Restoring p53 function by antagonizing its interaction with the negative regulator MDM2 is an appealing nongenotoxic approach to treating tumors with wild-type p53. Mutational inactivation of p53 is rare in neuroblastoma tumors at diagnosis and occurs in only a subset of multidrug-resistant neuroblastomas.
The antiproliferative and cytotoxic effect of nutlin-3, a small-molecule MDM2 antagonist, was examined in chemosensitive (UKF-NB-3) and matched chemoresistant neuroblastoma cells with wild-type p53 (UKF-NB-3 r DOX 20 ) or with mutant p53 (UKF-NB-3 r VCR 10 ). Activation of the p53 pathway was assessed by expression analysis of p53 target genes, flow cytometric cell cycle analysis, and apoptosis assays. Mice with established chemoresistant tumor xenografts were treated orally with nutlin-3 or vehicle control (n = 5–10 mice per group) and were used to evaluate effects on tumor growth, p53 pathway activity, and metastatic tumor burden. All statistical tests were two-sided.
Nutlin-3 induced a similar activation of the p53 pathway in UKF-NB-3 and UKF-NB-3 r DOX 20 cells, as evidenced by increased expression of p53 target genes, G 1 cell cycle arrest, and induction of apoptosis. No such response was observed in UKF-NB-3 r VCR 10 cells with mutant p53. Oral administration of nutlin-3 to UKF-NB-3 r DOX 20 xenograft–bearing mice led to inhibition of primary tumor growth (mean tumor volume after 3 weeks of treatment, nutlin-3– vs vehicle-treated mice: 772 vs 1661 mm 3 , difference = 890 mm 3 , 95% confidence interval = 469 to 1311 mm 3 , P < .001), p53 pathway activation, and reduction in the extent of metastatic disease. The growth of UKF-NB-3 r VCR 10 xenografts was unaffected by nutlin-3.
Nutlin-3 activates the p53 pathway and suppresses tumor growth in this model system of chemoresistant neuroblastoma, provided that wild-type p53 is present.
Survival rates of patients with high-risk neuroblastoma remain low because of resistance to conventional chemotherapeutic agents. p53 is a powerful tumor suppressor protein. Mutations in p53 that lead to its inactivation are rare in neuroblastoma. MDM2, a natural inhibitor of p53 that may have an increased activity in neuroblastoma, is a potential therapeutic target.
In vitro and in vivo analysis of the effects of the small-molecule MDM2 antagonist nutlin-3 on p53 activity and tumor growth in an established model system of multidrug-resistant neuroblastoma.
Nutlin-3 activated the p53 pathway, irrespective of sensitivity to chemotherapeutic drugs, in neuroblastoma cells with wild-type p53 but not in cells with mutant p53. Oral administration of nutlin-3 as a single agent reduced tumor growth and metastasis in the presence of wild-type p53.
Nutlin-3 may be a viable treatment option for advanced-stage and chemoresistant neuroblastoma with wild-type p53.
It is possible that the findings in this model system cannot be generalized to all cases of chemoresistant neuroblastoma with wild-type p53.
From the Editors
Neuroblastoma is the most common solid extracranial malignancy of childhood ( 1 ). Approximately half of all neuroblastoma patients are diagnosed with high-risk disease, which is associated with an overall survival rate of less than 40% ( 1 ). Acquisition of drug resistance is primarily responsible for treatment failure in these patients, because many tumors respond well to initial chemotherapy but eventually progress as intractable disease ( 2 ). Several factors are thought to contribute to the emergence of multidrug resistance in neuroblastoma. Loss of function of the gatekeeper protein p53 has been shown to confer a broad multidrug-resistant phenotype in neuroblastoma cells ( 3 , 4 ). Accordingly, increased frequencies of mutations in the TP53 gene as well as aberrations that result in inappropriately increased activity of the p53 inhibitor MDM2 have been observed in neuroblastoma cell lines that were established from patients at relapse ( 5 ). Elevated expression of drug efflux pumps (eg, multidrug resistance–associated protein 1 and P-glycoprotein) ( 6–9 ) and high expression of detoxifying enzymes (such as glutathione synthetic and utilization enzymes) ( 10 , 11 ) have been implicated as a second group of mechanisms by which neuroblastoma cells evade therapeutic intervention. An overall imbalance between proapoptotic and survival signals might also give rise to chemotherapy resistance by diverting cell fate away from drug-induced apoptosis. Examples of mechanisms resulting in such a reduction in global apoptotic sensitivity in treatment-refractory neuroblastoma cells include activation of the phosphatidylinositol 3′-kinase/Akt survival pathway ( 12–14 ), overexpression of the antiapoptotic proteins survivin ( 15 , 16 ) and Bcl-2 ( 17 , 18 ), and decreased expression of death receptors such as Fas ( 19 ) and TRAILR ( 20 ). Specific defects in apoptotic effector pathways as a mechanism of drug resistance in neuroblastoma cells seem to converge at the death-inducing signaling complex, which governs the execution of the apoptotic program following engagement of cell surface death receptors ( 21 ).
Although the occurrence of p53 mutations inevitably argues for development of p53-independent therapeutic strategies or approaches that can restore the function of mutant p53, all other mechanisms of drug resistance outlined above remain amenable, in principle, to therapeutic exploitation of the p53 pathway. This notion is supported by the different modes of action of p53, with its principal ability to activate the mitochondrial apoptotic cascade providing a means for bypassing defects in death-inducing signaling complex signaling, and by the concept that induction of apoptosis relies on crossing a threshold without needing to correct for specific hyperactive survival pathways. Therapeutic utility of p53 reactivation for cancer is derived from the ontogenic requirement in tumor cells for apoptosis suppression, imposed by the stressed state of malignant cells, whereas normal cells are not sensitized to undergo apoptosis in response to p53-inducing stimuli. Several genetic studies in mouse models have demonstrated that inactivation of p53 is not only critical for tumor development but also for the maintenance of established tumors, suggesting that tumors are in a permanent state of addiction to the loss of p53 function and thus pointing to p53 as an attractive target for therapeutic intervention ( 22–24 ).
Nutlins are a class of cis -imidazoline derivatives that selectively disrupt the interaction between p53 and MDM2 in an enantiomer-specific manner ( 25 ). The release of p53 from negative control by MDM2 results in stabilization of p53 and direct activation of the p53 pathway, avoiding the nonspecific genotoxic damage inflicted by many classic cytotoxic drugs. We have previously reported that selective antagonism of MDM2 by nutlin-3, the best-studied member of the nutlin family, is highly effective in unleashing the antitumor potential of wild-type p53 in a panel of neuroblastoma cell lines, resulting in G 1 cell cycle arrest, apoptosis, premature cellular senescence, and neuronal differentiation ( 26 ). However, preclinical validation of the therapeutic benefit offered by experimental cancer therapeutics, such as nutlin-3, should be performed in a setting in which conventional drugs have failed and new treatment modalities are most urgently needed. To our knowledge, no studies have hitherto directly addressed whether activation of p53 by targeted inhibition of MDM2 is influenced by the acquisition of a multidrug-resistant phenotype. This study was undertaken to investigate the in vitro and in vivo antitumor activity of nutlin-3 in a model system of multidrug-resistant neuroblastoma that allows for study of treatment effects in relationship to p53 status.
Materials and Methods
Cells and Drug Treatment
The MYCN -amplified cell line UKF-NB-3 was derived from bone marrow metastases of a patient with stage IV neuroblastoma who had relapsed. Chemoresistant sublines were established by long-term incubation of UKF-NB-3 cells with increasing concentrations of doxorubicin or vincristine and maintained in the presence of 20 ng/mL doxorubicin (UKF-NB-3 r DOX 20 ) or 10 ng/mL vincristine (UKF-NB-3 r VCR 10 ), as described previously ( 27 ). Solutions of doxorubicin and vincristine (both from Sigma, St Louis, MO) were prepared in water according to the manufacturer’s instructions. Cells were cultured in RPMI 1640 medium supplemented with 15% fetal calf serum, 2 mM glutamine, 100 IU/mL penicillin, and 100 μg/mL streptomycin (medium and supplements from Gibco–Invitrogen, Merelbeke, Belgium) in a humidified atmosphere with 5% CO 2 at 37°C. The identity of the cell lines was verified by using array comparative genomic hybridization and short tandem repeat genotyping. Etoposide was purchased from Sigma and used according to the manufacturer's instructions. Nutlin-3 (Cayman Chemical, Ann Arbor, MI) was dissolved in absolute ethanol and stored as a 10-mM stock solution in 100-μL aliquots at −20°C. Cells were exposed to various concentrations of nutlin-3 as indicated, with the final ethanol concentration kept constant in each experiment.
Analysis of Cell Viability
UKF-NB-3, UKF-NB-3 r DOX 20 , and UKF-NB-3 r VCR 10 cells were seeded in duplicate wells of a 96-well plate (10 4 cells per well), incubated for 6 hours to allow them to adhere to the surface, and treated with nutlin-3 (at 0, 2, 4, 8, 16, or 32 μM) for 24, 48, or 72 hours. Cell viability was determined by using the CellTiter-Glo Luminescent Cell Viability Assay (Promega, Madison, WI). Each condition was tested in three independent experiments. Sensitivity to conventional chemotherapeutic drugs was investigated using a modified 3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl-2H-tetrazolium bromide (MTT) assay, as described previously ( 28 ).
Real-Time Quantitative Reverse Transcription–Polymerase Chain Reaction
Total RNA was extracted from UKF-NB-3, UKF-NB-3 r DOX 20 , and UKF-NB-3 r VCR 10 cells (2 × 10 6 cells) that had been treated with 16 μM nutlin-3 or vehicle control (ethanol) for 24 hours by using an RNeasy Mini Kit (Qiagen, Hilden, Germany), with RNase-free DNase I treatment performed on an RNA extraction spin column. After treatment with RQ1 DNase (Promega) in solution, first-strand cDNA was synthesized from 2 μg of total RNA using an iScript cDNA synthesis kit (Bio-Rad, Hercules, CA). Relative mRNA expression levels were determined by using an optimized two-step SYBR Green I real-time quantitative reverse transcription–polymerase chain reaction (RT-PCR) assay ( 29 ). Primer sequences are available in the public real-time PCR primer and probe database (RTPrimerDB; http://medgen.ugent.be/rtprimerdb/ ) ( 30 , 31 ): glyceraldehyde-3-phosphate dehydrogenase (RTPrimerDB ID No. 3); succinate dehydrogenase complex, subunit A, flavoprotein (SDHA; ID No. 7); ubiquitin C (ID No. 8); cyclin-dependent kinase inhibitor 1A (p21 WAF1/CIP1 ; CDKN1A; ID No. 631); BCL2-associated X protein (BAX; ID No. 814); BCL2-binding component 3 (PUMA; BBC3; ID No. 3500); tumor necrosis factor receptor superfamily, member 10b (KILLER/DR5; TNFRSF10B; ID No. 133); and mouse double minute 2 homolog (MDM2; ID No. 3499). Each real-time quantitative RT-PCR assay was validated by using an in silico assay evaluation pipeline (RTPrimerDB) to determine the specificity of the primer pair, to evaluate the secondary structure of the amplicon sequence, and to identify single-nucleotide polymorphisms in the primer-annealing region, followed by experimental evaluation of PCR specificity and efficiency by using a standard curve that consisted of six consecutive fourfold serial dilutions (from 64 ng to 62.5 pg) of reverse-transcribed high-quality control RNA for quantitative RT-PCR–based gene expression analysis (Stratagene QPCR human reference total RNA; Stratagene, La Jolla, CA) ( http://medgen.ugent.be/CMGG/protocols/ ). PCR was performed by using a LightCycler480 (Roche, Mannheim, Germany) or an iQ5 (Bio-Rad) real-time PCR detection system in reactions of 7.5 and 15 μL, respectively, that contained 250 nM of each primer and an amount of cDNA that was equivalent to 20 ng of total RNA. Gene expression levels and PCR efficiencies, along with the SEs on the gene expression levels, were calculated by using qBase PLUS 1.0 analysis software ( http://www.biogazelle.com ) ( 32 ), which uses a delta–quantification cycle (delta–C q ) model with PCR efficiency correction and normalization to multiple reference genes ( 33 ). Expression levels of glyceraldehyde-3-phosphate dehydrogenase, SDHA, and ubiquitin C were used for normalization. All real-time quantitative RT-PCR assays were conducted in duplicate, and four repetitions of the experiment were performed using independently prepared cDNA templates.
UKF-NB-3, UKF-NB-3 r DOX 20 , and UKF-NB-3 r VCR 10 cells (2 × 10 6 cells) treated with 0 or 16 μM nutlin-3 for 24 hours were solubilized in lysis buffer (1% Triton X-100, 150 mM NaCl, 50 mM HEPES [pH 7.5], 1 mM EDTA, 2.5 mM EGTA, 1 mM NaF, 10 mM β-glycerophosphate, and protease and phosphatase inhibitor cocktails [Sigma]). Cell lysates were clarified by centrifugation at 15 300 g for 2 minutes. Total protein (60 μg) was fractionated on a 10% sodium dodecyl sulfate–polyacrylamide gel; transferred to a Hybond-P membrane (Amersham Biosciences, Piscataway, NJ); and probed with a mouse monoclonal anti-p53 antibody (1:1000 dilution; Santa Cruz Biotechnology, Santa Cruz, CA), a mouse monoclonal anti-MDM2 antibody (1:1000 dilution; Santa Cruz Biotechnology), a mouse monoclonal anti-p21 WAF1/CIP1 antibody (1:500 dilution; Santa Cruz Biotechnology), or a mouse monoclonal anti-PARP antibody (1:2000 dilution; Cell Signaling Technology, Danvers, MA). To confirm equal loading, the membranes were incubated with stripping buffer (62.5 mM Tris, 2% sodium dodecyl sulfate, and 100 mM β-mercaptoethanol) for 15 minutes at 50°C and then probed with a mouse monoclonal anti-vinculin antibody (1:10 000 dilution; Sigma). Horseradish peroxidase–conjugated anti-mouse IgG (1:4000 dilution; Cell Signaling Technology) was used as a secondary antibody. Protein–antibody complexes were visualized using ECL Plus (Amersham Biosciences) or Immobilon Western (Millipore, Billerica, MA) chemiluminescent detection reagents followed by exposure to autoradiography film (Hyperfilm ECL; Amersham Biosciences).
Cell Cycle and Hypodiploidy Analysis
Measurements of cell cycle phase distribution and sub-G 1 DNA content were performed using a DNA Prep Reagents kit (Beckman Coulter, Miami, FL) and flow cytometric analysis. Briefly, UKF-NB-3, UKF-NB-3 r DOX 20 , and UKF-NB-3 r VCR 10 cells (1 × 10 6 cells) were seeded in a 25-cm 2 flask, cultured overnight, and treated with 16 μM nutlin-3 or vehicle control for 24 or 48 hours. The cells were detached from the substratum by incubation with trypsin–EDTA, washed with phosphate-buffered saline (PBS; 137 mM NaCl, 2.7 mM KaCl, 10 mM Na 2 HPO 4 , and 2 mM KH 2 PO 4 [pH 7.4]), permeabilized in 100 μL of lysing and permeabilizing reagent, and incubated for 15 minutes at room temperature with 2 mL of staining solution containing 50 μg/mL propidium iodide and 4 kU/mL RNase type III-A. The cellular DNA content was measured with an FC500 flow cytometer (Beckman Coulter). The percentage of cells with sub-G 1 DNA content, which is indicative of apoptotic cell death, was determined by using CXP Analysis software version 2.2 (Beckman Coulter). The sub-G 1 population was isolated from the total cell population before cell cycle fractions (G 1 , S, and G 2 /M) were quantified by using MultiCycle software version 3.0 for Windows (Phoenix Flow Systems, San Diego, CA). Two independent experiments were performed.
DNA Fragmentation Assay
DNA from UKF-NB-3, UKF-NB-3 r DOX 20 , and UKF-NB-3 r VCR 10 cells (2 × 10 6 cells) treated with 0 or 16 μM nutlin-3 for 24 hours was extracted by using a Suicide Track DNA Ladder Isolation kit (Calbiochem, San Diego, CA). The DNA (1500-ng aliquots) was resolved by electrophoresis on a 1.5% agarose gel containing 0.5 μg/mL ethidium bromide and was visualized under ultraviolet light. Three independent experiments were performed.
Analysis of Caspase-3 and Caspase-7 Activity
UKF-NB-3, UKF-NB-3 r DOX 20 , and UKF-NB-3 r VCR 10 cells were plated in duplicate wells of a 96-well plate (10 4 cells per well), incubated for 6 hours, and treated for 24 hours with nutlin-3 (0, 2, 4, 8, 16, or 32 μM). The combined activity of caspase-3 and caspase-7 was measured using the Caspase-Glo 3/7 Assay (Promega) with correction for cell viability. Three independent experiments were performed.
In Vivo Studies
Five-week-old female athymic nude mice (n = 30; CBy.Cg- Foxn1 nu /J strain; Charles River Laboratories, Brussels, Belgium) were inoculated subcutaneously into the right flank with 1 × 10 7 UKF-NB-3 r DOX 20 cells (n = 20 mice) or UKF-NB-3 r VCR 10 cells (n = 10 mice) that were suspended in 100 μL of Matrigel (BD Biosciences, Erembodegem, Belgium). Approximately 2 weeks later, after the establishment of subcutaneous xenograft tumors of 100–300 mm 3 in size, the mice were randomly assigned to receive either nutlin-3 or vehicle control (n = 9–10, UKF-NB-3 r DOX 20 –inoculated mice per group; n = 5, UKF-NB-3 r VCR 10 –inoculated mice per group). Sample sizes for UKF-NB-3 r DOX 20 –inoculated mice were sufficient to detect twofold differences between treatment groups, which were allowed to have a within-group variation in tumor volume up to 50%, with a power of 80% at a statistical significance level of α = .05. No differences in tumor volume after nutlin-3 treatment were expected in UKF-NB-3 r VCR 10 –inoculated mice. Nutlin-3, prepared by custom synthesis (Cayman Chemical) and dosed at 200 mg per kg body weight in 2% Klucel and 0.2% Tween-80 (both from Fagron, Braine-l’Alleud, Belgium), or the vehicle control (2% Klucel, 0.2% Tween-80) were administered by oral gavage twice daily for 3 weeks (UKF-NB-3 r DOX 20 –inoculated mice) or until euthanasia was required by a tumor size that exceeded 2000 mm 3 (UKF-NB-3 r VCR 10 –inoculated mice). This dosing regimen was chosen because it was previously shown to be effective and safe in a murine osteosarcoma xenograft model ( 25 ). Dimensions of the neuroblastoma xenografts were monitored by caliper measurement three times a week, and tumor volumes were calculated using the formula 0.4 × a × b2 , where a is the largest tumor diameter and b is the smallest tumor diameter. The general physical status of mice was recorded daily, and body weight measurements were performed weekly. Euthanasia was performed by cervical dislocation. Whole blood was collected by cardiac puncture, and the tumor, liver, and lungs were removed and cut into half. One piece of the tissue was snap-frozen in liquid nitrogen and stored at −80 °C, whereas the other half was fixed in buffered formalin followed by paraffin embedding.
We performed a separate experiment that was aimed at expression analysis of the p53 pathway and immunohistochemical evaluation of caspase-3 activity. CBy.Cg- Foxn1 nu /J mice with established UKF-NB-3 r DOX 20 xenografts of approximately 700 mm 3 (ie, 24 days after tumor cell inoculation) were treated orally with three doses of nutlin-3 (200 mg per kg body weight) or vehicle control over a course of 24 hours (doses were given at 0, 12, and 24 hours; n = 5–6 mice per treatment group). The mice were killed by cervical dislocation 4 hours after the last dose, and their tumors were excised and cut into two pieces for snap-freezing in liquid nitrogen and fixation in buffered formalin.
All mouse studies were performed in accordance with the Council of Europe guidelines for accommodation and care of laboratory animals. Protocols were approved by the Institutional Ethical Commission for Animal Experimentation (reference number ECD 05/40).
Analysis of p53 Target Gene Expression In Vivo by Real-Time Quantitative RT-PCR
Total RNA was extracted from cryosections of UKF-NB-3 r DOX 20 xenografts that had been collected from mice treated with three oral doses of 200 mg/kg nutlin-3 or vehicle control for 24 hours (n = 5–6 mice per treatment group) by using an RNeasy Mini kit (Qiagen). DNase treatment, cDNA synthesis, and real-time quantitative RT-PCR analysis for expression of p53 target genes were performed as described above. Expression levels of glyceraldehyde-3-phosphate dehydrogenase, SDHA, and ubiquitin C were used for normalization.
Immunohistochemical Analysis of Caspase-3 Activity
Paraffin-embedded sections of UKF-NB-3 r DOX 20 xenografts from mice that had been treated with three oral doses of 200 mg/kg nutlin-3 or vehicle control for 24 hours were deparaffinized in xylene, followed by rehydration by transfer through a graded alcohol series. Subsequently, heat-induced antigen retrieval was performed with EDTA buffer (pH 8.0), and endogenous peroxidase activity was blocked with 0.3% hydrogen peroxide in PBS. After washing with PBS, nonspecific binding sites were blocked with 1% bovine serum albumin in PBS. The sections were then incubated for 1 hour with a rabbit polyclonal anti-human antibody against active caspase-3 (1:1000 dilution; R&D Systems, Minneapolis, MN), rinsed, and exposed to biotinylated swine anti-rabbit secondary antibody (1:400 dilution; Dako, Glostrup, Denmark) for 30 minutes. After a further washing step, the sections were incubated with horseradish peroxidase–conjugated streptavidin (1:300 dilution; Dako) for 30 minutes. The slides were washed with PBS, and color was developed using 3-amino-9-ethylcarbazole (Dako). Finally, the sections were counterstained with hematoxylin and mounted under glass coverslips with Aquatex (Merck, Darmstadt, Germany). Specificity of staining was ensured by inclusion of a negative control section in which the primary antibody was substituted by irrelevant rabbit Igs (1:800 dilution; Santa Cruz Biotechnology). Caspase-3–positive cells, defined as cells with red staining of the cytoplasm and scored in a blinded manner, were quantified by averaging 10 high-power fields (×400 magnification) per tumor, avoiding areas of necrosis, for all individual mice (n = 5–6 mice in each treatment group). Results are expressed as mean number of caspase-3–positive cells per high-power field.
Evaluation of Metastatic Disease by Real-Time Quantitative PCR and RT-PCR
For real-time quantitative PCR–based assessment of metastatic disease in mice carrying UKF-NB-3 r DOX 20 xenografts after 3 weeks of twice daily oral treatment with 200 mg/kg nutlin-3 or vehicle control, genomic DNA was isolated from approximately 20 cryosections of liver and lungs (n = 6 DNA samples for each organ site per treatment group) by using a QIAamp DNA Mini kit (Qiagen) and from blood samples (n = 4–6 DNA samples per group) by using a QIAamp DNA Blood Mini kit (Qiagen). Copy number levels were determined for the human MYCN gene and for the murine Hprt1 and Pthlh genes. Primer sequences are available in RTPrimerDB ( http://medgen.ugent.be/rtprimerdb/ ) ( 30 , 31 ): v-myc avian myelocytomatosis viral–related oncogene, neuroblastoma-derived ( MYCN ; RTPrimerDB ID No. 11), hypoxanthine guanine phosphoribosyl transferase 1 ( Hprt1 ; ID No. 45), and parathyroid hormone–like peptide ( Pthlh ; ID No. 3314). Real-time quantitative PCR with SYBR Green I detection chemistry and qBase PLUS 1.0 analysis were performed as described above for quantification of mRNA expression levels, except that we used 10 ng of genomic DNA as template for the PCR reaction and Hprt1 and Pthlh as the reference genes for normalization.
In parallel, total RNA was extracted from liver and lungs of UKF-NB-3 r DOX 20 xenograft–bearing mice that had been treated orally with 200 mg/kg nutlin-3 or vehicle control twice daily for 3 weeks by using an RNeasy Mini kit (Qiagen) (n = 6 RNA samples, each derived from approximately 20 cryosections, for each organ site per treatment group). Residual DNA was removed by DNase treatment, cDNA was synthesized, and real-time quantitative RT-PCR analysis was performed as described above. Expression levels were determined for the human MYCN gene and for the murine Ppia , Tbp , and Ubc genes. Primer sequences are available in RTPrimerDB ( http://medgen.ugent.be/rtprimerdb/ ) ( 30 , 31 ): MYCN (RTPrimerDB ID No. 180), peptidylprolyl isomerase A (Ppia; ID No. 3310), TATA box–binding protein (Tbp; ID No. 3877), and ubiquitin C (Ubc; ID No. 3874). Levels of Ppia, Tbp, and Ubc expression were used for normalization of MYCN expression.
This real-time quantitative PCR and RT-PCR approach allows for estimation of the relative amount of MYCN DNA and the relative expression of MYCN mRNA, respectively, at a given organ site. We considered both variables to be reflective of the neuroblastoma tumor load. A comparison of these indicators between vehicle- and nutlin-3–treated mice was performed for each individual organ system and, by using a previously described standardization procedure comprising log transformation, mean centering, and autoscaling ( 34 ), across the different organ systems.
Morphological Evaluation of Metastatic Foci
Histopathologic evaluation of metastatic disease was performed in a blinded manner by an experienced pathologist on hematoxylin and eosin–stained paraffin sections of liver and lungs from mice bearing UKF-NB-3 r DOX 20 xenografts after 3 weeks of twice daily oral treatment with 200 mg/kg nutlin-3 or vehicle control. Metastatic foci were defined as groups of at least three adjacent tumor cells. The frequency of metastatic foci was quantified for 10 randomly chosen microscopic fields (×200 magnification) per organ site per mouse (n = 6 evaluated mice in each treatment group) and expressed as the mean number of metastatic foci per microscopic field.
Analysis of cell viability data was performed using GraphPad Prism software version 5.01 for Windows (GraphPad Software, San Diego, CA). The nutlin-3 concentrations that caused 50% inhibition of cell growth (IC 50 values) and corresponding 95% confidence intervals (CIs) were computed by fitting a sigmoidal curve with variable slope to the dose–response data. The 95% confidence interval for the ratio of two IC 50 values was determined by using the half maximal effective concentration (EC 50 ) shift equation in GraphPad Prism. Differences in the fold induction in expression level after nutlin-3 treatment between cell lines and differences in outcome measures (tumor growth, expression of p53 target genes, apoptosis, MYCN DNA content, MYCN mRNA expression level, and frequency of metastatic foci) between treatment groups of mice were analyzed by the nonparametric Mann–Whitney test with the use of SPSS software version 16.0 for Windows (SPSS, Chicago, IL). All statistical tests were two-sided. P values less than .05 were considered statistically significant.
Model System of Multidrug-Resistant Neuroblastoma and Effect of Nutlin-3 on Cell Viability
To investigate the antitumor activity of nutlin-3 against multidrug-resistant neuroblastoma, we used a well-established model system of acquired drug resistance that closely resembles advanced-stage neuroblastoma. Long-term exposure of human neuroblastoma UKF-NB-3 cells to gradually increasing concentrations of doxorubicin or vincristine resulted in the generation of sublines, designated as UKF-NB-3 r DOX 20 and UKF-NB-3 r VCR 10 , respectively, that were profoundly resistant to a variety of chemotherapeutic agents and γ-irradiation and that displayed, compared with the parental UKF-NB-3 cells, increased survival in serum-free medium, increased proliferative and invasive capacity, and enhanced tumorigenicity ( 35 , 36 ). Overexpression of P-glycoprotein and constitutive phosphorylation of Akt contribute to the chemoresistant and increased survival phenotype, respectively, of both sublines ( 35 ), and decreased expression of the NCAM (CD56) adhesion receptor might cause the enhanced invasive potential of UKF-NB-3 r VCR 10 cells ( 36 ). UKF-NB-3 r DOX 20 cells and parental chemosensitive UKF-NB-3 cells are wild-type for p53, whereas UKF-NB-3 r VCR 10 cells carry a 404G>T (C135F) missense mutation in the TP53 gene ( 35 ). In contrast to the abundant overexpression of P-glycoprotein in UKF-NB-3 r DOX 20 and UKF-NB-3 r VCR 10 cells, gene expression levels of other proteins involved in the extrusion of toxic substances from cells, such as MRP1, MRP2, and LRP, are comparable to those in UKF-NB-3 cells (data not shown). Qualitative information on phenotypically relevant genes and the IC 50 values for some chemotherapeutic drugs commonly used to treat neuroblastoma are represented in Table 1 .
|Cell line||MYCN†||TP53‡||P-gp §||Akt §||NCAM §|| Mean IC 50 , ng/mL (95% CI) ║|
|UKF-NB-3||Amp||WT||Referent||Referent||Referent||7.3 (2.1 to 12.5)||0.3 (0 to 0.7)||116 (18 to 214)|
|UKF-NB-3 r DOX 20||Amp||WT||↑||↑||≡||72 (32 to 111)||199 (96 to 302)||341 (114 to 568)|
|UKF-NB-3 r VCR 10||Amp||404G>T (C135F)||↑||↑||↓||59 (19 to 100)||69 (37 to 100)||386 (156 to 616)|
|Cell line||MYCN†||TP53‡||P-gp §||Akt §||NCAM §|| Mean IC 50 , ng/mL (95% CI) ║|
|UKF-NB-3||Amp||WT||Referent||Referent||Referent||7.3 (2.1 to 12.5)||0.3 (0 to 0.7)||116 (18 to 214)|
|UKF-NB-3 r DOX 20||Amp||WT||↑||↑||≡||72 (32 to 111)||199 (96 to 302)||341 (114 to 568)|
|UKF-NB-3 r VCR 10||Amp||404G>T (C135F)||↑||↑||↓||59 (19 to 100)||69 (37 to 100)||386 (156 to 616)|
↑ = increased; ↓ = decreased; ≡ = unchanged; Amp = amplified; CI = confidence interval; IC 50 = concentration resulting in 50% inhibition of cell growth; WT = wild-type.
Genomic amplification status.
Mutation status. Mutation is indicated by nucleotide change followed by amino acid change.
Expression level (P-gp, NCAM) or phosphorylation level (Akt) relative to parental UKF-NB-3 cells.
Derived from three independent experiments, each performed in eight replicates.
Targeted inhibition of the p53–MDM2 interaction by nutlin-3 decreased the viability of UKF-NB-3 cells in a dose- and time-dependent manner, with IC 50 values in the low micromolar range (at 24 hours of treatment, IC 50 = 3.62 μM, 95% CI = 3.04 to 4.33 μM; at 48 hours, IC 50 = 2.49 μM, 95% CI = 2.23 to 2.77 μM; at 72 hours, IC 50 = 2.03 μM, 95% CI = 1.90 to 2.16 μM) ( Figure 1 , A). A pronounced reduction in cell viability after treatment with nutlin-3, indicative of an antiproliferative and cytotoxic effect, was also recorded for the chemoresistant UKF-NB-3 r DOX 20 cells with wild-type p53, which displayed slightly higher IC 50 values than the parental UKF-NB-3 cells (at 24 hours, IC 50 = 4.81 μM, 95% CI = 4.46 to 5.20 μM, 1.33-fold increase in IC 50 compared with UKF-NB-3 cells, 95% CI = 1.09- to 1.57-fold; at 48 hours, IC 50 = 3.55 μM, 95% CI = 3.43 to 3.68 μM, 1.43-fold increase in IC 50 compared with UKF-NB-3 cells, 95% CI = 1.28- to 1.57-fold; at 72 hours, IC 50 = 3.14 μM, 95% CI = 3.06 to 3.22 μM, 1.55-fold increase in IC 50 compared with UKF-NB-3 cells, 95% CI = 1.45- to 1.64-fold) ( Figure 1 , B). In contrast, and as expected for a cell line with mutant p53, nutlin-3 did not reduce the viability of UKF-NB-3 r VCR 10 cells unless a relatively high concentration was used (at 24 hours, IC 50 was above the upper limit of the concentration range studied [32 μM], at least an 8.83-fold increase in IC 50 compared with UKF-NB-3 cells; at 48 hours, IC 50 = 26.08 μM, 95% CI = 22.89 to 29.73 μM, 10.49-fold increase in IC 50 compared with UKF-NB-3 cells, 95% CI = 8.79- to 12.18-fold; at 72 hours, IC 50 = 21.22 μM, 95% CI = 18.95 to 23.76 μM, 10.46-fold increase in IC 50 compared with UKF-NB-3 cells, 95% CI = 9.11- to 11.81-fold) ( Figure 1 , C). These results indicate that neuroblastoma cells with a multidrug-resistant profile to conventional chemotherapeutic agents may remain sensitive to nutlin-3, provided that wild-type p53 is retained.
Effect of Nutlin-3 on the p53 Pathway, Cell Cycle, and Apoptosis in Neuroblastoma Cells
According to our understanding of p53 regulation by MDM2, exposure of cells with wild-type p53 to a selective antagonist of p53–MDM2 binding should result in stabilization and activation of the p53 protein, induction of expression of p53 target genes, and activation of cell cycle checkpoints and apoptotic pathways. We used a real-time quantitative RT-PCR assay to evaluate expression of p53-transactivated target genes that have been implicated in cell cycle arrest (CDKN1A [p21 WAF1/CIP1 ]), apoptosis (BAX, BBC3 [PUMA], and TNFRSF10B [KILLER/DR5]), and autoregulation of the p53 response (MDM2). Treatment with 16 μM nutlin-3 for 24 hours resulted in an increase in the expression of these genes in both chemosensitive UKF-NB-3 and chemoresistant UKF-NB-3 r DOX 20 cells ( Figure 2 , A). Fold inductions in expression level after treatment with nutlin-3 were similar in UKF-NB-3 and UKF-NB-3 r DOX 20 cells for the proapoptotic p53 target genes BAX ( P = .149), BBC3 ( P = .149), and TNFRSF10B ( P = 1.0). Expression of MDM2 was increased by nutlin-3 treatment to a slightly higher extent in UKF-NB-3 r DOX 20 cells than in UKF-NB-3 cells (6.08-fold increase vs 3.32-fold increase, difference = 2.76-fold increase, 95% CI = 1.78- to 3.75-fold increase, P = .021), whereas the nutlin-3–induced increase in expression of CDKN1A was somewhat more pronounced in UKF-NB-3 cells than in UKF-NB-3 r DOX 20 cells (30.74-fold increase vs 9.55-fold increase, difference = 21.20-fold increase, 95% CI = 17.50- to 24.89-fold increase, P = .021). In contrast, treatment of p53-mutant UKF-NB-3 r VCR 10 cells with 16 μM nutlin-3 for 24 hours did not result in an appreciable increase in expression of p53 target genes compared with vehicle-treated cells, except for a 2.53-fold (95% CI = 2.27- to 2.79-fold) increase in CDKN1A expression ( Figure 2 , A). As expected, fold inductions in expression level of p53 target genes after exposure to nutlin-3 were statistically significantly greater in UKF-NB-3 cells than in UKF-NB-3 r VCR 10 cells ( P = .021 in each case). To confirm these findings, immunoblotting was performed with primary antibodies against p53, p21 WAF1/CIP1 , and MDM2 after exposure of cells to vehicle control or 16 μM nutlin-3 for 24 hours. In agreement with the previous results, treatment with nutlin-3 led to accumulation of p53 protein and increased expression of p21 WAF1/CIP1 and MDM2 in UKF-NB-3 and UKF-NB-3 r DOX 20 cells but not in p53-mutant UKF-NB-3 r VCR 10 cells ( Figure 2 , B).
We next investigated the effect of nutlin-3 on cell cycle progression and apoptosis. Incubation of UKF-NB-3 and UKF-NB-3 r DOX 20 cells with 16 μM nutlin-3 for 24 hours induced accumulation of cells in G 1 phase and a reciprocal reduction of the number of cells in S phase ( Figure 3 , A). This nutlin-3–induced inhibition of the G 1 /S transition was not observed in nutlin-3–treated UKF-NB-3 r VCR 10 cells ( Figure 3 , A). Induction of apoptosis by nutlin-3 in UKF-NB-3 and UKF-NB-3 r DOX 20 cells was demonstrated by an increase in the sub-G 1 DNA content ( Figure 3 , B), PARP cleavage ( Figure 3 , C), DNA laddering (internucleosomal DNA fragmentation) on agarose gel electrophoresis ( Figure 3 , D), and dose-dependent activation of caspase-3 and caspase-7 ( Figure 3 , E). In contrast, administration of nutlin-3 to p53-mutant UKF-NB-3 r VCR 10 cells resulted only in a slight increase in caspase-3 and caspase-7 activity at the highest tested concentration of 32 μM ( Figure 3 , B–E).
These data confirmed the hypothesis that nutlin-3 is a selective and potent activator of the p53 pathway in neuroblastoma cells irrespective of chemosensitivity and prompted us to further explore this targeted compound in a xenograft model derived from the chemoresistant UKF-NB-3 r DOX 20 and UKF-NB-3 r VCR 10 cells.
Effect of Nutlin-3 on the Growth of Chemoresistant Neuroblastoma Xenografts
To assess the in vivo efficacy of nutlin-3 against established multidrug-resistant neuroblastoma tumors, UKF-NB-3 r DOX 20 or UKF-NB-3 r VCR 10 cells were inoculated subcutaneously into the right flank of athymic nude mice and allowed to grow to tumors of 100–300 mm 3 , and the mice were then treated with nutlin-3 or vehicle control. Nutlin-3 was administered by oral gavage at a dose of 200 mg/kg twice daily, with an intention to treat for 3 weeks. This treatment regimen was well tolerated and did not induce alterations in the physical status or behavior of mice, weight loss, visible abnormalities at necropsy, or any other obvious signs of toxicity. The growth of UKF-NB-3 r DOX 20 xenografts was suppressed in mice that were treated with nutlin-3 (n = 9) compared with mice that were treated with vehicle control (n = 10) ( Figure 4 , A). At completion of the 3-week treatment course, the mean volumes of UKF-NB-3 r DOX 20 xenografts were 772 mm 3 for nutlin-3–treated mice and 1661 mm 3 for vehicle-treated mice (difference in mean tumor volume between treatment groups = 890 mm 3 , 95% CI = 469 to 1311 mm 3 , P < .001). In contrast, and consistent with the mutant p53 status of the UKF-NB-3 r VCR 10 cells, growth of UKF-NB-3 r VCR 10 xenografts in nutlin-3–treated mice (n = 5) was indistinguishable from that in vehicle-treated mice (n = 5) ( Figure 4 , B). Mean tumor volumes in UKF-NB-3 r VCR 10 xenograft–bearing mice at euthanasia, which was performed 12 days after start of the treatment course because of rapidly growing tumors, were 2057 mm 3 for nutlin-3–treated mice and 1978 mm 3 for vehicle-treated mice (difference in mean tumor volume between treatment groups = 79 mm 3 , 95% CI = −968 to 1127 mm 3 , P = .251).
Effect of Nutlin-3 on the p53 Pathway and Apoptosis in UKF-NB-3 r DOX 20 Xenograft Tumors
We next examined whether the antitumor activity of nutlin-3 against UKF-NB-3 r DOX 20 xenografts was mediated by activation of the p53 pathway. For this purpose, mice with established UKF-NB-3 r DOX 20 xenografts of approximately 700 mm 3 were treated with three oral doses of 200 mg/kg nutlin-3 (n = 6) or vehicle control (n = 5) for 24 hours, and tumors were analyzed for the expression of p53 target genes by real-time quantitative RT-PCR. We observed a statistically significant increase in the expression of all five evaluated p53 target genes in UKF-NB-3 r DOX 20 xenografts from mice treated with nutlin-3 compared with those from vehicle-treated mice. Fold inductions in expression level after treatment with nutlin-3 compared with vehicle control were 10.05-fold (95% CI = 5.09- to 15.02-fold, P = .006) for CDKN1A, 1.76-fold (95% CI = 1.08- to 2.45-fold, P = .006) for BAX, 3.31-fold (95% CI = 1.97- to 4.66-fold, P = .006) for BBC3, 2.53-fold (95% CI = 1.09- to 3.96-fold, P = .011) for TNFRSF10B, and 1.76-fold (95% CI = 1.19- to 2.34-fold, P = .011) for MDM2 ( Figure 5 , A). Induction of apoptosis in the nutlin-3–treated UKF-NB-3 r DOX 20 xenografts was confirmed by immunohistochemical evaluation of caspase-3 activity (mean number of caspase-3–positive cells per high-power field after 24 hours of treatment, nutlin-3– vs vehicle-treated mice: 34.0 vs 19.7, difference = 14.3, 95% CI = 4.7 to 24.0, P = .006) ( Figure 5 , B).
Effect of Nutlin-3 on Metastatic Disease in UKF-NB-3 r DOX 20 Xenograft–Bearing Mice
Finally, we examined the antitumor activity of nutlin-3 against metastatic neuroblastoma lesions in mice with established UKF-NB-3 r DOX 20 xenografts. To assess metastatic tumor burden, we extracted genomic DNA from liver, lungs, and whole blood of UKF-NB-3 r DOX 20 xenograft–bearing mice that had received 3 weeks of twice daily oral treatment with 200 mg/kg nutlin-3 or vehicle control and used it to perform real-time quantitative PCR (n = 4–6 mice per treatment group). We used the gene copy number of the human MYCN gene as an indicator of metastatic tumor load. Copy number levels of the murine Hprt1 and Pthlh genes were used for normalization, allowing a quantitative comparison of tumor burden by means of the relative amount of MYCN DNA. Compared with vehicle-treated mice, mice treated with nutlin-3 had non–statistically significantly reduced levels of MYCN DNA in liver ( P = .078), lungs ( P = .078), and peripheral blood ( P = .088) ( Figure 6 , A). Because elevations in MYCN DNA content in these organ systems represent three independent manifestations of increasing metastatic tumor burden, we next performed a pooled analysis of these data by using a standardization method ( 34 ) that allows one to draw statistically reliable conclusions on real-time quantitative PCR datasets that have different biological backgrounds. Application of this standardization procedure, which is based on log transformation, mean centering, and autoscaling, revealed a statistically significant reduction in the relative amount of MYCN DNA across the different organ systems of mice treated with nutlin-3 compared with those of mice treated with vehicle control ( P = .002). A caveat in interpreting these results in terms of metastatic tumor burden relates to observations of circulating MYCN DNA in the serum of patients with MYCN -amplified neuroblastoma ( 37 , 38 ). To exclude the possibility that the nutlin-3–induced reduction in MYCN DNA content of liver and lungs was caused by a decrease in circulating MYCN DNA in these highly vascularized organs and not by a genuine effect on metastatic tumor load, we next evaluated the MYCN mRNA expression levels in liver and lungs by real-time quantitative RT-PCR. Expression levels of the murine Ppia , Tbp , and Ubc genes were used for normalization. UKF-NB-3 r DOX 20 xenograft–bearing mice treated twice daily with 200 mg/kg nutlin-3 for 3 weeks had decreased expression of human MYCN mRNA in liver ( P = .006) and lungs ( P = .037) compared with mice treated with vehicle control (n = 6 mice per treatment group) ( Figure 6 , B), indicating that nutlin-3 had antitumor activity against metastatic lesions. A pooled analysis of the MYCN mRNA expression data from both organ sites further supported the potential value of nutlin-3 in treating metastatic neuroblastoma disease, as nutlin-3 was highly efficacious in reducing MYCN mRNA expression across both organ systems compared with vehicle control ( P = .001).
Confirmation of these results was obtained by morphological evaluation of metastatic disease on hematoxylin and eosin–stained sections of liver and lungs from UKF-NB-3 r DOX 20 xenograft–bearing mice. Mice treated with nutlin-3 had fewer metastatic foci, defined as groups of at least three adjacent tumor cells, in both liver and lungs compared with mice treated with vehicle control (n = 6 mice per treatment group) ( Figure 6 , C). In the liver sections, the mean number of metastatic foci per microscopic field at ×200 magnification was 0.17 for nutlin-3–treated mice and 0.60 for vehicle-treated mice (difference in mean number of hepatic metastatic foci per microscopic field between treatment groups = 0.43, 95% CI = 0.07 to 0.80, P = .014). In the lung sections, the mean number of metastatic foci per microscopic field at ×200 magnification was 4.85 after nutlin-3 treatment compared with 9.80 after vehicle treatment (difference in mean number of pulmonary metastatic foci per microscopic field between treatment groups = 4.96, 95% CI = 0.87 to 9.05, P = .025). These results further strengthen the concept that nutlin-3 may represent a viable treatment option for patients with advanced-stage and chemoresistant neuroblastoma with wild-type p53.
Acquisition of a multidrug-resistant phenotype during chemotherapy represents a major hurdle to the successful treatment of neuroblastoma. Development of novel antitumor strategies is essential to improve the poor survival rates of high-risk patients ( 1 ). In this study, we evaluated the antitumor efficacy of nutlin-3, a potent and selective antagonist of the p53–MDM2 interaction, in a model system of multidrug-resistant neuroblastoma. Selection of the chemoresistant UKF-NB-3 r DOX 20 and UKF-NB-3 r VCR 10 sublines was achieved by long-term exposure of the parental chemosensitive UKF-NB-3 cells to cytotoxic drug doses in the range of clinical plasma concentrations and was associated with the acquisition of a spectrum of clinically relevant phenotypic characteristics, such as increased proliferative and survival capacity, enhanced invasiveness, and higher tumorigenicity ( 35 , 36 ). We demonstrated that treatment with nutlin-3 induces activation of the p53 pathway, G 1 cell cycle arrest, and apoptosis in chemosensitive UKF-NB-3 and chemoresistant UKF-NB-3 r DOX 20 cells to a similar extent but not in UKF-NB-3 r VCR 10 cells, which carry a missense mutation in the TP53 gene. Oral administration of nutlin-3 to mice carrying established UKF-NB-3 r DOX 20 xenografts was efficacious against both the primary tumor and against metastatic lesions, whereas no treatment effects were observed in UKF-NB-3 r VCR 10 xenograft–bearing mice. Although these data are limited to one model system, they strongly suggest that targeted disruption of the p53–MDM2 interaction by nutlin-3 may provide a new therapeutic option for treatment of metastatic and chemorefractory neuroblastoma tumors with wild-type p53.
Important considerations in interpreting these results in terms of clinical applicability include the frequency of p53 mutations and the nature of p53-inactivating lesions in neuroblastoma, the ability to identify patients who are eligible for treatment with nutlin-3, the extent to which nutlin-3–induced p53 is fully activated, the potential toxicity of MDM2 inhibitors to normal tissues, and the possibility of secondary treatment failure because of acquisition of p53 mutations or other defects in the p53 pathway. These issues are discussed in detail in the Supplementary Data (available online). The clinical promise of the nutlin family of MDM2 antagonists for the treatment of cancer is underscored by the recent introduction of one family member, R7112, in phase I clinical trials for patients with advanced solid tumors ( 39 ) or hematologic neoplasms ( 40 ). Similarly, JNJ-26854165, a small-molecule MDM2 antagonist that targets the association of MDM2 with the proteasome rather than the p53–MDM2 interaction, is currently in clinical development ( 41 ), and it is expected that MI-219, another selective inhibitor of the p53–MDM2 interaction, will enter into clinical trials in the near future ( 42 ).
The principal goal of this study was to determine whether nutlin-3 may offer therapeutic benefit for treatment of multidrug-resistant neuroblastoma tumors that carry wild-type p53. Although we did not attempt to maximize the therapeutic response in the xenograft mouse model, it is conceivable that the actual therapeutic gain provided by nutlin-3 might extend beyond the current results. In this study, we used a racemic mixture of nutlin-3, consisting of two enantiomers with a 150-fold difference in the ability to inhibit the p53–MDM2 interaction, to treat xenograft-bearing mice according to an oral dosing regimen of 200 mg/kg twice daily for 3 weeks, a schedule that has proven to be effective and safe in a murine osteosarcoma xenograft model ( 25 ). This treatment regimen is equivalent to a 3-week oral course of the active stereoisomer nutlin-3a at a dose of 100 mg/kg twice daily. The latter scheme has been shown to retard the growth of osteosarcoma xenografts in mice, similar to our results, whereas doubling the dose to 200 mg/kg nutlin-3a twice daily for 3 weeks was capable of causing regression of the osteosarcoma tumors without inducing systemic toxicity ( 43 ).
Another important issue to consider is the potential efficacy of combining nutlin-3 with other treatment modalities. Although concurrent or subsequent administration of antimitotic drugs with nutlin-3 might result in antagonistic effects because of the ability of MDM2 inhibitors to induce a p53-dependent checkpoint arrest in tumor cells with wild-type p53 ( 44 ), several appealing opportunities for combination therapy exist. For example, intrinsic DNA damage signaling sensitizes cancer cells to nutlin-3–induced cytotoxicity ( 45 ), and many studies have reported a synergistic rather than additive interaction between nutlin-3 and genotoxic chemotherapeutics or radiotherapy [eg, ( 46–48 )]. Interestingly, chemotherapy-induced apoptosis is also enhanced by nutlin-3a in p53-mutant cancer cells in an E2F1-dependent manner ( 49 ), suggesting that tumors with different p53 genetic backgrounds might be eligible for combinatorial regimens that incorporate nutlin compounds. A synergistic action with nutlin-3 has also been reported for gene therapeutic approaches ( 50 ) and for molecular therapeutics directed at other survival and growth-promoting pathways, including recombinant TRAIL (51), cyclin-dependent kinase inhibitors (52–54), inhibitors of the phosphatidylinositol 3′-kinase/Akt/mTOR pathway ( 55 , 56 ), γ-secretase inhibitors of the Notch pathway ( 57 ), a Bcl-2 family antagonist (ABT-737) ( 58 ), a pan-Aurora kinase inhibitor (MK-0457) ( 59 ), a proteasome inhibitor (bortezomib) ( 60 ), and an inhibitor of the Raf/MEK/ERK signaling cascade (PD98059) ( 61 ). However, the in vivo safety and efficacy of these combinatorial treatment strategies remains to be determined.
Some potential limitations of this study should be considered. Although the model system we have used is well established and recapitulates several clinically relevant characteristics of chemorefractory neuroblastoma tumors, it is possible that our findings cannot be extrapolated to all forms of chemoresistance in neuroblastoma. Clinical introduction of selective MDM2 antagonists will most likely take place within a combinatorial treatment setting, and this subject has not been addressed experimentally in this study. Furthermore, several issues related to the clinical translation of this novel therapeutic strategy are still unresolved, as we discuss in the Supplementary Data (available online).
In conclusion, we demonstrated that in this model system of multidrug-resistant neuroblastoma, nutlin-3 is a selective and potent activator of the p53 pathway irrespective of chemosensitivity. Oral administration of nutlin-3 as a single agent to mice with established chemorefractory neuroblastoma disease is well tolerated and effective against both primary and distant tumor sites in the presence of wild-type p53. These findings support the initiation of clinical trials of selective MDM2 antagonists for treatment of advanced-stage and chemoresistant neuroblastoma.
Research Foundation, Flanders (Fonds Wetenschappelijk Onderzoek [FWO], G.0198.08 and G.0225.09); Cycle for Life, Belgium; Centrum voor Studie en Behandeling van Gezwelziekten, Ghent; Ghent Childhood Cancer Fund; Concerted Research Actions, UGent (Geconcerteerde Onderzoeksacties, 12051203 and 01G01307); European Community under the FP6 (project: STREP: EET-pipeline, 037260). T.V.M. is a research assistant from the FWO (011F4004). L.F. is supported by a doctoral grant from the Special Research Fund of Ghent University (011D10503). J.T. is supported by a doctoral grant from the Institute for the Promotion of Innovation through Science and Technology in Flanders (Instituut voor de Aanmoediging van Innovatie door Wetenschap en Technologie in Vlaanderen, SB/33271). I.L. is a recipient of a VIB PhD fellowship. A.R. is supported by a grant from the Emmanuel van der Schueren Foundation. M.M. and J.C. Jr are supported by the friendly society Hilfe für krebskranke Kinder Frankfurt e.V. and its foundation Frankfurter Stiftung für krebskranke Kinder.