RADIOLOGÍA

Radiología. 2011;53(2):134-145
ISSN: 0033-8338
RADIOLOGÍA
RADIOLOGÍA
Publicación Oficial de la Sociedad Española de Radiología Médica
Incluida en Index Medicus/MEDLINE
www.elsevier.es/rx
Actividad acreditada en base a la encomienda de
gestión concedida por los Ministerios de Educación,
Cultura y Deporte y de Sanidad y Consumo al
Consejo General de Colegios O ciales de Médicos
con 1 crédito, equivalente a 4 horas lectivas.
www.seram.es
www.elsevier.es/rx
UPDATE
Review and update about medulloblastoma in children
M.I. Martínez León
Sección de Radiología Pediátrica, Servicio de Radiodiagnóstico, Hospital Materno-Infantil, Complejo Hospitalario Universitario
Carlos Haya, Málaga, Spain
Received 5 October 2010; accepted 11 November 2010
KEYWORDS
Medulloblastoma;
Pediatrics;
Central nervous
system tumors;
Magnetic resonance
imaging
Abstract Medulloblastoma is the most common malignant CNS tumor in children. Although all
medulloblastomas are classi ed as grade IV lesions, the wide histological and molecular variation
among these tumors means that the risk and prognosis involved also vary widely. Imaging studies
are important not only because the initial diagnostic evaluation indicates what type of surgery
will be performed and has prognostic value, but also because it in uences the postoperative
treatment approach, providing, among other details, information about the dissemination of
disease and remnants of the tumor after surgery, which are both risk factors in medulloblastomas.
Improvements in our understanding of the biological and molecular characteristics of
medulloblastoma promise a dramatic change in the accuracy of staging and treatment of this
tumor in the near future that are sure to bring about further improvements in survival.
© 2010 SERAM. Published by Elsevier España, S.L. All rights reserved.
PALABRAS CLAVE
Meduloblastoma;
Pediatría;
Tumores del SNC;
Resonancia magnética
Meduloblastoma pediátrico, revisión y puesta al día
Resumen El meduloblastoma es el tumor maligno más frecuente del SNC en pediatría. Presenta
una variabilidad histológica y molecular tan importante que hace que, aunque todos los tipos de
meduloblastoma se clasi quen como grado IV, el riesgo y el pronóstico de los mismos sean muy
amplios. Las pruebas de imagen son importantes no sólo porque la valoración diagnóstica inicial
indica el tipo de cirugía a realizar, que presenta un valor pronóstico, sino para el planteamiento
terapéutico posterior, teniendo en cuenta que determinan, entre otras, la diseminación y el
resto tumoral posquirúrgico, factores de riesgo en este tumor. Una mejor comprensión de las
características biológicas y moleculares del meduloblastoma promete un cambio dramático
hacia la precisión en la estadificación y el tratamiento tumoral en un futuro próximo,
favoreciendo aún más la actual mejora de la supervivencia.
© 2010 SERAM. Publicado por Elsevier España, S.L. Todos los derechos reservados.
E-mail: marcela338@hotmail.com
0033-8338/$ - see front matter © 2010 SERAM. Published by Elsevier España, S.L. All rights reserved.
Review and update about medulloblastoma in children
Introduction
Medulloblastoma is the most common pediatric central
nervous system (CNS)
Malignancy. 1-4 It occurs more frequently in males (ratio
1.5:1) and usually before 10 years of age. Although much
less common, the disease may also occur in adults, usually
in the 3rd and 4th decades of life. 2,5
This condition was initially described as cerebellar glioma
until Bailey and Cushing named it medulloblastoma in 1925. 6
Now it is included in the group of embryonal tumors
(grade IV) of the World Health Organization (WHO)
classi cation. 7,8
Clinical characteristics
Clinical manifestations depend on patient’s age and extension
of the disease, local or disseminated, and are usually brief—
less than three months—re ecting the aggressive biologic
behavior of the tumor. 2 Clinical signs include headache,
vomiting, papilledema, irritability, diplopia, nistagmus, and
increased head perimeter in younger infants, and are due to
increased intracranial pressure related to hydrocephalus
secondary to tumor obstruction.1
Clinical signs are also related to the vermian location of
the tumor that results in ataxia, frequently accompanied by
spasticity, and gait instability. A laterally located mass, more
common in older children and adults, manifests as limb
ataxia and dysdiadokokinesis. In older children, the first
symptoms are usually headaches that worsen in the supine
position, start in the morning and may improve after
vomiting. Pressure from the hydrocephalus on the dorsal
brainstem may result in Parinaud’s syndrome, vertical gaze
palsy and pupils reactive to accommodation, but not to
light. 1,9 Abducens nerve palsy is common and may result
from compression of the partially exposed nucleus of the VI
cranial nerve along the anterior margin of the IV ventricle,
or from traction or pressure of this nerve in its course along
the cranial base. 10
Manifestations of disseminated disease are related to
metastatic location and include signs and symptoms of
spinal cord compression resulting from spread to the spinal
canal or convulsions when there is hemispheric spread.
Associated syndromes
Medulloblastoma is associated with nevoid basal cell
carcinoma (Gorlin-Gotz syndrome), Turcot syndrome -type
2-, Li-Fraumeni syndrome, neuro bromatosis types 1 and 2,
Rubinstein-Taybi syndrome, Fanconi anemia and Nijmegen
syndrome. 1,4,11
Part of the information on the molecular pathways related
to medulloblastoma comes from the study of two syndromes
associated with a constitutional predisposition to
medulloblastoma—Gorlin syndrome and Turcot syndrome.1,4,12
Radiological ndings
Medulloblastomas arise from the midline vermis, the most
common location, and grow into the IV ventricle. Cerebellar
135
hemispheric location is more common in adults and older
children. These variations in location may be explained by
the different cellular origins of medulloblastoma, by the
different progenitor cell pools of the cerebellum, and by
the cellular signaling pathways that control its development
that probably represent different compartments from which
the various types of medulloblastoma arise. In addition,
initially the migration of the origin cells starts from the
posterior medullary velum toward a superior location, close
to the midline, giving rise to tumors centered in vermis.
Later in life, the cells migrate laterally so in older patients
the tumor appears in the hemisphere. Less common
locations include the IV ventricle (3 %) and supratentorial
compartment (2 %). 2,5 Computed tomographic (CT)
appearance is hyperattenuating due to its high cellularity.
Calcium is present in a low percentage (22 %) and peritumoral
edema is variable ( g. 1). The initial CT may demonstrate
tumor complications including the degree of hydrocephalus
secondary to obstruction and signs of intracranial
hypertension.
MRI appearance is hypointense on T1-weighted sequences
and hypo- to isointense on T2. This tumor is predominantly
homogeneous, with little necrotic, hemorrhagic or calcic
component. Enhancement is usually homogeneous after
contrast administration; however, poor and heterogeneous
enhancement has also been described 2,13 ( g. 2).
Brainstem infiltration is common (33 % of 144 cases
published by Park et al) 14 and complicates complete tumor
resection. Although uncommon, foraminal extension with
involvement of the cerebellopontine angle cistern, cisterna
magna or other cisternal compartments may occur 3 ( g. 3).
It has been reported that signal heterogeneity on
T2 sequences with “honeycomb” pattern is associated with
anaplastic medulloblastoma with a sensitivity of 100 %, but
lower speci city. 15 Similarly, a “grape-like” appearance has
been described in the medulloblastoma with extensive
nodularity (fig. 4). Since this subtype has neuronal
differentiation, so it was previously called “cerebellar
neuroblastoma”,16 some studies have reported higher uptake
of iodine- 123-metaiodobenzylguanidin in this variant
compared to other subtypes. 17
The presence of diffusion restriction is usually found,
appearing as high signal on b1000 images and low signal on
the apparent diffusion coef cient (ADC) maps. This is due to
the high intrinsec cellularity of the tumor, decreased
extracellular space and high nucleus/cytoplasm ratio. Given
the fact that the higher the cellularity, the more histological
aggressiveness, the ADC map may be used for tumor grading.
In this respect, low signal on ADC map in pediatric patients
is very characteristic of high grade embryonal tumors
including medulloblastoma 11,18-21 ( g. 2d and e).
Diffusion imaging is helpful in the initial de nition of the
tumor and in the assessment of the progression of those
neoplasms that do not enhance with contrast before or after
treatment. Chemotherapy increases the signal on the ADC
map since the cellularity decreases and the extracellular
space increases, but the coef cient decreases again in case
of recurrence. Thus, diffusion imaging provides
complementary data for monitoring tumor response to
therapy. 20,22 Pretreatment diffusion values may predict the
tumor response to radiotherapy and may differentiate
radiation-induced lesion from recurrence. 18
136
M.I. Martínez León
Figure 1 Presence of calcium within the tumor. a) Unenhanced CT scan shows small scattered calci cations within the tumor mass
corresponding to a medulloblastoma in a 10-year-old patient. In a pediatric patient, this type of calci cations must not be confused
with those that appear in the plexus of the IV ventricle at later ages. Increased density of medulloblastoma in relation to the
surrounding parenchyma. Signs of hydrocephalus secondary to compression, like development of temporal horns. b) Gradient echo
MRI sequence for evaluating the presence of calcium. A different patient with millimetric signal-void areas in the tumor periphery.
(Image provided by Dr. Elida Vázquez Méndez, Hospital Vall d’Hebron).
MR spectroscopy shows elevated choline (Cho) levels,
higher than those of other posterior fossa masses, indicative
of high turnover of tumor cells. There is also a marked
decrease of N-acetyl aspartate (NAA), resulting in a high
Cho/NAA ratio; and a marked increase of Cho/creatine (Cr)
ratio and significant decrease of NAA/Cr 1,13,23,24 (fig. 5a).
Sometimes, lipid resonance indicative of necrosis is present
in rapidly proliferating tumors, including malignant
embryonal tumors, although not specifically in
medulloblastoma. 24,25
It should be noted the significantly elevated taurine
concentrations compared to other tumors. 21,23,25-27 The
hypotheses that explain the presence of taurine include the
aggressiveness of medulloblastoma, an active metabolite
associated with increased cellular proliferation and the loss
of differentiation of tumor cells. 26 Taurine is detected at a
frequency of 3.36 (3.4) ppm. Although there are few studies
published, taurine is observed at lower concentrations in
the desmoplastic/nodular variant than in more aggressive
subtypes of medulloblastoma 23 ( g. 5b).
Spectroscopy may also help, as it does in other tumors, in
the differentiation of radiation-induced necrosis and tumor
recurrence. Post radiation necrosis shows low Cho peak and
normal NAA. 1,28
Perfusion assesses hemodynamic changes in the cerebral
microcirculation. Its main application in neoplasms is
preoperative tumor grading. 24 Relative cerebral blood
volume (rCBV) is signi cantly higher in high-grade than in
low-grade tumors—secondary to tumor neovascularization—
irrespective of the degree of integrity of the blood-brain
barrier (BBB). This information is not provided by
conventional contrast-enhanced MR imaging, in which the
presence and degree of enhancement correlate with both
vascular hyperplasia and BBB status 13,24 ( g. 6). This explains
why the degree of enhancement does not correlate with
tumor grading. For example, a low-grade tumor such as the
pilocytic astrocytoma enhances intensely, whereas a
high-grade tumor such as medulloblastoma may not enhance
as intensely but may show high perfusion with increased
rCBV. In addition, perfusion has the potential to identify
disease progression, characterized by increased rCBV values
in comparison with stable lesions whose rCBV does not
change signi cantly over time. 13,24
Increased 18F fluorodesoxyglucose (FDG) uptake is
observed on PET scans of the brain and spinal cord in
disseminated medulloblastomas. 29 Cerebrospinal uid (CSF)
dissemination has to be ruled out at the initial evaluation,
as medulloblastoma is the pediatric tumor with the highest
rate of neuraxis dissemination. 30 Metastatic spread is
present in about 30 % of patients at diagnosis, with higher
probability in younger children. 12 This examination has the
highest sensitivity and specificity in the perioperative
period, before oncologic treatment planning. Current
diagnosis of dissemination is performed by:
1. Spinal RM imaging, preferably preoperative or two weeks
after surgery. If no preoperative spinal study was
performed, it is important to delay the spinal MRI two
weeks to reduce false positives caused by subarachnoid
blood and postsurgical irritation. 31
Review and update about medulloblastoma in children
137
Figure 2 MR imaging characteristic of medulloblastoma. a) Sagittal T1 MRI. b) Axial T2 MRI. c) Axial Flair MRI. d) Diffusion MRI.
e) ADC map. Infratentorial midline mass that arises from the posterior medullary velum and extends into the IV ventricle.
Hypointense T1 signal, slight signal increase in T2 and signal similar to the adjacent parenchyma in FLAIR. Solid tumor with minor
signal heterogeneity secondary to eccentric cyst. There is no assessable edema. Amygdala herniation through the foramen magnum
secondary to tumor mass effect (long arrow). Mild supratentorial ventricular dilation in the sagittal plane at the III ventricle.
Marked diffusion restriction of medulloblastoma (compared to the absence of restriction in the parenchymal cyst) (short arrow) and
the corresponding low signal on the ADC map (and high signal of the intratumoral cyst).
2. CSF cytologic analysis via lumbar puncture approximately
15-20 days after surgery. In previous years, the site of CSF
sampling, lumbar or cranial (through previously placed
catheters for hydrocephalus) was a subject of much
discussion; lumbar sampling is currently preferred. 30,32,33
MR imaging of the neuraxis has greater diagnostic
accuracy than CSF cytologic analysis in the detection of
medulloblastoma; in this respect, statistical studies have
proven that CSF cytology does not show disseminated tumor
when MRI of the neuraxis is negative. 30
MRI of the neuraxis generally includes T1 sequences with
gadolinium gadopentate dimeglumine (Gd-DTPA), with or
without fat suppression, depending on the institution and
protocol, but excellent results have been published for
diffusion imaging, which seems to be more sensitive than
contrast imaging in the detection of disseminated
medulloblastoma. 21
Images showing intradural extramedullary spread usually
correspond to linear pial lesions, but also to isolated
subarachnoid nodules and drop metastases in the caudal
region of the spinal canal, or to actual tumor masses
occupying the canal, 29 all showing contrast enhancement or
restricted diffusion ( g. 7).
Although extraneural metastases of CNS tumors are rare,
they are characteristic of medulloblastoma with a
prevalence close to 7 %, with bone marrow location (77 %)
followed by lymph nodes (33 %). 34
138
M.I. Martínez León
Figure 3 Although it may happen, foraminal extension involving the cerebellopontine angle cistern, cisterna magna or other
cisternal compartments is unusual in medulloblastoma, allowing differentiation from ependymoma, an infratentorial midline tumor
that shows this feature. a) Axial T2 MRI fast sequence shows tumor extension into the prebulbar cistern (short arrow) and foramina
of Lushka and Magendie (long arrows). b) Axial T2 MRI shows a medulloblastoma partially occupying the left cerebellopontine angle
cistern (arrow).
Differential diagnosis
The main differential diagnoses of pediatric posterior
fossa masses include medulloblastoma, astrocytoma
pilocytic and ependymoma. Atypical teratoid–rhabdoid
tumors (ATRT) also need to be considered in certain
instances.
Consideration of medulloblastoma within the
differential diagnosis mandates an aggressive surgical
approach as residual tumor is one of the prognostic
factors. If medulloblastoma is suspected, preoperative
imaging must include the neuroaxis given the high
propensity to disseminate. 3 For this reason, differentiation
with the above-mentioned tumors is important, with a
different pre- and postoperative imaging approach.
Unlike medulloblastoma, ependymoma, pilocytic
astrocytoma and ATRT generally occupy cerebellopontine
angle and adjacent cisterns. Ependymoma and pediatric
medulloblastoma are usually midline tumors, whereas
astrocytoma and ATRT are usually eccentric.
Figure 4 Relation between imaging findings and histological findings. a) Contrast-enhanced T1 MRI sequence shows a
medulloblastoma with nodular, “grapelike” appearance. No dissemination is observed. The diagnosis of extensive nodullarity variant
was considered, but this patient was 7-year-old and this condition usually appears in younger children. Pathologic results
demonstrated desmoplastic/nodular subtype. b) b1, T2 MRI sequence; b2, Contrast-enhanced axial T1 MRI; b3, Contrast-enhanced
T1 MRI with magni cation along the sagittal plane. Patient with medulloblastoma with marked heterogeneity on T2 sequences
(small cysts, internal areas of necrosis and heterogeneous signal of the solid component of the tumor) as well as an enhancement
pattern that may be considered as “honeycomb”. Consideration was given to the possibility of an aggressive variant. In addition,
the lesion showed other features of poor prognosis including leptomeningeal dissemination at the time of diagnosis. The diagnosis
of anaplastic medulloblastoma was considered and subsequently con rmed by pathologic examination.
Review and update about medulloblastoma in children
139
Figure 5 Hydrogen MR (1.5 Tesla). a) Single-voxel acquisition with short TE (23). b) Multivoxel acquisition. Horizontal arrow:
choline (3.22 ppm). Short arrow: creatine (3.04 ppm). Long arrow: NAA (2.02 ppm). Thick arrow: taurine (3.36 ppm). Each gure
corresponds to a different medulloblastoma, but both spectra show elevated choline peaks, marked reduction of NAA and decreased
creatine. The presence of taurine is observed in the second study, corresponding to a desmoplastic/nodular variant, which exhibits
lower taurine peaks than the most aggressive variants of medulloblastoma.
The most likely alternative consideration for a
hyperattenuated midline mass due to the presence of
calcium is an ependymoma. In contrast to medulloblastomas,
ependymomas typically show more calci cation and display
a “plastic” morphology, extending and conforming from the
IV ventricle through the foramina of Luschka and Magendie
into the cerebellopontine angle cistern. 2,3,13
Diffusion MRI shows some similarities among the three
types of tumors, particularly in their pathological variants,
incluiding the lower restriction of desmoplastic
medulloblastoma compared to other subtypes of the tumor, 35
the increased restriction in two thirds of anaplastic
ependymomas (WHO grade III) and 50 % of classic
ependymomas (WHO grade II), or the restriction described
in a considerable percentage of solid nodules of pilocytic
astrocytoma.
Although diffusion is a useful modality for the
assessment of pediatric posterior fossa tumors (evident
restricted diffusion of medulloblastoma compared to
other tumors), additional features (age, location,
Figure 6 Vermian medulloblastoma in a 7-year-old patient that shows intense and homogeneous enhancement at T1 MRI with
Gd-GPDM (a) and elevated rCBV at contrast-enhanced perfusion imaging (b), indicative of loss of BBB integrity and increased tumor
angiogenesis, respectively. (Images provided by Dr. Elida Vázquez Méndez, Hospital Vall d’Hebron).
140
M.I. Martínez León
Figure 7 MR imaging of the neuraxis. a) Spinal contrast-enhanced T1 MRI without fat suppression in the axial plane. Preoperative
assessment shows linear and nodular thickening in the dorsal aspect of the spinal cord related to neuraxis dissemination. Anaplastic
medulloblastoma. b) A 6-year-old girl with posterior fossa medulloblastoma. Diffusion MR image shows increased signal secondary to
restriction in the upper chest region caused by dissemination in the CSF (arrows). (Image provided by Dr. Elida Vázquez Méndez,
Hospital Vall d’Hebron).
morphology, dissemination) are needed for tumor
differentiation. 18,19
Differentiation between some midline pilocytic
astrocytomas with signi cant solid component, heterogeneity
and intense contrast enhancement, and medulloblastomas
would be dif cult if it were not for diffusion imaging, which
depicts medulloblastomas with no restriction and intense
brightness on ADC maps. 5,36 On MR spectroscopy, the
presence of taurine in the medulloblastoma helps distinguish
from cerebellar astrocytoma, which shows no taurine. 25,27
Although both ATRT and medulloblastoma are embryonal
tumors, ATRT presents earlier (mean 1.3 years) with a
remarkably more aggressive course and lower response
to treatment; its differentiation is thus relevant for
prognosis. Despite their similar neuroimaging features,
the typical infratentorial location of ATRT is in the
cerebellopontine angle (more hemispheric and less centric
than medulloblastoma) and exhibits more intralesional
hemorrhage than medulloblastoma. Diffusion imaging is
not useful for differential diagnosis since both tumors are
hypercellular; therefore, showing marked restricted
diffusion 18. In short, if a posterior fossa mass in a young
child displays marked heterogeneity, restricted diffusion
and cerebellopontine angle location, ATRT is the first
diagnostic consideration. 36
Pathologic characteristics
The 2007 WHO classification of tumors recognizes
medulloblastoma, primitive neuroectodermal tumor (PNET)
of the CNS and ATRT as embryonal tumors; all grade IV
lesions. 7,8
Two new histological variants with different clinical
behavior have been incorporated: medulloblastoma with
extensive nodularity, associated with a favourable prognosis;
and the anaplastic type, with poor prognosis. Other subtypes
previously described are classic, desmoplastic/nodular and
large cells. 7,8,37 Medulloblastoma with extensive nodularity
is related to the desmoplastic/nodulary subtype, but the
former has better prognosis, appears earlier, whereas the
latter occurs in older children, and tends to hemispheric
location. Both tumors are histologically more favourable
than the other variants. The anaplastic subtype shows the
highest degree of atypia. Large cell variant also carries poor
Review and update about medulloblastoma in children
prognosis and shows similar cytologic features than the
anaplastic form. 1,7,38
Prognostic factors
Identifying the pathobiological correlation of clinical
behavior or the therapeutic response currently represents
the key for medulloblastoma research. 4
The diagnostic factors that have been considered to date
are:
1. The degree of surgical resection. Complete surgical
resection is considered the best prognostic factor.
Macroscopic tumor excision is defined as absence of
residual lesion at the rst postoperative (48-72h) MRI 39
( g. 8).
2. Presence of metastasis at diagnosis, based on radiological
ndings and CSF analysis 29. Staging is done according to
the modi ed Chang classi cation: M0, no dissemination,
and M1 to M4, disseminated disease. 40
3. Age. Children under three have poorer prognosis,
probably due to the extension of the tumor that makes
excision more difficult and to the difficulty of giving
radiotherapy at this age. 1
According to these parameters patients are divided into
two risk groups: 4,41-43
1. Average-risk patients, with the following features: older
than three years, non-metastatic disease at diagnosis
(M0) and total or nearly total resection (residual
tumor < 1.5 cm 2).
2. High-risk patients present at least one of these features:
younger than three years, metastasis at diagnosis or
residual tumor > 1.5 cm 2.
Histopathologic and molecular pro les are incorporated
into the three classic clinical parameters of risk since many
current studies consider that these pro les may be evaluated
for risk stratification and, thus, influence the biological
behaviour and response to treatment. 12,44
A major drawback is that the classi cation according to
the classic parameters does not differentiate high- from
low- histological risk patients within the same clinical
stage. 12 It is a noted fact that patients with the same clinical
stage, receiving similar treatment, may show widely
disparate outcomes, depending on the biological differences
within the tumor. In addition, two different studies, the
German trial HIT’91 45 and the Children’s Cancer Group 921, 46
established that overall survival was not significantly
different between children staged M0 and M1. It has been
even proven that brainstem invasion, previously regarded as
a high-risk indicator, does not affect prognosis. These are
only two of the many examples showing the impact of tumor
biology on prognosis.
The histological factors currently included in the
determination of the outcome are: 12,44,47-49
1. Histopathological variant: desmoplastic and extensive
nodularity subtypes show better outcome than anaplastic and
large cell subtypes. The classic form has intermediate risk.
141
2. Extent and degree of nodularity: only medulloblastoma
with extensive nodularity is associated with better
survival.
3. Anaplasia: the most aggressive biological feature.
Presence and degree of anaplasia are associated with the
poorest outcome.
Molecular and cytogenetic factors considered for
prognostic assessment are: 44,47
1. Chromosomal abnormalities: the most common is loss of
genetic material of chromosome 17p, which corresponds
to the location of a suppressor gene, so its deletion leads
to tumor expresion. 49 Sometimes this deletion occurs as
a component of 17q, forming an isochromosome (i17q).
17p loss and isochromosome 17q are more frequent in
classic and anaplastic subtypes than in the desmoplastic
one.
2. Genetic alterations: ampli cation of c-myc and N-myc is
considered a very unfavorable factor usually associated
with the anaplastic and large cell subtypes.
3. Embryonic development and oncogenesis: mutation in
the PTCH gene, involved in the SHH (sonic Hedghog)
pathway, is essential for the development of the
cerebellar granular layer. PTCH mutation is identi ed in
the Gorlin syndrome, but also in 10-15 % of sporadic
medulloblastomas. 50
4. Abnormalities in cellular signal pathways: research
studies are investigating the role of the expression of
neurotrophins such as Trk C and their role in neuronal
development, the expression of the epidermal growth
factor receptor (ErbB2) as well as the involvement of
the SHH/PTCH signals and Wnt/Wg pathways among
others.
Other chromosomal abnormalities and molecular
aberrations have been described but are beyond the
scope of this work. Readers are referred to specialized
literature. 4,43,47,48,51-53
Treatment
Tr e a t m e n t i s b a s e d o n t h r e e m a i n s t a y s : s u r g e r y,
chemotherapy and radiotherapy. Signi cant variations are
managed in relation to the different risk factors above
mentioned.
There are several ways of grouping patients in order to
include them into the different treatments. The proposal of
some authors is: 12,54
1. Average-risk medulloblastoma in children older than
three years. Craniospinal irradiation and radiation boosts
of the posterior fossa in conjunction with different
chemotherapy regimens depending on the protocol into
which the patient is enrolled or the protocol in process
(the last protocol was the HIT-SIOP PNET 4, CNS 2003 05,
now closed). 54 A signi cant percentage of these patients
are curable. 1
2. High-risk medulloblastoma in children older than three
years. Intensi ed radiotherapy and chemotherapy with
modi ed regimens compared to the previous group.
142
M.I. Martínez León
Figure 8 Medulloblastoma with suprantentorial dissemination at the time of diagnosis. High-risk medulloblastoma. Axial T1 (a) and
sagittal (b) sequences after Gd-DTPA administration. Contrast-enhanced subcentimeter leptomeningeal nodules: one in a left
paralateral deep temporal location (horizontal arrow) and another in a right midline frontal location (vertical arrow). Infratentorial
medline mass is also depicted.
3. Medulloblastoma in infants younger than three years.
These protocols are continuously under review due to
the relative rarity of this tumor in this age group, and
multicenter prospective trials are needed in this
respect. 55-58 Although radiotherapy presents particularly
adverse effects in these patients, they seem to bene t
from proton radiotherapy. 55
Treatment of recurrence usually depends on the age,
extent of the recurrent disease and type of treatment
received after initial diagnosis. 1,58,59
Follow-up imaging studies
Follow-up imaging studies are intended to identify early
relapses, so secondary therapies may potentially improve
the outcome. In addition, medulloblastoma follow-up may
be considered as part of the treatment given its high
recurrence rate. 39 Recurrence is the major cause of death in
patients with medulloblastoma. 2,60
Medulloblastoma is a very aggressive malignancy with a
high rate of local recurrence associated with incomplete
tumor resection that results in signi cantly reduced time to
recurrence. 39 Most common sites of recurrence are the
posterior fossa, followed by spinal and supratentorial
locations 61 ( g. 9); infrequently, systemic spread (bone and
lymph nodes) may also occur 34. This makes necessary the
follow-up of these children until adulthood, 62 although most
recurrences appear two years after the initial diagnosis in
the posterior fossa. 1,2,63 The longest latency period reported
is 19 year after diagnosis. 64
There are numerous follow-up imaging protocols. The
present work describes the one implemented at the GOSH
of London, result of an extensive and detailed neuroimaging
study on medulloblastoma follow-up. 39,65 This protocol
includes a postoperative cranial MRI acquired within 24-48h
(in other centres 72h). MRI is repeated at 2–3 months during
the rst year and then every 6-8 months in the following
years (after six but before 7.5 years) or until a recurrence
appears or the child is discharged. Equivocal results require
closer follow-up (every 2 or 3 months).
Initially, MRI of the neuraxis was performed only in case of
dissemination at diagnosis or detection of recurrence. In the
current protocol, all the imaging studies include complete
evaluation of the neuraxis because, although they found no
children with spinal recurrence if there was not intracraneal
recurrence—in other words, neuraxis relapse does not
appear isolated—, they believe that imaging the neuraxis at
every follow-up entails more bene ts than drawbacks. 39,66
Other protocols propose different timing for neuraxis
follow-up.
Sequelae
Most sequelae correspond to neuropsychological disorders
and neurocognitive deterioration, 67,68 more severe in
younger children. 57 Many medulloblastoma survivors have
severe sequelae secondary to the tumor or the treatment. 69
Patient’s age at the time of craniospinal radiation and
radiation dose are known neurotoxic factors. Endocrine
follow-up is important due to the risk of endocrine disorders
secondary to holocranial irradiation.
An interesting trial has tried to determine by diffusion
tensor MR imaging the loss of white matter anisotropy
caused by radiotherapy, concluding that the mean value of
white matter anisotropy is potentially useful to evaluate
neurotoxicity. 70
Secondary tumors are also described in medical literature
as side-effects of radiotherapy, as well as vascular
malformations and white matter lesions. 2,68,69,71
Review and update about medulloblastoma in children
143
Figure 9 Immediate postoperative MRI for assessment of residual tumor, one of the main prognostic factors. a) No residual tumor
on MR imaging at 48h: a1, sagittal unenhanced T1 MRI sequence shows residual blood in the surgical bed; a2, sagittal
contrast-enhanced T1 RMI shows no enhancement in a lesion that enhanced before surgery; a3, axial contrast-enhanced T1 MR
image shows any mass or enhancement in the surgical area. b) MR image shows residual tumor < 1.5 cm 2 at 72h: b1, axial
contrast-enhanced T1 MRI; b2, diffusion restriction of the residual tumor. The residual tumor appears as two right millimetric
nodules paralateral to the surgical bed, which together were considered < 1.5 cm 2. c) Early postoperative MR image shows residual
tumor > 1.5 cm 2: c1, axial contrast-enhanced T1 MRI; c2, diffusion restriction at the same level; c3, coronal contrast-enhanced
T1 MRI. Residual tumor was larger than the 1.5 cm 2 limit.
Conclusions
Five-year survival rates for patients with medulloblastoma
have improved dramatically from the ‘70s (2-30 %) to the
present (50-85 %). 1,2,12
Imaging plays a prominent role in the assessment of the
prognostic factors since the degree of tumor resection and
the presence of dissemination at the time of diagnosis are
evaluated by imaging techniques. The use of diffusion
sequences in the study of leptomeningeal spread represents
an advance of the technique that helps in tumor staging and
follow-up.
Imaging follow-up is essential given the high recurrence
rate, so much so that some authors call it “treatment
protocol” instead of “follow-up protocol”. A significant
change has been introduced into the prognostic evaluation,
so in addition to clinical factors (metastasis, residual tumor,
age), biological parameters (histologic variant, degree of
anaplasia and nodularity extent) as well as molecular and
cytogenetic parameters have been included, although still
under research but with a very promising future for the
assessment of tumor prognosis and treatment. 51
It seeems unjusti ed to include all medulloblastomas,
highly aggressive neoplasms, into WHO grade IV. The new
tend is to consider medulloblastoma not as a single entity,
but as a complex group of tumors biologically different but
with morphological similarities, which entails
individualization of the treatment in order to increase
survival rates and reduce long-term toxicity and
morbidity.
Authorship
Dr. Martínez León takes responsibility for the integrity,
conception, design, analysis and interpretation of the data,
as well as for the bibliographical research, drafting of the
manuscript and approval of the nal version.
Con ict of interest
The author declares no con ict of interest.
144
Acknowledgments
I would like to thank Dr. M. Dolores Domínguez Pinos for her
contribution to the acquisition of data and to Dr. Ofelia Cruz
Martínez, pediatric oncologist, for her scienti c advice and
her immeasurable help.
References
1. Dahll G. Medulloblastoma. J Child Neurol. 2009;24:1418-30.
2. Koeller KK, Rushing EJ. From the archives of the AFIP:
medulloblastoma: a comprehensive review with radiologicpathologic correlation. Radiographics. 2003;23:1613-37.
3. Eran A, Ozturk A, Aygun N, Izbudak I. Medulloblastoma: atypical
CT and MRI findings in children. Pediatr Radiol. 2010;40:
1254-62.
4. Ellison DW, Onilude OE, Lindsey JC, Lusher ME, Weston CL,
Taylor RE, et al. Beta-Catenin status predicts a favorable
outcome in childhood medulloblastoma: the United Kingdom
Children’s Cancer Study Group Brain Tumour Committee. J Clin
Oncol. 2005;23:7951-7.
5. Roberts RO, Lynch CF, Jones MP, Hart MN. Medulloblastoma: a
population-based study of 532 cases. J Neuropathol Exp Neurol.
1991;50:134-44.
6. Bailey P, Cushing H. Medulloblastoma Cerebelli: a common type
of midcerebellar glioma of childhood. Arch Neurol Psychiatry.
1925;14:192-224.
7. Louis DL, Ohgaki H, Wiestler OD, Cavenee WK, Burger PC,
Jouvert A, et al. The 2007 WHO classi cation of tumours of the
Central Nervous System. Acta Neuropathol. 2007;114:97-109.
8. In: Louis DL, Ohgaki H, Wiestler OD, Cavenee WK, editors.
World Health Organization Classi cation of Tumours of the
Central Nervous System. Lyon: IARC; 2007.
9. Moguel-Ancheita S, Ruiz-Mor n I, Pedraza-Jacob M. Síndrome
de Parinaud asociado a otros estrabismos. Cir Ciruj. 2006;74:
147-51.
10. Baldawa S, Gopalakrishnan CV. Bilateral abducent nerve palsy
as the initial clinical manifestation of medulloblastoma. Acta
Neurochir (Wien). 2010;152:1947-8.
11. Poussaint TY. Diagnostic imaging of primary pediatric brain
tumors. In: Diseases of the brain, head and neck, spine. Part 3.
Ed. Springer, Milan; 2008. p. 277–87.
12. Packer RJ, Vezina G. Management of and prognosis with
medulloblastoma. Therapy at a crossroads. Arch Neurol. 2008;
65:1419-24.
13. Panigraph A, Blüml S. Neuroimaging of pediatric brain tumors:
from basic to advanced magnetic resonance imaging (MRI). J
Child Neurol. 2009;24:1343-65.
14. Park TS, Hofman HJ, Hendrick EB, Humphreys RP, Becker LE.
Medulloblastoma: clinical presentation and managementexperience at the Hospital for Sick Children, Toronto,
1950-1980. J Neurosurg. 1983;58:543-52.
15. Yeom K, Andre JB, Rosenburg JK, Mobley BC, Vogel H, Fisher
PG, et al. Prognostic features of childhood medulloblastoma by
magnetic resonance imaging. Abstract presented to the 14 th
International Symposium on Pediatric Neuro-Oncology (ISPNO),
June 2010. Neuro Oncol. 2010;12(6):ii7. Available from: http://
www.ispno2010.com/index.html.
16. Giangaspero F, Perilongo G, Fondelli MP, Brisigotti M, Carollo C,
Burnelli R, et al. Medulloblastoma with extensive nodularity: a
variant with favorable prognosis. J Neurosurg. 1999;91:971-7.
17. Naitoh Y, Sasajima T, Kinouchi H, Mikawa S, Mizoi K.
Medulloblastoma with extensive nodularity: single photon
emission CT study with iodine-123 metaiodobenzylguanidine.
Am J Neuroradiol. 2001;23:1564-7.
M.I. Martínez León
18. Rumboldt Z, Camacho DL, Lake D, Welsh CT, Castillo M.
Apparent diffusion coef cients for differentiation of cerebellar
tumors in children. AJNR. 2006;27:1362-9.
19. Jaremko JL, Jans LBO, Coleman LT, Ditch eld MR. Value and
limitations of diffusion-weighted imaging in grading and
diagnosis of pediatric posterior fossa tumors. AJNR Am J
Neuroradiol. 2010;31:1613-6.
20. Kan P, Lui JK, Hedlund G, Brockmeyer DL, Walker ML, Kestle
JRW. The role of diffusion-weight magnetic resonance
imaging in pediatric brain tumors. Childs Nerv Syst. 2006;22:
1435-9.
21. Vázquez Méndez E, Mayolas Rifá N, Delgado Alvarez I, Roig
Quilis M, Alonso Farré J. Resonancia magnética en el estudio
neurológico del niño: nuevos aspectos. Neurol Supl. 2008;4:
58-64.
22. Poussaint TY, Rodríguez D. Advanced neuroimaging of pediatric
brain tumors: MR diffusion, MR perfusion, and MR spectroscopy.
Neuroimaging Clin N Am. 2006;16:169-92.
23. Panigrahy A, Krieger MD, Gonzalez-Gomez I, Liu X, McComb JG,
Finlay JL, et al. Quantitative short echo time 1H-MR
spectroscopy of untreated pediatric brain tumors: preoperative
diagnosis and characterization. Neurosurgery. 2006;27:560-72.
24. Rossi A, Gandolfo C, Morana G, Severino M, Garrè ML, Cama A.
New MR sequences (diffusion, perfusion, spectroscopy) in brain
tumours. Pediatr Radiol. 2010;40:999-1009.
25. Panigrahy A, Nelson MD, Blüml S. Magnetic resonance
spectroscopy in pediatric neuroradiology: clinical and research
applications. Pediatr Radiol. 2010;40:3-30.
26. Moreno-Torres A, Martínez-Pérez I, Baquero M, Campistol J,
Capdevilla A, Arús C, et al. Taurine detection by proton
magnetic resonance spectroscopy in medulloblastoma:
contribution to noninvasive differential diagnosis with
cerebellar astrocytoma. Neurosurgery. 2004;55:824-9.
27. Kovanlikaya A, Panigrahy A, Krieger MD, Gonzalez-Gomez I,
Ghugre N, McComb JG. Untreated pediatric primitive
neuroectodermal tumor in vivo: quantitation of taurine with
MR spectroscopy. Radiology. 2005;236:1020-5.
28. Chernov MF, Hayashi M, Izawa M, Usukura M, Yoshida S, Ono Y,
et al. Multivoxel proton MRS for differentiation of
radiation-induced necrosis and tumor recurrence after gamma
knife radiosurgery for brain metastases. Brain Tumor Pathol.
2006;23:19-27.
29. Gururangan S, Hwang E, Herdon JE, Fuschs H, George T,
Coleman RE. (18F) fluorodeoxyglucose-positron emission
tomography in patients with medulloblastoma. Neurosurgery.
2004;55:1280-8. discussion 1288–89
30. Meyers SP, Wildenhain SL, Chang JK, Bourekas EC, Beattie
PF, Korones DN, et al. Postoperative evaluation for
disseminated medulloblastoma involving the spine:
contrast-enhanced MR findings, CSF cytologic analysis,
timing of disease occurrence, and patient outcomes. Am J
Neuroradiol. 2000;21:1757-65.
31. Wiener MD, Boyko OB, Friedman HS, Hockenberg B, Oakes WJ.
False-positive spinal MR imaging findings for subarachnoid
spread of primary CNS tumor in postoperative pediatric
patients. Am J Neuroradiol. 1990;11:1100-3.
32. Terterov S, Krieger MD, Bowen I, Gordon McComb J. Evaluation
of intracranial cerebrospinal uid cytology in staging pediatric
medulloblastomas, supratentorial primitive neuroectodermal
tumors, and ependymomas. J Neurosurg Pediatrics. 2010;6:
131-6.
33. Loree J, Mehta V, Bhargava R. Cranial magnetic resonance
imaging ndings of leptomeningeal contrast enhancement after
pediatric posterior fossa tumor resection and its signi cance. J
Neurosurg Pediatrics. 2010;6:87-91.
34. Rochkind S, Blatt I, Sadeh M, Goldhammer Y. Extracranial
metastases of medulloblastoma in adults: literature review. J
Neurol Neurosurg Psychiatry. 1991;54:70-86.
Review and update about medulloblastoma in children
35. Yamasaki F, Kurisu K, Satoh K, Arita K, Sugiyama K, Ohtaki M.
Apparent diffusion coef cient of human brain tumors at MR
imaging. Radiology. 2005;235:985-91.
36. Koral K, Gargan L, Bowers DC, Gimi B, Timmons CF, Weprin B,
et al. Imaging characteristics of atypical teratoid-rhabdoid
tumor in children compared with medulloblastoma. AJR.
2008;190:809-14.
37. Brad DJ, Parisi JE, Kleinschmidt-DeMasters BK, Yachnis AT,
Montine TJ, Boyer PJ, et al. Surgical neuropathology update: a
review of changes introduced by the WHO classi cation of
tumours of the central nervous system. 4th Edition. Arch Pathol
Lab Med. 2008;132:993-1007.
38. Giangaspero F, Eberhart C, Haapasalo H, Pietsch T, Wiestler OD,
Ellison DW. Medulloblastoma. In: Louis DN, Ohgaki H, Wiestler
D, Cavenee WK, editors. WHO classi cation of tumors of the
central nervous system. Lyon: IARC; 2007. p. 132-46.
39. Saunders DE, Hayward RD, Chong WK, Wade AM. Surveillance
neuroimaging of intracranial medulloblastoma in children: how
effective, how often, and for how long? J Neurosurg. 2003;99:
280-6.
40. Harisiadis L, Chang CH. Medulloblastoma in children.
A correlation between staging and results of treatment. Int J
Radiat Oncol Biol Phys. 1977;2(9–10):833-41.
41. Packer RJ, Rood BR, MacDonald TJ. Medulloblastoma: present
concepts of strati cation into risk groups. Pediatr Neurosurg.
2003;39:60-7.
42. Sakar C, Deb P, Sharma MC. Medulloblastomas: new directions
in risk strati cation. Neurol India. 2006;54:16-23.
43. Gajjar A, Hernan R, Kocak M, Fuller Ch, Lee Y, Mckinnon PJ, et
al. Clinical, histopathologic, and molecular markers of
prognosis: toward a new disease risk strati cation system for
medulloblastoma. J Clin Oncol. 2004;22:984-93.
44. Rossi A, Caracciolo V, Russo G, Reiss K, Giordano A.
Medulloblastoma: from molecular pathology to therapy. Clin
Cancer Res. 2008;14:971-6.
45. Kortmann RD, Kuhl J, Timmerman B, Mittler U, Urban C, Budach
V, et al. Postoperative neoadjuvant chemotherapy before
radiotherapy as compared to immediate radiotherapy followed
by maintenance chemotherapy in the treatment of
medulloblastoma in childhood: Results of the German
prospective randomized trial HIT’91. Int J Radit Oncol Biol
Phys. 2000;46:269-79.
46. Zeltzer PM, Boyett JM, Finlay JL, Albright AL, Rorke LB, Milstein
JM, et al. Metastasis stage, adjuvant treatment, and residual
tumor are prognostic factors for medulloblastoma in children:
Conclusions from the Children’s Cancer Group 921 randomized
phase II study. J Clin Oncol. 1999;17:832-45.
47. Ellison DW. Childhood medulloblastoma: novel approaches to
the classification of a heterogeneous disease. Review Acta
Neuropathol. 2010;120:305-16.
48. McManamy CS, Pears J, Weston CL, Hanzely Z, Ironside JW,
Taylor RE, et al. Nodule formation and desmoplasia in
medulloblastomas-de ning the nodular/desmoplastic variant
and its biological behavior. Brain Pathol. 2007;17:151-64.
49. Cogen PH, McDonald JD. Tumor suppressor genes and
medulloblastoma. J Neurooncol. 1996;29:103-12.
50. Raffel C. Medulloblastoma: molecular genetics and animal
models. Neoplasia. 2004;6:310-22.
51. Verma S, Tavaré CJ, Gilles FH. Histologic features and prognosis
in pediatric medulloblastoma. Pediatric Develop Pathol. 2008;
11:337-43.
52. Ovani S, Etame AB, Smith ChA, Rutka JT. Genetics of
medulloblastoma: clues for novel therapies. Expert Review of
Neurotherapeutic. 2010;10:811-23.
53. Ellison D. Classifying the medulloblastoma: insights from
morphology and molecular genetics. Neuropathol Appl
Neurobiol. 2002;28:257-82.
145
54. http://www.cancerhelp.org.uk/trials/a-trial-looking-atdifferent-ways-of-giving-radiotherapy-for-children-and-youngpeople-with-medulloblastoma.
55. Bouffet E. Medulloblastoma in infants: the critical issues of the
dilemma. Curr Oncology. 2010;17:2-3.
56. Lafay-Cousin L, Bouffet E, Hawkins C, Amid A, Huang A,
Mabbott DJ. Impact of radiation avoidance on survival and
neurocognitive outcome in infant medulloblastoma. Curr
Oncol. 2009;16:21-8.
57. Johnston DL, Keene D, Bartels U, Carret AS, Crooks B, Eisenstat
DD, et al. Medulloblastoma in children under the age of three
years: a retrospective Canadian review. J Neurooncol. 2009;
94:51-6.
58. Bowers DC, Gargan L, Weprin BE, Mulne AF, Elterman RD, Munoz
L, et al. Impact of site of tumor recurrence upon survival for
children with recurrent or progressive medulloblastoma. J
Neurosurg. 2007;107(Suppl 1):5-10.
59. Gururangan S, Krauser J, Watral MA, Discoll T, Larrier N,
Reardon DA, et al. Efficacy of high dose chemotherapy or
standard salvage therapy in patients with recurrent
medulloblastoma. Neuro Oncol. 2008;10:745-51.
60. La Marca F, Tomita T. Importance of patient evaluation for longterm survival in medulloblastoma recurrence. Childs Nerv Syst.
1997;13:30-4.
61. Roka YB, Bista P, Sharma GR, Adhikari D, Kumar P. Frontal
recurrence of medulloblastoma ve years after excision and
craniospinal irradiation. Indian J Pathol Microbiol. 2009;52:
383-5.
62. Paulino AC. Collin’s law revisited: can we reliably predict the
time to recurrence in common pediatric tumors? Pediatr
Hematol Oncol. 2006;23:81-6.
63. Brown WD, Tavaré CJ, Sobel EL, Gilles FH. Medulloblastoma and
Collin’s law: a critical review of the concept of a period of risk
for tumor recurrence and patient survival. Neurosurgery. 1995;
36:691-7.
64. Amagasaki K, Yamazaki H, Koizumi H, Hashizume K, Sasaguchi
N, et al. Recurrence of medulloblastoma 19 years after the
initial diagnosis. Childs Nerv Syst. 1999;15:482-5.
65. Saunders DE, Thompson C, Gunny R, Jones R, Cox T, Chong WD.
Magnetic resonance imaging protocols for paediatric
neuroradiology. Pediatr Radiol. 2007;37:789-97.
66. Wootton-Gorges, Foreman NK, Albano EA, Dertina DM, Nein PK,
Shukert B, et al. Pattern of recurrence in children with midline
posterior fossa malignant neoplasms. Pediatr Radiol. 2000;
30:90-3.
67. Bernabeu J, Cañete A, Fournier C, López B, Barahona T, Grau C,
et al. Evaluación y rehabilitación neuropsicológica en oncología
pediátrica. Psicooncología. 2003;0:117-34.
68. Fouladi M, Chintagumpala M, Laningham FH, Ashley D, Kellie
SJ, Langston JW, et al. White matter lesions detected by
magnetic resonance imaging after radiotherapy and
high-dose chemotherapy in children with medulloblastoma
or primitive neuroectodermal tumor. J Clin Oncol. 2004;22:
4551-60.
69. Vázquez E, Castellote A, Piqueras J, Ortuno P, Sánchez-Toledo
J, Nogués P, et al. Second malignancies in pediatric patients:
imaging findings and differential diagnosis. Radiographics.
2003;23:1155-72.
70. Khong OL, Leung LHT, Chan GCF, Kwong DLW, Wong WHS, Caro
G, et al. White matter anisotropy in childhood medulloblastoma
survivors: association with neurotoxicity risk factors. Radiology.
2005;236:647-52.
71. Stavrou T, Bromley CM, Nicholson HS, Byrne J, Packer RJ,
Goldstein AM, et al. Prognostic factors and secondary
malignancies in childhood medulloblastoma. J Pediatr Hematol
Oncol. 2001;23:431-6.