Pediatr Blood Cancer 2014;61:457–463
Children <1 Year Show an Inferior Outcome When Treated According to the
Traditional LGG Treatment Strategy: A Report From the German Multicenter
Trial HIT-LGG 1996 for Children With Low Grade Glioma (LGG)
Cora Mirow, MD,1* Torsten Pietsch, MD,2 Susanne Berkefeld, PhD,3 Robert Kwiecien, PhD,4 Monika Warmuth-Metz, MD,5
Fabian Falkenstein, MD,1 Barbara Diehl, MD,1 Stephan von Hornstein, MD,1 and Astrid K. Gnekow, MD1
Background. Children diagnosed with LGG at an age <1 year are
reported to have an impaired prognosis in comparison to older
patients. Analysis of this subgroup could reveal the necessity to
develop risk-adapted treatment approaches. Procedure. Children <1
year at diagnosis (n ¼ 66, median age 7.3 months, 33 female, none
NFI) from the HIT-LGG 1996 cohort were analyzed for risk factors for
EFS, PFS and OS. Several children suffered from diencephalic
syndrome (DS, n ¼ 22) and primary dissemination (DLGG, n ¼ 9), 50
had a supratentorial midline (SML) location. Extent of resection was
complete/subtotal in 12, partial in 15, biopsy in 27. Tumors were
pilocytic astrocytoma WHO grade I (n ¼ 33), other WHO grade I
(n ¼ 14), pilomyxoid astrocytomas WHO grade II (n ¼ 3), and
neuroepithelial tumors WHO grade II (n ¼ 4). Results. One-year
EFS was 34.8%. SML-localisation, minor extent of surgery, pilocytic
astrocytoma, DLGG and DS were unfavorable predictive factors. No
additional non-surgical therapy was applied in 24, 36 were treated
with VCR/carboplatin chemotherapy, 6 with radiotherapy (5/6
brachytherapy). Ten-year-PFS-rate following non-surgical therapy
was 16.7%; DS and DLGG were unfavorable factors. Ten-year-OSrate was 72.8%, lower for children <6 months at diagnosis, with DS,
or with DLGG. At last follow up in August 2011, vision in 31 living
children was often severely impaired. Conclusions. Children <1 year
at diagnosis have a conspicuously impaired survival with current
treatment approaches. Age <6 months, diencephalic syndrome and
dissemination constitute risk factors for even lower PFS and OS.
Treatment adaptations are needed to improve outcome and
molecular genetics may explain tumor aggressiveness. Pediatr Blood
Cancer 2014;61:457–463. # 2013 Wiley Periodicals, Inc.
Key words: chemotherapy; glioma (low grade); infants; long-term follow-up; observation
INTRODUCTION
Study Strategy
Infants with pediatric low grade glioma (LGG) under 1 year of
age at diagnosis constitute a minority in most reports with little
epidemiologic data published. In a North American cohort of 6,212
pediatric glioma patients 103 (1.7%) children <1 year with LGG
have been mentioned [1]. The clinical course of these tumors of low
grade malignancy is influenced by localisation, extent of surgery,
histological subtype, association with neurofibromatosis type I
(NF1), and specifically age at diagnosis [2–9]. Earlier reports
demonstrated that children diagnosed with LGG at young age show
more frequent progression and inferior outcome than older
children [7,10–12]. With this report we detail clinical characteristics and the course of disease in infants within the HIT LGG 1996
study cohort following a comprehensive multicenter trial strategy [13,14]. We present data demonstrating that conventional
treatment strategies are insufficient for infants that demonstrate
diencephalic syndrome (DS), primary dissemination (DLGG) and
age <6 months at diagnosis.
At diagnosis best safe resection of the primary tumor was
recommended. Observation was scheduled for all children without
threatening clinical symptoms, while non-surgical therapy was
indicated at radiological progression or evidence of threatening
symptoms. For children 5 years chemotherapy was recommended
to delay radiotherapy, while radiotherapy was scheduled for
children >5 years. Chemotherapy consisted of an induction with
10 weekly doses of vincristine 1.5 mg/m2 and four doses of
carboplatin 550 mg/m2 at 3-week intervals followed by 11 cycles of
simultaneous vincristine and carboplatin at 4-week intervals. Dose
modification was advised for children <10 kg of weight to
PATIENTS AND METHODS
Eligibility
The HIT-LGG 1996 trial was a comprehensive multicenter trial
for children suffering from LGG in the German speaking
countries [14]. Inclusion criteria comprised age <17 years and
diagnosis of LGG according to the WHO-classification [15]. In case
of extensive optic pathway glioma, radiologic diagnosis by
magnetic resonance imaging (MRI) was accepted [14]. Time of
registration ran from October 1, 1996 until March 31, 2004 for
newly diagnosed patients or patients with an earlier date of
diagnosis but without prior non-surgical therapy. Informed consent
was given by the patient’s parents or guardians. Approval of local
and central ethic boards was obtained.
C 2013 Wiley Periodicals, Inc.
DOI 10.1002/pbc.24729
Published online 4 September 2013 in Wiley Online Library
(wileyonlinelibrary.com).
1
Children’s Hospital of Augsburg, Augsburg, Germany; 2Department
of Neuropathology, University of Bonn, Bonn, Germany; 3Institute of
Medical Biostatistics, Epidemiology and Informatics (IMBEI),
University of Mainz, Mainz, Germany; 4Institute of Biostatistics and
Clinical Research, University of Muenster, Muenster, Germany;
5
Department of Neuroradiology, University of Wuerzburg, Wuerzburg,
Germany
Grant sponsor: Deutsche Krebshilfe; Grant number: 70-2288-GnI
Conflict of interest: Nothing to declare.
This article was published online on 04 September 2013. Subsequently
an error was found in Figure 2 as mentioned below. The curves (not
subtitle and caption) in graph A and B were mistakenly swapped in the
original version. The curve in graph A wrongly showed the data for PFS
of patients with/without primary dissemination, not the data of the PFS
of patients with/without diencephalic syndrome as is written in the
caption and vice versa in graph B. This has now been corrected.
Correspondence to: Cora Mirow, Children’s Hospital, Klinikum
Augsburg, Stenglinstrasse 2, D 86156 Augsburg, Germany.
E-mail: [email protected], [email protected]
Received 25 May 2013; Accepted 17 July 2013
458
Mirow et al.
carboplatin 18.3 mg/kg and vincristine 0.05 mg/kg. Further dose
reductions of one-third were recommended for children <6 months
of age. Radiotherapy was scheduled as conventional external beam
radiation with a total dose of 54 Gy (1.8 Gy per fraction). In case
children <5 years were irradiated the dose was modified to 45.2 Gy
(1.6 Gy per fraction). For 125-iodine-radiosurgery (brachytherapy)
in small, well delineated tumors no age limit was given. Regular
clinical and radiological assessments were scheduled.
Biostatistics
All children <1 year of age at diagnosis registered in the study
were included in this analysis. Overall survival (OS) was calculated
from date of diagnosis until death (regardless of cause). Event-freesurvival (EFS) was calculated from date of diagnosis until the first
event, defined as the necessity to start non-surgical therapy,
radiologic progression or relapse or death. Progression-free-survival
(PFS) was calculated from the start of non-surgical therapy until
progression of the residual tumor, relapse after complete remission
or death of any cause. Patients who were lost to follow-up were
censored at last visit. The log-rank test was applied for detecting
differences between subgroups. Any statistical analysis in this paper
is explorative. Hence, no significance level has been fixed, and no
adjustment for multiple testing was performed. Survival rates were
estimated by the Kapla–Meier method, using SPSS 13.0.
Response Assessment
Contrast enhanced MRI was planned at week 12, 24, 36, 48 and
after week 53. Response was evaluated according to the recommended criteria of the SIOP brain tumor subcommittee [16].
Complete, partial and objective response and stable disease were
considered positive responses within this trial. Central neuroradiological review was available in the latter part of the study.
RESULTS
Patients
The study registered 1,182 children (1,031 protocol patients), of
which 80 were <1 year of age at diagnosis. Eleven were excluded
because first therapy did not follow the protocol-strategy. In three
patients data were incomplete. Thus, we report the results of 66
patients <1 year, 25 of these <6 months of age at diagnosis. None of
the patients was later reported to have developed NF1, one patient
had trisomy 21. Epidemiologic data and symptoms are detailed in
Tables I and II.
Tumors were located in the supratentorial midline (SML) in 50
patients with the majority in the chiasmatic–hypothalamic region.
Nine children suffered from primary, four from secondary
dissemination. Median time to last follow up was 9.7 years.
Surgery
Extent of first surgery was complete or subtotal in 12 patients.
Ten of these had tumors in the cerebral hemispheres. In one patient
with a chiasmatic glioma subtotal resection was performed. Partial
resection or biopsy was performed in 42 patients, including biopsies
of three hemispheric/cerebellar tumors with large extension to
either the SML region or the hippocampus. In 12 SML tumors
radiological findings were unequivocal for LGG; no biopsy was
Pediatr Blood Cancer DOI 10.1002/pbc
taken. Thirteen children were re-operated for tumor progression
during follow-up, and six were re-operated twice.
Histology
Initial central pathological review at the national German
reference centre, Institute of Neuropathology, University of Bonn,
was obtained in 30 of 54 resected tumors. In eight patients,
discrepancies between the local pathologist and the central reviewer
were detected.
Upon central re-review of all astrocytoma according to the
revised version of the WHO classification by the Brain Tumor
Reference Center of the German Society of Neuropathology and
Neuroanatomy, three cases were classified as pilomyxoid astrocytoma (Table I) [17–19].
Strategy Groups
Following initial radiological diagnosis or tumor resection 24
children were observed, and 42 were treated with non-surgical
therapy (Fig. 1).
Observation Group
Details of the observation group are given in Table I. In 10
children, the tumor progressed during observation. One patient with
a temporal lobe ganglioglioma had partial resection at diagnosis.
Progressing after 7 months, the tumor was again partially resected
and was stable until last follow-up in 2011. In two patients with
chiasmatic–hypothalamic tumors progression occurred 0.5 and
1 year after diagnosis. Due to good general condition the children
remained observed with stable tumors thereafter until last follow-up
in 2011. One infant had an astrocytoma grade II and the other tumor
was diagnosed radiologically. One radiologically diagnosed
optic pathway tumor progressed 6 years after clinical diagnosis.
Due to good clinical condition no treatment was initiated, but the
patient was lost to follow-up in 2004. In five other patients with
optic pathway tumors no chemotherapy was started at tumor
progression due to parents’ decision, patient morbidity or
clinicians’ decision. These patients all died within 5 years after
diagnosis. Histology in these patients was three pilocytic
astrocytoma (centrally confirmed), one pilomyxoid astrocytoma
(confirmed after re-review) and one radiological diagnosis. In 1
male centrally confirmed astrocytoma grade II had transformed to
anaplastic oligoastrocytoma when re-operated for progression 2.5
years after diagnosis. He received high grade glioma treatment and
was censored at the time of malignant tumor progression. He died
6.5 years later.
Radiotherapy
Five chiasmatic–hypothalamic tumors were treated with 125-Ibrachytherapy [20,21]. One patient with the diagnosis of a
dysembryoblastic neuroepithelial tumor (DNT) in the cerebral
hemispheres received conventional external beam radiotherapy
8 years after diagnosis.
Four of the six patients later progressed. One patient died
without further treatment 3 years following local tumor progression
with secondary dissemination. Two children went on to receive
chemotherapy within 2 years after seed implantation. The patient
with external radiotherapy was treated with chemotherapy
15 months after radiotherapy.
Children <1 Year With Low Grade Glioma
459
TABLE I. Patient’s Characteristic
Number of patients
Gender (m:f)
Age at diagnosis (months) (median, range)
Age <6 months
Age 6 months
Diencephalic syndrome
Primary dissemination
Histology
Pilocytic astrocytoma I
Pilomyxoid astrocytoma II
DIG/DIA I
SEGA I
Astrocytoma II
DNT
Ganglioglioma I
Radiologicaldiagnosis “LGG”
Extent of surgery
Complete
Subtotal
Partial
Biopsy
No surgery
Tumor localization
SML
Chiasm
Chiasmatic–hypothalamic
Hypothalamus
Thalamus
Tectum
Suprasellarnos
Cerebral hemispheres
Cerebellum
Tumor size (cm, median, range)a
Time to best response (months, median, range)
Time to first progression after start of therapy (years, median, range)
Last follow up 2011
Alive
CR
PR
SD
PD
Dead
a
Observation
Chemotherapy
Radiotherapy
All
24
12:12
6.6 (0.5–11)
11
13
4
1
36
17:19
7.3 (3–12)
14
22
16
8
6
4:2
10.4 (7–12)
0
6
2
0
66
33:33
7.3 (0.5–12)
25
41
22
9
5
2
5
1
2
1
3
5
23
1
1
—
2
—
2
7
5
—
—
—
—
—
1
—
33
3
6
1
4
1
6
12
5
5
4
5
5
9
2
—
18
7
—
—
2
4
—
5
7
15
27
12
10
1
5
1
—
2
1
12
2
4.6 (3–11)
35
6
26
1
1
—
1
—
1
4.0 (3–7)
3.6 (2–33)
1.4 (0.1–6.7)
5
1
4
—
—
—
—
1
—
3.0 (2–8)
3.3 (2–5)
1.8 (0.3–7.3)
50
8
35
2
1
2
2
13
3
17
6
—
11
—
7
27
1
2
22
2
9
5
—
—
4
1
1
49
7
2
37
3
17
Tumor dimension obtained from cranial MRI; largest diameter was available in 52 patients.
Chemotherapy
Thirty-six children received chemotherapy following protocol
recommendations, 27 started therapy within 3 months after
diagnosis. The major indications for treatment were DS and visual
impairment. The scheduled 1 year of vincristine/carboplatin
chemotherapy was completed by 24 patients, while it was extended
individually to 61, respectively, 73 weeks for two patients. In seven
patients chemotherapy was stopped prematurely due to progression,
allergy, death from early progression, or without medical reason. In
five patients the protocol regimen was switched due to progression,
allergy to carboplatin, or ototoxicity to various alternatives. Best
response to vincristine/carboplatin was PR in 14, OR in 4, SD in 11
and PD in seven children after a median time of 3.6 months. Seven
Pediatr Blood Cancer DOI 10.1002/pbc
tumors progressed during chemotherapy with the histology of
pilocytic astrocytoma in 5 (3 with increased Ki67/MIB1-proliferation-rate, 4/5 central review), desmoplastic infantile ganglioglioma
(DIG) in 1 (1/1 central review) and radiologic diagnosis of LGG
without biopsy in 1. Only one child achieved complete remission
after resection, five died, one has stable disease. Of the 11 children
with stable disease as best response to chemotherapy, 9 progressed
during follow up, of whom 2 died. Altogether 21 children of this
group needed at least one salvage therapy which included vinblastin,
vincristine/carboplatin, temozolomide, cisplatin-etoposide-ifosfamide
(PEI), 6-thioguanin-procarbazin-vincristine-CCNU (TPVC), cyclophosphamide and external beam radiotherapy. At last update
nine children had been treated with radiotherapy as 2nd, 3rd, or
further therapy.
460
Mirow et al.
TABLE II. Symptoms at Diagnosisa
Visual symptoms
Nystagmus
Squint
Amaurosis
Protrusiobulbi
Diencephalic syndromeb
Hydrocephalus
Seizures
Paralysis
Impaired consciousness
38
35
1
1
1
22
16
10
3
2
a
Multiple symptoms per patient possible; bDiencephalic syndrome
characterized by cachexia, nystagmus and the presence of a chiasmatic–
hypothalamic tumor [25–27].
Toxicity
Chemotherapy was well tolerated in most patients. Three
quarters of the reported toxicities were haematotoxicity with >50%
classification of chemotherapy toxicity (CTC) grade II–IV. Nonhaematological toxicity was moderate. Twelve children suffered
from infection CTC grade I–II and 2 from grade III, 4 from
neurology grade I–II, 2 from ototoxicity grade I–III, 2 from
nephrotoxicity grade I–II, 10 from gastroenterological toxicity
grade I–III, and 5 from carboplatin allergy grade I–II.
Survival and Response to Therapy
EFS was 34.8% (SE 5.9%) after 1 year. Patients with tumors in
the SML, associated with pilocytic astrocytoma, without surgery or
with biopsy or partial resection only, with DLGG or with DS had
events earlier than others. Event was first therapy in 33, progression
in 19, death in 1, while 13 were censored at follow-up. For those
later receiving non-surgical treatment events were first therapy in 33
and progression in 9.
Fig. 2. PFS according to presence/absence of diencephalic syndrome
(A, n ¼ 42, 5-year-PFS-rate 42% and 11%, P ¼ 0.079) and to presence/
absence of primary dissemination (B, n ¼ 42, 5-year-PFS-rate 35% and
0%, P ¼ 0.051).
PFS was 28.6% (SE 7%) after 5 years and 16.7% (SE 5.8%) after
10 years for the combined treatment groups. PFS was influenced by
DS (5-year-PFS-rate 41.7% (SE 10.1%) without vs. 11.1% (SE
7.4%) with DS, P ¼ 0.079) and primary dissemination (5-year-PFSrate 35.3% (SE 8.2% without versus 0.00 with dissemination,
P ¼ 0.051; Fig. 2), but not by age, sex, localisation, histology, or
extent of surgery.
OS was 72.8% after 10 years (SE 5.7%). Children <6 months
of age had lower OS than children between 6 and 12 months.
Ten years OS for children <6 months was 58.7% (SE 10.2%), while
it was 81.1% (SE 6.5%) for children 6 months (P ¼ 0.024).
Children with DS (10-year-OS-rate 42.7%, SE 11.1%) or DLGG
(10-year-OS-rate 44.4%, SE 16.6%) had lower survival than
children without (P < 0.001 for DS and P ¼ 0.021 for dissemination; Fig. 3).
Visual Outcome
Fig. 1. HIT-LGG 1996- Infants (n ¼ 66) trial strategy. DOD ¼ death of
diagnosis.
Pediatr Blood Cancer DOI 10.1002/pbc
Thirty-eight children presented with visual symptoms at
diagnosis. However, ophthalmologic data of these young children
was not regularly assessed before and after therapy. At last followup 5 of 31 surviving children had minor visual impairment that
allowed visiting regular school. Twenty-one patients had severe
visual impairment with bilateral amaurosis in 5, and 16 visited a
school for visually impaired children. Five patients were blind on
one eye with full to nearly full vision in the second eye.
Children <1 Year With Low Grade Glioma
Fig. 3. Overall Survival, related to age groups (A, n ¼ 66, 10-year-OSrate 59%, 73%, and 81%, P ¼ 0.024 for children <6 months and 6–12
months), to presence/absence of diencephalic syndrome (B, n ¼ 66, 5year-OS-rate 60% and 91%, P < 0.001 and to presence/absence of
primary dissemination (C, n ¼ 66, 5-year-OS-rate 56% and 84%,
P ¼ 0.021).
DISCUSSION
Previous reports suggest that LGG, particularly optic pathway
glioma, have a more aggressive clinical course in young children
[4,7,9,10,11,14]. Presenting detailed analysis of 66 infants with
LGG we indicate that infants with age of <6 months, diencephalic
syndrome, and primary dissemination show early progression and
death.
There is consensus to observe patients following resection of
cerebral hemispheric or cerebellar tumors without additional nonsurgical therapy [22,23]. However, only 10 of 24 patients in our
Pediatr Blood Cancer DOI 10.1002/pbc
461
observation group remained without progression following complete or subtotal resection.
The majority of our patients had large unresectable SML tumors
and severe clinical symptoms. Particularly, infants with DS needed
early intervention. Their portion is high in our cohort with 31%,
corresponding to the young age group [24]. While most reports
describe single cases, seven (median age at diagnosis 11 months)
and nine (mean age at diagnosis 25 months) patients with DS were
included in the reports of Poussaint et al. [25] and Gropman et al.
[26]. Prognosis was impaired despite a positive effect of
chemotherapy in some of these children [25–28]. But, tumor
recurrence and final death after stopping chemotherapy were
frequent as in our group.
Further-on, disseminated LGG has been linked to aggressive
tumor growth and increased morbidity [25,27,29]. Chemotherapy
can induce clinical stabilization and delay progression confirmed
for the respective cohort of the HIT-LGG 1996 study [29,30]. Our
infant group included nine patients with DLGG, six of whom
suffered from DS, as well. Since four of six children with DLGG
and DS died, this association proved specifically unfavorable.
Our results corroborate previous reports identifying younger age
as risk factor for survival or progression, though no causative
explanation was given in the extensive discussions [7,10,11]. Our
subgroup of 26 patients <6 months of age at diagnosis fared worse
if compared to infants between 6 and 12 months at diagnosis (5-year
OS 64% vs. 90%), since they accumulated additional risk factors:
Thirteen (50%) suffered from DS, and 7 (27%) from DLGG.
Complete or subtotal tumor resection was possible in only two
patients, one of which is blind today. While 12 patients remained
observed, eight of them despite progression, the majority (n ¼ 14)
received chemotherapy with four never responding.
Within the French cohort of 85 young children (median age
17 months, range 1 day to 123 months) with optic pathway glioma
treated with a multiagent chemotherapy (BBSFOP, seven threecourse cycles alternating carboplatin and procarbazin, etoposide
and cisplatin, vincristine, and cyclophosphamide, given every
3 weeks, for a total treatment time of 16 months.), age <1 year at
diagnosis was the only prognostic factor for PFS after start of
therapy besides the absence of NF-I [11]. Three-years-PFS was
34% for 26 patients <1 year of age as compared to 61% for the older
age group. In addition, all 11 children <1 year of age, who
responded less than PR to chemotherapy, progressed within
42 months after the start of chemotherapy. In the French series
none of 13 children with DS remained progression-free during
follow-up. Altogether, 65% of the French patients required one or
more salvage therapies comparable to our series (21/36, 58%).
Relapses occurred more frequently in patients younger than 1 year
at diagnosis and response to salvage treatment was impaired in
those with less than PR to first chemotherapy [31].
Massimino et al. reported 34 children with progressive LGG
treated with 10 cycles of cisplatin and etoposide at a median age at
diagnosis of 45 months. The group of 6 children <1 year of age at
diagnosis had a worse PFS than older children (3 years PFS 33% vs.
87% for children >1 year), yet no more clinical details were
given [12].
Furthermore, Fouladi et al. [10] reported 19 children <2 years of
age within the group of 70 children (median age 7.7 years) treated
with a carboplatin or cisplatin based regimen. The 5-year PFS was
inferior if the children were younger (43 12% for age <2 years vs.
68 6% for age >2 years, P ¼ 0.01), while OS did not differ
462
Mirow et al.
notably. Since PFS was calculated as time from diagnosis to first
failure, it is not fully comparable with the event free survival (EFS)
of 35% in our analysis.
The contemporary British cohort of 639 LGG-patients up to
16 years (median age 6.7 years), reported by Stokland et al. [7],
followed a comparable treatment strategy. Children <1 year of age
at diagnosis had the shortest time from diagnosis to first progression
as compared to the older age groups (5 years PFS 41% as compared
to 56%, 58%, 79%, and 77.2% for age 1–3, 3–5, 5–10, and >10
years, P ¼ 0.01), but at multivariate analysis age <1 year did not
reach significance. The infant group had predominantly tumors in
the SML, comparable to our cohort.
In the Children’s Oncology Group study 274 patients up to
10 years were randomly assigned to either carboplatin/vincristine or
TPCV [9]. Five years EFS, defined as time from diagnosis to first
progression/relapse/death/second malignancy, was 19% (7%) for
34 children <1 year of age and 51% (4%) for older patients. The
relative risk for progression/relapse was 3.4 times higher for
patients <1 year of age at diagnosis than for the older patients
(P < 0.001) and six times higher for death (P ¼ 0.001). No details
are given about the presence of DLGG and DS. However, besides
age thalamic tumor site and residual tumor >3 cm2 were
independently associated with impaired OS, while tumor size
was not prognostic in our series.
Thus, the reports from Stokland et al. [7], Ater et al. [9] and our
data stress impaired survival for children <1 year of age at
diagnosis in addition to a high progression rate, not reported from
other series.
The 2007 revision of the WHO-classification of tumors of the
central nervous system introduced pilomyxoid astrocytoma
previously described as “pilomyxoid variant of the pilocytic
astrocytoma” as grade II tumor based on the observation that
patients had a more aggressive course of disease [32–34]. One of us
(T.P.) re-reviewed all astrocytoma slides, but found only one
additional case, so we cannot confirm a remarkably high number of
pilomyxoid astrocytoma in our infant cohort. However, two of the
three patients with PMA died.
There is consensus to treat all infants, predominantly those with
large SML tumors, with chemotherapy to delay or avoid early
external beam radiotherapy with its deleterious late effects upon
neurocognitive development, and risks for endocrinopathy or
secondary malignancy [35,36]. The combination of vincristine and
carboplatin has been proven as a successful strategy, as have been
other multiagent protocols [5,7,9,11–14,37]. Currently, there is no
evidence of one strategy being more effective than the other, yet all
reported series as well as ours stress disappointing PFS and even OS
for children <1 year of age at diagnosis despite acceptable
treatment results for older patients. In addition, 21 out of our 36
surviving patients with optic pathway glioma are blind or have
severe visual impairment.
Assessment of visual acuity is difficult in infants and recent
papers have addressed the necessity of a standardized procedure to
compare visual outcome [38,39]. Limitation of the study is that
visual acuity was not assessed by a compulsory program. We
recommend including standardized instruments for assessment of
visual acuity of infants in future trials.
Thus, there is a need to improve treatment strategies for infants
with LGG, particularly those with optic pathway tumors. The
international trial SIOP LGG 2004 investigates prolongation and
intensification of chemotherapy, but results have not been
Pediatr Blood Cancer DOI 10.1002/pbc
released [40]. Besides common chemotherapy regimens, no
alternative strategy tailored to the specific needs of the infant
subgroup with LGG has been suggested.
While vascular endothelial growth factor (VEGF) is a target for
the therapy of high grade glioma already, multiply recurrent LGG in
young children (mean age 21 months) have been treated with a
combination of bevacizumab and irinotecan by Packer et al.
[41–43]. Objective responses were achieved predominantly in
patients remaining on treatment. However, side effects like
leukoencephalopathy and proteinuria were reported compared to
the moderate toxicity profile of common chemotherapy strategies.
Likewise, platelet derived growth factor (PDGF) presents an
angiogenetic target in multiply recurring hypothalamic LGG in
children of young age. In a study of six children (3–29 months)
imatinib achieved stable disease in all six patients. Yet, despite
successful tumor control two children treated at age 6 and
12 months showed developmental delay. The authors caution the
use in children <1 year of age [44].
In >80% of LGG, BRAF alterations activate downstream
MAPK (mitogene-activated protein kinase) effectors. Inhibition of
the MAPK-pathway is considered a potential treatment approach [45–47]. In a multicentre phase II study the response-rate
of sorafenib, a RAF/multiple tyrosine kinase inhibitor, was
determined for recurrent/progressive LGG in children 2 years.
Due to high rate of early progression 9 of 11 children were taken off
the study [48]. Further clinical trials are ongoing for MAKPtargeted therapy, like selective BRAF-inhibitor vemurafenib [49].
Unfortunately, systematic molecular analysis of LGG-tumors <1
year of age is lacking.
In conclusion, despite harboring CNS-tumors of low grade
malignancy, our patients <1 year of age at diagnosis treated
following contemporary strategies have an impaired prognosis.
Children diagnosed at age <6 months and/or with diencephalic
syndrome or primary dissemination frequently suffer from early
progression and often die despite multiple lines of multiagent
chemotherapy and radiation. New treatment strategies have to be
developed for this subgroup of pediatric LGG-patients. Within the
European working group on LGG, a meta-analysis confirmed the
generally poorer prognosis for infants in all national protocols and
our risk factor assessment. In the future, these tumors shall be
further characterized by molecular genetics following mandatory
biopsy. Although standard vincristine/carboplatin effectively
delays disease progression in the majority of patients, this standard
treatment has to be amended by sequential chemotherapies for those
infants with (early) progression and risk for death.
ACKNOWLEDGMENTS
The study HIT-LGG 1996 was supported by the Deutsche
Krebshilfe (Grant No. 70-2288-Gn1). We acknowledge Mrs.
Breitmoser-Greiner for secretarial assistance and Mrs. Soellner
and Mrs. Geh for data-management.
REFERENCES
1. Qaddoumi I, Sultan I, Gajjar A. Outcome and prognostic features in pediatric gliomas. Cancer
2009;115:5761–5770.
2. Alvord EC, Lofton S. Gliomas of the optic nerve or chiasma. J Neurosurg 1988;68:85–98.
3. Gajjar A, Sanford R, Heideman R, et al. Low-grade astrocytoma: A decade of experience at St. Jude
Children’s Research Hospital. J Clin Oncol 1997;15:2792–2799.
4. Janns A, Grundy R, Cnaan A, et al. Optic pathway and hypothalamic/chiasmatic gliomas in children
younger than age 5 years with a 6-year follow up. Cancer 1995;75:1051–1059.
Children <1 Year With Low Grade Glioma
5. Packer RJ, Ater J, Allen J, et al. Carboplatin and vincristine chemotherapy for children with newly
diagnosed progressive low-grade gliomas. J Neurosurg 1997;86:747–754.
6. Deliganis AV, Geyer JR, Berger M. Prognostic significance of type 1 neurofibromatosis (von
Recklinghausen disease) in childhood optic glioma. Neurosurgery 1996;38:1114–1119.
7. Stokland T, Liu JF, Ironside JW, et al. A multivariate analysis of factor determining tumor progression in
childhood low-grade glioma: A population-based cohort study (CCLG CNS 9702). Neurooncol
2010;12:1257–1268.
8. Nicolin G, Parkin P, Mabbott D, et al. Natural history and outcome of optic pathway gliomas in children.
Pediatr Blood Cancer 2009;53:1231–1237.
9. Ater J, Zhou T, Homes E, et al. Randomized study of two chemotherapy regimens for treatment of lowgrade glioma in young children: A report from the Children’s Oncology Group. J Clin Oncol
2012;30:2641–2647.
10. Fouladi M, Wallace J, Langston JW, et al. Survival and functional outcome of children with hypothalamic/
chiasmatic tumors. Cancer 2003;97:1084–1092.
11. Laithier V, Grill J, Le Deley MC, et al. Progression-free survival in children with optic pathway tumors:
Dependence on age and the quality of the response to chemotherapy-results of the first French prospective
study for the French society of pediatric oncology. J Clin Oncol 2003;24:4572–4578.
12. Massimino M, Spreafico F, Cefalo G, et al. High response rate to cisplatin/etoposide regimen in childhood
low grade glioma. J Clin Oncol 2002;20:4209–4216.
13. Gnekow AK, Kortmann RD, Pietsch T, et al. Low grade chiasmatic–hypothalamic glioma-carboplatin
and vincristine chemo- therapy effectively defers radiotherapy within a comprehensive treatment strategy
—Report from the multicenter treatment study for children and adolescents with a low grade glioma—
HIT-LG 1996—of the Society of Pediatric Oncology and Hematology (GPOH). Klin Padiatr
2004;216:331–342.
14. Gnekow AK, Falkenstein F, von Hornstein S, et al. Long-term follow-up of the multicentre,
multidisciplinary treatment study HIT-LGG 1996 for low-grade glioma in children and adolescents of the
german speaking society of pediatric oncology and hematology. Neuro Oncol 2012;14:1265–1284.
15. Kleihues P, Burger PC, Scheithauer BW. The new WHO-classification of brain tumours. Brain Pathol
1993;3:255–268.
16. Gnekow AK. Recommendations of the brain tumour subcommittee for the reporting of trials. SIOP brain
tumour subcommittee. International Society of Pediatric Oncology. Med Pediatr Oncol 1995;24:104–
108.
17. Louis D, Ohgaki H, Wiestler OD, Cavenee WK, editors. WHO classification of tumours of the central
nervous system. Lyon: IARC Press; 2007.
18. Kleihues P, Cavenee WK, editors. WHO classification of tumours: Pathology and genetics of tumours of
the central nervous system. Lyon: IARC Press; 2000.
19. Kleihues P, Cavenee WK, editors. Pathology and genetics of tumours of the nervous system. Lyon: IARC
Press; 1997.
20. Korinthenberg R, Neuburger D, Trippel M. et al. Long-term results of brachytherapy with temporary
iodine-125 seeds in children with low-grade gliomas. Int J Radiat Oncol Biol Phys 2011;79:1131–
1138.
21. Ruge MI, Simon T, Suchorska B, et al. Stereotactic brachytherapy with iodine-125 seeds for the treatment
of inoperable low-grade gliomas in children: Long-term outcome. J Clin Oncol 2011;29:4151–4159.
22. Pollack IF, Claassen D, al-Shboul Q, et al. Low-grade gliomas of the cerebral hemispheres in children: An
analysis of 71 cases. J Neurosurg 1995;82:536–547.
23. Abdollahzadeh M, Hoffman HJ, Blazer SI, et al. Benign cerebellar astrocytoma in childhood: Experience
at the Hospital for Sick Children 1980–1992. Childs Nerv Syst 1994;10:380–383.
24. Burr IM, Slonim AE, Danish RK, et al. Diencephalic syndrome revisited. J Pediatr 1976;88:439–444.
25. Poussaint TY, Barnes PD, Nichols K, et al. Diencephalic syndrome: Clinical features and imaging
findings. Am J Neuroradiol 1997;18:1499–1505.
26. Gropman AL, Packer RJ, Nicholson HS, et al. Treatment of diencephalic syndrome with chemotherapy:
Growth, tumor response, and long term control. Cancer 1998;83:166–172.
Pediatr Blood Cancer DOI 10.1002/pbc
463
27. Perilongo G, Carollo C, Salviati L, et al. Diencephalic syndrome and disseminated juvenile pilocytic
astrocytomas of the hypothalamic-optic chiasm region. Cancer 1997;80:142–146.
28. Sardi I, Bresci C, Schiavello E, et al. successful treatment with a low-dose cisplatin-etoposide regimen for
patients with diencephalic syndrome. J Neurooncol 2012;109:375–383.
29. Hukin J, Siffert J, Velasquez L, et al. leptomeningeal dissemination in children with progressive lowgrade neuroepithelial tumors. Neuro Oncol 2002;4:253–260.
30. von Hornstein S, Kortmann RD, Pietsch T, et al. Impact of chemotherapy on disseminated low-grade
glioma in children and adolescents: Report from the HIT-LGG 1996 trial. Pediatr Blood Cancer
2011;56:1046–1054.
31. De Haas V, Grill J, Raquin MA, et al. Relapses of optic pathway tumors after first-line chemotherapy.
Pediatr Blood Cancer 2009;52:575–580.
32. Scheithauer BW, Hawkins C, Tihan T, et al. Pilocytic astrocytoma. In: Louis DN, Ohgaki H, Wiestler OD,
et al., editors. WHO classification of tumours of the central nervous system. Lyon: IARC Press; 2007. pp.
14–21.
33. Tihan T, Fisher PG, Kepner JL, et al. Pediatric astrocytomas with monomorphous pilomyxoid features
and a less favourable outcome. J Neuropathol Exp Neurol 1999;58:1061–1068.
34. Komotar RJ, Burger PC, Carsons BS, et al. Pilocytic and pilomyxoid hypothalamic/chiasmatic
astrocytomas. Neurosurgery 2004;54:72–79.
35. Merchant TE, Conklin HM, Wu S, et al. Late effects of conformal radiation therapy for pediatric patients
with low-grade glioma: prospective evaluation of cognitive, endocrine, and hearing deficits. J Clin Oncol
2009;27:3691–3697.
36. Kortmann RD, Timmermann B, Tayler RE, et al. Current and future strategies in radiotherapy of
childhood low-grade glioma of the brain. Part II: Treatment-related late toxicity. Strahlenther Onkol
2003;179:585–597.
37. Massimino M, Spreafico F, Riva D, et al. A lower-dose, lower-toxicity cisplatin-etoposide regimen for
childhood progressive low-grade glioma. J Neurooncol 2010;1001:65–71.
38. Avery RA, Ferner RE, Listernick R, et al. Visual acuity in children with low grade gliomas of the visual
pathway: Implications for patient care and clinical research. J Neurooncol 2012;110:1–7.
39. Fisher MJ, Loguidice M, Gutmann DH, et al. Visual outmes in children with neurofibromatosis type 1assiciated optic pathway glioma folling chemotherapy: A multicentre retrospective analysis. Neurooncol
2012;14:790–797.
40. Gnekow AK, Falkenstein F, Walker D, et al. SIOP-LGG 2004—Cohort description of a comprehensive
treatment strategy for low grade glioma in children and adolescents including a randomised
chemotherapy trail and a radiotherapy trial. Neurooncol 2012;14:1265–1284 (abstract).
41. Bokstein F, Shpigel S, Bulmenthal DT. Treatment with bevacizumab and irinotecan for recurrent highgrade glial tumours. Cancer 2008;112:2267–2273.
42. Vredenburgh JJ, Desjardins A, Herndon JE, et al. Phase II trial of bevacizumab and irinotecan in recurrent
malignant glioma. Clin Cancer Res 2007;13:1253–1259.
43. Packer RJ, Jakacki R, Horn M, et al. Objective response of multiply recurrent low-grade gliomas to
bevacizumab and irinotecan. Pediatr Blood Cancer 2009;52:791–795.
44. Peyrl A, Azizi A, Czech T, et al. Tumor stabilization under treatment with Imatinib in progressive
hypothalamic–chiasmatic glioma. Pediatr Blood Cancer 2009;52:476–480.
45. Pfister S, Janzarik WG, Remke M, et al. BRAF gene duplication constitutes a mechanism of MAPK
pathway activation in low-grade astrocytomas. J Clin Invest 2008;115:1739–1749.
46. Forshew T, Tatevossian RG, Lawson AR, et al. Activation of the ERK/MAPK pathway: A signature
genetic defect in posterior fossa pilocytic astrocytoma. J Pathol 2009;218:172–181.
47. Jones DTW, Gronych J, Lichter P, et al. MAKP pathway activation in pilocytic astrocytoma. Cell Mol Life
Sci 2012;69:1799–1811.
48. Karajannis MA, Fisher MJ, Milla SS, et al. Phase II study of sorafenib in children with recurrent/
progressive low-grade astrocymas. Neuro Oncol 2012;14:1163–1170 (abstract).
49. Clinical trials gov. Service of the U.S. national institute of health. NCT 01748149 http://www.
clinicaltrials.gov/ct2/show/NCT01748149