Search

Neuroblastoma (NB) is one of the most common and deadly pediatric solid tumors. NB is characterized by clinical heterogeneity, from spontaneous regression to relentless progression despite intensive multimodality therapy. There is compelling evidence that members of the tropomyosin receptor kinase (Trk) family play important roles in these disparate clinical behaviors. Indeed, TrkB and its ligand, brain-derived neurotrophic factor (BDNF), are expressed in 50-60 % of high-risk NBs. The BDNF/TrkB autocrine pathway enhances survival, invasion, metastasis, angiogenesis and drug resistance.We tested a novel pan-Trk inhibitor, GNF-4256 (Genomics Institute of the Novartis Research Foundation), in vitro and in vivo in a nu/nu athymic xenograft mouse model to determine its efficacy in inhibiting the growth of TrkB-expressing human NB cells (SY5Y-TrkB). Additionally, we assessed the ability of GNF-4256 to enhance NB cell growth inhibition in vitro and in vivo, when combined with conventional chemotherapeutic agents, irinotecan and temozolomide (Irino-TMZ).GNF-4256 inhibits TrkB phosphorylation and the in vitro growth of TrkB-expressing NBs in a dose-dependent manner, with an IC₅₀ around 7 and 50 nM, respectively. Furthermore, GNF-4256 inhibits the growth of NB xenografts as a single agent (p0.0001 for mice treated at 40 or 100 mg/kg BID, compared to controls), and it significantly enhances the antitumor efficacy of irinotecan plus temozolomide (Irino-TMZ, p0.0071 compared to Irino-TMZ alone).Our data suggest that GNF-4256 is a potent and specific Trk inhibitor capable of significantly slowing SY5Y-TrkB growth, both in vitro and in vivo. More importantly, the addition of GNF-4256 significantly enhanced the antitumor efficacy of Irino-TMZ, as measured by in vitro and in vivo growth inhibition and increased event-free survival in a mouse xenograft model, without additional toxicity. These data strongly suggest that inhibition of TrkB with GNF-4256 can enhance the efficacy of current chemotherapeutic treatment for recurrent/refractory high-risk NBs with minimal or no additional toxicity.

Recent genomic and biological studies of neuroblastoma have shed light on the dramatic heterogeneity in the clinical behaviour of this disease, which spans from spontaneous regression or differentiation in some patients, to relentless disease progression in others, despite intensive multimodality therapy. This evidence also suggests several possible mechanisms to explain the phenomena of spontaneous regression in neuroblastomas, including neurotrophin deprivation, humoral or cellular immunity, loss of telomerase activity and alterations in epigenetic regulation. A better understanding of the mechanisms of spontaneous regression might help to identify optimal therapeutic approaches for patients with these tumours. Currently, the most druggable mechanism is the delayed activation of developmentally programmed cell death regulated by the tropomyosin receptor kinase A pathway. Indeed, targeted therapy aimed at inhibiting neurotrophin receptors might be used in lieu of conventional chemotherapy or radiation in infants with biologically favourable tumours that require treatment. Alternative approaches consist of breaking immune tolerance to tumour antigens or activating neurotrophin receptor pathways to induce neuronal differentiation. These approaches are likely to be most effective against biologically favourable tumours, but they might also provide insights into treatment of biologically unfavourable tumours. We describe the different mechanisms of spontaneous neuroblastoma regression and the consequent therapeutic approaches.

Neuroblastoma is characterized by biological and genetic heterogeneity that leads to diverse, often unpredictable, clinical behavior. Differential expression of the Trk family of neurotrophin receptors strongly correlates with clinical behavior; TrkA expression is associated with favorable outcome, whereas TrkB with unfavorable outcome. Neuroblastoma cells cultured in a microgravity rotary bioreactor spontaneously aggregate into tumor-like structures, called organoids. We wanted to determine if the clinical heterogeneity of TrkA- or TrkB-expressing neuroblastomas was reflected in aggregation kinetics and organoid morphology. Trk-null SY5Y cells were stably transfected to express either TrkA or TrkB. Short-term aggregation kinetics were determined by counting the number of single (non-aggregated) viable cells in the supernatant over time. Organoids were harvested after 8 d of bioreactor culture, stained, and analyzed morphometrically. SY5Y-TrkA cells aggregated significantly slower than SY5Y and SY5Y-TrkB cells, as quantified by several measures of aggregation. SY5Y and TrkB cell lines formed irregularly shaped organoids, featuring stellate projections. In contrast, TrkA cells formed smooth (non-stellate) organoids. SY5Y organoids were slightly smaller on average, but had significantly larger average perimeter than TrkA or TrkB organoids. TrkA expression alone is sufficient to dramatically alter the behavior of neuroblastoma cells in three-dimensional, in vitro rotary bioreactor culture. This pattern is consistent with both clinical behavior and in vivo tumorigenicity, in that SY5Y-TrkA represents a more differentiated, less aggressive phenotype. The microgravity bioreactor is a useful in vitro tool to rapidly investigate the biological characteristics of neuroblastoma and potentially to assess the effect of cytotoxic as well as biologically targeted drugs.

Background: In the INRG dataset, the hypothesis that any segmental chromosomal alteration might be of prognostic impact in neuroblastoma without MYCN amplification (MNA) was tested. Methods: The presence of any segmental chromosomal alteration (chromosome 1p deletion, 11q deletion and/or chromosome 17q gain) defined a segmental genomic profile. Only tumours with a confirmed unaltered status for all three chromosome arms were considered as having no segmental chromosomal alterations. Results: Among the 8800 patients in the INRG database, a genomic type could be attributed for 505 patients without MNA: 397 cases had a segmental genomic type, whereas 108 cases had an absence of any segmental alteration. A segmental genomic type was more frequent in patients >18 months and in stage 4 disease (P

Lestaurtinib (CEP-701), a multi-kinase inhibitor with potent activity against the Trk family of receptor tyrosine kinases, has undergone early phase clinical evaluation in children with relapsed neuroblastoma. We studied the interaction of CEP-701 with isotretinoin (13cRA) and fenretinide (4HPR), two retinoids that have been studied in children with high-risk neuroblastoma.In vitro growth inhibition was assessed following a 72-hour drug exposure using the sulforhodamine B (SRB) assay in eight neuroblastoma cell lines with variable TrkB expression. When appropriate, the combination index (CI) of Chou-Talalay was used to characterize the interaction of 13cRA (non-constant ratio) or 4HPR (constant ratio) with CEP-701.The median (range) IC(50) of single-agent CEP-701 across all cell lines was 0.09 (0.08-0.3) μM. The combination of 13cRA and CEP-701 resulted in additive to synergistic interactions in four of the five cell lines studied. Addition of 1 or 5 μM of 13cRA decreased the median (range) CEP-701 IC(50) 1.5-fold (1.1-2.8-fold) and 1.7-fold (1.5-1.8-fold), respectively. With 10 μM 13cRA, less than 50% of cells survived when combined with various concentrations of CEP-701. The combination of 4HPR and CEP-701 trended toward being antagonistic, with a median (range) CI at the ED(50) of 1.3 (1.1-1.5).The combination of 13cRA and CEP-701 was additive or synergistic in a spectrum of neuroblastoma cell lines, suggesting that these agents can be potentially studied together in the setting of minimal residual disease following intensive chemoradiotherapy for children with high-risk neuroblastoma.

TrkB acts as an oncogenic kinase in a subset of human neuroblastomas. Lestaurtinib, a multi-kinase inhibitor with potent activity against Trk kinases, has demonstrated activity in preclinical models of neuroblastoma.Patients with refractory high-risk neuroblastoma received lestaurtinib twice daily for 5 days out of seven in 28-day cycles, starting at 70% of the adult recommended Phase 2 dose. Lestaurtinib dose was escalated using a 3 + 3 design. Pharmacokinetics and plasma phospho-TrkB inhibitory activity were evaluated in the first cycle.Forty-seven subjects were enrolled, and 10 dose levels explored starting at 25 mg/M(2)/dose BID. Forty-six subjects were evaluable for response, and 42 subjects were fully evaluable for determination of dose escalation. Asymptomatic and reversible grade 3-4 transaminase elevation was dose limiting in 4 subjects. Reversible pancreatitis (grade 2) was observed in 3 subjects after prolonged treatment at higher dose levels. Other toxicities were mild and reversible. Pharmacokinetic analyses revealed rapid drug absorption, however inter-patient variability was large. Plasma inhibition of phospho-TrkB activity was observed 1 h post-dosing at 85 mg/M(2) with uniform inhibition at 120 mg/M(2). There were two partial responses and nine subjects had prolonged stable disease at dose levels ≥ 5, (median: 6 cycles). A biologically effective and recommended phase 2 dose of 120 mg/M(2)/dose BID was established.Lestaurtinib was well tolerated in patients with refractory neuroblastoma, and a dose level sufficient to inhibit TrkB activity was established. Safety and signs of activity at the higher dose levels warrant further evaluation in neuroblastoma.

Neuroblastoma serves as a paradigm for utilising tumour genomic data for determining patient prognosis and treatment allocation. However, before the establishment of the International Neuroblastoma Risk Group (INRG) Task Force in 2004, international consensus on markers, methodology, and data interpretation did not exist, compromising the reliability of decisive genetic markers and inhibiting translational research efforts. The objectives of the INRG Biology Committee were to identify highly prognostic genetic aberrations to be included in the new INRG risk classification schema and to develop precise definitions, decisive biomarkers, and technique standardisation. The review of the INRG database (n=8800 patients) by the INRG Task Force finally enabled the identification of the most significant neuroblastoma biomarkers. In addition, the Biology Committee compared the standard operating procedures of different cooperative groups to arrive at international consensus for methodology, nomenclature, and future directions. Consensus was reached to include MYCN status, 11q23 allelic status, and ploidy in the INRG classification system on the basis of an evidence-based review of the INRG database. Standardised operating procedures for analysing these genetic factors were adopted, and criteria for proper nomenclature were developed. Neuroblastoma treatment planning is highly dependant on tumour cell genomic features, and it is likely that a comprehensive panel of DNA-based biomarkers will be used in future risk assignment algorithms applying genome-wide techniques. Consensus on methodology and interpretation is essential for uniform INRG classification and will greatly facilitate international and cooperative clinical and translational research studies.

Background Chromodomain-helicase DNA binding protein 5 (CHD5) is an important tumor suppressor gene deleted from 1p36.31 in neuroblastomas (NBs). High CHD5 expression is associated with a favorable prognosis, but deletion or low expression is frequent in high-risk tumors. We explored the role of CHD5 expression in the neuronal differentiation of NB cell lines. Methods NB cell lines SH-SY5Y (SY5Y), NGP, SK-N-DZ, IMR5, LAN5, SK-N-FI, NB69 and SH-EP were treated with 1–10 μM 13-cis-retinoic acid (13cRA) for 3–12 days. qRT-PCR and Western blot analyses were performed to measure mRNA and protein expression levels, respectively. Morphological differences were examined by both phase contrast and immunofluorescence studies. Results Treatment of SY5Y cells with 13cRA caused upregulation of CHD5 expression in a time- and dose-dependent manner (1, 5, or 10 μM for 7 or 12 days) and also induced neuronal differentiation. Furthermore, both NGP and SK-N-DZ cells showed CHD5 upregulation and neuronal differentiation after 13cRA treatment. In contrast, 13cRA treatment of IMR5, LAN5, or SK-N-FI induced neither CHD5 expression nor neuronal differentiation. NB69 cells showed two different morphologies (neuronal and substrate adherent) after 12 days treatment with 10 μM of 13cRA. CHD5 expression was high in the neuronal cells, but low/absent in the flat, substrate adherent cells. Finally, NGF treatment caused upregulation of CHD5 expression and neuronal differentiation in SY5Y cells transfected to express TrkA (SY5Y-TrkA) but not in TrkA-null parental SY5Y cells, and both changes were blocked by a pan-TRK inhibitor. Conclusions Treatment with 13cRA induces neuronal differentiation only in NB cells that upregulate CHD5. In addition, NGF induced CHD5 upregulation and neuronal differentiation only in TrkA expressing cells. Together, these results suggest that CHD5 is downstream of TrkA, and CHD5 expression may be crucial for neuronal differentiation induced by either 13cRA or TrkA/NGF signaling. Electronic supplementary material The online version of this article (doi:10.1186/s12943-015-0425-y) contains supplementary material, which is available to authorized users.

Chromatin remodeling is one of the mechanisms by which gene expression is regulated developmentally. Chromatin structure is controlled at least in part by post-translational modification of histones, as well as by chromodomain proteins. We have identified a novel gene encoding a protein with chromatin remodeling, helicase and DNA-binding motifs. This gene, called CHD5, is the fifth member of the CHD gene family identified in humans. This gene is most homologous to CHD3 and CHD4, which encode proteins that are part of the nucleosome remodeling and histone deacetylation (NuRD) complex. CHD5 is preferentially expressed in total brain, fetal brain, and cerebellum. It is also moderately expressed in the adrenal gland, but expression is undetectable in almost all other tissues examined. CHD5 maps within a small region of deletion on 1p36.3 in human neuroblastomas, a common pediatric tumor. We examined a panel of neuroblastoma cell lines for CHD5 expression, which was consistently low or undetectable in all these lines. Expression was also examined in a panel of 137 primary neuroblastomas, and low expression was highly correlated with 1p deletion, MYCN amplification, advanced stage, and unfavorable histology. These findings suggest that this gene may play a role in the development of the nervous system, and it may also play a role in the pathogenesis of neural tumors.

TNF-related apoptosis-inducing ligand (TRAIL) is a potent inducer of apoptosis in adult malignant glioma and various other human solid tumor models but not in normal tissues. To characterize the TRAIL death pathway in childhood primitive neuroectodermal brain tumor (PNET), 8 human PNET cell lines were tested for TRAIL-induced apoptosis. TRAIL-sensitivity of the PNET cell lines was correlated with mRNA expression levels of TRAIL, its agonistic (TRAIL-R1, TRAIL-R2) and antagonistic (TRAIL-R3, TRAIL-R4) receptors, cellular FLICE-like inhibitory protein (cFLIP), caspase-3 and caspase-8. Three of 8 PNET cell lines tested were susceptible to TRAIL-induced apoptosis. Sensitivity to TRAIL-induced apoptosis did not correlate with mRNA expression of TRAIL receptors or cFLIP. However, all TRAIL-sensitive PNET cell lines expressed caspase-8 mRNA and protein, while none of the five TRAIL-resistant PNET cell lines expressed caspase-8 protein. Treatment with the methyltransferase inhibitor 5-aza-2'-deoxycytidine restored mRNA expression of caspase-8 and TRAIL-sensitivity in formerly TRAIL-resistant PNET cells, suggesting that gene methylation inhibits caspase-8 transcription in these cells. We conclude, that loss of caspase-8 mRNA is an important mechanism of TRAIL-resistance in PNET cells. Treatment with recombinant soluble TRAIL, possibly in combination with methyltransferase inhibitors, represents a promising therapeutic approach for PNET that deserves further investigation.

Activation of the neurotrophin receptor TrkA by its ligand nerve growth factor (NGF) initiates a cascade of signaling events leading to neuronal differentiation in vitro and might play an important role in the differentiation of favorable neuroblastomas (NB) in vivo. To study TrkA signal transduction pathways and their effects on differentiation in NB, we stably expressed wild-type TrkA and five different TrkA mutants in the NGF unresponsive human NB cell line SH-SY5Y. Resulting clones were characterized by TrkA mRNA and protein expression, and by autophosphorylation of the receptor. Introduction of wild-type TrkA restored NGF responsiveness of SH-SY5Y cells, as demonstrated by morphological differentiation, activation of mitogen-activated protein kinases (MAPK) and induction of immediate-early genes. Expression of TrkA in the absence of NGF resulted in growth inhibition of transfectants compared to parental cells, whereas NGF-treatment increased their proliferation rate. Analysis of downstream signal transduction pathways indicated that NGF-induced differentiation was dependent on TrkA kinase activity. Our data suggest that several redundant pathways are present further downstream, but activation of the RAS/MAPK signaling pathway seems to be of major importance for NGF mediated differentiation of NB cells. Our results also show that the signaling effector SH2-B is a substrate of NGF-mediated Trk signaling in NB, whereas it is not activated by NGF in rat pheochromocytoma PC12 cells. This might explain the differences we observed in TrkA signaling between neuroblastoma and PC12 cells. Further insight into TrkA signaling may suggest new options for the treatment of NB.

We previously isolated and characterized cDNA clones of DRT (EPHB2), encoding a receptor protein-tyrosine kinase of the EPH family. Northern blot analysis showed that EPHB2 transcripts are expressed in three sizes of approximately 4, 5, and 11 kb, suggesting that these transcripts are generated by alternative splicing and/or alternative use of polyadenylation sites. To explore this possibility, we isolated additional EPHB2 cDNA clones, including clone 5K-1, by re-screening the human fetal brain cDNA library. Nucleotide sequence analysis of clone 5K-1 revealed that it represents a variant transcript of EPHB2 (EPHB2v). Relative to the EPHB2 cDNA sequence previously reported, clone 5K-1 has two coding region deletions of 3 and 93 nucleotides. Nucleotide sequence analyses of EPHB2 genomic DNA fragments corresponding to these deletions suggest that the EPHB2v transcript is generated by alternative splicing. The 3' end of clone 5K-1 contains a polyadenosine stretch preceded by a potential polyadenylation signal, which is not used to generate the EPHB2 transcript. Taken together, these data indicate that EPHB2v is generated by both alternative splicing and alternative use of polyadenylation sites. The EPHB2v protein lacks one arginine residue that resides immediately following the EPHB2 transmembrane domain. In contrast, as a result of the frame shift caused by the 93 nucleotide deletion, the C-terminus of the EPHB2v protein is longer by 70 amino acids than that of EPHB2. We also show that the human neuroblastoma cell line SY5Y and NTera-2N neurons express high levels of EPHB2 and lower levels of EPHB2v. These structural variations found between the EPHB2 and EPHB2v proteins may reflect functional heterogeneity of EPHB2.

Amplification of the N-myc gene is a significant adverse prognostic factor in neuroblastoma, a common childhood tumor. In non-transformed cells, myc expression is controlled through an autoregulatory circuit, through which elevated Myc protein levels lead to down-regulation of myc transcription. The precise mechanism of myc gene autoregulation is unknown. Loss of c-myc autoregulation has been documented in transformed cells from a number of different lineages, but N-myc autoregulation has not yet been investigated. In neuroblastoma, the increased N-Myc protein produced by amplified tumors would be expected to silence N-myc transcription if the autoregulatory loop were intact. To determine whether N-myc autoregulation is operative in human neuroblastoma, and to localize cis-acting elements which mediate N-myc autosuppression, we transfected a series of N-myc 5' promoter constructs into a panel of human neuroblastoma cell lines carrying one or multiple copies of N-myc. The transfected promoter was equally active in single-copy and amplified lines. Significant promoter activity in the presence of abundant Myc protein in amplified neuroblastoma lines indicates that autoregulation is disabled in this subset of tumors. To investigate whether single-copy lines produce insufficient N-Myc protein to trigger autosuppression yet retain an intact autoregulatory circuit, we transfected neuroblastoma lines with 5' promoter constructs in the presence of a c- or N-myc expression vector. Overexpression of c- or N-Myc resulted in diminution of activity of both the transfected promoter and the endogenous N-myc gene in single-copy, but not amplified lines. Using a series of 5' promoter-deletion minigenes, we localized a cis-acting element required for autoregulation close to the transcription start sites. While the precise mechanism of autosuppression remains unknown, we demonstrated that Myc is incapable of silencing the adenovirus major late promoter (AdMLP) in neuroblastoma cells, indicating that Myc suppression of its own promoter and the AdMLP involve distinct components. These studies provide the first systematic investigation of autoregulation in neuroblastoma, and indicate that single-copy neuroblastoma lines produce insufficient N-Myc protein to activate downstream effector(s) of autosuppression; the autoregulatory circuit is otherwise intact. Amplified lines, in contrast, have lost autoregulation.

There is considerable interest in the role of the TRK family of neurotrophin receptors in regulating the survival, growth and differentiation of normal and neoplastic nerve cells. Indeed, there is increasing evidence that TRK genes play an important role in the biology and clinical behavior of neuroblastomas, tumors of the peripheral nervous system. Evidence from several independent studies suggests that high expression of TrkA is an indicator of favorable outcome, and there is an inverse correlation between TrkA expression and N-myc amplification. In addition, some primary neuroblastomas differentiate in vitro in the presence of NGF but die in its absence. We have evidence that coexpression of full-length TrkB and BDNF is associated with N-myc amplification and may represent an autocrine survival pathway. Conversely, truncated TrkB is expressed predominantly in differentiated tumors. Finally, TrkC is expressed in favorable neuroblastomas, essentially all of which also express TrkA. In summary, the study of neurotrophin receptor expression and function in neuroblastomas may provide important insights into the role that these pathways play in the pathogenesis and clinical behavior of this tumor. Ultimately, these pathways may provide attractive targets for the development of therapy aimed at inducing differentiation or programmed cell death in these tumors.

Neuroblastoma is the most common extracranial solid tumour of childhood. Amplification of the proto-oncogene, N-myc, confers a poor prognosis in neuroblastoma, while hyperdiploidy is associated with a favourable outcome. Little is known about the contribution of tumour-suppressor genes to the development or progression of neuroblastoma. We examined allelic imbalance at the locus of the tumour-suppressor gene, APC (adenomatous polyposis coli), on chromosome 5q using a polymerase chain reaction (PCR)-based assay. Nine of 24 (37.5%) informative neuroblastoma tumours showed allelic imbalance (AI) at this locus. Clinical data concerning N-myc amplification and DNA content were correlated with these results in the same patients. Allelic imbalance was found only in tumours containing a single copy of the N-myc gene and exhibiting hyperdiploidy. All nine patients with AI of chromosome 5q were alive after a median follow-up period of 46 months, while 7 of 15 (47%) of those lacking AI at this locus had died (P = 0.018). Allelic imbalance at three additional loci on chromosome 5 was demonstrated in tumours that exhibited AI at the APC locus, suggesting that endoreduplication of chromosome 5 had occurred. Fluorescent in situ hybridisation (FISH) analysis of tumour tissue from one patient exhibiting AI demonstrated two, three, four or six copies of the APC gene per cell, consistent with this hypothesis. These data suggest that allelic imbalance of chromosome 5 is involved in at least a subset of neuroblastomas and influences survival in patients with neuroblastoma. Images Figure 2 Figure 3

Preferential amplification of the paternal allele of the N–myc gene in human neuroblastomas

Tumors of the peripheral nervous system include neuroblastomas, pheochromocytomas, and neuroepitheliomas. Neuroblastomas and pheochromocytomas are adrenergic in origin and share certain genetic features, whereas neuroepitheliomas are thought to be cholinergic and are characterized by distinct genetic features. Neuroblastomas are characterized by deletion of the short arm of chromosome 1 (1p), amplification of the MYCN proto-oncogene, and hyperdiploidy in subsets of tumors. All three of these genetic features have prognostic value in subsets of patients. Allelic loss of 14q also occurs with increased frequency, but the prognostic importance of this abnormality is not known yet. Pheochromocytomas have not been studied as extensively, but allelic loss for 1p appears to be a frequent change, and no clear examples of oncogene activation have been identified to date. Neuroepitheliomas are characterized by translocation between chromosomes 11 and 22. Although they have a characteristic pattern of proto-oncogene expression, it is not clear that any of these oncogenes are activated specifically, and no sites of allelic loss have been identified to date. Thus, cytogenetic and molecular analysis of neuroblastomas, pheochromocytomas, and neuroepitheliomas are useful in distinguishing them from each other and from other tumors in selected cases. Furthermore, certain genetic markers are useful in predicting clinical behavior, especially for neuroblastoma.

Identification of candidate tumor suppressor genes in meningioma by regional expression profiling

Cellular oncogenes comprise a class of genes whose aberrant expression or function may be involved in the development of tumours1. Indeed, several naturally occurring animal and human tumours are associated with consistent alterations in the structure or genomic position of particular cellular oncogenes2–4. Recently, we isolated a DNA segment having limited similarity to c-myc (termed N-myc) from a human neuroblastoma cell line5. Although N-myc was present as a single copy in normal cells, it was selectively amplified up to 140-fold in tumour cells from human neuroblastomas5. Now, we have used somatic cell hybrids to show that N-myc is normally localized on the distal short arm of chromosome 2, and in situ hybridization to localize N-myc to chromosome 2p23–24. Further, in situ hybridization localizes amplified N-myc in neuroblastoma cells to homogeneously staining regions (HSRs) on different chromosomes. Thus, our results suggest that amplification and translocation of N-myc may be interrelated processes associated with human neuroblastoma, and demonstrate that lhe site of N-myc amplification is quite variable and bears no apparent relationship to either the normal single-copy locus or recognized sites of non-random chromosome alteration in human neuroblastoma.

Amplified cellular genes in mammalian cells frequently manifest themselves as double minute chromosomes (DMs) and homogeneously staining regions of chromosomes (HSRs). With few exceptions both karyotypic abnormalities appear to be confined to tumour cells. All vertebrates possess a set of cellular genes homologous to the transforming genes of RNA tumour viruses, and there is circumstantial evidence that these cellular oncogenes are involved in tumorigenesis. We have recently shown that DMs and HSRs in cells of the mouse adrenocortical tumour Y1 and an HSR in the human colon carcinoma COLO320 contain amplified copies of the cellular oncogenes c-Ki-ras and c-myc, respectively. Both DMs and HSRs are found with remarkable frequency in cells of human neuroblastomas. We show here that a DNA domain detectable by partial homology to the myc oncogene is amplified up to 140-fold in cell lines derived from different human neuroblastomas and in a neuroblastoma tumour, but not in other tumour cells showing cytological evidence for gene amplification. By in situ hybridization we found that HSRs are the chromosomal sites of the amplified DNA. The frequency with which this amplification appears in cells from neuroblastomas and its apparent specificity raise the possibility that one or more of the genes contained within the amplified domain contribute to tumorigenesis.