Keywords
Childhood - craniospinal irradiation - medulloblastoma - molecular characterization
Introduction
Medulloblastoma (MB) is the most common embryonal tumour in children. The mean age
at presentation is 9 years, though up to a third of these tumors present in children
under 3 years of age. The predominant symptoms are secondary to obstructive hydrocephalus.
Patients may also present with ataxia, cranial neuropathies or nerve root/spinal cord
compression. The symptoms may be gradually progressive over weeks to months. 10%–40%
of patients have central nervous system dissemination at diagnosis, with infants having
the highest incidence. Imaging of the entire craniospinal axis is, therefore, vital
at initial evaluation.
There have been major strides in the understanding of the biology and molecular characterisation
of MB which are now being used to personalize management. Conventionally, MB has been
classified based on histology (classic, desmoplastic/nodular [DNMB], MB with extensive
nodularity [MBEN], large cell/anaplastic). Besides, MB is currently classified into
four distinct molecular subgroups according to transcriptional profiling studies[1] - WNT, sonic hedgehog (SHH), Group 3, and Group 4 with prognostic implications.
The molecular profiling correlates both with clinical outcomes and histologic appearance.
The World Health Organization 2016 guidelines classify MB based on histology and genetically.[2],[3]
[Table 1] depicts the current classification.
Table 1
Classification of medulloblastoma
|
NOS: Not otherwise specified, MBEN: Medulloblastoma with extensive nodularity, SHH:
Sonic hedgehog
|
|
Medulloblastoma genetically defined
|
|
Medulloblastoma, WNT-activated
|
|
Medulloblastoma, SHH-activated and TP53-mutant
|
|
Medulloblastoma, SHH-activated and TP53-wild-type
|
|
Medulloblastoma, non-WNT/non-SHH
|
|
Medulloblastoma, Group 3
|
|
Medulloblastoma, Group 4
|
|
Medulloblastoma, histologically defined
|
|
Medulloblastoma, classic
|
|
Medulloblastoma, desmoplastic/nodular
|
|
MBEN
|
|
Medulloblastoma, large cell/anaplastic
|
|
Medulloblastoma, NOS
|
The Indian Society of Neuro-Oncology guidelines and the SIOP-PODC guidelines for MB
provide a pragmatic approach to the management of this condition in our setting.[4],[5]
Case 1
Standard risk medulloblastoma in >3 years old
A 5-year-old boy presented with headaches and clumsiness starting 3 months prior to
presentation. Headaches were worse in the morning and relieved by vomiting. Ophthalmic
review revealed papilledema and nystagmus. His development had been normal prior to
these symptoms.
Physical examination revealed ataxic gait. Urgent neuroimaging with magnetic resonance
imaging (MRI) [Figure 1] revealed a mass lesion in the region of the 4th ventricle with hydrocephalus. Screening
MRI of the spine revealed no evidence of leptomeningeal disease. He underwent midline
suboccipital craniotomy, external ventricular drainage (EVD), and tumor resection.
Histopathology confirmed classical MB with subsequent IHC-based assignation into the
non-WNT, non-SHH subgroup.
Figure 1: Standard risk medulloblastoma in >3 years old
Postoperative imaging [Figure 2] did not demonstrate any hemorrhage and confirmed near-total resection (NTR).
Figure 2: Standard risk medulloblastoma-postoperative
Three weeks postoperatively, lumbar puncture was done for cytopathological evaluation
and was negative for malignant cells. The patient was commenced on adjuvant radiation
54 Gy with 23.4 Gy to the craniospinal axis. Radiotherapy could be completed without
undue interruptions and delay.
Reassessment MRI performed 4 weeks after completion of radiation revealed no evidence
of residual disease. Baseline neurological, endocrine, and auditory evaluations were
performed. Adjuvant chemotherapy was commenced with vincristine, cyclophosphamide,
and cisplatin. A total of six cycles were delivered. The patient continues on surveillance
with MRI brain every 3 months in the 1st year and screening MRI of the spine every
6 months. Growth, development, and neurocognitive functioning is assessed every 6
months.
Optimal management of MB in the current era mandates precise risk stratification incorporating
stage, postsurgical status, histology, and molecular characterization.
Diagnosis and staging
The diagnosis is established with a contrast-enhanced MRI. Ideally, the entire neuraxis
should be evaluated prior to surgery. Postoperative imaging may be confounded by artifacts
especially blood. All patients must undergo cerebrospinal fluid (CSF) analysis 2–3
weeks after surgery. Up to 10% of patients may have malignant cells in CSF in the
absence of radiological evidence of leptomeningeal dissemination. CSF analysis done
earlier than 2 weeks post-operatively may be misleading. Bone marrow examination or
bone scans are not routinely indicated at diagnosis unless patients are symptomatic.
The tumor extent is defined based on the presence and extent of neural and extraneural
metastasis [Table 2]. The extent of resection is the most significant factor affecting prognosis. This
is defined on the basis of postoperative residual disease [Table 3]. Recent evidence,[6] however, indicates that the prognostic benefit of increased extent of resection
depends on the molecular subgrouping. Maximal safe surgical resection continues to
be the standard of care but repeat surgery to resect small residual portions of MB
is not recommended if the likelihood of neurological morbidity is high. There is no
definitive benefit to gross total resection compared with NTR. Specifically for patients
with WNT, SHH, or Group 3 tumors, there is no significant survival benefit with gross
total resection versus subtotal resection. For Group 4 tumors, gross total resection
confers better progression-free survival.
Table 2
Staging of medulloblastoma
|
CSF: Cerebrospinal fluid
|
|
M0: No dissemination
|
|
M1: CSF-positive cytology only
|
|
M2: Gross nodular seeding in cerebellar-cerebral subarachnoid space and/or lateral or
third ventricle
|
|
M3: Gross nodular seeding in spinal subarachnoid space
|
|
M4: Extraneural metastasis
|
Table 3
Classification based on postoperative residual disease
|
STR: Subtotal resection, GTR: Gross-total resection, NTR: Near-total resection
|
|
GTR: No radiographical evidence of disease
|
|
NTR: ≤ 1.5 cm2 residual disease after resection
|
|
STR: >1.5 cm2 of measurable residual disease
|
|
Biopsy: No tumor resection; only a sample of tumor tissue removed
|
Genetics
Genomic profiling currently identifies four distinct subgroups WNT, SHH, Group 3,
and Group 4. These molecular subtypes may further be classified based on the presence
of MYC or MYCN alterations, TP53, and other genomic alterations.[7],[8]
The WNT and SHH classifications define the oncopathogenic pathway, while Groups 3
and 4 retain generic designations. WNT tumors are a result of unregulated WNT signalling
leading to increased transcriptional activity and tumorigenesis. WNT MB lacking somatic
CTNNB1 mutation should ideally be tested for germline APC mutations (familial adenomatous
polyposis syndrome) as they have significantly lower long-term survival due to deaths
from second tumors, which may potentially be improved with diagnosis and surveillance.[9]
Oncogenesis in SHH MBs results from up regulation of SHH signaling as a direct consequence
of loss of function of the tumor suppressor of fused gene (SUFU) and patched-1 (PTCH1).
Young children with SHH MB should be screened for germline PTCH1 and SUFU and older
children with TP53 mutations for Li-Fraumeni syndrome.[9]
Children with MB with WNT pathway activation have an excellent prognosis.[10] The prognosis of patients with SHH pathway-activated tumors is influenced by the
presence or absence of TP53 mutations.[11] The outcome for the remaining patients is inferior to that for patients with WNT
pathway activation.
WNT, SHH, Group 3, and Group 4 MBs account for 10%, 30%, 25%, and 35%, respectively.
WNT tumors are usually seen in children and adults, Group 3 tumors are more often
seen in infants and children whilst SHH and Group 4 lesions are seen across all age
groups. The SHH tumors exhibit a bimodal age distribution, typically occurring in
patients <4 and >16 years of age. The biological behavior of SHH MB depends on the
age at diagnosis with implications both on management and prognosis.[12]
While having immense prognostic and therapeutic implications, molecular profiling
using genome-wide expression profiling studies is not routinely available to the vast
majority of patients in low-middle income countries (LMIC). However, in experienced
hands, immunohistochemistry based studies can classify the tumors into WNT, SHH, and
non-WNT/non-SHH subgroups. Although it is not possible to differentiate between Groups
3 and 4 using this approach, it still offers a pragmatic approach to classify these
tumours with therapeutic and prognostic implications.[13]
The role of radiogenomics in the evaluation of MB is evolving and may especially be
valuable in LMIC settings. For example, cerebellar hemispheric tumor is likely to
be SHH and a midline tumor without significant enhancement is likely to be subgroup
4 MB.[14],[15] In addition, many centers in LMIC do have access to FISH studies for MYC, MYCN,
and Sanger seqencing for TP53. These, together with IHC and radiology, may help in
risk stratification.
Histologic correlation
Although not absolute, there is a correlation between the molecular subgroup and histologic
type, for example, 97% of WNT MBs are of the classic histologic variant. In infants,
children, and adults, 89%, 25%, and 100% of DNMBs were of the SHH molecular subgroup,
respectively. LC/A tumors in infants are usually Group 3 lesions, but are evenly distributed
across molecular subgroups in other ages.[16]
Risk stratification
Risk assignation for children older than 3 years of age into average and high-risk
prognostic groups is based on the presence of metastatic disease and residual tumor
postresection of less or greater than 1.5 cm2. Patients having postoperative residual
tumor >1.5 cm2, evidence of radiographic metastases, or presence of leptomeningeal
disease/CSF seeding are classified as “High-risk,” with the remaining patients defined
as “average-risk.”[2] Children less than 3 years of age constitute a unique group in which current standard
of care is chemotherapy alone as a first-line adjunct therapy, with radiation therapy
(RT) omitted to avoid the very poor neurocognitive outcomes associated with craniospinal
irradiation (CSI) in very young patients.
Given the poor outcomes in patients with diffuse anaplasia,[17],[18] it is also recommended that patients with LC/A histology be classified as high risk,
irrespective of other adverse features. The current consensus guidelines[2] suggest integrating molecular subgrouping, clinical and radiological features into
low risk, standard risk, high risk, and very high-risk categories with distinct survival
outcomes [Table 4].
Table 4
Consensus risk-stratification for medulloblastoma in the molecular era
|
Risk category
|
WNT
|
SHH
|
Group 3
|
Group 4
|
Others
|
|
Low risk
|
< 16 years
|
|
|
|
|
|
(Expected survival >90%)
|
|
|
|
|
|
|
Standard risk
|
|
TP53 wildtype
|
No myc amplification and
|
Non-metastatic with
|
|
|
(Expected survival 75-90%)
|
|
No myc amplification No metastasis
|
Non-metastatic
|
Chromosome 11 loss
|
|
|
High risk
|
|
One or both:
|
|
Non-metastatic without
|
|
|
(Expected survival 50-75%)
|
|
Myc amplification Metastatic
|
|
Chromosome 11 loss
|
|
|
Very high risk
|
Metastatic
|
TP53 mutation
|
Metastatic
|
Metastatic
|
|
|
(Expected survival <%)
|
|
(Metastatic or non-metastatic)
|
|
|
|
|
Unknown
|
|
|
Non-metastatic with MYC amplification; anaplasia; isochromosome 17q
|
Anaplasia
|
Melanotic medulloblastoma Medullomyoblastoma Indeterminate between group 3 and 4
|
Case 2
High-risk medulloblastoma in >3 years old
An 8-year-old boy presented with headache, vomiting, ataxia, and head tilt for 2 weeks.
Neuroimaging revealed a well circumscribed lesion 4 cm × 4 cm × 3.8 cm in the region
of vermis [Figure 3].
Figure 3: High risk medulloblastoma-preoperative
The patient was taken up for surgery. He underwent suboccipital craniotomy with transvermian
approach splitting the inferior aspect of the vermis. The tumor was highly vascular
and noted to be involving the floor of the fourth ventricle, bilateral foramina of
Luschka, and left cerebellar peduncle. Immediate postoperative imaging was not feasible.
The patient underwent repeat MRI, 3 weeks postoperatively which revealed residual
tumor of 2.5 cc. MRI of the spine demonstrated two focal enhancing nodules in the
cervical and thoracic spine. CSF was positive for malignant cells. Molecular profiling
confirmed it to be Group 3 MB with myc-amplification.
He received CSI 35 Gy in 21 fractions with posterior fossa boost of 19.8 Gy with concurrent
daily carboplatin. He was commenced on chemotherapy 4 weeks' following completion
of radiation with vincristine/cisplatin/CCNU and cyclophosphamide. [Figure 4] MRI post radiation in child with STR.
Figure 4: High risk medulloblastoma-postradiation
Surgical management
Optimal surgery is the cornerstone of management. The extent of resection largely
depends on the anatomy of the tumor determining what can be done safely without incurring
significant neurological deficit. Many studies support a relationship of extent of
resection with progression-free survival. A retrospective analysis of 233 children
in a randomized controlled trial of differing chemotherapy regimens indicated that
residual tumor less than 1.5 cm2 was associated with improvement in 5-year progression
free survival (PFS) of >20% in patients with M0 disease and an 11% difference for
all patients irrespective of all other factors.[19] Subsequent studies have questioned this premise when taking into account biological
factors. A recent retrospective analysis of 787 patients demonstrated that the benefit
of extent of resection is largely attenuated after taking into account molecular subtype
and not significant when comparing STR (>1.5 cm2) versus NTR (<1.5 cm2) or GTR versus
NTR. For tumors involving the brainstem, investigators found no difference in outcome
between GTR and residual tumor <1.5 cm2[6] Aggressive resection of brainstem disease with potential of high morbidity is, therefore,
not warranted in view of the sensitivity of the tumor to radiation and chemotherapy.
Children typically present with features of raised ICP due to obstructive hydrocephalus.
However, routine preoperative ventriculoperitoneal shunt should be avoided as definitive
surgical resection readily relieves the obstruction. Ventricular diversion if needed
may be achieved by EVD or an endoscopic third ventriculostomy[20],[21] taking care to avoid rapid decompression of the ventricles and overdrainage. Corticosteroids
may be required in the preoperative period. Dexamethasone in a loading dose of 0.5–1
mg/kg intravenously (maximum dose 10 mg) followed by 0.25–0.5 mg/kg/day can be given
in divided doses.
It is imperative that wherever feasible, screening of the spine should be is done
prior to surgery. If not feasible preoperatively, it should be acquired postoperatively
2–3 weeks after surgery to reduce the chance of erroneous interpretation consequent
to postoperative enhancement of spinal leptomeninges.[22]
All patients must undergo lumbar puncture and CSF cytopathology to evaluate for dissemination
2–3 weeks' following surgery. Intraoperative samples taken from ventricles do not
suffice for this purpose.[23]
The postoperative clinical course can be complicated by the posterior fossa syndrome
or cerebellar mutism syndrome in 8%–24% of infratentorial tumor resections. It usually
presents in the first 2 days following surgery and is characterized by a triad of
cerebellar mutism, ataxia/axial hypotonia, and irritability and emotional lability.
These children are commonly apathetic, and/or hypokinetic. While pathophysiology is
poorly understood, possible mechanisms include disruption of the dentate-thalamo-cortical
pathway, vermian injury, postoperative vasospasm, axonal injury, and neuronal dysfunction.[24],[25]
Radiation
Postoperative adjuvant RT is an integral component of therapy for all children above
3 years of age. In view of the high propensity of the tumor to develop leptomeningeal
disease, CSI followed by boost irradiation of the tumor bed/posterior fossa is recommended
to achieve adequate disease control preferably delivered from a linear accelerator.
The recommended dose is 54–55 Gy delivered over 6–6.5 weeks using conventional fractionation.
The dose for CSI for rigorously staged standard risk MBs is 23.4 Gy in 13 fractions
followed by tumor-bed boost (30.6 Gy in 17 fractions) to a total tumor-bed dose of
54 Gy in 30 fractions over 6 weeks. Such therapy in conjunction with adjuvant multiagent
systemic chemotherapy results in excellent long-term survival outcomes[26],[27] but with reduced neurocognitive and endocrinological sequelae compared to the higher
doses of CSI. High risk disease is treated with full-dose CSI (35–36 Gy in 20–21 fractions)
plus posterior fossa boost (18–19.8 Gy in 10–11 fractions) to a total tumor dose of
54–55 Gy in 30–32 fractions over 6–6.5 weeks. Patients with diffuse leptomeningeal
dissemination should receive extended dose CSI (39.6–40 Gy in 22–24 fractions) plus
entire posterior fossa boost (14.4 Gy in 8 fractions). A boost of 5.4–9 Gy in 3–5
fractions to focal nodular metastatic deposits in the brain and/or spine can be delivered
concurrently during posterior fossa boost irradiation.
Adjuvant RT should ideally begin within 4 weeks of surgery, but definitely within
6 weeks post surgery. Routine use of steroids and GCSF is avoided but may be required.
Proton therapy, where available, may be preferred over photons primarily to limit
the burden of both short and long-term effects attributable to radiation.[28],[29] However, lack of access to this modality limits it to a very small minority of patients
in our setting.
Conventionally, radiosensitizing chemotherapy with concurrent weekly vincristine has
been used extensively and is tolerated well in young children. However, there is limited
evidence of its role and in view of the significant morbidity attributable to neuropathy
especially in older children and adolescents, a number of current protocols omit weekly
vincristine in this cohort.
For “high risk” lesions addition of daily carboplatin to weekly vincristine has been
evaluated.[30] Carboplatin is potent radiosensitizer and this approach appears to enhance outcomes
for this patients with metastatic disease. However, it is important to remember that
delivering uninterrupted RT is more important than adding Carboplatin as a radiosensitizer.
The ACNS0332 Phase III data suggest no benefit in response or survival with this strategy.
It is ideal to document neurocognitive, endocrinal, and auditory status prior to initiation
of adjuvant therapy to establish a baseline for future comparisons.
Chemotherapy
Adjuvant chemotherapy, in the current era is an integral part of the management of
MB in children.[26],[31] For children above 3 years, adjuvant chemotherapy should ideally start within 4
weeks of radiation, but definitely within 6 weeks. This period is required for hematological
recovery. Neuraxial imaging should be done for re-assessment of the disease status
prior to the initiation of adjuvant chemotherapy. A total of 6–8 cycles of adjuvant
chemotherapy should be administered generally cycled at 3–6 weekly intervals depending
on the regimen used. A number of regimens may contain platinum and it is therefore
prudent to monitor with audiometry during the treatment as well. The risk of ototoxicity
is higher if cisplatin dose in individual cycle exceeds 100 mg/m2 or cumulative doses
exceed 300 mg/m2.[32] The current evidence indicates that it may not be necessary to deliver cumulative
Cisplatin doses of upto 600 mg/m2 scheduled in older protocols. Lower doses may be
equally effective.[33] Attempt should be made to deliver optimal cumulative doses of cyclophosphamide (12
g/m2) and Cisplatin (300–400 mg/m2).[31] For children under 3 years of age, adjuvant therapy comprises of primarily chemotherapy
with delayed or no radiation to spare young children from devastating late effects.[34],[35],[36],[37],[38]
High-dose chemotherapy with stem cell rescue has been evaluated for MB. It is feasible,
and may offer an advantage in patients with metastatic or recurrent disease. It offers
no survival advantage in older children receiving CSI and in LMIC may be reserved
for patients with recurrent disease only. Prior CSI may render stem cell mobilization
difficult but plerixafor can overcome this.[39],[40],[41]
The Milan strategy for metastatic MB incorporated intensive chemotherapy with myeloablative
chemotherapy in selected cases along with hyperfractionated radiation yielded excellent
outcomes. However, similar results could not be achieved by other centres with this
strategy and it is no longer being actively pursued.[42]
Case 3
Medulloblastoma in <3 years old
A 21-month-old child presented with irritability, and increasing head size for 3 months.
Neuroimaging revealed a large 4th ventricular tumor with marked hydrocephlus [Figure 5]. He underwent midline suboccipital craniotomy with microsurgical gross total excision
of the tumor with duraplasty [Figure 6]. The right frontal 3rd ventriculostomy was done prior to tumour excision. Histology
confirmed MB of the desmoplastic nodular type. Molecular studies confirmed SHH activated
variant. Postoperative imaging revealed no residual disease.
Figure 5: Meulloblastoma in <3 years old
Figure 6: Medulloblastoma in <3 years old-postoperative
The patient was commenced on systemic chemotherapy as per HIT SKK regimen with 12
cycles comprising of carboplatin, methotrexate, cyclophosphamide, and etoposide. Ommaya
reservoir was inserted to deliver ventricular methotrexate in each cycle to a total
of 32 doses.[35]
The patient was reassessed after every 4 cycles to confirm continuing remission. Decision
was taken to omit radiation altogether. The patient continues to be free of disease
2 years' post completion of therapy.
Up to one-third of cases of MB occur during the first 3 years of life. Historically,
the survival rates for this cohort have been poor and did not exceed 25%–45% until
the past decade. The relatively unfavorable prognosis may partly be explained by more
frequent occurrence of metastases and the different biology of MB in young children.
Further, the immature brain is particularly susceptible to radiotherapy-induced neurocognitive
deficits warranting the omission of this modality.
Treatment strategies for young children with MB have been aimed at improving survival
whilst limiting neurocognitive sequelae. Treatment approaches have been focused on
delaying or omitting radiotherapy using conventional systemic chemotherapy incorporating
high dose and intra-ventricular methotrexate, high-dose chemotherapy with autologous
stem cell rescue and tandem transplant following induction chemotherapy with/without
high dose methotrexate.[34],[35],[36],[37],[38]
The pilot trial HIT-SKK'87 confirmed that postoperative chemotherapy may successfully
delay the start of radiotherapy.[34] Intraventricular methotrexate was introduced as a substitute for radiotherapy in
the subsequent HIT-SKK'92 trial and HIT 2000 study.[35] If complete remission was achieved, survival rates, especially for young patients
with DMB (5-year PFS and OS 85% and 95% respectively were very favorable. Neurocognitive
deficits were reduced as compared with the HIT-SKK'87 trial. The estimated survival
rates for the entire cohort (5-year EFS rate, 57% + 8%; 5-year OS rate, 80% + 6%)
compared favorably with results of older studies.
Exclusively chemotherapy-based approach as first-line treatment may contribute to
improved salvage strategies at relapse: while 50% relapses were successfully salvaged
in HIT-SKK' 92, only 1 of 10 treated children were salvaged in HIT-SKK' 87 trial.
These studies also confirmed that histology is a strong prognostic factor in this
age group. This is important as DMB/MBEN account for 40% of cases in this cohort.[38] Gross total resection appears to be more common and feasible for patients with DMB
or MBEN. All infants with SHH MB have a favorable prognosis regardless of histology.
For children with non-DMB/non-MBEN, for which predominantly local relapses lead to
less favorable survival rates, local radio-therapy has been introduced after chemotherapy
since 2006. However, infants with non-SHH MB mainly belong to Subgroup 3 and usually
succumb to metastatic disease, and focal RT approaches have failed to improve outcomes
in this cohort. The best results till date are with the COG0334 protocol on the high-dose
methotrexate arm added to the 99,703 tandem transplant backbone (to be published).
Therefore, careful consideration is needed prior to offering focal RT to infant with
subgroup 3 MB as this may not have any survival benefit.
The SJYC07 Phase II[43] study offered molecularly driven risk adapted therapy for young children with MB.
The study identified a good responder SHH subtype (iSHH-II) that exhibits excellent
progression-free survival in the absence of radiation, intra-ventricular or high-dose
chemotherapy in contrast to the poor responder iSHH-I that has much inferior outcomes.
Recent data incorporating genomics now mandates that future approaches for young children
will increasingly incorporate molecularly driven, risk-adapted approaches.
Follow-up after treatment
Surveillance imaging during and after treatment aimed at detecting recurrent disease
at an early stage in asymptomatic patients, has been arbitrarily determined and not
been shown to influence survival. Isolated spinal recurrences are infrequent and follow-up
imaging must be tailored based on risk assignment.[44] It is imperative that strategies for follow-up incorporate multidisciplinary review
inclusive of audiological, neuropsychological and endocrine evaluations.
Case 4
Recurrent medulloblastoma
A 7-year-old young girl had been diagnosed with classical MB at 5 years of age. She
had undergone gross total resection and received adjuvant radiotherapy and chemotherapy.
One year after completion of chemotherapy, she presented with recurrent disease locally.
Surgical resection was attempted followed by high-dose chemotherapy with stem cell
rescue. Unfortunately, despite this the disease progressed and she succumbed.
Options for recurrent disease are currently limited and largely unsuccessful. The
pattern of failure is likely subgroup-specific and can guide management decisions.
Eg: Group 3 and Group 4 usually have distant recurrence after RT-based treatment.
SHH-MB may recur both locally and with distant metastasis. Patients with germline
conditions are also at risk of second malignancies which need to be recognized and
appropriately managed.
Surgery, re-irradiation, and chemotherapy regimens including myeloablative chemotherapy
have been explored with poor results.[45],[46] A proportion of infants treated previously with only chemotherapy may be salvaged
with CSI with profound neurocognitive sequelae. Bevacizumab and irinotecan with or
without temozolomide has been shown to give objective response rate and is well tolerated.[47]
An evolving alternative approach is to target tumor angiogenesis with metronomic therapy
incorporating bevacizumab, thalidomide, celecoxib, fenofibrate, etoposide, and cyclophosphamide
and additional intraventricular therapy (etoposide and liposomal cytarabine) which
can be delivered with manageable toxicity.[48]
Future strategies
Specifically targeted chemotherapies targeting oncogenic pathways are a promising
future application of molecular subgrouping. Currently, molecularly targeted agents
for each of the four molecular subgroups are being evaluated in preclinical and clinical
models. The most well-studied of these is Vismodegib, that has demonstrated some utility
in both preclinical and clinical models and may be an option in adolescents and older
children.[49]
Alternative strategies include sensitization of MB tumor cells to chemotherapeutic
treatment. Thiostrepton, an antagonist of FOXM1 (an oncogene known to be upregulated
in a variety of malignancies), was shown to sensitize MB cells to cisplatin in vitro.[50]
In our setting, future strategies must focus on improving access to molecular classification
which is the cornerstone for improving outcomes and limiting toxicity. For example,
infant SHH even with metastasis may be curable using chemotherapy only approaches
and WNT tumors may be offered reduced dose CSI. It is therefore, imperative that ongoing
efforts in LMIC focus on this. Till genomic profiling becomes more widely available,
validation, and incorporation of IHC and radiogenomics could play a major role in
delievering risk-adapted treatments aimed at optimizing outcomes and limiting toxicity.
Summary and Recommendations
Summary and Recommendations
-
Optimal treatment of MB can yield high cure rates even in countries with limited resources
-
The current risk stratification incorporates clinical, histopathological and molecular
characteristics
-
Molecular characterization, while of immense prognostic and therapeutic significance
may not be feasible for the vast majority of patients in our setting
-
Immunohistochemistry-based approach to molecular characterization may offer a pragmatic
approach
-
Optimal resection is the cornerstone of treatment. However, in view of the radio and
chemo-sensitivity of the tumor, heroic attempts at complete resection associated with
neurological morbidity are unwarranted
-
Placement of VP shunts is not required and temporary diversion measures should suffice
in cases with marked hydrocephalus
-
Radiation is an essential part of treatment in older children. It should be or omitted/deferred
in younger children wherever feasible. Tumor bed boost is equivalent to posterior
fossa boost
-
Current evidence supports adjuvant chemotherapy in all patients for 6–8 cycles. High-dose
chemotherapy may offer an advantage in salvage settings
-
Ongoing approaches are focused on improving outcomes for high risk disease and limiting
the late effects associated with treatment
-
For LMIC these should be focused on improving access to molecular stratification.
