Influence of segmental chromosome abnormalities on survival in children over the age of 12 months with unresectable localised peripheral neuroblastic tumours without MYCN amplification.

BACKGROUND: The prognostic impact of segmental chromosome alterations (SCAs) in children older than 1 year, diagnosed with localised unresectable neuroblastoma (NB) without MYCN amplification enrolled in the European Unresectable Neuroblastoma (EUNB) protocol is still to be clarified, while, for other group of patients, the presence of SCAs is associated with poor prognosis. METHODS: To understand the role of SCAs we performed multilocus/pangenomic analysis of 98 tumour samples from patients enrolled in the EUNB protocol. RESULTS: Age at diagnosis was categorised into two groups using 18 months as the age cutoff. Significant difference in the presence of SCAs was seen in tumours of patients between 12 and 18 months and over 18 months of age at diagnosis, respectively (P=0.04). A significant correlation (P=0.03) was observed between number of SCAs per tumour and age. Event-free (EFS) and overall survival (OS) were calculated in both age groups, according to both the presence and number of SCAs. In older patients, a poorer survival was associated with the presence of SCAs (EFS=46% vs 75%, P=0.023; OS=66.8% vs 100%, P=0.003). Moreover, OS of older patients inversely correlated with number of SCAs (P=0.002). Finally, SCAs provided additional prognostic information beyond histoprognosis, as their presence was associated with poorer OS in patients over 18 months with unfavourable International Neuroblastoma Pathology Classification (INPC) histopathology (P=0.018). CONCLUSIONS: The presence of SCAs is a negative prognostic marker that impairs outcome of patients over the age of 18 months with localised unresectable NB without MYCN amplification, especially when more than one SCA is present. Moreover, in older patients with unfavourable INPC tumour histoprognosis, the presence of SCAs significantly affects OS.

Segmental chromosomal alterations have prognostic impact in neuroblastoma: a report from the INRG project.

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<0.0001). In univariate analysis, 11q deletion, 17q gain and a segmental genomic type were associated with a poorer event-free survival (EFS) (P<0.0001, P=0.0002 and P<0.0001, respectively). In multivariate analysis modelling EFS, the parameters age, stage and a segmental genomic type were retained in the model, whereas the individual genetic markers were not (P<0.0001 and RR=2.56; P=0.0002 and RR=1.8; P=0.01 and RR=1.7, respectively). CONCLUSION: A segmental genomic profile, rather than the single genetic markers, adds prognostic information to the clinical markers age and stage in neuroblastoma patients without MNA, underlining the importance of pangenomic studies.

Segmental chromosomal alterations lead to a higher risk of relapse in infants with MYCN-non-amplified localised unresectable/disseminated neuroblastoma (a SIOPEN collaborative study).

BACKGROUND: In neuroblastoma (NB), the presence of segmental chromosome alterations (SCAs) is associated with a higher risk of relapse. METHODS: In order to analyse the role of SCAs in infants with localised unresectable/disseminated NB without MYCN amplification, we have performed an array CGH analysis of tumours from infants enrolled in the prospective European INES trials. RESULTS: Tumour samples from 218 out of 300 enroled patients could be analysed. Segmental chromosome alterations were observed in 11%, 20% and 59% of infants enroled in trials INES99.1 (localised unresectable NB), INES99.2 (stage 4s) and INES99.3 (stage 4) (P<0.0001). Progression-free survival was poorer in patients whose tumours harboured SCA, in the whole population and in trials INES99.1 and INES99.2, in the absence of clinical symptoms (log-rank test, P=0.0001, P=0.04 and P=0.0003, respectively). In multivariate analysis, a SCA genomic profile was the strongest predictor of poorer progression-free survival. CONCLUSION: In infants with stage 4s MYCN-non-amplified NB, a SCA genomic profile identifies patients who will require upfront treatment even in the absence of other clinical indication for therapy, whereas in infants with localised unresectable NB, a genomic profile characterised by the absence of SCA identifies patients in whom treatment reduction might be possible. These findings will be implemented in a future international trial.

Central nervous system relapse in high-risk stage 4 neuroblastoma: The HR-NBL1/SIOPEN trial experience.

BACKGROUND: There is rising concern on the impact of new strategies, such as high-dose chemotherapy (HDC) and immunotherapy, on the pattern of relapse in high-risk neuroblastoma (HR-NBL). Our aim is to evaluate the incidence and identify risk factors for first recurrence in the central nervous system (CNS) in HR-NBL. PATIENTS AND METHODS: Data from patients with stage 4V HR-NBL included from February 2002 to June 2015 in the prospective HR-NBL trial of the European International Society of Pediatric Oncology Neuroblastoma Group were analysed. Characteristics at diagnosis, treatment and the pattern of first relapse were studied. CNS imaging at relapse was centrally reviewed. RESULTS: The 1977 included patients had a median age of 3 years (1 day-20 years); 1163 were boys. Among the 1161 first relapses, 53 were in the CNS, with an overall incidence of 2.7%, representing 6.2% of all metastatic relapses. One- and three-year post-relapse overall survival was 25 ± 6% and 8 ± 4%, respectively. Higher risk of CNS recurrence was associated with female sex (hazard ratio [HR] = 2.0 [95% confidence interval {CI}: 1.1-3.5]; P = 0.016), MYCN-amplification (HR = 2.4 [95% CI: 1.2-4.4]; P = 0.008), liver (HR = 2.5 [95% CI: 1.2-5.1]; P = 0.01) or >1 metastatic compartment involvement (HR = 7.1 [95% CI: 1.0-48.4]; P = 0.047) at diagnosis. Neither HDC nor immunotherapy was associated with higher risk of CNS recurrence. Stable incidence of CNS relapse was reported over time. CONCLUSIONS: The risk of CNS recurrence is linked to both patient and disease characteristics, with neither impact of HDC nor immunotherapy. These findings support the current treatment strategy and do not justify a CNS prophylactic treatment.

International consensus for neuroblastoma molecular diagnostics: report from the International Neuroblastoma Risk Group (INRG) Biology Committee.

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.

FKHRL1-mediated expression of Noxa and Bim induces apoptosis via the mitochondria in neuroblastoma cells.

Protein kinase-B (PKB) and its target, the forkhead transcription factor like 1 (FKHRL1)/FoxO3a, have been suggested as regulators of neurotrophin-mediated cell survival in neuronal cells. We analyzed human neuroblastoma cells and found that FKHRL1 was phosphorylated, suggesting its inactivation. To study FKHRL1 function, we infected SH-EP and NB15 cells with a 4OH-tamoxifen-regulated FKHRL1(A3)ER(tm) transgene. Activation of FKHRL1 promoted cytochrome-c release and caspase-dependent apoptosis. FKHRL1 induced TRAIL and the BH3-only proteins Noxa and Bim, implicating both extrinsic and intrinsic death pathways. However, expression of dnFADD did not inhibit FKHRL1-induced cell death, whereas Bcl2 protected against apoptosis. This excluded the death-receptor pathway and suggested that cell death decision is regulated by Bcl2-rheostat. Importantly, RNAi knockdown of Noxa or Bim decreased apoptosis, indicating that Noxa and Bim cooperate to mediate FKHRL1-induced cell death. We conclude that Noxa and Bim establish a connection between FKHRL1 and mitochondria, and that both BH3-only proteins are critically involved in FKHRL1-induced apoptosis in neuroblastoma.

Treatment of localised resectable neuroblastoma. Results of the LNESG1 study by the SIOP Europe Neuroblastoma Group.

Main objective of this study was to confirm that surgery alone is an effective and safe treatment for localised resectable neuroblastoma except stage 2 with amplified MYCN gene (MYCNA). Of 427 eligible stages 1-2 patients, 411 had normal MYCN and 16 had MYCNA. Of the 288 stage 1 patients with normal MYCN, 1 died of complications and 16 relapsed, 2 of whom died; 5-year relapse-free survival (RFS) and overall survival (OS) rates were 94.3% (95% confidence interval (CI): 91.6-97) and 98.9% (95% CI: 97.7-100), respectively. Of the 123 stage 2 patients with normal MYCN, 1 died of sepsis and 22 relapsed, 8 of whom died (RFS 82.8%, 95% CI: 76.2-89.5; OS 93.2%, 95% CI: 88.7-97.8). In stage 2, OS and RFS were worse for patients with elevated LDH and unfavourable histopathology. Of 16 children with MYCNA, 7 were stage 1 (5 relapses and 4 deaths) and 9 were stage 2 (3 relapses and 2 deaths) patients. In conclusion, surgery alone yielded excellent OS for both stage 1 and 2 neuroblastoma without MYCNA, although stage 2 patients with unfavourable histopathology and elevated LDH suffered a high number of relapses. Both stage 1 and 2 patients with MYCNA were at greater risk of relapse.

Localization of Agrobacterium rhizogenes T-DNA in plant chromosomes by in situ hybridization.

We have used in situ hybridization to determine the sites of insertion of Agrobacterium rhizogenes Ri T-DNA in the chromosomes of Crepis capillaris (2n = 6) transformed roots. Four transformed root lines were obtained by infecting Crepis stem segments with A. rhizogenes. Southern hybridization analysis indicated that each root line was the result of one or more independent T-DNA insertion events. In two root lines, one copy of T-DNA was present; the other two root lines each contained two copies of T-DNA. To localize these T-DNA inserts on Crepis chromosomes, metaphase spreads were perpared from each root line, and hybridized in situ to a biotinlabeled T-DNA probe. The results indicated that T-DNA was present in a different chromosomal location in each root line, and that each chromosome had been a target for T-DNA insertion at least once. In the root lines containing two T-DNA inserts, two patterns of integration were observed: in one case the T-DNAs were present on separate chromosomes; in the other case the two T-DNAs were close together (but not tandemly arranged) on a single chromosome. A comparison of these results and those obtained previously for a fifth Crepis-transformed root line demostrated that Ri T-DNA does not insert preferentially into a particlar chromosomal location.

Sequence based high resolution chromosomal CGH.

We aimed to directly align a chromosomal CGH (cCGH) pattern with the gene mapping data by taking advantage of the clustering of the GGCC motif at certain positions in the human genome. The alignment of chromosomal with sequence data was achieved by superimposition of (i) the fluorescence intensity of the sequence specific fluorochrome, Chromomycin A3 (CMA3), (ii) the cCGH fluorescence intensity profile of individual chromosomes and (iii) the GGCC density profile extracted from the Ensembl genome sequence database. The superimposition of these three pieces of information allowed us to precisely localize regions of amplification in the neuroblastoma cell line STA-NB-15. Two prominent cCGH peaks were noted, one at 2p24.3, the position 15.4 mega base (Mb), and the other at 2p23.2, 29.51 Mb. FISH and high resolution array CGH (aCGH) experiments disclosed an amplification of MYCN (16 Mb) and ALK (29.2-29.9 Mb), thus confirming the cCGH data. The combined visualization of sequence information and cCGH data drastically improves the resolution of the method to less than 2 Mb.

Adenovirus E1A does not induce the Ewing tumor-associated gene fusion EWS-FLI1.

Rearrangement of the EWS gene with FLI1 is thought to occur early in the pathogenesis of Ewing's sarcoma family tumors (EFTs) because the chromosomal aberration is pathognomonic for this disease. Recently, adenovirus (Ad) 5 E1A protein has been reported to induce this gene rearrangement in a variety of cell types. This finding, if generally substantiated, not only suggests an etiological role for viral agents in the generation of oncogenic chromosomal aberrations but would also significantly impact the use of adenoviral vectors for gene therapy. In contrast, we now report on the absence of EWS-FLI1 chimeric products from short- and long-term cultures of stably Ad-transformed cells lines and from transiently E1A-expressing cell lines. In addition, we demonstrate the absence of E1A from EFTs. We conclude that there is no role for Ads in EFT pathogenesis. Consequently, evidence for a viral genesis of tumor-specific gene rearrangements is not available.

The EWS protein is dispensable for Ewing tumor growth.

EWS encodes a ubiquitously expressed RNA binding protein with largely unknown function. In Ewing sarcoma family tumors (EFT), one allele is rearranged with an ETS gene. This is the first description of an EFT with a complete EWS deficiency in the presence of two copies of a rearranged chromosome 22 carrying an interstitial EWS-FLI1 translocation. Absence of EWS protein suggested that it is dispensable for EFT growth. By sequencing of EWS cDNA from unrelated EFTs, we excluded inactivation of EWS as a general mechanism in EFT pathogenesis. Rather, EWS was found to be uniformly expressed in two splicing variants of similar abundancy, EWSalpha and EWSbeta, which differ in a single amino acid. Three EWS negative cell lines were established, which will serve as valuable models to study normal and aberrant EWS function upon reintroduction into the tumor cells.

Comparative genetic study of intratumoral heterogenous MYCN amplified neuroblastoma versus aggressive genetic profile neuroblastic tumors.

Intratumoral heterogeneous MYCN amplification (hetMNA) is an unusual event in neuroblastoma with unascertained biological and clinical implications. Diagnosis is based on the detection of MYCN amplification surrounded by non-amplified tumor cells by fluorescence in situ hybridization (FISH). To better define the genetic features of hetMNA tumors, we studied the Spanish cohort of neuroblastic tumors by FISH and single nucleotide polymorphism arrays. We compared hetMNA tumors with homogeneous MNA (homMNA) and nonMNA tumors with 11q deletion (nonMNA w11q-). Of 1091 primary tumors, 28 were hetMNA by FISH. Intratumoral heterogeneity of 1p, 2p, 11q and 17q was closely associated with hetMNA tumors when analyzing different pieces for each case. For chromosome 2, 16 cases showed 2p intact, 4 focal gain at 2p24.3 and 8 MNA. The lengths of the smallest regions of overlap (SROs) for 2p gains and 1p deletions were between the SRO lengths observed in homMNA and nonMNA w11q- tumors. Co-occurrence of 11q- and +17q was frequently found with the largest SROs for both aberrations. The evidence for and frequency of different genetic subpopulations representing a hallmark of the hetMNA subgroup of NB indicates, on one hand, the presence of a considerable genetic instability with different SRO of either gains and losses compared with those of the other NB groups and highlights and, on the other hand, the need for multiple sampling from distant and macroscopically and microscopically distinct tumor areas. Narrowing down the different SRO for both deletions and gains in NB groups would be crucial to pinpointing the candidate gene(s) and the critical gene dosage with prognostic and therapeutic significance. This complexity of segmental chromosomal aberration patterns reinforces the necessity for a larger cohort study using FISH and pangenomic techniques to develop a suitable therapeutic strategy for these patients.

Among genes involved in the RB dependent cell cycle regulatory cascade, the p16 tumor suppressor gene is frequently lost in the Ewing family of tumors.

The pRB cell cycle regulatory cascade is frequently perturbed in neoplasia by overexpression of a component of the pRB-phosphorylating cyclin D1/CDK4 complex or by inactivation of pRB or the CDK4 inhibitors p16 and p15. We investigated the status and expression of p16, p15, CCND1, CDK4 and RB genes in the Ewing family of tumors. P16 loss was observed in 8 of 27 tumors (30%) and in 12 of 23 (52%) tumor cell lines from unrelated patients. There were no discrepancies in the p16 status between primary tumors and the corresponding cell lines and between cell lines established from consecutive tumor samples. p15 was codeleted in most cases but p15 mRNA was absent also in cell lines retaining the gene. In addition, posttranscriptional p16 inactivation was observed in two cases. Although no evidence for CDK4 or CCND1 amplification was obtained, expression of these genes varied considerably in the cell lines in a case specific manner. In wild-type p16 cell lines, pRB expression was lost in one case. Our data indicate that, despite the absence of cytogenetically detectable 9p21 chromosomal aberrations, p16 deletions constitute the most frequent secondary molecular aberration in Ewing tumors so far. These results are discussed in the context of the stage of disease and the clinical outcome of the patients. The potential prognostic impact of these findings remains to be further evaluated.

Overexpression of the pseudoautosomal gene MIC2 in Ewing's sarcoma and peripheral primitive neuroectodermal tumor.

Ewing's Sarcoma (ES), the second most frequent bone tumor in childhood and adolescence, and the probably closely related peripheral primitive neuroectodermal tumor (pPNET) share a unique cytogenetic translocation between chromosomes 11 and 22. Both of them expose high amounts of a glycoprotein on their cell surface, which can be specifically detected by the mAb HBA-71. The cDNA coding for the HBA-71 antigen was isolated by screening a cDNA expression library constructed from a pPNET-derived cell line. Nucleotide sequencing revealed the HBA-71 antigen to be the product of the pseudoautosomal gene MIC2 previously identified by the mAb 12E7 in haematopoietic cells. This antigen is a glycoprotein with a molecular weight of about 29,000 and is expressed in low amounts in most human cell lines and probably normal tissues and tumors with only a few exceptions. In T-cells the antigen is involved in cell adhesion processes. In ES- and pPNET-derived cell lines MIC2 expression is significantly enhanced. No gross changes in posttranslational modification could be observed. The high expression results in easy and specific detection of the antigen in immunocytochemical analysis of paraffin embedded tissue sections making HBA-71 a useful tool in tumor diagnosis.

Identification of a novel gene, CDCP1, overexpressed in human colorectal cancer.

We report the identification of a novel human tumor associated gene, CDCP1 (Cub Domain Containing Protein), which was identified using representational difference analysis and cDNA chip technology. The gene consists of eight exons, the upstream region of which neither contains a TATA- nor a CCAAT-box. However, a CpG island is located around the transcription start, which is found in approximately 60% of known genes. The CDCP1 gene was mapped to chromosome 3p21-p23 by fluorescence in situ hybridization. For expression profiling real time quantitative RT--PCR was performed using cell lines and laser capture microdissected colon cancer biopsies. CDCP1 mRNA is approximately 6 kb and highly overexpressed in human colon cancer and lung cancer. CDCP1 represents a putative transmembrane protein, containing three CUB domains in the extracellular part most likely involved in cell adhesion or interacting with the extracellular matrix.

A novel putative tyrosine kinase receptor with oncogenic potential.

We have detected transforming activity by a tumorigenicity assay using NIH3T3 cells transfected with DNA from a chronic myeloproliferative disorder patient. Here, we report the cDNA cloning of the corresponding oncogene, designated UFO, in allusion to the as yet unidentified function of its protein. Nucleotide sequence analysis of a 3116bp cDNA clone revealed a 2682-bp-long open reading frame capable of directing the synthesis of a 894 amino acid polypeptide. The predicted UFO protein exhibits characteristic features of a transmembrane receptor with associated tyrosine kinase activity. The UFO proto-oncogene maps to human chromosome 19q13.1 and is transcribed into two 5.0 kb and 3.2 kb mRNAs in human bone marrow and human tumor cell lines. The UFO locus is evolutionarily conserved between vertebrate species. A 4.0 kb mRNA of the murine UFO homolog is expressed in a variety of different mouse tissues. We thus have identified a novel element of the complex signaling network involved in the control of cell proliferation and differentiation.

Mutational hot spots within the carboxy terminal region of the LMP1 oncogene of Epstein-Barr virus are frequent in lymphoproliferative disorders.

We have recently identified in Epstein-Barr virus (EBV) positive Hodgkin's disease (HD) a variant of the latent membrane protein 1 (LMP1) oncogene characterized by four point mutations and a 30 base pair deletion. These findings led us to test whether such mutants were also present in other lymphoproliferative disorders (LPD). We analysed 98 EBV DNA positive cases (67 LPD, 15 benign conditions, 16 lymphoblastoid cell lines) by PCR for deletions within the LMP1 gene. DNA sequencing of the region coding for the carboxy terminal protein domain was performed on 24 cases. In 13 cases the same combination of 4 point mutations at positions 168,320, 168,308, 168,295 and 168,225 was identified. Of these cases, 12 had an additional point mutation at position 168,357 and eight at position 168,355, and nine had a 30 base pair deletion including nucleotides 168,285 to 168,256. These deletion mutants were identified in HD, angioimmunoblastic lymphadenopathy, B-immunoblastic lymphoma, peripheral T-cell lymphoma, and two lymphoblastoid cell lines. Our findings reveal a high frequency of non-random point mutations at preferential sites within the 3' (carboxy terminal) region of the LMP1 oncogene. The association of these mutational hot spots with LPD suggests that they are involved in EBV related lymphomagenesis and that they define a clinically relevant EBV strain.

Co-amplification of a novel gene, NAG, with the N-myc gene in neuroblastoma.

Substantial evidence implicates amplification of the N-myc gene with aggressive tumor growth and poor outcome in neuroblastoma. However some evidence suggests that this gene alone is not the sole determinant of outcome in N-myc amplified tumors. We have searched for genes that co-amplify with N-myc in neuroblastoma by means of two-dimensional analysis of genomic restriction digests. Using this approach, we have identified and cloned a novel genomic fragment which is co-amplified with N-myc in neuroblastomas. This fragment was mapped in close vicinity to N-myc on chromosome arm 2p24. It was amplified in 5/8 N-myc amplified neuroblastoma cell lines and in 9/13 N-myc amplified tumors. Using a PCR-based approach we isolated a 4.5 kb c-DNA sequence that is partly contained in the genomic fragment. The open reading frame of the cDNA encodes a predicted protein of 1353 amino acids (aa). The homology of the predicted protein, which we designated NAG (neuroblastoma amplified gene), to a C. elegans protein of as yet unknown function, and its ubiquitous expression suggest that NAG may serve an essential function. By Northern blot analysis we showed that amplification of the cloned gene correlates with over-expression in neuroblastoma cell lines. Amplification and consequent over-expression of NAG may, therefore, contribute to the phenotype of a subset of neuroblastomas.

International neuroblastoma staging system stage 1 neuroblastoma: a prospective study and literature review.

PURPOSE: To gain insight into the management of non-metastatic neuroblastoma by examining clinical and biologic features of International Neuroblastoma Staging System (INSS) stage 1 tumors. METHODS: Patients were staged by both the INSS and the Evans staging system and were evaluated for biologic prognostic factors. Patients with INSS stage 1 received no cytotoxic therapy. The literature was reviewed for clinical and biologic data about INSS stage 1. RESULTS: We evaluated 10 consecutive patients (median age, 17.5 months) with INSS stage 1; all remain disease-free (median follow-up duration, > 5 years). Tumors were in the abdomen (n = 6), chest (n = 3), or pelvis (n = 1). Neuroblastoma involved margins of resection in six tumors. Poor-prognostic biologic findings included tumor-cell diploidy (n = 2) and unfavorable Shimada histopathology (n = 2). Two patients were to receive chemotherapy for, respectively, a tumor deemed unresectable and a tumor classified as Evans stage III; second opinions resulted in surgical management alone in each case. Published reports confirm that some INSS stage 1 patients (1) are at risk for overtreatment, and (2) have poor-prognostic biologic findings yet do well. CONCLUSION: Surgery alone suffices for INSS stage 1 neuroblastoma, even if biologic prognostic factors are unfavorable, microscopic disease remains after surgery, and tumor size is suggestive of "advanced-stage" status in other staging systems. Attempts to resect regionally confined neuroblastomas should take precedence over immediate use of cytotoxic therapy; otherwise, some patients may receive chemotherapy or radiotherapy unnecessarily.

Survival from locally invasive or widespread neuroblastoma without cytotoxic therapy.

PURPOSE: To test the hypothesis that cytotoxic therapy is not needed at diagnosis to assure the survival of most patients with non-stage 4 neuroblastoma. METHODS: Patients with non-stage 4 disease received no cytotoxic therapy in the absence of N-myc amplification. The International Neuroblastoma Staging System (INSS) was used. RESULTS: Of 84 consecutive patients with previously untreated, newly diagnosed neuroblastoma, 31 (37%) had non-stage 4 disease. All 31 patients initially received no cytotoxic therapy because none of them had N-myc amplification. Nine stage 1 patients are relapse-free. This report focuses on the 22 patients with locally invasive or distant disease: two stage 2A with gross residual tumor postsurgery, 11 stage 2B with ipsilateral or midline lymph node involvement, four stage 3, and five stage 4S. Eight of the 22 patients were older than 1 year. Postsurgery, 13 patients had visible residual disease, and two others had markedly increased urinary catecholamine levels for more than 1 year. Recurrent or enlarging tumors regressed spontaneously (n = 2) or were excised 5 to 39 months after diagnosis (n = 4). One of the latter had chromosome 1p deletions (common in poor-risk neuroblastoma) that were not detected in the patient's original tumor resected 23 months earlier--findings consistent with clonal evolution or multifocal disease. The patient received chemotherapy. All 22 patients are alive 24 to 98 months (median, 64) from diagnosis. CONCLUSION: Our results suggest that non-stage 4 patients without N-myc amplification can be spared cytotoxic therapy because (1) residual postsurgical or recurrent biologically favorable neuroblastoma rarely evolves into lethal stage 4 disease; and (2) neuroblastoma in lymph nodes has no prognostic significance. These findings are remarkable because no other cancer includes subtypes that are curable without therapy to ablate residual disease.