In vitro and in vivo antitumor activity of the novel trinuclear platinum complex BBR 3464 in neuroblastoma.

PURPOSE: BBR 3464 is a promising new trinuclear platinum complex that has been shown to circumvent the resistance to cisplatin in a panel of tumor cell lines and xenografts with acquired or intrinsic resistance to cisplatin. The in vitro and in vivo antitumor activity of BBR 3464 was evaluated and compared with that of cisplatin in neuroblastoma. METHODS: In in vitro studies, the short- and long-term cytotoxicities, cell cycle perturbations, the ability to induce apoptosis, the intracellular platinum accumulation and DNA platination were evaluated in three neuroblastoma cell lines exposed to appropriate drug concentrations for 1 h. In in vivo studies, BBR 3464 was administered i.v. at doses of 0.30 and 0.35 mg/kg three times at intervals of 4 days (q4dx3), and cisplatin was administered i.v. according to two different schedules (at 2 and 4 mg/kg three times at intervals of 4 days and at 6 and 12 mg/kg as single doses). RESULTS: In a short-term growth inhibition assay, BBR 3464 was shown to be up to 100-fold more potent than cisplatin and it was even more potent in a clonogenic assay. The difference in the antitumor effect of BBR 3464 on the different cell lines was evident in both assays, while cisplatin exerted a comparable antitumor activity in all lines tested. Cell cycle analysis demonstrated a longer-lasting block in G2/M phase induced by BBR 3464 without the early S phase accumulation induced by cisplatin. The higher potency of BBR 3464 appeared to be unrelated to the induction of apoptosis, that was lower or at most comparable to cisplatin. Cellular platinum accumulation and platinum-DNA adduct formation following BBR 3464 exposure was higher than following cisplatin exposure. These differences may have resulted from a different mechanism of action and may explain the lack of cross-resistance with cisplatin. In xenografts of neuroblastoma, BBR 3464 was confirmed to be very potent as compared to cisplatin (MTD 0.35 mg/kg and 4 mg/kg for BBR 3464 and cisplatin, respectively). The efficacy of BBR 3464 was superior to that of cisplatin when both drugs were administered on a fractionated schedule (q4dx3), while BBR 3464 appeared equally active to 12 mg/kg cisplatin administered as a single dose. CONCLUSIONS: Our findings indicate that BBR 3464 has a definite antitumor effect in neuroblastoma lines and may be a candidate for early clinical trials in children with neuroblastoma.

The novel trinuclear platinum complex BBR3464 induces a cellular response different from cisplatin.

BBR3464 is a new platinum-based drug non cross-resistant with cisplatin. To characterise the cellular basis of BBR3464 cytotoxicity as opposed to cisplatin, we performed a comparative study of the two drugs in cisplatin-resistant neuroblastoma and astrocytoma cells. In both model systems, BBR3464 proved to be more potent than cisplatin and was able to overcome cisplatin resistance. The higher potency exhibited by BBR3464 correlated with an increased cellular platinum accumulation and DNA-adduct formation. At equitoxic doses, BBR3464 induced apoptosis to a lesser extent than cisplatin and failed to overcome the decreased susceptibility to cisplatin-induced apoptosis in cisplatin-resistant cells. Cell cycle analysis showed a dose-dependent G2/M arrest by BBR3464. In astrocytoma cells, cisplatin treatment resulted in the upregulation of p53, p21 and bax, while only p21 induction was observed after BBR3464 treatment. In cisplatin-resistant cells, the reduced sensitivity to cisplatin paralleled a resistance to the induction of p53/p21 pathway by cisplatin, while the same doses of BBR3464 induced p21 to a similar extent in the resistant cells as in the parental cells. In conclusion, BBR3464 induces a cellular response that is different from cisplatin, supporting the view that the two drugs act through different mechanisms. Our data indicate that BBR3464 may be a promising agent in the treatment of tumours unresponsive to cisplatin and with a non-functional p53.

Influence of cisplatin and doxorubicin on 125I-meta-iodobenzylguanidine uptake in human neuroblastoma cell lines.

The combination of 131I-meta-iodobenzylguanidine (MIBG) with chemotherapy has recently been employed in the treatment of advanced stage neuroblastoma with encouraging results. However, the mechanisms underlying the interaction between these two different modalities of treatment have not yet been explored. In this study, human neuroblastoma cell lines pretreatment with cisplatin and doxorubicin increased cellular 125I-MIBG accumulation in a dose-dependent manner. Cell cycle analysis showed that increased 125I-MIBG accumulation correlated with the drug-induced G2/M phase block. Northern blot analysis demonstrated an increase in gene expression of the noradrenaline transporter induced by doxorubicin, but not by cisplatin treatment. Increased 125I-MIBG accumulation was also observed in murine xenografts of the human neuroblastoma cell line SK-N-DZ or BE(2)M17 treated intraperitoneally (i.p.) with cisplatin or doxorubicin, respectively. These results suggest that the combination of 131I-MIBG and these drugs could selectively increase radiation doses delivered to neuroblastomas.

Cranial irradiation and permeability of blood-brain barrier to cytosine arabinoside in children with acute leukemia.

Cranial irradiation (CI) is an effective way to prevent central nervous system (CNS) leukemia in children with acute leukemia (AL). However, it is still unclear whether the antileukemic effect of CI is mediated by alteration of blood-brain barrier (BBB) permeability and consequent increased levels of systemically administered drugs or whether it simply results from a direct cytolytic effect on leukemic cells in the meninges at diagnosis. We evaluated the influence of CI on BBB permeability to 1-beta-D-arabinofuranosylcytosine (ara-C) in 23 children with AL undergoing CI for CNS leukemia prophylaxis. CI was administered at 18 Gy (16 patients) and 24 Gy (7 patients). ara-C levels were measured in cerebrospinal fluid (CSF) and plasma before, during, and after CI. Two doses were evaluated: 75 mg/m2/day (12 patients) and 480 mg/m2/day (11 patients). CSF and plasma ara-C levels were measured when steady state was achieved. CSF:plasma ratios, obtained before, during, and after CI, were compared by an ANOVA model for repeated measures and by Tukey's test. At the 75-mg/m2/day dose, the mean values of ara-C CSF:plasma ratios before, during, and after CI were 0.20, 0.27, and 0.27, respectively. At the 480-mg/m2/day dose, the mean CSF:plasma ratios before, during, and after CI were 0.09, 0.12, and 0.13, respectively. No significant differences were observed when CSF:plasma ratios were compared before and during, during and after, and before and after CI. Our results indicate that CI at the doses used for CNS prophylaxis in AL does not significantly alter the BBB as far as ara-C is concerned.

Coordinate regulation of STAT signaling and c-fos expression by the tyrosine phosphatase SHP-2.

The src homology 2 (SH2) domain-containing protein-tyrosine phosphatase SHP-2 has been implicated as an important positive regulator of several mitogenic signaling pathways. SHP-2 has more recently been shown to be tyrosine phosphorylated and recruited to the gp130 component of the ciliary neurotrophic factor (CNTF) receptor complex upon stimulation with CNTF. CNTF does not, however, have a proliferative effect on responsive cells, but rather enhances the survival and differentiation of sympathetic, motor, and sensory neurons. In this study, expression of an interfering mutant of SHP-2 in the neuroblastoma cell line NBFL increased CNTF induction of a vasoactive intestinal peptide (VIP) reporter gene, and in cultures of sympathetic neurons, it resulted in an up-regulation of endogenous VIP and substance P (SP) gene expression. Members of the CNTF family of cytokines transmit their signal by activating signaling pathways involving both STAT and Fos-Jun transcription factors. In CNTF-stimulated NBFL cells that constitutively express the SHP-2 interfering mutant, there was increased and prolonged formation of STAT/DNA complexes, but decreased AP-1 binding activity, that mirrored a down-regulation of c-fos expression both at the mRNA and protein level. Taken together, these data indicate that SHP-2 has dual and opposing roles in a signaling cascade triggered by the same ligand, as illustrated by its ability to differentially regulate the levels of activity of both STAT and AP-1 transcription factors.

The protein tyrosine phosphatase SHP-2 is expressed in glial and neuronal progenitor cells, postmitotic neurons and reactive astrocytes.

In this study we examined the distribution and developmental profile of the src homology 2 (SH2) domain-containing protein tyrosine phosphatase SHP-2 in the mouse brain. We found that SHP-2 is present in both mitotically active and postmitotic cells in the forebrains of embryonic day 12 (E12) mice. In a developmental study extending from embryonic day 12 to adulthood, Western blotting analysis demonstrated equivalent levels of SHP-2 protein at all of the ages examined. Expression of SHP-2 paralleled the level of enzymatic activity at the different developmental periods. In the adult brain SHP-2 was restricted to diverse classes of neurons, while the majority of glial cells did not express detectable levels of protein. However, reactive astrocytes in response to an ischemic brain injury showed SHP-2 immunolabelling. Our data suggest that SHP-2 may play a role in pathways of neuronal and glial progenitor cells, in a broad spectrum of neuronal responses in the adult brain and in the gliotic response to the injury.

The protein tyrosine phosphatase SHP-2 negatively regulates ciliary neurotrophic factor induction of gene expression.

Ciliary neurotrophic factor, along with other neuropoietic cytokines, signals through the shared receptor subunit gp130 [1-3], leading to the tyrosine phosphorylation of a number of substrates [4,5], including the transcription factors STAT1 and STAT3 and the protein tyrosine phosphatase SHP-2 [6,7] [8]. SHP-2 (also known as PTP1D, SHPTP2, Syp and PTP2C) is a positive regulatory molecule required for the activation of the mitogen-activated protein kinase pathway and the stimulation of gene expression in response to epidermal growth factor, insulin and platelet-derived growth factor stimulation [9-11]. We have previously shown that cytokines that signal via the gp130 receptor subunit activate transcription of the vasoactive intestinal peptide (VIP) gene through a 180 bp cytokine response element (CyRE) [12,13]. To characterize the role of SHP-2 in the regulation of gp130-stimulated gene expression, we examined the regulation of the VIP CyRE in two systems that prevented ligand-dependent SHP-2 phosphorylation. Inhibition of SHP-2, either by mutating the tyrosine residue in gp130 that mediates the SHP-2 interaction, or by expression of dominant-negative SHP-2, resulted in dramatic increases in gp130-dependent gene expression, through the VIP CyRE and more specifically through multimerized STAT-binding sites. These data suggest that SHP-2 has a negative role in gp130 signaling by modulating STAT-mediated transcriptional activation.

A human neuroblastoma xenograft model for 125-I-metaiodobenzylguanidine biodistribution studies.

We developed an animal model to evaluate the 125-I-metaiodobenzylguanidine (125-I-mIBG) biodistribution in tumor bearing mice. Six weeks old nude-atimic mice were subcutaneously injected with 30 x 10(6) cells of the human neuroblastoma (NB) cell line SH-SY5Y. TE-671, a rhabdomyosarcoma cell line, was used as a control tumor without a specific mIBG uptake mechanism. In order to prevent possible tumor rejection mediated by NK activity the anti asialo GM1 antiserum was administered intraperitoneally once a week for 4 weeks. The maximum anti asialo mediated effect was obtained by administering the first dose the same day as the cell implant. In this group of animals by 9 weeks 98% of mice had a measurable tumor. We have utilized this model to evaluate the biodistribution of 125-I-mIBG given as two different formulations: standard preparation with a specific activity of 84 mCi/mg and the no carrier added (n.c.a) formulation with a specific activity of approximately 8,000 mCi/mg. Our preliminary results indicate that the biodistribution of the two different formulations in the various organs are similar. Therefore it appears that n.c.a. mIBG should not cause an increased toxicity in possible normal target organs such as heart or adrenals. Additional experiments will be performed in this model to ascertain if there is a potential advantage of the clinical use of n.c.a. mIBG over the standard preparation.

In vitro and in vivo studies with no-carrier added radioiodinated MIBG.

An important drawback of radioiodinated MIBG prepared by isotopic exchange (maximum specific activity: 100 Ci/mmol) is that a considerable amount of carrier is present in the "ready to use" radiopharmaceutical, thus affecting MIBG uptake in the target cells. To improve the therapeutic utility of MIBG an interesting method was developed for the no-carrier-added (n.c.a.) radioiodination of MIBG via a halodesilylation reaction using 3-trimethylsilybenzylguanidine (TMSBG) as precursor. In order to investigate its possible clinical usefulness, n.c.a. [125I]- and [131I]MIBG was prepared according this procedure. HPLC analysis of the labeling mixture showed that the labeling yield reached values greater than 98% within 20 minutes at 70 degrees C. HPLC fraction containing [131I]MIBG was then recovered and tetrahydrofurane was removed under nitrogen stream. Biodistribution studies were performed in BALB/c normal mice and compared with those obtained using [131I]MIBG prepared by isotopic exchange. Preliminary results showed that at 4 hours myocardial uptake was significantly higher than that of c.a. MIBG (20.3 +/- 0.4 %ID/g). Adrenal uptake gradually increased over time; at 24 hours adrenal uptake was almost twofold that a 1 hour. MIBG uptake was also studied in the human neuroblastoma cell line SH-SY5Y which has a specific uptake system for MIBG. The kinetic parameters for MIBG in this line are: km = 0.27 +/- 0.03 microM, Vmax = 39.5 + 1.8 pmol/10(6) cell/10 min. Preliminary data from in vitro binding studies showed that significantly higher levels of specifically incorporated radioactivity can be achieved using n.c.a. [125I]MIBG instead of c.a. [125I]MIBG.

Cytotoxicity of paclitaxel and docetaxel in human neuroblastoma cell lines.

Taxanes are an important new class of anticancer agents that inhibit cell division by the unique mechanism of increasing the rate of microtubule assembly and preventing microtubule depolymerisation. Using the colony inhibition assay, we compared the cytotoxicity of paclitaxel and docetaxel in three human neuroblastoma (NB) cell lines, SH-SY5Y, BE(2)M17 and CHP100. Different exposure times (3, 6, 12, 24, 48 and 72 h) and different concentrations ranging from 0.1 nM to 10 microM were tested. Both paclitaxel and docetaxel show antineoplastic activity in human NB cell lines. Taxanes' antitumour activity varied among the different cell lines, CHP100 being the most sensitive and SH-SY5Y the least sensitive. Paclitaxel cytotoxicity appears schedule-dependent, with marked cell kill observed only for exposures of 24 h or longer. Docetaxel cytotoxicity was dependent upon prolonged exposure only in the SH-SY5Y cell line, while an exposure time of 3-6 h resulted in exponential cell kill in the other two cell lines. Docetaxel was more cytotoxic than paclitaxel with a mean ratio of (paclitaxel/docetaxel) IC50 values ranging from 2 to 11. For both taxanes, we observed good correlation between cytotoxic effect and percentage of cells blocked in G2/M phase. A cytotoxic effect occurred at concentrations comparable with those achieved in the plasma of patients treated with these agents in initial clinical trials. The full potential of prolonged infusion or repeated daily administrations of taxanes should be explored in clinical studies, and responses to taxanes in neuroblastoma should be assessed in paediatric phase II studies.

Release mechanisms of [125I]meta-iodobenzylguanidine in neuroblastoma cells: evidence of a carrier-mediated efflux.

[131I]metaiodobenzylguanidine ([131I]MIBG) is selectively taken up and stored by tumours derived from the neural crest, and is used for diagnosis and treatment of neuroblastoma (NB). The antitumoral effect of [131I]MIBG is closely related to the intracellular level of the radiopharmaceutical compound, which is dependent on uptake and storage/release mechanisms. While MIBG uptake is well characterised, storage and release mechanisms are still controversial. In order to better characterise [125I]MIBG release mechanisms, we studied the basal and stimulated efflux of [125I]MIBG in the human NB cell line, SH-SY5Y, preloaded with 0.1 microM [125I]MIBG for 1 h. We found that [125I]MIBG basal efflux is highly temperature-dependent, that [125I]MIBG release, induced by cell depolarisation with high potassium, is mainly calcium-independent, and induced by exchange with cold MIBG or noradrenaline, inversion of the sodium gradient across the cell membrane by veratridine by substitution of sodium chloride with equimolar concentration of lithium chloride. The exposure of NB cells to imipramine, an Uptake-1 inhibitor, also produces a net stimulatory effect on [125I]MIBG release. However, when used in association with other releasing stimuli, such as higher levels of intracellular sodium or external agonists, imipramine abolishes the consequent increase of [125I]MIBG release. Our findings suggest that stimulated [125I]MIBG release is mediated by a carrier, most probably the uptake carrier working in a reverse mode, while a minimal fraction of [125I]MIBG is released by an exocytotic mechanism.

Clinical pharmacokinetics of carboplatin in children.

The present study was undertaken to evaluate in children the plasma pharmacokinetics of free carboplatin given at different doses and schedules and to evaluate the inter- and intrapatient variability and the possible influence of schedule on drug exposure. A total of 35 children (age range, 1-17 years) with malignant tumors were studied. All patients had normal renal function (creatinine clearance corrected for surface body area, above 70 ml min-1 m-2; range, 71-151 ml min-1 m-2) and none had renal involvement by malignancy. Carboplatin was given at the following doses and schedules: 175, 400, 500, and 600 mg/m2 given as as a 1-h infusion; 1,200 mg/m2 divided into equal doses and infused over 1 h on 2 consecutive days; and 875 and 1,200 mg/m2 given as a 5-day continuous infusion. A total of 57 courses were studied. Carboplatin levels in plasma ultrafiltrate (UF) samples were measured both by high-performance liquid chromatography and by atomic absorption spectrophotometry. Following a 1-h infusion, carboplatin free plasma levels decayed biphasically; the disappearance half-lives, total body clearance, and apparent volume of distribution were similar for different doses. In children with normal renal function as defined by creatinemia and blood urea nitrogen (BUN) and creatinine clearance, we found at each dose studied a limited interpatient variability of the peak plasma concentration (Cmax) and the area under the concentration-time curve (AUC) and a linear correlation between the dose and both Cmax (r = 0.95) and AUC (r = 0.97). The mean value +/- SD for the dose-normalized AUC was 13 +/- 2 min m2 l-1 (n = 57).2+ The administration schedule does not seem to influence drug exposure, since prolonged i.v. infusion or bolus administration of 1,200 mg/m2 achieved a similar AUC (13.78 +/- 2.90 and 15.05 +/- 1.44 mg ml-1 min, respectively). In the nine children studied during subsequent courses a limited interpatient variability was observed and no correlation (r = 0.035) was found between AUC and subsequent courses by a multivariate analysis of dose, AUC, and course number. The pharmacokinetic parameters were similar to those previously reported in adults; however, a weak correlation (r = 0.52, P = 0.03) between carboplatin total body clearance and creatinine clearance varying within the normal range was observed.(ABSTRACT TRUNCATED AT 400 WORDS)

Uptake and storage of m-iodobenzylguanidine are frequent neuronal functions of human neuroblastoma cell lines.

The mechanisms of the uptake and release of m-iodobenzylguanidine (MIGB) have been studied in 5 neuroblastoma (NB) cell lines and in 4 clonal NB sublines with a homogeneous phenotype. A specific uptake system for MIBG was found in 8 of 9 NB cell lines or subpopulations. The uptake was characterized by temperature dependency, high affinity, saturability, sodium dependency, and imipramine sensitivity. The majority of NB cell lines that possessed a specific uptake system for MIBG were also able to efficiently store the incorporated drug. However, 3 NB cell lines were identified without the ability to retain high levels of MIBG, despite the presence of a specific uptake system. We also report that a clonal subline, SH-EP1, which has a nonneuroblastic phenotype, failed both MIBG uptake and retention. Conversely, the parental cell line, SK-N-SH, and the neuroblastic subline SH-SY5Y possessed both a specific uptake system and the ability to store MIBG. In addition, the induction of neuronal differentiation with retinoic acid increased the velocity of uptake and the storage efficiency for MIBG in the clonal subline SH-SY5Y. We conclude that MIBG uptake and storage should be considered to be frequent but independent neuronal functions of human NB cells.

Cerebrospinal fluid pharmacokinetics of carboplatin in children with brain tumors.

The pharmacokinetics of carboplatin in cerebrospinal fluid (CSF) and plasma was studied in five children with brain tumors (four medulloblastomas and one ependimoblastoma) who underwent preirradiation treatment with carboplatin. Carboplatin pharmacokinetics was studied following the administration of 600 mg/m2 as a 1-h infusion. Four children were treated a few weeks after surgery, whereas one child with an unresectable tumor was treated prior to surgery. All patients had a ventricular-peritoneal CSF shunt connected to a subcutaneous reservoir. Total platinum and free carboplatin were measured. The mean AUC values for free carboplatin in CSF and plasma were 2.29 +/- 1.20 and 8.18 +/- 1.27 mg ml-1 min, respectively. The mean ratio of CSF AUC to plasma AUC was 0.28 (range, 0.17-0.46). Both plasma peak levels and AUC values showed limited interpatient variability. On the other hand, carboplatin levels in CSF showed substantial interpatient variability, with a greater than 5-fold difference in peak levels and a 3-fold difference in AUC values being recorded. The interpatient difference in CSF pharmacokinetics may have been related at least in part to the different anatomical alterations induced by the surgical procedures or by the presence of a large tumor mass. In the four evaluable patients exhibiting macroscopic residual tumor, we observed one complete remission (CR) and two partial remissions (PR) following two cycles that consisted of two doses of 600 mg/m2 carboplatin given on 2 consecutive days (total dose, 1200 mg/m2) and were separated by a 1-month interval. These results may give some indication as to the optimal dose and schedule for carboplatin administration in the treatment of primitive neuroectodermic tumors (PNET).

Biology of metaiodobenzylguanidine interactions with human neuroblastoma cells.

[131I]Metaiodobenzylguanidine (131I-MIBG) is selectively taken up and stored by tumors derived from the neural crest and is utilized in the diagnosis and treatment of neuroblastoma (NB). Variable MIBG uptake has been observed, although the underlying mechanisms are not known. We have studied the uptake kinetics of 125I-MIBG and uptake characteristics of NB cell clones with different phenotypes (SH-SY5Y and SH-EP1, the neuroblastic and the substrate-adherent sublines of SK-N-SH respectively, BE(2)-M17 and LA-N-1n with neuroblastic phenotype). We have been able to correlate the MIBG uptake with the neuroblastic phenotype: a specific uptake system satisfying all the characteristics of the neuronal uptake-1 (temperature dependency, sodium dependency, high affinity, saturability and imipramine sensitivity) was observed in all the neuroblastic sublines. In contrast, MIBG accumulation was a passive diffusion phenomenon in the substrate-adherent clone SH-EP1. In addition, terminal neuronal differentiation induced in SH-SY5Y by retinoic acid caused a marked increase of the uptake and retention of MIBG. Our findings may be pertinent to an understanding of the variability of the MIBG uptake in vivo.

Cranial irradiation and cerebrospinal fluid levels of 6-mercaptopurine in children with acute leukemia.

We measured 6-mercaptopurine levels in the cerebrospinal fluid and plasma of 15 children undergoing treatment for acute leukemia. Plasma and cerebrospinal fluid samples obtained by lumbar puncture were collected before, during, and after cranial irradiation in order to evaluate a possible change in blood-brain barrier permeability to orally administered 6-mercaptopurine. Considerable interpatient variability has been observed in both plasma and cerebrospinal fluid 6-mercaptopurine levels. No statistical differences in the 6-mercaptopurine cerebrospinal fluid levels under the three different conditions could be detected. Our data suggest that cranial irradiation does not significantly influence the cerebrospinal fluid levels.

Modification of the 6-hydroxydopamine technique for the correct determination of superoxide dismutase.

Heikkila & Cabbat (Anal. Biochem. 75, 356-362 (1976] have proposed the autoxidation of the 6-hydroxydopamine as a method to test superoxide dismutase activity in biological samples. This method has several advantages but in some instances leads to incorrect determinations. We present here a necessary modification of the method to avoid bias.