to the adjuvant setting, where it has resulted in improved disease-free and overall survival when compared to 5-FU alone.27,29 The central role of TS in DNA biosynthesis and tumor biology has been clearly established. However, the clinical experience with 5-FU in a wide range of malignancies has been somewhat disappointing, with response rates in the range of 10%-15% with single-agent 5-FU, and 25%-30% when administered in combination with LV.29 For this reason, significant efforts have focused on designing novel, more potent inhibitor compounds of TS. These agents fall into two broad categories: folate analogues and nucleotide analogues. Herein, an overview of the new TS inhibitors (Figure 2) currently under evaluation in clinical trials is presented, with emphasis on their mechanism of action, toxicity profile, and range of clinical activity.

Folate Analogues CB3717 CB3717 is a quinazoline-based antifolate analogue that was initially investigated in the 1980s. It is a potent inhibitor of TS in vitro and displays a broad range of preclinical activity against human ovarian, liver, and breast cancer cell lines. Phase I studies documented broad-spectrum activity in patients with breast, liver, and cisplatin-refractory ovarian cancers. Unfortunately, several patients developed life-threatening nephrotoxicity in these initial clinical trials, which was found upon subsequent evaluation to be due to precipitation of CB3717 in the acidic pH of the renal tubules.30 This severe adverse event resulted in an immediate halt to further clinical development of this agent.

Raltitrexed (ZD1694, Tomudex®) Raltitrexed is a water-soluble analogue of CB3717 and a more potent inhibitor of TS. Because of its improved water solubility, this agent does not cause renal toxicity. This compound enters the cell through the reduced folate carrier and undergoes rapid polyglutamation by the enzyme folylpolyglutamate synthase (FPGS). In its monoglutamate form, raltitrexed is a mixed, noncompetitive inhibitor of human TS, with a Ki approaching 90-100 nmol. Polyglutamation to the higher glu-

tamate forms renders it a significantly more potent inhibitor of TS by up to 100-fold. Moreover, once metabolized to the higher polyglutamate form, its retention within cells is significantly prolonged.31 Phase I Raltitrexed Trials. Two phase I studies were conducted with raltitrexed, one in the United Kingdom32 and one by the National Cancer Institute (NCI) in the United States.33 In both studies, raltitrexed was administered as a 15-minute infusion every 3 weeks, a schedule based on preclinical in vivo animal data. The dose-limiting toxicities observed in these two trials included anorexia, fatigue, diarrhea, and myelosuppression. A reversible elevation of serum transaminases, as well as increases in alkaline phosphatase and serum bilirubin, were also observed. Based on the UK study, the recommended phase II dose was 3.0 mg/m2, while a dose of 4.0 mg/m2 was recommended in the United States. While the precise reason why patients in the US study were able to tolerate higher doses of the drug remains unclear, differences in the underlying nutritional status of patients from the two countries, especially as it relates to folate supplementation, may be an important factor. Phase II Raltitrexed Trials. Several phase II trials of raltitrexed have been conducted in patients with a variety of solid tumors. The best response rates (≈26%) were seen in previously untreated patients with advanced colorectal and breast cancers.34,35 Major toxicities observed were grade 3/4 diarrhea, leukopenia, asthenia, and reversible elevation of serum transaminases. Grade 3/4 nausea and vomiting were seen in 12% of patients, but this toxicity was fairly well controlled with antiemetic therapy. A maculopapular rash was noted in 14% of patients. Five patients died of severe drugrelated hematologic suppression and sepsis combined with severe GI toxicity. Phase III Trials of Raltitrexed Versus 5-FU/LV. The promising response rates observed in patients with advanced colorectal cancer led to two phase III randomized trials comparing single-agent raltitrexed to the Mayo Clinic regimen of 5-FU/LV. In this European study, 439 patients with previously untreated, advanced colorectal cancer were randomized to receive either raltitrexed 3 mg/m2 every 3 weeks or 5-FU 425 mg/m2 and LV 20 mg/m2 for 5 days, every 4-5 weeks.36 Response rates were similar in the two arms at 20% in patients receiving raltitrexed and 16.6% in patients receiving 5-FU/LV. Time to progression and median survival were also similar. Of note, the incidence and severity of toxicities were significantly reduced in the raltitrexed group. Based on the results of this study, raltitrexed was approved as first-line therapy for advanced colorectal cancer in several European countries and in Australia, Canada, and Japan. In a North American phase III trial,37 the same two regimens were compared. Objective responses were similar in both arms (14% for raltitrexed versus 15% for 5-FU/LV), but time to progression and overall survival were significantly longer for patients treated with 5-FU/LV (survival, 12.7 months versus 9.7 months). Because of these differences in overall survival between the two arms of the study, raltitrexed

Clinical Colorectal Cancer February 2002

• 223

Thymidylate Synthase: A Target for Chemotherapy was not approved for use by the US Food and Drug Administration (FDA) as first-line therapy in metastatic colorectal cancer. Combination Raltitrexed Studies. Presently, raltitrexed remains an investigational agent in the United States, and much attention is presently focused on developing it in combination with other approved anticancer agents. A phase I study of raltitrexed and the topoisomerase I inhibitor, irinotecan, showed the combination to be well tolerated, and 5 of 20 patients with advanced colorectal cancer in this trial had a partial response (PR).38 Of note, 4 of these 5 patients had previously received 5-FU chemotherapy. A phase II/III trial compared biweekly LV-modulated 5-FU (LV/5-FU) combined either with irinotecan or raltitrexed in advanced colorectal cancer patients, most of whom were previously untreated (34 of 159 patients were previously exposed to adjuvant 5-FU).39 Dose-limiting toxicities for the raltitrexed arm (raltitrexed 3 mg/m2 on day 1 followed on day 2 by LV 250 mg/m2 intravenously [I.V.] plus 5-FU 1050 mg/m2 I.V. bolus) were neutropenia (16%) and diarrhea (16%), with a response rate of 24% (2 complete responses [CRs] and 11 PRs). In contrast, the irinotecan arm (irinotecan 200 mg/m2 on day 1 followed by LV 250 mg/m2 on day 2 plus 5-FU 850 mg/m2 bolus) had a response rate of 34% (3 CRs and 15 PRs) with dose-limiting toxicities of neutropenia (46%) and diarrhea (16%). The response rate seen with raltitrexed compared similarly to the control arm of methotrexate with LV/5-FU (methotrexate 750 mg/m2 on day 1 followed by LV 250 mg/m2 and 5-FU 800 mg/m2 on day 2), which was associated with a response rate of 24%. Encouraging results have been recently observed in phase I/II studies with raltitrexed combined with oxaliplatin in patients with previously untreated metastatic colorectal carcinoma.40 Raltitrexed was administered at 3 mg/m2 followed by a 2-hour infusion of oxaliplatin 130 mg/m2. This combination showed a 47% response rate, with a median overall survival of > 14.5 months. Neutropenia, elevation of serum transaminases, peripheral neuropathy, and diarrhea were the main adverse side effects observed. Further evaluation of this regimen is ongoing. Raltitrexed as a radiosensitizing agent has been studied in phase I trials with radiotherapy in previously untreated but surgically resected patients with stage II/III rectal cancer.41 In these patients, a total of 50.4 Gy at 1.8 Gy per fraction over 5-6 weeks was administered with escalating doses of raltitrexed from 2.0-3.0 mg/m2. The main dose-limiting toxicities included leukopenia and diarrhea at 3.0 mg/m2. Based on this study, the recommended dose for subsequent phase II studies was 2.6 mg/m2 once every 3 weeks.

Pemetrexed (Multitargeted Antifolate, LY231514, AlimtaTM) Pemetrexed is an antifolate analogue in which a pyrrole ring replaces the pyrazine portion folate and a methylene group replaces the benzyl nitrogen in the bridging portion of the molecule. While this compound inhibits TS, it also in-

224 • Clinical Colorectal Cancer February 2002

hibits other folate-dependent enzymes, including dihydrofolate reductase, aminoimidazole carboxamide ribonucleotide formyltransferase, and glycinamide ribonucleotide formyltransferase. Like raltitrexed, pemetrexed utilizes the reduced folate carrier for entry into the cell and requires polyglutamation for maximal inhibitory effects on the various target enzymes. It has shown activity in vitro against colon, renal, liver, and lung cancers.42 Phase I/II Trials of Pemetrexed. Phase I studies have demonstrated that pemetrexed causes neutropenia, anorexia, thrombocytopenia, fatigue, GI toxicity, and a reversible elevation of liver enzymes.43 The initial dose and schedule selected for phase II studies was 600 mg/m2 I.V. every 21 days.44 This dose in two Canadian studies was subsequently reduced to 500 mg/m2 secondary to undue toxicity observed with the higher dose. In phase II studies focusing on NSCLC, pemetrexed as a single agent in previously untreated patients produced a 23% PR rate,45 with a slightly higher PR rate in studies combining pemetrexed with cisplatin (39%).46 A phase II study of pemetrexed in advanced colorectal carcinoma showed a response rate of 15%. This study used a dose of 600 mg/m2, and 22% of the patients required dose reduction secondary to significant hematologic toxicity.47 A second phase II study in colorectal cancer at a reduced dose (500 mg/m2) is being currently conducted by the NCI of Canada. Of interest are clinical trials studying pemetrexed in combination with carboplatin against malignant mesothelioma, such as the ongoing phase I trial by Calvert et al.48 Dose levels of pemetrexed/carboplatin ranged from 400 mg/m2 at an area under the curve (AUC) of 4, to 600 mg/m2 at an AUC of 5. Toxicities included neutropenia and leukopenia although they do not appear to be dose limiting. Of 22 patients enrolled thus far, 14 have had improvement in symptoms, and 10 have shown response by computed tomography scan. Preliminary phase II results have shown encouraging activity in patients with mesothelioma, breast, and gastric cancers.49-51

Nonpolyglutamatable TS Inhibitors The process of intracellular polyglutamation allows the antifolate analogues to be concentrated within cells and enhances their inhibitory effect on TS. However, it is now well established that a decreased ability to polyglutamate represents an important mechanism for the development of cellular drug resistance. This inability to polyglutamate can be caused by low cellular levels of FPGS or by increased activity of γ-glutamyl hydrolase (GGH), which cleaves glutamate residues from the antifolate analogue. With this in mind, significant efforts have focused on designing antifolate analogues that do not require polyglutamation for their cytotoxicity. The following compounds belong to this class of antifolate analogues.

GS7904L (Liposomal GW1843U89) GW1843U89 is an antifolate analogue in which a glutaric acid moiety replaces the glutamate in the molecule. It is taken up by the reduced folate transporter and is a very potent, spe-

Michal G. Rose et al cific inhibitor of TS that does not require polyglutamation for its cytotoxicity. Preclinical in vivo antitumor effects were seen in a broad range of human tumor xenografts including colon, breast, and ovarian cancers and osteosarcoma. In these studies, the administration of oral folic acid reduced or eliminated toxicity without preventing antitumor activity.52 A phase I trial performed with and without coadministration of oral folic acid revealed dose-limiting toxicities including pancytopenia, fever, stomatitis, and skin rash. All of the side effects except for myelosuppression were reduced in severity by coadministration of folic acid. Specific antitumor activity was observed in patients with gastric, bladder, and colon cancers.53 GW1843U89 has recently been reformulated through encapsulation in liposomes (GS7904L), and preclinical toxicology studies have revealed reduced toxicity to intestinal mucosa and bone marrow. Phase I clinical studies with this liposomal preparation have been initiated in Germany, and additional phase I trials are planned in the United States.

ZD9331 (VamidexTM) ZD9331 was developed through rational drug design as a highly specific TS inhibitor, and like its predecessor ZD1694, it is actively transported into cells via the reduced folate carrier.54,55 ZD9331 does not require polyglutamation by FPGS for its activation and, therefore, retains its cytotoxic activity in tumors expressing low levels of FPGS. Preclinical studies have shown ZD9331 to have a broad spectrum of antitumor activity in human cancer cell lines and tumor xenografts. In phase I studies, the dose-limiting toxicities were myelosuppression, nausea, vomiting, skin rash, and diarrhea.56-58 In phase II trials, ZD9331 has shown promising activity as second- and third-line therapy in patients with ovarian and colorectal cancer. These studies were performed after failure with platinum-based therapy in ovarian cancer or after progression on 5-FU/LV/irinotecan chemotherapy in advanced colorectal cancer.59,60 This agent is currently under investigation in other solid tumors, including NSCLC, gastric, and pancreatic cancers as a single agent and in combination with other cytotoxic agents. Phase I studies of ZD9331 in combination with topotecan, a topoisomerase I inhibitor, have been conducted in refractory malignant solid tumors. Side effects have included myelosuppression and grade 3 asthenia, with 1 death occurring at the maximum tolerated dose.61 Other phase II studies are currently underway in combination with gemcitabine, as are various phase I studies combining ZD9331 with cisplatin, carboplatin, or docetaxel.

AG331 and AG337 (Nolatrexed; ThymitaqTM) AG331 and AG337 (nolatrexed) are unique lipophilic, antifolate analogues that were designed on the basis of the crystal structure of human TS. They enter the cell via passive diffusion and are not dependent on the reduced folate carrier or other specific transport systems to cross the cell membrane. Their cellular half-lives are short, and, as a result, they must be administered via continuous infusion. AG331 caused se-

vere liver toxicity in initial clinical testing, and further clinical evaluation was terminated.62,63 In contrast, nolatrexed was much better tolerated in the initial phase I studies and showed promising activity against head and neck, pancreatic, and hepatocellular cancer. In a phase II trial, patients with squamous cell cancer of the head and neck were given nolatrexed as a continuous infusion over 5 days every 3 weeks with doses initiated at 1000 mg/m2. A CR was seen in 2 of 22 patients and a PR in 2 of 22 patients, for an overall response rate of 18%.64 The most common side effects included rash, mucositis, neutropenia, and thrombocytopenia, toxicities similar to those observed in the initial phase I studies.65-67 One drug-related neutropenic sepsis death occurred during the study.64 Phase III trials are currently underway for the treatment of unresectable hepatocellular carcinoma.

Oral Fluoropyrimidines Over the past 10-15 years, significant attempts have focused on enhancing the efficacy, tumor selectivity, and oral bioavailability of 5-FU. In this regard, several 5-FU prodrugs and drug combinations have been developed and are now in various stages of clinical development. The oral 5-FU prodrugs, uracil-tegafur, and S-1 have been reviewed in a previous issue of this journal.68

Capecitabine (Xeloda®) Capecitabine is an oral fluoropyrimidine carbamate that was rationally designed to generate 5-FU preferentially in tumor cells (Figure 1). It is absorbed intact by the GI mucosa and metabolized in the liver by the carboxylesterase enzyme to 5'-deoxy-5-fluorocytidine ribonucleoside and then to 5'deoxy-5-fluorouridine ribonucleoside (5'-DFUR) by cytidine deaminase, an enzyme more abundantly expressed in liver and tumor tissue than in normal tissues. Subsequently, 5'-DFUR is converted directly to 5-FU by thymidine phosphorylase, a thymidine-metabolizing enzyme expressed at higher levels in tumor tissues relative to normal tissues.69 In human tumor xenografts, capecitabine showed promising antitumor activity against colorectal and breast cancers, and the efficacy of the drug correlated well with the ratio of thymidine phosphorylase to dihydropyrimidine dehydrogenase (DPD) in tumor tissue.70 In phase I studies, the main side effects associated with capecitabine were similar to those observed with continuous 5-FU infusion: stomatitis, diarrhea, PPE, and myelosuppression.71 A multicenter phase II study of 162 patients with paclitaxel-refractory metastatic breast cancer demonstrated a response rate of 20%,72 and this study led to the initial FDA approval of capecitabine as third-line therapy for metastatic breast cancer, after anthracyclines and taxanes. Capecitabine as First-Line Treatment for CRC. Capecitabine has significant activity in colorectal cancer. Two phase III randomized studies comparing capecitabine to the Mayo Clinic regimen of 5-FU/LV in the first-line therapy of patients with metastatic colorectal cancer have been completed.73,74 In an integrated analysis of both trials, which included 1207 pa-

Clinical Colorectal Cancer February 2002

• 225

Thymidylate Synthase: A Target for Chemotherapy tients, overall response rates were significantly higher with capecitabine when compared to 5-FU/LV (26% versus 17%, respectively).75 Median survival and time to disease progression were equivalent between the two arms of the study. In terms of toxicity profile, capecitabine was much better tolerated, with a markedly reduced incidence of myelosuppression, stomatitis, nausea, and alopecia. There was a higher incidence of PPE observed with capecitabine therapy, but only 2 patients required short-stay hospitalization. The results of these clinical studies led to the recent FDA approval of capecitabine in May 2001 for the first-line treatment of metastatic colorectal cancer when fluoropyrimidine therapy alone is preferred. Capecitabine Plus Radiation for Locally Advanced Rectal Cancer. Capecitabine is presently being evaluated in combination with other chemotherapeutic agents and radiation therapy. With regard to radiation therapy, initial preclinical in vivo studies of human colon cancer xenograft models revealed improved antitumor activity with the combination of capecitabine and radiation when compared to either agent alone.76 Of note, radiation therapy has now been shown to induce the expression of thymidine phosphorylase, the enzyme that is critical for activation of capecitabine to 5-FU in tumor tissues. These initial preclinical studies led to phase I studies of capecitabine with radiation therapy, which included patients with rectal adenocarcinoma who received radiation after resection of primary or recurrent rectal cancer or for locally recurring or inoperable primary tumor. These studies administered radiation therapy (1.8 Gy/day for 5 days/week to 45 Gy, and a presacral boost of 5.4 Gy) with oral capecitabine (250-1000 mg/m2 b.i.d. on days 1-7/week from the beginning to end of radiation therapy). Major side effects included leukopenia, local skin toxicity, and diarrhea, with the dose-limiting toxicity of PPE encountered at the 1000 mg/m2 b.i.d. dose level (grade 3 in 2 patients). Preliminary data showed significant downstaging of tumors in 5 patients and 1 pathologically confirmed CR.77,78 An Australian study of patients with stage II/III rectal cancer using capecitabine (425-1000 mg/m2 b.i.d. on days 1-5 of each week of radiation therapy) and radiation (45 Gy and 5.4 Gy boost) is ongoing with no major toxicities or dose-limiting toxicities observed to date.78,79 Capecitabine/Irinotecan. The encouraging results of irinotecan with 5-FU80 as first-line therapy of advanced colorectal cancer has led to the evaluation of capecitabine (1000 mg/m2 or 1250 mg/m2 b.i.d., 2 weeks on, 1 week off, with a repeat of cycle every 50 days) combined with weekly irinotecan (70 or 80 mg/m2 as a 30-minute infusion on days 1, 8, 15, 22, 29, and 32).81 This regimen has yielded an overall response rate of 45%, with 11 PRs/CRs in a total of 29 patients studied. The dose-limiting toxicities observed with this combination included diarrhea and neutropenia. Capecitabine/Oxaliplatin. The combination of capecitabine and oxaliplatin in colorectal cancer is also under active investigation. This combination stems from the very promising clinical studies performed in Europe with oxaliplatin and chronomodulated 5-FU/LV. Preclinical studies have con-

226 • Clinical Colorectal Cancer February 2002

firmed a synergistic effect when capecitabine is used in combination with oxaliplatin. This combination is now being evaluated in both the phase I and II settings. In two phase I studies82,83 (oxaliplatin 130 mg/m2 on day 1 plus capecitabine 500-1250 mg/m2 b.i.d., for 2 weeks on, 1 week off) and one phase II study84 (oxaliplatin 130 mg/m2 plus capecitabine 1250 mg/m2 b.i.d., 2 weeks on, 1 week off), diarrhea (grade 3) and neurosensory changes were seen in 8%-14% of patients. Toxicity was easily managed by treatment interruption and dose reduction. The phase II study reported by Borner et al revealed a response rate of 44% in patients treated with first-line therapy and a 22% response rate in previously treated individuals.84 Stable disease was achieved in approximately 40% of all patients. These clinical results are quite promising, and phase III studies are in the planning stage.

Discussion Since the initial synthesis of 5-FU 45 years ago, significant advances have been made in the development of TS inhibitor compounds. However, many questions remain as to which factors are critical in identifying the optimal TS inhibitor compound for clinical application. The first generation of TS inhibitors, including 5-FU, was directed against colorectal and breast cancer as well as other GI malignancies. However, it seems that the newest generation of inhibitors has activity against a different spectrum of tumors, including head and neck cancer, NSCLC, mesothelioma, and hepatoma. How and why has this change occurred? An understanding of the mechanisms mediating this potential shift in tumor specificity of TS inhibitors might provide new insights into tumor cell biology and specific mechanisms of cytotoxicity of the different compounds. Before exerting their cytotoxic effects in the target tumor tissue, each TS inhibitor must enter the systemic circulation either through complete absorption from the GI tract or through direct I.V. injection. The next important step that these compounds must face is drug metabolism and pharmacokinetic elimination either through hepatobiliary and/or renal processes. Thus, issues relating to bioavailability, biodistribution, and drug pharmacokinetics play an important role as initial steps in determining the final antitumor activity of a given TS inhibitor. Once delivered from circulation to the target tumor cell, the drug must enter the cell through either specific carrier-mediated systems or through passive diffusion. It is conceivable, then, that the different spectrum of antitumor activity of the various antifolate TS inhibitors might, in part, be due to inherent differences in their respective abilities to be effectively transported into the malignant cell. Moreover, the levels of normal reduced folates, as well as thymidine in the general circulation, the tumor tissue microenvironment, and within the tumor cell itself, may need to be considered. Finally, the potential selective nature of these compounds might depend upon the differential level of expression of the various transport systems among various normal host tissues, as well as from altered levels of cellular reduced folates and thymidine nucleoside within the normal

Michal G. Rose et al cell environment. Once inside the malignant cell, 5-FU and the various antifolate analogues, such as raltitrexed, pemetrexed, and ZD9331, require metabolic activation to either nucleotide metabolites or higher polyglutamate forms for these compounds to exert their cytotoxic effects. It is reasonable to assume that malignant cells of diverse origins will express different levels of the critical activating enzymes. With regard to 5-FU and related compounds, the balance between activation and degradation by the enzyme DPD is critical for determining eventual biologic effect. As for the antifolate class of compounds, the balance between the level of expression of the activating enzyme FPGS and the breakdown enzyme GGH is critical in mediating the final polyglutamate status. Another important factor to consider is the level of expression of the target enzyme, TS, in various tumor types. There is growing evidence that the level of TS expression, either at the RNA or protein level, might be dramatically different in certain types of human malignancies. This differential level of expression may help to explain why a given antifolate analogue is effective against one tumor type but not another. While TS has been widely assumed to be a cytoplasmic enzyme, several investigators have now shown that TS might be preferentially localized to other compartments within the cell, including the nuclear compartment and the nucleolus.85,86 It is possible that the cellular localization of TS may play an important role in determining whether a specific TS inhibitor is able to exert its inhibitory effect. Finally, malignant cells might differ in the extent to which they rely on the thymidine salvage pathway for their thymidylate nucleotide precursors for DNA biosynthesis.

Conclusion The challenge for future work in this important area of research will be to identify those critical elements that determine the specific antitumor activity and therapeutic selectivity of a given TS inhibitor. For well over 40 years, the long-held view was that TS functions as a critical cellular catalytic enzyme that provides the essential nucleotide precursors for DNA biosynthesis. However, there is growing evidence that TS, in its capacity as an RNA-binding protein, functions as an important regulator of certain key aspects of cellular metabolism, including cell cycle progression, apoptosis, and perhaps even chemosensitivity.87-89 Thus, the design and development of future TS inhibitors must take into account the potential downstream cellular signaling consequences of TS inhibition. The rapid advances in biotechnology with DNA microarray systems, tissue microarrays, high-throughput screening, and advanced bioinformatics will be especially important in this regard. In particular, the availability of rapid, relatively inexpensive, valid, and reliable biotechnology for molecular profiling of a patient’s tumor will allow for the direct and specific application of a given TS inhibitor to individual patients. Research efforts must continue to focus on understanding the potential mechanisms by which cellular resistance develops to these agents. Such studies will be particularly important in

serving as the basis for the rational design of novel therapeutic strategies to prevent and/or overcome the development of drug resistance and to enhance the therapeutic activity of this important class of anticancer compounds.

Acknowledgements The authors wish to thank Dr Edward Chu for his critical reading of this manuscript, and the Yale Cancer Center and the VA CT Cancer Center for their support. This work was supported by grants CA16359 and CA75712 from the National Cancer Institute, Bethesda, MD.

References 01. Brandt DS, Chu E. Future challenges in the clinical development of thymidylate synthase inhibitor compounds. Oncol Res 1997; 9:403410. 02. Chu E, Allegra C. Antifolates. In: Chabner B, Collins J, eds. Cancer Chemotherapy. Philadelphia, PA: Lippincott-Raven; 1996:109-148. 03. Heidelberger C, Chaundhuri N, Danenbery P, et al. Fluorinated pyrimidines: a new class of tumor-inhibitory compounds. Nature 1957; 179:663-666. 04. Pinedo HM, Peters GF. Fluorouracil: biochemistry and pharmacology. J Clin Oncol 1988; 6:1653-1664. 05. Lokich J. Infusional 5-FU: historical evolution, rationale, and clinical experience. Oncology (Huntingt) 1998; 12:19-22. 06. Johnston PG, Lenz HJ, Leichman CG, et al. Thymidylate synthase gene and protein expression correlate and are associated with response to 5-fluorouracil in human colorectal and gastric tumors. Cancer Res 1995; 55:1407-1412. 07. Sobrero AF, Aschele C, Bertino JR. Fluorouracil in colorectal cancer-a tale of two drugs: implications for biochemical modulation. J Clin Oncol 1997; 15:368-381. 08. Aschele C, Sobrero A, Faderan MA, et al. Novel mechanism(s) of resistance to 5-fluorouracil in human colon cancer (HCT-8) sublines following exposure to two different clinically relevant dose schedules. Cancer Res 1992; 52:1855-1864. 09. Efficacy of intravenous continuous infusion of fluorouracil compared with bolus administration in advanced colorectal cancer. Meta-analysis Group in Cancer. J Clin Oncol 1998; 16:301-308. 10. Chu E, Allegra C. Mechanisms of clinical resistance to 5-fluorouracil chemotherapy. In: Hait W, ed. Drug Resistance. Boston, MA: Kluwer Academic; 1995:175-196. 11. Swain SM, Lippman ME, Egan EF, et al. Fluorouracil and high-dose leucovorin in previously treated patients with metastatic breast cancer. J Clin Oncol 1989; 7:890-899. 12. Doroshow JH, Multhauf P, Leong L, et al. Prospective randomized comparison of fluorouracil versus fluorouracil and high-dose continuous infusion leucovorin calcium for the treatment of advanced measurable colorectal cancer in patients previously unexposed to chemotherapy. J Clin Oncol 1990; 8:491-501. 13. Nishimura R, Nagao K, Miyayama H, et al. Thymidylate synthase levels as a therapeutic and prognostic predictor in breast cancer. Anticancer Res 1999; 19:5621-5626. 14. Johnston PG, Fisher ER, Rockette HE, et al. The role of thymidylate synthase expression in prognosis and outcome of adjuvant chemotherapy in patients with rectal cancer. J Clin Oncol 1994; 12: 2640-2647. 15. Edler D, Hallstrom M, Johnston PG, et al. Thymidylate synthase expression: an independent prognostic factor for local recurrence, distant metastasis, disease-free and overall survival in rectal cancer. Clin Cancer Res 2000; 6:1378-1384. 16. Lenz HJ, Leichman CG, Danenberg KD, et al. Thymidylate synthase mRNA level in adenocarcinoma of the stomach: a predictor for primary tumor response and overall survival. J Clin Oncol 1996; 14: 176-182. 17. Ishikawa Y, Kubota T, Otani Y, et al. Thymidylate synthetase and dihydropyrimidine dehydrogenase levels in gastric cancer. Anticancer Res 1999; 19:5635-5640. 18. Salonga D, Danenberg KD, Johnson M, et al. Colorectal tumors responding to 5-fluorouracil have low gene expression levels of dihydropyrimidine dehydrogenase, thymidylate synthase, and thymidine

Clinical Colorectal Cancer February 2002

• 227

Thymidylate Synthase: A Target for Chemotherapy phosphorylase. Clin Cancer Res 2000; 6:1322-1327. 19. Huang CL, Yokomise H, Kobayashi S, et al. Intratumoral expression of thymidylate synthase and dihydropyrimidine dehydrogenase in non-small cell lung cancer patients treated with 5-FU-based chemotherapy. Int J Oncol 2000; 17:47-54. 20. Horie N, Aiba H, Oguro K, et al. Functional analysis and DNA polymorphism of the tandemly repeated sequences in the 5’-terminal regulatory region of the human gene for thymidylate synthase. Cell Struct Funct 1995; 20:191-197. 21. Kawakami K, Omura K, Kanehira E, et al. Polymorphic tandem repeats in the thymidylate synthase gene is associated with its protein expression in human gastrointestinal cancers. Anticancer Res 1999; 19:3249-3252. 22. Villafranca E, Okruzhnov Y, Dominguez MA, et al. Polymorphisms of the repeated sequences in the enhancer region of the thymidylate synthase gene promoter may predict downstaging after preoperative chemoradiation in rectal cancer. J Clin Oncol 2001; 19:1779-1786. 23. Iacopetta B, Grieu F, Joseph D, et al. A polymorphism in the enhancer region of the thymidylate synthase promoter influences the survival of colorectal cancer patients treated with 5-fluorouracil. Br J Cancer 2001; 85:827-830. 24. Marsh S, McKay JA, Cassidy J, et al. Polymorphism in the thymidylate synthase promoter enhancer region in colorectal cancer. Int J Oncol 2001; 19:383-386. 25. Dong S, Lester L, Johnson LF. Transcriptional control elements and complex initiation pattern of the TATA-less bidirectional human thymidylate synthase promoter. J Cell Biochem 2000; 77:50-64. 26. Kaneda S, Takeishi K, Ayusawa D, et al. Role in translation of a triple tandemly repeated sequence in the 5’-untranslated region of human thymidylate synthase mRNA. Nucleic Acids Res 1987; 15:1259-1270. 27. Sotos GA, Grogan L, Allegra CJ. Preclinical and clinical aspects of biomodulation of 5-fluorouracil. Cancer Treat Rev 1994; 20:11-49. 28. Buyse M, Thirion P, Carlson RW, et al. Relation between tumour response to first-line chemotherapy and survival in advanced colorectal cancer: a meta-analysis. Meta-Analysis Group in Cancer. Lancet 2000; 356:373-378. 29. Bobbio-Pallavicini E, Porta C, Moroni M, et al. Folinic acid does improve 5-fluorouracil activity in vivo. Results of a phase III study comparing 5-fluorouracil to 5-fluorouracil and folinic acid in advanced colon cancer patients. J Chemother 1993; 5:52-55. 30. Calvert AH, Alison DL, Harland SJ, et al. A phase I evaluation of the quinazoline antifolate thymidylate synthase inhibitor, N10-propargyl-5,8-dideazafolic acid, CB3717. J Clin Oncol 1986; 4:1245-1252. 31. Jackman AL, Calvert AH. Folate-based thymidylate synthase inhibitors as anticancer drugs. Ann Oncol 1995; 6:871-881. 32. Clarke SJ, Hanwell J, de Boer M, et al. Phase I trial of ZD1694, a new folate-based thymidylate synthase inhibitor, in patients with solid tumors. J Clin Oncol 1996; 14:1495-1503. 33. Sorensen J, Jordan E, Grem J, et al. Phase I trial of ZD1694 (Tomudex), a direct inhibitor of thymidylate synthase. Ann Oncol 1994; (suppl 5):132 (Abstract #241). 34. Zalcberg JR, Cunningham D, Van Cutsem E, et al. ZD1694: A novel thymidylate synthase inhibitor with substantial activity in the treatment of patients with advanced colorectal cancer. Tomudex Colorectal Study Group. J Clin Oncol 1996; 14:716-721. 35. Smith I, Jones A, Spielmann M, et al. A phase II study in advanced breast cancer: ZD1694 (‘Tomudex’) a novel direct and specific thymidylate synthase inhibitor. Br J Cancer 1996; 74:479-481. 36. Cunningham D, Zalcberg JR, Rath U, et al. Final results of a randomised trial comparing ‘Tomudex’ (raltitrexed) with 5-fluorouracil plus leucovorin in advanced colorectal cancer. "Tomudex" Colorectal Cancer Study Group. Ann Oncol 1996; 7:961-965. 37. Pazdur R, Vincent M. Raltitrexed (Tomudex“) versus 5-fluorouracil and leucovorin (5-FU + LV) in patients with advanced colorectal cancer (ACC): results of a randomized, multicenter, North American trial. Proc Am Soc Clin Oncol 1997; 16:228a (Abstract #801). 38. Lewis N, Scher R, Weiner L, et al. Phase I and pharmacokinetic study of irinotecan in combination with raltitrexed. Proc Am Soc Clin Oncol 2000; 19:195a (Abstract #757). 39. Comella P, De Vita F, Mancarella S, et al. Biweekly irinotecan or raltitrexed plus 6S-leucovorin and bolus 5-fluorouracil in advanced colorectal carcinoma: a Southern Italy Cooperative Oncology Group phase II-III randomized trial. Ann Oncol 2000; 11:1323-1333. 40. Scheithauer W, Kornek GV, Ulrich-Pur H, et al. Oxaliplatin plus raltitrexed in patients with advanced colorectal carcinoma: results of

228 • Clinical Colorectal Cancer February 2002

a phase I-II trial. Cancer 2001; 91:1264-1271. 41. Botwood N, James R, Vernon C, et al. Raltitrexed (‘Tomudex’) and radiotherapy can be combined as postoperative treatment for rectal cancer. Ann Oncol 2000; 11:1023-1028. 42. Schultz RM, Chen VJ, Bewley JR, et al. Biological activity of the multitargeted antifolate, MTA (LY231514), in human cell lines with different resistance mechanisms to antifolate drugs. Semin Oncol 1999; 26:68-73. 43. Rinaldi DA. Overview of phase I trials of multitargeted antifolate (MTA, LY231514). Semin Oncol 1999; 26:82-88. 44. O’Dwyer PJ, Nelson K, Thornton DE. Overview of phase II trials of MTA in solid tumors. Semin Oncol 1999; 26:99-104. 45. Rusthoven JJ, Eisenhauer E, Butts C, et al. Multitargeted antifolate LY231514 as first-line chemotherapy for patients with advanced nonsmall-cell lung cancer: a phase II study. National Cancer Institute of Canada Clinical Trials Group. J Clin Oncol 1999; 17:1194. 46. Manegold C, Gatzemeier U, von Pawel J, et al. Front-line treatment of advanced non-small-cell lung cancer with MTA (LY231514, pemetrexed disodium, ALIMTA) and cisplatin: a multicenter phase II trial. Ann Oncol 2000; 11:435-440. 47. John W, Picus J, Blanke CD, et al. Activity of multitargeted antifolate (pemetrexed disodium, LY231514) in patients with advanced colorectal carcinoma: results from a phase II study. Cancer 2000; 88:1807-1813. 48. Calvert A, Hughes A, Calvert P, et al. Alimta in combination with carboplatin demonstrates clinical activity against malignant mesothelioma in a phase I trial. Proc Am Soc Clin Oncol 2000; 19:495a (Abstract #1936). 49. Scagliotti G, Shin D, Kindler H, et al. Phase II study of ALIMTA single agent in patients with malignant pleural mesothelioma. Eur J Cancer 2001; 37:21 (Abstract #66). 50. Theodoulou M, Llombart A, Cruciani G, et al. Pemetrexed disodium (Alimta, LY231514, MTA) in locally advanced or metastatic breast cancer (MBC) patients (pts) with prior anthracycline or anthracenedione and taxane treatment: phase II study. Proc Am Soc Clin Oncol 2000; 19:130a (Abstract #506). 51. Celio L, Bajetta E, Buzzoni R, et al. Efficacy and toxicity of pemetrexed disodium (Alimta) with folic acid (FA) in gastric cancer. Proc Am Soc Clin Oncol 2001; 20:166a (Abstract #660). 52. Smith GK, Amyx H, Boytos CM, et al. Enhanced antitumor activity for the thymidylate synthase inhibitor 1843U89 through decreased host toxicity with oral folic acid. Cancer Res 1995; 55:6117-6125. 53. Burris H, Smetzer L, Eckhardt J, et al. A phase study of the novel thymidylate synthase inhibitor 1842U89 with and without high dose oral folate. Proc Am Soc Clin Oncol 1996; 15:490a (Abstract #1563). 54. Jackman AL, Kimbell R, Aherne GW, et al. Cellular pharmacology and in vivo activity of a new anticancer agent, ZD9331: a water-soluble, nonpolyglutamatable, quinazoline-based inhibitor of thymidylate synthase. Clin Cancer Res 1997; 3:911-921. 55. Marsham PR, Wardleworth JM, Boyle FT, et al. Design and synthesis of potent non-polyglutamatable quinazoline antifolate thymidylate synthase inhibitors. J Med Chem 1999; 42:3809-3820. 56. Plummer R, Rees C, Judson IR, et al. Phase I trial of ZD9331 in adult patients with refractory solid malignancies administered by 30-min infusion on days 1 and 8 with the cycle repeated every 3 weeks. Eur J Cancer 1999; 35:285 (Abstract #1143a). 57. Trigo J, Rees C, Beale P, et al. Phase I trial of ZD9331, a non-polyglutamatable thymidylate synthase inhibitor given as a 5-day continuous infusion every 3 weeks. Eur J Cancer 1999; 35:286 (Abstract #1149a). 58. Goh BC, Ratain MJ, Bertucci D, et al. Phase I study of ZD9331 on short daily intravenous bolus infusion for 5 days every 3 weeks with fixed dosing recommendations. J Clin Oncol 2001; 19:1476-1484. 59. Rader J, Clarke-Pearson D, Moore MJ, et al. Phase II trial of ZD9331 as third-line therapy for patients with ovarian carcinoma. Ann Oncol 2000; 11 (suppl 4):83 (Abstract #368). 60. Schulz J, Douglass E. ZD9331 as second- or third-line therapy in patients with advanced colorectal cancer: A phase II multicentre trial. Ann Oncol 2000; 11 (suppl 4):62 (Abstract #270PD). 61. Benson A, Poplin E, Douglass E. ZD9331 plus topotecan in patients with refractory solid malignancies: A phase I dose-escalation study. Ann Oncol 2000; 11 (suppl 4):139 (Abstract #635P). 62. Clenendin N, Peterkin J, Webber S, et al. AG-331, a “non-classical” lipophilic thymidylate synthase inhibitor for the treatment of solid tumors. Ann Oncol 1994; (suppl 5):133 (Abstract #246). 63. O’Dwyer PJ, Laub PB, DeMaria D, et al. Phase I trial of the thymidy-

Michal G. Rose et al

64. 65.

66. 67.

68. 69. 70. 71.

72. 73.

74.

75.

76.

late synthase inhibitor AG331 as a 5-day continuous infusion. Clin Cancer Res 1996; 2:1685-1692. Belani C, Aqarwala S, Johnson J, et al. A phase II trial of ThymitaqTM (AG337) in patients with squamous cell carcinoma of the head and neck. Proc Am Soc Clin Oncol 1997; 16:387a (Abstract #1381). Calvete J, Balmanno K, Taylor D, et al. Preclinical and clinical studies of prolonged administration of the novel thymidylate synthase inhibitor, AG337. Proc Am Assoc Cancer Res 1994; 35:306 (Abstract #1821). Stuart K, Tessitore J, Rudy J, et al. A phase II trial of nolatrexed dihydrochloride in patients with advanced hepatocellular carcinoma. Cancer 1999; 86:410-414. Rafi I, Taylor G, Balmanno K. A phase I study of the novel antifolate 3,4-dihydro-2-amino-6-methyl-4-oxo-5-(4-pyridylthio)-quinazole dihydrochloride (AG337) given by 24 hour intravenous continuous infusion. Ann Oncol 1994; (suppl 5):131 (Abstract #238) Eng C, Kindler H, Schilsky R. Oral fluoropyrimidine treatment of colorectal cancer. Clin Colorectal Cancer 2001; 1:95-103. Schuller J, Cassidy J, Dumont E, et al. Preferential activation of capecitabine in tumor following oral administration to colorectal cancer patients. Cancer Chemother Pharmacol 2000; 45:291-297. Lewis NL, Meropol NJ. Oral fluoropyrimidines in cancer treatment. Cancer Invest 2000; 18:747-760. Mackean M, Planting A, Twelves C, et al. Phase I and pharmacologic study of intermittent twice-daily oral therapy with capecitabine in patients with advanced and/or metastatic cancer. J Clin Oncol 1998; 16:2977-2985. Blum JL, Jones SE, Buzdar AU, et al. Multicenter phase II study of capecitabine in paclitaxel-refractory metastatic breast cancer. J Clin Oncol 1999; 17:485-493. Hoff PM, Ansari R, Batist G, et al. Comparison of oral capecitabine versus intravenous fluorouracil plus leucovorin as first-line treatment in 605 patients with metastatic colorectal cancer: results of a randomized phase III study. J Clin Oncol 2001; 19:2282-2292. Van Cutsem E, Twelves C, Cassidy J, et al. Oral capecitabine compared with intravenous fluorouracil plus leucovorin in patients with metastatic colorectal cancer: results of a large phase III study. J Clin Oncol 2001; 19:4097-4106. Hoff PM. Capecitabine as first-line treatment for colorectal cancer (CRC): Integrated results of 1207 patients (pts) from 2 randomized, phase III studies. On behalf of the Capecitabine CRC Study Group. Ann Oncol 2000; 11:60 (Abstract #263PD). Sawada N, Ishikawa T, Sekiguchi F, et al. X-ray irradiation induces

77. 78. 79. 80. 81.

82.

83. 84.

85. 86. 87. 88. 89.

thymidine phosphorylase and enhances the efficacy of capecitabine (Xeloda) in human cancer xenografts. Clin Cancer Res 1999; 5: 2948-2953. Dunst J, Reese T, Frings S. Phase I study of capecitabine combined with simultaneous radiotherapy for rectal cancer. Proc Am Soc Clin Oncol 2001; 20:149a (Abstract #592). Ngan S, Zalcberg J, Kell A, et al. A phase I study of capecitabine combined with radiotherapy for locally advanced potentially operable rectal cancer. Proc Am Soc Clin Oncol 2001; 20:149a (Abstract #591). Yoon W, Choi J, Kim J, et al. Enhanced tumoricidal effect of preoperative chemoradiation using capecitabine for locally advanced rectal cancer. Proc Am Soc Clin Oncol 2001; 20:1046a (Abstract #2165). Saltz LB, Cox JV, Blanke C, et al. Irinotecan plus fluorouracil and leucovorin for metastatic colorectal cancer. Irinotecan Study Group. N Engl J Med 2000; 343:905-914. Schleucher N, Tewes M, Achterrath W, et al. Extended phase I study of capecitabine in combination with a weekly schedule of irinotecan as first-line chemotherapy in metastatic colorectal cancer. Proc Am Soc Clin Oncol 2001; 20:141a (Abstract #561). Evans J, Tabernero J, Cassidy J, et al. Safety profile and preliminary efficacy of capecitabine (Xeloda®) in combination with oxaliplatin in patients with advanced or metastatic solid tumours: Results from a phase I study. Ann Oncol 2000; 11:51 (Abstract #222P). DeSantis S, Zeuli M, Di Costanzo F, et al. Oxaliplatin and capecitabine in advanced colorectal cancer: a dose-finding trial. Proc Am Soc Clin Oncol 2001; 20:141a (Abstract #559). Borner M, Mueller S, Roth A, et al. Phase II study of capecitabine (CAP) + oxaliplatin (OXA) in first line and second line treatment of advanced or metastatic colorectal cancer (ACC). Proc Am Soc Clin Oncol 2001; 20:137a (Abstract #546). Poon PP, Storms RK. Thymidylate synthase is localized to the nuclear periphery in the yeast Saccharomyces cerevisiae. J Biol Chem 1994; 269:8341-8347. Samsonoff WA, Reston J, McKee M, et al. Intracellular location of thymidylate synthase and its state of phosphorylation. J Biol Chem 1997; 272:13281-13285. Chu E, Allegra CJ. The role of thymidylate synthase as an RNA binding protein. Bioessays 1996; 18:191-198. Chu E, Copur SM, Ju J, et al. Thymidylate synthase protein and p53 mRNA form an in vivo ribonucleoprotein complex. Mol Cell Biol 1999; 19:1582-1594. Ju J, Pedersen-Lane J, Maley F, et al. Regulation of p53 expression by thymidylate synthase. Proc Natl Acad Sci U S A 1999; 96:3769-3774.

Clinical Colorectal Cancer February 2002

• 229