HETEROGENEITY OF MYCN AMPLIFICATION IN A CHILD WITH STROMA-RICH NEUROBLASTOMA (GANGLIONEUROBLASTOMA)

Amplification of MYCN portends rapid tumor progression and poor prognosis in neuroblastoma. MYCN copy number has been described as homogeneous within a tumor and congruent in primary tumor and metastasis. We report a child with stage III favorable histology stroma-rich neuroblastoma (ganglioneuroblastoma) and a poor outcome with an apparent change in MYCN gene amplification by Southern blot. Initial biopsy revealed a ganglioneuroblastoma with predominance of differentiating cells designated as neuroblastoma, stroma-rich, intermixed (Shimada). Southern blot failed to demonstrate MYCN gene amplification. After front-line chemotherapy failed, a total resection was performed. In this specimen, Southern blot demonstrated MYCN amplification (15-20 copies) in the undifferentiated component and no amplification in the differentiated. Fluorescence in situ hybridization (FISH) analysis performed retrospectively on both tumor biopsies demonstrated MYCN amplification in the undifferentiated sections of both tumor specimens but not in the differentiated ones. This is the first well-documented case report of heterogeneous MYCN amplification in a child with neuroblastoma. Because key therapeutic decisions are based on the presence of MYCN amplification, physicians diagnosing and treating children with neuroblastoma need to be aware of the possibility that MYCN amplification may be heterogeneous within a tumor and may be missed using techniques based on pooled DNA such as Southern blotting. FISH may be a preferable method for determining MYCN amplification.     

Relationship between histopathological features, MYCN amplification, and prognosis: a UKCCSG study. United Kingdom Children Cancer Study Group

Histological sections from 231 patients with neuroblastoma were reviewed and morphological features and their relationship to age, stage, MYCN amplification (in 128 tumours by Southern analyses), and clinical outcome (based on Shimada risk grouping) determined. Stage 4 disease was associated with poorly differentiated and undifferentiated tumours (P = 0.001), an MKI of >2% (P< 0.001), and Shimada unfavourable histology (UHi) P< 0.0001. In univariate analysis MKI was significant in predicting a poorer relapse-free survival (RFS), low vs. intermediate and high (P< 0.001). Age, MYCN amplification, and Shimada UH also emerged as significant variables. There was a higher proportion of MYCN-amplified tumours with Shimada UH (P = 0.03), and this group had a decreased RFS (P = 0.002). In patients with Shimada FH, MYCN amplification did not significantly predict a poor prognosis. In those with stage 4 disease, Shimada classification was not significant in predicting survival (P = 0.97); the same was true for those over the age of 1 year (P = 0.66). In multivariate analysis, MYCN amplification and Shimada UH both emerged as independent prognostic factors. In conclusion, morphological features assigned some subsets of patients to prognostic risk groups. Most MYCN-amplified tumours have unfavourable histology and a poorer prognosis. However, in patients with stage 4 disease and those over the age of 1 year, other factors that may influence prognosis should be determined.     

Successful liver transplantation in an infant with stage 4S(M) neuroblastoma

We report a 2.5‐month‐old infant with bilateral adrenal neuroblastoma, stage 4S(M), with liver metastases and chemotherapy‐induced veno‐occlusive disease leading to cirrhosis requiring liver transplantation. Despite unknown tumour histology and MYCN‐amplification status, we proceeded with liver transplant. This decision was based on clinical suspicion that our patient was MYCN‐negative due to significant tumour regression, and was supported by evidence indicating that MYCN‐amplification is rare in infants with favourable‐stage neuroblastoma. This is the second case report of neuroblastoma requiring liver transplantation; however, in the previously reported case, the diagnosis of neuroblastoma was not established until after transplantation. We discuss this unique case to justify the potential use of life‐saving liver transplants in infants with neuroblastoma. Pediatr Blood Cancer 2013; 60: 515–517. © 2012 Wiley Periodicals, Inc.     

Biological aspects of neuroblastomas identified by mass screening in Quebec

BACKGROUND: Neuroblastoma has several characteristics that suggest that preclinical diagnosis might improve outcome. Therefore, the Quebec Neuroblastoma Screening Project was undertaken from 1989 to 1994 to examine infants at 3 weeks and 6 months by measuring urinary catecholamine metabolites. PROCEDURE: Over the 5-yr period, 45 tumors were detected by screening, 20 were identified clinically prior to the third week, and 64 were identified clinically at a later time. We analyzed available tumors for Shimada histopathology, tumor ploidy, MYCN copy number and serum ferritin. RESULTS: Of the tumors detected by screening, only 2 of 45 tested had unfavorable histology, 2 of 45 had diploid or tetraploid DNA content, 0 of 43 had MYCN amplification, and 4 of 44 had elevated serum ferritin. All of these patients are alive and well. The 20 patients detected prior to the 3-week screen had similar biological characteristics. In contrast, of the patients detected clinically after 3 weeks of age, 19 of 51 testedhad unfavorable histology, 25 of 66 had diploid or tetraploid tumors, 12 of 56 had MYCN amplification, and 14 of 54 had elevated ferritin. CONCLUSIONS: The difference between the screened and clinically detected cases was highly significant for each biological variable. Preliminary data on other biological variables, such as neurotrophin expression and allelic loss on 1 p in these patients are consistent with the above findings. These data suggest that mass screening for neuroblastoma at or before 6 months of age detects almost exclusively tumors that have favorable biological characteristics, many of which might have regressed spontaneously. Thus, continued mass screening for neuroblastoma at 6 months is unlikely to accomplish its intended goal, and should probably be discontinued.     

Natural history and biology of stage A neuroblastoma: a Pediatric Oncology Group Study

PURPOSE: To prospectively analyze the outcome of patients with Stage A neuroblastoma (NB) treated with surgery alone, especially with regard to the prognostic significance of age, tumor site, MYCN copy number, tumor cell ploidy, and histology. PATIENTS AND METHODS: The clinical course of 329 patients with Stage A disease registered on the POG NB Biology Study #9047 between February, 1990 and October, 1997 were evaluated. Age, tumor site, MYCN copy number, tumor cell ploidy, and histology were analyzed for their impact on event-free survival (EFS) and survival (S). RESULTS: The 5-year estimated EFS and S rates for the 329 patients were 91% (+/-3%) and 96% (+/-2%), respectively. The EFS rate was similar for infants younger than 12 months and children age 12 months or older, but age older than 12 months was predictive of lower S rates (P = 0.044). Patients with adrenal, abdominal non-adrenal, thoracic, and cervical tumors had similar S rates. The majority of patients had tumors with favorable biologic features, and only 3% had MYCN amplification. For infants with diploid tumors, the EFS rate was 82% (+/-16%), but effective therapy yielded an S rate of 100%. Rate of S was 80% (+/-26%) and 64% (+/-27%) for patients with unfavorable tumor histology and MYCN-amplified tumors, respectively. CONCLUSION: The outcome for patients with Stage A NB treated with surgery alone is excellent. Although EFS and S rates were significantly worse for patients with MYCN-amplified tumors, a subset achieved long-term remission after surgery alone. For patients with Stage A and MYCN amplification, additional factors are needed to distinguish the patients who will achieve long-term remission with surgery alone from those who will develop recurrent disease.     

Successful treatment of MYCN amplified, progressive stage 4S neuroblastoma in a neonate with hepatic artery embolization in addition to multimodality treatment

Stage 4S metastatic neuroblastoma (NB) has a favorable prognosis due to a high rate of spontaneous regression. Young infants risk lethal complications arising from hepatomegaly, which can develop rapidly despite treatment. MYCN oncogene amplification confers a significantly worse prognosis. We describe a 4-week-old neonate with MYCN-amplified stage 4S NB complicated by gross hepatomegaly causing rapidly progressive respiratory, hepatic, and renal failure. The child remains in remission 3 years after hepatic artery embolization, radiotherapy, standard, and high-dose chemotherapy. Embolization of the hepatic artery, with classical treatment, is feasible and safe at this age and may contribute substantially to the management of high-risk patients.     

Locoregional MYCN-amplified neuroblastoma

MYCN-amplification is strongly associated with other high-risk prognostic factors and poor outcome in neuroblastoma. Infrequently, amplification of MYCN has been identified in localized tumors with favorable biologic features. Outcome for these children is difficult to predict and optimal treatment strategies remain unclear. We report a 5-month-old who presented with an MYCN-amplified INSS stage 3, pelvic neuroblastoma. The tumor had favorable histology, hyperdiploidy, and lacked 1p36 and 11q23 aberrations. Although the patient met the criteria for high-risk neuroblastoma, because of the discordant prognostic markers we elected to treat her according to an intermediate-risk protocol. She remains event-free more than 18 months.     

Liver involvement in neuroblastoma: the Memorial Sloan-Kettering Experience supports treatment reduction in young patients

BACKGROUND: We reviewed clinical and biologic findings in a series of infants with neuroblastoma (NB) in liver. The aim was to gain insights into improving therapy. PATIENTS AND METHODS: Among 19 newly or recently diagnosed infants with NB in liver, 1987-2002, those with stage 4 involving bone received chemotherapy, while those without bone or extensive bone marrow (BM) involvement were observed or received limited treatment if NB caused life-threatening symptoms. We assessed results in the context of NB treatment risk stratification, which is based on age, stage, and selected biologic features (MYCN, ploidy, histology). RESULTS: Six of eight infants with bone involvement became long-term event-free survivors including 1/2 with MYCN amplification and four who received only 4-6 cycles of chemotherapy; at the end of treatment, four infants had abnormalities in liver +/- the primary site, but these resolved. All 11 infants without bone lesions became long-term survivors with either no cytotoxic therapy or only one cycle of chemotherapy (+/- radiotherapy to liver), including four who had stage 4 and one stage 4S patient who still had NB in BM at age 15 months. CONCLUSIONS: Treatment reduction should be considered for subsets of infants with non-MYCN-amplified widespread NB: stage 4 without bone or extensive BM involvement may not require cytotoxic therapy, stage 4S with symptomatic hepatomegaly may not require multiple cycles of chemotherapy, and classic stage 4 may do well with limited chemotherapy. Persistent liver abnormalities post-treatment may not require continued therapy to achieve a radiologic complete remission.     

Metastatic sites in stage IV and IVS neuroblastoma correlate with age, tumor biology, and survival.

PURPOSE: The goal of this study was to determine the incidence of metastatic sites in neuroblastoma and the extent to which metastatic sites correlate with age, tumor biology, and survival. PATIENTS AND METHODS: All 648 patients with stage IV and IVS neuroblastoma registered on Children's Cancer Group protocols 3881 and 3891 were analyzed. Metastatic site data were provided by treating institutions and reviewed in patients with central nervous system (CNS), intracranial, lung, or "other" metastases. RESULTS: The incidence of metastatic sites at diagnosis was 70.5% in bone marrow, 55.7% in bone, 30.9% in lymph nodes, 29.6% in liver, 18.2% in intracranial and orbital sites, 3.3% in lung, and 0.6% in CNS. Event-free survival (EFS) was decreased in patients with bone, bone marrow, CNS, intracranial/ orbital, lung, and pleural metastases, and improved in those with liver and skin metastases. In infants, MYCN amplification and unfavorable Shimada histopathology correlated with increased frequencies of bone and intracranial or orbital metastases. In older patients, MYCN amplification correlated with increased frequencies of intracranial or orbital, liver, and lung metastases. Multivariate analysis revealed that metastatic site is not an independent prognostic factor. CONCLUSIONS: Metastatic pattern in neuroblastoma differs with age and correlates with tumor biological features and EFS. These correlations could reflect changes in host or tumor biological features with age resulting in differences in metastatic capacity or tumor affinity for specific sites.     

新生儿神经母细胞瘤：发病特点、治疗刍议以及临床转归

目的 总结新生儿神经母细胞瘤的发病特点、随访结果,针对治疗方案提出本中心的观点。方法 对2003～2009年我院收治的14例新生儿神经母细胞瘤患儿进行回顾性分析,14例新生儿神经母细胞瘤中,男8例,女6例。发病年龄最大29d,最小2d,平均(19±3.5)d,中位数16d。Ⅰ期5例,Ⅱ期1例,Ⅲ期1例,Ⅳ期1例,4S期6例。腹膜后占位8例,纵隔4例,盆腔2例。分析每个患儿的发病特点,临床分期、生物学特征、治疗方案、及其与预后的关系。结果 8例患儿接受手术切除,其中6例完整切除,2例部分切除后辅以化疗。4例患儿活检后化疗,2例患儿单纯行经验性化疗,未手术/活检。所有患儿中只有4例患儿出现尿VMA的增高(28.6％),Mycn扩增3例,不扩增9例,UH:FH为4:8。两年的总体生存率(OS)是92.9％,局限性神经母细胞瘤患儿的生存率是100％。单纯手术组的生存率(7/8,87.5％),化疗组生存率(5/6,83.3％) (P=0.065);Mycn扩增组的生存率为50％,不扩增组的生存率为100％ (P=0.021);VMA阳性生存率为50％,阴性者生存率为100％(P=0.021)。死亡2例,其中1例死于化疗后的并发症,另1例死于肿瘤的复发。结论 N-myc不扩增和VMA不增高的新生儿神经母细胞瘤的预后很好,单纯手术组和化疗组生存率没有差别。手术、化疗以及其他的相关的并发症要比疾病本身所致的死亡例数要多,手术、化疗应仔细考量。手术治疗提倡改变既往观念,建议短期内密切观察和随访,尽量避免在新生儿期手术。     

Prognostic value of MYCN and ID2 overexpression in neuroblastoma

BACKGROUND: The molecular mechanisms driven by overexpression of the MYCN gene in neuroblastoma are not well known. Whether MYCN overexpression in the absence of genomic amplification, or ID2 overexpression has prognostic value remains controversial. PROCEDURE: Ninety-nine neuroblastic tumors from Memorial Sloan-Kettering Cancer Center and 12 neuroblastoma cell lines were analyzed by Affymetrix U95v2 Microarray System. The expression levels of the genes MYCN and ID2 were determined by RT-PCR and immunohistochemistry and the clinical value of the overexpression of both genes was determined. RESULTS: MYCN genomic amplification with overexpression was prognostic for survival (P = 0.0005). Stage 4 patients with MYCN amplification but without overexpression, had no increased likelihood of death, whereas cases with MYCN overexpression but no genomic amplification showed a low survival (P = 0.0096). ID2 did not correlate with MYCN expression (R = -0.23 for tumors and -0.27 for cell lines) or with survival (P = 0.8746). CONCLUSIONS: MYCN amplification was not related to clinical outcome in the absence of overexpression in neuroblastoma tumors. ID2 expression appears to be independent of MYCN expression and lacks prognostic value.     

Intensified chemotherapy increases the survival rates in patients with stage 4 neuroblastoma with MYCN amplification

PURPOSE: Patients with high-risk neuroblastoma who have multiple copies of MYCN fare much worse than do those without MYCN amplification; however, it has not been clarified whether intensified chemotherapy with or without blood stem cell transplantation can alter the extremely poor prognosis of patients with amplified MYCN. METHODS AND RESULTS: Between 1985 and 1999, 301 patients older than age 12 months with stage 4 neuroblastoma were treated. From January 1985 to February 1991, 80 patients with stage 4 neuroblastoma with and without MYCN amplification uniformly received induction chemotherapy with regimen A(1) (cyclophosphamide 1,200 mg/m(2) and vincristine 1.5 mg/m(2) on day 1, tetra-hydropyranyl [THP]-Adriamycin 40 mg/m(2) on day 3, and cisplatin 90 mg/m(2) on day 5). Among 22 patients with MYCN amplification, nine (40.9%) achieved a complete remission and seven (31.8%) underwent stem cell transplantation. Of 58 patients without MYCN amplification, 43 (74.1%) achieved a complete remission and 14 (24.1%) underwent stem cell transplantation. The 5-year relapse-free survival rates were 23.2% for stage 4 patients with MYCN amplification and 33.3% for those without MYCN amplification (P = 0.029); the 5-year overall survival rates were 32.8% for stage 4 patients with MYCN amplification and 42.8% for those without MYCN amplification (P > 0.05). From March 1991 to June 1998, patients with stage 4 neuroblastoma who had 10 or more copies of MYCN were treated with regimen A(3) (cyclophosphamide 1,200 mg/m(2) per day on days 1 and 2, THP-Adriamycin 40 mg/m(2) on day 3, etoposide 100 mg/m(2) per day on days 1 to 5, and cisplatin 25 mg/m(2) per day on days 1 to 5); those with fewer than 10 copies of MYCN received regimen new A (cyclophosphamide 1,200 mg/m on day 1, THP-Adriamycin 40 mg/m on day 3, etoposide 100 mg/m per day on days 1 to 5, and cisplatin 90 mg/m on day 5), which is similar in intensity to regimen A. Among 88 patients with MYCN amplification, 63 (71.6%) achieved a complete remission and 63 (71.68%) underwent stem cell transplantation. Of 133 patients without MYCN amplification, 93 (69.9%) achieved a complete remission and 71 (53.4%) underwent stem cell transplantation. The 5-year relapse-free survival rates were 36.0% for stage 4 patients with MYCN amplification and 32.2% for those without MYCN amplification (P > 0.05), the 5-year overall survival rates were 34.0% for stage 4 patients with MYCN amplification and 38.9% for those without MYCN amplification (P > 0.05). The difference in relapse-free survival rates was significantly different (P = 0.003) between patients with MYCN-amplified tumor treated before (regimen A(1)) versus after 1991 (regimen A(3)). CONCLUSIONS: With the use of the more intensive induction regimen A plus blood stem cell transplantation for MYCN-amplified patients, survival curves for those with or without MYCN amplification now appear similar. Higher doses of chemotherapy may ameliorate the effect of MYCN amplification in patients with high-risk neuroblastoma.     

New definition of low-risk neuroblastoma using stage, age, and 1p and MYCN status

Data from patients with localized and stage 4S neuroblastoma were analyzed to define a new, extended low-risk group that does not require postoperative chemotherapy. Nine hundred eight patients with stage 1 to 3 and 4S disease without MYCN amplification were included. The prognostic impacts of age, stage, serum lactate dehydrogenase (LDH) activity, histology, and alterations of chromosomes 1p, 11q, and 3p were analyzed. By univariate analysis, alterations of chromosomes 1p and 11q were correlated with poor event-free survival (EFS) and overall survival (OS). Chromosome 3p alterations were prognostic only for EFS. Age, stage, and histology were found prognostic for EFS and OS. Stage 3 patients older than 2 years showed the worst outcome and were excluded from multivariate analysis. By multivariate analysis, status of 1p (P = 0.005, hazard ratio [HR] 3.6) and 11q (P = 0.024, HR 2.8) proved prognostic for EFS but only 1p status (P = 0.009, HR 3.0) for OS. The new low-risk group was defined as no MYCN amplification and either stage 1, stage 2 without 1p alterations, stage 3 two years of age or younger without 1p alteration, or stage 4S. These patients had a better outcome (3-year EFS 88.0 +/- 1.3%, 3-year OS 97.4 +/- 0.6%) than stage 2 and 3 patients with 1p alterations and stage 3 patients older than 2 years (3-year EFS 51.7 +/- 6.5%, P < 0.001; 3-year OS 83.4 +/- 4.5%; P < 0.001). The authors conclude that postoperative chemotherapy is required only in a small group of patients with localized and stage 4S disease without MYCN amplification.     

Biological characteristics of neuroblastoma with partial deletion in the short arm of chromosome 1

BACKGROUND: Neuroblastoma shows remarkable heterogeneity, resulting in favorable and unfavorable outcomes. It is well known that almost all cases with MYCN amplification have a poor prognosis. We have previously reported that unfavorable tumors show high telomerase activity, whereas favorable tumors show low or nil activity. We also found that the unfavorable neuroblastoma often have a loss of heterozygosity (LOH) at the MYCL locus. PROCEDURE: To clarify the biological and clinical profiles of tumors with genetic abnormalities of the short arm of chromosome 1, we performed deletion mapping on 1p on 92 neuroblastoma tissues and corresponding noncancerous samples obtained from 92 cases for 24 micro- or minisatellite loci. RESULTS: LOH was detected in at least one locus of 1p in 43 (47%) cases. All samples were classified into four groups according to the deleted pattern: interstitial deletion (group I, n = 20), short terminal deletion (group ST, n = 6), large terminal deletion (group LT, n = 17), and without detectable deletion (group N, n = 49). All group I cases, whose SRO (shortest region of overlap) was at 1p36.1-2, survived disease free, and none of them showed MYCN amplification or high telomerase activity except for one case. On the other hand, in group LT cases, who showed a large terminal deletion from D1S162 (1p32-pter), including the SRO of group 1, only 5 out of 17 have survived disease free, and 13 showed MYCN amplification or high telomerase activity. The six group ST cases showed small terminal deletion from 1p36.3 with modest prognosis, similar to the group N. CONCLUSIONS: Thus, we propose three loci, 1p36.1-2, 1p32-34, and 1p36.3, as the candidate loci of neuroblastoma suppressor genes on chromosome 1p responsible for groups I, LT, and ST, respectively. Among them, the 1p32-34 locus may be associated with aggressiveness of tumor progression, possibly due to MYCN amplification and/or telomerase reactivation, while the remaining two loci may not.     

Survivin mRNA levels are associated with biology of disease and patient survival in neuroblastoma: a report from the children's oncology group

Alterations in apoptotic mechanisms favoring cell survival may be vital for modifying tumor behavior. Survivin, an inhibitor of apoptosis, and caspase 8, a proapoptotic enzyme, are key players in cellular apoptotic mechanisms. We investigated whether the levels of survivin and caspase 8 and the ratio between these 2 apoptotic factors correlate with tumor biology and predicts outcome in patients with neuroblastoma. Survivin and caspase 8 levels were quantified in 38 primary tumor specimens and analyzed individually and in relation to each other. High survivin expression and high survivin:caspase 8 ratios were associated with MYCN amplification, unfavorable histology, and high-risk group of disease (P<0.0008). High survivin mRNA levels were associated with worse overall survival (P=0.02) although the median follow up was only 22.6 months with a range of 1 day to 3.3 years. Low caspase 8 expression was associated with stage 4 disease, high-risk group, MYCN amplification, and unfavorable histology. Although the survivin:caspase 8 ratio was associated with these risk factors, the ratio did not improve the predictive value of survivin alone in this small series. Quantifying multiple apoptotic genes in neuroblastoma may supplement current risk stratification. Moreover, categorizing aberrant apoptotic gene expression in neuroblastoma may translate into novel therapeutic targets.     

Localised and unresectable neuroblastoma in infants: excellent outcome with primary chemotherapy. Neuroblastoma Study Group, Société Française d'Oncologie Pédiatrique

PROCEDURE: Infants with neuroblastoma (NB) were assessed according to INSS recommendations, including MIBG scan and extensive bone marrow staging to eliminate metastatic spread. Patients with unresectable tumour received chemotherapy, including two courses of carboplatin-etoposide (CE) and two of vincristinecyclophosphamide-doxorubicin (CAdO). Post-operative treatment was to be given only in infants with MYCN amplification. Between 1990 and 1994, 52 consecutive children were registered. RESULTS: Among the 44 patients who received CE as a first course, the response rate was (66%) and the primary could be removed in all children but one, who was in remission. The toxicity was mainly haematological and was always manageable. The 5 year overall survival (OS) and event-free survival (EFS) were 94 and 90 +/- 8%, respectively, with a median follow-up of 48 months. The outcome of infants with no MYCN amplification was excellent; OS and EFS were, respectively, 97 and 94%. CONCLUSIONS: Chemotherapy allows surgical excision and excellent outcome in infants with localised and unresectable NB. Less intensive Chemotherapy should be investigated in such patients.     

A case of composite neuroblastoma composed of histologically and biologically distinct clones.

We report a case of a 12-month-old girl with stage 3 neuroblastoma composed of 2 distinct clones in the adrenal primary tumor. One clone showed neuroblastoma, poorly differentiated subtype with a low mitosis-karyorrhexis index (favorable histology), and the other was neuroblastoma, poorly differentiated subtype with a high mitosis-karyorrhexis index (unfavorable histology), according to the International Neuroblastoma Pathology Classification. Fluorescent-labeled in-situ hybridization using the formalin-fixed, paraffin-embedded material showed that the MYCN oncogene was amplified in the latter clone but not in the former clone. Lymph nodes from ipsilateral and contralateral sides contained metastatic neuroblastoma of the latter clone. It is well documented that tumors in the Ganglioneuroblastoma, nodular (composite, Schwannian stroma-rich/stroma-dominant and stroma-poor) category are composite and composed of multiple clones. To our knowledge, however, this is the 1st case report of composite tumor with biologically favorable and unfavorable clones in the Neuroblastoma (Schwannian stroma-poor) category.     

MYCN AMPLIFICATION IN MEDULLOBLASTOMA

Prominent neuronal differentiation and MYCN amplification occur only in a small percentage of medulloblastomas (primitive neuroectodermal tumor of the cerebellum). In this article, we describe two medulloblastomas that showed a striking degree of neuronal differentiation with islands of mature neurons associated with abundant neuropil. In these differentiated foci, there were no mitoses or atypia, and the histology was reminiscent of a malformative or hamartomatous lesion. Both tumors were assessed for MYCN amplification by the novel technique of differential polymerase chain reaction (PCR) and fluorescent in situ hybridization (FISH). In case 1, MYCN amplification was in the 10 - fold range determined by differential PCR, while in case 2 the level of amplification was marked, with 20 - 30 copies compared with diploid controls. FISH analysis also confirmed the presence of MYCN amplification in both tumors. These two cases are of great interest, especially as they show MYCN amplification in medulloblastomas with marked degree of neuronal differentiation. Furthermore, in both cases, there was evidence of a slower and more indolent clinical course, suggesting a more favorable outcome in medulloblastoma with this degree of neuronal differentiation.