Phase I trial of intravenous aviscumine (rViscumin) in patients with solid tumors: a study of the European Organization for Research and Treatment of Cancer New Drug Development Group

P. Schöffski1,*, S. Riggert1, P. Fumoleau2, M. Campone2, O. Bolte1, S. Marreaud3, D. Lacombe3, B. Baron3, M. Herold4, H. Zwierzina4, K. Wilhelm-Ogunbiyi5, H. Lentzen5 and C. Twelves3

1 Department of Haematology and Oncology, Hannover Medical School, Hannover, Germany; 2 Department of Medical Oncology, Centre René Gauducheau, Centre Regional de Lutte Contre le Cancer Nantes-Atlantique, Nantes, France; 3 EORTC New Drug Development Office (NDDG), Brussels, Belgium; 4 Department for Oncology, University Clinic Innsbruck, Innsbruck, Austria; 5 VISCUM AG, Bergisch Gladbach, Germany

* Correspondence to: Professor Dr P. Schöffski, Department of Oncology, Universitair Ziekenhuis Gasthuisberg, Herestraat 49, B-3000 Leuven, Belgium. Tel. +32-16-346900/16; Fax +32-16-346901; Email: patrick.schoffski{at}uz.kuleuven.ac.be


    Abstract
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Background: Aviscumine is an Escherichia coli-derived recombinant type II ribosome-inactivating protein with potent antitumor activity in vitro and in vivo. It is the recombinant counterpart of natural mistletoe lectin-I. The current study was performed to determine the safety profile, dose-limiting toxicity (DLT) and maximum tolerated dose (MTD) of the intravenous (i.v.) administration of aviscumine in cancer patients. Translational research included the evaluation of pharmacokinetics and monitoring of plasma cytokine and anti-aviscumine antibody induction after administration of the drug.

Patients and methods: Aviscumine was given twice weekly as a 1 h central i.v. infusion in patients with advanced, refractory progressive, solid malignant tumors who had not been previously exposed to natural mistletoe preparations. They had histologically or cytologically verified disease, were ≥18 years old, had an Eastern Cooperative Oncology Group performance status ≤2 and adequate bone marrow, liver and renal function. DLT was defined as any non-hematological grade 3–4 toxicity (National Cancer Institute Common Toxicity Criteria version 2.0), neutrophil count <500/µl for ≥7 days, febrile neutropenia or thrombocytopenia grade 4. The MTD was defined as the dose at which >20% of patients experienced DLT during the first treatment cycle. The Continual Reassessment Method was used to determine the number of patients required per dose level.

Results: Forty-one fully eligible patients (19 male, 22 female) with a median age of 56 years (range 37–74) were enrolled. Colorectal, ovarian, renal cell and breast cancer were the most common tumor types. Dose levels of aviscumine ranged from 10 to 6400 ng/kg. The median number of cycles was two (range one to eight). Common clinical toxicities in cycle 1 were fatigue, fever, nausea, vomiting and allergic reactions. Fatigue grade 3 was dose limiting in one of six patients at 4000 ng/kg and reversible grade 3 liver toxicity (elevation in alkaline phosphatase, transaminases and/or {gamma}-glutamyltransferase) occurred in one of 10 patients at 4800 ng/kg and in two of five patients at 6400 ng/kg. The best response (RECIST criteria) was stable disease in 11 patients, lasting for two to eight cycles. The pharmacokinetic evaluation revealed a short {alpha} half-life of 13 min and linear kinetics on dose levels ≥1600 ng/kg. Aviscumine stimulated the immune system with a release of cytokines such as interleukin (IL)-1ß, IL-6 and interferon-{gamma}, and induced immunoglobulin (Ig) G- and/or IgM-anti-aviscumine antibodies of uncertain clinical relevance.

Conclusions: The recommended dose for further clinical trials is 5600 ng/kg twice weekly. Based on the short half-life of the recombinant protein observed in this trial, the exploration of prolonged infusion schedules of aviscumine is warranted.

Key words: aviscumine, CD75s, phase I study, recombinant mistletoe lectin, ribosome-inactivating proteins, solid tumors


    Introduction
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Mistletoe lectin-I (ML-I; Viscum album agglutinine I) is known to be the major active constituent of natural mistletoe extracts [1Go]. The lectin component can induce apoptosis in various transformed cell lines [2Go–6Go]. ML-I is a 66-kDa heterodimer that consists of a toxic A-chain, a site-specific type II ribosome-inactivating N-glycosidase [7Go] and a carbohydrate-binding subunit B responsible for cellular lectin uptake [8Go–10Go].

The 1923-nucleotide ML-I gene was sequenced in 1999 and the existence of a single intron-free gene was demonstrated. The production of pure, biochemically defined ML-I was then achieved by recombinant cloning and separate expression of the A- and B-chain in Escherichia coli BL21DE3 by recombinant DNA technique, yielding the active heterodimeric protein rViscumin (INN; aviscumine) [9Go, 10Go]. The expression vectors contain the exact coding sequence of the genomic DNA fragments of the subunits A (26 kDa) and B (30 kDa) as found in European mistletoe. The recombinant drug is obtained by a single processing step folding of the solubilized subunits and linking by a disulfide bond. Purification of the protein is achieved by two chromatographical steps yielding >98% homogeneous material. The manufacturing process finally results in an aqueous, buffered solution of aviscumine, which is stable at pH 7–9 (Figure 1) [10Go].



View larger version (55K):
[in this window]
[in a new window]
 
Figure 1. Composition of aviscumine.

 
Extract preparations from natural mistletoe (V. album) have been widely used as an alternative therapy in the management of patients with malignant disease for several decades, based on presumed immunostimulatory and antineoplastic effects [1Go, 11Go]. There is, however, an obvious discrepancy between the popularity of mistletoe extracts among cancer patients, the commercial success of the natural product and a striking lack of evidence-based data to support their use in oncology [12Go–15Go].

The availability of the pure, homogeneous protein was a prerequisite for reproducible structural and mechanistic studies, in order to gain a better insight into the mode of action of this product on tumor and immune cells, as the isolation of the glycosylated proteins from natural mistletoe only yields inhomogeneous material, probably due to post-translational modifications. Therefore, the recombinant drug was studied extensively in preclinical models. It was found to possess impressive antineoplastic and immunomodulatory properties in vitro and in vivo, and was tolerated well in animals [16Go–19Go].

Furthermore, the cellular receptor involved in the binding of the heterodimer could be identified [20Go, 21Go]. Aviscumine was shown to bind preferentially to terminally {alpha}2-6-sialylated ganglioside structures. In contrast to another well-known ribosome-inactivating agent, ricin, aviscumine was therefore characterized as a sialic acid-specific type II ribosome-inactivating protein. Neolacto-series gangliosides with a Neu5Ac{alpha}2-6Galß1-4GlcNAc-terminus (CD75s), rather than galactose-specific, were defined as the aviscumine receptors for the B-chain, leading to internalization of the holoprotein [20Go].

In further mechanistic studies, the A-chain was confirmed to be a potent hydrolase that selectively cleaves the N-glycosidic bond of the adenine-4324 residue in eukaryotic 28S ribosomal RNA. This effect leads to a catalytical inactivation of ribosomes, and thereby inhibits translation and protein synthesis as the presumed mode of action [7Go, 9Go].

In vitro, aviscumine was found to be a very potent cytotoxic agent, which induced apoptosis in a low concentration range (fM to pM) and necrosis at higher concentrations (nM to mM). The mean aviscumine concentration required to eliminate 70% of human tumor cell lines in vitro (Freiburg tumor panel) was 0.4 ng/ml, and was 2 ng/ml to inhibit tumor xenograft colony formation by 70%. On a molar basis, aviscumine was ~5000 times more potent than the standard therapeutic agent doxorubicin and 1500 times more potent than paclitaxel against some human tumor cell lines. Aviscumine was more active against doxorubicin-resistant breast tumor cells and vindesine-resistant pleural mesothelioma cells than against the respective drug-sensitive parental cell lines, indicating that the drug is not affected by common resistance mechanisms. Interactions between aviscumine and conventional anticancer drugs were investigated in vitro. Aviscumine was capable of enhancing the cytotoxic effects of vincristine, mafosfamide, idarubicin and cisplatin in the human leukemia cell lines K562 and KG1a.

The intraperitoneal (i.p.), subcutaneous (s.c.) or intravenous (i.v.) administration of the recombinant protein was shown to have growth inhibitory action in various heterotopic tumors and metastasis mouse models, including the C8 colon 38-carcinoma, Lewis lung sarcoma, Renca renal carcinoma, F9 testicular carcinoma, RAW 117-H10 P lymphosarcoma, L-1 sarcoma and B16 melanoma models [18Go]. A significant inhibitory effect on experimental urothelial carcinogenesis was seen after intravesical instillation into rats. Intratumoral treatment of human ectopic CXF 280, and i.p. administration in 60444 and 5776 colon cancers in immunodeficient mice, led to a strong, dose-dependent inhibition of tumor progression. Combination therapy of aviscumine with doxorubicin was found to be superior to the effect of each single agent alone. The i.p. administration of aviscumine into human ovarian cancer SoTü3-bearing SCID mice led to a significant prolongation of survival [17Go].

Immunological studies in vitro showed that aviscumine stimulated the release of interleukin (IL)-1{alpha} and IL-6 from human keratinocytes and fibroblasts, the release of IL-12, interferon (IFN)-{gamma} and tumor necrosis factor (TNF)-{alpha} from peripheral blood mononuclear cells, as well as the expression of the IL-2 receptor {alpha} chain and HLA-DR on peripheral blood T lymphocytes. Additionally, aviscumine increased the activity of human natural killer cells against lymphoma cells ex vivo [16Go]. In tumor-bearing mice, aviscumine was found to enhance the number or to induce the activated state of some blood cell subpopulations (CD4/8 + CD45+ cells, activated MAC3 + CD45+ cells).

Toxicology studies have shown that aviscumine can be safely applied in animals and lacks relevant genotoxic or mutagenic effects up to dosages of 1000 ng/kg i.v. Repeated i.v. or s.c. dosing did not reveal any specific target organ toxicity in rats and dogs up to 1000 ng/kg. Bleeding complications were observed on much higher doses, i.e. at 4000 ng/kg in rats and at 10 000 ng/kg in dogs after 1 week of daily i.v. application. Local reactions at s.c. injection sites were observed at concentrations from 50 ng/ml upwards.

The promising preclinical profile of the recombinant ribosome-inactivating protein was the scientific basis for designing this very first i.v. dose-finding trial of aviscumine in patients with refractory solid tumors.


    Patients and methods
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Objectives
This phase I, open-label, multiple dose, dose-escalation study was conducted in two sites in Germany and France. The primary objective was to assess the safety and toxicity profile of the twice weekly 1 h i.v. infusion of aviscumine, to evaluate the dose-limiting toxicity (DLT) and maximum tolerated dose (MTD), and to establish a recommended dose for further clinical trials. Secondary objectives were to study the pharmacokinetics and immunological properties of the antineoplastic compound in patients, including the repeated determination of plasma cytokine levels and monitoring of anti-aviscumine antibodies.

Patient selection criteria
Patients with a histologically or cytologically confirmed diagnosis of a progressive solid malignant tumor were eligible for the trial, provided the disease was assumed to be refractory to further conventional treatment. Standard selection criteria included age ≥18 years, Eastern Cooperative Oncology Group (ECOG) performance status 0–2, life expectancy of ≥3 months, no central nervous system involvement, adequate bone marrow (white blood cell count ≥3 x 109/l, neutrophils ≥1.5 x 109/l, platelets ≥100 x 109/l), liver [serum bilirubin within 1.5x the upper limit of normal (ULN), serum aspartate aminotransferase (AST), alanine aminotransferase (ALT) and alkaline phosphatase <2.5x ULN if no liver metastases, or <5x ULN in case of liver metastases] and renal (serum creatinine <120 µmol/l) function. The study excluded patients with previous exposure to natural mistletoe preparations to avoid interference with secondary trial end points. Prior chemotherapy, hormone and/or radiation treatment had to be completed at least 4 weeks prior to study entry. Patients were not allowed to receive any immunostimulating substances, biological response modifiers, colony-stimulating factors, systemic steroids or monoclonal antibodies within 4 weeks prior to and during the study. Patients had to use effective contraception if of reproductive potential and were not allowed to be pregnant or lactating. Written informed consent was obtained from all participating patients according to applicable German and French laws and regulations. Patient insurance was provided by the European Organization for Research and Treatment of Cancer (EORTC). The study was conducted in accordance with the Good Clinical Practice Guidelines as issued by the International Conference on Harmonization and the Declaration of Helsinki.

Study logistics
After obtaining informed consent, patients were screened for potential trial participation. Eligible candidates were registered by a telephone procedure at the EORTC Data Center prior to the start of treatment after verification of the eligibility criteria and selection of the target lesions. The treatment started within 1 week after registration. Aviscumine was given twice weekly until one of the following withdrawal criteria occurred: disease progression, unacceptable toxicity, patient's refusal or patient's best interest according to the treating physician.

Study treatment
Aviscumine concentrate was supplied in 10 ml vials by the sponsor (VISCUM AG, Bergisch Gladbach, Germany) and kept in the hospital pharmacy at 2–8°C. The drug was formulated as a pyrogen-free, phosphate-buffered solution at pH 8.0, and contained sodium chloride, polysorbate 80 and glutamic acid. The appropriate volume of aviscumine was calculated according to the body weight of the patient, diluted with physiological saline up to a total volume of 1000 ml and administered within 8 h after preparation of the infusion. Aviscumine was given twice weekly (days 1 and 4) as a 1-h i.v. infusion through a central line, to avoid the potential risk of local vascular effects in peripheral veins. In most cases, port catheters were used for this purpose. One cycle was arbitrarily defined as three consecutive weeks of treatment (six infusions). The treatment was given continuously without interruption between the treatment cycles. After the end of each cycle, the safety and tolerability of the treatment was assessed in detail. Patients with progressive disease were removed from the study; patients achieving stable disease or an objective response could remain on treatment. Supportive care was at the discretion of the treating physicians, but corticosteroids, immunostimulating agents, other anticancer drugs, monoclonal antibodies or hematopoietic growth factors were prohibited.

Dose escalation procedure
The dose escalation procedure in this phase I study was guided by the Continual Reassessment Method (CRM) [22Go, 23Go]. Dose escalation was done on an inter-patient basis. The aim of using the CRM was to decrease the number of patients on low, non-toxic dose levels in this trial. The protocol allowed the evaluation of intermediate dose levels. The number of dose levels was not fixed in advance. The initial version of the protocol foresaw the following steps: 10, 20, 40, 100, 200, 400, 800 and 1600 ng/kg per administration of aviscumine. The protocol was later amended, as the treatment was found to be well tolerated up to the dose of 1600 ng/kg, and higher dose levels were allowed.

Definition of DLT, MTD and recommended dose
DLTs were defined as any non-hematological grade 3–4 toxicity (with the exclusion of nausea, vomiting or fever), an absolute neutrophil count <500/µl lasting for ≥7 days, febrile neutropenia (defined as an absolute neutrophil count <500/µl lasting for ≥3 days associated with fever ≥38.5°C for 24 h) or thrombocytopenia of National Cancer Institute Common Toxicity Criteria NCI CTC grade 4. The MTD for this study was defined as the highest dose at which a maximum of 20% of the patients of a certain dose level experienced DLT during the first treatment cycle.

Clinical evaluation, laboratory tests and follow-up
Within 2 weeks prior to study treatment, the following baseline tests were obtained: tumor evaluation, tumor markers (if applicable), complete medical history, ECG and pregnancy test in fertile female patients. Within 7 days, the following parameters were determined: weight, height, complete blood count with differential and platelet count, serum chemistry (blood urea nitrogen, creatinine, total protein, albumin, bilirubin, lactate dehydrogenase, creatin kinase, alkaline phosphatase, ALT, AST and electrolytes) and coagulation panel (PTT and PT). Immediately prior to the first administration of aviscumine, the following parameters were obtained: physical examination, ECOG performance status, temperature (°C) and vital signs. Vital signs were repeated at 15, 30, 60 and 120 min, and then 4 and 8 h, after the first i.v. infusion. On a weekly basis during all treatment cycles, the following parameters were assessed: physical examination, toxicity assessment, ECOG performance status, vital signs, complete blood count, serum chemistry and coagulation panel. A detailed toxicity evaluation according to NCI CTC was performed prior to each treatment cycle. The tumor imaging and marker assessment was repeated every 6 weeks.

Criteria for evaluation of toxicity and response
All adverse events were graded according to the NCI CTC (version 2.0). The causality of clinical events was categorized as unrelated, unlikely, possible, probable, definitely related or not assessable. Diseases and clinical symptoms already known at the beginning of the study were not classified as adverse events if they were encountered again at subsequent examinations, except in the case of a relevant increase in intensity or incidence. All patients who had started the treatment were included in the overall toxicity assessment. Serious adverse events were defined according to the ICH Good Clinical Practice Guidelines and reported within 24 h to the EORTC. This included all events occurring during or within 30 days of termination of the study treatment. The response to the treatment with aviscumine was assessed according to the Response Evaluation Criteria in Solid Tumors (RECIST) criteria [24Go].

Statistical considerations
The experimental design for this study was an adaptation of the CRM, which is based on the concept of entering every patient at the best current statistical estimation of the MTD given the previous observations. The trial was conducted in two consecutive steps: step 1, until mild toxicity was observed; and step 2, toxicity evaluation regarding the DLT and MTD. Cohorts of only one patient were to be treated per dose level, as long as there was no toxicity exceeding grade 1. New patient inclusion was blocked until the tolerance of the patient to the first cycle was fully documented. In the event of moderate drug-related toxicity (grade 2), the cohort size was increased from one to three patients for all further groups and the dose escalation was slowed down. At that point, at each dose level, after the first patient had been registered, new patient inclusion was blocked until the tolerance of the first patient to the first cycle was fully documented. In order to proceed to a next dose level, none of three patients treated on that dose level was allowed to have experienced a grade 3 or 4 toxicity related to the treatment during the first cycle.

As soon as a DLT was observed, the second stage started. In this stage, the dose escalation strategy was based on a one-parametric statistical model, based on the relationship between the dose level di and the probability of DLT at that level, {phi}({alpha}i, a)={alpha}ia, where {alpha}1=0.04, {alpha}2=0.07, {alpha}3=0.20, {alpha}4=0.35, {alpha}5=0.55, {alpha}6=0.70 was a recoding of the dose levels di. Using a maximum likelihood estimator, estimates of the probability of toxicity at each available dose level were updated after every event. Given these estimates, the next patient was included at the dose level whose estimate was closest to the targeted percentile, 20%. After each estimate a confidence interval of the probability of toxicity at the current recommended dose was calculated.

Pharmacokinetic analysis
The translational research component of the protocol included plasma sampling for pharmacokinetics and immunological effects of aviscumine. Pharmacokinetic sampling was carried out during and after the first and 11th administration of aviscumine. Samples were taken 15 min prior to the i.v. infusion and 1 h, 1 h 15 min, 1 h 30 min, and 2, 4, 8 and 24 h after the administration. The samples were shipped on dry-ice to Chimera Biotech (Dortmund, Germany) for analytical purposes. Pharmacokinetic analyses were performed using an immuno-PCR method [25Go].

Patient plasma samples were analyzed directly or diluted prior to analysis with blank plasma samples to an aviscumine concentration of up to 6100 pg/ml. Quantitation was carried out in comparison with an aviscumine reference curve in each individual patient's blank plasma. Plasma samples were transferred into 96-well titre plates with capture antibody (polyclonal rabbit anti-aviscumine antibody; VISCUM AG), and aviscumine was detected with a polyclonal antibody rabbit–biotin conjugate and streptavidine–DNA conjugate (Chimera Biotech). PCR was carried out with Taq polymerase (bioTaq; Biomaster GmbH, Köln, Germany) and digoxigenin-ddUTP (Roche, Grenzach–Wyhlen, Germany), and a specific competitor (Chimera Biotech) for 28 cycles. Digoxigenin was determined with anti-digoxigenin alkaline phosphatase conjugate and Atto-Phos Substrate Kit (Roche) by fluorogenic detection.

For the detection of anti-avicumine antibodies in plasma, an enzyme-linked immunosorbent assay (ELISA) was used with the murine monoclonal anti-aviscumine B-chain antibody 36-2-0 (VISCUM AG) at a concentration of 5 µg/ml plus aviscumine at a concentration of 5 µg/ml as antigen. Plasma samples were diluted to 1:100, 1:1000 and 1:10 000 with phosphate-buffered saline. Detection of the IgM and IgG titre was carried out with anti-human IgM (µ-chain specific) POD conjugate and with anti-human IgG (Fc specific) both from goat (Sigma-Aldrich, Munich, Germany). This analysis was carried out by PHAST (Homburg, Germany).

Cytokine studies involved the repeated determination of plasma IL-1ß, -6, -10, -12, IFN-{gamma} and TNF-{alpha}. The samples were shipped to the Department of Internal Medicine at University Clinic of Innsbruck, where commercially available ELISAs were used to determine the plasma levels. The results of the translational research were analyzed by descriptive and exploratory methods.


    Results
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Patient characteristics
Between August 2000 and February 2003, 41 fully eligible patients entered the trial. The median age was 56 years. A broad range of solid tumors was represented, with a predominance of adenocarcinomas. Colorectal, ovarian, renal cell and breast cancer were the most common tumor types. The vast majority of patients were pretreated with surgery and chemotherapy, others had received radiotherapy, hormone treatment and/or immunotherapy (Table 1).


View this table:
[in this window]
[in a new window]
 
Table 1. Patient characteristics

 
Treatment administration
The median number of treatment cycles administered was two (range one to eight). There were no dose modifications in cycle 1. During the further study treatment, 10 patients had dose modifications due to non-hematological toxicity, one patient had a dose modification due to both hematological and non-hematological toxicity, and four patients had schedule changes for non-drug-related reasons.

The dose levels visited ranged from 10 to 6400 ng/kg. Only one patient per cohort was required on dose levels 10, 20, 40, 100, 200, 400, 800, 1600 and 2400 ng/kg, respectively, as toxicity in cycle 1 did not exceed grade 1. The CRM allowed these nine dose levels to be visited within only 7 months, and to escalate the dose 240-fold in only nine patients without compromising patient safety. With the occurrence of grade 2 toxicity from dose level 3200 ng/kg onwards, the dose escalation was slowed down according to the model. The observation of a sporadic grade 3 event (fatigue in a patient with renal cell carcinoma and advanced lung metastasis) during the first treatment cycle with 4000 ng/kg aviscumine, triggered the statistical model to evaluate six patients on this dose level, but no further DLTs were observed in this cohort. Another 10 patients were treated with the 4800 ng/kg with the occurrence of one DLT at this dose level, and seven patients were exposed to 5600 ng/kg without occurrence of a DLT. Further dose-limiting events were then observed in two out of five patients treated with a dose of 6400 ng/kg. The majority of patients discontinued the treatment due to disease progression; three patients were taken off the study due to liver toxicity (Table 2).


View this table:
[in this window]
[in a new window]
 
Table 2. Treatment administration

 
Toxicity
All patients were evaluated for toxicity, including seven patients who left the study during the first treatment cycle due to disease progression (four) or toxicity (three). In general, aviscumine was very well tolerated. The most frequent clinical toxicities were fatigue, fever, nausea, vomiting and allergic reactions, and with few exceptions these were grade 1/2 events (Table 3). Hematological toxicity was uncommon. Thrombocytopenia and leukopenia were limited to grade 1 during cycle 1; anaemia was a more common feature but was related to the underlying malignancy in the majority of cases. The 1-h i.v. infusion of aviscumine was not associated with granulocytopenia. The majority of serum chemistry abnormalities in cycle 1 were attributed to the malignant disease (Table 4). There was not a single case of grade 4 toxicity in cycle 1 in this trial, and no evidence of cumulative toxicity in patients continuing treatment beyond a first treatment cycle.


View this table:
[in this window]
[in a new window]
 
Table 3. Non-hematological toxicity (maximum toxicity in cycle 1 per patient, all dose levels, related events only)

 

View this table:
[in this window]
[in a new window]
 
Table 4. Laboratory abnormalities (maximum abnormality in cycle 1 per patient, all dose levels, related and unrelated events)

 
DLTs, MTD and recommended dose
Fatigue grade 3 was observed in a 73-year-old male patient with metastatic renal cell carcinoma during the first treatment cycle. The event was attributed to the study treatment, as there was no fatigue at baseline. However, the patient died within 17 days after this event due to rapid progression of his malignant disease, so in retrospect the causality assessment for this event was questionable. Reversible grade 3 increases in alkaline phosphatase, AST, ALT and {gamma}-glutamyltransferase (GGT) were observed in one patient on the 4800 ng/kg dose level during cycle 1, and a very similar event was seen in one further patient treated with aviscumine 6400 ng/kg. Another patient on this dose level had a reversible grade 3 increase in GGT (Table 5). According to the investigator's judgement, these events were not related to the underlying malignancy, co-morbidity or co-medication, and thus qualified as DLTs. With two out of five patients on 6400 ng/kg having reversible grade 3 liver toxicity, this dose level was defined as being the MTD. As the incidence of DLT on this dose level exceeded the protocol-defined 20% cut-off for severe events and the dose level of 5600 ng/kg was safely administered in seven patients, the latter dose was defined as the recommended dose for phase II studies.


View this table:
[in this window]
[in a new window]
 
Table 5. Dose limiting events (all patients, cycle 1, related events only)

 
Response assessment
Tumor evaluation was carried out every other cycle. Twenty-nine patients were evaluated for response after cycle 2 and six patients were evaluated after cycle 4. The best overall response during the conduct of the trial per patient is shown in Table 6. Eleven patients achieved disease stabilization for two to eight cycles, but there were no minor, partial, complete or tumor marker responses.


View this table:
[in this window]
[in a new window]
 
Table 6. Best overall response (RECIST)

 
Pharmacokinetics
Aviscumine was found to have a short T1/2{alpha} of 13 min. The mean plasma concentration of aviscumine over time after the first i.v. administration is shown in Figure 2. Plasma levels >2 ng/ml (mean IC70 of aviscumine in cell lines) were achieved for only 2.3–7.5 h per infusion. At dose levels ≥3200 ng/kg, the Cmax and area under the concentration–time curve (AUC) increased in linear fashion with increasing dosages of the drug (Figure 3). Further details on the pharmacokinetic evaluation of aviscumine are shown in Table 7.



View larger version (16K):
[in this window]
[in a new window]
 
Figure 2. Mean plasma concentration of aviscumine (dose levels 3200–6400 ng/kg, cycle 1).

 


View larger version (9K):
[in this window]
[in a new window]
 
Figure 3. Correlation of Cmax versus dose of aviscumine (all dose levels, cycle 1).

 

View this table:
[in this window]
[in a new window]
 
Table 7. Pharmacokinetics of aviscumine (arithmetic mean and range, dose levels 3200–6400 ng/kg, cycle 1)

 
Immunological effects and antibody induction
Aviscumine stimulated the secretion of various cytokines. Cytokine releases were more frequent after the first infusion compared with the 11th infusion of the drug (Table 8). After the first infusion, all patients showed an increase of IL-1ß, -6, -10, IFN-{alpha} and/or TNF-{alpha} in plasma; after the 11th administration of the drug, five patients showed no stimulation. Baseline plasma cytokine levels prior to the first and the 11th administration of aviscumine were not different. IL-6 was the cytokine reaching the highest plasma levels and being stimulated most frequently. The release of cytokines in plasma was found to be independent of the administered dose of aviscumine, with the exception of TNF-{alpha}, which showed higher values after administration of low doses of the drug. The increase in cytokine plasma levels was not associated with disease stabilization (data not shown). Owing to the low patient numbers and different dose levels in this phase I trial and high interindividual variation, these findings are of unclear clinical relevance.


View this table:
[in this window]
[in a new window]
 
Table 8. Frequency of increase in cytokine plasma levels after administration of aviscumine

 
The i.v. administration of the recombinant protein induced the production of antibodies of immunoglobulin (Ig) M and IgG class. IgG antibody induction was seen with doses of aviscumine exceeding 3200 ng/kg. The clinical relevance of this antibody induction is unclear.


    Discussion
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Aviscumine is an innovative antineoplastic compound with interesting pharmacology and a unique mode of action. The recombinant heterodimer enters the cell via specific binding of the B-chain to CD75s, a membraneous ganglioside with unknown physiological function, which is frequently expressed in common solid tumors. This binding leads to internalization of the holoprotein. The A-chain of the drug catalizes the inactivation of ribosomes and thus inhibits translation and protein synthesis. In vitro, the drug was found to be very active against various cell lines, was more potent than many conventional agents, was also active in chemoresistant models and exhibited synergy with anticancer agents. In vivo studies confirmed the drug's broad antineoplastic activity. In addition, there was some evidence of immunomodulatory activity of the compound, including cytokine stimulation and quantitative effects on immune cell subpopulations. Furthermore, the drug was well tolerated in animals. Based on these findings, we designed the very first study for the i.v. administration of the recombinant protein in patients with advanced cancers refractory to conventional treatment.

The administration of a twice weekly 1-h i.v. infusion of aviscumine was found to be safe up to a dose of 5600 ng/kg body weight. The MTD was 6400 ng/kg, and was associated with reversible hepatotoxicity as dose-limiting event in two out of five patients. The toxicity profile in this clinical trial was quite acceptable, and according to the adverse events data it was concluded that the recommended dose for further clinical trials should be 5600 ng/kg.

Aviscumine was found to stimulate the immune system with a secretion of cytokines in plasma, independent of the dose of the drug administered. Furthermore, the recombinant drug induced anti-aviscumine antibodies in plasma in doses above 3200 ng/kg body weight. These findings are of unclear clinical relevance, and should be studied further in clinical trials.

The pharmacokinetic data revealed linear kinetics of the recombinant protein on the higher dose levels, ≥3200 ng/kg body weight. The initial half-life of the drug of 13 min indicates a fast degradation of the compound. This corresponds to in vitro studies which revealed that degradation could be expected due to a proteolytic process. Distribution volumes remained in a low range, especially not exceeding the patients weight for the extracellular body fluid (~18% of total body weight), indicating a lack of accumulation. The mean IC70 against the growth of tumor xenograft in vitro models was 2 ng/ml. This concentration could be exceeded during and after the infusion for a mean duration of 2.3–7.5 h with acceptable tolerability. This duration above plasma levels of 2 ng/ml and the short T1/2{alpha} of only 13 min led us to the conclusion that the 1-h infusion twice weekly is probably not the optimal dosing regimen with regard to the exposure of the tumor to adequate drug levels. Therefore, the investigation of prolonged infusion schedules seems to be more appropriate and warrants further investigation in clinical trials, which are already ongoing. Based on preclinical data and the observed toxicity profile of the agent, combination trials with conventional chemotherapeutic agents should also be considered.


    Acknowledgements
 
The authors wish to thank study nurses Bianca Wawzik and Claudine Bourdin, data managers Radoslaw Kowalski, Natalia Waldt and Filip Delaei, and clinical research associates Claudine Bourcier, Christophe Goris and Nathalie von Hove for their very valuable contribution to the trial. Parts of this research have been presented previously at the EORTC-NCI-AACR Conference on Molecular Targets and Cancer Therapeutics, Frankfurt, Germany, 2002, and the AACR-NCI-EORTC International Conference on Molecular Targets and Cancer Therapeutics, Boston, MA, USA, 2003.

Received for publication June 9, 2004. Accepted for publication July 9, 2004.


    References
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
1. Hajto T, Hostanska K, Gabius HJ. Modulatory potency of the ß-galactoside-specific lectin from mistletoe extract (Iscador) on the host defense system in vivo in rabbits and patients. Cancer Res 1989; 49: 4803–4808.[Abstract]

2. Janssen O, Scheffler A, Kabelitz D. In vitro effects of mistletoe extracts and mistletoe lectins. Cytotoxicity towards tumor cells due to the induction of programmed cell death (apoptosis). Arzneimittelforschung 1993; 43: 1221–1227.[Medline]

3. Büssing A, Suzart K, Bergmann J et al. Induction of apoptosis in human lymphocytes treated with Viscum album L. is mediated by mistletoe lectins. Cancer Lett 1996; 99: 59–72.[CrossRef][ISI][Medline]

4. Hostanska K, Hajto T, Weber K et al. A natural immunity-activating plant lectin, Viscum album agglutinin-I, induces apoptosis in human lymphocytes, monocytes, monotypic THP-1 cells and muring thymocytes. Nat Immun 1996; 15: 295–311.[Medline]

5. Möckel B, Schwarz T, Zinke H et al. Effects of mistletoe lectin I on human blood cell lines and peripheral blood cells. Arzneimittelfoschung 1997; 47 (II 10): 1145–1151.

6. Bantel H, Engels I, Voelter W et al. Mistletoe lectin activates caspase-8/FLICE independently of death receptor signaling and enhances anticancer drug-induced apoptosis. Cancer Res 1999; 59: 2083–2090.[Abstract/Free Full Text]

7. Langer M, Möckel B, Eck J et al. Site-specific mutagenesis of mistletoe lectin: the role of RIP activity in apoptosis. Biochem Biophys Res Commun 1999; 264: 944–948.[CrossRef][ISI][Medline]

8. Endo Y, Tsurugi K, Franz H. The site of action of the A-chain of mistletoe lectin I on eukaryotic ribosomes. The RNA N-glycosidase activity of the protein. FEBS Lett 1988; 231: 378–380.[CrossRef][ISI][Medline]

9. Eck J, Langer M, Möckel B et al. Cloning of the mistletoe lectin gene and characterization of the recombinant A-chain. Eur J Biochem 1999a; 264: 775–784.[Abstract/Free Full Text]

10. Eck J, Langer M, Möckel B et al. Characterization of recombinant and plant-derived mistletoe lectin and their B-chains. Eur J Biochem 1999b; 265: 788–797.[Abstract/Free Full Text]

11. Beuth J, Ko HL, Gabius HJ et al. Behaviour of lymphocyte subsets and expression of activation markers in response to immunotherapy with galactoside specific lectin from mistletoe in breast cancer patients. Clin Invest 1992; 49: 658–661.

12. Kleijnen J, Knipshild P. Mistletoe treatment for cancer. Review of controlled trials in humans. Phytomedicine 1994; 1: 255–260.

13. Kienle GS, Berrino F, Bussing A et al. Mistletoe in cancer–a systematic review on controlled trials. Eur J Med Rex 2003; 8: 109–119.

14. Weis J, Bartsch HH, Hennies F et al. Complementary medicine in cancer patients: demand, patients' attitudes and psychological beliefs. Onkologie 1998; 21: 144–149.[CrossRef][ISI]

15. Ernst E, Cassileth B. How useful are unconventional cancer treatments? Eur J Cancer 1999; 35: 1608–1613.[CrossRef][ISI][Medline]

16. Hajto T, Hostanska K, Weber K et al. Effect of recombinant lectin Viscum album agglutinin on the secretion of interleukin-12 in cultured human peripheral blood mononuclear cells and on NK-cell-mediated cytotoxicity of rat splenocytes in vitro and in vivo. Nat Immun 1998; 16: 34–46.[CrossRef][ISI][Medline]

17. Schumacher U, Feldhaus S, Mengs U. Recombinant mistletoe lectin (rML) is successful in treating human ovarian cancer cells transplanted into severe combined immunodeficient (SCID) mice. Cancer Lett 2000; 150: 171–175.[CrossRef][ISI][Medline]

18. Schaffrath B, Mengs U, Schwarz T et al. Anticancer activity of rViscumin (recombinant mistletoe lectin) in tumor colonization models with immunocompetent mice. Anticancer Res 2001; 21: 3981–3988.[ISI][Medline]

19. Hostanska K, Vuong V, Rocha S et al. Recombinant mistletoe lectin induces p53-independent apoptosis in tumour cells and cooperates with ionising radiation. Br J Cancer 2003; 88: 1785–1792.[CrossRef][ISI][Medline]

20. Müthing J, Burg M, Möckel B et al. Preferential binding of the anticancer drug rViscumin (recombinant mistletoe lectin) to terminally {alpha}2-6-sialylated gangliosides. Glycobiology 2002; 12: 485–497.[Abstract/Free Full Text]

21. Müthing J, Meisen I, Bulau P et al. Mistletoe lecting I is a sialic acid-specific lectin with strict preference to gangliosides with terminal neu5 Ac{alpha}2-6Galß1-4GlcNAc residues. Biochemistry 2004; 43: 2996–3007.[CrossRef][ISI][Medline]

22. O'Quigley J, Shen LZ. Continual Reassessment Method: a likelihood approach. Biometrics 1996; 52: 395–405.

23. O'Quigley J, Reiner E. A stopping rule for the Continual Reassessment Method. Biometrika 1998; 85: 741–748.[Abstract]

24. Therasse P, Arbuck SG, Eisenhauer EA et al. New guidelines to evaluate the response to treatment in solid tumors. European Organization for Research and Treatment of Cancer, National Cancer Institute of the United States, National Cancer Institute of Canada. J Natl Cancer Inst 2000; 92: 205–206.[Abstract/Free Full Text]

25. Adler M, Langer M, Witthohn K et al. Detection of rViscumin in plasma samples by immuno-PCR. Biochem Biophys Res Comm 2003; 300: 757–763.[CrossRef][ISI][Medline]





This Article
Abstract
Full Text (PDF)
E-letters: Submit a response
Alert me when this article is cited
Alert me when E-letters are posted
Alert me if a correction is posted
Services
Email this article to a friend
Similar articles in this journal
Similar articles in ISI Web of Science
Similar articles in PubMed
Alert me to new issues of the journal
Add to My Personal Archive
Download to citation manager
Search for citing articles in:
ISI Web of Science (4)
Disclaimer
Request Permissions
Google Scholar
Articles by Schöffski, P.
Articles by Twelves, C.
PubMed
PubMed Citation
Articles by Schöffski, P.
Articles by Twelves, C.