 |
|
 |
|
 |
 |
|
 |
 |
 |
 |
 |
 |
 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
PEDIATRICS Vol. 106 No. 5 November 2000, pp. 1045-1053
, §, ¶,, ¶,
, ¶,, and
From the * Department of Pathology, Division of
Neuropathology, § Department of Neurosurgery, Department of
Radiology, and ¶ Kaplan Cancer Center, New York University Medical
Center, New York, New York; # Department of Neurosurgery, Montreal
Children's Hospital, Montreal, Quebec; ** Department of Neurosurgery,
Children's Medical Center, Medical College of Georgia, Augusta,
Georgia; and Institute for Neurology and Neurosurgery, Beth
Israel Medical Center, New York, New York.
 |
ABSTRACT |
|---|
Top
Abstract
Methods
Results
Discussion
Conclusion
References
|
|---|
Objective. We discuss the clinical aspects, pathology, and molecular genetics of 7 patients with primitive neuroectodermal tumors (PNETs) arising in the brainstem that were treated at our institution from 1986 through 1995. Most neuro-oncologists avoid performing biopsies in children with pontine tumors. This article raises the question as to whether biopsies should be performed, because treatment recommendations might differ if a PNET was diagnosed rather than a pontine glioma.
Patients and Methods. We reviewed the clinical neuro-oncology database and the files of the Division of Neuropathology at New York University Medical Center from 1986 through 1995 and identified 7 histologically confirmed PNETs arising in the brainstem among 146 pediatric brainstem tumors. The clinical, neuroradiological, and neuropathological data were reviewed. Postmortem examinations were performed in 2 cases. Formalin-fixed, paraffin-embedded tumor tissues were also available in 6 of 7 patients that were tested for p53 gene mutations using single-strand conformation polymorphism analysis. We also tested 9 cerebellar PNETs, 9 brainstem gliomas, and 3 normal brains for p53 gene mutations as controls.
Results. All 7 patients presented with focal cranial nerve deficits, and 2 were also hemiparetic. The median age at diagnosis was 2.7 (1-8 years). Magnetic resonance imaging (MRI) characteristics included a focal intrinsic exophytic nonenhancing brainstem lesion that had low T1-weighted and high T2-weighted signals. Hydrocephalus was present in 5 patients at diagnosis, 3 of whom had leptomeningeal dissemination. Meningeal dissemination occurred later in the course of the disease in 3 other patients. Five children required shunts at diagnosis and another 2 at recurrence. Despite therapy, all 7 PNET patients died within 17 months of diagnosis with a mean survival of 8 (4-17) months. No mutation in the p53 gene was detected.
Conclusions. Brainstem PNETs tend to arise at a younger age than brainstem gliomas and medulloblastomas. The MRI pattern suggests a localized rather than a diffuse intrinsic nonenhancing brainstem tumor. Like other PNETs, brainstem PNETs have a high predilection to disseminate within the central nervous system. The absence of p53 mutations is similar to other PNETs. Despite their origin close to the cerebellum, brainstem PNETs exhibit a more aggressive behavior and result in worse clinical outcomes than do cerebellar PNETs. Key words: primitive neuroectodermal tumors, brainstem, children, p53 mutation.
Intrinsic brainstem tumors account for 10% of all
pediatric brain tumors.1,2 The majority (85%) are
composed of high-grade fibrillary gliomas, which arise predominantly in
the pons and less frequently in the medulla.3-6 They
typically have a diffuse infiltrative pattern of growth, and 95% of
the children die from the disease within 3 years of diagnosis
regardless of therapy type. Low-grade gliomas (15%) are focal tumors
that usually arise in the medulla or midbrain. They are sometimes
amenable to surgical resection and are treated with radiotherapy or
chemotherapy. These have a better prognosis.4,7
Primitive neuroectodermal tumors (PNETs) are malignant embryonal tumors
occurring most commonly in the cerebellum of young individuals8-11 and consist of primitive neuroepithelial
cells that appear as small blue cells.9,10,12 Depending on
their differentiation, PNETs can express markers of astrocytic (glial
fibrillary acidic protein [GFAP]), neuronal (class III CNS PNETs arising outside the cerebellum23 (~10% of
cases) are classically designated according to their primary location,
eg, pineoblastoma, retinoblastoma, cerebral
neuroblastoma,24 but also include tumors designated as
ependymoblastoma and medulloepithelioma.12,25 They are
also recorded in the spinal cord,19,20,23 and occasional
case reports indicate brainstem PNETs.5,26,27 PNETs
respond readily to radiotherapy and chemotherapy. With current therapy,
as high as 50% to 70% of patients with cerebellar PNETs can be
expected to survive disease-free for 5 years after diagnosis.28,29 Although several chromosome structural and
numerical abnormalities have been described in cerebral and cerebellar
PNETs,30,31 mutations of the p53 gene have been only
rarely reported in PNETs.32-34 In contrast, brainstem
gliomas have a high frequency of ( We present our clinical, neuroradiologic, and pathologic analysis of 7 institutional cases of PNET arising within the brainstem. We also
performed single-strand conformation polymorphism (SSCP) analysis,
which failed to detect any p53 mutations in the 6 cases studied.
Cases were ascertained by performing a computerized search of
our pediatric neuro-oncology database at New York University Medical
Center from 1986 through 1995 for the diagnosis of PNETs arising in the
brainstem. Of a total of 146 pediatric brainstem tumors, 97 had some
form of surgical procedure yielding diagnostic tissue. The typical
reasons for operating on brainstem tumors at our institution consisted
of the presence of a focal or exophytic tumor or leptomeningeal
dissemination on the preoperative imaging studies. Seven of the 97 surgical cases were pathologically confirmed PNETs.
Clinical data regarding age, sex, symptoms at presentation, medical and
surgical management, treatment response, follow-up, and outcome were
obtained. For each case, the initial neuroimaging studies including
computerized tomography (CT) and/or magnetic resonance imaging (MRI)
were reviewed for tumor epicenter, focality, size, extension,
presence of any exophytic component, signal and enhancement
characteristics, and the presence of hydrocephalus and leptomeningeal
or ventricular dissemination. MRI was performed in all cases.
T2-weighted images were obtained in all but one of the cases (case 4).
Gadolinium enhanced studies were available in 6 cases. We adopted a
previously reported definition of focal tumors.37 These
have circumscribed borders on both T1- and T2-weighted images, are
limited to one anatomic segment of the brainstem, and occupy less than
one half of the involved brainstem segment.37 Because
cerebellar PNETs extend into the brainstem in 30% of cases,38 we excluded any such case. Two of our cases,
however, slightly extend beyond one anatomic segment of the brainstem,
but it is slight enough that we are expanding our definition to include these cases. We analyzed the type and extent of the neurosurgical intervention performed. We reviewed the pathologic findings in detail.
In each case immunohistochemical analysis for GFAP (Dako, Carpinteria, CA), vimentin (DAKO), synaptophysin
(Boehringer, Indianapolis, IN), neurofilament protein
(Dako, 2F11), epithelial membrane antigen (Biogenex, San
Ramon, CA), cytokeratin AE1/AE3 (Biogenex), desmin (Biogenex), and
muscle actin (Enzo, New York, NY) were performed if not already on
file. Postmortem examinations were performed in 2 cases (cases 1 and
4). Standard antigen retrieval was used in all of our
immunohistochemical studies.
Screening for p53 Mutation by SSCP and Sequencing
Paraffin-embedded, formalin-fixed tissue was available for DNA
extraction from all 7 cases. Tissues were obtained from surgical specimens in 4 cases and from autopsy material in 2 cases. We also
studied 9 surgical specimens of cerebellar PNET and 9 brainstem gliomas
as controls. Brainstem gliomas included 3 low-grade astrocytomas (LGAs), 3 anaplastic astrocytomas (AAs), and 3 glioblastomas multiforme (GBMs). One of the GBMs was obtained from a postmortem examination. We
also analyzed 3 nonneoplastic controls obtained from resection of
cerebral tissues in epileptic patients. Extraction of
DNA39 and SSCP40 analyses were performed as
previously described. Because the majority of p53 mutations in human
cancers,41 including gliomas, are distributed within exons
5 to 9,42 only these exons were screened by SSCP.
Oligonucleotide primers used for these p53 exons were previously
described.43 Sequencing was performed only when an
abnormal band was detected by SSCP as previously
described.43
Statistical Analysis
The hypothesis that a lower frequency of p53 mutation is
associated with brainstem PNETs compared with brainstem gliomas was tested by the Clinical findings, surgical management, and outcome of the 7 patients are summarized in Table 1. Their
median age was 2.7 (1.0-8.0) years; there were 5 females. The mean
duration of symptoms was 2 (1-6) months. One patient (case 2) had a
subscapular rhabdoid tumor diagnosed previously. All 7 patients
required a ventriculoperitoneal shunt; 5 at diagnosis and another 2 at
recurrence/progression. The median survival was 7 (3-17) months and
all 7 patients have died. Only 2 patients responded to therapy for 4 and 6 months. All 7 patients relapsed and had progression at the
primary site and 6 had additional documented leptomeningeal
dissemination. The seventh patient died abroad without imaging studies
of the spinal canal.
TABLE 1
-tubulin,
synaptophysin, neurofilament protein), epithelial (epithelial membrane
antigen, cytokeratin), or muscular (actin, desmin, myo D-1)
types.9,10,12 Moreover, the multipotential nature of the
PNET cell is well-documented by morphologic evidence of ependymal,
oligodendroglial, melanocytic, mesenchymal, and photoreceptor
differentiation,8,13,14,16-18 regardless of the location
of these tumors.8-10,12 The most common site of origin is
the cerebellar vermis, where such tumors have long been known as
medulloblastoma.19,20 PNETs constitute up to 25% of
central nervous system (CNS) tumors in children and have a high rate of
proliferation and a tendency to metastasize throughout the
leptomeninges.12,21,22
66%) p53
mutation.35,36
METHODS
Top
Abstract
Methods
Results
Discussion
Conclusion
References
2 test.44
RESULTS
Top
Abstract
Methods
Results
Discussion
Conclusion
References
Clinical Findings, Surgical Management, and
Outcome
Neuroradiology
Neuroimaging findings at diagnosis are summarized in Table 2. Neuroimaging studies identified intrinsic focal lesions of the brainstem. All had low signal on T1-weighted images and bright signal on T2-weighted images (Fig 1). Five had their epicenter in the pons and 2 in the brachium pontis. Two cases slightly extended beyond 1 anatomic segment of the brainstem (ie, pons). Thus, these cases did not perfectly fit the previously reported definition of focal tumors37 but were still considered as focal because of the scantiness of the extension beyond the boundaries of the pons. Two cases had tumors that occupied slightly more than one half of the anatomic segment,37 ie, pons, were also called focal because they had, like all the other tumors, well-defined borders on both T1 and T2 images. Gadolinium was administered to 6 patients with none showing tumor enhancement. All tumors had exophytic components to a variable degree. Leptomeningeal dissemination was present in 3 cases at presentation. The tumor seen in the subarachnoid space did not enhance in 2 of 3 cases. Hydrocephalus was seen in 4 cases.
|
|
Histopathology
Homer-Wright rosettes were seen in 1 tumor, multinucleated tumor giant cells were seen in 3 tumors, and 1 showed central lumen rosettes (Fig 2D) ie, Flexner-Wintersteiner rosettes. The results of the immunohistochemical analysis are summarized in Table 3 and are illustrated in Figs 2B and 2C. Four cases had astrocytic differentiation, 2 neuronal differentiation, 2 epithelial differentiation, 1 muscular differentiation, 1 neuroblastic differentiation, and 1 ependymoblastic differentiation. In 2 cases electron microscopy showed that tumor cells had few organelles and no intercellular junctions. Case 1 showed no differentiation by hematoxylin-eosin stain or immunohistochemistry.
|
|
Initial Management
Neurosurgical procedures consisted of a partial resection in 5 cases and a biopsy in 2 cases. For diagnosis, tissue was obtained from the primary tumor in 5 cases, while 2 of the biopsies were performed at sites of distant leptomeningeal dissemination in the cauda equina. Although only 3 patients presented with radiologic evidence of leptomeningeal dissemination, 5 required ventriculo-peritoneal shunts at diagnosis. Treatment consisted of radiotherapy (36 Gy craniospinal and 54 Gy posterior fossa) and multiagent chemotherapy in 4 patients and chemotherapy alone in 3 (CCG infant protocol 9921).
Clinical Course
All 7 patients died at 3 to 17 months after diagnosis, with local recurrence in 7 and diffuse leptomeningeal dissemination in 6 patients. In case 1 although a reduction in tumor size was documented at the site of the primary tumor, it was clear that the tumor continued to grow in the metastatic sites in the craniospinal radiation field. In case 3, a full course of craniospinal radiotherapy was administered and a short response was observed, but within 6 months, local and metastatic recurrences developed that responded poorly to chemotherapy. In case 4 the patient did not respond to chemotherapy and only partially to radiotherapy and died shortly after termination of radiotherapy. One patient (case 7) was initially empirically treated as a brainstem glioma, but when leptomeningeal dissemination was detected in her cauda equina (Fig 1F), a biopsy at the site of one of the lumbar subarachnoid tumors revealed a PNET and chemotherapy was added.
Autopsy Observations
Autopsy neuropathological examinations in cases 1 and 4 revealed extensive leptomeningeal subarachnoid tumor in both the cranial and spinal meninges as well as the primary tumor. The leptomeningeal tumor was manifested both by nodular masses and by diffuse neoplastic infiltrates, the latter especially around each brain base and brainstem. Histologically, in each case, the meningeal neoplasms were typical PNETs with high mitotic rates and foci of ependymal/ependymoblastic differentiation (Fig 2E). In case 1 the intraaxial brainstem tumor was similar, albeit with zones of spindle cell astrocytic differentiation and neuronal differentiation. In case 4, however, the intraaxial pontine mass resembled a LGA, with zones of necrosis associated with vascular changes suggestive of radiation effects.
p53 Status
Results of SSCP for the exons 5 through 9 of p53 are summarized in
Table 4 and are illustrated in Fig
3. None of the 8 cerebellar and 5 brainstem PNETs that amplified had mutations for p53. Mutations were
found in 5 of 8 tumors (62%) from patients with typical brainstem
gliomas. Four gliomas had abnormal mobilities by SSCP (Fig 3).
Sequencing in an AA showed a mutation in exon 8 (Fig 4). A
2 test was performed and yielded a power of
.29. More studies need to be performed with a larger sample; however,
preliminarily it suggests that the absence of p53 mutation may be an
important consideration in diagnosing PNETs.
|
|
|
| |
DISCUSSION |
|---|
|
Top
Abstract
Methods
Results
Discussion
Conclusion
References
|
|---|
We have described clinical, radiologic, pathologic, and some molecular findings of 7 brainstem PNETs that have an exophytic component and are amendable to biopsy. Although they resemble medulloblastomas and other PNETs in terms of their histology, absence of p53 mutations and metastatic potential, their lack of response to chemotherapy and radiotherapy set them biologically apart from cerebellar PNETs. Although our report clearly documents that PNETs of brainstem origin exist, they must be regarded as rare neoplasms. Previous reports of brainstem PNETs include one series of 4 cases originating in the pons,26 another series of 2 in the medulla and 2 in the region of the third cranial nerve,27 and an additional report of 2 cases of brainstem PNETs.25 This supports our recommendation that PNET be included in the differential diagnosis of focal brainstem tumors.5,26,27 Our report is intended to make clinicians aware of the importance of obtaining biopsy material for histopathologic, immunohistochemical, and molecular studies for diagnosis and treatment of brainstem tumors in children.
Clinical Diagnostic Criteria
The features of these 7 cases suggest 5 criteria that are more commonly associated with brainstem PNET rather than brainstem glioma: 1) age <3 years at diagnosis; 2) well-circumscribed focal pontine tumor with low signal on T1, bright signal on T2, and no enhancement after contrast injection; 3) the presence of diffuse subarachnoid spread at diagnosis; 4) the existence of another (non-CNS) primary tumor that suggests a multiple primary tumor syndrome, such as a PNET/rhabdoid syndrome45; and 5) the presence of hydrocephalus at diagnosis.
In addition, it is important to be aware of the difficulty in making a differential diagnosis between rhabdoid tumors and PNETs. One patient that was initially included in our study had a tumor with foci of larger cells with an epithelioid appearance resembling those of an atypical teratoid/rhabdoid tumor.46 This was initially diagnosed as a PNET but was later confirmed to be a rhabdoid tumor, and we, therefore, chose not to include this case in our study.
Clinical Findings
Age at diagnosis is a potential difference between brainstem gliomas and PNETs. The median age of children with typical diffuse brainstem gliomas at diagnosis is ~6 years.1,47-50 Our patients had a median age of 2.7 years at diagnosis. However, of the 7 children described, 3 (42%) were >3 years of age at the time of diagnosis, and 2 of these were in the more typical age range for pontine gliomas of 71/2 and 8 years. With the limited number in our patient population, it is difficult to know how firm one can be about this age prediction.
Neuroradiology
Children with typical pontine gliomas also demonstrate low signal on T1, bright signal on T2, and often have no enhancement after contrast.3,37 Typical pontine astrocytomas are infiltrative diffuse tumors that may enhance with contrast material on neuroimaging studies.49,51 Brainstem PNETs have well-defined borders on MRI or CT, and thus the major distinguishing characteristic is the sharp borders. PNETs arising elsewhere are usually circumscribed.52,53 Although in 2 cases the tumors slightly expanded beyond one anatomic segment of the brainstem (ie, pons) or crossed the midline, all had circumscribed borders on both T1 and T2 images. In general, the appearance of PNETs on MRI is nonspecific and variable.52,54 The most common appearance is of a hypointense mass, compared with normal brain.52,54 A hypodense tumor on T1-weighted images filling the pons may suggest a glioma. However, our cases as well as 3 of 4 previously reported brainstem PNETs26 were hypointense on T1-weighted images. Thus, our report enlarges, in part, the neuroradiologic differential diagnosis of hypodense tumors filling the pons.
The appearance of PNETs on T2-weighted images is variable. Meyers et al53 observed that cerebellar PNETs are isointense to slightly hyperintense to the cerebellar cortex on T2-weighted images. Zimmerman et al54 found that 20 of 42 posterior fossa PNETs were isointense to the cortex on T2-weighted images. The imaging characteristics of 4 brainstem PNETs were found to be similar.27 The absence of enhancement in our series is atypical for PNETs of other sites52,54 and may reflect a unique difference in tumor vascularity or the relatively small size of the tumor at diagnosis. The pathologic mechanisms that control contrast enhancement of brain tumors include breakdown of the blood brain barrier and tumor neovascularization.15,55 PNETs show variable degrees of vascularity,56 and thus have variable degrees of enhancement and may lack contrast enhancement when hypovascular. This might be a feature of PNETs in the brainstem. For example, brainstem medulloepitheliomas, which are variants of PNETs, have been described as having low signal on T1 and as being bright on T2, with no enhancement after injection of contrast material.25 Moreover, only 1 of 4 previously reported brainstem PNETs showed contrast enhancement.26 Thus, the features of the cases that we are reporting are not inconsistent with those of previously described PNETs.11,25-27,52-54
Leptomeningeal Dissemination
Leptomeningeal disease is considered unusual in most children with pontine gliomas, but few studies have prospectively imaged the spine in these children. Therefore, the incidence of subarachnoid spread may be underestimated in these patients. Leptomeningeal spread at diagnosis is distinctly uncommon among patients with brainstem gliomas. The incidence of clinical and radiologic seeding at recurrence, however, ranged from 33% to 50% in several series of brainstem glioma patients.57 The usual pattern of recurrence in pre-MRI reports has been local47 with a variable incidence of concomitant leptomeningeal dissemination. The reported subarachnoid spread in malignant brainstem gliomas ranges from 4% to 50%.47,48,50,57,58 The reported rate of subarachnoid spread in PNETs at presentation is variable but usually higher. For example, it was reported as 28.8%,58 30%,59 and as 43% in postoperative myelogram before radiotherapy.60 Leptomeningeal dissemination was also prominent in 2 of our cases in which an autopsy was performed. Leptomeningeal dissemination occurs in 42.9% to 100% of reported PNET cases.58,61,62 Thus, the rates of leptomeningeal dissemination of 42.9% (3 of 7) and 85.7% (6 of 7) in our patients, at presentation and during the course of disease respectively, are within the reported rates of tumor spread for PNETs in general.
Multiple Primary Tumor Syndrome
One of our patients had a multiple primary tumor syndrome. The association between non-CNS rhabdoid tumors and PNETs has been previously described for kidney tumors and CNS tumors. These included 5 PNETs (3 cerebellar medulloblastomas, 1 pineoblastoma, and 1 cerebral neuroblastoma45).
Hydrocephalus
The presence of a communicating hydrocephalus in patients with malignant CNS tumors is often related to radiographically apparent or occasionally occult leptomeningeal dissemination. The presence of symptomatic hydrocephalus at diagnosis, requiring a V-P shunt in 5 of 7 of our patients, is distinctly unusual in patients with newly diagnosed brainstem gliomas.3
Surgical Neuropathology
There were no unique hematoxylin-eosin patterns or immunohistochemical profiles of brainstem PNETs in our 7 cases, compared with PNETs in other sites. The tumors consisted of mostly undifferentiated cells, but there was histologic and immunohistochemical evidence of astroglial, neuronal, ependymal, muscle, or epithelial differentiation. Three of our cases showed multinucleated giant cells. This occurrence in PNETs has been previously documented.63
Postmortem Findings
In both cases where an autopsy was performed, leptomeningeal dissemination was evident. In case 4, although PNET was found throughout the subarachnoid space, the pontine tumor presented a radically different appearance, namely one of a low-grade fibrillary astrocytoma. This raises the question of separate, coexisting, independent primary tumors. Although this interpretation seems impossible to disprove, we do not favor it because the original biopsy from this site showed the pattern of PNET seen to be disseminated at autopsy. More likely, the neoplasm in the pons at autopsy represented a zone of advanced astrocytic differentiation of the PNET as previously reported.64
p53 in Gliomas and PNETs
Mutations of the p53 gene are rare in PNETs32-34 and other embryonal tumors such as neuroblastoma.65 In contrast, p53 mutations have been described in 5 of 7 (71%)35 and 8 of 13 (61%) cases of pediatric brainstem gliomas. We assayed for exons 5 to 9 of the p53 gene, the hot spots in which 90% of p53 mutations in human cancers including astrocytomas have been observed.41,42,66 SSCP has been extensively used for detection of p53 mutation67 and is unlikely to miss mutations.68 Therefore, the presence of p53 mutations in the limited number of brainstem gliomas that we investigated and the lack of these mutations in brainstem PNETs suggest a distinguishing characteristic between these 2 types of tumors.
Management Considerations
The clinical management of PNETs differs significantly from that of brainstem gliomas. For the most part, diffuse pontine gliomas are clinically diagnosed without resorting to a surgical biopsy and patients are presently treated with involved field radiotherapy alone or in combination with chemotherapy. For PNETs patients receive craniospinal radiotherapy with boosts to the primary site and any bulky CNS metastases. For patients with high-risk disease either subarachnoid spread at diagnosis or subtotally resected tumors, chemotherapy is also used.69 This overall strategy produces long-term progression-free survival in over 50% of children with PNETs arising in the cerebellum. However, children who developed primary PNETs in sites other than the cerebellum seem to have a relatively poorer prognosis.70 Our patients fall into this latter group and all 7 PNET patients have died within 17 months from diagnosis with local recurrences and diffuse leptomeningeal dissemination. Thus, brainstem PNETs seem relatively resistant to the 2 conventional treatment modalities that have helped other children with medulloblastomas, suggesting that these tumors may be more virulent than their counterparts in the cerebellum. Optimal therapy is yet to be defined.
| |
CONCLUSION |
|---|
|
Top
Abstract
Methods
Results
Discussion
Conclusion
References
|
|---|
The clinical, radiologic, and pathologic features and p53 analysis results in 7 cases of exophytic brainstem PNETs identify a separate type of tumor than the usually encountered brainstem glioma. The important clinical aspects include: 1) the localized, nonenhancing MRI characteristics; 2) the predilection for leptomeningeal dissemination; 3) the young age at diagnosis; and 4) the overall poor prognosis. Our present management of patients with a brainstem tumor includes an MRI without and with gadolinium of both the brain and spinal canal. If the child has an entirely intrinsic diffuse tumor of the brainstem with its epicenter in the pons and/or medulla without any evidence of leptomeningeal dissemination, a clinical diagnosis of a glioma is made and the child is treated with high-dose focal irradiation. We encourage biopsies in cases where clinical and radiologic features described above are present. This could also help further understanding of the pathobiology of brainstem tumors in children. Our cases broaden the differential diagnosis of intrinsic brainstem masses in children. These cases emphasize the importance of a histopathologic diagnosis in brainstem tumors with unusual presentations but do not support the routine biopsy of all intrinsic brainstem tumors.
| |
ACKNOWLEDGMENTS |
|---|
This work was supported in part by Grant CA 50434 from the National Institutes of Health.
We thank Stephanie Bruegman for her help in preparing this manuscript, and Henry Cohen, MD, and Joanne Scalzitti, PhD, for statistical analysis.
| |
FOOTNOTES |
|---|
Received for publication Jan 29, 1999; accepted Apr 17, 2000.
Reprint requests to (D.Z.) Department of Pathology, Division of Neuropathology, New York University School of Medicine, 550 First Ave, New York, NY 10016. E-mail: dz4{at}is3.nyu.edu
| |
ABBREVIATIONS |
|---|
PNET, primitive neuroectodermal tumor; GFAP, glial fibrillary acidic protein; CNS, central nervous system; SSCP, single-strand conformation polymorphism; CT, computed tomography; MRI, magnetic resonance imaging; LGA, low-grade astrocytoma; AA, anaplastic astrocytoma; GBM, glioblastoma multiforme.
| |
REFERENCES |
|---|
|
Top
Abstract
Methods
Results
Discussion
Conclusion
References
|
|---|
This article has been cited by other articles:
 |
|
 |
|
  S. S. Donaldson, F. Laningham, and P. G. Fisher Advances Toward an Understanding of Brainstem Gliomas J. Clin. Oncol., March 10, 2006; 24(8): 1266 - 1272. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Law, K. Kazmi, S. Wetzel, E. Wang, C. Iacob, D. Zagzag, J. G. Golfinos, and G. Johnson Dynamic Susceptibility Contrast-Enhanced Perfusion and Conventional MR Imaging Findings for Adult Patients with Cerebral Primitive Neuroectodermal Tumors AJNR Am. J. Neuroradiol., June 1, 2004; 25(6): 997 - 1005. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
||
|
||
Home | My Pediatrics | Journal Information | Current Issue | Past Issues | Subscriptions & Services | Contact Us Click here for faster international access © 2000 American Academy of Pediatrics. All rights reserved. |
||
|
||
