Dose escalation study of tezacitabine in combination with cisplatin in patients with advanced cancer




The authors performed a dose escalation study of cisplatin and the novel deoxycytidine analog, tezacitabine, to determine the maximum tolerated dose of the combination.


Twenty-three patients with advanced cancer and good performance status were accrued to 3 dose levels of tezacitabine (150–270 mg/m2) and cisplatin (50 mg/m2). Using a 28-day treatment cycle, both drugs were administered on Days 1 and 15.


Hematologic toxicity was the most frequently observed side effect and was dose limiting. Grade 3 or 4 neutropenia and thrombocytopenia complicated 75% and 31% of all cycles, respectively. Nonhematologic toxicities were mild. Among 18 evaluable patients, 2 with upper gastrointestinal tract tumors achieved partial responses and 4 had stable disease.


Based on dose-limiting neutropenia and thrombocytopenia at the highest dose level, the recommended Phase II doses are 200 mg/m2 of tezacitabine and 50 mg/m2 of cisplatin. Cancer 2003;97:1985–90. © 2003 American Cancer Society.

DOI 10.1002/cncr.11273

Tezacitabine [(E)-2′-deoxy-2′-(fluoromethylene)cytidine; FMdC] is a deoxycytidine analog, which inhibits ribonucleotide reductase and DNA polymerase.1, 2 The intracellular phosphorylation of tezacitabine results in the diphosphate nucleotide analog, which inhibits ribonucleotide reductase, and the triphosphate form, which inhibits DNA synthesis. The major cytotoxic effect of the drug is triggered by its incorporation into replicating DNA,3 resulting in an inability of DNA polymerase to progress beyond the incorporated nucleotide analog or the 3′ to 5′ exonuclease activity of DNA polymerase or p53 to remove it.4 This results in cell cycle arrest and apoptosis by a p53-independent mechanism. Although this mechanism of action is similar to that of gemcitabine, tezacitabine is relatively resistant to deamination by cytidine deaminase.5 To the extent that deamination may decrease delivery of nucleotide analogs such as gemcitabine to intracellular targets, this distinction may represent an advantage for tezacitabine.

Tezacitabine has shown activity against various human tumor cell lines.6, 7 In vivo, tezacitabine has demonstrated dose-dependent inhibition of human colon, prostate, neuroblastoma, glioblastoma, cervical, breast, and nonsmall cell lung (NSCLC) tumor xenografts in nude mice.6, 8–11 Radiosensitization with tezacitabine has been reported in vitro with cervical carcinoma cells and in vivo with colon carcinoma xenografts.12, 13 Preclinical experiments suggest that induction of DNA damage with cisplatin rendered noncycling tumor cells sensitive to tezacitabine in vitro by the incorporation of the phosphorylated drug into the repaired strand.11 However, in a mouse xenograft model of NSCLC, administration of tezacitabine 2–8 hours before cisplatin resulted in synergy (unpublished data). An additive effect was seen with simultaneous administration or a 24-hour interval between tezacitabine and cisplatin infusion.

Tezacitabine has been studied as a single agent in four dose escalation studies, each using a distinct treatment schedule (Table 1).14 Among 21 patients enrolled in a trial using an every 2-week schedule, a partial response (PR) and prolonged stable disease (SD) were observed in patients with gastrointestinal malignancies. A side effect was dose-limiting hematologic toxicity. The recommended Phase II dose, using a Day 1 and Day 15 schedule, was 270 mg/m2. We performed a dose escalation study of cisplatin in combination with tezacitabine in patients with advanced cancer. Given the shared hematologic toxicity of the two agents, the starting dose of tezacitabine was escalated from 150 mg/m2 to 270 mg/m2. The primary endpoint of the study was to determine a dose of each agent, in combination, suitable for further study. We also documented responses in patients with evaluable disease.

Table 1. Previous Studies of Tezacitabine
RouteNo. of patientsScheduleDose range (mg/m2)
  1. i.v.: intravenous; p.o.: oral.

i.v.27Every 3 weeks16–630
i.v.10Twice per week × 3 weeks, every 3 of 5 weeks16
i.v.12Once per week × 3 weeks, every 3 of 4 weeks16–112
i.v.21Every 2 weeks32–270
p.o.10Daily for 5 days every 3–4 weeks2–8


Twenty-three patients were enrolled on the study from February 2000 to September 2001 at the University of Pennsylvania Cancer Center (Table 2). Eligible patients had histologically confirmed advanced solid tumors and either measurable or evaluable disease. Adequate baseline bone marrow and organ function were defined as follows: leukocyte count (≥ 3000 per mm3), absolute neutrophil count (ANC; ≥ 1500 per mm3), hemoglobin level (≥ 9 g/dL), hematocrit value (> 30%), platelet count (≥ 100,000 per mm3), serum bilirubin (< 2.0 times the upper limit of normal [ULN]), aspartate aminotransferase (AST; < 5.0 times the ULN), serum creatinine (≤ 2.0 mg/dL), or calculated creatinine clearance (≥ 50 mL per minute).15 Patients were at least 18 years of age with an Eastern Cooperative Oncology Group (ECOG) performance status of 0 or 1. Patients were not excluded based on the number of previous therapies. Their expected survival was at least 3 months. Fertile participants were required to use adequate contraception. All patients were given information regarding the purpose and conduct of this study and signed written informed consent was provided by each patient in accordance with federal, state, and institutional guidelines.

Table 2. Baseline Characteristics
  1. ECOG: Eastern Cooperative Oncology Group.

Total no. of patients enrolled23
Age (yrs) 
ECOG performance status 
Primary disease site 
 Soft tissue sarcoma3
Previous treatments 
  No. of previous regimens 
  Chemotherapy and radiation7

To be included in the study, patients had to have recovered from the acute effects of any previous therapy or have stable symptoms of Grade 2 or less (using Common Toxicity Criteria) for 28 days. Patients were excluded if they had symptomatic brain metastases, if they had received previous treatment with tezacitabine, or if they had been administered chemotherapy, radiotherapy, surgery, or other investigational drug within 28 days of the first treatment or during the study. Patients with symptomatic metastases, apart from the measurable tumors being followed for response, were permitted to receive limited-field palliative radiotherapy. Patients were excluded if they had an active infection or significant medical problem that might limit their ability to receive treatment. Pregnant, lactating, and breast-feeding women were also excluded.

Patients received tezacitabine and cisplatin according to a predetermined dose escalation scheme (Table 3). Both drugs were given on Days 1 and 15 of a 28-day cycle. Tezacitabine was diluted to 10 mg/mL in normal saline and infused over 30 minutes, between 4 and 8 hours before the administration of cisplatin. Cisplatin was infused over 1 hour in 250 mL normal saline. One liter of normal saline was given for 2 hours before cisplatin and 500 mL was given after infusion. Immediately before cisplatin, patients were given 12.5 mg of mannitol.

Table 3. Dose Escalation Scheme and Chronologic Order of Patient Accrual
Dose levelNo. of patientsCisplatin (mg/m2)Tezacitabine (mg/m2)

A dose-limiting toxicity (DLT) was defined as any one of the following: 1) an ANC less than 500 per mm3 for 5 or more days, 2) an ANC less than 500 per mm3 with fever requiring antibiotics, 3) a platelet count less than 25,000 per mm3, 4) Grade 3 or 4 thrombocytopenia with a bleeding episode requiring platelet transfusion, or 5) nonhematologic toxicity of at least Grade 3 using the CTC. The maximum tolerated dose (MTD) was defined as one dose level below the level that induced DLT in greater than 33% of patients. The last patient enrolled to a given dose level was observed for at least 2 weeks before enrolling patients to a higher dose level.

Each dose level was intended to accrue three patients. In the event of DLT in one of the three patients, provision was made to accrue an additional three patients to that dose level. If two or more patients experienced a DLT, then the MTD was determined. An additional patient was added to dose level 3, as this patient required prompt treatment and the next dose level had not yet opened for accrual.

Before the initiation of therapy, each patient was evaluated with a history, physical examination, tumor measurement using appropriate radiographic technique, assessment of ECOG performance status, and complete blood count (CBC) with differential and serum chemistries. Each of these, with the exception of tumor measurement, was performed before the administration of each subsequent cycle. To monitor hematologic toxicity, a CBC was performed twice weekly on all patients for the first two cycles. In the event of Grade 4 toxicity, a CBC was performed every 2 days until the toxicity resolved to Grade 3 or less. For patients who demonstrate minimal or no hematologic toxicity during the first two cycles, a CBC was performed weekly during subsequent cycles.

Patients were evaluated for response after every other treatment cycle. A complete response was defined as complete disappearance of all measurable and evaluable disease for at least 4 weeks without the appearance of new lesions. A PR was defined as a decrease of 50% or more in the sum of the products of the perpendicular dimensions of all measurable lesions without the appearance of new lesions. An SD was defined as no change or a less than 25% increase or decrease in the size of indicator lesions. Progressive disease (PD) was defined as an increase of 25% or more in the ratio of the sum of the products of the perpendicular dimensions of all measurable lesions to the smallest sum observed. Clear worsening of evaluable disease or the appearance of any new lesions also was considered to be PD.


Patient Characteristics

The baseline characteristics of the 23 patients enrolled in the study are presented in Table 2. The lung and liver were the most common primary sites of cancer. The majority of patients had received previous chemotherapy, with eight patients having received two or more previous regimens. The ECOG performance status was nearly equally divided between 0 and 1. The current study population was determined to be a good-risk patient population.

Dose Escalation

At the first dose level, the second and third patients enrolled received treatment only on Day 1 of their first cycle. Both were removed from the study before Day 15 of the first cycle because of PD. An additional two patients were enrolled at the first dose level to provide adequate toxicity data before dose escalation. There were no DLTs reported at this dose level. Three patients were enrolled to the second dose level (cisplatin dose of 50 mg/m2 and tezacitabine dose of 200 mg/m2) without any DLTs being observed.

At the third dose level, the first patient developed dose-limiting thrombocytopenia (platelet count < 25,000 per mm3) and the second patient had Grade 4 neutropenia for more than 5 consecutive days. The first patient had received one previous chemotherapy regimen whereas the second patient had received four previous treatments. Due to the occurrence of this second DLT at the third dose level, subsequent patients were accrued to the second dose level.

To confirm the MTD, 12 additional patients were enrolled at that dose to ensure adequate toxicity data. Three of the 12 additional patients experienced hematologic DLTs, 2 with neutropenic fever and 1 with Grade 4 thrombocytopenia. One had received two previous chemotherapy regimens, one had one previous investigational chemotherapy, and one had no previous therapy. Three other patients received treatment on Day 1 of the first cycle only. One patient was hospitalized for atrial fibrillation related to his underlying disease. One died of hypoxia and oliguric renal failure related to PD and another died of progressive ovarian carcinoma. Each of these patients was hospitalized for management of these side effetcs. The duration of hospitalization in each case precluded further treatment on the study.


Hematologic toxicity predominated in this pretreated group of patients (Tables 4, 5). At the first dose level, four of five patients experienced transient Grade 3 and 4 neutropenia that was not dose limiting. The two cases of Grade 3 thrombocytopenia at this level were short-lived. At the second dose level, two-thirds of the patients experienced Grade 4 neutropenia, with only one case lasting for 5 or more days. There was one case of dose-limiting Grade 3 thrombocytopenia. At the third dose level, there were two hematologic DLTs among three patients (i.e., one neutropenia and one thrombocytopenia). Grade 4 neutropenia and Grade 4 thrombocytopenia were each a side effect in 23 (45%) and 2 (4%) of 51 total cycles of therapy, respectively.Grade 4 anemia was reported to have occurred once during the first cycle only.

Table 4. Hematologic Toxicity of the First Cycle: Number of Events
Dose levelPatients per dose levelNeutropeniaAnemiaThrombocytopenia
Grade 2Grade 3Grade 4Grade 2Grade 3Grade 4Grade 2Grade 3Grade 4
  • a

    Number of dose-limiting toxicities shown in parentheses.

Cisplatin 50, Tezacitabine 1505031300120
Cisplatin 50, Tezacitabine 200151310 (1)a52024 (1)a0
Cisplatin 50, Tezacitabine 2703012 (1)a101011 (1)a
Table 5. Hematologic Toxicity of All Cycles: Percentage of Events
Dose levelNo. of cyclesNeutropeniaAnemiaThrombocytopenia
Grade 2Grade 3Grade 4Grade 2Grade 3Grade 4Grade 2Grade 3Grade 4
Cisplatin 50, Tezacitabine 15080%50%13%63%13%0%13%50%0%
Cisplatin 50, Tezacitabine 2003462456%2112012260
Cisplatin 50, Tezacitabine 270933333356221101122

A patient with hepatocellular carcinoma (HCC) at the second dose level was admitted on Day 6 of the first cycle due to abdominal pain, hyperbilirubinemia, and neutropenia. A computed tomography scan of the abdomen revealed a cavitary lesion in the liver. Cultures from liver biopsy material and blood revealed an Escherichia coli infection. Two days after admission, he became hypoxic and developed oliguric renal failure. In accordance with the family's wishes, supportive measures were withheld and the patient died. Another patient with ureteral carcinoma and a history of frequent urinary tract infections was briefly hospitalized for a urinary tract infection on Day 2 of the first cycle.

Constitutional symptoms constituted the three most frequently observed nonhematologic toxicities attributed to chemotherapy. Nausea, fatigue, and emesis occurred during 63%, 57%, and 43% of cycles, respectively. Two instances of emesis, one instance of fatigue, and one instance of nausea were Grade 3; none were Grade 4. Grade 1 or 2 neuropathy was a side effect in 8 (16%) of 51 cycles. There was no Grade 3 or 4 neuropathy. Seven cases (14% of cycles) of Grade 1/2 fever were observed. Mild and reversible elevations of AST and alanine aminotransferase levels were reported in only three cycles (6%). Grade 1 edema was observed on five occasions. Grade 1/2 elevations in serum creatinine levels were observed in five patients. However, there was no evidence of cumulative nephrotoxicity with subsequent cycles. All other symptoms were infrequent and mild. Of the laboratory abnormalities seen, mild and reversible hypomagnesemia and increased serum creatinine levels were the most common. A single case of Grade 3 hypomagnesemia and Grade 3 hyperbilirubinemia were observed.


Eighteen patients were evaluable for response. Two PR were observed. One was a 61-year-old woman who had not received previous treatment for adenocarcinoma of the gallbladder. She was treated with two cycles of cisplatin and tezacitabine at the first dose level. She chose to discontinue therapy at that time. A 59-year-old man with adenocarcinoma of the pancreas, accrued to the second dose level, also had achieved a PR. He had been previously treated with gemcitabine, then with constant infusion of 5-fluorouracil (5-FU) with radiotherapy. He received eight cycles of therapy over the course of 12 months, before PD.

Four patients had SD. Two patients with previously untreated NSCLC received six cycles at the third dose level and four cycles at the second dose level each. A patient with HCC who had received one previous chemotherapy regimen received four cycles at the second dose level. A patient with ureteral carcinoma who had received two previous chemotherapy regimens, radiotherapy, and intravesicular bacillus Calmette–Guerin had SD for four cycles of therapy.


We performed a dose escalation study of cisplatin and tezacitabine to determine the MTD of the combination. Using a 28-day treatment cycle, both drugs were administered on Days 1 and 15. Based on hematologic DLTs at the third dose level, we identified cisplatin at a dose of 50 mg/m2 and tezacitabine at a dose of 200 mg/m2 as the MTD and the recommended Phase II dose.

Tezacitabine has been studied as a single agent and in combination with 5-FU. Rodriguez et al.14 summarized the results of four single-agent, dose escalation studies of intravenous tezacitabine in abstract form, each using a distinct treatment schedule (Table 1). Seventy patients were enrolled in the four trials, nearly 50% of whom had colorectal carcinoma. Approximately 67% of the patients had received previous chemotherapy and 50% had treatment failure with at least 3 previous regimens. Fever occurred in 96% of patients, nausea in 40% of patients, asthenia in 39% of patients, and emesis in 27% of patients. Neutropenia was observed in 69% of participants. The frequency of thrombocytopenia was not reported. The CTC grade of these events was not given. One PR (one patient with metastatic colon carcinoma) and one prolonged SD (one patient with metastatic cholangiocarcinoma) were observed in the study using an every 2-week schedule. Pharmacokinetic studies revealed monophasic clearance with 86% of the intact drug recovered in the urine after 24 hours. The recommended Phase II doses for each schedule were not reported. However, using the Day 1 and 15 schedule, 270 mg/m2 has been used in an ongoing, single-agent Phase II study in colorectal carcinoma patients.

Clark et al.16 reported the preliminary results of a Phase I trial of tezacitabine in combination with constant infusion 5-FU in abstract form. Eighteen patients with advanced cancer were accrued to four dose levels with tezacitabine (150–350 mg/m2) and 5-FU (fixed at 200 mg/m2). Tezacitabine was given on Day 1 and 5-FU was given by constant infusion on Days 1–7 of a 15-day treatment cycle. Based on DLTs at the two highest dose levels, additional patients were added at 200 mg/m2 of tezacitabine and 200 mg/m2 of 5-FU to confirm the MTD. The nature of the DLTs was not reported. Two patients with esophageal carcinoma (one with squamous cell carcinoma and one with adenocarcinoma) achieved a PR.

Our results are in agreement with the preliminary results of the single-agent and combination trials of tezacitabine. Severe neutropenia was observed with similar frequency. The MTD of tezacitabine in our combination regimen (200 mg/m2) was one dose level lower than the MTD found in a single-agent study that administered the drug by the same schedule (270 mg/m2). Clark et al.16 also determined the MTD of tezacitabine to be 200 mg/m2 when combined with 7 days of constant infusion 5-FU at 200 mg/m2. That regimen was also repeated every 14 days. Mild peripheral sensory neuropathy, a nonhematologic toxic side effect, occurred with the addition of cisplatin to tezacitabine. Nausea, emesis, and fatigue occurred with greater frequency in our study than in the single-agent tezacitabine trials. Hematologic toxicity is not more severe using this combination at the recommended Phase II dose compared with the experience with single-agent tezacitabine at 270 mg/m2.

The combination of cisplatin and gemcitabine is active in NSCLC and in mesothelioma, as well as breast, ovarian, head and neck, pancreatic, and cervical carcinomas.17–22 Preclinical data suggest that the relative resistance of tezacitabine to cytidine deaminase and its interaction with ribonucleotide reductase may overcome resistance to gemcitabine. Tezacitabine and cisplatin are tolerable when given in combination and show antitumor activity in some diseases that are traditionally resistant to either of these drugs. The toxicity profile is characterized by relatively less thrombocytopenia and transaminase elevations than are characteristic of gemcitabine. We observed low-grade fever at a much lower frequency than that reported in single-agent Phase I trials.

The results of the current study demonstrate that the combination of tezacitabine and cisplatin is tolerable and is associated with predictable hematologic toxicity at the recommended Phase II doses on a Day 1 and 15 schedule. This is an attractive regimen for the treatment of platinum-sensitive tumors such as NSCLC and ovarian carcinoma, for which the addition of tezacitabine may result in greater activity. We observed activity in pancreaticobiliary tumors, which are relatively platinum resistant. Phase II trials in these diseases will further explore the tolerability and clinical activity of this regimen.