Phase I and pharmacokinetic study of TZT-1027, a novel synthetic dolastatin 10 derivative, administered as a 1-hour intravenous infusion every 3 weeks in patients with advanced refractory cancer

P. Schöffski1,*, B. Thate1, G. Beutel1, O. Bolte1, D. Otto1, M. Hofmann1, A. Ganser1, A. Jenner2, P. Cheverton2, J. Wanders2, T. Oguma3, R. Atsumi3 and M. Satomi3

1 Department of Hematology and Oncology, Hannover Medical School, Hannover, Germany; 2 Daiichi Pharmaceuticals UK Ltd, London, UK; 3 Daiichi Pharmaceutical Co., Ltd, Tokyo, Japan

Received 27 October 2003; revised 10 November 2003; accepted 23 December 2003


    ABSTRACT
 Top
 ABSTRACT
 Introduction
 Patients and methods
 Results
 Discussion
 REFERENCES
 
Background:

TZT-1027 is a synthetic dolastatin 10 analog with antineoplastic properties in various cell lines and tumor xenografts. The purpose of this phase I study was to evaluate the safety and toxicity, maximum tolerated dose, pharmacokinetics and pharmacodynamics, clinical and metabolic antitumor activity of TZT-1027 when given as a 1-h intravenous infusion every 3 weeks in patients with refractory solid tumors.

Patients and methods:

Patients had a histologically verified refractory tumor with measurable disease, were ≥18 years old, had an Eastern Cooperative Oncology Group performance status <2 and adequate bone marrow, liver, renal and cardiac function. Dose-limiting toxicity was defined as platelets <25 x 109/l, neutrophils <0.5 x 109/l for >5 days, febrile neutropenia ≥38.5°C with grade 4 (National Cancer Institute–common toxicity criteria) neutropenia, or grade 3/4 non-hematological toxicity excluding nausea and vomiting. The last dose was the dose where ≥2 out of six patients experienced dose-limiting toxicity in cycle one. The maximum tolerated dose was one dose level below with less than two of six patients with dose-limiting events.

Results:

Twenty-one non-selected, fully evaluable patients were enrolled. The majority were male (19) and the median age was 55 years (range 39–67). Dose levels of TZT-1027 ranged from 1.35 to 3.0 mg/m2. The median number of cycles was two (range 1–4). Dose-limiting toxicities were observed in three patients at the 3.0 mg/m2 dose level, including neutropenia, fatigue and a short lasting, reversible peripheral neurotoxic syndrome. The most common toxicities per patient were fatigue, anorexia, alopecia, nausea, constipation, leukopenia and neutropenia. Based on RECIST criteria, the best response was stable disease in seven patients. The pharmacokinetic evaluation revealed a T1/2 of ~7 h and linear kinetics.

Conclusions:

The recommended dose of TZT-1027 for the 3-weekly administration is 2.7 mg/m2. Neutropenia, fatigue and a reversible peripheral neurotoxic syndrome are dose-limiting with this schedule. TZT-1027 may be associated with neurological side-effects in patients previously exposed to neurotoxic compounds such as oxaliplatin.

Key words: dolastatins, phase I study, solid tumors, TZT-1027


    Introduction
 Top
 ABSTRACT
 Introduction
 Patients and methods
 Results
 Discussion
 REFERENCES
 
TZT-1027 is a synthetic tetrapeptide structurally related to dolastatin 10 [1]. Dolastatin 10 was isolated in 1987 from an Indian Ocean mollusc, the sea hare (Dolabella auricularia). The mother compound has antitumor activity against various murine tumors as well as human tumor xenografts [25].

TZT-1027 (Figure 1) is a mitotic spindle poison that interacts with tubulin in the same domain as the vinca alkaloid binding region. The drug affects the binding of vinblastine in tumor cell lines, but the mode of interaction of both agents with tubulin is not identical [68]. In various human cancer cell lines, TZT-1027 showed superior cytotoxicity to other antitumor agents [9]. Administered intravenously to mice, activity was seen in solid tumors such as colon 26 adenocarcinoma, B16 melanoma and M5076 sarcomas, as well as p388 leukemia, including cell lines resistant to vincristine, cisplatin and 5-fluorouracil (5-FU). Tumor xenografts in nude mice showed regression of human breast, lung, colon and stomach cancers, when treated with the synthetic agent. The intermittent administration of TZT-1027 increased the life span of tumor-bearing mice in various models.



View larger version (8K):
[in this window]
[in a new window]
 
Figure 1. Chemical structure of TZT-1027. N2-(N,N-dimethyl-L-valyl)-N-[(1S,2R)-2-methoxy-4-[(2S)-2-[(1R,2R)-1-methoxy-2-methyl-3-oxo-3-[(2-phenylethyl)amino]propyl]-1-pyrrolidinyl]-1-[(S)-1-methylpropyl]-4-oxobutyl]-N-methyl-L-valinamide.

 
After a single intravenous administration to mice, rats and dogs, the plasma half-life (T1/2) was between 5 and 8 h and unchanged TZT-1027 was the major component in plasma. In human plasma, TZT-1027 is predominantly bound to {alpha}1-acid glycoprotein ({alpha}1-AGP). The main drug-metabolizing enzyme for TZT-1027 is the cytochrome p450 isoenzyme CYP3A4, but ~90% of the intravenous dose is excreted unchanged in feces in animal models. After repeated intravenous administration in rats and dogs, the major target organs of toxicity were the bone marrow, gastrointestinal tract, lymphoid tissue, heart, liver, spleen and thymus. TZT-1027 was also found to be embryolethal, teratogenic and genotoxic. There was some evidence of cardiotoxicity in rats and monkeys, but none in dogs, and no relevant neurotoxicity.

On the basis of this interesting preclinical profile, the first clinical phase I study was initiated in 1994, and others are currently ongoing. A single administration study in Japan enrolled 23 patients. The major toxicity was hematological in the form of neutropenia, which occurred in four patients, two experiencing grade 3 toxicity. Alopecia, fatigue and anorexia were the principal non-hematological toxicities [10]. A repeated dosing study with a weekly times three schedule replaced the single dose study in Japan in 1997 [11]. Toxicity was similar, with leukopenia and neutropenia being the principal toxicities. Neutropenia was seen in 27 out of 40 patients and 12 of these had grade 3 or 4 toxicity. All grade 4 neutropenias occurred at doses of 1.5 mg/m2 or higher. Non-hematological toxicity was mild and did not exceed grade 2 in most patients. The study resulted in a recommended dose of 1.8 mg/m2 per week. An objective partial response was observed in one patient with thymoma, which lasted for 6 months. A third phase I study was conducted in Hungary in 2000, involving exclusively patients with non-small-cell lung cancer. The study was still ongoing at the time of the start of this study and the drug was given every 3–4 weeks to 49 patients pretreated with chemotherapy or radiotherapy.

Preclinical and clinical data indicated that a suitable dosing schedule for the present study would be once in a 3-week course. According to the toxicity profile in the disease-oriented non-small-cell lung cancer study performed in Hungary, a safe starting dose for this protocol was considered to be 1.35 mg/m2 [12].

The primary objective of the present phase I study was to determine the maximum tolerated dose (MTD) for a 3-week dosing schedule. Furthermore, we wanted to determine the safety profile and dose-limiting toxicity (DLT) of the investigational product by assessing the quantitative and qualitative toxicities of TZT-1027. Secondary objectives were to assess the pharmacokinetics and pharmacodynamics in human subjects and to evaluate the potential antitumor activity of TZT-1027 in a population of patients with refractory cancers, to provide further guidance for the design of future disease-oriented trials.


    Patients and methods
 Top
 ABSTRACT
 Introduction
 Patients and methods
 Results
 Discussion
 REFERENCES
 
Patient selection
At the time of screening for trial participation, patients had to meet the following criteria: histologically and/or cytologically proven solid tumor, measurable or clinically evaluable disease, no standard therapy of proven benefit available or failure to respond to conventional therapy, age ≥18 years, Eastern Cooperative Oncology Group (ECOG) performance status 0–2, left ventricular ejection fraction within normal limits, good bone marrow function with absolute neutrophil counts ≥1.5 x 109/l, platelets ≥100 x 109/l, hemoglobin ≥8.5 g/dl, adequate liver and renal function, defined as AST and ALT <2.5 x the upper limit of normal, total serum bilirubin <25.6 µmol/l and serum creatinine <176.8 µmol/l. Patients were eligible if they had fully recovered from any previous chemotherapy- or radiotherapy-induced toxicity or surgery. All patients signed a written informed consent form. Patients were excluded in cases of a concurrent antitumor therapy within <4 weeks (6 weeks for prior treatment with nitrosoureas, mitomycin C or carboplatin), prior wide-field radiotherapy within the previous 4 weeks, extensive bone marrow infiltration by tumor, symptomatic brain metastases, active or uncontrolled infection, history of cardiac disease or rhythm disorders which had not been stable for 6 months, pregnancy or lactation, inadequate contraception, history of psychiatric disorder or patients having taken an investigational drug within 28 days prior to the administration of TZT-1027.

Study procedures
The clinical trial protocol, investigator’s brochure and patient informed consent form were approved by the local Ethics Committee at Hannover Medical School prior to the screening of the first eligible patient. The study was performed according to the Declaration of Helsinki (Edinburgh, Scotland, October 2000) and the ICH GCP guidelines, and was fully monitored and audited.

Pretreatment and follow-up examination
Baseline evaluations were performed within 14 days prior to receiving TZT-1027 and included medical history, physical examination, routine neurological examination, 12-lead electrocardiogram, height and weight, vital signs, performance status, concomitant medication, assessment of prior anticancer therapy, full blood count, extensive biochemistry including {alpha}1-AGP, urinalysis, and pregnancy test for females of child-bearing potential. Pretreatment evaluations within 28 days prior to receiving TZT-1027 were chest X-ray, tumor assessment by appropriate imaging studies to assess measurable and non-measurable disease (X-rays, magnetic resonance imaging and/or computed tomography scans) and assessment of left ventricular ejection fraction by multi-gated analysis scan. During therapy, blood counts, biochemistry and urinalysis were performed on a weekly basis. In case of grade 4 [National Cancer Institute–common toxicity criteria (NCI–CTC)] neutropenia, blood counts were carried out at least twice weekly. Prior to the start of each 3-weekly course of treatment, physical examination, weight, vital signs, electrocardiogram, performance status and toxicity were assessed. The conventional tumor assessment was done every two cycles. Standardized whole-body 18F-fluoro-deoxy-glucose (FDG)-PET scans were performed at baseline and in cycle one and two in a subset of patients treated on the higher dose levels for detection of subclinical metabolic responses. The end-of-treatment evaluation included physical examination, weight, vital signs, electrocardiogram, determination of the ejection fraction, performance status, hematology, biochemistry, urinalysis, pregnancy test, concomitant medication, adverse event and tumor assessment. All toxicities were followed until complete resolution. The performance status was evaluated according to ECOG standards, toxicity was assessed according to NCI–CTC version 2.0, RECIST criteria were used for response evaluation and the metabolic effects of TZT-1027 were assessed according to EORTC criteria [13, 14].

Drug administration
TZT-1027, a clear, colorless aqueous solution for injection, was stored at room temperature (15–25°C) and protected from light. The required dose of TZT-1027 was diluted in a PVC bag with 250 ml of 0.9% sodium chloride solution. The diluted product was administered without delay over a period of 1 h by intravenous infusion into a peripheral or central vein using a programed peristaltic pump on an out-patient basis. The actual dose was based on body surface area as calculated from the height and body weight prior to each cycle. Antiemetic or other concomitant medications were not routinely applied, but could be used in case of treatment-related toxicity. TZT-1027 was administered every 3 weeks as a 1-h intravenous infusion.

Dose escalation procedure, definition of dose-limiting toxicity and maximum tolerated dose
Based on the previous experience with TZT-1027 in human studies, the starting dose was defined as 1.35 mg/m2. A minimum 1-week safety interval was foreseen between the first and second patient. The third patient was treated when the first patient had completed the first course. This methodology was applied on all dose levels. The dose was escalated sequentially in cohorts of at least three evaluable patients, until the stopping dose was reached. DLT was defined as platelets <25 x 109/l, neutrophils <0.5 x 109/l for >5 days, febrile neutropenia ≥38.5°C with grade 4 (NCI–CTC) neutropenia, or grade 3/4 non-hematological toxicity excluding nausea and vomiting without adequate antiemetic prophylaxis. Dose escalation was carried out when three patients of a certain dose level had completed the first course in the absence of a DLT. The dose escalation schedule consisted of dose levels of 1.35, 1.8, 2.4 and 3.0 mg/m2. If one out of three patients experienced a DLT in cycle one to three, further patients were to be treated with the same dose. If no further patient experienced a DLT, the dose was increased as agreed between investigator and sponsor according to protocol. If two or more patients on a dose level had a DLT in course one, this was the limiting dose. In this case, further patients were to be treated at a previous or intermittent dose to define the recommended dose for the further clinical use of TZT-1027 with this schedule. Dose reductions were not permitted. Any regular concomitant treatment that the patients had been taking within 30 days prior to study entry could be continued during the study and was recorded in the case report forms. Patients did not receive any other antitumor therapy (cytotoxic, biological or radiation) while participating in this study. Prophylactic therapies in the first course of administration were prohibited. Growth factors or transfusions were not administered. If patients showed no subjective or objective sign of progressive disease as defined by RECIST, treatment could be commenced with further chemotherapy cycles.

Plasma and urine pharmacokinetic sampling and assay
Pharmacokinetic evaluations were to be performed in all patients in courses one and two of treatment. Five ml of venous full blood was drawn and placed on ice and water. The plasma was separated immediately after collection. The specimens were centrifuged at 1800 g for 10 min. The plasma layer was transferred into labeled plastic specimen storage vials. Collected urine was stored for further analyses. Both separated plasma and the urine were placed in –20°C freezers and then transferred to a –70°C freezer. Specimens were accumulated and shipped on dry ice to MDS Pharma Services Inc. (Montreal, Canada) for further analysis.

Plasma samples were collected on days 1 and 22 of treatment prior to the infusion, 30 min before the end of infusion, at the end of the drug administration, and at 30 min, 1, 2, 4, 6, 8 and 24 h after the end of the TZT-1027 infusion. The plasma collections were used to determine the following pharmacokinetic parameters by non-compartmental analyses: time of maximum concentration (Tmax), T1/2, maximum concentration (Cmax), area under the curve (AUC), volume of distribution (VSS) and clearance (Cl). The linearity of TZT-1027 plasmakinetics in relation to dose and the relationship between toxicity and Cmax, AUC and Cl were assessed.

Pharmacokinetic and pharmacodynamic analyses
A sensitive and specific analytical method for determining plasma concentrations of TZT-1027 has been reported based on liquid chromatography-mass spectrometry [15]. Daiichi Pharmaceutical Co., Ltd, validated a high-performance liquid chromatographic method using mass spectrometric detection for the determination of TZT-1027 in human plasma and urine over a concentration range of 0.252–503.500 ng/ml. For the analytical procedure, plasma samples and urine samples (0.5 ml) were spiked with the internal standard solution (0.1 ml of 500 ng/ml TZT-1027-d4). After addition of 0.5 ml of saturated sodium hydrogen carbonate solution and 7 ml of diethyl ether anhydrous, the samples were shaken for 15 min and centrifuged at 3000 r.p.m. for 10 min at room temperature. The organic phase was transferred to a tube and evaporated under nitrogen at 40°C. The sample was resolubilized into 0.1 ml of acetonitrile/2 mM ammonium acetate (pH 3.5) (7:3, v/v), transferred to an injection vial and injected onto a high-performance liquid chromatographic TSK Gel Amide-80 column (100 x 4.6 mm, 0.5 µm) with guard column (Amide-80, 15 x 3.2 mm, 0.5 µm). Separation was carried out by acetonitrile/2 mM ammonium acetate (pH 3.5) (7:3, v/v) at a flow-rate of 1 ml/min (Split ratio 1/5, 0.2 ml/min to mass spectrometry). TZT-1027 was quantitated by peak area ratio to its internal standard by mass spectrometry using a selective reaction monitoring mode (MS/MS API 3000) [16, 17]. Correlations between hematological toxicity and pharmacokinetic parameters were explored using linear regression.


    Results
 Top
 ABSTRACT
 Introduction
 Patients and methods
 Results
 Discussion
 REFERENCES
 
General
The protocol was approved by the local Ethics Committee on 15 August 2001. Between 26 September 2001 and 3 September 2002, 21 fully eligible patients were enrolled to the single-center trial.

Patient characteristics
Nineteen males and two females were included; the median age was 55 years with a range of 39–67. The median number of previous treatment regimens administered was four (range 1–12). The patients represented a typical, non-selected phase I population in this institution (Table 1). The most frequent tumor types were colorectal cancer, renal cell cancer and soft tissue sarcoma. All patients were pretreated with surgery and chemotherapy, five patients were pretreated with radiotherapy, and three each with immunotherapy and hormone treatment. The most common previously administered antineoplastic agents were 5-FU (13 patients), oxaliplatin (13 patients), irinotecan (11 patients), etoposide (five patients), cisplatin (four patients), ifosfamide (four patients), capecitabine (three patients), gemcitabine (three patients) and doxorubicin (two patients).


View this table:
[in this window]
[in a new window]
 
Table 1. Patient characteristics
 
Drug administration
Five different dose levels of TZT-1027 were administered, ranging from 1.35 to 3.0 mg/m2 (Table 2). Three to six patients were treated on each dose level. The total number of treatment cycles per dose level ranged from six to 13 cycles. The median number of cycles per patient was two, with a range of one to four.


View this table:
[in this window]
[in a new window]
 
Table 2. Drug administration and dose levels of TZT-1027
 
Dose-limiting toxicity
DLTs were observed exclusively at the 3.0 mg/m2 dose level in three out of five patients during the first treatment cycle. A 66-year-old male with colorectal cancer, pretreated with three lines of chemotherapy including fluoropyrimidines, irinotecan and oxaliplatin, experienced uncomplicated neutropenia grade 4, lasting 9 days. A 48-year-old female with colorectal cancer pretreated with two three-drug combinations based on infusional 5-FU/folinic acid plus irinotecan and oxaliplatin experienced reversible mandibular cramps grade 3, fatigue grade 3, paraesthesias of the pharynx and throat grade 3 and insomnia and agitation grade 3 during the early days of the first course of treatment with TZT-1027. A 67-year-old male with colorectal cancer pretreated with a similar chemotherapy sequence, who was at increased risk for neurotoxicity due to insulin-dependent diabetes mellitus, developed severe pain of the extremities grade 3, paraesthesias grade 3 and fatigue grade 3 in course one. All non-hematological DLTs observed in this trial were short lasting and fully reversible within 3–5 days and none led to hospitalization.

Non-hematological toxicity
The most common non-hematological toxicities per patient in cycle one were fatigue, anorexia, alopecia, nausea and constipation (Table 3). One patient had local phlebitis after each administration of TZT-1027 to peripheral veins. The majority of patients received the drug through central venous lines via port catheters, thus the true incidence of phlebitis cannot be assessed. These patients had port catheters inserted for other infusional chemotherapies prior to participation in the TZT-1027 trial.


View this table:
[in this window]
[in a new window]
 
Table 3. Non-hematological toxicity (all cycles). Most frequent adverse events summarized by patient
 
The most frequent grade 1–4 serum chemistry abnormalities occurring during cycle one were hyponatremia (six patients), increase in aspartate aminotransferase (four patients), hypochloremia (four patients), creatinine increase (three patients) and hyperbilirubinemia (two patients). None of these events required clinical intervention (detailed data not shown).

Hematological toxicity
The most prevalent drug-related hematological toxicity in cycle one was leukopenia (Table 4). It occurred in 13 out of 21 patients. Neutropenia was observed in 11 patients and was grade 3/4 in nine patients. Only one of these events qualified as a DLT, as the duration of neutropenia was generally short. Thrombocytopenia was rare. Anemia was common but almost exclusively related to the underlying malignancy and was already present prior to treatment (Table 4). Transfusions were not required during this trial.


View this table:
[in this window]
[in a new window]
 
Table 4. Hematological toxicity (all cycles). Most frequent toxicity summarized by patient
 
Response assessment
According to RECIST criteria, the best objective response was stable disease in seven patients. The exact duration of stable disease is not known because most of these patients were removed from the study after two treatment cycles.

Fourteen patients had progression during the early course of their treatment. Based on EORTC criteria, none of the evaluable patients achieved a metabolic partial or complete response to TZT-1027, based on serial 18FDG-PET scans performed at baseline and during cycle one and two in a subset of 14 patients (data not shown). Twelve patients had elevated tumor markers at baseline. Among 11 patients with colorectal cancer or cholangiocellular carcinoma with elevated carcinoembryonic antigen and/or CA 19.9, none showed a marker response. One patient with non-seminomatous testicular cancer, who had failed high-dose chemotherapy and autologous stem cell transplant, had a short-lasting, clinically insignificant decrease in serum {alpha}-fetoprotein, which lasted for <3 weeks.

Pharmacokinetics
The baseline pharmacokinetic sampling on day 1 was carried out in all patients. Four patients were not available for the day 22 sampling procedure. Reasons for not performing the repeated sampling were death due to disease progression (two patients), only one cycle administered (one patient), and patients’ request (one).

A summary of pharmacokinetic data for all patients in this trial is shown in Table 5. The Cmax of TZT-1027 was reached at the end of the 1-h infusion, varying between 124.6 and 333.1 ng/ml. The T1/2 was 6.52–7.42 h, with no major difference over the administered dose ranges. The Cl varied between 2.42 and 5.31 l/h/m2, with a Vss of 24.6–64.8 l. The AUC ranged from 326 to 1219 ng·h/ml.


View this table:
[in this window]
[in a new window]
 
Table 5. Summary of pharmacokinetics of TZT-1027
 
With the exception of patients treated on dose level 2.4 mg/m2, pharmacokinetics were linear. The dose-dependent average plasma concentrations over two courses of intravenous administration of TZT-1027 are shown in Figure 2. On all dose levels, plasma AUCs were higher in cycle two as compared to cycle one (see also Table 5). Data analyses of the pharmacokinetic parameters were limited to descriptive statistics due to the small patient population. A dose relationship could not be established due to the sample size limitations in each dosing level.



View larger version (19K):
[in this window]
[in a new window]
 
Figure 2. Pharmacokinetics of TZT-1027. Average plasma concentration over two courses of intravenous administration of TZT-1027. Samples were taken on day 1 and day 1 of the following first and second treatment cycles. Patients were treated with doses of 1.35, 1.8, 2.4, 2.7 and 3.0 mg/m2. LLOQ, lowest level of quantification.

 
Pharmacodynamics
The correlations between hematological toxicity and pharmacokinetic parameters showed the absolute neutrophil count (ANC) decreased with increasing AUC and Cmax (Figure 3). A correlation between {alpha}1-AGP versus Cl in Figure 4 shows Cl values decreased with increasing plasma {alpha}1-AGP levels. The described correlations were poor, most likely related to the limited number of samples.



View larger version (10K):
[in this window]
[in a new window]
 
Figure 3. Plasma pharmacodynamics of TZT-1027. Correlation between absolute neutrophil count (ANC) percentage decrease from baseline and exposure of TZT-1027 (AUCinf and Cmax) in courses one and two of treatment.

 


View larger version (10K):
[in this window]
[in a new window]
 
Figure 4. Plasma pharmacodynamics of TZT-1027. Correlation between the transport protein {alpha}1-AGP and clearance of TZT-1027.

 

    Discussion
 Top
 ABSTRACT
 Introduction
 Patients and methods
 Results
 Discussion
 REFERENCES
 
TZT-1027 is a synthetic derivative of dolastatin 10 [18], a mitotic spindle poison that interacts with tubulin in the same domain as the vinca alkaloid binding region. The spectra of antitumor activity of the dolastatin 10 derivative and the vinca alkaloids are not identical in vivo, despite a similar mechanism of action [19, 20]. More recent work on the mode of action of the innovative compound has shown that the drug induces DNA fragmentation and apoptotic chromatin condensation, and that the tumor cells are arrested in the G2/M phase of the cell cycle. The induction of apoptosis by TZT-1027 is independent of the presence or absence of caspase-3 or bcl-2 [9]. According to in vitro studies performed with tumor tissue obtained from patients with lung and renal cell cancers, the activity of TZT-1027 is influenced less by the p53 mutation status than DNA-damaging agents, which may be very relevant for many clinical cancers [21].

Based on cDNA macroarray gene expression studies, the drug modulates the expression of a variety of genes in malignant and non-malignant cell lines, including genes encoding cell cycle and growth regulators, receptors, invasion regulators, cytokines and angiogenic factors [21].

In the murine colon 26 adenocarcinoma model, TZT-1027 was found to attack tumor vasculature, as tolerable doses of the drug induced tumor-selective hemorrhage within 1 h after administration. The vascular damage was followed by immediate induction of apoptosis of the tumor cells. In cultured human umbilical vein endothelial cells, TZT-1027 induced marked cell contraction with membrane blebbing within 30 min, which could be blocked by inhibitors of protein kinases.

These preclinical findings indicate that TZT-1027 has both a conventional antimitotic activity similar to that of the vinca alkaloids, but may also have further unique molecular antitumoral properties, including vascular effects [22]. On the basis of this interesting preclinical profile, a series of clinical phase I studies have been initiated by Teikoku Hormone Mfg. Co., Ltd, and later by Daiichi Pharmaceuticals Co., Ltd, Japan. Until the end of June 2001, 88 patients have been treated in three separate studies, two performed in Japan and one in Hungary. So far, none of these trials has been fully published.

The primary objective of our present phase I study was to establish the MTD for a 3-week dosing schedule in patients with refractory solid tumors. Furthermore, we wanted to determine the safety profile and DLT of the investigational product by systematically assessing the side-effects of TZT-1027. Secondary objectives were to study the pharmacokinetics and pharmacodynamics in human subjects and to evaluate the potential antitumor activity of TZT-1027 in a population of patients with relapsed cancers, to provide further guidance for the design of future disease-oriented trials.

The starting dose of 1.35 mg/m2 chosen in this trial was based on preliminary results from the ongoing Hungarian lung cancer study, where grade 4 neutropenia was first observed at doses of 1.8 mg/m2 [12]. Hence we were able to avoid treating too many patients on sub-toxic dose levels and we were able to increase the dose to a maximum of 3.0 mg/m2.

The most frequent non-hematological, reversible side-effects included alopecia, fatigue, nausea, anorexia and constipation, which are common complications of antineoplastic treatment. Treatment-related pain, mainly abdominal, was observed repeatedly, and could possibly be interpreted as a vascular effect, as TZT-1027 induces tumor-selective hemorrhage in animal models, and most of our patients had abdominal tumor manifestations. The coincidence of abdominal pain and constipation in this trial could also be related to neurotoxicity.

The most frequent hematological events were leukopenia and neutropenia, which confirms the experience with dolastatins in other early clinical studies. In the majority of cases, the neutropenia was mild, short-lasting and in no case complicated by febrile or infectious complications. Growth factors or systemic antibiotics were not required.

As an unexpected finding in this trial, DLT was not limited to hematological events. DLTs occurred exclusively at the 3.0 mg/m2 dose level, and two out of three patients with DLT reported grade 3 non-hematological side-effects during cycle one. These events included severe fatigue and variable manifestations of a presumed neurotoxic syndrome of mandibular cramps, pain of the arms and legs, paresthesia, insomnia and agitation. Electrolyte imbalances were ruled out. Of note is that all three patients with DLT had been pretreated with oxaliplatin. None of them had clinical evidence of neuropathy at baseline prior to the first administration of TZT-1027, with the exception of one patient with known diabetic neuropathy. Peripheral neurotoxicity is the most frequent DLT of oxaliplatin [23]. Acute neurotoxicity is characterized by the rapid onset of cold-induced distal dysesthesia and/or paraesthesia. Sensory symptoms may also be accompanied by cold-dependent muscular contractions of the extremities or the jaw. The symptoms, often occurring during or shortly after infusion, are usually transient and mild. A persistent sensory peripheral neuropathy may also develop with prolonged treatment, eventually causing superficial and deep sensory loss, sensory ataxia and functional impairment. Studies have shown patients with acute sensory symptoms to display little or no axonal degeneration, suggesting a specific effect of oxaliplatin on sensory neurons and/or motor neurons or muscle cells that is not observed with other platinum agents. The similarity of the acute symptoms induced by oxaliplatin with those caused by several drugs or toxins acting on neuronal or muscular ion channels suggests that these symptoms may result from a specific interaction of oxaliplatin with ion channels located in the cellular membrane.

An exacerbation of oxaliplatin neurotoxicity has recently been described in patients following surgery, probably through a redistribution of a pool of intra-erythrocytic oxaliplatin biotransformation products into the plasma [24]. Possibly TZT-1027 has a similar interaction with oxaliplatin, leading to some kind of ‘recall’ neurotoxicity.

Severe neurotoxicity itself is an unexpected finding for TZT-1027, even though the drug acts on tubulin. In the Japanese phase I trial, sensory abnormalities were limited to two grade 1 events among 40 patients [10]. However, it should be taken into account that this study only reached a maximum dose of 2.1 mg/m2 whereas in our study this toxicity was only observed at dose level 3.0 mg/m2.

In contrast to vincristine or paclitaxel, two well-known neurotoxic spindle poisons, TZT-1027 does not induce neuropathological alterations in laboratory animals such as rabbits or mice [25]. The presumed toxicities of TZT-1027 with oxaliplatin could be studied in similar models, and these preclinical toxicology studies are currently considered by the manufacturing company, based on our clinical findings.

In our patient population, the recommended dose for further clinical trials is 2.7 mg/m2 for the 3-weekly administration of TZT-1027. No DLTs were observed in six patients exposed to this dose. It cannot be excluded that higher doses can be administered in patients who have not been pretreated with specific neurotoxic agents. Our findings support the hypothesis, that TZT-1027 may exert some degree of neurotoxicity in patients previously exposed to neurotoxic compounds. The pending final results of the other phase I studies should be considered for selecting an optimum schedule for phase II trials.


    Acknowledgements
 
We acknowledge the excellent assistance of Bianca Wawzik, study nurse, Radoslaw Kowalski, data manager, Michaela Mühlefeld, study monitor, and all staff at Daiichi Pharmaceuticals UK Ltd, London, Daiichi Pharmaceutical Co., Ltd, Tokyo, and MDS Pharma Services, Montreal, Canada, involved in this trial.


    FOOTNOTES
 
* Correspondence to: Dr P. Schöffski, Department of Hematology and Oncology, Hannover Medical School, Carl-Neuberg-Str. 1, D-30625 Hannover, Germany. Tel: +49-511-532-4077; Fax: +49-511-532-8077; E-mail: schoeffski.patrick@mh-hannover.de Back


    REFERENCES
 Top
 ABSTRACT
 Introduction
 Patients and methods
 Results
 Discussion
 REFERENCES
 
1. Miyazaki K, Kobayashi M, Natsume T et al. Synthesis and antitumour activity of novel dolastatin analogs. Chem Pharm Bull 1995; 43: 1706–1718.[ISI][Medline]

2. Pettit GR, Kamano Y, Herald CL et al. The isolation and structure of a remarkable marine animal antineoplastic constituent: dolastatin 10. J Am Chem Soc 1987; 109: 6883–6885.[ISI]

3. Pettit GR, Singh SB, Hogan F et al. The absolute configuration and synthesis of natural dolastatin 10. J Am Chem Soc 1987; 111: 5463–5465.

4. Pettit GR, Kamano Y, Herald CL et al. Isolation of dolastatin 10–15 from the marine mollusk Dolabella auricularia. Tetrahedron 1993; 49: 9152–9170.

5. Hamada Y, Hayashi K, Shiori T. Efficient stereoselective synthesis of dolastatin 10, an antineoplastic peptide from a sea hare. Tetrahedron Lett 1991; 32: 931–934.[CrossRef][ISI]

6. Kobayashi M, Natsume T, Tamaoki S et al. Antitumor activity of TZT-1027, a novel dolastatin 10 derivative. Jpn J Cancer Res 1997; 88: 316–327.[ISI][Medline]

7. Kobayashi M, Natsume T, Watanabe J et al. Activity of a novel antitumor agent, TZT-1027. Nippon Yakurigaku Zasshi 1999; 114 (Suppl 1): 230P–235P.[Medline]

8. Natsume T, Kobayashi M, Fujimoto S. Association of p53 gene mutations with sensitivity to TZT-1027 in patients with clinical lung and renal carcinoma. Cancer 2001; 92: 386–394.[CrossRef][ISI][Medline]

9. Watanabe J, Natsume T, Fujio N et al. Induction of apoptosis in human cancer cells by TZT-1027, an antimicrotubule agent. Apoptosis 2000; 5: 345–353.[CrossRef][ISI][Medline]

10. Niitani H, Hasegawa K. Phase I studies of TZT-1027, a novel inhibitor of tubulin polymerization. Ann Oncol 1998; 9 (Suppl 2) (Abstr 360).

11. Yamamoto N, Andoh M. Phase I study of TZT-1027, an inhibitor of tubulin polymerization, given weekly x 3 as a 1-hour intravenous infusion in patients with solid tumors. Proc Am Soc Clin Oncol 2002; 93 (Abstr 420).

12. Horti J, Juhasz E, Bodrogi I. Preliminary results of a Phase I trial of TZT-1027, an inhibitor of tubulin polymerization, in patients with advanced non-small cell lung cancer. Proc Am Assoc Cancer Res 2002; 93 (Abstr 2744).

13. Therasse P, Arbuck SG, Eisenhauer EA et al. New guidelines to evaluate the response to treatment of solid tumors. J Natl Cancer Inst 2000; 92: 205–216.[Abstract/Free Full Text]

14. Young H, Baum R, Cremerium U et al. Measurement of clinical and subclinical tumor response using [18F]-fluorodeoxyglucose and positron emission tomography. Review and 1999 EORTC recommendations. Eur J Cancer 1999; 35: 1773–1782.[CrossRef][ISI][Medline]

15. Ochiai H, Mori H, Murata H et al. Validation of an analytical method for a potent antitumor agent, TZT-1027, in plasma using liquid chromatography-mass spectrometry. J Chromatogr B Biomed Sci Appl 2001; 762: 155–163.[CrossRef][Medline]

16. Oguma T, Atsumi R. TZT-1027: validation of a high performance liquid chromatographic mass spectrometric method for the determination of TZT-1027 in human plasma. Daiichi Pharmaceutical Co. Ltd. Study Report No. 20010582, 26 December 2001.

17. Oguma T, Atsumi R. TZT-1027: validation of a high performance liquid chromatographic mass spectrometric method for the determination of TZT-1027 in human urine. Daiichi Pharmaceutical Co. Ltd. Study Report No: 20010583, 26 December 2001.

18. Pitot HC, Frytak S, Croghan GA et al. Phase II study of dolastatin-10 (dola-10) in patients (pts) with advanced renal cell cancer. Proc Am Soc Clin Oncol 2002; 20 (Abstr 2409).

19. Fujita F, Koike M, Fujita M et al. Antitumor effects of TZT-1027, a novel dolastatin 10 derivative, on human xenografts in nude mice. Gan to Kagaku Ryoho 2000; 27: 451–458.[Medline]

20. Natsume T, Watanabe J, Tamaoki S et al. Characterization of the interaction of TZT-1027, a potent antitumor agent, with tubulin. Jpn J Cancer Res 2000; 91: 737–747.[ISI][Medline]

21. Natsume T, Nakamura T, Koh Y et al. Gene expression profiling of exposure to TZT-1027, a novel microtubule-interfering agent, in non-small cell lung cancer PC-14 cells and astrocytes. Invest New Drugs 2001; 19: 292–302.

22. Otani M, Natsume T, Watanabe JI et al. TZT-1027, an antimicrotubule agent, attacks tumor vasculature and induces tumor cell death. Jpn J Cancer Res 2000; 91: 837–844.[ISI][Medline]

23. Gamelin E, Gamelin L, Bossi L, Quasthoff S. Clinical aspects and molecular basis of oxaliplatin neurotoxicity: current management and development of preventive measures. Semin Oncol 2002; 29 (Suppl 15): 21–33.

24. Gornet JM, Savier E, Lociec F et al. Exacerbation of oxaliplatin neurosensory toxicity following surgery. Ann Oncol 2002; 13: 1315–1318.[Abstract/Free Full Text]

25. Ogawa T, Mimura Y, Isowa K et al. An antimicrotubule agent, TZT-1027, does not induce neuropathologic alterations which are detected after administration of vincristine or paclitaxel in animal models. Toxicol Lett 2001; 121: 97–106.[CrossRef][ISI][Medline]